. Vol Class J\o Cost | |^■ l ■^■^■ l ■^l l .^■l■^l l l l ■ l ■l■ l .^■ l l^. l ^i^il ^llli^ ■■■^■^■'■ll^l i i l |^ ^ « > » ^ ^ ^ suois 'sauBjqn in3tl3auuo3 jo itjisjaAiun 3H1 •3JBD miM auinyoA Slip aipuEn .... ■ "^ ■ v' Mil, ' , ^ ' Digitized by the Internet Archive in 2009 with funding from Boston Library Consortium Member Libraries http://www.archive.org/details/geologyofoldhampOOemer DEPARTMENT OF THE INTERIOR MONOGRAPHS OF THE United States Geological Survey VOLUME XXIX WASHI^-GTON GOVERNMENT PRINTING OFFICE 1898 io^^*2a / Vk . UNITED STATES GEOLOGICAL SURVEY CHARLES D. WALCOTT, DIRECTOR G-EOLOGT OF OLD HAMPSHIRE COUNTY, MASSACHUSETTS COMPRISING FRANKLm, HAMPSHIRE, AND HAMPDEN COUNTIES BT BElSTJAMlIlSr KETSTDALL EMERSON WASHINGTON" GOVERNMENT PRINTING OFFICE 1898 CONTENTS Letter of TRA^SMITTAL xxi Chapter I. — Introduction 1 Area covered 1 Historical sketcli 2 Chapter II. — Topography 8 Chapter III. — -Geological outline and general comparative sections 12 General comparative section of rocks in Maasacliusetts , 16 Chapter IV. — The Algonkian 19 Geological description 19 The Hinsdale area 19 The Coles Brook anticline 21 The Tolland area 24 Petrographical description • 24 Lower or Hinsdale gneiss, Hinsdale station 24 The Hinsdale limestone, Hinsdale 25 The Coles Brook limestone 27 The blue- quartz gneiss, Peru 28 The Lee gneiss 29 E^sumiS 30 Chapter V. — The Lower Cambrian gneisses 31 The Becket conglomerate-gneiss >^ 31 Contact u.pon the Washington gneiss below 31 Description of the rock 32 Distribution 33 Petrographical description 34 Crushing tests 36 The gneiss at Shelburne 38 The Monson gneiss and associated rocks 41 The Pelham and Wilbraham area 42 The gneiss 42 Petrographical description 43 The actinolite-quartzite 45 Petrographical description 46 T Vi CONTENTS. Chapter V. — The Lower Cambrian gneisses — Continued. Page. Tlie Monson gneiss — Continued. The Polham and Wilbraham area — Continued. Saxonite and serpentine in Monson gneiss 47 • The Pelham asbestos quarry 47 Petrographioal description 52 The Pelham serpentine 55 The Shutesbury serpentine 55 The New Salem serpentine 55 The Orange and Moosou area 56 General description 57 Petrographical description 59 The Monson quarry 60 Strength of the Monson gneiss 63 Conglomerate structure of the Monson gneiss and sudden expansion of the rock in quarrying 63 A complex mineral vein in the gneiss 65 Chapter VI.— Lower Silurian sericite-schists and amphibolites on the west side of the valley. 66 The Hoosac schist = the albitic mica-schist 66 The Monroe area 67 The Middlefield area 70 Section along the Boston and Albauy Railroad 71 Relation to the Becket gneiss 72 The Grranville area 73 Hornblendic bands in the albitic mica-schist 75 The Shelburne Falls anticline 75 The Eowe schist^ the lower sericite- or hydromica-schist 76 Franklin County 76 Hampshire County - 70 Hampden County 77 Thickness 78 The Chester amphibolite and serpentines . 78 General description 78 Franklin County - 79 The Eowe serpentine 79 The East Portal fault - 80 Hampshire County 81 The Middlefield serpentine - 81 Hampden County 85 The Chester amphibolite and serpentine 85 The Blandford serpentines and pyroxenite 85 The Granville and Enssell enstatite-serpentines 90 The Westfield serpentine and marble 92 Faults and serpentinization 95 Petrographical description 96 The amphibolites 96 The serpentines and associated magnesian rocks 97 Bladed serpentine ; antigorite- (or bastite-) serpentine 98 CONTENTS. Vll Page. Chapter VI.— Lower Silurian serioite-schists and amphibolites on the west side of the valley — Continued. The Chester amphibolito and serpentines — Continued. Petrographical description — Continued. The serpentines and associated magnesian rocks — Continued. Olivine- and enstatite-serpentine 101 R€8um6 114 Table of analyses of serpentines 116 The Chester emery bed 117 History of discovery and working of the bed 117 A description of the emery mine of Chester, Hampden County, Massachusetts, by Charles Upham Shepard 122 General description 135 Association and paragenesis of the minerals of the emery vein 143 Re'sumt? of paragenesis - 147 General explanation and correlation of the Chester amphibolite series — 147 Original condition of the enstatite-serpentine and limestone complex 147 The Savoy schist = the upper sericite-schist 156 Distribution 156 Boundary upon the rocks below 156 General description of rocks 157 Comparison with the Rowe schist below ..- 158 Detailed description and sections 158 The Shelburne anticline 162 Petrographical description 162 Intrusive rocks - 163 TheHawley schist 163 Distribution 164 Detailed description 164 Petrographical description 166 The possible igneous origin of the Hawley schist 169 Mineral deposits 170 The pyrite beds 170 Copper ores 171 The great Hawley fault and the magnetite and hematite deposits, the rhodonite and rho- dochrosite beds, and the garnet-schist or coticule 171 The Goshen anticline 175 Chapter VII. —The graphitic mica-schist series on the west side of the valley 177 The Goshen schists or flags 177 General description 177 Unconformable contact on the rocks below ; outliers in the Hawley schist 179 Petrographical description 181 The Conway schists, or the corrugated mica-schists 183 General description 183 Subordinate beds in the Conway schist 185 The gneiss beds 185 The whetstone-schist 186 Petrographical description 187 viii CONTENTS. Chapter VII. — The graphitic mica-schist series on the -west side of the valley — Continued. Page. The Conway schists, or the corrugated mica-schists — Continued. Suhordinate beds in the Conway schist — Continued. The limestone beds 188 The amphibolite beds 189 TheConwaybed 189 The Whately bed 190 The Whitmores Ferry bed 190 Petrographical description of limestone and amphibolite; the limestones, the anvils, passage of limestone into amphibolite 191 Analyses of the amphibolites 195 Projection of the limestone and amphibolite of the Conway schist through the Leyden argillite in Whately 196 Contact metamorphism of the limestone by granltite ; argentine 197 Cleavage in the Conway schists 199 Fossils (?) of the Conway schists 200 The Leyden argillite 201 Description 201 Quartzite in the argillite 202 Petrographical description 202 Stratigraphy 203 Boundary on the Conway schists 203 Argillite on the western border of the ' ' graphitic mica-schist " (Goshen schist) 204 Relative .age of the Conway schist and the Leyden argillite 204 Contact metamorphism of the Leyden argillite bordering the tonalite of Hatfield.. 205 The sericite-gneiss 206 The chiastolite-schist 209 Chapter VIII. — The bands of Silurian schists on the east side of the valley 211 The Northfield semisyncline ■- 212 General description 212 The Gulf road sections 213 Sections north and south of the old Warwick road 215 Pegmatite dikes and minerals - 216 The Wendell branch syncline 217 The Leverett-Amherst area 218 The amphibolite and mica-schist series along the east side of the Connecticut basin from Leverett southward 218 North Leverett (Greenfield quadrangle, southeast comer) 219 Leverett Center — . 220 The Savoy schist, or whetstone-schist 220 The Amherst feldspathic mica-schist (Conway schist) 222 Leverett 222 Amherst — 222 The Pelham-Shutesbury syncline 225 The great central syncline 227 Warwick and Orange 227 Topography 230 CONTENTS. IX C'liAi'iKK VIII. — Tbr biinils ofSiluriuu sc-hists ou the east side of the valley — Continued. Ta^c The great central syucline — Coutiuued. .South Oranyo and Now Salem 230 I'rescott and Knfleld 232 Structure 232 Petrographical descriptions 233 The eastern sy ncline 234 Orange and Athol 234 General description 234 Metamorphism of the amphibolite hand as it is involved in the grauitite of the Athol hatholite, and its later change to steatite 236 Ware 237 General description 237 Petrographical description 238 The Hardwick gneiss 239 Petrographical description 240 Palmer 241 Monson - 241 The zone of contact around the Belchertown tonalite 243 The pyroxenic amphiholites 243 The fibrolite-schist inclusions 246 Petrographical description 246 The Wilbraham syncline 248 The Monson syncline 249 The East Greenwich-Enfield syncline 251 Efeum6 251 Argument for the identity of the schist series east of the Connecticut with those ou the west 251 The passage eastward into the Brimfield fibrolite-schist 252 Chapter IX. — The Bernardston series of Upper Devonian rocks 253 Literature 253 History 254 Upper Devonian age of the Bernardston fossils 259 Description of the region 260 The relation of the Bernardston series to the argillite 261 The AVilliama farm section; the fossiUferous limestone; proof that the whole series is Devonian 262 Description of the range from Bernardston to South Vernon 272 The feldspathic quarfczite 282 The Bernardston series east of the Connecticut 284 The original character of the series and its metamorphism 285 Petrographical description 287 The quartzite series 287 Amphibolite associated with the limestone in the gneissoid quartzite 290 The mica and amphibolite series - 291 Rooks at the mouth of Millers Eiver 295 X CONTENTS. Page. Chapter X.— The ampliibolites described in the preceding chapters 300 Analyses and sections ^'"' Porphyritic character of the amphibolites 304 Chapter XI. — The eruptive rocks 307 Introduction 3"' Historical notes on the mica-granites 312 Biotite-muscovite-gran ito ^^^ Areas west of the Connecticut 31'^ Distribution ^1* Petrographlcal description ■'^l^ Chemical analysis 315 The Athol area 316 Secretions and inclusions 317 The Hardwick gneissoid granite and granitlte 317 Biotite-granite, or granitite 318 Contact metamorphism of the granitite and schists 318 The Middlefield porphyritic granitite 318 The Coy's Hill porphyritic granitite 319 Description and distribution 319 Cordierite-granitite 321 Muscovite-granite, or pegmatite 322 Probable extreme modification of the pegmatite by crushing 323 Albitic granite and pegmatite dikes containing rare minerals 323 Distribution and description 324 The great tonrmaline-spodumene dike 324 Dikes in Goshen 3-'6 Dikes in Chester, Blandford, and Huntington 327 Dikes east of the Connecticut 327 Garnet in pegmatite with complex paramorphic border of zoisite-hematite, epidote- fibrolite, and muscovite 3-^8 The crushing of minerals in the albitic granite 329 Hydrothermal changes in the albitic granite veins 329 Ordinary meteoric alteration 330 Aplite - 331 Quartz-gabbro and quartz-diorite, or tonalite 331 Historical 331 Basic secretions : Hitchcock's suggestion of the theory of " schlierengUnge " 331 Distribution 335 Analyses of tonalite 336 Petrographical description 336 The crushing and alteration of the tonalite along the Pelham fault 339 Petrographical description of the altered tonalite 341 Diorite 342 Garnet-biotite-norite Cortlandite 346 QAQ Age of the granites Resume as to the genetic relations of the granites 348 Contact effects of the eruptive rocks 349 CONTENTS. XI Page. Chaptbr XII.— Tlio Trias ■ 3.51 The Coniiocticut Rivoi- sandstono 3.51 General section of Triassic rocks 354 The Sugar Loaf arkose, or the feldspathlc sandstone and conglomerate 354 Contact and distribution 355 The Mount Toby conglomerate, or the slate and quartzite conglomerate 358 Contact and distril>ution 358 The outcrops of crystalline rocks in the midst of the Mount Toby conglomerate ... 361 Action of ice in the Trias 363 The Longmeadow sandstone 364 Fragments of white trap without augite in the sandstone above the Holyoke sheet. 365 Disturbances in the sandstones and inclusions of trap fragments just below the posterior sheet 367 The boundary of the sandstone 368 Analyses 369 The Granby tuff, or the diabase-tuff 369 The Chicopee shale, or the calcareous shale 370 The continuation of the State-line fault in a crushed band at the Holyoke dam and the secondary minerals found in the fissures 370 The diabase 372 The formation of the basin and the distribution of the sediments by strong tidal currents.. 372 The possible connection of the foot tracks with the trap sheets 379 Artesian wells - 380 Pseudomorphs of caloite and dolomite after hopper-sh aped cubes of salt 389 The use of the Triassic sandstone as a building stone 391 Paleontology 394 Plants 394 Insects 398 Fishes 398 Ichnology '*^'^ Eecent progress in Ichnology, by C. H. Hitchcock 400 Reptiles *05 Chapter XIII. — The Triassic eruptive rocks '107 Historical '^07 The three epochs of eruptive activity ; general account 4:10 Diabase dikes and stocks in the gneiss east of the Trias *ll A microscopical diabase dike from Pelham, and olivine and glass-bearing dikes from Monson - *1° The bedded or contemporaneous eruptives '^^° The Deerfield sheet - ^^^ Contact on the sandstone below; the underrolling of the crust and the alteration of the diabase by heated waters to a pitchstone-breccia and a diopside-plagioolase rock '^^^ General character 4-^4 Greenfield quarry exposures and contacts - ^"^ Petrographical description 43^ Diabase-pitchstone ^32 XU CONTENTS. Chapter XIII. — The Triassic eruptive rooks — Continued. Page. The bedded and contemporaneous eruptives— Continued. The Deerfield sheet — Continued. Contact on the sandstone below, etc — Continued. Petrographical description — Continued. Glass-breccia 433 Anygdaloidal sandstone 435 Contact material 436 LithophysEB 436 Chemical discussion 436 Origin of the glass and minerals 437 Contact of the sandstone upon the diabase 439 Fall River fault 439 The unity of the sheet 440 Petrographical description 441 Paragenesis of secondary minerals 444 The Holyoke sheet 446 The faults at Jfount Tom and southward 449 General characteristics of the sheet 451 Normal contacts of diabase on sandstone 452 Contacts of underroUed diabase inclusions of limestone 452 Petrographical description 453 Normal contact of the sandstone on the diabase 455 Contacts of sandstone on diabase which is kneaded full of limestone and shale 456 Section of trap filled with limestone fragments on the Westfield-Holyoke Railroad. . 456 Magmatic differentiation 459 Origin of the clay and marl deposits 459 On the uuderroUiug of the solidified surface of the trap 460 Petrographical description of the normal diabase 461 Chemical composition of the trap 463 The upper or posterior sheet and its feeding dikes 464 The great widening of the trap area and the feeding throat beneath 467 Sills intruded in the sandstones below the posterior sheet 469 Delaney 's quarry, near the north line of Holyoke 470 The Roaring Brook fault and the disappearance of the posterior sheet 473 The blending of the tuff with the surface of the posterior bed 474 A tufiTaceous sandstone containing white trap 474 The posterior dike across Hampden County 475 The Talcott sheet 476 The tuff and tuffaceous agglomerates 476 The Deerfield bed 476 TheGranbybed 476 The isolated mass of tuff north of the seventh core 479 Source of the material of the tuff bed 480 A hollow bomb from Delaney's quarry, Northampton 480 Petrographical description 480 The newer series of cores and short dikes 481 CONTENTS. Xiii Chapter XIII. — The Triassic eruptive rocks — Continueil. Page. Tlie newer series of cores and short dikes — Continued. Belchortown 481 Grauby 482 South Hadley 483 The ninth core of diabase, with granitic inclusions 483 Petrographical description 484 The eleventh or Black Rock core 489 Petrographical description 492 Northampton 494 Summary of the history of the Connecticut River sandstone 49.5 The use of the trap as road material 500 Chapter XIV. — Mineral veins 502 Chapter XV. — The Pleistocene period 508 Literature 508 The interval between the Triassic and the Glacial period 508 Deposits 508 Pre-Glacial weathering 509 Pre-Glacial drainage and erosion 510 Pre-Glacial course of the Connecticut and Its tributaries 513 Character and amount of the erosion during later Mesozoic time as compared with that of the Glacial period 515 Chapter XVI.— The Glacial period 518 The present rook surface and the amount of Glacial and post-Glacial material on the same. . 518 Glacial grooves and striae 522 Glacial notches 529 Pseudo-glacial strise on Devonian argillites 531 Potholes 532 Thetm 533 Introduction 533 The upland drift 535 The fine valley drift of the east side of the valley 537 The coarse valley drift 541 Distribution of the coarse valley till west of the river 542 Drumllns 543 Moraines and bowlder trains 549 Interglacial sands . - : 550 The upper till 558 Remarkable bowlders 559 Chapter XVII. — The Champlain period 562 Glacial lakes east of the Connecticut River 562 Introduction 562 Ice barriers 565 The Brlmfield Lake 565 The Monson esker 566 The Monson drainage 567 The eastern Palmer and Monson Lake 567 Xiv CONTENTS. Chapter XVII. — The Champlain period — Continued. Page. Glacial lakes east of the Connecticut Eiver — Continued. The Ellis Mills drainage 569 The Palmer Lake 569 The Ware and Swift Eiver lakes 569 The Chicopee Eiver drainage 575 The Belcherto wn Lake 575 The Pelham Lake and esker 578 The Iladley Lake drainage 584 The Leverett Lake and the Notch east of Mount Toby 584 The Locks Pond Lake 588 Notches through the Holy oke range and the range north of Moody Corners 586 TheGranby Eoad Lake 587 The Notch 587 The low place and Moody Corners Lake 587 The Pelham Eiver and the "Moraine Terrace" sands along the eastern valley side, just above the level of the high terrace 588 The Sunny Valley Lake 592 The sands along the west side of Mount Tom Eange and in the Westfield basin above the level of the high terrace 592 Chapter XVIII.— The Champlain period (Continued) 593 Glacial lakes west of the Connecticut Eiver 593 The Granville Lake 593 TheNorth Granville Lake 593 The Westhampton Lake 594 The Williamsburg Lake 595 The Beaver Brook Lake above Leeds 595 The Deerfield Eiver lakes 595 The Deerfield Eiver and its tributaries on the north ■- 597 The Conway Lake 598 The Bear Eiver Lake --- 600 The Ashfield Lake 601 The Buckland Lake 602 The last important halting place of the ice front across the basin of the Deerfield Eiver. 604 Glacial lakes north of the Deerfield Eiver 604 High level deltas 605 The character of the terraced flood deposits of the Westfield Eiver 607 Chapter XIS. — The Champlain period (Continued) 609 The Connecticut Eiver lakes 609 Introduction 609 Detailed description of the flood deposits in the Montague basin 615 The northern lobe of the lake , 616 The Bennetts Brook plain, or moraine terrace 617 The extension of the flood gravels westward through the Bernardston Pass 619 The old course of Fall Eiver 621 The bench on the east side of the river in Northfield and Erving 622 The Millers Eiver delta ; the canyon and old course of the Connecticut 625 CONTENTS. XV Chapter XIX. — The Chiimplaiu })orio(l — Continued. Page. The Conuectiout River lakes — Continued. The Iliidley Lake g29 The uortli end of the lake in Greenfield and the channel of connection with the main valley 629 The Green River glacier 630 The Factory Village channel 632 The high terrace plains in the south of Greenfield and the north of Ueerfield 632 The lake bench from Deertield River south 634 The Deerfield delta 634 The West Brook delta 635 The Mill River delta in Northampton 637 The lake hench on the east side of Hadley Lake in Leverett and Amherst 639 The delta of Cushmans Brook at North Amherst and the isolation of the East Street hasin in Amherst 640 The hench surrounding the East Street basin 641 Shore notches in the sides of drumlins 642 The high terrace or bench along the west side of the Amherst ridge 644 The bench around Mount Warner 648 The bench along the north slope of the Mount Holyoke and Mount Tom range 649 The Westfield plain 650 The greater elevation of the terraces in the Westfield than in the Springfield Lake ; possible western elevation 654 Geology of Westfield and vicinity, by J. S. Diller 654 The Springfield Lake 657 The "gorge terrace" of Dry Brook Hill in the north part of South Hadley 661 The high terrace of the west side of the river from the Holyoke notch southward . , 662 The similarity of the Belchertown notch to the notch east of Mount Toby 663 The moraine across the southern part of the Granby plain 664 Kettle holes and the old bed of the Connecticut i 664 Kettle holes and the structure of the high-terrace sands; their origin from the melting of ice beneath the terrace gravels 665 Lake bottoms 672 The Montague Lake 672 The Hadley Lake 673 The Springfield Lake 677 Detailed sections of terraces and lake bottoms, showing several advances of the ice front _ _ .,...' _ 677 The Camp Meeting cutting 677 Section of clays in Hatfield, showing great disturbance and pressure cleavage 691 The Wapping cutting ; 695 Chaptek XX.— The Champlain period (Continued) 69? The Champlain clays 697 Introduction 697 The Montague Lake 697 The Hadley Lake 698 The Springfield Lake 701 Contact of the clays upon the till 701 xvi CONTENTS. Chapter XX. — The Champlain period — Continued. Page. The Champlain clays — Continued. The Springfield Lake— Continued. The structure of the clays 703 The surface of the layers 704 The lateral passage of the clays into the high terrace sands 705 The passage of the clays into the sands above 705 Explanation of the structure of the clays 706 The time occupied in the deposition of the clays 707 Action of icel)erg8 and floes upon the clays 707 Secondary structures in the clays 709 Joints 709 Concretions - 711 Fossils of the Champlain clays 718 Chapter XXI. — The terraces of the Connecticut and the modern deposits 722 Introduction 722 The intermediate terrace and harrier at Lily Pond in Gill; an abandoned waterfall 724 The low-level terraces and flood plain of the Connecticut in the basin of the Montague Lake. 725 The later terraces or meadows of the Connecticut in the Hadley Lake 726 The structure of the terraces 727 The river sands 727 The muck sands 728 The peat deposits and the plant remains 728 Loess 729 The terraces of the Connecticut in the Springfield basin 729 The incomplete terraces as illustrations of the stages in the growth of terraces 731 On the oscillations of the Connecticut from its earliest position 733 The oxbows of the Connecticut 734 On the deflection of streams toward the right hank 734 Eiver terraces around a receding waterfall 735 The terraces of tributaries 736 An old oxbow of Fort Eiver 737 Fossils of the terrace period 738 The Pleistocene beetles of Port Eiver, Massachusetts, by S. H. Scudder 740 The repulsion of tributaries 746 Dunes and wind loess 747 Mineral springs - 749 Thick modern fissure deposits of quartz surrounding roots in the base of the Holyoke trap sheet. 752 Chapter XXII. — Supplement to the author's mineral lexicon of Franklin, Hampshire, and Hampden counties 754 Chapter XXIII. — Chronological list of publications on the geology and mineralogy of Frank- lin, Hampshire, and Hampden counties 762 Index 783 LLUSTRATIONS, Page. Plate I. Coign of Williston Hall at Amherst College, sbowiug conglomerate-gneiss from the Mouson (luarry 64 II. Thin sections 106 Fig. 1. Sahlite changing to tremolite 106 2. Dolomite changing to serpentine 106 3. Enstatite crystal altered to serpentine, cut parallel to (001 ) 106 4. Garnet, with complex border, from pegmatite 106 III. Thin sections , 208 Fig. 1. Leydeu argillite changed to chiastolite schist in contact on tonalite 208 2. Covdierite twins, frohi cordierite-granite 208 3. Diorite, from Packards Mountain, Prescott 208 4. Contact of diabase-amygdaloid and clayey limestone 208 IV. Map of the Devonian rocks of the Bernardston series and of the fanlted syuclini'. of Silurian schist in Northfleld Mountain 260 V. Sections of amphibolites derived from limestone 302 VI. Sections of amphibolites probably derived from limestone 306 VII. Tourmaline dendrites in granite, Leeds 316 VIII. Vertical wall of diabase at the quarry for road material in the east of Greenfield 424 Villa. Details of trap ridge east of Greenfield 426 VIII6. Inclusion of mud in upper surface of trap sheet 428 VIIIc. Thin sections of material from Greenfield and Meriden "ash bed" 430 IX. Geological map and sections of the Mount Holyoke-Mount Tom range, with the pos- terior diabase sheets, the tuff, and the A'olcanic cores 446 X. Batterson's quarry, in north part of South Hadley, showing veneering of sandstone on Black Rock core 488 XI. Map of preglacial -drainage and drift strite 510 XII. Sections in Amherst House cellar, showing interglacial beds 550 XIII. The great serpent esker in Pelham 578 XIV. Diagram of the lake-shore and lake bottom profiles of the Connecticut lakes 656 XV. Sections at the Camp Meeting cutting, on the north line of Northampton, showing the readvances of the glacial ice. 678 XVI. Surface of ice-contorted clay, smoothly cut with a knife, east of J. Ryan's house, Hat- field - 690 XVII. Joints and faults in laminated clay, produced by the weight of the ice 692 MON XXIX ii s;vii Xviii ILLUSTBATIONS. Page. Pl. xviii. The Wapping and Camp Meeting cuttings 694 Fig. 1. Section of ane-graiued contorted sands at the Wapping cutting on the Canal Kailroad, in Deerfield 694 2. Section on the west side of the Camp Meeting cutting 694 3. Detail from point above (i on Plate XV, below fourth ice-worn surface 694 XIX. Champlain clays, distorted by floe ice, Northampton 708 XX. Calcareousconcretionswith wormtracks,Champlainclays,HadleyandNorthampton. . 716 XXI. Profile of the Connecticut Eiver from Vernon, Vermont, to Hartford, Connecticut, showing high and low water and the river bottom. 722 XXII. View across the Connecticut Eiver, showing the notches formed by the river at the Lily Pond, in Gill, and its escape around The Narrows 724 XXIII. Pleistocene beetles of Fort Eiver, Massachusetts; S. H. Scudder 742 XXIV. Geological sections along Hues I to IV, dj'awn on the northwestern portion of the generalmap (PI. XXXIV) 782 XXV. Geological sections along lines V to VIII, drawn on the western portion of the general map (PI. XXXIV) 782 XXVI. Geological sections along lines IX to XII, drawn on the southwestern portion of the general map (PI. XXXIV) 782 XXVII. Geological sections along lines XIII to XVI, drawn on the northern portion of the general map (PI. XXXIV) 782 XXVIII. Geological sections along lines XVII to XIX, drawn on the central portion of the general map (PI. XXXIV) 782 XXIX. Geological sections along lines XX to XXIII, drawn on the southern portion of the general map (PI. XXXIV) 782 XXX. Geological sections along liues XXIV to XXVII, drawn on the northeastern portion ofthe general map (PI. XXXIV) 782 XXXI. Geological sections along lines XXVIII to XXXII, drawn on the eastern portion of the general map (PI. XXXIV) 782 XXXII. Geological sections along lines XXXIII to XXXVI, drawn on the southeastern por- tion of the general map (PI. XXXIV) : 782 XXXIII. "Anvils," formed by the unequal erosion of blocks of impure limestone (plate wrongly numbered) 1"2 XXXIV. Geological map of Franklin, Hampshire, and Hampden counties In pocket XXXV. Map of the .surface geology In pocket Fig. 1. Algonkian section at Coles Brook 22 2. Detailed section of the limestone at Coles Brook 23 3. Southwest wall of Pelham asbestos quarry in 1890 4:8 4. West wall of Pelham asbestos quarry 4:9 5. Section at Osborn's soapstone quarry, Blandford 87 6. Map of emery veins in epidote-amphibolite at north end of bed on the bank of the Westtield Eiver, Chester 136 7. Section of old emery mine, Chester - 141 8. Stellate marble; Westtield Marble Company's quarry, Eussell 152 9. Plan of altered dikes and quartz veins in chlorite-schist, Charlemont 169 10. Contorted layer of garnetiferous quartzite (coticule), from mine on Forge Hill, Hawley. . 174 11. Map showing the protrusion of the limestone ofthe Conway schist through the Leyden argillite, Whately 1^^ ILLUSTRATIONS. XIX Pago. Fig. 12. Surface of h\ark limestone with oolltol■t(^(l (juarfcz voins, Wliately. '. V.W 13. Section ou lailinad oast of Erving station ^17 14. Section of schists west of Belchoitown 244 15. Map of Devonian locks on the Williams farm, Bernardston 263 16. Section of Devonian rocks from the Williams farniliouse 250 rods northwest 264 17. Section of the Williams farm quarry 264 18. Section at uortli cud of limestone, Williams farm 266 19. Section across Bernardston series on Purple blind road, Bernardston 278 20. Sketch uuip of rocks near the mouth of Millers River, Erviiig - - . 295 21. Sketch of rocks at mouth of Millers Eiver, looking northeast from B, iig. 20 295 22. Section on east bank of the Connecticut above mouth of Millers Eiver, at A, fig. 20 . -- 296 23. Section at the Holyoko dam, showing the passage of a fault through the shales 371 24. Thin sections of sand and glass breccia from the base of the Greenfield sheet at the City quarry and of trap from Cheapside 422 25. View of the posterior trap sheet and its feeding dike at Little Mountain, in Forest Park, Northampton ^66 26. Section of Uelaney's quarry on the Connecticut Eiver Eailroad in Northampton, near north line of Holyoke 4:70 27. Section of contact of Black Rock plug and the Mount Holyoke diabase bed 490 28. Holyoke notch from Hadley meadow ; pre-Glacial rock terr.aces 510 29. Glacial groove on compact diabase, Prospect House, Mouut Holyoke 527 30. East slope of a large glaciated groove behind the bowling alley on Mount Holyoke 530 31. Section showing the striai on the surface of sandstone continued on the surface of the till, Hoe Factory, Northampton 540 32. Pelhnui Lake section 578 33. Section of terminal moraine covered by high-level flood gravels of Westfield River, at Russell - 607 34. Sand bowlders, in terminal moraine, crushed by the ice while frozen 607 35. Section throiigh the eroded front of the great delta at Montague City 629 36. Section of the Green River delta at north end of Green River basin 631 37. Section of the shore beds of Hadley Lake south of College Hill, Amherst 646 38. Enlarged section of the south side of the cutting shown in fig. 37 647 39. Detail of clay layer crumpled by the current, from fig. 38 648 40. Sections south of Millers Falls station, showing- kettle-holes formed by ice melting from beneath the sands - 666 41. Section south of Millers Falls, showing kettle-hole formed by ice stranded ou the sur- face of the sands 668 42. Section of north half of kettle-hole below D wight's station, Belchertown 669 43. Section at the south end of North Pond, Belchertown, showing part of a kettle-hole at the north end and of an erosion slope at the south end 670 44. Section of kame sands at the north end of the " big fill," south of Itwight's 671 45. Block of frozen "pink sand," showing fine system of joints - 681 46. Pharyngeal hone of a fish, from the Champlain clay, Holyoke 721 47. Sketch of the point of the Northampton meadow from Mount Holyoke. showing that the meadow is a composite of many islands 726 48. An old oxbow of Fort River cut by the Connecticut below Hadley 737 LETTER OF TRANSMITTAL. Amherst College, Amherst, Mass., March 4, 1895. Sir: I have the honor to transmit herewith the manuscript of a geo- logical description of the three counties in Massachusetts through which the Connecticut River runs, and which include nearly the whole of that por- tion of its drainage area which lies within the limits of the State. The studies here presented began in 1873. The results were offered to the United States Greological Survey in 1887, and were accepted at that time. A small portion of the area has since been reexamined, under the direction of the Survey, and the map has been extended a Httle beyond the limits of the State north and south, to cover the whole of the area repre- sented on the topographic sheets employed. In this work I have been assisted by Mr. William Orr, jr., of Springfield, who has mapped part of the limestone and amphibolite bands of the Conway schist, and by Mr. Fred- erick B. Peck, who traced the western boundary of the Shelburne anticline and worked on the southern border of the area mapped on the Granville sheet. As the work was mostly done before the appearance of the topographic maps issued by the Sm-vey, many allusions to names found upon county atlas maps remain. Very respectfully, your obedient servant, B. K. Emerson, Geologist. Hon. Charles D. Walcott, Director United States Geological Survey. GEOLOGY OF OLD HAMPSHIRE COUNTY, MASSACHUSETTS, COM- PRISING FRANKLIN, HAMPSHIRE, AND HAMPDEN COUNTIES. By Benjamin Kendall Emerson, Pkofessor of Geology in Amherst College. CHAPTER I. INTRODUCTION. AKEA COVERED. Old Hampshire County, which formerly stretched across the State of Massachusetts between Berkshire on the west and Worcester on the east, has been less fortunate than these and has lost Franklin County on the north and Hampden on the south. Amherst lies in the center of this area, and hence it has come about that for many years the region has been the field of my geological studies. The rocks strike north and south and run quite across New England and beyond, so some artificial limits had to be chosen in these directions, and the limits of the State were as convenient as any. On the east and west, the area lying between the plateau of Worcester County on the east and the full development of the Berkshire Hills country on the west pos- sesses a good degree of geological unity, the Cambrian gneiss of its eastern and western boundaries being almost certainly continuous beneath the whole area and supporting several series of schistose rocks, which cidminate in the Bernardston highly metamorphosed but fossiliferous beds of Devonian age. The area includes, also, the northern half of the Triassic terrane, which reaches nearly to the north line of the State, while the sudden widening of the valley of the Connecticut just at this northern point, with the lowering MON XXIX 1 1 2 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. of its borders, occasions a much greater degree of complexity in its post- Glacial deposits, the great series of Glacial lakes on itfj eastern side being just within the limits of the State, and the division of the valley into two portions by the sandstone and trap ranges from Mount Tom southward being wholly confined within the same limits. So that the area has given me a section of sufficient length for my purpose in those rocks which are of great extent meridionally, and a goodly number of problems of which all the factors are within its limits. HISTORICAL SKETCH. While many a quaint and appreciative remark may be gathered from the records of the explorers and early settlers of the Connecticut Valley concerning the great natural beauty of the new country, I have after much search found nothing which had reference to its geological structure. Considering the little that was then known, even among the learned, con- cerning geology, we do not wonder at this. It is more a matter of regret that they so generally failed to retain the Indian names of the prominent landmarks, or to replace them by significant or euphonious svibstitutes. Except the name of the Connecticut^ itself, I know of few Indian names retained from the beginning in their proper application, and but few descrip- tive and appropriate names which have come down to us from the fathers. Among these are "The Notch" and "The Low Place" in the Holyoke range, and "Sugar Loaf," named, I doubt not, by the Hadley farmers who rowed over to mow the Hatfield meadow, whence its conical shape is most striking and suggestive. It is true that in late times the names Agawam and Chicopee have been applied to towns, Mittineague and Willimansett to villages, while in a, few cases the Indian names of brooks seem to date far back, as Chicopee, Quinebaug, Quabaug, and Scantic rivers, Pecowsick and Watchaug brooks, and Massasoit Pond. President Hitchcock attempted to baptize several of our peaks with Indian names; i. e., Nonatuck, Norwottuck, transferred from the Northampton Meadows to the peak overlooking them, and Metawampe, from the name of an Indian who deeded the region to the whites. With ' Quiu m tuk=long- tidal river: Coll. Conn. Hist. Soc, Vol. II, p. 8. Quon eli ti cut=tlie long river: Trumbull Hist. Coun., Vol. I, p. 32. HISTORICAL SKETCH. 3 these exceptions the Indian names of the region have largely passed into the possession of hotels and manufacturing companies. There is, on the other hand, a great poverty of names for all the natural features of the country, "mount" and "hill," "brook" and "river," serving rather indis- criminately for all elevations and streams. One longs for the rich vocabu- lary of Spain and Scotland. Again, the names given are often trivial and constantly repeated. All the larger streams have an east, west, and middle branch, and I remember hearing one brook called the "West Branch of the Middle Branch of the Westfield River." There are several "Swift" rivers, "Roaring" brooks, "Muddy" brooks, and eleven "Mill" rivers (and brooks) within the limits of the three counties. The early settlers had little appreciation of the natural beauties of the landscape, or they would not have offended the poetical ear of President Hitchcock by naming our finest peaks Mount Toby and Bull Hill, and have left so many striking objects unnamed entirely. Certain peculiarities of nomenclature have grown up in the valley, as the naming of mountain gorges "gutters" (e. g.. Running Gutter in Hatfield and Rattlesnake Gutter in Leverett), of alluvial bottoms "meadows" (Hadley Meadows), and of deep narrow valleys "gulfs" (Gulf road in Northfield). In 1810 Prof Benjamin Silhman, of Yale College, visited the lead mine in Southampton at the request of the proprietors and drew up a report for their use. This does not seem to have been printed by them separately, but was published by the author the same year in the second number of Brace's Mineralogical Journal, in which also a paper descriptive of some of the minerals found at the mine was published, from the pen of Dr. William Meade. The pubHcation of Cleaveland's Mineralogy (1816) and of the first volume of Silliman's Journal (1818), and the influence of Amos Eaton in Albany, mark the beginning of a strong movement toward the study of mineralogy and geology in New England. The first articles of Edward Hitchcock 1 appear in these years, one of them, "with a sketch by Mrs. Hitchcock," marking the beginning of a scientific partnership which was to last so long, and which has made this region classic ground for the geologist. From this time on for a half century nearly all that became known con- ' Obituary: Proc. Am. Acad. Arts Sci., Boston, Vol.VI, p. 291; Hist. Conn. Valley, 1879, Vol. II p. 617. ' 4 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. cerning the geology of the Connecticut Valley was discovered by him, and the whole body of knowledge on the subject was systematized in his suc- cessive reports. So frequent mention will be made of his work in the following pages, and its progress may be followed so fully in the Chrono- logical List, in Chapter XXIII, that special mention may be omitted here and attention called to the many physicians, teachers, and laymen who became enthusiastic mineralogists and scoured the hills so thoroughly that it is now exceedingly rare that one finds a new locality for minerals within these bounds. Prominent among these was Dr. David Hunt, of Northamp- ton, to whom President Hitchcock acknowledges great obligation for assistance in mineralogy as early as 1818, and of whom Amos Eaton said that he had every mineral in this part of the State at his call.^ Dr. Jacob Porter, of Cummington; Emerson Davis, principal of the Academy of Westfield; Dr. William Atwater, of Westfield; Simeon Colton, of Monson, and Dr. Ebenezer Emmons,^ of Chester, who commenced his scientific work here, were among the professional men who pursued min- eralogy with great energy, and the last of this band of men, Mr. W. Morris Dwight, died in extreme old age in Williamsburg only a few years ago. Prof Amos Eaton, Dr. George Gibbs, Prof. Chester Dewey, and Prof J. T. Webster extended their studies over this region from without, so that already in 1825 Mr. A. 0. Hubbard, writing from Yale in commendation of Mr. Hitchcock's "excellent description of the Connecticut Valley," says the region "is becoming, or rather has already become, the rallying point of all the mineralogists in Massachusetts." There appeared in Silliman's Journal for 1827 an article on the lead mines and veins of Hampshire County by Mr. Alanson Nash. Prof. C. U. Shepard was then assisting in the publication of this journal, and he once described to me the difficulty he had in deciphering the crabbed script of the author and in bending his sentences to the common rules of grammar. Little knowledge of the distribution of the lead veins has been added, however, to what is contained in that article, and several of the veins described by him I have not been able to find, though I do not doubt their existence. He was the forerunner of a great body of natural prospectors — men without learning, books, or assistance, who, from a strong love of the I Index, 1820. = Sketch of life, by J. B. Perry: Proc. Boston Soc. Nat. Hist., Vol. XII, p. 214; also by Jules Marcou: Am. Geologist, Vol. VII, p. 1, with fine portrait. HISTORICAL SKETCH. 5 quest, roamed over the hills hunting for minerals, and became as acute and skillful in the search as theii' neighbors did in hunting and fishing. I do not know that Mr. Nash was a cobbler, but I suspect so, for I have found both here and in Europe that, perhaps from the intermittent character of their employment, men of this trade are exceptionally apt to develop the taste for collecting minerals. Of the long list of these men who have forwarded mineralogy in an unambitious way I will mention only Mr. B. Hosford, of Springfield, who, at the suggestion of Professor Shepard, first dissected one of the Lancaster chiastolites, which was figured in Dana's Mineralogy, and whose study of the salt crystals in Westfield I have reported in Bulletin No. 126 of the United States Geological Survey, and Mr. William Newell, of Pelham, long time cobbler in Amherst. Students long before and after my time in college will remember his love of minerals and his reticence concerning his " locali- ties." There was a pint of fine amethysts in the collection at Amherst which he had gathered from the gravels of Amethyst Brook. If they had been solid gold they would have poorly paid him for the time spent in searching for them ; being amethysts, however, they satisfied him much better. The first mineral from Hampshire County to receive notice abroad was the albite of Chesterfield (cleavelandite), which, as kieselspath, was described by Ilausmann in 1817. The first article on minerals from this area by Professor Shepard appeared in 1824, beginning a half century of work as profitable for the advance of mineralogy in this region as that of President Hitchcock was for the progress of geology. The halting places in the history of the geology of the valley are the dates of the publication of the principal works of President Hitchcock, as follows : 1818. Remarks on Geology of a Section of Massachusetts. This was followed by a period of collecting minerals and recording their localities, and by the beginnings of geological work, especially by Prof. Amos Eaton. 1823. Sketch of Geology of the Region of the River Connecticut. This was succeeded by a continuation of the collecting and recording period. 1833. Report on Geology of Massachusetts. 1835. Report on Geology of Massachusetts, second edition. This was followed by the most interesting episode in the history of the geology of the Connecticut, the discovery and description of the very numerous and 6 GEOLOGY OF OLD HAMPSHIRE GOUETT, MASS. perfect Triassic tracks found up and down the valley, with which discovery the names of Dexter Marsh/ Dr. James Deane,^ and Dr. Roswell FiekP were also connected. Previous to the year 1884 I was for a long time accustomed to arrang-e the successive senior classes of Amherst College on the lawn before the house of Dr. Field, in Gill, and the old man would come out and give the boys a lecture on the "true theory of bird tracks," claiming, and I think with justice, that he first discovered the quadrupedal character of the animals which made the tracks. 1841. Final Report upon the Greology of the State. This summarized the geology of the region to date, adding, however, very little to the report of 1835, while the discussion of the "bird tracks" went on vigorously until, in 1844, the Report on Ichnology brought together all that was known on the subject, with abundant illustrations — indeed, vastly increased what was before known, though it did not close the subject, since articles descriptive and controversial continued to flow from the pens of all those mentioned above, as well as more elaborate works from the two sons of President Hitchcock, while the last scientific article published by the President himself (1863) was concerning New Facts and Conclusions Respecting the Fossil Footmarks in the Connecticut River Valley. 1860. Illustrations of Surface Geology. With this, one of the pioneer works in a field which has since become most popular, the great work of President Hitchcock on the geology of the Connecticut closed. 1863. Reminiscences of Amherst College. This book contains an aftermath of opinion on the geology near Amherst. I may here mention, in conclusion, several persons who have advanced the science of mineralogy in the region, or at least have gathered valuable collections for the use of other. Mr. James T. Ames, proprietor of the well-known foundries at Chicopee, was led, perhaps from his connection with the Chester emery bed, to gather a collection very valuable for the illustration of the local mineralogy ; and Dr. H. T. Lucas,* who had a large share in the discovery of the emery at Chester, has been identified with the exploitation of this and many other mining properties in Hampshire County for many years. Mr. M. A. Brown, formerly of Northfield, has done very useful work in exploring the mineralogy of eastern Franklin 'For sketch of his life see History of the Connecticut Valley, Vol. II, p. 585. 2Ibia.,p. 520. 'Ibid., p. 576. ^Ibid., Vol. I, p. 1064. HISTORICAL SKETCH. 7 County. ^Ii". (-)liver M. Clapp,' of Amherst, recently deceased, was an ardent collector during his long life. The finest collection for the illus- tration of the local mineralogy, excepting, of course, the great collection of Professor Shepard, was that made by Mr. Josiah D. Clark, for a long time a teacher in Brooklyn, but a native of Northampton, who watched carefully the progress of the work at the last opening of the Loudville mine, during the war, and secured very abundant and wholly unique suites of all the rare things found there, as well as valuable material from all other localities of western Massachusetts. He sold his collection at an exceptionally low price to Smith College, from a desire that it might remain entire and in Northampton. The burning of the great Shepard mineral collection in Walker Hall of Amherst College in 1882 may stand as a next and sad epoch in the history of mineralogy in the Connecticut Valley, a loss in many ways irreparable, for the rich store of material for the illustration of the local mineralogy can never be wholly replaced. Fortunately Professor Shepard had published largely concerning this material, and I had taken quite full notes of almost all the collection, which have been incorporated in the following report. In December, 1887, the collections made by Professor Shepard after the sale of his collection to Amherst College were presented in his name to the college by his son. Dr. C. U. Shepard, of Charleston, South Carolina, and this goes far toward restoring the monument to his memory, and very far toward filling out the local collections at Amherst, which should be, of coiu'se, unsurpassed for the region in the center of which the college is situated. It is proper to call attention to the fact that the list of publications upon the geology and mineralogy of the State (Chapter XXIII), in which I have included those upon topography, is the true history of the progress of these studies here during the present century, and that in the preceding pages I have purposed only to emphasize some names that would otherwise be overlooked, and to indicate some salient points in the history which seemed to me to deserve mention. 'Hist. Conn. Valley, Vol. I, p. 241. CHAPTER II. TOPOGRAPHY. The great central plateau of Worcester County, averaging about 1,000 feet above the sea, lowers a httle toward the west, and is accented as it passes into the area under consideration by deep north-south longitudinal valleys, the streams here taking for long distances a north-south course, and it is cut deeply by two great transverse valleys — those of the Millers and Chicopee rivers — which gather all the drainage from the east. With this modification the plateau is continued westward until its border forms the eastern edge of the Connecticut Valley. ' The rim of the valley on its west side is the border of a similar broken plateau of about the same height, deeply cut by longitudinal valleys whose waters also reach the Connecticut by two transverse valleys — those of the Deerfield and Westfield (or Agawam) rivers — which are farther south than the corresponding valleys on the east, each by about the same distance. The plateau rises along the western portion of the three counties into the Berkshire Hills. It will be noted that the Connecticut Valley includes about all of the broad, low area underlain by Triassic rocks. These two plateaus were probably once parts of a continuous plain that extended across the Connecticut and other valleys far beyond the limits of the area studied. This plain was formed by erosive agencies which degraded the rocks nearly to sea level. It seems to have been well established by Professor Davis that this degradation took place during the Cretaceous period, and that a later Tertiary elevation enabled the streams to cut down their valleys and clean out the wide lowlands in the soft rocks that border the Connecticut and the eastern branches of the Swift River in Enfield. The peaks and ridges of more resistant rock that rise in these low- lands still reach almost to the level of the old plains, and are remnants of it. 8 TOPOGRArHY. 9 Of the longitudinal valleys the most i)eculiar is the l)asiu in Greenwich and Enfield, in the eastern portion of the region, in which the branches of the Swift River join and move southward. It is a broad, low, sand plain, studded with isolated, high, rocky islands and stretching from north to south through these towns. The streams enter and leave it by narrow channels, while the plain continues south through Ware, and was once, I susjiect, continuous with the deep, straight valley which extends through the middle of Monson and on into the valley of the Willimantic; and it was in its middle part (in Palmer) clogged up with till during the Grlacial period, so that the Swift River, which on this supposition formerly ran southward across Palmer and Monson into the Housatonic, has in post- Glacial time found its way westward, breaking through the side of the basin to join the Connecticut. If this be so it explains at once why the basin is so disproportionate to the size of the present river, and why it is on all sides walled in by high ground, except the narrow gorge by which the Swift River escapes from it. It also explains the very straight Monson Valley, in the middle of which, just at the State line, the waters run south to the Willimantic and north to the Quabaug at Palmer. The Connecticut Valley stretches across the center of the area from north to south, with a width of about IjV miles at the north, which increases to 8 J miles opposite Greenfield, 10| miles opposite Amherst, and averages 15 miles in the southern portion of the State. It is divided lengthwise into two portions of about equal width by the remnants of the red sandstone and the long trap ridges of Deerfield Mountain and the Holyoke range; and, except the short canyons of the two western tributaries, the only breaks in this dividing wall are at its north end in Bernardston and in the long distance opposite Amherst, between Sugar Loaf and Mount Holyoke. Post-Glacial deposits occupy the full width of the Connecticut Valley in great complexity and beauty. From the northern line of the State the eastern border of the valley, sloping rapidly to the bottom, runs nearly due south across the State, notched sharply by the gorges of the Millers and Chicopee rivers, and rarely opening out into a rounded high-lying valley, as in Pelham, opposite Amherst, or breaking down into an elevated plateau, as in Belchertown. On the west the high gi-ound crosses the State line but a little way back from the river, and for a few miles the valley preserves the same narrow limits 10 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. and simple cliaracter which mark its more northern course. The crystal- hne rocks are then set back 7 miles to the west, along- the northern border of Greenfield, and the rocky boundary thence goes south, with sharp east- ward slope, notched only by the Deerfield gorge, to be again set back by about the same amount along the north of Northampton. It then runs south again, interrupted only by the Westfield River, to and beyond the south line of the State. On both sides the brooks and the roads (which usually follow the brooks) come down sharply from the uplands, and railroads can enter and leave the valley only by the four tributaries mentioned above. Just south of where the western boundary first turns westward, in Greenfield, a great block of red sandstone hills, occupying the whole town of Gill, separates the valley into two parts, the river occupying the eastern portion and the narrow, high Bernardston Pass connecting it with the north end of the western portion. From the southwest corner of this mass the Deerfield trap sheet runs southward, forming Deerfield Mountain, its ver- tical western scarp making the eastern boundary of the western lateral valley, which preserves its width southwardly through Deei-field, while east of it the valley of the Connecticut proper expands into the Montague basin, the ridge being much narrower than the block of hills in Gill, which makes the northern border of this basin. On the south the great mass of Mount Toby shuts in this Montague basin, the river passing in a narrow valley between it and the south end of the Deerfield range, which ends abruptly with Sugar Loaf, into the much broader Hadley basin, while a deep, nar- row valley around the east side of Mount Toby also connects the two. By the breaking down of the Deerfield range the Deerfield Valley opens widely into the broad Hadley basin, which here has the full width of the Connecticut Valley, 1 4 miles, between the crystalline borders on the east and west, though Mount "Warner, a mass of crystalline rocks, stands mid- way to partly continue the ba,rrier. South of Amherst the Holyoke range rises abruptly athwart the valley, lea,ving a narrow passage on the east into the Springfield basin, like that around the east end of Mount Toby, while it is broken through for the escape of the river just as the latter comes through a narrow passage between Mount Toby and Sugar Loaf on its entrance to the basin. The Holyoke range extends south along the western border of the TOPOGRAPHY. H Sprin<>-field basin, \vliilo its steep western slope is the eastern boundary of a lateral valley, similar in size and position to the Deerfield Valley, of which, indeed, it may be looked upon as the continuation, and this valley extends across Southampton and Southwick and, as the Farmington Valley, is con- tinuous to the Sound. The Springfield basin is also continued beyond the limits of the State, and, though contracted at the Enfield Falls, is not terminated until it reaches the narrows at Middletown, Connecticut. On the east the longitudinal valleys, especially the Enfield Valley, are largely due to the folding of bands of newer and harder schists down into the gneiss and the subsequent deeper erosion of the latter. On the west, where the whole area is occupied by closely folded schists, one can only rarely see any connection between the valleys and the dm-ability of the bottom rocks. The topography of the northwest portion of Franklin County is, how- ever, very plainly influenced by its stratigraphy. The Deerfield River, on entering the State, runs southward with the strike of the Hoosac schist. It then bends and ciits across this strike at right angles, and then turns south- west again with the strike, and repeats this zigzag several times, and at last, reaching the great fault at the portal, it turns sharply east across the sericite-schists. All the orographic lines in Rowe — the mountain ridges and the intervening valleys — are for the same reason directed southwest, par- allel to the abnormal strike of the rocks thereabout. The deep depression in which Shelburne Falls lies is plainly the result of the great quaquaversal by which the gneiss is here exposed, and is the expression of its lesser durability. Across the western half of Hampshire and Hampden counties the drainage is southeast, and is only in a minor degree controlled by the north- south structure of the rocks. The east branch of the Westfield River flows from Cummington south to its mouth with the strike, curving around the Groshen anticlme, and its gorge above West Chesterfield Hollow and the gorge of the Westfield Little River are the wildest in the State. CHAPTER III. GEOLOGICAL OUTLINE AND GENERAL COMPARATIVE SECTIONS. A long series of Archean outcrops runs from north to south across the western portion of the high ground between the Housatonic and the Con- necticut valleys, and barely enters the western border of the area here described. This high ground is the continuation of the Green Mountain range across Massachusetts. Cambrian conglomerate-gneisses (Becket gneiss) wrap around these patches of Archean, gi-aduate westward into the Stockbridge limestone, and dip eastward beneath the great sericite-schist series, which may be placed parallel to the Berkshire and Greylock schists on the west. These highly metamorphosed and much foliated sericite- schists stand vertical in appressed folds for a long distance eastward and then go beneath the extensive graphitic schist series, coming up farther east in anticlines from beneath the latter. A remarkable band of amphibolites, with enstatite-bearing limestones and enstatite, pyroxene, and olivine rocks, all largely changed to serpentine, and with emery, runs down the middle of the sericite-schists. It seems to me possibly the equivalent of the Bel- lowspipe limestone of Greylock ; and the Bolton limestone, farther east, is upon about the same horizon. The upper series of graphitic schists (the Goshen and the Conway schists) is less metamorphosed, and shows much of the original lamination, though masked by cleavage and foliation. It contains many beds of limestone in every stage of change to amphibolite. It is a graphitic muscovite-schist, abounding in garnet, staurolite, and transverse spangles of biotite. It graduates into the corrugated and cleaved Leyden argillite (phyllite) along the eastern border of the elevated 12 GEOLOGICAL OUTLINE, 13 area defined, above, and upon it rests, at the lower level of the Connecticut Valley, the complex Bernardston series — conglomerates, quartzites, lime- stone, mica- and hornblende-schists, and gneiss — which is proved by the presence of many fossils to belong in the Upper Devonian. A complex series of faults, with much westward overthrusting, bounds the elevated area on the west. A series of echeloned faults also di-ops the bottom rocks of the Connecticut Valley on the east and makes the elevated area a "horst" and the valley bottoms "graben," in the nomenclature of Suess.^ A great stock of tonalite, or quartz-diorite, occupies the eastern border of the area and encroaches on the Connecticut Valley. This has come up through the thick Whately amphibolite bed. It graduates westwardly into the gi'anitite, or biotite-granite. This has emerged in the region of the broad Whately limestone bed. This is followed outwardly by a great group of dikes, of every size, of granite or muscovite-biotite-granite. This is in the region of the muscovite-schists without limestone. Each of these rocks seems thus to be distinctly influenced in its chemical constitution by the rocks it has penetrated and dissolved. On the periphery are great quartz veins, and the remarkable tourmaline- and cleavelandite-bearing dikes, with minerals containing rare elements. Farther west all the sericite-schists and Cambrian gneisses are free from later igneous rocks except the great isolated granitite dike in Middlefield. The valley of the Connecticut may in a general way be called a broad syncline, so far as the crystalline rocks are concerned. It is rather a broad area of greater crushing and disturbance, which has favored greater erosion, and over its bottom the crystalline rocks lie often horizontal or in small anticlines and synclines, while on its borders they dip toward the center, often with high angles. In attempting to trace the history of the valley, it will perhaps always be impossible to assign their proper weight to the erosive agencies mentioned above in comparison with another agency which has been of prime importance in the formation of the valley. I mean that which has produced the great faults and the sinking of the areas between the faults. The principal southwest- 1 E. Sueas, Das Antlitz der Erde, Vol. I, pp. 166,264. 14 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. by-soutli fault, which appears so plainly on the map, forming the eastern bonndary of the valley across Northfield and Montague, is probably pro- longed in the Holyoke range fissure from Mount Tom southward. From a point north of Mount Toby a fault branches from the main one and is continued down the east side of the valley, a series of great faults running south by east at the eastei'n border of the valley; and much the same seems true of the western side, and especially the two settings- back of the valley border seem due to the two east-west faults. How. far the valley bottom has been depressed between these faiilts I can not determine, but the great thickness of the red sandstone, as shown by artesian wells, would indicate that the sinking must have been consid- erable after, and perhaps during and before, the deposition of the Trias. The region is thus a great "graben" — a band of country sunk between parallel faults; and the great Grreenwich-Enfield basin has, at least in part, the same character, though here erosion has been the more important agent, and in its northward extension into New Salem and Orange the sole agent. In both pre-Triassic and Triassic time the Connecticut Valley has been a region of extensive faulting and the pre-Triassic faulting extends con- siderably east of the present bottom of the basin, especially in the Northfield region. All the rocks of the area west of the Connecticut reappear in the eastern region. The Bernardston rocks are present only in a few outcrops in Northfield and farther south, while the Leyden argillite appears in the south bank of the Connecticut just below the mouth of Millers River, and seems to run down the valley beneath the Trias and to appear west of the pond in the center of Leverett. It is also represented lithologically in the center of the middle syncline in Monson. The salient features of the eastern area are — (:Z) The eruptive rocks, consisting of (a) the great block of diallage- granite, or tonalite, and quartz-gabbro in Belchertown and the surrounding towns, around which the crystalline rocks are thrown into great confusion; (&) the block of diorite in New Salem and Prescott, which seems to have produced very little confusion in the surrounding rocks; (c) The Coy's Hill porphyritic granitite ; (d) the large granite areas in Leverett and Amherst. GEOLOGICAL OUTLINE. 15 (:i) The Monson rjneis.s. — The great plateau of Cambrian gneiss which, starting in Northfiehl, runs through Wendell, Shutesbury, and Pelham, and ends against the Belchertown tonalite, furnishes the key to the structure of the region. It is another "horst" — a great area of ancient crystalline rocks bounded by faults outside which the ground has everywhere sunk away. It is, moreover, a region of very gentle dips, unlike the western hill countr)^ The rocks, horizontal in the center, dip slightly toward the borders on the east and west. It is bounded by north-south faults on either side, which extend wholly or nearly across the State. These faults are lines or bands of extreme crushing, and outside them the rocks have been compressed in sharp folds, as if they had been thrust against the vmyielding shoulders of the great "horst." The normal Monson gneiss is, however, the ordinary biotite-gneiss. One of the faults mentioned runs at the foot of the high grounds along the east border of the broad Connecticut Valley, from Northfield south, through the notches at the east foot of Mount Toby and at the east end of the Holyoke range, and so on through Granby and Wilbraham. As noticed above, this fault forms also the eastern boundary of the Connecticut Valley "graben." The other fault runs in a corresponding position, along the east border of the valley of the west branch of the Swift River, through Wendell, New Salem, Prescott, and Enfield. The faults are marked by great crush- ing of the rocks, by the development of curious, "fault rocks" — bastard granites and green and buff hornstones — and by the cementation of the crushed rocks by comby vein quartz and specular iron. Within the "horst" the Monson gneiss contains a thick bed of a fine-grained actinolite-quartzite or at times fine biotite-quartzite or biotite-gneiss. (5) The schists. — Outside these faults the upper schists are present in the same series as west of the river, ^^z: (a) A feldspathic mica-schist or two- mica-gneiss, at times a quartzite or quartz-conglomerate, is the equivalent of the feldspathic mica-schist or Hoosac schists and the lower sericite-schist or Rowe schists of the west side. This is named after the more persistent and important bed of the western area, the Rowe schist. (&) A hornblende- schist=:the Chester amphibolite. (c) A micaceous quartzite, very generally 16 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. containing a hydrated mica or a greeu chloritic mineral, which is at times certainly derived from garnet ; or the series is developed as a whetstone- schist — that is, as a thin-bedded, finely biotitic, arenaceous quartzite. It is the upper sericite-schist, or Savoy schist, (d) A mica-schist, in great thick- ness at times, exactly like the finely corrugated biotite-spangled garnet- schist of Conway and Groshen, as in Northfield Mountain, on the Shutesbury- New Salem line, and in Monson. Usually it is a coarse, barren, muscovite- biotite-schist, like most of the Conway schist, but always without limestone, which seems to be replaced by hornblende-schists. It is the equivalent of the Groshen schist and the Conway schist. It is named after the more important member and the one it most resembles — the Conway schist. Instead of appearing in broad areas, succeeding each other from west to east^i. e., from below up, as they do in the western hills — the schists appear here in sharply compressed synclines which run across the State, disjointed by faults and thrown into confusion by the presence of eruptive rocks. Four such great synclines can be traced across the State, within the limits of the three river counties, though their identity is disguised by the fact that metamorphic changes superinduced upon original variations in composition have varied greatly both in kind and degree. One may especially adduce the fibrolitization which has progressively affected the mica-schist from west to east and from north to south. For the reasons given above it will be more convenient to follow a geographical rather than a geological order in the discussion of the eastern schists and to take up the diff"erent synclines in succession. GENEEAL COMPARATIVE SECTION OP KOCKS IN MASSACHUSETTS. In the first column of the accompanying general section I have placed the section for northwestern Massachusetts, as determined by the labors of Professors Pumpelly, Dale, and Wolff", ^ though they must not be held responsible for the exact parallelism here attempted. The distinction between the Becket conglomerate-gneiss below and the Cheshire quartzite can not here be always maintained, and the quartzite graduates both laterally and vertically into the limestone. The area east of the Connecticut and extending slightly into Worcester 'Geology of the Green Mountains in Massachusetts: Men. U. S. Geol. Survey, Vol. XXIII, 1894. GEOLOGICAL OUTLINE. 17 County forms a strong contrast to that west of the river, and is an area of transition to the much simpler structure of Worcester County. Meta- morphism and the part taken by post-Carboniferous eruptives increase reyuUirly eastward. Such rocks are Avanting- in the first column and in tlie western half of the country covered by the second. In the eastern half iif the latter the granites begin and rapidly become important; in the second other rocks are associated, and in the area of the last column they cover more than half the surface. In a column devoted to the Massachusetts coast region eruptive rocks would be still more predominant. General section showing correlation of rocks in Massachusetts. [The names given in tlie second and third columns are those used in this monograph. Those in the foui-th column will he used in a forthcoming memoir on the geology of Worcester County.] Taconio Kange and Houaatonic Valley. Berkshire Hills and Connecticut Valley. East of Connecticut River. "Worcester County. Black Eock diabase Tlie whole Triassic (intruaive). series is repeated Chicopee shale. east of the river. Longmeadow sand- stone. Granby tuff. g Mount Holyoke dia- base (iuterbedded). Sugar Loaf arkose. Mount Toby conglom- erate. Unconformity. • m ^ Granite. Granitite, G r a nite. Granitite. Granite. Granitite. H% Pegmatite. Albitic Pegmatite . Albitic Pegmatite. Albitic granite. Tonalite. p;ranite. Tonalite. granite. Tonalite. g|.§ Diorite. Diabase. Diabase. Olivine- "^^^ Cortlandite. gabbro. "Wehrlite. i "Worcester argillite 1 (phyUite). 1 ' u o Harvard conglomer- ate. "Worcester quartzite. EerDardston mica- Bernardston m i c a - scbist. schist. Bernardston araphib- ^ olite. g . Bernardston quartz- Bernardston quartz- t- ite. ite. fi Bernardston 1 i m e ■ stone. Vernon gneiss. Vnconforinity. MON XXIX- 18 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. General section showing correlation of rocks in Massachusetts — Continued. Taconic Kange and Houaatonic V alley. Berkshire Hills and Connecticut Valley. East of Connecticut Kiver. "Worcester County. r Leyden argillite. Leyden argillite. Conway dark corru- Conway schist, chang- Brimfield rusty fibro- gated mica-schist ing eastwardly to lite-schist. u ■with garnets. Brimfield rusty fi- Gosben flaggy biotitic brolite-schists. muscovi le -schist with quartzite and limestone beds. I Unconformity. Haw ley actinolitic chlorite-schist, 1, ampbibolite, pyrite, and hematite beds. m Greylock schist. Savoy chloritic seri- cite-schist. Savoy schist. Bellowspipe 1 i m e ■ Chester ampbibolite, Chester ampbibolite. Fasten whetstone- q3 stone. with emery, serpen- schist. o tine, steatite, and saxonite. (Changes east of the Carboniferous into Berkshire schist. Kowe quartzose seri- cite-schist with ampbibolite beds. Kowe flaggy schist. \ Bolton gneiss, in- cluding Boltonlime- 8 tone.) Stockbridge 1 i m e - Hoosac albitic sericite- stone. sebist. Stockbridge 1 i m e - Cheshire white gran- Pelham quartzite. Grafton quartzite and i stone, lower part. ular quartzite. conglomerate. "Vermont formation Becket white con- Monson conglomerate- Sutton gneiss. i (quartzite and glomerate-gneiss. gneiss. u gneiss). Unconformity. Stamford gneiss (por- phyritic gneiss with blue quartz) . "Wasbiugton blue quartz-gneiss. Tyringbam stretched Ifl'ortbbridge gneiss. d biotite-gneiss. ',3 EastLee black biotite- 1^ o hornblende- gneiss. < Hinsdale coarse chon- drodite-liniestone. Hinsdale granitoid gneiss. CHAPTER IV. THE ALGONKIAN.i GEOIiOGICAI/ DESCRIPTION. THE HINSDALE AREA.^ Before my work had extended to the western border of the region covered by this study, my attention was called, in the winter of 1882, by Prof. J. D. Dana, to two interesting outcrops of undoubted Archean rocks in Hinsdale, a gneiss and a limestone containing chondrodite and a ^Deculiar peach-blossom-colored mica, determined by him to be probably rhodo- clirome. Although these localities lie beyond the western border of the river counties, the same rock extends into the southwest corner of Middlefield, and Professor Dana's discovery was very acceptable to me as furnishing a possible base to work from in the complex region under examination. The two localities in question are at the first cutting west of the railroad station in Hinsdale and at the first cutting south of the railroad station in Washington, and as they give a much fuller exhibition of the series thaa the limited portion of the same which enters Middlefield, they are made ih the main tlie basis of the description following. The greater portion of the town of Hinsdale is occupied by an oval anticline, elongated north and south and overthrown to the west. This extends, much contracted, across "Washington, and bending southeastward and narrowing still more it enters Middlefield and runs along the south line of the town to a point a mile beyond Becket station. The newer gneisses, all down the east side of the anticline, dip normally eastward away from the older, but here — that is, where the narrow band of Algonkian extends east along the Westfield River— a sharp east-west wrinkle forms in the newer gneiss, and the older gneiss buckles up through the newer. 'Azoic (Lyell), Eozoic (Dawson), Archseau, Dana. " This will be described in detail in a monograph on the Archean of Berkshire County. 19 20 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. Counting from below upward, the Algonkian rocks may be divided into four groups: 1. Hinsdale gneiss. — This is a. group of gray biotite-gneisses, generally quite coarse and with the jet-black biotite in distinct, elongate patches, granitoid and yet well foliated. The broad, fresh cleavage surfaces of the feldspar are often strongly curved from pressure. These gneisses weather with exceptional rapidity and seem to be calcareous. 2. Hinsdale limestone. — The coarsely crystalline chondrodite-limestones form a concentric band around the older gneisses, marked by a series of abandoned limekilns, for the rock was economically important before the opening of the "Western Railroad." 3. Lee gneiss. — This is a heavy black hornblende- or hornblende-biotite- gneiss. i. Washington gneiss. — A broad band of rusty graphitic blue-quartz gneiss forms the outer circle of this Algonkian nucleus. It is in the main a biotite-gneiss, but with little mica, and rusty from the decomposition of hornblende, pyrite, pyrrhotite, and a ferruginous dolomite. In the whole circuit graphite is a never-failing accessory, especially in the upper por- tion. The graphite mine at Washington, except for the size of some of the constituents, suggests the Ticonderoga graphite mines. Very coarse calcite, graphite in broad, thin, hexagonal plates, coarse white sahlite, large green pyroxene and hornblende masses, groups of finely terminated pistachio- green pyi-oxenes, brown sphene, and garnets, followed paragenetically by coarse calcite with phlogopite, and this by quartz, are some of the points of resemblance. , Another equally persistent and characteristic constituent of these gneisses is a blue quartz in flat laminse 1 to 3"°™' in thickness, which has often so deep a tint of rich purplish blue as to furnish beautiful cabinet specimens, and is so abundant as to form more than three-fourths of the mass of the rock. Everywhere in the outer circuit of the Algonkian rocks a band having the above peculiarities lies below the lowest beds of the Cambrian con- glomerate-gneiss, viz, blue quartz formed in place, disseminated graphite, beds of the heavy black hornblende-gneiss, and a general abundance of hornblende and a very general rustiness, all associated with intervening bands of a common biotite-gneiss. THE HINSDALE AKEA. 21 From the Washing-ton station the older gneiss narrows and occupies the sides and bottom of the narrow canyon, which continues toward Becket station. The canyon, caused by the projection of this narrow lobe of the older rocks, is one of the most curious and interesting topographic features of the region. The lesser capacity of resistance to erosion of the older gneisses and limestones has caused the broad depression in which Hinsdale lies, and the southward projection of the same rocks has determined the long, nan-ow canyon in which the waters of the Westfield River, gathering in Washington, flow southward, thus providing the only chance for railway communication between the Connecticut and Housatonic valleys. At Becket station, south of the river, everything is newer gneiss. Just north, in the village, appear the hornblende beds of the upper Algonkian band, and following the road north to the pasture overlooking the village, one finds abundant outcrops of the blue-quartz gneiss and the contact on the conglomerate-gneiss striking southeast and dipping northeast — that is, in the normal relation to each other. This allows them — the older gneisses — to appear in a band on the north side of the brook, which band seems to contract and come to its apex just at the point where the i-ailroad enters Middlefield, so that thence southeastward it appears to be wholly wanting at the surface, or is perhaps only concealed in the bed of the river. It however makes its presence below manifest by a continuation eastward along the river of the overturned anticline without the core of exposed Algonkian, until, at the junction of Coles Brook with the river, the Algonkian chon- drodite-limestone, accompanied by heavy dark gneiss, buckles irregularly up through the conglomerate-gneisses. THE COLES BROOK ANTICLINE. Just a mile northwest of Bancroft station, Middlefield, the Boston and Albany Railroad cuts off a loop of the Westfield River, and Coles Brook enters this loop. The railroad runs in a deep cut a long distance before reaching this loop, and the cut continues through the loop and most of the way to the station. At bridge 143 the Cambrian white or conglomerate gneiss in synclinal posture mounts up on the older gneiss, which I have called the East Lee or black gneiss from its large development just above the limestone in the Lee-Tyringham region. It is especially contorted and cut by pegmatite at the junction, and consists of a great thickness of a 22 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. wavy-bedded gneiss of fine grain and almost black from the abundance of the black biotite. (See fig. 1.) Contorted white veins one-fourth inch to 2 inches wide run through the rock, high up in the cutting, with most tortuous course, and on the south side I could recognize the spot whence came a great block which has long lain in front of the geological museum at Amherst and whose origin I had been unable to learn. These "veins" are formed by the expulsion of the biotite from their area, the white quartz-feldspar mass being continuous within and without their limits. The black gneiss abuts, apparently by a fault, cer- tainly by a wholly abrupt transition, upon a band of the coarse white, almost micaless, Hinsdale gneiss, 23 feet wide below, but narrowing above. This is followed by a bed of white, thin-bedded, highly crystalline chondrodite-lime- stone, with thin films of serpentine, forming a beautiful verd antique, which is separated by 108 feet of the same black Lee gneiss from a second band of a similar limestone, of which only 29.5 feet are exposed. This is followed in the brook bed at bridge 142, and on through the cutting, by a large mass of the dark gneiss, carrying beds of hornblende-schist, until we come, at the fourth telegraph pole from bridge 142, upon the fine unconformity where the conglomerate-gneisses mount upon the dark Lee gneiss. Between this point and the Bancroft station the cut- tings expose a long extent of contorted and twisted rocks, where the beds swing round from horizontal to vertical within a few feet. Gradually a low dip eastward predomi- nates, and this becomes steeper, and a band of hornblende- gneiss 10 feet wide sets in, and at the eighth telegraph pole from the station, just at the signal house, a boss of coarse actinolitic rock derived from the older limestone protrudes. All east of the unconformity is Becket gneiss, ^ except the few hornblende-gneiss masses and the last- mentioned boss of actinolitic rock, which are brought just above the railroad level by the undulations of the Becket gneiss. ^. THE COLES BROOK ANTICLINE. 23 The Coles Brook uuticliue extends north into Middlefield more than a mile mid a halt', following the brook bed for a hundred rods. It is well exposed just south of Factory village, by the roadside, for a long distance south of the schoolhouse. Large bosses of the coarse limestone appear here, flanked on the west by the Lee hornblendic gneiss. A still more instractive section is exposed halfway between the two localities mentioned above, where the road going west from Factory Brook up onto the high ground crosses the limestone near the site of H. Hawes's house (now destroyed). In the bare hill opposite this site the limestone and the green actinolitic rock derived from its alteration are abundantly Hinsdale Xiirrtestarte Pegmatite ISnedale Gneiss N30'E.75W N6'E.90' ZSFT I I Fig. 2. — Detailed section of the limestone at Coles Brook. exposed in vertical strata, and the white Becket gneiss can be seen mantling over it in clear unconfoi'mity, starting with steep west dip on the west side of the bill, becoming horizontal on the top, and dipping- easterly down the east side. The true bedding is in places replaced by a secondary vertical structure. A coarse, rusty muscovite-biotite-gneiss, with graphite and tour- maline in quite large prisms (the equivalent of the Washington gneiss far- ther west), accompanies the limestone on either side, extending east to the bend in the road and west to the house at the top of the hill. Just east of this is a bed of typical Becket gneiss. In the yard of the ruined house the mantle of the Becket gneiss is so nearly continuous that a boss of white limestone a foot wide projects from the ground, and only a few feet on either side the Becket gneiss dips away from it. Interest attaches to the fact that the Becket gneiss is so strongly meta- morphosed as to form a quany stone of first quality only in a narrow band along either side of the limestone belt, as if the violent upthrust of the pre- Cambrian rocks along this narrow axis had exerted an influence upon the Cambrian gneisses for some distance outward, producing in them a marked 24 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. granitoid structure, wliicli disappears insensibly as the beds recede from this center. THE TOLLAND AREA. A large outcrop of Algonkian limestone occurs on the roadside in Eiverton, Connecticut, near the southwest corner of the Granville quad- rangle,^ associated with a coarse fibrolitic two-mica-gneiss, which is charac- teristically Algonkian in the Sandisfield quadrangle, next west of this, where it is a variant of the blue-quartz gneiss. This fibrolite-gneiss enters the Grranville quadrangle at its southwest coi'uer and runs north by east, in a tapering syncline, to a point north of Black Pond in Tolland. The rocks often resemble coarse mica-schists, and are scarcely distin- guishable from the coarse schists and schistose gneisses on the horizon of the Hoosac and Rowe schists, which lie next east of the Becket gneiss, except that they contain fibrolite and lie beneath the Cambrian gneisses, and in the next quadrangle west can be traced into undoubted connection with the blue-quartz gneisses and the chondroditic limestones. East of the middle of the town of Tolland, at 0. E. Slocum's,^ is a great quantity of large bowlders of a peculiar coarse hornblendic gneiss, often brecciated, with black hornblende, colorless quartz, and orthoclase. Some masses are medium-grained, some coarse, with hornblendes 4 to 5 inches long and 1 inch square at base. This rock is mentioned by President Hitchcock, but I could not find it in place. It probably was derived from the Algonkian anticline to the west. PETEOGEAPHICAIi DESCRIPTION. LOWER OR HINSDALE GNEISS, HINSDALE STATION. The coarse gneiss just above ' the limestone is granitoid in texture and contains in abundance a fresh black biotite in large scales, which in its upper layers are aggregated into concretionary masses, flattened-out lentic- ular nodules made up wholly of fine scales of biotite and epidote. These 1 The four-cornered division of the earth's surface represented on one of the sheets of the Topo- graphic Atlas of the United States is called a quadrangle. ^ The manuscript of this work was mostly completed before the atlas sheets of the United States Geological Survey were issued, and the citation of names refer to those upon the county atlases of F. W. Beers. • ' Stratigraphically helow, as the rocks are overturned. i THE HINSDALE (JNEISS AND LIMESTONE. 25 iiodnlos are placed in laiuiuatiou i)laues about 30""°' apart, the interspace, except for rare thin iihns of the same, being made up of a dead-white mix- ture of much feldspar and little quartz, mostly fine grained, but with here and there large curved cleavage faces of orthoclase exposed. It contains pvrite in small pentagonal dodecahedrons and submicroscopic zircons of dark clove-brown color. Under the microscope the rock is much dusted with minute inclusions which give it an opaque white appearance. The quartz contains a few short, straight, black microlites, unlike the long rutile needles of the granites. The trains of cavities are very abundant, and often run through several grains of quartz, suggesting crushing. The orthoclase shows all stages of decomposition into epidote. At the begin- ning the epidote gathers in small crystals in the two cleavage planes. The microcline is filled with the same short, black microlites as the quai'tz, and shows most beautiful microcline structure. The only place where a rock of this type appears in the old Hampshire County area is in the coarse mass which adjoins the Coles Brook limestone on the west. It contains, as does the pre-Cambrian gneiss of the Tyringham Valley farther south, a white orthoclase in large cleavage plates, which exhibits a rich blue opalescence. THE HINSDALE LIMESTONE, HINSDALE. Fifty rods west of Hhisdale station the limestone occurs with an exposed thickness of 25 feet (the top not seen) and dips 30° E. Eight hundi'ed and thirty feet farther west, at a stone mill, a gray epidotic gneiss occurs, with strike 30° S. and dip 65° E. The limestone is a white to pink, rather coarse (grains 3-5™"), highl}^ crystalline rock, with a certain translucency in the grains which distin- guishes it immediately from all the other limestones of western Massachu- setts and allies it to the limestones of the Adirondacks. It carries coccolite, phlogopite, biotite, actinolite, chondrodite, pyrite, and magnetite. Grenerally the coccolite or the chondrodite, or both, are so abundantly and evenly scattered through the mass that it deserves the name coccolitic limestone or chondroditic limestone, and the accessory minerals are so arranged as to give the mass a distinct foliation, especially when the chondrodite and biotite predominate. 26 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. An analysis performed in the laboratory of Amherst College by Mr. F. H. Fitts g-ave the following results: Analysis of Hinsdale limestone. Insoluble in HCl . CaO MgO FeO CO, Total Per cent. 21.96 41.31 1.87 0.68 34.71 100. 53 The chondrodite is disseminated through the rock in yellow patches elongate and parallel to one another, and as it in places changes into black patches by the admixture with it of a green mica and magnetite the resem- blance to the boltonite from Bolton, Massachusetts, is striking, especially in specimens of the latter which are changing to serpentine. In large masses it is a rich deep-red, like the chondrodite from the Tilly Foster mine. Under the microscope the patches of the minei-al are seen to be made up. of crys- alline grains fresh and free from inclusions, and enwrapped by scales of at pale-green micaceous mineral, without any indication that the one mineral has been derived from the other. The mineral shows strong dichroism, honey-yellow to deep red-brown. Toward the surface of the ledge the chondi-odite weathers to a honey-yellow opaque mass. The phlogopite is in small, thick crystals with rounded bordei's, having exactly the same bronzy color as the phlogopite from Templeton, Canada. Its crystals are generally surrounded by a band of scales of greenish-gray biotite. Both minerals are fresh, and there is no indication of a transition of one to the other. The mineral is optically negative and has the same axial angle as the Templeton phlogopite. The biotite is disseminated in black scales through some parts of the rocks ; at times as isolated crystals with rounded contours ; at times bordering the phlogopite in greenish-gray, matted scales, or the chondrodite in thinner, deeper-green scales. All these occurrences are nearly uniaxial and negative. The pyroxene occurs in dark-green grains of coccolite scattered through the limestone and in small, stout, limpid emerald-green prisms in the pink variety of the rock. The magnetite and pyrite are in small crystals and crystalline grains. THE COLES BROOK LIMESTONE. 27 the former often associated with the choudrodite, the hitter always in small complex crystals. THE COLES BROOK LIMESTONE. The limestones of this locality are first noted by President Hitchcock in his Final Report^ as occurring- in the west part of Middlefield on Pon- toosuc turnpike and on the railroad at the mouth of Coles Brook and 1 mile east. Both beds are said to extend south into Becket, one, the easterly (?), appearing- in the southeastern part of the town, on the "Billy Messinger" farm; also 2 miles farther south, on the old Becket turnpike. It is a more or less crystalline, white, impure magnesian limestone. A delicate variety of serpentine is mixed with the limestone, forming a beautiful verd antique, and in the south of Becket tremolite, talc, and titanite occur in it. The following analyses are given: ^ Analyses of Coles Brooh limestone. [TSo 1, Coles Brook ; No. 2, 1 mile east of Coles Brook (a) ; No. 3, Becket, southeastern part ; No. 4, Blandford.] CaCO:!. MgCO:, FejO:, . SiO.2... Sp. gr. 56.25 31.56 1.12 U.07 100. 00 2.78 88.02 9.91 0.15 1.92 100. 00 2.71 .58. 31 28.61 1.24 11.84 100. 00 2.82 51.66 39.48 0.91 7.95 100. 00 2.77 a This locality, 1 mile east of Coles Brook, can not be located. The analysis contains so little magnesium and silicon that I suppose the specimen came from a bowlder of the Stockbridge marble. I have added the analysis of a limestone from Blandford from the same table, which proves to be a bowlder, doubtless from the Becket locality, and, like it, contains tremolite and talc. These large bowlders occur abun- dantly, and the one whicb is noted in the 1 841 report as a ledge of lime- stone, in the northwest portion of the town of Blandford, was found to be a bowlder by Mr. S. A. Bartholomew, who used it and many others in his limekiln and traced them northwest to the outcrops in Becket. The micro- scopic description of the Hinsdale station limestones given above will apply wholly to these, and the change of choudrodite into the serpentine may be followed better here. The former rock is, however, coarser, and the ' Geol. Mass., 1841, pp. 81, 85, and 567. 2 Ibid., p. 80. 28 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. included minerals are in larger individuals, and so better fitted for mineral- ogical study. THE BLUE-QUARTZ GNEISS, PERU.' This is a rusty, fine-grained gneiss, with little mica (biotite), in distant flat sheets of small scales, and with greasy blue quartz in grains and flat plates 1 to 3"" thick, which often coalesce into parallel layers of considerable extent. These layers are plainly secondary infiltrations in a fine-granular ground which has the aspect of a fine sandstone or quartzite. Under the microscope this ground proves to be an exceedingly fine-grained mixture of quartz, orthoclase, microline, and, in abundance, minute scales of mus- covite, and it is such a structure as may have been produced by the crushing' of a granite and the change of most of its feldspar into muscovite. The blue quartz contains a few miniite broken rutile needles, rarely cavities containing small, rapidly moving bubbles, and many sheets of very fine pores or grains of some mineral. These are rudely parallel. There are a few distant fissures. A fragment heated for a long time with the bellows blowpipe retained its color without perceptible change. It shows, with plane-polarized light, small traces of undulatory polarization, and the whole of each of the bands of the blue quartz, however large, polarizes as a single individual. The sections were cut at right angles to the foliation, but with what direction in that plane I do not know. It is interesting that in each case they are cut at right angles to the optical axis, and the slide can be moved from one end to another of the blue-quartz bands — 1-2™™ wide, 15°"° long — and the optical figure remains sharply defined, regular, and unchanged, which would seem to militate against the explanation of the color as due to strain. It is, however, a very remarkable fact that these slides still show the lavender color distinctly with transmitted light when examined with the lens or the eye alone, in spite of the fact that it is of so pale and dilute a charac- ter that one would not expect to see it in so thin a film. Moreover, narrow bands, at times branching, run across the colored layers, in which the color is wholly wanting; and these bands, when examined in polarized light, are made up of a fine mosaic of quartz fragments. It is thus plain that the blue color is due to the state of tension in which the quartz is held, and disappears when this tension is reheved by rupture across the mass. 1 Residence of H. A. Messenger. PETEOGRAPHICAL DESCRIPTION. 29 THE LEE GNEISS. Amphlhollte from Wasliiiigton.^ Black, fine-grained, distinctly bedded rock. Microscopic hornblende abundant in small, thin plates, of medium absorption and pleoclu'oism. c=b>a; empale indig-o; lv=olive; a=pale ocher. Menaccanite in large, shapeless masses, with broad border of leucoxene, abundant; little biotite. The common feldspathic mosaic forming the groundmass of the rock is so covered up by the hornblende blades and of so fine grain that it is not possible to determine the variety of plagioclase which is present. There is not the sHghtest trace of cleavage or twinning, and thus there is small ground to suppose the rock to have been greatly influenced by shearing forces. At the same time, the separate rounded or polygonal grains of which the mosaic is composed show quite uniformly, when examined with plane- polarized light, a form of undulatory polarization which I have called in the following notes concentric polarization. A single grain becomes black, first at the border, and the darkening advances regularly toward the center, and it sometimes requires a rotation of 45° to render the whole fragment dark. At times such a fragment is cracked into several parts without disturbing the regularity of the above process. In the absence of cleavage and twinning it is not possible to think of this as a result of strain from the external forces which have deformed the rock. It also is without the banded zonal arrangement which usually accompanies changes of chemical composition, and where a distinct crystal has been broken up into such a mosaic the fragments show this peculiarity in a striking manner. It is a structure characteristic of the whole series of amphibolites described in the following pages, and especially of several forms which are certainly derived from limestones. This amphibolite preserves no residual structures pointing to an eruptive origin. It is a long, interbedded stratum, parallel with and near to the Hinsdale limestone, and it is a distinct associate of this rock and reappears with it in the Coles Brook band. It occurs also as a continuation of the limestone seen on the Alderman farm in Becket, where in one place the limestone is changed into white tremolite- schist for 7 feet in from the contact and in another into black amphibolite. It is also seen at the interesting outcrop in Middlefield described above; ' C. F. Lyman's pasture, east of the graphite mine. 30 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. and the same is true throughout Berkshire County.^ I think it probable that the rock was derived from an impure Hmestone, but must leave its ori- gin in doubt, because no lithological criteria can be found that will distin- guish amphibolites derived from lavas or tuffs and those derived from impure limestones. In the succeeding sections treating of the amphibolites, to which reference may be made (see Chapter X), only those from the east of the river in Leverett have shown distinct residual characteristics peculiar to eruptives and comparable to those found in the altered eruptives of the Baltimore (Maryland) area and of several foreign localities. RESUME. These oldest gneisses are coarse, often very coarse, often granitoid, and the cleavage surfaces of the large microclines are strongly curved. Allanite is very generally distributed, at times abundant. The chondrodite-phlogopite-limestones are characteristic. The upper gneisses, often strongly foliated, are marked by the universal distribution of graphite, at times so abundant as to tempt mining, and by the abundance of the peculiar blue quartz, of hornblende, and of iron rust. Allanite is even more abundant here. I Professor Kemp haE recently called attention to the fact that similar black hornblendic rocks are constant attendants of the pre-Camhrian limestones of the Adirondacks, fringing the beds both above and below. Geol. Moriah and Westport : Bull. N. Y. State Museum, Vol. Ill, 1895, p. 329. CHAPTER V. THE LOWER CAMBRIAN GNEISSES. THE BECKET CONGLOMERATE-GNEISS. This gneiss skirts the western border of Ham^jshire and Hainpden counties, inclosing narrow strips of Algonkian rocks in Middlefield and Tolland, and stretches westward across the first two tiers of towns in Berkshire County, around many pre-Cambrian areas, to enter into most complex and obscure relations to the Stockbridge limestone and associated rocks of western Berkshire. It is thus much more amply developed beyond the boundaries of the river counties than within them, and I have for convenience given it a name from the town in Berkshire where it may be best studied. It rests upon the older gneiss in great beds of highly altered quartz-conglomerate, as at the Hoosac Tunnel central shaft and at the Dalton Clubhouse, and graduates in its upper portion into the Cheshire quartzites, so largely used for glass-making. The rock is unconformable upon the lower series. With many exceptions, especially where it folds round the older rock, as given in detail below, the strike is the prevailing one of the region, varying but little from north and south, and the dips are high. CONTACT UPON THE WASHINGTON GNEISS BELOW. As it passes down the eastern side of the area of older rock in Hins- dale it dips away from it with some irregularity, which is confined to the immediate vicinity of the contact ; farther away it regains the normal north- south strike and a dip which varies but little from verticality for long dis- tances. As it swings around the southern end of the underlying gneiss it dips away from it with low angles, changing from east thi'ough south to west, and it is at the same time so far affected by the strong east-west com- pression which has molded the whole region that it is thrown into a series of subordinate folds with axes radiating outward and pitching from the old gneiss, which has thus assumed the role of a foreign and more resistant 31 32 (GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. body during the later folding of the newer gneiss. This is well seen along the railroad from above Becket to Bancroft station, in Middlefield, where one passes four principal anticlines with their intervening synclines, as well as many subordinate flexures, all pitching southward. DESCRIPTION OF THE ROCK. The prevailing rock is a rather fine-grained biotite-gneiss, always in some degree friable and breaking crisply, and without the extremely firm texture of the older series. Sometimes it is, while seemingly quite fresh, so softly saccharoidal as to crumble under the pressure of the fingers, and ao-ain so brittle that a blow of the hammer will punch a square hole in the middle of a thin slab without cracking it. It shows clear gray shades, dependent for their depth upon the amount of the biotite present, which is in clear black to dark-brown scales, both the feldspar and the quartz being colorless, limpid, and much fissured. It contains very few accessory min- erals and only small and unimportant veins of coarse granite. It varies from a very thin-fissile rock — "scaly," the quarrymen call it — to a fine- grained granitoid-gneiss, furnishing a quarry stone of the first quality, equal to anything in New England for all kinds of monumental work. It is best exposed for study along the Boston and Albany Railroad below Becket station, and a brief description of this section will give a o-ood view of the range of variation in the rock, although it must be noted that the section is not taken at right angles to the dip, and that it contains several repetitions of the same strata, as the folds around the older gneiss are traversed. From Becket station east to the Middlefield line the older, rusty, pre- Cambrian gneiss with small segregated granite veins continues, passing three brido-es, and changes here immediately into a light-colored, fine-grained granitoid gneiss, which continues a long distance to the next (fourth) brido-e, becoming gradually bedded. The change takes place across the strike, and the rock dips 70° E. ; the passage being apparently from lower to hio'her beds. A little farther, east of the next (fifth) bridge, and thus still higher up, a stratum of thin and wavy bedded muscovite-gneiss occurs, which is quite exceptional so far east in this series. Then for a long dis- tance a "scaly" biotite-gneiss, often subporphyritic and rusting from the abundance of the pyrite which is disseminated through it, runs on in great THE BEOKET CONGLOMEEATE-GNEISS. 33 folds, the general strike coincidiug with the course of the railroad, until the large quarries on the north side of the road are reached. These were worked in 1887 by the Clark Hill Granite Company, Mr. J. H. Adams, of Dalton, being the principal owner. 1 am indebted to the superintendent, Mr. Hopkins, of Becket, who opened some of the first quarries in the region, for much information con- cerning the working of the quarry. Besides supplying much rough stone to the railroad and shipping many paving stones to Holyoke and other cities, this quarry furnishes a fine, light-colored granite of medium grain, obtainable in large blocks and suitable for all the uses of architecture, and a finer- grained, darker stone of very even grain, which, if it can be quarried in as large blocks as the bed promises from surface indications, will be very valu- able as a monumental stone and for all the finer classes of work for which granite is employed where its somewhat somber shade, when polished, is not objectionable. The "granite" extends far north into Clark Hill, on the south slope of which these quarries extend for a long distance, parallel to the railroad, and crosses the river to the south into Becket, where also are quarries. Some small segregated veins and lenses of pegmatite cut the rock at the quarry. The bedding of the granitoid gneiss of the quarry can be clearly seen,, and is nearly horizontal, corresponding with the more plainly foliated rock adjacent, along the railroad, which seems certainly to grade into the quarry rock. Between the next two bridges is again a great development of the same granitoid gneiss, followed by a thin, flat-bedded gneiss, banded in gray and reddish layers. Another band of the fine-grained granitoid gneiss separates this in the western entrance of the Coles Brook cut from the heavy, dark gneisses of the Algonkian. (See section, fig. 1, p. 22.) DISTRIBUTION. The rocks of this series occupy the western part of Middlefield, which is in Hampshire County, but beyond the limit of the map, and stretch across Becket, which is in Berkshire County. The broad band of workable granitoid gneiss seems to be continuous across the whole length of Becket, and it is used extensively by the Chester Granite Company, which obtains its materials from quarries in the eastern part of Becket, not far south of the MON XXIX 3 34 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. road running west from Chester station. This company has ah-eady put upon the market a large quantity of stone of the first quahty. The rock is, when pohshed, a clear, dark gray — too dark for many purposes — and when left with a rough surface is almost white, producing a marked contrast where the two kinds of surface are juxtaposed. The "sap" of the stone in the quarry is thin and white, showing it to be very durable, and the pyrite, which exists in small grains, seems not to be subject to oxidation, unlike that in the thin-bedded portions of the same rock. If it shall prove equally changeless in the worked surfaces after long exposure, the deposit is of great importance, as flawless blocks of the largest dimensions can be obtained, and the extent of the quarry rock is very great. The gneiss enters the area of the map again at the northwest comer of Blandford and extends, with similar characteristics, down the western side of the town, widening to the east so as to occupy the whole width of Tolland and half that of Granville. Following the band across Blandford, one finds it supper portion, nearest the mica-schist, to be everywhere thin-fissile, rusty, contorted, and more or less shot through by granitic veins; and where it widens out to the south the increased area seems to be occupied by these upper thin-fissile biotite- gneisses and worthless rocks, and west of Tolland the granitoid gneiss either passes down or has run out entirely. In some places in Tolland the rock approaches so closely the most feldspathic vaiiety of the next series — at the blacksmith shop in the village even containing large garnets — that I have questioned whether one or more folds of this series are not included in the older gneisses. It extends south into Connecticut as the western part of Percival's K 2,^ from which, on the east, the mica-schist is not separated. Far to the east the same gneiss rises again from beneath the hydromica-schists east of South Mountain, in the southern portion of Granville. It is here a gran- itoid gneiss of the common type, which extends southward into Connecticut, and is marked 1 2 upon Percival's map. PETROGRAPHICAL DESCRIPTION. 1. Granitoid gneiss from Clark Hill quarries, Middlefield. "Finest quarry stone." •J. G. Percival, Kept. Geol. Conn., 1842, p. 113. THE BECKET CONGLOMERATE-GNEISS. 35 A line-g-rained biotite-granitoid gneiss of gray color, with shade of brown. The deep-brown biotite is scattered through a fresh colorless mix- tui-e of quartz and feldspar. Titanite is so abundant as almost to deserve place as an essential constituent. The lens shows a wholly even-grained, very dusty mass. Under the microscope the quartz is characterized by the small number of inclusions it contains, rarely fluid pores with large, slow-moving bubbles in the largest grains. The long rutile needles are wholly absent; stout, flat muscovite microlites occur. Orthoclase appears in large, clear grains. Microchne is the most abundant and the most recent feldspathic con- stituent. Biotite in deep greenish-brown, jagged grains fits itself to all the other constituents, and so is of later formation. Muscovite appears in small quantity under the microscope. Titanite appears in large, well-formed crystals, wine-yellow, and in abundant smaller, irregular-clustered grains. Minute zircons, highly refractive, elongate, with rounded outlines, are not rare. Magnetite and titanic iron are wholly absent. 2. Granitoid gneiss from Clark Hill quarries, Middlefield. Coarse quarry stone. A medium-grained, light-gray muscovite-biotite-granitoid gneiss, whose clearer color, as compared with the preceding, is produced by the increase in the size of grain of the other constituents, while the mica does not increase in size or quantity. The lens shows larger, limpid grains scattered in a disconnected, granular, dusty, and micaceous groundmass, which is identical with the whole mass of the preceding variety. The quartz rarely includes rutile needles, and contains, especially in the larger grains, sheets of pores, often negative crystals, a few with large motionless or slow-moving bubbles. The orthoclase is in subporphyritic masses, rendered turbid, as usual, by an opaque white substance (kaolin?), which also occurs as an exquisite dendritic growth thrust out among the fissures between the quartz grains and appearing black by transmitted and silvery white by reflected light. In one quadrangular section of orthoclase cut about parallel to oo P cb 36 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. (010) a great number of sheets of fluid pores with moving- bubbles appear, arranged part parallel to P (001) and part parallel to go P (110). Microcline is abundant, with microperthitic structure. A plagioclase near albite occurs. The biotite is deep red-brown. Muscovite appears abundantly, in microscopic scales. Titanite is very abundant in congeries of grains, in one or two cases inclosing a grain of menaccanite. Zircon appears in regular square prisms P (111) «> P (HO), colorless. 3. Granitoid gneiss, Becket. The best quarry stone of the Chester Granite Company. Of slightly coarser grain than the best stone at the Clark Hill quarry, and of clear gray color — a muscovite-biotite-gneiss. The lens shows larger limpid grains in a porphyritic granular groundmass, which contains all the biotite and is somewhat dusty. The larger grains are mostly quartz, without rutile needles, and with minute fluid inclusions showing motionless bubbles of elongate shapes. Orthoclase occurs in rare, large grains, much dusted. Microcline is in secondary growths cementing a great number of grains together. It is very fresh. Plagioclase is rather rare. The biotite is deep greenish-brown. Titanite is visible with a lens, but is present in only small quantity in the slide. No zircon or magnetite occurs. CRUSHING TESTS. Prof. J. F. Kemp has given ^ some valuable facts in regard to the granite quarried by the Hudson and Chester Granite Company at Becket, Massachusetts. He says: "An analysis, which is the mean of two closely agreeing duplicates, was made by Prof. L. M. Dennis, of Cornell University, and the soda is given by difference, because in the NH^Cl and CaCOg used in the determination of the alkalies some sodium was shown by the spectroscope. ' Trans. New York Acad. Sci., Vol. XI, p. 4. CKUSniNG TESTS OP THE BECKET GRANITOID GNEISS. Analysis of granite from quarries at Becket, Massachusetts. 37 Moisture at 110° C Loss on ignitiou... SiO, Fe,0;, A1..0;, MuO CaO MgO KjO S NajO by difference Per cent. 0.08 .74 69, 465 2.30 17.50 Trace. 2.57 .305 4.07 .04 97.07 2.93 "Crushing tests were made on five sample cubes with the Emory testing machine in the School of Mines, and as preparatory to this the specific gravity was found on four cubes at 2.688, 2.687, 2.684, and 2.688. After three weeks' soaking these cubes absorbed water, respectively, 0.0021, 0.0021, 0.00224, and 0.0026 per cent. The cubes were first ground and polished so that the faces next the jaws of the crusher were parallel within a limit of error of 0.005 inch. The cushion employed between the cubes and the jaws was blotting paper. The crushing tests gave the following results: Crushing tests of granite from quarries at Becket, Massachusetts, Height inches.. Breadth inches . . Thickness inches . . Area sq. inches.. Maximum compression pounds . . Crushing strength per sq. inch . . 2.033 2.0 2.1 4.2 113, 200 26, 952 II. 1.983 2.13 1.99 4.23 122,000 28, 841 III. 2.059 2.02 2.03 4.1 106, 000 25, 853 IV. 2.011 1.97 2.03 4.0 101, 400 25, 350 2.009 2.03 2.03 4.12 108, 700 26, 383 ' ' The cubes exploded without previous cracking. This strength is excep- tionally high, as the general run of granite is far less. It does not, however, equal the elaiolite-syenite of Little Rock, Arkansas, which was tested by the late J. Francis Williams (see Annual Report of Arkansas State Geologist, 38 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. 1890, Vol. II), where results of over 30,000 pounds were obtained; but it is far beyond the requirements of building. "Cubes of the rock were also boiled in acid. One which was thus treated for five hours in boiling dilute HCl (1 part HCl of specific gravity 1.20 to 20 parts HgO) lost 0.59 per cent in weight. A second cube treated in the same way in boiling dilute H2SO4 (1 part H2SO4 of specific gravity 1.84 to 20 parts HjO) gave a loss of 0.48 per cent. Both these results indi- cate a great resistance to natural solvents. Two large cubes were placed in a muffle and maintained at a bright red for half an hour. One was allowed to cool just below redness and then plunged in cold water. It caused one crack that extended half through. The other cube was allowed to cool slowly in the air, and showed a thin external crumbling layer. When these results are compared with somewhat similar tests of other granites, as set forth by Mr. Gr. P. Merrill in his valuable work, Stones for Construction and Ornament, and with others in Vol. I of the Final Report of the Geo- logical Survey of Minnesota, and others by Dr. A. W. Jackson in the recent annual reports of the State mineralogist of California, it appears that the Chester^ stone endured well." THE GISTEISS AT SHEIiBUR]SnE. The oval area of gneiss on the Deerfield Eiver, at Shelburne Falls, has long attracted attention as a very striking illustration of erosion.^ It is a regular quaquaversal. The gneisses in the center of the area are in the main horizontal, though much contorted. Toward the borders they dip under a bed of hornblende-schist, which frames them beautifully, and this schist in turn dips outward on all sides beneath the mica-schists, and these dip outward also, with gradually increasing inclination. The erosion which wore through the newer beds domed over the gneiss has cut more rapidly into it, so that the gneiss occupies now the bottom of a deep circular basin and rises high up the sides of the surrounding hills, where it is capped by the newer beds. This basin is cut across by the Deerfield River and its tributary, the North River. The rock is very largely a biotite-gneiss of medium grain, granitoid and light-gray, as at the quarry by the railroad on the western boundary of the 1 This should be Becket ; the quarries of the company are in Becket and the workshops in Chester. 2E. Hitchcock. "Ten thousand feet of vertical thickness have disappeared." Elementary Geoloji.v, I860, p. 121. THE GNEISS AT SHELBUENE. 39 outcrop. Here it is not to be distinguished from the Becket or Monson gneiss. Under tlie microscope it is so fresh that the quartz and feldspar aj-e scarcely visible without polarizer. Below the falls the gneiss is greatly- dislocated, and many varieties alternate in much confusion. A white bio- tite-granitoid gneiss is followed conformably by a similar but thin-bedded rock. These are faulted against a greenish gneiss containing many inter- mixed fragments of schist, and against this rests the contorted hornblende- gneiss which furnished the beautiful bowlder now adorning the vestibule of the geological museum at Amherst, which was figured by President Hitchcock.^ The rock is made up of thin bands of a very hornblendic gneiss, alternating with equally thin bands of a white gneiss, and the whole folded with a remarkable complexity. On the south side of the stream the black hornblende rock rests upon the biotite-gneiss exactly as it does on the top of Bald Mountain (now called Massaemet), and it is not impossible that the deep basin has been formed by a sinking of its bottom about 1,200 feet. Bald Mountain is the eastern border of the basin. Toward the southwest of the area the rock is a thin-bedded biotite- hornblende-gneiss with few garnets and with pyrite. At the contact under the bridge on the road to Charlemont the rock is a rather fine-grained, thin-fissile biotite-gneiss, with few red garnets and some thick, compact quartzose beds. Above this is a very cortorted horn- blende-gneiss. On the road south from Shelburne Falls along the east side of the river, and near the south border of the gneiss, the latter wraps around a great mass of hornblende-schist, as if it were a granite rather than a gneiss. It is with some reserve that I identify this gneiss with the Becket and Monson gneisses. The gray gneiss can not be distinguished from the upper portion of the Monson gneiss, except that it is not "stretched." The thin- bedded hornblendic gneiss in many ways suggests the idea that it is devel- oped from the hornblende-schists which surround and once capped the gneiss, and it is unlike the hornblendic layers in the Monson gneiss. I have been brought to weigh these matters with care because of a more serious difficulty. At the Groshen antichne, next south, the calciferous mica-schists are broken through, and we have the normal section in descending order: 1. Corrugated schists = Conway schist. ^ Calciferous mica-schist. 2. Flags=Goshen schist. ) 3. Chloritic and hornblende-schists=:Hawley schist. IE. Hitchcock, Elementary Geology, 1860, p. 26. 40 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. Going west across Groshen and Cummington, we find the same series repeated and carried still lower, thus: 4. The upper sericite-schist=Eowe schist. 5. The hornblende- serpentine band=Ohester amphibolite. 6. The lower sericite-schist=Savoy schist. 7. The feldspathic mica-schist=Hoosac schist. 8. The Becket gneiss. Now in the Shelburne anticline one passes directly from the corrugated schists (1) to the gneiss (8), with only the intervention of a single horn- blende band, often not more than 50 feet tliick, and this bed thus replaces the Groshen flags and the whole sericite-schist series. It is true that the first bed of limestone above the hornblende-schist is white and slightly actinolitic, but it has a border of hard, black hornblende- garnet rock, so characteristic of the limestones of the Conway schist. An inspection of the map will show that the normal succession of the beds occurs across Shelburne exactly as across the towns north or south of the Shelburne gneiss, from which one is inclined to hesitate between three sup- positions: (1) That the Shelburne rocks are the sericite-schist (4 to 6 above) grown feldspathic; (2) that all the beds of the flagstone and sericite-schist series, so abundantly developed just to the west, have thinned out to the east, so that they are represented only by the thin hornblende band; and (3) that the granitoid gneiss is an intrusive rock grown gneissoid by pressure. I am inclined to accept the second supposition, as the hornblende-schist is almost certainly the continuation of the Hawley schist, and one may assume that the gneisses formed an island larger than the present exposure during the deposition of the sericite-schists and the flagstones. The diminished thickness of these two series east of the Connecticut harmonizes with this assumption. The coloring adopted on the map accords with this hypothesis. Contacts. — Groing south along the west side of the river into Conway, 20 rods north of L. W. and B. A. Andrews's house, one passes for a long distance over a thick-bedded, white biotite-gneiss, and finds this changing, in the hillside west of the road, into a thin-bedded hornblende-biotite-schist with garnets and pyrites. The transition is sudden to the hornblende-schist above, and the two rocks are not separated by any fissure, but are welded together intimately. The schist is a thin-bedded hornblende-schist with few garnets, black, THE MONSON GNEISS AND ASSOCIATED KOCKS. 41 lustrous, with some beds , gneissoid and some marked by the absence of liornblende from spots which appear Hke porphyritic feldspars but are composed of a granular feldspathic mass. At the top of the hornblende- schist the contact is also visible, and the change is sudden into a rather coarse, slightly rusty, gray muscovite-schist with few garnets. Directly across the river, back of J. Dole's house, the rather coarse white gneiss is followed immediately by an arenaceous hornblende-schist, gneissoid as before, and this is separated from the mica-schist above by a small mineral vein. THE MONSOK GNEISS AND ASSOCIATED BOCKS. Amos Eaton says,^ referring to the gneiss range east of the river: "This range evidently passes under the Connecticut River, accompanying the granite and covered by other strata, and rises with it on the western side," and I have, myself, no hesitation in associating the bands of gneiss which cross the State east of the Connecticut with the Becket gneiss on the west of the river, on both lithological and stratigraphical grounds. They are, however, nowhere known to come into visible contact, and in default of this final proof of their identity I may consult convenience and give this rock also a separate name and treatment. It is the C 4 of Percival.^ Beginning north of the great bend of the Connecticut, opposite Middletown, it runs north, and in a quaiTy at Portland, to which I was kindly guided by Prof. WilHam North Rice, of Middletown, it is so exactly like its con- tinuation farther north that in hand specimens and in mass it could not be distinguished from the products of the quarries of Monson or Pelham. It enters the State from the south in two narrow bands, separated by newer rocks, and the eastern band is limited on the east by the deep sand-fiUing of the central valley of Monson. The two bands of this rock, separated by an infolded complex of hornblende- and mica-schists, and bounded also on the west by a repeti- tion of the latter, may be followed across Monson and Wilbraham into Palmer. Here they are all twisted together in extreme metamorphism to form the hornblendic border of the intrusive tonalite (syenite, Hitchcock), from which they extricate themselves in the latitude of Belchertown village, 'Index, 1820, p. 119. ^J. G. Percival, Kept. Geol. Conn., 1842, p. 233. 42 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. and the gneiss then extends continuously across the State to Northfield, where it is partly covered by newer rocks before reaching the State line. The eastern band runs north to Orange, where it disappears completely within the limits of Massachusetts. On returning to the study of the Monson gneisses, after long experience with the change of the Cambrian conglomerates into the white gneisses in the Berkshire Hills, the traces of the same change struck me in the stretched gneisses of Monson and Pelham. The traces of pebbles may now and then be clearly seen, and I present a reproduction of a photograph of the north- east corner of Walker Hall, one of the buildings of Amherst College, which shows this clearly (PI. I, p. 64). The rock is from Monson, and in 1890 a great wall of conglomerate was exposed in the quarry just north of the trap dike, but it was all quarried away in 1892. In many cases the flat patches of lighter color and of long elliptical shape which appear on the cleaved foliation faces of the gneiss seem to be the remains of pebbles wholly flattened out into films, as was suggested by President Hitchcock in his remarkable investigation of distorted pebbles.^ THE PELHAM AND WILBRAHAM AREA. THE GNEISS. ' The broad anticline of this area enters the towns of Northfield and Warwick from New Hampshu-e, and though its surface is at first covered in part by isolated areas of newer rocks, it soon expands to a greater width than any other gneiss in the counties, and maintains this width nearly across the State, interrupted by the protrusion of the Belchertown tonalite. It is in Northfield a fine quarry stone, especially marked on foliation faces by small squarish blotches of jet-black hornblende, and it continues to be good quarry stone in large part clear across the State. It differs curiously from the other areas in that it is, across the central portion of the State, a broad anticline with all its central portions almost horizontal and at the edges bending down quite sharply beneath the newer rocks. A further distinction of this area is found in the presence of a great bed of an actinolite-quartzite, which will be S23ecially described, and in the presence of three great intrusions of an olivine-enstatite rock, which, with its complex contact phenomena, will be also the subject of a separate chapter. ' Geology of Vermont, Vol. I, 1861, p. 28. THE PELHAM AND WILBRAUAM AREA. 43 To the south the rock is coarser than in the other areas, and in contact with the great mass of the Belchertown tonalite is considerably altered. In Wilbrahara its attitude is nearly vertical, and it forms the core of an anticline which is slightly overturned to the east, as the dips are high to the west. At Power's mine, in Greenwich, on the high hill overlooking the house of S. B. Estey, considerable blasting has been done upon a vein of coarsely granular magnetite, containing much coarse red garnet and pyrite — an entirely worthless deposit. PETROGRAPHICAi DESCRIPTION. 1. Granitoid gneiss from Massachusetts Agricultural College quaiTy, Pelham. This may be taken as a type of the Monson gneiss. A very clear, fresh, gray, stretched biotite-gneiss. It is a most crisp and friable stone, showing no trace of decomposition, the fresh black biotite appearing in the mixture of limpid quartz and feldspar. Titanite is an abundant constituent, and rarely a trace of epidote appears in the neighborhood of the biotite. The lens shows the jet-black biotite scattered in an almost limpid granular mass, with faint trace of porphyi'itic structure and slight nacreous dusting. Under the microscope the quartz shows swarms of minute inclusions, with groups of larger cavities having moving bubbles. One grain alone was filled with long rutile needles, and this had a slightly reddish shade. Orthoclase occurs in larger crystals than the other constituents and includes rounded quartz grains. It is quite abundant. Microcline is abundant and of late formation, crystallized out so as to cement a great number of quartz grains. Albite occurs rarely. The biotite is in separate black scales, and with the lens is seen to be abundant, much notched and often extended to include several quartz grains. Titanite is in angular grains of the same size as the other constituents, and in distinct crystals, pale greenish-brown exteriorly and deep red- brown in the interior, the boundary between the two colors being generally distinct, but in one case a red-brown crystal is inclosed by a pale-yellow one, the two being of common orientation and the outer bounded by fewer faces. 44 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. Zircon is quite regularly disseminated in colorless to pale-yellow crys- tals one-fiftli to one-tenth tlie size of the other constituents and regularly crystallized in stout prisms, some almost cubical. The forms P, 2 P 2, co P oo could be seen in one crystal. It is included in all the other constituents. Magnetite is absent. 2. Biotite-giieiss from Bassett's quaiTy, Northfield. A fine quarry gneiss, light-gray. On the foliation faces distant, squar- ish, thin plates of hornblende make the rock appear as if ink-spattered. The feldspar appears as glassy as the quartz, which is common in all these gneisses, though it shows traces of change into muscovite under the microscope. There are present orthoclase, microcline, and albite; a fragment of the latter gave extinction -f 15° on go Poo , and the tri clinic feldspar in all these gneisses give commonly an extinction of -+-4° on either side of the twinning sutures on P. Fine minute zircons are present, but no titanite. 3. Hornblende-gneiss from southwest Shutesbury, opposite W. Thresher's, adjoining trap dike. It is a sandy-granular rock of very fine and even grain, and of very dark-gray color. It is a rock quite common in the Monson gneiss, and found also in the Becket gneiss, in the northeast of Tolland. It becomes much more abundant in the eastern area, in its southern exten- sion into Connecticut, where it is Percival's C 3Mn its eastern portion. Microscopical character : The background is made up of little quartz, little albite (extinction 6° on either side twinning lines), and much limpid orthoclase, without cleavage, and determined only by its positive biaxial character. The abundant hornblende molds and incloses the other constituents; it shows peculiar basal cleavage in fine, close, straight lines. Its absorp- tion and pleochroism are exceedingly strong jc>tr>a. c=deep blue ; tt=deep olive; a=bright yellow; much deep-green biotite and large light-red garnet, many plates of tremolite, miich black and red ore, and a single group of leucoxene grains. 4. Biotite- gneiss from east foot of Mount Hygeia, upper quarry. A white gneiss, making heavy beds above the normal gneiss of Pelham, dif- fering from it by the small amount of black biotite in distant scales and the abundance of small red garnets. The quartz contains no rutile needles, and is in rounded grains that 'Eept. Geol. Conn., p. 222. THE PELHAM AND WILBKAHAM AKEA. 45 suggest water-wear. These are cemented by newly deposited quartz and feldspar. It contains cavities, which are often negative crystals with very large, motionless bubbles, and other long trains of cavities, showing in great numbers smaller bubbles in rapid motion, not affected by being heated to 70° C. Orthoclase predominates. Albite and microcline are present. Biotite occurs in deep brownish-green scales. There is little muscovite. A single square prism of deep-red rutile was seen in the slide. Zircons are rare. Single large grains of menaccanite were seen, changing to leucoxene. THE ACTINOLITBQUARTZITE. The central portion of the Pelham gneiss area presents two peculiari- ties as compared with the other similar areas, viz, the series of olivine- enstatite rocks and the great quartzite beds here described. The biotite of the gneiss disappears at a certain level and reappears again as suddenly, leaving a great bed, perhaps 300 feet thick, between two beds of the Monson gneiss which can not be distinguished from each other. The intervening quartzite bed varies from a fine-grained quartzite to an equally fine-grained quartz-feldspar mass, with needles of tremolite or pale grass-green actinolite, just visible to the eye, scattered through the mass. It becomes at times a more distinctly bedded rock, and almost continuous films of the same pale-green actinolite appear on the foliation faces. Small garnets are quite commonly disseminated, and at times distant, minute scales of an amber mica replace the actinolite. Distribution. — The outcrop of the rock is quite peculiar and depends upon the great flatness of the dome of the gneiss synchne in Pelham. The bed is exposed by the double scalping of the undulating surface of this syncline, and appears, therefore, in one closed ring in Shutesbury and in a loop open to the south in Pelham. Beginning in the northwest corner of Belchertown, it runs north along the eastern slope of the Pelham range, passing just east of Pelham post- office and just west of the poor farm, and continues north through the center of Shutesbury and a little beyond it; then it turns sharply southwest, and its dip, which had been low east, becomes westerly. It then runs southwest into Pelham again and ends in the high peak of Hygeia. Its 46 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. extension is shifted more tliau a mile eastward to the waterworks dam by a fault, and it continues from this point southwardly, passing- east of the "asbestos mine." 'The other great area occupies the eastern portion of Leverett, is cut off on the north by a fault at the Rattlesnake Grutter, and projects southwardly across Shutesbury into the western portion of Pelham, where it ends under Mount Hygeia, approaching near the other band. PETKOGRAPHICAL DESCRIPTION. 1. Adinolite-tremolite-gneiss from Northfield, east of R. H. Minot's, and adjoining the great north-south fault. A greenish-gray, stretched, ligniform rock, the abundant needles just visible to the eye. In a fresh, granular quartz-orthoclase base the abundant parallel needles of pale-green actino- lite and tremolite appear. They show a delicate, close prismatic cleavage, distant, strong basal parting, and strong absorption and pleochroism. Green and brown biotite are abundant, and there is much magnetite. This is the only occurrence of the rock in the western portion of the Pelham area, and it is soon cut off on one side by the fault and on the other by granite. 2. Micaceous quartzite from Pratts Comer, southwest Shutesbury. In a white, fine-grained, only subgranular quartz mass are scattered small, rounded, red scales of biotite. No feldspar. Under the microscope the quartz mass shows only rarely a fissure, and is so homogeneous, colorless, and free from foreign bodies that it is invisible in common light. "With crossed nicols it shows a most complex mosaic of interlaced grains The red scales of biotite are all in parallel planes, and inclose zircons which are surrounded by a marked deep-brown pleochroic border. The zircons are also scattered through the quartz in fine crystals. Red-brown, stout rutile prisms occur surrounded by a granular, colorless leucosene. 3. Actinolite-qtiartsite from north side of brook and about 100 feet east of the dam of the Amherst waterworks in Pelham. To the eye the rock is like a fine-grained, white sandstone or crisp, friable quartzite, with scat- tered needles of pale-green to almost colorless actinolite. The lens rarely detects a grain of feldspar. It is whiter and contains less actinolite than the Mount Hygeia rock, but is closely like it. Under the microscope THE PELHAM AND WILBRAHAM AREA. 47 the fresh o-ranular quartz is free from fluid pores and acicular microlites. Feldspar is uot distiuguishable. The nearly colorless actiuolites are parallel, and contain large, rounded grains, common also in the quartz, which are strongly refringent and polarize brilliantly; they may be zircon. Other grains clustered along the actinolite crystals seem to be epidote. 4. ActinoUte-quartzite from east bluff of Mount Hygeia, Pelham. A granular quartz, white and of medium grain, with parallel needles of color- less to aquamarine actinolite. In the granular quartz ground the actinolite needles are irregularly arranged; here and there is a scale of biotite. There are large zircons and microlites inclosed in quartz and actinolite ; also grains of titanite. SAXONITE AND SERPENTINE IN MONSON GNEISS. THE PELHAM ASBESTOS QUARRY. This locality has been long known as furnishing large masses of a hard asbestos, and the mineral has been extensively quarried. Its interest from a mineralogical point of view was greatly increased by the discovery in 1869, by Mr. A. B. Kittredge, of corundum in hard nodules in the biotite, which occurs there in great abundance. Later, Pro- fessor Shepard, observing the difficult fusibility of the "asbestos," analyzed it and found it to have the composition of bronzite, but gave it the wholly superfluous name asbestite. He also analyzed a tough, black, granular mineral which occurs in large masses in the deeper parts of the several excavations and found it to have the composition of oli^dne, but named it pelhamine, a name equally supei-fluous, as the mineral is optically as well as chemically identical with olivine, and its black color is due to dissemi- nated magnetite and chromite. The pits by which the bed is exposed are scattered for a distance along an eastward-sloping hillside, and as the dip is 40° W., while the strike of the inclosing Monson gneiss is due north, the lenticular mass is exposed by erosion in a plane at right angles to its dip, giving a length of about 200 feet and a greatest thickness of 40 feet. This is a great lens or short dike — probably an old volcanic core — of the highly basic igneous rock saxonite, in the highly acid conglomerate 48 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. gneiss; and during the strong metamorphism which has transformed the conglomerate into a gneiss, the saxonite has been largely changed into anthophyllite, and, what is of higher interest, a broad selvage — a "reaction rim " on a grand scale — has formed by the mutual influence of the basic core and the acid surrounding. This selvage consists largely of minerals containing little or no silica — apatite, corundum, magnetite, tourmaline, anorthite, and biotite. It wraps around the saxonite with great irregularity, often folding deeply into its mass. Measured downward from the hanging wall, the mass is greatly decom- posed for a depth of from 3 to 12 feet, and as the pits have been sunk in this decomposed portion and work stopped as the hard unaltered rock was '^^^±¥*€rt^*"''^^^^^ 8 FT Fig. 3.— Southwest wall of Pelham asbestos quarry in 1890. A, Pelham gneiss; B, anorthite rook; C, black tourmaline masses ; D, biotite layer ; D', vermiculite layer (Termiculite and steatite from biotite and actinolite) ; B, saxonite ; F, anthophyllite derived from saxonite. approached, attention has been directed almost entirely to this decomposed portion, which will be discussed a little later. The pits everywhere disclose at the bottom the black rock, which consists of granular olivine, chromite, magnetite, and bronzite (saxonite), without admixture of any other minerals, and of this rock the whole intruded mass originally consisted. At the top of the wall in the large central cutting the contact of the superincumbent gneiss upon the olivine rock can be well studied, and it is very peculiar. (Fig. 3.) The gneiss penetrates the olivine rock in a great club-shaped apophysis, the lamination of the gneiss being first bent down THE PELHAM AND WILBKAHAM AREA. 49 toward rlu' latter iuul then somewhat confused, but distinctly traceable far iiUi) it and dyiui;- oTit gradually by the slow disappearance of the biotite and ([uartz, until the whole of the great projection is made up of a mass of snow-white, extremely fine-granular, massive anorthite, carrying toward its borders a large quantity of black tourmaline in great irregular bunches, which, at the apex of the mass, afi'ords blocks of pure, coarsely crystalline tourmaline over 2 feet across. The fluorine of the biotite has gone into the tourmaline. Many small crystals of allanite are shot through the anorthite, and when broken across the latter mineral shows the usual puckered surface radiating from the allanite. ■- ■•■;> 8 FEET FiS. 4.— "West wall of Pelham asliestoa quarry=rigbt half of fig. 3, qnarried deeper. Letters as in flg. 3. G, cortmdiim nodules; H, apatite. The tourmaline breaks into large imperfect crystals, often 8 to 12 inches long, and in cavities shows terminations always Avith broad P planes. It contains, in cavities between crystals, zoisite, apatite, and beau- tiful geniculate twins of rutile, together with perfect apatites alone in other cavities and in the mass itself. Under the microsco^ie it often shows an exquisite micropegmatitic intergrowth with the anorthite, the latter taking the place of the quartz in graphic granite, while the tourmaline is extin- MON XXIX 4 50 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. guished as a single individual over broad surfaces and in large disconnected reticulate portions. At the line of contact of the tourmaline and the anorthite many microscopic deep-red rutiles occur. This anorthite projection, which is a thickened part of the outer layer of the "reaction I'im," is separated from the olivine rock by a thick band of a deep bronze-colored biotite in large scales, which is wrapped around and extends beyond this projection. As it separates the gneiss at the north end of the bed from the olivine rock, it may very probably be a con- tinuous layer around the latter. Nothing can be seen of the lower contact. This biotite band, usually 4 to 8 inches thick, reaches in places a thick- ness of nearly 4 feet, and incloses hard nodules of a blackish-green matted hornblende and of the finest emerald-green parallel-fibered actinolite, and other similar nodules which contain large, imperfect crystals of gray corundum with central spots and streaks of rich sapphire-blue (see fig. 4), which are wrapped in a greenish chloritic mineral of large axial angle and marked pleochroism, probably clinochlore. Still other nodules contain large friable masses of a fine grass-green actinolite. It is interesting to note how the extremely basic character of the olivine rock is continued outward in the biotite-corundum rock and beyond in the anorthite-tourmaline rock, and to observe how uniform this collocation of minerals is in all parts of the world — a subject to which I recur after describ- ing the Chester emery bed (Chapter VI). The broad border of decomposition products of the olivine rock men- tioned above is of the highest interest, and for its understanding reference may be made to the accompanying fig. 3 (p. 48). Between the biotite (d) and the unchanged olivine (e) is a layer, generally about 3 feet thick, which, nearer the outcrop (at the left of the figure), is 13 feet thick, and consists of olivine changed in part to a pale-yellow, friable, granular villarsite, and in part to an earthy mass of ochery appearance. Through this runs an irregularly anastomosing network of veins of fibrous antho- phyllite (f), which reach at times a thickness of 8 inches, at times run out to extreme thinness and disappear. They are for the most part made up of a woody mass of fibers, which are placed at right angles to the walls of the vein and meet on a suture at the center. In the thicker veins the visibly fibrous poi'tion exists only a few inches fi-om the walls on either side, and the central portion is made up of a compact, woody THE PELIIAM AND WILBRAHAM AREA. 51 mass, splittiiiji' in ;i direction at right angles to the walls, and pearl-gray when not blackened by manganese. More rarely the vein filling is completely asbestiform and the fibers cross the vein from side to side; very often they are all bent somewhat to one side or the other as they approach the wall, being compressed by their own growth. In other parts of the excavation these veins have swollen to much greater width, and great ligniform masses, 20 to 30 inches in length, have been excavated. This is the "asbestos" of the quarry, and many hundred tons have been excavated and sold for grinding into paint and for asbestos papers. The resemblance of this structure to the well-known microscopic olivine network is extremely striking, and it would seem difficult to avoid the conclusion that the anthophyllite here must be of secondary oi-igin and a derivative from the olivine, probably under conditions of considerable pressure and heat, and therefore at an early period in the history of the changes which the deposit has undergone. Its exact resemblance to the transverse fibrous vein fillings of calcite, gypsum, and chrysotile will hardly admit for it any essentially different explanation. The anthophyllite occurs also in large, rather coarse-matted fibers. It polarizes very brilliantly and is quite fresh and limpid, the gray color being due to fine magnetite dust. At the northern excavation and at the large opening there are sparingly disseminated in the fresh oUvine rock squarish plates, J to ^ inch across, of a pale bronzy enstatite or bronzite, making an ordinary olivine-enstatite rock. This is a primary bronzite. Masses of a bright emerald-green actinolite in matted fibrous arrange- ment of the single crystals were produced from the large opening, but their relations to the other minerals can not now be observed. At a new excavation made during the year 1883, near the south end of the bed, a long band of this mineral was struck just below the drift, and resting upon the thick decomposition layer of anthophyllite, in the midst of which several thin layers of the actinolite also appeared. The biotite containing nodules of the dark-green hornblende here also folded deep into the saxonite, as at the large cutting. The anthophyllite layer was followed in the bottom of the excavation by the usual black, undecomposed olivine rock. The biotite has also been attacked on a large 52 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. scale by a decomposition which has resulted in the fonnatiou of a mass of soft, greasy scales of vermiculite (D' fig. 4), which, when boiled with sul- phuric acid, yields a residue of white scales of pure silica. This has been named pelhamite by Prof J. P. Cooke, and thus the town of Pelham has lent its name to two equally poor minerals. PETROGEAPBICAL DESCRIPTION. 1. Saxonite, or olivine-enstatite rock (pelhamine, Shepard). This is a very fresh mixture of olivine and enstatite, both dusted thi-ough with black ore, largely chromite. It is a dull-black rock of very great toughness. The olivine grains have often many crystalline faces. The enstatite is in rare, small plates, with parallel sides and ii'regular ends, and with a fine wavy lamination, which is often marked by lines of black ore generally concen- trated in some part of the plate, especially the center. Although neai'ly colorless or pale bronzy in common light, it has marked pleochroism. It is plainly rhombic, and grades into the asbestiform decomposition product in veins running tlu-ough the section. The distant, strongly marked transverse cleavage so common in enstatite is wanting. 2. Secondary asbestiform anthopliyllite occurs in the altered saxonite in clear gray masses parallel or matted fibrous, in the former case so fine- grained as to resemble silicified wood, in the latter made up of a mass of short needles without radiated structure. It has very harsh feel. With a lens it seems to be entirely fresh, transparent, and colorless, the gra}?- color being due to disseminated magnetite, which is visible, and may be removed fi'om the powder by a magnet. Under the microscope it presents a mass of colorless needles and blades with delicate longitudinal striation, which breaks off here and there against a transverse cleavage. The needles are broken across by a distant fracture not exactly at right angles to the length. Long, fine, straight needles, breaking up at times into a row of grains, are present, and thoixgh not very abundant, are concentrated more in the center; they appear black, but at times red with high powers. In crystals cut across the blades the form and cleavage of hornblende can be detected, and I was able to separate and measure one needle, obtaining 55.30°. They polarize brilliantly, and always strictly as rhombic crystals, and this is the case with the silky asbestos. 3. The ]}lagioclase-feldspars of the contact zone. Professor Shepard analyzed the two varieties of massive triclinic feldspar found in the great THE I'ELIIAM AND WILBEAHAM AREA. 53 apopliysis ])cuetratinj>' tlie saxonite: (a) tlm white saccliavoidal portion ibrining- its extremity and nearest to the oHvine; (h) the Ijhiish-white, coarser- grained portion which formed the neck of the mass and passed into the gneiss. For the former he fonnd the composition of anorthite, and for the Litter that of andesite. The Litter portion, as it approaches the common gneiss, is less pnre than the other, containing much biotite, but with the microscojje the characters of anorthite were presented clearly by both varieties; here and there, however, the larger crystals were very distinct and were clearly andesite. The mass is like the feldspar accompanying the "fringe rock" of the Chester emery bed. The portion called andesite by Shepard is compact to fine-granular, translucent, bluish-white, fresh-looking, showing slight flesh color from the abundance of small disseminated biotite crystals, and having seams and irregular masses of black tourmaline scattered through it. Occurring largely in the latter, and more sparingly disseminated in the feldspar, are minute crystals of zircon: The anorthite grains are often almost entirely single individuals; here and there a few very fine distant twin laminae are interposed, but these run out in a short distance, and in many cases the whole surface is covered by distant laminge lying at right angles to each other. The maximum extinction was 31° to 34°. Under the microscope the feldspar shows through a lacelike network of brightly polarizing films or raveled-out scales of muscovite, and this increases until in slides cut from seemingly quite fresh material the feld- spar can scarcely be distinguished in the mat of mica scales. The biotite is optically uniaxial, and is often decomposed wholly or in part, the sides being dissolved into a congeries of colorless scales, or the change attacking one or several of the laminae and proceeding quite across the specimen; and much of the new mineral has wandered out and surrounds the biotite crystals in large spots, which, with reflected light, are seen to surround the remnant of the original crystal like a growth of glistening white mold, and these white spots are visible to the eye all over the slide. The zircons are white, with a faint tinge of red and a high adamantine luster, or deep amber color to pale red by reflected and reddish olive-green by transmitted light. The white crystals are most regularly-formed, long, square prisms with sharp termination P and 3 P, and apparently 3P 3. The 54 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. red crystals show at times a sharp prismatic cleavage, being often very irregular lobed masses and often very regular crystals with shining faces. The andesite (Shepard's type) crystals are about 20 by 10 by 10"™, not bounded by distinct faces, but embedded in a granular, compact mass of anorthite, translucent and bluish- white, with many small scales of biotite intermixed. Sections cut parallel to P, go P &, and oo P oo were much decomposed and impregnated with scales of muscovite, placed principally in the planes of principal cleavage and in especially large plates parallel to 00 P do. It is poly synthetically twinned parallel to oo P do in broad continuous plates of equal width; extinction on co P do, — 13° to— 14°; on OP, — 4° to — 5°, agi-eeing thus exactly with andesite 4. Serpentine from Pelham, about 100 yards west of the "asbestos'' mine. Dull black-green serpentine, changed an inch deep into white talc and showing deep in the interior "ph8estine"-like aggregations of talc, from alteration of the bronzite. Slides cut — (a) from the outer talc layer, (6) from the layer of transi- tion, (c) from the interior deep-green serpentine — showed: (a) A mass of wavy talc scales containing remains of bronzite, with long jet-black hairs, slightly curved, placed parallel to the vertical axis of the bronzite and ending on the mass of talc scales into which the mineral is decomposed; (V) a matted mass of actinolite fibers of pale-green color, very strongly dichroic; (c) a mass of actinolite fibers and talc scales, with small portions of serpentine and in places with grains showing the olivine network, the whole having the outward aspect of a common serpentine. The specimens labeled "Black serpentine and talc, Pelham, Mass.," No. 132, in the Massachusetts State Survey collection of President Hitch- cock, is from the above locality. 5. Epidote-gneiss from Pelham; asbestos quarry. An even, fine- or medium-grained mixture of flesh-colored orthoclase, gray quartz, and bright pistachio-green biotite, with small black grains of tourmaline. Under the microscope the epidote is scattered in long crystals exactly like the plagioclase in a diabase. Minute veins are entirely filled with epidote; only orthoclase is present as a feldspathic constituent, and around nuclei of this broad bands of epidote needles are arranged parallel to the former cleavage planes or outlines of the feldspar, showing the nucleus to be only a remnant of a larger crystal. THE PELHAM AND WILBKAHAM AREA. 55 TIIK rm.EIAM SERPENTINE. About 325 feet west of the asbestos mine, on a small flat which inter- rupts the western slope of the hill on the eastern side of which the mine is situated, a great boss of serpentine rises through the till, and a little south of it a second, of which it can only be said that they occur within the limits of the Monson gneiss. The rock is a deep dull-green, opaque when wet, and containing chromite in some abundance. Over a large portion of its surface it is changed for some distance inward into a white talc, and as this change follows the surface of the rock it is plainly a change of the serpentine into talc since the erosion of the Glacial period. THIS SHUTESBUKY SERPENTINE. A second locality identical with the "asbestos" mine in Pelham occurs a mile south of the village of Shutesbury, in a pasture south of the house of C. Leonard. Fragments, some of large size, lie over the surface in a space a few yards square, turned up by plowing. One large mass of rusty-brown, half-decomposed olivine rock, shot through by white anthophyl- lite fibers and full of chi-omite, is not to be distinguished from similar masses at the Pelham locality. The fibrous asbestiform and woody varieties of anthophyllite are repeated here also, and masses of a green chloritic mineral occur. The dej^osit is surrounded on all sides by outcrops of the Monson gneiss, but its exact relation and size can not be determined. THE NEW SALEM SERPENTINE. This locality is situated on the west slope of Rattlesnake Hill, about 300 yards northeast of A. A. Haskell's house. The country rock is a rather coarse biotite-granitoid gneiss, striking north-south and dipping 90°. The old digging is covered, and no contacts can be seen. The olivine rock is at most 50 feet wide and may be 150 feet long. The nearest outcrops of the gneiss are wholly normal and do not betray the presence of the foreign body. This is apparently a lenticular mass, its greatest diameter coinciding Avith the strike. The greater portion of the rock taken out is deep dull- black olivine, with small glistening scales of a micaceous mineral, appar- ently clinochlore. The rock weathers to a pale isabella-yellow from the removal of the black ore and the hydration of the olivine. Associated with it in some quantity is a fine fibrous light-gray anthophyllite, largely altered to an imperfect steatite. I was guided to the spot by an aged man 56 GEOLOGY OP OLD HAMPSHIKE COUNTY, MASS. who had owned the land for more than half a century, and I asked him how the rock had been discovered in the thick woods. He told me that when he was a small boy his father had cleared the hillside, a desolate slope consist- ing largely of uncovered ledges, and his older brother, while harrowing in oats on the spot, noticed that the harrow teeth made no noise over one por- tion of the ledge, but gouged deeply into the rock. He thereupon took a large piece of the rock home and put it into the fire, but could not melt it. His pyrognostic experiments do not seem to have proceeded much further, but long after, abovit thirty-five years ago, the owners dug a deep trench into the mass, dumping a great quantity over the bluff, but did not find anything of value for use -as soapstone, only a small portion of the rock having completed the change to steatite. THE ORANGE AND MONSON AREA. This band of gneiss extends nearly across the State as a naiTow anti- cline, and near its north line in Orange the axis of the anticline dips down northwardly beneath the fibrolite-schists. Because it yields more readily to erosion, the gneiss occupies the bottom of a deep amphitheater open to the south, its bottom deepest outwardly, just at the foot of the sharp, higli schist hills beneath which it sinks. At its northern end the gneiss is quite granitoid and much disturbed by small intrusions of pegmatite. Around Orange village it is a fine quarry stone. Much of it is a dark biotite-horn- blende-gneiss, much a lighter gneiss containing angular fragments of the darker variety, and very tortuous. At the railroad east of Orange village the light-colored granitic gneiss folds around great fragments, or groups of fragments, of the dark hornblende- gneiss, which have been but slightly moved and cemented by the lighter o-neiss. In this it resembles the Shelburne Falls gneiss. Two east-west faults, 17 feet apart, here include a much darker and more hornblendic o-neiss. It contains prehnite and stilbite in fissures. All down its western border in Orange its contact with the schists above is more like that of an eruptive with an overlying sedimentary than like that between two sedi- mentary beds. At L. Mayo's it is very granitic and is intermixed with the lower schists in a confused way. In the village of Orange, between Main and High streets, it is in direct contact with the hornblende -schist, and it continues in contact with the schist across into New Salem. In this THE ORANGE AND MONSON AliEA. 57 town au Geology of the Connecticut : Am. Jour. Soi., Ist series, Vol. VI, 1828, p. 19. = Geology of Massachusetts, 1835, p. 332. 3 Boston, 1871. 72 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. filled with flattened lenses, placed witli the bedding, of quartz and of quartz and feldspar ; and from this point the road runs for nearl}^ a mile through an almost continuous cutting of the light-gray, flat-bedded hydromica- schists belonging to the next series — the Rowe schists, which are without accessory minerals of any kind and preserve a monotonous uniformity. The dip is nearly vertical, and the section line is nearly at right angles to the strike. RELATION TO THE BECKET GNEISS. It has been already stated that at the brook junction east of Bancroft station the change from the compact, flesh-colored, granitoid gneiss of the Becket series below to the porphyritic hydromica-schist is abrupt; The contact between the two is exposed for a good distance and is a fissure, the rocks not being welded together, and the discordance in strike is consider- able and in dip very large. There is certainly unconformity and probably faulting at this point, but the nearness of the point to the area where the Becket gneiss is so irregularly wrapped around the southern end of the Hinsdale gneiss, and the fact that the Becket gneiss is peculiarly irregular in structure clear up to the junction, make it possible that this unconformity is only local, and at all events deprive it of a decisive weight in setthng the question of real unconformity. The junction can be well followed north from the railroad to the center of Middlefield, and the series retains exactly its character, showing a great development of feldspathic hydromica-schists and imperfect, sandy gneisses, and above these a much greater mass of barren, gray, and green-blotched schists, belonging to the Rowe schists. The transition between the two series — the gneiss and the feldspathic schist — is best studied between the village and the Fair Grounds in Middle- field Center. Just below the point where the roads join at M. Smith's, south of the Fair Grounds, the Rowe hydromica-schist, while retaining exactly its dip and strike (strike N. 10° E., dip 70° to 80° E.), its flat-fissile, schistose appearance, its gray surface spotted with green, and the multitude of small corrugated and twisted quartz lenses, becomes indistinctly porphyritic, the feldspar here and there cementing together a group of sand grains. As one goes lower (i. e., westerly) this alternates many times in thick and thin beds with the common hydi'omica-schist, often chloritic, until the beds which strike through the Fair Grounds become a quite well-characterized THE IIOOSAC SCHIST. 73 gneiss, but still tilled with the small tortuous quartz veins, and differing from the Beeket gneiss below by the presence of two micas, the muscovite being the prevailing variety. In the village itself, but a few rods farther west, the true Becket gneiss appears and occupies all the region westward with exactly the same strike and the same high dip, and though the exact line of contact is not exposed, there is nothing to suggest unconformity. Everything here points to a gradual passage of the gneiss up into the hydromica-schist. On the other hand, the rocks here all stand vertical side by side and have been subjected to the greatest compression, and the traces of an unconformity of considerable importance may well be masked. However, going south from the railroad across Becket, Blandford, and Tolland, along the winding junction line of the two formations, one finds marked evidence of a considerable unconformity, in that while the newer formation conforms in strike to the undulations of tlie boundary line, dip- ping away from it to the east, the strike of the older is in all this distance uniformly N. 40° to 45° E., almost at right angles to the boundary, and thus to the strike of the newer rocks. I conclude that the unconformity between the two formations is general, and that the feldspathic character of the lower half of this forma- tion is due to its derivation from the older gneisses, against which it rests in the form of a coarser, feldspathic material, while the upper portion was a more arenaceous sediment, largely deprived of its alkaline constituents, and this conclusion seems to me strengthened by the study of the same junction on the east side of the Connecticut. THE GRANVILLE AREA. This area comprises Blandford, Tolland, and Granville, in Hampden County, and Hartland and Granby in Connecticut. South of the railroad section given above, along the south line of Mid- dlefield, the feldspathic mica-schist continues across Becket in Berkshire County to its southeast corner, and there it enters the Granville quadrangle at its northwest corner, and at the same time Hampden County. Its relations, especially to the Rowe schist above, can best be studied on the road west from Chester, where the pale greenish-gray hydromica-schist (Rowe schist) succeeds the hornblende-schist as one goes west from the Emery mine, and is well exposed at the iron watering trough. Just beyond the first 74 GEOLOaY OF OLD HAMPSHIEE COUNTY, MASS. bridge in Becket the gray, garnetiferous, feldspathic mica-schist sets in and continues to the sawmill, where the Becket gneiss appears in a large quarry. Still farther south, in the west comer of Blandford, the Rowe schist narrows and occupies only the width of the North Meadow Pond, but is still a well- defined band of sericite-schist, while the Hoosac schist retains its width and appears in the high hills west of North Blandford. Two miles farther south, at Blair Pond, the rock from the Becket gneiss below to the liornblende-serpentine band above, and including thus both the Hoosac and Rowe schists, is a rather coarse mica-schist, not sericitic, but quite feldspathic, and in places abounding in staurolite crystals. The country begins in this latitude to abound in granite stocks and swarms of dikes, and the feldspathic character of the schists seems to depend largely on an impregnation from this granite, and the feldspathic constituent is arranged in flat blotches on the foliation faces, rather than in abundant small porphyritic crystals, as is the case farther north. The lithological dis- tinctness of the Hoosac schists and the Rowe schists disappears, from the loss of the hydrated mica in the upper bed and of the porphyritic albite in the lower, and I have not tried to separate the two beds in the Granville quadrangle. By the development of three anticlines in these schists, in the two outer of which the Becket gneiss comes to the surface, and by the troughing out of the hornblende-schists in the intervening synclines, this complex expands eastwardly to cover the whole of the Granville quadrangle, wrapping around the separate area of gneiss in East Granville and Granby. (See map, PI. XXXIV.) Granite continues abundant, and the rock becomes in the whole south- ern portion of the Granville qviadrangle a very coarse muscovite-biotite- schist, showing on foliation faces continuous films of large muscovite plates, or muscovite and biotite regularly intergrown, Avith, at times, feldspar or pegmatitic quartz-feldspar masses in the interstices, in place of the usual granular quartz. Toward the base of this complex on its western border, and in better development around the Granville gneiss, is a rock of very attractive appearance. It is a white, gneissoid rock of rather coarse grain. In the limpid, gramilar quartz mass the rather distant scales of silvery musco- vite, pale-red biotite, and pyrite are compressed into perfect parallelism, so that on foliation faces a very bright, silvery luster occurs. Considerable well- THE HOOSAO SOMLST. 75 striated plugioclase appears in limpid grains in tlie granular quart/, ground, and the rock is the gneissoid development of the albitic Hoosac schists, whose places it takes, though it did not seem constant enough to furnish a basis for the division of the rock in mapping. HORNBLENDIC BANDS IN THE ALBITIC MICA-SCHIST. Along the eastern portion of the area, on the east slope of Sodom Mountain, in Granville, bands of nodules of a pale-green actinolite-garnet rock occur, of a. type which, so far as I have observed, has always been derived from limestone. On the west slope of the same mountain is a narrow band of flat, fissile, garnetiferous hornblende-schist of gneissoid structure. The horn- blende is in black, shining grains, and the mass of the rock is black, but is closely spotted with round,' whitish spots 4-6"" in cross-section, in which the hornblende is in larger crystals but much less abundant. Farther south the same rock contains garnets of the same size and arrangement as the whiter spots, so that it seems the hornblende may have been kept out of these spaces by garnets which have since disappeared, to give place to a later development of larger crystals of hornblende. THE SHELBURNE FALLS ANTICLINE. Nearly everywhere around the Shelburne Falls anticline hornblende- schist seems to rest directly upon the gneiss, and in several places it can be seen to do so, but on the west side, near J. W. Whitney's, there occurs just below the hornblende-schist a white quartzite containing distant scales of biotite, magnetite octahedra, and rutile needles. This may be taken as a possible remnant of the hydromica-schist series. . (See section 3 of the Hawley section sheet, PI. XXIV.) Accessory minerals. — Excepting garnet, which occurs locally in the greatest abundance in large crystals (12-20""), generally with trape- zohedral form, the formation is very poor in accessory minerals. Staurolite occurs in quite good crystals, in both forms of twinning, on the road west of Blair's pond, in Blandford. Cyanite appears in gray crystals just where the formation crosses the State line on the south, and near the south line of Blandford on the West Granville road. 76 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. THE ROWE SCHIST=THE LOWER SERICITE- OR HYDROMICA-SCHIST. For two-thirds of the way across the State, starting from the north, the stratum between the albitic mica-schist below and the interrupted horn- blende-serpentine band (the Chester amphibolite) is a thick but extremely monotonous bed; and as for the purpose of working out the architecture of the region every valid distinction needs to be utilized, I have marked this bed separately from the Hoosac schist as far south across the State as practicable. It is the rock of the first 7,000 feet of the Hoosac Tunnel. In Hampden County, as already indicated, it becomes feldspathic, and can not be easily distinguished from the band below, as it is followed south from that region. FRANKLIN COUNTY. The schist enters Franklin County from the south, across the line between the east portal of the tunnel and the great serpentine deposit at E. King's, nearly a mile east, and, with high dips to the southeast, bends around the south end of the Grreen Mountain gneiss, and extends, with a width of a mile, northeast into Vermont. At the tunnel portal and east to the serpentine it is a very quartzose, pale-green, hydromica-schist, stretched so that it has often a ligniform structure. It contains a few garnets, trapezohedra, and many flattened lenses of quartz, which rarely contain dolomite. Followed northeast, where it crosses the Rowe-Monroe road it is very chloritic in its upper portion, and at the base is a dark, rusty mica-schist, resembling the Conway schist. A thin section was cut from the rock 4,000 feet from the east portal of the tunnel ; it is a light-gray, schistose rock of greasy feel, a true sericite- schist, from whose powder the magnet removes much magnetite. It shows under the microscope a mosaic of fine quartz grains, dusted with magnetite and wrapped around with muscovite and pale-green chlorite scales. HAMPSHIRE COUNTY. As the Rowe schist crosses Middlefield it has the same monotonous character. It is, however, more garnetiferous, and the garnets are very generally chlorite-bordered, and on foliation faces blotches of chlorite appear mixed with the hydrated mica. THE ROWE SCHIST. 77 It is bt'st studied in the ccmtiiuiiitiou of the section along- the Boston and Albany Railroad, beginning at the point reached on page 72, at the second bridge east of Middlefield station. There is from this point a nearly- continuous cutting for almost a mile through these light-green, quartzy sericite-schists, here -wholly barren and monotonous. Just beyond the fourth bridge many beds of a flat-fissile, epidotic amphibolite and of sericite-schist are exposed, as follows, eastward from the bridge : Section of Bowe sckist containing amphibolite. Feet. Sericite-schist 78 Amphibolite 33 Sericite-schist 23 Amphibolite 3 Sericite-schist 7 Amphibolite 3 Sericite-schist 30 Amphibolite 150 Sericite-schist 59 The same rock extends, poorly exposed, with a single small band of am.phibolite to the Chester amphibolite at the Chester line; whole thick- ness, 820 feet. This is the first case where any amphibolite occurs below the Chester amphibolite, and it is here that the remarkable overfolding or overcrushing of the vertical beds of thin-fissile amphibolite occur, which has been figured by President Hitchcock, who refers it to crushing by ice.^ HAMPDEN COUNTY. The rock is best studied along the Chester-Becket road, westwai'd from the Chester emery mine, where miich rock cutting has been done to protect the highway from the mountain brook along which it runs. It is a soft, greasy sericite-schist, often becoming very quartzose and then of firmer texture. It enters the Granville quadrangle (and at the same time Hampden County) at its northwest corner, and continues with a width of half a mile to the pond at North Blandford. Two miles farther south, as noted in the description of the Hoosac schists, the whole area across from the Becket gneiss to the Chester amphibolite is biotitic and feldspathic and not marked ' Elementary Geology, p. 139. 78 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. by liydrated mica. As the iuterveuing area is almost wholly covered, the line is drawn upon the map with much doubt. THICKNESS. The section gone over from the Middlefield station to the top of the series at the Chester line is, measured directly across the strike, 6,970 feet, which, with an average dip of nearly 80°, would give a thickness for the series of 6,897 feet, provided there be no repetitions from close folding or secondary structure simulating bedding— a thing one would be very unwill- ing to admit. This includes, it will be noticed, the two beds between the Becket gneiss and the big hornblende bed; that is, the Hoosac and Rowe schists. THE CHESTER AMPHIBOLITE AISTD SERPEKTIKES. The albitic mica-schist and the lower sericite-schist already described, the hornblendic band which is separately discussed in this chapter, and the upper sericite-schist and the chloritic schist next to be treated (that is, the Hoosac schists to the Hawley schists, inclusive) are certainly one conform- able series of beds, and form a group well demarcated from all above and below. The correlation of the strata has been attended with great difficulty, owing in large part to the fact that southward along the line of strike the hydration of the mica becomes less, and at last becomes inappreciable, while the chlorite also disappears and the feldspar increases in quantity, so that what in Hampshire County is well-characterized hydi'omica and chloritic schist becomes in Hampden feldspathic mica-schist, or even quite well- marked gneiss.^ I have therefore found the broad band of amphibolite, associated abundantly with serpentine and talc, although interrupted, to form an exceedingly useful horizon clear across the State. Carrying as it does the unique emery vein at Chester, it is also of great interest in itself. GENERAL DESCRIPTION. The amphibolite is a dark-green rock, either flat thin-fissile or ligni- form, and rarely massive. It is almost always epidotic. Along its eastern (that is, its former upper) surface at various points occur great masses of ser- pentine or serpentine and steatite (the latter above the serpentine), or rarely •This change is caused by the great quantity of granite in and south of Blandford, from which the schists are greatly soaked witli feldspar. THE CHESTER AMPHIBOLITE AND SERPENTINES. 79 of steatite alone. These lenticular masses have eaten their way into the amphibolite for various distances, and it is suggestive that they always appear along the upper surface of the amphibolite, or on the upper surface of separate bands where, as is often the case, the latter rock does not occupy the whole space assigned to it on the map, but has intercalated subordinate layers of sericite-schist. The Chester emery bed occupies the same position along the eastern border of the amphibolite. The character of the serpentine bands which accompany the amphibolite changes in Blandford. At Osborn's quarry is a bed of sahlite-serpentine, one of olivine-serpentine, and the first of a series of enstatite-serpentines, which, as the bed is followed, becomes of greater relative importance and gradually almost replaces the amphibolite and is itself at last almost replaced by coarse dolomitic limestone. The band is in its whole extent conformable with the sericite-schists and runs across the country with dip varying very little from 90° and in strike conforming to the winding of the schists. FRANKLIN COUNTY. THE ROWE SERPENTINE. In the northern portion of the State the band enters the town of Rowe from Vermont, exactly at its northeast corner, and extends southwest across the town as a heavy bed, 1 to 20 rods wide, of a black, thick-bedded, epidotic amphibolite. It seems to be continued far north to the important actinolite bed at Newfane. It is well exposed at the bottom of the hill south of the house of J. Streeter, jr., and runs about a mile west of Rowe Center, where, 49 rods northeast of A. C. Bliss's, it carries on its east border a heavy bed of steatite (bed No. 1^), which is very hard and chloritic. From this point it takes the same curve as the Deerfield River to the west, and forms the crest of a ridge until, at J. C. Cressy's, it crosses the road running down to Hoosac Tunnel. It is here 30 rods wide, is very fine-grained, black schist in its western portion, and on the east is an epidotic quartz-hornblende-schist, and there are one or two other bands a few rods east in the hydromica- schist. It continues down the hill, and where it cuts across a sharp bend in the road it changes suddenly almost entirely into serpentine and steatite (bed No. 2), only 7 feet of the amphibolite remaining on the western border. 1 For convenience of reference I have numbered the beds of steatite and serpentine described in this section. Chester amphibolite . < 80 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. Here the following section is exposed, from below upward, i. e., from west to east: Section at Bowe serpentine quarry. Feet. Inch. Rowe schist 1. Sericite-schist. r 2. Hornblende-schist 7 3. Talcose schist 1 4. Serpentine, showing structure of amijhibolite 450 5. Talcose schist 46 v6. Chlorite-schist 7 Savoy schist 7. Sericite-schist. Total thickness, from 2 to G inclusive 510 1 No. 4 is a serpentine altered from amphibolite; 3 and 5 are masses of schistose talc, representing a further stage of the alteration. It abounds in dolomite. It has been extensively quarried, but the large buildings erected for working it are abandoned. THE BAST PORTAL FAULT. The series maintains a width of about 30 rods down to a point 100 feet above the Deerfield River, and the continuation of the band can not be found across the river on the south; but it is shifted a mile to the west, to Mount Serpentine, by an important fault running in the bed of the "Westfield River at this point. It is thus carried beyond the limit of Franklin County, but as it is still within the Hawley quadrangle its further course is traced to the point where it enters Worthington, in Hampshire County. The steep mountain rising west of Rice's tavern, at the east portal of Hoosac Tunnel, and easily distinguished from the others that surround the valley by its bare precipitous walls, is Mount Serpentine, and is a great mass of the rock which has given it its name. Its north face seems to be the face of the fault here described; its east wall appears to be foi'medby the peeling off of the vertical schist from the massive serpentine. The mountains as seen from the valley are projections of the plateau, notched by the successive brook gorges; and taking the road running up southwest from Rice's tavern, one finds at the first brook-crossing a magnificent bowlder of serpentine, and can climb by this brook to the serpentine overhead. By following the road to the first house, and then going 50 rods west, one comes on the serpentine, after passing a band of TUB CHESTER AMPHIBOLITE AND SERPENTINES. 81 amphibolite 6 rods wide and 10 rods of swampy ground, probably in whole or part underlain by steatite (bed No. 3). Then comes tlie dark- green serpentine, which is here 35 rods wide, and it is followed immedi- ately by black, thin-fissile amphibolite. This is so clearly a repetition of the series at the soapstone quarry given above that a fault, substantially as shown on the map, is quite certainly present. Where the beds cross the road to the south they are covered, and continue so as far south as I could penetrate in this wilderness. At latitude 42° 35§', longitude 72° 55f' , the amphibolite appears again, and the line of boundary is drawn approximately from the strike between these points, as the amphibolite could not be found and the sericite-schists above and below the amphibolite are hardly distinguishable. Indeed, at the cross-roads a mile north of the last locality the sericite-schist is almost continuously exposed, but careful search failed to disclose any amphibolite. From the last locality the amphibolite makes a bend to the east and cuts across the sharp curve in the road next south. On entering the Chesterfield quadrangle, the amphibolite, where it crosses the road near Swift River, in Windsor, is changed to steatite (bed No. 4), and at Jordans- ville the schist is well exposed in the brook southwest of the village. HAMPSHIRE COUNTY. THE MIDDLBFIEUD SERPENTINE. Reentering the county, the amphibolite appears just west of the village of West Worthington, and can be traced thence southward. At H. Smith's, in the northwest of Middlefield, it has on the east a fine deposit of serpen- tine (bed No. 5), bordered on the east by talc. Along the east side of a band of the common amphibolite rests a mass of dark-green serpentine, and next east a great mass of steatite, often carrying large nodules of the finest dolomite surrounded by delicate-green talc, and on the east sericite-schist folds around the great boss of steatite, as if it had been present — or, rather, as if the rock of which it has been formed had been present — as a foreign and resistant body during the compression of the schists. The steatite is here QQ feet wide, and it furnishes the best material in its upper half It is opened in a quarry 41 feet wide and 82 feet long, and is separated from the amphibolite opposite the quany by only 16 feet of covered space; so MON XXIX (I ■ 82 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. here the seipeutine can have at most only this thickness, though it swells a few yards north to treble this thickness at the expense of the steatite, which runs on north for a distance of 492 feet, with a thickness of 10 feet, and enlarges again into a pocket of harder soapstone. The deposit extends southward across the road, and is then opened again in a large quarry on the land of Mr. Howard. The New York Metropolitan Company has quarried 200 tons, paying a royalty of 50 cents a ton, and the material was ground at a mill in the valley to the east. About as much more had been gotten out earlier, but no work was in progress at the time of my visit (1877). This is a type of all the serpentine and talc deposits — a lenticular mass of serpentine replacing the amphibolite in its upper layers, and, as it were, eating into its mass and suggesting strongly that it has been formed at the expense of the schist and itself changed later for a varying distance downward into talc. In a recent interview published in the Springfield Republican,^ the discovery by Dr. H. S. Lucas of another bed of emery, or the continuation to the north of the Chester bed, is announced. It is at a point a mile east of Middlefield and a mile and a half nearly due north of Chester, on land of Frank Smith, and the land has been purchased by Dr. Lucas. It is asso- ciated with hornblende-schist, as is the Chester bed, and is quite certainly the continuation of this bed northward. The specimens from the new locality shown me by Dr. Lucas contain grains of blue corundum. Southward on the strike the outcrops are not abundant, but they are sufficient to show that the amphibolite is probably interrupted for a considerable distance, though it may be continued as a narrow band, some- what shifted by faults. Two miles southeast of the cheese factory it appears again in great force, and immediately to the east of it the serpentine (bed No. 6) appears in still greater force. The two expand rapidly to a width of 200 rods and run as a prominent range of hills over the town line into Chester, dropping down suddenly to the brink of the Westfield River. On the east and west the vertical sericite-schists, 200 rods apart, inclose this great double bed of amphibolite and serpentine, and are continuous across the river to the south, and the western half of the bed, the amphibolite, also continues across, its eastern half, the serpentine, being replaced by amphib- olite in the bed of the river. The boundary between the two, which may 1 "Another vein of corundum:" Springfield Republican, December 12, 1895. THE CHESTER AMPHIBOLITE AND SERPENTINES. 83 be found just opposite a shoddy mill near the river, runs in a great curve N. 30° W., so that the serpentine encroaches still more on the amphibolite and at last occupies nearly its whole width. An inspection of the map (PI. XXXIV) shows that the course of the river, where it separates ser- jjentine and amphibolite, is in southward continuation of this curve, and that the serpentine is lodged as a great lens, a mile and a half long and nearly a half mile wide, in the amphibolite. The boundary line between the two runs up the hillside in a narrow gorge, its bottom everywhere encumbered with bowlders, and the amphibolite and serpentine could not be found nearer each other than 10 feet. At that distance there was no trace of transition from one into the other. Search was made for the bound- ary between the serpentine and the sericite-schist on the east for a mile north through the dense woods, but they could not anywhere be found in actual contact. The contact line was, however, a straight one, following the line of strike of the schist, while the schist, ordinarily a very flat-fissile rock, was for all this distance, and, indeed, for the full length of the ser- pentine lens, and in a thickness of above 350 feet, thrown into the most extreme contortions and twistings, the like of which I have hardly seen among any of the rocks of the region. This I take to be another indication of the formation of the serpentine before the final folding of the region. It is likewise interesting that along this line the serpentine was in many places, and it seemed continuously, separated from the sericite-schists above by a thin layer of amphibolite, and the serpentine, when traced to within a single foot of this, was complete serpentine. The mass of the latter would seem to be, then, strictly speaking, inclosed in the amphibolite. As already noted by President Hitchcock, this serpentine mass shows abundant signs of stratification, and I may add that this not only agrees with the dip and strike of the adjoining amphibolite, but shows closer agree- ment still with the latter, extending to the exact thickness of the laminae, the angles and distance of the jointing, etc.; and further, that this structure is one brought out in the serpentine again only by the action of atmospheric agents, below the surface the serpentine appearing wholly compact. The serpentine is the common rather light oil-green variety, and, especially where a fine splintery fracture is developed, it has a dry grayish-green color. It weathers to a deep red brown, and the great ragged hill, bare of vegetation and covered with an almost unbroken layer of immense bowlders 84 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. upheaved by frost, is a very striking object, suggesting immediately the idea of igneous action. I visited the place once when, after heavy rains, the fine brook which runs down from the high ground in a great gorge lined with bowlders of the weathered serpentine, and the succession of beautiful waterfalls, derived a peculiar charm from their setting in the warm browns and greens of the rugged serpentine masses. The serpentine locally is rich in chromite, and a considerable excava- tion made in mining for it exists in the woods near the southeast extremity of the bed. Small veins of precious serpentine, much picrolite, and crusts of hydromagnesite of some thickness occur. It also furnished to Dr. Emmons the well-known pseudomorphs of serpentine after chrysolite, formerly called serpentine after quartz, or hampshirite, the exact locality of which I have not been able to recover,^ and was doubtless the origin of the large masses of yellow chalcedony found in Chester by the same geologist. These pseudo- morphs are large, distinct crystals more than an inch long. They are six- sided prisms terminated by six faces which have some resemblance to the ending of a quartz crystal, in which two opposite faces predominate, but giving the angles of chrysolite. They are covered by a straw-yellow, secondary serpentine of a compact but slightly radiate-fibrous structure (picrosmine) ; it is homogeneous and almost apolar under the microscope. An analysis was made for me by Miss Helen P. Cook, of the chemical department in Smith College. Analysis of pseudomorphs of serpentine. SiOo MgO Fe^Ci [AI2O3 trace] Ignition, 6^ hours, 55° to 150° C. Ignition, open flame and blast . . Per cent. 40.27 40.00 4.74 0.92 13.38 99.31 Secondary shrinkage joints in serpentine. — The detached blocks of the serpentine have often suffered secondary decomposition, so common with 1 These are fully described and figured, and the proof of their derivation from chrysolite is given, in A mineralogical lexicon: Bull. U. S. Gaol. Survey No. 126, 1895, pp. 92, 146, under "Hampshirite." THE CHESTER AMPHIBOLITE AND SERPENTINES, 85 serpentine, to a depth of 10-15""°, accompanied with loss of color, hardness, and volume; and as a result of this last the surface is often covered with a tine system of regular slirinkag-e joints, one set of straight fissures about 20"'"' apart being cut by another at an oblique angle, the latter about 50""° apart. In places the blocks have all separated from the underlying unchanged mass and lie loosely upon it. HAMPDEN COUNTY. THE CHESTER AMPHIBOLITE AND SERPENTINE. Following the heavy hornblende band across Chester, where it forms in the north the high, sharp ridge called Gobble Mountain, and in the south the still higher Round Mountain, one finds in the higher part of the first hill a considerable deposit of serpentine (bed No. 7), situated, like the others, at the upper surface of the hornblende, but offering nothing peculiar. Farther south, in the bottom' of the brook gorge between these hills, at the old emery mine, is another deposit (bed No. 8), which is at a level of several hundred feet below the other. The excavations at the mine exposed the following section from east to west across the vertical strata: Section at the old emery mine near Chester. Savoy acMst Sericite-schist. Feet. 'Steatite inclosing a few small serpen- tine nodules 4-16 Emery and magnetite bed 6J-10 Fringe rock 1 inch to 10 . Hornblende-schist. Chester amphibolite . . < The small nodules of serpentine, often as large as one's hand, are isolated in the mass of the talc and are permeated by veins of the same material, and doubtless represent the original material from which the talc was formed. The serpentine is the usual variety, dark-green when wet, but, partly from its fine splintery fracture, gray-green when dry. Another variety is rich olive-green, and carries much malachite. The talc is pale-green, foliated for the most part, and often crowded with dolomite crystals. THE BLANDFORD SERPENTINES AND PYKOXBNITE. The heavy hornblende bed continues with undiminished width across Chester, and is much covered by drift as it crosses into Blandford, where it is 86 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. again well exposed. Northwest of S. A. Bartholomew's house, at his soap- stone quaiTy, some layers of mica-schist are intercalated in the amphibolite. The quany, from which much soapstone (bed No. 9 ) has been taken for grind- ing, is inclosed in walls of chloritic mica-schist, and lies in the prolongation of a bed of the ordinary amphibolite, which is exposed just north of the opening. There is exposed in the north end of the excavation a layer, 1 foot thick, of light-green talc with scattered needles of actinolite, and east of this, one (the same thickness) of a green, soft, scaly chlorite, with here and there larger jjlates of clinochlore with very divergent optical axes, and magnetite octahedra. Farther south, in the bottom of the quarry, it can be seen that the steatite bed widens rapidly southward to 10 feet, and a mass of light-green fibrous actinolite appears, from which the whole steatite mass seems to have been derived, as it still retains the radiated and matted acicular structure of the actinolite. A few rods south of the steatite quarry, and just west of the village of North Blandford, is the great mass of serpentine (bed No. 10) marked upon Walling's map of the county as "The Crater." The name is said to have originated with Dr. Hitchcock, when he thought the rounded, isolated mass, with a large cavity in its center, proof of the volcanic origin of ser- pentine. It seems to me not improbable that the cavity in question may be an artificial excavation, and it is certain that in early times considerable digging was done there for chromite. It is an oval mass, 328 feet long and 200 feet wide. On the west is a stratum of amphibolite 20 feet thick, which strikes north-south along the side and wraps round the north end until it strikes N. 28° E. This seems to indicate that the change to serpentine took place before the final compression of the rock, or that the original rock was different and less compressible than the amphibolite. Below is the sericite- schist. The serpentine from this locality is easily distinguished from any other by its compactness, its black-gray color, the abundance of dissemi- nated magnetite, and the nickel-green crust from weathering. Along the strike of the rocks southeast by south the ground is much covered and no further outcrops have been found, though the region has been thoroughly searched in prospecting for emery, until the Osborn soap- stone quarry is reached; but several bowlders reported to me by Mr. Bartholomew, viz, serpentine west of the north end of Blair's pond, and soapstone northwest of Pebble's brook, and also west of the Blair's pond road, indicate other deposits in the intervening space. THE CHESTER AMPHIBOLITE AND SERPENTINES. 87 The Osborn soapstoue quarry lies west of the house of Mr. W. H. Griswold. Passing west over a few rods of sericite-schist, with two granite dikes and a thick stratum of serj^entine, and more schist, all with strike N. 40° W., dip 46° E., one comes upon a bed (No. 11) of black serpentine 50 feet thick, which can be followed south a considerable distance along the line of strike and ends abruptly against chlorite-schist along a line at right angles to the strike. It is also underlain by the chlorite-schist, and following the line of strike of this south a few yards, across covered ground, one comes upon the large quarries of a soapstone which has completely the structure of the coarse radiated actinolite from which it has been derived, and fresh and partly altered masses of the latter are also abundant, together with large ^„C-^^ N ■*0°W fS'E N 35 °W 80°£ N35°tV'*5''E riG. S.-Section at Osborn soapstone quarry, Blandford. S S = Salilit6-serpentine; S = Steatite and enstatite-serpen- tme; 0S = 01ivine-serpentine; A = Amphibolite ; P = Pegmatite; conntiy rook = sericite-schist. masses of coarsely foliated chlorite — a clinochlore with very wide optical angle. The steatite bed is separated by a thin stratum (1 inch) of black mica and ©ne of equal thickness of heavy black hornblende-magnetite rock from a dike of granite. A small brook runs from this point west through the woods, down over sericite-schist, to the bottom of the valley, where it cuts a great bed of pecuhar, streaky, black to gray serpentine (SS; bed No. 12), derived from a very coarse-grained pyroxenite or sahlite rock, which still shows cleavage faces 20-30 """^ square. This bed seems to haA^e been overlooked before, and it is doubtless the source of many of the bowlders found in the south- eastern part of the town. The old quarry has been opened during the past summer (1895) quite extensively with improved machinery. The whole width of the steatite 88 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. bed lias been exposed and a deep trench blasted through the schists to the west for drainage. The following section is now exposed (fig. 5, p. 87): Commencing at the bottom of the hill to the west, one climbs up 15 rods over coarse chloritic sericite-schists abounding in large quartz lenses and quartz-filled garnets sometimes an inch across. The schists dip 80° E. At 50 feet above the meadow the lower bed of coarse, rudely bedded, black serpentinous rock appears (SS). It shows broad, black, lustrous cleavage surfaces of much-altered sahlite, and no specimens could be found where this mineral was still unchanged, such as were procured in the bed of the brook at the former visit. These cleavage surfaces make up the whole surface, or are somewhat separated and the interstices filled with white calcite and magnetite and shot through with tremohte. An analysis of the least-altered forms of this rock, which still retains enough of the unaltered sahlite to enable one to make out its optical con- stants, gives the complete formula of serpentine, and is interesting as show- ing, as do all the other rocks of the series, a constant content of nickel and chromium. The analysis was made by Dr. W. F. Hillebrand. Analysis of serpentine from Osborn's soapstone quarry, Blandford, Massachusetts. SiO.2 TiOi Al.Oa CroOs Fe^Os FeO NiO MnO CaO SrO BaO MgO K3O NajO Li,0 HjO below 110° H2O above 110° PsOs CO2 Per cent. 40.77 None. 1.16 . .28 3.56 1.47 .17 ^.09 None. None. None. 39.37 .10 .14 Trace. .49 12.48 Trace. None. 100. 08 THE CHESTER AMPHIBOLITE AND SERPENTINES. 89 Next east is a bed, 150 feet thick, of finer-grained chloritic sericite- schist, without garnets, and containing a subordinate bed of jet-bUxck, flat- bedded amjjhibohte (A), which is made up almost wholly of shining-black needles, the larger porphyritic in a network of the smaller. Eight feet of coarse pegmatite are followed by the same thickness of schist, and this by 12 feet of pegmatite, which is separated by a thin layer of reddish schist from the soapstone (S), which is 60 feet thick. On the west border is a thick bed of coarse tourmaline in a matted mass of large clinochlore, with a 6-inch bed of coarse biotite adjacent. The eastern selvage is of coarse transverse chlorite in broad plates, which is often crushed to schist. The outer sheets of the main soapstone bed are of coarse, matted trem- olite, often radiated and plumose, and more or less changed to talc. In the eastern portion of the soapstone bed is a 10-inch layer of fine actinolite, and at the border these actinolite needles change directly into tremolite. The central third of the steatite bed consists of black enstatite- serpentine, more or less tremolitic and partly changed to steatite, but still quite hard. This is the first bed of this enstatite rock met with, and it becomes increasingly important as the series is traced southward. Layers of an apple-green serpentine fill fissures in this mass. The superintendent informed me that a 2-foot layer of a black amphibolite, exactly like that descnbed above, ran through the soapstone parallel with the strike in a part of the quarry which was under water. The vein of steatite makes a sharp bend of 90° to the east, and bends directly back 90° to the north, and along the east side, where the latter bend is effected, the eastern schist wraps irregularly over the steatite and around a white albite lens, which is enclosed in a thin layer of black, coarse hornblende rock. This bend explains the cutting off of the bed of black serpentine mentioned above, and shows that beyond this sudden fault-like bend the band is less altered to steatite. Indeed, the steatitic alterations may be due to local disturb- ance, as the development of serpentine farther north seems to be caused by faults. The next eastern bed is a reddish, quartzose, fine-grained biotite-schist 150 feet thick. This is followed up the hillside by 150 feet of a massive, dark- green serpentine (OS), which at the base shows much half-changed olivine in granular masses, separated by a later tremolitic growth, followed by talc, all of which is beautifully shown under the microscope. This is the only 90 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. trace of olivine that has been found west of the river, except the Middlefield pseudomorphs mentioned above. A bed of pegmatite a rod wide separates the serpentine from schist, wliich continues up the hill 412 feet to the Grriswold house. THE GRANVILLE AND RUSSELL ENSTATITE- SERPENTINES. From the section north of Borden Brook, in the south of Blandford, where the hornblendic complex is made up of several amphibolite beds alternating with beds of sericite-schist, the series continues due south into Grranville, and is for a short distance interrupted by pegmatite, but attains in Liberty Hill a thickness of 1,237 feet of clear, black amphibolite without interlaminated mica beds. It curves east and then west and retains this great width for a mile, and continues southwest as two bands of amphibo- lite, each about 15 feet wide. These soon run out southward, and no trace of them could be found in the well-exposed bluffs east of West Hartland. Where it bends most easterly it contains the heavy bed of steatite (bed No. 13) a mile southwest of West Granville, at the bottom of the bluff east of E. Williams's house. Here some work has been done upon a deposit of steatite, which has been derived from a bed of fine, radiated tremolite; it still retains the structure, and part of it the hardness, of hornblende, and therefore the bed is not a promising one to work. Many bowlders of the black enstatite-serpentine occur near Mr. Williams's house, which must come from another bed of the rock near at hand, as the two rocks seem to be connected genetically, since the tremolite is exactly like that found with the serpentine of the next locality. Just after crossing- the State line and Hubbard Brook the amphibolite band carries a bed of black enstatite- serpentine (bed No. 14), of which about 5 or 6 feet is exposed. On the southeast flank of Liberty Hill, in West Granville, a branch of the amphibolite separates from the main bed, as mentioned above, and, bending round sharply, runs north with much diminished thickness, not exceeding 6 feet, to a point west of East Granville, where it bends noii;h again and carries the remarkable bed of enstatite-serpentine (bed No. 16) which occurs in a densely wooded swamp 100 rods east of the house of J. Downey. A ridge 20 feet wide and rising 24 feet is exposed for a considerable distance, and, as usual, the serpentine is associated with amphibolite. It is a black serpentine, made up of crystals an inch square on the end and more than 2 THE CHESTER AMrHlJ50LlTE AND SEKPENTmES. 91 inches long, pseudomorph after enstatite, and it carries considerable dolomite disseminated, which does not effervesce with HCl. Traced northward a few rods it becomes a compact, gray, thin-bedded tremolite-schist, which lies in contact with an equally thin-bedded, white crystalline limestone which eifei-vesces readily. Southward it is found in many bowlders around the cemetery, and here the limestone contains very fine specimens of a rich- green actinolite, and it crops out farther south on Trumble Brook. The band can be traced north from Downey's, by the abundant bowlders of the black serpentine, to the pasture back of H. Cooley's. The overljnng rock in the Cooley pasture is a coarse muscovite-biotite-schist, carrying much cyanite in flat, colorless blades 1 to 1 J inches long, but 20 feet of covered space, possibly occupied by amphibolite, separates it from the serpentine. The serpentine bed (bed No. 16) is about 50 feet thick, and is exposed 175 feet in length. Over the weathered surfaces of the ledge the great enstatite crystals project in a close network. These crystals are great plates one- half to 1 inch in thickness, 3 to 4 inches wide, and in average 6 inches long, while some measure 14 inches in length. They are now changed to a dull-black serpentine, but still retain the lustrous enstatite cleavage. In the naiTow meshes between these large plates is a rather coarse-granular, limpid dolomite, dusted with small magnetite octahedra and broad plates of colorless to oil-green talc. The band can be traced northwest from this point by many large bowlders, and another locality occurs where the rock appears in place southwest of the point where "Wildcat road" bends south. Bowlders of the same rock occur northwest, in the bed of the Westfield Little River, at the great bend a mile below "Pothole Rock." From this point no traces of the bed have been found along the line of boundary drawn across Russell to the Atwater ledge, except where this line crosses a little-used road, not on the map, which runs west from the sharp bend in the road a mile above Atwater's to meet the dotted road. Careful search has been made in the intervening, heavily wooded country, and the presence of the rock as a continuous band is indicated by the abundant large bowlders strewn over the country for miles southeast. The next outcrop is the one mentioned above as Atwater's (bed No. 17), from the extensive exposure in the high hill 1 mile N. 30° W. of the house of F. B. Atwater, in the south corner of Russell and overlooking the Westfield plain. It was quarried quite extensively by Mr. Atwater's father as "black marble." The bed is 92 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. exposed with a width of about 53 feet, when it is cut off by granite. It shows distinct foHation, and is bordered on the west by a biotite-hornblende- schist, which becomes in places a distinct gneiss. It is a black serpentine abounding in the green foliated bastite to which the name marmolite has been given. THK WESTFIELD SERPENTINE AND MARBLE. The next outcrop of the serpentine is south of this point, across the Little River, in the western edge of Westfield (bed No. 18). It has been opened by the owners, the Westfield Marble and Sandstone Company, and reported upon during the last year (1895) by Prof W. 0. Crosby, who considers the deposit valuable for verd-antique marble. He reports the following section from east to west, with explanations: Section in the loestern edge of WestfieM. Feet. 1. Yein of coarse granite (pegmatite) 10 2. Soapstone and serpentine, with partings of micascbist and veins of pegmatite 15 to 20 3. Massive serpentinic marble (verd antique), with large crystals 15 to 20 4. White marble, with thin layers or partings of serpentinic marble 15 5. Banded serpentinic marble, consisting of very thin alter- nating layers of white marble and serpentine 15 to 20 6. Shaly serpentine and marble and banded serpentinic marble 20 7. Massive black and green serpentine 50 8. Soapstone and serpentine, concealed- 10 9. Fibrolitic ^ mica- schist and granite trace. The most interesting and valuable bed in this series is the verdantique marble (ISTo. 3). This is a very solid bed, and of fairly uniform character, considering the coarse structure of the marble. The serpentine, which has evidently resulted from the alteration of actinolite, is in the form of slender crystals from 1 to 3 inches in length, lying at all angles in a matrix of white crystalline limestone. Near the east side of the bed the structure is finer and somewhat banded, as in bed No. 5. This verd-antique marble is a striking and, so far as I know, unique stone, of ornamental character; and I can see no reason why it should not give satisfaction in use. Although it would, I am confident, prove serviceable in exterior work, it is to be especially recommended for interior work. It is susceptible of a good and lasting polish, and this, together with its unique, breccia-like structure, should insure a demand for the stone when it is properly brought before the public. It is probable ' This is oyanite. THE CUESTER AMPHIBOLITE AND SERPENTINES. 93 that at a somewhat greater depth bed No. 4, which coukl be very easily worked with No 3 wou d yield so,„e good white marble. A part of the banded marble iu beds 5 and 6 IS of a decidedly ornamental character and well adapted for some kinds of decorative work. there are indications that the marble continues beyond it, this has not been prove^l So far as kuovvn, the marble is entirely wanting north of the river. South of the foi 200 to 300 feet, when they are again cut off by a mass of granite. The quarry, to which I was guided by Professor Crosby, is situated in the extreme western part of Westfield, and is reached by leaving the electric cars at the crossroads east of the old Atwater place and going three-quarters ot a mile south, passing two houses south of the Little Eiver bridge and gomg west by a field road, which runs northwest about a half mile 'to a pomt 380 feet above the sea and overlooking the valley of Little River The quarry throws much light on the problem of the origin of the enstatite beds. It contains three distinct beds of first importance- The first, the "black marble," like that of the old Atwater quarry, is a black enstatite rock of coarse g.-ain and wholly massive structure and shghtly bronzy luster (the enstatite cleavage showing in faces one-half inch m width by 2 to 4 inches in length), now in various stages of serpentinous change, and mottled with fohated masses of bastite (marmohte), derived from the enstatite, which are of high luster and rich apple-green color The second bed is a black spotted marble-a white or grayish, rather coarsely crystalline, magnesian limestone-containing much shining tremo- lite, effervescing moderately with strong hydi-ochloric acid, and spotted with elongate crystals of the same black altered enstatite, one-quarter to one-half mch wide and 2 to 6 inches long. These make a very regular reticulate or open network over the prevailing white surface of the marble, formino- a remarkable rock. At times black squares of the mineral, with lighter and less changed centers, are interspersed with the narrower rods, and the latter radiate from the centers and sides of these squares with some regularity and connect them into a stellate pattern (see fig. 8, p. 152). Again the squares may wholly replace the rods. These two beds are of massive structure and furnish large blocks which take a fine polish and promise to be of economic importance. The third bed, which connects the other two, is a thin, flat-foliated pale-green to white marble, with films and flat small lenses of pale-green 94 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. to rich dark oil-green precious serpentine. The surfaces show at times narrow bands and squares of serpentine with centers of straw-yellow and borders of oil-green color, which form an attractive verd antique, plainly the last identifiable stage of the flattened-out enstatites that the rock formerly contained. It is further clear that this central portion never contained so much enstatite as the border beds. The country rock to the west is a coarse muscovite-biotite-schist, with few garnets and a great abundance of coarse-bladed gray cyanite, which stands out in reticulated surfaces on the weathered slabs and will furnish interesting cabinet specimens. They can be obtained in abundance by following the path along the serpentine bed south to the first wood road and then going up west along the wood road to a prominent ledge on the south of the road. Here also the foliation faces of the coarse schist contain rounded and flattened disks, 1 to IJ inches long, which suggest pebbles. Earely one can be seen by its cleavage to be in part feldspar, but most are a quartz- biotite mixture. Also, to the west of the south opening made by the com- pany, which is reached by going up a wood road from a miner's shanty, the schist shows rounded and somewhat oval, white surface forms, which strongly suggest the trace of pebbles, but they are not distinguishable in the midst of the coarse schist when it is freshly broken. The rocks stand vertical, but I suppose these western schists to be older than the serpentine. The cyanite follows the serpentine and amphibolite for many miles across G-ranville, and in Barkhamsted, Connecticut, furnishes the finest cabinet specimens. The cyanite-schists are succeeded from west to east by the following beds: Section at the main mine. Feet. 1. Black enstatite-serpentine 60 2. Green laminated crystalline limestone 48 3. White actinolitic marble 3 4. Black mottled marble 30 5. Tremolitic soapstone 8 6. Coarse muscovite granite 8 The eastern country rock is not here exposed, as the high terrace gravels cover the area to the east. It is a schist like that on the west, but without cyanite. A band of rich-green actinolite three-fourths of an inch wide runs across both of the principal bands of the quarry. THE CHESTER AMPHIBOLITE AND SERPENTINES. 95 The soapstone has been worked by the liidiam. Ilalf-made pots are still to be seen on the surface, and an Indian arrow was found, on blastino- 12 feet down in a narrow crevice in the limestone. To the north of the quarry the bed of granite cuts off the serpentine band, but it reappears after an interruption of a rod. Followed a few rods south, the western contact of the bed is exposed by digging. It is a thick bed of rich-green, coarse-radiated actinolite mixed with biotite, and the wall on the left is granite with some plagioclastic fringe rock containing biotite and tourmaline. Nearly the whole thickness of the bed here is black serpentine, but Professor Crosby pointed out to me a continuous valley, generally quite swampy, which may be occupied by the limestone and caused by its solution. At the southern outcrop mentioned above (and at several other places) a black, flat amphibolite accompanies the serpentine on the east, but does not seem to be in great force. Here the radiated tremolitic character of the soapstone is specially manifest. It forms a heavy bed on the east, followed westerly by a thin-foliated verd antique, made of bands of blackish-green serpentine and white marble, with about 20 feet of the black serpentine to the west before the schist is reached. Farther south the rock crops out in the bed of Westfield Little River (bed No. 19), and in Westfield (bed No. 20) north of the Granville road, near the west border of the New Red sandsone;^ south of this road it crops out in the hill back of S. Drake's house (bed No. 21), where it is very coarse- grained, exactly like the East Granville locaUty, and from the weathering out of the calcite, which fills the interstices, it is very rough-surfaced; and, finally, it is seen in the bottom (bed No. 22) of Munn's brook, near the line between Granville and Southwick. It is in place where the brook emerges from its gorge in the hills. The prevailing rock is a black enstatite-serpen- tine; amphibolite is subordinate. The line of strike then carries the bed beneath the sands of the Westfield plain and it is not seen farther south. FAtJLTS AND SERPENTINIZATION. The great Hoosac fault displaces the rocks a mile on either side of the Deerfield River, below its bend at the mouth of the Hoosac Tunnel. It seems possible that this fault plane had a determining influence in the great devel- » Hitchcock, Geology of Mass., 1841, p. 159. 96 GEOLOGY OF OLD HAMPSHIKE COUNTY, MASS. opment of serpentine from tlie hornblende-schist bed, since this extensive development of serpentine extends north and south from this fissure, and the fault perhaps aided the work by bringing to the bed for great depths an abundance of water, and may have further intervened by localizing the earthquake forces, which may have shattered the rocks for a distance on either side of the fissure, thus aiding the chemical activity of the water. The sharp bend of the stream and its long course parallel to the direction of the fault show that the fault early controlled the direction of the river, and it probably did this because the softened rock was more easily eroded. The next large area of serpentine — and these two areas are vastly larger than the others — is in Middlefield, in the only other large transverse valley in the State. Here, also, I have mapped a fault in the valley bottom, and it seems probable that here also the fault may have had something to do with the hydration of the hornblende to serpentine, as well as with the position of the transverse valley. I have noted also a sharp bend, which is almost a fault, at the Osborn quarry. The other serpentine and stea- tite deposits are comparatively unimportant in size, or show trace of olivine and enstatite. This relation did not attract my attention until the field work was ended, or 'other similar coincidences might have been detected. PETROGRAPHICAL DESCRIPTION. THE AMPHIBOLITES. 1. Epidotic amphibolite. — Blandford; North Blandford road at watering trough. Typical jet-black, fissile schist, the shining hornblende needles just visible to the eye. Drusy surfaces of epidote and adularia upon fissures. Under the microscope the abundant hornblendes appear as broad plates with strongest extinction and pleochroism; c>» !>:>■», c=blue-green, Ii=olive, a=yellow. Extinction, 22° 30'. Interspersed everywhere among the hornblende needles are abundant grains of pistachio-green epidote. There is a sparing groundmass of rounded, untwinned albite grains, show- ing positive bisectrix. Magnetite is abundant, but no leucoxene. Rutile occurs in the feldspar. 3. Feldspathic amphibolite. — Blandford; Osborn's soapstone quarry, at west junction of soapstone bed and granite. (See p. 87.) From a thin bed of black, very heavy feldspathic amphibolite, 20-30°"" wide, with fringe of coarse, black transverse biotite lO""" wide adjoining granite, and therefore THE CHESTER AMPHIBOLITE AND SEEPENTINES. 97 carrying orthoclase. With the microscope hornblende in broad plates with strongest pleochroisin and absorption; jc>lj>a, jc=deep blue, tr=--deep olive, a=straw yellow. Extinction 17° 30'. Orthoclase is abundant ; plagioclase occurs with extinction 24°. Little magnetite and no leucoxene. Epidote is abundant. 3. AmpJiibolite.— Chester; cutting near railroad station on the west. (See PI. VI, fig. 4, and p. 160.) A finely banded rock; interrupted sheets of" white feldspar grains rather distantly placed in ground of jet-black hornblende needles of high luster, all parallel to the common direction of the stretching. Under the microscope, stout, long blades of very deep-green horn- blende, with distant basal partings and almost no prismatic cleavage visible, show the strongest pleochroism and absorption I have ever seen, the formula for which is the same as in the last case. Between the bands of these stout blades a coarse, limpid mosaic of plagioclase gi-ains occurs, and very little magnetite appears in the slide. The plagioclase is an oligoclase with positive bisectrix and extinction at +9° to the trace of the basal cleavage on M [010]. THE SERPENTINES AND ASSOCIATED MAGNESIAN ROCKS. In the long hornblende-serpentine band which stretches across the State from Rowe to Granville, looping back in the last town so as to be repeated three times in an east-west line, the serpentines and associated rocks present a great variety, both as to present status and as to origin, and one can distinguish hornblende-serpentines, pyroxene-serpentines, enstatite- serpentines, and olivine-serpentines. There are associated with these serpentines beds of clinochlore, tremo- lite, actinolite, corundum, magnetite, steatite, talc, and, in smaller quantity, deweylite, dolomite, magnesite, clii'omite, chalcedony, picrosmine, diaclasite, bastite, and "phsestine." Of these the first five represent in general the results of other lines of change than that which has ended in serpentine, or at times a stage antece- dent to the change into the latter mineral. The talc and steatite can be traced back to tremolite or actinolite with or without the intervention of a serpentine stage, or, the purer talc especially, to serpentine of any origin; the secondary dolomite, magnesite, magnetite, MON xxix 7 98 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. chromite, and chalcedony are surplus products of the serpentinous change. Deweylite and picrosmine are, as it were, special varieties of the main pro- duct, while diaclasite, bastite, and "phaestine" (or marmolite) are derived directly from enstatite. Of course remains of original calcite, dolomite, magnesite, pyroxene, epidote, and olivine also appear. BLADBD serpentine; ANTIGORITE- (OR BASTITE-) SERPENTINE. 1. Serpentine with dolomite. — Rowe, Massachusetts. Large bed at E. King's, east of the tunnel, p. 79. Compact, even-grained, splintery fracture, dark-gray, with trace of green on fractured or sawed surface, deep oil-green to light apple-green on polished or wetted surface ; very translucent in splinters ; abundant grains of pyrite and magnetite scattered through the mass. Dr. A. J Hopkins detected chromium in small amount; for analysis, see p. 116. Slides show with pocket lens little magnetite, and preserve a uniform pale-green color, even when ground extremely thin. Under the microscope is seen a network of interlaced serpentine blades of unusual range in size, the smaller elongate, irregularly outlined as usual, rhombic, and polarizing with bluish- white color; the larger broad, flat plates, with straight, longitudinal cleavage lines, polarizing white of the first order at border and deepening to yellow at center, the largest filling quarter of the field (x 70) and deepening in color, through yellow to bright magenta. These serpentine plates are sometimes arranged radially, showing a black cross; at other times they are arranged apparently according to the cleavage of a former mineral, and are accompanied by black rod-like microlites in the same direction. Another mineral, talc, appears in small veins and broad irregular patches, as well as i-eplacing to various depths certain laminae in the broad bastite plates. It gives an aggregate polarization in bright, softly blended colors, with wavy and sharp zigzag outlines. Leucoxene occurs, surround- ing the black ore. Dolomite appears at times in regular rhombohedi'a, and is generally in rounded grains, often with only faint traces of cleavage and always without trace of twinning, 2. Serpentine. — Chester. From the large Middlefield- Chester bed, at brook-crossing on Chester road near the base of the mass. (See p. 81.) Dull-black, with shade of brown; same color when wet; conchoidal fracture; massive. THE CHESTER AMPHIBOLITE AND SERPENTINES. 99 With a lens the slide shows patches of separated grains of a yellowish, shining mineral, and between these patches run broad veins of the amor- phous greenish serpentine and many magnetite grains. Under the microscope the broad veins break up into a mass of very fine bluish-white blades, and the same lie among and separate the brightly })olarizing grains, not after the manner of the olivine network, but so that the grains seem to be scattered and woven into the mass of needles as for- eign bodies, a single needle often lying lengthwise in a crevice between grains. They are exactly like the grains of epidote in the epidotic amphibo- lites. Large tracts of these grains polarize together and show a single axis Avitli rings of color. The colors are also not so bright as olivine usually is, and it is probable that the mineral is epidote. 3. Serpentine. — Chester. From the tipper portion of the same bed, at its south end on Chester-Middlefield road. Eock shows original bedding in laminse 20-2.5°"° thick, and fine intricate jointing, the latter structure brought out by weathering, while the rock still cleaves along the planes of the first structure and shows on these planes a brownish-gray, shining surface and a texture that is suggestive of the mica membranes of the sericite-schists above, rather than of the hornblende-schists, in the continuation of which it lies. The rock breaks with a harsh, fine- splintery fracture, is of rather light greenish-gray color, translucent and mottled with black when wet. With lens the section shows, beside the large masses of magnetite, wavy lines of fine grains of magnetite §°"° apart, which run out and are replaced by others. These are seen, in sections transverse to the bedding, to be determined by the cleavage planes mentioned above, and represent the original fine foliation of the rock. The mineral shows under the microscope broad bands of fiber set transversely, and many large areas of disconnected epidote grains, all polarizing together. 4. Serpentine. — Chester. From the same bed at the north line of Ches- ter, 3J feet from upper surface of serpentine bed and contact of sericite- schist. Slaty rock, dark-gray, dull yellowish-green and translucent when wet, with reddish-gray sheen on cleavage surface, as if from mica. The slide shows no large grains of ore, only fine magnetite dust arranged in lines running in various directions, and no unchanged grains of any piimary crystalline mineral. 100 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. Under the microscope very fine serpentine needles appear arranged with a certain paralleHsm over broad surfaces, a structure inherited from the fohation; in some places they are larger, arranged irregularly, and, being thicker, show bright colors. Veins of chrysotile occur, with a certain portion of the transverse needles replaced by magnetite. For analysis, see page 116. 5. Serpentine nodule in talc. — Old emery mine, Chester. Light-gray with shade of green; pale-green, translucent when wet; very fine-splintery frac- ture, producing a whitish surface, as if dusted with fine grains. The slide is pale-green, shows large grains of iron ore, and bristles with magnetite dust arranged along cleavage laminae. Many spots of dolomite are visible with lens. Under the microscope it is seen to be made up wholly of a network of fine serpentine blades, without any an-angement significant of the primary mineral from which they were derived. 6. Serpentine. — Chester. Another interesting serpentine occurs at the emery mine. A pale apple-gi-een to oil-green rock, translucent, with shin- ing luster, and having entirely the aspect of an indurated steatite; H = 2.5. It breaks into thin, shaly fragments, bounded by wavy, slickensided sur- faces, caused by pressure, and resembles deweylite. It has, however, specific gravity 2.51, and under the microscope has the structure of a platy sei'pentine and polarizes in white to yellow of the first order, and it is decomposed by hydi'ochloric acid. The serpentines thus far described, with the exception of those from the base of the large Middlefield bed, are characterized in greater or less degree by the following peculiarities : (a) A harsh, splintery fracture. (h) Secondary magnetite wanting or unimportant, and where present arranged often in long, wavy lines of the original lamination, as proved by the cutting of slides at right angles to this lamination visible on weathered surfaces of the rock. (c) As a consequence of (V), a great degree of translucency in the rock when wet or polished, while the succeeding enstatite-serpentines are very genei'ally black and opaque when moistened. (d) The mass of the rock is made up of distinctly polarizing serpentine (antigorite) in bladed crystals, which stand in relation to the splintery fracture; and there is lack, at least nearly complete, of an amorphous serpentine substance and of clirysotile. THE CHESTEK AMPHIBOLITE AND SERPENTINES. 101 (e) The abundant grains of a brightly pohirizing, granular minei'al seem to be sometimes ejiidote inherited from the amphibolite, sometimes titanite formed around grains of menaccanite which have wholly disap- peared, while no trace of olivine can be detected. The stratigraphical indications are thus reinforced by the lithological study, and the conclusion is rendered probable that these serpentines are in large part derived from the amphibolites with which they are associated. The lack of any direct proof of the presence of olivine is, however, very far from proof that it was wholly absent from these beds. Indeed, this mineral is so closely connected with the formation of serpentine in so many cases that one may suspect its former presence here, and the observations of Kalkowsky ^ on the presence of olivine in hornblende-scliists led me to search for it in the schists adjacent to these beds, but without avail. These rocks plainly also resemble Von Drasche's^ serpentine-like rocks (now called antigorite-serpentines), and can be easily distinguished from the olivine-serpentines. I have not, however, found any certain trace of diallage in them, and the minerals which are present separate them quite distinctly from those described by him. OLIVINE- AND lONSTATlTE-SBRPENTINE. 7. '^Serpentine, Chester." — XIII, No. 53, Massachusetts Survey Col- lection. Rock dull-black, black when wet; many large grains of magne- tite. The rock weathers superficially to carbonate. With lens the slide is pale-green, and shows the secondary magnetite occupying planes of lamination and a fine system of joint planes nearly at right angles to these and very regular. Under the microscope an olivine network, inclosing in one slide frag- ments of unchanged olivine, in another lacking these altogether, runs through the slide without being influenced at all by the lamination and joint planes mentioned above. This network runs through a base of nonpolarizing serpentine. I am not able to locate this specimen, but suspect that it comes from the base of the large Middlefield-Chester bed, or from the smaller bed upon the top of North Mountain. Serpentine. — "The Crater," North Blandford. The rock of the crater is easily distinguished both macroscopically and microscopically from that 'Die GneisBformation des Eulengebirges. p 37; Tsohermaks mineral. Mittheil., 1871, p. 1. ^Ueber Serpentin und Serpentinahnliche Geeteine: Tsohermaks mineral. Mitthi-ll., 1871, p. 1. 102 GEOLOGY OE OLD HAMPSHIKE COUNTY, MASS. of any otlier locality in tlie range. It is dark-gray, scarcely shaded with green, spotted full of primary magnetite in large grains, and weathering uniformly through light green to fawn color. It is very compact and tough and much jointed. A quaHtative analysis detects about 1 per cent of chro- mium in the specimen from which slides were cut (Dr. A. J. Hopkins). With the lens the slide shows no trace of magnetite dust, but is frosted all over with shining grains of a yellowish-white crystalline mineral. Under the microscope the ground is a confused tangle of bluish-white, rhombic needles of extreme fineness, and the shining grains, polarizing brightly, are scattered in it so much like foreign grains that I suspected the slide to have been badly cleaned of the corundum used in polishing, and cut new ones carefully, but with the same result. The mineral polarizes with about the brightness of pyroxene; the angular grains are fresh to the edge and show no cleavage; some of the larger show a single axis with rings of color. It does not gelatinize with hydrochloric acid. For analysis, see page 116. 8. Olivine-serpentine. — Osborn's soapstone quarry, Blandford. Eastern bed. (See fig. 5 and page 87.) Except one specimen from Chester, whose exact location is not known to me (Massachusetts Survey Collection, XIII, No. 63, described on p. 101), this is the only bed in the long series of outcrops west of the river which contains olivine in abundance. In all the beds hitherto mentioned its occur- rence could at best be rendered only probable, though I have little doubt that it was formerly present in many cases. In all the beds discussed below the absence or rarity of olivine is equally certain, and the derivation of the serpentine from pyroxenite or coarse enstatite rock is quite clear. Indeed, much of the rock is so Httle changed that it could be as properly called enstatite rock as serpentine. The great mass of the rock where freshest is dull-black, opaque when wet, with the marked shining, greasy luster characteristic of those serpentines which still contain olivine in abundance. It gelatinizes abundantly with acid, and the solution contains magnesium and iron, with trace of calcium. A layer of surface decomposition of a drab or grayish olive-green color and 10-20"" thick covers the surface, and is sharply demarcated from the black interior. It is caused largely by the removal of the black ore, and the rock within its Umits has much more the look of ordinary olivine than in the THK CllESTEU AMPHIBOLITE AND SERPENTINES. 103 black center. The weathered layer is distinctly softer, and although the change to serpentine is not more advanced than in the interior, the olivine fragments polarize much less brilliantly than in the black portion. They may be referred to villarsite. In both the interior and the weathered crust occur distantly scattered spherules, about 10°"" across, of a finely radiated tremolite-asbestos. They are not bounded by a true spherical surface outwardly, but long, delicate needles, just visible with a strong lens, project far beyond the average sur- face. The impression is very strong that these latter are, as it were, feelers thrust forward into the mass from a center of alteration. Many of the spheres are changed wholly or partly into talc, the change starting at the center and following up the other to the periphery, and, especially in the outer layer, resulting in the entire change of the sphei-ule into talc; and as the steatite bed into which the serpentine grades has the same radiated fibrous texture, it has apparently been derived from the latter after the same manner. These radiated tufts bear also some resemblance to the radiated asbestos zone surrounding the garnets of the Saxon "garnet-serpentines," described by Dathe,^ though here no garnet center can be observed, and the radiating mineral is much coarser than would accord with the description of the Saxon occurrence. Indeed, garnet, so abundant in the next higher forma- tion, is here curiously absent from the amphibolites and associated rocks clear across the State. Slides of the freshest black portion of the rock appear under the lens to be made up of angular grains of olivine, often quite complete crystals, without admixture of anything else except a black ore arranged in rudely parallel, interrupted lines. The bleached outer layer shows nothing differ- ent, except that the black ore is removed and the whole soaked full of iron rust. Under the microscope the slides show the finest olivine network; the broad meshes of chrysotile are beautifullj^ developed and occixpy about a third of the area. The olivine is without inclusions, except small chromite octahedra, and rarely long series of straight, black needles, which are arranged parallel to the vertical axis and at right angles to the length of the series with the regularity of a micrometer, except that some lines are 1 Oliviiifels, Serpentm, and Eklogit des sachischen Granulitgebietes : Neues Jahrbuch, 1876, p. 225. 104 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. prolonged far beyond the others. Highly magnified, they are reddish and translucent. In other portions of the slide the characteristic enstatite- serpentine occurs, with exactly the structure of fig. 3, PI. II. In the outer, bleached layer the lines of magnetite dust in the suture of the chrysotile bands have disappeared and in the form of limonite have soaked through the fibrous chrysotile. With further process of change the latter loses its fibrous character, and every trace of the origin from olivine disappears. At times also the olivine grains here retain their position in the clu-ysotile network and their appearance, but have lost their bright colors of polarization. The change of the olivine to tremolite seems to be of earlier date, and has often advanced to the replacement of nearly all the olivine. It appears to have occurred earlier than and independently of the superficial weathering, because the tremolite is found in equal abundance in the fresh interior and in the weathered surface. Where the olivine is weathered, as in specimens taken from large bowlders on the West Granville road, just where the glacial currents would cany material from this bed, the specimens are not distinguishable from corresponding specimens from the Pelham and Shutesbury beds (see p. 47), and I had expected confidently to find the fibrous mineral to be antho- phyllite, and indeed its powder gives fragments which foi- the most part extinguish longitudinally, but at times other fragments extinguish with an angle of 11° to 15° to the length, and show a single optical axis placed laterally, exactly as in hornblende. A cleavage parallel to go P oo (100) seems to be unusually well devel- oped, and most fragments rest upon it and so extinguish longitudinally, but show only one axis, as in hornblende. It is to be noted that this is from a unique bed lying east of the main hornblendic band. (See p. 89.) 9. Pyroxene-serpentine. — Lowest bed, near Osborn's quarry. (Fig. 5, p. 87.) The lowest bed at the margin of the north-south brook in the woods below Osborn's quany attracted my attention immediately as some- thing quite unlike any other rock connected with the serpentine series, to which it manifestly belonged; indeed, unlike any rock with which I was acquainted. The weathered surface is for the most part rough and warty, dirty white, and covered with shining scales of talc. In some places this layer, 5-10""" thick, is covered by a white, powdery layer of magnesite. PLATE II. 105 PLATE II. THIN SECTIONS. Fig. 1.— Sahlite changing into tremolite; the hitter heginning to change into serpentine. From the lower hed at Oshoru's soapstone quarry, Blaudford. X47. See pp. 87, 104. Fig. 2. — Dolomite changing to serpentine. Granville. X28. See p. 110. Fig. 3.— Enstatlte crystal altered to serpentine, cut parallel to (001). Drawn with polarized light and with the light bands placed at 45° to the plane of polarization. x60. See p. 110. Pig. 4. —Garnet, with complex border; from pegmatite. Northfield. Xl4. See p. 328. 106 MONOORAM A . ATI 8 THIN SKCTIONS -Bablite c-bani^injr into tremol ■il, lo-i. i;toue nuuiry, IWai.'. tine. Granville. .. itcrcil to serpentine, out parallel to (001). Drawn witli polarized light ht bands placed at 45° to the plane of polanVnti"" -fill j^i ii p no lex border; from pegmatite. Northfield. X U. S. GEOLOGICAL SURVEY MONOGRAPH XX(X PU 11 THIN SECTIONS. THE CHESTER AMPHIBOLITE AND SERPENTINES. 107 fP riic interior is ;i dark oil-greeu, flecked with white or yellow; opaque black when wet, and witli pecuHar greasy luster i-eflected from large cleavage surfaces, which run through the whole mass. These surfaces reach a size of" 20 x 40""", have a pearly luster, and are, in many cases where the rock is deei)lv weathered, bleached to an isabella-yellow or changed to a white mass like kaolin. They are covered with an acute-angled network from a second cleavage, like that of hornblende, but more acute, and a satiny sheen runs over the face from the presence of fine tremolite needles, arranged parallel to this cleavage and gradually encroaching upon the original mineral, which proves upon microscopical study to be sahlite. Slices cut parallel to the perfect parting (see fig. 1, PI. II), which proves to be P (001), show a fine, regular network of tremolite needles, which polarize with an obliquity of about 15° and coincide in position with the acute cleavag-e of the orig-inal mineral mentioned above. Where this secondary tremolite has not come to occupy the whole space the meshes are occupied by a colorless sahlite, showing in traces an interrupted pris- matic cleavage and a delicate lineation parallel to go P oo (100), with traces of a second at right angles to this. In the figure this fine lineation is of necessity too coarsely represented, it being visible only with high power, and it is given specially to show the extent of the unchanged sahlite. The latter polarizes with .extreme brilliancy, and characteristic sudden changes of tint appear over its sm'face, arising from the brittleness of the thin laininse due to the very easy parting on P (001), which renders it difficult to polish it to a true surface. The mineral is positive, and the optical axial plane is at right angles to the fine lineation — i. e., is in oo P co (010) — and a single axis appears, and this plane bisects the acute angle of the tremolite network, which meas- ures about 54°. This would make the cleavage, which has determined the position of the tremolite fibers, approximate to co P 2 (120), the counterpart of oo P 2 (210) — the prism of hornblende when reckoned upon the pyroxene axes. An examination of tlie Bolton (Massachusetts) sahlite shows that dis- tinct traces of the same cleavage existed in fresh specimens. Here, as is not unusual, it is rendered much more distinct in the process of decomposition. In slices cut at right angles to the perfect basal cleavage or parting the strong equidistant lines of separation are the marked feature, and these lines quite far apart are the seat of most advanced change. 108 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. The perfect sahlite cleavage is also retained after tlie change to serpen- tine is far advanced, and shines out when the piece is held in a particular position as single faces luster-mottled with small opaque spots of serpentine. In the process of change, black iron ores in rods and lines of dots appear in the interstices of the tremolite needles, as Avell as in the prismatic and pinacoidal cleavage of the sahlite; and after the serpentinous change (which commences in the transverse cleavages of the tremolite and advances in a network somewhat like that in olivine) has completely transformed the whole into a confusedly polarizing mass, the acute-angled network is as clearly marked in ordinary light as before by the black lines, and in places traces of the rectangular pyroxenic cleavages can also be seen. In the most completely changed portions blades of actinolite, either later formed than the tremolite or more resistant than it, show marked ple- oclu-oism for so nearly colorless a mineral — pale blue-green to ochei'-yelloAv. 10. Enstatite-serpentine and steatite. — Hartland, Connecticut, just south of the town line, where the road from West Granville to Hai'tland crosses Hubbard Brook. On passing 100 feet up the slope southwest of the bridge the black enstatite-serpentine occurs in force. It is of finer grain than the other beds to the north, with which it otherwise agrees exactly, and it is largely changed to steatite. 11. Tremolite rock and enstatite-serpentine. — J. Downey, Granville. Fol- lowing a wood road east into the densely wooded swamp from a point just north of the house of Mr. J. Downey, I came upon a very interesting out- crop, which represents the first occurrence of serpentine upon the amphibolite band, where, after turning north, it swings around the gneiss of Granville. Along east of a band of the common amphibolite there crops out a low ridge of limestone, at times quite pure, light-gray, and thin-fissile, but taking more and more very fine tremolite into its composition, until it comes to be a flat-fissile, pale-green tremolite-schist, almost as fine- grained as nephrite, which it somewhat resembles. It polarizes brilliantly, has extinction in maximum 27°, and shows a few straight, black microlites and a few large grains, also visible to the eye, of a black magnetic ore. It gives on analysis only traces of AlgOg, FeO, and CaO, and is an almost pure silicate of magnesia. South of this there are no exposures for a short distance, and in the strike of the tremolite rock rises a great knob of enstatite rock. It is a THE OnESTER AMPHIBOLITE AND SERPENTINES. 109 coarso, ragged rock, made up of cr3^stals of enstatite, often 10 x (! x 4™ in size; and no other original constituent can be detected except dolomite, which is inclosed in large, rounded grains in the freshest enstatite. Indeed, in much of the mass the whole is made up of the large, imperfect, interlaced prisms. These are thickly coated by a greenish-gray talc-like product of decomposition, which also penetrates in thick layers along the perfect cleav- age until tlie whole is changed into bastite and ultimately into talc. Where the change is more advanced, great sheets and remnants of the bastite, gray-green in color, lie in a mass of black serpentine, or in a mixture of this and a yellow dolomite. Often, however, the enstatite rock seems to degenei'ate into a talc-like mass without an}?^ trace of serpentine, and the masses of black serpentine and dolomite may have originally contained some other mineral besides the enstatite, though I could obtain no proof of this. Under the microscope the enstatite, cut parallel to the perfect cleavage, shows in the freshest portions only a few black mici'olites, but it is much cut up by a network of yellow serpentine ; and here large octahedra of magne- tite appear. With convergent light it polarizes in bright colors, and, of course, shows no axes, and on moving the slide spaces are found which show the axes as in diaclasite, accompanied with bright colors, and these parts are not distinguishable in ordinary light from the unchanged ensta- tite. Moving the slide a little farther, one sees the axes as in bastite and in paler colors, and in common light these parts have the appearance of ser- pentine. The divergence of the optical axes is very small for the bastite, certainly less than 30°. Embedded also in the black serpentine are, rarely, large scales of a deep-green clinochlore, with divergence of the optical angles of about 10°. By the roadside near the cemetery, southwest of the last locality, are found bowlders of all the varieties of rock mentioned from this outcrop, and several others of interest which seem also to come from this place, although this can not be made certain. One great mass of limestone is in places banded in dull black, from the large amount of magnetite in the limestone. It shows also, under the microscope, a large number of fine actinolite needles and rarely a grain of coccolite, and this piece is so exactly, in other parts, like the tremolitic limestone found in place north of the serpentine knob that it is scarcely possible that one can err in assigning to them the same 110 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. origin. It contains, however, bands several centimeters broad of dark- green, matted, fibrous actinolite. Associated with this in another large bowlder is an entirely distinct limestone, a white, coarse, granular calcite, all the grains showing with the lens abundant twin striation, while the limestone mentioned above is too fine-grained to allow its structure in this respect to be seen, and under the microscope it is not twinned at all. It must be calcite, however, as it effervesces freely with cold acid. In the limestone now described a distinct stratification is effected by the interposition of fibrous masses of emerald- green actinolite upon the foliation planes, and in the midst of the granular calcite fine grains of coccolite and magnetite occur. 12. ^^Dolomite-serpentine, Granville^ — XIII, No. 26, Massachusetts Sur- vey Collection ; PI. II, fig. 2. This is a black serpentine, containing much white to greenish, granular dolomite, and is identical with the bowlders described above from near the cem.etery in Grranville. Remnants of the gi-ay-green enstatite in every stage of change to phgestine appear, and prove, under the microscope, identical with those described above, and the traces of enstatite structure can also be distinctly seen in the completed serpentine. The most interesting change here is that of the dolomite into chrysotile, many stages of the Eozoon structure being beautifully represented. The slide (PL II, fig. 2.) shows a network of yellow serpentine, amorphous in common light, running through the dolomite and generally following the cleavage. The dolomite network appears where the car- bonate has wholly disappeared. The dolomite fragments are surrounded by a quite broad, dark band, consisting of short, stout rods of the unchanged dolomite which project into the serpentine. The dolomite shows exceptional absorption, and this dark band absorbs and extinguishes with it. Outside this band the serpentine veins polarize with wavy extinction and low colors, and show the moat delicate fibrous structure, with central suture. 13. Enstatite-serpentine. — H. Cooley, Granville (PI. II, fig. 3). The section cut parallel to the base of the large crystals of enstatite changed into serpentine shows a series of bands which appear in pairs separated by a narrow line of magnetite. These are the light bands seen in the figure, and broad surfaces could have been selected where these bands were more closely parallel than in the one drawn. The drawing was made with crossed nicols. .1 0&S^ THE CHESTER AMPHIBOLITE AND SERPENTINES. 1 1 1 and the light bands are placed at 45° to the plane of the instrument. Parallel with this plane they are black. They show an extremely fine, transverse, fibrous structuj-e. The intervening lens-shaped fields, dark in the drawing, are black in this position at 45° when the bands are white; when rotated to 0° they sliow white, radiate-fibrous tufts on a black ground of nonpolarizing serpentine. The deposition of magnetite in certain of the places of perfect cleavage was accompanied by the very regular change to fibrous serpentine growing out from the planes and forming the white bands. Then the gi-owth of secondary serpentine between some of these bands has seemingly wedged them apart, and given them their curved forms and produced the lenticular fields of serpentine. 14. "Serpentine (bowlder). Blandford." XIII, No. 11, Massachusetts Survey Collection, 1841 ; No. 880, 1835. — Compact, fresh surface, bluish- black, mottled with dull brownish-black; contains much magnetite; is unlike any other serpentine known by me from this region, and comes, doubtless, from the bowlders noted by Dr. Hitchcock on the east line of the town.-' With the lens the slide is seen to be mottled with large green spots, which were doubtless formerly enstatite and which retain its structure, though com- pletely changed; and in other spots traces of an olivine network can be made out with much probability. The rock is, however, for the most part com- pletely changed, and shows everywhere the softly colored polarization of talc. 15. Enstatite-serpentine. — Atwaters, Russell. Following the hornblende band northward to the point where it bends round through the southeastern corner of Blandford, where I suppose the above specimen No. 11 of the Massachusetts Survey Collection was obtained, we come upon the great out- crop in the high hill in Russell overlooking the Westfield plain, where many years ago the rock was quarried extensively as black marble by Mr. Atwater, the father of the present proprietor. The black serpentine presents no peculiarities by which it can be dis- tinguished microscopically from the other localities in Granville and Russell. Because of the deep quarrying the finest specimens can be obtained from this place, and the bastite is of a beautiful apple-green color, instead of the pale gray seen elsewhere, and was found by Tschermak to have the low angle of 30° for the optical axes. ' E. Hitchcock, Geol. Mass., 1841, p. 617. 112 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. In a considerable number of thin sections the enstatite structure was found everywhere strongly marked by broad, distant, granular bands of black ore, with very numerous straight, narrow lines made up of rods and grains of the same black ore and running at right angles to the broad bands. No trace of enstatite could be found unchanged, and the broad apple-green plates, which are often quoted as marmolite, from these localities had passed for the most part beyond the bastite stage into a network of serpentine needles, in which isolated bastite plates still remain. Irregular grains and large patches of carbonate, much corroded and polarizing with a soft amber color, with faint irised border, occur every- where; also miscroscopic veins of satin spar, sometimes insinuated in large number between successive plates of the bastite. No trace of olivine structure could be discovered in any of the slides from this locality; and in the localities described later, where the rock is less changed, it could be seen that all of the rock exposed was made up of large enstatite crystals so closely apposed that there could have been at best only a trace of olivine present; and on the broad cleavage surfaces of the enstatite no trace of included olivine grains could be seen. The structure in the completed serpentine was everywhere the rectangular network, as characteristic of the enstatite-serpeutine as the olivine network is of the latter mineral. It is beautifully illustrated by Dr. Wadsworth in pi. 7, fig. 2, of his Lithological Studies,^ from a specimen obtained "four miles from Westfield Center, Westfield, Massachusetts." This must have come from the Atwater ledge, which is just 4 miles west of Westfield callage, but lies across the line in Russell. As the rock from which the serpentine was derived was a nearly or quite pure enstatite rock. Dr. Wadsworth's assignment of it to the peridotites can not be accepted. 16. " LigJd-green, compact serpentine. — Russell." XIII, No. 25, Massa- chusetts Survey Collection. This is a superficial layer a few millimeters in thickness, which also runs in veins into the black serpentine, and is super- ficially covered by a rusty white layer. It is probably from the surface of the above bed. It presents under the microscope a tremolitic structure throughout — radiated, fibrous, a late stage of the change into serpentine. 17. '■^Serpentine (bowlder). — Russell." XIII, No. 50, Massachusetts Survey Collection. This is certainly an erratic derived from the great bed 1 Mem. Una. Comp. Zool. Harvard Coll., Vol. XI, pt. 1, 1884. THE CnESTEE AMPHIBOLITE AND SEEPENTINES. 113 in Middlefield, and probably found in the valley of the Westfield in the north part of Russell. 18. "Black serpentine, talc, actinolite. — Westfield." XIII, No. 24, Massachusetts Survey Collection. Subgranular, dull black, very little talc, derived from enstatite. The actinolite mentioned above is in part enstatite partially changed to bastite, in part fine radiating tufts of tremolite, green from the back- ground of serpentine. The rock is traversed by veins of snow-white, fibrous calcite 20-30°"° long and 2™" wide, with satiny transverse fibers and central suture. It shows under the microscope large masses of unchanged pjrroxene with coarse co P cleavage, and long, black microlites, often crossing each other rectangularly in three directions. It changes outwardly into coarse, radiated, fibrous tremolite (cleavage 124°), which is altered along prismatic and transverse cleavage into serpentine. 19. ''Serpentine and calcite. — Westfield." XIII, Nos. 27, 28, 29, 30, Massachusetts Survey Collection. The first two are wanting in the collec- tion. No. 29 is a contact piece of dolomite, with light-green and straw- colored serpentine running out into it from a mass of oil-green serpentine with fine, broad veins of chrysotile and many characteristic eozoonal struc- tures. It shows beautifully every stage of the change of dolomite into a colorless, almost perfectly amorphous serpentine, showing no needles and only faint patches of color with crossed nicols, and in many cases these serpentine grains retain perfectly the cleavage and the repeated twinning planes of the dolomite. No. 30 is a white, bedded limestone with distant, thin partings of ser- pentine, probably originally an actinolitic limestone. Traces of hornblende with extinction 14° could be seen. 20. ''Massive garnet. — Westfield." XIII, No. 40, Massachusetts Survey Collection. — This is a granular mixture of quartz, garnet, and pyroxene, and can have been introduced here only as one of the rocks bordering upon the Atwater serpentine bed. 21. "Compact scapolite (?).— Westfield." XIII, No. 32, Massachusetts Survey Collection. — A bluish- white, translucent, partly sparry, partly cryp- tocrystalline mass, showing the distinct, very fine, triclinic striation of a plagioclase, exactly like that associated with the serpentine at the Chester emery mine and at the Pelham asbestos quarry. MON XXIX 8 114 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. 22. Enstatite-serpentine. — Westfield. Prolonging the line of the Atwater bed in Russell southward, it is found that many large bowlders occur where it crosses the road to Grranville, and farther south an immense block lies in the woods on the hill back of S. Drake's house, in the west parish of Westfield, which is made up of very large enstatite crystals in every stage of change to bastite, and the whole mass is exactly like that described above at J. Downey's, in West Granville. Farther south, along Munn's brook, where it cuts a deep gorge in Sodom Mountain, at the head of the gorge, near the house of H. H. Pur- chase, and at the mouth of the same, are many great bowlders of a similar black serpentine with large crystals of enstatite changing or changed into bastite or white talc. All these occurrences, to this last by Munn's brook, are so similar that a single description of slides cut from the great bowlder at H. H. Purchase's, Granville, may serve for them all, and this occurrence is so like that near J. Downey's, in West Granville, that the description can be brief, as I shall note only the important differences. Slides of the freshest enstatite show a reddish-yellow color and polarize brilhantly. Some of them contain very abundantly the stout, straight black rods; in others they are as rare as in the former locality. Magnesite in rounded grains and distinct rhombohedra is found abundantly in the freshest enstatite, apparently as a primary constituent. A qualitative examination determined the absence of calcium. With cold hydrochloric acid fragments of the serpentine presented no change; with boiling acid there was a long-continued effervescence, and there remained an interlaced mass of altered enstative crystals. 23. "Diallage in serpentine. — Sodom Mountain." XIII, No. 48, Massa- chusetts Survey Collection. The specimen is wanting in the collection at Amherst, but it must have come from one of the latter localities along the gorge of Munn's brook, through Sodom Mountain, and must have been an enstatite and not a diallage rock, the latter mineral not occurring in the serpentine range. In the long line of outcrops from Zoar to North Blandford the serpen- tine is characterized by deep oil-green colors, marked translucence, and freedom from secondary magnetite. It is composed of fine serpentine blades mingled with softly irised films of talc, and still contains the scattered grains THE CHESTER AMPHIBOLITE AND SERPENTINES. 115 of hi"-lily refractive epidote, which have passed unchanged from the parent rock, and often show traces of its stratification. It would seem, then, to have been derived from the common epidotic hornblende-schist (amphibolite) with which it is associated. In a great number of slides no trace of olivine or enstatite, or of the structures characteristic of these minerals, could be detected except in the bowlders from Chester (Massachusetts Survey Col- lection, XIII, No. 53), whose exact locality can not be determined, and those from the base of the large Middlefield bed, from which place the Chester bowlder may have come. The presence of so large masses of chi'omite and of nickel requires explanation, and indicates that some parts of these beds were once chrysolitic. The specimen labeled XIII, No. 43, of the Massachusetts Survey Collection, from Chester, probably from the south end of the large Middlefield bed, where is an old " mine " of this ore, is a mass of chromite 50"™ on a side. Moreover, the discovery that the so-called quartz pseudomorphs from the Middlefield bed are serpentine pseudomorphs after olivine^ must receive consideration in this connection. They are, how- ever, large, perfectly terminated crystals, some of which have broken off from the walls of the cavity to which they were attached, and they can not be taken as normal idiomorphic constituents of an ultra basic eruptive. They are better explained as fissure minerals in a crystalline limestone, like the Snarum crystals. I conclude, therefore, that nearly all of these serpentines are derived from the amphibolites, and find the stratigraphical evidence in this direction strengthened by the lithological evidence. Osborn's quarry at Blandford is a dividing point between the trans- lucent hornblende-serpentines on the north and the black enstatite-sei-pen- tines on the south. To the east of the central steatite bed, which is quarried, is the isolated and unique oli vine-serpentine bed; to the west is the equally peculiar sahlite-serpentine bed. The black serpentine and dolomite mass which forms the center and least changed portion of the central steatite bed is the first of a series of such deposits which locally replace the amphibolite in its further progress south. They are rare and subordinate to amphibolite in the broadened portion of the bed across West Granville, but after the bed has bent northward at East Granville they increase in relative importance and are associated with 1 See BuU. U. S. Geol. Survey No. 126, 1895, p. 91. 116 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. considerable beds of white crystalline limestone, dolomite, and tremolite schist, and in the further prolongation of the bed across Russell and West- field the enstatite-serpentines and the half-altered beds of very coarse enstatite rock associated with great beds of crystalline limestone become entirely predominant. The black, fissile amphibolites accompany the ensta- tite-serpentine beds to the end. The presence of dolomite as an original constituent, both as inclusions in the unchanged enstatite and in the interstices of the enstatite rock, as well as in great beds, indicates the derivation of the whole series from large beds of dolomitic limestone. The subject is summarized on page 147, after the description of the emery bed. Below is given a table of all the analyses of serpentines known to me from this area. The material used in the analyses 1 to 6 was furnished by the author, and, excepting 2 and 9, great care was taken to detect all the rarer constituents, especially nickel, cobalt, and chromium. Analyses of serpentines. [Nos. 1, 3, 5, 6, and 8, by Mr. George Steiger, in the laboratory of the United States Geological Survey ; No. 3, by Miss H. P. Cook, instructor in chemistry in Smith College ; No. 9, by Prof. C. IT. Sbepard ; No. 2, by Melville, quoted from Dana's Manual, p. 672; Nos. 4, 7, and 11, by Dr. W. F. HUlebrand, of the United States Geological Survey.] 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Si02 40.42 None. 1.86 2.75 4.27 .43 .28 .53 Trace. Trace. .66 None. 35.95 } .16 44.22 40.27 38.62 None. .35 3.44 3.99 39.14 None. 1.18 4.46 3.14 33.87 None. .77 2.81 4.25 40.77 None. 1.16 3.56 1.47 37.82 Trace. .61 7.92 1.15 44.16 7.05 36.94 Trace. .50 6.04 1.94 54.04 None. .52 1.51 3.90 TiOj AI2O3 .53 6.61 }5.74{ FeO CrnO^ .39 .21 .33 .47 Trace. None. None. None. 41.45 1 None. .38 \ .33 .04 None. None. 38.67 None. .28 .09 None. 39.37 \ .14 Trace. .49 12.48 .19 .45 .05 .33 .40 None. Trace. None. None. 38.35 None. .14 .23 .11 None. None. 34.40 .08 NiO CoO MnO .10 .40 CaO None. None. 37.94 JTrace. 37.44 BaO MgO 37.54 40 40.61 , .08 1 .10 Trace. .36 10.91 KjO NajO Li H2O— 100 H2O+IOO .21 10.51 Trace. Trace. 1.44 .36 11.26 .69 13.61 .34 9.48 None. .02 None. .38 7.00 .20 Trace. 10.82 .75 12.50 11 .71 12.07 .20 Trace. 1.85 .70 3.07 None. 1.32 P„0= Trace. .62 Trace. None. Trace. C02 99.47 100. 52 100.31 100. 08 100. 01 99. 42 100. 08 99.38 99.65 99.33 100. 02 No. 1. Eicb, dark-green serpentine. Eowe. Quarry near E. King's. No. 2. Picrolite. Florida. No. 3. Straw-yellow, fibrous serpentine, glazed, enveloping olivine pseudomorpbs. Middlefleld. From the speci- men figured in Bull. U. S. Geol. Survey No. 126, PI. I. THE CHESTER AMPHIBOLITE AND SEEPENTINES. 117 No. 4. Normal dark-green^ slightly oily eerpeutino, from the ceuter of the large Middlefield bed, taken from where the road crosses the Cho3t«r-Middlefiold line. No. 5. Black-groen serpentine, weathering to pale niekel-greon, with much chromite. North Blandford. From "The Crater." No. 0. Gray, splintery serpentine enveloped in talc. Chester. From the east wall of the old mine. No. 7. Serpentine from the lower bed at Osborn's quarry, Blandford, which still retains the cleavage of sahlite and iB places considerable remnants of the mineral. No. 8. Enstatite changed to serpentine. Granville. H. Cooley's. No. 9. Black serpentine with bastite. Eussell. No. 10. Black serpentine containing marmolite (bastite). Kussell. Atwater's quarry. No. 11. Slightly altered, nearly colorless enstatite, from Downey's, in Granville j added for comparison. The constant content of nickel, cobalt, and chromium in all these analyses where it has been searched for is very interesting and may be taken as an indication of the eruptive origin of the whole series, which would, however, involve the derivation of large beds of white crystalline limestones, both dolomitic and quite purely calcareous, from the same basal eriTptive rocks. THE CHESTER EMERY BED. HISTORY OF DISOOVEKY AND WORKING- OP THE BED. Not the least interesting element in the peculiar geology of the west- ern part of Chester is the great magnetite-emery bed which lies along the upper (eastern) line of junction of the hornblende-schist with the sericite- schist and extends from the Westfield (better Agawam) River southward nearly to the south line of the town and nearly as far as the great horn- blendic band retains its maximum thickness. The history of the discovery of this bed has often been told, and deserves to be retold. The credit of the discovery and its first announce- ment belongs to Dr. C. T. Jackson. I remember how Professor Shepard, when taking my college class through the cabinets in 1865, stopped at the old State geological collection made by President Hitchcock dm-ing his survey of Massachusetts, and took down the specimen of magnetite col- lected from the Chester bed and pointed out to us the emery which it contained, to show us how near Dr. Hitchcock had been to ntimbering this among his many discoveries. Dr. Hitchcock had described several beds of magnetite for the first time in his final report.^ They were located in the western part of Chester, in hornblende-schist, and none of them exceeded 1 foot in width. For the next events in the history of the locality I must have recourse 1 Geology of Massachusetts, 1844, pp. 194, 612. 118 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. to the " Biogi-apliical Sketch" of Dr. Herman S. Lucas, of Chester, inserted at the end of the town history of Chester.^ From the well-known manner in which this book was compiled, and from internal evidence in the sketch itself, one is led to attribute to it somewhat of an autobiographical character. The paragraphs bearing upon the history of the emery beds are as follows : But perhaps tlie most remarkable event in his career was the discovery of what was for some time supposed to be a vast deposit of iron ore in the mountains around Chester. This occurred in 1856, and arrangements were at once made for the opening and working of the mine ; and in the course of about one year 1,200 tons of mineral were taken out and transported to the furnaces of Stockbridge, Lenox, and Hudson. The financial crisis of 1857 compelled a discontinuance of the business, and it was not renewed until 1863. In the last-mentioned year, in company with his brother, John B. Lucas, and Henry D. Wilcox, he resumed the business. A blast furnace and forge were erected in Chester and the manufacture of iron was commenced, but the ore proved somewhat intractable and the results were not satisfactory. In the mean- time the Doctor made a more thorough examination of the mineral, and on the 6th of September, 1864, discovered that it contained a large percentage of emery, a mineral hitherto nearly unknown in the United States; in fact, this is believed to have been its first discovery in America. Coming at a time when the country was engaged in a terrible war with internal enemies, the discovery was doubly valuable. Heretofore the emery used in this coun- try had been imported from the Turkish dominions, and as the English and French Goverments had a monopoly of the mines near Smyrna, in Asia Minor, and on the Greek island of Naxos, in the archipelago, the United States Government was debarred from procuriiig its necessary supply, except under unusual difflculties. In this dilemma the Chester emery was utUized and the Government works were supplied from it for a considerable time. This mineral had repeatedly been examined by various scientific gentlemen, and specimens were placed in the collections at Amherst College and in that made by Professor Hitchcock for the State and labeled magnetite. In 1868 Dr. Lucas, with Messrs. Charles Alden and H. D. Wilcox, formed what was known as the Hampden Emery Company, and erected a mill on the river below Chester village for the manufacture of emery. In 1874 questions touching the ownership of the mines involved the Doctor in litigation, which necessitated a change in his business, and from that time he has obtained his mineral mostly from the Turkish mines. The Doctor continued in business in his own name until May 1, 1878, when Nathan Harwood became associated with him. The importance of this discovery is well illustrated by the remark of a great English statesman, that "The discovery of an emery mine was of more value than that of many gold mines." 'History of the Connecticut in Massachusetts, vol. 2, p. 1064; Lewis H. Evert, Philadelphia, 1879. THE CHESTEE EMERY BED. 119 I thiuk that one may hazard the conjecture that the last sentence in the above was quoted from memory, and thus rather imperfectly, from the opening sentence of the article cited next below, though there is no other mdication that the biographer of Dr. Lucas was acquainted with the part taken by Dr. Jackson in the discovery of emery at Chester. It will be well, therefore, to let the account of the matter given by Dr. Jackson himself follow:^ It has been said in England that " a good mine of emery is worth more to a man- ufacturing people than many mines of gold." Such being the case, it affords me great pleasure to be able to announce the discovery of an inexhaustible bed of the best emery in the world in the middle of the State of Massachusetts, in Chester, Hampden County quite near to the Western Railroad, which, with its ramifications, leads to the largest armories and manufactories of metallic articles in this and the adjacent States. „ ... For more than two years the existence of important beds of magnetic iron ore, originally discovered by Dr. H. S. Lucas, has been known, and endeavors were made by that gentleman to organize a company for the purpose of smelting these ores. In consequence of this agitation I was employed by John B. Taft, esq., on the 19th of October, 1863, to examine the locality and to make report of my results to him. On examination of my specimens of minerals after returning to Boston, and my notes for sectional profiles of the rocky strata containing the iron ore, I found that the minerals margarite and chloritoid, in talcose, hornblende, and mica slate rocks, indicated the occurrence of emery, the association of the rocks and minerals being identical with conditions known to exist in the localities of emery in Asia Minor. I therefore called the attention of the owners of the property to these facts, and directed that search be made for emery, and that every mineral resembling it should be sent to me for examination. Little attention was paid to this prediction at the time, nor until I had invited Dr. Lucas, who resides in Chester, by personal represen- tations and solicitations, to make the required search, the characters of emery being fally described to him. . On his return to Chester he soon learned that the miners were complaming of the great hardness of the supposed iron ore, and that no less than forty drills were dulled in boring a single hole for blasting. He then sent me pieces of this hard rock, in the belief that it was the emery I had predicted. On examination it was found to scratch quartz and topaz readily and to have all the properties of emery. A chemical analysis proved it to be identical with the emery of Naxos. The owners, resident in Boston, being notified of this discovery, went with me to the locality on the 11th of October last, when a full exploration of the premises was made There are several large beds of rich magnetic iron ore at this locality, and the emery being magnetic (as it always is) has caused it to be mistaken for magnetic iron ore, and many tons of it had been smelted wjth^the_c^nate ^ iron and hematite in :^^^s^J of Emery in di^ster^Blachusetts, by Charles T. Jackson, M.D., Geologist and State Assayer: Am. Jour. Sci., 2d series, Vol. XXXIX, May, 1865, p. 87. 120 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. the Berkshire County irou .furnaces without a suspicion, notwithstanding its refrac- tory nature, that the ore was emery, with only a small admixture of iron ore. The mineralogical and geological data of the article are quoted under "Emery" in the Mineralogical Lexicon covering the field of this mono- graph^ and under the general description of the vein (p. 135). The conclu- sion of the article is as follows: It may be proper to add that John B. Taft, esq., of Boston, in behalf of his associates, owners of the emery mine, has the sole management of the business con- nected with the mine. I would express my obligations to Mr. J. L. Smith for the valuable information contained in his articles on the emery of Asia Minor and on the associated minerals of the emery localities published in Vols. X and XI of this journal; also to Dr. H. S. Lucas, of Chester, for kind assistance in the field.^ It seems thus that the veins became known to Dr. Hitchcock between 1836 and 1841, and that Dr. Lucas, who, as an ardent student of the min- eralogy of his native town, was doubtless acquainted with Dr. Hitchcock's published work, examined the beds, became convinced of their economic value and began work upon them in 1856. He renewed his work in 1863, bat as an iron industry it did not prove profitable, and the property passed into the hands of a Boston company represented by Mr. John B. Taft, and in which Dr. Lucas was interested. Largely, perhaps, on account of the refractory nature of the ore. Dr. Jackson was employed to examine the mine, and, relying upon the earlier investigations of J. Lawrence Smith, predicted the occurrence of emery from the associated minerals and urged Dr. Lucas to search for it. In the meantime the miners had practically discovered the emery, much to their sorrow, and I have been informed bv two who Avorked in the mine at the time that they were well persuaded that the brown mineral was what dulled their tools and were accustomed to call it emery. Armed with this practical and scientific information. Dr. Lucas investigated the ore anew, determined the emery, and sent the specimens to Dr. Jackson, upon which he made his mineralogical tests and, most important of all, his chemical analyses.^ 1 Bull. U. S. Geol. Survey No. 126, under "Corundum." 2 In a recent report of an interview with Dr. Lucas (Springfield Eepublioan, "Another vein of corundum," December 12, 1895) the history of the discovery is repeated much more nearly in accordance with the account of the matter I have given than with the biographical sketch quoted THE CHESTER EMERY BED. 121 This o-ave rise to the emery mining', and, in 1868, to the formation of the Hampden Emery Company, in which Dr. Lucas, Mr. S. A. Bartholomew, of Blandford, and Dr. Jackson were interested. A few years later this com- pany deeded, in apparent good faith, what they supposed to be the main vein, of which they had previously bought the mining right, to the Chester Iron Company, afterward the Chester Emery Company, a stock company controlled by Mr. James T. Ames, of Chicopee, Massachusetts, of the Ames Manufacturing Company. The older company deeded "commencing in the middle of the vein of iron ore and running at right angles to the same 5 rods, thence parallel to the same to its south end, thence 10 rods at right angles to the same, thence parallel to the same to the north end, thence 5 rods at right angles to the same to the place of beginning, being about 4 acres." The new company went to work upon its purchase, while the old company continued to work ujDon a vein farther west, which it still held. In a short time it was found that the new company was working upon a line of bowlders derived doubt- less from the true deposit to the west, upon which the old company was still working vigorously. Theretipon arose an important lawsuit, the Chester Company claiming the true vein, while the old company claimed that a blunder had been made by all parties, and that the attempt to apply the deed to the western vein would give 7 acres instead of 4. After protracted litigation the case was decided for the purchasers, and the vein came into the hands of the Chester Company and was worked by it, extensive buildings being erected and expensive machinery obtained. In 1879 thirty-five men were employed and 210 tons of emery were produced, valued at $20,000.^ The mine was worked apparently without much profit, since in 1883, after the death of Mr. Ames, the whole property, said to have cost above $80,000, was pm-chased by Dr. Lucas for a sum reported to be about $12,000. On the adverse issue of the lawsuit Dr. Lucas had, with customary energy, turned his attention to the Naxos emery, and curiously, from the above. Dr. Lucas here claims to have discovered the emery in 1864; it is said that the miners "could make very little headway against the rocks, which, they told Dr. Lucas, were so hard that they could not keep their tools sharp;" that Dr. Jackson told Dr. Lucas that the margarite was sometimes found with emery, and " it was this that gave Dr. Lucas his clew." 'Hist. Conn. Valley, Vol. II, p. 1063. 122 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. fact that the crude emery could be entered free while the manufactured article was highly taxed, and from the further fact that sailing vessels trading with the Mediterranean were compelled to return in ballast and were thus willing to deliver the crude emery in New York with little or no charge for freight, he was able to create a lucrative industry. Latterly his sup- plies have been drawn principally from the Southern States. In 1883 no work was in progress except at the north mine, where six men were employed, and during the year even this work was stopped. In 1890 the old mine was reopened and has since been worked continuously, and the working has reopened the area where the fine diaspore occurs. In 1894 an adit was driven into the hill on the north side of the road along the eastern wall of the vein. In 1865 Prof C. U. Shepard published a report upon the mine,^ mainly mineralogical, but containing notes on the geology of the vein. The repoi-t was published in such form that it is not now obtainable, and it seems to me desirable that it be published in full in this place. A Description of the Emery Mine op Chester, Hampden County, Mass. U. S. A., BY Charles Upuam Shepard, Massachusetts, professor of Natural History in Amherst College. (Printed by Taylor & Francis, Eed Lion Court, Fleet street, London, 1865.) Correction. The statement in this report respecting the suggestion of Dr. Lawrence Smith was based upon a misapprehension of the facts. The whole credit of the emery discovery at Chester is due to Dr. C. T. Jackson, who made a personal examination of the locality with reference to the iron ore, and from his knowledge of the minerals associated with emery inferred the existence of that mineral in this locality and advised an examination for the purpose of ascer- taining whether it did not exist there. Dr. Jackson, among other sources from whence he had derived information on the general subject, had in his possession the articles of Dr. Smith on the emery of Asia Minor and on the associated minerals, published in Vols. X and XI of the American Journal of Science. But Dr. Smith had no knowl- edge of the Chester mine or minerals until after the published reports of the discovery of the emery by Dr. Jackson. Charles Upham Shepard. Boston, November 7, 1865. 1 A Description of the Emery Mine of Chester, Hampden County, Massachusetts. 16 pages. London. THE CHESTER EMERY BED. 123 Emkry Mine. The discovery of this mine so recently as the autuma of 1864 within the bound- aries of the State of Massachusetts, where so much attention has been given to mineralogy and geology, seems somewhat singular; the more so, perhaps, as its occur- rence is so near the machine shops and armories in which the consumption of emery is very considerable. Among the reasons of its delayed discovery may be adduced its situation in a mountainous and thinly inhabited section, which until recently has attracted but little scientific or economic notice. Emery, moreover, being itself rich in irou and largely associated with magnetic iron ore (magnetite), is extremely liable to be confounded with the latter substance, and this was the case with it at Chester, whence about 1,000 tons were raised for iron making before its true nature was ascer- tained. The discovery would probably have been still longer deferred but for the happy suggestion of Prof. J. Lawrence Smith,' that the occurrence of the margarite at Chester should lead to a direct search for emery, this mineral being one of the invariable concomitants of that highly important substance. This suggestion was at once successfully acted upon by Dr. Lucas and jST. C. Sawyer, esq. The mine is situated nearly in the center of the Green Mountain chain as it traverses the western border of the State, at a point not far from halfway between the Connecticut and Hudson rivers. It is included in the metamorphic series of rocks, here consisting of vast breadths of gneiss and mica-slate, with considerable inter- polations of talcose slate and serpentine. The general direction of the stratification is N. 20° B. and S. 20° W., the relation to the horizon varying from vertical to a dip of from 75° to 80°, sometimes east, sometimes west. The immediate vicinity of the mine presents a succession of lengthened rocky swells with rather precipitons sides, having summits between 750 and 1,000 feet above the level of the principal streams by which the hills are traversed. The longer axis of the elevations generally coincides with the direction of the strata. The emery vein traverses in an unbroken line the crests of two of these adjoining mountains and scarcely deviates as a whole from the magnetic meridian. Each moun- tain is estimated to have a length of 2 miles, thus giving 4 miles extent to the metallif- erous stratum, for such it may truly be called, consisting, as it does, so largely of the metals iron and aluminium. The Westfleld River, here a small stream of about 4 rods in width, flows directly across the northern end of the vein, while a branch of the same river, having half its size, separates the two mountains and very nearly divides the vein into two equal portions. The height of each mountain is estimated at 750 feet. The emery vein, whose average width may be taken as 4 feet, is situated near the junction of the great gneiss formation constituting the western flank of the mountains, with the mica-slate forming their eastern slope. To speak more exactly, however, it lies just within the gneiss, having throughout a layer of this rock from 4 to 10 feet in thickness for its eastern wall. Nor does the mica-slate advance quite up to this outside layer of the gneiss, but, in place thereof, an extensive intrusion of 1 To Dr. Smith we are indebted for the first scientific survey of the emery mines of the Grecian Archipelago and Asia Minor. 124 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. the talcose slate occixrs, having an average thickness of 20 feet on the South Mountain and widening out on the North Mountain to a breadth of nearly 200 feet as it reaches the terminus of the vein in the bed of the Westfield Eiver. The gneiss, more especially in the vicinity of the vein, is a very peculiar rock. It abounds in thick seams of a coarsegrained very black and shining hornblende, and where this is not found it is much veined and penetrated by epidote. The stratification is much contorted also, and when the surface of the formation happens to be weathered or water-worn its basseting edges strikingly resemble in color some of the serpentine marbles. It is also noticeable that in it quartz is everywhere singularly deficient. Traces of a white calcareous spar (calcite) are now and then visible upon the joints of the gneiss, with occasional specks of yellow copper, together with malachite stains, but no corundum, emery, or magnetite particles have thus far been detected as constituents of the gneiss itself. It is quite otherwise, however, with the talcky rock exterior to the wall of gneiss, for that formation in all its different varieties of talcose slate, soapstone, chloritic aggregates (with included seam of indianite), talcky dolomite, etc., which together constitute the stratum separating the gneiss from the mica-slate, contain here and there disseminated grains of either emery, corundum, or magnetite, but, like the gneiss again, are strikingly free from quartz or uncombined silica in any of its forms. Indeed, this generally abundant substance is altogether, wanting, not only in the emery vein, but in the talcose formations consti- tuting its eastern boundary. It makes its appearance, however, in abundance in the mica-slate as soon as the talcose rocks are passed, showing itself not only as the usual constituent of the slate, but in more or less continuous seams from a few inches thick up to above 6 inches and sometimes a foot in width. Where the seams are thin and discontinuous the included masses thin out at each end before disappearing, the sharp edges being curved in opposite directions so as to form frequent white patches upon the surface of the rock in the shape of the letter S.' Corundum and emery (the former consisting of pure alumina and the latter of the same earth in combination with the protoxide of iron), have been found hitherto almost exclusively in carbonate of lime (marble or saccharoidal limestone), from the substance of which as a medium or vehicle free from silica they were precipitated in crystals, nodular masses, or veins. Here, however, carbonate of lime is wanting (if we except a partial development of impure dolomite in one place at the top of the South Mountain) ; but a parent rock or menstruum for the formation of corundum and emery is supplied in a talcose slate series equally deficient in free silica, this being a compound which, if coexistent with alumina and protoxide of iron, would seem to be incompatible with the formation of either corundum or emery, inasmuch as under the play of the ordinary chemical affinities, several very different species would be more likely to result.^ ' It is in the princij)al veins of this white quartz that very large crystals of ilmenite (washing- tonite) were found at one spot vrithin a mile from the northern end of the vein. ^An analogous abeyance of quartz characterizes the aluminous group of the spinels, the occur- rence of which is much restricted to limestone and talcose slate ; and since alumina is rarely abundant, even in granular limestone and talcose slate, we appear to have an explanation of the general scarcity of the corundom and spinel species in the mineral kingdom. THE CHESTER EMERY BED. 127 particles not exceeding in size those of certain varieties of steel or flne-grained cast iron. The fracture is effected with the greatest diHiculty, and takes place as readily in one direction as another. The surface, moreover, has an exceedingly hard feel. Its color is a darker brown than that of the preceding varieties. It also presents in some specimens a faint violet-blue tarnish. It has often been mistaken for magnetite, though its harshness of feel and verydiflacult frangibility easily distinguish its masses from that ore. It has a tendency to occur throughout the vein in wedge-shaped, con- torted masses, as well as in elliptical balls of all sizes, from a few pounds weight up to a hundred pounds. {d) Emery magnetite. — This is a massive magnetite containing a variable intermix- ture of emery. It closely resembles magnetite, but it is distinguished by its superior hardness, its purplish tarnish, and more difScult frangibility. It does not appear to be abundant, though it is often liable to be met with throughout the course of the vein. (e) Stony emery. — This in general is the chloritoidal rock or substance of the vein in places where it is not replaced by one of the preceding varieties. It is a slaty, tough, greenish-gray, rather heavy aggregate, containing everywhere flne-grained emery in proportions varying between 10 and 20 per cent. Other minerals also are present occasionally, such as tourmaline, epidote, margarite, ottrelite, magnetite, etc. A variety of stony emery in thin, highly contorted, schistose layers enveloping the compact emery frequently presents itself. Its color is a delicate greenish white, and it is often interlaminated by seams of pinkish margarite— the entire aggregate being, nevertheless, rich in emery. 2. Magnetite. ISext in abundance among the constituents of the vein stands the present species, that richest and most precious of all the ores of iron. Its composition being so strictly accordant with that of emery, their joint occurrence would, on chemical grounds, be looked for almost as a matter of course. Like the emery itself, it here occurs massive, thin-veined, granular, and disseminated. The massive variety is found perfectly pure and unmixed, having a structure between the coarse-grained (shot ore) of the Lake Champlain region and the finegrained, compact ore of Franconia and Danne- mora, Sweden. It sometimes exhibits in the fracture a slightly purplish tint not observable in any other magnetite with which I am acquainted. It is also a shade blacker than most magnetites. It is wholly free from pyrites and all traces of rust, and consequently is bright and fresh in luster throughout. It is magnetic with polar- ity, but does not give rise to examples of the native magnet. It presents itself in considerable quantity at several places on the course of the vein. For example, at the top of the North Mountain it constitutes a continuous seam from 10 to 15 inches thick in a chloritoidal vien, itself 4 or 5 feet wide, and made up of the disseminated variety, presently to be mentioned. This vein of magnetite forks off from the great emery vein on its eastern side at an angle of about 30° and then pursues its course between the talcose slate strata, within which it has already been opened for 128 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. 10 or 12 yards, having an easterly dip of 70°; and thence continues for an undeter- mined distance, the surface of the ground not yet having been cleared of loose rocks and of trees in such a manner as to define' its extent. This branch vein has already been sunk upon at this spot to a depth of 25 feet. Little or no emery is present. A mass of at least 10 tons weight of the massive variety of magnetite lies loose in the wood (with which the northern slope of the mountain is covered), 600 feet on the direct coarse of this vein and near to an opening of what seems to be its direct con- tinuation. The product of the latter excavation, however, is not the present massive variety of ore, but this again occurs in quantity at the opening quite at the summit of the South Mountain, forming a part of the great emery vein which there for a dis- tance of several rods widens out to at least 15 feet. It here occurs in a series of several more or less interrupted seams, often affording masses 6 or 8 inches thick and quite pure, though in immediate proximity to the emery. The massive magnetite passes into the thin-veined where the seams are half an inch and less in thickness, traversing the corundophilite or chloritoidal mineral, among whose particles are frequent chinks or cavities constituting a very fissile open rock. When these seams or veins become much broken or interrupted we have what may be called the disseminated variety. All three coexist in the same vein and often graduate insensibly to each other, the ore itself being in each variety entirely identical. The stony magnetite exists in other places, and is disseminated in smaller grains through a firmer gangue, identical in character with that constituting the stony emery. It is well seen at a place above referred to, where a vein has been worked upon 600 feet to the north of the summit of the IsTorth Mountain, and whence 50 tons of ore have been raised for the furnace. It here exists in the proportions of 50 to 60 per cent throughout the vein, which is 3 feet wide. As the vein is free from emery, it may prove to be a prolongation of that bearing magnetite at the top of the mountain. Dr. Jackson has detected the presence of titanium in the emery of Chester, ren- dering it probable that it will also be found in the accompanying magnetite. Should this prove to be the case, it will only be in mere traces, and will probably increase the value of the ore for iron making. It is quite certain that all the compounds, both of sulphur and phosphorus, are entirely wanting throughout the formation. ■f 3. Corundum. This species consists of the well-known pure anhydrous alumina, and is rare even in mines of emery. It is nevertheless occasionally found in those of Turkey, occurring in the form of thin seams, small grains (often of a blue color), and rarely in crystals diffused through the emery stone. It occurs rarely also at Chester, but thus far has been met with only in seams or veins one-half or three-quarters of an inch thick, though exhibiting a surface of nearly a square foot. It is grayish white, highly crys- talline, like that from the Carnatic. The seams occur in the granular and compact emery. THE OIIESTKR EMEUY BED. 125 To complete this general description of the locality it may not be deemed super- fluous to add the interesting geological fact that in two places the surface of the emery vein (near the summit of each mountain), for a distance of several rods in each case, has been deeply grooved and smoothed by glacial action. That the friction pro- ducing this effect must have been enormous is apparent from the size and depth of the channels, no less than from the initial hardness of the mineral worn away; and that it could not have been the result of running water is demonstrated by recurring to the example of river action in the Westfield River upon another portion of the same vein, where no such smoothing effect has been produced; but in place we have merely an eroded, pitted surface from which the coarse crystalline particles of the hard emery are left projecting, precisely as garnet and staurotide are seen on merely weathered faces of mica- slate. Passing now from the geological features of the region we enter upon a brief notice of the vein itself and its mineralogical contents. The principal gangue or matrix may be said to be chloritoidal. It can not properly be called chlorite slate or even chloritic trap, inasmuch as the green chloritoidal mineral it contains is considerably removed in character from the species chlorite. This opinion is based not so much upon its wanting the color and argillaceous odor of chlorite as upon the consideration that where crystallized it is found to be harder and heavier than that mineral and further differs from it by containing less magnesia and more alumina and protoxide of iron than belong to chlorite. In fact, it is much nearer to corundophilite, a mineral thus named by me from its being the almost constant attendant of corundum. It is not certain, however, that true chlorite is absolutely wanting in the vein, or at least in the contiguous talcose slate, and inasmuch as masonite and ottrelite, varieties of the species chloritoid, are often present, I shall generally speak of the gangue or vein stone as chloritoidal rock. Minerals in the Vein. 1. Emeky. Not a little confusion has hitherto prevailed as to the mineralogical and chemical nature of this substance. A common opinion has been that it is a mechanical mixture of corundum and magnetite, while some have imagined it to be a triple compound of alumina and' the two oxides of iron. Dr. Jackson, in view of his own analyses of emery, conceives it to be a combination only of alumina and the protoxide of iron.' He found — Chester. Nasos. Alumina 1. 60.4 39.5 2. 59. 05 40.95 62.3 37.7 Protoxide of iron ' See Am. Jour. Sci., 2d series, Vol. XXXIX, January, 1865. 126 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. and he suggested that it be considered a distinct species in place of being included as a variety under corundum.^ His conclusion would obviously be acquiesced in were it not for the strong resemblance in strife and cleavage between the emery and common corundum, making it impossible for us to separate the sabstances crystallographically from one another. It would, however, be singular if two minerals differing so widely in their other physical qualities should be specifically identical. Nothing like a perfect crystal of emery has yet been found at the mine, but it is quite remarkable that the mineral is here generally coarsely massive or in large separate individuals, often of the size of kernels of Indian corn (maize), whose cleav- age is perfect, and which present on their planes the delicate strife so characteristic of adamantine spar from the Oarnatic. The color, moreover, is perfectly uniform, a reddish-brown with a faint coppery luster. Its specific gravity is superior to that of corundum by nearly five-tenths, while its power of abrading, as accurately determined in the Turkish and Grecian varieties by Dr. J. L. Smith, is less than half that of the sapphire, though in all its varieties, and especially that of Chester, its scratching power or true hardness is sufiicient to scratch topaz. It is constantly magnetic. In chemical composition the Chester emery is equally uniform, and in constitu- tion may be considered strictly isomorphous with the magnetite, which species indeed coexist in the same vein or in closely contiguous offshoots therefrom. Supposing alumina to have replaced the peroxide of iron (ferrous acid), we then have instead of ferrite of iron (magnetite), the aluminate of the same base (emery), the chemical expression of the first being FeiF, that of the second Fe^. The composition of emery in 100 parts, being deduced from this formula, gives a composition scarcely different from the results of actual analysis. Its percentage of metallic iron is therefore a fraction over 29. It hence becomes apparent how natural was the mistake of regarding it as an ore of iron. Several varieties of emery at Chester, growing out of the size of particles, their mode of aggregation, and mixture of other minerals require to be pointed out. (a) Granular emery. — This occurs in flattened grains, from the size of kernels of Indian corn down to that of peppercorns, disseminated through corundophilite. The grains rarely touch each other and are distributed through the rather open green mineral, with their flat faces parallel to the foliation of the gangue. Hence this variety cleaves without difficulty into slaty fragments a few inches in thickness. It also breaks crosswise without much difficulty. The pure emery forms from one-half to three-fifths the bulk of the aggregate. (6) Veined emery. — This variety arises from the occasional contact and partial union of the individuals (by their edges mostly) of the preceding variety. The veins are much interrupted and are rarely above half an inch thick. The granular and veined varieties sometimes pass into each other. (c) Compact emery. — This variety, though not absolutely compact in the mineral- ogical sense, is nevertheless a very close, fine-grained mineral in its structure, the 1 Should it hereafter be found proper to separate emery from corundum the name of emeriie might not be an unsuitable designation for the new species. — [Shepard.] THE CHESTEE EMERY BED. 129 4. DiASPOBB. This is liydrated alumina, perhaps the most strictly characteristic accompani- ment of emery in the Grecian Archipelago and Turkey. It occurs at Chester precisely as in those regions, viz, in needle-shaped crystals and bladed masses, chiefly upon the cross joints of the emery blocks, though sometimes embedded (in compressed round masses) quite within its substance. It is generally colorless, though sometimes of a pinkish or violet tint. Perfect crystals of the usual form are not wanting where the usual open spaces exist in the inasses. Crystals also of corundophilite and very rarely of brookite are found embedded in it. 5. Margaritb. Scarcely less characteristic of emery, and also of corundum, is the present mineral, a species whose general aspect suggests that of mica, from which it difi'ers in possess- ing a greater hardness and a lower dose of silica with a corresponding increase of alumina added to an almost total absence of either of the alkalis.' The margarite presents itself frequently and with a richness of crystallization and color nowhere else known. It is always in near proximity to the purest masses of emery — sometimes traversing it in veins, at others coating, more or less perfectly, large and small rounded masses of it with layers an inch or more in thickness. The laminte of the margarite are arranged transversely in respect to the direction of the seams, i. e., they stand at right angles to the walls of the veins. Sometimes an open space exists in the middle of the margarite seam, when the mineral exhibits very rarely regularly terminated crystals with which also crystals of corundophilite are associated. Emery grains are likewise to be detected everywhere among the margarite. Its color is almost univer- sally of a pinkish tint. In a few instances, however, where it occurs in detached scales mixed up with a yellowish epidote In the massive emery, it assumes a grayish color and might be mistaken for ordinary mica, a species which I have nowhere recognized in the formation. 6. Ottrblitb (Masonitb, Chloritoid). This species belongs to the same mineralogical group as the preceding, but differs from it in many physical properties no less than in chemical composition. It is in disseminated scales of a blackish-green color, whose breadth is rarely more than a quarter of an inch. They present considerable resemblance to mica where seen on weathered surfaces of the vein or on open joints of the rock, but are easily distin- guished by their greater hardness and want of elasticity. It is chiefly confined to the stony emery. In composition it differs considerably from the margarite, having above 20 per cent of protoxide of iron, together with 6 per cent of water, and stands in relation to mica somewhat as emery does to corundum. 1 This uonalkaline feature of the contents of the vein, together also with the paucity of silica, seem to he essential conditions of all the emery veins. It is curious to remark how completely all the other micas, as well as the feldspars, with the exception of the indianite, itself not abundant, are excluded from the formation. MON XXIX 9 130 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. 7. COEUKDOPHILITE. The present mineral has not yet been fully analyzed. An imperfect examination made of a few grains of it found along with the sapphire of Buncombe, ISTorth Carolina, lead to the conclusion that, like ottrelite, it is a silicate of alumina and protoxide of iron, but with little lime and magnesia. Des Cloizeaux has described it as one of the micaceous minerals, but it rather appears to belong to the clinochlore group. Its crystallization is near to mica and its hardness is between 2 and 3. Its laminae are inelastic and almost brittle. In color and in the arrangement of its particles, even when giving rise to a slate, as it often does, it does not resemble the well-known mineral chlorite. It is of all others the most abundant gangae mineral of the purer varieties of both emery and magnetite. 8. Indianite. Exterior to its vein, on its eastern side and a few feet within the talcose slate, at a place on the South Mountain near the smaller Westfield Eiver, runs a layer or stratum from 6 inches to 2 feet in thickness, called by the workmen " the fringe rocTcP It consists of a soft, columnar mineral, but it is difQcult to say whether it is chlorite or corundophilite, or whether it may not be a mechanical mixture of the two. The columns or fibers, if such they can be called, are perpendicular to the sides of the vein and are made up of superimposed scales of the mineral. The columns have been rendered tortuous and wavy by lateral pressure. Through the middle of this stratum runs, with occasional interruptions, a vein of indianite varying from 2 to 10 inches in thickness. The mineral is massive, finely granular, of a yellowish color, and contains grains of corundum, whereby it is easily capable of scratching quartz. 9. Tourmaline. This is also a highly prevalent mineral throughout the entire course of the vein, though perhaps most abundant on the l^orth Mountain. It is more frequent near the sides of the vein, though at some places it is interlaminated through its entire mass, showing itself on the cleavage surfaces. The crystals are often several inches long and from one-iifth to three-fourths of an inch in diameter, being arranged in fascic- ular and radiating groups with their longer axes conforming to the stratification of the rock. The crystals are usually six-sided prisms with smooth surfaces, but always lacking regular terminations. Their color is brownish black. 10. Epidote. Though not abundant, it is nevertheless frequently observed, especially in the vein on both sides of the smaller Westfield River, near the mill. It is in light yellowish-green crystals, 1 or 2 inches long by one-eighth to one-fifth of an inch in diameter, the crystals being arranged parallel to the lamination of the vein and being often associated with grayish scales of margarite, ottrelite, and with emery. A beau- tiful radiated pistachio- green epidote, accompanied by diaspore, has also been observed in the same vicinity coating the cross joints of the vein rock. THE OHESTEE BMEEY BED. 131 11. Washingtonite (Ilmenite). This species is rarely met with in black foliated, much-curved laminae betwixt the double seams of margarite. On the whole, however, its occurrence is very limited compared with that in the adjoining mica-slate, to which reference has already been made. 12. Brookite. Only a few crystals of this rare titanic acid have thus far been noticed, and these were found in close connection with diaspore. 13. Chaxcopyrite (Yellow Copper Ore). But few grains of this ore have been seen. It was found, like the washingtonite, in margarite, and also upon the joints of the gneiss near the emery vein— in the latter case attended sometimes by stains of malachite. The foregoing are all the species thus far found as proper to the vein, with the exception of two apparently rare instances — one in small brown and copper-colored prisms somewhat resembling tyrite, the other in orange-colored specks (slightly decomposed) upon the joints of the emery, and sometimes disseminated through the chlorital gangue, both of which await examination. Outside of the vein with the talcose slate, besides the sparsely diffused grains of emery and magnetite, a greenish- white laminar talc in thin seams occurs sometimes, penetrated by a greenish yellow actinolite. But the most important mineral economic- ally is that modification of the talcose slate recognized under considerable variations of character as soapstone. It is here found in immense quantity at several points on the course of the vein, but nowhere, perhaps, in a more promising condition for being wrought than near the works upon the South Mountain. It here quarries with much facility in virtue of the natural joints by which it comes out in blocks of from 4 to 6 feet superficially, with a thickness of at least 1 foot, often 2 or more feet. It has the further recommendation of being free from those foreign minerals so frequently interfering with its easy division into slabs in the process of sawing. It is a point of some importance to notice the correspondence between the minerals enumerated in this paper and those described by Prof. J. Lawrence Smith in his report ^ as occurring at the Turkish and Grecian localities of emery. He concludes his account of these with the following observations : " I do not risk much in saying that the hydrate of alumina (diaspore), as well as the silicates emerylite (margarite), chloritoid, and tourmaline, and the ores of iron (magnetite) and titaniferous iron (ilmenite), wiU be found almost everywhere with the emery and corundum." It will also occur to the chemical geologist and mineralogist that we are now furnished with an explanation of the unfrequency of the corundum and spinel families of minerals, since their formation presupposes the existence of alumina, not only in excess, but attended by the absence of silica; while for the formation of emery there » See Am. Jour. Sci., 2d series, Vol. XI, January, 1851. 132 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. is demanded the same unusual chemical conditions, accompanied by the abundant presence of protoxide of iron. The contrast between the conditions of emery at the only two regions of its known occurrence grows out of the difference in the parent rock or originating formation. It is saccharoidal limestone (white marble) in Asia and talcose slate in America; for although the emery vein at Chester is situated just within the limits of the gneiss, it can scarcely be doubted, since the emery and magnetite are both found in the slate and neither in the gneiss, that these minerals originated in the former at a period, of course, when the strata were horizontal and the talcose slate was uppermost. Conclusion. The preceding statements afford the fullest view I am able to present of the emery mine at Chester as the result of several days of diligent examination of the locality and surrounding region. The vein, though fully traced and secured by its present proprietors for the distance of 4 miles, has not been opened except at comparatively few points; but the workings have been sufficiently extensive to demon- strate the perfectly inexhaustible supply of the emery, not to say also of the magnetite and soapstone. The vein, moreover, following as it does the crests of two precipitous mountains, and holding a vertical position with a width of 3 to 5 feet within well defined walls throughout, presents facilities for exploration that must be obvious to everyone. Experience has not yet been sufficient to determine what will be the cost of raising either the emery or the magnetite; but it may be stated that many hun- dred tons of both have been mined at a cost not exceeding $2 per ton, and it is confidently believed that, when the most advantageous localities for working have been decided upon, a reduction of this charge may be effected. The fact that the mine is capable of being worked for generations without incurring the smallest expense in freeing it from water is a consideration of much importance; while a further advan- tage to the property arises from the fact that the smaller Westfield Eiver bisects the vein midway of Its length where the North and South Mountains come together. Already valuable use is made of this water power in extensive mills here erected for crushing the emery and splitting the soapstone, and a further use maybe made of the same stream for the erection of iron works to any extent desired. The Albany and Boston Railroad passes within a quarter of a mile of these mills — the road from there to the depot being level, or at most having only the descent of the stream. The distance (going east) to Springfield, on the Connecticut Eiver, is 30 miles, and the road to that place is crossed at Westfield, two-thirds of the way thither, by a north-and-south road, whereby the coast at New Haven can be struck at a distance of 62 miles, thus affording by the Hudson Biver another connection with New York City, as well as the entire region of the West. Indeed, it is easy for any- one, with a map of the United States before him, to assure himself that the Chester mine is situated at the very focus of our system of railroad communication, as well as of the manufacturing industry of the country. THE CHESTER EMEEY BED. 133 The quantity of manufactured emery at preseut annually consumed in the TJnited States is over 1,000 tons, and the consumption is rapidly on the increase. This supply has hitherto been derived in part from London and partly frpm Alden's manufactory at Ashland, near Framingham, Massachusetts, which establishment has imported the crude stone direct from Smyrna, though it is now beginning to derive its material from Chester. The estimation in which the American product is held is sufficently vouched for in the following certificates, coming as they do from sources of the highest respecta- bility. The letters are addressed to J. B. Taft, esq., of Boston, who is trustee for the parties in interest to the property : Office of Master Armorer, U. S. Armory, Springfield, Massachusetts, November 16, 1S64. Dear Sue : The samples of emery sent here for trial have been tested and decided to he auper- exeellent. The test has been made without the slightest knowledge of its character by either Mr. Chamberlain or the men. The Nos. 46 and 70 are not of the right grade, but the evenness of all the numbers and their catting qualities can not be equaled by any Turkish or American emery. Yours, truly, (Signed) E. S. Axlin, M. A. John B. Taft, Esq. Boston, Massachusetts, December 1, 1864. We are very much pleased with the emery you left with us to try. We find it far superior to anything we have heretofore used. We have bought the best that we could possibly find for years, without regard to price, and we find by actual experiment that yours will do one-third more work than the best London emery. We have made a very careful experiment, and can give a more partic- ular report if necessary. If you can supply us with the fine grades to compare with the samples left, you can have all our orders as soon as we use what stock we have on hand. Very respectfully, (Signed) Hassam Bkos., Makers of Fine Cutlery and Surgical Instruments, 146 Washington Street, Boston. John B. Taft, Esq. Chicopee, April 20, 1865. Dear Sik : The two packages of emery, as samples, Nos. 46 and 70, were received and tried — the 46 on a lead wheel for cutting out grooves of blades and the 70 on a leather-covered wheel for shaping the large grooves of saber blades. Either test requires good emery to do the work successfully. We had not the means of making a comparative test, but the best thing I can say for it is that the work- men would use no other kind if they could have emery like the sample. This is of the very best quality we have had an opportunity to use. Yours, very truly, (Signed) Jas. T. Ames. J. B. Taft, Esq. Milling Shop, April 25, 1865. Sir: The following is a report on the relative qualities of American and English emery, No. 70, as tested in this department. We have given it three different trials. The number of wheels set with emery at each trial was 12, 6 with American and 6 with English. The wheels used for the first trial were reversed for second trial, those being set with English emery in second trial that were used with American in first trial, and vice versa. You will please notice the work as performed by different men with the same emery. 134 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. Testa of American and English emery. "Workmen. First trial. Second trial. Third trial. ■Wheels. Bayonets. "Wheels. Bayonets. Wheels. Bayonets. Am. Eng. Am. Eng. Am. Eng. Am. Eng. Am. Eng. Am. Eng. 2 2 2 2 2 2 38 45 24 18 24 25 2 2 2 2 2 2 23 31 33 25 52 42 2 2 2 2 2 2 35 32 68 36 47 55 Total 6 6 107 67 6 6 87 119 6 6 135 138 RECAPITULATION. Hecapitolation . American English ■Whole number of "wheels. 18 18 Bayonets polished. 328 329 Yours, respectfully, (Signed) W. G. Chamberlain, Foreman. [Indorsement.] EespectfuUy referred to Mr. Taft for his information. (Signed) T. T. S. Laidley, Major of Ordnance. U. S. Armory, Springfield MiLLiista Shop, April 27, 1865. Sir: The folio-wing is a report on the result of an experiment made in this department on American and English emery. No. 70 : Butt plates. Bayonets. No. of wheels. Tangs of butt plates polished. No. of wheels. Bayonet backs polished. No. of wheels. Bayonet fronts polished. 4 4 256 211 5 5 223 225 5 5 253 250 A quantity of Nos. 80 and 90 American emery is desirable for further experiments. ■Very respectfully, (Signed) W. G. Chamberlain, Foreman. [Indorsement.! Respectfully for-warded to Mr. Taft for his information. Several -workmen this morning asked for the American emery to put on their -wheels. This is the best indication of its quality. Please send as soon as possible some of Nos. 80 and 90. (Signed) T. T. S. Laidley, Major of Ordnance. TnB CHESTER EMEEY BED. 135 There would appear to be one quality iu the Chester emery particularly recom- mendiug- it over tlie Asiatic variety. It is this, the foreign emery is liable to oxida- tion, while the American is not, it remaining bright and clean after being moistened and exposed to air. The damage by oxidation is not only a partial loss of hardness, but the coating of the particles by rust interferes materially with their adhesion to the wheel ; consequently a wheel charged with the American emery best retains its charge and accomi)lishes the most abrasion. That this is a well-ascertained difference between the substances from the two localities is apparent from the invariably rusted appearance of the crude stone coming from the East, whereas the produce of the Chester mine and even the loose stones lying about the vicinity betray not the slightest tendency to oxidation. I am informed also by Mr. Alden, the emery manu- facturer at Pramingham, that he has long been aware of a 2 per cent gain in weight to his manufactured Naxos emery — an increase which under the circumstances may fairly be ascribed to the fixation of atmospheric oxygen in the production of iron rust. It is singular, indeed, that the same mineral, though from different localities, should not exhibit the same phenomenon when subjected to similar conditions; but numerous examples of other minerals are familiar to the mineralogists, presenting the same capricious instability of constitution. London, 21 Norfolk street, Strand, 1865. GENERAL DESCKIPTION. The Westfield River (the Agawam) runs east across the strike until, entering the northwest corner of Chester, it svpings round the north end of the broad hornblende band already described, forming the boundary between this and the Middlefield serpentine, and running south through the township nearly with the strike, it occupies a somewhat wider valley, in which is the village of Chester. This valley is excavated in the softer sericite-schists, and the greater durability of the vertical hornblende-schists (amphibolite) finds expression in the sharp ridge of the North Mountain — or Gobble Hill, as it is called with less euphony by the inhabitants — which, seen from north or south, rises like a tower and is a prominent landmark. A small brook coming in from the west in a deep, narrow valley separates it from the South Moiintain, which rises to greater height, but is more rounded and falls away southward to the common level of the high ground iu Blandford. The great height of these hills, about 750 feet above the village, 1,583 and 1,797 feet above the sea, is due, as said above, to the amphibolite band, and to the south, where this breaks up into several beds intercalated with sericite-schist, the ground falls off. 136 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. Starting at the north end of the great amphiboHte bed and at or near its junction with the sericite-schist on the east — that is, I suppose, its former upper sru-face — ^the first exposure of the emery, and the most interesting one offered for study at present, occurs in a ledge projecting into the Agawam River on its left bank near the most northerly railroad bridge over the river in Chester, north of L. Otis's house. Fig. 6 represents this reef, which projects into the river, where the wear of liigh water keeps a fresh surface suitable for study. The country rock is an epidotic amphibolite (a), contorted and thin- laminated. The tortuous lines in the drawing represent the foliation, and along the western side of the principal vein, so far as it retains its greater thickness, the laminae bend around, often quite sharply, so as to end abruptly against the emery vein, the lamination being at times continued tln-ough the "fringe rock" (c). On the west of the naiTOwed portion of the vein, as well as along the whole eastern side of the same, the lamination of the schist accommodates itself quite accu- rately to the irregular boundary of the vein. Around the smaller vein to the east the structure of the schist is still more complex, and in part, especially in the small mass which is wholly inclosed in the vein, the lamination is entirely obliterated, and filaments from the vein are spun out into the schist until they become as thin as a knife blade. A heavy vein of white quartz {d) runs parallel to the main vein in its contracted portion, at a distance from it of 1 to 2 feet, and bunches out several times to a width of a foot or more. The emery vein (&) where it comes out from the bank is scarcely a foot wide and is growing thinner. It expands northerly, at first quite '^'z^ Fig. 6. — Map of emery veins in epidote-amphibolite at nortit end of bed on the bank of the "Westfield River, Chea- ter, a, Epidote-amphibolite ; b, magnetite-emery beds ; c, biotite fringe rock; d, quartz veins; c, tourmaline. THE CHESTER EMERY BED. 137 suddenly and then more gradually, to 12 feet, and appears again in the river in an isolated rock Avith a somewhat greater width. The customary "fringe rock" (c) borders the vein on both sides from an inch to a foot wide, the width being rudely proportioned to the width of the emery vein.. It is a soft schist, made up wholly of biotite. The emery vein is a chloritic magnetite containing in abundance bronze-colored grains of emery, and, along the borders of the thicker portion of the main vein and of the eastern vein, a considerable quantity of brown- black tourmaline in delicate stellate forms (e). This extreme contortion of the amphibolite is rare in the region, and I may call to mind that, following the line of strike across the river from this point, one comes directly upon the line of junction of the serpentine (which has replaced the amphibolite) and the sericite-schist, and that the latter is also contorted to an equally extreme degree. From the outcrop upon the river bank one follows the vein southward up through a notch in the mountain, where, about 800 feet south, it has been opened and some iron ore taken out, and then up along the eastern slope of the mountain, just under the crest, to the new mine, about a mile north of the village, where alone work was in progress in 1883. The part of the vein rich in emery was about 1 to 3 feet wide where I saw it, and the corundum was regularly disseminated porphyritically in rich bronze-colored crystals 5-15""" across, affording a very rich ore. The soft, green chloritic "fringe rock" was developed in great force and cuts the emery bed in the bottom of the opening as a heavy horizontal cross- vein. It was filled with bright fresh cubes of pyrite and crystals of tourmaline 10-30'"°' long and 2-3""'" in diameter, which were all regular hexagonal prisms, with rather dull unstriated sides. They are often radiated and fasciculate. With a lens sHdes of the rock show wavy bands of a pale-salmon color, which alternate with bands and lenticular patches of bright green. The former are very fine fibrous, and show the aggregate polarization of talc. The latter is in coarser scales, often radiate, and they polarize from green to black. They have low absorption and pleochroism: x; = pale blue-green; Xi = same; a = bright yellow; extinction inclined 8° from the cleavage. Magnetite is abundant; also deep-brown grains of chromite, the former often interlaminated with chlorite. The tourmaline is in sharp hexagons; 138 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS, G? = black, £ = deep prussian-blue, with black border. Sharply refringent grains of titanite, with dark border surrounding them, occur in the chlorite. The opening was otherwise poor in the usual accessory minerals of the vein, only traces of margarite occurring. Here also a branch vein nins off to the east and the mica-schist is greatly contorted. Going south toward the old mine in the valley, one passes, about a third of the way on, the large abandoned Sackett mine, opened about 50 feet on the vein, which is here about 12 feet wide and has been extensively woi'ked for magnetite. The thin "fringe rock" can be seen attached to the hornblende-schist on the west wall of the opening over a broad surface. In the magnetite only a limited amount of emery is visible, and no other minerals could be found. Dr. Jackson says: On tlie North Mountain the emery is more largely crystalline and less mixed with magnetic iron; it is more like corundum, but still contains the combined protoxyd of iron, characteristic of true emery. Three large beds of rich magnetic iron ore, distinct from the ore accompanying the emery, occur, the ore yielding 54^ per cent of metallic iron. This ore is mined and is smelted into bar iron by forge fires, and is also sold to mix with the hematites and carbonates of iron at the Lenox and Stockbridge furnaces. The next opening on the vein is at the foot of North Mountain, north of the Becket road and directly opposite to the main works. (See fig. 7, p. 141.) Here, at the head of a small ravine, an adit was driven in 500 feet during 1893, and brought out a great quantity of the fringe rock, filled with beautiful rosettes of the hexagonal tourmaline. Of the further con- tinuation of the vein across the brook and up the north face of South Mountain Dr. Jackson writes as follows: The principal bed of emery is seen at the immediate base of the South Mountain, where it is 4 feet wide and cuts through the mountain near its summit at an angle of 70° inclination, or dip, to the eastward. Its course is N. 20° E., S, 20° W., and its known extent 4 miles. Near the summit of the mountain the bed expands to more than 10 feet in width, and in some places is even 17 feet wide. The alternations of rock in two sections are as follows, beginning to the eastward [that is, at the top J : 1. a, Mica-slate; 6, 15 feet soapstone or talcose rock; c, 2 feet crystallized talc; ^,talcose slate; e, 1 foot granular quartz ;/, chlorite slate; j/, 4 feet emery; h, chloritoid and margarite ; t, magnetic iron ore ; j, hornblende rock highly crystalline. 2. a, Mica-slate; &, 6 feet magnetic iron ore; c, talcose slate; >c>ix. The colorless })lienocrysts are small, single twins of albite, extinguishing at about 2° on either side the center. With a lens thev are seen to be frosted full of small grains of epidote, and scales, apparently of muscovite. With polarized light these make a mosaic of such brilliant color that only a trace of the polarization of the feldspar is visible. In a section parallel to oo P oc (100) a negative obtuse bisectrix was observed. The large rhombs of ankerite with rust-marked cleavage appear everywhere. Magnetite is abundant in small, rounded grains without alter- ation products. The background is a mosaic of small, short hornblende blades and magnetite, epidote, and feldspar grains. It forms most beautiful microscopic slides. Black, fissile amphiholite from Worthington, west of G. Sherman's. Nearly all a network of large blades of hornblende, with finely marked prismatic cleavage This has medium pleochroism and the formula jC = blue-green, b = yellow-green, a = yellow: c>>I»>a. Titanite and appar- ently zircon appear. The analysis of this rock was made by Mr. L. G. Eakins. Analysis of amphiholite from Worthington. SiO,. TiOj AUOs FejOa FeO- MnO, BaO. CaO MgO KsO. Na^O H2O. P2O-,. Per cent 48.53 .51 16.35 2.03 10.52 .17 trace 9.83 9.71 .32 1.36 1.7 .07 100. 19 The Heath amphiholite bed. — Across Heath from north to south, within the border of the Goshen schist, runs a band, standing nearly vertical and 168 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. 50 rods wide, of a dull, dark-gray, slightly greenish amphibolite, thin-fissile and highly porphyritic. It closely resembles the Guilford band in its wider portion across Vermont, west of Brattleboro, where the latter is porphyritic. The porphyritic spots are due to feldspar, which excludes the hornblende, but is itself full of impurities, especially biotite. The latter is wanting in the main mass. It lies near the western border of the Goshen schist and passes through the railroad cut east of Charlemont station. The upper amphibolite bed of the Hawley schist is at times porphyritic, and I have assumed that this Heath bed was a repetition of that upfolded through the Goshen schist. The small content of lime and magnesia may be thought to militate against this assignment and indicate a relationship to the Guil- ford and Whately bands, which occur in the Goshen schist far to the east, and this is perhaps the best assignment of the bed. An analysis of the rock has been made by Mr. L. G. Eakins, from a specimen taken from near W. M. Sanford's : Analysis of Heath amphibolite. The white gneiss. — A single curious band of white, thick-bedded gneiss runs down through the middle of the Hawley schist area. It passes through the southwest corner of Halifax, Vermont, and can be well studied on the high hill south of the road. It is a very prominent bed on the road going THE HAWLEY SCHIST. 1(59 u}) to the Davis mine, where, u mile behi\v the mine, a crossroad tnrns off' to the east. It has blue quartz, single twinned feldspars, and very little biotite, and is 60 feet thick. THE POSSIBLE IGNEOUS ORIGIN OF THE HAWLEY SCHIST. The theory that a ferromagnesian formation like the present may be in whole or part of igneous origin is very attractive, and I know of no sed- imentary series which could be more easily transformed l)y wholly intelli- gible metamorphic processes into the present one than the Triassic beds of tlie Holyoke range, with their interbedded traps, tuff's, and feiTuginous sandstones, to which respectively the amphibolites, chloritic schists, and fas- ciculate sericite-scMsts can be compared. There remains now no distinc- tively eruptive structure in these beds. The pseudo-por- phyritic character of the am- phibolites proves to be caused by the exclusion from the white spots of the black horn- blende needles by a second- ary feldspar growth, now more or less SaUSSUritic so T^g. 9. — Plan of altered dikes and quartz veins in clilorite-scbisfc, Cliarlemont. a, ankerite-chlorite-scliist; &, sandy muacovite-gneiss ; that the white feldspar ground 0, altered dikes, now green ankerite-chlorite-scliist; d, Mue-quartz Teins. appears. A single very strik- ing occurrence, looking like a small dike branching across the bedding, occurs in Charlemont (fig. 9), going up over the bare rocks west of A. P. Maxwell's (now Vincent's), a mile north of the village, to the highest rocky bluff visible in the woods from the house. On the east is the common soft ankerite-chloi'ite-schist (a), and a sharp boundary line sepa- rates this from a white feldspathic muscovite-schist or gneiss of sandy texture (&) Distinct dikes (c) of ankerite-chlorite-scliist of slightly dif- ferent texture from the country rock (a) appear in the latter and run out into the white gneiss, branching and expanding into irregular forms. The country rock is distinctly faulted by the dike, and a later fault cuts across the whole and throws it, and this is filled with vein quartz (d). There is a distinct foliation in the dike, which is in part parallel to that of the country rock and in part divergent therefrom, as indicated in the figure. 170 GEOLOGY OF OLD HAMPSHIRE COUJirTY, MASS. MINERAL DEPOSITS. THE PYRITE BEDS. Toward the north, to some extent in Hawley, but more abundantly north of the Deerfield River, lenticular beds and impregnations of pyrite occur, carrying a small percentage of chalcopyrite. The success of the Davis mine in Rowe has greatly stimulated the search for similar deposits and their exploitation. Much time and money seem to have been fruit- lessly expended, and often by those who could ill afford it, without, so far as I could learn, consultation with any person competent to advise upon such matters. Many considerable beds have been opened, but none have proved remunerative except the Davis mine, near the east line of Rowe. This is a great lens, lyiug with sericite-schist as its western (foot) wall, and on the east chlorite-schist, which is soon replaced by sericite-schist. The strike is N. 30° E., the dip 70° E. The bed has been opened for 700 feet on the strike and to a depth of 425 feet. On the north it swells in a short dis- tance from 6 feet to 24 feet and maintains this thickness nearly to the south end. It is an enormous mass of almost pure, coarse-granular, shining yel- low pyrite, with some chalcopyrite, blende, garnets, and gahnite — the two latter fine and rare. I am indebted to the proprietor, Mr. H. J. Davis, for the facts concerning the workings of the mine up to the summer of 1 892. It is supplied with the most approved appliances of every kind. A little village has sprung up in this desolate corner of a decaying town, and much taste and energy were manifest everywhere. A new post-office (Davis) has been created, and surveys were being made for a narrow-gauge road from the mine to Charlemont. Everything thus indicates that the mine is suc- cessful to a degree exceptional in the history of mining in Massachusetts. In 1892 the greatest depth of No 1 shaft Avas 601 feet; greatest length of adits, 998 feet; total depth of all shafts, 621 feet; whole length of adits, 5,989 feet; greatest width of the vein, 61 feet. Total product of mine to January 1, 1892, 334,552 tons Considerable work has also been done on Rice's brook, a mile above Charlemont village, and an engine has been set up, but the euterj^rise did not prove remunerative. The most abundant and promising deposit after the Davis mine is in RHODONITE, KHODOCHKOSITB, AND COTICULE. 171 tlic liiji'li l)liiffs overlooking the road east of M. V. Cressy's "second pasture," w lic'i-c t'oi- ;i lon^- distance along the strike, in an area 20 rods in width, the chlorite-schist is crowded with pyrite in large, rougli-faced cubes two-thirds (if an inch across. One layer nearly a foot thick has been opened. COPPER ORES. The pyrite beds usually carry a small percentage of chalcopyrite. Copper is said to have been mined in a small way northwest of M. Stetson's and northeast of P. Packard's, in Plainfield. In an opening made b^• jM. V. Cressy in the pasture north of H. Baker's, where the sericite- schist is mucli impregnated with granitic material, bornite is quite abun- dantly disseminated in small grains, partly changed to malachite; and farther north in Charlemont, back of the house of G. Veber, on a blind road running north from the river road, bornite appears in masses an inch across. THE GREAT HAWLEY FAUI.T AND THE MAGNETITE AND HEMATITE DEPOSITS, THE RHODONITE AND RHODOCHROSITE BEDS, AND THE GARNET-SCHIST OR COTICULE. The mineral rhodonite has been found for many years in large bowlders in Cummington, near the Bryant homestead, and it has, in fact, been called cummingtonite from that circumstance. Withui a year or two state- ments have appeared in print to the effect that the ledge from which the bowlders had been derived had been found on the Bryant road in Cum- mington. On investigating the trenches I found that they did not reach any ledge there at all, and taking the direction of the glacial striae I soon found the ledge 2^ miles distant to the northwest, in Plainfield. On the road running north, up the hill from West Cummington, at the house of T." Williams, now occupied by Henry C. Packard, about 35 rods south of the house and 10 rods west, a garnetiferous sericite-schist (Savoy schist) on the west abuts against the chlorite-ankerite-schist on the east, and a band 8 rods wide is filled with veins of quartz, quartz and magnetite, and quartz and rhodonite. Some of the latter are 3 feet wide, of the finest and deepest color, often blackened at the surface. All the varieties found in the bowlders on the Bryant road in Cummington can be found in place here. The line between the two has the direction of the glacial striae of the region, and this locality is doubtless the source of all the Cummington 172 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. rhodonite. Much mming was done here in 1848, but, it was abandoned because of the California gold excitement. By following the road north a mile and a half to a point where it turns sharp east, measuring 60 rods south from this bend, and going 10 rods east into the open field, one comes to another opening on a vein or group of veins 10 feet wide, mostly quartz- rhodonite veins, the unaltered rock faintly pink and the whole greatly blackened by oxidation. These two openings are called, locally, the man- ganese mines. They mark the line of a great fault, which runs south through the area of iron-manganese in Hawley. An inspection of the map will show that the amphibolite bands coming south are cut off with an acute angle against this fault, and the contrast of the highly ferruginous ankerite- chlorite-schists (Hawley schist) on the east and the barren quartzose sericite-schists (Savoy schist) is everywhere very striking. Across Forge Hill, in West Hawley, this contact line bends consider- ably to the east, but the crushed band, largely filled with quartz veins, more or less ferruginous, is so wide here . that I have represented the state of things by doubling the fault line across this area. I was guided the whole length of the iron deposits on Forge Hill, south of the old Hawley mine, by Mr. M. V. Cressy, who owns most of the land and has examined the country for iron more carefully than anyone else. At the most southern opening marked on the map, and the one where the dipping needle was most affected, the schist was impregnated with magnetite for a thickness of 12 feet in the digging, and about 2 feet of this would j)ass as a lean ore. From this point the vein or veins can be followed north for a long distance, and opposite the south end of the amphibolite band and in the line of the straight fault marked on the map considerable digging has been done and the magnetite, here exceptionally abundant, is accompanied by much flesh- colored quartzite, apparently colored by rhodonite and rhodochrosite. The schist is full of magnetite for many rods to the east, and a well-marked hematite vein occurs here, with the quartz-rhodonite mixture accompanying it. About 10 rods south of this the epidotic amphibolite comes to an end and the ankerite-chlorite-schist abuts against the quartz-sericite-schist. The vein can be followed north by disseminated ore to the Cressy "second pasture," a mile south of the old mine. Here a deep shaft has been sunk on the vein at the junction of the two rocks and masses of pure magnetite were lying at its mouth, and the accompanying vein quartz here and along RHODONITE, RIIODOGHROSITB, AND COTICULE. 173 a liiu! a tew feet east was pink from the intermixture of rhodonite and rhodochrosite. Ihneuite and fluorite also occur here in quartz veins near the ])oint wliere, a few rods west up the hillside, a new shaft has been sunk and has exposed much beautiful corrugated liematite-schist. A half mile farther north on the vein, at Mr. Cressy's "first pasture," the excavations have so exposed the vein as to throw much light upon its character and origin. Tlie following section is exposed, commencing on the east: Section in West Hawley. 1. Dark-green ankerite-clilorite-schist (Hawley schist). 2. Compact vein quartz, tinted iiesli-color from rhodonite, or rusted to deep brown and black; vertically color-banded, and with very regular rhombic jointing, 3 feet. 3. Solid, rusty, granular magnetite, 3-6 inches. 4. Granular quartz full of white quartz veins, carrying ilmenite, which seems to be the sericite-schist crushed, deprived of its mica, and recemented, 18 feet. 5. Sericite-schist full of scales of hematite replacing the mica and fine-granular rhodonite, 4 feet. 6. Sericite-schist (Savoy schist). Everything indicates the crushing of a wide body of the rock and the deposition of magnetite along the main fissure, while manganese was carried far out from the fissure into the crushed rock, with much vein quartz and ilmenite; and still farther away from the main fissure, and perhaps at a lower temperature, hematite replaced the mica of the schist. The old Hawley mine, a half mile farther north, is sunk on a vein miming N. 10° E. and dipping about 80° E. This vein is situated in the sericite-schist, 5 rods west of the junction of this schist and the ankerite- chlorite-schist, from whose abundant store of iron the vein was doubtless filled when the mountain-making movements opened the fissures and stimulated the chemical activity of the heated waters. The mine was much worked many years ago, and though the shaft was filled, the deep open workings have remained open and have furnished the many elegant specimens of the "micaceous iron" from Hawley found in all cabinets. This is the ore in the upper portion of the veiii, and has been opened for 80 or 90 feet south from the shaft, which is 50 feet in depth, with a maximum thickness under 2 feet. A cross-cut of 10 feet cuts two more narrow veins. It is a well-foliated rock, generally finely corrugated, and 174 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS, made up almost entirely of small, shining scales of hematite. 1 think it is a pseudomorph after the sericite-schist in which it is intercalated. Below, the vein changes into a firm, compact magnetite, which is often interlaminated with the quartz-rhodonite mixture. The mine has been recently (1891) opened by Mr. Ferd. S. Ruttmann with improved appliances. The flesh-colored quartzite, plainly colored by manganese, which occurs as a veinstone on Forge Hill and south all along the vein, has the aspect of a common granular quartzite, but under the micro- scope its appearance is very striking. The slide looks exactly as if a layer of half-dried blood cor- puscles were spread upon it. They average .05""° across, but vary from half to double this size. They are flattened and circular, but not always entirely regular, have darker centers, and are of reddish color. They are' completely apolar, but have at times a polarizing grain at center. Some of them are cbm- \ plete trapezohedra, and they are manganesian j garnets. A few grains of quartz and a rare scale \ of chlorite complete the slide. This description is 1 of the rock at Forge Hill. \ From the mine on Cressy's land, on the south, « the arrains are much smaller, .016™™. There is a \ little more quartz, otherwise they are the same. \ The rock is thus a coticule or quartz-garnet rock, i tinted flesh-color by manganese, and is a product \ of the same heated waters which have filled the \ vein with hematite. ^ It is at times marvelously contorted, as illus- trated by fig 10, taken from a band at the mine on Forge Hill. This in appearance is a fine-grained, pink sandstone. The folds are so nearly sheared apart that a piece falls asunder easily in coarse bars, and the whole surface is frosted over with specular iron. Few traces of the iron can be found farther north, but south Ir>a; jc = green, lj = olive, a = yellow. The large white porphyritic spots, 2-3"™ across, are so loaded with opaque white dust, muscovite scales, etc., that it is generally only possible to make out a mosaic of untwinned feldspar and quartz grains, and, in the absence of cleavage and twinning, to make sure that the mineral is in part biaxial. In one large grain, cut parallel to M (010), an optical axis emerged at the lower left-hand border, indicating anorthite, and where twinning occurred the extinction angle was very large, giving the same indication. (d) Bim of a similar ^^ anvil" from Plainfield. (In the collection of Amherst College. See PI. V, fig. 1, p. 302, for section.) In the matted, green, fibrous hornblende, greatly darkened by rust and coal dust, are many scales of a greenish mica, garnets with the same radial inclusions as in the West Chesterfield schist (p. 182), curious long red prisms of rutile, matted fine white needles with longitudinal extinction, apparently zoisite, and a fine plagioclase, extinction 26°, loaded with coal dust, but with clear border. In other cases sections cut at right angles to both cleavages gave an extinction of 38° to 45°, indicating a very basic feldspar It is significant, as connecting these beds with the porphyritic amphibo- lites, that rounded clear spots of impure plagioclase appear, from which all the dark constituents are excluded. The slides of black hornblende-schist or amphibolite last described, cut from the thin plates of the rock which borders the limestone beds, and which have manifestly been de^'ived from the limestone, furnish abundant proof that some amphibolite beds may originate from limestone. The thin beds of amphibolite of exactly similar habit with the above and found in the Conway schists have clearly the same origin, the change having reached the center of the former limestone from each side. These beds have commonly a thickness of 6 inches to 1 foot. (e) The amphibolite at the brook crossing in Whately. (See PI. V, fig. 3, THE CONWAY SCHISTS. 193 p. 302, for section.) The amphibolite at the brook crossing is thrust forcibly up tlu'ough the ai'gilUte, together with the black Conway limestone, as described on page 196. It shows in many ways a transition between the uaiTOW bauds of hornblendic rock which form selvages to the limestone beds, as described above, and the larger bed which is the subject of the next chapter. It is a dull, dark-green, massive rock, which shows with a lens the usual interlacing network of actinolite blades, with rare open white spots composed of a granular feldspar, much changed to mica The feldspar is optically positive and has extinction +6^° on M (010), and so is an oligoclase. The hornblende is peculiar in two ways. It has a brown center and grades through green to colorless at the ends of the blades The brown, is like the cummingtonite found in the Conway schists farther west. It has low absorption colors. The brown shows c = greenish brown, h = red- brown, a= pale brown; the gi-een, ,c = blue-green, h = pale green, a = pale yellow: c>lj = a. The blades are fibrous and often twinned, and give extinction 14° to 17°. The second peculiarity of the hornblende blades is that the brown centers often show dark-brown bands situated in the basal parting and sending out long, straight needles in both directions parallel to the vertical axis, which makes them look like combs with teeth directed both ways. These straight needles are also abundant everywhere in the hornblende and in the feldspars and seem to be rutile. A "hof" surrounds the larger comblike accumulations and dims their outline Other hornblendes are built up around red biotites filled with coaly matter. A few grains of calcite occur, and the black ore grains show no trace of leucoxene. This agrees so nearly with the calcite-derived amphibolite described above that one must assign to it the same origin. Its close association with the limestone strengthens this conclusion. It is, however, not cer- tain that this bed is part of the large bed next described, though highly probable. (/) The great Whately ampMholite. This bed, which extends as a broad band across Whately and Williamsburg, is for the most part a very fine- grained, black, fissile rock, and in sections cut from the north end of the bed the hornblende is present in a network of long blades with strong MON XXIX 13 194 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. absorption and pleochroism ; extinction, 16° 30'. These lie in a mosaic of untwinned feldspar grains. Menaccanite and leucoxene are abundant, and in slides from the south end of the bed in Williamsburg (south of P. M. Gillett's) each grain of menaccanite is surrounded by a single crystal of leucoxene, and these often have the regular wedge-shape of titamte. East of C. Bardwell's the rock abounds with white spots which prove to be a quartz-feldspar mosaic, greatly crowded with many impurities, but with clear borders. The hornblende is in sheafs and bundles of fine fibers, which partly coalesce into stout crystals, so that the transverse parting runs across the bundle and the center polarizes as a single individual. The cen- tral portion of these large crystals is full of coaly particles. Pleochroism and absorption are weak. Biotite and rarely a congeries of grains of leucoxene occur. This occurrence agrees clearly with the hornblende- schist derived from the calcite beds described above, and I assign this origih to all the hornblendic beds in the Conway schist, particularly as limestone is abundant and all other traces of basic eruptives are wanting. All these rocks share with the accompanying micaceous schist the pecul- iarity that the centers of the larger phenocrysts are full of coaly matter or fine quartz grains, indicating that both have together passed through two stages in the metamorphic process. This peculiarity is wanting in the similar amphibolites of the Bernards- ton series (see p. 291), with which I would compare these rocks. The latter series, though of later age geologically, is more metamorphosed and differs in the more abundant development of the clear mosaic of untwinned plagioclase, but in no other way. They have the same field relations, the amphibolite being always interbedded in the schists. They have the same abundant actinolitic hornblende, biotite, ilmenite with leucoxene, rutile with dark border, and basic plagioclase, and range from massive to slaty varieties. At Mrs. M. Taylor's, in Whately, the rock is fine-grained and thin-fissile. Its long, thin hornblende needles have low absorption and pleochroism, and lie in a feldspar mosaic. A great number of titanite grains inclose one or more grains of black ore. For section, see PI. VI, fig. 1, p. 306. (^) The Whitmores Ferry amphibolite. At Whitmores Ferry, in North Sunderland, in the midst of the Triassic shales, arises an outcrop of a dark THE CONWAY SCHISTS. 195 amphibolite, easily mistaken for the bituminous shale, and exactly resem- bling the above amphibolite, especially in having the highly refringent grains, each inclosing one or more rounded grains of a dark ore, which is here slightly brownish. The agreement, macroscopic and microscopic, is so peifect, and the rock is chemically so nearly identical with the Guilford band of amphibolite in the Conway schist farther north, as shown by the analyses below, that one can not doubt that the Whately amphibolite is continued in the line of its strike northeast beneath the sandstone to the Whitmores Feny outcrop.-' The association with whetstone-schist in both places strengihens the probability of their identity. The agreement of the three analyses given below is sufficiently close to strengthen the opinion maintained above that all these hornblendic rocks have been derived from limestone beds. (li) The Guilford band. (For section, see PI. VI, fig. 2, p. 306). An inspection of the map (PI. XXXIV) will show this extensive bed, beginning in Conway and running continuously across the northern half of the State and widening suddenly as it crosses into Vermont. Its stratigraphical relations do not preclude the supposition that it may be a great dike. Lith- ologically it is like many beds of the Chester and Hawley series. Its cleavage surfaces show many black hornblende needles of high luster in a mat of finer needles. Its hornblende crystals are not filled with grains of coal or earlier constituents, as are the other Conway beds, and also the Bernardston beds. The most marked peculiarity of this band is the presence in the slide of many deep red-brown rutiles clustering around black ore grains. The feldspar seems to be albite, with which the content of soda agrees. ANATYSES OP THE AMPHIBOLITES. I. Gruilford, Vermont. Shining-black, thin-bedded amphibolite. II. Whitmores Ferry, Sunderland. Outcrop in the midst of the Trias. Very fine-gramed, dull-black, shaly amphibolite. III. Groshen. Base of largest "anvil," from which slides described above (p. 191) were taken. A quartz-hornblende rock, formed by the alter- ation of the limestone by reaction of solutions derived from the inclosing schists. For general discussion of analyses see page 300. 1 See p. 361. 196 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. Analyses of the amphiboUtes, by L. G. Eahins. SiOj TiO, AI2O, Cr.O, Fe^Oj FeO- MnO BaO CaO. MgO K,0. Na.O H2O. PsOs 49.16 1.03 16.43 trace 3.92 7.19 .23 .02 9.21 8.19 .41 3.70 .45 .16 II. 49.86 1.58 15.50 100. 10 2.99 8.01 .07 trace 8.89 7.79 .72 3.26 1.51 .11 III. 55.64 .50 16.27 1.22 7.20 .28 100. 29 9.23 5.58 .19 .91 3.11 .23 100.36 PROJECTION OF THE LIMESTONE AND AMPHIBOLITE OF THE COKWAY SCHIST THROUGH THE LEYDEN ARGILLITE IN WHATELY. Following the road west from the hotel in "Whately, one comes in a few steps upon a bridge over a small brook, and to the north across the brook a fresh sui-face has been exposed in the bluff by blasting. An inspec- tion of the wall reveals small spots of pyrite as the probable cause of the blasting, and, what is of greater interest, one soon finds that a small boss of the black limestone and the amphibolite of the Conway series, both of which are in place a considerable distance to the west, has been here thrust up through the argillite with great force. The argillite dips away from the limestone on both sides and mantles round its end, as shown in fig. 11. A few rods up the brook, on the other side of the road, several similar bucklings of the limestone and hornblende rock up through the argillite may be seen. This shows, of course, that the Conway mica-schists are carried far beneath the argillite and thus are older than it. The amphib- olite and the limestone are identical with those farther west in the Conway schist and are described above. As an indication of the force with which the limestone was thrust up through the newer rock, there follows a THE CONWAY SCHISTS. 197 desci-iption of a vein of hard, vitreous quartz 7 inches iu width, which ruus across the face of the limestone; it is represented in fig. 12, p. 198. The laro-e vein is twisted and the limestone is kneaded into the com- pact quartz and drawn out into long filaments carried down into the center of the vein and pinched oft' in it, and the smaller veins are contorted still more remarkably. The limestone has received a marked fluidal stracture in the apophyses, which penetrate the quartz throughout its whole mass in curving bands, which fit themselves with more or less success to the complex surface of the vein. Under the micro- scope the limestone shows all stages in the development of a cleavage by slipping caused by the pressure (Ausweichungsclivage of Heim.^) . Portions of a thin section cut at right angles to the cleavage plane break up into a series of very long, thin wedges, placed with their cutting edges pointing alternately in opposite directions. Each wedge shows a fluidal structure, expressed Fio- H— Map showing tlie protrusion of the limestone of the Conway schist through the Leyilen argillite. "Whately. by the bending of the lines of coal particles toward its head. This slipping of the wedges alternately to right and left concentrates the coaly particles somewhat along the boundaries of the wedges, by which they themselves become more distinctly defined, and at last confluent into a new plane, marked at once by a cleavage and a color banding. CONTACT METAMOEPHISM OF THE LIMESTONE BY GRANITITE. ARGENTINE. A very interesting exposure occurs on the river road from Leeds to Haydenville, near the junction of the biotite-granite (granitite) and the muscovite-granite. The former is very confusedly melted into the Arg////fe. Hornblende Schist Limestone. Covered. 'Untersuchungen uber den Medianismus der Gebirgsbildung, Basel, 1878. 198 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. remnants of mica-schist which occur as inclusions in the great granite mass, and at one point appears a grayish-white, massive, fine-grained rock which proves under the microscope to be a labradorite-pyroxene-calcite rock. Treated with acid, it leaves a glassy, friable mass, in which scales of graphite and needles of bright-green actinolite are visible. Under the microscope it shows an abundance of calcite, multiple twinned; labradorite, extinction 14°, often doubly twinned; and large color- I"IQ. 12.— Surface of black limestone with contorted quartz veins. Whately. Scale, ^. less pieces of pyroxene, extinction 41°, inclosing many grains of the other constituents. Rounded grains of titanite occur. This may be referred to the graphitic limestone of the Conway mica-schist altered by the granite, and this, in connection with the long distance across the granite area that one can follow the hornblende-schist (see p 190), leads one to conclude that the mica-schist fragments in the granitic area in Williamsburg are also remnants of the Conway schist, and that the "argentine" occurrence in the midst of TUE CONWAY SCHISTS. 199 the granite still farther south, on the Westhampton Hue, is another remnant of the same limestone from the Conway schist which formerly mantled over the grauite.^ The inclusion is still partly micaceous limestone. CLEAVAGE IN THE CONWAY SCHISTS. In the flags, or Goshen schists, the original lamination seems to be gen- erally preserved. In the Conway schists a distinction can be made between the eastern half of the schists in the granitic area, where the impregnation of granite and quartz and the great contortion leave one at times in doubt as to the origin of the foliated structure, and the western or lower portion, where the fine crenulation or corrugation produces a ligniform structure in which strike remains distinct but dip becomes quite uncertain. Without searching far one can generally find a banding of coarser and finer material — a bed of limestone or whetstone-schist — and then generally will find the foliation to agree with the original lamination. This is beau- tifully seen at the dam in Huntington village. Standing at a distance, the laminae, from 2 to 14 inches in width (average 6 inches), are each bounded by a black band at the bottom, 2 to 3 inches wide, which shades off" above into the lighter portion, the whole making exactly the impression of a lam- inated sandstone, the lower part of each being fine-grained and clayey, the upper part coarse and sandy. On inspection the lower portion is found to be dark from the abundance of garnet, biotite, staurolite, and cyanite, while the light portion is sandy and contains only scattered garnets. What seemed at a distance to be true was doubtless once true, and the lower portion of each layer, being argillaceous, has given rise to the alumi- nous minerals wanting in the sandy portion of the layer. A part of the dark color also depends upon the fact that the new-formed minerals have often inclosed much coaly matter that might otherwise have been carried off. At other places precisely the same structure enables one to detect a well-developed cleavage. This is the case along the western of the two roads going south from Chesterfield Center, and on the east-west road a mile south of the village. This is finely illustrated also along the east side of the road going south from Stevens's mills, in Worthington, in a field abounding in most beau- tiful roches moutonndes. The rock is a dark, corrugated mica-schist. The ' See "Argentine" in Mineralogioal Lexicon : Bull. U. S. Geol. Survey No. 126, 1895, p. 43. 200 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. lamination is marked by alternations in color, in bands 1 to 6 inches tliick, exactly as described above. A dike of tourmaline-granite coincides in position with this structure ; strike N., dip 70°-80° E. The cleavage is indicated by a bedding which comes out by weather- ing, and along which alone the rock splits easily, and membranes of mica are developed. This strikes N. 40° E. and dips 25°-30 W. There is in this last structure a cleavage changing into a foliation. In general cleavage is subordinate in these schists, and usually where it occurs the strike of the primary and secondary structures very nearly coincide. FOSSILS (?) OF THE CONWAY SCHISTS. In many places cavities coated with rust are found in the quartzose and slightly calciferous beds in the schists, which I have no doubt represent fossils, but which, in every case that has come to my knowledge, are so poorly preserved that it is possible to explain them as due to the removal by solution of some mineral, possibly calcite. The mode of occurrence suggests, however, that a large number of small, flat bivalve shells, 5-25°"" long, were deposited, all lying flatwise and about equidistant in the sands which have now become the whetstone-schists. In a bowlder found on the railroad in Worthington these cavities were flattened, nearly round, 5-7°"" long. West of E. B. Drake's, in the northwest part of Chesterfield, the cavities are about 15-25™" long, flattened oval, and in many cases two such impres- sions lie side by side joined by a straight line, strongly suggesting the opened valves of a leperditia like L. haltica. They are flat, rust-covered cavities, and in one fresher part of the rock are represented by darker spots, all arranged parallel to the bedding plane of the rock, and having the same shape as the cavities. These dark-gray spots seem to be only spots in the sandstone. They effervesce much more abundantly than the rest of the rock, and seem to be flattened concretionary patches of a calcite darkened by carbon. At the Clarke tourmaline ledge an exactly similar occurrence is found, only the cavities are a little larger. At B. Shaw's whetstone quarry, in Cummington, is a bed of the whet- stone about IS""" thick, full of closely approximated tubular cavities 2-3"" THE LEYDEN ARGILLITB. 201 in diameter and parallel, which suggest scolithus, but which are parallel to the bedding. All these specimens were submitted to Mr. C. D. Walcott and other paleontologists, but they could not decide that any of them were certainly ot" organic origin. THE LEYDEN ARGILLITE. DESCRIPTION. The rock is in its whole extent of uniform texture and structure — a dark-gray and very fine-grained slate with glistening cleavage surfaces, dull-black when broken across the ends, and generally crumpled and corru- gated to the extreme of complexity. It is remai-kably barren of all acces- sory minerals, and this has been taken as a characteristic to distinguish it from the Conway schists, though in places small garnets and biotite scales are scattered sparingly over its cleavage surfaces. Slaty cleavage is devel- oped in it in every degree. Thin sandy layers often show the original bedding after the rock has been crumpled up into shai-p folds and the cleavage perfectly developed outside these layers, and the rock can still be separated along these into thick plates fluted and folded in the sharpest curves, and at the ends of the plates the slaty cleavage is seen to cut across the slab and to divide it into thin, flat laminae regardless of its convolutions. Moreover, the importance of the shearing force in the development of cleav- age can often be beautifully seen, the fine, close-set and equidistant corru- gations becoming sharper and changing from folds into faults, and the elements between these faults being flattened out, with some degree of flow of the material, into the cleavage plates. A remarkable block found (not in place) at the outcrop nearest and to the west of the lower quartzite of the Williams farm in Bernardston may find mention here. A mass of chlorite-slate 3 inches wide cuts across the argillite exactly like a dike. It is bounded by parallel planes and is sharply demarcated from the argillite, and while both are cleaved perfectly at right angles to the plane of the dike, the plane of cleavage in the green schist makes a small angle with that of the argillite. One can not well avoid surmising that a small diabase dike has been here very curiously metamor- phosed, but the microscope shows only matted chlorite and muscovite scales, quartz, and geniculate rutiles, the latter visible also with the lens. 202 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. The argillite is characterized by a great abundance of quartz nodules and bosses, often of great size, which, though not wanting in the Bernards- ton series, are there comparatively unimportant. On the road from Greenfield to Charlemont, above Fall River bridge, the argillite is a fine-cleaved roofing slate for a long distance by the road- side, almost as fine a slate as that at the Gruilford quarries in the town next north in Vennont. QUARTZITE IN THE AE6ILLITE. On the road north from Bernardston, at C. Cushmore's, is a heavy layer of a dark, thick-bedded quartzite about 33 feet thick, and a little far- ther north, at I. K. Brown's, is a crumpled, thin-bedded quartzite. Just over the State line to the north, near the Guilford slate quarries, the argillite is replaced by a fine-grained quartzite, which President Hitch- cock called a novaculite-schist and found to be a quarter of a mile thick.^ PETROGRAPHICAL DESCRIPTION. The mass of rock is made up of minute, elongate, brightly polarizing muscovite microlites, often raveled out at the ends and with wavy sides, in an amorphous background. Clay-slate needles are only doubtfully present. Stout elongate forms, opaque by transmitted and curdled white by reflected Kght, seem to be leucoxene derived from menaccanite. Magnetite and calcite are wanting. There is much coaly matter in swarms of black dots, and rarel}^ a biotite scale placed in the plane of cleavage. Often a strongly marked pseudo-fluidal structure, expressed by the position of the elongate muscovite crystals, indicates clearly the mode in which pressure has produced this cleavage. Microscopically the rock is thus a very fine-grained, argillitic mica- schist or phyllite, and it differs much from the true argillites, e. g., the cleaved slates of Snow den, Wales, or the slates of Hoosick Falls, New York, with which I have compared it. I have followed custom in applying the name argillite to the band of rock, somewhat in a geological sense. The rock sometimes contains small garnets in considerable number, and these are often changed wholly or partly into small white balls of kaolin, or kaolin and hematite. The kaolin was infusible and gave blue color with 1 Vermont Report, Vol. I, 1861, p. 490. THE LEYDEN ARCilLLITE. 203 cobalt, and no reaction for potash. The mass of kaolin does not quite fill the cavities. These occur north of the Devonian limestone in Bernardston. STRATIGRAPHY. The rock is crushed into sharp folds and finely coiTugated, and where sandy layers are wanting the primary structure may be replaced wholly by the cleavage; in other places it is brecciated and thrown into con- fusion. Everywhere the strike and dip vary suddenly and within wide limits. About N. 20° E. may be the average strike, and 60° E. the average dip. A comparison of hand specimens, and especially of sections of the three rocks, shows that the "argillite," while a distinct mica-schist, is far less thoroughly metamorphosed than the schists in the Helderberg series, and from this criterion alone one would consider it the newest rock in the whole ai'ea. That it is newer than the mica-schists to the west and older than the Helderberg series seems to me in the highest degree probable, and also that the two older groups are Paleozoic; but I can find no very con- vincing ground for their assignment to a definite horizon in the Paleozoic. BOUNDARY ON THE CONWAY SCHISTS. At Beaver Meadow, in the northeast corner of Leyden, one finds the point of contact just at the foot of the mill dam. The black, baiTen argil- lite has strike north to south, dip 70°-80° E., all the way up from Fall River, a mile east, and often shows true cleavage. Here several thin, rusty beds appear, and quite suddenly the rock becomes slightly coarser and full of very small spangles and transverse crystals of biotite ; and three thin beds of black limestone occur in quick succession. The boundary is best drawn at the first bed of limestone, just at the dam, but for 300 feet below the rock is black, fine-grained, finely double-corrugated, and difi'ers mainly in the minute mica spangling from the argillite lower down the brook, and for a little way above this limestone much of the rock can scarcely be distinguished from the argillite. It is, however, a little coarser, rusty on cleavage faces, and spangled on transverse fractures. It is thus a rather gradual transition, and President Hitchcock was often in dotibt about the existence of any boundary whatever. Exactly the same transition occurs between the two beds at all places 204 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. where the boundary can be studied. At the base of the argilUte one finds minute spangles of mica; a few feet below comes limestone, and then the rock quickly grows coarser, mica-spangled, and garnetiferous. This is well seen toward the south end of the boundary line, in the southeast corner of Coleraine, near the house of D. Nelson. In the Whately area the transition is almost exactly the same, but a heavy band of white quartz marks for a long distance the exact boundary, and there is probably a fault there. I have found nothing along this bound- ary which would suggest the existence of unconformity between the two beds. argillite in the western boeder of the "graphitic mica-schist " (goshen schist). An inspection of the map of the Vermont survey of 1861 shows a broad band of argillite, bordered on the west by Devonian limestone, extending south from Lake Memphremagog, and thinning south and disappearing midway the State. Along the west border of the Groshen schists, where they enter Massa- chusetts, in Heath, is a band of thin, black slate that looks exactly like the metamorphosed Carboniferous slate from Worcester, and which seems to be the continuation of the Memphremagog slate. Farther south it is indistin- guishable from the ordinary Goshen schists, except across Worthington, where a band, 50 rods wide at the base of the schists, is a fine-grained, barren, flat-fissile schist, unlike the garnetiferous schist above and the horn- blende-schist below. These beds are described in some detail in following down the western border of the Groshen schist (see page 179). I have treated them as the base of the Goshen schists, and think this the most probable view. There is no satisfactory reason for identifying the two argillites. The western seems inconstant, and does not appear in the Goshen anticline. relative age op the CONWAY SCHIST AND THE LEYDEN ARGILLITE. An examination of the comparative sections on page 258 will show that the first discrepancy of importance there indicated is in regard to the relative positions of these two series, the argillite being regarded as the lower and assigned to the Huronian by Professor Hitchcock. THE LEYDEN ARGILLITE. 205 The Leyden argillite, as it runs north through Vermont, borders and everywhere rests upon the Conway schist, where they are not vertical.^ In the discussion of" the argilHte in the Vermont survey it is placed, without hesitation, above the schist; indeed, is still associated with the limestone and assigned to the Devonian.^ In the Geology of New Hampshire, Professor Hitchcock has shown that the "calciferous mica-schist" dips beneath the argillite clear across the State. The much more pronounced metamorphism of the schists, the abun- dance of great granite veins containing rare minerals, as well as the long series of minerals found in the schists themselves, may be contrasted with the baiTenness and low degree of metamorphism of the argillite as indicat- ing that the schist is the older rock. The microscopic description of the two rocks may be compared from this point of view. The locality at the brook west of Whately village (see page 196) is also a decisive one in reference to the question of the relations of the two rocks under consideration. That the triangular area of argillite occurring here is a continuation of that in the Bernardston area is quite certain, in view of their complete identity, and has not been doubted by anyone; and that the black limestone, with its border of hornblende rock, is the common limestone of the lower formation is eqiially clear; but the latter is here tlu'ust up through the argillite in a knob, like a button thi'ough a button- hole, and the argillite mantles around it and dips away from it on all sides, and this is far out in the middle of the argillite, showing that the latter is underlain by the Conway mica-schist, which dips under it on the west. The relations of the two are indicated upon the sketch map (fig. 11, p. 197). A few rods farther south, and on the opposite side of the road, the limestone again buckles up twice through the argillite. CONTACT METAMORPHISM OF THE LEYDEN AEGILLITE BOKDEEING THE TONALITE 01' HATFIELD. A band about 1,300 feet wide, bordering the tonalite on the west, commencing in the woods west of the school south of Whately village and extending southwest across Hatfield to its southwest corner, shows on the exact contact a narrow band of green sericite-gneiss, and outside this a ' Geology of Vermont, Vol. II, 1861, Pis. XV and XVI. 'Ibid., p. 497. 206 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. very broad band of cliiastolite-schists, grading through pimpled schists into the ordinary slate. The argillite is itself, in its normal condition, a highly crystalline rock, approaching the mica-schists and widely removed from the more normal "argiUite," like that of Hoosick Falls, New York. Much of it is pimpled on cleavage surfaces and comparable with the knotenglimmerschiefer of the Germans. THE SKRICITB-GNBISS. • This rock may be best studied above West Brook village, on the south line of Whately. In the pasture just north of F. Bardwell's the contact of the two rocks can be followed for a long distance, and the argillite extends in a long point south into the granite ; and farther south, in the line of con- tinuation of this point, are several masses of the argillite wholly surrounded by tonalite. The southern of these rises in a vertical wall just east of a small pond in the pasture, and here the exact contact can be studied. The specimens described below were taken from this place. The rock at contact is a true sericite-gneiss. The foliation faces have a dull-green, serpentine-like surface, slickensided and with greasy feel. Broken transversely the thick sericite layers fold around small, white feld- spar grains; other layers run into white quartzite on one side and into a more micaceous and less feldspathic rock on the other. Both varieties resemble exactly the Taunus sericite rocks and are unlike the sericite or hydromica- schists of the west border of the county, where the mica scales are much more distinct. Under the microscope the fine-matted felt of a micaceous mineral (sericite) makes a background in which are scattered many wisps of green chlorite ; bright, highly refracting, rounded grains exactly resemble zircon, and large, almost wholly decomposed feldspars. The latter are wholly opaque by transmitted light and rusty white by reflected light, and often show regular eight-sided crystalline cross-sections. When very thin and very highly magnified these sections allow the light to pass through in thin, distant, parallel slits, arranged at times at right angles, at times at an angle approaching that of the prismatic cleavage in feldspar. This seems to come from thin bands of the feldspar still undecomposed. The zircon contains larg,e bubbles. PLATE III. 207 PLATE III. Fig. 1. — Ley den argillite changed to ohiastolite-sohist in contact on tonalite. Only the bLick cross of the chiastolite remains. The crystals have been changed into a mass of muscovite scales inclosing many staurotite crystals. Belmont, Hatfield. X7. Natural light. (Seep. 209.) Fig. 2. — Sections of twins of cordierite from cordierite-granite. Brimfield. X20. Drawn with crossed nicols. (See. p. 321.) Fig 3. — Diorite from north end of Paokards Mountain, Prescott. x25. Natural light. (Seep. 342.) Fig. 4. — Contact of diabase-amygdaloid and clayey limestone, from the upper surface of the Holyoke sheet. The curving of the layers of the fine mud as it flowed into the open steam holes can be seen on the left. The large cavity was clogged by a trap fragment and afterwards filled by infiltrated calcite. Rounded drops of the mud and rounded holes filled by infiltration can be seen in the trap C. Dibbles, South Holyoke. X28. (See p. 456.) 208 scaloB inciosinj IS liliert by inliltratpti U. 8. OEOLOOICAL SURVEY MONOQRftPH XXIX PL. Ill THIN SECTIONS. THE LEYDEN AEGILLITE. 209 Besides these exomorphic effects of the contact, the tonalite shows distinctly an endoinorphic influence of the schist npon itself. It is finer f^rainc'd than usual, though it is granular to the eye, and the deep flesh- red feldspar stands out on a background green from the abundance of chlorite. It is rudely foliated, and the foliation surfaces are dull-green, like the schist itself, and in transverse sections the microscope reveals thin, wavY layers, winding in between thick layers of the feldspathic material, which seem to be made up of the sericitic matter from the schist crowded into the fissm'es. The main mass of the argillite followed north from the contact retains all the complex contortions common in the rock, but it is soaked full of quartz, or quartz and feldspar, the parallel bands being in some cases sepa- rated as much as 30"" by the intrusion of these new constituents. There is also much coarse muscovite, and the rock is in places greatly brecciated. When it is followed farther north small staui'olites appear and the next band is reached. Going a short distance west along the road to the bridge over West Brook, and then south 165 feet along the brook, one finds a fine contact of the argillite and the tonalite exposed. The rock is here more arenaceous, and is indurated to a homfels. THE CHIAST0LITE-SCHI8T. This rock (see fig. 1, PI. III.) may be studied most conveniently on the southeast slope of Belmont — a great symmetrical drumlin, bare of trees, in the northwest corner of Hatfield. The original bedding of the rock is here clearly marked by bands of sandstone about an inch wide, separated by argillite layers of twice this thickness. The whole is extremely contorted, and the well-marked cleavage oversprings the sandy layers in almost every case. The original clay layers are now a fine, dark-gray mica-schist, to which one would hardly still apply the name argillite, and in some places it is coarsely muscovitic. The schist is full of chiastolite crystals, square prisms about 4°"" wide and 40™™ long, enfolded m the layers of the schist, as is Usual with this mineral. These are now uniformly changed into a shining-white muscovite in matted scales (with traces of the black cross everywhere remaining), in which small andalusite crystals occur so abun- dantly as often to occupy half the space. They are in stout prisms nearly MON XXIX 14 210 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. a millimeter across, often twinned and well terminated, of red-brown color, and with shining faces. They project in every direction into the muscovite, and have been plainly manufactured from the material of the chiastolite hj some second metamorphism. The muscovite gave deep blue with cobalt, and a purple flame when fused with gypsum, and fused with difficulty to a white enamel. It gave the axial divergence of muscovite. The staurolite, measured with reflecting goniometer, gave oo PA co Pz=129°. oo PA 00 P 06 ^115° 11', and twins after | P 06 , could be determined optically under the microscope. The andalusite crystals are orange-yellow under the microscope, but a central portion with boundaries parallel to the surface, even when that surface is plainly one of fracture, is colorless in most cases and has a soft, slightly wavy striation, which a high power shows to be due to the presence of an immense number of stout tubular bodies, slightly reddish, with rounded ends, often slightly twisted and varying in diameter; at times, indeed, passing into formless bodies. They are so numerous as to give the rock a spongy appearance, and are parallel to one another and to the verti- cal axis of the staurolite. They are 0.025"° long, COOS"""" across. Being placed parallel to the axis of the inclosing mineral, they extinguish with it; but in diagonal position the larger ones show color for themselves, and they are probably quartz. Many sections of the staurolite are broken up into separate fields from twinning, and the rods have a separate direction in each of these fields.^ The rock contains, also, groups of small garnets. It is a biotite- muscovite-schist. In a quartz-muscovite background many long-notched blades of a dark-brown biotite and much coaly matter are arranged in a pseudo-fluidal structure and wrap around the chiastolite crystals. ' Lassaulx, Ueber Staurolite: Tschermaks mineral. Mittheil., Vol. Ill, 1872, p. 173, pi. 3. Compare the uncolored figures where the rods are stouter and more distant than here. CHAPTER VIII. THE BANDS OF SILURIAN SCHISTS ON THE EAST SIDE OF THE VALLEY. As noted in the geological outline and the generalized section in Chapter HI, the representatives of the Silurian series from the Hoosac schists to the Conway schists are present east of the river in several narrow synclinal bands resting in the Monson gneiss, which are most conveniently described in geographical rather than geological order. The series is greatly simplified and is divisible into only four or five members — a mus- covitic or sericitic and biotitic quartzite below; next a band of hornblende- schist (amphibolite) ; above this a thin-bedded biotitic quartz-schist, which I have called the whetstone-schist, as it is much quarried for scythestones; then a garnetiferous and graphitic schist. These are, respectively, referred to the Howe schist, the Chester amphibolite, the Savoy schist, and the Conway schist of the western side of the valley. Along the eastern border of the region the series is still more simplified by the disappearance of the hornblende-schist, and the lower bed, which includes the Rowe and Savoy schists, is developed across Worcester County as a monotonoiis, thin- bedded micaceous quartzite which I have named the Paxton whetstone-schist, while the upper bed, the Conway schist, grows more metamorphosed east- wardly and southerly and becomes rusty, strongly fibrolitic, coarsely graphitic, and in places feldspathic. This I have named the Brimfield schist in Worcester County. I have, then, to describe the following areas (see geological map, PI. XXXIV): 1. The Northfield semisyncline. 2. The Wendell branch syncline. 3. The Leverett-Amherst area. 211 212 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. 4. The Pelliam-Shutesbury syncline. 5. The great central syncline. 6. The eastern syncline. 7. The zone of contact and disturbance around the Belchertown tonalite. 8. The Wilbraham syncline. 9. The Monson syncline. 10. The East Greenwich-Enfield syncline. THE KORTHFIELD SEMISYNCLINE. For a long time it seemed to me probable that the rocks at the mouth of Millers River (see p. 295) and those here under discussion were a con- tinuation of the Bernardston series, and thus of known age; and because of the importance of the question I have studied these areas with great care and describe them in somewhat greater detail than usual, and com- pare them with the Bernardston series, in order that the grounds for accepting or rejecting the correlation suggested above may be clearly seen. The fold here described lies along the east line of Northfield, in the Warwick quadrangle. The comparison of this series with the Silurian beds west of the river forms the first step in the correlation of the beds east and west of the river. The gneiss a in the sections below (p. 2 1 3) is identical with the Becket gneiss^ The beds h and c are close lithological representatives of the Rowe schist. The bed A agrees well with the Chester amphibolite. The whetstone-schist e is closely like the Savoy quartzose schist, while the bed / is the exact counterpart of the Conway schist in all its peculiarities, even to the presence of spodumene and cleavelandite dikes. GENERAL DESCRIPTION. The rocks were first compressed into a great syncline in the Monson gneiss, the axis of the syncline pitching to the north, and then a north-south fault occurred along this axis, and the rocks on the east were upheaved by about the thickness of the series (1,890 feet), and so far eroded that only a remnant of the lowest bed remains on the eastern half Then several transverse faults cut across the beds, and one is notable from the amount of drag which the beds on the north side of it have suffered at their south THE NOKTIIFIELD SEMISYNCLINE. 213 ends from friction ao-ainst tlie wall of the fault on the south; for an inspec- tion of tlic niiip will sliow tliat to the north of this transverse fault tlie strike of all the beds bends from a north-south direction round to an easterly direction. It is remarkable, also, that to the north of this fault all the beds of the western flank of the syncline, as well as the remnant of the eastern flank, are inverted and now dip uniformly to the west. The topography of the region is to an exceptional degree dependent upon its geological structure. Each of these transverse faults is now the gorge of a brook. The upper beds of the series — the Conway mica-schists — are the most resistant to erosion, and form the high hills, which are pushed forward or recede as the block of mica-schist of which each is made is pushed forward or back by the faulting. The amphibolite is more rapidly eroded, and it forms a deep furrow across the town, in which runs what is appropriately called the Grulf road, the word gulf being used in this sense in several places in western Massachusetts. The basal qiiartzite is also resistant and mantles over the gneiss of Brush and Crag mountains in sharp, angular ridges, which can be seen and recognized so far off as the station at Millers Falls as peculiar and not like the forms of the gneiss. The Gulf road men- tioned above runs south from Northfield to Erviug, at the east base of Brush and Crag mountains, and continues a long way on the hornblende-schist of this sei'ies, and here the whole may be best studied. 'THE GULF ROAD SECTIONS. Two miles south on this road a side road goes up onto the mountain westerly to the house of Mrs. J. Robbins, and a little farther south a similar blind road runs east to the house of R. H. Minot. The whole series is well exposed along this line, and it is described in the following section, beginning at the west end: The granitoid biotite-gneiss (a), which makes the mass of Brush Mountain, forms the base of the section. It is the northern portion of the large Pelham area of the Monson (Cambrian) gneiss. The line of boundary between the basal quartzite (the Rowe schist) and this gneiss runs beneath the Robbins house, making a large cvirve to the east, and the two rocks are unconformable. This is shown by the fact that the gneiss has strike N. 40°-50° W., dip 25°-3o° E., while the quartzite above has strike N. 15° W., 214 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. dip 10°-15° E.; and as one goes east from the boundary down the steep hill and across the qnartzite the dip of this latter rock incr(3ases gradually to 45°, showing that the quartzite mantles over the gneiss. The qnartzite (6) is here quite micaceous, the mica being, as usual, a shining-white muscovite, or often a sericite. In places coarse patches of biotite scales also occvir. Southward along the crest of the hill this bed is in one place distinctly conglomeratic, pebbles of quartz about an inch across and much flattened by the compression of the rocks, making up the mass of the rock in a great vertical clifP looking west. West of C. T. Swan's, where the 1,200-foot con- tour crosses the road, 200 rods south of the Robbins house, on the mountain crest, it is a very vitreous quartzite, resembling an aggregation of the quartz nodules in common mica-schist. Some beds here also abound in a shining- white mica, and others carry a little biotite. The thickness opposite C. T. Swan's house is 575 feet. This is followed by a very coarse, wavy, very micaceous, often sericitic, garnet-bearing schist of white color (c). It is 40 feet thick on the Robbins road; in the section opposite C. T. Swan's house, 65 feet. (5) and (c) are the equivalents of the Rowe schist. The amphibolite (d), or the Chester amphibolite, is a greenish-black rock of fine grain, separating into thin plates which have a ligniform struc- ture from the perfect "stretching" of the rock. It is usually of even grain and free from all accessories. Nodules of albite and ilmenite occur rarely. It is, on the Robbins section, about 500 feet thick; on the Swan section, 330 feet. The whetstone-schist (e), or the Savoy schist, is a gray, arenaceous biotite-schist or micaceous quartzite. The biotite is in thin scales, not concentrated upon foliation planes, but scattered sparingly and evenly through the rock. Near the top, at R. H. Minot's house, is a very rusty layer full of coarse garnet and hornblende. In the Swan section its thick- ness is 612 feet. Then follows a coarse muscovite-schist (/), often very micaceous. It is affected by both a fine corrugation of the foliation surfaces and a general twisting and contortion of the folia themselves. It is graphitic and abounds in garnets and staurolite, the latter especially abundant toward the base. Its thickness in the Swan section is 354 feet, but here the whole thickness is not present because of the fault; a little farther south, opposite the schoolhouse, it is 445 feet. This is identical wath the Conway schist. TIIIO NOIITHFIELU SEMISYNCLINE. 215 In tlie Swan section cuutiuued east aloiif)- tlie Minot mad the fault and the contact of tlie mica-schist (./'), dipping 30° E., with the basal quartzite dijipino- 15° W., can be clearly seen. The latter is here largely a two- jnica-gneiss of arenaceous structure, with shining-white inuscovite. In the section opposite School No. 10, where the road branches a little way south of Swan's house, a bed of granite 8 feet thick occupies the place of the fault, and to the east of it is a thick-bedded quartzite, which at top becomes a coarse, white, sandy muscovite-schist with wavy folia and carry- in"- o-arnets. The whole has a thickness of 307 feet, and represents the basal quartzite (b) and a little of the mica-schist (c) above it. It belongs to the eastern flank of the syncline, and is brought up by a fault whose throw must be at least equal to the thickness of the strata c to e, or 1,890 feet. The western flank of the syncline dips normally E. 30°-35°, while the remnant of the eastern flank is overturned upon the mica-schist and dips easterly against a great dike of pegmatite. SECTIONS NOETH AND SOUTH OF THE OLD WARWICK ROAD. The mica-schist (/) continues north as a high ridge which terminates in the prominent hill south of Gr. Alexander's, called locally Tom Field's hill, whose crest and western slope are underlain by the corrugated schists, while the fault runs along just east of the highest part of the hill. The schists sink down northwardly to the east-west fault Avhich follows closely the line of the old Warwick road, upon which A. Moore's house stands. North of this fault the main longitudinal fault is continued north with little or no interruption, passing just east of A. Moore's house, but the whole series of schists, which forms the western half of the anticline and which has been already described, is overturned so that it dips everywhere 50°-70° to the west. This continues to the next road on the north, the present Northfield-Warwick road, and all the members of the series are unchanged except the bed of rusty garnet-hornblende rock at the Minot house, which becomes a persistent and thick bed of hornblende-schist in the upper portion of the whetstone-schist. Another fault cuts off the southwest portion of this area, and this part is placed in normal relation to the north end of the gneiss area to the southwest, striking east and west, and dipping north away from the gneiss. The next transverse fault to the north follows the Northfield-Warwick 216 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. road already mentioned. To the north of this line the whole series of the schists is moved horizontally to the west for a considerable distance, dragging upon the fault so that the strata bend around from the normal north-south direction to a direction due east at the faiilt. The series main- tains, however, the inverted position and constant westerly strikes of the portion south of this fault. The explanation of this complex system of faults seems to be that the great Pelham gneiss mass on the west ends just where the first of these tranverse faults appears, while to the east of the great north-south fault an equally high gneiss area extends north across the whole town of Warwick. When the east-west compression acted on these beds they were on the south of the first transverse fault, supported by the north end of the Pelham gneiss in Brush Mountain, while on the north, finding firm support only at a lower level, they were thinist westward and overturned. PEGMATITE DIKES AND MINERALS. A curious point of resemblance between the Conway schists of the west side of the county and the same schists in this section— the upper beds (/) of the series — is the appearance of large granite dikes carrying spodumene, cleavelandite, tourmaline, columbite, and beryl. One great dike of this character appears in the yard of M. A. Brown, on the Win- chester road and just over the town line in Vermont, and is there filled with poor crystals of spodumene. On the top of Strowbridge Hill, a half mile south, I found the same dike, or its successor on the line of strike, filled with cleavelandite and a little tourmaline ; and the same distance again to the south along the line of strike is the fine columbite locality discovered by Mr. M. A. Brown. This may be reached by following the lane back of L. A. Moody's house, east through the woods nearly to the Warwick road. Farther south, on the Minot section, the same coarse granites carry immense beryls, and just where the beds cross the town line to the south the granite abounds in spodumene. This is one of those curious and inexplicable matters of paragenesis, and it derives its problematical character from the fact that the pegmatites cutting all the other beds are wholly wanting in those minerals containing rare elements, except those penetrating the comparatively recent Conway schist, which at distant localities on both sides of the Connecticut River carries them abundantly. SILUKIAN SCHISTS. 217 THE VVKXWELIj HKANCII SYNC LINE. Tn the southwest corner of Warwick, at Harris's pond, a subordinate syncHue hrauclies oft' from the great central synchne next to be described. It is directed first west, bends round south in Barber's hill, in which it passes tlu'ough a corner of Erving, and crosses the river and extends south into Wendell, where it ends abruptly against a fault. At the point where it branches, west of Barber's pond, the uppermost bed in the syncline is a dark, graphitic mica-schist (Conway) with abundant transverse biotite and with many staurolites and small garnets. It is thus exactly like the corresponding u^jpermost beds (/) in the Northfield syncline already described, and so forms an important link in the chain of evidence in favor of the identity of the series I am here describing with the similar series across the Connecticut Valley, with which I have associated it. There is a fine section of the beds of this series exposed in the railroad Cfjestvr/impfT/ho/if^. Granite t:hesferAi77phibolif^'^ CfiesferAtnphibolifv. Conway Sc/?/st. Savoy Schi'si: (fflHETSTOHEj JFiG. 13. — Section on railroad east of Erving station. cutting east of Erving and opposite the piano factory (fig. 13), though the beds are thrown into such confusion that no conclusions can be drawn con- cerning their sequence. Entering the cutting from the west, several large outcrops of amphibo- lite appear through the sands, and just beyond is a great boss of granite curiously molded together with amphibolite, which is changed to biotite- schist at its contact with the granite, which carries upon its back a great mass of a gray whetstone-schist extremely contorted. This is followed by a great body of amphibolite, in places much contorted. It contains albite and calcite in veins, and nodules of epidote often 15"" long. To the east this is followed by a coarse, gray mica-schist with garnets (co 0) and small staurolites. All these beds resemble closely the corresponding ones of the North- field section, which ends just north of this point, and this serves to connect the two and unite both with the western area. It serves also to illustrate 218 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. the extreme complexity which characterizes these beds in their progress south through Wendell until they are cut off by the fault east of Wendell Center. This may be well studied by going up the Osgood Brook road and turning onto the high hill north of S. Stevens's house. Everywhei'e the beds are thrown into great confusion and filled with granite dikes, so that the representation on the map, though the result of much work, gives only a general view of the main facts. The whole southern portion of the area is underlain by whetstone-schist, which has been quarried here for scythe- stones, and for this reason the hill is locally called Whetstone Hill. THE liEVEBETT-AMHEEST AREA. THE AMPHIBOLITE AND MICA-SCHIST SERIES ALONG THE EAST SIDE OF THE CONNECTICUT BASIN FROM LEVERETT SOUTHWARD. The bottom of the Connecticut Basin, as the area of transition between the closely folded rocks with vertical dips on the west and the undulating, almost horizontal gneisses on the east, is underlain by a broad band of extremely disturbed rocks, faulted, soaked full of granite and quartz veins, and, especially along a line extending quite across the State and situated at the immediate foot of the eastern plateau, most thoroughly crushed, brec- ciated, slickensided, and filled with veins of hematite, albite, quartz, and epidote, or mineral veins of the "baryta-lead formation." It is just along this line of maximum disturbance that a series of rocks which forms a repetition of those described in Northfield (p. 212) runs south from the mouth of Millers River at the great bend of the Connecticut. The same succession — feldspathic quartzite, or two-mica-gneiss (V),^ amphibolite (c/), whetstone-schist (e), and spangled mica-schist (/) — can be made out, but with difficulty, and all the members are much altered and thrown into great confusion, so that the assignments made upon the map, though the result of long study, are given with much hesitation. For convenience the amphibolite and the quartzose bands, the quartzite below and the whetstone-schist above, are described together, while the equivalent of the spangled or Conway mica-schist — the Amherst feldspathic mica-schist — is discussed apart. ' The italic letters a-f refer to section given on pp. 213-214. THE LEVEKETT AMnEUST AREA. 219 NORTH LEVERETT (GREENFIELD (iUA])RANGLE, SOUTHEAST CORNER). Still turtlier south, and just south of the railroad crossing over Locks Pond Hrook, the series appears in Stoddard Hill and forms a narrow band running- south between the gneiss and the red sandstone. It is largely cut by granite, which has replaced it over great areas, and this rock shows often the peculiar structure adverted to in the description of the Millers River section (p. 295). The materials, of a very coarse pegmatite, form a dis- tinct!}" foliated mass from the parallel arrangement of the large muscovite scales, and agree in dip and strike with the surrounding schists. It seems also, where it comes in contact with the different beds of the series, to have absorbed larg'e quantities of their material into its mass, being near the amphibolite a fine two- mica-granite of coarse but very even texture, as above the cemetery in Leverett, and finer-grained, more quartzose, and almost free from mica in the neighborhood of the quartzite, as north of the cemetery in North Amherst. Furthermore, the granite seems to have assumed a schistose character where it has intruded itself into the place of the more schistose members of the series, as if by a kind of pseudomorphism it had inherited their structure. The series in Leverett is divided into two portions by an exceptionally large mass of granite. The northern portion presents a section, from east to west as follows: [a) Monson gneiss, (6) basal quartzite, (c) mica-schist, h and c together representing the Rowe schist, (cl) Chester amphibolite, (e) granite, here occupying the position of the whetstone (Savoy) schist, (/) spangled or Conway mica-schist — all dipping westward from the gneiss. East of E. Gr. Reynolds' the quartzite is feldspathic and like the Bernardston upper quartzite. The mica-schist (c) and the amphibolite {d) agree com- pletely with the corresponding beds of the Northfield section. In Stoddai'd Hill, 325 feet east of the railroad, the latter is a coarse hornblende-schist, in places very biotitic, in places massive. The mica-schist (/), which I identify with the Conway mica-schist, is the first outcrop we meet, going south, of a rock which, from its expansion across Amherst, I have called the Amherst feldspathic mica-schist. Its appearance here in the same position as the Conway mica-schist of North- field is one of the reasons for identifying the whole mass with the rock so named across the river. The subject is fully discussed on page 222. 220 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. The rock here is a coarse, very rusty, garnetiferous and feldspathic mica-schist. It is surrounded by g-ranite, and floats, as it wei-e, in it, and is largely injected with granite veins, which at times so greatly predominate that one must describe the area as occupied by granite containing parallel filaments and thin sheets of schist. The latter do, nevertheless, preserve the dip and strike of the main mass, while the granite has also its constant rude dip and strike in the same sense as if the process had here been carried a step farther, and the granite, being injected into and opening out the laminae of the schists and cooling between them, had retained a lamination from them after they had been wholly or almost wholly absorbed into its mass. The schists agree so closely with the Conway mica-schist where it comes into the granitic areas on the west of the river, directly opposite, that I have no hesitation in following the stratigraphical indications and associat- ing them together. LEVERETT CENTER. Southwest of the great mass of granite another long strip of the rocks of the series runs from A. Field's, on the road east of Mount Toby, southeast through Leverett Center and South Leverett and on into Shutesbury, to end in Mount Boreas at Adams Mills. Just above Leverett Center the gneiss is notched into it by a series of faults. The amphibolite runs down the eastern border of the strip. It is for the most part a thin- fissile rock, often stretched and ligniform, of dark- green color, made up of magnetite, feldspar, and hornblende, the latter in elongate needles, and all parallel to one another and to the line of stretching. It is at times, as south of A. Field's, a tremolite-schist. The mineral is in short, stout prisms, without feldspar, quartz, or ore. Rarely the lower mica- schist (c) appears between it and the gneiss, but the whole series is in the greatest confusion and is also largely covered by till and sand. THE SAVOY SCHIST, OR WHETSTONE-SCHIST. The center, and by far the larger portion of the series, is taken up by an arenaceous rock, slightly micaceous, and at times slightly hornblendic, which is at times crushed to pieces and jointed and cut by many quartz and specular iron veins, the rock itself being thoroughly silicified and ren- dered compact and hornstone-like. It is often exactly like the corresponding THE LEVEllETT-AMHEKST AKEA. 221 stratum in the Northfield Mouutaiii. It abounds often in a green, cliloritic mineral, and along the road east of Mount Toby it is abundantly brecciated" aud cemented by hematite. Along the east side of Mount Toby the whetstone -schist appears in the base of the mountain. Its first outcrop is just south of the Mount Toby station, aud it can be followed from this point south to the first brook, where the contact of the Mount Toby conglomerate upon the whetstone is 40 feet above the railroad (436 feet above sea level), and on to the second brook, where the contact is 12 feet above the railroad. This greatly lessens the probable thickness of the conglomerate of Mount Toby. In the extreme northeastern corner of Amherst there is by the roadside a small outcrop of thin-fissile, stretched hornblende-schist in the whetstone, exactly resembling that found in the Northfield section. It can be traced northwest past the brook-crossing east of A. Adams's house, and connects with the bed at Leverett Center. It is a beautiful rock under the micro- scope. The perfectly parallel hornblende blades have the strong pleoch- roism and absorption of the Chester amphibolite, large, rounded grains of magnetite are frequent, and the whole is placed in a background of untwinned feldspar grains. An analysis of the rock by Mr. L. G. Eakins is given below : Analysis of liornblende-schist from Amherst. ■ SiOi . Per cent. 47.56 1.24 16.13 trace 1.80 9.39 .08 trace 6.67 9.21 1.58 2.52 3.51 .21 TiOi Al.Oa. . . Cr.,03 Fe^Os FeO. ... MnO .. . . BaO CaO Mo-O KjO Na^O HO P.,0, 99.90 222 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. THE AMHERST FELDSPATHIC MICA-SCHIST (CONWAY SCHIST). « LEVERETT. West of the band of whetstone-schist in South Leverett begins the broad area covered by the Amherst feldspathic mica-schist and granite in inextricable confusion. It is in this northern portion so purely granite, and the shreds of schist are so impregnated with granitic material, that I have marked but a small portion as schist upon the map. AMHEBST. The Conway mica-schist of the western side of the Connecticut Valley, where it approaches the great masses of granite from Williamsburg to Montgomery, can be seen along both dip and strike to become more coarsely crystalline and feldspathic, while the plumbaginous material dis- appears or crystallizes into graphite and thus colors the rock less. The garnet and staurolite also disappear in large measure, and a rock results closely comparable to that which underlies the towns of Amherst and Hadley Furthermore, the same mica-schist in Horse Mountain, on the western line of Hatfield, dips west and formerly mantled over the hornblende- granite at its eastern foot. It is here not greatly different from much of the Amherst rock, and I assume that it reappears in the nearest outcrops on the east of the river in Mount Warner, where it forms a much-disturbed syncline, and then extends across Amherst, on its eastern border dipping west — that is, away from the hornblendic band which underlies it, and which I have already traced across Pelham. Starting thus from the exact lithological identity of the Conway mica- schist of Northfield on the east and that of Coleraine directly opposite, west of the Connecticut, I have shown that the coarse mica-schists of the north- west of Leverett occu^jy the same stratigraphical position as the Conway schist in Northfield, and then have traced the Leverett schists southward into continuity with the Amherst schist. The latter is then shown to be identical with the altered representative of the Conway schist on the west of the river just opposite, and the same parallelism can be proved clear across the State. THE LEVERETT-AMIIEHST AEEA. 223 It remains to discuss the rock itself and see how far it still shows i)oints of resemblance to the calciferous mica-schists. The r(H'k appears in Mount Wai'ner, in an area north of South Am- herst, beneath the till in the ridge from Amherst village to North Amherst, and in the rocky region along the north line of the town and extending over into Leverett. It is everywhere greatly cut by granite dikes and thoroughly impregnated with granitic material, especially in the latter area, where it exists only as shreds in an almost continuous expanse of granite. This is clearly the eastern border of the great granite area which has its center in Williamsburg, on the west of the valley, and extends thence east beneath the Trias and finds its eastern border closely coincident with the Conway mica-schist in which it has its whole development. Description. — The rock is in composition a gneiss, in texture a coarse schist, so that Dr. Hitchcock sometimes gave it one name and sometimes the other. It varies from a coarse muscovite-schist, made up almost wholly of mica in large scales, to a schistose gneiss, at times containing large rounded masses of fine microcline. It is always rusty, and very generally contains pyrite, so that the water from several wells along the western border of Amherst, when low, curdles inilk and gives strong reaction for sulphuric acid, and in new openings fissures of the rock are covered with fine sheets of pyrite of very recent origin. Along the western edge of the ridge, appearing in my well on the Northampton I'oad, and in that of President H. H. Goodell farther north, as also in Mount Warner, is a band the lamination surfaces of which are spangled with large, rounded, equidistant plates of silvery muscovite filled with fine radiated needles of fibrolite, a peculiarity which appears on a much more extensive scale in the more easterly bands of the mica-schist. This fibrolite occurs where the road over Mount Warner rounds a rocky spur at the southwest corner of the mountain, and this is the most western appearance of fibrolite in the schists. In excavations at the north end of Prospect street, in the hill east of North Amherst railroad station, and in the large outcrop near South Amherst, there are intercalated beds of an eclogite-like rock, a massive quartz-garnet- hornblende rock containing shining scales of graphite. The garnet is light- red, and is intimately mixed with the quartz to form a groundmass which the hornblende penetrates in stout, parallel rods, transverse to the bedding, 224 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. of such size and arrangement that, on weathering, the rock presents very closely the appearance of a scolithus sandstone, and I tried for a long time to persuade myself that this was the case. I am now inclined to connect these beds with the beds of tough hornblende rock carrying black garnet which appear in the Conway schists, either alone or as a selvage to the beds of black limestone, as both have the same composition — quartz, garnet, hornblende, and graphite. The hornblende changes often to serpentine. A very similar rock appears on the eastern border of the tonalite in Hatfield (near the house of J. Glasner), and is probably a product of the contact action of the latter upon a limestone bed of the Conway schist. The exposures are not sufficient to make its relations clear. Correlation of Amherst schist. — As a feldspathic mica-schist the rock resembles the feldspathic varieties of the Conway schist on the west side of the valley, especially in its southern extension, as about Russell. In this assignment I have been influenced by stratigraphical considerations, by the very general content of graphite, by the common traces of calcite, by the probable derivation of the eclogite-like rock from bands of arena- ceous limestone like those common in the calciferous mica-schist, and by the fact that these Amherst schists closely resemble the calciferous mica-schist immediately opposite, in Williamsburg, where it is most influenced by the granite and develops into a fibrolite-schist like the neighboring bands of the same schists on the east. Minerals in the Amherst schists. — Apart from the baryta-lead veins, described under mineral veins in Chapter XIV, there have occurred the following minerals in the schists: (1) Essonite and graphite; west slope of Mount Warner. (2) Heulandite in perfect, deep-red crystals, with rosettes of a newly formed pyrite ; head of Prospect street, Amherst, and at the college grove well with pyrophylhte. oo P do (010), — 2 P o6 (201), 2 P oo (201), P (001), 30 P (110). (3) Pyrophyllite after feldspar, fibrolite, and biotite. In a well at the northwest corner of the college grove the rock was a biotite-schist, much impregnated with granite, which swells to lenses of the coarsest pegmatite many feet in length. Associated with these granite lenses are layers and large masses, which often run off' into veins across the schists, of a granitoid mixture of quartz, little feldspar, and much green biotite THE PELHAM-SMUTESBUllY SYNCLINE. 225 (rarely brown) in large scales which are completely filled with radiating tufts of fibrolite. The schists contain graphite abundantly disseminated in small scales, often hexagonal, and some layers are finely colored masses of purple, almost amethystine garnet, pyrite and apatite in distinct crystals. The orthoclase t)f the granite is now in every stage of change into an amorphous mineral; Hzzl; color, pale mountain-green to deep olive-green, or light to dark wax-yellow. The yellow is translucent on edges, and this increases in water. In the flame the green mineral rapidly becomes white. The yellow variety becomes flesh-colored, like a decomposed feld-spar. Both give a fine blue with cobalt, and fuse at 3 to 4 to white enamel. The fibrolite is also often attacked in the same way, the change proceeding from the centers of radiation of the needles, which are first beaded with browner spots, showing aggregate polarization, and then wholly changed, and at last involving the biotite also, while the garnet is the last to be afi"ected. The quartz clears up under the blowpipe and efi'ervesces with soda. The quartz, through all the adjoining schist and granite, is of the same waxy luster as the amorphous mineral, and has become brittle (H^4), and gives with cobalt a beautiful, filmy, superficial blue, deeper in spots. The change seems to be initiated by the decomposition of the pyrite, and it seems possible that the curious appearance of the quartz is due to hydi'ofluoric acid set free from the micas during their decomposition, but it is at times a deeper change into an aluminous silicate. The topographical surroundings of this interesting locality are such as to render it probable that the Trias conglomerate was barely planed off from its surface during the Glacial period, so that it is a remnant of an ancient and peculiar form of decomposition which took place beneath the conglomerate. THE PELHAM-SHUTBSBURY SYNCLINE. Across Pelham the great block of Monson gneiss (a) which occupies the whole town is nearly horizontal, with low dip to the east on the east side and to the west on the west side. On the east side of this extremely flat anticline we have, commencing with the central (that is, the lowest) beds at the quarries in the center of Pelham, the true friable subporphyritic MON XXIX 15 226 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. biotite-gneisses (a); then, at and just east of East Pelham, the actinoHte quartzite; then a second narrow band of the biotite-gneiss (a); then a broad band of the muscovite- (&) and hornblende- (d) schist here discussed; and finally, at top, a recurrence of the true biotite-gneiss (a)} On the east the series dips with so small an angle and so regularly eastward, and the members can be seen passing under each other so normally, that it is very difficult to avoid the conclusion that they are in regular succession and that all are a part of the Monson gneiss sei'ies, and this was at first my opinion. On the other hand, the series bears in several particulars strong resemblance to the mica-schists and amphibolite as devel- oped to the east in New Salem. As one goes down the long hill east from Pelham Center, after reach- ing the first road tui'ning south, one finds many outcrops of a thin-fissile, quartzose two-mica-gneiss, which varies from a thin-fissile quartzite with much coarse muscovite spread upon the rather distant foliation faces to a fissile biotite-gneiss with muscovite distributed as above, or, finally, a shin- ing-white, purely muscovite-gneiss, or — and this last comes to be the prevailing rock going either north or south — a very coarse, rather rusty miuscovite-biotite-gneiss or schist. This agrees closely with the basal beds (V) of the other section. Slight traces of the hornblendic rock (d) occur down this slope, but northward, across Purgee's brook, a heavy bed of the hornblendic rock {d?) appears in the bluff north of D. Shore's house. The series can be followed from this section south 5 miles to a point west of Enfield Center and north 5 miles across Pelham and Shutesbury into Wendell, maintaining a width of about a mile, which, from the low dip and its position on a hillside sloping with the dip, does not represent a great thickness In all this distance the rock is everywhere cut by great granite dikes or is greatly impregnated with granite, so that many beds seem like piu-ely granitic (pegmatitic) material made schistose by pressure. Above the amphibolite (e Fig. 14.— Section of schists west of Belohertown. A/nphiboAye. mica-schist of dark greenish-gray color and so full of small cubes of pyrite that it is deeply decomposed. Dikes of coarse pegmatite from 1 to 25 feet wide cut through it, and they also contain pyrite and are kaolinized to great depth. The mica-schist is 10 feet thick. Below this is a thin-bedded, light-gray quartzite, slightly biotitic on cleavage faces. It is much fissured, and filled with combs of quartz, films of hematite, and calcite, and slickensided. Below this, on the west side of the road, is a massive, crumbling amphibolite, which seems to underlie the quartzite. The section is here sufficiently undisturbed to show the amphib- olite in normal relation to the upper beds. Along the road farther south, in the field east of T. S. Haskel's, is an outcrop of a coarse sahlite-amphibolite, like that at Kelleys Crossing, which joins Monson gneiss on the east and is cut off by a great granite vein upon the south. It is still coarser than that farther north, and the pyroxene crystals are larger. It furnished the material for the microscopical description of the rock below. It is a massive, friable, granular mass of CONTACT AROUND BELCHERTOWN TONALITE. 245 greeu p\TOxene grains, with here and there a great bhxck hornblende crystal appearing porphyritically in the mass, its shining surfaces luster-mottled bv many grains of" the pale-green pyroxene, which are here better crystal- lized and smaller than in the main mass. The pyroxene is colorless, without ])ina<'oidal cleavage or inclusions of any kind. This is the rock that was called augitic syenite by President Hitch- cock, and slides were cut from the specimen in the survey collection (XVIII, 92). The rock contains large leek-green crystals of pyroxene, large black hornblendes, and a scanty granular groundmass of plagioclase. In slides the dark-green hornblende, which is at times brown centrally, is luster-mottled on its broad cleavage surfaces with pyroxene, which is faintly reddish, of high refraction and coarse cleavage. The large pyrox- enes are intergrown with irregular portions of hornblende with the axes a and h of the two minerals parallel. In sections normal to h the cleavage lines coincide and a revolution of 17° to 19° brings the hornblende to extinction, and of 43° in the same direction, the pyroxene. The two min- erals are so interwoven that they give almost an aggregate polarization. Small, brown octahedra appear in the hornblende. In general the amphib- olite is not made pyroxenic, but is only crushed and filled with quartz veins. It is the usual flat-fissile, dark, fine-fibrous rock. Samples from an artesian well, bored on the grounds of Mr. Myron P. Walker, in the center of Belchertown, taken at the depths indicated, gave the following results: Record of an artesian-well boring in Belchertoton. 80 to 100 feet, pegmatite. 115 feet, granite, witli little amphibolite. 130 feet, granite, with little amphibolite. 145 feet, granite. 160 feet, granite. 175 feet, yellow granite, with much muscovite. 190 feet, gray granite, with amphibolite. 205 feet, gray granite, muscovite, and amphibolite. 220 feet, gray granite, muscovite, and amphibolite. 249 feet, much coarse biotite. Still farther south, on the west slope of Baggs Hill, in Granby, appear dark greenish-gray, membranous, feldspathic mica-schists, associated with a quartzite which is at times blackish, at times greenish, and abounds in quartz crystals and pyrite. 246 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. THE FIBROLITE-SCHIST INCLUSIONS. The most instructive occurrence to prove the eruptive character of the tonaHte and to ilhxstrate its contact phenomena is found in the broad sheet of coarse fibroUte-schist which runs two miles southwest from "Slab Citv," in the east of Belchertown, to end at the house of V. H. Pease. In the mid- dle of the road that runs along its southern border at the western Clough house — this and the Pease house being the only ones on this road — at a watering trough, a brook crosses the road, coming down over the rocks, and 30 feet above the road one sees the contact of the tonalite and the schists above it, and at the trough the quartzite appears as a gramilar quartz- eijidote rock. The bright yellow-green epidote is in rounded crystals, each surrounded by a white spot, from which the iron has gone to supply the epidote crystal. At the northeast end of the inclusion, at G. Robinson's, a dark biotite schistose gneiss, like that found at Baggs Hill, dips normally under the fibrolite-schist. The biotite is black, with a shade of green, and makes continuous films through the granular quartz mass. Below this gneiss are beds of a thin-fissile, slightly micaceous quartzite. This fixes the position of the fibrolite-schist as the equivalent of the upper mica-schist, as does the fact that it lies in continuation of the mica- schists in Enfield, and the latter are the only beds sufficiently argillaceous to have furnished material for so much aluminous silicate. These same mica-schists grade eastward into fibrolite-schist and continue across Worcester County, but they are rarely so coarse as here, PETKOGRAPHICAL DESCRIPTION. Fibrolite-cJilorite-scJiist, from bowlder in cutting on Massachusetts Cen- tral Railroad, South Belchertown, but coming doubtless from the contact zone of the granite; a stretched gneiss-like rock of gray color, with shade of green and showing much fibrolite. Under the microscope radiated fibrous tufts of a green, chloritic mineral inclose much graphite in notched plates, and this chlorite is associated with an abundance of large garnet grains free from the same inclusions, and these together frame large grains of quartz full of rutile needles. The quartz polarizes as a mass of grains and is plainly secondary. The fibrolite is abundantly woven through the whole. CONTACT AROUND BELCIIERTOWN TONALITE. 247 I'lbrofite-biotitc-schist, from south end of the main belt of schist in the granite. This is a coarse schist, showing an abundance of muscovite and biotite, rusty," and containing large spots of garnet and coarse fibrolite blades, often 3-5""" wide. 'J'he microscope shows many black scales, part of which are blood-red specular iron, and part seem to be graphite, as they are grown together in long lines and have rounded outlines. There are many rutile needles in the quartz. Garnet-staurolite rock, from large bowlder in the first cutting of the Massachusetts Central Railroad south of Belchertown, and coming doubtless from the band of fibrolite rock to the north. This rock represents the extreme of metamorphism reached by the rocks bordering the granite. It is a highly crystalline rock of medium grain. Large patches of garnet and quartz and much biotite are visible to the eye, and the lens detects much staurolite, graphite, and a few shining surfaces of fibrolite. Under the microscope nearly half the surface is occupied by stauro- lite; the garnet patches are seen to be made up of congeries of small grains, and these two separate quite widely the quartz patches, which are crowded with fibrolite and rutile microlites and are thus plainly secondary quartz. All these minerals include plates of graphite scales — single or grown together in long series. Epidote rock, from Belchertown. This is an interesting product of the contact metamorphism of the tonalite upon the schist. It occurs at the watering trough near the house of J. Clough, in the southeast part of Belchertown. The rock has a mottled look; a white groundmass winds among rounded spots of a dark yellowish-green color, made up of biotite and epidote. The rock grades into biotite-gneiss. The epidote is the most abundant constituent, and with a strong lens one can make out the fresh, shining-, model-like crystals, regularly dis- seminated, and semiopaque centrally. With the microscope they are seen to be filled with grains of quartz, of elongate, irregular shapes, and very large in proportion to their host, which crowd the central portion and radiate outward. It contains, also, chlorite scales. Biotite, regularly disseminated and strongly dichroic, molds itself to the epidote, as does the rare quartz. Apatite occurs in regular crystals, forming pleochroic rings in the biotite. 248 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. All the constituents are perfectly fresh and almost entirely free from fluid inclusions and microlites, and the absence of these, as also of zircon, rutile, garnets, and iron ore, is remarkable. Hand specimens are on one side biotite-gneiss, on the other epidosite, and the two seem normally interlaminated ; but the latter must be of later and very different origin, and may be in effect a vein stone, in which, per- haps, the chlorite scales are remnants of the earlier rock, which has been almost wholly resorbed to make place for the new minerals. THE WILBRAHAM SYlSTCIilNE. South of the deep transverse valley of the Quabaug and its continua- tion in the Chicopee River, the simplicity of the geology is as marked as is the complexity of the region north of the same valley. Three great syn- clines of the schists run south across the towns named above, forming as many high ridges. The Wilbraham sjmcline looks down on the sands of the Connecticut Valley on the west and upon the deep gneiss-bottomed valley of East Wilbraham on the east, and across this valley rises the West Mountain of Monson, made up of a second syncline of the same rocks and looking down on the deeper and narrower Monson Valley, which is under- lain by the same gneiss. Across this valley on the east the third syncline rises to form East Mountain, which is bordered on the east by a less strongly marked and yet distinct valley, underlain by a third repetition of the Monson gneiss and, followed farther east, by the Brimfield gneissoid mica- schists, forming a foiirth syncline. (See sections, PI. XXXII, and map, PI. XXXIV.) The Wilbraham syncline is concealed in its western half beneath the Triassic sandstones, which rest against the western foot of the ridge, and the slope of the ridge on the west is so steep that it is probable that the fault, so well marked farther north, is continued at its base, and that the rocks have sunk to form the broad Connecticut Valley. It is a closed fold, slightly overturned to the west, and its rocks closely resemble the corresponding beds on the west of the Connecticut Valley in Granville. The gneissoid quartzite or muscovitic gneiss, the equivalent of the Rowe schist, which usually intervenes between the Monson gneiss and the horn- blendic beds, seems to be wanting here, and the hornblendic beds rest directly on the white biotite-gneiss. The upper beds of this gneiss are very fine-grained and magnetitic and probably represent the Rowe schist, but THE WILBRAHAM AND MONSON SYNGLINES. 249 no bolln(lal■^• could be drawn below to separate it from the Monson gneiss proper. The hornblende-schist (Chester amphibolite) is a jet-black rock, satiny on the surface from the eifect of the great number of fine needles of horn- blende which make up nearly its whole mass. The whetstone-schist (the equivalent of the Sa\'oy schist) is a gray, granular, friable quartzite, vary- ing from thin-fissile to massive, often a shining muscovite-quartzite, or abounding in distant flakes of chlorite. It is covered on the western flank of the syncline until the range crosses into Connecticut, when it appears on the west flank of Perkins Mountain. The Conway mica-schist is a coarse, light-gray muscovite-schist, gen- erally barren, but caiTying at times a few garnets. Along its western base it is much crumpled and silicified, as if from the influence of the fault. On passing into Connecticut the regularity of the syncline is interrupted. The amphibolite band which forms the ridge of Pine Mountain, Rattlesnake Hill, and Perkins Mountain, in Somers, is suddenly cut off" in the south shoulder of Perkins Mountain by the gneiss. The latter rock, which up to this point has dipped a little north of west, here swings around sharply, dip- ping steeply north and northeast, so as to cut off the whole series up to the mica-schist, and, reversing its direction, it runs south again, dipping normally beneath the Conway schist, of course with a fault boiuidary. THE MOKSON SYlSrCLIKE. The west Monson syncline is a perfectly symmetrical closed fold of the scliistose series in the gneiss, and its character will be understood by com- paring the detailed section below with the cross-sections on PI. XXXII. The section given below commences with the older rock — the Monson gneiss — on the east, at a point 1,830 feet east of the sharp turn in the road at the house of A. Bliss, jr., a mile northwest of Peaked Mountain, and runs west : a.^ Monson gneiss. b. Eowe schist. Gneissoid quartzite, with very little feldspar, muscovite, and a green mica or chlorite, with beds of gray biotite-quartzite, chlorite-schist, and horn- blende-schist appearing a little farther north, opposite the house of J. Burley ; 361 feet. d. Chester amphibolite. Epidotic quartz-hornblende-schist, thin-bedded above and changing into chlorite-schist; 459 feet. e. Savoy schist. Cbloritic mica-schist, with subordinate beds of muscovite- gneiss, changing above into arenaceous mica-schist (whetstone-schist) and still higher ' These italic letters refer to sections described on pages 213-214. 250 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. iuto quartzite and quartzose sericite-schist, with pale-greeu hydrated muscovite and large distant garnets; 3,791 feet. /. Conway schist. A coarse, lead-gray, barren mica-schist occupies about half the thickness of this bed and is succeeded above by a corrugated mica-schist of fine grain, very dark, from the large amount of graphite in it, and abounding in small garnets (oo P) and dark red-brown biotite, set transversely to the bedding; 1,188 feet. The similarity of this series to the corresponding one across the Con- necticut Valley is striking. Each subdivision between the Becket gneiss and the Leyden argillite is represented, though vpith diminished thickness. The Savoy schist (e) is well exposed in the first cutting west of the Palmer station on the Boston and Albany Railroad. Here there is trace, apparently, of a corrugation of the quartzite, upon which the vertical foliation may be superinduced as a secondary structure. This would throw doubt upon the thickness given above. Following the Somers turnpike a mile and a half west from the south end of State-line Pond, in Connecticut, at the south end of the long ridge of Peaked Mountain one comes upon the finest quartz- conglomerate in the Rowe schist. It is in a great ridge on the north side of the road, at a ruined house northwest of the schoolhouse. The mica-schist (/), the central portion of which agrees strikingly with the Conway schists clear across the town, is best studied where the road from iPalmer to Hampden crosses it, a mile northwest of Flynt's quarry. The uppermost beds are so fine-grained and plumbaginous that they recall the Leyden argillite, and this is exactly the horizon at which it should occur. If the section be continued westward it repeats itself exactly in inverse order, though here the amphibolite is very generally porphyritic in appear- ance — a structure which is due usually to the absence of hornblende from small spots regularly disseminated, so that the whole granular groundmass shows; but many bowlders of the rock found in the southwest corner of Monson are of fine, porphyritic diorite-schist with fresh, poorly cleaving feldspars in close-set, rounded grains. It seems to me probable that a narrow fragment of rocks of this series starts east of Flynt's quarry, near the "rock house," and extends north thrditgh Bunyan Mountain, either faulted down into the center of the anti- cline or brought there by a subordinate downward fold of the schist which foi-merly mantled over the gneiss. It was of too limited extent to find place upon the map. CONTACT AROUND BELCHERTOWN TONALITE. 251 THE BAST GREElSrwiCII-ENFIELD SYNCLINE. A naiToM' s>-iu'liue comes out from beneath the sands east of Green- wich village and near the east line of Greenwich. Traces of it appear to the north, mostly covered by sand, along- the roadside east of Warner's pond. It makes tlie high hill which extends down the east line of Greeiiwich and Enfield, and is well exposed along- the road running east from Enfield. Here, near W. N. Avery's, the fibrolite-schist is a nearly pure bucholzite, in thick layers, in a fine-grained feldspathic quartzite without brown mica, gi'aphite, or garnet. As it lies immediately above the amphibolite it occu- pies the position of the whetstone-schist. The center of the series is occupied by the rusty mica-schist, while on the west there is a dull-greenish graphite- garnet-muscovite-schist and a granulite with its gai-nets bordered by green, and both these beds indicate the presence of the basal beds below the amphibolite, but not in thickness sufficient to be put upon the map. In Ware this terrane is well exposed between the town farm and the schoolhouse to the west. It is shifted to the west by the great fault in the south of Ware, and across Palmer it forms the high Pattaquattic Hill and the range of high ground south across the town. On the south flank of this hill, northwest of J. Can-igan's, the black mica-schists are locally so crowded with the large rounded "augen" of feldspar that the separated folia of the schist, 2-5"" thick, w^ind in and out among the latter and occupy not more than a fourth of the space in a cross-section of the rock. In Palmer it can best be studied along the road running east from the Center, and its first branch to the northward, especially in the hill east of B. Olney's. Here a distinct band of quartzite appears above the amphib- olite. It runs out soon after reaching Monson. EESUME. ARGUMENT FOR THE IDENTITY OF THE SCHIST SERIES EAST OF THE CONNECTICUT WITH THOSE ON THE WEST. In the north of the State the beds in the first band east of the river agree most closely with the con-esponding beds west of the river, and some of them, as the Conway schists, agree exactly in a multitude of characters. Southward the strata change greatly, but in the latitude of Amherst the 252 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. corresponding strata east and west of the river still resemble eacli other closely. Farther south the beds revert to the types prevalent in the north- ern part of the State, but with many minor peculiarities, and still the resem- blance is close, bed for bed, between those on the east and those on the west of the river. THE PASSAGE EASTWARD INTO THE BRIMFIELD FIBROLITE-SCHIST. On the north of Warwick the Conway schists agree exactly with those in the Northfield syncline next west, and thus with the type west of the river. As the beds are traced south along the strike they gradually become fibrolitic, and the same change takes place slowly across the strike as one goes eastward into higher beds. Because of the abundance of granite, metamoi-phism is more pro- nounced in the central part of the State, and here fibrolite occurs in the schists all the way to the Connecticut. Again, where the nonfibrolitic Conway schist passes out over the Belchertown tonalite it becomes most coarsely fibrolitic. Finally, as the beds are followed still farther east across Worcester County the great increase of granite promotes a corre- sponding increase of fibrolite in the highly aluminous Conway schists, and they are named for convenience the Brimfield schists. CHAPTEE IX. THE BERNARDSTON SERIES OF UPPER DEVONIAN ROCKS. LITERATURE. 1. 1819. B.Hitchcock. Geology of a section of Massacliusetts on Coiiuecticut Eiver, etc. Am. Jour. Sci., 1st series, Vol. I, p. 105. 2. 1823. E. Hitchcock. Geology, etc., of the regions contiguous to the Connecticut River, with map. Ibid., Vol. VI, p. 1. 3. 1825. E. Hitchcock. Same article, separate publication. 4. 1832. E. Hitchcock. Eeport on the Geology of Massachusetts. Part I, Economic Geology, with map. Ibid., Vol. XXII, p, 1. 5. 1833. E. Hitchcock. Eeport on the Geology, etc., of Massachusetts,with atlas. 6. 1835. E. Hitchcock. Eeport on the Geology of Massachusetts. 2d edition, 702 pp.; map. 7. 1841. E. Hitchcock. Final Eeport on the Geology of Massachusetts. 831 pp.; map. 8. 1844. E. Hitchcock. Explanation of the newly colored map of Massachusetts. 22 pp. 9. 1844. E. Hitchcock. Geological map of Massachusetts, on same sheet as the Borden Trigonometrical Survey Map. 10. 1851. E. Hitchcock. On the geological age of the clay slate of the Connecticut Eiver Valley. Proc. Am. Assoc. Adv. Soc, Vol. VI, p. 298. 11. 1858. E. Hitchcock. Geological map of Franklin County, on Walliug's wall map of Franklin County. 12. 1861. E. Hitchcock and C. H. Hitchcock. Eeport on Geology of Vermont. Vol. I, p. 447; Vol. II, p. 598. 13. 1870. C. H. Hitchcock. The Geology of Vermont, in the Geology of Northern New England. 14. 1871. C. H. Hitchcock. Geological map of Massachusetts, in Walliug's Atlas of the State. Boston. 15. 1873. J. D. Dana. On rocks of the Helderberg era in the valley of the Con- necticut; the kinds including staurolite slate, hornblende rocks, gneiss, mica-schists, etc., besides fossiliferous limestone. Am. Jour. Sci., 3d series. Vol. VI, p. 339. 253 254 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS, 16. 1877. C. H. Hitchcock. Note upon the Connecticut Valley Helderberg. Ibid., Vol. XIII, p. 313. 17. 1877. C. H. Hitchcock. The Geology of New Hampshire. Vol. II, p. 428, with map and sections. 18. 1877. J. D. Dana. Note on the Helderberg formation of Bernardston, Massa- chusetts, and Vernon, Vermont. Am. Jour. Sci., 3d series, Vol. XIV, p. 379. 19. 1883. R. P. Whitfield. Observations on the fossils of the metamorphic rocks of Bernardston, Massachusetts. Ibid., Vol, XXV, p, 368. 20. 1890. Ben K. Emerson, A description of the Bernardston series of metamorphic Upper Devonian rocks. Ibid., Vol. XL, pp. 263, 362. HISTORY. 1819. "Argillite sometimes alternating with mica-slate, siliceous slate," "undoubtedly primitive." Almost perpendicular, inclining a few degrees to tlie west. (1,^ p. 105.) The hornblende-schist of this series is associated with the Triassic "greenstone." (1, p. 109.) ■ 1823. Extends from Leyden, north to Rockingham, Vermont; occurs again at Woodbridge, Connecticut; often tortuous and slightly undulating, especially when passing by imperceptible changes into mica-slate. It embraces numerous beds and "tuberculous masses" of white quartz. It also alternates with mica-slate, and a peculiar coarse limestone forms beds in the argillite. The map separates the argillite from the mica-slate on the west by a continuous band of limestone and extends it eastwardly to include all the mica-schists which have been associated with the Bernards- ton limestone in later time, while the mica-schists on the eastern side of the river are associated with the mica-slate west of the argillite, (2, p. 36.) The hornblende rock is separated as primitive greenstone in the north of Gill and south of West Northfield. (2, p. 31.) 1832. The limestone and magnetite beds which had been worked forty or fifty years before, but had produced poor iron, are described briefly in their economic aspect but without geological data. (4, p. 27.) It was supposed to form a bed in the argillite. Compares it in value with a gold or silver mine. 1833. Fossils discovered in the limestone and figured (6, atlas, pi. 14, p. 47) ; and the limestone, though not seen in contact, supposed to lie uncon- formably upon the argillite. The quartz rock lying above the limestone, ' The numbers 1, 2, 3, etc., refer to the numbers above under the head of Literature. BEKNAKDSTON SERIES OF UPPER DEVONIAN. 255 but not seen in contact, is noted. The complex folding-s of tlie arf)-illite iu-e described and fig'ured. (6, pp. 289, 295.) Concludes that the encrinal limestone is newer than the argillaceous slate. 1841. Doubts are expressed as to the encrinal character of the fossils. The relations of the argillite, limestone, and quartzite described as before. (7, pp. 54, 556, 560.) 1851. Because of the discovery of an upper stratum of slate (the upper schist described below) beneath which the limestone passes, it is decided that the whole of the argillite together with all the series to the top of the upper schist of the section on page 258 is Devonian, in accordance with the determination of the crinoids by Prof James Hall, who thought them to be of the age of the Onondaga limestone. (10, p. 298.) 1858. Bernardston and West Northfiekl are divided about equally by a north-south line between argillaceous slate on the west and hornblende- schist on the east, with the number for mica-slate entered on the area of the latter, but not subdivided from the rest. Limestone and iron ore marked. (11-) 1861. While the preceding history has dealt entirely with, the work of President E. Hitchcock, I understand, though it is not distinctly stated in the chapter in question, that the report of the Vermont survey was based upon the studies of Prof C. H. Hitchcock, and I have so indicated above Under the heading "Upper Helderberg Limestone" is given the best section yet published of the rocks in question, containing every bed of importance except one, and indeed one bed, D, a clay-slate immediately above the limestone, which does not exist. All the beds from the argillite up are made conformable, but no other indication is given in the chapter as to how much of the series is assigned to the age indicated in the heading. The upper quartzite (that is, the quartzite east of the limestone on the Williams farm) is suspected to rest unconformably upon the argillite in Ver- non, the intervening members being absent, and on page 598 of Vol. II this quartz rock, with the gneiss into which it grades, is assigned to the Devo- nian age, from its identity with the iipper quartzite of the Williams farm sec- tion. Attention is called to the fact that an upper schist resting upon the quartzite — it is described as a "distinct clay-slate, thicker-bedded and harder than most clay-slates" — is not elsewhere seen resting upon the quartzite. This I have not found to be true. (12, Vol. I, p. 447; Vol. II, p. 598.) 256 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. 1870. Prof. C. H. Hitchcock classes the argilHte as Upper Silurian, and the Bernardston series is "doubtfully referred to the Devonian." "Both above and below ai-e quartzites not of gi'eat thickness, and also slates." (13, p. 4.) 1873. Prof J. D. Dana pronounced the argillite to be an older formation lying unconformably below the other members of the series, as supposed by E. Hitchcock in 1833 for the argillite in relation to the limestone — an opinion receded from on the discovery of an upper band of slate — and by C. H. Hitchcock in 1861 for the overlying quartzite. From the close resemblance of the mica-schist and quartzite on the other side of the Fall River Valley to that on the Williams farm, he assigns to the age of the Helderberg these and the new rocks associated with them, viz, staurolite, mica-schist, hornblende rock, and feldspathic quartzite, which comes at last closely to resemble trae gneiss. He concluded that the Coos group of Professor Hitchcock, if correctly traced out, was the continuation northward across New Hampshire of the Helderberg rocks, and that the two bands of hornblende rocks marked upon Prof E. Hitchcock's geological map of Massachusetts as extending across the latter State, with their continuation southward in Connecticut, as described by Percival, where they pass beneath the New Red sandstone near Middletown, and emerge again west of New Haven, were possibly to be assigned to the same horizon. 1877. In 1877 Professor Hitchcock, first in abstract in the American Journal (16), and later in the Geology of New Hampshire (17), gave the result of a new investigation of the region in question, which diverges in a remarkable degree from his own and his father's conclusions and from those of Professor Dana. Accepting the conclusion of the latter that the argillite is an older and unconformable bed beneath the strata in question, he claims that the limestone "does not certainly dip beneath the quartzite," but "maybe a remnant of a once extensive deposit covering both the other formations mentioned, and what remains is in an inverted position," and thus is newer than all the other rocks of the region. This decided change of opinion caused a discrepancy in the volume already cited, as, in the earlier part, the series is stated to consist of several thousand feet of quartzite, limestones, schists, etc., and probably hornblende-schists. (17, p. 18.) BEBNAKDSTON SEKIES OF UPPER DEVONIAN. 257 lu his uiiitured conclusions (17, p. 428 ff) the gneissoid rocks which in tht' Vermont report are stated to appear to pass imperceptibly into the (piartzites, and to rest invariably upon them, and therefore to be newer (12, Vol. II, p. 598), are classified as Bethlehem gneiss, and thus assigned to the Laurentian. The band of this gneissoid rock crossing the State line west of South Vernon is marked on the map (17, PL XVIII) as Bethlehem, but in the atlas to the same volume, prepared later, it is colored as Coos qiiartzite, but left in the section at the foot of the sheet as gneiss. The liornblende-schist is next described, and its extension southward through Gill pointed out, and it is referred to the same horizon as the Shelburne Falls, Massachusetts, baud, and both are assigned on the scale of colors of the map to a position below the Huronian. The argillite is described as Cambrian clay-slate — that is, as Primordial Silurian. The remainder of the series on both sides of Fall River and east of the Connecticut thi'ough Northfield — quartzite, mica-schist, and staurolite- slate — is assigned to the Coos group, and this is placed, in the stratigraph- ical column at the end of the book, beneath the calciferous mica- schist, and to the whole is given a position in the Paleozoic series above the Cambrian and below the Lower Helderberg. Professor Hitchcock calls attention to one very important matter — the absence of staurolite, hornblende rocks, and feldspathic quartzite from the Williams fann section, and their presence, with the absence of limestone, on the other side of the narrow Fall River Valley. In his final column of the rocks of the State (17, p. 674) a thickness of 500 feet is assigned to the Helderberg, which is not clear if only the limestone is to be assigned to that age. During the same summer I visited this region with Professor Dana and we went over the ground between Bernardston and South Vernon together, examining the Williams farm section carefully. I then called his attention to the lower stratum of schist beneath the limestone, and soon after detected fossils in the quartzite over the latter. These we found to be quite abund- ant. On his return Professor Dana gave the results of this examination and controverted the conclusions of Professor Hitchcock in a somewhat polemical paper (18), giving in some detail the earlier opinions of the latter, and deciding that, since the quartzite was both fossiliferous and conformable upon the limestone, the two could not be brought into their MON XXIX 17 258 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. present position by faulting or inversion, and since the schist occurs both below and above the limestone in apparent conformity, one or other of the beds must be newer than the latter. Both of the members of the Coos gi-oup being thus proved to be of the same age as the limestone — the quartzite by containing fossils and the schist by conformity with the latter and with the limestone — numerous examples of visible and conformable conjunction of the hornblende-schist and gneiss with both the mica-schist and the quartzite in the area between Bernardston and South Vernon are given, together with instances of the passage of the one rock into the other, in proof that these rocks are there all of Helderberg age. Order of succession of rocks in the Bernardston region. Hitchcock. Dana. Emerson. •i % ■3 P-i Lower Helderberg limestone. 1. u IS a Fossiliferous limestone. Mica-schist. Staurolite-schist. Hornblende-schist. Quartzite. Feldspathic quartzite. Gneiss. [The order of the above beds was not fixed. The list in- cludes all of the preceding list ex- cept calciferous mica-schist and ar- gillite]. i 1 a S 1 1. Garnetiferous mica- schist, becoming staurolltic east and north, with four or more beds of amphibolite. 2. Quartzite, con- glomeratic at base; becoming gneissoid east and north, with mag- netite and crinoi- dal limestone beds carrying Chemung fossils. Calciferous mica-schist. Staurolite mica-schist. Ph 2 CD o O Mica-schist, often stauroliferous. Quartzite. Cambrian slates ^ar- gillite. i 1 i § 1 Bethlehem gneiss. Unconformity. Argillite. a TJnconformity( ?). 3. Argillite. 4. Calciferous mica- schist. BERNARDSTON SEKIES OF UPPEK DEVONIAN. 259 I'rot'ossor Daua says (18, p. 381, note) that Pi-ofessor Hitchcock's "later conclusions have been influenced by his faith in the lithological test of ereolosfical ae:e and his unbelief in the existence of s'neisslike metamoi'- phic rocks of later date than the Cambrian;" which I can not think wholly just, since the lattei', upon his atlas map, classes the wholly gneisslike band upon the north line of the State and the quartzite, together with the stauro- lite-schist — indeed, all the rocks in question except the hornblende-schist — with the Coos group, and places this among the Paleozoic rocks in his final scheme at the end of the Geology of New Hampshire, Vol. II. The dis- crepancies are, however, sufficiently great between the interpretations of the two authors, and I have placed in parallel columns their views and my own of the true order of succession of the rocks in the area in question. See also the section on page 285. 1883. Mr. Whitfield (19) concludes, from an examination of the fos- sils, that the limestones may be Middle Silurian; the shales (i. e., the thin-bedded, rusty quartzite immediately above the limestone) were most probably Middle Devonian. 1890. The results reached by the writer were published in the Ameri- can Journal of Science (20); but as several errors unfortunately escaped his notice, the substance of the article is reproduced below. in a corrected form. THE UPPER DEVONIAN AGE OF THE BERIVARDSTOIV FOSSILS. Prof John Mason Clarke has been so kind as to reexamine the fossils, and as he is familiar with the locality his conclusions may be considered as settling the age of the series with a large degree of probability. All the fossils of the upper bed of shaly quartzite occur also in the upper part of the the limestone, and it is not possible to separate this continuous limestone mass. Professor Clarke writes me as follows (January 28, 1895) : The impressions left by the fossils are so distorted, obscure, and closely packed together that a little imagination can construe them into species of all sorts of ages, but I feel reasonably secure of the following points : First. The prevalence of a large spirifer, with moderately strong dental plates, like S. granulosus Conr. of the Hamilton group, or S. disjunctus Sow. of the Chemung. Second. The presence of Microdon, probably abundant among the distortions, but recognized in a single instance. The species is uncertain, may be Hamilton, Ithaca, or Chemung. 260 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS Third. A well-defined Palseoneilo, witli coarse surface striae (I obscured the im- pression somewhat by taking a squeeze from it). There are species throughout the Middle and lower Upper Devonian of a similar character. Pourth. Well-defined Oamarotcechias, like Rhynchonella sappho and B. congre- gata, too obscure for specific identification. Fifth. A recognizable fragment of a large Actinopteria. Sixth. The "tricircled encrinite" of Eaton and Vanuxem. Though only a crinoid column, this is, 1 think, the safest horizon-marker visible; at least its association with the other fossils mentioned helps to a close approximation to the age of the fauna. Its stout calcareous body has often preserved it from the distortions which have wrecked the associated fossils, and its characteristic expression as it occurs in the lower Chemung and Ithaca beds is well reproduced here. Thus I believe we are justified in assuming this fauna from the shaly quartzites to be of Upper Devonian age; no change from the old conclusion, but a better fortified opinion. DESCRIPTIOlSr OF THE REGION. The terrace sands of the Connecticut River are narrow upon its western side, where the river crosses the State line, and they continue with little increase of width for 4 miles southwesterly, and then, as they enter Ber- nards ton, their boundary upon the older rocks turns abruptly west and runs for 7 miles a little south of west, past the village of Bernardston and along the north line of Greenfield. (See PL IV.) Bernardston village stands just in the middle of this line and at the mouth of a narrow valley, up which a lobe of the alluvial sands reaches northwardly for nearly 2 miles. On the west this valley is bounded by the high ridge of West Mountain, made up of the contorted argillite, which stretches in a narrow band far north across Vermont and disappears below the river sands on the north line of Greenfield, appearing again only in the limited outcrop just west of the village of Whately, 15 miles farther south, and in one newly discovered at the mouth of Mill River. Everywhere the slope of West Mountain shows only the black argillite, except in a single band back of the house of Mr. F. Williams, a mile north of the village, where, apparently resting upon the argillite, occurs the fossiliferous series. The section has a width going up the hillside on the line of dip of only 3,445 feet. The outcrops of the argillite to the north and south show that there can be only a very limited amount of the newer series preserved upon the hillside, while the heavy accumulation of till generally prevents one's seeing its limits or its contact 4 >v BERNAEDSTON SERIES OF UPPER DEVONIAN. 261 upon the rock below. It approaches the argilhte quite closely upon tlie west, and in the line of strike can not bo more than 9,850 feet long. Over against West Mountain on the east, across the narrow valley of Fall River, rises a range of hills, bounded on the south and east by the ten-ace sands, which is composed of a similar series of rocks in similar succession. The principal difference between the two is that on the east a dark hornblende rock, often massive, takes its place in the series, while the limestone and magnetite beds of the typical section are present only in traces or in altered form, and all the other members are somewhat more metamorphosed. Staiu'olite here occurs in the schists, feldspar crystals and biotite in the quartzites, and they are thrown into complex folds and greatly faulted. They lie, in fact, along the center of the great syncline of the Connecticut Valley, which is an area of maximum disturbance of the rocks quite across the State. These discrepancies become less important when it is noticed that hornblende exists in considerable quantity directly above the Williams farm hmestone, and the second bed of the same limestone in South Vernon is encased in hornblende-schist, and several of the hornblende-schist beds can be proved to be altered limestone beds. Across the river in Northfield the white saccharoidal quartzite extends to the base of Northfield Mountain, and is there bounded by a north-south fault, while only a single outcrop of schist is exposed. THE REIiATION OF THE BERlSrARDSTON SERIES TO THE ARGILLITE. It was originally assumed by President Hitchcock that the argillite and the schists of this series were conformable. Prof J. D. Dana,^ finding the argillite about a half mile west of the limestone to have a much higher dip, decided that they were unconformable to and much older than the upper series, and this conclusion was accepted by Prof C. H. Hitchcock.^ In tracing the distribution of the quartzite, I have given five localities where the boundary of the quartzite and argillite is well exposed (p. 273), and I could increase the num.ber, and in each case there is apparent conformity and a uniform passage from the common argillite into argillite with minute garnets and minute biotite spangles, fine-grained black quartzite grading 'Am. ,Joar. Sci., Vol. VI, 3d series, 1872, p. 343. ^Geol. New Hampshire, Vol. II, 1887, p. 433. 262 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. into coarser quartzite, and conglomerate. The argillite is extremely corru- gated and often cleaved, and observations of dip a rod from the contact are of no value in settling a question like this. THE WI1.LIAMS FARM SECTION. THE F08SILIFER0US lilMESTONE. PROOF THAT THE WHOLE SERIES IS DEVONIAK. The long band of the recks of the Bernardston series along the lower slope of West Mountain has been brought into its present position by extensive dislocations, and is plainly cut off by two transverse faults which run approximately in the brook gorge north of the limestone and in the larger gorge of Fox's brook half a mile south. The area between, contain- ing the fossiliferous limestone, is the one here described. (See PL IV and fig. 15.) Passing up the hillside back of Mr. Williams's barn, the first bed and the upper one on the section (fig. 16, p. 264) is a dark muscov^'te-schist (1), which is exposed in a single small quarry and separated by a depression which runs with the strike, and which I have assumed in the section to be occupied by the same schists and to have been formed by their erosion. The outcrops are almost continuous across the quartzite (2) and the lime- stone (4) which follows to the second outcrop of schist, where a similar depression separates the latter from the second band of quartzite, which I have in like manner supposed to be occupied by this schist. Section of the Williams farm rocks. Feet. 1. Garnetiferous mica-schist 73 2. Micaceous quartzite and conglomerate 443 3. Magnetite, maximum 3J 4. Limestone 19^ Quartzite concealed beneath the limestone ( ?). Fault. v. Mica-schist 115 2'. Quartzite and conglomerate, if conformable with the mica- schist 666 Argillite. (The beds below the fault are a repetition of those above.) The argillite (fig. 16, west end). — Beginning nearly a mile northwest of the Williams house, and just north of the point where the road over West Mountain bends sharply west, a long ridge of the typical, excessively con- BBRNAKDSTON SERIES OF UPPEK DI<;VONIA]Sr. 263 toi"te(l arg'illite oxteiids northerly. Eastward, a drumlin conceals its contact with the newer rock. It is probably a conformable contact of the argillite and the (juai-tzite; as I have found it so everywhere in the region. The -western outcrop of the mica-schist (!'). — Where the series outcrops for the first time after crossing the di'umlin a small area of the mica-schist of this series has recently come to my notice. It is a garnetiferous mica-schist, A, Main Quarry. h. Thickest Ma^neiiis.. c, Fxcav^iof? connecting Uinestone and Quat'tzitc. d, Excavation CKposing fault of Mica-Schist against Limestone. e, Same as d. if^ Contact of Quartzito on Argi/iite. ARGILLITE— Outcrops /narked with darker shade. Fig. 15. — Map of Devonian rocks on the Williams farm. like the more eastern ontcrops, and it lies plainly in a small syncline of the qiiartzite 10 rods south of the western end of a row of great chestnuts which crown the hill. The western exposures of the qiiartdte (^'). — The discovery of the schist just described makes plain the structure of these quartzite outcrops with their western dip. As the schist is in a small syncline, the quartzite makes a corresponding anticline on the east of this outcrop of the mica-schist. The 264 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. rock is dark-gray quartzite, at times a conglomerate, weathering very rough, with strike and dip very irregular and uncertain, with many slight slips and crushings — indeed, often completely brecciated and recemented with limpid quartz. Locally it passes into a black siliceous slate by the micro- scopical development of biotite and the accumulations of coaly matter. A APirct: lurg^Otsfmlft. Fig. 16. — Section of Devouian rocks from the Williams farmhouse 250 rods northwest to the sharp hend in the road over ^\''est Mountain, along the section line on map, fig. 15. few scales of the former mineral can be seen with the lens. Up the hillside from the limestone along the line of dip, two small ledges of the rock appear, as may be seen from the section, widely separated from each other and from the rocks above and below. It is not diificult to find among tlie less crushed portions of each ledge ■2 jc> a; jc = emer- ald green. It = olive, a — yellow; extinction at 21°. Leucoxene is in aggi'e- gates of grains nearly colorless or with red-brown centers ; rutile occurs in square prisms. Fine, large, pale-reddish titanite crystals show ^^ositive bisec- trix and axial figure parallel to co P o6. The whole colorless background is made up of limpid granular plagioclase, often twinned but more often free BEENARDSTON SERIES OF UPPER DEVONIAN. 291 from twinning' or cleavage, and then sliowing the strongest concentric extinc- tion. Extinction in twinned i)ktes, 14°-34°. Quartz is scarcely present. 13. Aniphibolite from South Vernon, 10 rods north of limestone at E. Gr. Scott's. A thin-fissile, greenish-gray rock, showing with lens many scales of black biotite and rarely a sharply defined, light-brown prism of epidote. In section nearly the whole field is covei'ed by a tangled network of hornblende blades which show quite marked pleochroism. The colorless background is for the most part feldspar, rarely showing twin striation, and, as so often happens in these hornblende schists, wholly fresh and without cleavage. Many scalariform or coraloidal grouj)s of leucoxene occur, rarely with a trace of black ore at center, but each separate crystalline grain itself red-brown at center and colorless superficially. No other ore occurs. THE MICA AND AMPHIBOLITE SERIES. 14. Mica-schist from Bernardston, Williams farm, from the bed of schist west of the limestone. A dark-gray to black, very fine-grained, even- bedded slate, with its glistening surface pitted here and there by hemi- spherical hollows, from which small red dodecahedral garnets have fallen and marked by minute white spots of shining muscovite scales just visible to the eye, which often appear in fine double lines sharply parallel to each other and inclosing a narrow dark center. They appear thus like minute chiastolites. Under the microscope the rock shows a fine, scaly, colorless ground, dusted abundantly with coaly matter, and with polarized light showing^ in abundance minute muscovite scales and needles. These are embedded in a ground which shows aggregate polarization and is partly apolar and apparently opal. Kaolin could not be detected, nor "clay-slate needles." The much fissured garnets have often a black boundary, from the accumu- lation of the coaly matter expelled from the space they occupy, and within this a broad decomposition band of chlorite In twisted scales, which often extends nearly to the center of the crystal. They contain large grains of quartz Irregularly arranged. The centers of the minute chlastollte-like forms mentioned above prove to be small, very Impure, transversely placed blotltes with flat sides and raveled ends, having on each of their long sides bands, broader than them- selves, of clear muscovite scales placed at right angles to the broad faces of 292 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. the biotite against which they rest. The large amount of coaly matter in the centers of the biotites indicates that the rock was more carbonaceous when the biotite was formed than at present. The biotite incloses garnet, which was thus first formed, and the muscovite has also folded round the garnets, forming small cups from which they fall, and has also, as above, arranged itself symmetrically to the biotite, and is thus shown to have been third in order of appearance. Leucoxene appears in yellowish-white grains more rarely than in the argillite. Staurolite appears in some abundance in stout, shapeless masses nearly large enough to be seen with a lens — red-brown by reflected light, nearly oi^aque by transmitted light. They polarize distinctly, showing in the thinnest places a mosaic of bright red and blue, indicating twinning, and also traces of the square and prismatic sections of single crystals. Some crystals giA^ng the proper angles of staurolite are white by reflected light, from decomposition, and this I have seen macroscopically in the schists around Vernon. Here the staurolite was removed in every degree from the network of quartz, until only a few brown grains remained, and at last only a cellular network of white quartz. 15. Mica-schist from above, and east of the quartzite, Williams farm. A dark-gray, fissile muscovite-schist, with pimpled surface of somewhat coarser grain than the preceding. ' • The ground is exactly the same colorless, scaly, coal-dusted mass as in the lower schist, and presents with polarized light exactly the same appear- ance upon a slightly larger scale. It difi^ers by the development of the transversely placed biotite into quite large crystals, visible to the eye when the rock is broken across the bedding, and these crystals form most of the pustules which rise on the cleavage surface of the plates. They are bounded on the basal planes, as in the lower schist, by a line of larger muscovite plates, but this is not at all so constant as in the former case. Scales of muscovite are often intercalated in the biotite with magnetite and pyrite. The mineral is a true biotite (meroxene), with p if^Z-:^> w. ^"5 Millers /f/ver — Fig. 21. Sketch of rocks at mouth of Millers Elver, looking northeast from B ou map, flg. 20. Scale, 1: 2000. H. S.= hornblende-schists; Q. = qDartzite; GN. = gnei33; GE.=granite; M. S.=mica.schist; B. GN.=hiotite-gnei8s; Older Gx.— Mo.nson Cambrian gneiss. unbroken exposure of rocks of great interest, which I have associated with the Bernardston series, at times with much confidence and at times very doubtfully. It becomes, in fact, a question how far the original sediments may have been different from proximity to the gneiss instead of the argil- lite, and how far, also, the immediate presence of the gneiss during the thorough metamoi-phism of the sediments in question may have conduced 296 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. fer:i;u:^/?/fl>li£Wy^5 to a I'esult different from that reached where the underlying mass was a nonfeldsj)athic schist. I am inchned to give great weight, perhaps the greatest weight, to the occurrence of the same succession in beds of about the same thickness — quartzite, mica-schist, limestone — the latter changing into hornblende rock, and to the tracing of the beds into such close proximity rather than to the exact texture of the beds themselves. The following description will emphasize the differences in the latter regard: Below the fall at the mouth of Millers River, and on the north bank of the latter, at a small crevice in the cliff, a fault is plainly seen (fig. 21), the biotite-hornblende-gneiss which formed the apron of the dam dipjjing 10° W., against a flinty quartzite which dips 40° W. Following the j.^ E. outcrop along the river to ^i^^ffoApy its north end, at a point 164 feet south of "The French King,''^ we find a marked promontory — an island except at low watei- — of the same jas- pery quartzite, with high westerly dip, which is sep- arated by a narrow dike of coarse granite from the much older horizontal Becket gneisses which crop out in the bank of the river and continue for a long distance north. The imconformity is indicated in the above figure (fig. 22). The rock at the promontory is a very peculiar quartzite, very thin-laminated and corrugated like the grain of gnarled oak. Layers, which sometimes swell to 10""", of black, flinty quartz, wavy and interrupted, alter- nate with bands of white to oil-green, compact quartz, producing a structure which resembles the banding of some eruptive rocks more than ordinary bedding. The rock can be followed south for 600 feet along the bank. The ribboned quartzite changes into a coarse mixture of blue, greasy quartz 'A great bowlder of Triasaic conglomerate which lies in the middle of the river at the head of the rapids, and derives its name, according to tradition, from the fact that the bateaux of the French and Indians, during the French wars, were stopped here by the rapids, and one adventurous French- man pressed on to this rock and broke a bottle of wine over it, claiming the country in the name of the French King. COA/iSE 6RAN/r£ -^l ~ L I6HT- G/fAy 6A/S/SS Fig. 22.— Section on east bant of Connecticut Kiver above montb of Millers Eiver at A on sketch map, fig. 20. BEENARDSTON SEKIKS OF UPPER DEVONIAN. 297 and telclsj)ar, very rusty aud carrying pyrite and g'aleua, and rarely niusco- vite in broad si-ales. Going 33 feet along the strike, one finds the rock changing to a massive, dark leek-green hornstone, which continues a long distance, becomes in places black, and assumes a small columnar structure, and at last returns to the coarse mixture of quartz, flesh-colored orthoclase, aud muscovite, the latter often in 2:)lates about an inch across — the whole coarsely but distinctly bedded in laminge 1 to 2 inches thick. This is succeeded in ascending order by a well-developed, coarse muscovite-schist 13 feet thick, which dips beneath a bed of very siliceous limestone about 40 feet thick, very rusty externally, in the interior white to flesh-colored at base, but soon becoming dark-green to black above, and very hornblendic. In places it is a pure amphibolite, but it is generally mottled with white calcite. It is cut by two dikes, 3 to 7 feet thick, of coarse granite. Then begins a great bed which seems to rest uj)on the hornblende-calcite rock, but the exposure leaves this indistinct. This bed begins at base as a greenish, apparently calciferous quartzlte (it rusts deeply), and makes the mass which projects into the Connecticut at a point just north of the mouth of Millers River. At base some parts are conglom- eratic, quartz pebbles one-half to 1 inch long occurring. This rusty layer is about 20 feet thick. Then a thin layer of amphibolite, like the other, caps the quartzite for a short distance (72 feet) along the water's edge, and the latter rock, the quartzite, runs on in great undulations for 656 feet toward Millers River, its average strike agreeing closely with the course of the Connecticut at that point (S. 60° W.) and its dip being 42° W. The high, bare cliff's give almost unbroken exposures between the two exposures of the fault at the extremities of the section, and leave the stratigraphy uncertain at only one point. The upper quartzite is thin-fissile in its upper layers, bluish at times, and repeating all the flinty varieties seen at "The French King" (16 feet). At the great point just north of the mouth of Millers River, where the shore-line swings round to the east as one passes up the bank of the latter stream, this grades below into a perfect feldspathic gneiss of medium grain, with a little greenish mica (20 feet), which passes below into coarse granitic gneiss or a gneiss breaking in laminae nearly an inch thick but composed of the coarse orthoclase and large muscovite scales of a common coarse granite. The muscovite scales are often an inch broad, and are generally in the plane of foliation (30 feet). 298 GEOLOGY OF OLD HAMPSHIEB OOUFTY, MASS This stratum is followed below, just at the junction of the rivers, by a heavy bed of a very coarse, very micaceous muscovite - schist. This mica-schist incloses a great number of lenses of coarse granite, its laminae separating to inclose them, and the strike of the rocks continues closely parallel to the course of the Connecticut. This stratum passes beneath the water at the place of junction of the two rivers, and the granite lenses rise in twelve small islands which stretch across the mouth of Millers River (105 feet). The mica-schist changes suddenly below into a dark, much-jointed and yet fissile biotite-gneiss. The biotite is bronze-colored, but green superfi- cially. This is followed finally by the basal quartzite, containing at first bands of coarsely feldspathic quartzite with scattered red garnets like those common in granite, and broad flakes of muscovite, and with quite large gran- ite lenses. Below it is for the most part a green to flesh-colored, compact quartzite with feldspathic and biotitic layers, and resembling closely that at the north end of the exposure at "The French King," and faulted, as already detailed, against the older gneiss just below the bridge over Millers River. (See fig. 21, p. 295.) The order from the fault upwai'd is, thus, at both ends: (1) quartzite, (2) mica-schist, (3) amphibolite and limestone; and it is doubtful whether the order continues upward, (4) rusty quartzite, (5) amphibolite, or whether the two upper are repetitions of the lower members of the series. Crossing Millers River, the fault line runs through the high Mine Hill west of the village of Millers Falls, and, especially the band of schist and gneiss impregnated with granitic material, makes the crest of the hill. This band is greatly brecciated and its fissures are filled with magnetite, which has suggested the name, while on the western slope a greenish compact quartzite or hornstone caps the feldspathic beds. Southward across the Montague plain all the outcrops are to the east of the line of junction of the two formations, and thus lie in the older gneiss, until, on the southern border of the great sand plain and just northeast of the village of Montague, there occurs a great mass of the same gray to pale-green, greatly-jointed and brecciated quartzite, quite massive and hornstone-like in texture. A few rods to the east, across the railroad, the older gneiss rises above the sands, dipping beneath the quartzite. South and west everything is covered by the Triassic sandstones, but on the north IJEKNAKDSTON SERIES OF UPPER DEVONIAN. 299 faco ot" the hill west ot" Montague a coarse gneiss showing traces of con- glomerate structure rises out of these sandstones and is most nearly allied to the coarser rocks at the mouth of Millers River, with which I have asso- ciated it on the map. (See p. 362.) Along the fault line northward from the mouth of Millers River across Nortlifield, the older gneisses rise from beneath the sands in many places up to, but not beyond, the line of the fault, while the few outcrops of the Bernardston series approach the same line on the west until, ojjposite Northfield village, in L. A. Moody's lane, the Devonian quai'tzite, identical with that west of the river on the Williams farm, here greatly brecciated and its fissures filled with hematite, approaches very near to the older rocks, and the same thing is true northward along the road which skirts the moun- tain to and beyond the State line, where the fault enters the older rocks, as shown on the map, PI. XXXIV. The quartzite, greatly brecciated, and veined with hematite, can be traced in several places into near proximity to the older series. It is a fine- grained rock which shows no signs of derivation from the varying members of the older series against which it successively rests. It fails to conform to them in dip and strike, and these facts, combined with its thorough breccia- tion along this whole line, make it probable that the Bernardston series is bounded on the east by the extended fault which has been traced north and south from Millers River. CHAPTER X. THE AMPHIBOLITES DESCRIBED IN THE PRECEDING CHAPTERS. A]SrALXSES AND SECTIOKS. I have in the preceding chapters described a great variety of amphibo- Utes, and have assigned most of them with more or less confidence to the list of altered sedimentary rocks. Forms which are associated with basic eruptive rocks or retain any traces of the structure of those rocks I have classed with eruptive rocks. Varieties which form selvages to limestone beds, or are in every way closely associated with limestones and inter- bedded with the schists in which they occur, I have considered to be derivatives from impin-e limestones. Without recapitulation, I may call attention to the discussion of the amphibolites associated with Algonkian limestones, on pages 29—30, and of the origin of the Chester amphibolite, on pages 147-155, and to the description of the. amphibolites of the Hawley series, on pages 166-169, and of those more certainly derived from lime- stones in the Conway schist, on pages 189-196, and in the Bernardston series, on pages 275-282, 290-294, 297. I bring together here for com- parison all the analyses of amphibolites from the region, and a series of sections commencing with forms which are certainly sedimentary and end- ing with those concerning whose origin there Avill be difference of opinion. A curious porphyritic structure which is common in the amphibolites is also briefly discussed. I have described many beds of amphibolite that may, with more or less certainty, be derived from limestone. I have perhaps given too much weight to this mode of derivation, and may have included beds of igneous origin. It is very striking how small the chemical variation is in rocks differing widely in age, habit, and probable origin, and the close agreement of all these analyses with that of an average disbase will not escape atten- 300 PLATE Y. 301 PLATE V. SECTIONS OP AMPHIBOLITES DERIVED PROM LIMESTONES. Silurian and Devonian age. Drawn with lower niools to show the pleochroism of the hornblende. The upper nicol is used to bring out the outlines of the colorless mosaic. The plane of polari- zation is parallel to the long side of the plate. The hornblende crystals are generally full of colorless grains like those of the ground. All show remnants of calcite. x 20. Pig. 1.— Calcareous garnet-amphibolite. Conway schist, Plainfield. From the base of the large "anvil," figured in PL XXXIII and described on page 191, and formed by the metamorphism of the impure limestone which still makes the shaft of the anvil. The rock has the aspect of a coarse hornblende-schist, but effervesces with acid. The coarser-grained portion of the color- less mosaic is clastic, identical with that of the adjacent mica-schist, and the garnets have the same symmetrical arrangement of the coaly impurities. The finer part of the colorless ground is plagioclase. One large grain is marked by triclinic striation and most of it is crushed. The biaxial character of many grains could be determined. Extinction 37°, indicating anorthite. The ragged hornblende plates inclose many colorless grains, generally quartz or calcite, but sometimes small colorless zircons with deep halos of darker color. The dull portions, heavily dusted with carbon grains, are remnants of corroded calcite. Swarms of leucoxene grains surround ilmenite. (See Analysis I, p. 303.) Pig. 2.— Amphibolite. Whately. Prom bridge west of the Whately Hotel. A black, massive amphibolite; forming portion of limestone bed in Conway schist and folded into argillite. The long hornblende blades are often centrally brown, with colorless ends, and loaded with transverse black bands which send out comb-like teeth parallel to the vertical axis. There is some biotite. Red rutile surrounds black ore, and small colorless zircons (?) appear, surrounded by dark halos. There is a mosaic of feldspar without twinning and generally without cleavage. (See pp. 192, 196.) Fig. 3.— Amphibolite. Bernardston, near R. Park's. Prom the Devonian Bernardston series. A black, massive rock, made of broad, stout, interlaced blades of hornblende. These blades grade into radiate tufts of hornblende needles. The hornblende crystals inclose many color- less grains, often of branching and irregular forms, resembling the grains of titanite which surround some of the black ore masses in the slide. Some of the grains of the colorless mosaic are twinned and have generally the small extinction angle of albite. Others show secondary growths around rounded centers. One fine fibrous and punctate fragment seems to be organic, resembling a brachiopod shell. It is too small to show in the drawing. (See Analysis VII, p. 303, and comparative discussion, p. 275.) Fig. 4.— Garnet-graphite-amphibolite. Bowlder from Leverett, but coming with great probability from one of the coarser Bernardston beds or from the great Guilford bed in the Conway schist, figured on PI. VI (fig. 2). A complete block of massive amphibolite. The stout interlacing blades of hornblende contain in their meshes little granular mosaic, which decomposes readily and gives a beautiful surface. The fresh surface effervesces. A few gar- nets appear. The striking peculiarity, indicating the derivation of the rock from a graphitic limestone, is that the surface of very many of the hornblende crystals show shining scales of graphite, which only rarely appear in the photograph, x h 302 us GEOLOGICAL SURVfY MONOGRAPH XXtA. PL.V. X20 X20 X20 >.ii CHARLOTTE F. EMERSO N, FECIT. ""SadwiT^lIieEaEiIlfliniTCo' AM PHI BO LIT ES ANALYSES OP TYPES OF AMPHIBOLITE. 303 tiou. No. IX is, however, certainly, and Nos. I, IV, and V are almost certainly, derived from limestone. There is therefore no reason from the chemical side why they should not all be so derived. Analyses of the amphiholites described in the preceding chapters. [Analyst, L. G. Eakius.] '- I. II. III. IV. V. VI. • VII. VIII. IX. SiO.. 51. 38 1.07 18.01 45.48 .77 19.43 trace .13 6.58 trace .01 10.66 11.08 .11 2.28 3.17 .14 .20 51.56 1.97 14.82 49.86 1.58 15.50 49.16 1.03 16.43 trace 3.92 7.19 .23 .02 9.21 8.19 .41 3.70 .45 .16 48.53 .51 16. 35 51.72 1.39 16.51 47.56 1.24 16.13 trace 1.80 9.39 .08 trace 6.67 9.21 1.58 2.52 3.51 .21 55.64 .50 16.27 TiO. AI2O3 Fe-Os FeO 3.30 8.53 .19 trace 6.27 5.08 .18 5.34 .56 .18 4.30 7.21 trace trace 7.09 7.36 .17 4.21 1.47 .09 2.99 8.01 .07 trace 8.89 7.79 .72 3.26 1.51 .11 2.03 10.52 .17 trace 9.83 9.71 .32 1.36 .79 .07 1.72 9.56 trace trace 8.89 6.58 .34 2.74 .51 .23 1.22 7.20 .28 MnO CaO 9.23 5.58 .19 .91 3.11 .23 Mo-0 KcO Na.O H,0 PcOb 100. 09 100. 04 100. 25 100. 29 100. 10 100. 19 100. 19 99.90 100. 36 I. Heath; W. M. Sanford's. Porpliyritic amphibolite in Goshen schist. II. New Salem. Amphibolite associated with steatite. III. Leverett; gothic house east of village. Massive, coarse, altered diabase. IV. Whitmores Ferry ; Sunderland. Thin, shaly, aphanitic amphibolite of Conway age project- ing through Triassic sandstone. V. Guilford, Vermont ; shining-black, flaggy amphibolite; long bed in Conway schist. VI. Worthington; in Hawley schist ; nearly pare, matted, black hornblende needles. VII. Bernardston; E.Park's. Black, heavy, massive hornblende rock. Devonian. VIII. South Leverett. Ligniform, deep-green amphibolite. IX. Goshen ; base of the "anvil" formed by solution of the block of impure limestone, and thus certainly derived from limestone of Conway age. (See p. 191.) In the accompanying plates (Pis. V, VI) I have brought together types of all the amphibolites described in the preceding chapters for com- parison with the table of analyses given above. The sections on PL V are certainly derived from limestone. The first two sections on PL VI have probably the same origin. The third section, from the Hawley schist, and the fourth, which is from a bed adjacent to the great Chester amphibo- lite, are of uncertain origin. 304 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. POEPHYRITIC CHARACTER OF THE AMPHIBOLITES. The hornblendic bands of tlie Bernardston series are often pseudo-por- phyritic, especially west of the Mount Hermon School. This structure is widel3^ distributed in older amphibolites from the Hawley to the Conway series, both inclusive, especially in the Gruilford and Heath amphibolites and the large upper band of the Hawley schist (see p. 166). It reappears again in the Conway schists in Grranville, and is seen in the most striking form in South Monson (see p. 249). The dark surface of the rock is inter- rupted by white spots 2-10'"™ in length, more or less angular and of some- what uniform size. A quite close inspection will often leave the impression that they are formed simply by the expulsion of the hornblende needles from the area, and are a portion of the granular base of the rock, but a bright light Avill show at times the flash of a common cleavage over the whole or half of the surface. With polarized light tlie same effect is produced. A simple mosaic of feldspar grains appears, but by using a very low power it can generally be seen that the groundmass is held together by a single large feldspar growth, so crowded with foreign bodies that it can hardly be separated. I compare these feldspar growths to those described on page 287 in the Bernardston quartzites, or the small porphyritic albite crystals in the Hoosac schists, and consider them the earlier generation (as compared with the hornblende needles) in their present position. They often include minerals of early growth, as biotite and dolomite (which are now wanting in other parts of the rock), as well as the common groundmass, and have by their early presence prevented the iron-bearing mineral from occupying their place. They are now often saussuritic, made up wholly or largely of highly refringent epidote, or zoisite grains, very possibly as the result of a paramorphic change at the time of the development of the hornblende. The whole process is one more intelligible as occurring in a calcareous red sandstone than in a metamorphosed diabase, and it is very common in the amphibolites, which occur in thin, extended, conformable sheets, grade more or less into limestone, and show no tendency to form sei"pentine and steat^'te, and it is wanting in the gabbro-like beds and in the great Chester amphibolite, which- is associated with olivine and enstatite rocks, serpentine, steatite, and emery, and which may thus be derived, at least partly, from beds of distinctly eruptive origin. PLATE VI. 305 MON XXIX 20 PLATE VI. SECTIONS OF AMPHIBOLITES PROBABLY DERIVED FROM IMPURE LIMESTONES. Silurian. Drawn with lower nicols to show pleochroism. The upper nicol is used to bring out the outlines of the grains in the colorless mosaic. The plane of polarization is parallel to the long side of the plate. The hornblende grains are generally free from grains of the plagio- clase. X 20. ■piG. 1. — Magnetite-amphibolite. Conway schist, Whately. From the north end of the large band near house of Mrs. M. Taylor. A black, slaty rock, easily mistaken for a black shale. A rock of very fine and even grain. Many of the ore grains are surrounded by small groups of titanite grains, many grains of which are scattered evenly in the whole field. A narrow vein filled with large and long plagioclase grains, like the ground, crosses the slide. (See Analysis IV, p. 303, and for description p. 194.) Fig. 2. — Rutile-epidote-amphibolite. South line of Guilford, Vermont. A shining-black schist, slightly larger needles, black and lustrous, appearing among the others. A limpid granular mosaic, in small quantity, no multiple twinning, rarely cleavage, one probable determination of albite. Hornblende ragged-bordered, but without inclosures, with strong pleochroism and cleavage. Much deep-red brown rutile clustering around black ore grains. Considerable pale- yellow epidote. (See Analysis V, p. 303, and for description p. 195.) Fig. 3. — ^Epidote-amphibolite. Hawley series. Heath, near E. Gleason's. With large porphyritic hornblendes and feldspars. A fine-grained, limpid feldspar mosaic, thick set with minute but quite well-formed hornblende and rounded epidote grains. The large hornblendes show remarkably strong pleochroism when viewed with a lens without the intervention of nicols, and twinning, both of which properties are indicated in the figure. They are centrally filled with minute foreign bodies. The feldspar crystals are centrally filled with highly refringent, slightly yellow grains of epidote, and the same are scattered through the ground. The distinctly bounded feldspar crystals are mostly broken into a mosaic. One shows trace of carlsbad twinning, but no cleavage or multiple twinning can be seen. The abundant grains of magnetite show no trace of change. The red grains are much rusted ankerite. (See p. 166.) j'lQ. 4. — Amphibolite. Chester. From the cut nearest to the station. A black rock banded with white layers, which are thin and interrupted, the mass of the rock made of shining, jet-black needles just visible to the eye. Under the microscope the rook is very fresh, the hornblendes are in long, ragged-ended, parallel blades with few inclusions and strong pleochroism. Extinc- tion 21°. The coarse, limpid ground mosaic is made up of quartz and albite grains, polarizing brilliantly and hardly distinguishable from one another, except that the quartz gives the black cross and the feldspar is positive and biaxial. This mosaic resembles exactly that of the adjacent sericite-schist. There is no trace of any other mineral except these three. (See pp. 97, 160.) 306 U S GEOLOSICAL SURVEY MONOGRAPH XXIX. PL. VI. aackoK ftWilhehna titho.RPfi Co AMPHIBOLITES _ CHAPTER XI. THE ERUPTIVE ROCKS. INTRODUCTION. The species of igneous rocks occurring within the area of the crystal- line schists are: 1. Granite, in the strict sense, or biotite-muscovite-granite, the most widely distributed. 2. Granitite, or biotite-granite, generally porphyritic. 3. Pegmatite, or muscovite-granite. 4. Albitic granite in secondary veins in the pegmatite, remarkable for their content of rare elements. 5. Aplite. 6. Quartz-gabbro. 7. Tonalite, or quartz-diorite, wholly or in part derived from No. 7, and with it forming the syenite of President Hitchcock. 8. Diorite. 9. Diabase. 10. Cortlandite. Within a square twenty -five miles on a side, with Northampton at its center and its eastern line along the foothills on the east side of tlie broad Connecticut Valley, in Belchertown and Pelham, the country consists for the most part of large areas of granitic rocks of the above types. Where schists cross the region they are contorted and granite-impregnated, and rest upon the granite in separate sheets, often of small size, or narrow bands, and all, down to the smaller fragments, retain their dip and strike, even when surrounded on all sides by the massive rock. A large portion of the area outlined above lies beneath the sands and sandstones of the Connecticut Valley, and the line of Triassic eruptions 307 308 GEOLOGY OP OLD HAMPSHIKB COUNTY, MASS. bisects it from north to south, so that the great depression of the valley seems to be connected with these ancient granitic intrusions. Outside the area defined above granite dikes are few and small, the only large ones being the Middlefield and Coys Hill dikes. Topographically, and in the interest of cartographic work, they may be divided into four groups: (a) the great stocks and dikes of muscovite-granite, with their accompanying swarms of smaller dikes, the whole surrounding the groups b and c below ; (If) the extended central areas of quartz-gabbro and tonalite (syenite of President Hitchcock), which are without accompanying dikes; (c) the biotite-muscovite-granite, which occupies great areas topographically as well as lithologically intermediate between the other two, and is with dif- ficulty separated from the former (a) because it is itself cut by an enormous number of veins of muscovite-granite, or pegmatite, not distinguishable from that of the group a itself, so that it could often as well be assigned to the one as to the other on the map; (d) the porphyritic biotite-granites, which are widely separated from the above group. At the two opposite corners of the granitic region are two great squar- ish masses of quartz-hornblende rocks (tonalite), which send out no dikes, and which have produced a much more intense contact metamorphism than the mica-granites. The Hatfield tonalite area is immediately succeeded on the west by a fine-grained biotite-granite almost like the Monson gneiss, but which from the beginning carries a small, constant quantity of muscovite. To the west it soon begins to be cut by pegmatite dikes, and at the Mill River in Leeds, a mile west, their number is already considerable. In the next mile west the belt of granite which stretches from Loudville to "Williamsburg has, as it were, a substratum of the fine-grained biotite- (or two-mica-) granite, but so cut up by successive generations of the coarser muscovite- granite that it almost disappears beside the latter. Then still farther west and south, and on much higher ground, the great rounded granite stocks, which stretch from Montgomery to Conway and rise to form some of the highest hills on our western horizon — Pomeroy Mountain, in West Hampton, and Moores Hill, in Goshen — are desolate regions of a coarse muscovite-granite, rarely slashed by great blades of biotite, in which one finds here and there large areas or, as on the top of . Moores Hill, an isolated block of the fine-grained biotite-granite. o THE ERUPTIVE ROCKS. 309 Up to this point the description covers an area of unbroken granitic rocks of various types, superficially separated by shallow bridges of schist. Like the roots of a great tree inverted, there radiate from this central mass numberless dikes of every size, the connection being proved in many cases and probable in all. These dikes are of two kinds, the fine and uniform grained biotite- (or two-mica-) g-ranites or granitite, and the coarse to very coarse muscovite-granite veins — pegmatite. The former are generally, the latter sometijnes, interbedded in the schists for long distances. Toward the periphery of the area the pegmatite dikes carry secondary veins of albitic granite with many rare minerals. Within the area cut by the dikes and suiTOunding it in a broad halo the country rock is filled with quartz veins and pegmatite lenses of every size, derived, I doubt not, from the granite. On the eastern side of the river there stretches north from the Bel- chertown tonalite ("syenite") area a region where the schists are so crowded with pegmatite veins that they (the schists) sink into unimportance. This continues across Amherst, and in Leverett is followed by a large area of almost unbroken granite. The discovery and description of the peculiar type of eruptive masses to which the name " laccolites " ' has been given by Gilbert — great mushroom- like bodies of lava thrust up into the bedded rock to a certain level and then, expanding into a cake-like mass between the beds, pushing up the superin- cumbent strata into a low dome, but not reaching the surface — suggested to E. Suess^ the name "batholites" for the similar but more extensive masses of granite which occupy a position in the crystalline schists analogous to that of the laccolites in the newer rocks. It is in this connection that the obser- vation of Hitchcock is interesting, that the great masses of granite seem to be set free by the denudation of the schists above them, and the furth'er observation — ^which I have had occasion to make repeatedly — that where the schists are so cut up by the interlacing granite dikes that the latter make up far the greater portion of the surface, and even where long isolated sheets stand vertical or nearly so in the great granite masses, the prevalent strike and dip of the suiTounding schists are strictly maintained, indicating ' G. K. Gilbert, Kept. Geol. Henry Mountains. A. C. Peale, On a peculiar type of eruptive mountains in Colorado: Bull. U. S. Geol. and Geog. Surv. Terr., No. 3, p. 551. F. M. Endlich, Erup- tive rocks of Colorado: Tenth Ann. Kept. U. S. Geol. and Geog. Surv. Terr., p. 199. * E. Suess, Das Antlitz der Erde, p. 219. 310 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. that they are downward projections of the roof of the bathoUte, which has been removed just to their average plane of junction. Such a great batho- hte is well seen in Mount Tekoa, in Montgomery. To one standing on the high ground on the west line of Westfield and looking north the contrast of the white granite and the black schist is strongly marked. On the right the great dome of granite makes the sky-line. To the left its curved surface passes down beneath the mass of the schists of Mount Tekoa. The latter at first mantle up over the dome conformably, and higher up end very obliquely on the contact plane, and are greatly cori'ugated and cut by many large dikes sent off from the main mass. Again, these schists and their limestones, entangled in the granite, have been subjected to the same kind and degree of contact metamorphism as the broad band surrounding them. The schists became feldspathic and the limestones coarsely crystalline, as described under the head of the Con- way mica-schist, page 197, while the hornblende-schists became pyroxenic (as described on p. 243) or feldspathic with or without the development of pyroxene. I look upon the larger masses as great granitic reservoirs' which have partly forced and partly melted then- way up through the schists to the place where they are found, absorbing much of the material of the latter in their progress and sending upward and outward a complex radi- ating network of dikes. I consider the two great stocks of "tonalite" described below to be partially denuded domes of these great granite batholites, which have melted so much of the gneiss and hornblende-schist into their mass that their composition has been greatly changed, but which, penetrated more deeply, would change to ordinary granite. Two bands of hornblende-schist may be traced right up to the Belcher- town stock on the south, and reappear again with their attendant beds upon the north, and a single very thick bed can be followed up to the Hatfield bed on the north, and in traces dipping toward it along its western side. The hornblende-schist west of Belchertown village, cut by numerous dikes of granite, becomes impregnated with feldspar, and its fragments have their hornblende largely changed to green pp-oxene for a foot from the contact plane (this at Kellys Crossing), and farther south beds of augitic ' See J. W. Jncld, The ancient volcanoes of the Hebrides : Jour. Geol. Soc. London, Vol. XXX, 1874, pp. 220-300. THE EEUrTIVE ROCKS. 311 granite nearly 4 feet thick border the hxrger granite veins, or are inter- cahited in the amphiboUte, and at hist the whole greatly resembles the "syenite" — here a diallage-biotite-gabbro. The eastern hornblendic band comes south as a sharp synclinal fold of honiblende-schist, embracing a band of mica-schist, and becomes changed to resemble the tonalite, while the inclosed schist continues far south into the tonalite, metamorphosed into a highly crystalline fibrolite-schist. On the west side of the river broad bands of hornblende-schist and limestone can be traced to contact with the tonalite, and isolated fragments ap2:)ear on the latter across its whole length. Farther west, beyond the influence of the hornblende-schist, the tonalite changes to biotite-granite, and still farther west to muscovite-granite. Biotite-granite becomes the prevailing' rock of the batholites, where they are contained in the Conway garnetiferous schists. Two circumstances are very peculiar in the distribution of the rock. The first is the barrenness of the great central masses as compared with the richness in minerals of the smaller bordering dikes ; the other, the degree to which the granite is confined to the mica-schist and avoids the gneiss which bounds it east and west and in all probability underlies it. This association is so marked that when a narrow strip of the Conway mica-schist appears east of the river in Northfield there are associated with it dikes of pegmatite having secondary veins of albitic granite carrying cleavelandite, spodumene, columbite, and beryl. The western line of Pelham and its prolongation northward and south- ward through Leverett and Belchertown is the eastern boundary of the disturbed area, and in the gneiss east of it granite dikes are few and unim- portant, rarely, as at the Monson quarry, can-ying garnet and beryl. I have given much thought to those theories which would trace the granite down to the subjacent gneiss which, entirely melted, is supposed to have been " extra vasated" into the subjacent rocks; but I find no good reason for inferring any intimate relation between the gneisses of the region and the pegmatite. Many chemical and microscopical peculiarities of the gneiss militate against that relation, such as the large content of quartz, calcium, and iron and the small content of potassium, the uniform distribu- tion of biotite and titanite, and the absence of tourmaline and muscovite. Further (exception being made of the small secondary veins with 812 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. their abundant minerals), the notion that any considerable ijortion of the pegmatite is "endogenous" — i. e., is a granite veinstone of aqueous origin — is met by its enormous amount in comparison with the schists from which it is supposed to be derived, and the fact that it has plainly conduced to the enrichment of the schists by rendering them feldspathic and abundantly quartz- veined rather than to their impoverishment. At the same time, the appearance of the granites solely in the mica-schist area is a fact for which I have no explanation, except that the granites have come up along the axes of the larger synclines. HISTORICAL KOTES ON" THE MICA-GRAIsITES. 1819. E. Hitclicock. Geology of Deerfleld, etc. Am. Jour. Sci., 1st series, vol. 1, p. IOC. 1820. A. Eaton. Index, p. 95. 1823. E. Hitchcock. Geology of the Connecticut River. Am. Jour. Sci. , 1st series, vol. 6, p. 18. 1824. C. Dewey. Geology of Berkshire County, etc. Ibid., vol. 8, p. 4. 1824. E. Emmons. Notice of localities. Ibid., p. 250. 1827. A. Nash. Lead mines of Hampshire County. Ibid., vol. 12, pp. 238-258. 1828. E. Hitchcock. Mineral localities. Ibid., vol. 14, p. 219. 1833. E. Hitchcock. Geology of Massachusetts, p. 465. 1835. E. Hitchcock. Ibid., No. 17, p. 473. 1841. E. Hitchcock. Ibid., Final Eeport, p. 682. 1855. B. Emmons. American Geology, p. 64. 1866, C. TJ. Shepard. Am. Jour. Sci., 2d series, vol. 42, p. 248. 1876. W. O. Crosby. Report on Geological Map of Massachusetts, pp. 30, 38. 1879. A. A. Julien. Spodumeue and its alterations. Annals N. Y. Acad. Sci., vol. 1, p. 346. In 1819^ President Hitchcock mentions the granite as extending from Southampton to Hatfield, with veins of lead ore — not distinguishing the Hatfield "syenite." In 1820 Eaton uses the term "the Northampton Range," and notes the direction and extent of the lead vein from Montgomery to Leverett as proof of the continuity of the above range of granite. He also notes the three granite veins, with tourmaline, in Goshen and Chesterfield. In 1823 President Hitchcock designates the granitic area extending 1 The year number may serve as reference number to the article cited above. HISTORICAL NOTES ON THE MICA-GKANITES. 3 13 troiu Soutluuuptou through Williamsburg "the Southampton granite," and considers it "an original fundamental deposit of this rock." He says: I would here suggest whether the mica-slate of this region that contains beds of granite may not be a newer formation reposing immediately upon that granitic nucleus which probably forms the basis rock in New England. And wherever this mica-slato and upper granite is worn away or there is a projection in the nucleus the basis rock may appear. He describes further the Amherst-Leverett range of gi-anite, extending it to the mouth of Millers River. The gi-anite veins abundantly cutting across all the other rocks of the region are discussed and figured. These, as, for example, the main body of the great Chesterfield tourmaline-bearing vein, "are doubtless contemporaneous — that is, such as were consolidated at the same time with the rocks they traverse" — a curious idea, based, I presume, on the fact that the Chesterfield dike is interbedded in its schists with apparent conformity. Graphic gi-anite in Deerfield and Goshen, porphyritic granite in a range five or six miles long in Chester (which is a mistake for Middlefield), on authority of Dr. Emmons, and " pseudomorphous granite" are described. The latter is a coarse pegmatite, in which thin blades of biotite of the size and sliape of the blade of a dinner-knife penetrate the rock in every direction and meet at every angle, but never intersect. In 1824 Dewey correctly locates the great Middlefield porphyritic granite vein, doubtless on the authority of Emmons, and the latter describes and figures many veins in Chester. In 1827 Nash notes that often in ascending a mountain mica-slate forms the base, granite the apex, and that the great masses of granite are wholly destitute of minerals, and only the veins in mica-slate contain these. In 1833 President Hitchcock gives a very complete and very clear description of granite, restricting it to the variety without hornblende, illus- trating its complex relations to the mica-slate by forty-eight figures, enumer- ating the minerals contained in it, and giving a long argument in favor of its eruptivity. He says: "Upon the whole, the granite lies remarkably low in respect to other rocks, and one can not avoid the inference when he examines its situation, in almost all cases, that the abrasion of the stratified rocks may have brought the granite to light." In 1835, and again in 1841, he publishes the same description with 314 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. scarcely any change, except to call attention to the remarkably complex system of veins at Grreat Falls, in Russell, He recurs as follows to the idea quoted above: "And I have sometimes inquired whether, if the whole sur- face were denuded as deep as that part occupied by granite, we should not find this rock spreading over a great part of the State." In 1876 W. 0. Crosby calls the granites of western Massachusetts "exotic montalban granites, whitish or gray, seldom red or greenish, as the Huronian granite is always micaceous, seldom hornblendic," and yet in the next section he classes the "syenite," which is generally reddish or greenish, and hornblendic, with the montalban granite. I am not certain that I understand this classification. He says further: "The Williamsburg granite represents, I conceive, the extension southward of the Shelburne anticlinal, onl}^ carried a step farther to produce the extravasated granite." An inspection of the map will show that the Goshen anticline is the continu- ation of that at Shelburne. There is no anticlinal structure connected with the Williamsburg granite and mica-schist. In 1879 Julien publishes a most valuable article on the minerals of the granite veins related to spodumene, containing much concerning the secondary veins themselves which I have incorporated in my own descrip- tion beyond. BIOTITE-MUSCOTITE-GRAlSriTE. AREAS WEST OF THE CONNECTICUT. Characteristic for this rock is its fine, even grain. Biotite, the prevailing mica, is scattered in small, separate, jet-black scales in a fresh, bluish-white mixture of quartz and feldspar. This gives it a deceptive similarity to the granitic forms of the Becket gneiss, from which it is distinguished by its greater firmness and by a small, constant content of muscovite. It resem- bles the granite of Concord, New Hampshire. It may be best studied at the quarries east of Florence. In its finest varieties, as at the Loudville mine, it is almost a petrosilex ; in its coarsest, as at the quanies above, the grains reach 2-3™™. It is wholly wanting on the east side of the river, around the Belchertown tonalite, which is in immediate proximity to the Monson gneiss. DISTEIBUTION. Just east of the Florence quarries, and extending from the house of Mrs. Haley to that of W. N. Moore, this granite adjoins the tonalite. In all BIOTITE-MUSCOVITE-GRANITE. 315 tliis (listiiiu'o the exact contact is covered, but the rocks can be studied at points a few feet from it, and the change from the one rock to the other seems to bo quite abrupt. From this boundary it extends westward to the Mill River, and it is abundantly exposed along the road beside the river from Leeds to Williams- burg. As already indicated, it is found to be more and more replaced by dikes of pegmatite as one goes out to the border of the area and up to the higher levels. It makes always the impression that it was the original rock, and that the pegmatite was injected into it at a later time, pei'haps only slightly later. Around the periphery of the area its dikes are very abundant in Goshen and Chesterfield, and less so in Conway and Blandford. Its dikes are so uniformly interbedded in the schist around the Goshen anticline that I for a long time mapped it as gneiss, luitil at the south end of South street in Chesterfield, near C. Damon's, I found it cutting across the beds of the schist. In these dikes it is of 'a little finer grain and more friable than in the main stocks. PETEOGRAPHICAL DESCRIPTION. In the middle quarry west of Moore's and east of Florence it is medium-grained, very fresh biotite-granite, with little muscovite, veiy feldspathic, and showing abundant triclinic striation. The quartz is rare and occurs in rounded grains, as if resorbed. It contains fluid cavities in enormous quantity, of grotesque forms and in large sheets, often with bubbles, some moving rapidly, some slowly, and some being stationary. They contain water and carbon dioxide. Large, rigid needles of rutile also occur. The feldspar, mostly triclinic, is centrally decomposed into a brown, opaque mass of kaolin scales. The narrow, fresh border seems almost as if it were a secondary growth. Extinction, 18° on either side. Orthoclase and microcline are also present, but in small quantity, and the large amount of plagioclase allies it to the tonalite. CHEMICAL ANALYSIS. Analysis I, following, was made by Mr. L. G. Eakins from a speci- men of the best quarry stone of coarser grain from Moore's quany, Flor- ence, from which also the slides were cut. It is remarkable how exactly this analysis agrees with that of the lighter variety of the Monson gneiss, 316 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. witli which variety this rock agrees most closely. I have repeated, for comparison, the analysis of the Monson rock (II) : Analyses of granite from Florence and gneiss from Monson. I. II. SiOj 73.27 .10 15.51 .33 1.14 Trace. 2.74 .15 1.66 4.79 .68 Trace. 73.47 TiOj ALO3 15.07 V 1.15 Fe,G3 FeO MnO CaO 4.48 .12 ..88 5.59 MgO ICO ... . Na,0 H,0 T2O5 S .' Trace. Trace. Cu Total 100. 37 100. 26 TOURMALINE DENDRITE ON THE SURFACE OF A BLOCK OF GRANITE FROM LEEDS. On the surface of a block of the light-gray granite quarried at Leeds is a layer, one-eighth of an inch thick, which differs greatly from the mass of the granite. The latter is a clear, gray granite of medium grain whose mica is mostly a jet-black biotite. The thin layer is a slightly coarser muscovite-granite, and over a large surface beautiful dendritic growths of jet-black tourmaline have formed. (PL VII.) It is clear that boracic-acid emanations have passed through a fissure in the newly formed granite, and have promoted at once the formation of tourmaline and the replacement of the biotite of the granite by muscovite. THE ATHOL AREA. This enters the county only in the east portion of Orange, east of the west branch of Tully Brook, and the boundary is continued northward across the west portion of Royalston. The granite, from more rapid erosion, forms a steep valley, out of which rise the steep-sided Big Tully and Little BIOTITE MUSCOVITE-GE ANITE. 317 Tully mountains. The (li'ainage established itself toward the sides of the basin and left these mountains in its center, as in the Orange-Enfield basin. The result is that the contacts with the schists are everywhere concealed beneath the brook deposits, and farther north by the till. From the highest groiuid on the road north from North Orange a fine view is obtained of the deep basin, with the white granite showing in the flanks of the Tully Mountains and all the ground above the sand level a "felsenmeer" of great woolsack bowlders of granite, while the bold hill in the extreme northeast of Orange shows by its jagged ridges of rust-brown rock that it is made up of the higher fibrolite-schists. The rock is the same almost purely biotitic granite as in the other bands, in the northern portion beautifully " stretched" and slighl>ly garnetiferous. Along the side of Little Tully Mountain the biotite is mingled with epidote in porphyritic blotches. SECRETIONS AND INCLUSIONS. I have described below (p. 332) the black biotitic secretions which occur in the tonalite on either side of the river, and which resemble exactly those found in this granite. They are formed by the accumulation of biotite around centers. Other inclusions are more or less angular, and are finer- grained and less micaceous than their host, or coarser-grained and black from excess of biotite and hornblende. These seem to be portions of the rock itself which have solidified before the rest and have been broken up and floated to their present position, with more or less re-solution. There is in the first Massachusetts survey collection one specimen from Whately which contains a true inclusion of a foreign rock — a highly pyritous muscovite-schist. THE HARDWICK GNEISSOID GRANITE AND GRANITITE. Reference may be made to the section in Chapter VIII having the above caption for a preHminary description of this rock (p. 239). It covers a much greater area in Worcester County than here, and its relations will be more fully discussed in a memoir on the geology of that county. The rock could have been described with perhaps greater propriety in this chapter than with the Brimfield schists. The Coys Hill granitite seems to me somewhat older than the other 318 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. granites of this section, as it is more clearly affected by the last foldings, and it cuts the Brimfield schist and the Hardwick granite-gneiss. The latter is thus materiall}^ older than the other granites, and more nearly contemporaneous with the Brimfield schist, in which I suppose it to have been intraded before the final folding of the series, and with which I have therefore described it. BIOTITE-GRAJSriTE, OR GEAIflTITB. Within the central granite area I have found but one dike of a purely biotite - granite, an exceptionally fresh, coarse, subporphyritic rock with white orthoclase, much plagioclase, and amber quartz. It occurs 820 feet west of the outl.et of Burnell's pond in Chesterfield. The granite described above — that extending west from the tonalite through Florence, and that around Williamsburg — is often in hand speci- mens a purely biotitic rock; but it generally contains at least a small amount of musco^dte. This muscovite has always in the freshest rock the character of an original component, and is so regularly present that I have classed these rocks as two-mica-granites. CONTACT METAMORPHISM OF THE GRANITITE AND SCHISTS. The granitite is a highly feldspathic rock, and it has had great influ- ence U2Don the rocks bordering it on either side. The rusty fibrolite-schists become garnetiferous gneisses, jDorphyritic with a great number of rounded masses of clear, fresh, transparent ortho- clase, which oi*'en furnish good moonstones, and were foi'merly quoted as adularia from Brimfield and Sturbridge. They appear also in the dark Hardwick granite, where the porphyritic granitite approaches it, and they continiie to appear in the fibrolite-gneiss far south of the most southern point to which the granitite can be traced, across Brimfield and Monson, as if they marked its subterranean continuation. They are often crushed at the border into a fine, sugary mosaic, and this cataclase structure is at times continued clear to the center. THE MIDDLEFIELD PORPHYRITIC GRANITITE. The great dike of granite in Middlefield, about 6 miles long, is widely separated from all other outcrops, and is unlike all the other masses of BIOTITE-GKANITE. 319 granite in the region. It is purely a biotite-granite, small -porpbyritic in all its central portions. The feldspars are about three-fourtbs of an inch long, rai'ely show carlsbad twinning, and are microcline without albite bands. A few rounded spots, apparently of albite, break the continuity of the cleavage surface. These feldspar crystals are at times bounded by a layer of secondary muscovite plates, and this is the only appearance of musco\'ite in the granite. The biotite is aggregated in groups of rather dull-black plates, with epidote, garnet, and rarely white apatite needles accompanying it. The yellowish-white background is a somewhat friable mixture of much gran- ular orthoclase and little bluish quartz, which is characterized by the presence of small, elongate cavities. At the border the porphyritic feld- spars and the biotite aggregates disappear, and the friable ground with small distant spots of biotite and the small cavities remain unchanged. THE COYS HILL PORPHYRITIC GRANITITE. DESCKIPTION AND DISTRIBUTION. The Middlefield dike on the west of the area is matched by this still larger dike on the extreme eastern border. It begins in Winchendon, and runs south 25 miles across Phillipston, Barre, New Braintree, West Brook- field, and Warren before it enters the Palmer quadrangle at its northwest corner, and ends in Brimfield. Its whole length is 33 miles; its average width is one-half mile. Only the portion in the Palmer quadrangle is here studied. It is a highly feldspathic, very coarse-porphyritic, garnet- iferous granitite, which presents almost everywhere a distinct gneissoid structure from the parallel arrangement of the large feldspars. It is proved to be an intruded rock by the fact that it runs for 20 miles in the Brim- field fibrolite-gneiss; then, just as it enters the Palmer quadrangle, it crosses very obliquely the Hardwick black granite, and enters the eastern band of fibrolite-gneiss. This is further proved by the fact that where it sends a great lobe into the western fibrolite-gneiss the boundary between the two is a broad sigmoid curve, having a general east-west direction, while the granitite on the north and the rusty fibrolite-gneiss on the south of this line have the same foliation structure, which strikes N. 25° E. and dips 60°-70° 320 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. west with great uniformity, and thus cuts directly across the boundary between the two rocks at every angle. The sudden disappearance of graphite, pyrite, and fibrolite, and rustiness and the great increase of feld- spar, separate it from the Brimfield gneiss. This gneissoid structure, furthermore, distinguishes this dike from the other granites of the region and indicates for it a greater age. The other granites have often been injected into the vertical foliation planes of the schists after these had been completely formed, and do not show any trace of having been subjected to the pressures which have given these structures to the schists, while here the granitite and the schists have been subjected to the same compression. A crushing of the feldspars occurs in the bordering portions of the granitite itself, and is well shown where, across the brook north of Fenton- ville, in Brimfield, the western boundary runs up the mountain side. Here the large feldspars are only slightly rounded and the sugary border of crushed feldspar still retains the angular boundary of the former crystal. The change increases until only the transparent centers remain, and this causes a marked whitening of the whole rock and ends with the formation of a light-gray, granular granitite, hardly to be distinguished from the Monson gneiss. This forms a selvage to the dike a hundred feet wide near H. Sherman's, a mile southwest of West Warren, and a large quarry has been opened upon the same rock on the west slope of Colonels Mountain, in the northeast corner of Palmer. The rock can well be described by supposing the lai'ge porphyritic Carlsbad twins which are scattered through the rusty fibrolite-schist of East Monson and Sturbridge to develop so abundantly that a complete augen- gneiss should result, the biotite and the garnet remaining the same as in the fibrolite-schist, and only the fibrolite, graphite, and pyrite disappearing, which they do almost uniformly. I have been thus led at times to consider this rock an extreme of the granitic impregnation which has affected the fibrolite-schists in this region, and not an intruded plutonic rock pure and simple. A granite dike 33 miles long and only 2,500 feet wide is rather anomalous, especially in a region where the granites are in great blocks of a wholly different type. We are here, however, at a point where the type changes. Farther east porphyritic grauitites are very common. BIOTITE-GRANITE. 321 CORDIERITE-GRANITITE. At Brimfield, in the roadside near the north line of the town, is a coarse granitic rock made np almost wholly of o-ranular feldspar, in which hu-ge, rounded crystals of the same are embedded. In both forms the feld- spar is largely transparent. Thin films of biotite, mostly changed to (chlorite, are shot through with tufts of fine fibrolite. Garnets and flakes of graphite are irregularly disseminated. Large, granular masses of nearly black, fresh cordierite occur, which are at times amethystine. The feldspar proves to be almost wholly microcline, with finest micro- perthitic structure (which is the cause of the moonstone luster) and with crushed borders, and it contains unusually large and well-defined zircons. The quartz contains many long, curved rutile needles. The cordierite is exceptionally fresh; rarely there spreads in fissures a delicate, feathery growth of limonite, and the mineral is altered for a small distance into a yellow, serpentine-like mass having aggregate polar- ization. It contains in great numbers regular hexagonal plates of hematite, placed in two planes at right angles to each other. Interposed laminae occur at times in twin positions. (See fig. 2, PL III.) There are two sets, making an angle of about 61° with each other. They are long, rigidly straight and parallel, narrow plates, sometimes slightly tapering or truncated at the end by an oblique plane. Sometimes a broad untwinned area sends a great number of these thin bands far into the untwinned area of another crystal. At times the bands interlace and include many diamond-shaped fields of the host. They are unlike plagioclase bands in that they are sur- rounded in polai'ized light by a white band. This is because the plane of boundary runs obliquely to the plane of the section, and the complementary colors of two parts neutralize each other. The fibrolite runs up, branching and rebranching like a plant, and at the end of each branch bright-green plates of chlorite are attached like leaves. In some cases it seems as if the square prism of the fibrolite were changed to chlorite. In fig. 2 of PI. Ill the unshaded portion shows the axial figure of the fii'st crystal (I) eccentrically as indicated. This crystal was large, and from a second, smaller crystalline portion (II) blades generally rigidly straight and with straight boundaries were sent out into (I). These plates MON XXIX 21 322 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. are bounded by sides of the prism oo P, which is the twinning plane. They show a single axial ring, and the axial plane makes an angle of 48° with that of (I), instead of 60° 50', as would be the case if they were cut par- allel to P. The other cr}^stal (III) is represented by a series of blades which make an angle of 61° with the first series. They are also bounded by <^ P, which is the twinning plane, and their optical axial plane makes an angle of 58° oia the other side of (I). The figure is drawn with crossed nicols and with the crystal (II) at the point of extinction. MITSCOVITE-GBAKITE, OR PEGMLATITE. One may omit any detailed description of a rock so well known. It is a granular mixture of quartz, muscovite, and a potash-feldspar mostly microcline, or orthoclase more or less mixed with microcline, and veined with albite; generally coarse-grained, and often very coarse — a giant granite with its feldspars several inches on a side and its mica a foot or more across. Its greater masses and its smaller dikes are almost wholly barren of any accessory minerals. In its dikes of medium or larger size several minerals — as beryl, biotite, tourmaline, garnet — occur sporadically, and much more rarely "secondary veins" of foliated albite contain these and others of the rarest minerals in greatest beauty and profusion. The great masses which stretch from Montgomery to Conway and the many dikes which go out from them, the great mass southeast of Mount Toby in Leverett and the north of Amherst, and the dikes extending thence south through Amherst and Belchertown are the main outcrops of this rock. These dikes have been extensively worked in Blandford and Hunting-- ton for mica, feldspar, and quartz, especially upon the property of the Pontoosic Flint Mills, in the north part of Blandford. There is opened here one of the most beautiful veins of giant granite in the county. The upper and the lower walls are occupied by a selvage 2 to 3 feet wide of the coarsest muscovite, which projects inward with crystals a foot square, mingled with, feldspar. There follows above and below a layer 1 to 2 feet wide of great feldspars a foot on a side, which projects freely inward. The center, 2 to 6 feet wide, is of clear, smoky quartz, and in one direction the whole vein changes to quartz. These veins are very rare in the tonalite, but going out from these into the granites they increase gradually in ntimber and at last swarm in PEGMATITE. 323 several <;-om'riiti()us iiiid of every size until the original granite alinowt cli;s- aj)jK'ars, and they appear in great numbers also far beyond the great granite stocks, in the schists. While nuiscovite is so rare as to be almost an accessory in the granitite, biotite is not wholly wanting here, but appears always with the distinct habit of an accessory, in great blades touching each other so as to form a rude cellular structure. This was called pseudomorphous granite by Hitchcock. I'EOBABLE EXTREME MODIFICATION OF THE PEGMATITE BY CRUSHING. The North Amherst granite. — This is a peculiar rock, appearing in the hills southeast and west of North Amherst station, and again at the foot of the west slope of the Pelliam ridge east of D. Hawley's, beside a brook. It has at times a quite marked foliation (almost certainly a secondary structure), distant planes being thickly covered with quite coarse muscovite- biotite films, while in cross-section it appears wholly free from mica and has a subporphyritic look. Opaque, subangular portions of feldspar or quartz-feldspar are scattered quite distantly in a highly crystalline and con- tinuous quartz mass which seems a secondary constituent cementing the brecciated fragments of a highly feldspathic granite from which most of the hornblende or biotite has been removed during the violent changes the rock has undergone. In thin section a few twisted fragments of much altered biotite appear. The feldspars are brown with alteration products — rust, kaolin, and musco- vite — and rarely determinable ; more plagioclase can be made out with the lens in the mass than under the microscope in thin section. The grains are much cracked and crushed, and show undulatory extinction and wavy twin laminae. Everything accords with its position along the main fault area marked by strong crushing. ALBITIC GEAIiTITE AJND PEGMATITE DIKES CONTAINIlSrG RARE MINERAES. It is a remarkable fact that the rare elements appear only in pegmatite dikes on the extreme periphery of the great granite area, and that they are not found in the biotite-granite or within the great central region of granite. This is enforced by the list of localities below, and may be brought into 324 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. casual connection with fumarole activity, using the term in a wide sense, and indicates that the pegmatite dikes were the later products in the com- plex series of granitic types present in the region. It has some analogy with the fumarole products accomjjanying the trap eruptions in which, as I have described (p. 423), albite, clearly water-deposited, rests on delessite in amygdaloid cavities, as do also datolite, axinite, and rarely tourmaline — boracic acid minerals. The cleavelandite, which is a variety of albite, and abundant tourmaline match the minerals named above, but the analogy does not extend to the rarer elements. DISTRIBUTION AND DESCRIPTION. Because of several peculiarities in the separate occurrences, I have given below a description of each, commencing at the northwest and going around the area by the south. I. THE URBAT TOTJKMALINB-SPODUMENE DIKE. Macomber's spoduniene ledge, Clarke's tourmaline ledge, the West Chesterfield Hollow, and the well-known Walnut Hill spodumene ledges (the last in what is now Huntington, the others in Chesterfield) are all portions of one continuous or nearly continuous, vertical, interbedded dike of coarse pegmatite, which is faulted and its south half thrown east at West Chesterfield Hollow. A. A. Julien^ says : At Macomber's ledge the coarse orthoclase granite of the main vein contains films of margarodite and few imperfect green beryls, while in the secondary vein the succession seems to have been, first, quartz, muscovite, granular albite, tourma- line, and spodumene; then cleavelandite, quartz, manganese, garnet, and zircon; and, finally, smoky quartz with green and blue tourmaline. The larger crystals of most of these minerals penetrate through all the layers and their growth seems to have been continuous. At Clarke's ledge the main granite vein is of the same general constitution as at Macomber's, rarely showing a few large beryls. In the secondary vein no spodumene occurs, but the succession is in the same order. First, on either wall a saccharoidal albitic granite, with little quartz and mica and a few scattered, imper- fect black tourmalines and garnets, then coarse cleavelandite, with blue, green, red, and rarely brown tourmaline, and small quantities of the rarer minerals, microlite, columbite, cassiterite, zircon, cookeite, lepidolite; all these, especially the tourmaline, increase in quantity toward the center of the vein, which is filled up by an irregular sheet of smoky quartz. \ Spotlumene and its alterations: Annals N. Y. Acad. Sci., Vol. I, p. 351. PEGMATITK D IKES CONTAINING KAliE MINERALS. 325 Here the aclii.sts havu bt'on worn iiway from the niaiu (Uke, which stands up in a vertical wall 33 feet high. In many places a veneering of schist remains attached, and when it is removed the impression of the schist is sharp and clear on the surface of the pegmatite. The la}'er of schist against which the pegmatite rests shows no signs of its influence, thus differing from the schist in contact with the albitic granite at the Barrus farm, described below, where the contact metamorphism is pronounced and the granite and schist are fused together. The secondary vein, as described above, is seen high up on the face of the vertical side of the main dike, and seems to me to have been deposited in a vertical transverse fissure in the latter, which fissure extends from the east face only about halfway across the dike and to an intermediate distance up and down. This fissure seems to have been formed in the newly consolidated pegmatite and to have been filled by a magma of peculiar composition, much hydrated, rather than by a simple solution. Thus, commencing with this tonahte and ending with the quartz veins, I suppose there is an unbroken series from igneous fusion to aqueous solution. At Chesterfield Hollow the granite of the main vein is of the usual character, bat shows no beryl and little mica. The successive deposition of minerals in the secondary vein is, first, orthoclase in huge crystals, large plates of muscovite, some- times 6 to 10 inches in diameter, and grayish-white quartz. Within this comes an irregular mass of a coarse albitic granite, with green muscovite, spodumene, greenish-white beryl in masses sometimes 10 to 25 pounds in weight, and a zircon rich in uranium iu minute double pyramids rarely three-sixteenths of an inch in diameter. Usually this albitic granite passes gradually into a mixture of quartz and cleavelandite iu bunches of snow white plates inclosing less muscovite— manganese garnets in large and abundant but imperfectly crystalline grains, zircon, spodumene, and yellowish-white beryl in irregular masses. Finally the core of the vein consists of an irregular sheet of smoky quartz, penetrated by long prisms of spodumene, green beryl in small and good crystals, muscovite in hexagonal plates, often well crystallized and up to 2 or 3 inches across, as well as in sheets, scattered scales and wavy films which in part seem to be altered to margarodite, columbite, and zircon in rare but perfect crystals. This succession of minerals in the secondary vein is not as regular as might be inferred from the foregoing description, in which it is intended to indicate only the general tendency toward a definite arrangement. At Walnut Hill, in Huntington, the material of the main vein is similar to that of the preceding locality. In the secondary vein the rich deposit was found to be a very coarse albitic granite, rich iu black tourmaline in huge masses, mus- covite, and garnet; then followed cleavelandite, white quartz, and spodumene iu the 326 G^BOLOGY OF OLD HAMPSHIRE COUNTY, MASS. well-known fine crystals associated with black and blue tourmaline, triphylite, cryto- lite, garnet, apatite, niuscovite, and greenish-white beryl, and the central sheet of smoky white quartz received the terminations of the spodumene crystals, together with a little beryl, moscovite, and cyrtolite.' II. DIKES IN GOSHEN. The first dike in Groshen, that on the Manning- farm, west of the Ash- field road, near the north line of the town, can not be well studied, as only disconnected masses can be obtained and the boundaries of the dike are not visible. The veinstone consists of a coarse aggregate of albite, iudicolite, garnet, and spodumene, whose crowded and imperfectly outlined grains indicate a more rapid crystallization than in the other localities. At the Barras vein, a mile to the west, the mass of the vein seems to be repre- sented in j)lace by a coarse aggregation of white quartz, orthoclase, and muscovite, and occasionally greenish beryl, accompanied in places by a contiguous vein of red- dish-white quartz, while the scattered bowlders of albitic granite appear to be frag- ments of a central band or secondary vein whose slow crystallization is suggested by the beautiful aggregate of snow-white cleavelandite and grayish-white quartz which forms the matrix of the rarer minerals. Of these the most abundant are the spodumene, mostly in rectangular prismatic masses up to 18 inches in length, and tourmaline in black, green, or blue-black (indicolite), generally massive, but sometimes in good crystals. Less commonly were found beryl, green and white (goshenite), in grains or sometimes fairly crystallized, with good terminations, garnet, rose-colored muscovite, and, still more rarely, columbite and cassiterite in minute crystals. Appar- ently there has been also in parts of the vein a final deposition of masses of smoky quarts enveloping smaller crystals of these minerals, but particularly of green beryl and indicolite.^ Here the secondary vein came in part in contact with the countr}^ rock, and the latter, which is a whetstone-schist just at the contact, has been for several inches (at least 4) fully impregnated with silica, albite, and tour- maline in fine, black needles. Halfway between the Manning and the Barrus ledges, by the road- side, south of J. B. Taylor's, much blasting has been done recently (1889) by Mr. Barrus for spodumene. It was proposed to export the mineral for the manufacture of lithium. The spodumene is abundant in poorly bounded crystals and coarse crystalline aggregates associated with little tourmaline, ' A. A. Julien, Spodumene and its alterations: Annals N. Y. Acad Sci., Vol. I, p. 351. ■^A. A. Julian, ibid., p. 350. PEGMATITE DIKES CONTAINING KARE MINERALS. 327 "•ariR't, aiul iiiuscovite. The three loccihties hist luentionetl he in a line, trending about N. 70° W., and may be parts of one vein. A new locality discovered by me is in a pegmatite ledge a mile south of the Barnis ledge and overlooking Lily Pond. Here i^rismatic pseudo- morphs of coarse inuscovite after spoduraene, 2 niches long, occur in limited number. III. DIKES IN CHESTER, BLANDFORD, AND HUNTINGTON. The locality mentioned by E. Emmons^ as occurring a mile north of Chester village was stated by him to contain spodumene, smoky quartz, muscovite, cleavelandite, and indicolite. This ledge I was not able to find. Mr. A. A. Julien seems to have had better success, but to have found no spodumene there.^ Not far from this locality the granite veins have furnished large and perfect manganesian garnets. Farther south, on the northeast line of Blandford, a very coarse peg- matite, much quarried for mica, quartz, and feldspar, the property of the Pontoosic Flint Mills (see page 322), has furnished beryls of great size, the largest as big as a powder keg, with large garnets. A granite in the churchyard in Blandford also carries beryl. .Just south of the first house on the Westfield- Russell road after entering Russell the pegmatite abounds in manganesian garnets of lai'ge size and great perfection, which are found in every cabinet. IV. DIKES EAST OF THE CONNECTICUT. On the other side of the area the small pegmatite veins at the Monson quarry have furnished very fine beiyls and many manganesian garnets. The finest bluish-white cleavelandite occurs in New Salem. In Northfield, where the Gulf road crosses the south line, large beryls occur in the peg- matite, and farther north, a mile west of the Moody homestead, is the inter- esting locality of columbite in a pegmatite vein in the mica-schists, and a mile north on the strike of the schists is a secondary vein of the fine radiate-foliate cleavelandite of very considerable size, exactly like the Goshen-Chesterfield schists, in which I could find no other minerals. Still farther north, on the strike and therefore in the same schists, is a 'Am. Jour. Sci., Ist series, Vol. VIII, 1824, p. 243. =A. A. Julien, Spodumene audits alterations: Annals N. Y. Acad. Sci., Vol. I, p. 221. 328 GEOLOGY OP OLD HAMPSHIRE COIHsTTY, MASS. pegmatite vein abounding in albite and spodumene and closely resembling the occurrence at the Manning farm, described above (see p. 326). This lies just across the north line of Northfield, near the house of M. A. Brown, on the Winchester road. GARNET IN PEGMATITE WITH COMPLEX PARAMORPHIC BORDER OF ZOISITE- HEMATITE, EPIDOTE-FIBROLITE, AND MUSCOVITE. In the coarse inuscovite-granite that occurs on the Gulf road in the southern part of Northfield and cuts the Goshen mica-schist which farther east becomes fibrolitic and is called the Brimfield schist, several interesting garnets have been recently discovered by Mr. C. H. Webster. (PI. II, fig. 4.) They are nearly an inch across, of deep-red color and of trajDezo- hedral form, with a narrow dark-red band surrounding them which is in places spotted with green. Outside this is an opaque white border, 3-7™°" broad, which looks like saussurite. The garnet under the microscope is evenly cleaved and almost free from inclusions. There are a few rounded blebs of the quartz-muscovite mixture or of the zoisite-hematite mixture which is found in the border. This border layer is largely composed of zoisite in stout grains, which shows low polarization colors, high refraction, a divergence of optical axes of 45°, and is optically positive. The hematite is deep-red to black, and occupies regular and close cleavage fissures in the zoisite, often so regular as to recall the cross-section of a tabulate coral. This layer is joined to the garnet by a suture, which is very intricate, so that lobes of the two minerals penetrate deepl)- into each other and rounded blebs of the zoisite are cut off in the garnet. While thus intricately joined interiorly, the zoisite mixture forms exteriorly quite smooth crystal faces for the garnet. Outside this first layer is a delicate and very thin layer of green epidote, which folds into every irregularity of the last layer and extends continuously over all the surface of the latter and around all inclosed por- tions of the same. It has rather moderate absorption and extinguishes in proper relation to the cleavages for epidote. The outer layer is a mass of muscovite blades felted with a fine-fibrous mass of fibrolite needles arranged in beautiful plumose and tufted groups in a general way radially to the surface of the garnet. ALBITIC GRANITE. 329 THE CRUSHING OF MINERALS IN THE ALBITIC GRANITE. la tlie suiiiiuer of 188;j the pegmatite dike oil Walnut Hill, in Hunt- ing'ton, was reopened for nie by Mr. Frank L. Nason. The spodumene crystals obtained were large — larger than most of those obtained previously. They were clear-gray, without the shade of flesh-color of those before obtained, and were covered with dendrites, which also penetrated every- where into theperfect cleavage. Several fine twins occurred, but for the most part they were not well terminated. The largest crystal was 28 by 7i by oh inches. The crystals bear abundant evidence of the violent pressure to which they have been subjected since their formation, several large, perfectly terminated crystals a foot long being several times obliquely sheared off and the parts slipped one-eighth to one-fourth of an inch and recemented; and the largest crystal, whose dimensions are given above, is broken across or sharply folded into "monoclinal flexures" more than forty times. Other large crystals are bent over as much as 45° in a great curve, one sharply a full 90°, and without a crack.^ The feldspar (microcline) occurs in masses as large as one's head, often in part green. The cleavelandite is not distinguishable from that of Chester- field. Tourmaline appears in large, rude, black crystals. Granular ixiasses of honey-yellow manganesian garnet (intermixed with feldspar) as large as an egg have by their decomposition furnished the material for the abun- dant dendi'ites. These latter masses are at times punched into the great spodumenes as if these had been plastic as wax. There is in the collection at Amherst a crystal of tourmaline from the Clarke ledge, once figured by President Hitchcock,^ which, is broken across fifteen times and the parts moved into a position en dchelon and recemented by quartz, and I have a crystal of beryl from Huntington similarly affected. HYDROTHERMAL CHANGES IN THE ALBITIC GRANITE VEINS. Pseudomorphs. — Julien has described a most interesting series of pseudo- morphs in these dikes, produced by alkaline (mainly sodic) silicate solutions, by which spodumene is changed into cymatolite, killinite, albitic granite, muscovite, albite, and quartz, the lithia being replaced by the other ' For figures of these crystals see Minerological Lexicon, iiuder " Spodumene" : Bull. U. S. Geol. Survey No. 126, 1895, p. 159. =E. Hitchcock, Geol. Mass., p. 702. 330 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. alkalies. Immediately afterwards Brush and Dana proved the cymatolite to be a mixture of albite and muscovite formed from the spodumene with the intervention of a lithia- nepheline (eucryptite), and made it probable that the killinite was mainly hydrated muscovite, and thus the series of pseudomorphs is reduced to one type, namely, spodumene changed to albitic granite, whereby, from the gradual suppression of any one or two of the constituents, forms made up of either quartz, mica, albite, or mica and albite result. Pseudomorphs a yard in length and nearly a foot across are made up of a coarse mixttire in various proportions of albite, muscovite, and quartz, with manganese, garnet, zircon, beryl, etc., occasionally inter- posed, and we may add also microcline. ORDINARY METEORIC ALTERATION. By ordinary carbonated waters there has been a gradual removal of a part of the lithia and more soluble protoxides, almost universal, with the consequent effect upon the physical characteristics of the mineral shown by the loss of weight, luster, greenish color, and translucency. The zircons have absorbed water and lost part of their uranium, which has sep- arated as autunite, torbernite, and, by a further decomposition, uranocher. The garnets afford ocher and pyrolusite in dendritic films. The triphylite by absorption of water and higher oxidation of some constituents has assumed its present altered form, so that only rarely do small blue nuclei of the unaltered mineral remain. The spodumene and cymatolite both at last degeiierate into clayey material, sometimes pink and allied to kaolin or montmorrillouite.' The kaolin beds at Blandford village illustrate on a large scale the results of the agencies described in the last section. Grreat beds of coarse granite in every stage of alteration are exposed in the diggings; in some parts the feldspars are onlf softened and made friable, in others they are pure soft kaolin, and the mica-schist Avhich is tangled among the big veins is rotted to a soft, rusty earth. All the fissures in the altered mass are blackened by deposits of manganese oxide. It is quite certain that this deep-seated alteration of the granite is mainly pre-Glacial and owes its preservation to its position on the southeastern slope of the hill upon which the village is built. The material has been used extensively at Russell for the manufacture of brick of fine quality and tile, but recently the buildings have been destroyed by fire. ' A. A. Julien, Spodumene and its alterations: Annals N. Y. Acad. Sci., A''ol. I, p. 353. APLITE, QUAKTZ-PORPHYKY, AaS'D TONALITE. 331 APIilTE. Coarse peg'inatite dikes are rare in the tonalite areas. Their place is taken ])y aphte dikes, which in many places are very abundant in the toiiaUte, but are always very narrow. The rock is a fine-grained quartz- t'cldspar mixture, almost without hornblende or mica. Two miles north of Leyden Center a road runs west down to Green River. Following this road a halt' mile west, one reaches a place where it bends sharply south around a projecting spur of rock. A dikelike mass which has in part the aspect of an aplite and in part that of a quartz-porphyry crosses the road at this point. It is 13 rods wide, and stands vertical in the Conway schist and strikes north with it. The dike ends in the bluffs to the south, but can be followed a long way north. It is a pale-gray, rather small- porphyritic rock, and is the only rock of this type in the area. I have sometimes thought it an exceptionally massive arkose-gneiss. QUARTZ-GABBEO AND QUARTZ-DIORITE, OR TOIS^ALITE. The syenite of President Hitchcock seems to me to have been origi- nally a quartz-diallage rock, but it is now for the larger part a hornblendic rock; indeed, west of the river the presence of diallage can only rarely be rendered certain. The low percentage of silica and the almost complete absence of orthoclase exclude it from the syenites. HISTORICAL. BASIC SECEETIONS: HITCHCOCK'S SUGGESTION OF THE THEORY OF "SOHLIEREN- GANGE." In 1819 President Hitchcock mentions "syenite" as the prevalent rock along the Connecticut on the east side — a statement scarcely correct — and notes that the proportion of hornblende is rather small and that mica is often present. "Porphyritic syenite is common in this quarter and steatite occurs in its eastern part." The first statement is afterwards retracted, and the second I can not explain.^ In 1823 the same author describes the rock more fully from its two localities, Whateiy and Belchertown. He notes first the interesting- fact that in coming from the westward across Northampton ' ' one passes over the most decided granite until he comes within 4 or 5 miles of the village. ' Geology of the Connecticut River: Am. Jour. Sci., 1st series, Vol. I, 1819, p. 106. 332 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. He will then find the texture of the rock to be finer, and in some instances it contains a portion of hornblende, while the proportion of quartz is dimin- ished somewhat and the feldspar frequently becomes red. Coming nearer Northampton, however, we find the hornblende more and more abundant until we arrive at the eastern edge of the range, where we find a rock containing little else than feldspar and hornblende." He notes the abundant veins of granite in the brook 2 miles south of the church in Whately, and the many minute faults of the rock and its peculiar conglomerated character. "The rock here contains numerous embedded masses of other primitive rocks, as gneiss, mica-slate, quartz, hornblende, and a finer kind of syenite, all almost uniformly rounded." He mentions a third locality of syenite west of the road, a mile north of the village of Whately, associated with "greenstone-slate nearly allied to hornblende-slate, and unstratified primitive greenstone. It consists of nearly equal proportions of feldspar and hornblende — the former white and compact or very finely granular, entirely destitute of foliated structure or pearly luster."^ A study of this rock in thin section shows it to be an amphibolite. The white spots of supposed feldspar are parts of the finely granular ground mass, which are free from green hornblende, and it is not diff'erent from the other beds of amphibolite named above, with which it is associated. (See p. 191.) In the Reports of 1833^ and 1835 an extended account of the rock is given, which is reprinted almost verbatim in the Report of 1841,^ with the suppression of a single very suggestive theory. He describes the rock as a quaternary compound of feldspar, horn- blende, quartz, and mica. He describes the "conglomerated sienite" of Whately in detail: It is iu fact a real conglomerate, and in some places the nodules are so numerous that it has very much the aiDpearance of the coarse pudding-stones of the newer rocks. The nodules vary in size from the diameter of half an inch to that of 6 or 8 inches. They are not smoothed, like the pebbles in the more recent conglomerates, by mechan- ical attrition, but they appear like masses of rocks that have been partly melted down by heat. Upon the whole, I think I have ascertained the presence of hornblende- slate, mica-slate, and quartz rock in these nodules. When the rock is broken they are knocked out without difficulty, like the pebbles of a common conglomerate. • Geology of Connecticut River: Am. Jour. Sci., 1st series, Vol. VI, p. 29. ^Eept. Geology of Massacliusetts, p. 463 'Ibid., p. 668. TONALITE. 333 Tlu' inclusions niny iuivo been rounded by slow solution in the melted niiU'iua, this solution attacking corners and edges most rapidly. That the iuclosino- rock is an igneous rock is certain from its microscopical character and its contact effect. That the i)ebble-like masses are true foreign "inclu- sions" does seem pi'obable from their lithological variety and difference from the inclosing rock. They closely resemble pebbles rounded by water, but so many cases of such inclusions rounded by melting (as granite in lamprophyre and quartz in various magmas) have been described that this may be the explanation.^ I am, however, more inclined to consider them secretions in the mass of the rock itself, as they seem to belong wholly to two types which occur commonly in this way. The one is a coarse, black hornblende-biotite aggregate, such as is often seen in small masses anywhere in the rock. Slides of this showed it to agree with these concretionary masses, and not with any amphibolite known as an independent rock in the region. The other is a fine-grained granite or eurite, like that so common in the veins that cut the rock abundantly. Both these rocks seem to have separated from the magma, to have been accumulated here in unusual amount, and to have been roimded by resorption according to the methods discussed in the articles cited above. The locality where these forms are found is at the upper dam at West Brook, on the north line of Hatfield. A third most interesting variety of the syenite is described as follows : Augitic syenite. — The presence of hornblende in this variety and the absence of mica have led me to call it augitic syenite rather than augitic granite, although in position it is associated with granite. There are two varieties. The first is com- posed of black hornblende, greenish augite, and yellowish feldspar, all the ingre- dients except the feldspar exhibiting a very distinct and lively crystallization. This variety occurs in the northern part of Belchertown. The other variety, which I have found only in bowlders in Amherst, consists of augite and feldspar, the former being so arranged in the latter as to present the appearance of letters.^ H.e insists, further, on the low level occupied by the syenite in the valley and upon its columnar structure, and describes in considerable detail the segregated veins which occur so abundantly. He presents a ' See, for several citations, R. Pohlman, Emschliisse von Granit in Lamprophyr : Neues Jahrbuch fiir Mineral, etc., 1888, II, p. 87, and note on page 92. ■■'This is a contact modification of the amphibolite bordering the granite and is described on p. 243. 334 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. discussion of these last in the Report of 1833, suppressed in that of 1841, which, as he develops it, has some resemblance to the theory of "schlie- rengange," as developed by E. Reyer. The greater part of the veins in our syenite consists of material foreign to the nature of the rock and introduced subsequently to its original production. I do not say subsequently to its consolidation, for it has appeared to me possible that while a molten mass of rock — say syenite — was in an incipient state of refrigeration matter of a similar kind still more intensely heated might have been injected into it, so as to form veins. Other veins associated with faulting he refers to the filling of fissures formed in the solid rock. Under the head "Mineral contents" he refers a mineral crystallizing in four-sided ^Drisms to rutile. It is allanite. He mentions quartz crystals and "gashed quai'tz," quartz "full of thin fissures, as if made by random cuts of a knife," and explains it as due to the growth of quartz about some mineral now removed. This mineral was doubtless barite. The minerals of the Hatfield baryta mine are also enumerated. Under the head "Theo- retical conclusions" the hypothesis is advanced that "the syenite Avas formed by the melting down of the hornblende-schist," and in proof of this he pre- sents a rude diagram of the state of things at the north end of the syenite in Whately. In this diagram the syenite is represented as being succeeded to the north by hornblende-schist, the two being in contact and the bedding of the schist being continued in a rude stratification of the syenite. He adduces also the fact that the hornblende-schist on its western border, a mile north of the syenite, is massive, columnar, and feldspathic, and theorizes that there was here heat "sufficient for the production of feldspar, but not for its crys- tallization;" that with a greater degree of heat syenite would have been produced; with a still greater, the production of hornblende would have been impossible and granite would have resulted. The diagram and description of the relations of the syenite and the hornblende-schist in Whately are based upon a serious error of observation. The hornblende-schist Avhich outcrops in great force north and south of the south line of Whately is toward the north separated by argillite from the broad band of hornblende-granite with which the diagram connects it. C. H. Hitchcock, in 1871,^ classed the rock as a Laurentian gneiss. 'Explanation of geological map of United States, in Waiting's Atlas of Massachusetts. TONALITE. , 335 It is put iis iui "exolic Montalbmi granite" on the "centennial map" of W. 0. Crosby,'- because it is micaceous as well as liornblendic and because it contains a center of true g-ranite (according to President Hitchcock's map of 1844) in the southwest comer of Belcliertown — a groundless argument, since the granite in (juestion is simply a great pegmatite dike which cuts the tonalite. • DISTRIBUTION. On the west side of the river the rock commences in "Whately, a short distance southwest of the village, where it is seen in contact with the Leyden argillite, producing a marked contact metamorphism (p. 205), and runs south in a long, bare ridge ("The Rocks"), hke a great dike, into Northampton, where it ends in Elizabeth Rock. It is 6 miles long and 2 miles wide. East, west, and south broad areas of sands and sandstones separate it from its neighbors. To the west of its south end it grades into a great area of biotite-granite identical with itself except in the absence of hornblende. On the east side of the river a great squarish mass occupies the south- west portion of Belchertown, extending into Granby and Ludlow, its con- tacts, unfortunately, greatly obscured by the heavy post-Glacial sands. It is a great batholite and in many places strong contact metamorphism can be observed at its borders and in broad sheets of schists that float out in the center of the great mass. (See p. 243.) North from the northwest corner of the area of tonalite across Belcher- town and Pelham, and so on north in the foothills, is a line of oiitcrops of much crushed rock which seem at times like amphibolite shot through by many small aplite veins and at times like the tonalite. The Shays flint is a peculiar facies of this rock which resembles a petrosilex. As it runs along the western border of Mount Hygeia it is quite gneissoid, but appears . in Leverett in typical development as a beautiful dark-green granitoid rock shot through with epidote veins. Also, going north from the northeast cor- ner of the Belchertown mass along the corresponding eastern foothills of the next valley to the east, across Prescott and New Salem, a similar line of tonalite outcrops occurs, ending with the great block of diorite on the north line of Prescott. These are both lines of strong faulting and crushing. ' Report on a Geological Map of Massachusetts, p. 31. 336 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. ANALYSES OF TONALITE. Analyses of the tonalite were made in the laboratory of Amherst College (1) by William Orr, jr., and (2) by F. H. Fitts. A third (3) was made by L. Gr. Eakins, of the United States Geological Survey: Analyses of tonalite. 1. 2. 3. SiOa 56.69 .62 15.48 6.22 56.18 1.60 1 22.79 55.51 .91 . 16.51 I 1.68 4.57 .11 .02 6.73 6.73 3.19 2.46 1.53 .17 TiOa A1,0, Fe.O FeO MnO BaO CaO 7.59 6.53 3.41 3.43 6.49 6.53 3.40 3.27 MgO NaO K„0 H2O PiOj 99.97 100. 26 100. 12 Analyses 1 and 2 were made from the same hand specimen of the slightly amethystine, fresh, medium-grained tonalite, which showed with the lens dark, bronzy diallage, and bright-green hornblende and amethys- tine quartz. The specimen came from just north of Three Rivers, in Belchertown. Analysis 3 was made from the beautiful epidotic-veined rock from the crossroads east of South Leverett, described on page 339, which was more altered than the others. PETROGRAPHICAL DESCRIPTION. MacroscopicaL — The rock is a wholly granitoid, medium- to fine-grained and very even-grained mixture of quartz, orthoclase (*?), plagioclase, biotite, and hornblende, the latter being at times replaced by a dark-bronze dial- lage in the Belchertown area, a mineral which occurs very rarely in the Hatfield region. The passage of diallage into hornblende can be well observed, and reasons will be given below for the assumption that the rock TONALITE. 337 was orig'hiiiUy erupted as a diallage-biotite-gabbro. It becomes aplianitic in places, but is uever porphyritic or pegmatitic. It is a tough, compact rock not easily disintegrated, but weathering- white. It is light-gray, often greenish, the two colored constituents rarely predominating, so as to give it a dark-gi-ay shade, but often weathering so as to give it a greenish tint. In other cases the feldspar weathers red, and it always has a somewhat compact appearance, the cleavage hardly appearing. In the south of Belchertown a beautiful variety occurs abundantly. The quartz is amethystine, the diallage dark-bronze colored, the hornblende bright-green. Very generally the decomposition of the hornblende has furnished a large quantity of chlorite, which then gives a green shade to the rock. Microscopical. — A description of the quartz-gabbro from South Bel- chertown specimens will be given first, as the least altered form of the rock. In thin sections the quartz shows fluid inclusions with moving bubbles. Long, fine, rigidly straight, opaque needles of rutile occur in great abun- dance, and are often divided into many widely and regularly separated parts, all perfectly aligned. The feldspar is nearly all triclinic, with extinction of adjacent bands at 12° to 14°. Ortlioclase could not be proved to be present. The diallage is in separate, quite well-formed crystals of pale- green color, but so loaded with the customary red and black inclusions as to give it a deep-brown color. In sections parallel to oo P co (100) these are, in abundance, shape, and arrangement, exceedingly like the Labrador hypersthene, and the vertical striation is clearly developed. In sections parallel to oo P co (010) the red plates are not nearly so much shortened parallel to the vertical axis as in the hypersthene, and are so abundant as nearly to obscure the green color of the diallage. In one regular octagonal basal section, while the diallage cleavage is finely developed, and a cleavage less perfect and at right angles thereto is clearly seen, the prismatic cleavage is entirely wanting. In another twinned very clearly after the common pyroxene law, on oo P co (100), all the tln-ee cleavages are developed, the co P co (100) cleavage being much the best. The freshest of these crystals are surrounded by a narrow zone of gi-een, rounded plates of hornblende, in which the black inclusions remain, but the red do not. MON XXIX 22 338 GEOLOG-T OF OLD HAMPSHIEE COUNTY, MASS. In other crystals this change has proceeded in every degree until the crystal becomes wholly changed to a fibrous hornblende (uralite), in which the black inclusions of the diallage still remain, and indeed appear often to be considerably increased at the expense, it would seem, either of the red inclusions or of the diallage itself, and to possess in the hornblende a very distinct zonal arrangement. Furthermore, the hornblende itself is in most cases changed more or less into an aggregate of rounded green scales of chlorite, but sliglitly dichroic, which has in many cases eaten into the center of the hornblende in great patches, in others has surrounded it in a regular layer. Finally, very peculiar and delicate plumose aggregations of elongate, round-ended scales of biotite are gathered in tufts at spots along the outside of the chlorite and attached to it, or a group of such tufts radiates from a center in which, in each case, remains of the green chlorite scales appear. The whole arrangement suggests very strongly the derivation of the biotite from the chloritic mineral parallel with the decomposition of the feldspar. In another case a flat patch of the green chlorite scales seems to change gradually into a mass of brown biotite scales, some of the small plates having the green color and weak pleochroism of chlorite at one end and the brown color and strong pleochroism of biotite at the other, and these latter pass into a single large biotite crystal, so that one can hardly avoid con- cluding that the biotite has been derived, in part at least, from the diallage through the hornblende and chlorite stages. The earliest stage may, of course, have been with diallage surrounded by biotite, and the change to chlorite may have proceeded both ways from the boundary. Many slides cut from various parts of the area show no remains of the diallage, but only the fibrous hornblende containing the zonally arranged black inclusions, and having chlorite and biotite arranged with regard to it exactly as in the slides where the diallage is present. We may thus con- clude that the diallage was once widely and abundantly present in the rock. Apatite occurs in exceptionally large crystals in the hornblende. To the above description of the pyroxenic varieties are added some special notes upon the commoner and more altered biotite amphibole granite, or tonalite, and upon one or two rare varieties. The quartz is everywhere distinctly subordinate to the feldspar, and molds the latter. At Elizabeth Rock, in the north of Northampton, it is TONALITE. 339 ver)' lull of ciu'ities with moving- bubbles, many of the cavities containing water and carbon dioxide and a moving bubble of the latter. The feldspar is uniformly very much more decomposed than the appear- ance of the rock would lead one to suspect. Sometimes the change is into kaolin, sometimes into muscovite. The change is always central, and at times a sharply defined diamond-shaped area of change occurs in a square crvstal. This change is so general that it can only be determined that the feldspar is for the most part triclinic, with extinction at small angles. The hornblende is often twinned, and extinguishes at high angle — 19°-21°. Epidote, in minute groixps in the chlorite, and titanite are abundant in the Whately bed. Allanite is frequent, especially in the Hatfield bed, in crystals visible to the eye, and surrounded by the peculiar radiate puckering or splintery fracture common around this mineral. Under the microscope it is at times suiTOunded by epidote. In the Hatfield mass the biotite is subordinate and the rock agrees exactly with the tonalite of the Tyrol. In Belchertown it is more biotitic. In the latter area, in the region around Three Rivers, the quartz is ame- thystine and contrasts beautifully with the green diallage. This variety shows under the microscope a beautiful granophyre structure. Farther east, in South Belchertown, large bowlders on the railroad show a coarsely por- phyritic development of the biotite, each of the large scales being surrounded by a white border, and the quartz in this variety is violet, like the pre-Cam- brian gneiss in the western portion of the State. THE CRUSHING AND ALTERATION OF THE TONALITE ALONG THE PELHAM FAULT. The outcrops of the tonalite which appear in the line of the great fault at the foot of the eastern plateau from Belchertown to Leverett are greatly altered by the movements which have taken place along that line. Follow- ing the road west from South Leverett to the point where an unused road goes east to the old cemetery, one finds a large outcrop of a beautiful dark- green chloritic tonalite, in which the reddish feldspar contrasts finely with the dark hornblende, and the contrast is heightened by a network of fine, 840 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. dark-greeu, epidotic quartz veins. Between this and the Monson gneiss to the east the lower coarse mica-schist crops out. In thin section the feld- sj)ars are largely triclinic, and the constituents are larger and clearer than elsewhere in the series, but much crushed. The epidotes are especially- large and well crystallized. Titanite, which is not wanting in the other sections, is especially abundant here. An analysis of this rock is given on page 336. The band is covered by heavy sands across Shutesbury, but reappears at "Mount Boreas," above Pratts Corner, and a mile east of the pond a mile south of South Leverett. The western half of the hill is made up of the whetstone-schist, so crushed that dip and strike can be determined with difficulty, and the eastern slope by an equally crushed quartzose amphibolite, while below, by the stream, is a tonalite which is somewhat gneissoid, and is doubtless the continuation of the ornate rock mentioned above, though from the greater amount and the colorlessness of the quartz and feldspar it has not its attractive appearance. This rock continues across Pelham in a narrow band resting in the foothills against the older gneiss and separated by a broad area of sands from the feldspathic mica-schist of the center of Amherst. It is a highly hornblendic granitoid gneiss, much intersected by epidotic quartz veins and often very chloritic. The large mass of leek-green hornstone known locally as Shay's flint,-^ from the tradition that it was used for flints during Shay's rebellion, was found on Amethyst Brook, in Pelham, just where this band crosses it. It was a bowlder, and its oi'igin Avas unknown. Some years ago I found the same material in place where the band of hornblende -gneiss (altered tonalite) crosses the south line of Pelham. It forms beds in the latter sometimes as much as 20 feet thick, and at times crosses the bedding. It is a cryptocrystalline quartz, colored green by chlorite derived from the decomposing hornblende of the granite, passing from green to flesh color and weathering white and grading into ordinary gneiss, and it is a result of the thorough crushing and silicification to which the rock has been sub- jected. The veins colored by epidote are not essentially different from these, though they do not reach such large dimensions, and where the fissure is not entirely filled they show beautiful plane, polished surfaces of ' For tlie history of the rock, which has been called plasma, prase, and green hornstone, see under "Quartz," in A mineralogical lexicon: Bull. U. S. Geol. Survey No. 126, 1895, p. 135. TONALITE. 341 the chalcedonic quartz, colored a lig-lit i)istachio-green by epidote. These surfaces are not "slickeusides," but are as if varnished, and are caused by the crystallization of the fine-grained material. They are analogous to the smooth surface of botryoidal chalcedony or limonite. A similar petrosiliceous variety occurs in Whately — a pale leek-green, subgranular mass, of hornstone-like appearance, with a few crushed mus- covite plates. It shows no biotite or quartz. The luster is generally dull, but here and there the sheen of a feldspar cleavage appears, and this always shows tri clinic striation. It appears at the Hatfield lead mine in thin layers on fissures. (XVIII, No. 57, in Massachusetts Survey Collection.) PETEOaRAPHICAL DESCRIPTION OF THE ALTERED TONALITES. '' Shay's flint," Pelliam, the typical rock. Under the microscope this shows a regularly mottled aggregate polarization which has some resem- blance to clastic structure, but more to that of agate or chemically deposited quartz; and as it resembles exactly the purer parts and the veins of the same hornstone from Pelham, in situ, I have no doubt that it is chemically deposited silica, rendered impure by kaolin and a little green chlorite. It is in large part apolar, and therefore opal. Tonalite, Pelham, west line north of S. Jewett's. Dark hornblende abundant, feldspar flesh-colored. In section very feldspathic, the feldspars (mostly triclinic) greatly kaolinized; all constituents reach the extreme of crushing — the hornblendes opened along cleavage planes; the feldspars crushed and parts moved; the twin striation greatly twisted, and the undulatory extinction greatly obscuring the twinning; hornblende shows jc = green, lj = olive, a = yellow; x;=t)(>a. Green hornstone, like "Shay's flint," from locality where first found by me in place in Pelham, at S. Jewett's. This is a quartz mass, filled with scales like kaolin, which are opaque by transmitted and white by reflected light, and permeated by veins which have the same scattered scales. It resembles exactly, both with and without polarized light, the true "Shay's flint." Some slides show a beautiful microbrecciation from crushing. They contain magnetite and a little green chlorite. Under the polarizer there appear now and then larger, rounded, transparent grains, which may be 342 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. the original quartz grains. The structure seems, however, in general much more like that of agate. It contains much opal. Pelham, south line, 40 rods east of western road. A coarse, schistose, hornblende-gneiss; coarse, wavy cleavage siirfaces of hornblende make up foliation faces, luster-mottled by roinided grains of fresh white feldspar, in which cleavage is feebly developed. In section the hornblende is in large, fresh plates, exactly like the few developed in the section last described ; it shows deep colors, weak pleochroism, and is much cracked and twisted by pressure. The feldspar is very fresh, and shows a great variety of triclinic striation — very broad to very narrow bands with perfectly parallel sides, and tapering, interrupted, and offset bands ; also bands wavy and contorted by pressure and associated undulatory extinction. In one crystal, cut at right angles to P (001), all the laws of twinning are beautifully developed. Belchertown, northwest corner, 40 rods east of R. Thayer's. A green granitoid rock of medium grain, mottled with flesh-red from decomposed feldspar; distinctly foliated. In section broad hornblendes much crushed, feldspars crushed, showing undulatory extinction, much kaolinized, many triclinic, with small extinction angle; much chlorite and epidote, the latter often with distinct crystal faces externally, but with rounded zonal struc- ture internally, the spherical center extinguishing first and then successive zones in order to the surface, with revolution of 17°. A little farther south, on same band, north of house of A. Groodale, the wholly crushed and altered rock is hornstone-like, with a dull mottling of greenish and flesh color. In section the bisilicate is almost wholly removed, and the quartz-plagioclase mass is wholly crushed, with wavy extinction and twisted twin laminse. This is the south end of the "Shay's flint" band. Followed 40 rods east, its contact on Pelham gneiss is seen. There is a hornblende-biotite-gneiss for a rod at the contact, and the Pelham gneiss is full of granite dikes. DIOBITE. North Prescott and New Salem. — A great oval area of diorite, 3 miles long from north to south and about a mile wide, lies across the line sepa- rating the above towns. It is a resistant rock, and makes the whole of Packards Mountain in the latter town. It is surrounded on all sides by the gneissoid quartzites, which dip uniformly to the west, undisturbed by the intrusive rock. On all sides as one approaches the mass the quartzite DIORITE. 343 grows more gneissoid fi'ora contact iuflvieiice, but this is not marked. The rock is normally dark-gray or nearly black, with a shade of brown, and seems at first sight to be fine-grained; but when held to the light it is seen to be made up of squarish surfaces, from a half to three-fourths of an inch across, of jet-black to dark-green hornblende, very beautifully luster- mottled by fresh, white, striate, broad lath-shaped plagioclases, and show- ing rarely a grain of quartz, garnet, or a black ore. It is in places bedded, and on the west, in the hill above Cooleyville, one ti'aces the amphibolite into immediate proximity to the diorite, where it is thickened imusually, is massive, and greatly resembles the diorite. It may be a compacted and altered ash bed, associated with the eruptive rock. In the southwest portion of the mass, near A. Pierce's, in Prescott Hollow, the diorite is a coarse, white, feldspathic, slightly saussuritic rock, with only small, distant patches of a dark silicate, now changed to a mixture of actinolite and biotite. The freshest material for microscopical study was obtained from a great bowlder on the north side of the road west from Prescott Center, near the last house in the village. (See PI. Ill, fig. 3.) It presents a very attractive appearance under the microscope. A portion of a single horn- blende crystal occupies the whole field, notched by the regular crystals of feldspar, which run in every direction. It shows a maximum extinction angle of 22°, and is therefore near labradorite. It is quite fresh, and full of acicular needles. The hornblende is deep-green, extinction 20°, with slight pleochroism, c=ii Terrace sands Very fine bufl' sand. . . J Rather fine-grained, light-red, micaceous sandstone, muscovitic Same, but much mixed with sand Light-red, fine-grained, muscovitic sandstone Same Same Red, fine-grained, muscovitic sandstone Dark-gray, shaly, muscovitic sandstone Red, very fine-grained, muscovitic sandstone Brick-red, very fine-grained, muscovitic sandstone Same Same Gray, very fine-grained, muscovitic sandstone Dark-gray shale Same Brick-red shale Brick-red, shaly, muscovitic sandstone Same Brownish-gray, shaly sandstone Dark-gray, shaly sandstone Same Same Pale-buff, shaly sandstone , Chocolate-colored, fine-grained, micaceous sandstone , Like 260 , Light-gray, fine-grained, micaceous sandstone Dark-gray shale Same Same Reddish-gray, fine-grained, micaceous sandstone Dark-gray, fine-grained, micaceous sandstone Dark-gray shale , Dark-gray, fine-grained, micaceous sandstone Reddish-gray, fine-grained, micaceous sandstone Brick-red, fine-grained, micaceous sandstone Same Same Gray, fine-grained, micaceous sandstone Light-gray, fine-grained, biotitic sandstone Brick-red, muscovitic sandstone Same Feet. r 25 30 [ 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 400 410 420 430 440 450 ARTESIAN WELLS. 383 WELL AT HOLYOKE. Artesian well of Parsons Paper Company, at Holyoke; near west end of clam; October, 1884. E. W. Cliapin, superintendent. The samples in this series were especially clean, and the method of boring favored the production of large fragments. All the samples are very fine, and the distinction di'awn between shale and sandstone is a very close one. The specimens were carefuU}' bottled and labeled with depth and date, and given to the Amherst College cabinet by Mr. Chapin. Record of artesian-well boring at Holyol-e. Kiud of rock. Sand, etc. (not reported) , Surface of rock : a dark-gray shale, black when wet; much efferves- cence with HCl ; melts to light-yellow glass Same black shale ; small drusy surfaces of calclte crystals, apparently E^ with edges out by — 2Raud apex by — JE Fine sandstone, dark chocolate-brown, slightly calcareous, micaceous. Same sandstone, slightly redder and more flaky ; same calcite crusts as 102 Same very fine, micaceous sandstone, dark-gray ; some grains colorless ; sandstone with biotite and muscovite scales Fine, light-gray, micaceous sandstone; abundant calcareous cement... Mixture of 132 and 150 ; calcite crusts Dark-gray shale ; calcareous, pyritous grains Fine-grained, chocolate, micaceous sandstone, a little coarser than any- thing preceding ; calcareous Dark-gray, highly calcareous shale; silky, white, acicular efflorescence on some grains Same as 164 Same as 160 , Same as 190, but more calcareous; like the finest-grained, thin-bedded sandstones, which often show insect tracks Same as 200 Mixture of 215 and 230 Bluish-black, slightly micaceous shale ; very calcareous Dark-gray, micaceous and pyritous shale Fine, black, calcareous shale Same as 240 Dark-gray, calcareous shale Dark-chocolate, shaly sandstone Mixture of 260 and gray, shaly sandstone Same as 270 Mixture of dark- and light-gray shale; many grains show efflorescence of iron Feet. 85 102 113 115 130 132 140 150 160 164 176 190 200 215 223 230 235 240 244 255 260 270 280 285 384 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. Record of artesian-well boring at Holyohe — Continued. Kind of rock. Dark-gray shale mixed with red. Same Fine, coal-black shale : calcareous . Same Chocolate, shaly sandstone. Same Feet. Dark-gray shale. Same Bright-chocolate shale. Same Dark-gray shale Black shale Light-gray, calcareous shale Dark-gray, calcareous shale Dark-chocolate, iiue-grained sandstone. Same Same ; more micaceous Black shale Gray shale Dark-gray shale with layer of light-buff, highly calcareous sandstone.. Dark-chocolate, shaly sandstone Same ; micaceous Dark-gray shale with admixture of 475 Dark-gray shale ; layers of fine-grained, light-gray, micaceous sandstone. Dark-gray shale ; micaceous and only slightly calcareous Reddish- gray, fine-grained, shaly sandstone Chocolate-colored, fine-grained, muscovitic sandstone 290 300 308 318 330 338 347 355 360 365 375 380 385 390 400 408 420 435 440 460 472 475 485 490 500 505 510 Parsons Paper Company, Holyoke, Mass. A separate partial series from the same well as the last. Samples deposited in collection at Amherst College. Record of artesian-well boring at Holyohe. Kind of rock. Red-brown, fine-grained, marly sandstone, very ferruginous Black shale Very coarse, rusty sandstone Coarse, dark sandstone Same, granitic Black shale Same Dark-brown sandstone Feet. 250 365 400 420 425 435 445 455 ARTESIAN WELLS. Record of artesian-well boring at Holyoke — Continued. 385 Kiud of rook. Diirk-l>rown sandstone Same, granitic Ferruginons marl DarU-brown sandstone Same, coarse gneiss grains Black, shaly sandstone Darli-gray, shaly sandstone Brown sandstone , Dark brown-gray, marly sandstone Brown sandstone Coarse, rusty, granitic sandstone.. Feot. 460 480 490 500 530 565 570 590 615 645 685 WELL AT NORTHAMPTON. Nortliampton, at Belding's silk mill, south of the railroad station. Com- menced in 1885. Depth, 3,700 feet; mouth of well, 125 feet above sea level. In New Red sandstone. Samples furnished by the borers of the well and deposited in the Amherst College collection. Record of artesian-well boring at Northampton. [Abbreviations: q, quartz; f, feldspar; m, muscovite; b, biotite; g, garnet.] Kind of rock. Sand Clay Red sand (probably till) Red sandstone, borings Coarse, buff sand, white to amethystine quartz, flesh- colored feldspar grain Similar, but finer and more rounded grains Same Same ; few scales muscovite and hornblende Same Same ; many grains deep red-brown from rust covering, which has been usually worn off by the attrition of the driU Very fine, buff sand, quartz, orthoclase, and abundant muscovite scales Like 730 Same Buff sands, quartz, feldspar, little muscovite. Average grain, in millimeters. itoli itoli itoli 4 toll i to li i Feet. 140 150 535 682 692 710 730 750 780 910 930 950 970 MON XXIX- -25 386 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. Record of artesian-well boring at Northam'pton — Continued. [Abbreviations: q, quartz; f, feldspar; m, muscovite; b, biotite; g, garnet.] Kind of rock. Buff sands, quartz, feldspar, little muscovite Same Finer, darker-brown, much mica Coarser, light-buff, granitic, q. f. m Same, q. f. m.b Same Same Same Same Light-buff, granitic, q. f. m. b Same Same Fine, light-buff, granitic, q. f. m. b Fine, light-buff, granitic, q. f. m. b. ; coal Light-buff, granitic, q. f. m. b Same Same Fine, light-buff, granitic, q. f. m.b Same Same Same Same Same Same Same Same Medium, buff sand, q. f. m Medium, buff sand, q. f. m. ; few worn grains black slate Same Medium, buff sand, q. f. m. ; white slate Same Medium, buff sand, q. f. m. ; black slate Same Same Coarse, buff sand, q. f. m. ; black slate Medium, buff sand, q. f. m. ; garnet, slate Medium, buff sand, q. f. m. ; slate Medium, buff sand, q . f. m. b. ; slate Same Medium, buff sand, q. f. m. b. ; slate trace Same Average grain, in millimeters. Feet. 1 990 1 1,010 i 1,030 Uto2 1,050 Uto2 1,070 lito2 1,090 itoli 1,110 ■itol* 1,130 |tol| 1,130 1 1,170 1 1,190 1 1,210 * 1,230 i 1,250 1 1,270 i 1,310 itol 1,330 i 1,350 ito2 1,370 *to2 1,390 itol 1,420 *tol 1,440 + to2 1,460 ito2 1,490 ito2 1,510 ito2 1,530 itol 1,550 itol 1,570 itol 1,590 itol 1,610 Itol 1,630 itol 1,650 Itol 1,670 itol 1,690 lto3 1,695 itol 1,710 •itol 1,730 itol 1,750 itol 1,770 itol 1,790 itol 1,810 AKTESIAN WELLS, Record of artesian-well boring at Northampton — Continued. [Abbreviatious : q, quartz; f, feldspar; in, musoovite; b, biotite; g, garnet.] 387 Kind of rock. Meiiium, bnflf sand, q.f. m. b. ; garnet Medium, buff sand, q. f. m. b. ; garnet, slate trace Same Same Medium, buft' sand, q.f. m. b. ; slate trace Same Same Same Same Same Same Medium, buff sand, q. f. m. b. ; magnetite Medium, buff sand, q. f. m. b. ; slate trace Same Same Same Same Same Same Same Same Same Same Same Same Same Same Same Same Same Same Same Same Same Same Same Medium, buff sand, q. f. m Same Same Coarse, brown sand, q. f. m Same Average jE^rain, in millimeteTs. itol Itol itol Itol itol I to 2 ito2 I to 2 ito2 I to 2 itol itol itol itol itol itol itol itol itol itol itol itol itol itol itol itol itol itol itol itol ito 1 itol itol itol itol itol itol itol itol 1 to5 1 to3 Feet. 830 850 870 900 920 940 960 980 020 030 040 060 100 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 230 340 350 360 370 380 400 388 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. Record of artesian-well boring at Northampton — Continued. [Abbreviations: q, quartz; f, feldspar; m, muscovite; b, biotite; g, garnet.] Kind of rock. Coarse, brown sand, q. f. m . Same Same Coarse, brown sand, q. f. m., 1 grain 10™™ ; second piece, deep-red, fine, micaceous sandstone Coarse, brown sand, q. f. m Medium, brown sand, q. f. m Same Fine, bufif8and,q. f. m Medium, red sand, q. f. m Medium, reddish sand, q.f. m Same Same Same Same Same Medium, buff sand, q. f . m Same Same Medium, buff sand, q. f. m. ; more rusty Medium, brown sand, q. f. m. ; more rasty. Fine, brown sand, q. f . m. ; more rusty Same Medium, buff sand, q. f. m Medium, buff sand, q. f. m Same Same Fine, brown sand, q. f. m Same Same Same Same Same Medium, brown sand, q. f. m. Same Same Same Same Same Same Average grain, in millimeters 1 to 2 1 to3 1 to3 1 to3 1 to3 1 to2 1 to2 itol 1 to2 to 2 to 2 to 2 to 2 to 2 to 2 to 2 to 2 to 2 1 to2 itol itol 1 to2 1 to2 1 to2 1 to2 itol itol -.. i ito i i-to i ito i ito 4 to 2^ to 2 to 2 to 2 to 2 4tol itol Feet. 2,410 2,420 2,430 2,440 2,450 2,460 2,470 2,480 2,490 2,500 2,510 2,520 2, 530 2,540 2,550 2,560 2,570 2,580 2,754 2,770 2,774 2,826 2,851 2,872 2,901 2,933 3,024 3,100 3,200 3,250 3, 300 3,350 3,500 3,525 3,536 3,550 3,650 3,675 3,700 PSEUDOMORPHS OF CALCITE AND DOLOMITE. 389 WELL ON WKSTFIELD LITTLE RIVER SOUTH OF WESTPIELD. Artesian well at Crane Brothers' paper mill, on Westfield Little River, south of Westfield; bored by Daniel Dull, New York. Sunk 1,110 feet in conglomerate; unsuccessful and now closed up. A soft, black, pulverulent layer reported. PSEUDOMORPHS OF CALCITE AND DOLOMITE AFTER HOPPER- SHAPED CUBES OF SALT. It has been reported for many years that chiastolites occurred in the sandstone in West Springfield, but I can not find that anything has been published upon the subject. Specimens containing small white crosses of about the size of ordinary chiastolites and having some resemblance to them were brought to me some years ago by a student, who informed me that they were discovered by Mr. B. Hosford, of Springfield. These specimens were lost in the fire which destroyed the Shepard collection. Later, through the kindness of Mr. J. S. Diller, I received another specimen with permission to sacrifice it, and I had several slides cut from it. It shows white squares and triangles on a black ground of fine-grained, shaly, bituminous sandstone, but this ground is not marked off from the rest of the surface of the sandstone by any square or round boundary representing the cross-section of a prismatic crystal in which the white lines should be diagonal, so that the resemblance to chiastolite is only superficial. These slides are figured in the Miner- alogical Lexicon^ of the three counties. On touching the white areas with acid an abundant effervescence occurred, and under the microscope they proved to be made up of calcite, quite white and coarsely granular down the central portion of the bands and very finely granular and gathered in minute rounded concretions just visible with the lens on either side of these central bands, the concretions grouped with more or less of the dark ixiaterial of the sandstone inter- vening, so as to give the whole a brownish shade. The calcite was not confined to these bands, but impregnated large portions of the sandstone, so that, when polished, parts where there was no calcite remained dull and other patches took a fine polish. It is plain that cubical crystals of salt 1 Bull, U. S. Geol Survey No. 126, 1895, under "Salt." 390 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. with excavated hopper-shaped faces had been embedded in the mud, dis- solved out, and their j)lace taken by the calcite, which has largely impreg- nated the sandstone, but which shows the white color only where it occupied the cavities of the salt hoppers. The locality as given me by Mr. Diller is along the south bank of the Westfield River, in West Springfield, near the water's edge, and just below the large dam some distance above (west of) West Springfield village. Later a specimen was found at Holyoke, near the west end of the rail- road bridge, and is now in the Smith College collection. It is larger and much more delicate than the Westfield specimens. It is figured and described in detail in the Mineral Lexicon.^ The piece must have come from a very short distance northwest, and I have observed single hopper-shaped casts in the shale at the cutting within the city of Holyoke, and similar forms on the shale at Ashley's pond, farther west, and at many other localities in the shale. In 1895 a large share of the finest specimens collected by Mr. Hos- ford came into my possession — the best piece of all through the kindness of his daughter. This is a finely ice-polished slab of black shale, covered with small white figures, three or four to the square inch, in great variety, formed by the various cross-sections of single and aggregated cubes, whose faces were excavated into hopper shapes to various depths. Three-rayed, four- rayed, and six-rayed forms were most common. The center of each ray, marking the trace of the six planes which connect the cube edges, is gener- ally very dark, so that it stands out against the white calcite, and where the faces are only slightly excavated, so that the calcite is now nearly a square, the resemblance to a chiastolite is striking. This darker band is calcite colored by petroleum or coaly matter, and in some cases it is a quite wide band of pure asphaltum. It would seem that the solution of the salt and its replacement by white calcite progressed slowly from the outside at a low temperature. At the last the central band of salt was removed and calcite took its place when somewhat more elevated temperature prevailed, so that bituminous matters were distilled into the empty spaces along with the last calcite. In other specimens cubes are found with only slightly excavated faces, which are made of quite coarsely crystalline calcite, irregularly colored by bitumen. Other pieces have slickensided faces, with surfaces of fine-fibrous graphite ' Bull. V. S. Geol. Survey, No. 126, 1895, under "Salt." THK TKIASSIC SANDSTONE AS A BUILDING STONE. 391 (which is rcMUiirkable, as there is here no other trace of marked heat action), together with veins of coarse-fibrous calcite, grains of galena, and films of gypsum. Other cubes are flat-faced, but a little elongate, and made of fine-grained calcite. At times the rays are broadly bordered by delicate feathery growths of white limestone, which shows a fine, concretionary, almost oolitic structure imder the microscope. The thin-bedded rusty sandstone from the island at Turners Falls, which contains the ferruginous concretions, contains also remarkable salt pseudo- morphs — skeleton cubes with each bar nearly an inch long. The interspaces are now filled with limonite, which was doubtless at first an iron carbonate. THE USE OF THE TRIASSIC SANDSTONE AS A BUILDING STONE. The Sugar Loaf arkose is somewhat used for rude masonry, such as embankments, walls, bridge piers, etc. The large qxiarry on the northwest shoulder of Mount Tom furnished the stone for the piers of the railroad bridge over the Connecticut River at Northampton, and had been long worked for similar purposes. The rock is too coarse for architectural use; if it were not its light color would make it very valuable. The Longmeadow sandstone, under the name " brownstone," has been for a long time in high repute as a building stone of the greatest value, and it has been exported to great distances and employed upon the most expen- sive buildings. The report Mineral Resources of the United States for 1890^ states that the sandstone produced in Massachusetts during that year was valued at S649,097, and of this amount $563,179 was furnished by Hampden County, and came from the quarries extending south from Six- teen Acres, in Springfield, to East Longmeadow. The following, copied from an article in the Springfield Republican of May 9, 1884, and verified by me in all important particulars, gives a good account of the industry at that date : The Norcross Brothers are the largest shippers of stone from East Long- meadow, having last year loaded 115,000 cubic feet of brownstone for building purposes on about 900 freight cars. In addition to this amount, 35,000 cubic feet ■was quarried during the year, but kept in the yard to furnish winter work for the stonecutters. Two quarries, located within a mile of the East Longmeadow depot, the Saulsbury and Kibbe, furnish all but a small part of tlie product and give 'Issued by the United States Geological Survey, p. 402. 392 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. employment to about 200 men for a greater portion of the year. There is also a third, yielding a finer-grained, harder stone, which occurs, however, in a thinner stratum and at greater depths below the soil, so that it is now used only on special orders. All the stone is brownish-red iu color, does not flake on exposure to the weather, is free from stratification, and evenly hard throughout— that from the Kibbe having a somewhat richer red hue than the others. Tbe quarries are located on nearly flat ground, and the sandstone croppings are over 10 feet thick, with a dip toward the southeast of about 10 degrees. As the line of the dip is followed the stone becomes finer in texture and harder, and the stratum also thickens, until, when covered by 20 feet of soil, it forms a layer from 20 to 30 feet through Above the stone is found a mass of slaty brownstone, and below it is the same material, although explorations indicate the existence of another stratum of good rock at a small distance below the first. But little powder is used in quarrying, most of the work being done with picks and wedges. Blocks weighing in the rough from 5 to 6 tons are frequently taken out and sometimes shipped uncut, and one block of 12 tons weight has been successfully quarried and raised. Water causes much trouble and expense, and in the Saulsbury workings a steam pump, throwing 60 gallons a minute, is employed for an average of twelve hours a day to keep down the flow from springs and surface drainage. The quarry work lasts from April to Decem- ber, and during the winter months a force of laborers is employed in stripping the rock and removing the soil and waste to old workings. About half of the stone quarried is dressed before shipment. The firm is now using Longmeadow stone either iu solid walls or as trimmings on the following contracts : The Union Theological Seminary, a four-story 200 by 125 foot building, on Park avenue, ISTew York, which will cost $300,000 when finished in May: the St. James Episcopal Church, to cost $125,000, and cover a space of 120 by 72 feet on Madison avenue, New York; the Jefferson Physical Laboratory for Harvard College, a four-story building, 70 by 212 feet, with the peculiarity that in portions of it no iron, even in the form of nails, will be used on account of pos sible magnetic action; for the University of \^ermont, at Burlington, a library building of Kibbe sandstone, to cost $100,000; on Eighth street, St. Louis, Missouri, an eight-story 64 by 130 foot building, to cost $225,000, for the use of the Turner Eeal Estate and Building Association ; at Lawrenceville, New Jersey, eight buildings, to cost $325,000, for the Lawrenceville Academy. The Norcross Brothers quarry three shades of stone, the trade names by which they are known being " Maynard," a bright-red stone; " Kibbe," a dark-red; and " Worcester," a brown. There are a number of Springfield men interested m getting out stone for buildings, and the East Longmeadow quarry of James & Marra, of this city, lies near the Norcross Brotliers works, and the stone obtained from it much resembles the Kibbe rock in quality, although of a slightly lighter color. The quarry was first worked about sixty years ago by a man named Saulsbury, but only small amounts of stone were taken out until it passed into the hands of Nathaniel Billings la 1882 the present owners bought the property of him, and have since added to THE TRIASSIO SANDSTONE AS A BUILDING STONE. 393 it, iiutil tliey now own 174 acres of good stone laud and two more (luanies, both of which have been opened, but are now unworked. In the Billings workings the rock was L'O feet thick where first qnarried, but by following its dip of about 10 degrees to the southeast the owners tind it increased to 40 feet of unstratitied and little seamed stone. About 20 feet of earth cover the layer of stone at present worked, and below it is found a deposit of soft, shaly rock. Water is a trouble- some feature of this quarry, and a steam pump is kept at work much of the time. During nine mouths 85 men are employed around the works, 10 of whom are stonecutters; the same firm keeps 25 cutters at work in the Franklin street yard in Springfield. About 100,000 cubic feet of rock was shipped from the quarry last year, and nearly half of this amount was dressed before it was sent away. The largest contracts for stone either completed within a year or now being finished are: An order for 20,000 feet for Judge Tree's house in Chicago, Illinois; for the Union League Clubhouse, Chicago, 35,000 feet, and for the Second Congregational Church, Holyoke, 35,000 feet. The Springfield quarry, located within the city limits, 4 miles out on the Hampden road, owned by W. & E. W. Pease, was first worked in 1882, and lies on a tract of 30 acres, bought from John Eockford. The ledge first quarried was 20 feet thick at the croppings and of fine quality browustone, but at a few feet below the surface a large spring was struck, which made operations too expensive. In the second opening two ledges, each 12 and 14 feet thick and separated by a layer of shaly stone, are worked, and 20 feet of sand is at present removed to get at the deposit, which dips toward the southeast at an angle of about 30 degrees. Water has not yet proved troublesome. About 50 men are employed, and last year nearly 100,000 feet of stone was shipped over the New England Eoad, mainly to the eastern part of this State. The Palmer depot and the new Taftsville mill are conspicuous examples of buildings trimined with stone from this quarry. The Carlisle Stone Company owns a browustone quarry not far from Sixteen Acres, and last year employed 26 men and shipped to Boston by way of the Indian Orchard depot 25,000 cubic feet of rock, of which only a small proportion was dressed. A tract of 60 acres, including the present quarry, which was first worked four years ago, was bought by the company in 1881, and the stone obtained since that time bas been of fine quality, although of a lighter red color than Longmeadow stone. The stratum is 18 feet thick, dips about 15 degees to the east, and is cov- ered by 12 feet of sand and 2 feet of hardpan. No shaly rock is found, but flinty bowlders occur, and water causes considerable trouble in the spring months. M. A. Glynn works a quarry at East Longmeadow, about a mile north of the depot, and obtains a fine quality of browustone, which he sells undressed to several New England dealers. The Glynn quarry was opened ten or twelve years ago, but was worked only a little. It was bought, with 7 acres of land, by the present owner a year ago from the Enfield Shakers. The rock is covered by 5 feet of earth, without hardpan or shaly material, and is of uncertain thickness, having been worked only to a depth of 16 feet as yet. Water is not troublesome. Last year 8 quarrymen were employed and 12,000 cubic feet of stone were sold. 394 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. Mr. Greorge P. Merrill^ gives the following data: Price per cubic foot, $1 to S2; cost of pointing, 10 to 15 cents; ax-hammering, 30 cents. An extra price is charged for blocks of more than 25 cubic feet. Strength of the sandstone per square inch, 8,945, 8,812. The stone sells now (1893) for 60 to 70 cents per cubic foot deliv- ered at the station, while the Berea stone sells for $1 per foot delivered in Springfield. PAIiEONTOLOGY. In the following section I have given, by means of the synonymy, the history of opinion concerning each species known to me, and each locality where the species has been found. I have not, however, given special attention to the foot-tracks, and present only the latest list of species prepared by Professor Hitchcock. PLANTS. Halymenites shepardi E. Hitchcock. 1833. Fueoides sp. E. Hitchcock. Geol. Mass., p. 233, pL 13, flgs. 38, 39. 1835. Same. 1811. Fueoides shepardi E. Hitchcock. Geol. Mass., Pinal Eept.,p. 455, flg. 95. These forms occur so abundantly throughout the central areas of fine-grained sandstones and shales that I have often called these the fucoidal sandstones. The best locality, in addition to those mentioned below, is at the water shops in Springfield. I append Hitchcock's description, premising that branching is not so rare as it would indicate. Bauds of the sandstone several feet thick are at times filled evenly full of these rods, and inter- vening bauds are empty, as at the mouth of Fall River: This relic varies in size from one-tenth of an inch to an inch in diameter. More commonly it runs through the rock in a direction corresponding to that of the laminae, in which case it is considerably flattened. Sometimes it passes obliquely through the layers, and very commonly crosses them at right angles, in which last case it has a cylindrical form. It is rare to see a specimen of any considerable length that is not more or less curved, and I have never met with one that was branched at all. I have noticed specimens a foot or more in length, and they may be much longer than this, since I have not met with any large mass of rock containing them. The sandstone in which they are found is rather fine and quite soft and easily disintegrates. They occur near Hoyt's quarries, 1 mile west of the village of Deer- 'Cat. Nat. Mus., Washington, pp. 54, 499. PALEONTOLOGY. 395 (iekl, and also a few rods south of the county jail in Greenfiejd, close by the stage road, and on the road to TJernardstou, a mile north of the village of Greenfield. The vegetable matter in these remains is wholly replaced by sandstone. By breaking the specimens transversely a curious structure is revealed. It may be described by saying that the cylinder is made up of convex layers of sandstone piled upon one another; and I observe that la the same rock all the specimens have the convex sides of these layers in the same direction, so that on one side of the rock you will see numerous button-like protuberances and on the other side correspond- ing concavities. (No. 258.) Bat I do not know which side is uppermost in the rock, iu situ.' I allow the above to stand, altliotigli the forms now seem to me to be tiibulai- ferruginous concretions, the result of the circulation of iron-bearing solutions in the sands. After forming the concretions the solutions have gone on to cement the intervening sand into a red sandstone. Of the other figures presented in the Geology of Massachusetts in 1841 as plants, fig. 89, p. 451; fig. 91, p. 453; figs. 92 and 93, p. 454; and figs. 3 and 5 on pi. 28 (cited as 29 in the text) are dubious impressions, which are very common in the sandstones. Some may have been caused by fucoids ; others, as fig. 3, by the dragging of the roots or branches of float- ing trees rising and sinking with the waves. Fig. 94, p. 454, represents ferruginous concretions ; fig. 1, pi. 28, is a track. For fig. 92, the name Fucoides connecticutensis is suggested on p. 453. Clathropteris platyphylla Bronsfn. 1841. " Peculiar vegetable relic," like a fern. E. Hitchcock, Geol. Mass., p. 452, fig. 90. Teste, E. Hitchcock, jr. 1854. C. rectiusctdus. E. Hitchcock, jr. Description of a new species of Clathrop- teris, discovered in the Connecticut Valley sandstone. Am. Jour. Sci., 2d series, XX, p. 22 ; figured in the text. 1858. G. rectiusculus. E. Hitchcock. Ichnology of Massachusetts. PI. V, fig. 1; PI. VII, fig. 1. 1890. G. lylatyphylla Brongn. J. C. Newberry, Fossil Fishes and Fossil Plants of the Triassic Eocks of New Jersey and the Connecticut Valley: Mon. U. S. Geol. Survey, Vol. XIV, p. 94, PI. XXII. Locality: Bassett's quarry, on the west face of Mount Tom, in East- hampton, just below the Holyoke trap sheet, iu coarse, buff arkose. The type specimen is in the museum of Williston Seminary, at Easthampton. A large series in the Amherst College cabinet, where are also specimens from the quarry of Roswell Field, in Gill; also from the banks of the ' E. Hitchcock, Geol. Mass., 1841, p. 456. 396 GEOLOGY OF OJjD HAMPSHIRE COUNTY, MASS. Connecticut in Moptague, 2 miles southwest of the latter place, as noticed by E. Hitchcock in 1841. The latter localities are a coarse, gray arkose. I am convinced, from an inspection of European specimens in the museum at Munich, of the identity of this species with C. platyphylla. Dr. Newberry cites, also, Westfield, Massachusetts, Durham, Comaecti- cut, and Newark and Milford, in New Jersey. Maceot^niopteris magnifolia Schimper. Small leaves for this species, about 3 inches long, and a little more cordate than the figures. In black shale; Turners Falls. Maceot^niopteris sp. 1843. Tceniopteris vitata. E. Hitchcock. Trans. Assn. Am. Geol.. Vol. I, p. 294. From a bowlder of dark-gray sandstone on Mount Holyoke. The impression is nearly 2 feet long.^ I can not find the specimen in the Amherst collection. Similar large leaves occur in the north part of Montague, on the road going down to the bridge to Grreenfield. AsTROCAEPUs viEGiNiENSis Fontaine. A very poorly ^^reserved specimen of a large frond with strong rachis and long, straight piimse. On buff arkose, like that under Mount Tom, containing Clathropteris. From the collection of President Hitchcock, who said it came from the valley, but could not give the exact locality. Pachyphyllum simile NewbeiTy. 1857. Walehia variabilis E. Emmons. American Geology, p. 108, fig. 76. 1890. Pachyphyllum simile Newberry. Fossil Fishes and Fossil Plants of the, Triassic, p. 88, PI. XXII, fig. 2. Includes the larger and longer and sharper-leaved twigs of coniferous plants. Locality: Turners Falls, in black shale. Pachyphyllum brevipolium Emmons sp. 1823. Unknown relic. E. Hitchcock. Geology of Connecticut Elver. Am. Jour. Sci., 1st series, Vol. VI, p. 80, pi. 9, fig. 5. 1832. Lycopodites sUHmanni De la Beche. Manual of Geology, 2d ed., p. 419. 1841. Possibly a Voltzia. E. Hitchcock. PI. 28 (cited 29), fig. 2. 1843. Possibly a voltzia. E. Hitchcock. Trans. Ass. Am. Geol., Vol. I, p. 294. 1857. Walehia brevifoUa E. Emmons. American Geology, p. 108, figs. 74, 75. 1 E. Hitchcock, Trans. Assn. Am. Geol., Vol. I, 1843, p. 294. PALEONTOLOGY. 397 1858. Cone and twig. E. Hitchcock. Ichuology of Massachusetts, IM. VII, fig. 2. 1890. racliypliijllitm hrevifolinm Newberry. Fossil Fishes and Fossil Plants of the Triassic, PI. XXII, figs. 3-3c. The L. silUmantii is quoted above, from Hadley, Connnecticut, doubt- less a mistake lor Massachusetts, and the phxut was carried from here to Europe.' It Avas described (1823) from the fish locaHty at Sunderlaud. It occurs at Turners Falls; and I have found it quite abundantly at the cut just south of the south line of Holyoke, below Holyoke dam, and at the adjacent cut on the raih-oad to Westfield; also in the northwest of Mon- tague, where the road goes down the hill to Greenfield. Its small cypress- like twigs often spread over slabs 2 or 3 feet square. Its small cones, about an inch long, are figured in the last two works cited above. ScHizoNEURA PLANicosTATA Rogers sp. 1883. 8. planicostata Fontaine. Older Mesozoic Flora of Virginia: Mon. U. S. GeoL Survey, Vol. VI, p. 14, PI. I, fig. 1. 1890. S. planicostata, J. S. Newberry. Fossil Fishes and Fossil Plants of the Triassic, p. 87. Palissya? sp. Many flattened fragments of branches or stalks of plants occur, especially in the arkose. These are transversely jointed, from shrinkage in the process of change to bituminous coal, and are faintly striated longi- tudinally. Larger trunks occur at times as cylinders of sandstone crossing the laminations of the sandstone, 12 to 20 inches in diameter. President Hitchcock mentions stems of plants "converted into vesic- ular amygdaloid," and he figures a specimen from a bowlder in Amherst, which he evidently supposes came from the upper portion of the Greenfield trap sheet.^ Trunks of this kind are doubtfully referred to the above coniferous genus by Dr. Newberry. The specimen is a tapering, rough- surfaced rod, of rounded, cordate cross-section, 2 feet long, 2^ by IJ inches at one end, and 1| by 1 inch at the other. The inclosing rock is a dark greenish-gray diabase, of the type of the freshest, medium-grained rock of the Deerfield bed. The tube is made up of a slightly finer diabase, with steam cavities filled with delessite. There is no trace of tuff sti-ucture in the rock or in the slides of either portion. It is a case where a branch was enclosed in 'E. H. Lee, Geol. Eept. 1833, p. 233. 2 6eol. Mass., Final Kept., 1841, p. 457, fig. 96. 398 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. the liquid lava, burned, and the cavity immediately after filled with lava from above. I have collected such specimens at Kilauea. This rock came probably from near the south end of the Deerfield sheet, where the river crosses it, and where several amygdaloidal beds show that the sheet is made up of a number of successive flows, one quickly following another. LopEEiA SIMPLEX Nowberry.-^ Professor Newberry has given this name to the plants whose stems appear commonly as simple cylinders about an inch across. They occur abundantly in Springfield, and were filled at one locality by a sand that differs from that which inclosed them by its freedom from mica scales and its pale-green color. INSECTS. MORMOLUCOIDES ARTICULATUS E. HitchcOck. 1858. M. articulatus E. Hitchcock. Ichnology of New England, pp. 7, 8, pi. 7, figs. 3, 4, with letter of Professor Dana. 1862. Palephemera medieva E. Hitchcock. Am. Jour. Sci., 2d series, Vol. XXXIII, p. 452. 1867. M. articulatus S. H. Scudder. Proc. Boston Soc. Nat. Hist., Vol. XI, p. 140; Geol. Mag., Vol. V, p. 218. 1871. M. articulatus A. Packard. Bull. Essex Inst., Vol. Ill, p. 1. 1886. M. articulatus S. H. Scudder. The Oldest Known Insect-Larva, from the Connecticut Eiver Eocks. Mem. Bost. Soc. Nat. Hist., Vol. Ill, p. 431. These remarkable forms were found in considerable numbers in the fine black shale at Turners Falls. A series of slabs containing each a great number of indviduals is preserved in the museum of Amherst College. A full history and description of the species and abundant illustrations are given in the last article cited above. Professor Dana first decided that the form was a neuropterous larva. Mr. Scudder and Professor Packard concluded that it was a coleopterous larva. In the last work Mr. Scudder returns to the first conclusion, that it is probably the larva of a sialidan neuropteron. FISHES. The monograph upon the fossil fishes of the Trias,^ by Dr. Newberry, should be consulted by anyone wishing to become acquainted with what 'Fossil Fishes and Fossil Plants of the Triassic: Men. U. S, Geol. Survey, Vol. XIV, 1888, p. 93, PI. XXV, figs. 1-3. '^ Idem. PALEONTOLOGY. 399 is known conceruiu''' tlie fossil Hshes of this region, and the numerous and accurate phites will enable him to determine the name and character of an\- specimen found. Dr. Newberry says : Fishes seem to be equally abundant in the Connecticut lliver basin. At Dur- ham, Couuecticut, and Turners Falls, Massachusetts, they are particularly numerous and well preserved, while they have also been obtained at Middletown, Sudbury, Chicopee, Amherst, and Hadleys Falls.' In this list Sudbury must be changed to Sunderland, and Hadleys Falls to South Hadley Falls ; and Amherst must be canceled, as only coarse arkose occm-s in Amherst, and no fishes have been found there. At Turners Falls, on the east bank of Fall River, a few rods above the bridge, at the southeast corner of the island, a few feet above the point where the dam abuts, and on the mainland directly north of this spot, in the line of strike at the foot of the bluffs and near the water's edge, many specimens can be obtained by digging in the black shales. At Whitmores Ferry, Sunderland, in the north part of the town, in rocks exposed only at low water, numerous impressions may be found. Good specimens, carefully and skillfully developed, can be purchased of the owners of the mill adjacent. The slabs are left out during the winter and split by the frost, so as to expose the impressions of fishes to the best advantage. Hadleys Falls, mentioned by Newberry, must, I think, be South Hadley Falls Canal, as fishes were found during the digging of this canal, and are now deposited in the museum of Amherst College. The specimens from this locality do not seem to have been examined by Professor New- berry, as he does not cite any species from there. Those in the Amherst Musuem were by oversight not submitted to him. Chicopee Falls has not afforded anything, so far as I know, for many years. The excavations made during the building of the dam and mills may have supplied the specimens which fell into the hands of Mr. Red- field, and furnished the material for the new species which Dr. Newberry has named for this town. There are no specimens from this place in the Amherst collection. I have given below a list of the forms which have been identified in Massachusetts, and a word concerning the history of the more interesting ' Loo. cit., p. 21. 400 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. species. Excepting the rare form found at Chicopee, it will be seen that they are all referred to two ganoid genera, Ischypteras of Sir Phillip Egerton, which is characterized by the great strength of the fin rays, and Catopterus, distinguished by the posterior position of the dorsal fin. Most of the specimens found belong to the two species Iscliyptems tenuiceps and Catopterus gracilis. Ischypterus ovatus. W. C. Eedfield. Sunderland (Eedfield), Turners Falls (New- berry). Ischypterus marshii W. 0. R. Sunderland (Eedfield). Ischypterus micropterus N. Sunderland. Ischypterus tenuiceps Ag., sp. Turners Falls. Sunderland. Figured by E. Hitch- cock. Geol. Mass., 1841, Vol. II, p. 459, PI. XXIX, figs. 1, 2. Ischypterus macropterus W. 0. E. Sunderland. Ischypterus parvus W. 0. R. Sunderland. Figured by Hitchcock, Geol. Mass., 1835, Atlas XIV, fig. 44, and 1841, PI. XXIX, fig. 3. Ischypterus latus J. H. E. Sunderland, Ischypterus elegans. Sunderland. Catopterus gracilis J, H. E. Sunderland, Catopterus parvulus W, C, E. Sunderland, Acentrophorus chieopensis N, Chicopee Falls, ICHNOLOGY. Since the publication of the Ichnology of Massachusetts and its Sup- plement, which President Hitchcock looked upon as closing the most original scientific investigation of his life, but little has been done to advance the knowledge of this the most peculiar contribution of the Con- necticut Valley to geology, except what has been published by Prof C. H. Hitchcock, who has kindly permitted me to print in this place a portion of an article upon the subject, containing his latest views upon the classification of these foi'ms, from the proceedings of the Boston Society of Natural History, Vol. XXIV, 1889, p. 117. The article has been cor- rected by Professor Hitchcock (1892). Ebcent Pbogkess in Ichnology. By C. H. Hitchcock. The study of the Ichnozoa, or the animals that made the tracks, naturally, divides itself into three parts: First, an examination of the ichnites themselves; sec- ond, the restorations of the animals from their bones, and third, comparisons of the PALEONTOLOGY. 401 impressions made by livinj; animals with the Triassic imprints. I will at present speak only of the first. Allow me to present, at the outset, a complete list of the Triassic Ichnozoa, arranged in convenient classes. It will not be needful to state the reasons why cer- tain species of the Ichnology are dropped. The number, after several erasures, haa increased from 150 of the Ichnology to 170.> ICHNOZOA or THK TRIAS. Marsripial. Cunichnoides marsupialoideus E. H. Birds, Pachydaotylous. Brontozoum glganteum C. H. H. approximatum C. H. H. minusculum E. H. divaricatum E. H. tuberatum E. H. exsertum E. H. validum E. H. sillimanium E. H. Amblonyx giganteus (?) E. H. (?) Birds, Lepiodaciylous Argozoum redfieldianum (?) E. H. dispari-digitatum E. H. Amblonyx lyellianus ( ?) E. H. Grallator cursorius E. H. parallelus E. H. tenuis E. H. gracilis C. H. H. cuneatus Barratt. formosus E. H. Leptonyx lateralis E. H. Argozoum pari-digitatum E. H. Dinosaurs. Anomoepus major E. H. isodactylus C. H. H. intermedins E. H. curvatus E. H. minor E. H. cuneatus C. H. H. minimus E. H. gracillimus C. H. H. Gigantitherium caudatum E. H. minus E. H. Hyphepus iieldi E. H. Corvipes lacertoideus E. H. Tarsodactylus expansus C. H. H. caudatus E. H. Apatichnus crassus C. H. H. holyokensis C. H. H. circumagens E. H. bellus E. H. Plesiornis quadrupes E. H. pilulatus E. H. sequalipes E. H. mirabilis E. H. Plesiornis giganteus C. H. H. n. sp. C. H. H. CliimsBriclinus ingens C. H. H. barrattii E. H. Anticheiropus hamatus E. H. pilulatus E. H. Platypterna deaniana E. H. tenuis E. H. delioatula E. H. recta E. H. varica E. H. digitigrada E. H. Ornithopus gallinaceus E. H. gracilior E. H. Tridentipes ingens E. H. elegans E. H. elegantior E. H. insignis E. H. uncus (?) E. H. Trihamus elegans E. H. magnus C. H. H. >A catalogue of the Ichnozoa, as they were known in 1871, was prepared by me for Walling and Gray's Official Atlas of Massachusetts. MON XXIX 26 402 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. Polemarchus gigas E. H. Plectropterna minitans E. H. gracilis E. H. angusta E. H. lineans E. H. Triaenopus leptodactylus E. H. Harpedactylus gracilis E . H. gracilior E. H. crasBus E. H. n. sp. C. H. H. Xiphopeza triples E. H. Toxiclinus insequalis E. H. Orthodaotylua floriferus E. H. iutrorergens E. H, Otozoum moodii E. H. caudatum C. H. H. parvum C. H. H. Batrachoides nidificans E. H. Palamopus clarki E. H. Macropterna vulgaris E. H. divaricans E. H. gracilipes E. H. Ancyropus teteroclitus E. H. Chelonoides incedeus E. H. Helcnra caudata E. H. Beptiles and Amphibia. Orthodactylus linearis E. H. Antipus bifiduB E. H. iiexiloquus E. H. Stenodactylus curvatus E. H. Axaclmichnus dehiscens E. H. Isocampe strata E. H. Typopus abnormis E. H. gracilis E. H. Anislchnus [C. H. H.] deweyanus E. H. gracilis E. H. gracilior E. H. Comptichnus obesus E. H. n. sp. C. H. H. Batrachians. CheJoniana. Cheirotheroides pilulatus E. H. Shepardia palmipes E. H. Lagunculipes latus E. H. Selenichnus falcatus E. H. breviuBculus E. H. Exocampe arcta E. H. ornata E. H. minima E. H. Helcura surgens E. H. anguinea E. H. Amblypus dextratus E. H. Sexapod Arthropoda. Giammepus erismatus E. H. Acantbichnus cursorius E. H. alternans E. H. alatus E. H. anguineus E. H. trilinearis E. H. punctatua E. H. rectilinearis E. H. divaricatus E. H. saltatorlus E. H. Bifurculipes laqueatus E. H. soolopendroideus E. H. Bifurculipes curvatus E. H. elachistotatus E. H. Copeza triremis E. H. propinquata E. H. punctata E. H. cruscularis E. H. Hexapodichnus magnus E. H. horrens E. H. Conopsoides larvalis E. H. ourtus E. H. Harpipes oapillaris E. H. Sagittarius alternans E. H. Harpagopus dubius E. H. Stratipes latus E. H. Hamipes didactylus E. H. Saltator blpedatus E. H. caudatus E. H. Halysichnus laqueatus E. H. tardigradus E. H, Cunicularius retrahens E. H. Inferior Arthropods, including larval forms and worms, Spbferipes larvalis E. H. magnus E. H. Lunula obscura E. H. Ptericbnus centipes E. H. Unisulcus marshi E. H. intermedius E. H. minutus E. H. magnus C. H. H. PALEONTOLOGY. 403 Bisulcns iindulatus E. H. Trisnlons laqueatiis E. H. Cocliloa archimedea E. H. Hopliobnus equus E. H. polodruB E. H. ^nigmichnus multiformis E. H. MoUuaca. Incertce sedis. CoohliohnuB anguineus E. H. two n. sp. Grammiolmus alpha E. H. Ampelichuus sulcatus E. H. Climacodichnus corrugatus E. H. Of lower arthropods and worms there may be half a dozen new species and two new genera. Summary : Marsupial 1 Pachydactylous birds 17 Leptodactylous birds 18 Dinosaurs 28 Reptiles and amphibia 27 Batrachians 16 Chelonians 6 Hexapod arthropods 24 Lower arthropods and worms 16 MoUusca 6 IncertsB sedis 6 Total 165 The class of Birds is still retained for convenience, although the bones found in the west seem to point to reptiles as most probably the animals thus designated. It is still a fact that such special reptilian characteristics as would be exhibited in walk- ing are absent in the genera Brontozoum and Grallator, while those creatures called Dinosaurs are thus referred, either because of the marks of front feet, heels to the hind feet, or of tails. The bird group is also characterized by long legs, while most of the Dinosaurs had short legs, as indicated by their numerous steps. I do not change the reference of a group to Chelonians, though it is not satisfactory. The Arthropoda are most likely to be referred to the lower classes; yet the presence of only 6 feet in the impressions leads us to speak of them as Hexapods. They may not be true insects, but larval forms, requiring further investigation before satisfactory references can be made out. Further statement of the reasons for referring various imprints to their lowly owners would involve a discussion of the third part of the subject, which can not be undertaken now.^ It will be proper to state a few facts about museums and localities before describ- ing the new species. THE AMHEBST MUSEUM. A few slabs have been added since 1865, and the arrangement of the rooms has not been changed since the printing of the catalogue. One slab shows a Brontozoum with two toes on one foot and three upon the other, as if the owner had lost a toe by ' Of modern authors, A. Gt. Nathorst has treated of the invertebrate tracks most fully in his M^moire sur quelques traces d'animaux sans vert^br^, etc., et de leur port^e pal^ontologique, 1880. His bibliography notices several American authors, but he has evidently not seen the Ichnology of Massachusetts. 404 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. fighting or by accident. After the discovery of Apaticlmus holyohensis, I was able to point out several illustrations of the new species at Amherst, which had been over- looked in the preparation of the Ichnology and Supplement. museum: at south hadlbt. The Mount Holyoke Seminary and College has taken great interest in Ichnology and possesses an admirable collection. Among the more important ones are the type specimens of Apatichnus holyohensis and of six or eight new species from Wethers- field Cove, besides Anotncepus cuneatus and A. isodaotylus from the Dickinson quarry at South Hadley. The data for improved descriptions of Brontozoum divaricatum and Plectropterna elegans are present, as well as long rows of Otozoum moodii, Bronto- zoum giganteum, and B. approximatum. The slabs occupy a large room in the base- ment of the Lyman WiUiston Hall, while smaller specimens have been placed in the adjoining apartment. The institution possesses several slabs from the Dickinson quarry, about a mile north from the buildings. These are composed of a hard sandstone which preserves the impressions and casts with unusual distinctness. The marks of the heels of the hind feet, the front feet, and the tails of Anomcepus are very plentiful. About sixty species of Ichuozoa are placed upon these tables, and a careful description of every slab exists in the manuscript form. This catalogue is like the one prepared by myself in 1865 for the Amherst collection and printed in the Supplement. Of the above species, tlie JEnigmichnus multiformis is certainly the marking of a diifting tree whose roots or branches trailed in lines strictly parallel along the bottom, these lines often changing to a row of dots or a moniliform line from the rising and sinking of the plant with the waves. These parallel lines cover a space several feet wide and could hardly be formed by any animal. Further, Professor Hitchcock has omitted a sec- tion of fish-tracks given in the Ichnology with a genus Ptilichnus, or " fin- track," thought by President Hitchcock to be the marks of the fins of fishes. The markings are very uncertain and may well be withdrawn from the list. I have, however, found in the collection many fine, thin slabs with a curious marking upon them, which bear the name " Aroid Plants, Sunderland" in President Hitchcock's writing, but about which he seems to have published nothing. The resemblance of these markings to the markings which would be made by the ventral fin spines of a fish drifting slowly backward, and by a few slight rapid strokes stemming the current at regular intervals, is certainly sti'iking. PALEONTOLOGY. 405 REPTILES. Megadactylus Owen. Megadact\lus polyzelus E. Hitchcock, jr. 1858. "Bones of a reptile." Jeffries Wyman. Ichnology of New England, p. 186. 1863. Megadactylus polyzelus. E. Hitchcock, jr. Supplement to Ichnology of New England, p. 39. 1871. Megadactylus polyzelus. E. D. Cope. Synopsis of the extinct Eeptilia and Aves of the United States: Trans. Am. Phil. Soc, Vol. IV, p. 122a, PI. XIII. 1884. Amphisaurus [Megadactylus). O. C. Marsh. Am. Jour. Sci. Sup. XXVII, p. 338. 1889. Anchisaurns. O. 0. Marsh. Am. Jour. Sci. Sap. XXXVII, p. 331. This rare and remarkable fossil has had a peculiar history. The bones ■were thrown out by a bjast in excavating- a well for the casting of a big gun at the water shops of the United States Armory, in the south part of Spring- field, and only a part of the skeleton was preserved and presented to Presi- dent Hitchcock. These bones were first studied by Jeffries Wyman, and determined by him to be those of a reptile. His letter contains many acute observations. He notes the hollowness of the bones, a peculiarity suggest- ing birds and pterodactyls, but decides against the reference of the bones to either of these. The unequal length of the toes suggests a jumping animal. The bones were then carried to London by Dr. Edward Hitchcock and submitted to Prof. Richard Owen, who determined them to be those of a Saurian reptile, but added otherwise nothing to the diagnosis of Wyman.' His one sentence concerning the bones is interesting. They belong to "a Saurian reptile with an unusually thin wall of bone in the limb bones, which, however, might have been occupied by unossified cartilage, as in the young crocodile and turtle; but if they were filled with oil or light maiTow, it would point to a course of development toward pterodactyls or birds. The phi-ase is purely hypothetical, and I mean to express no more than a degree of resemblance, supposing marrow and not gristle to have filled the large cavities." Later, Dr. Hitchcock worked out the bones with a graver and named the animal in the article quoted. The specimen was then carefully studied, figured, and described by 406 GEOLOGY OP OLD HAMPSHIEE COUifTY, MASS. Cope, who refeiTed it to the Triassic Dinosauria and called special attention to the very peculiar ischium. In 1876 I earned the bones to New Haven, where they were studied by Professors Huxley and Marsh, and casts of them were taken by the latter. Professor Huxley was inclined to think them, identical T^ith one or other of two genera of reptiles found by Stutchbury in the Trias near Bristol, England, and preserved in the museum of that city — Paleosaurus and Thecodontosaurus — but because of some loss or change of labels it was not possible to tell which of the bones preserved in the museum should be called by the first of these names and which by the second. At a later time Professor Marsh gave a new name to the genus, as indicated in the synonymy above, and still later, finding that this name had been preoccu- pied, he gave the fossil another name. The bones include a nearly perfect foot, the ischium, femur, caudal ver- tebrae, and many imperfect fragments. A few very imperfect fragments of the bones of a similar species were found earlier, and are preserved in the museum, but without locality; and I have found many imperfect fragments of bone in the indurated sandstone of the contact zone of the easternmost volcanic core in Belchertown. This induration has prevented the percola- tion of water, which has doubtless carried away many bones formerly embedded in these coarse sandstones. CHAPTER XIII. TRIASSIC ERUPTIVE ROCKS. HISTORICAL. As early as 1815 President Hitchcock described the "Basaltick Col- umns" of Titan's Piazza in the first volume of the North American Review.' He gave the "greenstone" only a word in the Greology of Deerfield,^ not distinguishing it from the hornblende-schist of West Northfield. It is described at some length in the. Geology of the Connecticut^ as "secondary greenstone," without reference to its mineralogical constitution. He notes that it is more amygdaloidal in its upper portion, describes the contact of the upper sandstone on the trap in Sunderland and Deerfield, and interprets the fault at the mouth of Fall River, described below (p. 437), as a repe- tition of the trap. In his earlier report upon the Geology of Massachusetts* he gives a very full account of the "greenstone," touching upon its lithological pecul- iarities, its distribution, mineral contents, and origin, an account which has lost httle of its value, and which, because of its great length, I shall only briefly summarize here, as the main points are cited beyond. He now considers the "greenstone" to be made up of feldspar and hornblende, and remarks that he has not met with a genuine and distinct dike of trap in the sandstone. In the later edition of the above work (1835), and in the Final Report,^ the same account is reprinted almost verbatim, the only additional informa- tion given relating to the small dikes in the gneiss on the east of the sand- stones. An inspection of the maps accompanying the above reports shows clearly that the trap was laid down most accurately on the map of 1823, ' Page 337. !2 Am. Jour. Soi., Ist series, Vol. I, 1819, p. 105. 3 Am. Jour. Sci., 1st series, Vol. VI, 1823, p. 44. "Geol. of Mass., Amherst, 1833, p. 404. ^Geol. of Mass., Final Kept., Amherst, 1841, p. 640. 407 408 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. especially as regards the Deerfield bed, and that with the poor maps and the small scale used the delineation became more and more inaccurate. Upon the map of 1844 the Mount Tom range is represented much more accurately, a posterior range is given in West Springfield, and three long patches of "trap tufa and tufaceous conglomerate" are laid down. Upon the map in the Ichnology, 1857, a further advance is made by showing that the Holyoke range consists of two bands of trap with a narrow intervening band of sandstone. The section through Norwottock on the border of the above map is incorrectly colored to indicate three bands of trap ; the south- em band should receive the color of the "trap tufa." Furthermore, in all the copies of the work I have seen the trap tufa has the same color as the crystalline rocks upon the borders of the map, while in the legend a deeper shade of the color is assigned to it. This error has perpetuated itself in a curious way. Upon the small geological map attached to the map of Hampshire County of H. F. Walling (1858) a pink band of crystalline rocks is made to run across from Belchertown to the river south of the Holyoke range. The map is said to be "by Edward Hitchcock." One may infer, I think, that he had very little to do with it. In the small map appended to Reminiscences of Amherst College (1863), President Hitchcock gave his final results regarding the rocks in question. He now makes the Holyoke range a single broad area of trap extending east to overhang, with undiminished width, the northern of the Belchertown ponds, and lays down two great areas in Pelham, the one made out by coloring as trap the whole area between the two occurrences detailed below, and the other based upon the area of great bowlders east of Amethyst Brook. The geological map of Prof. C. H. Hitchcock in Walling's Atlas of Massachusetts (1871) gives a posterior bed in the eastern half of the Holyoke range and the northern part of the Mount Tom range, presenting thus the best results of his father's work. In 1875 Prof. E. S. Dana presented to the American Association the preliminary results of studies upon the Triassic diabases, undertaken by himself and Mr. Gr. W. Hawes, and in the same year Mr. Hawes printed a series of analyses of these rocks, including one from Mount Holyoke. Although in the main devoted to the Connecticut rocks, these studies reached results applicable to this area, viz: the greater freshness of the TEIASSIO ERUPTIVE ROCKS. 409 diubuso from the o-iieiss, its composition of pyroxene find the triclinic feld- spars labradorite and anorthite, and the rare presence of oHviue. By companson oi' the altered with the unaltered varieties, it was seen that this alteration has not been attended by further oxidation of the iron, and therefore could not have been accomplished by any surface action, since the oxidation of protoxide of iron is one of the chief causes of surface alteration, while in this case one mineral containing- protoxide has been changed into another protoxide mineral. It would therefore seem, certain that the alteration took place at the time of ejection, as had been urged by Professor Daua.^ Later, Mr. Hawes^ made a separation (by Thoulet's solution) and analysis of the feldspars in diabase from New Jersey, determining them to be labradorite and andesite. He further calculates, on the basis of analyses in his preceding paper, the mineral composition of the "West Rock dike" near New Haven, finding it to contain the feldspars anorthite, albite, and orthoclase, with augite, titanic iron, magnetite, and apatite. Professor Dana,^ in a very pungent critique of this paper, objects that the anorthite came from a later transverse (east-west) dike in the West Rock dike, and so can not be combined with the gross analysis of the latter, as was done by Mr. Hawes. In 1882 the author published a paper, mainly mineralogical, on the Deei-field dike,* in which the contact metamorphism on the sandstone below and the amygdaloidal character of the trap sheet in its upper portion and the unaltered condition of the sandstone above, which is molded into all the interstices of the trap, are adduced in support of the view that this trap body is a contemporaneous sheet and not a true dike. The presence of a beautiful fault at the mouth of Fall River was noted. The proofs of its existence would seem to have been given with too much brevity, as they failed to convince the author of the paper to be mentioned next, and they will be given more fully in the sequel. In the following year appeared a very valuable article, by Prof. W. M. Davis,^ on the " Triassic trap rocks of the eastern United States," spe- ' Am. Jour. Sci., 3d series, Vol. VI, p. 104. = Proc. U. S.Nat.Mu8., 1881, p. 129. 3 Am. Jour. Sci., 3d series. Vol. XXII, p. 230. •■Am. Jour. Sci., 3 series, Vol. XXIV, 1882, p. 195. 6 Bull. Comp. Zool. Harvard Coll., Vol. VII, p. 251. 410 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. ' cially valuable because of the reproduction of all the sections and dia- grams explaining earlier views of the structure of the trap ridges and of the full discussion of previous theories. Of the abundant original observa- tions only two groups relate to the Massachusetts area — one to the Turners Falls, the other to the Mount Tom region. He considers part of the trap masses to be contemporaneous beds and part to be true dikes, but adduces only cases under the first category from Massachusetts. The Deerfield bed he makes to be three beds, echeloned one posterior to the other. I have found it to be a single bed, faulted several times at the north end, as will be detailed below. In 1892 the author published a preliminary paper on the quarry lor road material at Greenfield and described the under-rolling of the trap and the formation of breccia-like beds.^ In the summer of 1896 the author presented a paper before the Greolog- ical Society of America entitled, " Diabase pitchstone and mud inclosures of the Triassic trap of New England."^ The paper explains the manner in which water and mud, frothing up into the trap from the sea bottom over which it was flowing, produced pitchstone and shattered the mixture and recemented it with an aqueous deposit of albite and bisilicates. There was also described the sweeping of fine mud out over the surfacQ of the Holyoke sheet by convection cuiTents and its under-rolling to form the base of the bed. THE THREE EPOCHS OF ERUPTIVE ACTIVITY; GETSTERAXi ACCOUDST. 1. The rapid transgression of the Triassic waters over the area had spread a great thickness of coarse granitic debris when two fissures allowed the passage of great volumes of basic lava to form the Deerfield and Holyoke diabase sheets. Sedimentation went on undistm-bed. Generally the first layers spread on the surface of the sheets were the same or nearly the same as those on which the trap rests. In the Holyoke bed one can see in small degree the influence of the shallowing of the waters, and the beds above are of finer grain. The fissure for the Deerfield bed must have been beneath the present outcrop or the lava must have come from the dikes in the gneiss along the eastern border of the basin. The fissure of the 'Am. Jour. Sci., 3d series, Vol. XLVI, p. 146. 3 Bull. Geol. Soc. America, Vol. VIII, 1897, pp. 59-96. THREE EPOCUS OF TltlASSIG ERUPTION. 41 1 Holyoke bed was probably a mile soiitli and east of the present outcrop, along- the line of later trap intrusions. The beds slightly baked the sand- stones below and are amygdaloidal and ropy-surfaced above. They often took up great quantities of the rock over which they flowed, and the fact and direction of flow are shown by the marked difference between these fragments and the subjacent rock. Much sedimentary material is in places kneaded into the surface layers of the trap — either before it became solid or in a breccia layer — and is then carried underneath by the under-rolling of the solid and yet plastic front of the advancing sheet. 2. A great core, representing a second epoch of volcanic activity, now forms Little Mountain, which lies between Mount Tom and the river below Smiths Ferry, and from it flowed a thin but double sheet south beyond the limit of the State and north at least to the river, a half mile south of the Holyoke gap. 3. Immediately following this came an explosive outburst which spread tuff south to Holyoke and east across the whole basin to Belchertown. East of the river this rests on arkose; west, on the upper trap sheet. Its masses are largest (3 feet in length) at Smiths Ferry and decrease slowly east and south. The results of the last period of volcanic activity appear in a line of crater tln-oats and short intrusive dikes extending from the river to the east edge of the basin, parallel to and a mile south of the Holyoke range. Two are of very large size and one is a diabase full of quartz and feldspar grains. They make a small angle with the tuff sheet, so that some lie south and some north of it and some penetrate it in whole or part. DIABASE DIKES AISTD STOCKS IN THE GNEISS EAST OP THE TRIAS. A series of small dikes appear in the gneiss east of and a short distance from the sandstones. I do not find reason to consider them continuous over so long a distance north and south as they would appear to be from Perci- val's excellent map of their distribution in Connecticut, nor does any trace of the similar western line of dikes marked by him extend northward into Massachusetts. They are typical diabases, much fresher and of finer grain than the large masses in the sandstones, but scarcely offering any appreciable distinction from the finer grades of the latter. On their borders, however, and in small 412 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. dikes from a half inch to an inch across, which are at times abundant in the gneiss, they reach a degree of fineness never seen in the central dikes, and are in part or wholly made up of glass, and contain olivine, which allies them to the newer outflows in the main valley. 1. The most northerly of these dikes cuts gneissoid rocks on the east bank of the Coimecticut, a few rods below the mouth of Millers River. The dike is about 3 feet wide and runs south from the water's edge and disap- pears in a short distance beneath the terrace sands. It is a compact, very fresh, dark-gray rock, with few porphyritic feldspars 1 ™™ long and extinc- tion 21° on either side of the twimiing suture, the smaller feldspar 0.12™™, the light-yellow augite peculiarly granular and without crystalhne outline. Magnetite is veiy abundant. This occurrence is cited by President Hitch- cock in his first report,^ and incorrectly assigned to Erving in the Final Report.^ 2. The next dike is intruded along the bedding of the gneiss, in the vertical wall which forms the north bank of Millers River, east of the bi'idge in the village of Millers Falls. As the gneiss has a low dip to the west, the dike, which is about 7 feet wide, reaches the water's edge just west of the bridge, where its crossing the stream gave rise to the falls from which the village gets its name. The rock was not distinguishable in thin sections from that of the preceding occurrence. 3. The next outcrop was a knob of remarkably fresh ice-worn rock exposed in the cutting made in 1881 in the relocation of the raih-oad tracks a few rods south of the Millers Falls station. The diabase was exposed in a rounded ice-worn boss, 10 or 12 feet across, without contacts. A few yards to the east, and 2 yards lower, gneiss was exposed, in which rock the diabase was doubtless intruded. 4. President Hitchcock notes greenstone in Montague, on the west border of gneiss, 2 miles northeast of the meetinghouse. It separates in plates directed east and west and standing vertical.^ This locality is beside the railroad, a mile south of Millers Falls, south of J. Hannegan's house. A ridge 325 feet long, 82 feet wide, and 20-30 feet high nms N. 35° E., surroimded by the terrace sands. The last three outcrops may form parts of one long dike. 1 Geol. of Maes., 1835, p. 417. ^Ibid., 1841, p. 648. 8 Geol. Mass., Final Report, 1841, p. 648. DIABASE DIKES AND STOCKS IN THE GNEISS. 413 The above dikes near Millers Falls are of ideal freshness; very rarely one sees in a single large feldspar a slight central clouding, like a delicate ileck of cotton. They are rather light-gray, extremely tough, and yet brittle as glass. The constituents are of exactly the average dimensions given in the general description of the diabase, page 438. The augite is yellow to amethystine, dichroic, and, although perfectly fresh, it appears, from the strong cleavage and abundant inclusions, only translucent, and looks in the slide as if a quantity of pulverized material had been spread over the network of feldspars. This enables one to distinguish it from other occurrences. 5. Across Montague and Leverett no other outcrops occur. In Pelham, on the south side of the Shutesbmy road, west of where it crosses Amethyst Brook, a great outcrop of the same fine-grained diabase occurs in the actin- olitic quartzite. It is a squarish mass about 82 by 130 feet, its longest diameter north and south. Following the stream up from this point to where a brook comes in from the north, one finds a great number of large bowlders of diabase in its bed, some of large size. There is probably a considerable bed in the pasture a few rods east of the junction of the brooks. Further, the fine amethysts which occur as rolled specimens in the bed of the brook probably indicate the presence of diabase here. A mile northeast also, in the deep brook gorge north of Ward's quarry, occur a great number of very large diabase bowlders, as well as much farther east in the eastern portion of the town, along the roads that run down from Pelham Center to the Swift River Valley. 6. If the line connecting the above outcrops in Pelham be prolonged N. 40° E. into Shutesbury, it strikes a great outcrop of diabase at the point where it crosses the road going north from Pelham Center, ojaposite the house of W. Thrasher. It is exposed with a length of 25 rods and a width of 75 feet, and runs N. 40° E. It is a fine to very fine, very fresh diabase of the common structure, the finest-grained portion showing a globulitic groundmass as inclusion in the feldspars, and small olivines. 7. If the line be prolonged N. 40° E. across to the river road, another outcrop of diabase occurs on the hillside northwest of the house of S. H. Stowell. An inspection of the map will show that all the above series of outcrops occur along a northeast fault which has opened the entrance to the upland basin of Pelham and caused the sharp southern slope of Mount Hygeia. 414 GEOLOGY OF OLD HAMPSHIEB COTJNTT, MASS. 8. Again, where the road east from Pelham post-office, halfway down the hill, turns from south to southeast, a large outcrop of trap occurs on the north side of the road. 9. Groing about 115 rods on the first western road running south from the West Village of Pelham, and turning east into the woods, one finds an east-west vertical dike, at one place nearly 6 feet thick, but running west with a thickness of only 1 foot, which sends off many small branches into the gneiss, one of which furnished the material for the study on page 416. 10. On Coys Hill, in the southeast part of Ware, north of the point wnere the road crosses the town line, a vertical dike of diabase occurs in the high bluff northeast of the road across the ravine. It is horizontally bedded, 50 feet wide, and can be followed a half mile south, first with strike N. 40° E., then swinging round to N. 30° E., when it crosses the town line into Worcester County. It is fine-grained and is beautifully exposed, with its attendant swarm of small dikes in the adjacent gneiss. It is now quar- ried for road material just east of the station, where it is 5 rods wide. 11. A great accumulation of bowlders of the aphanitic diabase in Belchertown, north of the schoolhouse, near E. Willis's, another near the center of Wales, and another in the northeast of Belchertown indicate in each case the proximity of an area of the rock covered by drift deposits. 12. About 650 feet east of the house of J. Bardwell, near the west line of Belchertown, occurs an isolated outcrop of trap, forming a hill of great, broken masses of the rock. It is about 33 by 100 feet, and gneiss occurs in the near vicinity on every side, though the immediate contact could not be observed. The rock is the dark bluish-gray aphanitic variety common in the gneiss. 13. Just south of Flint's quarry, in Monson, a heavy dike of trap is cut through by the quarry railroad. It can be followed but a short distance to the north, when it is cut off by a fault and offset to the east, and its con- tinuation, with the evidence of the faulting, can be found in the south bluff of the ridge next east. From this point it can be followed northeast more than half a mile, till it disappears beneath the sands in the Monson Valley. It is about 410 feet wide. The small dike next described is apparently an offshoot from it, and the great number of trap bowlders found over the high ground in the east part of Monson are clearly derived from it, and their distribution makes it plain that the dike extends much farther northeast and southwest than can be seen. It is now quamed for road material. DIABASE DIKES AND STOCKS IN THE GNEISS. 415 14. A vertical dike of black, fine-grained, horizontally jointed diabase runs east and west tlirougli the great quarry at Monson. It is nowhere more than 16 inches wide, and as it goes upward it has a curious warp to the south. It sends oflf many small dikes, which are specially discussed in the next section below (p. 416). 16. Farther south in Monson, in the crest of the bluff west of S. Mac- intosh's house, is a dike of similar rock, 50 feet wide, running N. 65° E., which can be traced for some distance in the face of the cliff, cutting the amphibolite. ■ 16. Still farther south, on the east slope of Peaked Mountain, west of the house of J. Bliss, jr., occurs a dike of about equal size and of similar character. These dikes in Monson were already traced by Percival. Trap bowlders are very abundant along the western slope of the high ground east of the central valley of Monson, from one end of the town to the other. 17. Another plug occurs just over the State line in Stafford. It runs N. 10° E., is 60 rods long and 200 feet wide. It is high up on the east slope of the hill which lies across the brook west of where the Hampden- Stafford road crosses the State line. The shdes show a trace of decompo- sition. The feldspars of first generation have broad bands with wavy extinction from strain; the second are very complex twins. 18. A mile S. 10° W. of this, where the road from the State-line Pond to Somers rises to the top of a high hill, another dike is exposed just south of the road. The contact, in granite, is exposed on the west. The strike is N. 40° E. It is 56 feet wide, 200 feet long, has steep slope on the north and a swamp on the south. The sections show unusually fresh and sharply and regularly outlined plagioclase of only one generation. 19. A third stock of trap occurs a half mile S. 10° W. of this, which crosses the next east-west road. It is 45 rods long from north to south, 25 rods from east to west. The gneiss is continuously exposed around its east, north, and west sides. It is a compact, light-gray trap. In these sections the large plagioclase crystals of first consolidation have the central portion out nearly to the border changed into a cottony mass of plumose, micalike, elongate, ragged scales, while the clear border shows at one end a marked wavy extinction and the other end extinguishes sharply at 25° on either side of the twinning suture. This is an unusual change to some micaceous or zeolitic mineral, instead of to kaolin. 416 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. 20. In the northeastern corner of the Palmer quadrangle a dike of the same fine-grained, dark-gray diabase was discovered by my assistant, Mr. C. S. Merrick. It is nearly a mile west of the point where the Boston and Albany Railroad leaves the quadi-angle, and appears in the crest of the hill at the 900-foot contour.-* It is about 100 feet wide and strikes N. 20° E., and is plainly a part of the dike No. 10, which can be traced north across Ware and New Braintree. A MICROSCOPIC DIABASE DIKE FROM PELHAM, AND OLIVINE AND GLASS- BEARING DIKES FROM MONSON. The two great diabase masses of the Triassic in Massachusetts, the Deerfield and the Holyoke dikes, are amygdaloidal at surface and aphanitic at base, but everywhere normally crystalline, and everywhere, even when seeming quite fresh, much decomposed. The series of smaller dikes of the same rock, when run in the gneiss, parallel to and a few miles distant from the eastern border of the sandstone, which were traced across Connecticut and Massachusetts by Percival and Hitchcock, are in texture exactly similar to the former, showing a typical diabase texture, but always very much fresher. They often send off a great number of apophyses, which sink to very small dimensions and run out in all directions and to considerable distances through the gneiss, which, ordinarily very friable, is here so indurated that thin flakes can be broken off and slides prepared containing one or more of these minute dikes. An interesting slide of this character from Pelham contains a dike 0.9™™ wide and 20™™ long. It is a tachylyte, shading from dark gray at one side to jet black at the other, and under the microscope is a colorless glass loaded with a fine dust, apparently magnetite. The shading into black is due to the occurrence of this material in much greater quantity at one side of the dike, as if it had been formed horizontally and the magnetite had sunk to the bottom. The rest of the surface has a mottled look, like a miniature representation of a tiger's skin. This comes from the fact that minute angular fragments of quartz and feldspar, which are scattered through the mass, are surrounded by a halo of the same black dust, outside which a broad ring of the glass is comparatively clear. This gives the whole an apparent spherulitic structure, and this structure is really present and the glass is in a state of tension around the foreign grains, as 'Percival, Geol. Conn., map. DIABASE DIKES AND STOCKS IN THE GNEISS. 417 is seen bv the tact that the clear rings j)ohirize t'eebl)- and show traces of a black cross. On the upper side, i. e., opposite the black border, tlie lic^uid rock forced its way in several places between the grains of the bounding rock. In one place it flowed in with a width of 0.5™™, showing a delicate fluidal structure, the lines of black dust being drawn into a series of regular par- abolas, exactly as in a diagi'am of the surface flow of a river around a curve. Another, narrower, runs far into the gneiss and passes lengthwise of a large biotite crystal in a gliding plane, with a width of 0.02™™. The contact effects of the small dike on the gneiss are also interesting. Not only is the former filled with minute fragments of the inclosing rock, as already noted, but in places along the side is finely crushed and dis- turbed, and cemented again by eruptive material. Crystals of triclinic feldspar have their laminae interrupted and echeloned by a series of fine faults, and in the immediate neighborhood of the dikes they were so influ- enced by heat that the laminae, instead of being as usual (and as they are here farther away) perfectly straight and sharply defined in polarized light, become wavy and bend over into the direction of flow of the lava, and the bands of color pass gradually into each other. In other cases, in a feldspar apparently fresh, on approaching extinc- tion a band of black passes in from the border to the center and disappears. The large biotite, through which the narrow vein passed, seemed entirely fresh, but in polarized light it was seen to be markedly affected, apparently by compression, so that it broke up into patches of color, arranged along the sides of the intruding vein. Smaller crystals of biotite were twisted, so as to show a brilliant aggregate polarization in long inter- woven lines. The large quartz grains, usually entirely uniform, were broken up into irregular patches of brilHant color, and showed marked undulatory extinction. Specially fine cabinet specimens of the small dikes mentioned above can be at times obtained from the Monson quarry — hand specimens of the light-gray gneiss, with three or four dikes narrower than one's finger cross- ing them, and at times bending round so sharply as to inclose a thin wedge of the gneiss, thinner even than the small dikes themselves. From the aphanitic border of the largest dike there, which is only MON XXIX 27 418 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. about a foot wide, I cat sections which showed, under the microscope, many porphyritic ohvine crystals sharply outlined, some nearly fresh, but most well advanced in the change to serpentine. Some were penetrated by thick branching lobes of a brown glass, which in one case took up more than half the surface of the section of the crystal and was accompanied by two separate globules of the same glass with motionless bubbles. This is the first certain occurrence of olivine in the traps of the Connecticut Valley in Massachusetts, and this, with that mentioned on page 411, the first occur- rences of a glassy modification of the rock. The position of the olivines and their large size suggest that they may have been formed at great depths and floated up to their present position. Another of the minute dikes, 2"°" wide, in the gneiss from Monson was cut. It had for part of its boundary a border of crushed gneiss, the triclinic feldspar showing undulatory extinction, and the dike sent off into this a veinlet 0.1"™ wide. It was of finely granular, devitrified tachylyte, with a lighter border one-third millimeter wide. The feldspars in it were from one-third to one thirty- eighth millimeter in length. The well-shaped oli- vines allowed measurement of (021) A (021) = 98° (calculated 99° 06'). THE BEDDED OR CONTEMPORANEOUS ERUPTIVES. THE DEERFIELD SHEET. This, the most northern occurrence of eruptive rock in the Trias, begins near the northeastern border of the latter, back of C. M. Conant's house, in the west edge of the village of Grill, and extends west by south past the house of J. Blake, where it is slightly faulted and where it has a thickness of about 40 feet, which it maintains for a long distance. It is compact at base and slightly porous at surface, and has low southeast dip with the conglomerate in which it is intercalated. At its crossing of the Gill-Tumers Falls road it is again slightly faulted, and the bed was traced only to this fault in my previous study of it.^ It turns here and runs down to the mouth of Fall River, where it is again faulted. It is moved about 165 feet to the west and an opening made, through which the Fall River reaches the Connecticut. From this point it runs down the west side of 1 The Deerfield dike and its minerals: Am. Jour. Sci., 3d series, Vol. XXIV, 1882, p. 195. THE DEEIIFIELD SHEET. 419 the Coniiocticut, tlirouy-h Greeiilield and Deei-field, and, turning- eastward, crosses tlic river and ends in Mount Tob}^ It is at first included in the Long-niea(h)w sandstones, and continues south in them until, at its south eud, it runs otl' into the conglomerate of Mount Toby. It has thus the characteristic elongated U shape which appears on a scale so much larger in the Holyoke range. It is worthy of note that the high western border of the valley, which I shall elsewhere try to prove to have been caused by faulting, corresponds in direction with both these sheets, being set back in Greenfield and Northampton so as in each case to present a reentrant angle to the northwest corner of the trap ranges, with sides parallel to the corresponding portions of the ranges. The 'bed is about 21 miles long, and where the Deei-field River breaks through it it is about 100 feet thick; at Fall River, 165 feet. Where it is cut by the artesian well of the Montague Paper Company (see p. 380) it was penetrated 110 feet, which, with the dip of 40°, would give a thick- ness of 84 feet. THE ALTERATION OF THE DIABASE BY HEATED WATERS TO A PITCHSTONE- BRECCIA AND A DIOPSIDE^PLAGIOCLASE ROCK. Going southward from the bridge over Fall River, one finds in the I'oadside, just before coming to the mouth of the stream, a contact of the diabase upon the sandstone below — a granitic sandstone, coarse to medium in grain, which is baked for an inch into a black horustone and changed for a foot into a strong quartzite. The trap above is little aifected. Opposite Mrs. G, P. Heyward's, in Greenfield, and underneath the lookout tower which stands on the crest of the trap ridge, a crushing machine has been set up to supply the city with road material. For a long distance the vertical wall has been cleared and a most interesting contact is exposed. (See PL VIII.) Climbing up from the sand flats, over 60 feet of fucoidal sandstones with strike N. 10° E. dip 40° E., one finds, at the base of the great trap sheet and resting on the sandstone below, a layer 60 feet thick, made up of rounded and angular blocks of trap, of all sizes up to 3 feet thick, the whole mass penetrated by veins of fine red and black sandstone, often 6 inches 420 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. thick, which branch upward for about 7 feet from the main mass of the sandstone and are full of small steam holes at the top. A fine-grained and finely porous, reddish trap is continued downward from the compact trap above in all the interstices between the blocks, cementing them together in the same way that the sand below cements the larger blocks, and the two cements meet along a horizontal line. (See PI. VIII. The person seen in the plate stands on the sandstone and touches this line with his finger.) As the great mass of lava flowed over the bottom of the bay, its con- gealed and much fissm-ed crust at the front of the flow, like an unrolling carpet, gradually passed beneath the advancing mass, and the mud rose up into all the fissures in the crust, while the heat baked it into a porous rock and the still liquid lava within oozed into the cracks above to meet the mud. The above partial description of this most interesting locality was made at a time when the quarry had exposed only a portion of the surface to study.^ A more careful examination of the place brought out these facts: The basal portion of the bed is made up of angular blocks of trap, and these blocks are often interlocked and a common structure passes from block to block, showing that it is the portion of a bed of trap in place and not a tuff or agglomerate of transported blocks. The blocks are of the common, rather coarse-grained trap of the sheet, but are distantly and coarsely vesicular, some of the spherical cavities being an inch across; and what is most striking, many of the blocks have rows of these cavities around their borders in whole or in part, and these cavities are tubular at times and closely set at right angles to the fissure which separates the block from its neighbor. At times two adjacent blocks have a similar arrangement of tubular cavities on either side of the crack. The arrangement of these tubes at the surface of the blocks shows that the slow expansion of the steam was effective after the mass had cracked into great blocks. Perhaps the increased heat from its under-rolling and penetration by the liquid lava may have been effective here. Moreover, some of the blocks sm-rounded by the finer trap are quite spherical, as if they had been partly remelted after being envel- oped in this newer trap. Again, it is a very partial description of the upper portion of the wall to say that a finer trap has oozed down to meet the iipcom- ing red sand and cemented the blocks of trap. There is a well-defined line ' See Am. Jour. Sci., 3d series, Vol. XLIII, 1892, p. 146. THE DEEKFIKLI) SHEET. 421 !il)(>ut 7 teot above the saiulstoue alon<^' vvlucli the veins of red sand blend with a iinc'-<;'rainod, reddish material quite unlike the coarse blocks of trap, and this reddish material cements the blocks of earlier trap together for a few feet hig-lier, and higher up the blocks grow more distant and smaller and disappear in the mass of the newer material, which is cracked into small fragments, so that the whole closely resembles a tuff, but is not a tuff, if the idea of trans- portation of fragmeutal igneous material by air or water be essential to the definition of a tuff. It is a breccia of sand, trap fragments, and glass, pro- duced by explosions of the water introduced with the mud. In places it loses the red color and becomes greenish. On examining the whole face of the cliff, it is seen that this tuff-like condition continues up half the height of the bed, and its upper boundary continues north and south for a long dis- tance. This is visible in the plate. A careful examination of the zone of contact of the sandstone veins and the newer trap shows the latter to be compact or finely porous, as contrasted with the blocks of trap, which are very coarse amygdaloidal. The newer trap or glass-breccia is reddish, because it is an intimate mixture of trap and red sand, and for 20 feet up, as far as one can climb at the quarry, the mixture of the filaments of sand and trap are most inti- mate, and on a polished surface it is seen that the delicate anastomosing films of the trap penetrating the sand could have reached their present posi- tion and condition only in a liquid state, while the thin layers of sand are as intimately mixed in the trap. Under the microscope (see fig. 24, A, B, p. 422) the thicker portions of the sand filaments (left side of figures) in specimens taken about 20 feet from the base of the sheet are of the same texture exactly as in the broad intruded masses of sandstone below, but are blackened around their border by the caustic action of the adjacent lava, and as they grow thinner they become black across their entire width. This seems to be caused by the coating of the sand grains with hematite derived from the iron of the red mud and recrystallized by the heated waters. These borders bristle out- wardly also with beautiful hexagonal plates of blood-red hematite, and the same plates are found also in the sand and in the surrounding rock. The second constituent of the rock is the trap, here in somewhat abnormal development. It is in small fragments and minute filaments, penetrating the sand in every way. It contains the large, angular, and 422 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. well-formed plagioclase twins of earliest consolidation, which are, as usual, dusted with impurities, or decomposition products, and a few fine rods of later growth, and these constituents are entirely like those found in the normal trap. They are, however, not accompanied by magnetite and augite, / ^ w ^v^ FiQ. 21.— Thin sections of aand and glass breccia ( JL, B) from the base of the Greenfield sheet at the city quarry and of trap (O) from Cheapside. Drawn by Charlotte F. Emerson. ^1 X 30. .B X 85. Ox 30. as in the normal trap, but are included in an olive-green streaky and hardly differentiated magma, which is often thrust in among the sand grains where the feldspar can not follow. Large trap fragments appear at the right of A and B. The third constituent of the rock, and a most interesting one, is of aqueous or igneo-aqueous origin, it being plainly formed by the action THE DEERFIELD SHEET, 423 of the vviiters of the mud ou the heated lava under ])ressure. It appears as narrow limpid bands in A and B, often interjected between the other con- stituents. It is made up mostly of a clear feldspar, in blades and plates sev- eral times twinned, of very fresh appearance, and polarizing in bluish whites, with the stron<^-, wavy or central extinction which characterizes the water- deposited albites of the cavities of the red diopside-diabase of the dike at Cheapside (fig. 24, C). This feldspar is also closely like the ordinary pla- gioclase of the amphibolites and albitic schists of the metamorphic series farther west. There is also a pyroxenic mineral of a quite peculiar character associated with this feldspar, and like it plainly of secondary origin. It has extinction a = emerald-green, h — clove-brown to violet, ,c = red-brown. In this rock small groups of stout, colorless diopside crystals occur, often bristling on the surface of the sand filaments like the hematite (which is another constituent of this rock), and in one case a well-formed arrow- headed twin of this mineral was observed. The considerable development of the green pyroxenic mineral gives much of the tuff-like rock a green color and the appearance of being greatly weathered diabase, and this somewhat abnormal variety forms narrow and interrupted bands between the filaments of the red mud and small fragments of the trap. These latter have the primary and secondary feldspars weathered and inclosed in an olive-green groundmass. The hematite plates penetrate to the very center of these fragments. The mild was thus most intimately blended with the liquid trap in which the lath-shaped feldspars had already been crystallized. It furnished water for the hydration of the groundmass into an olive-green nonpolarizing glass, and some of the same superheated water produced the abnormal igneo-aqueous deposit which unites the normal trap with the sand filaments. Several years after the foregoing description was written I made a comparative study of the above occurrence and similar tuff-like beds in Meriden, Connecticut, during which many slides were examined and an analysis of the glass at Meriden was made. This gave me much clearer ideas of the part taken by the water in forming and shattering the glass (which proves to be a basic pitchstone) to make the fine sand and trap- breccia mentioned above, in carrying up portions of the basal bed to become the bomblike masses, and in promoting the formation of a rock resembling a crystalline schist. I therefore reprint here the substance of 424 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. this article, so far as it relates to the Greenfield bed, at the expense of a little repetition.^ The flow of the submarine lava bed seems here to have been unusually rapid, and the under-rolling to have been a somewhat subordinate phenom- enon; still, the convection currents rising- from the front of the bed seem to have generally chilled it, so that a somewhat thin layer of compact, heavy, fine-grained trap was solidified and under-rolled to form a basal bed protect- ing the liquid mass above. When the sheet had advanced over the muddy bottom so far that the imprisoned vapors could not escape laterally, some slight and local disturbance broke up this basal layer more or less, the heat reached the water-soaked sand below, and steam and mud frothed up into the mass of the still liquid lava in great quantity, carrying many blocks of the basal bed. These abnormal conditions promoted the forma- tion of unusual varieties of trap. The absorption of water caused the formation of much basic pitchstone, while repeated smothered explosions shattered and commingled the heterogeneous products. GENERAL CHARACTER. For a thickness of 30 to 70 feet and for a distance of several miles in the vicinity of Grreenfield the basal portion of the trajD sheet is a mixture of sand, fragments of various sandstones, fragments of various kinds of diabase — some with glass base, some with hyalopilitic base, and some resembling andesites, all unlike the monotonous Triassic diabase — and abundant fragments of glass, all cemented by glass, and variously shattered and recemented, and the interstices filled by a water-deposited mixture of albite, diopside, calcite, segirine-augite, and hematite. The main mass of the trap sheet is normal and continuous above this confused mass, and in many places the basal portion of the sheet can be seen to be a. continuous mass of trap beneath the breccia, so that the latter must have been formed in the midst of the sheet itself The sheet is a normal, contemporaneous sheet, often showing- a ropy flow structui'e at the surface. GREENFIELD QDAERY EXPOSURES AND CONTACTS. For a mile north of the quax'ry beneath the observation tower east of Greenfield one can walk along the line of contact of the trap on the ' Bull. Geol. Soc. America, Vol. VIII, p. 64. s » H g s < s ^ PLATE Villa. 425 PLATE Villa. DETAILS OF TRAP RIDGE EAST OF GREENFIELD. Fig. 1. — Photograph of the south face of a large bowlder at the foot of the cliif below the quarry. The rounded and angular blocks are trap, and they are inclosed in the sand and glass mixture, which often shows fluidal structure. Commencing to the right of the watch chain, which is to be seen in the lower left-hand corner of the picture, and continuing upward for twice the length of the chain is a series of four rounded blocks connected by narrow necks, and sending out narrow, angular lobes — forms which can not have resulted from explosions throwing masses of lava into the air. The effect of the pile of great round blocks with comparatively small amount of interstitial matter can only partly be given by the photograph. See page 431. Fig. 2. — Enlargement of the part of fig. 1 which lies to the right of the watch in the photograph. A band of the sand and glass mixture extends across from the upper right corner and separates a large, rounded block above from a double block below, whose parts are joined by a narrow neck near the center, while the part to the right sends down a long, curved lobe into the breccia below. This shows one of the forms which can not have been "bombs" in the ordinary sense. See page 431. 426 PLATE VIIIj. 427 PLATP] VI 1 1ft. INCLUSION OF MUD IN UPPER SURFACE OF TRAP SHEET. Fig. 1.— a block of trap from the contact of a sheet of sandstone 12 feet long and a foot wide which was included in the trap a few feet below and parallel with the surface. The lower surface of the specimen was in contact with the sandstone. The whitest spots are steam holes filled by secondary calcite. Vhe trap is full of drops and lobate masses of the gray mud. From the north end of the east wall of the cut. Dibbles Crossing, Holyoke. About two- thirds natural size. From photograph. See page 456. Fig. 2. — Polished surfaces of pieces from the south end of the cut, to show the intimate mixture of the shattered trap and the light-gray mud. The mud fills many of the steam holes in whole or part. Natural size. From photograph. See page 456. 428 U. ti. QEOLOOICAL SURVEY MONOQHAPH XXIX PL. Vlll/; INCLUSION OF MUD IN UPPER SURFACE OF TRAP SHEET. PLATE VIIIc. 429 PLATE VI lie. THIN SECTIONS OF MATERIAL FROM GREENFIELD AND MERIDEN "ASH BED." Fig. 1. — Red htmatitic trap with secondary albite in perfect twinned crystals lining the interior of steam holes. Two large half-filled cavities and three smaller ones, wholly filled, appear. The large porphyritic plagioclase to the right is mottled from decomposition. Green- field, near Cheapside Village, at the electric railroad cut. See page 442. Magnified 20 times; crossed nicols. Fig. 2. — The interstitial aqueous deposit of plagioclase (probably albite), diopside, and segerine- augite. The plagioclase has a dusty, altered center, caused by an early change to calcite and a limpid exterior of later formation, which resembles the secondary plagioclase of fig. 1. The diopside is marked by strong boundaries and distant cleavage. The segerine-augite is in dark patches. The darker bordering portions are altered to serpen- tine with development of cleavage. At the lower border patches of the black sand appear. At the top and right edge are isolated spberulites. Greenfield quarry, 20- feet above base of bed. See page 434. Magnified 35 times ; crossed nicols. Fig. 3. — Scoriaceous sandstone. The dark parts are the rusty sandstone, red in the interior of the bands, and blackened by heat exteriorly. They show mud flow. The light parts are irregular, limpid, plagioclase grains. The mud has shrunk away at the top from a first growth of this kind, leaving a thin film of black grains, and in the narrow space a more limpid, plagioclase growth occurs. In the center of the older growth is a highly refring- ent mineral (datolite?), showing a micropegmiititic structure with the plagioclase. See page 435. Greenfield, Cheapside cut. Magnified 20 times. Fig. 4. — Greenish-brown glass with yellow borders, which are devitrified in series of small spherulites with dark centers. The glass has been shattered, while the fragments were slightly plastic. The fragments are in place in the slide, and the cavities are partly filled by a secondary water-deposited albite growth. See page 432. From Meriden "ash bed," near top on south path. Magnified 35 times. Fig. 5. — Hyalopilitic diabase from the Meriden "ash bed." Base formed of tufted, feathery, and fasciculate groups of beaded threads. Large olivine at right, large augite full of glass inclusions on left. Contact of basal bed on glass breccia. See page 436. Magnified 35 times. 430 U. b. CiEOLOtilCAL SURVEY MONOGRAPH XXIX PL. VIIIC 4 5 THIN SECTIONS OF MATERIAL FROM GREENFIELD AND MERIDEN "ASH BED.' TUB DEEKFIELD SHEET. 431 sandstone with the vertical wall of the traj) risinj>- above. Here there seems to ha\e been no distinet basal bed, but the \\hole mass was cooled nearly to the crystallizing point when the sand rose up into it at almost equal intervals, and the streams of the sand and glass breccia formed by the water rise in g-reat streaks or "schlieren," anastomose, and pass with iiuidal structure around the great rounded blocks of the normal traj), which make somewhat more than half the wall. At the quarr}^ is a more distinct basal bed of trap 7 or 8 feet thick, more or less shattered and displaced, and the sand can be seen continuous with the underlying sandstones rising in rifts in this basal bed and frothing out into a scoriaceous sandstone, where it meets and blends with the breccia above. This breccia is 60 feet thick — a greenish mass of shattered glass and trap, full of filaments of red sand shining with hematite scales. The rounded, bomblike masses of the compact and crystalline trap which are contained in this breccia grade superficially through hyalopilitic trap into the green glass, and while compact at center are toward the sur- face full of radiating steam pores. They seem to have been often carried aloft by the explosions into the still liquid glass, partially melted, and made siiperficially plastic by reheating, so that the steam has been able to struggle to the surface from the outer portion. Where they are large and angular they have been cari'ied but a little way from the base where they were formed; where they are small and spherical they are far-carried and much resorbed in the glass mass. Among these blocks are many long sheets and rounded masses con- nected by narrow necks, which could not have been blown into the air and have fallen as common bombs. (See PL Villa, p. 426.) A little way north of the quarry one can climb up the whole face of the trap by a steep path, and 60 feet from the base can study the top of the breccia. Here are unusually large masses of sand frothed up into an amygdaloidal sandstone and filled with water-deposited silicates like the Cheapside rock (see PL VIII c, fig. 3), and above this the trap is normal and crystalline and full of steam holes for a few feet, and then grades into the common compact columnar trap of the upper part of the sheet.^ ' In reporting my brief account of this case, Professor Dana has destroyed the meaning of the whole by an error. He says that the trap sheet rests on coarse sandstone-breccia 12 to 16 feet thick, instead of coarse trap-breccia. (Manual of Geology, 1895, footnote on p. 805.) 432 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. PETEOGRAPHICAL DESCRIPTION. DIABASE-PITCHSTONE. The diabase-pitclistone iii its purest form is a dark liver-brown pitch- stone, dull-green or mottled brown and green by reflected light, and red- brown by transmitted light. It is often an apple-green glass with the same dichroism. It has resinous pitchstone luster, and so differs from most tachy- lytes. The microscope shows a very minute, regular network of cracks, often developing into a perlitic structm-e around crystals and spherulites, which explains this luster. The deep-brown glass streaked with very deep brown is wholly amorphous and hardly to be distinguished from the Kilauea glass in common light, and, like it, it is not affected by acid. The pheno- crysts are of similar size and distribution, but with polarized light the feld- spar rods are always, and the large colorless pyroxenes sometimes, changed to granular calcite, easily removed by acid; the olivines, to fibrous serpentine. The fresh glass is full of small grains (cumulites), white by reflected light, red-brown by transmitted light, which are made of aggregates of minute grains (globulites). Even where the glass seems compact it often separates into small sheets and portions, showing minute curdled surfaces, and under the microscope the same wrinkled surfaces can be seen where small cavities have collapsed or where the fragments have flowed or have been drawn out in threads. The glass has been shattered into angular fragments by sudden explo- sion while still able to flow under slow pressure. Each of the fragments is then bordered by a layer of even thickness of paler-brown and equally nonpolarizing glass — an effect of the heated waters on the iron content. The larger fibrous spherulites in the glass are usually perfect circles or ovals, but they are sometimes distorted by flow or pressure. They are often bordered by several concentric bands of lighter and darker brownish- green glass, each band having a concentric radiate structure. The central part is colorless and beautifully radiate-fibrous, showing perfect black cross. The fibers are optically positive and polarize like a plagioclase. They are not affected by boiling acid or alkali. Sometimes the centers are filled by a greenish granular mass, which scarcely polarizes, showing only scattered light points. The spherulites are often broken and found in parts in the breccia, and the layers separated and crushed, so that the glass seems full of fragments of eggshells. THE DEERFIELD SHEET. 433 A fibrous devitrification sometimes afi'ects all the fragments of a slide, each oue beinc now a pale-yellow devitrified glass of a finely tufted or fibrous sti-ueture radiating- from many centers. The fibers have the same optical properties as do the spherulites. The inclosing glass is more granu- larly devitrified, polarizing in dots. The o-lass sometimes rindergoes a peculiar calcification, which seems to me rather a metamorphic change produced by the heated waters than a later decomposition by cold atmospheric waters. A fragment of glass will be red- brown at the center, pale-brown farther out, and perhaps colorless at its border; its angular boundaries will be sharply defined and the phenocrysts equably disseminated through the whole, and with common light the whole seems unchanged glass. It will, however, polarize in whole or part in broad patches of bright and softly blended colors and show everywhere the luiiaxial figure of calcite. Acid removes it readily and leaves only a powdery remnant. The outer colorless part is generally devitrified in plumose patches or in series of minute fibrous globes in the greenish fibrous devitrified glass. The calcite disappears rapidly with acid, leaving an opaque-white granular residue, while the colorless glass becomes opaque- white in lines and streaks, showing a concealed fiuidal structm-e. It is noteworthy that among all the reactions carried out here so little quartz is set free. Under the influence of the heated and carbonated water the glass, rich in calcium and alkalies and poor in silica, tends to split into calcite and acid feldspars. This explains the formation of spherulites and the fibrous devitrification of the glass, with the abundant development of calcite. QLASS-BBECCIi.. Under the microscope a fragment of the greenish tuff-like mass, taken 20 feet from the base of the bed, was composed as follows: The first thing that attracted attention was the fine red sand, each gi-ain being covered with iron rust. Where this was in thick masses it was still red in the interior, but on the exterior was black from the recrystalliza- tion of the iron rust by the caustic effect of the melted lava, in which it had been disseminated in threads and sheets. In the interstices between these dark sand portions many minute angular grains of diabase, like that found in the basal bed, were scattered. These had been broken up by an early explosion and earned up from the base with the sand. The whole had been MON XXIX 28 434 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. cemented by an olive-green glass, containing a few crystals of plagioclase and scattered spherulites, penetrating among the sand grains and to the very center of sand areas, which would otherwise have been called sand- stone fragments. The whole thus formed has been again shattered, and is now cemented by a hot-water deposit of albite, calcite, diopside, and gegirine-augite. Beautiful large hexagonal plates of hematite bristle over the trails of sand grains, and in all the other constituents except the basal trap fragments. Sometimes cavities of later formation are filled by radiat- ing chalcedonic growths, with centers of calcite and ankerite and copper pyrite. The water-deposited plagioclase (PI. VIII c, fig. 2, p. 430, the colorless center) has the appearance and the optical character of the small but per- fect albites (PI. VIIIc, fig. 1, and fig. 24, C, p. 422) which line the steam holes in many places in this bed, and often rest upon the earlier diabantite. These I have proved by optical and specific-gravity tests to be albite.^ It has also a curious resemblance to the albite of the "albitic" schists and amphibolites, and the whole mixture has some resemblance to a crystalline schist. The gegirite-like mineral (PI. VIIIc, fig. 2, the dark grains) is in shapeless grains and shows a strong prismatic cleavage like that of augite. It is intergrown with the feldspar, calcite, and diopside in such a way as to show that they were all deposited together. The absorption in this min- eral is very strong: a = deep blue-green, Ii := violet to olive-brown, some- times with shade of green, c = brownish yellow. A single twin with an extinction of 38° on either side of the suture was found, and the maximum of the blue-green absorption was also at 38° on either side of the suture, and this blue absorption represented the greatest elasticity. The mineral has thus the negative^ sign and the strong absorption of segirite and the optical figure in the position of augite. It is therefore allied to the segirine- augite of Rosenbusch, but the absorption parallel to a is clear blue-green and not grass-green. Large patches of the mineral are changed to a yellow- green serpentinous mineral, which under crossed nicols is almost black, but with scattered points of light. iMineralogical Lexicon, under "Albite": Bull. U. S. Geol. Survey No. 126, 1895. 2 By an. oversiglit the mineral is said to have the positive sign in the article cited, and the absorp- tion color is given as blue. This is only true in some sections between a and b, which blend the blue, "reen, and violet. THE DEEliFIELD SHEET. 435 The (liopside is iu stout, small crystals or in long, stout prisms, some- times In-oken. They are enveloped by the segirine-augite without common orientation. AUYQDALOIDAL SANDSTONE. One of the columns of sand rising from the sandstone and penetrating the basal bed at the Greenfield quany expands 9 feet from the base, where it passes above the basal bed into the glass-breccia, and its central portion presents a scoriaceous appearance. It is a red sandstone filled with more or less rounded spots of a white silicate, which I have no doubt, from my examination of other similar cases, is mainly a granular plagioclase. The same thing is developed much more extensively at the top of the breccia, on the path going up over the cliff north of the quarry. Here for several feet in thickness the rock is a red sandstone closely filled with small cavi- ties. The whole makes the impression of a rather coarse, red amygdaloid with white amygdules. A still more attractive form of the same rock is found in the cut of the electric road at the Deerfield River, a mile south of Cheapside (see PI. VIIIc, fig. 3). Here a light-red sand rock is filled with the fresh white amygdules. Under the microscope the sandstone between the white fillings has a beautiful fluidal structure, thus heightening the resemblance to an amygdaloid. The cavities are superficially blackened by the recrystalliza- tion of the iron oxide. The white filling is mainly a fresh matted network of plagioclase blades, which shows distinct triclinic striation rather more frequently than is iisual in this water-deposited feldspar. They are ragged- edged from interference due to rapid crystallization. In the center of the cavities is another mineral into which the feldspars penetrate with a micro- pegmatitic structure or which runs out among them. It polarizes with bright yellows, and I suspect it to be datolite, as a mineral with the high glassy luster of datolite can be seen with the lens in the centers of some cavities. It shows no cleavage, and it has a rough surface like olivine, which agrees with the high refractive index of datolite. Other slides of this occurrence showed a curious radiate-fibrous structure with coarsely beaded fibers and extinction up to 40°, and some smaller stout, square prisms with flat ends. They present all the peculiarities of wollastonite. Another peculiarity is that the cavities seem to have been filled with the mixture described above, after which the sand has shrunk away from the -436 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. filling for a considerable distance along one or more sides, leaving a film of the black sand grains attached, and then a more limpid feldspai- has grown in the narrow cavities thus formed. CONTACT MATERIAL. A slide cut within the porous outer portion of the trap from the contact of one of the bomblike masses of trap with the glass-breccia showed only a very feldspathic and vesicular diabase. Specimens cut from the fused border between the two showed a rock with the aspect of an augite-andesite. The well-shaped feldspars of two generations and the equally well-shaped olivines were inclosed in an opaque red-brown base, which in thinnest places revealed its hyalopilitic or fibrous structure. (PL VIIIc, fig. 5, p. 430.) Its outer surface had at times a rounded and lobed, fused sm-face, and just under the surface a single row of steam holes filled with silica, all indicating a superficial remelting. LITHOPHTS^, In one large specimen from near the base of the bed north of the quarry at Greenfield the breccia was full of well-formed lithophysse a half inch to an inch and a half in diameter. The cavities were half filled with cm-died masses of a lighter rock. CHEMICAL DISCUSSION. In his article on the lavas of the Sandwich Islands and other volcanic islands of the Pacific,^ Cohen states that all the basic glass found was anhydrous, and in general a basaltic pitchstone has not been described. I have studied slides of many tachylytes, and only that of Ostheim, in Hessen, with its green superficial color and liver-brown interior color, resembles these glasses. I have not seen any analysis of this rock giving water determination. It is deeper brown than most of the glass here studied, and contains large, round, oval spherulites with still deeper color, with radiate structure, and drusy surface. The other basaltic obsidians quoted by Zirkel do not contain more than 2.75 per cent of water. The following analysis of basic pitchstone from the Meriden "ash bed," by Mr. H. N. Stokes, of the United States Geological Survey, was made on a pure liver-brown glass identical with that here described. ^Neues Jahrbuch fiir Mineralogie, Vol. LVIII, p. 57. THE DEER FIELD SHEET. 437 It has specific gravity of 2.87, aud melts easily to a black magnetic and frothy glass. Basic pitchstone from ^'■ash bed^' northeast of Meriden. SiO- TiO, COj P.O5 F AI2O3 Fe^Os Feo NmO BaO SrO CaO MgO KjO NajO LijO (at 110° ... ^ (above 110° Per cent. 46.86 1.13 2.19 .15 trace 13.96 5.23 4.67 trace .03 trace 9.42 7.69 2.02 1.85 trace 1.29 3.43 99.92 ORIGIN OF THE GLASS AND MINERALS. It remains to consider the cause of the extensive development of glass in the midst of the trap as a result of the introduction of water and sand in so great a quantity. It might seem probable that the introduction of so much quartz would have perixdtted some solution, so that the glass, being more acid, would more easily take the vitreous form. The percentage of silica is, however, somewhat less than in the average of the diabase, and a study of a great number of slides failed to show any trace of quartz or tridymite, except in a late vein filled with coarse calcite and analcite. Slides boiled with concentrated HKO failed to show any change. It is more probable that water has been absorbed in such quantity as to have contributed to the observed result. While obsidians are water-free, pearlstones average 3 per cent of water, and pitchstones 7 per cent, while the corresponding porphyries average only IJ per cent. It is remarkable, considering the quantity of water which must have 438 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. been carried into the mass with the mud, that there is almost no trace of amygdaloidal development. Only one fragment of a trap inclosed with others in a breccia contained small steam holes. The collapsed cavities with wrinkled interiors and the absence of the common steam holes are explained by the absorption of the water by the magma, and this absorption explains the unusually large development of basic glass in connection with this exceptional occurrence. Above the compact and columnar trap which rests on this hydrated glass is the usual coarsely amygdaloidal surface layer of the trap, whose moisture seems to have no comiection with this development at the base of the bed. It was, however, in this surface amygdaloid in the Deerfield bed that I found perfect secondary albite crystals resting on diabantite in the amygdiiles. The great abundance of calcite and its intimate admixture with the other constituents are remarkable. I have elsewhere given reasons for thinking it in great part formed during the consolidation and cooling of the glass It is consonant with this that the feldspars formed during this cooling, especially those in the spherulites, are quite acid, while Hawes found very basic feldspars an abundant constituent of normal trap. When these secondary feldspars are boiled with strong hydi'ochloric acid and treated with fuchsin there is no trace of decomposition, and the optical characters indicate a very acid feldspar. The COg brought into the mass by the waters from the coal-bearing sandstones below may have taken possession of a large portion of the Ca, leaving the Na to go into the newly made feldspar. The similarity of this aqueous feldspar to that in a metamorphic schist is remarkable, and it is interesting to find diopside and segerine-augite and hematite formed with it, thus making a very peculiar crystalline schist in a very peculiar position. It is again remarkable that diabantite and its serpentinous decomposition product are rare in these glasses and the associated traps. This militates against the idea that the peneti-ation of the ground waters into the liquid trap is the cause of its chloritization. The lava bed flowed over the muddy bottom quite rapidly, and the heated mud and water have frothed up into the still liquid mass, causing an intimate blending of sand and lava for a thickness above the base of the bed of from 30 to 75 feet and for a distance, parallel to the advancing front of the sheet, of several miles. THE DEEKFIELD SHEET. 439 The suddeu introductiou of so large a volume of water has caused the mass to cool as a spherulitic glass with a mimite crackling, which gives it a pitchy luster and a large content of water (4.72 per cent), thus forming a basic pitchstone, which does not seem to have been described before. As a further direct influence of the water on the lava, many abnomial forms of trap were made locally. The liquid mud rose in the liquid lava with many explosions, shattering the abnormal mixtui-es already solidified, and blending them in still more complex mixtures while the newly solidified glass was still slightly plastic. The whole is cemented by the remnant of the glass, or an aqueo-igneous stage follows the igneo-aqueous, and a more distinctly hot-water product, consisting of albite, diopside, hematite, calcite, and segerine-augite, forms the cement. This glass-breccia is proved to be an integral portion of the trap sheet by the fact that there is a heavy basal bed of crystalline trap resting upon the sandstone, and the breccia grades downward into this bed, as it does also upward into the overlying crystalline trap which forms the major portion of the overflow. Sometimes this basal bed is shattered and its parts are carried up into the glass and rounded and filled with superficial steam holes by remelting. CONTACT OF THE SANDSTONE UPON THE DIABASE. On either side of the mouth of Fall River, and for a mile south, con- tinuous outcrop of the upper contact is visible at low water. The rather soft, deep-red, shaly sandstone is wholly imaltered and never included in the trap, while it folds around all small protuberances of what was, doubtless, the old ropy surface of a lava flow, its laminae thickening in the bottom of the pro- tuberances till they have evened up the surface of the ropy lava, and at times fragments of the traps are wholly included in the sand. The sand even fills the opened steam holes. Just north of the point where the wood road goes east from the Sunderland Hotel there is another fine contact of the sandstone on the trap, near its south end. FALL RIVER FAULT. On following down the trap from its north end to the Connecticut, one finds that it halts abruptly at the water's edge east of the mouth of Fall River and faces an island of sandstone which lies just in its line of strike ; but on following the bed up from the south, one discovers that it 440 GEOLOGY OF OLD HAMPSHIRE OOTJNTT, MASS. ends abruptly in a vertical, northward-facing wall 165 feet west of the ter- mination of the northern portion already noted, and on the other side of Fall River. We notice that the sandstone resting on the trap near the northern terminus of the west ridge is exactly the same soft, deep-red shale, and that it rests upon a trap with the same amygdaloidal texture and the same min- eral contents as at the south terminus of the east ridge. At the same dis- tance from this contact in either ridge the rock becomes suddenly filled with the same chopped-straw-like forms, which may be fucoids, or indusia of the insect found farther east in the sandstone, or, more probably, concretions; and measuring a second distance we find a thin bed of gray conglomerate interposed in the sandstones in both cases. The sandstone series thus agrees minutely on either side the line, and with the traps must have been faulted with the dip about 165 feet. The rock is intercalated in the sandstone and dips eastward with it. It would seem to follow this dnection only a little way before coming to the Fall River fault, as an artesian well sunk on the east bank of the river by the Montague Paper Company (see "Artesian wells," Chapter XII) went down 900 feet below the level of the dam, while immediately opposite on the west, and separated only by the width of the river (about 1,430 feet), the trap dips toward the well with an angle of 32°, which would make it appear in the well at 894 feet below the surface, whereas it does appear at 585 feet, making an upthrow on the east of the fault plane of 209 feet. THE UNITY OF THE SHEET. From its north end to a point just below the lower suspension bridge at Turners Falls, the trap ridge is an inconspicuous object seen from the surface of the high sands on the west, and here it is for a distance entirely covered by them. When it reemerges it has a greater width and has changed its direction to southerly. This is my interpretation of the facts at this point, and I find myself here again at variance with the conclusions of Professor Davis, cited above. It is certain that there is no proof that the trap from the south runs by the northern strip on the west, so that the latter could be called a posterior range to it. There is also no conclusive proof that the two parts of the dike are united under the sands. I think it most probable that they are. The two are lithologically identical, as are the sandstones above them, and the region abounds in faults. THE DEEEFIELD SHEET. 441 Southward the trap rises higher aucl continues, with lofty, nearly vertical walls on the east and west, between the river and the town of Greenfield. President Hitchcock quotes "trap tuff" as constituting, a mile east of Grreenfield, "a large portion of the ledge of greenstone, which is in places a hundred feet thick." This is the great pitchstone-breccia at the base of the bed described above. Across the deep notch of the Deerfield River the sheet rises and thickens in Deerfield Mountain and looks down with vertical wall upon the village of Deerfield at its foot. It shows just east of the village the finest columns in the State, 2 to 3 feet in diameter, and in places distinctly curved.^ Farther south, just before crossing the river, the great sheet shows, from below upward, four horizons of heavy amygdaloids, indicating, doubt- less, that it is a composite of as many great lava flows in this portion of its extent. To the south, in Mount Toby, where it is thinner, it is amygdaloidal in nearly its whole thickness, while at its north end it is compact at base and heavily amygdaloidal in its upper portion. PETROGRAPHICAL DESCRIPTION. NORMAL DIABASE. The rock is a typical diabase, ranging from aphanitic varieties to those where the white, flat feldspars are 2 to 4°"" square, and from compact to very coarse amygdaloidal. The different veins are of very uniform texture and always in an advanced stage of decomposition, though appearing quite fresh; plagioclase, apparently of two species, augite, magnetite, and olivine are uniformly present. Apatite can not be detected. The common plagioclase, probably labradorite, is always by far the most abundant constituent, and the angle of extinction of its long rodlike crystals is commonly 12°. Several varieties of the rock are subporphyritic by the development of white spots, made up of groups of stout crystals of a second triclinic feldspar, apparently distinct from the first, whose angle of extinction is 21°. Both feldspars are thoroughly decomposed, commonly from the center, and sometimes show only aggregate polarization. The augitic constituent has for the most j)art gone over into a mixture of green and brown chloritic minerals, but. here and there an exceptionally arge crystal remains in whole or in part intact. IE. Hitchcock, Geol. Mass., 1841, p. 642. 442 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. The rock at the new cutting south of the Deerfield River and south- ward is very fine-grained, breaking with conchoidal fracture, dark-gray and compact at the base of the dike, and there distinguished by an abundance of the well-known feathery aggregations of magnetite grains, while in the whole upper portion it is coarsely amygdaloidal, the amygdules filled com- monly with diabantite, calcite, or both — when one penetrates below the deep layer of rusty scoriaceous rock from which all the secondary minerals have been removed — and here the magnetite is never arranged in feathery groups. At the old cutting on the other side of the Deerfield River, a few rods north, the rock becomes more granular in texture, and grayish- and reddish-white varieties occur, subporphyritic and abounding with flattened steam cavities, filled now with diabantite. These colors are arranged in layers, giving the rock an indistinct fluidal structure. These varieties con- tinue northward and are exposed in great force for nearly a mile of fresh cuttings where the road from Grreenfield to Turners Falls crosses the dike, and from the Suspension bridge, at the end of this road, along the river side for a mile north, to the mouth of Fall River and beyond. Through all this area prehnite and the products of its decomposition occupy the amygdaloidal cavities in very great quantity, accompanied everywhere by traces of copper minerals in place of calcite and chalcedony, which abound farther south. The masses of native copper found in the till must come from here. The most interesting variety is a very coarse one, abundant on the Greenfield road, which contrasts pleasantly with the somber gray of the prevailing types. Broad white plates of the feldspar stand out upon a dark-red background of decomposed augite, the' whole sprinkled with amygdules of prehnite and diabantite. That this coarse variety is younger than the greenish-gray subporphyritic trap is clear from a large slab from the middle of the slope on the Grreenfield road, showing a contact of the two, upon which the latter is cut off immediately and sharply and without change, while the former has a layer of deep-red, very fine-grained rock IS™"' wide adjacent to the contact plane. It seems to me, however, to represent only a slight difference in age and to be probably a case of "schlieren," in the sense of E. Reyer.^ An exceptional rock occurs abundantly in bowlders on the south side of 1 Theoratisclie Geologie, 1888, p. 80. THE DEERFIELD SHEET. 443 the Deerficld River, but I have not met it on the north or in place. It is a clear, lig-ht-g-ray rock, Avith roundish blotches of white, and it looks like a weathered leucitophyre. Under the microscope the blotches are seen to be made up of aggregated stout crystals of plagioclase, and the rest of the mass between of rodlike plagioclase and magnetite, with almost no augite. The rare amygdules in this rock are filled with a fine silky, radiated mineral, apparently an altered prehnite resting upon diabantite, or more rarely lined with glassy crystals of albite, with datolite, pyrite, or globules of zincblende. RED DIOPSIDB-DIABASK, WITH SECOITDAKY ALBITB. Much of the basal part of the Deerfield bed just north of the Deerfield River is a peculiar rock, remarkably different from the usual monotonous trap of the region. It has been radically metamorphosed by hot water during- its cooling. It is a fresh, fine-grained, bi'ick-red rock, full of small cavities and scattered larger ones, both lined or filled with exquisite albite crystals large enough to be easily studied with a lens (fig. 24, C, p. 422). The feldspars "of first consolidation" in the body of the trap, which are near oligoclase, have been floated to their present place in delicate feathery groups. They retain their sharp crystal outlines and trace of cleavage and multiple twinning on two bands, but have been changed to a sericitic mass of subparallel scales and needles of two kinds, very minute needles polar- izing in low colors 0.015°"™ long and 0.0003°"" wide, and brightly polarizing scales 0.04™™ long. They seem to be kaolin and mica. The ordinary brown interstitial augite is wanting, but a few much twinned idiomorphic diopside crystals occur. The above minerals are free from the very abundant hematite which in grains and dendritic growths fill the second generation of feldspars and make most of the slide opaque, and which entirely replace the iisual black ores and colored augites. The second generation of feldspars is often in sheaves of parallel fibers, one or more generally proj ecting far beyond the rest. They are heavily loaded with the red rust, but often have clear borders or the rust is in a cross occupying the diagonals to the square sections. Many of the cavities are filled with a fresh albite mosaic, and this at times closely resembles the limpid feldspar mosaic of the amphibolites, being often without twinning and showing the same concentric polarization. This want of twinning is largely due to the development of the albite in 444 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. mica-like scales with their M (010) faces arranged parallel to the section plane, so that they show no twinning, but give with convergent polarized light a negative bisectrix. In some of the large cavities a broad-bladed mineral, probably barite, has formed in many separate and parallel plates, and all these have been coated with albite and then removed by solution. Chalcopyrite also appears in these cavities. There is no diabantite nor any trace of ordinary weather- ing in the slide; and it is probable that heated waters acting on a magma in which the first feldspars were floating have decomposed these, changed all the iron into hematite, thus preventing the formation of the dark augite and the black ores, and have then deposited the residuum of the feldspathic material in the steam holes. Specimens can be obtained where the trap has recently been blasted to make way for the electric road and the fragments dumped on the steep slope extending down to the Deerfield River. The difference of the rock from the normal diabase is shown by order of crystallization of the constituents of this and of the normal diabase. Diopside- diabase. Normal diabase. First plagioclase. Magnetite, Diopside. First plagioclase. Hematite. Second plagioclase. Second plagioclase. Augite. Steam holes. Steam holes. Third plagioclase. Diabantite. It is interesting to see here the development of the sericitic growth and the albite mosaic without the formation of hornblende. The rock here incloses fragments of fine sandstone exactly like those found at the Grreenfield quarry about 3 miles north on the same dike, and, as there, it is greatly baked and fused with the trap. (See p. 419.) PAEAGBNESIS OF SECONDAKT MINERALS. During the summer of 1880 a heavy cut was made through the trap on the south side of the Deerfield River for the extension of the Canal Railroad, which opened up veins canying the usual trap minerals in great abundance and beauty. The veins run nearly vertical, with a thickness not above 4 inches, and they were exposed to a depth of 60 feet. Later a similar cutting along the north side of the stream and directly opposite afforded many large cavities filled with the finest transparent datolite of unusual size, but lacking THE DEERFIELD SHEET. 445 wholly the variety shown on the sonth side. I have included a detailed study of these minerals in the Mineralogical Lexicon,^ and give here — 1. The paragenesis of the stilbite-chabazite veins — 1. Radiated stilbite. 1- Prehnite. •J. Chabazite. 2. Heulaudite. 3. Oalcite. 3. Prismatic stilbite. 4. Pyrite; or 4. Chabazite. 5. Oalcite. 2. A general table of the paragenesis of the minerals found. The old- est is first, and the overlap of the words corresponds approximately to the overlap of the minerals: Diabantite. g Albite. 1 Prelinite. p< Bpidote. 1 Axinite. E Tourmaline. "3 ^ Calcite. n 1 Fluor. 1 Sulphides. Pi DO Datolite. ■3 Spbene. Calcite. Sulphides. 'Natrolite. Stilbite. s Heulandite. Analcite. a Calcite. Fluor. p Sulphides. o Chabazite. Oalcite. Fluor. Pyrite. rSaponite. p. Chlorophseite. f tlie thin-flssile, sandy shale, as if balls of putty had been separated by being folded in thick wads of wet wrapping paper. Above this intimate mixture a few angular fragments of scoria are inclosed for a foot or two in the thin-bedded sandstones. This layer can be followed north 10 miles wherever the upper surface of the trap is exposed. Another contact of the sandstone upon the trap occurs on the West- field-Holyoke highway, just where it crosses a brook, and this is the most southern point where the trap contains limestone inclusions at its surface. President Hitchcock plainly refers to a further effect of the trap farther south on this line, in West Springfield, at a place which escaped my obser- vation, when he speaks of the limestone in contact with the trap being converted to "tripoli" and in part made brittle as glass.^ MAGMATIO DIFPEEENTIATION. Many fragments of the trap which were inclosed in the mud while still molten are bordered with black from the concentration of the iron in feathery groups of twinned octahedra of magnetite. This illustrates on a small scale a process which has been the subject of much study — the differ- entiation of a molten magma into a more basic portion, which seeks the cooled outer surface, and a more acid one, which remains at the center. When this process is carried to its limit the centers of the fragments become white and free from iron and iron-bearing minerals, and the frag- ments of white trap described on page 365 seem to have been thus formed. They are found only in this contact layer and in the sandstone immediately above it. ORIGIN OF THE CLAY AND MARL DEPOSITS. It is hard to explain how, over a portion of the surface of the great sheet, so large a quantity of laminated marl can have been deposited and then become so regularly and deeply intermixed with the trap. It seems most probable that the central currents carried the mud out over the sheet while it was still moving, and filled its brecciated surface, and that the mud flakes sank down at times into the still-liquid trap in such quantity that they were merely indurated and cemented by the small quantity of the diabase. 'Geol. Mass., 1835, p. 433; 1841, p. 659. 460 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. The thick trap sheets flowed out over the muddy bottom of the bay, and their heat produced strong upward convection currents and corre- spondingly strong indi'afts from the sides, which carried muddy waters out over the surface of the trap while it was still flowing and covered it with a quantity of calcareous mud out of proportion to what would have been carried in the same time by the normal currents. I have seen sheets of newly solidified lava careen and slide beneath the liquid mass at Kilauea, and the sheets of mud and lava may have thus become variously mingled here, producing the results described above. The surface of the Holyoke trap sheet is filled with fine mud just as far north as the fine Chicopee shales extend; and farther north, where the sheet flowed over coarse gravel, nothing of the kind occurred, because the coarse gravel could not be thus carried out over the thick sheet. ON THE UNDER-ROLLING OF THE SOLIDIFIED SURFACE OF THE TRAP. The appearance of the same layer at the base of the trap sheet is explained by the under-rolling of the newly solidified surface of the sheet, as when a carpet is unrolled on the floor what was on top descends along the front and comes to lie inverted beneath. Thus the porous mud-filled surface came to form, inverted, the base of the bed, and to rest, though filled with fine mud, upon the coarse sand onto which the sheet had advanced.-' 'I have already reported very briefly upon this occurrence (Am. Jour. Sci., 3d series, Vol. XLIII, p. 147); too briefly, it would seem, as the facts given were wholly misunderstood and incorrectly quoted by Professor Dana and made to do duty in proof of the laccolithic origin of the Mount Tom trap sheet. In his Manual of Geology, on page 805, he says : "The limestone had been torn off from a layer not visible in the section." This was the very point I was trying to disprove, by showing both that there was no bed in the older rocks of the region from which any such material could be derived and that the shapes of the inclusions were not such as would be possible in solid rock torn oft' from the walls of the fissure through which the lava flowed, since it was in thin filaments and flowed in to fill all the open steam holes of the trap fragments. On the next page, 806, he says : "A laccolithic origin and the abrasion of the underlying sand- stone are indicated by the occurrence of breccia beneath the trap, and especially by the limestone chips in the lower part of the mass of the trap, and also over its upper surface, as described by Emerson. A bed of limestone was evidently divided by the advancing tongue of melted trap, part being left below and the rest above. As Emerson observes: 'The facts prove that the heavy trap flowed over the sandstone, abrading and tearing it.'" This was plainly quoted from a very dim recollection of the article in question. There is no breccia beneath the trap. The inclusions can not be called chips, and there is not the slightest evidence that the melted trap has split asunder a bed of solid limestone. I have not made, in the article cited or elsewhere, the observation quoted in the last sentence, since the facts all prove exactly the opposite. I know of no facts favoring a laccolithic origin of the Holyoke trap sheet. THE HOLYOKE SHEET. 461 At every point where the surface of the trap sheet can be inspected, fniiii where it crosses the Connecticut to where it crosses the West- tieUl-Holyoke Raih'oad, it has included a great number of fragments of marly limestone and indurated clay, and the trap and limestone are often kneaded together. Within the same limits the base of the trap repeats all tlie peculiarities of the surface. It is amygdaloidal for about the same thickness and in the same way; the same dove-colored limestone occurs blended with the trap in the same way; and the subjacent arkose is almost wholly unaffected by heat. The 300 feet of trap have not pro- duced so much effect as is often seen upon the border of a 10-foot dike. This is best studied at the river's edge at the north foot of Titans Pier. On the other hand, where the molten surface of the trap sheet has come in contact with the sands of the sea bottom, as at Titans Piazza, 100 rods north, the trap is aphanitic at the contact, but pierced by great vertical steam holes, and the sandstone is greatly baked. It seems that the broad submarine trap sheet moved slowly westward, its incrusted surface being covered by a fine marly clay deposit which was in places desiccated and _ molded together with the still plastic trap, and that the surface was car- ried forward to be rolled over the front and become the bottom along a length of about 10 miles. The limestone and marlite inclusions of the surface and base of the trap have been described in detail above and their identity established, and similar cases of under-rolling of the Deerfield dike and of the posterior dike have been given elsewhere.-' PETROGRAPHICAL DESCRIPTION OF THE NORMAL DIABASE. GENERAL DESCRIPTION. The rock is so monotonously uniform in all its characteristics that much repetition will be avoided by giving first a general description of the com- mon type and then following this by a special discussion of the peculiarities of separate occurrences. The rock from the "Iron Grate," or Thermopylae, where a passage has been blasted through a projection of the Holyoke sheet for the river road to South Hadley, near Titans Pier, coming from near the middle of the sheet, is an especially fresh-looking variety, and may serve as the new type for general description. 1 See pages 419, 470. 462 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. It is a dark-gray, almost aphanitic rock, with broad conchoidal fracture and without any tinge of red or brown in its color. It is faintly subpor- phyritic, and with a lens the scattered, minute, squarish feldspar cross sections appear, and at times a triclinic striation can be observed. At times, also, one detects a black cleavage surface of augite, but only with difficulty. Under the microscope the rock is seen to be a typical diabase, the network of elongate feldspars inclosing the shapeless masses of augite. Two generations of feldspar, augite, magnetite, and apatite make up the constituents. The larger feldspars of earlier generation are distantly scattered in the field in squarish crystals or crystal groups, and measure about a mil- limeter across, and this is by far the commonest size in all the slides I have examined. They often show indication of transportation, being broken, or showing undulatory extinction, or having an external band which extinguishes diiferently from the central. A delicate zonal struc- ture is at times present, or the center is full of opaque grains and the outer .portion limpid. These latter structure forms are more common in the dikes than in the two large beds. The twinning striation is often interrupted and distant, so that quite broad patches belong to a single individual. At the type locality these large crystals are exceptionally fresh for specimens out of the large trap beds. They are, however, largely decomposed into a mass of shapeless, brightly polarizing scales, apparently micaceous, while more commonly both the generations of feldspar are decomposed into a fibrous saussuritic mass. In a long series of observations of the extinction angle of porphyritic crystals from every part of the valley, more than half the angles obtained were about 31°. This would indicate strongly that the feldspar was anorthite, which would agree with the results obtained by Mr. Hawes (cited below, p. 464) in an analysis of the porphyritic crystals of a dike cutting West Rock in New Haven. I may recall, also, Hawes's suggestion that the more difficult fusibility of anorthite may favor its earlier crystallization. The second generation of feldspar, which forms the latticework, is lath-shaped, often with ragged ends and notched and irregular sides, and averages 0.1™™ in length, though it is subject to more fluctuation than the larger group. Its extinction angles vary from 12° to 26°, which would best comport with the composition of labradorite. The rock under special THE nOLYOKE SHEET. 463 discussion is one of the freshest-looking in the valley, and yet it is some- times impossible to find in a slide a single feldspar on which one can observe the extinction, so decomposed are they, and the sei'ies of which the extremes are given above are taken from the whole length of the valley. The augite is strictly subsequent to the lath-shaped feldspars and pi'esents little that is specially noticeable, though oftentimes it is less decomjiosed than the feldspars. It differs thus in the large sheets from the diabase of the tuff above and of the newer dikes, where the augite is often porphyritic and contemporaneous with the earlier feldspars. I have in many places noted olivine with a query; but on reviewing the whole series of slides I have not been able to find either the unchanged mineral or any serpentine or hematite patches which would seem to have been derived from it at the locality under consideration or in either of the large trap sheets. In the dikes in the gneiss and in the newer dikes in the sandstones it occurs, and it may be wanting in the large beds only because of their advanced state of decomposition. Magnetite is uniformly distributed, always rather but never very abundant, generally quite well crystallized. The delicate featherwork of beaded octahedra is especially abundant at the base of the great bed at the contact on sandstone just north of Titans Piazza. Apatite, never abundant, is rarely to be detected except piercing magnetite. There is no trace of groundmass discernible between the constituents; rounded or pear-shaped blebs of glass appear in the older feldspars. Cavities filled with diabantite, rust, calcite, and zeolites are not wanting, even in the wholly compact rock we have chosen for discussion, but they are very minute. Sections from the upper surface of the dike where it is cut by Dry Brook in the northwest of South Hadley exhibit very beautiful amygdules, showing from without inward diabantite, calcite, and radiated natrolite. CHEMICAL COMPOSITION. In 1838 President Hitchcock analyzed the much decomposed and amygdaloidal trap from the east end of Mount Holyoke with the result shown in column 1.^ In 1875 Dr. G. W. Hawes published analyses of 1 Economic Geology, p. 135. 464 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS, the compact trap from Mount Holyoke (columns 2 and 3), and a mean of the same (column 4):^ Analyses of trap from Mount Holyoke. 1. 2. 3. 4. SiO.2 53.70 13.00 52.70 14.11 9.78. 1.87 0.45 9.36 6.42 2.54 0.89 52.65 14.17 9.80 2.03 0.44 9.39 6.35 2.57 0.87 52.68 14.14 9.79 1.95 0.44 9.38 6.38 2.56 0.88 AhO, FeO PejOs 21.00 0.19 0.70 0.15 MnO CaO MeO NajO KjO H2O 8.50 2.76 Is 1.61 1.58 1.60 100. 00 99.73 99.85 99.80 THE UPPER OR POSTERIOR SHEETS AND ITS FEEDING DIKES. This bed (see PI. IX, p. 446) runs from a point on the Connecticut River about a mile below the Mount Tom station, parallel with and about a half mile east of the Holyoke range, to and beyond the south line of the State, while its great irregular feeding dike is about 2 miles south of Smiths Ferry and just east of Mount Tom, where on the map the outcrop swells out suddenly. It is locally known as Little Mountain, and forms the culmi- nating point of Forest Park, to which the Electric Road runs from Spring- field and Holyoke. The trap sheet shows the low easterly dip of the sandstone, in which it lies at a horizon about '600 feet above the Holyoke bed, though north of Mount Tom the two beds seem to be much nearer because of the Mount Tom fault, which at the cut south of the Mount Tom Electric Railroad station brings them apparently within 30 feet of each other. The bed is thick, but does not seem to extend east of the Connecticut, where the tuff rests directly on the sandstone. Yet an inspection of the map may leave 1 Am. Jonr. Soi., 3d series, Vol. IX, 1875, p. 186. ^ Called thus by Percival in The Geology of Connecticut, because the trap ridges face west and subordinate ridges often appear before and behind the main one. THE POSTERIOR SUEET. 465 the impression that it extends, at least in a fragmentary way, far east. The long eastern projection of the Black Rock plug (see PI. IX, p. 446) and the one east, and the string of smaller })lugs elongate east and west, seem to be parts of it. They are, however, true intrusions, and their elongation seems rather to indicate the existence of a common ancient and deep-seated rissuro through which they, have been extruded. This is proved by the fact that they cut directly across the beds of the sandstone below the tuff, the tuff itself, and the sandstone above, while west of the river the tuff rests directly upon the posterior sheet. The sheet appears first as a great reef projecting into the Connecticut a mile below Moimt Tom station, its northern portion fine-grained and col- umnar, its southern coarse and in great blocks, and is doubtless continuous beneath the sand southwest to the interesting outcrop at Lymans Crossing (the first crossing below Mount Tom station), where a wall of trap is exposed in the railroad cut. The northern jaortion of the cut is rudely columnar trap, with an irregular surface dipping about 35° SE. Resting upon this surface is a coarse trap agglomerate, consisting of blocks a foot across and a fine sandy paste, in which many flakes of graphite appear. This is the normal relation of the tuff to the posterior sheet for a long way south. A rod south of this tuff is an outcrop of trap which, from its great fresti- ness and compactness, and from its containing inclusions of coarse amyg- daloid from the tuff, I associate with the Burnt Mill plug just south, which interrupts the sheet at this point. A few rods south of the crossing a brook crosses the road, and on it is the ruin of Aldrich's leather mill, burnt many years ago. The brook flows east along the course of a transverse fault, and at and below the dam can be seen very finely the outcrop of an intruded trap mass, which clearly cuts across the sandstones, bakes and twists them, and extends west along the north side of the mill pond. (See p. 494.) South of the brook and the fault the outcrops are continuous, and the posterior sheet can be seen to be wholly independent of the core which crops out north of the stream at the dam. Commencing at the railroad culvert over the brook, the sandstone can be seen on the south side of the brook in contact with and beneath the trap of the posterior sheet and having the unusually steep dip of 60° SE. beneath the trap because of the fault. From this point the sandstone can be followed along the south bank of the brook contuiuously, past the mill and the pond. It dips regularly to the southeast MON XXIX 30 466 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. - i> i a 6f?^ wiro beneath the trap (Avhich has only sHghtly indurated it), and thus sepai-ates it completely from the intrusive trap of the plug- north of the brook. The removal of the mill and dam has improved the outcrop greatly and disclosed a quite sharp anti- cline of the sandstone beneath the ti-ap in the south bank of the brook, which, as the dip of the sandstone is very low to the east, and the slope of the hill is in the same direction, has the effect to very greatly increase the width of the exposure of the trap, although the sheet is only 35 to 40 feet thick here. The half of this anticline as formerly exposed seemed to show the trap resting directly on the basset edges of the sandstone beds, and this, taken with the unexplained greater width, made it seem probable that the trap had broken through here.^ This anticline seems to cause the greater Avidth of the trap outcrop south nearly to Smiths Ferry. From the burnt mill the outcrop of the upper bed extends southward as a prominent ridge just west of the river road and separated b}^ a deep valley from the corresponding ridge of the Holyoke- Mount Tom bed to the west. In the steep west- ward declivity of this ridge the contact of the trap on the sandstone beneath can be found in many places, and the sandstone is indurated for a small distance downward and rarely sends up a steam hole into the trap above. The trap is covered by the heavy tuff beds, which seem in the neighborhood of the burnt mill to be blended with the trap itself, as if it had fallen upon the latter while it was still molten, so that it is hard to mark the true bound- ary, but no trace of such blending could be detected in slides cut for the pui-pose. The narrowing of the trap upon the map is due to the westward advance of the tuff upon it, by the elevation of the ridge, so that it outcrops in the steep westward-facing bluff. ' E. Hitchcock, Geol. Mass., 1S35, p. 429; 1841, p. 656. THE POSTERIOR SHEET. 467 Just northwest of the Smiths Ferry niih-oad station the trap is faulted shg-htly, the south side being moved a few rods westward, and in the low place in the ridge thus formed the sandstone approaches within 80 rods of the railroad. THE (iUEAT WIDENING OF THE TRAP AREA AND THE PEEDIVG THROAT UENEATH. A mile south of Smiths Ferry the trap widens to a triangular surface, a half mile on a side, and the ridge reaches its greatest height, rising westerly from the river to its crest in Little Mountain (now marked by the highest lookout tower in Forest Park) and sinking bj'- a vertical wall to the valley which separates it from the Mount Tom trap ridge. (See PI. IX, p. 446, northeast of Mount Tom.) Along the western edge of the expanded area the thin trap sheet still rests normall}^ on the sandstone, and on the eastern edge is covered by the tuff, and its great width is due to the fact that it dips with the slope of the hill east from its crest. In the deep inlets of sandstone running down into the trap from the north the latter can everywhere be seen to lie normally on the sandstone, with little baking, and along the border from this point around to the west the same conditions hold for a long way south, until one comes to the point where the wood road coming up from the reservoir crosses the brook and goes up onto Mount Tom, and where the posterior ridge itself rises to its greatest height in Little Mountain. At this point the face of the core is finely exposed for study, as indicated in fig. 25. The trap comes up from the depths with but a small portion (30 feet) of its width exposed, sending out great dikes into the sandstone north and south. The southern dike, starting with a width of 8 feet, was followed 50 feet. North of the core a small dike is seen inclosed wholly in the sandstone, and a wide dike branches from the main mass and can be followed a long way north before it is concealed by the talus. At the surface the trap flows out over the sandstone, greatly indurating it, and becomes the sheet which we have followed from the north to this point. The whole is like a great toadstool; the stem is the core which forms Little Mountain. The west- ern and most of the southern part of the "umbrella" is broken off by erosion ; the eastern part is the sheet dipping east beneath the tuff. The exposed wall of trap shown on the left in the figure seems to be a 468 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. portion of the western wall of the plug, just grazed by the Mount Tom- Holyoke fault, and from its point of contact with the sandstone on the north the boundary of the plug seems to run first north and then about east beneath the continuous area of trap. The southern half of its bound- ary can be much more closely located. Continuing south from the south contact of the trap which forms part of the core and the sandstone in fig. 25 along the highest portion of the bluff, and turning round the face of the high bluff at its south end where it overlooks the reservoir, one finds the sandstone to be only a thin veneer- ing on the face of the walls of the great throat, and one can dig at the surface and see that the trap extends down behind the sandstone. The surface boundary of trap and sandstone is, along this line, almost the boundary of the core also, and erosion has spared little of the western half of the overflow. This boundary skirts the eastern vertical face of the bluff for a little way north, and as this bluff soon turns to face south, and runs east less steep and elevated, the surface boundary of trap and sandstone, turns and runs parallel with it, but not quite so near the edge of the bluff, and becomes the south boundary of the sheet as the latter extends east from the plug. The baking of the sandstone from the point Avhere the plug was first reached has been exceptionally marked, but along this wall it is more intense than anywhere else in the valley and can be clearly perceived 12 to 15 feet from the trap. Along the middle of this south wall, which continues east from the plug, near the top, a foot- wide dike of trap is intruded between the nearly horizontal layers of the sandstone beneath the sheet. It is unusually decomposed, to a pistachio-green porous mass, with spheroidal structure. About 20 feet below this a great horizontal dike or sill starts, just at the reentrant angle made by the southward projection of the high bluff — that is, just where the plug ends and the wall of sandstone facing south begins and seems to branch off from the main trap mass. It starts with a width of 2 feet and runs down east, widening soon to 12 feet, and continues with the bottom concealed, and at its end it bends up suddenly, with the sandstone on its back, into a vertical position. It is exposed about 150 feet and is very fine-grained, black, and horizontally fissured for 2 feet at surface, and is an exceptionally fresh, ringing, small-columnar rock in the center. It sends two narrow dikes, an inch to a few inches wide, up into the overlying sandstone. TUE POSTERIOR SHEET. 469 These pass upward in fissures for about a foot and then bend the unbroken layers of tlio sandstone above into an arch, forming minute laccoliths, and clearly indicating that the sill was injected under strong pressure. SILLS INTEUDED IN THE SANDSTONKS BELOW THE POSTEEIOR SHEET. Besides the dikes and sills which penetrate the sandstones so abundantly in the immediate ^^cinity of the Little Mountain core, other small sills appear immediately beneath the posterior sheet at so great a distance that they can not be brought into very close connection with the core itself. The most northern of these is N. 60° W. of the Smiths Ferry station and about 6 feet below the top of the sandstone. There is a sill 2 feet wide which can be followed 20 feet, and 2 feet below this is another only 1 foot wide. The sandstone has strike N. 40° E. and dip 22° E. About a mile south along the blufP, at a point S. 65° W. of Smiths Ferry and west of the marked drumlin which conceals the tuff, a larger sill appears, 10 feet below the trap, which is 4 feet wide and 8 or 9 feet long. The sandstone is much disturbed beneath it. Along the boundary of the sheet farther south no other sills are found in the sandstone below until the western border of the plug is reached and the very abundant dikes and sills appear around its western and southern side, which have been described and figured above. There is a turnstile by the road, and steps going down to the railroad, a mile and a half below Smiths Ferry, and tlie field road southwest from here leads out over a ridge to an amphitheater, now called Forest Park, from which all the points here described are easily identified. The ridge is the continuation of the trap sheet soiithward. The beautiful horizontal 12-foot sill described above (p. 468) is in the north wall at one's right, and if one crosses the basin to the next ridge overlooking the reservoir, and west of the terminus of the Electric Road, the high bluff of the plug projects south toward the point where one stands, and above the screes of trap fragments the sandstone veneering can be seen abutting against the trap in the thick woods. Turning south from the east end of the 12-foot sill mentioned above, on the southeast of the plug, and going to the bottom of the basin near the brook, one finds a place where the sandstone is crushed into sharp folds a foot or two across and baked by the trap, which has penetrated it irregu- larly, but apparently only in small amount. 470 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. Along the eastern margin of this basin the outcrop of the trap sheet is plainly visible, covered by tuff and marked at its base by a prominent talus. As it is followed south its boundary makes a great loop to the east, where the ridge is cut through by a brook, but rises again in a higher l^luff south of the brook, now marked by a high trestlework lookout-tower. The contact on the sandstone below is everywhere normal — the compact trap rests on unbaked sandstone — until the boundary swings aroiind the south end of this prominent hill to a point which rises sharply in a bluff 30 rods north of where Roaring Brook crosses the road. This is the brook that comes up from the south and bends sharply to enter the Connecticut southeast of Mount Tom. Delaney's quarry, described below, is situated just south of its mouth, between the road and the railroad. 0/u>/rswji>sron£ »„ "/^ ToserHEn. SANOSro/^^- AMyeoAio/o / ^ J / > ^ / ^ ' ^ / / J ^ / / -7- TiG. 26. — Section of Delaney's quarry, in Northampton, near the north line of Holyoke, on the Connecticut Eiver Railroad, The base of the trap sheet in this bluff is scoriaceous and filled Avith sheets and filaments of limestone and shale exactly like the surface of the trap a few rods south at the quarry mentioned above, so that I am compelled to assume that a portion of the surface has here been under-rolled to make the base The conditions here are so peculiar that they require detailed discus- sion, which may best begin with a detailed description of the quarry east of the fault, returning then north to the south bluff section, which can be best explained by a comparison with the conditions at the quarry. DELANEYS QUARRY, NEAR THE NORTH LINE OF HOLYOKE. This is a good example of a deeply submerged lava surface onto which much mud was washed while it was still plastic (see fig. 26). Many masses of the mud, varying from thin filaments a few inches long and a small THE POSTEKIOR SHEET, 471 fraction of an int'li thick to broad layers, were washed onto the trap and sank into its mass, so that the upper 3 or 4 feet of the trap is kneaded full of the dark, compact shales, which have at times glazed an', is surrouiuled by a broad layer of fibrous structure, the fibers radiatiuyaud wholly amorphous. MayiictUe is rare and in small grains only. The groundmass is made up of angular and rounded grains 0.001- 0.005™'" across, which can at times be seen to be twins, and they seem to be, in part at least, augite, as they show an extinction at 42°. A specimen from the north edge of the dike has the large feldspars so filled by these minute augites that they occupy the whole space as closely as they do in the surrounding groundmass. Indeed, it appears as if a portion of the groundmass having a regular crystalline outline had been preserved intact from all decomposition, so that the interstices of the grains have not been filled with the fine dust of limonite, kaolin, etc., which renders the rest of the ground clouded. With polarized light the grains are seen to be optic- ally orientated in the feldspar, as they extinguish together, and the feldspar bands can be distinctly seen shining through. The groundmass is for the most part the same in the portions included in the large feldspar crystals as outside, but some inclusions are red-brown and apparently glass. I can not detect with certainty any glass in the groundmass itself The inclusions are plainly from granite : quartz with sheets of pores, some containing mov- ing bubbles, and rutile needles, microcline, centrally decomposed albite with extinction angle 4°, and orthoclase. Fragments of granite with feldspars wholly altered, and of an amphibolite quite fresh and closely resembling the fine-grained rock at the northeast corner of Amherst, also occur. A specimen from the first cleared field north of the brook, externally hke the last, shows both the feldspars and the augite perfectly fresh and colorless, sharply defined, and distantly scattered in the ground. There is so much hematite that it takes up a considerable portion of the surface. One quartz inclusion is surrounded by a colorless radiated fibrous layer, and outside this by a broad band of hematite. The hematite so often sun-ounds the foreign inclusions in a rock otherwise fresh that one is tempted to assume it to have been a cement covering the grains before their envelop- ment in the lava. A great number of bodies are present having exactly the shape of olivine crystals and a bright-yellow or red color. The yellow scarcely polarizes at all, some few fibers or isolated spots showing faint color, and it seems to be a yellow serpentine pseudomorph. The red shows a peculiar 488 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. aggregate or patchy polarization, and seems to be a hematite pseudo- morph after ohvine. Another specimen from the vicinity of the last has a very difiFerent structure. In a granular groundmass (0.01-0.02™°^) there are regularly disseminated, well-formed octahedra of magnetite, visible with a lens in the slide, and abundant diabantite-filled cavities. Small lath-shaped plagioclases and augites are distantly scattered and inconspicuous. The rock resembles that of the dike from the house south of the ruined leather mill below Mount Tom station, on the west of the river. The above descriptions had been written before I received the first accounts of Mr. Diller's discovery of quartz-basalt at the Cinder Cone, in California. On sending him fragments of the rock here described, he wrote that the quartz resembled closely that of the Cinder Cone, and was more abundant. It will be seen from my own descriptions that the idea that the quartz was original in the rock had not occurred to me. It does certainly resem- ble the Cinder Cone quartz very closely, and it is hard to see how a great quantity of foreign sand could be included in an erupted dike, and espe- cially how it could fail to bring with itself moisture enough to make the rock vesicular. The shapes of the grains and the high greasy luster are not like granite-quartz. I have not been able to verify my obser- vation that the slides contain microcline and mica, as the slides are not now a,ccessible, but the presence in the diabase, among many quartz grains, of a large fragment (5"'" across) made up of quartz and orthoclase is certain. The quartz was exactly the same rounded, bluish, greasy quartz as the rest, and the flesh-colored feldspar gave the optical tests of orthoclase, so that I feel quite certain that the unusual constituents have come in as foreign inclusions. The structures produced by the introduction of this large amount of foreign material into the liquid trap resemble those described from the Greenfield bed in Chapter XIII (p. 419), where the sand has risen up into this lava from below. This locality was studied many years ago, before the Grreenfield and Holyoke beds were understood, and I can not say what modification of the above description might come from a new examination of the place with new light. The blue color of the quartz may be due to tension produced by heating, or the quartz may be derived from Algonkian blue-quartz gneiss. (See page 29.) o 488 (I- is iHji as till; .^. SO that I foreign inck The str oreif(n mat« rial Cireenfield b this lava fron b^-^' the Greenfie raodificatiou of the place ' ; ■ '■ ■•^Hiin onrycrft-^ u,j,>; I §ir^of tlie|j j^ 5, (fi.Ol-O.OiJi ^ -6'Li'32'Metite,^i.' imui icmaiue p.-^-cai has a very differ* nt there are regulaily e with a lens in \ tli^haped plagioola ,e.s lie rock resembles i the ill Cone, seen '. \ 5 i the rock y r (joiif verify my obs and m|ica, a| the slides are i.ot the diabiise, ai« ua ict. d in 1. dai of tlie a'' with uvjH' y fei A ^S- A i"^ \ ^ e CM \ ^^ 5 \ 31 ^ ) II V lonjj- iiiaiu ^ ade iiplof ijuliitz and in-tiiocl; $>- '• rounded, blsish, ffrf^asy qua <5^ g:c tlii op^;al \^sts of orthocla 2 &■ ! .•. aistitcftmit^ liav<' oorno iipii of this /large amount oi jiost' deBcribed from <|li iie sandfhas i-isen wv • •;ny|years ago. 1mi,» ,can M(»t ..ix V a Yie^^' •St in f/ 1/ in produced by heat I uartz gneiss. (See pa i ■<-■ m % •m: O tu o 8 > E THE NEWER CORES AND SUORT DIKES. 489 THE TENTH CORE. The w()(mI road that runs north into the mountain west of Moody Corners branches after crossing the eleventh trap mass, and the western branch in a few rods runs out on the sands of the large Glacial lake described below. Here, at a pair of bars giving entrance to the field, near a small brook, begins a long outcrop of trap, which continues 50 rods west, forming the bluff which made the south shore of the lake. The coarse arkose surmounts it on the south, and it is by the downward pitch below the sands of this sandstone on the east and the west that the outcrop of the trap in those directions is limited, while the sands conceal its northern limit. Near its western end, where a stone wall runs across the sands, at the foot of a marked bluff, the sandstone resting on the trap can be seen to be well baked by it, and as the trap is wholly fine-grained and without steam holes it is plainly intrusive. THE ELEVENTH OR BLACK ROCK CORE. Looking southeast from the Mountain House, on the top of Mount Holyoke, one sees a prominent ridge of dark rock running parallel to the mountain — indeed, duplicating it on a smaller scale, repeating its easy southern slope and sharp northward-facing bluff and making with it the great sweeping curve. It differs radically from it in its origin, the larger deposit having been, as I have shown, a bed spread out over the subjacent sandstone, and this an injected dike cutting across the latter. (See PL IX, p. 446.) This bluff, as seen from the mountain, is called "The Black Rock," and I have chosen this name to designate the core, and also the whole series of the newer trap intrusions. Seen from the west side of the river above Smiths Ferry, it simulates exactly a volcano with sharp slopes and central depression. The core is best studied at Batterson's quarry, in the northwest corner of South Hadley, near the last house (E. H. Lyman's) before the town line is reached. As seen in the accompanying view (PI. X), the nearly horizon- tal sandstones are a remnant resting with their edges against the diabase. The latter not only cuts across the sandstone at this point, but sends into it apophyses of finer grain than the main mass, which have altered the sandstones in places for 4 feet from the contact and have fused themselves into firm union with the latter at their junction. The thin-bedded mica- ceous sandstones are delicately plicated by the intruded trap. 490 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. Following' the vertical wall of the diabase north 335 feet from the quarry — a wall which is the contact surface against wliich the sandstones formerly rested — one comes upon a most interesting point, where this wall is continued as sandstone, a fine contact being exposed, and the boundary line of the diabase and sandstone goes into the hill at a right angle. Climb- ing to the top of the bluff, one can follow this contact east, the sandstone at a distance of 4 feet from the diabase being baked into a dark-blue, hornstone-like rock. When the boundary bends round from east to north the thin-fissile sandstones have the imusual position, strike N. 70° W., dip 40° E., being thus thrown off from the eruptive rock. Continuing, the boundary returns westwardly, and thus embraces a great projection of the FiQ. 27. — Section of contact of Black Kock plug and the Mount Holyoke diabase bed. sandstone which extends far into the diabase, and then turns round to the east, parallel to the direction of the older bed. For a long distance one can follow up the bed of Dry Brook I'unning on the back of the older diabase, while its left (south) bank is a vertical wall of sandstone dipping southward and ending abruptly against the diabase of the Black Rock dike, as indicated in fig. 27. At the point where the first outcrop of the sandstone on the brook appears, about 590 feet from the contact in the vertical wall last described, occurs a curious metamorphosed limestone-breccia, with garnet, near the base of the sandstone. This nearness of the two diabase bodies continues, and one goes east a long distance thi-ough a valley with its right or north side THE NEWER COKES AND SHOliT DIKES. 491 till- back of the main diabase bed, rising gradually north to the Holyoke House, its left or south side the vertical wall of the Black Rock dike, its face veneered to a varying. distance upward with the remains of the sandstone. When one comes out where one can look down on the cleared sand flats of the post-Glacial lake mentioned above, one sees that the boundary of the Mount Holyoke bed continues east, while that of the great crater swings round southeast and extends to the deep gorge of the little brook which drains the basin of the lake above mentioned, and has cut deeply through the diabase to enter Elmer Brook, just north of H. White's. The diabase continues to rise high and to carry a thin remnant of the sandstones in con- tact with its vertical face, which sandstone shows contact effects and can often be plainly seen to abut against and not to underlie the volcanic rock, toward which it dips. Wliere the boundary of the diabase runs southeast the sandstone preserved its east-west strike in the main, but in places dips toward the diabase with, the abnormally high angle of 80°. At the west end of the mass the diabase appears in the road at the Lyman house, and its westward extension is concealed by sands. This is also the case with its southern border. The outcrop at the point where Elmers Brook ci'osses the road is so brecciated and its fissures are so filled with druses of small rhombohedra of hematite that it is probably near the southern contact. On following the southern edge along to a point about north of the schoolhouse, where the road to South Hadley starts, it is seen that sand- stones appear on the south of the trap, strike N. 65° E., dip 15° S. — fine- grained, calcareous sandstones, blue-black as if baked or loaded with vol- canic ashes, and rusting slowly inwardly, like the diabase, and between them and the diabase is a band, apparently 10 to 15 feet wide, of the most perfect tuff, made up wholly of angular trap fragments of the size of a pea, with here and there one as large as an acorn, all greatly decomposed. The exact relation of the tuff to the other beds could not be made out. The boundary can be. closely followed eastward to the brook east of White's wood road, to which the northern boundary has already been followed. The exact contact can not be seen, but the dark rusting sand- stones dip south away from the diabase, while the latter rock in the imme- diate proximity to the contact (6 to 8 feet distant) is compact, coarse- grained, and not porous. 492 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. Approaching the brook the boundary bends northeast, as that on the north side had bent southeast, and down the slope to the brook the diabase is amygdaloidal where nearest the sandstone. The boundaries have approached each other so that in the deep side of the narrow brook gorge only a fourth of the width of the great mass appears, but this is well exposed, and a great talus of fragments of a quite coarse diabase makes up the greater portion of its width. The bottom of this narrow gorge is covered with sand. In its opposite east wall one can trace from the south the southward-dipping tuffs, and from the north in fine cliffs the light-buff sandstones with the same dip, to where they approach the eastward continuation of the diabase ; and although the contacts are covered, it can be pretty plainly seen that from both sides the sedimentary rocks abut against the diabase. The latter is very fine-grained and has only a small fraction of the .width it had on the other side of the gorge. This gorge is 1 mile northwest of Moody Corners. Going east it cuts through the tuff, and where this is coarse and both are decomposed it is very difficult to separate them. In one place the dia- base is quite coarse, light-colored, and greenish from the abundance of diabantite, like that just east of White's wood road, and like the rock of the Deerfield bed at the Deerfield Notch. Followed still farther east, where it is crossed by the wood road north from Moody Corners, the diabase is on the north dark, fine-grained, and bounded on the north by sandstones which for a long distance east abut against the high wall of the diabase, as already described. Its boundary against the tuffs on the south is less clear. Where the road crosses, the distinctly columnar diabase rises in a ridge about 35 feet wide, and yet in this is a mass of tuff nearly a meter across, containing fragments of granite. To the south a narrow swamp separates it from a rock which seems to be a coarse volcanic agglom- erate made up of angular fragments often 10 to 16 inches across, which in much-weathered exposures can hardly be distinguished from the normal diabase. PETROGRAPHICAL DESCRIPTION. 1. A section taken from the second outcrop by the roadside going in from Mr. Lyman's house and Batterson's quarry is the typical gray diabase, not distinguishable by the lens from the Iron Gate rock taken as a type above (page 461), and the microscope reveals little distinction between the two, either in structure or stage of decomposition. Pyrite occurs in excep- THE NEWER COUBS AND SHOKT DIKES. 493 tional almiidiuu'd. Tlie feldspars are wholly decomposed and the rock is full of spots of diabautite. 2. In various sections cut from specimens taken at different distances up to 3,300 feet from the edge of the dike no distinction could be observed, but in one taken from very near the center the augites were in large, dis- tinct crystals, very abundant, and plainly anterior to the feldspars. 3. Slides taken from the south edge of the dike, where Elmers Brook finally leaves the trap, showed a large development of the finely granular groundmass (grains 0.005°™) so common in the rock of the tenth dike. 4. Sections were cut from the long, narrow, eastward prolongation of the dike where the Moody Corners wood road crosses it and at its inter- section by the two roads next east. They resemble the type closely. The augite is in the main subsequent to the feldspars, but is a little more dis- tinctly individualized in long blades. Olivine changed to an olive-green serpentine and distinct traces of the unaltered mineral occur sparingly. 5. In sections cut from the edge of the small apophyses sent off by the main mass into the sandstone and exposed in Batterson's quarry, we get additional proof that the larger feldspars are of earlier consolidation. These porphyritic feldspars are of the common size, 1 to 2™™ across, and are asso- ciated with deep-green, well-formed olivines in an extremely fine-grained groundmass, so that it seems that they had already separated out in the magma before its injection into the narrow fissure in the sandstone, in which it cooled so rapidly that the customary ophitic structure was not produced but was replaced by the semicrystalline development described below. The main groundmass is a felted mass of finest fibers 0.0016""" across, quite possibly feldspar microlites, which are not rigidly straight, but wavy, often beaded, and are clearly margarites ; generally, however, they polarize distinctly. These fibers have a radiated arrangement, which gives the whole groundmass a spherulitic structui-e. The fibers polarize sheafwise, although they are not parallel. The presence of olivine in the fresh fine-grained diabase dikes in the granite, and especially in the minute dikes I have described (p. 416), as also its presence in the newer diabase of the volcanic plugs, particularly in that one which has been described as so full of quartz grains (p. 483), may seem, when contrasted with the absence of olivines in the great Deerfield and Holyoke beds, to indicate that the two former occurrences are to be 494 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. associated together as of the same age and contrasted with the two latter and older occurrences; but olivine occurs in considerable abundance in the base of the Holyoke bed at the west foot of Provens Mountain in Westfield. It is strange, however, that where the inclusions of quartz are most abun- dant olivine is also most abundant. NORTHAMPTON. THE TWELFTH OR BURNT MILL CORE. Beneath the railroad bridge over the brook which crosses the railroad a mile above Smiths Ferry the contact of the sandstone dips steeply south- east beneath the postei'ior trap, and this contact can be followed, clearly exposed, along the south side of the brook beneath the road bridge and past the ruined leather mill, and along the south side of the mill pond. The conformable posterior trap bed extends southeast of this line. (See p. 465.) Just north of the brook and below the dam one sees the outcrops of an intrusive dike or plug, which comes up apparently along a transverse fault that can be traced west across the mountain from this point. The trap cuts across the strata of sandstone that underlie the posterior trap bed at every angle, and adjacent to the north end of the dam bakes the sandstone for 6 or 8 feet, so that the line of contact between trap and sandstone is seen with difficxilty. A long ridge of the trap extends west along the northwest side of the pond, and appears also on its south side, and can be followed thence south for 20 rods. At the water's edge trap and sand are confusedly blended. In the bluff above, just south of the head of the pond, appears a well-marked fault. The posterior trap abuts on the sandstone. It is the continuation of the Mount Tom fault, and the newer trap comes up at the intersection of the two faults. A large part of the trap is fresh, compact, and breaks with sharp splintery fracture; a portion of the surface is crumbly and much weathered. This represents a part of the surface of the old laccolith. The trap of the plug may be followed down the brook on its north side to the road, and just east of the road several offshoots from it appear on both sides of the brook. On the north is a 4-foot dike, on the south a 4-inch dike at the water's edge and a 1-foot dike a few feet up. The latter con- tinues under the road to the dam. A few rods farther east one comes upon the most southern of two outcrops of trap at the Lyman railroad crossing, which is doubtless the eastern edge of the plug, since it is com- THE NEWER CORES AND SMALL DIKES. 495 posed of the same type of trap aud contains as inclusions portions of coarsely araygdaloidal trap, derived doubtless from the posterior trap sheet, or the tuff, through which it has been intruded. THE SMITHS FERRY CORE. Directly opposite the Smiths Ferry station and at the edge of the low ten-ace sands just south of the extensive dog kennels the tuff is interrupted by an area of trap, about 6 rods on a side, which seems to be intruded through the tuif, since it has angular masses of the tuff 6 inches across included in its mass. It does not rise above the level of the tuff, as is usual with the more compact plug trap, and I at first considered it a portion of the surface of the posterior sheet exposed where the tuff had been worn tlirough, but the inclusions of trap are foreign to the posterior sheet and the erosion of the old lake shore-line may have lowered the plug at this point. (See, however, page 474.) CORE AT THE ELECTRIC RAILROAD CROSSING OF ROARING BROOK. Where the Holyoke Electric Railroad track leaves the main road to go to Forest Park it crosses Roaring Brook. Just south of this crossing is a surface of trap, exposed by the excavations for the road, which is weath- ered and breaks up into spheres at the surface and rests on the sandstone in the brook just below. This is the south exposure of the posterior trap sheet. Just across the brook the raih'oad cutting exposes a splintery trap of fine grain and perfect freshness, which seems to be the southernmost of the small plugs accompanying the posterior sheet. SUIVOIAEY OF HISTORY OF THE COISTSTECTICTTT RIVEK SAlSTDSTOlSnE. The mountain-making forces which folded up the Appalachian chains acted against the mass of the Archean rocks in the Adirondacks, as seen by the great curve which these chains make as they run southward beyond its influence. The outlines of the Connecticut Basin were laid in pre- Devonian time, since the Bernardston Devonian is bordered by shore con- glomerates which coincide with the borders of the basin and the later Hmits of the Trias. The sinking of the great block south of the Connecticut shore-line, which broke this curve of the Appalachian chains, prepared the way, perhaps, for the second admission of the waters into this narrow channel, which in shape and position resembled the Bay of Fundy. 496 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. The tides of this bay were ou a scale which for a bay of this width have no counterpart at the present time. They passed strongly np the west side and down the east side, and were here reenforced by the prevail- ing west winds, so that they formed very coarse conglomerates on the east. The currents sweeping up the west side, past shores and over bottoms of coarse and deeply decomposed granites, swept granitic debris far north, over the area of black schists. Those passing south swept argillite and quartzite debris far south along the pegmatite shore-line of the east side, and the peg- matite material south along the tonalite and schist terrane, and only where the bay widened was there in its center an elaboration of quartz sand to form the brownstones. In the narrower parts of the bay the two shore con- glomerates meet along a central suture, so to speak, and this ends so abruptly at the north end of the basin as to suggest that it extended much farther north and was perhajDS a strait opening into a larger area to the north. The network of faults which bounds and intersects this basin permitted the sinking of its bottom, in which movement the block from Amherst to Northfield participated in a much less degree than did the rest. In the great transgression which followed, the waters slowly rose upon the bottom and the slopes of the basin and found a great abundance of material ready for transportation and redistribution, because the rocks had become deeply disintegrated during the long period of emergence which was now brought to a close. I have examined both shores of the Bay of Fundy as well as the fiords of Norway and Scotland, and in comparison the work done here seems to me to indicate stronger currents, a larger amount of material, more rapid change of level of the sea, and more rapid deposition, than can be found in any modem examples. The sea seems to have risen over the flats and slopes of deeply softened rock more rapidly than it could remove the material, and therefore advanced without forming a fixed and deep-cut coast line. It often moved the softened d^bi'is in such large quantity to its present resting place that it is scarcely sorted or rounded even when quite coarse gravel. Indeed, the study of this Triassic transgression has thrown more light upon the ancient and more widespread Cambrian transgression (Chapter V) than I have gained from the examination of more modern instances. It is very remarkable how entirely the finest clayey material was SUMMARY OF HISTOKY OP THE TRIASSIO BEDS. 497 \vlioll\- removed troTU the basin during the deposition ol" beds of so great thickness. At a certain time in the midst of this rapid deposition came the great eruptions, apparently synchronous, of the Deerfiekl and Holyoke beds. The bottom of the bay at this time presented a surface covered in (Hfterent parts by beds of every degree of coarseness along its borders and grading toward the center into finer beds, as is indicated by the character of tlie substratum on vs^hich the trap beds rest. Along beneath the trap of the Holyoke range from west of Mount Holyoke to beyond Mount Tom much argillaceous limestone is inclosed in the trap at its base, and in Holyoke, at Ashley's pond, the same limestone occurs in the inclusions at the surface of the same trap sheet. I explain the above structure by the imder-rolling of the surface of the sheet. A limited amoimt of calcareous mud was washed by the strong convection currents onto the submerged surface of the advancing sheet (which was superficially solidified) and blended more or less with this sur- face, which by the continued advance of the mass became in part under- rolled, like the surface of an unrolling carpet, thus protecting the sand below from baking, and bringing the highly vesicular trap loaded with limestone to the base of the bed. A similar structure occurs at the base of the trap sheet east of Greenfield, but not at its surface. This was caused by the frothing up of the muddy bottom into the liquid trap. It is quite probable that these trap sheets were poured out through fissures — the Holyoke sheet through an east-west fissure passing beneath the line of small volcanic cores a mile south of the main outcrop of the main bed, and continuing as a north-south fissure passing tlii-ough the same series of plugs west of the Connecticut. The focus of most intense and long-con- tinued action was about a mile south of the point where the river cuts through the main ridge and where these fissures intersect. The Deei-field fissure can not be exactly located. It was beneath or east of the present outcrop, because artesian wells at Turners Falls cross the trap sheet just east of this outcrop, and there is no trace of intrusive trap west of the present western bluffs. Its focus was probably just east of Greenfield. These sheets produced no disturbance in the distribution of sediments and almost no tuffs, and the arkose which covers them is often buff and nearly free from iron and lime. . The sands which spread over the basin soon covered the great trap MON XXIX 32 498 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. sheets, and then a second period of volcanic activity followed. The lava broke out at a point on the fissure through which the older lava had come, and flowed in a broad sheet down the bottom of the valley. The place where the lava came up is preserved in Little Mountain, east of Mount Tom, and the outcrop of the lava sheet extends north and south from here, and has been described as the posterior sheet. The next episode in the history, following immediately on this outflow of lava, was an explosive eruption of diabase forming the beds of tuff from the Belchertown ponds to Holyoke. The center of eruption seems to have been at the focus mentioned above, not far from Smiths Ferry, since the bombs there are a foot across and decrease in size in both directions. The tuffs produced a marked change in the fucoidaP beds (the Longmeadow sandstone) above, into which they grade, as compared with the older beds beneath the tuff. By shallowing of the water the beds are rendered finer, and they are made deep-red and calcareous from the decomposition of the tuff. These red beds extend south from the lunate band of tuff, but this tuff only accelerated and intensified a process which extended far south beyond its influence, and which had its cause in the width of the basin, its shallowness, and the presence of northward cuiTents along its west side and southward cvirrents along its east side. These cm-rents kept the sides of the basin deeper and made them a seat of coarser sedimentation, and between them was a central area of conflicting and shifting currents— a sort of Sai'gasso Sea, in which the finer fucoidal sandstones were deposited and so frequently exposed by the retreating tides that almost every part shows mud-cracks, rain-drops, tracks, thin films of coal, or some trace of exposure. The northern or Montague basin reached only this stage, the broad development of the central fucoidal sandstones, and there was no tuff outburst or development of shales. In the wider southern or Springfield basin a central area of still greater quiet developed with the widening of the channel, in which at last marly sediments were retained within the area, marked by numerous salt pseudomorphs, and in wliich a later circulation of the waters has concentrated the lime into bands of concre- tionary limestone. ' These paragraphs were written when the rodlike markings in the "fucoidal" sandstones were supposed to be plant remains. I now think them to be concretions, as explained under "Plants," p. 395. There may thus be added to the effects of the impregnation with iron the abundant rod-shaped concretions which have been mistaken for plant remains and called "fucoids." SUMMARY OP HISTORY OF THE TRIASSIO BEDS. 499 The beds coutaiuing' reptile tracks are almost without exception above the great trap sheets, and in most cases not very far above them vertically. Some of the localities situated far to the east have been brought up by faulting. These central exposed mud flats seem to have been caused by shallowing of the waters, which resulted from the flowing of the great sheet out over the bottom. The present dips are the result of three actions difficult to separate: (1) Deposition upon an inclined plane, especially that between the central shallower portion and the deeper portion on the border. This seems to be the case across Hatfield and Deerfield, on the western side of the basin, where the finer central beds dip slightly west toward the coarser beds near the shore, and across South Hadley, Springfield, and farther south, where the finer central beds have a low dip eastward toward the shore beds. In these cases the beds have been moved but little since their deposition. (2) A slight excess of sinking on the eastern side or an increment in the strength of the eastern currents, or both, by which the finer central beds were in their eastern portion encroached upon and covered by a broad transgression of the eastern conglomerates, so that all down the east side the fine-grained beds dip normally beneath the coarse. (3) Later tilting, largely to the east, but bending to the south in the Holyoke range, and generally of the monoclinal type, the important excep- tion being at the mouth of Millers River, where there is a great syncline whose axis pitches sharply a few degrees south of west. By this later monoclinal tilting the covered bed of fucoidal sandstone is brought up several times in the mass of Mount Toby. A third period of volcanic activity occurred in the southern basin about the time of the close of sedimentation and the final tilting of the sandstones. Nearly a score of volcanoes formed a chain running from the Belchertown ponds first west to the Connecticut River and then south to Holyoke, apparently caused by the reopening of weak points along the great fissures which had supplied the material of the earher sheet. One of these shows indications of having been a laccolith sending out a long fissure-filling in the sandstone. Another is diabase filled with granitic inclusions. The rest are small plugs. The final tilting was much more severe in the northern portion of the 500 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. basin, and has here a large component of north-south motion, and thus of east-west strikes. The limit of this severe action is the east-west fault mai-ked by the line of craters just mentioned. Many baryta-lead veins in the sandstone and the crystalline rocks on its borders seem to have been formed at this period. In his review of the question whether the rocks of the Connecticut Valley were deposited in a separate basin, as has been maintained in the preceding pages, or were connected across western Massachusetts and Connecticut and eastern New York with the New Jersey Triassic, which culminates in the Palisade range. Professor RusselP still maintains the opinion he had advanced in earlier papers. The localities described in the preceding pages where coarse conglomerates and coarse, unworn arkose beds rest on the crystalline rocks along the western border, and the relation of these beds to the great granite areas directly west, make it quite certain that the upper end of the bay in Massachusetts had its western shore- line quite exactly at the present western border of the Triassic beds. These latter are mainly formed from bottom to top of the coarse ddbris of mus- covite-granites such as now form their western border, while if they had transgressed but a few miles westward they would have covered entirely all this coarse granite area, and there is no similar area from which the arkose could have been derived farther west, where the shore-line must have been. THE USE OF THE TRAP AS ROAD MATERIAL. It is well known that the trap furnishes the best material for road making, and as the legislature of Massachusetts has wisely entered upon an extensive and carefully arranged scheme looking to the extension of macadamized roads throughout the State those places where the trap is found in large quantity and of good quality and near to railroads will be of economic value. The city of Springfield has for a long time worked a quarry at the point in Westfield where the Boston and Albany Railroad crosses the main or Holyoke trap sheet. Recently (1895) the Massachu- setts Stone Crushing Company has established an extensive plant on the south side of the Deerfield River at Cheapside, with a capacity of about 100 tons per diem. The company has spur tracks to the Turners Falls branch of the Canal Railroad and can distribute its product readily by rail ^ Correlation Papers, The Newark System, Bull. U. S. Geol. Survey No. 85, 1892, p. 101. TRAP AS ROAD MATERIAL. 501 fi'om Nortliampton or Greeniield. This work is on the main Deerfield tru}) sheet. The New England Trap Rock Company, of Westfiekl, has opened another laa-ge quarry on the west line of West Springfield, on the Ilolyoke- Westfiekl road, with a spur track to the railroad which connects these towns. It has two large crushers with a capacity of 700 tons per diem, and can deliver the rock on the cars at 65 to 75 cents per ton. A crusher has also been set up at the west foot of Mount Tom, in Easthamp- ton, which supplies this town with road material. It is directly under the most picturesque portion of the palisaded bluff, and although it is at present working in the trap talus, later the operations may seriously mar the north wall of the mountain. The great dike beside the spur track of the quarries of W. N. Flynt & Co., in Monson, has been opened by that company, and will supply material of the very best quality. The city of Northampton has for several years established its cruslaing works near the north line of the town and at a distance from railroads, and works the Hatfield tonalite or hornblendic granite, which is a partly decom- posed rock, more brittle than the trap, and in many ways an inferior rock for road ballast. The city has access beside the road or railroad within its own limits to several better ledges than the one it works now. In 1897 the city of Ware opened a quarry on the Coys Hill dike on the mountain side east of the railroad station. The dike is here 5 rods wide, favorably situated for quanying, and will furnish the best material in inexhaustible quantity. The other localities where the rock occurs in good quantity and qiiality and convenient of access to the railroad are : Where the Fitchburg Railroad crosses the Deerfield trap sheet, on the north side of the Deerfield River and directly opposite to the present works; along the Connecticut River Railroad below Mount Tom station, especially at the first crossing, and at Delaney's quarry on the north line of Holyoke; at Tatham Cutting, in West Springfield; and finally, all the later volcanic cores marked on the map, and especially the dikes of the trap marked in the crystalline rocks on the east of the Triassic area, furnish a rock more fresh and firm than the trap of the main sheet. . West of the Triassic no beds of trap are found and the hornblende- schist of Chester or the Becket gneiss will be the best substitute for local use. CHAPTER XIV. MINERAL VEINS. The only mineral veins in the area are of the "baryta-lead formation," though in some of the fissures occupied by these veins there seems to have been an antecedent "fluorspar-calcite formation." Many of these veins seem to have been first filled with fluorspar and calcite and various ores. These are now scarcaly represented except by the many pseudomorphs of quartz after fluor and calcite. The circulating waters bearing silica first dissolved out or replaced the fluor and calcite. This forms the beginning of the second stage of vein filling, and the veins soon became quartz-barite-galena deposits, with chalco23yrite and sphalerite at times replacing the galena. It is quite possible that the fluorspar-calcite formation dates from the time of the post-Carboniferous folding, and entirely probable that the barj^ta-lead veins coincided with the folding of the Triassic rocks, since they occur both in the Triassic sandstones and in the older rocks. All the minerals which occur in the veins mentioned above are described in detail in the author's Mineralogical Lexicon of the three counties^ and in the supplement to the same in Chapter XXII of this monograph. The other beds in which mining is done — the emery bed and the pyrite and hematite beds in the Hawley schist — are in the main contemporaneous ' beds, interstratified with the schists which contain them, and the workable ores were either originally present, as in the emery bed, or were formed largely by replacement of other beds, so that it has been more natural to discuss them in connection with their coimtry rock. Westliampton ; the Londville vein. — On July 27, 1679, the little plan- tation of Nonotuck, now Northampton, held a town meeting and voted, "after much discourse and agitation," that the town have a general interest ' Bull. U. S. Geol. Survey No. 126, 1895. 502 MINERAL VEINS. 503 in a k'ud iniue newly discovered within its limits. There is but one other entry on this subject in the early records of the town, viz : At a legal meeting Oct. 16, 1679, they then having further Conference about the lead mine which Robert Lyman found out, they then voted that all such persons as would Join in the Carrying on of that design. Should meet on the 23d of this instant at Sun one hour high at night, and to them or to those persons that shall then appear the Town do hereby give up all their right in that mine lying about six miles otf, at the west side of the Town. It can not be learned what came of this vote, but bullets were cast from lead smelted here during the Revolution.^ The shaft was opened before 1769, and again in October, 1809. It was reported upon by Ben- jamin Silliman in October, 1810, and the report was printed as an article in Brace's American Mineralogical Journal.^ The shaft was then 60 feet deep, the adit 25 to 30 feet, and the vein was "a very magnificent one, 6 to 8 feet in diameter." In 1815 the adit was 726 feet; the shaft entering 500 feet from the mouth was 90 feet deep.^ In 1818 Amos Eaton described the rocks of the adit carefully. It was then 800 feet deep, 666 feet in sand- stone, 134 feet in granite-schist and serpentine, containing veins carrying quartz, fluor, calcite, chalcopyrite, and one small vein of galena.* In 1823 the adit was 990 feet long and had cost $20,000." In 1827 druses containing more or less calcite crystallized among the crystals of quartz had occurred in the last 200 to 300 feet of the adit^ and a company opened a new mine with a drift on what was supposed to be the same vein 3 or 4 miles southwest; the vein being 6 inches to a foot wide. In the next year the vein was opened one-half mile north.'' In 1832 Presi- dent Hitchcock mentions with apparent regret that work had been stopped on the adit at 900 feet, largely because the pi-ice of lead had decreased greatly, from western competition, and expresses the belief that the vein would have been struck in a few feet. The mine was opened again in 1855 "with prospect of success."^ The mine was again opened in about 1862, and I remember visiting it ' Evert's History of the Connecticut Valley in Massachusetts, Vol. I, 1879, p. 17. ^Ibid., Vol. I, p.63. ^E. Hitchcock: North American Eeview, Vol. I, p. 335. "Amos Eaton: Am. Jour. Sci., 1st series, Vol. I, p. 137. '■E. Hitchcock : Am. Jour. Sci., 1st series, Vol. VI, p. 201. "A. Nash, The lead mines of Hampshire County: Am. Jour. Sci., Ist series, Vol. XII, p. 258. 'E. Hitchcock: Am. Jour. Sci., 1st series, Vol. XIII, p. 218. 8 E. Emmons, American Geology, p. 183. 504 GEOLOGY OP OLD HAMPSHIEB COUNTY, MASS. then, in my freshman year, and rowing the length of the adit. The Manhan Silver-Lead Company was formed, which erected extensive buildings and installed expensive machinery. I" have been informed that the enterprise failed because of the fall in the price of lead at the close of the war, and that the machinery, costing about $60,000, was sold to the Chester Emery Company for about $17,000. The vein produced lead with about 12 J ounces of silver per ton from galena. Sphalerite, chalcopyrite, pyrite, and bornite occurred more rarely; barite and quartz in abundant ciystals was the gangue. As decomposition products, malachite appeared with wulfenite, cerussite, and anglesite, and the finest pyromorphite occurred. Pseudomorphs after calcite and fluorite *j indicated the former more abundant presence of these gangue minerals. y'^ — , The most interesting article that has been published on the lead veins of Hampshire was by a wholly self-taught man, Mr A. Nash, and this seems to ha;^e been his only essay in authorship. Professor Shepard, who then did editorial work on the American Journal of Science, told me that it took much editing to make the paper intelligible. Much of what follows comes from that paper.^ Whatehj. — This vein is in the southwest part of the town, on the summit of a high mountain of granite. The vein is 3 to 4 feet wide; considerable galena occurs in a quartz g-angue; the range and vein strike northeast. (Nash.) I have searched for this without success. WJiately. — In the northwest part of the town. The vein runs north and south. It has been traced 100 yards to the edge of Conway. The ends of the vein are in mica-schist; the middle is in granite; 6 to 7 feet wide. The gangue is quartz; the ore, galena only. (Hitchcock.^) Shows graphitic slickensides; crushed veins with quartz, calcite, and green fluor. Conway. — Southeast part, 3 miles from meetinghouse, and southeast of the manganese vein. It contains quartz and galena. (Nash.) Maybe the same as the last. ' Chesterfield. — A copper mine is put down on Nash's map east of the Lily Pond Brook, but not mentioned in the text. Goshen. — Sixty rods east of Congregational meetinghouse; galena in crystallized masses of quartz on the ground; no vein seen. (Nash.) I Am. Jour. Sci., 1st series, Vol. XII, p. 238; map. ''Am. Jour. Sci., Ist series, Vol. VI, p. 204. MINEEAL VEINS. 505 Williamsburg. — Northeast part; vein not seen; larg-e blocks of (quartz occur in great profusion in a range several rods wide and one-quarter of a mile long; the quartz is radiated and rich in galena and chalcopyrite. (Nash.) WiUiamshurg. — Vein runs northwest, then north, and then northeasterly into Whately; quartz partly green and amethystine with pyrolusite and galena, which increases northerly. (Nash.) WiUiamshurg. — Extending into Whately one-half mile east of the above. Contains galena and pyrolusite in quartz. (Nash.) Williamsburg. — ^At northwest corner of Northampton, near the argentite locality (see Mineralogical Lexicon,^ under "Calcite"). It contains pseudo- moi-phs of calcite and fluorspar; the vein extends down the brook one-third of a mile on the east side. (Nash.) SMburm.—^orih. of J. Dole's, 1 mile west of Shelburne Center, at southeast border of gneiss on contact of hornblende-schist and mica-schist; vein 2 feet wide, containing pyrite, galena, blende, malachite; runs N. 25° E., dip 40° E. Greenfield. — At junction of diabase and upper sandstone, on the west bank of the Connecticut, 100 rods below the mouth of Fall River. It goes north obliquely into the diabase and south across the sandstone in the river bed. The principal vein is 5 to 6 feet wide. It strikes north-south; dips 90°; malachite is common, the sulphuret is rare. There is a second vein about a mile below, and narrow veins with fine slickensides occur in other places between. (E. Hitchcock.^) Turners Falls.— West side of the island at the falls; strike north-south; dip 90° ; produced fine large masses of chalcopyrite and much siderite ; is in brecciated sandstone. Hatfield. — Vein appears in the bluff of tonalite about 2 miles west of the town, 60 rods north of the road to Williamsburg. It can be traced N. 60° W. for about 30 rods. A slanting shaft has been sunk from the base of the bluff; the vein is 1 foot at surface and 3 feet at bottom. Farther west the vein has been opened about 20 feet deep ; it is here 4 feet wide at surface and 8 feet at bottom. Back from the vein the tonalite seems very fresh, but under the microscope its feldspar is always much kaolinized. ' Bull. U. S. Geol. Survey No. 126, 189.5. = Am. Jour. Sol., 1st series, Vol. VI, p. 207. 506 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. For several feet on the border and in the numerous " horses " it is changed to a plainly kaolinized white mica-granite, most or all the hornblende being removed. It contains rarely green fluor and calcite, R',-^R, weathered to dull gray and both in every stage of change to hollow pseudomorphs of quartz. The main filling of the vein — following the above calcite-fluor stage — was quartz, itself covered by barite, which so abuts against the quartz with its prismatic faces that the latter seems to be the newer mineral, which is rarely the case. Galena, blende, pyrite, chalcopyrite are the ores. A second generation of calcite, R, R^, occurs in the quartz. Cerussite, mala- chite, pyromorphite, limonite, and pyrolusite are the decomposition products. Leverett. — One mil6 northwest of the meetinghouse, on land of Mr. Field, once considerably worked, but abandoned on account of its unprom- ising appearance. (Nash, 1827.) Was worked by a company organized in New York a few years ago, but did not pay.-' Strike north-south; dip 90°. (E. Hitchcock.) The vein is in mica-schist and granite. It is several feet wide, and contains galena, chalcopyrite in masses of the size of one's fist, blende in the best crystals obtained from any of the veins, and pyrophyl- lite. The gangue is baryta. Hollow quartz pseudomorphs after pyro- phyllite occur. Leverett. — South line, "White Rock quarry." Only few inches wide at surface, but widening below. Galena and chalcopyrite abundant at sur- face, but rare below ; worked but few feet down, there 1 foot wide ; nearly pure barite. (Nash.) Later a long adit was driven in, but caved many years ago. Leverett. — Cut south of railroad crossing next east of last mine; narrow veins of barite, with little galena. Northampton. — At the quarry east of Florence, in Northampton (south of W. N. Moore's house), the biotite-muscovite-granite is cut by joints run- ning N. 60° E. and dipping 60° N. These joints are about a foot apart and in this and in the next quarry to the east are often marked by fine slicken- sided surfaces. Between two of these fault planes a sheet of the granite is finely crushed and the parts recemented, producing a great crush fault which runs beyond the limits of the quarry in both directions. The fissures thus produced were occupied first by calcite, which is now present only in a few crystals coated with transparent cubes of fluor, but is further represented by negative crystals in barite and quartz. Barite followed the ' Evert's History of the Connecticut Valley in Massachusetts, Vol. II, 1879, p. 73.3. MINERAL VEINS. 507 cuk'ite and sliot out through nil the cavities in broad plates of extreme thinness. This Avas followed by an abundant deposition of quartz, both as drusy surfaces and as pseudomorphs after calcite, and by barite. There is idso an abundance of a chocolate-colored tabular quartz, slashed full of fissiu-es from which the blades of barite have disappeared, which is a most perfect pseudomorph after the peculiar tabular form of calcite called argen- tine, which occurs also on the other border of the great granite area. The quartz is followed by prelinite in broad surfaces of large crystals, simple or slightly rosetted. The prehnite was followed by laumontite in fine large crystals possessing the wholly peculiar form characteristic of this mineral, but now represented only by hollow incrustation pseudomorphs in albite, which latter appear as minute, limpid, very characteristic twins. The whole forms thus a very peculiar but very clearly observed pai-agenesis. Bussell. — Mineral veins appear in the northwest part of Russell, show- ing drusy quartz and galena.' Specimens are deposited in the Massachusetts State Survey collection made by E. Hitchcock. Huntington. — Angel's mine, Norwich, now Huntington. Showing blende in large masses and a beautiful drusy quartz pseudomorph after barite and calcite,^ according to the specimens in the survey collection. iCat. Agr. Museum, 1859. Rept. Agriculture Mass., Appendix, p. LXIX, No. XIX, 202, 203. « Loo. cit., 200, 201, 204-211. CHAPTER XV. THE PLEISTOCENE PERIOD. LITERATURE. 1818. B. Hitchcock. Geology of Deerfield. Am. Jour. Sci., 1st series, Vol. II, p. 107. 1823. , Geology of Connecticut River. Ibid., Vol. VI, p. 80. 1827. A. If ash. Lead Mipes, etc., of Hampshire County. Ibid., Vol. XII, p. 248, 1833. E. Hitchcock. Geology of Massachusetts, pp. 33, 135. 1835. . Geology of Massachusetts, 2d edition, p. 174. 1841. . Geology of Massachusetts, Final Eeport, i°, pp. 306, 332, 357. 1850. . Proc. Am. Assoc. Adv. Sci., Vol. Ill, p. 155. 1852. . Ibid., Vol. VI, p. 264. I860. . Illustrations of Surface Geology. Smithsonian Contributions to Knowledge, Vol. IX, pp. 1-155. Also issued separately. 1863. . Reminiscences of Amherst College, pp. 260, 311. 1871. J. D. Dana. On the Connecticut River Valley Glacier. Am. Jour. Sci., ' 3d series. Vol. II, p. 233. Vol. V^1873, pp. 198, 217. Vol. X, 1875, pp. 180, 280, 353, 497. Vol. XII, 1876, p. 125. Vol. XXIII, 1882, p. 87. Vol. XXV, 1883, p. 440. Also published separately. 1877. J. S. Diller. Westfleld during the Champlain Period. Am. Jour. Sci., 3d series, Vol. XIII, p. 262. 1877. . Westfleld Times and News Letter, Vol. X XX VII, March 28, Sep- tember 19. THE INTERVAL BETWEEN THE TRIASSIC AND THE GLACIAL PERIOD. DEPOSITS. Within tlie area here under survey the materials for a reconstruction of the history of the later Mesozoic and the Tertiary are extremely scanty. With the exception of a single trap talus beneath the lower till there are no known deposits left to represent these long ages. The excavations for the Turners Falls branch of the Canal Railroad were earned along the south side of the Deerfield River where the latter passes through the notch in the Deei-field range to reach the Connecticut, and exposed at a height of 508 PKE-GLACIAL CONDITIONS. 5()9 50 feet above the river a great talus of trap fragments — a pre-Glacial "Devils Garden," as these desolate slopes were called by the fathers — resting against the vertical wall of trap, which here rises about 100 feet above the level of the stream. The talus was exposed for a length of 90 feet and for a height of 30 feet, and it apparently extends down to the level of the river, 50 feet below, but this was not observed. Covering this talus and extending up over the trap was a layer of very compact till, 30 feet thick, of reddish color, made up mostly of sandstone with few bowlders of mica-schist from the western hills and with none of trap. At least nine- tenths of the bowlders, down to those not above 2 inches on a side, were finely striated — a quite unusual proportion. A fresh vertical section of this till produced by caving was marked for a long distance by wavy lines of apparent bedding so perfect that at a dis- tance I had supposed the beds to be the thin-laminated Champlain clays, but the lamination proved to be an unusually perfect pressure cleavage in the till, in planes dipping 60° to 70° NW., at right angles to the direction from which the ice was moving in the canyon, as marked by the striae upon the trap immediately above. These data prove that the ice breasting the long westward-facing vertical wall of the Deei-field trap range was pressed into this notch in the range with exceptional force, from which we may deduce that the prevalent southward motion of the ice in the valley was due to its deflection from the normal northeast direction by the north- south walls of the valley and of the divide ranges. A further interesting deduction is that the notch of the Deerfield range is, in its present form, of pre-Glacial origin, and since the river flows through without exposing rock at bottom the gorge was then of even greater depth than at present. The Deerfield Indians aflirmed that it was begun by a squaw with a clam shell. One other deposit, probably of Tertiary age, is described with the "Camp Meeting cutting," near end of Chapter XIX. It is a thoroughly sorted, pink beach sand, and it appears below the glacial beds on the north line of Northampton. PRE-GLACIAL WEATHERING. The only important case of the preservation of any portion of the deeply decomposed surface rocks which must have characterized the country before the advent of the ice, as they are now characteristic of 510 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. non-glaciated countries, is the great bed of kaolin preserved from the ice under the lee of the great hill on which Blandford is built (see p. 330). Another area is northwest of Roaring Brook bridge, on the south line of Northampton (p. 474). PRE-GLACIAL DRAINAGE AND EROSION. The above deduction concerning the age of the Deerfield notch may serve as an introduction to the discussion of the other similar notches in the valley and of its pre-Glacial drainage and erosion. (See map, PI. XI.) Not only does the Connecticut pass tlii'ough a like notch in the Holyoke trap range near its highest point in a deep, short valley bordered by fine rock-cut terraces (fig. 28), while it could have passed down the western lateral valley (see topography of the valley, p. 8) without rising more than 145 feet above its present height, but the Westfield and Farmington rivers also, like the Deerfield, after passing out of their gorges in the crystalline rocks, run across the low sand plains of the western lateral valley, make a Fig. 28.— Holyoke notch from Hadley meadow ; pre-Glacial rock terraces. wide loop southward, and return to find in each case opposite the mouths of these gorges a notch in the high trap ridge through which they join the main stream, while in each case they could with a sHght rise have passed southwardly across the sand plains, the Deerfield to join the Connecticut around the south of Sugar Loaf, the others to reach the Sound at New Haven. Indeed, this peculiarity of the valley system of the Connecticut early attracted the attention of President Hitchcock, who, after having described it with a sketch map in the Geology of the Connecticut,^ writes in the Greology of Massachusetts (1841, p. 328): The valleys through which the Connecticut and its tributaries flow are among the most remarkable in the State. The ordinary laws of physical geography seem here to be set at defiance, so much that a late ingenious writer doubted whether I had correctly represented the geology of the Connecticut because the course of the rivers and the direction of the mountain ridges were described as having so little correspondence with the rock formations. ' Am. Jour. Sci., Ist series, Vol. VI, 1823, p. 1. 11 (A Z o ll. o U) z (0 H _l (0 u Z S o D Q. Q. CC u U o o Z L. Q 1- o z 1- (/) < < X m LJ < o a. o LJ < m u z o X 4 -1 u o c < GC LI (0 z h 3 X Li Z L. i "i Ul V H z Li z 1- -J - _l Z Q. z < H 1- < ^ > .• 3 o - O i Q. < o w LJ D O <8 Li CC 5 Ll ^ z < o LJ z z o Si T *^ 1- z ffi (L 1 fr Ll O Q o O L. W h U LJ LJ > I h z z a a. a. bj (/) iij < LJ a: (T > Q Q. LJ o o (/) LJ Li 1- ^ W z CC QC u < => (0 g O LJ o CC 0. LJ 0) (T hi o 1- CC z h ^ CC -1 LJ m 3 < s O D oc < Z < u LJ u LI z z 0. H 1- PUE GLACIAL DRAINAGE AND EROSION. 511 President Hitchcock draws the inference from the facts detailed above "that the Connectictit River did not excavate its own bed through these mountains, for liad the bamer at Northampton been more than 134 feet above its present bed it must have emptied into the Sound at New Haven. We nuist seek some other cause, therefore, for the origin of the passage between Holyoke and Mount Tom." I draw the opposite conclusion, and belie\e the history of the erosion of the valley to have been as follows: The streams occupied their present valleys in the crystalline rocks before and during the Triassic (at levels, of course, much higher than the present), and entered the Triassic estuary near where their gorges now end at the border of the sandstone. On the recession of the waters the Connecticut followed down the deepest line in the middle of the long bay and the tributaries took a dii-ect course down the slope to this line of greatest depth to join the main river. I imagine that the dislocation of the sandstones took place after this drainage was established, and so slowly that the streams were not seriously disturbed, but cut down through the sandstones till they reached the trap sheets, and then through these until the four gorges were carved. Many facts point to the conclusion that these valleys were cut much deeper than the present bed of the river, and down, indeed, to or below the present level of the sea. Piles driven in clay at the Northampton bridge went 10 feet below sea level. The Belden artesian well, south of the North- ampton station, struck rock 25 feet below sea level, and soundings showed the clays to have great depth beneath the main street crossing; these may represent an old course of the Mill River. Borings of the United States survey of the Connecticut River between Chicopee and Longmeadow were can-ied to points 19 and 21 feet above sea level and 43 feet below without meeting rock, and 1^ feet above sea level striking rock.-' In each of the four gorges here specially under discussion no rock appears in the stream beds. All the points cited above lie along the old channel, at places specially sheltered from glacial erosion. From this one may conclude, in passing, that the falls along the Con- necticut are located in portions of its course which do not coincide with this ancient one. •T. G. Ellis, Report of survey of Connecticut River: Ex. Doc. 101, Forty-fifth Congress, second session, 1878, p. 122. 512 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. Parallel with this deepening of the stream beds the unequal erosion of the whole area wore back the banks of the streams in easy slopes where they crossed the soft sandstones, brought out the trap ridges in sharp relief, and left thus the short canyons or notches through these ridges. Thus on the approach of the Glacial period a surface had been reached which coincided much more closely with the present cultivated surface of the valley than with its present rocky floor, except along the western lateral valley — that is, the broad depression west of the trap ranges — and north of the Holyoke range, where the surface of the sandstones was then probably higher than the present surface of the later sands. Then came the Glacial ice, destroying the whole drainage system and removing an enormous quantity of the soft sandstone. Its work was favored and localized in two Avays by the position of the trap ranges running down the middle of the valley. Where these ran north and south with bold westward-facing bluffs, as in the Deerfield and Mount Tom ranges, the ice coming from the north- west was deflected southward and scoured out the soft sandstone at the western foot of these ranges, and where the Holyoke range runs clear athwart the valley east and west the ice, by its recoil as it lifted over the range, plowed out the sandstone all along its northern base down to a level much lower than could have been well effected by ordinary aqueous erosion- Thus the river channels between the ends of the ravines in the crystal- line rocks and the notches in the trap ridges were obliterated because they were contained in the comparatively soft sandstone, and we have finally to seek the reason why the streams, upon the decrease of the floods which accompanied the retreat of the ice, in every case found their way again through their old notches instead of taking the more direct and natural course down the deep western lateral valley, from which the ice had removed the sandstone to so low a level. The broad river occupied then almost precisely the boundaries of the earlier Triassic estuary, and the tributaries entered it along the border of the western highlands. Across Massachusetts the great river was, indeed, rather a series of lakes than a river, in that it was filled mainly from the sides by the great confluent deltas of its tributaries, which were pushed out to a distance proportionate to the importance of the stream that furnished each, while down its center clays and fine sands were deposited in less thickness. Thus it came about that the great body of sand delivered to the main stream by each tributary was spread diagonally across between the western hills and the divide ranges, rKli-GLAOIAL DRAINAGE AND EROSION. 513 and in somevvhut larger proportions in the area just below the mouth of the tributary as a resultant of the transporting power of the main stream and the tributary, so that upon the lowering of the waters of the mam stream and theii' retreat from the western lateral valley each tributary found its way southward dammed up by its own delta deposits, and, ponding back behind them, flowed again through its old notch to join the diminished waters of the Comiecticut. The heavy sands which fill up the lateral valley below each of these tributaries, from the western border across to the divide range, do in fact show, both by the derivation of their material and by their structure, that they are the ancient deltas of these streams. The thread of the current of the main stream, driven clear across toward the eastern foot of the divide range by the great delta of the Millers and Chicopee rivers, had continued to pass through (a) the narrow passage between Deei-field Mountain and Mount Toby and (h) the Holyoke notch, two portions of its old channel, partly, perhaps, because these lay in the main artery of the pre-Glacial drainage, but more because they were out in the center of the lake, far from all lateral streams and their deposits, and on the recession of the waters the western or lateral valley was filled up to such a height by the Westfield River that the Connecticut was compelled to shrink down to this line and reoccupy its old notch in the Holyoke range. THE PRE-GLACIAL COURSE OF THE CONNECTICUT AND ITS TRIBUTARIES. The pre-Glacial bed of the Connecticut across Massachusetts lay below the present sea level. (See map, PI. XI, p. 510.) Hence, where the river passes over rocky bottonis with rapids and waterfalls it has been expelled from its pre-Glacial bed by Glacial and Champlain deposits. In each case the old bed of the stream is marked by a broad band of depres- sions in the high terrace sands — kettleholes — partly empty and partly water-filled. The ice seems to have persisted in the deep channel until it was covered by the flood sands and then to have melted to form the depressions. This is most marked south of Millers Falls across the Mon- tague plain, in the great' loop of Millers River and the succession of ponds extending southward, of which Lake Pleasant (its bottom about 67 feet below the plain) is the largest. From the State line to Northfield farms the river has regained its old bed. South of this point the great delta of Millers River crowded it 6 miles west to the foot of the trap ridge and MON xxis 33 6i4 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. compelled it to cut in tlie sandstone the French King gorge^ and the canyon of the Lily Pond. The river regains its old bed in the passage between Mount Toby and Sugar Loaf, where its fine pre-Glacial rock-cut terraces which flank Mount Toby have received the name Sunderland Park, and its course across the Hatfield-Northampton meadows is closely given on PL XI, p. 510. After its passage through the Holyoke notch its course is uncertain, and there I give two alternatives on the map. The reasons favoring the eastern course are that it lies along a line of deep depressions in the broad sand plain, and shows no rock outcrops where erosion has gone deepest. The reasons favoring the other course are that it passes over the borings of the Grovernment surveys of the Connecticut River, one of which went 30 to 40 feet below sea level, and, like the other, is a line where the deepest erosion discloses no rock. (See "The Spring- field Lake," in Chapter XIX. It is certain that between the Holyoke notch and the latitude of Springfield the river ran far east of its present course, because it now cuts tlu'ough rock all the way from Mount Holyoke to the Holyoke Falls. The justification for the course given the Deerfield River and the Westfield River has been presented above. An inspection of the map will suggest that the Deerfield River may have run southeast from its ravine through the finely rock-terraced notch between Sugar Loaf and North Sugar Loaf, and that its present notch in this range may have been cut by the Grreen River, but the drawing on the map represents the most pi'obable status. The Sugar Loaf notch is not deep enough for the Deerfield River, which probably ran south of Sugar Loaf. The notch which separates Sugar Loaf from North Sugar Loaf is plainly water-formed, and Whately Glen is its most probable upstream continuation among the crystalline rocks, as indicated on the map. In the same way the main gap in the center of the Holyoke range, to which the name "Notch" is especially I'estricted in the valley, was the result of water erosion and was the site of a great waterfall before the Glacial period. It is deeply cut in the trap, with vertical walls, and its continuation in the sandstone immediately south of the trap sinks very suddenly to a much lower level, forming the Orchid Garden, celebrated among botanists. I think this notch was in continuation of the "Freshman ' See footnote on pnge 296. PRE-GLACIAL EROSION. 515 River." It was temporarily reoccupieil duriiii^' the recession of the ice, recei\iu<>- the overflow of" a Glacial lake which formed ou the north flank of the Holyoke range, banked on the north by the ice of the Hadley basin. There are two striking gorges in tlie west of the town of Holyoke, both cutting the trap very obliquely, one occupied by Wrights Brook (which enters Hitchcocks Pond), while the Westfield and Holyoke Rail- road passes through the other. These gorges seem to be portions of the bed of a stream that gathered on the east flank of Mount Tom and ran south into the Westfield River. Another notch of unknown depth cuts the trap ridge just where it crosses the State line into Connecticut. This I have connected with the large brook which comes down from Sodom Mountain, in Granville, and have called it on the map the Southwick notch. Though the evidence is much less clear, it seems probable that the narrow canyon skirting the east front of Mount Toby was cut by Locks Brook. Its bottom has now the shape of an abandoned water channel. It is probable that the portion of the channel of Locks Brook which ran in sandstone between the end of its gorge in the crystalline rocks and the beginning of the canyon was removed b)^ ice erosion. At the end of the Glacial period the ice, halting in the Montague basin, deflected the brook again southward into this canyon. THE CHARACTER AND AMOUNT OF EROSION DURING LATER MESOZOIC TIME AS COMPARED WITH THAT OF THE GLACIAL PERIOD. From the preceding discussions of the crystalline rocks and the Tri- assic sandstones it is certain that the broad Connecticut Valley was an orographic feature of first importance formed in the crystalline rocks before the deposition of the sandstones, its borders coinciding closely with the present boundaries of the latter. Prof W. M. Davis^ has suggested that there may have been a pre-Triassic penei^lain over this area. The places where the crystalline rocks break through the Trias are at such different levels in places very near one another that this is not probable. This valley was then deeply filled by the sands of the Trias, indeed above and beyond the present lips of the basin, and has been since so thor- oughly eroded a second time that only remnants of this filling remain. It seems quite certain, that the walls of the valley during and at the close 1 Bull. Geol. Soc. America, Vol. II, p, 549. .Jour. Geol., Vol. IV, p. 678. 516 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. of the Trias were much higher than now or that the sandstones once extended much farther east and west. Mount Toby, wholly made up of sandstone and conglomerate, is 1,275 feet above sea level (1,170 feet above the river), but to obtain its true height as a measure of the height of the Triassic at its maximum we must add a considerable but unknown amount for subaerial and glacial erosion. It is possible that we must subtract something also to offset the elevation of the mass during the disturbances which have tilted the rocks. I imagine it would be an overestimate of the latter movement if we should assume it to have been great enough to counterbalance the depression of the old sur- face by erosion. If, then, we take the present height of Mount Toby or Mount Tom as that of the sandstone at the close of its deposition and run a contour line at this level along the sides of the valley to obtain the orig- inal boundaries of the sandstones, on the assumption that the valley walls were then about their present height, this line would lie so far back from the present border of the sandstone and run into so many sheltered valleys that we should encounter greater difficulty in explaining why the sandstones are wholly absent from these broad areas on each side of their present limits than in assuming a very considerable degradation of the walls of the valley since it served as an estuary for the accumulation of the Triassic sediments. Indeed, we may say directly that the present border of the sandstones represents closely the old border of the estuary, because the coarse angular conglomerates and rudely sorted feldspathic sandstones can have been transported but a very short distance, and, as their mineralogical character indicates, must have been derived largely from beds immediately adjacent, which would have been covered if the waters of the estuary had extended as far as the supposed contour line, and that, therefore, the crys- talline border of the valley must have been sufficiently higher than now to form retaining walls for the accumulated Triassic gravels. Within these limits the coarse sandstone rose, as above indicated, to a height above that of Mount Toby, filling the whole valley to that level. The increased ele- vation may have amounted to many hundred feet. It would be interesting to follow the course of the erosion by which the present ridges have been sculptured out of this mass and to divide the long quiet work of the waters in later Mesozoic and Tertiary times from the work of the harsher agencies of the Glacial period. PRE-GLACIAL EEOSION. 517 I think it nvAy be deduced from the facts given above that the greater portion of this erosion was performed by the first agent, but that the ice wore into the soft sandstone considerably, and in some places enormously; so that, if the Pleistocene deposits were removed from the valley, the rocky floor below would bear small resemblance to the surface upon which the ice be"-an to act. I imagine that the present surface of these latter deposits would much more nearly coincide therewith. Indeed, along all the west side of the valley f)"om Deerfield to Southwick and beyond, and north of the Holyoke range, the sandstones may well have been considerably higher than the present cultivated surface of the valley. This is deduced from the consideration that if the present drainage represents closely the pre-Glacial, as shown above, the sandstone should rise by easy slopes from the streams and be highest in the areas between them, or in some way show an intelligible relation to them. But from this point of view the deep depression in the sandstone west of the trap ridge in Deerfield and north and west of the Holyoke range would render such a drainage impossible, and must be a later work, which can only have been done by the ice. This exceptional erosion of the ice depended largely upon the soft nature of the sandstones and the peculiar position of the trap ridges. From the top of Mount Holyoke I have seen the valley fog rest against the hiUs east and west and, rising to my feet, spread, with a surface level as the sea, up and down the valley as far as the eye could reach. If it had risen a few hundred feet higher I believe its mass would have rudely equaled the pre-Glacial erosion of the Triassic, while I imagine the present Pleistocene deposits in the valley would scarcely equal the amount removed by the ice. As for the crystalline rocks which flank the broad Connecticut Valley on either side, the fact that the newer crystallines are covered by the Trias in the bottom of the valley and yet are abundantly present in the coarse Triassic conglomerates, while the older Cambrian gneisses are broadly exposed on the east but are not represented in the adjoining Triassic con- glomerates, shows that there has been large erosion over the eastern plateau since the Trias. The suggestion of Professor Pumpelly that secular disin- tegration may have deeply prepared these rocks for glacial erosion must be taken account of, and renders it impossible to assign to pre-Glacial and Glacial agencies their proper share of work. CHAPTER XVI. THE GLACIAL PERIOD. THE PRESENT ROCK SURFACE AND THE AMOUNT OF GLACIAE AND POST-GLACIAL MATERIAL ON THE SAME. If the unconsolidated deposits — sands, clays, and gravels — ^were removed from tlie valley we should see a rocky floor, everywhere almost the exact surface upon which the ice last lay, except where, from the north- ward-facing cliffs of the Holyoke range, the frosts have since eaten into the much fissured trap and formed the talus of sharp fragments which rests against its base, and in limited areas where the streams flow on rocky beds. The whole horizon would be unchanged. The high ridge which stretches south from Mount Toby, and upon which North Amherst, Amherst village, and South Amherst are built, would be little changed until, coming south- ward, we reached Mount Pleasant, the southern portion of which would be lowered to the level of the street at its western base, and College Hill, Mount Doma, and Castor and Pollux^ would also be absent. A ridge of rock woiTld also stretch southward from Mount Warner, much below the present surface. The three depressions which, running noi'th and south, bound these two ridges, would be much deepened, the East street depression by at least 50 feet; the middle one, separating the two rocky ridges, to an unknown depth ; the western, in which the Connecticut now rmis, to at least 110 feet below low water of that river, and thus down somewhat below the level of the sea. On the west of the river, in Northampton, the changes would be more extensive, as south of Elizabeth Rock and Roberts Hills and east of Loudville all the elevated country. Round Hill, the Hospital Hill, and the rest, would be removed, and the rock floor would be found everywhere down near or below the present level of the river, except along Mill River near the West street bridge. Under the Northampton meadows it may well be a hundred feet below the river level. I have already indicated the probable condition of the valley when the ice began to work upon it. ' Names given Ijy President Hitchcock to drumlins south of College Grove and north and south of South Amherst. .518 THE PRESENT KOCK SUIiFAOE. 519 ard the cause of the extreme inequality of its effects over difiPereut portious of the basin is to be found primarily in the unequal resistance offered by the different rocks of which it is (composed, and secondarily in the influence of the i)rojecting masses of harder rock in deflecting the ice and shielding- the softer rocks in their lee. Here the trap, so easily dissected by the frost, proved most able to resist the onset of the ice. The ridge of trap which makes the backbone of Deerfield Mountain survived after the sand- stone had been worn down on either side and protected the Sugar Loaf ^ in its lee, and, with Mount Warner, farther south, projected into the ice as it wore dee})ly on their flanks. So Mount Toby, built of a conglomerate more durable than the sandstone beneath, and protected by Deerfield Mountain, stemmed the ice and sheltered the long ridge which runs south from it, so far that a fragment of the soft incoherent sandstone still lies along its eastern slope in Amherst village. But the Holyoke range, coming up from the south, swings around eastward in a great curve, commencing at Mount Tom, and from Mount Holyoke on runs eastward to its end, and the great trap sheet which makes its strength is so placed as to present the maximum resistance to the ice moving from the north and northwest — that is, it dips every- where as a continuous sheet from the crest of the ridge southward where the chain runs east and west, and as the ridge swings round to run south- ward the dip of the sheet swings round to the east. It received the pressure of the ice, then, as a log set to brace a falling building receives its weight. The ice, lifting over this sharp obstruction set right athwart its course, wore into it with great severity, and by its recoil as it raised its mass over the opposing range wore to a very exceptional depth in the area just in front of the latter, which had been filled with the soft sand- stone, forming the broad, deep furrow which runs along the northern and western base of the range, beneath the Easthampton, Northampton, and Hadley meadows, and in the southern part of Amherst, in which furrow the tlu'ee deep north-south depressions I have described above ended.^ ' The table-moantain form of Sugar Loaf is probably due to a capping of trap from the southward projection of the Deerfield trap sheet, ■which endured to near the close of the Glacial period. It is called an "Eddy Peak " by Prof. J. D. Whitney (1888) ; see bibliography in Chapter XXIII. = It is an interesting fact that a line at the north foot of the east end of the Holyoke range forms a boundary north of which granite bowlders are abundant in the till, while south of this line they are rare. This is because the ice mass was greatly shattered as it lifted over the ridge, drop- ping its bowlders, while it eroded strongly on the crest; or it may represent the closing period when the ice wore over the Leverett granite and halted at the north foot of the Holyoke range. 520 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. It would be tedious to detail all the observations upon which this descrip- tion of the present surface of the rocky substratum of the valley is based. One may trace on the map the crystalline rocks emerging from below the sandstones of Mount Toby and appearing at the surface in scattered out- crops southward to South Amherst, and the same thing may be seen, though less clearly, south of Mount Warner. At East street a well 50 feet below the lowest ground there failed to reach the ledge, and at the Northampton bridge piles were driven in the clays 110 feet below low water of the river without reaching bottom. The most remarkable effect upon the present contour of the basin of this general erosion of the ice was in excavating hollows so deep and capa- cious that — especially where they lay aside from the direct line of the cur- rents of the floods in the subsequent period — they have remained only partially filled to the present time, notably in the case of the East street val- ley and the southern part of the middle depression, which lies west of the village of South Amherst. South of the Holyoke range the protecting influence of the ridge is as plainly seen as its agency in reenforcing the power of the ice on its north and west, and the sandstones stand much higher and appear abundantly above the surface of the later deposits and doubtless make a continuous substratum for the latter, while north and west, I imagine, the erosion over much of the deeply covered area must have cut down through the sandstones to the crystalline rocks below. The low rock-floored valley bottom, everywhere nearly at and often much below the present river level, stretching across from the Pelham Hills to the western line of Northampton and broken only by the Amherst ridge and Mount Warner, not only sent a lobe southwardly tlirough Easthampton, but another of exceptional depth up through the Deerfield Valley to the north line of that town, which was continued still farther north in a strange, narrow depression running up the west side of Grreenfield and ending abruptly at its north line — a depression which was left unfilled in Cham- plain time. North of Mount Toby the Montague basin would be also largely increased toward the north by the removal of the drift. The immense sand desert between Millers Falls and Turners Falls and all the hills except one that rise above it would be removed, leaving a great depression, THE PRESENT ROCK SURFACE. 521 mucli of it, l)elow tlie present river level, with an old bed of the Connect- icnt running- down its middle and extending- north from Millers Falls to the State line with considerable increase of width. And the removal of the great swarm of drumlins which crowd the area west of the river in the northern portion of its course would materially affect the contours in Grill and Bemardston. On the higher ground west of the valley the removal of the loose deposits would not so materially affect the surface except in the extreme west of Hampden County, and especially in Blandford, where over broad areas the till reaches great thickness and rises in drumlins of the first magnitude. East of the Connecticut Valley the same remark holds, 'except for eastern Hampden and southeastern Hampshire, where the removal of the heavy sands of the great series of Glacial lakes described beyond would greatly modify the surface and would probably show the deep Greenwich- Enfield Valley to be continuous across Ware and thence, via the Beaver Brook and Ware River, to Thorndike, and thence straight south to Palmer station and on through the deep Monson. Valley and the narrow gorge of the Willimantic to the sea. (See map, PL XXXV.) The Ware River also seems then to have run directly south to Palmer to join the Swift River. This basin stretching from Orange south across the State to its south line at Monson is peculiar in many ways. It is underlain by the Monson gneiss and widens and narrows with the width of this rock. While the broad band of this same rock which lies next west of this forms high ground, this forms a deep flat-bottomed valley, in the center of which rise high, isolated, dome-shaped hills of gneiss, which may have been preserved by a capping of the same quartz schists which form the high walls of the basin. The whole basin seems to be the result of deep disintegration of the gneiss. 522 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. GLACIAL GROOVES AND STRI^. (See map, PI. XI, page 510.) List of glacial grooves and strice in Franklin, Hampshire, and Hampden counties ; hearings corrected. [E.H. = E. Hitchcock, Geology of Massachusetts, 1841, p. 387; Keport on Certain Points in the Geology of Massachu- setts, 1853, House Doo. No. 39, pp. 34-44. B. K. E. = the present author.] Locality. Rock FEANKLIN COUNTY. Kowe; near meetinghouse and in part. Rowe; near meetinghouse Eo-we; north part, road to Whitingham, Vt. ; spot nearly as high as Hoosac Mountain. Heath ; west part . Heath; near center, highland. Charleniont Village Shelburne ; northwest part, high hill Bernardston ; Williams farm Northfleld; Gill station , Northfield ; west side of river near Hon. S. C. Allen's. Northfield ; where ferry road crosses rail- road. Northfield ; north part of village Northfield; north of commencement of road in Big Meadow. Northfield; Beers Plain Northfield ; on Strowbridge Hill Northfield ; southern part of village Ashfield; on north road to Goshen, on hill northeast of school. Ashfield; G. B. Hall, east of Ashfleld plains. Ashfield ; east of Howsville Remarks. Sometimes 2 to 12 inches wide and several deep. Sometimes on a slope of 10° southerly imd a still greater northerly slope. The last but one cuts the last. Frequent Authority. Mica schist . Quartzite Gneiss. Hornblende rook. Sandstone do .- Granite . Mica-schist — do Conglomerate. Mica- schist ... Conway schist. do S. 5° W. S. 10-12° W. S. 10° W... s. 10= "w... S. 20°W.-. S. 8°E E Ashfield ; south of South Ashfleld do Greenfield ; opposite S. Bullard's Sandstone . Greenfield ; low down on the trap east of I Trap Poets Seat. | Greenfield; new road to Gill do Greenfield; 200 feet north of the west i do end of this road. Greenfield ; road down to Riverside Sandstone . Do do N.-S S. 10° AV.. S. 30° E... fS.5°E-... IS. 20° W . . S.40°'W .. S. 30° W . S. 5° \7 . . . Parallel with valley - Troughs 1 to 2 feet wide- Distinct Arranged in order of age I newest above. In valley among rugged hUls— first entry newer. S. 5° "W . - S. 30° "W" . S. 18°E. S.25° E. Grooves 4 inches wide. E.H. E.H. E.H. B. K. B. E.H. B. K.E. E.H. B. K. E. B. K. E. E.H. B. K. E. B. K. E. B.K.E. B. K. E. B. K. E. E.H. B. K. E. B.K.E. B. K. B. B. K. E. B. K. E. B. K. E. B. K. E. B. K. E. B.K.E. B. K. E. (ILAGIAL 8TliI^. TAst of glavial grooves and xtriw, etc. — Coiititiued. 523 Locality. KBANKLIN COUNTY— CODtinUOll. Greenliclil : HouthofC. andJ.S. Newton's. Greunticlil : road to lieoch Hill Greenliehl ; north part of Petloral street. Greenfield ; west of factory village Greeniirld ; one-half mile northeast of center. Grei'ntiold ; near mouth of Mill Brook. . . Gill ; school south of center Gill : IJ miles north of Lily Pond Gill ; northwest of factory village ErA'ing : under Dressers Mountain on south . Warwick ; southeast near iron-ore beds . . !Newyaleni; 200 rods south of academy. Book. Sandstone . ....do ....do ....do Sandstone . do Conglomerate. do Sandstone Gneiss Warwick: near meetinghouse Deerfield; south end of trap nearest to river. Deerfield: mouth of gorge of Deerfield Elver. Deerlield ; southeast part Deerfield ; northwest part Deerfield; gorge of Deerfield River in trap range. Deerfield ; at west entrance of above gorge Montague ; southwest corner Montague ; south part Greentield : north edge of city, S.J. Lyons Montague ; road up to Turners Falls from lower suspension bridge. Sunderland ; north bend of north wood road onto Mount Toby. Sunderland ; north part, near cave Sunderland ; nortbwest of cave , HAMPSHIRE CODNTY. Plainfield; south of S. Barton's Plainfield; northwest corner Cummington; northwest comer, oouth of Deer Hill. Ciunniiugton, School No .10 Cummington end of blind road at I. Farling's. Goshen; southeast part, deserted road west of Hubbard's ledge. Mica-schist . Hornblende-schist Trap Sandstone . Conglomerate. Mica-schist ... Trap Sandstone Conglomerate. do Sandstone . .do . IConglomerate- do Sericite-schist . Amphibolite . - . Sericite-schist . Conway schist Sericite-schist Mica-schist . Direction. S. 30° W . S. 10° B . . S. 8° E . . . S. 25° W . S. 8° B . . . S. 15° W . S. 20° W . S. 200 w . S. 25° W . S. 40° E . . S.130 B.. •S.150 E.. .8.25° E.. S.13° E.. S.10° W. S. 15° W. S. 8° E. S.8° E. E S S. 12° E.. ,S.8° E... S. 15° E. . S.5° W.. S.20° W. S.50° W. •S.5° W.. Is. 10° E.. S. 8° W.. S. 10° W. S. 35° E . S.36° E. S.60° E. S. 10° E. S.21° E. S.45° E. S.25° E- S.45° E- E Kemarks. Striae and grooves 3 wide, 8 inches deep. Very distinct Very distinct In Millers River Valley out of Connecticut Valley. Sometimes several feet wide and a foot deep. ^On high ground East slope of Deerfield Mountain. In the Connecticut Valley. Finely preserved Frequent Striae runningup hillside 45° Groove 2 feet wide Groove 5 feet wide, rising slightly. In bottom of deep valley running S. 60° E. Moutonn6 over a broad sur- face. Authority. B.K.E. B.K.E. E. H. B. K. E. E.H. B.K.E. B. K. E. B. K. B. B.K.E. B. K. E. B.K.E. E.H. B. K. E. B. K. E. E.H. E.H. B. K. E. B.K.E. B.K.E. E.H. B. K. E. B. K. E. E.H. B. K. E. B.K.E. B. K. E. B. K. E. B. K. E. B. K. E. B. K. E. 524 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. List of glacial grooves and strim, etc. — Continued. Locality. Kock. Direction, Kemarka. Authority. HAMPSHIRE COUNTY — continued. Groshen ; farther south Worthington ; west part "Worthington ; west part, road to Middle- field. AVortbington Center "Worthington ; Stephens's mills Worthington; I milenorthweat of center. Worthington; first crossroads west of center. Worthington Center Amherst; nortbeastspurof Mount War- ner. Amherst; roadside one-half mile north of the site of President Clark's house. Amherst; northeast of center Pelham ; west slope of ridge 1 mile north of Hygeia. Middlefield ; near meetinghouse Middlefield ; one-half mile south Middlefield; east part, near soapstone quarries. ^Northampton ; high upon road to Eyrie House; Mount Nonotuck. Northampton; southwest corner Kings- ley's mill. Northampton ; at Eyrie House on top of the mountain. Northampton ; below Smiths Ferry bird- track locality. Northampton ; 1 mile south of mountain, 125 rods northwest of where road to Westfield branches. Hadley ; second peak west of G-ap road . . Hadley ; same peak, west fore knob Hadley ; same peak west of this and 40 feet lower, straight pass 15 feet wide, wholly smoothed and scratched. Hadley ; Mount Holyoke House Granby; mouth of forge pond G-ranby ; north part Granby; Moody Corners Southampton ; south of center Southampton ; east of village Easthampton; Mount Tom Easthampton ; quarry, west shoulder of Mount Tom. Holyoke ; south of Mount Tom Belchertown; 42° 20', 72° 25' Micarschist . Sericite-schist . ....do -do. .do. -do. .do. ...-do . Gneiss. Granite. Gneiss. . Sericite-schist . -...do do Arkose Granite Trap Sandstone . Trap -do. -do . .do. .do . Sandstone do do do Conglomerate - Trap Conglomerate. Sandstone . Tonalite . . . S. 76° W. S.22° E. S.22° E. 8.40" E. S.50O E. S.650 E. S. 43° E . S. 59° W . S.150E.. N. 50° W. S. 30° E . . S. 150 -w . S.220 E. s S S. 22° W . S S. 25° E . S. 60° E . •S.50 W-- ,S.2° E... S.150 B.. S.8° E... S.50 W.. S. 73° W. S.120 W. S.8° E... S6.0O B.. S. 60° E. N.80°E. Both follow direction of val- ley. 300 feet be]ow top of hill. Stoss side observed.. Groove 2 feet wide, 4 inches deep, li feet wide jSee figs. 29, 30, pp. 527, 530 - - B. K. E. E.H. B.H. B. K. B. B. K. E. B.K.E. B.K.E. B. K. E. B. K. E. B. K. E. B.K.E. B. K. E. E.H. B. K. B. E.H. b.b:.e. E.H. B. K. E. B.K.E. B. K. E. B. K. E. B. K. B. B.K.E. B.K.E. B.K.E. E.H. B.K.E. E.H. B.K.E. E.H. B. K. E. B.K.E. G. H. Bar- ton. GLACIAL STRI^. List of glacial (jrooves and striw, etc. — Continued. 525 Locality. Rock. Direction. Kemurka. Authority. HAMPDEN CODNTT. Cheater: top of Round and of Gobble hills Blandford ; one-balf mile north of meet- inghouse. Blandford; North stroet, north of meet- inghouse. Emery Serioite-schist do do S.8°E?.... S. 8°E ? Deeply grooved and smoothed. C.U.Shep- ard. E.H. B.K.E. B. K. E. B.K.E. B.K.E. E.H. B.K.E. B.K.E. E.H. B.K.E. B.K.E. B.K.E. B.K.E. E.H. E.H. E.H. E.H. E.H. E.H. E.H. E.H. E.H. E.H. B. K. E. S. 20° E 8.35-10° E. S. 25-40° E. Blandford; North street, farther north, crotch in road. do .. do S.40° E.... 8.73° W... 8.15° W... Beautiful deep flutings, 8 to IS inches across. (Drumlins here run south. B.K.E.) Westfieldj sonthweat from Cowle's quarry, south of mouth of gorge of Little River. Sandstone Conglomerate Trap 8.30° E Sandstone Hornblen de- schi 8 1 8.73° W... 8.40° E Tolland; northeast of Noyes Pond Tolland ; south of mouth of Noyes Pond. 8 20° E do 8.10° E do 8.15° E Granville; middle, 1 mile west of meet- inghouse. Granville; north part.Tnouth of deep gorge in Sodom Mountain opening into Connecticut Valley in Southwick, near west end, near house of Mrs. Jones. do 8.22° E Mica-schist do 8.60° E.... 8.62° E.... 8 20° B 630 feet above sea Thought by Dr. Hitchcock to have been a local gla- cier. do GranviUe; 2 miles northwest of east Tillage, Blandford road. Do do . ...do 8.62° E do do 8 12° E ... do N.80°E.. . 8 80° E .. Thought hy Dr. Hitchcock to have been a local gla- cier. do do Russell; southwest, north bank Little River, 1,100 feet above sea. Russell; southwest, north bank Little River, at mouth of gorge. Agawam and Suffield, Conn.; one-half mile east of east foot of trap ridge. do Sandstone N.80°E... Thought by Dr. Hitchcock to have been a local gla- cier. On the map, PL XI, p. 510, all the glacial strise tabulated above are entered, together with some from the map in the Surface Geology of Presi- dent Hitchcock.^ The latter are transferred as accurately as possible, and appear without indication of the divergence from the meridian, as that is not ' Surface Geology, Amherst, 1860, pi. 8. 526 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. given upon the original map. So far as the southern ]3art of the State Is concerned, this map is the same as the one accompanying the report of 1853, cited above; but the striae with abnormal westerly direction, at Kingsley's mills, in the southwest corner of Northampton (southwest of Florence on the map of the 1853 report), are omitted from the later map. Hence we may suppose that President Hitchcock came to doubt the observation as to the stoss side being on the east. I have entered the strige with the probable direction — namely, east. The map (PI. XI) is very interesting as a composite of (1) the general direction of the ice across the area under consideration; (2) the simulta- neous deflection of the ice at base in the long trough of the Connecticut Valley ; (3) minor deflections around prominent obstructions and in gorges, also afi^ectiug only the base of the great ice sheet; (4) later deflections of frontal lobes of the ice by prominent valleys ; (5) fanning out of frontal lobes. (1) The normal direction is S. 35-40° E., and this is more regularly manifest beyond the confines of the map on the west. (2) The influence of the Connecticut Valley in deflecting the ice south- ward seems felt for a considerable distance out over the western plateau. (3) The southward deflection in the valley is well exhibited and is most remarkable on the top of the Holyoke range. (4) The local influence of the Greenfield and Deerfield trap ridges of the Cummington Valley and the Fall River and Deerfield River notches are well marked, producing in both the latter cases striae at right angles within and without the gorges. The same is seen at both ends of the Mount Tom range. (6) The curious fanning out of the striae north of Westfield, S. 73° W. and S. 60° E., seems to indicate a valley lobe of the ice extending south between Mount Tom and Mount Pomeroy to the west, and expanding to the south where the valley widens toward Westfield. A similar fanning out is indicated in the main Connecticut Valley by the direction S. 50° W. in Agawam and Suffield. The southwesterly direction above Shelburne Falls and the easterly direction down all the valleys opening from the western highlands into the main valley were due to later lobes in the retreating ice front. The most remarkable groovip.g and fluting is found along the whole crest of the Holyoke range on the hard trap. Near the Prospect House, GLACIAL STRIJB. 527 on ALouut llolyoko, are some of the most remarkable grooves I have seen. One northeast of the house, between tlie two iron boundary posts, is at tlie north end 2 feet wide and shallow, at the south end 1 foot wide and 8 inches deep. Several other grooves almost equally marked occur near this. Their direction is S. 2° E. A curious groove (fig. 29) comes out from mider the house on the southwest side and runs in a southwesterly dii-ection. This is exposed best in the bottom of an unused reservoir, and can be traced for a length of 12 feet. It is about 2 J feet wide by 10 inches deep, the greatest depth being at the east side, which is overhanging, being fluted regularly like a letter S. This seems to me to have been caused possibly by water run- ning beneath the ice and to be a true " lapiaz," as they occur beneath the ice of the Alpine glaciers. It must, then, have been polished ^^ ,, . ....^. ^ by the ice at a later time. ,/ High up on the road to the Eyrie House, on Mount Nono- tuck, deep strise run S. 22° W., deflected westward in the direc- tion of the Easthampton valley, while on the summit above broad, ^"^- 29.— eiacial groove on compact aiabase, Prospect House, - Mount Holyoke. deep grooves abound, running north and south. South of the Holyoke range, at Smiths Ferry, the strise run S. 25° W., on trap. At Batterson's sandstone quarry, south of Moimt Holyoke, at E. H. Lyman's house, the fine-grained sandstone is grooved and fluted and sca-atched most beautifully over a broad surface (see PI. X, p. 488). The ice met the vertical and overhanging face of the sandstone and fitted itself, so exactly to it that scratches and polishing occur on surfaces placed at all angles to the horizon, even upon the under side of projecting ridges. In the vertical westward-facing wall the basset edges of the horizontal sandstone beds are polished like glass, and one thin, softer, shaly bed is cut in deeply to form a long groove 10 J inches deep and only 3 inches wide at the mouth, but polished to the bottom, where it is but a half inch wide. The direction of the scratches is here very irregular, going to all points both in altitude and azimuth. The prevalent direction, how- ever, over the broader, flatter surfaces is south. 528 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. I hardly need call attention to the magnitude and the peculiar char- acter of the force which has done this work, grinding down all the pebbles of a conglomerate, hard and soft alike, to a common level, as can be well seen on the road which goes over the north shoulder of Mount Toby just after leaving the Sunderland road, and cutting grooves in the trap, a rock so tough that one rarely attempts to drill a hole in it, preferring, when it is necessary to remove it, to do the work by building fires upon it and drenching the rock with water, by which means it is crumbled and slowly removed. Two men and a holder drill only 8 feet a day in trap. These grooves are of all dimensions, ranging from fine lines, visible only in oblique light with a lens, to broad troughs. Even more striking is the polishing of the surface of the great emery vein in Chester, which for a distance of several rods near the summit of each mountain has been deeply grooved and polished by glacial action. That the friction producing this effect must have been enormous is apparent from the size and depth of the channels, and that it could not have been the result of running water is demonstrated by recurring to the example of river action in the Westfield River upon another portion of the same bed, where we have an eroded, pitted surface from which the coarse crystalline particles of the hard emery are left projecting.^ Another point deserving, perhaps, further consideration here is the great degree of irregularity in the direction of the striae, since these give accurately the direction of the motion of the ice at the time they were made. For many of these differences of direction we may assume, as above, (p. 526) that they were variations in the direction of the motion of the ice at different times. For most we must assume that the great ice sheet was affected by the greater irregularities of the bottom over which it flowed, just as — to use the illustration given by Prof J. D. Dana (to whom we owe this explanation and its application to the anomalous north-south direction of the ice in the Connecticut Valley) — a mass of pitch flowing down an inclined board upon which strips had been nailed at various angles to the line of inclination would in its under parts be deflected behind the strips and flow in the direction of the grooves thus produced. Ice, in short, though moving with extreme slowness, comports itself like a fluid and obeys the laws of hydraulics. Thus the line of motion for the great ' C. U. Shepard, Report on Chester Emery Mine, p. 5. GLACIAL NOTCHES. 529 mass of tlie ice over Hampshire County was S. 35° E., while the lowest portions in the broad depression of" the Connecticut Valley moved with that valley from north to south, and even west of south along the Mount Tom range. While this explanation is surrounded with difficulties, it does explam in a very satisfactory way many peculiarities of the character and distribution of the till in the valley, as will be made clear in the next sections. GLACIAL NOTCHES. Another remarkable series of phenomena, which we may possibly refer, in whole or part, to the direct action of the ice upon the rocky floor over which it moved, is to be found in the succession of notches of varying depth which cut the Holyoke chain transversely in its east-west portion and give it the appearance of a sierra in miniature. One of these cuts the ridge to its base, forming the notch through which the river flows. Two cut down deep into the heart of the mountain, forming low cols, thi'ough the western of which the road runs.^ Others are shallower, and one may find a quite complete series connecting them with the ordi- nary glacial grooves and scratches. The larger notches are themselves scratched and polished, and the direction of the scratches coincides with the axis of the notches themselves. Another circumstance harmonizes with the idea that they were formed by a force like that of moving ice, the direction of whose action was in great degree independent of the relative hardness and direction of the ridge. The Holyoke range lies like a blowpipe with the mouthpiece pointing south and the point directed east. So long as the chain runs east and west the grooves cross it at right angles, running, as did the ice in the valley, north and south, while as the ridge swings round from west to south the succeeding notches run parallel to the first and cut the chain more and more obliquely TUitil the last coincides with the southward prolongation of the mountain and splits it; and one looking at the trap from the west — in Southampton or Easthampton, for instance — sees the almost vertical cliff of trap bounded above by a line which deviates little from horizontality, instead of the serrate sky line of the Holyoke range proper as seen from Amherst. 'This used to be called the East Crack, the deep notch just east of the Holyoke Mountain House being known as the West Crack, and there was once a road through this also ; and the deepest depres- sion between these was the Low Place. MON xxix 34 530 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. This rule is not without exceptions, since the pass which separates Mount Tom from the next peak trends a Httle north of east, and the next passage north trends east and west. President Hitchcock argued as follows concerning the matter : ^ If these notches had been determined by anything in direct relation with the trap of the mountain, the most probable cause would have been a Assuring of the bed of trap during its upheaval, and as this Assuring would have occurred most naturally at right angles to the axis of the chain, the Assures would have con- verged on a point south of the mountain, somewhere about the northwest corner of Ludlow. So he concluded that if tlie first notches he noticed (those cutting at right angles) were caused by Assuring, those farther west would be also at right angles to the chain there and parallel with the dip there; and when he found this was not the case, he explained them as a strange result of the great north- em diluvial current which did duty then in place of the ice cur- rent of more modern theories.-' The larger notches seem to have been caused hj the system of faults which cut the range, and to have been enlai'ged by pre-Glacial streams (see PL XI, p. 510) in case of two or three of the deeper ones. Where, as is often the case, these faults fail to run north and south, the notches may have been remodeled by the ice and given a new direction, and the great number of smaller notches, all parallel with the direction of the ice, do not seem explicable as a result of water action, but rather as the work of the ice acting on the irregular rim of the trap sheet, which emphasized irregularities where this rim ran athwart the course of the ice, as in the Holyoke range, and smoothed them down whfere the rim ran with the ice, as in Mount Tom. Fig 30 East slope of a large glaciated groove behind the bowl ing alley on Mount Holyoke. ' Geology of Massachusetts, 1841, p. 389. GLACIAL NOTCUES. 531 'I'hus it', atU'V cxaniiuiiif'' the marked grooves under and noi'tli of the Prospect Mouse on Mount Holyoke, one goes a few rods east to the groove, about 12 feet deep and of equal width, just beliind the bowling alley (see fig. 30), one will find it hard to draw the line between them. And if, after exam- ining the grooves and striae on the second peak west of the notch, one goes down west into the deep groove about 40 feet across, the similarity in direc- tion and shape will lie foun.d very striking; and such cases are quite common. PSEUDO-GLACIAIj STEI.^ on DEVONIAN ARGILLITES. While examining the garnetiferous mica-schists at Purple's quarry, in the east part of Bernardston, I was attracted by a peculiar striation which occurred upoii a broad, flat cleavage surface of the nearly horizontal slates and continued beneath the superincumbent beds. The surface in question was just at the north edge of the water which fills the abandoned quarry, and was certainly in place and undisturbed, and I raised the slates which rested upon it and followed the striation beneath, for a foot or more inward without seeing anything which suggested to me that these upper layers were not also in place and undisturbed. Clear impressed lines, from those so fine as to be seen only with a lens up to those a millimeter in diameter, covered the broad, flat surface — in average about an inch apart — the larger showing a delicate longitudinal striation. These grooves vary in length between quite wide limits — 1 to 6 inches. The larger number are straight, or nearly so ; very many form easy open curves, single or double. Over most of the surface two distinct systems, making an angle of 40° with each other, were apparent, the one having the longer and finer lines and most of the long curving lines, the other being somewhat broader, shorter, and more rigidly parallel and straight. Their length varied very little from an inch, and they were often slightly gouged out at the end. On putting several parts together, so as to get a broad surface, the finer lines of the first system are seen to bend and continue in the second system. The whole impression was quite like that of the rain-marks on a car window before and after starting. Faint traces of a third system at right angles to the first are also present. The direction in the rock was not taken, as the marks were supposed to be of mineral or organic origin; many of them strikingly resemble in size, curvature, etc., impressions of Graptolithus 532 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. minutus. A suggestion of long needles of hornblende or chiastolite also occurred. On further examination the minute garnets on the surface were found to be polished down and scratched like the rest. At one portion of the surface unmistakable glacial striae were found adjoining the problematical grooves. That the marks were formed by movement of an upper layer of the slate on the underlying ledge seemed clear, and that the garnets fixed in the bottom of the upper moving stratum furnished the grooving tools. The change in the direction was caused by a change in the direction of the moving mass, some portion of the bottom becoming fixed and forming a pivot around which the rest revolved. That the mass was moved only a slight distance from its original position was also clear. Whether this motion was caused by glacial ice, by the expansion and contraction of the rock, or by earthquake action, I can not decide. POT-HOLES. President Hitchcock notes ^ the absence of pot-holes among the results of the diluvial currents which were supposed to have originated the till and the glacial striae, and concludes therefrom that these phenomena were not the work of rivers but of widespread currents without falls of much magni- tude. He describes later a great series of pot-holes west of Shelburne Falls, on the road to Charlemont, in an old bed of the Deerfield River, 85 feet above the present stream, which may have belonged to a pre-Grlacial bed of the river or may be of Glacial age. Pot-holes occur, of course, along the channel of the Connecticut and its tributaries, in the former especially below its falls in the canyon formed by their recession, in the latter on the bottoms of the deep gorges they have cut through the crystalline rocks. Striking illustrations are to be seen in the Westfield River, at the Crescent Paper Mills, in the extreme north of Russell. Just below Russell station also a great dike of granite formerly obstructed the stream, but has been cut through, and here are many pot-holes. One interesting one was half removed as the stream cut down its bed, and the remaining half is still to be seen in the wall, about 10 feet above the water. It is regularly urn-shaped, with bent constricted neck, and is about 6 J feet deep. ' Final Report, 1841, p. 392. THE TILL. 533 liy far the finest development of river pot-holes is in the almost inac- cessible canyon of the south branch of Westfield River, one of which is 25 feet deep and 20 by 10 feet at the mouth. They exist abundantly along the coiu-se of Deerfield River, in many cases high above the present level of the river, as noted by President Plitchcock. I counted more than 50 on a single reef of sandstone which projects into Deei-field River at the most northerly point reached by the stream before it turns towai'd its notch in the trap range. One is found by the road to the south side of Catamount Hill, in Colerain, 2 feet deep and 1^ feet wide. The only pot-hole, however, which I can without hesitation assign to the Glacial period I found by the roadside under the steep southern face of Sugar Loaf, in South Deerfield. It is in red sandstone at a point 130 feet above sea, and is 2 feet deep and 2 feet wide. From its position it must have been formed during some phase of the Glacial period, as it lies apart from any probable stream bed, and the surface of the sandstone around it is striated. I have surmised that these usual accompaniments of glacial action, which we should especially expect to find in so irregular a region, may have been many times formed and again eroded and destroyed by the ice, and that tliis may be the origin of many of the spherical, ovoid, and flat-ellipsoidal pebbles of quartz which occur here in considerable numbers in the true till and which agree quite exactly in form with the polishing stones of a pot-hole. However, they may belong to a coarse-pebble beach of inter-Glacial age, synchronous with the pink sands described below. THE TILL. INTRODUCTION. Pure ice moving over the country would by its thrust tear off project- ing portions of the subjacent ledges, but could not alone polish and scratch the rocks as we find them now. The agents of this work were the stones themselves, which, torn from their places and frozen in the ice, trans- formed it into an immense rasp and increased its eroding power many fold. By the melting and freezing of the lowei- surface and by the slow intestinal motion, as well as by the sudden fissuring of the mass, its lower portion would become filled with a large and varying quantity of loose, rocky material. Also, where, by secular decomposition, as indicated on page 374, the rocks had become softened to great depth, the whole, soaked with water,, 534 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. might be frozen into a solid mass, and the snows gathering on this, it might with httle change become the base of the glacier and be moved on bodily Mnch of the same material was pushed along beneath the ice, its parts crushed and ground against each other, whereby all the softer rocks were soon reduced to an impalpable mud, the larger and the harder ones enduring longer, but all at last yielding to the same fate, unless, worked up into the ice itself or gathered in thick accumulations beneath it, they were shielded from the more violent action of its mass. The ice was, however, constantly providing itself with new material, and soon wore the fragments into the peculiar shapes so characteristic of glacial accumula- tions, three- and four-sided forms, with irregular ends more or less elongate as the rock was more or less schistose, the sides flat or broadly convex, joined by rounded edges and scratched in various directions. These peculiar forms, called by the G-ermans " dreikantner," are as characteristic of the till as graptolites of the Silurian. Thus the ice elaborated in immense quantity a peculiar subglacial material of varying but always characteristic composition, and spread it with unequal and sometimes with very considerable thickness upon the rocky surface. For the ice did not everywhere and ahvays rest with its rasping surface upon the rock and grind into it without interaiission. Over a given surface it might wear for a long time continuously, but by this means a new surface would be gradually produced, partly by the unequal force of the ice, partly from the varying hardness of the rock, and this would react upon the ice, producing slight variations in its subordinate currents, transferring its intenser action to another area and alloAving it to deposit material over the first area. At a later time the maximum of eroding power might be transferred back to its former position and the accumulation so laboriously brought together would be again swept away. In this way one may explain some of the cases where the rock surface shows striae in two directions, for the local movement of the ice might be somewhat different at widely separated times. Very commonly the ice heaped up its accumulations in the rear of some obstruction in a long ridge projecting from the obstructing rock in the direc- tion in which the ice was moving, as the water arranges sands. At other places, especially in broad open portions of a valley, the ice molded its fine clayey moraine material into massive hills, called drumlins, rounded and THE TILL. 535 eloii|)osit, iind one needs only to exjunine n Iresli exposm'e of it and see how all its parts are thrown together in confusion, without any assorting- accordino- to the weight and size of the stones — here a large bowlder ])rojecting, there many small ones grouped, and again over broad surfaces the dark-gray compacted clay occurring almost free from stones of con- siderable size and lacking all signs either in the color or the grain of a lamination or an assortment into parallel layers — one needs only to make these observations and then for comparison examine the clay banks or sand and gravel beds so well exposed in the river banks, remembering that they are instructive only in a somewhat fresh exposure, to be convinced that all the characteristics of water action — the delicate sorting and arranging, like with like, according to size and weight — are here markedly absent, and that it is quite impossible to explain the bed as formed in this way. If one has reached this conclusion by carefully compai'ing the two formations and has the opportunity to examine many sections of the drift where it is a fine sandy clay, he will be almost startled to find isolated patches which seem, to show a true and delicate lamination — a series of fine, horizontal, parallel fissures, a few millimeters apart, usually gently undu- lating. At times the undulations of adjoining lines meet at equidistant points like a flat-meshed net, or like the cleavage of hornblende, so that the clay is separated into a bundle of flat, sharp-edged blades. These lines fade away, however, in all directions into the general formless mass, and constitute not a lamination in the technical sense — a result of deposition in water — but a pressure cleavage caused by the same force which had compacted the whole stratum. The effect of considerable pressure in producing cleavage, or a tendency to split at right angles to the direction of the force applied, may be seen in a variety of instances, and its recog- nition has thrown light upon important problems of geology, such as the delicate banding of glacier ice and the smooth splitting of roofing slates. Gun-cotton pressed into cakes, or thick pasteboard calendered under heavy pressure, may be separated easily into thin layers, and even the splitting of a common cracker or the flaking of pastry is a structure pro- duced by the pressure of rolling out the dough and developed afterwards in the baking. This structure was well seen in the waterworks ditch opposite Phoenix Row in Amherst, and in the canyon of Deerfield River through the divide i-ange, described in the first section of Chapter XV, p. 509. 540 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. In another way, though rarely, the till may simulate the bedding of sedimentary deposits, where it is thickened into dome-shaped drumhns, and it will be seen later on that these are common in the valley. I have once or twice seen a rude separation into thick, irregular layers molded into each other and distinguishable only in a view of a broad surface at a distance. This occurs at the section mentioned at the bridge in Leeds. It would seem that the ice pushed one layer after another into the accumulating mass and so gradually built it up. In the foregoing discussion of the various phenomena of the valley drift I have assumed its siibglacial origin, though many of the details per- haps would fit equally well with the idea that the mass was pushed out from the front of the ice as it retreated northward with various oscillations. The great compactness of the whole and the pressure cleavage would hardly . . ^Safla/sfv/)*rW/Wistr/ateaLsiJrfacf IVater/eveJ be/oiv c/sm | cont/nuedont/iesurfaceoff/tet/'/la-^ob. 'Ground moraine lof^^r. -G. 31— Section on the left bank of the Mill Elver at tlie hoe factory, N'orthampton, taken after the washout of 1878, which carried the dam away, showing the striae on the surface of sandstone continued on the surface of the till below. be explicable upon this assumption, and I have now to describe two sections which render it certain that the whole mass is of subglacial formation. On the night of December 10, 1878, the Mill River, flooded by the very abundant rains and by the breaking away of several dams on its head- waters, rose in Northampton to a height greater than on the occasion of the flood of May 16, 1874, which caused so great a loss of propei'ty and life, and was less destructive only because the earlier flood had done its work so thoroughly. It carried away the western part of the dam at the hoe factory in Northampton and wore deejDly into the western bank, exposing the section seen in fig. 31. The dam had been built on a reef of coarse red sandstone which ran diagonally across the stream from northwest to southeast, the stream flowing here from north to south, and the section runs in the latter direction. The THE TILL. 541 surface of the saiulstoiie is rounded and retains everywhere the glacial scratches perfectly. These are broad, deep grooves, uniformly directed S. 30-40" E. On the south side the sandstone was uneven and ended abruptly in a nearly vertical wall, against which rested a mass of dark-gray till of stony compactness, the surface of which was an exact continuation of the broad, convex, striated surface of the sandstone, showing that the ice had passed over them both together and planed them down to a common level. Ao-ain, in changing the grade of the Canal Railroad, near the South street bridge in Northampton, a section was exposed where the compact stony clay abutted on the east against the red sandstone, continuing the curvature of the convex roche moutonnde surface of the sandstone in the same way. In each case the drift and the sandstone were covered by the Champlain clays in such a way as to show that the exact surface of the drift upon which the ice rested had been covered with the clays immediately after the disappearance of the latter, a point I have developed more fully in discussing the clays and their relation to the valley drift at the beginning of Chapter XX. THE COARSE VALLEY DRIFT. On the west side of the river in Northampton the bay formed by the retreat westward of the crystalline rocks is much deeper, the drift accu- mulated there is more abundant, and the drumlins are on a larger scale. The deposit is, however, much more masked by the later accumulation of sand in the flood period, by which the whole surface is brought up to the level of the highest terrace. While the deposit is probably the exact equivalent of the valley drift already described from the east side of the river, I have thought it best to describe it separately, both because it occurs in a separate portion of the basin and because it presents several points of difference when compared with that. While the paste is clayey and well compacted, it is generally much coarser, bowlders above a foot in length often making up three-fourths of the mass, and masses above 3 or 4 feet in length being in places very abundant. The three outcrops already described in the preceding section from the west side of the river agree in their fineness and exceeding com- pactness with the fine valley drift with which they are associated, and, like it, certainly rest directly upon the older rocks. I have not been able to ascertain if this was the case with regard to the coarse valley drift here 542 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. under discussion. The latter differs from the former also in the origin of its material. On the east the bowlders, except the abundant far-traveled quartzites, are mostly fi'om Mount Toby and the granite south of it. I have rarely found a piece of the spangled mica-schist of the great western range, although the bowlders of 2:)silomelane and yellow cavernous jasper found across Amherst seem certainly to have come from the locality in Conway on the northwest. On the west all the rocks to the north and west are abundantly represented by large bowlders, and very large masses of the Vermont quartzites are also abundant ; one taken from near D. Denniston's now adorns the old Whitney homestead, on King street, in Northampton, and is about 6 feet in diameter.^ I have been inclined to connect the excep- tional coarseness and abundance of the subglacial debris gathered here with the peculiar direction of motion impressed upon the lower portion of the ice by the trend of the great valley. As the ice moved toward the valley from the northwest it came upon its western rim well charged with bowlders from the area it had crossed, and was below deflected southward by the trend of the valley, and still farther deflected to the west of south and obstructed by the transverse Holyoke range, and its morainic material was gathered in a sort of eddy under the western cliffs or swept southward in the valley, and so failed to reach the eastern side of the basin. DISTRIBUTION OF THE COARSE VALLEY TILL WEST OF THE RIVEE. The most northerly exposure of the bowlder clay in the river side is at the westernmost point of the great Hadley bend, where the river has worn into it, and the bowlders, accumulating upon the shore, have formed a natural "riprap" and thrown the current across against the Hadley side, where it will in time cut off the point of the bend and leave its present channel. This exposure seems to be the northern end of a long ridge or series of drumlins which runs in a general way southward across Northampton, mostly covered by the later sands. It is exposed on Slough Hill, west of the north end of King street, and deeply cut into by the Canal Railroad at the Black Pole bridge. Its further prolongation, Round Hill, is a mass- ive drumlin. Under the Forbes Library, Smith College, and the asylum the bowlder clay rises to the surface and reaches just the same level as the surrounding sands which form the level surface of Elm street. These two 'It has been placed over the grave of Prof. Josiah D. Whitney, late professor of geology in Harvard University. THE TILL. 543 liills (if drift are separated to tlie depth of the present bed of Mill River, as is shown by the height of the bowlder clay in the section at the hoe factory, ah-eady described. West of the asylum Sunset Hill and the long wooded hills between which runs the road to Loudville form the most elevated and extensive accumulation of drift in the valley. The first of these hills, starting from the northwest corner of the asylum grounds, runs southwesterly and is continned across the Loudville road by the higher and more massive hill which is conspicuous at a distance from the number of great bowlders of whitened quartz-diorite (tonalite) that cover the broad benches on its southeastern side. The road to Easthampton skirts this hill for a long distance on its southern side, and the sands of the high terrace abut upon its other slopes. This completes the chain of ridges, and a glance at the map will show how they are swung in a broad curve, from Elizabeth Rock to the West- hampton Hills, across the mouth of the deep bay formed by the recession of the crystalline rocks. Within this bay two other prominent drumlins take the same west-of-south direction — the wooded hill east of Florence and the long steep elevation above Bay State on the north. And finally, all the broad wooded area west of Mill River opposite Bay State is a con- tinuous tmdulating ai'ea of bowlder clay, and from a point below Florence nearly all the way to the asylum Mill River is wearing into it, and its western bank is covered with abundant bowlders from which the stream has washed out the finer material, and here are the best permanent sections m the bowlder clay to be found in the valley. Farther south, across Southampton and Westfield, the Champlain sands occupy the greater portion of the valley bottom, and where the till appears it is usually with an undulating surface and is made up of rather fine-grained, reddish material, derived mainly from the red sandstone, very compact, with bowlders nearly all under 1 foot in greatest length, and thus is very different from the upland till. This is notably the case in all the west half of Southwick. DRUMLINS. Perhaps the most notable contribution to science made by the Second Geological Survey of New Hampshire was the recognition of this interesting and peculiar form of drift hills, coupled as it was with a careful mapping of their distribution and a satisfactory explanation of their origin. 544 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. The credit due for this piece of excellent original work is certainly not impaired by the fact announced by Prof. W. M. Davis, in his historical rdsumd of the literature of the subject,^ that the same forms had been observed, mapped, and correctly explained by Mr. M. H. Close,^ in Ireland, at an earlier date. As, moreover, the name "lenticular hills," proposed by Messrs. Hitchcock and Upham, is certainly not a very satisfactory one, and seemed, indeed, not wholly satisfactory to its authors, it is a matter of con- gratulation that the earlier paper proposes a name from the Irish, which, with the Scotch, is so much richer in names for the varieties of surface form of the land than is our own dialect. Indeed, if the word could come into general use it would be a valuable addition to our synonyms for hill forms, while its more precise use follows a custom already set in this department of study. The comparison of drumlins with the sand banks formed beneath flowing water seems quite satisfactory. I have also been interested to compare them with roches moutonndes, with which they are associated in origin beneath the ice. Fig. 31, p. 540, is a representation of a vertical bank of clay and till resting against red sandstone. A broad roche moutonnee of the red sand- stone, beautifully striated, was exposed, and abutting against the southern vei'tical and unstriated wall of the rock was a till of almost equal com- pactness with the rock itself, bounded above by a curved surface, which was the exact continuation of that of the sandstone. The curve sank under the water above and below. This may with some propriety be called half roche moutonnee and half drumlin, and illustrates the close similarity of the cause originating the two rock forms — the differential pressure of the ice upon its substratum. In several other cases rock takes part in the formation of the drumlin, at times as a nucleus with steeper slopes than those of the drumlin itself, but appearing along its crest; yet this is the exception here as elsewhere. The distribution of these hills along the valley, as shown in PL XXXV, is interesting, and may throw some light upon the question of the north-south motion of the ice in the valley as compared with the northwest- southeast motion on the higher ground on either side. 1 Am. Jour. Sci., 3d series, Vol. XXVIII, 1884, p. 407. 2 Jour. Royal Geol. Soc, Ireland, 1886, p. 1207. DRUMLINS. 545 If one could remove ;ill the newer deposits — sands and t-luy.s — which still await our discussion, and then raise this great stratum of stony clay which overspreads the valley, as one lifts a plaster mask from the face, it would be found that its under surface had been exactly molded to every line and curve of the rocky substratum; but its upper surface would have the effect of a comic mask, swelling with unequal thickness over every prominent feature, distorting and concealing its true form, and sending up great protuberances due wholly to a thickening of its own mass and not molded on any projecting ledge below. The protuberances formed thus by the local thickening of the drift sheet appear now as dramlins — massive domed hills, in shape like an inverted canoe, with the long axis pointing in the direction of the glacial motion, from north to south. Where they are most symmetrical they slope away rapidly and equally toward the east and west, more gradually but equally north and south, and very naturally suggest the name "hogbacks," by which they are often known inland, or "whalebacks," as they are called near the sea. They rise hke islands out of the sands, which wrap around their bases to a maximum height of 150 to 200 feet above the present low ground of the valley, and often the thick- ness of the till composing them seems to be greater than that. The two hills just north and south of the village of South Amherst named Castor and Pollux by President Hitchcock, from their close simi- larity — another to the east of the former, and two others farther south and west, are all cast in the same mold. Farther north the hill south of Col- lege Grove— named the Occident by President Hitchcock— the College Hill, and, finally, all the group of hills occupying the space between Amherst, East Street, and North Amherst villages are of the same origin and pre- serve in varying degrees the common form. In the case of all the hills around South Amherst, except Castor, there are no neighboring outcrops by which one can judge of the elevation of the subjacent ledges and so fix the thickness of the drift stratum forming the hill. The surface of the rock may be concave beneath them and the thick- ness of the till much greater than their height above the valley bottom. In Castor the gneiss and granite appear high up on the shoulder of the hill on the east and the west, and if it runs under the drift at the same level the thickness of the latter would be about 30 feet, which is probably more than the real thickness. MON XXIX 35 546 GEOLOGY OF OLD HAMPSHIEE COUJSTTY, MASS. The ledg-e seems to me to be very deep below the surface of the Occi- dent, the hill south of College Grrove, but no certain data are attainable. The well on the east slope of this hill at R. W. Greene's was sunk in "hardpan" 46 feet, as he informed me. In the College Hill the Octagon cellar showed that the surface was the true till The college well is 25 feet deep, and about 45 feet below this the well already mentioned was sunk 55 feet, and another on the southwest slope of the hill, at the east end of Short street, was sunk to the same depth, giving the till an ascertained thickness of 100 feet, and the whole thickness is certainly much greater. The high lull north or northwest of East Street has in perfection the form of a drumlin, and is doubtless of common origin with those already described. There is, however, at its top a broad expanse of red sandstone, which appears nowhere upon its slopes, either at the surface or in wells. It has, therefore, a nucleus of rock of different configuration and with sharper slopes than the present hill. Between the Center and North villages, finally, and bounded on the west by the road and on the east by the railroad between these places, is a group of these hills, so blended that the symmetry is somewhat lessened, in which different summits resemble the several hills already described. Under the hill on which Professor Tyler's house is built, and its prolongation eastward, the red sandstone is everywhere near the surface and the till is thin. West of Mount Pleasant the gneiss is near the surface; it is 50 feet below the surface under the house at the south end of Mount Pleasant, and rises to the surface a few rods north of the site of the residence of the late President Clark. An examination of the map will make it plain that these hills rest upon a concealed ridge of older rocks running south and a few degrees west of south from North Amherst City to South Amherst, and that they lie in the lee of the high ground consisting of crystalline rocks which projects westward north of the former village, and, finally, that there is a close similarity in the arrangement of the drumlins on both sides of the river, those on the west lying in a line curving to the westward and in the shadow of the projecting heights of Elizabeth Rock, as described in the preceding section. The most striking series of drumlins in the valley is found in Bernards- ton and Gill. They are of the largest size and of most symmetrical form. A fine view of them may be had from the railway in Northfield village. DEUMLINS. 547 across the river U) the west. They cover the first range of hills above tlio highest ten-ace and rise one behind the other, their long, curving lines overlapping re]ieatedly and forming an ideal drumlin landscape. Standing on top of the tallest of these hills east of the village of Bernardston and lookin"- southward, one can see the train of drumlins crossing the plain, where they are in part submerged in the Champlain sands, and then rising high upon the great mass of Triassic sandstone which forms the town of Gill, though not reaching its top. The surface of the sandstone beyond and higher up is molded into drumlin-like forms. Descending the south- ward slopes of the sandstone mass, or following the eastward side of the valley southward, one finds no drumlins except a single small but well- formed one beside the railroad just north of the station in Whately. Nor is any trace of them to be seen north of or up the north slope of Mount Toby, which holds a situation in the valley quite similar to the Grill mass. It is a peculiarity of these hills in Bernardston that while they in many places obscure the geology of the region fatally, the interspaces are over considerable areas almost driftless, so that, outside the regular oval base of the hill, fragments on the surface are quite safe indications of the ledges which lie but a little distance below. As indicated upon the map the boundary of the crystalline rocks wMcli form the western border of the valley follows the east line of Ber- nardston near the river and then turns west along the south line of that town and Leyden, and again south along the west line of Greenfield, Deerfield, and Whately, to Northampton, where it is again set back by the width of the latter town, and runs thence southerly to the south line of the State. Along this sloping border of the valley between Greenfield and Northampton runs a train of drumlins, some having their bases nearly 100 feet above the level of the high terrace sands (Northampton high terrace 305 feet, Greenfield 357 feet, above sea), while others are more or less submerged in these sands; indeed, in several cases wholly submerged and beautifully regular drumlins have been exposed in the extensive railroad cuttings up this side of the valley. In one most interesting case at the Camp Meeting cutting on the north line of Northampton (see PI. XV), what seemed to be a broad terrace of coarse sand contained, to the dismay of the contractors, a fine drumlin of rocky hardness which had to be blasted away in front of the steam shovel, and was capped by 548 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. three later glacial deposits with as many intervening sands, which could be followed for 3,350 feet in the open cutting. Several of these hills of most regular shape are so built up on the steeply sloping rocky valley side (the valley runs here north and south) that looking up from below one seems to have before one a drumlin of the largest size, while looking down from higher up the hillside one sees only a small ridge interrupting the eastward slope. These hills are directed southerly, as are the neighboring strise, but they He near the western boundary, between the areas of southeast and of southern motion of the ice, like a line of bars between two currents meeting from different direc- tions; and in Northampton, where the ice was deflected in the valley southwestward, the drumlins have the same direction and swing in a great curve across the reentrant angle in the rocky border. In the town of Amherst nearly every hill is a drumlin, and in several cases they are laid side by side in pairs and coalesce laterally. It is fur- ther interesting that this group of drumlins in Amherst runs right up to the steep northern base of the Holyoke range, which here traverses both the valley and the direction of the ice, and whose crest of trap is finely covered with north-south striae. I have mentioned above that the drumlin exposed in the Camp Meeting cutting, a little higher in the valley, is covered by three sepa- rate glacial beds, representing, doubtless, as many oscillations in the ice at the time of its retreat, which shows — what, indeed, hardly admitted of doubt — that the drumlins were formed beneath the thick ice of the general glaciation; and the position of this last group, carried with north- south axes right up to the foot of the steep Holyoke range, which itself is striated in the same direction, bears strong evidence against the exist- ence of a separate Connecticut Eiver glacier which should explain the north-south striation of this valley. Indeed, these north-south drumlins are carried up so high on the sides of the valley that when one imagines ice of the smallest thickness needful to build them and compress them to their present rock-like density, one sees that the ice would have risen above the boundaries of the valley and have overflowed fan-like, as in the great lobes found in the Western States. The facts seem, then, to accord better with the theory proposed by Professor Dana of a differential motion of the lower portion of the ice in the valley, and the long line MORAINES AND BOWLDER TRAINS. 549 of flrumlins carried down the western border of the valley mark the line alon<>- wliicli the ice was deflected southwardly into its new direction. South of the Holyoke range and east of the Mount Tom range the drum- lins are broader, flatter, and fewer in number than farther north. On the hills east and west of the valley drumlins are rare oi wanting. I have noted only one train — this of hills of the largest size — which enters the northwest corner of Blandford from Becket, with direction S. 35° E. A very fine one is situated a little southwest of the center of Granville. MORAINES AND BOWLDER TRAINS. The great ridge of bowlders of tonalite which passes the Catholic church in Thorndike and extends southwardly, going to the west of the group of high hills southwest of this village and appearing in exceptional force near E. Brown's house, just west of Palmer village, and crossing the river to mount the high hill just south (Bald Peak, in Monson), is a portion of a true ter- minal moraine of a lobe of the ice which shut up the gorge through which the Quabaug River passes northwestward from Palmer village, and fur- nished the barrier for the Palmer Lake (see PI. XXXV and Chapter XVII). In the latter part of its course its bowlders are exclusively of Monson gneiss and of very large size, one 26 by 16 by 7 feet. From the large dike of granite in the center of Middlefield a well- marked bowlder train is carried across Chester, passing through the center of the town and traceable for a distance of 5 miles. Just west of the road running north from the village of Leverett a prominent hill of granite is continued for a long distance southward by a mass of bowlders so densely packed that it seems like the continuation of the hill itself, and in the northern part of Worthington is a similar crag-and-tail arrangement of colossal bowlders of mica-schist carried southeast from a prominent hill, so closely packed that one can jump from one to another for a long distance. Stretching southeast from the great band of fine-grained granite west of Burnell's pond in Chesterfield is an immense accumulation of large, often immense, bowlders. It continues to the southeast corner of Chester- field and on into Westhampton. A marked bowlder train starts from ti/e dike of pecuUar porphyritic granite northwest of Leyden and extends past the center of the town and on a little east of south into Greenfield. 550 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. Bowlders of the peculiar brown, porous, and drusy chalcedony and pyrolusite from Conway are so abundant across Amherst and Granby as to deserve mention. One mass on the eastern peak of the Holyoke range measures 6 by 6 by 4 feet. THE INTERGLACIAL SANDS. An elderly lady in Amherst says that when she first visited the town of Amherst there was a remai'kable spring, never failing-, near H. M. Burt's residence, opposite the A^<1> house, the water from which flowed down eastward across the common and into a quagmire overgrown with alders, in which several pigs were drowned during her visit. Since then Mr. Burt's well occupies the position of the spring, and is remarkable for its volume of water, which often rises to within 2 feet of the surface at the very crest of the ridge. Farther north on the same ridge the well at J. L. Lovell's house is also remarkable for its abundant flow of water, it being almost impossible to empty the well. Again, in lowering the Northampton road opposite College Hall in 1878 a layer of yellow stratified sands, the finest 0.3 to 0.6™™, the coarsest 0.5 to 1™™ in grain, from 6 inches to a foot and a half thick, much contorted, was exposed, which was covered by a thin layer (from 6 inches to a foot and a half in the section, but rising to a greater thickness farther north) of . a hard, blue till and underlain by an ashy till carrying many striated bowlders, one mass of conglomerate being 3 feet long. The sand layer continued to both ends of the section, about 5 rods. The same section occurred at two excavations farther north on the same ridge, on the grounds of Mrs. Davis and William W. Hunt. I did not con- nect these facts or find suitable explanation for them until I had studied the exceptionally interesting section furnished by the digging of the Amherst House cellar. In digging the cellar a block of earth 92 by 104 feet and 12 feet deep was removed, and at the same time the ditches of the Amherst waterworks were opened, having a depth of from 5 to 8 feet and extending from a point just in front of the cellar eastward to the dam in Pelham, a distance of nearl)^ 3^ miles, a mile north to the Plant House, 1,400 feet south to the railway station, and 1,200 feet west to the brow of the hill on Amity street. The cellar section is illustrated by the figures of PI. XII, drawn care- fullv to true scale. Fig. 1 is taken from the northeast corner of the cellar. 12 U. S. GEOLOGICAL SURVEY. MONOGRAPH XXIX. PL. XII. ^■■. ^r""^^^- I "■■^••\~': ■ ' 4 '=. :.;i4?a,v ': D ''-> ^' ■■■■■/ f*^:-. .-^-^^ r::> - CLAV FILLING ^^...^^^ OF CREVICE ^^ 3 5. THIRD TILL. 4, SECOND SAND. 3 SECOND TILL. 2. INTEROLACIAL SAND. I. LOWER TILL. 8CRATCHED BOWLDERS. SECTIONS OF AMHERST HOUSE CELLAR, SHOWING INTERGLACIAL BEDS. INTEIIGLACIAL SANDS. 551 lookiii"- soutliwc'stward The lowest stratum (1) i)resent is the lowest till or vallc>' drift, which forms the floor of the cellar and is seen rising to the sur- face in the south wall (fig. 2). It forms all the remainder of the south, all the east, and nearly all the north wall (fig. 6); and in the waterworks ditches which radiated from this point it occupied the whole depth for a thousand feet north, south, and east; and to the west, where the ditch ran parallel to the north side of the cellar, it repeated exactly the section developed in the latter (fig. 6). The ground here is 311 feet above tide, and slopes away in all directions, so that the till soon sank under the highest stratified deposits of the subsequent flood period, which reached here nearly 300 feet above tide. Above this level it had never been covered, and the boundary of the till traced upon the map represents only the uncovered part. The bottom of the deposit is here nowhere exposed, but farther east, opposite the old Amherst Bank building, the New Red sandstone comes to the surface and has this till on its back, and farther north the gneiss does the same at the entrance to the Agricultural College farm, and in both cases the stratum has shrunk to a foot in thickness. The cellar deposit has already been made the type of the detailed description of the valley di-ift (page 537). Upon this base rests a layer of stratified sand (2) 5 feet thick, upon this a bed of compact till (3) 1 to IJ feet thick, next 1 foot of sand (4), and the whole is capped with a 7-foot bed of till (6). The lower sands (2) were deposited immediately upon the irregular, hummocky, apparently eroded surface of the till, the lowest layers, some- times gravelly, folding over smaller irregularities and projecting bowlders and gradually obliterating the depression. The upper and larger portion was cross-stratified on a large scale, the laminae dipping west from 5° to 40°, and where the structure was least disturbed a high dip, about 30°, pre- dominated. Here and there a delicate flow-and-plunge structure could be seen. The whole stratum consists of clean, well-washed sand, whitish where not colored by a later infiltration of iron, varying from a fine sand which retains water and has an average grain of 0.09""° to a coarse granitic sand having a grain of 0.5 to 1""°. Thin seams of gravel separate the layers of sand here and there. Comparing many samples with the ordinary sands which compose the higher terraces of the valley, I found them to agree quite well under the microscope, but the glacial sands had been more rounded by attrition in water and were better sorted than the later flood 552 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. sands. The bed was, however, everywhere disturbed by the pressure and thrust of the ice which had moved over it from north to south. On the western face (fig. 5) the laminae, which, being exposed in the direction of the strike, had run nearly horizontally, have been squeezed into wavy folds, and often show beautiful illustrations of reversed faults, the upthrow overlying the downthrow, the faults always dipping to the north. On the southern face these faults were much more numerous, and as the work of removing the earth progressed they were constantly chang- ing. On one face of 10 feet I counted twenty. On the surface repre- sented in the figure they are present in great number, and two are especially marked, one faulting the bed 3 feet ; these dip 60° W. In other portions the bed was thrown into entire confusion. Over a large portion of the section a beautifully dehcate incipient cleavage has been superinduced in the sands by the pressure, and its existence is made manifest only by the concentration of iron rust in sharply distinct layers 1^ inches apart, which run parallel to the level base of the till above, passing across the laminae of the sands and distinguishable instantly from the ordinary lines which mark the lower limit of infiltrating water. Their position in the upper portion of a thick permeable layer and beneath an impervious one would make them difficult of explanation in that way. It seems to me that the pressure has produced in the sands distinct traces of a plane-parallel structure, which has favored the movement of the percolating waters in a definite plane, and with this also the deposition of the iron from the water. This structure, I have no doubt, was produced within the sands when frozen. At its base the stratum of sand is closely blended with the till, and although the transition is effected in the space of an inch, there is no sharp line of separation. Above, the stratum is planed down to a horizontal line, the laminae being cut sharply across, and the middle layer of till rests upon the surface thus produced like a plank, with a clearly defined line of demarcation between it and the sands it covers. It is a horizontal fault. It seems to me certain that when the ice moved over this mass of sand, now so yielding and incoherent, the latter was frozen into a solid and rocky mass, and that it was thus eroded and faulted and cleaved, and where the freezing was less entire was swept into the common chaos of the till above. In many cases the upper layers of the till contain well-rounded sand INTERGLAOIAL SANDS. 553 l)(>wl(lci's (I iuclii's in greatest diameter, which can be exphiined only as ;il)ovf 'uuHcated. They are now jtockets of a nuxch flattened elHpsoidal t'orni, fiHod with a l)utf sand hke that fornaing the layer below. 'I'liis sand stratum was again finely exposed in the water-main ditch on Ainit\- street (a few rods north), on a line running east and west, and thus with the dip of the laminse of the sands. It rested, as before, upon the irregular surface of the till below, and was covered here and there by frag- ments of the second till, partly removed in grading the road. The sands were exposed for a distance of 350 feet, commencing at a point opposite the noi'thwest corner of the cellar. Here they began as a thin, gravelly bed, and, the till beneath dipping westward, they soon reached a thickness of more than 6 feet, and their whole depth was not exposed for 60 feet. Then the till rose nearly to the surface for 60 feet, and for the rest of the distance the till appeared only here and there in low hummocks in the bottom of the ditch, until at last the sands ran out to the surface on the slope of the hill between the first and second layers of the till, opposite Professor Crowell's house. The sands agreed in all particulars with those already described in th.e cellar section, presenting the same gradation from a fine, whitish, clayey sand through buff sands to fine gravel, the same flow-and-plunge structure, and false bedding with westerly dips, all in places more or less obliterated in the contortions produced in connection with the deposition of the second till. Again, the ditch almost continuously cut across the same sands, overlain and underlain by till and in places confusedly intermixed with the second till, as it continued north on the North Amherst road up to the western base of Mount Pleasant and 80 rods north of the cellar, where the road goes down a small slope, at the gate of the Mount Pleasant grounds. These sands agree exactly with those before described, and are doubtless a con- tinuation northward of the same stratum which I have traced from College Hall and which here crosses the road and runs eastward into Mount Pleasant. What course it takes from here on is uncertain, as it conforms itself to the irregular surface of the underlying till. It seems to me prob- able that it rises high enough toward the north or to the east to produce the head and strong flow of water in the wells on the ridge mentioned above. That this water sets from the north to the south was shown very clearly by the fact that for 40 feet south of an old well which had been sunk 554 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. under the former Amherst House through the sand stratum (2) and into which the sewage of the house had been directed for many years, the buff color of the sands was wholly discharged and the sands were clotted into a greenish mass, and that this effect extended southward without diminution as far as the excavation continued. On the north the sands retained their buff color up to the well itself This well was located in the middle of sec- tion 5, PI. XII, and went below the bottom of the section. It was clear that the decomposing organic matter had reduced and removed the iron from the sands for a long distance south, and that the sands had thus lost their power of piu-ifying the water which set southward. A colleague who had had abundant opportunity for judging remarked to me that he never liked the flavor of the water in Mr. Burt's well, located a few rods south. I have now traced these sand strata along the western slope of the ridge which joins College Hill and Mount Pleasant from the first section northward more than a mile, with a width of 50 to 350 feet and a thick- ness which for a considerable distance was fully 6 feet. They run farther north and south — how much farther can not be said. They appear with undiminished thickness in the northern exposure and may be seen in the Central Railroad cut on the south. They crop out in a slope produced by later erosion, and the position of these sand strata between layers of till seems to me to have in part determined the position of the Lincoln avenue plain from the Northampton road to the Agricultural College and north- ward. Vertically one can see in every section how the sand has been scalped by the ice, and when one considers how exceptional a grouping of favorable circumstances must have been required to shield these inco- herent and exposed sand beds beneath the ice and retain any portion of them intact, one will, I think, be inclined to consider what remains as but a feeble remnant of the beds as originally deposited. Again, the texture of the beds, the large scale of the cross-bedding, the flow-and-plunge structure, and the close resemblance to the flood deposits of the valley in later times, make it probable that they were of similar origin, the one being deposited in the flood waters subsequent to the first retreat of the glacier, while the other and later beds were laid down by the floods which accompanied the final melting of the ice. The sands then furnish strong evidence, if not conclusive proof, of an interruption in the continuity of the presence of the ice in the valley and of INTEEGLAGIAL SANDS. 555 its retreat fnun the seaboiinl to a point north of Amherst, while the abun- dant infoi-niation conceniiug the character of the glacial deposits in Maine and New Hampshire, ])ublished by Prof. C. H. Hitchcock, incline me to the opinion that the recession continued at least to the foot of the White Mountains. The middle layer of the till (3) in the cellar section is a compact, stony clay, showing" no distinction in color, compactness, or texture froin either the upper or the lower layer when exposed in fresh section. When frozen it showed itself a little more sandy toward the north end of the section, evi- dently because it had borrowed part of its material from the sands upon which it rests. Its sharp horizontal line of demarcation from the sands below I have already described. Its upper surface is, on the contrary, most irregular. It sends many long, tortuous projections into the sands above, which are bent over and spun oiit southward as the smoke of a chimney is by strong wind, and indicate clearly the direction of the motion of the ice. This structure is more manifest in the section itself than it can be made in the drawing, and recalls the "fluidal" structure of many volcanic rocks. Oftentimes filaments of the di'ift lie wholly inclosed in the sand, strung along in the direction and in the prolongation of one of the pi'ojections, from which they have manifestly been separated. The upper layer of sand (4) is about 1 foot thick, and is somewhat finer than the average of the lower stratum — about one-fom-th inch — but agrees with it under the microscope in degree of rounding of the grains. It shows nowhere distinct traces of its former texture, this having apparently been wholly replaced by a fine horizontal lamination, which seems to me rather a pressure cleavage superinduced by the weight of the ice upon the mass when frozen, while below it is confusedly interwoven with the till on which it rests. Above it joins the third layer of till along a line nearly horizontal, although the sand and the till are thoroughly molded together. This is a second horizontal fault. Toward the north end of the section it ends abruptly, being cut off at right angles to its length, and the layers of till above and below it come together, separated only by a thin seam of sand, which in places disappears entirely. Distinct traces of a second stratum of sand were to be seen in some of the other sections I have described, and while the sand and second till were often so confusedly interwoven that all indications of a second sand 556 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. bed might well have been obliterated, I am inclined to think that such a second layer was deposited on a second till. At the same time I explain this second layer in the cellar section as a sheet of sand which originally formed a part of the lower stratum (2), and which, when frozen, was moved as an immense bowlder into its present position. Thus it would be classed with the true sand bowlders — regular flattened elipsoidal sand pockets with their longer and shorter diameters averaging about 8 and 4 inches, respec- tively, which occur not rarely in both the upper layers of the till, and which can be explained only by supposing them rounded to their present form when compactly frozen. Another curious phenomenon which points in the same direction was observed near the south end of the west wall of the cellar. A fissure had opened an inch and a half, commencing at the upper surface of the sand layer and running down through the middle drift layer and for a little way into the lower sand stratum, and this fissure had been filled with alternating layers of clay and sand, about seven in all, which correspond on each side of the center and present a curious imitation of a mineral vein. (See figs. 4, 5 of PI. XII.) This would seem also to find its explanation most natu- rally in the assumption that the sand layers (2 and 4) were frozen when the fissure was formed and that the latter was kept open so long that successive lavers of muddy water trickled down tlu-ough it. The sand bed abounds in small masses, 1 to 2™" in size, formed of a few grains of sand cemented with limonite. The upper layer of the till (5) differs in no respect from the lower, and, 'like it, was removed with chisels and heavy sledges. It had a thickness of 5 feet in the section, but the ground had been lowered here by the same amount, so its whole thickness was more than 10 feet as it appears a few feet west of the cellar. The massive and compact character of the stratum was shown by the effect upon it of the extreme cold of the winter of 1879. The mass above the sand expanded with the freezing so energetically that it projected like a cornice 10 inches in the west wall of the cellar, which had been cut away vertically. I am thus inclined to explain the phenomena I have described by assuming that after the deposition of the first till there was a retreat of the ice, during which heavy sand beds were deposited in the valley, followed • by a second advance of the ice, which then plowed up and destroyed the INTERGLACIAL SANDS. 557 greater portion of these beds, even luoviug- iuid partially molding into the till beneath it great sheets of the sand, as in the case of the bed (4) just described, more commonly destroying its identity entirely. Nevertheless, I think one would be strongly inclined, from a study of the cellar section alone, to assume a second retreat of the ice for the formation of the second sand bed, and a third and final advance, during which the third layer of the till was deposited. Or, finally, one has an alternative ; namely, to explain all these sand beds intercalated in the till as deposited by subglacial streams during the progress of a single glaciation of the country. The fact of a retreat and second advance of the ice seems abundantly proved for western Em-ope, and many observations in this country point in the same direction, especially those made toward the borders of the ice sheet, since traces of a double glaciation would naturally be more abundantly preserved there than farther north, whence the ice a second time occupied the country in such force as to obhterate most traces of the incoherent deposits made in the interim. Again, the compact, unsorted, and clayey character of the till above and below the sands shows that for the most part there was here no free ckculation of the waters below the ice, and we should expect the waters to have escaped along the bottom of the valley and not along its side 300 feet above the bottom. On the other hand, the sand beds occupy just the same position fringing the valley and have just the same structure as the flood beds which attended the final disappearance of the ice, and seem to me to bear the same relation to the retreating ice of the earlier epoch. ^ In 1881 the deep railroad cutting south of College Hill exposed the same sand beds at a distance of 1,463 feet south of the first locality cited above on the Northampton road, displaying the following section: Section in railroad cutting south of College Hill. I II III Stratified gravel 0to6 6 Laminated clays 6 ^ Gravel 3 4 Till, olive-green to brown *5 6to8 Sand - lto3 4 Till, blue 1 to3 4 Sand 3 6 Till, blue; bottom not exposed. 1 1 prefer to leave this section as it was written in 1879, although now the case in favor of a second Glacial epoch seems to me less strong than then. 558 GEOLOGY OP OLD HAMPSHIRE COUNTY MASS. The lower sand bed maintained a constant thickness for 450 feet east and west, going below the surface at the east end of its exposure and dis- appearing at the end of the cutting (opposite the northeast corner of Col- lege Grove) with the same thickness. It agreed in all particulars with the lower sand in the cellar section above. The upper layer of sand was exposed for 325 feet east and west, measured back from the end of the cutting. It is greatly contorted and twisted in every conceivable way into the upper layer of till, and in one place it is wholly interrupted for 65 feet and appears in long patches and filaments of sand, one above the other, in the mass of the upper layer of till. At one place also a third layer of sand is intercalated in the mass of the till halfway between the two more extended layers of sand, with a thick- ness of 3 J feet and a length of 80 feet, and ending abruptly. This seems also to have been a great slab of frozen sand from the lower bed, while the extent of the upper bed here makes it possible that there were two inde- pendent sand beds deposited, which it did not seem necessary to assume from the former sections. THE UPPER TILL. I have called the stratum below the interglacial sands the first till, as the product of the first glaciation, and that above the second till, it having been formed during the second advance of the ice, reserving the name upper till for a deposit to which Prof. C. H. Hitchcock has called special attention and to which he has given this name. It is conceived by him to have been derived from the material taken up into the mass of the ice itself, and to have sunk down, when the ice melted, in a coarse, uncompacted, and unstratified sheet upon the lower till, which had been compacted beneath the ice. While the lower till is compact, with few small bowlders, well scratched and not far-traveled, and is bluish in color, having been protected from the air by the thick ice, the upper till is loose, contains many large bowlders, angular and far-traveled, and is reddish from oxidation. I have been able to recog- nize this distinction only partially in the valley. The immediate deposi- tion of the Champlain clays upon the surface of the lower till proper at several places in the valley shows that the upper till was not uniformly spread upon the latter, and where, as along the northern part of High street in Amherst, an upper loose bufi" layer from 1 to 6 feet thick covers the blue compact till, I do not find the bowlders to be more angular or far-traveled than below, and am inclined to explain the peculiarities of the surface layer BOWLDERS. 559 as due to the surface oxidation and disintegTation by frost. The same was true at the Central Raih-oad cutting south of the college, where beneath the clays the till was in its upper part olive-green to brown, and blue -green below, but with no further distinction in matter of compactness, coarseness, or derivation of bowlders. I have seen several cases where the up^^er layer was blue and the lower reddish. REMARKABLE BOWLDERS. President Hitchcock has described^ and named six of the most notable traveled bowlders to be found on the east of the river in the valley and on its border, remarking that bowlders of the largest size do not occur in this vicinity. Those named are: THE NORTHERNEK. An irregular mass of the coarse conglomerate of Metawampe ( Mount Tom), weighing nearly 100 tons, * * * lodged on the gneiss rock of Pelham Hill in the bed of a large brook close by a small cascade, where it was pointed out to me by Mr. Newall. ROCK OREB. Near the top of Mount Warner, a little east of the summit and in the cleared pasture, lies a large bowlder of imperfectly prismatic trap or greenstone. Its weight we estimated at 78 tons. The rock of the mountain is granite and mica-schist, and no trap in place is found to the north till we reach the north part of Sunderland and the south part of Deerfield, say some 10 miles distant. From that range this bowlder undoubtedly came. * * * ROCK ETAM. Northwest of Rock Oreb, say a quarter of a mile in the woods, and far down the northwest slope of the mountain, is another and larger bowlder of the same variety of trap. We estimated the weight of the Hadley Btam to be 385 tons. * * * Another bowlder of the same columnar trap projects from the ground on a lower bench of the mountain southeast of Eock Oreb. Its exposed portion is half as large as the latter. THE MAGNET. At the western foot of the steep part of Holyoke, and a little south of the place where the railroad goes up the hill, lies a large bowlder of trap precisely like those just described on Mount Warner. It is 15 feet high, and, by a loose estimate, I think it must weigh 300 tons. It is remarkable for exhibiting on its north face a vast number of magnetic poles sufficiently strong to completely invert a common magnetic needle, forming in fact several continuous lines of poles. I spent some days several years ago in tracing them out. These facts furnish a reason for the name which I venture to propose for it, viz. The Magnet. I formerly supposed that 'Reminiscences of Amherst College, pp. 264-265. 560 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. it had been broken off from the trap ledges immediately above it, but its exact correspondence with Oreb and Etam in characters, and want of resemblance to the trap of Holyoke, make it more probable that it was brought into its present position by drift agency and originated in the same region as those on Warner. THE SENTINEL. As we ascend Mount Boreas, looking northerly up the valley on its east side, * * * we see a prominent bowlder lying near the base. We find it to be com- posed of gneiss and lying on gneiss, although the stratification on both is very indistinct. It weighs something less perhaps than 200 tons, although not accurately measured. THE KOCKING STONE. Some years ago a bowlder of several tons weight, capable of being rocked a little by one man, lay on a farm then owned by Mr. Grout, about a mile north of Pelham Center, on the road to Shutesbury. OTHER BOWLDERS. The finest bowlder ever found in the valley is the one now lying" in front of the Woods cabinet, where it was bi'ought by the class of 1857, as the inscription npon it indicates. Its former north end now faces south. (See PI. XXXIII.) It was uncovered in lowering the road in front of the residence of the late Edward Dickinson, and, judging frona the excavations here for the waterworks, it was derived from the lower till. It is a large, coarse, red sandstone, in size 78 by 66 by 33 inches, the four sides planed down to a flat convex surface and striated longitudinally, the ends for the most part still rough and irregular. It exhibits exactly, on a large scale, the form of the most perfectly polished glacial stones. The striae of the upper surface when it was first exposed ran north-south, as do the striae in the valley, and it may be that the ice passed over it after it was fixed in the till, thus polishing its fourth side, which was naturally at first mistaken for a ledge. A full description of it was published by President Hitchcock.-^ The largest specimen of the buff quartzite, which is so abundant in smaller masses throughout the valley, is the one mentioned on page 542, in the yard of the Whitney homestead, on King street, in Northampton, which came from the Denniston place, near Florence. This quartzite, I think, came into the valley farther north from Vermont and then drifted down in the valley with the altered direction of the ice. 'Am. Jour. Sci., 2cl series, Vol. XXII, 1857, p. 397. BOWLDERS. 561 A siujilc l)()\vl' up I'nnu tlio terrace (tluf l)ank of tlie river when the teiTace was part of its bottom), and limited toward the river by a descending scarp which was the river bank at a later time, generally when the terrace made i)art of its flood plain. Corresponding to this, we shall find the plain covered with the meadow loam laid down in the floods of the river, and under this loam the strong river-bottom sands of an earlier date, the last underlain unconformably by older deposits which the river had not reached and eroded while it flowed above them. On the other hand, we shall find the highest terrace or bench bounded outwardly by a slope which, as to its material and structure, has no relation to the river. At most, the river has undermined this more or less exten- sively at the water level, and, by caving, an escarpment of till, sandstone, or gneiss has resulted. The terrace itself, widening iiito extensive sand or gravel plains where the alluvial cones or deltas of the side streams were thrust out into the lake, narrows in places remote from these, and its level is often represented by shelves in the sandstone scarcely covered by sands, or in the till deeply concealed by gravels concentrated from the till itself. Pursuing the same level, we soon come upon the continuation of the normal sand beds which make the bulk of the bench. Inwardly, however — that is, toward the center of the lake — especially around all the Hadley basin and its prolongation in the Deerfield and East- hampton valleys, the terrace is for the most part bounded, not by an escarpment of steep and constant pitch — an abandoned river bank — but by the slope of passage from shallow to deep water. This is sharpest and most constant on the face of the large deltas (but here of less angle than in the former case, as the highest angle at which sands come to rest under water is less than that assumed in air), less and less marked in other places, until at last the case occurs where from the rocky bank the sands pass with gentle and continuous slope to the deepest central line, where was the thread of the current, and rise in the same way to the opposite bank. This slope of passage I have called a scarp of deposition, or, as locally synonymous therewith, the delta front, in contradistinction from the ordi- nary scarp of erosion. On the map the normal high terrace or bench (1 s h, PI. XXXV) and its widening into great delta flats are not separately indicated. One passes by a scarp of deposition to the broad area of the old lake bottom (1 b t), which was synchronous with the bench itself. 614 GEOLOGY OE OLD HAMPSHIEE COUl^TY, MASS. The three great water areas indicated ah-eady were sufficiently broad and sufficiently separated to justify one in calling them lakes, and these two terraces would then be called the lake-shore and the lake-bottom deposits (1 s h and 1 b t). This is further justified by the lakelike mode of accumu- lation of the sediments in these areas, and allows me to use the term "old river bottoms" for the abandoned beds of streams in old oxbows. The second terrace or the old lake bottom, unlike the other terraces, is a surface depending, not upon the level of the water at the time of its forma- tion, but upon the water level and the amount of material. The valley widens southwardly, which has the same effect as if the supply of material decreased in this direction. As a result, the lake-bottom level sinks gradu- ally as one proceeds toward the south relatively to the lake bench, w the deposition scarp which sejDarates the two increases in height. The third terrace, counting from the shore line, is generally the uppermost flood plain of the normal Connecticut River. This brinsrs about the curious result that the second and third terraces change places as we go south, the change taking place between the Mon- tague and the Hadley lakes; that is, the Montague Lake was a fiUed-up lake, and as we go inward from its shore line we pass by a slight scarp of deposition to the remnant of the lake bottom at a level but little loAver than that of the bench itself. We descend next by a scarp of erosion to a marked terrace (t*) that crosses the northern line of the State with a heiffht of 310 feet, which I have often called the intermediate terrace or the Lily Pond terrace, formed during the early decline of the flood by the rocky barrier at the Lily Pond in Grill (see PI. XXII, p. 724). This is at times broken into two or more terraces. We descend then finally by an erosion scarp to the group of terraces but little above the present flood plain of the river, and still lower to the incomplete terraces which lie below that level, both which groups have been formed by the river in its j)resent size and condition. The intermediate terrace (t*) was thus excavated in the lake-bottom beds — that is, inside the lake bottom. Farther south, in the broader Hadley Lake, the filling had not pro- gressed far enough to obliterate the lake, and the equivalent of this third terrace (t*) is found as the first terrace below the bench, generally slightly marked and excavated in the upper portion of the deposition scarp which had connected the shore and deep-water deposits of the highest floods — that FLOOD DEPOSITS IN MONTAGUE BASIN. 615 is, outside the lake-bottom deposit. In the first case the order is, (1) lake bench, (2) lake bottom, (3) 310-foot terrace (t*); in the second, (1) lake bench, (2) the contiiuiation south of t*, (3) lake bottom. In other words, the Hadley Lake continued through the whole period, and its lake-bottom beds are, strictly speaking, a little later in age than those of the northern basin. It seems by far best to represent all the lake- bottom deposits b}^ one color, as I have done. DETAILED DESCRIPTION OF THE FLOOD DEPOSITS IN THE MONTAGUE BASIN. This description is in continuation of the interesting account of the terraces of the Connecticut in New Hampshire given by Mr. Wai-ren Upham in the Geology of New Hampshire, Vol. Ill, page 19. I may men- tion that I accept the criticism of Professor Dana^ of the view taken by Mr. Upham, that the deltas thrust out into the main valley are often above the highest "normal" terrace of the flood time, and consider these deltas as marking by their levels the true height of the flood waters, and look upon the lower level of the highest terrace which connects these deltas as explained by a lack of material in the intermediate spaces. I can not, however, accept the other criticism of Professor Dana that the esker traced down the valley by Mr. Upham has no existence as an eai'lier structure antedating the flood gravels of the open valley. The Montague basin is narrow — about a mile wide — where it enters the Warwick quadrangle in Vernon (PI. XXXV, C), and it retains this width across the area, connecting at the highest water stand westward around Mount Hermon with the northern or Greenfield lobe of the Hadley Lake. As it enters the Greenfield quadrangle at Millers Falls it widens to above 6 miles, and is connected again at flood level by several narrow passes in the trap ridge with the northern lobe of the Hadley Lake at Greenfield. It connects by the narrows at Sunderland with the main Hadley basin. It was a nearly filled-up lake. The main stream quite filled its rather narrow valley down to Millers Falls, where it widens, and here the heavy contribu- tions of the Millers River filled the whole widened valley. The distinction between the shore flats (1 s h) filled to the highest effective level of the 'Am. Jour. Soi., 3cl series, Vol. XXII, p. 431. 61 6 GEOLOGY OF OLD HAMPSHIRE COUl^TT, MASS. waters and the unfilled portions (1 b t), 10 to 50 feet lower, can be clearly made oiit. Karnes rise out of the shore flats, which are often kettle-holed and plainl}^ deposited in the presence of remnants of the ice ; but there are no continuous and important "moraine terrace" beds fringing the eastern rocky slope and raised above the flood level of the waters. These shore- ward plains sink by easy construction scarps to the bottom flats in which the erosion terraces (t* to t*) have been cut. THE NORTHERN LOBE OF THE LAKE. From the hill which overlooks the hotel in South Vernon, Vermont, just on the State line, one sees the river for a long way northward flowing in a narrow channel bounded on both sides by high lands which slope rapidly to the stream and leave place for only narrow terraces. Nearer, the sand flats spread westward from the river around the base of a prominent hill (t) which rises to the north, and bending north suiTOund this hill. The sands are very thick, and seem to rise a little above the highest probable flood level for this latitude, about 400 feet, which would indicate that they were brought in behind this hill while the ice filled the main valley and were not wholly planed down by the later stream. Around the south spur of the hill east into the open valley the sands sink rapidly to the lake bottom at 322 feet, as they failed to receive further protection in the lee of the hill, and the plain of fine sand sinks riverward to 307 feet, and is continued in a remnant which lies just north of the station with a height of 297 feet, cut off from the rest by an old channel of the river. The old lake bottom commences again just opposite the hotel in South Vernon, it having been cut away by erosion at the State line, and extends southward as a broad, level plain, down the center of which the road to Bemardston passes. On its outside it rests for more than a mile against the rocks, which rise first abruptly and then more gradually, and present a rugged and irregular surface, thinly covered by loose till. On this surface the river has deposited nothing. Where the Bernardston road mounts from t* to the top of this terrace a section showed — Feet. 1. Very fine, loamy, iinlamiuated sand 6 to 8 2. Well- washed granite gravel, pebbles one-fourth inch 7 3. Fine sand in great thickness. THE BENNKTTS UUOOK I'LAIN. 617 The upper striiiiun extended over the wliole surface of the plahi and seems to have l)een deposited when the river reached this level only in its floods. FoUowino- down this jjlain (1 h t) for more than a mile one is con- fronted by a great escarjjment which stretches obliquely across the road from the rocky hillside to the river bluff, and rises 80 feet above the lower plain, or 380 feet above the sea, and reaches 400 feet when it rests against the rocks. Seen from the hills across the river, its upper edge is sharply preserved and its horizontally fluted slope is clearly a portion, preserved intact, of the riverward face of a great submerged bank and not a stream- cut scarp. The road rises to the surface of this high plain next to be described. THE BENNETTS BROOK PLAIN, OR MORAINE TERRACE. The plain stretches far southward into Bernardston and Grill, expanding rapidly to more than a mile in width. It is the true high terrace or bench (1 s h) of the Montague Lake. The surface is as level as any river terrace for more than a half mile back from the edge overhanging the river, and for a long way south. A small reef of rock projects above the general surface near its northern end, and the gravel is scooped out in front and along the sides, the grooves running out southward into the common level of the plain exactly as the sands are hollowed out around the pier of a bridge. Shallow, empty watercourses run over its surface and toward the river. With these exceptions the plain shows a true level as one rides along the road or crosses it at any point going east toward the river, a distance in many places more than a half mile. If, however, one goes westward to the mountain, taking, for instance, the field road to A. Whitehead's, north of the Lily Pond, in about 100 rods flat hollows begin to appear, at first only 5 to 6 feet deep and 20 to 30 feet in radius, but growing deeper and closer together tintil the whole surface is covered by regular kettle-holes about 20 feet deep and separated only by narrow ridges which rise everywhei'e just to the level of the plain, and the road goes up and down as if it were built along the edge of a saw. Farther on the cols between these hollows grow lower, and by degrees the kettle-holes merge into broad, irregular depressions, several of which are occupied by ponds 50 to 75 rods long and about 40 to 50 feet below the general surface. These ponds were almost wholly dried up in the dry time when I examined them, and showed 618 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. flat sand bottoms out over which peat meadow was spreading. The road we have followed ends in a depression, elongated in the direction of the old shore line and extending to the mountain, which has been further hollowed out by the brook that now runs in it. The striking peculiarity in connection with the appearance and grad- ual development of this system of kettle-holes is that they are excavated in a quite level plain, and from a distance one would have no suspicion of their existence. At first they do not interfere with the manifest levelness of the surface, and as they grow deeper the ridges between them are flat- topped and of the common level, and only as the depressions are crowded together do the ridges become at first sharp-edged and then sink into passes between the hollows, until, against and running southward parallel with the mountain, there is a broad space where almost everything has sunk irregu- larly below the common level. Along this line the surface of the plain is made up of finely rounded gravel, with cobblestones 6 to 12 inches across, and the exposures in the roadside where the highway descends on the north are of the same material for perhaps 20 feet downward from the surface. Farther south Bennetts Brook runs across the plain to the river, at the bottom of a gorge 140 feet deep, bounded by a single steep sand slope on either side, without as yet cutting down to the ledge. Following the northern road down the slope to the ferry, one finds that the great plain is here, on its front edge, also made up above of finely rounded gravel of great thickness, consisting of cobblestones 6 to 12 inches in length. Below a point 80 feet above the river, or 280 feet above the sea, fine, horizontally laminated sands underlie the gravels, and similar fine laminated claylike sands appear at the same height in the road going southwest up from the same meadow. The surface of the bench remains . unchanged to and beyond the railroad crossing. Here, just on the south hue of West Northfield, the configuration of the surface was originally much modified by the great quadrangular mass of Mount Hermon, Avhich rises in the midst of the plain. The surface of the latter was depressed by the sweep of the waters around this obstruction, especially on the west, where they entered the narrow passageway between this hill and the border of the basin, a passage through which the road and the railroad now go, and this is expressed by the sinking of the plain eastward from 375 feet at the railroad crossing to about 330 feet at the eastern brow of the terrace. JUNCTION OF MONTAGUE AND HADLEY LAKES. 619 Nearer tin- ohstriu'ting lull a brook lias taken advantage of the depres- sion and, as hai)pens very often in similar circumstances — so often, indeed, as to make it tlu' rule — has worn down between the hill and the terrace gravels, slipping down, as it were, upon the northward-sloping side of the sand- covered drmulin and eroding for the most part in the sands of the terrace. The plain is, however, clearly continuous through this pass around the west side of tlic hill; on the east side it has been removed or terraced down to lower levels by the river. A distinct esker ridge, elevated about 20 feet above the level of the plain, and older than it, runs along southeastwardly through the pass and near the mountain side, ending opposite the second crossing. South of the village of Gill for a long distance lower terraces abut directly upon the steep rocks, and only traces of the high terrace bench occur where the road comes down from the hill to the Noi-thtield Farms ferry. A section in the latter showed about 20 feet of clay, its surface about 70 feet above the river and 150 above the sea Above this are sands. From this point the vertical rock wall of the canyon bounds the river, and the high terrace disappears, except in traces, until one reaches Turners Falls. THE EXTENSION OF THE FLOOD GRAVELS WESTWARD THROUGH THE BERNARDSTON PASS TO JOIN THE HEAD OP THE HADLEY LAKE IN THE NORTH OF GREENFIELD. North of Mount Hermon the mountain side, against which on the west the great bench we are following has rested, swings abruptly westward and continues — maintaining its height — westward to Bernardston. Between this hillside and Mount Hermon the gravels of the Bennetts Brook plain extend westward through a narrow pass, 200 rods wide, which I have for conveni- ence called the Bernardston Pass. This is occupied by the highway and the railroad running past Bernardston. The gravels are naturally lowered in the narrow portion of the pass, but rapidly regain their high level of 392 feet as the pass widens southwestwardly and the sands expand into a broad, very level plain which widens north up the fiord valley of Dry Brook. It doubtless owes its abundant sands largely to the great stream which flowed down this valley, and its freedom from kettle-holes is due to the fact that this stream continued to flow after the main current had ceased to flow westward through the pass and the ice had completely jxielted away beneath the sands. 620 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. On the south the plaui ends very abruptly over the basin in which lies the village of Gill, and Dry Brook has been kept up to the level of the plain by a reef of schist which rises in its front edge and over which the brook falls rapidly into the rounded valley below. The latter is a high basin, with sides and bottom of till, and how far it has been filled up to the flood-plain level and then cleared otit again by Dry Brook and its many tributaries it is hard to decide. Westward across Dry Brook the high plain is soon again supported on the south side by high ground, as well as on the north, and soon begins to develop kettle-holes and merges into a kame area exactly as described in the last section. Its sm-face dips very shghtly westward, it beinp- 396 feet hidi at its eastern side, 392 feet at its western border, where it begins to break up into kettle-holes, and 389 feet farther west in the middle of the kame area. To one looking down on this broad area of intricately reticulated gravel ridges, short kames, and interrupted plains, the whole forming a typical "kame landscape," it seems clear, from the configuration of the sur- face and the trend of the broken ridges, that the cim-ent flowed west into the Greenfield basin. A restored surface carried through the highest por- tions of the ridges sags along the middle and cuts the high ground north and south like a shore line. The material also in the pass consists largely of pebbles — mostly under 6 inches in diameter, but some a foot long, in part quite well worn but in part only battered — of the common gneiss and quartzite which abound in the main valley farther north. Westward beyond the narrows the gravels grow much finer. In the western part of the kame area in and south of the village of Bernardston the pebbles are almost exclusively of the dark mica-schist and the black argillite which occur wholly northwest of this point. This is notably the case in the "Bernardston picnic grove," south of the railroad station, where is the north end of a continuous esker which extends a mile or more southwest into the Greenfield basin. Here the pebbles rarely exceed 4 to 6 inches, and are as finely worn into flattened discoid and ovoid forms as on a sea beach. A kame ridge where the road branches north, just before it goes down over the bridge to enter Bernardston village (opposite E. M. Slate's), gave THE OLD COUESE OF FALL KIVER. 621 this section: Coarse sand and gravel, 2 feet; medium buff sand, 4 feet; fine, even-bedded sand, 5 feet. Tlie ridge was 20 feet wide and the layers crossed it horizontally, as if they had been eroded on either side. A little farther west, down the hill toward the bridge, the gravels were found to be coarse and scarcely bedded at all. Everything shows that the floods swept west through the Bernardston Pass and, joined by the waters coming down the extensive upper valley of Fall River at Bernardston village, passed into the Greenfield basin. THE OLD COURSE OF FALL RIVEK. ' Commencing high up in the valley north of Bernardston, Fall River is bordered by a broad, flat plane (1 P) that has been cut in a heavy sand deposit which once filled the bottom of the valley, forming the lake bench, and which in part still remains intact on either side of the alluvial bottom of the river. At the bridge in the village of Bernardston this plain leaves the river and skirts the west edge of the kame area, being bounded on the west by West Mountain, and extends southwesterly into Greenfield, where it merges with the lake bottom of the Greenfield basin, above which the esker ridges project for a distance and then are finally submerged. The river, on the other hand, does not follow this lower plain, as would seem natural, but runs from the bridge due south, in a deep, narrow channel, cut in the much higher kame area, and then among the drumlins and the sandstone ridges to the Connecticut. It seems to me certain that when the waters of the Connecticut became confined to the main valley to the east, the stream coming down the Ber- nardston Valley continued to run southwestward by the now abandoned channel, and cut down and flattened the kame material into the broad, flat, lakelike watercourse which now remains, and which forms now, near the town line, a low divide from which a small brook runs back into the main stream and another on to join the Leyden Brook, in the west of Greenfield. The continued melting of the ice beneath the kame area at last di-opped the sands so low that the stream suddenly found a new course opened to it directly south into the Connecticut. It can be clearly proved that the flow of the glacial waters in the Bernardston Valley commenced long before the ice had lowered so far that any connection with the main valley can have existed, for the esker (k) 622 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. that starts at the mouth of Fall River Valley, in the Bernardston picnic grove, runs south, rising over a col between two drumlins east of the road to Greenfield, at a much higher level than the plain to the west, here described as the old course of Fall River, which was at that time still filled with ice. It can be seen further from this section that the abundant flow continued after the waters had ceased to flow from the main channel through the Bernardston Pass. An inspection of the map will make it clear that the deep Fall River Valley must for a long time have been a main artery of drainage. When the waters went through the Bernardston Pass the ice had mostly melted far north up the main valley, but a remnant was submerged beneath the sands along the border of the stream in West Northfield, causing the kettle-holes of the western border of the Bennetts Brook plain (see p. 617), and through the pass the waters spread their gravels over a, considerable but gradually diminishing body of ice. At the same time the great volume of water which came down the valley of Fall River also flowed over ice, and thus were formed the esker ridges of argillite pebbles which project out from this valley and blend with the gneiss gravel brought through the pass from the main valley. A great mass of ice filled the basin of Grill, and thus completed the walls of the pass and prevented the flood from filling this basin, as they naturally would have done. When the flood had so far receded that the waters" of the main stream no longer went through the pass, the waters of Fall River continued to flow into the Greenfield basin, carrying a large volume of the kame sands southward into this area and smoothing out the broad plain which still extends between the two, until, by the sinking of the ice, its southeastern border was breached and it found exit across the kame gravels south into the drift region of Gill by way, apparently, of its reopened pre-Glacial bed. THE BENCH ON THE EAST SIDE OF THE RIVER IN NORTHFIELD AND ERVING. The hills are set back on the east side of the valley at about the same place as on the west side, and the high sands expand eastward across Hins- dale and the corner of Winchester, in New Hampshire, up the valley of the large Perchee Brook and continue southward with a width of 200 to 400 rods across Northfield and. Erving. The rock surface is everywhere THE BENCH IN NORTHFIELD AND EEVING. 623 quite hig'li, often up to or iil)ove the 300-foot contour, and the layer of till above this is generally thin ami not molded into drumlins as on the west side. Hence the bench sands are generally not of great thickness. They repi'csent mainly the deltas of Perchee and Northtield brooks. At the head of the deep recess formed by the southwestward trend of the valley's rim in the corner of Winchester, New Hampshire, is the apex of the delta of Perchee Brook (1 s h, PL XXXV, C), at 392 feet above sea. It consists of coarse deposits, with many rounded bowlders of porphyritic granite, even up to 2 or 3 feet in diameter. The brook runs at the foot of the rocky ridge nearly to the State line, and all its delta is on its south side. From its apex two roads run toward Northfield. The eastern runs south at the foot of the cliffs and marks the eastern shore iintil, at L. Lyman's, it turns into the plains toward Northfield street. The western follows the brook until, just over the State line, it goes down from the bench to the next terrace level (1 f) at 320 feet. Between these two roads runs a great island of till in the midst of the delta plain. Just at the foot of this hill, on the side facing the head of the delta, is a triangular pond, 800 feet on a side, depressed 30 feet below the level of the plain, its concave base embracing the island and its apex pointing toward the head of the delta. From the other end of this island a sandy esker ridge (k) extends southwest for a long distance, and just south of Mr. D. L. Moody!s main school building a cutting showed about 10 feet of well-sorted sands; but I was informed that a little below coarse bowlder beds occur. What is most remarkable in the deposits of this delta and its continua- tion south in the high terrace is the great accumulation of fine sand. Soon after leaving the hills the brook has cut deeply into these sands, and all the brook sections in the neighborhood are in like material. Following the brook down to where it descends sharply over the rocks to the river plain, these sands are seen to rest on clay at a height of 290 feet above sea. Here a line of springs marks the base of the sands, and immediately below aban- doned clay pits occur, as they do southward at various lower jDoints in a gorge cut by a tributary of this brook and farther south by the roadside, showing the clays to be continuous below the level of 290 feet. Following the terrace southward, shallow depressions begin to appear in it, and oppo- site the village street it has developed abundant well-formed kettle-holes 624 GEOLOGY OP OLD HAMPSHIRE COTOTYTlrASS. and is made up of coarse gravel, containing cobbles up to 6 inches in length. Its front edge has a height of 360 feet where it sinks down by a steep scarp to the level of the Northfield village plain (1 b t). At its foot a brook runs noiihwest into the Connecticut, which has cut a notch in it, but has made no delta projecting out onto the terrace below, showing that when it was effect- ively eroding the main stream was also strongly eroding, and carried on all its contributions. This is the case, also, with all the tributaries down to Millers Falls. South of this brook the high terrace (1 s h) is continuous, but narrows rapidly, and by the side of the road going up to F. Johnson's, just north of the single di'umlin marked on the map, a section occurs in coarse gravel much contorted. From this point the great sand masses of the next lower level — the old lake bottom (1 b t) — which are here nearly a mile wide and extend southward for over 2 miles in the great "Beers Plain," have been thi'own up in a wilderness of sand dunes, thus obliterating almost all trace of the scai-p which once connected the two levels. The plain of Northfield village, at the third level — 305 feet (t*) — is thinly covered with sand. Immediately below is till or ledge, but south- ward the rock lies much lower, while the level of 300 feet is maintained by a great volume of sands. Southward these sands rest upon finer material. Just over the railroad, on the road west from the station, 20 feet of coarse sand, dipping S. 20°, rests upon very fine, horizontally bedded sands with a single layer of fat clay 18 inches thick. The former stratum was laid down while the stream was forming the terrace (t*); the lower is the uneroded portion of the lake-bottom beds. Their present eroded surface is 250 feet above sea, and they are exposed with a thickness of 42 feet, and no bottom is seen; nor do the sands vary. Just south of the village street, where two brooks come together and run under the railroad, the same sands rest, at a height of 270 feet above sea, upon blue banded clays, the fat layers being one-third to two-thirds of an inch thick, and the intervening layers of sandy clay 6 inches thick. Four miles farther south, at the ferry at Gill station, the clay layers are one-half of an inch wide and are separated by layers of fine sand 2 feet thick. Farther south, below Northfield Farms, the Four-mile Brook has cut through heavy clay beds rising about 260 feet above sea. The above figures show that the basin was filled up with fine bedded THE MILLERS EIYER DELTA. 625 sands {intl clays to 290 feet at its north end and 260 feet at its south end, a descent of 30 feet in 7 miles, and the surface of the liigh terrace shows about the same descent. As this slope is wholly inconsistent with the accumulation of thick beds of iine laminated clays, some part of this differ- ence may be assig'ued to a post-Grlacial elevation increasing northwardly, of which we shall lind many other indications. The order of events in the basin seems to have been, in brief, as follows: The broad shoreward gravels of highest level began to be brought into the basin before the last remnants of the ice had been melted, those on the west side largely by the main stream, those on the east side by the tributaries. Then far to the south the great delta of Millers River, as detailed below, was thrust across the narrow outlet of the basin, ponding back the waters and allowing the deposition of the great thickness of fine sands and clays. The coarser delta deposits were continued out over the finer, unconform- ably in a sense, and completed the filling of the valley. Where the highest floods failed to plane the earlier beds down fully they remain as kame ridges. When the floods ceased to rise over these highest flood plains before the ice had wholly melted beneath them the latter are kettle-holed. THE MILLEES KITER DELTA. THE CANYON AND OLD COURSE OF THE CON- NECTICUT. The section of the flooded Connecticut which we have above described might very properly be treated separately as the Northfield Lake. It would include just that portion of the valley which is portrayed on the Warwick sheet. The valley expands at the north border of this sheet, and soon contracts again to the north (PI. XXXV, C). The high terrace which we have followed south along the east side of the valley as a narrow bench of sands applied to the high, rocky valley side, widens suddenly south of Northfield Farms, extends entirely across the valley proper, and abuts on the west against a steep ridge, called Mine Hill. The river does not, as heretofore, erode its channel down the middle of this plain, but escapes southwestwardly from the corner of the basin tloi'ough a deep gorge of its own cutting, between the ridge of crystalline rock mentioned above and the Triassic conglomerate. MON XXIX 40 626 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. A "diy brook" has cut its notch part way across this plain, just west of the raih'oad, trying ineffectually to replace the river, and the contours on the north slope of the plain bend south toward the brook gorge. Millers River, emerging from its jjortal in the eastern rock border of the valley, makes almost immediately a remarkable curve, turning first south and then 180° round through west to north, and then runs north, skirting the Mine Hill ridge, to meet the "dry brook," and then with sharp western turn it cuts through this I'idge to join the Connecticut. This is one of the most beautiful spots in the State. The Connecticut comes down from the north in its vertical-walled canyon, its waters foaming in rapids around the great pudding'-stone bowlder amidstream, still called the "French King," from a tradition that in the old French wars an expedi- tion dropped down the river to this point and a venturesome officer pushed his canoe to the head of the rapids and broke a bottle of wine on the great rock, claiming the land for the French King. The broad stream then bends sharply northwest and flows strongly in its deep gorge, while just at the bend Millers River comes down over the rocks in a picturesque fall, flanked by a ruined mill. The fall has scarcely worn back at all from the mouth of the stream, and the whole impression is one of recency. Looking down on this Montague plain from one of the high hills east of Millers Falls, one easily restores the beds eroded by Millers River, and then the plain is seen to be the northern j)ortion of its great delta, expand- ing northward up the narrower part of the valley of the Connecticut. In following this plain down from its north end, opposite the point where the main stream enters its rocky gorge, a distance of about a mile, one finds that the sands grow coarser and coarser and grade into gravel, and opposite the point where Millers River leaves its rocky canyon in the eastern wall of the valley — that is, at the head of the delta — many of the beds are of very coarse gravel alternating with sand beds, showing the coarsest flow-and- plunge structure. Moreover, the plain slopes southward quite rapidly, its elevation being 362 feet north of Millers River and 350 feet south, at points 3,000 feet apart. That the delta deposits of the tributary could have been extended north against the current of the main stream more than a half mile proves that the current of the main stream could not have been very strong, and the southward slope of the surface of the delta indicates that the land was, there relatively depressed toward the north and has since risen. THE MILLEKS RIVEK DELTA. 627 From Xortlitield Farms the Connecticut River I'uns in a canyon, with sandstone on the right ])ank and cr3^stanine rocks on the left, and at the moutli ot' Jlillers River it tnrns west and northwest for about 5 miles to Tiu-uers Falls, cutting off a corner of the Grill sandstone massif, and then runs south, skirting the diabase ridge of Greenfield. It thus gives place for a great ex})ansion of the delta of Millers River, about 5 miles square, a broad elevated sand desert — the Montague plain — which on the south sinks by a marked delta front to the low basin in which lies the ■\allasre of Montao-ue. From Turners Falls back to the moiith of Millers Ri^'er one descends from the north edge of this plain by a single great erosion scarp to the level of the river, or to the sandstone ledges into which the stream has cut, thereby preventing any further erosion of the delta beds. In all this latter distance it formerly extended north across where the river now runs and rested against the sandstone, and above Factory village a broad remnant of it still remains ; and at the mouth of Fall River, opposite Turners Falls, it extended into the basin of Greenfield through the gap in the trap range, and sent a large body of sand by this passage into the Hadley Lake. The river poured with full current through this pass, and it must have been a slight chance which determined it in the direction of its present course and prevented it from choosing a channel down the west side of the trap ridge through Greenfield. The Connecticut River was thus driven westward around the great delta and compelled to cut a canyon between the sandstone and the crys- talline rocks from Northfield Farms to the mouth of Millers River, and in the sandstone on to and beyond Turners Flails, nearly down to the mouth of the Deei-field River. The old bed of the Connecticut runs due south from Northfield Farms past Millers Falls, and thence southwest to join its present bed at the mouth of Sawmill River, in Montagiie. This course is marked by a line of kettle- holes continued in the channel of the dry brook mentioned above along the plain north of Millers River, by the sharp bend of the latter, and by the deep erosion basin that extends south from it. Farther on it is continued by the line of large kettle-holes of which Green Pond and Lake Pleasant are the most important, and by the course of Pond Brook and Sawmill River. Its eastern rocky border is exposed at the falls which give the name to the village of Millers Falls, in the north of Montague, and at the bottom of the deep cuttings of the railroad ju.st below the Millers Falls station. The 628 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. cuttings for the relocation of the tracks of the raihoad running south- west from Millers Falls gave fine sections radiating out from the head of the delta. Besides most instructive sections of kettle-holes, described further on, the opening gave a fine view of the whole structure of the delta (see fig. 41, p. 668). At a point near where the two raihoads sejaarate, the cutting was 20 feet deep and showed the sands resting on glaciated surfaces of gneiss and diabase, without the intervention of till or clays. The section showed an extensive body of sands, often exposed 12 to 16 feet in thickness, and cross-bedded in great sheets which dip south away from the head of the delta and represent the advancing front of the latter. Above this a horizontal layer of gravel, averaging about 3 feet in thickness, and diminishing in thickness and coarseness outwardly, made the surface. This represents the concentration gravel manufactured out of the cross- bedded sands of the delta by the floods of the river as they swept over its surface after its front had passed farther outward. Where kettle-holes had sunk dm-ing the flood time, this gravel thickened below to fill the depression, and had plainly been pushed into the depres- sions from the direction of the head of the delta, the gravels being cross- bedded in their thickened portions, with radial dip. All along the eroded front of the delta overhanging Turners Falls the clays, resting directly on till or sandstone, rise to a height of 250 feet above sea and are capped by the delta sands. The clays have a maximum thick- ness of 59 feet and are thin-laminated, with the layers 1 to 1 J inches thick. The clays change upward into the sands by repeated alternations of sand and clay. At the top of one stratum of clay 1 foot thick a single layer was contorted and compressed into acute folds bent over southward and covered by a foot of sand, as if moved by the friction of the waters by which the thick layer of nonlaminated sand was brought in. All above and below was undisturbed. The illustration, fig. 35 (p. 629), indicates the relation of the beds at the large brick pit south of Turners Falls. The delta sinks southward into the deep land-locked hollow in which IS the village of Montague, and along the bald face of the mountain to the east of the village the tei'race is represented only by a narrow bench cut in the till, and farther south cut in the high sands which fill the Mount Toby THE HADLEY LAKE. 629 e-oro-e. The hi>'li hill of sandstono which rises west of the village is cou- nected sox;th with iMouut Toby by a sandstone ridge at about the height of the hio'li tiMTacc, and it is therefore certain that the old bed of the Con- necticut can not have gone, as an inspection of the map would suggest, due southwest to join the present bed at the Sunderland line. The Montague depression may have been eroded by the pre-Glacial Connecticut in a great bend directed southward. It was more probably cut out of the soft sand- stone by the ice dividing on Mount Toby. Farther south, around the west side of Mount Toby, in the narrows which separate the Montague from the Hadley Lake, as well as along the west side of the river from the entrance of the gorge below Northfield Fig. 35— Section through the eroded front of the great delta at Montague. Farms to Sugar Loaf Mountain, the Triassic rocks everywhere approach closely to the present river and the high terrace sands are preserved for the most part only in sheltered recesses. THE HADLEY LAKE. THE NOETH END OF THE LAKE IN GREENFIELD AND THE CHANNEL OF CONNECTION WITH THE MAIN VALLEY. In the last chapter I have traced the waters from the main valley through the Bernardston Pass into the north of Greenfield, where, at the flood time, they widened somewhat into a small temporary lake, whose outlines, as it extended west across the town, are indicated on the map by the extent of the colors marked 1 s h and If, where they are drained by the three branches of Mill Brook. After the waters had ceased to flow across from the main valley an abundant supply still came down the valley of Fall River and pushed out into this Greenfield Lake a marked delta, and the broad bottom of this 630 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. stream with its delta is marked on the map (1 f ^) by a color different from that devoted to the lake bottom. The progress of this delta was arrested (as detailed under a preceding heading, p. 621) by the breaching of the high terrace sands and the passage of the Fall River south to the Connecticut. Clay (1 b c) appears at the surface of the lake bottom at one place, back of the schoolhouse, near the residence of A. Graves. It is abundant and is the only occurrence in this area. THE GREEN BIVEE GLACIEB. High ground borders Greenfield north and west. In the eastern half of the town all depressions are filled with flood sands, which we have traced into the area through the Bernardston and the Fall River passes. The western half is a deeply sunken area. The two bodies of sand noted above expand westwardly, wrapping around French HilP on the north and south, and end very strangely on the west in a high bluff which overlooks the broad, low basin of Green River and Glen Brook. One goes down from the edge of this bluff by a steep scarp 60 feet to the bottom of the basin, and neither the scarp nor this broad bottom seem to me to be the work of Mill River, which now flows in it, bounded on either side by its own alluvial bottom and terraces. This valley, which I believe to have been filled with ice while the lake deposits were gathering, stretches along the whole west side of Greenfield. Not only is the mass of sand which must have been removed, if this basin had been filled up at the flood time, out of all proportion to the amount of work done by the other streams in the terrace period, but the bottom of the basin and its eastern scarp is an irregular, kamy, kettle-holed surface, entirely unlike the surface of the erosion terraces of this and the other tribu- taries of the Connecticut; and the true terraces which border the stream, cut at and below the level of this broad, irregular bottom, correspond in number and extent with those of the other streams. Again, on the west the rocky and till-covered border of this basin slopes rapidly to its bottom, and opposite each valley notch a great delta heading at a level but little below that of the high terrace, and with its semicircular front untouched by erosion, is thrust far out into the basin, showing conclusively ' The hill 500 feet high in the north part, just east of the railroad. THE GREEN KIVER GLACIER. 631 that the bottom of tlie basin and these high deltas were formed at the same time, whifli must have been near the end of the time of the high water stand, wlien the ice had finally melted after having prevented the filling of the valley. The dissected delta of the Green River itself where it leaves its rocky gorge and enters the basin is shown in fig. 3G. But one traces with great clearness the broad watercourse, with its abundant sands, fi-om Bernardston across the north of Greenfield to where the extended sand flats end suddenl}^ and sink by a great, irregular scarp into this basin, and a little farther south the similar watercourse from Factory Village, near Turners Falls, passes across the middle of Greenfield, and stands in the same relation to the southern part of this deep elongate depression. It must thus have been filled had it stood empty in the way ~^S FEET - FI8. 36.— Section of the Green Eiver delta at the north end of the Green Elver basin, where the stream comes out of its rocky canyon, showing that the delta was sent but little into the lake, and its front not eroded. of these two abundant streams, and I can therefore only suppose that here, in the northwest corner of the valley of the Connecticut, and in this long depression between the mica-schist mass of Charlemont and the red sandstone, a lobe of the ice, sent down the Green River Valley from the high ground in Leyden across the whole length of Greenfield, lingered till after the floods had ceased to come through the two passes mentioned above, and after Fall River had ceased to flow west into the Greenfield Lake. I do not think that the ice stood high above the level of the flood waters in the flood time ; but, like the great bodies of ice described by Dall in Alaska, it was submerged beneath the sands as a great continuous body filling the valley and, on melting, allowing its load of sands to drop about 50 feet to their present position. 632 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. THE FACTORY VILLAGE CHANNEL. The map shows very clearly the broad watercourse which bends north from Turners Falls and then turns sharply southwest and runs, its banks and bottoms well preserved and uneroded, to where it widens out into the broader sand plain of the south part of Grreenfield. This passageway was set free by the ice earlier and was occupied by the Connecticut longer than the passage farther north through the Ber- nardston Pass, and a vastly greater body of material was brought into the Deerfield side valley by this way than by the northern one. THE HIGH TERRACE PLAINS IN THE SOUTH OF GREENFIELD AND NORTH OP DEERFIELD, At the end of the Champlain period a broad unbroken plain extended from the south part of Greenfield southward through Deerfield, out of which the channel of Green River and the great basin of the Deerfield River have been eroded. Tlu-ough the southern part of Greenfield and the north of Deerfield, to near the point where the Deerfield River leaves its rocky gorge, the deposits forming this plain are laminated clays, often 20 to 33 feet thick, overlain by sands reacliing a thickness of 80 feet, often hori- zontally laminated in their lower portions and cross-bedded on a grand scale above. The section exposed on the south side of the road from Greenfield to Franklin Park,^ in the hillside immediately beyond the bridge, is very striking. In the bed of the brook the reefs of bright-red sandstone rise above the water and run under the bank. On this, in the vertical wall facing the stream, is exposed 20 feet of till, dull red and made almost entu-ely of comminuted sandstone. This is covered by 20 feet of horizontally bedded clay, in layers 1 inch thick on an average, and as one goes up the hillside the clays are seen to be capped by a great thickness of fine sands, hori- zontally and distinctly laminated, at least 55 feet thick. The upper 20 feet is made up of sands with flow-and-plunge structure and cross-bedding on a grand scale. The section is exposed for 200 feet, and the sands dip with varying and suddenly-changing angle 0-30°, always toward the east. These latter sands vary from fine to coarse. ■ '■ The luroad, perfect plain (1 s h) south. west of Greenfield and extending to the Deerfield River. THE HIGH TERRACE IN GREENFIELD. 633 West across the high plain (Fraukhn Park), from the top of tliis sec- tion to where the raih-oad again cuts into it, the sands rise in heavy beds h\ a long and slightly cni'ved sweep from north to south. These two sec- tions lie just south of the south end of the great Green River depression mentioned above. To the west the wall rises unbroken, and there is no channel down which a considerable stream could have come. It seems to mo that the sands have here been built up to this high level by the water from the Bernardston Pass and Factory village channel coming down over the ice which filled the Green River basin. It is difficult to see how they can have come from aay other direction, and equally diffi- cult to see how they can have been built up here to a broad plain of the height of the high terrace while the above basin. remained open and unfilled to the north. The clays appear abundantly in the south half of Greenfield, Avhere they are used for brick making, and rest on sandstone or till. Farther south, in the southwest corner of the Deerfield River basin, where a brook has cut back in the rim of the basin, is a great exposure of these clays, which for a distance of about 12 feet down from the surface and about the same in from the basset edges of the horizontal beds, have weathered to fine buff clays, while the interior is the ordinary blue clay. Farther south the upper surface of the clays is marked for a long dis- tance by a line of springs in the bluffs along the west side of the basin. Before reaching the mouth of the gorge of the Deerfield River, however, the clays change into fine sands, and the upper sands also grow finer, and in the southern bluffs of the erosion basin the whole thickness of the old delta of the Deerfield River is made up as illustrated in the Wapping cutting (see PI. XVIII, p. 694) by 50 feet of the very finest sands, and this continues to be the character of the great body of sands which fill the Deerfield Valley south through Deerfield and Hatfield. It seems probable that the delta of Deerfield River was thrust across the valley to abut against Deerfield Mountain upon the east, and was elevated more rapidly than the deposits to the north in Cheapside and Greenfield, so that a quiet area of deeper waters existed here, in which the clays were laid down; and later, the current increasing, the horizontal sands were carried in over them, probably through the pass from Turners Falls; and at last the heavy floods of the hightest water stand through the 634 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. Beruardstoii Pass and across the north of Greenfield brought the coarser sands down over the Grreen River glacier and spread them to build up the broad plain of Franklin Park THE LAKE BENCH FROM DEEKPIELD EIVER SOUTH. THE DEERFIELD DELTA. South of the erosion basin of Deerfield River the bench (1 s h) consists of the southern half of the great delta of the Deerfield — that portion which has escaped the later erosion of the river itself. It spreads out, fanlike, as a broad, flat alluvial cone from the mouth of the rocky canyon of the Deerfield, where it has an elevation of 320 feet, and slopes very gradu- ally to its front edge, which is about 30 feet above the lake bottom, and then drops by a steeper grade to the level of the latter. Its outer boundary is in places not sharply marked, as broad bars molded by the current of the main valley from the abimdant detritus furnished by the Deerfield are spread in front of it and render the lake bottom unusually irregular. A cutting of the Canal Railroad, 18 feet deep, passing from the outer border directly to the apex of the cone above Stillwater bridge, showed in beautiful detail the whole structure of the broad delta. It is made up entirely of well-washed sands, everywhere coarser above and finer below. The upper layer varies from 3 to 7 feet, and is made u;p of coarse sand and fine gravel, well washed and rounded, laid down in broad, lentic- ular layers, as a whole horizontal or conforming to the slight slope of the surface. Below are fine, whitish, perfectly sorted sands in two grades, fine and very fine. The former are thrown down in layers 1 to 2 feet thick, with delicate flow-and-plunge structure, and dipping at all angles up to 30° SE. — that is, radially from the old mouth of the river. These layers are separated by other layers, from 2 to 8 inches thick, of the very fine, moist, compact, almost clayey sand, which are thrown down upon rippled surfaces of the coarser, and show a flow-and-plunge structure of extreme delicacy. In an exceptional case a layer of the very fine sand occurs a mile out in the valley, dipping 15° SE., which, although bounded for a long distance above and below by horizontal surfaces and contained in undis- turbed layers of the coarser sand, is contorted in a very complex way, and TUE WEST BKOOK DELTA. 635 thin layers of a coarser saiul included Avithin it are so twisted into the mass that they can be followed for only a short distance. A mile south, in the Northampton quadrang-le, in the delta at the road south from Mill River village, where Bloody Brook joins Mill River, I found in the same })Osition a layer identical with this in all respects, and it may be continuous between the two places, and represent a time when the river was clog'g'ed with ice, so that its current was stopped and an unwonted thickness of the very fine sands deposited and thrown into confu- sion by the stranded ice. The layer resembles so exactly the thicker one described from the Wapping cutting (PI. XVIII, p. 694) that one diagram would serve for both. Farther out, near the outer edge of the bar, the coarse sand and gravel layers thicken downward and pitch sharply southeast in broad, cross-laminated layers, and the finer sands have disappeared or gone below the level of the cutting. It is plain that these latter, which lie below and continue everywhere below the level of the railroad certainly for many feet, represent the front of the delta as it was pushed out into deep water Their varying dip corresponds to the varying slope of the face of the delta, and I am inclined to believe that the thick layers of fine sand (1 to 2 feet) represent the product of a single flood, upon whose rippled surface rests in each case the finer deposit of the succeeding winter. The front of the delta narrows southward and is continuous, at the same level, with the delta, also very large, of Mill River, upon which is the village of the same name. This is more complete, though Mill River escapes through it in a broad, low plain of erosion, and skirts the hill for a long distance south. Then, for a still longer distance south, across the line into Whately, the bench is wholly wanting. At present the broad lake- bottom plain stretching across from South Deerfield abuts against the steep cliffs with no change of level. THE WEST BROOK DELTA. From Roaring Brook down through Whately the hills have an easier slope and were covered with much drift material, out of which the waters have formed an irregular bench, which is only in part built up to true level. This continues almost to the south line of Whately, where, near West Brook, the bench (1 s h) is again well developed and is very complicated and 636 GEOLOGY OF OLD HAMPSHIRE COUJ^TY, MASS. interesting. Long before reaching the brook it rises to the height of 318 feet and widens rapidly into a broad sand plain, across which the brook, emerging from the high lands at the road crossing near a magnificent drumlin (called Belmont) that rises on the north a hundred feet above the plain, runs, over a bed of coarse gravel which is very little lower than the surrounding level, and at the front of the plain falls rapidly over a reef of compact hornblendic granite (tonalite) into the valley below. Just south this reef rises in a narrow ridge and runs parallel to and about half a mile distant from the western rim of the valley, southward through Hatfield, to end in Elizabeth Rock in Northampton. At the highest water stand it was a long island in the lake, or rather two islands, as it is broken through at a point in the middle of its length, tlu-ough which the "Running Grutter" enters the main valley. Into this lateral valley the waters of West Brook carried the greater part of the detritus they were bringing down, and the plain we are following continues at the same high level, quite even and sandy, for a mile farther south, bounded on the west by the steep, rocky rim of the valley and on the east by this island; and from the south the sands of the high bench in Northampton enter the side valley west of Elizabeth Rock and pass up it for almost the same distance, while outside, on the east of the rocky island which is called "'The Rocks," in Hatfield, the fine sands of the broad lake bottom (1 b t) abut at a much lower level directly against the bare cliffs. On the shrinking of the flood waters A^est Brook found its way, not down the western side trough into which nearly all its sands had been carried, but, like so many other streams in the valley, by a detour to the north around the north end of the granite ridge. In a similar way Broad Brook, which heads in the broad sand plain north of Florence, runs a long way north up the trough we have just followed south, and breaks through " The Rocks" in the center of the ridge to join the main valley, searching out for itself the most northerly outlet possible. This is sufficiently explained by supposing that the current of the stream, combined with that of the main stream, kept the sands at a slightl}^ lower level opposite its mouth than lower down, where they were spread in the long trough of quieter waters, so that on the lowering of the water in the main valley the tributary found its way through lower ground around to the north of the bar; still, the many times this occurs in the valley, under THE MILL RIVER DELTA. 637 various circumstances, points to a cuiumon cause, and is, I think, connected with the lowering of the upi)er portion of tlie valley, thus lessening the pitch to the southward. Brooks from the north and south now join and break through the barrier near the south line of Hatfield, and have carried out much of the sand, so that one can not decide whether the high sands formerly filled it entirely. It is certain that the sands of West Brook spread very slowly southward, and that the waters entering by the central break in the ridge spread north and south, throwing down clays up to high level, and that the high delta sands encroached upon them from the north as the growth of the delta went on. THE MILL RIVER DELTA IN NORTHAMPTON. Farther south, on the north line of Northampton, the western rim of the valley, which has come down southward from the northwest corner of Greenfield, swings southwestward and runs back of Florence, by the bridge at Leeds, to Loudville, where it turns at right angles and runs for two miles southeast before it regains its southward course. The bay thus formed was studded with a great number of islands, all of till, for the rocky floor lies everywhere deep below the surface. They are the drumhns already described. Into this bay flowed the waters of four large streams, two of which are dignified by the name of river, and they, together, filled the bay and sent great quantities of detritus out into the valley, to be carried south- ward by the main stream. Their common delta has been greatly cut away by the streams them- selves in their subsequent oscillations- as they followed the margin of the great river downward during the period of shrinkage, and one must know the country well and draw much on the imagination to reconstruct the broad plain as it formerly spread across from Elizabeth Rock to Loudville and out from Leeds to the border of the Meadows. Mill River has been espe- cially destructive, and, as its mouth advanced from Leeds to its present place, it has worn out all the broad basin in which it flows, and its tribu- taries have cut out the pecuHar depression of the " Bay State." One must think of all this area raised to the level of and merged into the Florence plain in order to reconstruct this, by far the largest delta deposit of the high bench upon the west side of the river. Along the road from Florence to West Farms, and then to Loudville, one rides for several miles over a sand plain (1 s h) about 305 feet above sea, 638 GEOLOGY OF OLD HAMPSHIEE COUISTTY, MASS. abutting against the cliffs on the northwest. Its border against the granite bluffs is exceptionally well preserved, but in places is deeply kettle-holed in the portions adjoining the rocks. It stretches, except where interrupted by drift islands, with gentle slope southward for a long distance, to descend at last more abruptly to the village of Easthampton, its scarp being ter- raced, but apparently not much cut back, while in Northampton it has suffered much more serious erosion during the formation of the lower ten-aces. The apex of the delta of Mill River in Northampton is where the bridge crosses the rocky bed of the stream before entering Leeds. It widens sud- denly at Florence. Its extent, apparently out of proportion to the drainage area of Mill River and the other streams that formed it, is due lai'gely to the fact that its sands are spread out among the lenticular drift hills by which the great bay in the crystalline rocks was filled. (See p. 643.) The cutting along the New Haven and Northampton Railroad made to obtain material for raising the railroads through Northampton gave repeated sections north of the railroad, extending from the brook crossing east of Florence to the crossroads next east, a distance of a quarter of a mile. In all the western part of this section (which runs east and west) the sands are cross-bedded on the grandest scale, the layers in the long cut, which was 15 feet high, having a uniform and high westerly dip. In two cases the material suddenly grew fine, and heavy clayey layers are intercalated in the coarse buff to reddish sands. In the eastern portion of the section — the part south of the cemetery — the beds bend over and dip east, and are here greatly disturbed and mixed with glacial material by stranded glacial ice. An inspection of the map will show that the long drumlin called Strawberry Hill, just north of Florence, and the prominent drumlin north of the Bay State, nearly cut off this area from direct communication with the waters coming out of the Mill River gorge, and that these cross-bedded sands must have grown as a broad sand spit extending south from Fortifi- cation Hill to the north and made up of -material swept south across the Camp Meeting grounds and around the east side of this hill, so that they were thrown down with strong westward dip on the inner (western) and sheltered side of this bar, along the outer side of which the icebergs stranded. TUE MILL RIVER DELTA. 639 The southern portion of the delta is composed of the confluent deposits of ]\Iill Kiver and the north branch of the Manhan. The great •ilacial lake in Westhanipton (p. 594) served as a catchment basin for sands which were carried ultimately by Roberts Meadow Brook and the Manhan to augment the high terrace at this point. While the sands in Northampton are in many beds clear gray, showing under the micro- scope many rounded grains of black mica-schists like those of Goshen and Chesterfield, in others they are reddish from the abundance of garnet grains in them, both peculiarities indicating their origin from the garnetif- erous mica-schists in the drainage area of Mill River. The sands of the southern portion of the plain are more largely granitic and are derived from the great granitic area of Westhampton. This may be taken as one of the jDroofs of the assertion that the high terrace was mainly brought in from the sides of the basin. The great sand plain is continued across to the North Branch, is in all this distance more than a mile wide, and sinks in several great terraces to the clayey lake bottom at Easthampton, and as it nears the south line of Southampton it enters the western of the three passages by wliich the waters passed out upon the Westfield plain, and just on the town line it received the abundant deposits of the southwest branch of the Manhan at Russellville, and across the basin since eroded by this stream it was plainly continuous with the north end of the Westfield plain. Just where the western channel widens by the dropping down of the hill east of East Farms into this broad, open plain the abundant contribu- tions of the branch of the Manhan last mentioned were received and spread clear across the channel, up nearly to the normal high terrace level — the deep water of the lake bottom shallowing southward in the channel and coming to an end just opposite the mouth of the branch, and marking out thus the channel whereby, on the recession of the waters, the Manhan was compelled to take a course north across Southampton and Easthampton to join the Connecticut at the head of the oxbow. THE LAKE BENCH ON THE EAST SIDE OP THE HADLEY LAKE IN LBVERETT AND AMHERST. Through the Narrows in Sunderland the bench (Ish) is well marked along the west slope of Mount Toby, and turning the corner of the mountain it rests against its south side. It is characterized by fine sands in great quantity, dependent upon the fact that the region is far from the mouth of 640 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. any river, the sands having been carried a great distance by the main stream. Against the north end of Smiderland street it is represented by a horizontal shelf cnt in the sandstone. The terrace then widens in the extensive plain of South Leverett which rests against the sandstone moun- tain on the west and against the crystalline rocks on the east, and runs up into the gorge on the east side of Mount Toby. At its head, near the rail- road-crossing north of the station, it is a coarse gravel with pebbles 6 inches in diameter, and it has a height here of 310 feet above sea. It slopes gently to "its front, where it has a height of 290 feet above sea, and is made up of coarse sand. By recurring to the description of the old course of the Locks Pond Brook down through the Mount Toby gorge to empty into the Hadley Lake at this point (see p. 584) when the ice still filled the Montague basin to the north, the reader will understand my conclusion that the main por- tion of the great mass of gravel gathered here was swept into its place by the Locks Pond Brook, deflected soiithward, and only smoothed down to its present level by the waters of the Hadley Lake. 1 imagine that this deflection of the brook by ice filling the Montague basin may have taken place when the ice had abandoned all the Hadley Lake except its northern lobe in Greenfield. Southward, the high terrace is only indistinctly marked against the till for a long distance, as no brooks brought in material here. THE DELTA OF CUSHMANS BROOK AT NORTH AMHEKST AND THE ISOLATION OP THE EAST STBEBT BASIN IN AMHEKST. On reaching North Amherst we find the high terrace (1 s h) developed in great force and, because of the rising of the block of hills north of Amherst Center as a great island in the lake, with considerable complexity. A great depression, closed on all sides, extends along the eastern line of Amherst, ending on the south at Dwight's station, having the village of East Street in its center and being bounded on the north by the delta of Cushmans Brook. It is plain that when Cushmans Brook began to flow into the lake there was free communication between this depression and the main area of the lake to the west, across the space now occupied by the delta, and that for a time the sands brought in by the brook were swept southward KENCH SUKROUNDING BAST STREET BASIN. 641 al.ni-;' the west slope of the I'elhaiu Hills, forming the extensive sand deposits which flank these hills for a long way south. At last, however, the delta extended across to the rocky hill north of the North Amherst cemetery and excluded the main current from this eastern basin, and from this time on the sands of Cushmans Bi-ook were swept around west of the Mount Pleasant block of hills, building up a great terrace, or rather sand bar, which extends south to the Agricultural College. The college build- ino-s stand on it, and it ends at the south border of the college farm. THE BENCH SURROUNDING THE EAST STREET BASIN. By the extension of the delta of Cushmans Brook across the north end of this basin a separate body of water resulted, connected with the main lake only by narrow channels among the drift hills south of Amherst Center. The hiffh terrace, continuous southward from the extended delta flat at North Amherst City along the flank of the Pelham Hills, is a marked object from College Hill. It appears here, as around much of the valley, as the highest line of cultivation, and above this horizontal line the hillside is heavily wooded. It is a broad sand flat, its material derived partly from the sands brought down from the Leverett Lake deposits (see p. 584) and partly from cutting into the kettle-holed sands carried along the side of the Pelham Hills before the departure of the ice and left at a level higher than that of the lake (m t, PI. XXXV, C). Fort River, opposite Amherst, coming out of the Pelham basin, adds somewhat to its width, but less than one would expect, the main portion of the sands brought down by this stream having been at an earlier period carried southward, as detailed on page 578. South of this stream the terrace is a marked bench cut in the sands thus carried along the slope at a higher level than its own (m t), and it swings round the west side of the great drift hill north of Dwight's station and continues east as a horizontal bench notched in the south face of the delta of the earlier and higher stream (see p. 589). It was thence continued south and west as a bench cut in the older sands across the entrance of the Belchertown Pass, for at this time the waters certainly did not go through this pass, as the lowest point in the sands across north of the Belchertown ponds is about 30 feet above the high terrace in this latitude. It is continued MON XXIX 41 642 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. westward along the north foot of the Holyoke range, still as a bench cut in the irregular sands which are heaped so high along its flanks. SHORE NOTCHES IN THE SIDES OP DEUMLINS. Along its western side the East Street basin is bounded by a continuous line of di'umlins, and the high terrace is marked by a horizontal fluting cut in these drift hills. As all the hills south of Amherst village stood as islands in the lake, while narrow channels connected the East Street basin with the rest of the lake to the west, this fluting surrounds them on all sides, and the same was true of the great block of hills north of the village until, by the extension of the delta of Mill River, or Cushmans Brook, it was joined to the mainland and made a peninsula. This horizontal fluting is well shown in the drumlin which rises north of the Methodist Church in Amherst. Starting from the top of the hill, one follows down on either side its regular curved slope for a distance, when it suddenly grows much steeper, and then, at the 300-foot contour, begins a much easier slope. One comes down to this contour line on till, but here begins a shore gravel bed, at first thin, but thickening outward, as its surface has a lesser slope than the old surface of the drumlin upon which it rests. So long as this East Street basin was open to the north, the water moved through here with considerable velocity in flood time and swept such material as it could erode from the drift hills themselves southward along their slopes (there were no brooks in these isolated hills to bring down material and build up deltas), and so the bench along this side is scantily represented by slojDing sheets of gravel concentrated from the till. Just north of the New London Northern Railroad station, for several hundred feet west of and above the railroad, the bench widens into a con- siderable sand plain, recently built over. The sands dip south in great sheets, which were pushed over the south front of a deltalike bar and carried south through the notch in which the railroad runs. Across the village of Amherst the waters of the two basins were con- tinuous. Farther south the fluting is carried along College Hill below the church and the gymnasium. It surrounds the long isolated drumlin south- east of College Hill, and the section through the south end of this hill made by the Central Railroad showed that a great hooked bar of gravel was SHORE NOTCHES IN DRUMLINS, 643 carried oxit south from the nucleus of till with an anticlinal structure like a nest of inverted canoes, a type repeated in connection with all the other isolated drumlins farther south. At first the axis of the bar seems to have been shifted now to the right and now to the left, only part of the deposit of each position being retained 2)ermanently. Then the layers are continuous, flat on the top for 30 to 50 feet, and dip east and west. On the west side it was built up with easier slope and finer material, as the bar was being carried south across the some- what land-locked bay south of College Hill, where it opened eastward into the East Street lake, and the main current, sweeping down the East Street channel, not yet closed on the north, wore a deep fluting into the east side of the di-umlin and carried the material south in great sheets of coarse gravel, often 3 to 8 feet thick, to form the eastern slopes of the canoe- shaped layers, while, if we follow these sheets over to their western slopes, we find them made up of much finer sand, at times slightly gravelly. At the bottom of the western slopes the sheets run west horizontally for a little distance and then mount up gradually onto fine clays, which latter in turn sink with slight dip eastwardly beneath the sands and below the level of the cutting. This shows that the water stood at this high level for a long time, allowing the fine clays to accumulate (which happened at a higher level in this sheltered bay than in the deep East Street basin), before the bar was pushed south over them. The village of Soxith Amherst is built on such a bar carried as a ridge from one drumlin to another, and the road running south from the village keeps on the bench around the east side of the great di-umlin south of the village, and follows the bar that projects southwardly from it to join the high terrace at the "Bay road" along the northern flank of Holyoke. South of College Hill is a deep depression, just mentioned, sheltered on all sides by drift hills, and never filled up, and another, much more extensive, lies west of the village of South Amherst. On the decline of the waters a stream draining the East Street lake found its way between drift hills into the first, and from this into the second, of these partially isolated bodies of water, and through the western line of drift hills into the main basin, and cut its way down through the drift so slowly that separate terraces were formed around the East Street lake, where the streams entered it from the Pelham Hills. Ultimately these 644 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. streams, uniting in the bottom of the drained lake, completed the erosion of the present sinuous course of Fort River, across the drift area south of the college, apparently to a level somewhat lower than the present bed of the river, without striking rock, for the stream now flows over a muddy bottom, and there is no trace of the sheet of bowlders which it must have concentrated out of the till. It has, however, sufficient slope for a water power, and the ponding back may have caused it to cover this up, as below the dam it flows over a bottom of coarse bowlders. THE HIGH TERRACE OR BENCH ALONG THE WEST SIDE OF AMHERST RIDGE. So long as the water passage from the main basin into the north end of the East Street basin was open, and the sands of Cushmaus Brook (or Mill River) were carried down along the flank of the Pelham Hills, the work of the lake waters along the west side of the Mount Pleasant block of hills, and along the west side of College Hill, and its prolongation north- ward to the head of Prospect street, and of Mount Doma farther south, consisted mainly in the concentration of a coarse, well-washed and well- rounded beach gravel out of the till, of which all these hills are composed. Because of the narrowing of the channel by the hills named above, and by Mount Warner, farther west in mid-channel, the current- was here somewhat accelerated, and, aided also by the prevailing west winds, wore with exceptional force into the hillsides along the line we are now trac- ing, cutting deep into the till along the 300-foot contour, or a little lower, as the effective erosion level was often somewhat below the highest water stand, and forming thus a broad horizontal or outwardly sloping bench in the till, over which sheets of the concentration gravel spread in bars and low ridges. The exceptionally steep slope above the 300-foot contour, often, indeed, slightly concave, which I have called the horizontal fluting, is best devel- oped along the west flank of Mount Pleasant and its continuation north past the Plant House and through the chestnut woods farther north. All the plain south of the Plant House has been formed thus by erosion, and the hill formerly extended here as far west as the new road to North Amherst across the College farm. The gravel spread over this plain in great sheets has been largely used for sidewalks, taken mostly from the pits just south of the Plant House. THE HIGH TERRACE IN AMHERST. 645 where a few feet of dig-ging exposes the till below. This 2:)lain sinks away to the iK'xt lower level on the west, that on which the Agricultural College Ijuildings are placed, because the old surface of the till had this configura- tion and was not filled uj), the outer (western) portion of this latter plain being, however, made up of thick sands through which the brook has cut between the college buildings. This sand is the southern tongue of the delta of Cushmans Brook, carried along the western flank of the Mount Pleasant hill after this delta had grown across so as to abut against the north end of this hill, and had thus built out the great sand plain which stretches north therefrom, and the main current of the brook, rounding the hill itself, carried the sand south along its western flank, at a level much below that of the high-water stand of the lake. Farther south, Mount Pleasant breaks down suddenly, and a short dis- tance to the west a rocky projection at the head of North Prospect street rises 30 to 40 feet above the old high-water stand. This mass of rock, which has now been mostly covered up, used to be called Pikes Peak, and for convenience I will continue to employ that name. Between Mount Pleasant and Pikes Peak the water had free communication with the East Street basin across the village of Amherst. The water line followed the 300- foot contour around the south spur of Mount Pleasant, extended as a rounded bay up its eastern side, skirted on the south the hill on which Professor Tyler's house stands, and so swung around northeast to join the broader terrace above the railroad. (See p. 642.) From Pikes Peak the water line extended south just west of and at the next level below Prospect street for the whole length of this street, turned southeast through the grounds of the president's house, crossed South Pleasant street and ran at the foot of the sharp slope south of the Octagon, skirted the College Hill on the south and east, and on the north ran just north of the Lucius Boltwood house, now Hitchcock Hall, and along the south border of the common, and bending north and crossing Pleasant street it ran north just west of this main street of the village, past the hotel front, to the point o£ beginning at Pikes Peak. Thus an L-shaped island, with the College Hill as its horizontal and the Prospect street ridge as its vertical portion, rose above the level of the flood waters, which came up almost exactly to the level of the post-office steps. It must be remembered that the level of the college chapel was once continuous under the Octagon, the library, and the XW house, and that the deep notches 646 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. of the roadways are the result of subsequent excavation, and that the surface is chang-ed by grading north of Walker Hall and the Octagon. Through this passage between Mount Pleasant and Pikes Peak there was a steady set of the current which has built up the broad area of finely washed and sorted gravels Avhich stretch across and down through the pass and which are well exposed all round the Catholic Church. Southeastward they stretch as a flat of finer sands, with a layer of concentration gravel capping it, across from Professor Tyler's hill to College Hill. The two stone churches and the high-school building are on this sand plain. College street lies so near its border that the houses on the north side have cellars in sand; those on the south side have wet cellars, as they cut through the thin border of the sand and get the drainage which conies down from the College Hill on the sm-face of the impervious till beneath. The current swept the sands across in a line from the Catholic Church to the high-school building and the common. An area in the recess of the N ^UegeJIiiL. s ~^ Cen/rdlRR.. TOZ. Fig. 37 Section of shore beds of Hadley Lake soutli of College Hill, at Amherst. The cutting was 18 feet deep. L-shaped island, the south half of the common, was not filled up quite to the true level and was miderlain by till at no great depth, and so was orig- inally a very swampy place. It has been filled in considerably, and along most of the street to the east and the whole of the street to the west of it the artificial filling has been so great that the waterworks ditches did not reach the undisturbed sands. Along the whole west side of the L-shaped island the level of Lincoln street is the level of the high terrace. It is a bench cut in the till, very broad, and but little covered by sands, since all that the main stream obtained from the delta of Cushmans Brook was swept in across the village to the East Street basin. Thin cappings and bars of sand are applied to its surface and to the slope down to the lake bottom, and can be well studied from the side of Mount Warner. Along Lincoln street the cuttings of the waterworks struck till for more than half the distance, and along every street which crosses this TEIE HIGH TERRACE IN AMHERST. 647 shore lino I have at one time or anc.ther had opportunity, hi cuttings of the water t>r <>iis companies, to locate exactly this old shore line and plain. CoUeo-e Hill breaks down like Mount Pleasant, and southwest, at Professor Harris's house, begins another drunilin, named Mount Doma (by President Hitchcock), from its regular shape. Between the two the waters passed southeast into the depression south of College Hill, and a broad, thin sheet of gravel stretches through the pass, and is well exposed in the cutting of the Centi-al Railroad. Everywhere through this pass the till is but a little distance— at most 6 feet— below the surface, as at the bridge over this cutting on Woodside avenue. Fig. 38— Enlarged section of the sonth side of cutting shown in flg. 37. The section represents the aouth side of the railroad out beneath the bridge shown in flg. 37. Fig. 37 shows a section south from the Octagon, on College Hill, through the ciitting of the Central Railroad, at the point where the highway crosses it. It is interesting as showing sands under the clays and separating them from the till. This is the only instance of the kind I have seen in the valley. The clays thicken off into the deep water south and southeast, and northward grade to sand layers, and these to the beach gravels which make this broad flat and which are spread over the bench cut back in the till, by which cutting the sharp slope south of the Octagon was produced. The varying currents from the west are finely shown by the detailed sections figs. 38, 39. The quiet water allowed the clay layers to form, and then the strong current crumpled them. 648 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. Farther south, the terrace swinging round either side of Mount Doma is continued in an exceptionally long spit of gravel which bends round southwest and continues to Fort River, and beyond the river a peculiar ridge of sand, sloping down gradually to the depression west of South Amherst and westerly to the main basin, is carried south to the high ter- race skirting the north flank of Mount Holyoke. This seems to me a bar thrown across the mouth of the deep bay which occupied the second depres- sion, mentioned above (p. 643), by the current of the main stream coming down through the channel between Mount Warner and the Amherst ridge. THE BENCH AROUND MOUNT WARNER. As one looks at this isolated rocky hill from Amherst a northern por- tion, horizontal and at the level of the high terrace, attracts attention, and investigation shows this to be a broad, rudely horizontal rocky bench but slightly covered with loose material. To assume that this perfectly terrace- -I SOMm. Fig. 39. — Detail of clay layer cnimpled \f^ the current, from fig. 38, to show how the layer was carried along hy the friction of a current from the west. like portion of the mountain was planed down to the level of the high ter- race by the flood waters would be to assume that this flood period was immensely longer than we have been accustomed to think it, and longer than the other phenomena connected with it would seem to warrant. An inspection of the map will show that south of the mountain a great tail of sand extends southeast to the Northampton road. Just under the south end of the mountain a pond occupies the place where the waters meet- ing from both sides around the mountain stagnated and thus prevented the sands from building up quite to the highest level, but farther south a broad, perfectly level sand plain projects at the level of the high terrace southeast- ward, indicating the direction of the current. (See map, PL XXXV, C.) I imagine it to have been deflected somewhat by the prevailing west wind. This tail sinks like a delta southward and runs out on the clay bottom of the lake, reaching nearly the Northampton road. On the west side it flanks the mountain for a long way north, but is so blended with dunes carried up from below that its original relations can not be clearly made out. THE LAKE BENCH NORTH OF HOLYOKE llANGE. 649 This bntad, Hut, siuid- covered jjlateiui in continuation of the soutli end of ^Idiint Warner has the exact heiglit of the old Hadley Lake. Its direction (southeast) was a great puzzle to me, and I tried to explain it by supposing- that the south current and the west wind produced a resultant southeast direction in the great sand spit. Recently (1888) excavations along the road south from the Catholic cemetery have shown that all along the south- east front of the plateau the till lies almost at the surface and makes the explanation more probable that the whole mass of the deposit is due to ice, and that the north-south valley movement of the ice is here, where the valley is xinusuall}^ wide and open, replaced by the usual upland (N. 30° E.) movement, and this agrees with the strong pressure of the ice along the west face of Deerfield Mountain. Only the surface and slopes of the plateau were then molded later by the water and covered and flanked by sand bars. THE LAKE BENCH ALONG THE NORTH SLOPE OF THE MOUNT HOLYOKE AND MOUNT TOM RANGE. I have already (p. 586) called attention to the fact that great masses of irregular sands are in places heaped up against the flanks of these ranges at heights much above the highest water level of the Hadley Lake. Where, as along south of Amherst, the high terrace is a bench cut in these sands it sinks gradually, and often without any marked change of slope, into the lake bottom, as if there had been here no marked current, but an undertow had drawn the sands in large quantity down into the deeper water. Farther west, south of Hadley and in the Holyoke notch, the current was more marked; but the material at the disposal of the stream was less in amount and the terrace is a narrow bench, often of till, and from the entrance of the notch down to Titans Pier the waters cut back the till in a broad bench and then wore into the trap and sandstone, producing a ver- tical wall which the talus of fallen trap has not yet obliterated. Across the river the same conditions hold. Above the highest terrace level, as determined by its coincidence with the Florence plain, higher levels of coarse sand occur and the lake bench slopes inward to where it is cut off by the later erosion of the Connecticut, or when we get beyond this, as in Easthampton, it continues its gradual slope to the middle of the basin, or to the line of the deeoest water of the broad stream which flowed down 650 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. across the Hampton plain. Along this portion of its course, between Nona- ttick and Mount Tom, fine bowlder beaches mark the outer boundary of the high terrace. THE WESTFIELD PLAIN. I have followed tlie high terraces on either side of the broad Hadley Lake and found them much more intimately connected with the Southamp- ton Valley than through the gorge of Mount Holyoke with the Springfield basin. They are confluent with the broad Westfield plain, one of the most interesting deposits of the river. The broad, unfilled lake, 15 miles wide opposite Northampton, nar- rowed across Southampton to a width of 3 miles, and on the south line of this town two long ridges. White Loaf and the high hill to the west of it. East Farms Hill, rose as islands in its course, and the waters passed on south by three narrow channels — respectively 180 rods, 120 rods, and 360 rods wide, counting from west to east — into the "Westfield plain, the filled-up portion of its ancient bed. These passes formed a waste gate through which the overflow of the river went with velocity accelerated by the narrowing of its passageway. It swept the abundant kame sands (m t) which had been spread at the western foot of the Mount Tom range and over White Loaf through the eastern and middle channels, and this is the proximate source of the trap pebbles traced far south across the plain by Mr. Diller.-' The sands of the Manhan were spread by it over the western portion of the plain- As a result, we haA^e coarse gravels concentrated from the kame gravels in the eastern gorge, stretching far south across Hampden plain and growing gradually finer, and in the same latitudes on the western side of the plain the sands are much finer, being derived from the sands of the Manhan. That the sand here had this origin in local kame deposits is manifest from the fact that along the whole course of the Holyoke-Tom divide there are no streams flowing into the river to bring sediment, and through all this length the high terrace is for long distances wanting or marked only by a narrow shelf cut into older deposits, and certainly nothing was brought from the upper waters of the river across the broad, low clay bottom of the lake in Easthampton. The delta deposits of the Loudville branch, swept along the west side of the basin, had, south of Southampton village, shrunk to a naiTOw shelf, ij. S. Diller, Geol. of Westfield: Westfield Times and News Letter, Sept. 19, 1877. THE WESTFIELD PLAIN. 651 and \vt, iiniuediately after passing the two obstructing hills, the waters liaA'e filled the broad valley (which is over B^ miles across at its narrowest point) well-nigh to the height of the highest terrace, everywhere from side to side, and for 10 miles south. Where the waters swept around the two hills mentioned above, broad grooves appear in the plain, hugging the sides of each hill (If), and joining and running out southward on the plain for a long distance, with a tail of higher sands between them; and from the south- east corner of White Loaf a heavy bar of coarse gravel (1 s h) runs out southeast, and east of this was left the great depression of the Hampden ponds. White Loaf ends near the north line of Westfield, but the East Farms Hill is continued in a low, broad reach of till down halfway to Westfield village, dividing the plain, but in all its southern portion hardly rising above the surface of the highest waters, and bounded by a marked bowlder beach. On the west side was the real thread of the current of the broad river, and this was early utilized for the Farmington Canal. Just on the north line of Westfield the main stream received the waters of the Manhan, and the increased eroding power derived from their junction is seen in the lower- ing of the plain for a mile south of the town line. This was aided, also, by the narrowing of the channel through this distance. Then the valley quite suddenly doubles in width and the low thread continues along its eastern side, hugging the East Farms Hill, and the plain is gradually built up to a much greater height along its western half, and for a long distance south the height of its western edge is 300 feet, and it slopes east very gradually 40 feet and then quite rapidly 25 feet more to the bottom of the deep-water channel. Southward, the highest point in this channel is a mile south of East Farms, where the south end of the East Farms Hill drops down and the two valleys come together. North of this all the brooks which come from the hills of West Farms and East Farms gather in this deep-water cur- rent bed and run north. The terracelike slope which borders this deep-water channel on the west bends round (north of F. W. Griswold's) to the west and runs west to the slope of Pochassic Mountain. The corresponding slope which bounds the channel on the east bends east at the same point, and the westward-running last-mentioned slope, if prolonged eastward, would just meet and be continued 652 GEOLOGY OF OLD HAMPSHIKE COUNTY, MASS. by tliis similar slope, which stretches east from the Catholic cemetery across the Hampden plain. The highest sands (1 s h) on the east and west sides of the plain run south with regular slope and then drop quite abruptly in this terracehke construction scarp. The deep cuttings of the Westfield and Holyoke Rail- road and the many openings on the north edge of the Westfield basin do not give any sign that the Westfield ever wore up to the foot of this scarp. Everything indicates rather that the whole plain north and south of the later-eroded basin of the Westfield was the result of one continuous opera- tion, and that this scarp was formed east and west across the channel of the main stream just where the waters of the Westfield River joined its waters, and the outlet through the Divide Range gave a means of communication with the eastern lake, and thus the carrying power of the main stream was sud- denly lessened along this hne, and the scarp was the index of that lessening. The diminished current carried finer material, and in the steep erosion scarp by which one descends from the south edge of the plain to the Westfield River basin, a mile south of the Catholic cemetery, we have many deep sections showing a great thickness of sands so fine that the owners have often attempted to utihze them for brick making, but without success. On the south of this broad original depression which guided the Westfield rivers finally back to the gorge in the Divide Range and to the Connecticut, the fine sands continue in "Poverty plain," west of Little River, rising from 229 feet on the edge north of the Westfield basin to 264 feet on the south of the basin of the Little River, in the center of Poverty plain — an enormous waste of desolate sands whose increased height comes from the sands of the Westfield rivers swept down around the high drift hills of the "Fox district." The broad "Avenue plain" between the two Westfield rivers is a very interesting portion of the original plain of the flooded river. It is now about a mile wide and 4 miles long, and stretches from where it rests against the drift border of the valley between the two Westfield rivers, at a height of 290 feet, eastward to the cemetery in Westfield, descending 16 feet per mile (Diller), and bounded north, south, and east by the deep erosion basins of the two rivers. It is made u.p very largely of quite coarse and well-washed gravels, even out at its eastern end, which are exposed in many natural sections and gravel pits, notably just east of the cemetery, where the well-sorted and rounded gravel is 12 to 14 feet thick and rests THE WKSTFIIOLl) PLAIN. 653 Upon sands. These sheets of gravel stretched, 1 have no doubt, right across the area now occupied by the basin of the Westfield River, and were con- tinuous with the fine gravels just northwest of and above the railroad station. Here tliore is a tlK)roughly classified bed of 4 to 6 inch pebbles, all well rounded and made up very largely of the peculiar hard Laurentian gneiss of Washington and Hinsdale and of the Berkshire quartzites, both brought down from the lieadwaters of the Westfield River. Mr. Diller calls attention to the depression of the east end of this Ave- nue plain 1 7 feet below the adjacent plains. I believe this plain to have been formed as it now is during the flood time of the main river, and to owe its slope to the heavy flood of the Westfield River, which kept this passage between Pochassic Mountain and the West Parish Hills Scoured out, and cari-ied out over its bottom the broad sheets of coarse gravel which reach east to the village of Westfield. The position of these gravels over the underlying sands is the normal one all up and down the valley wherever a delta is advanced into deeper water, and the two beds are parts of the result of a single operation. The flood of the Westfield then, as now, pre- ceded that of the main stream, and thus annually swept its channel clear and gradually built up its heavy gravel beds. Poverty plain is continuous across Westfield and into Southwick. It begins to contract in width on the town fine, and from Southwick Hill southward has a width of little more than a mile and a half. The con- finement of the waters in these narrow limits, by increasing their eroding power, seems responsible for the long, shallow depression of the Congamuck or Southwick Pond, and for the curious course of Great Brook, which, starting from the middle of the pond on its west side, runs north among the drift hills, and, leaving them, takes a diagonal course across Poverty plain, passing within 100 rods of the head of the pond, and finding what I imagine was the thread of the current of the main stream and following it back until it joined the Westfield near the divide gorge. The thread of the current passed out of the deep water over South- ampton village and by the west pass down to and across the place where Westfield village now stands, and then, on receiving the waters of the Westfield rivers, bent east to near the gorge, whence it followed the present course of Great Brook to and across the whole length of Southwick Pond, and so southward across the Farmington basin and by the course of Mill River into the sound at New Haven. 654 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. An inspection of tins Westfield-Sotithwick plain as represented on the map will, I think, convince one that it was constructed by a broad, very shallov/ body of water, often broken into separate threads meandering' across the plain, which were separated from one another by long intervening bars and spits, bounded by construction scarps, at times quite steep and fluted on the convex side of the curving channels, but often of long and easy slope. THE GREATER ELEVATION OP THE TBERACBS IN THE WBSTFIELD THAN IN THE SPRINGFIELD LAKE. POSSIBLE WESTERN ELEVATION. Professor Dana has noted that the highest nol-mal terraces in the west- ern valley are 60 feet higher than in the eastern. Mr. J. S Diller has discussed the matter in an interesting article which was published in the Westfield Times and News Letter, September 19, 1877, and which is here reproduced : THE GEOLOGY OF WESTFIELD AND VICINITY. By J. S. Diller. Professor Dana has shown that at Tariffville, Connecticut, where the Farming- ton Eiver flows through the Divide range, the terraces upon the west side of the range are about 50 feet higher than those upon the east side. At the Westfield gap, through the Divide range, the upper terrace on the west side of the range is 264J feet above sea level, but on the east side the highest terrace is 50 feet lower. It has been shown by Professor Dana that during the Champlain period the highest flood level over Springfield was 240 feet above the sea level. We have shown in a pre- vious article that during the same period the highest flood level on the west side of the Divide range was 280 feet above sea level. The flood at Westfield was at least 48 feet higher than that at Springfield. The question at once arises. Why was the water so much higher on the west side of the range? The answer most frequently given is that the gaps through the Divide range were closed, thus damming the water back and raising it to a greater height west of the range. In the Westfield Eiver gap, upon the south side of the river, there are two terraces. The lower one extends directly through the gap, at a height of 199 feet above the sea. This terrace is made up of stratified deposits, con- taining a large portion of clay. The beds extend, with the terrace, directly through the gap. The continuity of the beds is evidence that the gap was open when the deposits were made. These lower deposits, we have reason to believe, were made during the early part of the Champlain period. It therefore appears that during the early part of the Champlain period the gap was not completely closed by either drift or trap rock. It should here be remarked that there is, on the right bank of the river, just east of Morley's bridge, in the gap, a ledge of trap whose top is 21 feet THE GKOLOOY OF WESTFIELD. 655 above tbe highest modern flood level at that place. The ledge breaks the coutimiity of the lowest beds of the terrace, and may have once formed a considerable dam in the gap. Above the ledge the beds are continuons through the gap, and are evidence that there the gap was oi)en. If the gaps in the Divide range were not closed during the Ghamplain period, the height of the water must have been due to some other conditions. There were two conditions on which the height of the water seems to have depended, viz: (1) The narrowness of the gaps through the Divide range, and (2) the difference in slope of the valleys east and west of the range. Dr. Davis, in his History of Westfleld, says that the Westfield Eiver at Westfleld, during floods in 1819 and 1826, rose U feet. Mr. L. F. Eoot, civil engineer of this place and of the Canal Railroad, has recorded a rise of 12 feet during the great flood of 1SG9. Mr. Austin Williams made marks upon a tree near the north end of Morley's bridge, showing the height of the water there during an ice flood in 1855, and also during the flood of 1869. In 1855 the water rose 27 J feet, and in 1869 it rose 26 feet. It thus appears that when the river rose 12 feet at the village it rose 26 feet in the gap. Some of the excess in height was due to the inflowing water from Little Eiver, but by far the greater part is dne to the smallness of the gap through the range. By measuring the gap it has been determined that a flood nineteen and one-half times as great as the highest modern flood would flow through the gap at such height as to cover the top of the highest terrace. The overflow from the Connecticut and Manhan rivers entered the Westfield Valley by two large streams, neither of which were less than three-fourths of a mile in width, and one having a depth of 40 feet in its most shallow portion. Add to the water poured into the Westfield Valley by these two streams the immense floods of the Westfield rivers and it will be seen that for such floods the gap through the Divide range was a small outlet. The smallness of the gap evidently had much to do with Increasing the height of the water west of the Divide range. Supposing the stratified drift were removed from the valleys on both sides of the Divide range, we would see that the northern portion of the valley on the west side has much less slope than the corresponding portion of the Connecticut Valley on the opposite side of the ridge. The valley west of the range is crossed by the red sand- stone divides which separate the Westfield Eiver Valley from the Manhan Eiver Valley on the north and the Farmington Valley on the south. Such divides are not found in the Connecticut Valley on the opposite side of the ridge. The lowest parts of the valley west of the Divide range are those across whicli the Westfield and Farmington rivers flow. These lowest portions are considerably higher than the lowest parts of the Connecticut Valley directly opposite, else the Westfield and Farmington rivers would not flow into the Connecticut. It is evident that at the close of the Glacial period the average slope of the valley west of the Divide range was much less than that of the opposite portion of the Con- necticut Valley. The two valleys filled, during the Ghamplain period, with water from the Connecticut Valley, in the region of Northampton, acted much like two parallel 656 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. troughs having their source in the same place and at the same level, but having dif- ferent slopes. The water in the one having the least slope must be above the level of the water in the other at all points directly opposite. The difference in slope of the two valleys, together with the narrowness of the gap in the Divide range, seem to be the cause of the greater height of the water in this vicinity. It seems that the following considerations should be weighed in seeking for an explanation of this curious difference of level: (1) The Springfield basin is about four times as wide as the Westfield, and thus much more material would be required to fill it up to the same level. (2) Because of the northwestern recession of the ice the eastern floods sent the mass of their sands down through the Mouson-Wilhmantic Valley or lodged them in the great series of catchment basins I have described above as the eastern series of glacial lakes. (3) The same recession of the ice, continued northwestwardl}^, caused the heaviest floods to pour into the lateral or Westfield Valley by all the transverse valleys coming in from the west, and of these the Westfield River was the. most important, because it runs back northwest across the whole plateau of the Berkshire Hills and at Dalton opens broadly into the great Housatonic Valley, and because it remained the main trunk of the ice drainage until the ice had receded from those hills; and while the ice front was in the region of Pittsfield the di-ainage of a portion of the Upper Housa- tonic was deflected into this valley, producing the interesting sand plains in the upper valley at Hinsdale and bringing down bowlders from this region to spread over the Westfield plain. The combined effect of these three conditions seems sufficient to explain the lower level of the eastern plain, and instead of saying that "the flood at Westfield was at least 48 feet higher than that at Springfield," I should say that the waters were 48 feet shallower in the Westfield basin than over Springfield. Where kame sands were heaped up in the Springfield basin the high terrace is notched in them at nearly the same height as in the Westfield basin; as, for example, on the extreme east of the basin in Wilbraham or north in Holyoke. At the notch in the Divide Range occupied by the West- field River the exact surface of the lake bottom has, of course, been removed by the later erosion of the river; but at the next notch south, at Risings, just on the State line, the surface is well preserved and is very instructive. It is what might be expected on the assumption of a narrow anpJf, QMOJ-WDU wvid en. ll 0OU3UI>J^i QSMHimy^ OB OOZ M.0pz>& lu^ucyj 'PJ. 911CUQ OiVipUJ D 5^ §- Ts 1 ■*o ■0 o .c o ■*o o o <0 -C S. .c (0 0 S i !«5 Q> o 0) ■s" -^ ^ ra <: ra c ■*o V- -J IB C ,S- ID U -s; « nt 40 feet above the liiC~~:''-'^;S^:55>. molded into the complex curves seen ^^tr^^-^^^i^-: ■'"^.^^^C-^ m the more clayey sands above, which ^^^^^0p^^^£&>^^'^3^^^^ we shall see to have been mdepend- ^^^^^^:yzK:-':pl^:^::ip^:^i-':-^^ ently and at a later time subjected to a *^^^^^fe^^-SSS^i^^^^^B similar crushing. The surface of the ^^^.vii-i.^ -■ - ~ Sands rises with an easy even slope F-a-^.-Block of frozen ■• pink sand/' showing Ane sy^^ J 1 tern of joints. The lionzontal linea are the bedding of and sinks with the same smooth lines tl^esand. The vertical are strongly marked joint planes. The block was 15 inches square. below one's sight. A cm'ious appear- ance shown (fig. 45) by a frozen block of the sand may even be due to the development by the weight of the ice mass, or by torsion in connection with its thi'ust, of a latent pressure cleavage. A frozen block from the north end of the east opening of the sands had carved out upon it by the wind a series of perfectly parallel cracks, 3 to 4"™ apart and about at right angles to the bedding, and these, together with the bedding, had been very beautifully dissected out by the wind. The sands here dip southward, and the ice coming from the north had ridden over the upturned edges of the laminse, so that the plane of these fissures was at right angles to the direction of the thrust of the ice. I have little doubt that the ice moved over this sand bed while it was frozen, and that this is the reason why the sands are so often and so sharply faulted and broken and not thrown into twisted, contorted folds,- as happened later to 682 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. the upper sands when the ice was plainly thrust out into the waters of the lake — ^then risen higher — and plowed up its bottom. While I was studying the sands the workmen were breaking off masses of the frozen sand with wedges and heavy hammers to make way for the steam shovel, and the rock thus formed was one of great hardness. The depression which separates the two parts of the sand on the west side is lined by a thick layer of unstratified pebbles concentrated from the pink sands, and seems to me a "pot-hole" in the sand, caused by the waters of a moulin of the glacier, though it may be an old brook bed. The eroded surface of these sands is the third erosion plane occupied by the ice. The second tiU.~(V\. XV, E to T.) Returning now to the south end of the great drumlin, we iiud the second till, which rests upon the pink sands, to have a thickness of 2 to 3 feet, and to be sharply demarcated on the east from the pink sand and the bowlder bed below, the boundary being a straight line, and separated above from a third layer of till by the thin, disconnected remnants of a second sand, which thickens rapidly southward, so that where the second till goes out of sight below, it is separated from the third till above by 15 feet of sand. This second till is the hardest stony clay, wholly indistinguishable from the oldest till of the drumlin, against which it seems to rest in a wedge-shaped end, although no distinct line of demarcation can be seen between them. The ice seems still to have rested upon the surface of the older till, or to have eroded down to it, and a train of large bowlders appears in the second till a little way from the great till, quite plainly derived from it, as farther on they are wanting in the second till. Farther south, where the ice rode up over the pink beach sands as already described, there is but slight trace of till— a few large bowlders resting on the sands — that which we have followed from the oldest till seeming to have been deiived from the erosion of the drumlin; and here the material has failed or been removed at a later time by water, as has much of the pink sand, which one can follow by its color as it is swept southward and now lies between the layers of clay of later deposition, showing that ice and water worked together here. The second sands. — (PI. XV, H to L.) At the sovith slope of the di'umlin the sands which cover the second till and separate it from the third appear only as a thin, disconnected film, rising to a thickness of 8 to 10 inches on the east side, while on the west they are continuous and THE CAMP-MEETING CUTTING. 683 nearly 2 feet tliick iind rise up uutu the back of the flrumlin northwardly, where thev are sheared off abruptly by the tlnrd till, which here blends with the second. Southward, the third till, which passes down the south slope of the druinlin, sinkini^- deei)ly into these sands, rises with an easy gradient up to the surface of tlie sands on the east side of the cutting, its further extent being now cut otf by erosion; and on the west side rising in the same way to within a foot of the surface and then extending 50 feet over the sands, and fiinall)^ thickening downward to 6 feet and ending abruptly in the sands in a club-shaped mass, the sands that rest against its south face being continu- ous with those beneath it and like thein in every way. From this point the sands make the whole thickness of the wall, 24 feet, for a distance of 224 feet to the ravine, and crossing this (25 feet wide), the sands, with the bottom nowhere exposed, run under a bed of clay, the same as that north of the drumlin, and go on with a slight dip southward and fold over the pinlc sands already described. They are here much thinned, and dip beneath the surface near the south end of the cutting. These are, for the most part, coarse to very coarse, reddish sands, laid down by a strong and steady southward current in layers which are horizontal for long dis- tances or slightly inclined southward. Only for a few rods on the west side and just south of where the third till rises upon them are they clean, white, better-sorted sands, the cross-bedding dipping sharply north for a time and then as sharply south,^ and their eroded surfaces are covered by a h^yer of well-worn beach |)ebbles. The bedding is everywhere, except in the white sands, sharjjly marked by thin layers of very fine sand 2 to 6 inches apai't, which are persistent for long distances, and which farther south, where the sands have run beneath the clay, become layers of true clay, and toward the top of the sands approach nearer by the thinning out of the intervening sand layers, and so effect a transition into the clays. For a long distance south of the brook this arrangement is well devel- oped. Layers of sand, beautifully rippled at surface and about 6 inches thick, are capped by layers of clay, one-fourth of an inch thick, which takes an accurate cast of the ripples below and makes the upper surface more or less nearly horizontal. 1 These are back-set and front-set sands, in the terminology of Professor Davis. Bull. Geol. Soo. America, 1890 p. 195. 684 GEOLOGY OF OLD HAMPSHIRE COU:S[TY, MASS. lu the frozen wall the sand layers were deeply worked out by the wmd, and the clay layers projected one above another like the eaves of a fluted iron roof By the thinning of the sand layers the whole deposit loses in thickness and the superincumbent clay sinks lower. This plainly indicates a strong current in the summer floods, a quiet one during the winter, and a depth of water sufficient to so far remove the banks of the stream to the west that the floods brought only thin layers of sand out over the clay to this point, layers which have mostly dwindled to nothing before we reach the south end of the cutting. The sands, as indicated above, run up on the south slope of the drumlin in a thin film, and could in one section be traced almost or quite continu- oiTsly across it, to join a thick bed of similar sand, whicli extends to the north end of the section, where it is cut off by erosion. It is capped, as in the layer south of the di-umlin, by the same thick deposit of clay. These sands are finer than those already described, especially near the drumlin, manifestly because they were laid down in the sheltered area behind it. Northward they grow coarser, and at the extreme north are gravelly, and iron-shot where the water circulated below the clays, with coarse cross- bedding dipping south. Through most of the distance the beds are (or were) horizontal, and show repeated oscillations of coarser and finer layers, and everywhere most delicate cross-bedding. Upward, the whole gradually becomes finer, clay layers -making their appearance, which at the end effect a somewhat sudden transition into the clay above; in short, the sands agree in all points with the corresponding sands south of the drumlin. The third till. — It will make clearer the complex series we are studying if I call attention to the four surfaces on which the ice has rested. The first is the surface of the drumlin. The second is the surface of the second till, which has eroded the pink sands ; and as the till layer seems largely derived from the broad surface of the di-umlin, this layer lessens and the ice rests almost directly on the sands in the continuation of the surface southward. Consideration of the third till, which is the subject of discussion here, may be omitted for a moment. The fourth and last surface occupied by the ice is very clearly defined along the whole length of the section. It is the hori- zontal upper surface of the clays above the second sands from the north end of the section to the drumlin, and is continued along the eroded surface of the drumlin, and is the surface of the fourth till from the beginning of the THE CAMP-MEETING CUTTING. 685 same at the south end of the (h-iiiuHu or southward. This fourth till is made up of material derived from the drumliu and uiolded with the clay and sand below, and so "-rades southwardly into the contorted clays uncontaminated with o-lacial debris, whose planed-off upper surface is the continuation of the fourth surface occupied by the ice in the whole distance south of the centi-al o-oro-e to the place Avhere this ice-worn surface sinks below the level of the section. Because of this blending of the clays with the second sands beneath them by the fourth ice, the relative importance of the third surface occupied by the ice in the midst of the second sands can not be clearly made out. It is seen in the sloping layer of till that extends down through the second sands, ending at the north edge of the brook gorge, and is marked "third ice-worn surface" on the main section. Some part of the deep erosion of the second sands between this point and the south end of the drumlin seems due to this third ice advance. In the opposite side of the cutting, 50 feet west of and parallel with the above, the third layer of till ends abruptly in the sands, soon after thicken- ing to 7 feet and rising nearly to the surface in a way peculiar and difficult to explain. It is here a compact, stony clay, in which, near the end, I coimted twenty bowlders 12 to 16 inches long, all of glacial shapes and many striated. As seen in the wall of the cutting, it ends in three long, sharp teeth projected southward, receiving between themselves correspond- ing projections of the sands; and these projections are made up of laminae, which begin against the till and extend from it with j)erfect regularity, exactly as if it were the fluited face of a sea cliff and the sands had been laid down against it. There seems to be no question here of a thrusting of the ice into the sands after their accumiilation, but it would seem that the third till represents a second advance of the ice after a slight retrogression, and that here it piished itself over the sands of the lake or estuary as before, with the difference that now the water stand was higher and the snout of the glacier was thi-ust out into the lake, gouging and crumpling the beds at its bottom. It ended here for a time and then retreated, leaving the till, which it had gathered mainly from the drumlin, covering the sands to this point. The deposition of the sand continued uninterruptedly except so far as the space was occupied by the ice, and the sand increased arotmd and over the till as soon as the ice disappeared. 686 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. Turnins' to the second cut on tlie east side, which was 50 feet farther out in the lake and parallel to the two last described, we find the till homoge- neous in the lower 2 feet of its thickness. Then it runs up over the sands and thins to a foot in thickness, and is then prolonged in a stratified bed of the same dark greenish-black sandy clay, Avhich ends abruptly (thickening slightly before its ending) in a sharp point, the last portion being beauti- fully cross-bedded and apparently the product of a single plunging wave from beneath the ice. This is inclosed above and below in the light-yel- low coarse sands, which beneath are undisturbed so far back as traces of lamination occur in the stony clay above, and are conspicuously con- torted farther back beneath the amorphous and ice-carried portion of the same bed. At the fourth cutting parallel to those last discussed (the most west- erly), where the till runs up on the sand, it splits into three or four layers, each successive one running up with sharper angle and being separated by thickening sheets of flood sands; and the till reaches here its greatest height. Some layers of the till bend irregularly and sink deeper into the sands and extend farther south, but are cut off by the brook erosion before the connection southward is made. (PI. XVIII, fig. 2, p. 694.) The transition of the sands to clays beneath this till indicates a deep- ening of the waters southwardly, by which the ice was more or less buoyed up allowing a portion of the sands described above to accumulate beneath it after which the ice dropped again upon the sands. This was repeated several times, and at one of these times a mass of water from beneath the ice swept into the sands the curious point of remanie di'ift described above, and finally the ice was floated away to the south as icebergs, allowing the sands to continue their accumulation over the till it had left. The clays above the second sands. — These are, from one end of the cuttino- to the other, the common Champlain clays of the valley, formed from the wash of till, and where not disturbed are thin-laminated in layers 8 to 12™™ thick, each layer buff colored and sand}^ in the upper third, and composed of fine fat clay in the lower two-thirds. The fourth till and its effects upon the clays and sands leloiv. — Starting from the north end of the opening, the surface of the clays is an almost perfectly level surface of erosion on to the drumlin. The ice has passed over it, planing it down to this level, twisting and contorting it and the THE CAMr-MEETING CUTTING. 687 sands Ix'iR'atli it into tlio greatest confnsion, kneailing tlieni tog-ether, press- in"- tlic clav in yrcat bosses down into tlic sands, in some places destroying the lamination of the claN' entirely; in others, where the alternation of line sand and fat clay was more clearly marked, ])roducing- in each layer masses where a smoothed snrface resembled marbled paper. This contor- tion increased to its maximum where the two beds, here inextricably mixed, mounted up the north slope of the drumlin and were sheared off on a plane which is almost coincident with the snrface of the ch'umlin and which is continued south as the upper surface of the fourth till. It w^ould seem that the ice pushed out into water of considerable depth, and so, partly buoyed up, w^as able to move over the plastic clays, producing a minimum of erosion and depositing no till on the clay ; but the drumlin acted as a resistant substratum, and between the two the stratified beds were sheared off entirely, the hill itself was scalped, and the combined material was trailed along over the remnant of the sands down the slope (a train of great bowlders occupying its lower portion) for a distance south- ward from the drumlin and plainly derived from it. Masses ranging from filaments to large sheets of the sands or clays, or beds containing alterna- tions of these two, are contained in a formless mass of till of great compact- ness, which rests with a flat under surface upon the sands below. I have figured a surface of this bed (PI. XVIII, fig. 3, p. 694) where it is just beginning to mount again upon the sand (above D, PI. XV). The upper layer of till is crushed into the sand layer, its bowlders plowing into it and producing folds and faults ; while below, a thick bed which once consisted of clay with thin sand partings is as a whole kneaded into such a confused mass that, over the broad, smooth siu-face sculptured by the wind, wherever the sand layers come to the surface they were eaten out into intricate convolutions, like the interior of the ear. That portion of the sands caught between the second and fourth layers of till has all structure crushed out of it; but as the upper till layer rides up onto the thick mass of the sands, the line between them is sharply defined, being gently convex below; and as the sands thicken, signs of bedding gradually disengage themselves from the confusion of the mass, and one sees the effect and direction of the tlu-ust of the ice marked with wonderful clearness in the contoi-ted layers of the sand. Within the sands the layers are quite horizontal and undisturbed, and as one follows each back toward 688 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. the till, it is after a few undulations thrown into a series of sharp zigzags or short distinct faults, and the layers thus brought into a vertical position together or bent over southward are then dragged along beneath the ice, running a few feet parallel to it and blending together into a confused layer a foot or two thick, in which no structure is visible. On the opposite side, after the last cutting, the appearance was very similar, except that the layers were thrown into still greater confusion, and for 4 rods all the upper half of the sand, 13 feet in thickness, had been pushed 15 or 20 feet southward, the layers now standing on their heads and thrown into folds as complex as the sutures of an ammonite; and farther on the whole mass has been wedged in between the layers of the sands in advance, heav- ing them up and occupying in a contorted mass a great triangular space beneath them. The till rides over the whole, and every layer of the con- torted mass as it comes up from below, as well as of that thrown up by the underthrust portion, bends over southward beneath it as the smoke curls over the chimney edge in a strong wind. As the till continues southward over the sands it moves parallel with their lamination and disturbs them very little, and at last, as it thickens downward, it cuts across them at a low angle, and the layers just below run on continuously and show no signs of any effect from the ice. Where the fourth till rising over the sands is cut off by erosion it is already largely composed of the contorted clays. Southward, across the ravine, the surface occupied by the ice sinks into a gentle depression and rises over the pink sands and goes down below the level of the ciitting near its south end. It is a surface and noth- ing more, and in this long distance south of the brook ravine, as well as in the equally long distance north of the di'umlin, no trace of till is found upon it. Only in the remaining space, from the drumlin south, the stratum of till is carried forward along this plane, and it is unfortunate that its ending is not to be observed in a satisfactory way, owing to erosion. It is, therefore, not strictly proved that this plane is continuous, but the identity of the beds north and south of the drumlin makes this highly probable, and an inspec- tion of the section will show it to be the only simple supposition, any other requiring an additional recession and advance of the ice. Everywhere below this plane the clays are variously contorted, as in the reach north of the drumlin; often clay and sand are curiously molded THE CAMP-MEETING CUTTING. 689 t()>;i'tlic'r, and iiuiuecliatel}' abo\e it the newer clay or saiul is wholly uudi,s- turbed to the surface of the terrace. Also, in many masses of the laminated clays a beai;tiful jiressure cleavage has been developed, a series of fine, closely apin-oximated slip faults making- a large angle with tlie lamination, and dip})ing sharply northward in the direction toward which the pressure came. The section shown in PI. XVII (p. 692), taken from a similar locality, uiight have been many times exactly duplicated in the first 300 feet south of the brook. I assume this work to have been done by the advance of a glacier into the water, and not by icebergs, because only a single great body of ice moving over the soft mass of clays could have planed them down to so true a level except when the protuberance of the di-umlin caused an irregularity in its action, and the great disturbance of the clay and subjacent sands for a depth of above 20 feet over so large a space would indicate a mass of very considerable thickness which was pushed over the surface and not simply carried forward by the current. Except for these reasons, I do not see why a continuous mass of floe ice might not have done the work, for the scratched bowlders in the till layer seem to have been derived from the earlier till of the drumlin. The up]ier sands and clays. — Above the line of disturbance a heavy layer of coarse sands, grading southwardly into laminated clays, smoothes over the irregularities of this surface and builds up the terrace to its completion. Commencing at the north end, the sands dip sharply south- ward and represent plainly an advancing delta front or growing sand bank, the sands having been pushed over its surface and deposited upon its southern slope. In the central portion, over the more irregular surface of the fourth till, these sands — which, where the till ends abruptly, I have already described as deposited continuously against the southern termination of the latter, and as thus being continuous with the sands below the till — are carried on in broad, more nearly horizontal sheets, with finely developed flow-and-plunge structure. Southward, beyond the brook ravine, these horizontal sands are capped by a thick upper layer of cross-bedded sands which dips sharply south and which probably represents the further advance of the delta or bank from the north, the intermediate connecting portion having been removed by MON XXIX 44 690 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. erosion. In this lower horizontal bed the alternation of coarser and finer layers is very marked, and southward the finer layers gradually change to clay, while the coarser grow thinner and finer and at last run out or blend with the clay layer forming its lower and sandier third. At the same time the boundary between the horizontal and cross-bedded portions of the bed rises slowly southward, since each layer of the latter coming to the bottom of its slope bends sharply to a horizontal position, and, gradually dwindling to become the sandy portion of a clay layer, runs on between layers of the clay, which, coming from the south, bend at the foot of the slope, rise up for a distance upon it, and grade into a layer of finer sand which forms the upper part of the sloping sand layer. Thus, going south, the horizontal layers gain at the expense of the cross-bedded layers above, and the clays gain at the expense of the coarse delta sands, and one has the clearest illustration of the blending of the shore sands and the deep-water clays of the Champlain epoch, and evidence of their synchronism with the later events of the Glacial epoch. r£sum:6. The facts detailed in the above section enable us to construct the following pictures of the succession of events over this area: 1. The formation of a drumlin as a part of the ground moraine of the first or great glacier. 2. The recession of the ice to allow of the formation of the bowlder bed which lies at the foot of the drumlin and may be a terminal moraine, or may be due to water action concentrating it from the drumlin itself. 3. The formation of a true sea beach of great extent — the pink sands. 4. A second advance of the ice, rising over the drumlin and eroding the frozen beach sands. 5. The second recession of the ice, and the deposition by the flood waters, from its melting, of a great body of sands. 6. The rise of these waters so that an equally great body of clays was deposited upon the sands. 7. A third minor advance of the ice over these clays, molding them into the sands below, removing them entirely over the drumlin, and south of it for a long distance gouging deeply into the sands and covering them with a layer of till derived largely from the drumlin, finally riding up onto U. 9. QEOLOOICAL fiUHVEV MONOGRAPH XXIX PL. XVI ^ ■^''^ ^ I ' rf — \ \,^ SURFACE OF ICE-CONTORTED CLAYS SMOOTHLY CUT WITH A KNIFE, EAST OF J. RYAN'S HOUSE, HATFIELD, NATURAL SIZE. SECTION OF CLAYS IN HATFIELD. 691 the c-lays ag-aiii, and being buoyed up and carried off southward by the flood waters, which still covered the clays in considerable depth. 8. The continued deposition of the flood sands in waters somewhat lowered on the recession of the ice, so that coarse sands, with flow-and- plunge structure, are laid down, obliterating the irregularities of the sur- face and completing a ten-ace of apparently simple structure. The drumlin (1) must represent the work of the general glaciation. The bowlder bed and the pink sands (2 and 3) must represent an inter- glacial period of sufficient length to allow of the long-continued and imin- terrupted presence of a large lake or estuary, and to make this possible the ice must have receded far north of this point in the valley. The events of 4 to 7 indicate a second advance of the ice, with minor oscilla- tions, during the last of which the end of the valley lobe of the glacier was tlu-ust out into the waters which then filled the valley and by which the laminated clays were being deposited, while at the same time the high terrace gravels were gathering along the shores, a work which on the disappearance of the ice continued to the completion of the terrace. The discovery of isolated pockets of glacial ddbris and disturbed patches at various places in the clays farther south, which must be referred to icebergs or ice floes (described in the following section on the Cham- plain clays of the Iladley Lake), and of arctic plants also in the same clays,, completes the picture of the events of this time, and indicates that the Champlain clays and sands were here in part synchronous with the Glacial period. SECTION OP CLAYS IN HATFIELD SHOWING GREAT DISTURBANCE AND PRESSURE CLEAVAGE. About 3 miles northeast of the section last described (within the same portion of the Connecticut Valley, bounded on the west by the crys- talline rocks and on the east by the long ridge of Mount Warner, so that it is in a sense a continuation of the Deerfield Valley), at the southern foot of the red sandstone hill which rises north of the village of Hatfield, a small opening was made in the clays, which reproduced exactly the upper level of disturbance of the Camp Meeting section. The section was 33 feet east of the first house westward from the hotel on the lu-st road south from the ferry. 692 GBOLOaY OF OLD HAMPSHIEE COUNTY, MASS. Above a liorizontal line the clays were horizontal and normal in every respect; below this they were extremely contorted, as indicated in PI. XVI, showing a smoothed surface about 4^ inches square. Where the contor- tion was less pronounced, about 2 feet below the plane of disturbance, in a layer with a thickness of 1 foot and a length of 15 feet, a beautiful pressure cleavage was developed, superinduced upon the original lamina- tion (PL XVII) in the whole mass of the clays along parallel planes 2 to 4""" apart and dipping 32° N. Every stage in the series, from a slight, sharp monoclinal fold affecting all the laminae along a single plane to the shearing off of the laminse by small parallel slip faults, could be followed, and the clays, parting easily along the lamination and cleavage planes, broke up into a mass of long pencils. The more marked fissures are about 6 to the inch; between these again are finer ones, making the whole number about 18 to the inch. Along these the clay is very often slightly faulted; in one case the slip amounts to an inch. The newly formed cleavage layers have undergone decided compression and distortion, which is brought out clearly by the difference in color of the upper and under portion of the original laminae, so that in tracing the dai'k bands across the several cleavage planes we find them moi'e or less separated into parts placed slightly en Echelon by the con- tinued faulting in one direction, and these pai'ts variously elongated and flattened out parallel to the plane of cleavage. We may assume the plane of this incipient cleavage to be noi'mal to the direction of pressure. A second system of distant fault planes occurs at right angles to the first, which are more distant from each other and have greater tln-ow, showing that the parts of the bed slipped slightly on each other in the direction of the pressiu-e. The locality is 125 feet above sea, and thus somewhat lower than the preceding' section, but it lies out in the valley, where the clays did not reach so great a height as on its borders, and the disturbance must have been very nearly — I think it may be assumed to have been exactly — syn- chronous with the last disturbance of the previous section. Its position under the lee of a prominent hill, protected from icebergs and floe ice, would also point to a continuous mass of glacial ice as the agent of its formation. U. S. QCOLOQICAL BURVEY MONOORAPH XXIX PL. XVII MOTfON OF THE ICE JOINTS AND FAULTS IN LAMINATED CLAY, PRODUCED BY THE WEIGHT OF THE ICE. PLATE XVIII. 693 PLATE XVIII. THE WAPPING AND CAMP-MEETING CUTTINGS. Fig. 1. — Section of the fine-grained, contorted sands at the Wapping cutting on tlie Canal Railroad, in Deerfield. Fig. 2. — Section on the west side of the Camp Meeting cutting, between the south end of the drumlin and the hrook, showing the fourth advance of the ice into the second sands. It is opposite to the portions G and I on the east section, 3 rods distant from it. It is represented by the discon- nected mass of till above I. Fig. 3. — Enlargement of jiart of PI. XV (p. 678) at a point halfway between G and H, and above the second sands, where two large bowlders appear. It shows the passage of the fourth ice over the older clays, here nearly all eroded, and the thrust of the bowlders into the clays, and the kneading together of the clays and subjacent sands. The wind erosion of the sand has pro- duced deep, ear-like depressions in the lower part of the frozen wall. 694 18 . S. GEOLOGICAL 6URVEV MONOGRAPH X }ici/fjin^ CH.L, JJe^r fteZd. . COy£^/7SO COl^£R£D ^ AND r^ SANO THE WAPPING AND CAMP MEETING CUTTINGS, SECTIONS OF TERRACES AND LAKE liOTTOMS. 695 THE WAITING CU'I'TINC. CONTORTKn SANDS AT TlIK CUTTINCi AT WAI'I'INO, IN DKKRFIELI). This section, tig-urecl on PI. XVIII, fig-. 1, was exposed in the works for the extension of the New Plaven and Northampton Raih-oad northward, 2 miles south of Deerfield Station, at the same time with the Camp Meeting cuttinff described above.^ The deha of the Deerfield River in the Connec- ticut Lake (p. 634) extended out as a broad, flat alluvial fan from the mouth of the river gorge in the crystalline rocks at West Deei-field, reaching clear across the lateral Deerfield Valley to the foot of the trap ridge. The northern half of this delta has been removed by the later erosion of the Deerfield River, and from the northern rim of the remainder one looks down north into the deep basin thus formed. As the railroad approached this from the south its grade was lowered and a deep cut was made, so that it could pass across this basin by a high embankment. The section given was taken from the west wall, and is thus the north-south section through the middle of the delta, and just opposite the mouth of the gorg-e of the river. It is made up of fine to very fine, well-sorted sands, in layers 1""™ thick and made more distinct by the infiltration of iron. Coarser layers, drying white, and thus standing out prominently, occur 3 to 4 inches apart, and still coarser and thicker ones of the same character about one-half inch apart; these are represented by the heavy lines in the drawing, and they render the contortions visible for a long distance. The length of the section is 278 feet, the greatest height 45 feet above the railroad. Toward the north end the whole thickness of the sands in the section is crumpled, manifestly by a strong force coming from the north, the disturbance of the sands being greatest in the northern half of the section and in the lower portion of this half, and appearing more in detached patches in the southern part. The cutting was carried along the west side of a brook valley running north, and the rain washed a gulch from the cutting down to the level of this brook, exposing the fine sands for 20 to 25 feet below the base of the section. The disturbance lessens downward and the base of the sand is more clayey and rests unconformably upon coarse, reddish, cross-stratified sands derived manifestly from the sandstone, while the upper sands are as plainly derived from the crystalline rocks of the west. The line of junction of the two sands is extremely irregular, the lower beds having been much eroded before the deposition of the upper, but the ' See E. Hitchcock, Geol. Mass., 1841, p. 363. 696 GEOLOGY OF OLD HAMPSHIRE OOUl^TY, MASS. junction is such as can have been formed only by the forcible kneading together of the beds. A little farther south, on the border of the delta at the house of Captain Briggs, the red sands come to the surface in a long knoll. They are finely cross-stratified and dip south with an angle as high as 35°. At the cutting these under sands are much jointed and faulted, as if they had been subjected to pressure before the deposition of the upper sands. I identify these lower sands with the pink sands of the Camp Meet- ing cutting (p. 680), and believe them to be a remnant of beds deposited after the first recession of the glacier, while the Deerfield Valley was still sealed up by the ice, which escaped the erosion of the second advance of the ice and on its recession were covered by the sands of the Deei-field delta. It is not clear to me what could have caused the extensive disturbance of the upper sands. This disturbance is to be seen in the upper side of the road running parallel to the cutting, several hundred yards to the east. It is at its maximum at the north end of the opening, where the sands have been removed by the erosion of the river and extended an unknown dis- tance to the north in the beds before their removal. On the north side of the Deerfield RiA^er basin, a mile west of Cheapside, a complete section of the sands from the surface down to the till was quite normal and undisturbed, as also on its northwest border and on the island of the terrace sands which rises in the middle of the basin. Small detached areas of disturbance in the delta sands are common from the head of the latter south to the south line of Deerfield, plainly caused by stranded ice, but here a force of much greater magnitude was certainly concerned. I have described on page 630 the deep, long depression along the west line of Greenfield in which Grreen River flows and which was occupied by the west lobe of the ice that fotmd place in the valley while the flood sands brought in across Greenfield through the Bernardston strait were building up the high terrace in Greenfield, and this lobe of ice, extended south, would have come in contact with the delta of the Deerfield from the right direction to have plowed up the sands as we now find them. The high terrace sands are, however, undisturbed right across its supposed track west of Cheapside, and though these sands may have been swept in a little later, their presence renders this explanation only remotely probable. Another possible explanation is that the axis of the delta of which these beds form a part lay to the north of this spot, and along this axis the greater thick- ness of the beds caused, by their weight, a flowing of their fine sands. CHAPTER XX. THE CHAMPLAIN PERIOD (Continued). THE CHAMPIiAllSr CLAYS. INTRODUCTION. On the retreat of the ice and on the occupancy of the basm by the flood waters clays began to be deposited over all the bottom, far from the shores, where the current was not too strong, and sands and gravels accumu- lated off the mouths of all the tributaries, and were moved along the shore lines by the shore currents and out into the deeper water by the undertow. The two deposits are therefore strictly contemporaneous, and their laminse are intercalated with each other at their point of junction (see p. 690). The sands were pushed in deltas rapidly out over the clays, so that their place of junction is a plane with small shoreward dip. It is strictly synchronous with this earlier portion of the flood sands, since, as detailed in the last section, the increased velocity of the flood earned sands out over the clays in every portion of the lake bottoms, even in the most sheltered, like the East Street basin in Amherst. The scanty otitcrops of the clay, mostly along river gorges, are marked by a line of purple dots (1 b c, PI. XXXV), this color having thus a litho- logical value, while the other post-Tertiary colors on the map have rather an orographic value. The great importance and magnitude of this terrane can be seen best in vertical sections, as upon the map it is represented only by thin lines along the river courses and road cuttings and on the steep slopes of ter- races; elsewhere it is covered by the succeeding beds of sand. CLAYS IN THE MONTAGUE LAKE. The clays appear in great force above and below Northfield village, where the brooks cut back in the lake-bottom beds, and here they rise 233 feet above sea level. Southward they appear frequently in brook cut- tings in the bottom beds, and at Northfield Farms they rise to 270 feet 697 698 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. above the sea. The fat layers are one-fourth to oue-half inch thick and 6 inches apart. This great height was due to the rapid advance of the Millers River delta across the channel farther south, which checked the current to the north. How rapid this was is seen by the section, fig. 41 (p. 688), where far out in the delta the sands rest directly upon the rock. As the delta was extended westward its sands were doubtless carried up grgidu- ally over the clays, for in the long erosion scarp cut in the western face of this delta from Turners Falls around nearly to Montague village, only a small thickness of sand rests upon the clays, which rise to a height of about 213 feet above the sea and rest upon till or sandstone with a thickness of about 34 feet. The layers average about 1 inch, one-half fat clay, two- thirds fine sand. (See fig. 35, p. 629.) CLAYS IN THE HADLEY LAKE. I have mentioned an isolated occurrence of clay ]30orly exposed at a schoolhouse in the north of Greenfield. Around Greenfield village the clays are in great force and rest upon till, as seen at the clay pit in the village and on Fall River where the road to Franklin Park crosses it. These clays were continuous through the notch of the Deerfield River, and perhaps also connect farther north, through the passage at the mouth of Fall River, with the Turners Falls clays. Southward they crop out abundantly around the erosion basin of the Deei-field River, to near its soiith end, opposite the mouth of the river gorge, where, from the increased current and the increased material broiight in by the river, the clays are replaced by a great thickness of sand, which, in the center of the basin, becomes exceedingly fine, with distant clayey partings, as seen in the Wapping cutting (PL XVIII, p. 694), where these fine sands rest discord- antly on the problematical reddish sands which are there described. Farther south, through Deerfield and Hatfield, the sand plains are nowhere cut through to the clays below until the region of complex oxbows of the Connecticut west of Hatfield village, described later (p. 734), is reached, where, in the terrace scarp, the clays appear in great force ; only at one place, at a clay pit near the pistol factory, is the substratum — here coarse till — exposed. Southward, beyond the influence of the Deerfield, the whole broad bot- tom of the lake is underlain by a continuous stratum of clay of unknown, CLAYS IN THE HxVDLEY LAKE. 699 Init in tilac't's certaiiiK' of very great, thickness, and tlie clay has done more than all the other beds to obliterate tlie vei-tical irregularities impressed uiion the basin by the ice. It still underlies tlie whole flood plain of the Connecticut, and although the river in its oscillations has cut in the clay a broad and deep channel, it has not cut through to the base of the clay stratum, except opposite North Hadley,- where a reef of sandstone projects throuo-li, and at the knee of the great bend, where the river has worn into a submero-ed drumlin. This great bed of clay continues southward to the Westtield River, where the conditions of the Deerfield are exactly repeated, and the clays are replaced by the fine delta sands. It extends everywhere under, and sometimes very far under, the shore terrace, notably in the case of the Mill River delta in Northampton, where the clays spread under the delta deposits clear up to the " Bay State," near Florence, where they are worked in large brick pits and rest on till with a thickness of 23 feet. There are also large pits near the asylum. It reaches apparently its greatest thickness under the Northampton and Hadley meadows and in the East Street basin in Amherst. At the Belden silk mill, near the station in Northampton, the clay was reached beneath a few feet of sand, and its bottom was 140 feet below the surface — that is, about 12 feet below sea level. Beneath the clay was 10 feet of red sand. The clay was pierced 112 feet at the piers of the overhead bridge at the Northampton station. The trial piles at the Northampton bridge over the Connecticut, heavy timbers well jointed and hooped, were driven 113 feet below low water — that is, about 13 feet below sea level — without finding bottom, and after the pile had rested in its place for the night the first blow in the morning advanced it as much as the last of the night before, which would have hardly been possible in any material except a very plastic clay. The piles for all the piers of the bridge were driven 30 feet below the river bottom in the same clays after passing through the river gravels. About 1,500 feet north of the bridge the clays rise in the high western bank of the river about 72 feet above low water. This is just opposite and only a few yards from the south end of the Camp Meeting cutting (see p. 677), and the clays between are continuous. Thus their maximum thickness may be about 218 feet. About the same distance south of the bridge they are exposed for a long way in the river bank, at the south end of Hadley street, a locality furnishing fossil leaves (see p. 738) and an 700 GEOLOGY OF OLD HAMPSHIRE OOUl^TY, MASS. abundance of concretions. They have been extensively worked near the asylum in Northampton, at Rich's brickyards east of Southampton, and at Pomeroy's yards west of that town. They extend east from the river with the full width of the space between Mount Holyoke and Mount Toby, rounding Mount Warner, and being very near the surface in all the area south of Mount Warner to near the foot of Mount Holyoke, where wells 81 feet deep (Mr. Shipman, Lawrence plain, south of Hadley) and 40 feet deep (A. Losey, Nuttinsville, Amherst) were sunk in sand and fine gravel without reaching the base of the sands. Over the bottom of the central depression the clays, being near the SLirface, are often sandy. The clay is worked for bricks at Plainville (north of Hadley) ; and in the bottom of the depression, where Amity street crosses the brook, it was formerly worked. Near by, at Mr. Stebbins's barn, a well went through the following: Feet. Fine sand 2 Clay 7 Sandy clay 13 At Mr. S. Harrington's, in North Amherst, a well 90 feet deep did not reach the bottom of the clay. Tlie clays rise high up oil the slope of the Amherst ridge and thin out under the shore gravels. The}^ are cut through by all wells along the slope below, 246 feet above the sea, and reached their greatest observed height in the col between the central depression and the basin south of College Hill, where they were exposed by the Central Railroad cutting from the bridge at Main street east to the end of the cutting, with a maxi- mum height of 260 feet above sea. (See p. 645.) These clays thicken out east into the land-locked basin south of College Hill, where in Champlain time they accumulated to great thickness, with little covering of sand. In the East Street basin the clays were also developed to great thick- ness, and over all the area south of the road to Pelliam lie very near the surface. At the third house east of the bridge on this road (Mr. Hubbard's) a well was bored 50 feet in clay covered by 8 feet of sand, and the water rises to within 2 feet of the surface. In a well on the south side of the same road near the middle of East Street (Mr. Clutia's) this section was exposed: Feet. Fine sand : 7 Clay 23 Fine quicksand 3 Till 3 THE SPIUNGPIELD LAKE. 701 Fartlier soutli, where the road from the village crosses Fort River at tlio l)rick 'jitf*, the following- section was taken from the exposure in the pit and iVoiu a well adjacent: Feet. Fine sand 6 Cliiy 35 Till Pockets of pebbles were found in the clay, and the water, very sul- phurous and irony, came to within 5 feet of the surface. Fossil leaves occur here. CLAYS IN THE SPRINGFIELD LAKE. There are no brickyards in Agawam and West Springfield, though the clay crops out at Riverside. There are extensive brickyards along the east side of the river at the following points: Above the Holyoke bridge; at Willimansett; in the northern part of Springfield; and especially beside the Boston and Albany Railroad in the southern part of Springfield, and across the line in Longmeadow. Eastwardly the clays are deeply covered by the thick sands of the Chicopee River delta, which extend across Wilbraham and Springfield. CONTACT OF THE CLAYS UPON THE TILL. The section exposed at the hoe factory in Northampton, and illustrated in fig. 31, p. 540, not only shows the contact of the till upon the sandstone and the upper surface of the former, upon which the ice rested, but also demonstrates that the deposition of the clay followed immediately upon the disappearance of the ice, under circunastances which indicate that the ice cotild not have melted in place upon the till; nor could the till have been exposed to subaerial erosion before the clay began to be deposited. In the former case a loose deposit of upper till must have intervened over the till and sandstone alike; in the latter, the till would have been eroded below the level of the sandstone, and the common uniformly curved surface would not have been preserved. It seems to me probable that at this time — ^the end of the Glacial period for this basin — the waters stood over this place, which is about 135 feet above the sea — and of course over the whole basin — at a height so great that the ice was at last buoyed up and floated away, and' the clays began immediately to be deposited upon the surface thus abandoned. 702 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. The clays were grayish-blue, very fine, "fat" clays, agreeing exactly with those worked in the large brickyards a few rods south. When both the clay and the till were wet the sharp, curved line of junction was inconspicuous at a little distance, the whole surface presenting a uniform dark bluish-gray color, but above the line a cane could be easily tlu'ust into the clays for its full length, while the blow of a hammer would not make much more imjoression upon the till than upon the neighboring sandstone. The section is situated far within the limits of the high terrace and is exposed by the deep erosion of this terrace by the Mill River, which, in cutting down to this level, has carried its gravel beds over the whole, making the upper horizontal stratum in the diagram. If we restore the terrace here to its condition before it was affected by the erosion of the river, we shall need 25 to 30 feet of clay resting on the till and covered by 35 to 40 feet of sand to bring the level up to 200 feet above the sea, which is the height of the terrace over this area. The clays are exposed with this thickness in the face of the high terrace on both sides of the stream. When the ice disappeared, however, and the deposition of the clay began, this was a deep depression between the "drumlins" of till upon which the hospital and Smith College stand, opening southward into the main basin. In another section, from the Canal Railroad, exposed just west of the South Street bridge in Northampton, the clays rest also directly upon the stony till, and although greatly disturbed by stranded ice and mixed with material dropped from it, there is everywhere at least a foot of the fine clay, undisturbed, intervening between the till and the horizon where coarser iceberg material appears. In many of the clay pits the base of the clay is reached, and it is always in contact with the till. In the Central Raih'oad cutting south of College Hill, in Amherst, the following section was exposed beneath the bridge (figs. 37, 38, and 39, p. 645) : On the till, which appeared just above the bottom of the cutting, but arose westwardly to occupy nearly its whole thickness, rested coarse, cross-bedded sands, which had been swept fi'om the west over its surface, and which reached a thickness of 3 feet; upon these rested clay, reaching a thickness of about 7 feet, in the lower half banded in layers 1 inch thick, with fine sand partings; many layers resting below between undisturbed TllK STRUOTURE OF THE CLAYS, 703 laA'ors wore contorted tor a sliort distance in a most complex way, as indicated in tijj;-. 38, and on a larger scale in fig-. 39. It seems clear that the friction of the current was sufficient to slide the layer of tenacious cla}- upon its substratum of fine sand for a short distance and cniiiiplc it up, for each of these crumpled layers is covered by an unusually thick and somewhat coarser film of sand. The layers grow thinner toward the surface, and the upper 3 feet is an unctuous, nonlami- nated clav. It is capped by the coarse beach gravel, which rises to the surface. This is the only occurrence of" sand beneath the clay I have seen in the basin, though the fine sands of the Wapping cutting (see p. 695) are so exactly equivalent to the clays that the red sands upon which they rest uncomforiuably may come in the same category. It is also the highest point reached by the clays (251 feet), and here the till was for a short time swept by a strong current from the main valley into the East Street basin before the deposition of the clays began. THE STRUCTURE OF THE CLAYS. The upper horizontal laminse in PI. XVI (p. 690), from Hatfield, illus- trate the minute structure of the Champlain clays. Over all the central portion of the basin they are uniformly thin, even bedded, and horizontal, show a regular alternation of fat and lean portions, and on drying separate easily into layers, each of which consists of a sandy part below and a fat part above, which grade into each other. The brickmakers call the "fat" portion clay and the "lean" portion sand, distinguishing more closely than the geologist. On the river bank at Hadley the lower and much the larger portion of each layer is an extremely fine sandy clay, drab colored when wet, pale buff when dry, composed of a fine, sharp, quartz sand, 0.15 to 0.24"" in size, and of kaolin in irregular elongated particles, affecting reniform and sausage- like shapes from flocculation. This passes rather abruptly, by the lessening of the percentage of quartz grains, into an upper and finer portion, which is generally one-fourth to one-fifth the thickness of the lower portion, of dark bluish-gray color when wet and olive green when dry. It contains a small proportion of kaolin, the rest being very fine quartz grains. Its average grain is 0.0008 to 0.0016"™ for the kaohn. In a specimen taken from the bank of Fort River, below Mill Valley, in Amherst, where the olive-green upper portion was 0.7""" thick, the coarser 704 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. lower pai't 0.3"", measurements of grains from the top of the layer gave 0.0018 to 0.00735"™, from the middle of the lower portion 0.00735 to 0.0294"", and from the bottom 0.0735 to 0.147"". At the Hatfield locality (PL XVI, p. 690) in each layer the lower two- thirds is much coarser than the upper third, and is in reality an exceedingly fine sand, under the microscope appearing like a quartz sandstone, the grains angular, 0.0037 to 0.0075"" in diameter. Besides quartz, there occur feldspar, mica, and a few acicular microlites. The lower portion was olive green when wet, drab when dry. The upper portion showed, both wet and dry, a darker shade of the same color, but the difference was much more marked when it was wet. Under the microscope it appeared like the other portion, except that it was much finer ; but there were present many minute opaque particles of koalin, oblong or sausage-shaped, which showed the Brownian movement finely. The size of the quartz grains was 0.0011 to 0.002"". THE STJBPACE OF THE LAYERS. In some cases the layers are joined so closely that one can hardly dis- tinguish the line separating two laminae from that dividing the finer and coarser portions of a single one. Grenerally there is at least a thin film of rust, showing that the waters have sought out the planes of separation between the layers, as affording them easier passage, and the clays on dry- ing split readily along these planes. On these delicate surfaces one detects rarely the undulating tracks of worms or the small coriaceous leaves of arctic plants. On other surfaces a delicate ripple marking appeared, regularly arranged — broadly elliptical depressions several inches long and of so slight depth that their presence might easily have been overlooked if they had not been brought out by a film of reddish sand, which filled the hollows and was mostly wanting upon the surrounding ridges. The depth of the depressions was often only equal to the thickness of a single grain of the fine sand. This surface sand pre- served, also, the delicate water-drift structure impressed upon it by the current. The rijjple marks and these drifted sands together register, in each case where they occur, a flood so considerable as to give the whole body of water in the lake a current strong enough to enable it to drift along the bottom slieets of the red sands from the border beds farther north out to this point in the very middle of the lake. THE STKUOTUllE OF THE CLAY. 705 THE LATERAL PASSAGE OF THE CLAYS INTO THE HIGH TERRACE SANDS. At the soutli end of tlie Camp Meeting cutting (PI. XV, j). fi78) the whole thickness of the cutting was in clay and the plane of junction dipped north with a low angle, so that the clays ran far under the sands and dis- ai)peared below the level of the cut. The sands were part of a delta or bar front, advancing southward and dipping sharply in this direction in quite thick layers which at the bottom of the slope became horizontal, thin- ning rapidly and running out between the clay layers, becoming finer o-rained and disappearing or merging with the coarser portion of a layer ot the clay. On the other hand, some layers of the clay ran up the slope between the sand layers for a distance, becoming coarser and merging with an upper and finer portion of the sand layer. THE PASSAGE OF THE CLAYS INTO THE SANDS ABOVE. The delicate partings of sand described above (p. 704) increase in number and in thickness as one approaches the upper surface of the clay, and finally effect the passage of the one into the other. In the river ba.nk below Hadley, the locality which for the most part furnished the type of the preceding descriptions, the upper portion of the clays has been carried away by the river, and its sands rest unconformably upon the eroded surface of the clays. The true passage beds are best exposed at the extreme south end of the Camp Meeting cutting (PI. XV, p. 678). Nine feet below the upper surface of the clay these partings are one- sixteenth of an inch thick, of coarse red sand, and are very frequent, so as to give the blue clay a reddish tinge. This continues upward for 3 feet, when a 4-inch layer of coarse red sand intervenes, which is followed by a band 5 feet thick, where the red sand and clay, alternating in fine but regular layers, are in about equal quantity. The whole is capped by another thick layer of red sand, which grades into the ordinary buff flood sands, here only 4 feet thick. Opposite the Hatfield Hotel begins a long, narrow remnant of the old lake bottom, which, by a curious freak of the river, has been left intact, while the river has cut away on all sides of it. This preserves the old sur- face of the clays and the passage beds into the sands above. There is here, within 4 feet, a very gradual passage from the fine clays into fine, white MON XXIX 46 706 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. sands, locally reddened with iron in the lower part, where the waters have stood on tbe surface of the clays. Over all the broad plain south of Mount Warner, over which the road from Amherst to Northampton runs, the passage beds, seem to me to be present, and the clays seem still to retain their full height. EXPLANATION OF THE STRUCTURE OF THE CLAYS. In introduction of this explanation a further peculiarity of the clays may be here considered. While the "fat" portions of the clay layers are very uniform in thick- ness and grain, the variation in the thickness of the layers depends upon a thickening or thinning of the sandy portions of these layers, which may or may not be accompanied by a corresponding change in the grain of the latter. At times the fat laminae separate and take in between them 12 to 16 inches of a sand but little coarser than that of the coarse portion of the layers at the Hadley locality, as is the case in a large portion of the Wap- ping cutting. At other times the grain increases to medium or coarse. The fat laminae seem to be purely a sediment of matter held in suspen- sion when there was scarcely a trace of current, the lean laminae to contain in gradually increasing proportion the fine material carried over the bottom by the friction of a slow current, which was regularly intensified for the formation of the thin films of sand which separate the layers. One finds these clays as regular as a pile of thin deals over all the basin, and I imag- ine that each layer represents a year's work of the flooded river. The fat layers were thrown down in the winter impartially over every poi-tion of the lake bottom, and with the breaking up of the ice in spring the flood swept it off those portions where it had strong current, at times just crumpling it, as shown in figs. 39 and 40, p. 647, but over the deep lake bottom only rippling its surface, the fat tenacious clay resisting erosion slightly, while the coarse material brought in by the tributaries was pushed in sheets out over the delta flats and dumped over their fronts, and in small quantity carried out over the clays. In exceptional floods thin films of these sands were carried down across the very middle of the lake, as at the Hadley locality, and came at the beginning of the spring, for the coarse sand rests directly in rippled hollows of the surface of the finest clay. In this sand are found the twigs and reeds and leaves brought down by the tributaries, and the TUE STKUOTUKE OF THE CLAY. 707 sands fjrado up'vard into the lean portion of tlie layer, whicli represents tilt' uuifonu high water of the glacial river during the summer and which is a true "o-letchermilch," and this in its turn grades vipward into the fat deposits produced by the clarifying of the waters during the succeeding winter. This would conspire with the fact that the mass of the coarse material of these deposits has been brought in from the sides and moved but little downstream, to indicate a low pitch for the valley during the time of the glacial stream. THE TIME OCCUPIED IN THE DEPOSITION OF THE CLAYS. The considerations of the preceding section afford data for a calcula- tion of the time occupied by the deposition of the clays, which is presented as interesting rather than specially valuable. If we take the clays exposed in the south of the Camp Meeting cutting and in the river bank adjacent, a thickness of 72 feet is exposed down to the water level, which would give, at an average of two-fifths of an inch per layer, 2,155 years. If we take the boring at the Northampton bridge, 113 feet, we have 3,390 years. As these two neighboring sections are measured, the one up and the other down, from the river level, we may add these two numbers to obtain a maximum time for the deposition of the clays — 5,545 years. The erosion of the Deer- field and Westfield basins and the wearing back of Turners Falls in the red sandstone a distance of 3 miles, with a width of about 60 rods and a depth of about 40 feet, and of South Hadley Falls in the same sandstone for a mile, with somewhat greater width and depth, will each give a measure of the time that has elapsed since. ACTION OF ICEBERGS OR FLOES UPON THE CLAYS. Contorted clays. — At a railroad cutting just west of the South Street bridge in Northampton, already noted (p. 541) as showing sandstone and till planed down together into a drumlin, the clays rest normally on both, and a short distance eastward there begins a peculiar distorting, crump- ling, and comminuting of the latter. At its worst the clays are thoroughly chopped up into small pieces, which are mingled in entire confusion. This was exposed for a distance of about 33 feet, with a thickness of 2 feet. Eastward about 50 feet, across a space where the exposure was only sufficient to show that the clays were continuous and much disturbed, the}' 708 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. were again well exposed in a fresh cutting-, and here the clay is kneaded into fantastic shapes, squeezed into holes in the drift below, and a large mass of coarse, reddish drift has been dumped into it, and the two are in places well molded together. Twenty feet farther on the clays were per- fectly normal and horizontal. The intervening space was well exposed, and one could see how the clay disentangled itself from the mass of coarse material and gradually reassumed its horizontal lamination. Below and above the disturbed portion the clays are quite horizontal and undisturbed. This locality is at the base of a promontory in the ancient lake, around which the thread of the current bent as it swept southwestward over East- hampton; and the ice floes from the north, stranding here, have plowed up the clays and mingled them with the coarse material with which they were themselves loaded. In the curve by which the current bent around the projecting drumlins in Northampton several similar disturbed patches isolated in the otherwise horizontal clays h?,ve been exposed, as in the digging of the sewer at the south end of King street, where they were so distorted that they were mistaken for till by a good observer. They were described as being thor- oughly puddled. A mile farther northwest, at the great cut on the railroad to Williamsburg extending from the Bay State Brook east to the crossroads north of the railroad, the same thing is shown for many rods in the eastern portion of the cutting. The sands are irregularly disturbed, and at several places discoimected pockets of bowlders and glacial clay appear, wholly inclosed in the distorted sands. In the same area of disturbance a mile farther southwest, the fine exposures in the great clay pits south of the Insane Asylum are illustra- tions of the same action. A horizontal line is marked for many hundred feet in the vertical walls of the excavations at the same level with the plane of disturbance farther north. Above this line the clays are undisturbed and about 12 feet thick before they merge into sand; below they are kneaded into the most tortuous forms, and at times all trace of structure is gone. As in the block above the watch seen in the accompanying plate (XIX), traces of more than one passage of the ice are manifest, and in the largest of the blocks shown in the figure the extreme convolution of the plastic layers on the one side, and the faulting and incipient slaty cleavage on the other, are well shown. The convoluted layer in the block to the right is com- SECONDAltY STRUCTUEES IN THE CLAYS. 709 pressed to one-fifth of its former leng-th. Going- northwest, we soon come, at Sunset Hill, on the great drumlins that formed the shore of the ancient lake, and it is clear that the disturbance could not have been caused by ice coming down the valley of Mill River, which lies behind these. SECONDARY STRUCTURES IN THE CLAYS. Joints. — Where the clay stands in vertical walls in the river banks it is in time rudely . fissured parallel and at right angles to the exposed sur- face, and as the horizontal seams of sand weaken the cohesion of the mass in the third plane the river in the spring flood often moves off bodily great cubical masses of the clay and heaps them up lower down. Several years ago, on visiting the bank of the river below Hadley, I found a broad, horizontal surface thus exposed at about low-water level, which was jointed with extreme regularity and beauty. The principal lines ran parallel to the edge of the bank, perfectly straight and parallel to each other and an inch apart. The second set, also parallel to each other, were an inch and a half apart, and made an angle of 60° with the first set. The lines of the second set were not always continuous, sometimes failing between two contiguous lines of the first set, but continued beyond in the same direction. These lines represented fissures which extended tloi'ough one layer of the clay one-third of an inch thick, dividing the clay into blocks of mathematical regularity. Toward the edge the blocks had been moved by the current a short distance from their original position, manifestly very soon after the superincumbent block of clay had been lifted oii, for they were, when I examined them, so soft that they could not be touched without destroying their form, and yet as they lay they retained their perfect regulai'ity. One could not help thinking that in olden time it would have been called a fairies' pavement, as still in Scotland the claystones are called fairy stones. Below, where the large massses of clay had lodged, I found the small blocks piled in considerable abundance, but all softened and fused together, and in subsequent years I have always found them in abundance under the same circumstances. Later I found the same jointing in the large clay pit at the Bay State in Northampton, where surfaces 3 and 4 feet square were regularly jointed, exactly as on the river bank. When a vertical surface had been left for some time and the workmen then attacked that portion of 710 GEOLOGY OP OLD HAMPSHIRE COUNTY, MASS. the pit again, the large blocks of clay when dislodged would slip apart easily along the j^lanes of bedding, where the films of sand lessened the cohesion, and expose broad surfaces of the tessellated pavement. It was very plain that the greater ease with which the moisture could escape along these sand layers was the determining cause of the appearance of the structure along these planes. The moisture escaped so gradually and the clay was so nearly homogenous that the shrinkage tension could distri- bute itself equally throughout the mass and finally relieve itself by a system of fissures at angles of 60° and 120°, of great mathematical regularity. President Hitchcock mentions^ three localities, one on the Agawam and two near the Deerfield Eiver, where these joints also occurred in the same clays, and considers them to be due to a crystallization of the clay and "to be a more simple operation of the same general cause which produced the concretions." In an elaborate paper entitled "On the structure of rocks called joint- ing,"^ Prof W. King says: Hitchcock states that " unconsolidated clay beds in West Springfield and Deer- field, in Massachusetts, are intersected by numerous and distinct joints, while those above and below are unaffected. This clay has certainly never been subjected to any great degree of heat, being of very recent origin.'" It is to be apprehended that there is some oversight in this statement. This seems to be a wholly groundless assumption on the part of the author, made in support of the theory advanced in the paper cited. I may add that the fissures extend vertically downward through the fat laminae as if cut with a knife, and pass down through the sandy laminje with a curved surface. The torsion theory of Daubrc^e will hardly apply, as the joints are found in limited areas having relation to recent erosions, or in bluffs pro- duced by digging. I have searched the clays for many years for fossils and concretions, and these joints have been wanting in so great a number of cases where all the conditions were favorable that they can not well be referred to any such general cause. All the cases occurred in bluffs where the wall below was strong and well supported and there would seem to 1 Geology of Massachusetts, 1841, p. 418. ^Trans. Royal Irish Acad., Dublin, vol. 25, p. 606. ' Elementary Geology, p. 22. CONCEETIONS. 711 be small place for any iuflueuce of torsion. Yet from the removal of the clay to produce the bluff, and from the quite sudden drying of the surface i)f the bluff, there might be a slight creeping of the clays still below the level of the streams, or the wet floor of the clay pit might produce torsion, which would be influential in producing the forms observed. Concretions. — In the Journal Book of the Royal Society for 1734 is a manuscript catalogue of objects of natural history found in New England, by John Winthrop, magistrate of the Connecticut colony and great-grandson of the first governor of Massachusetts. It mentions "clay generated in the form of horse shoes from the bottom of Connecticut River." ^ It would be difficult to find a boy brought up near the Connecticut who had not in the early summer gathered claystones on the bank of the river which had been washed out of the clay in the spring floods, and wondered at their abundance, their smooth and apparently artificial surface, and their regular form — spherical, spheroidal, ellipsoidal, or flattened into disks, sometimes variously elongated, lobed, or grotesquely imitating animals and works of art. And he would be hardly satisfied with the common explanation that they were formed from hardened pieces of clay by the wearing of the water. This was Dr. Hitchcock's first opinion, and in 1823^ he gave a full mineralogical description of them. They thus very early attracted the attention of the geologist, and in 1835 President Hitchcock describes them with care, and asks the questions : "But are concretions the resiilt of crys- tallographic laws? If so, why are not crystals produced?" The tubular feiTuginous form he describes as a fossil of uncertain character.^ In the report of 1841 he devotes 16 pages and 5 plates to a discussion of concre- tions, and presents a classification of them according to form; and though he no longer looks upon any of them as fossils, he considers them exceed- ingly difficult of explanation and thinks one must assume them to be the result of the action of galvanic electricity, and associates with them, as a result of the same general causes, the prismatic blocks of clay produced by shrinkage joints which I have described on page 709. President Hitchcock returns to the subject in 1861, in the report on the Geology of Vermont, and devotes 8 pages and 3 plates to clay- 1 Am. Jour. Sci., 1st series, Vol. XLVII, p. 282. 2 Geology of the Connecticut River: Am. Jour. Sci., 1st series, Vol. VI, 1823, p. 229. 3Geol. Mass., p. 182. 712 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. stones, and still expects someone to develop the fundamental principles of their formation and "do for them what Abb^ Hauy did for crystallography." He quotes several pages from Prof. C. B. Adams's second report of the Geology of Vermont on the same subject, among other tilings: "It is obvious that the description and theory of concretions constitute a subject which, although j)erhaps less extensive than crystallography, is as properly entitled to rank as a distinct science." He quotes, also, from Professor Adams, a new classification of these forms, much more complex than his own, and containing several Greek words newly coined for the purpose. These elaborate classifications seem worthless, and remind one of Rafinesque's paper describing and naming nine new species of thunder and lightning, for all the variety in the forms depends solely upon causes wholly external to the concretion itself, namely, to the constantly varying permeability of the clay in its different parts and the decomposition of its constituents. In tracing the history of these forms one must notice, first, that the clay beds in which they occur differ materially from those beds of clay formed by the decomposition of massive feldspar in situ, which are often quite pure kaolin — a hydrated silicate of alumina. These Champlain clays, on the contrary, contain only a small portion of true kaolin, and are, in the main, an exceedingly fine, largely feldspathic sand, resembling somewhat the finest silt washed from a stamping mill ; they are, in fact, the finest portion of the material ground up by the glacier, and the waters which bore it southward may have been in part a veritable gletchermilch, issuing directly from beneath the ice. It may have been carried a long way southward in the valley, and thus have been derived by the ice partly from the Vermont rocks, among which limestone is prominent. It is certain, also, that the clays contain abundantly particles of min- erals, as lime feldspars, which, by their decomposition, afford calcic carbon- ate. And the waters with which the clays are saturated would, by virtue of the carbonic dioxid they contain, dissolve and carry in solution the carbonate derived from one or both these sources. The waters are con- stantly percolating, with a slow, capillary motion, through the clays, especially after the beds have been cut through here and there by streams and the edges of the laminae have been exposed, moving always from the moister toward the drier portions; and as the conditions in this respect often change, the direction of their motion woixld also change. CONORETIOiSrS. 713 Aji'ain, since, as I have described above, every layer is, as a rule, cai)ix'= ^ = hs: ^UoifrtS nci\ &o , s i ■« ■ \^\ ? n 70 1 {> i. 1 - > ra^\ ri 1 <«, 1 70 i t t> \ 1 ,r ft \ a Ii lls^ ' " 1 t^ i i \ \ 1 i 1 i — _j — ■ i 50 u ij « ,, fi 1 \ < I ? i if] ■< A 1 ! f _ 40 u >. r£V^7-M IM ' 1— i L \ L f" lA \%. i-f" ^1- == 1 rv [^ / K) to 1 r| i *H-r J^ f \ I, . ■^ /^^ \h r ^^\ p^< J- n D f\l\ / \ ] 10 B ^ei£ «2£ 4Mefc= j= p=: c= )^ j^ ^ 5 ,' ^ I? K C w*! A y3 ^' iL L^-TV -^ *^ jVv At/its 1 1 1 1 1 i i 1 1 i 1 1 yMll4S \ " ' z 3 s •z /3 « " • T / m »— 3- PROFILE OF CONNECTICUT Rl VEFf FROM HARTFORD. CONN..T0 VERNON, VERMONT » il « 1* U <7 « « » H THE TEKKACES OF THE CONNECTICUT. 723 A streuiu teiuls to iiuTuase its meanderiugs until friction on bank and bottom of its increased length uses up all the force derived from its descent during flood time. But when this happy equilibrium is reached the river goes beyond it and grows too long. It then, for relief, cuts off an oxbow in a sluggish stretch, as the Connecticut has often done between Sugar Loaf and Holyoke. This gives the section of the stream new life and eroding- power by as nuTch as it is shortened; and hence, since the great oxbow was cut off at the Northampton Meadow there has been more complaint of the loss of land by erosion across the Hadley and Northampton meadows than anywhere else on the river. The erosion has been especially severe at the upper and lower ends of Hadley street, and the location of the two bridges at Northampton has done much to direct and deflect the stream, especially promoting the erosion above those bridges on the east side and the growth of the islands on the west. At the extreme western apex of the great bend the stream has worn into a hill of coarse di-ift, out of which it has con- structed a natural riprap, which is restored as often as broken, and a period is put to the stream's wear in that direction, so that everything points to its cutting across parallel to Hadley street unless careful precautions are taken — more careful, it seems to me, than have been thought necessary. By the continued work of the agents here briefly mentioned, some of which are more fully discussed in the section on incomplete terraces (p. 731), the Connecticut has swung to and fro across the abandoned lake bottom as a cable swings through the water. The sands have melted away before it and filled in behind it, holding it to a constant width. In the Springfield Lake it has cut down very deeply into the lake sands, especially below Holyoke, forming many and complicated teiTaces. In the Hadley Lake it has lowered only very little since it began to flow as a river, forming few and broad terraces. At the Northampton bridge the track runs off the bridge at the west end and cuts the lake bottom, and from the east end one looks down on the lowest complete terrace, less than 20 feet below the level of the bridge. In the Montague Lake the downward erosion was arrested by the waters striking the Lily Pond sandstone reef, in Gill, and after they had rounded this reef they cut down rapidly to present level, forming an extra terrace not marked farther south. 724 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. THE INTERMEDIATE TERRACE AKD THE BARRIER AT THE lilLX POND IN GIEL, AN ABANDONED WATERFAXL. Mr. Warren Upham, in his Survey of the Terraces of the Connecticut River in New Hampshire/ described a "second apparently connected series of terraces which mark one of the principal flood plains formed by the river during- its work of erosion." It is "most clearly continuous below the south line of Brattleboro, but seems to be traceable from White River Falls." In the center of the State I was not able to trace any well-marked series corresponding with that described by Mr. Upham, but commencing at the north line it runs down the river, well marked and continuous, to the beginning of the canyon below Northfield Farms, and it had long been a problem to me why the terrace so broadly worn into the older sands in •the north was so faintly represented farther south. A study of the sandstone ridge at the Lily Pond^ quarry of Triassic "bird tracks" in the summer of 1882 made it clear to me that here had been the site of a waterfall of the Connecticut which had worn back two short canyons- about 100 feet long, in the northern and deepest of which the Lily Pond lies, and that the two had included a rocky island between them, just as is the case at present with Turners Falls, and on a larger scale with Niagara. This held up the waters to the level of the 300-foot terrace above this point. After an amount of erosion which must have represented a considerable lapse of time, the stream, wearing into the sands of the great delta on the south, cut round the edge of the ridge to the left and sunk suddenly to nearly its present level, abandoning (a) its course through the Lily Pond and Bartons Cove, and (b) the other branch starting from the other notch in the ridge and running parallel with the first, and, like it, still represented by a "cove" extending back some distance along the abandoned channel. The river took thus a more circuitous course through the " narrows," and had still to cut down somewhat to reach its present level, as the prolongation of the sandstone ridge appears just above the water level on the other side of the stream, coming out from under the thick delta sands. This is doubtless the reason why the width of the stream is so small at this point. ' Geology of New Hampshire, Hitchcock, Vol. Ill, p. 58. = This is the third Lily Pond mentioned in this chapter; one is in South Vernon, Vermont, the other in West Northfield, and this is just east of the Factory village, in the town of Gill. THE TERRACES OF THE CONNECTICUT. 725 The n'k'w o-ivuu in ]*1. XXII is taken from the edg-e of the high ter- race ii mile north of Willis Hill, in Montague, looking- north across the Con- necticut dm-ing- the spring flood. The stretch of the river between the "narrows" and the "horserace" is double the usual width, and it extends south covering the broad flats shown on the map. The broad notch (a b) in tlie sandstone ridge to the north, across the river, is the notch by whicli the waters formerly passed to fall deeply into the canyon concealed to the north. The small southward projection on the map, of the crescent-shaped pond, which is the Lily Pond, represents this canyon. The contours on the map are here incorrect, for the ground rises along the ridge to the east. The second notch (c) is opposite the next pond to the left; the place where the river turned the obstacle (d) and cut down to the point of the sand- stone ridge is the narrows on the map. THE LOW-LEVEL TERRACES AND FLOOD PLAIN" OF THE CONNECT- ICUT IN THE BASIN OF THE MONTAGUE LAKE. The subsidence of the waters of the Connecticut lakes to the present Connecticut River was very rapid, interrupted above the Lily Pond falls during their existence (see PL XXII), ' but completed perhaps still more suddenly here by the turning of the Lily Pond reef by the waters, as described on the preceding page. As a result, one goes down — through the whole length of the Mon- tague Lake, which was well filled up in the flood time, except in its southern portion — ^by a great scarp to the series of erosion terraces of the modern river, the highest of which rise but a few feet above the level of the flood plain. I have colored these on the map with diff"erent shades of yellow the darkest for the highest and oldest terrace, farthest from the river (t*), the lightest shade but one for the present flood plain (t^), and a very light yellow for well-marked but incomplete terraces below the completed flood plain (t*). Abandoned oxbows (o x) and old river courses (o b) now play an important part and are colored by lines of the same shade as the terrace coeval with them.- These later terraces form the "meadows" of the Connecticut. The Northfield Meadows and the romantic recess opposite, and the beautiful Pine Meadow above Northfield Farms, are the only ones of considerable extent carved in the northern lake, for from the latter place the river 726 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. any evidence of an intermediate water stand, the rapidity of the decHne of flows between rocky banks to the mouth of the Deerfield River and is bordered only by narrow terraces until it reaches the Hadley Lake basin at Sunderland. THE LATER TERRACES OR MEADOWS OF THE CONNECTICUT* IN THE HADLEY LAKE. The Sunderland, Hatfield, Hadley, and Northampton meadows, the most famous farming- lands and the earliest-settled portion of old Hampshire Fig. 47.— Sketch of the point«of the Northampton Meadow from Mount Holyoke, to show that the meadow is a composite of many islands. Except when seen just before sunset, the meadow seems an almost perfect plain. County, make up the area built by the Connecticut since its shrinkage to near its present dimensions, in its passage from Sugar Loaf to Mount Holyoke. The old lake bottom lay so low, especially in all the area north of Mount Warner, in Hatfield and North Hadley, that after one has followed down the slope from the high lake bench to and across this bottom to the scarp, a few feet in height, above the oldest of these later terraces (a scarp which registers the farthest outward swing of the river), and has failed to find THE TEKUAGES OF THE OONNECTICDT. 727 the water to essentially its present volume becomes ([uite manifest. The niea(U)\vs are broad prairies of the richest soil, the g-ift of the river, and seen from Mount Holyoke or Sugar Loaf when the crops are on, as they are farmed without fenciug, they spread in a carpet of wonderful Ijeauty at one's feet and take their j)lace in a landscape which owes much of its charm to the inunediate proximity of the prairie and the mountain. The plain that seems so perfectly level when seen from above proves on closer inspection to be made up of a series of broad, low ridges (fig. 47), like the long, low swell that comes in on the coast after a distant storm, and the curved grooves which separate these ridges run approximately parallel to the bank of the stream, but with greater or less curve. This is due to the composite nature of the terrace itself, as explained in a general way on page 722 and illustrated in its details in the discussion of incomplete terraces on page 731. Each of these low bars represents one of the ele- ments out of which the terrace is built, and has passed through the stages of bar, island, and " glacis terrace,"^ as it has a,dded itself to the previously formed plain, while the groove on the outside of each ridge (out from the river) is the unfilled remnant of the waterway which separated the island from the fonner shore. The surface of the broad terrace plain north of North Hadley and extending up to Sunderland shows this most strikingly, and when seen from the hill just north of Hatfield each separate island of whicb the ter- race was built by the westward swing of the river can be picked out. THE STRUCTURE OF THE TERRACES. The river sands. — The two scarps which form the riverward limit of the old lake bottom and the outer boundary of the terrace system on either side of the river, and represent the outermost limits of the oscillations of the stream, afford the best natural sections of the lake-bottom beds and com- monly expose at least the upper portion of the clays and their junction with the sands above, a junction very often marked by a line of springs. Between these scarps the river-bottom sands rest in the trough cut in the clays by the river, and the stream rarely nins directly on the subjacent clays. These sands are of medium grain, well washed, straticulate, with southward dip, and often, in addition, cross bedded with sharp southward or more moderate ' Hitchcock, Surface Geology, 1860, p. 5. 728 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. northward, eastward, or westward dips, according to their position upon the bar of which they form part. The much sands. — In 1838 President Hitchcock wrote :^ Luther Eoot, in digging a well in Sunderland, 80 rods from Connecticut Eiver, at bottom cut through a thick stratum of quicksand smelling of sulphuretted hydrogen. This sand proved to be very fertile. The same happened from a well in South Deerfield, on land of Mr. Eufus Eice. The bed was 6 feet from the surface. On searching, the bed was found on the bank of the river in Sunderland. It is the first stratum that retains water. President Hitchcock traced it through the Connecticut, Deerfield, and Westfield river valleys. It is, when wet, slightly green and soapy, but is a fine sand. It contains non oxide and vegetable fiber, and many analyses are given showing " soluble and insoluble geine" (as the substances that may be exti'acted from vegetable mold were then called), sulphate and phos- phate of iron, and silica. In his final report^ he returns to the subject at great length, compares the fertilizing part to the slime deposit of the river and expects much from its ixse upon lands. He calls it " muck sand," but notes that it is commonly called quicksand. Where I have been able to study this it has proved to be the finer deposit thrown down in the channels between islands and the shore to which they were in process of joining themselves, which channels are generally silted up at the upstream ends first and remain then long filled with stag- nant water. They are called "intervals" on many New England rivers. Peat deposits, plant remains. — In his first article on the geology of the valley^ President Hitchcock writes: In the meadows, logs, leaves, butternuts, and walnuts are found undecayed 15 feet below the surface, and stumps of trees have been observed at that depth stand- ing yet firmly where they once grew. In the same meadows a few years since several toads were dug up from 15 feet below the surface, and 3 feet in gravel, which soon recovered from a torpid state and hopped away. From the plain east of the south end of Sunderland street, beneath 7 feet of sand, hemlock logs with bark and leaves, beech nuts, and pine burs, have been very frequently dug up, as reported to me by Dr. Trow, of that town. These remains occur sparingly in the river sands everywhere as water-logged fragments, and more abundantly in old stream beds and in the 'Economic Geology of Massachusetts, p. 93. 2 Geology of Massachusetts, 1841, p. 107. 3 Geology of Deerfield : Am. Jour. Soi., 1st series, Vol. 1, 1819, p. 108 ; also Final Report, 1841, p. 366. THE TERRACES OF THE CONNECTICUT. 729 sheltered o-rooves descl'ibed above in connection witli the "muck sand.'' In diyfi-ing- wells in tlic loner i)art of Northampton along Maple street, on the north side of Mill River, and near the road leading- to Hockanuni Ferry from I'leasant street, the deposit has been found IG to 20 feet below the river — a line, bluish loam, with leaves, branches, and roots, butternuts, but- touballs, hendock knots, and a piece of coal. The same deposit was exposed at the foot of King- street in Northampton. Loess. — The most important stratum which goes to make up the ter- races is the wholly unstratified loess which everywhere caps thetn. It is most impoi-tant economically as giving the meadows their fertility, and deserves attention as a true water-formed river loess. Except for the lack of any large per cent of calcic carbonate, which, as there is almost no lime- stone in the drainage area of the Connecticut, is not surprising, and for the resultant rarity of land shells in the bed, its agreement with the Rhine loess is complete. It caps the river sands, and up and down the river presents a cornice, often 8 feet thick, of a fine, dark, wholly unstratified loam, ])ierced full of vertical root holes and breaking with vertical walls. It is the accumulated silt of the annual floods of the river, each layer being worked over by wind and frost and by the boring of worms and roots until the whole becomes entirely massive; and a rudely columnar structure is produced by the multitude of root holes, which become passages for water after the rotting of the roots, and so lessen the cohesion in this direc- tion that a vertical cleavage results. This loess layer appears capping the surface in the section (fig. 48, p. 737). It is finely shown in the curving bank above Northampton bridge, where the river is wearing with great rapidity into the Hadley Meadow and is forming already a great semicircle. Here the loess forms a perpendicular wall below which the sand slope is cut into great steps by the river as it sinks from high water, so that the whole resembles a Roman circus. The loess is here 5 feet thick. Over the Hatfield lower meadow it is 6 feet thick; over the upper meadow about 2 J feet. Over the Northfield Meadows the loess is 6 to 8 feet thick, and is especially strongly developed in the West Northfield Meadows. THE TERRACES OF THE CONNECTICUT IN THE SPRINGFIELD BASIN. The fact that the basin was left by the ice so nearly filled up to the level of the later lake, and the fact that the contributions to the lake were almost wholly from the east side, caused the thread of the cui-rent through 730 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. the lake to hug the western shore; and as the river took the place of the lake, it occupied the same position and cut very soon down into till or sand- stone, and so was unable to swing in broad oscillations, as in the deeper clay-filled Hadley basin. From the notch to Smiths Ferry a narrow terrace, or, for a distance, two narrow terraces, border the river on the west. On the east the river is wearing into the great gorge terrace of Dry Brook Hill, and a single sand slope of 1 88 feet touches the water's edge. From the south end of this hill to Holyoke the first position of the river was much farther east, and it has swung west to its present place and built on the east side an early flood plain, long since abandoned, and the river has now cut its bed deep in the sandstones and is thus prevented from oscillating. Doubtless if the dam below were removed the water woitld run in rapids over this ground, as it does over the rocks above Turners Falls. There is in all this distance scarcely a trace of any low terrace on the west side of the river. Across Chicopee there is a fine, low terrace bounded on the east by a high scarp of the high terrace, which everywhere shows till in great force beneath the sands of the old lake. From the Chicopee River south to the south line of the town the high terrace scarp comes forward to the river. Across Springfield there is developed a complicated series of river terraces. An incomplete terrace borders the stream opposite and above Hampden Park. The business portion of the city is built upon the normal flood plain of the river. Above this are two well-marked terraces, which send back deep lobes to the north and south of the armory grounds, up old water courses, and a remnant of one of these intermediate terraces is preserved in the hill north of the Memorial Church, cut off possibly by an oxbow, the only one found in this basin upon the main stream. The low terrace contracts to nothing on the south line of the town and widens again in Longmeadow. On the west side of the river the low terraces expand south of Holyoke into the broader meadows of West Springfield and Agawam. The sei'ies in all this distance is quite complicated, matching the oppo- site side of the river. There is across West Springfield an early flood plain raised well above the river, and around the entrance of the Westfield River the incomplete terrace occupies broad areas from which the water is largely kept out by artificial embankments. South of this tributary, across Agawam, the system of later terraces is developed with a beauty not exceeded in the THE TERRACES OF THE CONNECTICUT. 731 whole lengtli of tlie State. We have, beginning back at tlie mountain, the broad stretch of the higli lake flats (1 f), sinking into a more limited area of lake bottom (1 b t), and cut into this is a series of later terraces, four in nund-)er (t't^), much broader than the corresponding ones north of the Westliehl River, and combining with the terraces of this latter stream to form a most beautiful succession of broad meadows, bounded back from the river by sharp slopes, which swing in great (turves — representing former curves of the stream — up which one mounts to reach higher terrace flats as well characterized as those below. THE INCOMPLETE TERRACES AS IliLUSTRATIONS OF THE STAGES IIST THE GROWTH OF TERRACES. All up and down the river broad sand flats may be seen extending out into the stream at a level but little above low water and on the concave side of bends, as north of the knee of the Hadley bend and at the first concavity below the Northampton bridge. Generally only one bank of the river is wooded at a given section of the stream. Going up or down stream, one comes to a stretch where the growth ceases and is replaced by a caving bank, beyond which the bushes begin again. From the bushy banks the shallows extend far out, and the conditions are favorable for the formation of islands. Against the caving bank lies the thread of the stream. Each set of these sand flats and shallows is connected diagonally across the stream with a corresponding set on the other side, and at low water a series of disconnected deep water-pockets lies in the line of the thread of the current, alternating against the right and left banks of the stream, and so much of the water seeps through the sands of the shallows between the pockets that the bed is not scoured out at all between these long, curved deep-water stretches. It was a remarkable and interesting discovery of Gen. Theo. G. Ellis, of Hartford,^ that at high water a large portion or the whole of this system, of bars is scoured out, and on the recession of the flood is replaced exactly in its old place and with its old dimensions, as a curtain held up by the wind sinks to its old place as the wind falls. This is true, of course, in so far as the banks of the stream above and below are unchanged, for these bars are the mechanical solution of a complex equation in hydraulics and change with any change of the factors. 'Survey of Connecticut River: Ex. Doc. 101, Forty-fifth Congress, second session. 732 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. If bushes or turf strand on the bar and take root, it is protected and its increase is accelerated, and it grows in flood time above the low-water level and i-ises as an island or promontory, and the tendency of the stream to scour out everything at high water generally keeps open a channel between it and the mainland. This is the condition of Ellwells Island, just north of the west end of Noi-thampton bridge. In the time of canal navigation, sixty years -ago, the channel, 33 feet deep, ran right under the present island, and in digging for the pier of the new bridge old boat hooks were fovnid under its south end. The island generally joins the mainland by the silting up of the upstream end of the side channel, and a deep, stagnant inlet runs up from its south end. This is the condition of two broad peninsulas opposite the "oxbow" below Hockanum which have formed since 1840. The continued growth of the new addition to the flood plain takes place by material brought in over it during floods, and this decreases in rapidity as the ground rises, and soon the checking of the current as it rises over the flat makes itself manifest in the increased deposition along the outer border of the flat, and a "glacis terrace"^ results, sloping sharply to the water and gradually backward. The "glacis terrace" is thus. a case of arrested development of a terrace. The groove which separates the new from the old remains preserved for a long time and often permanently. Again, as the waters rise over the growing terrace, they are arrested first over its upstream portion and thus build up this end most rapidly. This is most beautifully illustrated in the terrace which begins at the North- ampton bridge and extends south to the south end of Hadley street, and is bounded by the road which leaves the main road at the bridge and joins Hadley street at its south end. This road runs along the edge of the former bank of the river, and at its south end one looks down upon the lower plain, still separated by a shallow inlet which runs up from the south. North- ward, the lower plain gradually rises, the inlet shallows and disappears, and the lower terrace is a complete "glacis terrace." Still farther north the lower plain continues to rise, and the scarp which separates the two becomes less in height until at the bridge the two have come so nearly to the same level that one might easily overlook the fact that the newer terrace extends 1 Hitchcock, Surface Geology, 1860, p. 5. THE OSCILLATIONS OF THE CONNECTICUT. 733 for some little distance alxnc the hvidg'e. I have already had occasion to describf the meadows as tunned by a continued repetition of this process. (See ii- 47. ]). T2(;.) ON THK OSCILIjATIONS OF THE COK]ST:CTICUr FROM ITS EAKLIEST POSITION. From the north line of the State to the Sunderland bridge the river everywhere cut down rapidly to rock and has not swung widely to east and west, but has been condemned from the beginning to rock cutting. The river at the beginning- took its course across the Hadley Lake bottom along the deepest line, which it has obliterated. I imagine that this line was very near its present position. It probably swung first eastwardly to its eastern limit, at the "halfway house" on the Hadley road. It is more certain that from this eastern limit of its oscillation it has moved west regularly and silted up its bed behind quite rapidly and completely; this is shown by the fact that the Hadley and North Hadley-Sunderland meadows are composed of series of elongate and coalescing islands, as detailed on ]3age 726. It has swung, then, west across the Hatfield and very far west across the Northampton meadows, and regained again a more central position by cutting off its oxbows. During its swing westward, across the Noi'thampton Meadow and back, it has lowered itself by about 7 feet more than its own depth, since at the foot of Hadley street its old bottom sands rest upon the eroded surface of the Champlain clays at a height of 7 feet above the low water of the river. This height may be some- what increased if we allow for the influence of the Holyoke dam. From the Holyoke notch south to the Holyoke dam the river early became entangled in rock and has cut only vertically. From the dam south the earliest position, or, more accurately, the earliest restorable position, may be found by following down the outside edge (counting from the river) of the oldest terrace of erosion (t^). This, the highest terrace of this later series, is found only on the west side of the stream from Holyoke south, and then is for a long way present on the east side, across Springfield, then being transfeiTed to the west side, across Agawam. This represents the sinuous position of the stream from the Holyoke Falls southward at a time when it had first established its course across the lake bottom, and from which it has swung to form its later and lower terraces, ending in its present temporary position. 734 GEOLOGY OF OLD HAMPSHIEB COUNTY, MASS. THE OXBOWS OF THE CONNECTICUT. lu the Montague Lake the valley was too narrow, the rock comes too near the surface, and the earlier deposits were too thick to allow of broad bends and cut-offs. Several old river beds there seem rather to have been formed by the building up of an island in midstream and the after limita- tion of the current to one side of it without filling up the abandoned portion. Over the broad bottom of the Hadley Lake the stream had more free- dom, and in the Hatfield and Northampton meadows are two most interest- ing series, containing in one case four and in the other thi'ee old cut-off oxbows, and between is the great Hadley bend, where the river runs about 6 miles to advance southward 1 mile, and threatens to take a straight course down through Hadley street. (See map, PI. XXXV, in pocket.) In Hatfield the oldest oxbow runs down west of the village. A part of the unfilled bed of the second is the Great Pond. The third is repre- sented by a sickle-shaped pond east of the road going north from the village, and the completion of the fourth has in very recent years transferred a fragment of Hadley to the west side of the river. In Northampton a sickle-shaped pond, at the western edge of the meadow, represents the oldest cut-off. The second remains in a smaller pond near the western curve of the third — the oxbow par excellence — which is still a ring-shaped pond, in communication with the main stream beneath the bridge of the Connecticut River Railroad. This was cut off during the flood of 1840. Figures of the river, as seen from Mount Hol- yoke before 1840, with the fine curve of the stream from 1840 to 1845, after the cut-off and before the silting up of the mouths of the oxbow, are given in the publication, Northampton, Meadow and City.^ ON THE DEFLECTION OF STREAMS TO THE EIGHT BANK. The Connecticut River between Mount Toby and Mount Holyoke, about 8 miles in a straight line, flows across the broad, level bottom of the ancient lake through thick, very fine-grained, and very homogenous deposits. It is thus, together with its tributaries, favorably situated to give evidence concerning the possible influence of the earth's rotation upon the erosion of streams according to Ferrell's law, that a stream under the influence of the 1 F. N. Kneeland, Northampton, 1894, p. 36. TEUKACES AROUND A WATERFALL. 735 earth's rotatidu always tends to weai- its riylit bank. Accordingly, it is interesting" that the i-iver has constantly made and cut off oxbows on the west — that is, the right side — and never on the east side. It has successively cut off four bends in Hatfield and three in the south part of Northampton, ami has also made the great Hadley bend, which it has long threatened to change, into an oxbow, and it has never made great bends out to the eastward. The same testimony also comes in a striking way from the tributaries I have for several years given, as practical work for advanced students, the mapping of portions of these tributaries of the Connecticut, which run for long distances ovit over the old lake bottom, and on counting up the sharp bends and oxbows on the right-hand side of the stream the proportion was as great as 30 to 1 in favor of this side as against the opposita RIVER TERRACES AROUND A RECEDING WATERFALL. The flood plain of a river tends to reach the full height of highest flood, and on approaching a fall this height diminishes greatly, as the waters as they go over the fall, because of their increased rapidity, rise to only a small fraction of their normal height. At the foot of the fall or at the mouth of the canyon below the fall the flood plain begins again at a level as much below that above the fall as the descent of the waters demands. Thus at Turners Falls the flood plain above "the falls is only 7 feet above the level of the waters, and the height above the waters before the erection of the dam was probably not many times greater, while the flood height of the river here is 30 feet. If now the falls recede, leaving remnants of this low flood plain, it will hang over the canyon with a height above the river equal to its original height plus the height of the fall; and this is the case at Turners Falls and at Holyoke, where the old flood plain is con- tinuous from above the falls south along the sides of the gorge formed by the recession of the falls. If, further, the stream by its oscillations below the falls builds a flood plain at the lower and newer level, we have the curious result that the flood plain above the falls will extend downstream above the flood plain below the falls, the two thus overlapping at two different levels. Distinct traces of this appear at both of the falls on the river, especially at South Hadley Falls, where the flood plain of the river is continued out over the lower one for a long distance. This makes a difficulty in coloring 736 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. the terraces, and I have on the map continued the low-terrace color down the canyon side in dots to distinguish it. THE TERRACES OF TRIBUTARIES. The deep, land-locked basins cut in the lake beds by the tribiitaries of the Connecticut are among- the most remarkable orographic features of the valley. Each stream has first built out the great delta plain and then excavated its basin on the sinking of the main stream. They have this peculiarity in common. Each stream emerges from its gorge in the crys- talline rocks, rims across its former delta, and passes through a short rocky gorge just above its mouth, and the stream has thus been fastened at two points like the string of a musical instrument, and has vibrated between these two points to form its closed basin. It has eroded with great violence because it has had the rapid fall across the crystalline rocks. The Deerfield and Westfield River basins are the most extensive and interesting examples, the one occupied by the most romantic and sleepy old town in the valley, the other by a typical, unattractive, manufacturing town. These rivers have reoccupied their old gorges in the trap ridges, as already explained (p. 512). The others, Green River, Millers River, Mill River in Northampton, and Cushmans Brook at the Golden Gate in Amherst, have by chance strack rock bottom as they cut down thi-ough their deltas, and thus the mouths of their basins are closed below, as are the first two. The basins are bounded on all sides by high scarps, and over the low meadow bottoms are many abandoned channels caused by ice obstniction in spring, which in the Deerfield bottom are developed into a most compli- cated network. A prominent, flat-topped hill, called Pine Hill or Pine Nook, its surface on a level with the adjacent high terrace, rises in the midst of the Deerfield Meadow and has doubtless been cut off by an old oxbow of the river, and a smaller but similar one, which has been called an Indian mound, but has the structure of the smTounding delta sands, stands in the basin of Mill River above Florence. TUE TEKUACES OF THE CONNECTICUT. 737 AX OLD <)XIJO\V OF FORT lUVEU. From the south end of Hadley street one may follow the Champlain clays continuously for a long distance south in the river bank. Near where they sink below the water a terrace scarp belonging to Fort River is cut off in the bank of the Connecticut. The last house passed in going south from Hadley and before crossing the bridge over Fort River stands on the edo-e of the completed flood plain of the Connecticut and looks down over this scarp to a lower plain, formerly part of the flood plain of Fort River, which here runs parallel to and just east of the Connecticut. In fig. 48, a represents the southern termination of the Champlain clays, which a few feet north furnished the leaves described on page 718, and still farther north abound in clay stones; 5, the bottom sands of the Connecticut when it flowed at a level higher than at present by an amount somewhat greater CONN ft.FLQOO ^LMN -^-^^..^^ \'-.''\ - - : ''.^.-^^^ '''''^^'^■^^^^■-J.^.^ZjT^- .■ ZT^^ ~ IZ7{IzIZT^J^^'^~^''^^^^^^^-'^^^ ccNN, /f s^NOs b. '^^ ^^^^p ^l ' S— ri — '-^; " ^'^_ .,..._ ^._ '^Tl— —^ '• '^^' '^^^' ^ -^ :r-,-^r:-^-^-^=^^^-^^^~ •^\'rr ^\ ' -r m~ mY \ u u ^ mt^mi£w\im a^ " ■ ' ■ i'- 'i iV ' Y i 'i f i ' rti ' i ' ' ^i Ji 'T B~"' ^ four ft, SAMOS than its own depth. These are coarse to medium grained straticulate sands, which rest unconformably upon the clay and extend with a thickness of 20 feet to the point where the old Fort River terrace scarp is cut off in the present river bank. Here these sands end, their horizontal beds abut- ting unconformably against c and e, except that at lowest water their lower beds can be traced beneath c for the whole length of the exposure. The scarp, partly exposed and partly submerged, against which these sands end, registers the farthest northward swing of Fort River in throwing out an oxbow here on its west side; c, which is a fine, horizontally bedded and straticulate sand, is the bottom sand of Fort River as it swung across its flood plain; cl and cV are two cross-sections of the old oxbow of the trib- utary, now cut into by the main stream. At d the stream plainly flowed toward the west — that is, toward the reader; at (i', toward the east; and the Connecticut has cut across this old oxbow, as indicated by the dotted lines. These old river beds are the exact equivalents of the present bed of Fort River — a stratified deposit of leaves, twigs, logs, and. seeds in fine MON xxix 47 738 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. whitish clay, abundantly pierced by the vertical pipestem concretions, espe- cially in its upper portion, where it grades into e, a thick stratum of loess, which is 10 feet thick over all the lower plain, and still thicker where it projects downward to fill the old river beds. It rises up the terrace scarp with a thickness of 3 feet, and is continuous over the upper plain with a thickness of 6 feet. This represents the accumulated deposits of the Con- necticut in flood time, laid down since Fort River abandoned its bed at d and d! . This stream now runs immediately adjacent, with its surface coin- cident with that of the Connecticut. When it occupied this old oxbow it flowed at a level 13 feet higher, and this represents certainly more than half of the amount by which the Connecticut has lowered its bed in the bottom of Hadley Lake since it shrunk to its present size. This would assign to the fossils found here an age about intermediate between those of the Champlain clays below and the present time, or somewhat nearer to the present flora than to the older; and the habits of the fossils themselves agree with this, and indicate a climate like that of northern Vermont or Canada. It is interesting that a fragment of charcoal from some light, open- grained wood was found in the midst of the matted leaves of the leaf bed and was certainly of the same age with them. It was about as large as a walnut. FOSSIIiS OF THE TERRACE PERIOD. VERTEBRATES. Mastodon americanus. — In 1872 Dr. Edward Hitchcock, jr., writes: "I have seen and identified a mastodon's molar which was found in the town of Coleraine, Massachusetts. It was shoveled out of a muck bed on the farm of Elias Bardwell." ' The tooth is still in Mr. Bardwell's possession. MOLLTJSKS. In digging in a marl pit which has formed by the filling of a small pond on the surface of the till on the farm of Fred Conant, at East Shel- burne, large quantities of white fresh-water shells are at times thrown out. They are very well preserved, and consist of the following species: Lymnea elodes Say. — Length, 30 """. Common. Planorbis trivolvis Say. — Large diameter, 25™™; small diameter, 18 Common. ' Am. Joiir. Sci., 3cl series, Vol. Ill, p. 146. mm FOSSILS OF THE TEERAOE TEEIOD. 739 Plaiiorhis jMirvus Saj-. — Abundant. Diameter, 6.5 '". Flsidmm variahilc Fi'ime. — Abundant. Length, 2"^"'; width, 2i'"'^. PLANTS. Banuncuhis aquatilis L. — A single well-preserved plant. This and the following, with one exception, are from the old oxbow in Hadley, described on page 737. Acer saccliarimmi Wang. — Leaves. Eare. Primus virginiana L. — Seeds very abundant; leaves abundant. "River banks. Common, especially northward" (Gray). Platamis occidentalis L. — Leaves, large branches, and balls found in great abundance. Matted masses several inches thick and many feet broad consist almost entirely of leaves, many of the largest size. Large branches, often very much flattened and still covered with the characteristic bark, occur fre- quently. In several cases delicate hollow globes of sand, like globes of lace or Chinese hollow ivory balls, have been formed by the penetration of the fine sand to the surface of the central ball, and its spreading in the regular interstices which surroimd each point of attachment of a seed, where the grains have been slightly agglutinated and left as a globe of lace on the rotting of the seed ball. The extreme northern range of the species is Lake Champlain and Montreal.-^ Jiiglans cinerea L. — Dwarf nuts, 1^ to If inches long, f to ^ inch wide; less deeply sculptured than the form now common here. In one case nine specially prominent ridges are present. In another the ridges are more rounded, broad, and irregular than now. The species now extend south to Georgia, and north through Canada, but this dwarf form would seem to indicate 'a station near its south border. Garya cmiara, Nutt. — At the old oxbow occurred an impression of an exterior inclosing a cast of the interior of a single specimen in rusty clay. Also well-preserved nuts in abundance were given me by Dr. Edward Hitchcock, as found at extreme low water below the mouth of Fort River, opposite the fourth pile of Mclndoes's boom, counting from the north — rounded, thin-shelled nuts, averaging somewhat larger than nuts of the same ' Michaux, Sylva, vol. 6, p. 56. 740 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. species from Burlington, Vermont. Sizes: Burlington, 17.5 to 19°""; Con- necticut River, smallest, 19 to 20.5°""; largest, 22 to 27"'". "Barely 1 inch (25.5"") long, thin walled" (Grray). "Northern boundary, Vermont" (Michaux). I am informed that but a single tree of the species is now- found in the county, and this upon the Hatfield Meadow. Quercus alba L. — A few well-preserved leaves. Querais coccinea Wang., var. ambigua. — Leaves and abundant acorns in all stages of growth. "The gray oak appears, by my father's notes, to be found farther north than any other species in America." (Michaux, Sylva, vol. 1, p. 98.) "Along the northern borders to Lake Champlain and northward." (Gray, Manual, 1872, p. 434.) Fagus ferruginea Ait. — Next to the sycamore, the most abundant plant represented. Leaves of full size, large fragments of the wood and bark and nuts alike abundant, the latter very large and surpassing in size those now found in the vicinity. "Common, especially northward." (Gray's Manual, p. 455.) "Almost exclusively confined to the northeastern United States and to the provinces of Canada." (Michaux, S3dva, vol. 5, p. 22.) Bet'ula alba L. — Large branches with bark marked exactly as in the common white birch. Besides these many other indeterminate plants were studied — willow leaves, grape vines, grasses, liliacese, lycopodium, hchens, various seeds, and even a flower. THE PLEISTOCENE BEETLES OF FORT RIYEB, MASSACHUSETTS. By Samuel H. Scuddbr. The insects found by Prof. B. K. Emerson in the old bed of Fort River in Hadley, Massachusetts, near its entrance into the Connecticut, have no special interest beyond the fact that they are the first insects found in such deposits in New England. They consist wholly of Coleop- tera, and represent five species and four families, viz: Carabidae, Dytiscidse, Elateridee, and Chrysomelidae, the latter having two species. At least three of the insects, perhaps all, belong to species not now known to exist, but so far as can be told with any certainty, all belong to existing genera, though some doubt may reasonably be claimed for the single species of PLATE XXIII. 741 PLATE XXIII. PLEISTOCENE BEETLES OF FORT RIVER, MASSACHUSETTS. (The original drawings are by J. Henry Blake.) Fig, 1. Cymindis extorpescens ; elytron f. 2. Corymbites asthiops (Herbst)f ; prothorax ^. 3. Dytiscidfe sp., perhaps a Matus; metasternum f. 4. The same; a portion further enlarged to show the surface sculpture ^. 5. Donacia elougatula; elytron f. 6. Saxinis regularis ; portion of the right elytron highly magnified to show the surface sculpture ^. 7. The same; dorsal view of the beetle f. 742 U. a. OCOLOOtCAi. 8URVEV MONOGRAPH XXIX PL. XXIII 11 m ^ PLEISTOCENE BEETLES OF F. . RIVER, MASSACHUSETTS. TLEISTOCENE 15EETLES OP FORT RIVEE. 743 Dvtiscida> and one of the two species of Chrysomelida;. This is rather surprising-, but is what has been found to some degree in American Pleisto- cene dejiosits, the insects of which appear to show less close relations to their successors on the spot than is commonly the case in Europe, and in consequence relatively little light can be shed upon the climatic conditions of the time by their remains. In the present case the information is meager and gives no certain clue. The existing species most nearly allied to the Pleistocene Cymindis (Carabidse) occurs from Massachusetts to Florida, and is more common in the South than in the North; our single species of Matus (Dytiscidse) is found in Canada and in the Northern States from Massachusetts to Iowa, but also in Missouri and Florida;^ Corymbites (Bthiops (Elateridae) occurs from Massachusetts to Pennsylvania and Ohio; the Donacia (Chrysomelidse) most nearly allied to the fossil species described below appears to be one known from the Pleistocene of Italy; while the species of Saxinis (Cluysomelidse) most closely related to the Pleistocene form here figured is a northwestern species, coming from Vancouver, Oregon, and California, and also from Utah, Colorado, and Wyoming. It is plain, then, that a considerably larger assemblage of forms must be obtained to give any evidence of value. The following are the species found: Family CARABIDtE. Cymindis extorpescens. PI. XXIII, fig. 1. A single elytron, representing a species aboiit as large as C. cribricoUis Dej., but more nearly allied to C. elegans Lee. in the reduction of the inter- stitial punctures to a single row, seems to be entirely distinct from any of our species of that genus in the reduction of the striae to a series of short longitudinal dashes separated from one another by their own length, while the interstitial punctures are more lightly impressed, arranged in single straight rows, and separated by twice their own length. The elytron is piceous, with a very faint bluish reflection. Length of elytron, 6.35°""; width, 2.5°"°. 'For information on the distributiou of American Coleoptera I always rely upon the ready and efficient aid of my friend, Mr. Samuel Henshaw, of Cambridge. 744 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. Family DYTISCID^. DytiscidcR sp. PI. XXIII, figs. 3, 4. The metastei'num of a species of Dytiscidae is among the fragments obtained. It is nearly complete, and, of all with which I have been able to compare it, most resembles that of Matus, particularly on account of the depth of the median groove, the form and relative abbreviation of the rounded intercoxal prolongations, and the shape of the coxaj, and it agrees very well in size with our single species, M. licarinatus. As, however, it is not nearly so long in proportion to its breadth as in that species, and nar- rows remarkably from behind forward, it hardly seems possible to refer it to that genus, and I find no other with which I can so well compare it. Instead, also, of being faintly and rather sparsely punctulate, as in M. bicarinatus, the surface is feebly, longitudinally, and undulately striate, and of a dull piceous color. In the general form of the metasternum it more nearly resembles an Agabus. Length of metasternum, 3"™; breadth posteriorly, 5""°. Family ELATERID^. Corymhites athiops (Herbst)l PI. XXIII, fig. 2. The prothorax of an elaterid of a piceous color is referred here with some doubt. There are but two or three of our species which have a pro- thorax large enough to compare with it, but the size and general propor- tions, and especially the punctation of the surface, agree perfectly with C. mtliiops. It differs from that, however, in the greater slenderness of the produced posterior outer angles, the sides are more strongly convex on the posterior half, and it is not narrowed to nearly the same extent anteriorly. This last point makes its reference here very doubtful, but until further remains are found it seems best to place it here with a mark of doubt. Length of prothorax along the median line, 5.35°"°; greatest breadth, 5.25"". PLEISTOCENE BEETLES OF EORT IIIVER. 745 Family CHRYSOMELID^. Tribe Donaciini. JDonacia elongatula. PI. XXIII, %. 5. A siuf'-le nearly perfect left elytron appears to represent a species not hitherto known, but apparently most nearly allied to I), lignitum Sord., from the Italian Pleistocene. It is somewhat more than three times as long as broad, tapering from the middle to the nontruncated apex, before which the outer margin is more strongly but very regularly curved, with no sudden change of direction. Besides the marginal groove, there are in the basal half ten parallel strise with delicate longitudinal punctures, but in passing from the base to the apex the two middle unite just before the middle to form a single stria, and just beyond the middle they are joined by the fourth from the inner margin. No others unite until shortly before the apex, when the third and fourth from the inner margin unite and terminate, and halfway from here to the apex all but the outer ones approach and termi- nate, the outer ones acting similarly at the very apex. The surface is shining piceous. Length of the fragment, 7.25'°'^; probable length of elytron, T.e"""; breadth in middle, 2.2°"^. Tribe Clythrini. Saxinis regularis. PI. XXIII, figs. 6, 7. The most complete specimen found in these beds is a chrysomelid, with the last abdominal segment exposed and callous, which with its form indicates one of the Clythrini, It is slightly larger than and of a similar form with 8. saucia Lee, though it differs decidedly from it in the details of the form and structure of the elytra. The prothorax is crushed and mis- shapen, so that nothing more can be said of it than that it differs from that of Saxinis in its lesser breadth, being decidedly narrower at base than the elytra, and on this account it is exceedingly doubtful if it should be placed 746 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. at all in that geniis, or, in fact, in any other of the American Clythrini. The elytra differ from those of S. saucia mainly in having the abbreviated apex less rounded and more squarely truncate, the inner a^jical angle espe- cially being far more angulate; as there, both outer and sutural borders are delicately margined, and the general proportions of the elytra are much the same (in this respect resembling it rather than our other species of Saxinis), but they are a little and gradually narrowed on the apical half, while nearly equal in S. saucia; besides the punctures which adjoin the outer margined border, the elytra have ten very straight and regular equidistant series of delicate punctures, which are short oval, those in each row sej)arated from their neighbors by more than, usually about twice, their own length, and the general surface is sparsely covered with excessively delicate hairs scarcely longer than the punctures. The general color is a uniform shining piceous with a slight greenish, metallic tinge, the metallic green being decided in the punctures. The last abdominal segment shows a slight dull median ridge. Lengthof body, 8™""; of elytra, 6.5 ™™ ; breadth of base of prothorax, 3.1 '"'"; of each elytron, 2.5 ="". THE REPULSIOlSr OF TEIBTJTARIES.i Oscar Peschel,^ from his orographic studies, notes the tendency of a tributary to run a long distance near and nearly parallel to its primary. In all the tributaries which enter the Connecticut across the broad lake deposits between Mount Toby and Mount Holyoke this is very marked. They all run out through the old bordering bench (1 s h) in deep gorges, then take a straight course down over the old lake bottom (1 b t), following its slope, but when they reach the oldest terrace flat formed by the river in its oscillations after the shrinking of the lake, they bend abruptly south and continue as far as possible to run nearly parallel to the main stream, and when they enter the latter it is by a sudden bend at right angles. This will be clearly seen by an inspection of the map (PI. XXXV, in pocket), or of the North- ampton and Belchertown sheets. First, the brooks north of Sunderland village, on the east side of the ' See PI. XXXV, in pocket at end of volume. s Vergleichende Erdkunde, 1878, p. 141. THE REPULSION OP TRIBUTAEIES. 747 river, do not show the pecuHarity, since the erosion terraces ai'e there nar- row or wanting, but the five brooks sovith show it most clearly. Cuslimans Brook (called Mill River on the new map) runs down west of Mount Warner. The next two brooks south do the same, and then Fort River, the last of the series, illustrates the rule in the most striking- manner, and indeed formerly ran much farther south than now, parallel with the Connecticut, and entered the latter above Hockanum at the boat landing of the Mount Holyoke House. This has here plainly the following explanation: The water sank very suddenly in the lake, and tbe oldest position of the present river of which any trace remains was the eastern edge of the ten-ace system. On this sinking of the lake water the streams followed it by the shortest course, cutting gorges in their old deltas, and at one time each one joined the main stream at the point where it at present cuts the boundary between the lake bottom (1 b t) and the terrace system (tM^). As the Connecticut swung west and built up its terrace behind it the tributary elongated and kept its com-se across this newly formed terrace, and since this terrace flat or flood plain was built up as a series of bars which grew to be islands, behind each of which there is for a long time a long groove opening south (see p. 726), the brook occupied this and entered the main stream round its south end, and at last this operation, many times repeated, gave the streams their present course. It was the observation and study of this law several years ago which caused me to doubt the then prevalent idea held by those most competent to judge, that the Connecticut Valley had been filled up to the height of its high ten-ace — the lake bench — and then excavated, and led me to map the terraces, as I have done, into (a) a high bench or string of deltas bordering the valley; (b) a succession of lake bottoms sloping from the above center- ward and broadening in each of the wider stretches of the valley, and (c) a comparatively small area occupied by the "oscillation terraces" of the river proper — the "meadows." 748 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. DUlSrES AKD WIND LOESS.' President Hitchcock notes ^ the dunes in Montague and Hadley, and in the east part of Hadley south of the road, and their motion southeast- wardly.^ The lake bottom in Northfield is strikingly cut up by great dunes over the whole of the Beers plain, and farther south in Montague one can see where they have crept upon the west slopes of the islands which rose in the midst of the old delta of Millers River, the broad Montague plain. The low lake bottom in Hatfield, made up as it is of very fine sands, is also greatly affected by old dunes, and many of the scattered farm buildings are here built upon dunes, while a line of still moving sand drifts runs up through the center of the plain, and is indicated on the map. But the most remarkable exhibition of dunes in the valley is where the prevalent westerly winds strike the scarp which, on the east side of the river, separates the flood plain of the Connecticut from the lake bottom. This sharp, westward-facing scarp has been longest exposed to the winds, and is made up of very fine sands, and taking the eastern of the roads which runs from Sunderland to North Hadley, one crosses an almost con- tinuous line of great sand drifts until this road joins the next westerly one, and the line of dunes is continued southwai'd, and along the west side of Mount Warner has pushed high up the side of the hill. Farther south the scarp is notched in many places by old or still active dunes, one of which is in sight on the south side of the road from Amherst to Northampton, just before it enters Hadley. Wind loess. — All along the west slope of the Amherst ridge, especially opposite the lower openings in the ridge, as across the old cemetery or south of College Hill, a layer of fine unstratified loam or loess has been brought by the prevailing westerly winds from the broad lake bottom of fine sand which extends west from the bottom of the ridge. This layer is from 2 to 2 J feet thick, and extends over the whole ridge, resting on the shore sands and gravels, and higher up on the till, and extends for a long distance down over the east slope. I have traced it everywhere over the ridge in the network of cuttings for the gas and water pipes, the sewers, and the ' See PI. XXXV, in pocket at end of volume = Geology of Massachusetts, p. 130; Jour. Boston Soc. Nat. Hist., Vol. I, p. 80. 3 Geology of Massachusetts, Final Report, p. 326. MINEIIAL SPRINGS. 749 railroad, ami found it present in every undisturbed opening'. It was espe- cially well marked in the ditches dug across the Agricultural College farm iu 1896 for laying water pipes. MINERAL, SPRINGS. The noteworthy springs in the region may be classified as follows: 1. Springs fro^n the gneiss. In Shutesbury, just west of the village, is a spring which was discovered about 1808, and a hotel was built at the place, which is still called the Pool Tavern, although it has long been used as a private house and the well dug over the spring has caved in.-^ More celebrated are the "Orient Springs," in Pelham, so named by President Hitchcock. These springs rise on the strong transverse fault which crosses Pelham and Prescott. The large building built at the spring in 1861 was never a success, and it was burned in 1883. It is a quite strong sulphur spring. 2. Springs from the mica-schist. The abundant pyrite in these schists has everywhere produced springs which are strongly mineralized. In Amherst, especially along the west of the ridge, where the schists come near the surface, many wells contain so much copperas that the waters blacken tea and curdle milk. In Hawley the Moody Spring, in the southwest part of the town, is said to possess strong medicinal properties and to be a specific for salt- rheum and other cutaneous diseases A similar chalybeate spring in the southern part of Ashfield has a local reputation. The Mount Mineral Spring, Shutesbury, was known as a chalybeate spring as early as 1828.^ The Mount Mineral Spring Company was incor- porated in 1867. A fine hotel was sustained for some years, but burned in 1876, and the property has since been abandoned. Appended is an analysis of the water, furnished me by the present owner of the property. It is a pure alkaline chalybeate water containing manganese in solution. 'Evert's History of Connecticut Valley in Massachusetts, Vol. II, p. 758. 2 E. Hitchoook, Am. Jour. Sol., 1st series, vol. 13, 1828, p. 217. 750 GEOLOaT OF OLD HAMPSHIRE COUNTY, MASS. Analysis by S. Dana Hayes of water of Mount Mineral Spring; in one United States gallon of 231 cubic inches; June 14, 1878. Sulph. pot Snlph. lime Clilor. soda Bicarb. soda Bicarb, lime Carb. and crenate of iron. Carb. mang Alumina Silicic acid in solution Total. Parts per 100,000. 1,476 301 3,360 1,398 2,351 5,537 364 701 15, 488 Grams. 0.868 .175 1.949 .811 1.364 3.108 .223 trace .407 8.905 3. Water of artesian wells in the Triassic. All the artesian wells dug in the valley have much saline matter in solution. I was informed by Prof. C. U. Shepard, as the result of his analysis of the water of the South Hadley well, that common salt was present in large amount, and the abun- , dance of the salt pseudomorphs in the sandstones in which the well was bored indicates that this comes from the sea water entangled in the sand- stones at the time of their deposition. The appended analysis of a sample from the more northern well shows that the water has nearly the constitution of a bittern. One United States gallon of water contained in solution 102.54 grains of saline matter, which consisted of the substances named below. Analysis by Prof. G. A. Goessmann of one gallon of water from the artesian well at the Montague Paper Company's mills at Turners Falls, Massachusetts; made at Amherst, November 2, 1874. Potasaa Soda Magnesia Lime Chlorin Sulphuric acid Silicon Total ... G-raina. 0.352 2.994 3.690 36. 951 .363 58. 191 trace 102. 541 MINERAL SPKINGS. 751 4. Sprhii/s of the fjhicinl hikes in the iijildi/rls. — Rpriiig's rising from tlie baso *>t' till.' liravy sands of glacial lakes in the uplands rest oia the till, and those from the base of similar sands of the Connecticut Lake rest on the Chami)lain clays. These are hardly to be called mineral springs. The former furnish the sources' of many of our mountain brooks. The latter, K'ing nearer the villages, are better known. Of these are the slightly chalybeate spring at South Hadley Falls, the fine, strong spring which gushes out of the blufi" west of Hatfield village, and several issuing from the bluffs that surround Deerfield. In Springfield the Wesson Spring, Avhich supplies the water of Court Square and a fountain at the corner of Willow and Stockbridge streets; the Walker Spring, at the corner of Maple and Stockbridge streets, and the Ingersoll Grrove Spring, a hundred feet south of Dartmouth terrace, the water of which is sold largely in the city, are of this character. The rain waters which have fallen upon the surface of the high terrace on which the higher portion of the city is built sink through these sands to the horizontal and impervious surface of the clays beneath and emerge at the edge of the blufi". The Ingersoll Grove Spring was reported upon by the State board of health,^ and the result of its analysis is given below (I) in connection with the analysis (II) in the same pamphlet of the Massasoit Spring, described below. The Massasoit Spring is of ideal purity; the other gives plain indi- cation of the presence of the barn, sewer, and streets, which are reported in the immediate vicinity of the spring. Analyses of waters of Ingersoll Grove and Massasoit springs. [Parts in 100,000.] I. Ingersoll Grove. II. Massasoit. liesidue on evaporation 8.70 .000 .0008 .43 .5000 .0002 .0275 2.73 .0 5.50 .000 .000 .09 .0600 .0001 .0160 2.86 .0 Ammonia : Free . Nitrogen as — Nitrates - . - Nitrites Oxygen consumed Hardness . 1 Examination of spring waters ofifered for sale in Massachusetts : Twenty-third A.nn. Eept. State board of health, pub. doc. 34, 1891 (also separate publication), pp. 356, 362, 364. 752 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. The Massasoit Spring- on the "Bear Hole" farm, in the western part of West Springfield, issues from the base of a very high bluff of sand that forms the east wall of the deep channel which the Black Brook has cut in the broad sand plain. The spring is said to show the uniform temperature of 45° F. tln-oughout the year. An analysis by Prof Charles Mayr, pub- lished in the pamphlet advertisement of the spring, is here given: Analysis of icater of Massasoit Spring. Grains in 1,000,000. Grains in 1 gallon. Sodium chloride (salt) 6.0 23.0 8.0 4.2 4.0 12.0 trace 0.360 1.380 .480 .252 9.240 .720 trace Silica Organic substances . . Potash, iron, alurnina, phosphates, nitrates Total 57.2 3.432 Although the spring was discovered only in 1886, very attractive buildings have been erected and it has become a well-known place of summer resort, and the water has been put on sale in Springfield for table use. The brook just to the west runs over the surface of the trap, which dips with great thickness beneath the sands from which the springs flow, but it is not probable that a deep-seated water coming up through the sandstone and trap would be so pm-e. It is probable that the waters come wholly from the sand itself, and that the exceptional purity comes from the fact that they have been filtered through a hundred feet of this sand. THICK MODERIf FISSURE DEPOSITS OP QUARTZ SURROUNDING ROOTS IK THE BASE OF THE HOLXOKE TRAP SHEET. In 1891 a great block of the trap fell from the vertical wall at the point on the river above Titans Pier, where the trap contains limestone, and dis- closed a mass of translucent chalcedonic quartz nearly as large as a man's head, which was pierced with tubular openings 0.5 to 1 ™™ in diameter and at least 4 inches in length, generally nearly but not rigidly parallel, and so PISSUEE DEPOSITS OP QUARTZ, 753 closely gi'ouped that the separating walls of silica were quite tliiu or partly wanting. At times they Avere quite wide apart or in small groups. These tubes are lined with limonite and sometimes nearly tilled with it. At times a separate cylinder or open tube of limonite is found free in the cavities. The limonite can not be wholly removed from the cavity, but impregnates the silica ft)r a small but definite distance in from the surface of the cavity. The most striking circumstance is that the silica in one portion of the mass grades with imperceptible boundary into a mass of distinctly banded, siliceous, dove-colored limestone, or ankerite, as it oxidizes into a porous ocher. It seems tolerably clear that the general explanation of this must be that a mass of rootlets penetrating a fissure of the trap became coated with limonite and that then a deposit of silica, at first impregnating the limonite there, went on to fill the whole fissure, while in part of the latter a mixture of calcite and silica completed the work. Where the delicate cylinders and tubes of limonite rest free in the cavity we may suppose that limonite was deposited within the bark of the rootlet, replacing or surrounding the shrunken pith. Indeed, a portion of this bark remained in the tube at one place and was in part removed and burned. MON XXIX 48 OHAPTEE XXII. SUPPLEMENT TO THE AUTHOR'S MINERAL LEXICON OF FRANKLIN, HAMPSHIRE, AND HAMPDEN COUNTIES.^ 1895. Albite. Blandford; Osborn's soapstone quarry. Fire, fresh, white-translucent crystals an inch across. In flat plates from growth in fissures and large development of basal plane, which is deeply striate parallel to the intersection edges with the primary prism. Twinned by the pericline law and with few plates interposed according to the albite law. Forms present, b (010), c (001), m (110), // (450), / (130), C (150), M (110), v (450), z (130), X (101), y (112), e (021), p (111), o (111). (See p. 85.) 1896. Albite. Chester. At the adit of the new mine opened north of the road opposite the old Emery mine. The mineral occurs in perfect simple white crystals an inch in leugth. They inclose titanite and are coated with prochlorite. 1892. Allanite. Belchertown. Cited from Belchertown. E. S. Dana. Sys. Min., p. 1058. 1892. Ankerite. Middlefield. E. S. Dana. Sys. Min. Localities, p. 1059. Doubtless from the steatite bed. All the specimens I have examined from these beds were dolomite. 1892. Anthophyllite. Blandford. E. S. Dana. Sys. Min., p. 1058. This is the brown actinolite from Osborn's soapstone quarry. 1892. Anthophyllite. Chesterfield. E. S. Dana. Sys. Min., p. 1058. This is the hair-brown, coarsely fibrous mineral from the bluff above Burnell's pond, which is identical with the cummingtonite or amphibole-anthophyllite occurring in Cummington, a little way farther north, in the Conway schists. ' See Bull. U. S. Gaol. Survey No. 126, 1895. 754 SUPPLEMENT TO MINERAL LEXICON. 755 1892 Antiiophyllite. Chesterfield. E. S. Dana. Sys. Min. Localities, p. 1058. I have never found tliis mineral in Chesterfield, nor any of the minerals with wliich it is associated; nor do I recall any other citation of the mineral from this town. I suppose it to be the brown cummingtonite from the bluffs west of liuruell's pond. 1858. Anthophyllite. Enfield. Specimens labeled "anthophyllite gneiss" in the State collection. E. Hitchcock, Nos. 96, 97, under gneiss. Oat. State Col. Mass. Agr. Kept., p. 15. A dark-brown, bladed mineral. The powdered fragments all extinguish lon- gitudinally, as if it were a rhombic mineral. 1895. Anthracite. Holyoke. In Chicopee shale of Triassic age below the Holyoke dam. In thick masses coating siderite. It has rounded surfaces, showing that it was introduced into the fissure as a bitumen. It is in very brittle layers, which give a yellow flame for an instant and then glow without further flame. (See p. 370.) 1896. Apatite. Blandford. Occurs in the Osborn soapstone quarry, in rich, deep oil-green crystals an inch long, iutercrystallized with chlorite. (See p. 85.) 1895. Apatite. Chester. Crystals 1 to 3"™ in length occur on and in the diaspore. (See p. 143.) 1895. Aeagonite. Chester. A beautiftil fibrous satin spar occurs in the serpentine at the old mine, in sheets a foot square and IJ inches thick. (See p. 143.) 1895. Barite. Holyoke. Cavities 4 inches long and one-third inch wide and an inch deep, with rec- tangular ends or ends beveled like barite crystals, occur in the Chicopee shale below the Holyoke dam. (See p. 370.) 1892. Bastite. Westfield. B. S. Dana. Sys. Min. Localities, p. 1060. Cited as Schiller Spar (Diallage). This is a bastite derived from enstatite, from Munns Brook. 1818. Beryl. Emerald. Chesterfield, Goshen, Northampton. Chesterfield furnishes them in great abundance, from the weight of an ounce, or less, to six pounds. Hexagonal prisms; diameter sometimes twelve inches; light green (Waterhouse). Northampton and Goshen (Hunt). All coarse granitic beryl. Samuel L. Mitchill. Phillips Mineralogy, with additions on American Min- erals. 756 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. 1892. Beryl. Russell, Warwick. E. S. Dana. Sys. Min. Localities, p. 1059. 1841. Calcite. Iceland Spar. Wales. In gneiss. E. Hitchcock. Final Eept. Geol. Mass., p. 638. 1897. Cerdsite. Hatfield. At the lead mine the mineral occurs in small globular forms, with drusy surfaces, and some of the globules are hollow. 1896. Clinochlok. Blandford. Broad encrusting masses, 3 inches thick, of a uniaxial chlorite, which is granular at the base but grows coarser upward, and grades into stout crystals one-half inch across, which project freely from the surface. Prom the Bland- ford soapstone quarry, formerly Osborn's quarry. (See p. 85.) 1896. Corundum. Emery. Middlefield. Found one-half mile north of the soapstone quarry at 0. Smiths's. Traced north from bowlders by Dr. H. S. Lucas. Shows quite large veins of blue corundum. (See p. 81.) Springfield Eepublican, December 12, 1896. 1897. Corundum. Pelliam. Crystals with perfect polished O face. Heating developed O cleavage. Others with polished prism faces deeply fluted horizontally by the oscillation of the unit pyramid face. (See p. 47.) 1853. CuMMiNGTONiTE. Hornblende. Cummington. Fibrous, resembling anthophyllite; color, ash-gray; in mica-slate. Analyses of cummingtonite. SiOo . Ar^O' FeO. MgO MnO CaO . Na20 KjO. H^O., Per cent. Per cent. 51.09 trace .95 32.07 10.29 1.50 .75 trace 3.04 99.69 50.74 .89 33.14 10.31 1.77 trace .54 3.04 100. 43 J. L. Smith and G. J. Brush, Eeexamination of American minerals : Am. Jour. Sci., 1st series, Vol. XVI, 1829, p. 48. SUPPLEMENT TO MINERAL LEXICON. 757 1892. CuMMiNGTONiTE. Ainpliibole-Antliophyllite, Iron-Magnesium, Amphi- bolo. Ciunmiug-ton. Sp. gr. = 3.1 to 3.32. E. S. Dana. Sys. Min., pp. 390, 395. Cites above analyses. 1895. Datolite. Northampton. Delaney's quarry, on railroad near north line of Holyoke, in Triassic diabase. Discovered by liev. J. Prevost. In a crushed zone in tlie diabase embedded in calcite. Very flne crystals of most brilliaut luster, the largest nearly a half inch across, and with the slightly green tint which is common in the Bergen crystals, but here a little more yellow than there. The forms are of a type new in the valley, resembling fig. 1, with the addition of the base, or fig. 4, page 503, of Dana's Sys. Min., 1892, but with e and /< greatly increased at the expense of m andw,; c is a large composite face. The forms present are c» P (110), J P (Il2), O P (001), a> P 00 (100), f P cib (023), P db (Oil), ^- P (113), — P (111), ^ P a> (102), — 2 P 2 (121), — J P ^ (102), ^- P CO (013), a, P cc (010), the last four small. Specimens in the Smith and Amherst College cabinets. (See p. 470.) 1895. Datolite var. Botryolite. Greenfield. At Cheapside, in a new road cut through the trap ; in white globular masses in steam holes in red trap. (See p. 443.) 1896. Enstatite. Grranville. The mineral, in large, square, colorless prisms from the large bed at Downey's (see p. 90), has been analyzed by Mr. W. P, Hillebrand, with the following result: Analysis of enstatite from Granville. Per cent. SiOj 54.04 none .52 .14 1.51 3.90 .23 .11 none none none 34.40 V .08 .70 3.07 none 1.32 TiO» AI2O3 ^ CraOs . .. Fe 3 3 . FeO NiO - MnO CaO SrO BaO . MgO K2O Na^O " LiaO H2O below 110° C H«0 above 110° C. P2O1 CO.. IOCS. 02 758 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. 1892. Enstatite. Westfield. E. S. Dana. Sys. Min. Localities, p. 1060. Cited as Scapolite. This is the enstatite, or the compact feldspar associated with it, from the serpentine locality on Munns Brook. (See p. 90.) 1896. Epidote. Hunting-ton, on farm of W. L. Angell. In a fissure in gneiss associated with quartz, biotite, albite, and calcite. (Pen- field and Pirsson.) Lighter -colored crystals bent and broken. Si02 = 37.99. AI2O3 = 29.59. FeOj = 5.67. FeO = 0.53. MnO = 0.21. OaO = 23.87. H2O = 2.04. f sp. gr. 3-.367. Contains minimum Fe. Paces: u(100),_c(001), m(llO), w(2l0), e(lOl), i(l02), r(TOl), k(012), o(Oll), n(Ill), g(221), y(211). Twins (100). Optical constants given. Double refrac- tion diminishes with the iron. E. H. Forbes, Zeit. Krys. u. Min., Vol. XXVI, p. 138. 1859. FiBROLiTE. Palmer. Cited as cyanite under mica-schist in catalogue of State collection, Nos, 216 and 218. E. Hitchcock. Sixth Ann. Eept. Dept. Agr., p. 14. This is the coarse fibrolite from bowlders which are in place in the schist area included in the Belchertown tonalite. (See p. 243.) 1892. GrALENA. Westhampton. E. S. Dana. Sys. Min. Localities, p. 1060. Not elsewhere cited. 1892. GrEDRiTE. Orange; east of North Orange, on the west slope of Big Tully Mountain Wrongly cited from Warwick in Mineral Lexicon (Bull. U. S. Geol. Survey No. 126), p. 86. 1892. Heulandite. Chester. Dana. Sys. Min. Localities, p. 1058. Cited as Stilbite. This is cited from " E. Emmons, Mineral localities: Am. Jour. Sci., 1st series, VoL VII, 1824, p. 254. Was recognized to be heulandite by Prof C. U. Shepard from the specimens in Emmons's cabinet. Boston Jour. Phil., Vol. Ill, p. 608. 1896. Lazulite. Chittenden, Vermont. In quartz-muscovite rock. There is a specimen in the collection of Harvard University from the above locality, where it was found by Mr. C. H. Whittle. This is probably the locality from which the unique specimen found in Green- field came. See under Lazulite, in Bull. U. S. Geol. Survey No. 126. 1895. Olivine var. Villaesite. Blandford. At the base of the upper (eastern) serpentine bed at the Osborn soapstone quarry, in lenticular remnants in the serpentine associated with some still unchanged olivine. (See p. 85.) SUrPLEMENT TO MINERAL LEXICON. 759 1896. Pkochlorite. Chester. On albite iu druses at adit north of road at old miue, in fine, large masses. (See p. 143.) 1734. Pyrite. Northampton'? "Marcasites," Pyrites. "Fragments of greenish sulphurous marcasite from Mount Tom and Kolyoke, each side Connecticut River." — John Wintlirop, P. R. S., Ex. Vol. XV, Journal Book of Royal Soc. Am. Jour. Sci., Ist series, Vol. XL VII, 1844, p. 289. 1892. Pyrolusite. WiUiamsburg. E. S. Dana. Sys. Mln. Localities, p. 1060. 1897. Pyroxene. Diopsicle. Bald Mountain, Shelburne Falls, Massachu- setts. In a dark, impure limestone. The crystals are themselves full of inclosed limestone and effervesce strongly. They are in stout prisms up to an inch and a half in length and a half inch across, greenish-white in color, strongly lustrous on the prism faces and glossy; color, pale green. They are nearly square prisms and recall the Canaan white pyroxenes. This mineral shows under the microscope the brilliant colors and the strong prismatic cleavage of pyroxene, and a basal parting with many interposed twin laminae. The extinction reaches 33°. The specimen probably comes from a limestone bed of the Conway schist, which has been strongly and peculiarly metamorphosed by contact with granite. 1888. Quartz. Rose quartz. Blandford. Abundant by roadside near E. H. Osburn's. 1892. Quartz. Amethyst. Greenfield. The cavities in the red diopside-diabase, described on page 443, from the cut through the trap ridge made for the electric railroad, contain small amethysts of great beauty, which are inter penetration twins of model-like perfection. The twinning plane is O (0001). 1897. Rhodonite. "Cunningham" (for Cummington.) Gr. P. Merrill. Stones for building and decoration, p. 174. Cites Kunz, Min. Rec, 1887. 760 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. 1895. Salt. Holyoke. Small cavities, which, seem to be ilattened and slightly distorted from the cubical form, occur below the Holyoke dam in the Triassic Chicopee shale. (See p. 370.) 1895. SiDERiTE. Holyoke. In trench below Holyoke dam, on fissure surfaces of Chicopee shale of Tri- assic age ; broad surfaces, coated with drusy crystallization, crystals one-fourth to one-half of an inch across, yellow-gray to warm reddish- yellow, with brilliant luster. Forms E, eo E 2, in equal development, which makes the attached crys- tals simulate dodecahedrons, so that they can almost be taken for garnet. Followed by gypsum ( '?), barite, calcite, anthracite, pyrite. (See p. 370.) 1879. Serpentine. Picrolite. Florida. Specimen in collection of Harvard University. Analysis by W. H. Melville. Analyses of serpentine. Per cent. Per cent. SiO. 44.22 J 6.61 I .53 44.22 ]■ 7. 91 Fe^Oa Al^Oa MgO 37.54 .36 11.26 37.40 .36 11.22 H2O(100i) H2O (al)ove 100) 100. 52 100. 11 /\ M. E. Wadsworth, Proc. Boston Soc. ISat. His., Vol. XX, p. 286. (See p. 73.) 1897. Serpentine. Variety picrolite. Pelham; asbestos mine at the bottom of the large digging and in the midst of the unchanged olivine rock. A thick seam of a leek-green columnar and polished serpentine, plainly produced by pressure and slipping. A thick layer of slickensided columnar serpentine. (See p. 47.) 1825. Spodumene. Vicinity of Deerfield. (Groslien or Chestei-field.) Light green, brittle, exfoliates with blowpipe. Yields prism of 100°; con- tains 8 per cent lithia. Found in a collection of minerals, but precise locality not known. George Bowen, Proc. Phil. Acad. Sci,, Vol. Ill, p. 285. 1825. Spodumene. S. Eobinson. Oat. Am. Min. Citation of above. 1892. Spodumene. Chester. E. S. Dana. Sys. Min, Localities, p. 1058. This citation depends on the doubtful report of the species in Chester, in small quantity in granite, by C. Dewey. Geology of Berkshire: Am. Jour. Sci., 1st series, Vol. VIII, 1824, p. 243. SUPPLEMENT TO MIJSTEKAL LEXICON. 761 1896. Spodumene. tloslieu. Tslv. Alvaii Barnis writes me as follows coucerning the spodumeue localities ill tlie north part of (loslieii, near Taylor's mill, and at Alannings: Now York parties have bcou at work on the siiodiiinoiir matter for lithia lor the past eight years, oil' and on. They had iio dilBculty in g('tting it into a sohition, but had tronhlo in making the separation. They wrote me a few days ago that they had succeeded in doing it all right and wouhl soon report results, for which I am still waiting. Wo find the spodumene in i>laco at two points, as iudioatcd on the map, one leading north and south and the other half a mile to the east, rnuuiugoast and west. There seems to l>e au abundance of it. 1896. Talc. Soapstone. Blandford. Reported from the north end of Blair's pond. (S. A. Bartholemew.) Also as au inclosure in horubleade-schist on the road going north from North Bland- ford past Bartholemew's quarry, 100 rods east of the road on the west side of Kound Hill. 1896. TiTANiTE. Chester. At the new adit north of the road at the old mine, in druses in and on albite, and covered by prochlorite ; wine-yellow; common flat forms, often twinned; fine crystals, S-G™"" long. (See p. 143.) 1852. Tourmaline. Chesterfield. The colored tourmalines are rarely terminated. A fine crystal is figured having the faces oo E 2, cc E, O E, E, with the basal plane making nearly the whole termination of the crystal. 0. U. Shepard. Treatise on Mineralogy, p. 220. 1896. Tourmaline. Huntington. A mile north of Knights ville, at the 700-foot contour, on the east side of the river. — A. Barrus (private communication). 1896. Tourmaline. Huntington. Beautiful flattened tourmalines occur in muscovite at the quarry in pegma- tite, near Knightsville. 1896. Tourmaline. Dendritic Tourmaline. Northampton. In fissures in the fine-grained muscovite-biotite-granite from the village of Haydenville ; an exquisite, delicately traced dendritic growth of tourmaline. The surfaces of the fissures are perfectly flat, wholly fresh, and the rock for 1 or 2 millimeters in is whiter from the absence of biotite, while the surface on which the dendrite is has also a slight excess of biotite in larger crystals than in the rest of the rock, and a few brown-red garnets. (See PL VII, p. 316.) 1896. Zoisite. Chesterfield. The locality is found by following the brook which enters East Branch a mile south of Bisbee mill, five-eights of a mile east, and then going 30 rods south into a spur of the hill marked 1455. — A. Barrus (private communication.) CHAPTER XXIII. CHRONOLOGICAL LIST OF PUBLICATIONS UPON THE MIN- ERALOGY AND GEOLOGY OF FRANKLIN, HAMPSHIRE, AND HAMPDEN COUNTIES. 1734. Selections from an ancient catalogue of objects of natural history formed in New England more than one hundred years ago, by John Winthrop, P. E. S. Journal Book of Royal Society, vol. 15, p. 451; Am. Jour. Sci., vol. 47, 1844, p. 282. The paper was copied from an ancient manuscript. (See under Olay- stones.) 1796. J. Morse. The American universal geography, 3d ed., Boston, p. 410 ; copper ore, Leverett; black lead, Brimfleld. 1810. J. Morse. The American gazetteer, 2 vols., Boston. 1810. B. Silliman. Particulars relative to the .lead mine near Northampton (Mass.). Bruce's Journal, vol. 1, p. 63. 1811. WUliam Meade, M. D. A description of several combinations of lead lately dis- covered at Northampton. Addressed to the editor. Bruce's Jourrial, vol. 1, p. 149. 1815. E. Hitchcock. Southampton lead mine; Basaltick columns on Mount Ilolyoke. North American Review, vol. 1, p. 334. 1816. Parker Cleaveland. Mineralogy and geology. Boston, 8°. 1817. J. F. L. Hausmann. Kieselspath von Chesterfield, Mass. Getting Gelehrte Anzeigen, p. 1401. 1818. E. Hitchcock. Description of Turners Falls on Connecticut River; with sketch by Mrs. Hitchcock. Portfolio. Philadelphia. 1818. B. Hitchcock. Remarks on the geology and mineralogy of a section of Massa- chusetts on Connecticut River, with a part of New Hampshire and Vermont; 12 pages; dated October, 1817; map in 1st and 2d editions omitted in reprint; contains list of minerals. Am. Jour. Sci., 1st series, vol. 1, p. 105. 1818. Samuel L. Mitchill. An elementary introduction to mineralogy, by William Phillips, with notes and additions on American minerals, by Samuel L. Mitchill, Professor of Mineralogy, etc., in the University of New York. Cyanite and beryl, from Hampshire County. 1819. Amos Eaton. Account of the strata perforated by, and of the minerals found in, the great adit to the Southampton lead mine; 4 pages. Am. Jour. Sci., 1st series, vol. 1, p. 136. 1819. B. Silliman. Localities of minerals, etc. ; "Molybdenais found in Shutesbury * * * on land of William Eaton;" 1 page. Ibid., p. 238. 762 LIST OF PUBLICATIONS. 763 1819. (ieorgo Gibbs. On the tounniilines and other minerals found at Chesterfield and Goshen, Mass.; G iiagcs. Am. Jour. Sci., 1st series, vol. 1, p. 346. 1811). K. Hitchcock. Supplement to the remarks on geology, etc., of a section of Massachusetts; 3 pages. Ibid., p. 430. 1820. Amos Eaton. Index to geology of Northern States. 12°, Troy, N. Y. 1820. Chester Dewey. Localities of minerals; 3 pages. Am. Jour. Sci., 1st series, vol. 2, p. 23G. 1821. Stromeyer. Chemische Untersuchungen. Gottingen. Aalysis of Chesterfield albite, p. 307. 1821. Dr. William Atwater. Extract of letter; Hill of serpentine In Westfleld; 1 page. Am. Jour. Sci., 1st series, vol. 3, p. 238. 1821. Editor's note. On fossil fish (Sunderland), with catalogue of specimens sent by E. Hitchcock; 2 pages. Am. Jour. Sci., 1st series, vol. 3, p. 365. 1821. T, Dwight. Travels in New England and New York. 8°. Vol. 1, pp. 34-35. 1822. Editor's note. Micaceous iron; Northampton and Hawley. Am. Jour. Sci., 1st series, vol. 4, p. 53. 1822. Editor's note. "This fluor spar is of a grass or emerald green, a rare color in this country ; not found except near Northampton, by Dr. David Hunt." Ibid., p. 188. 1822. Prof. C. Dewey. Miscellaneous notices relating to American mineralogy and geology; Crystallized steatite in Middlefield ; 3 pages. Ibid., p. 274. 1822. Parker Cleaveland. An elementary treatise on mineralogy and geology; 2d edition ; 2 vols. 1822. Editor's note. Miscellaneous notices on mineralogy; Adularia, Brimfield, Mass. (Prof. Amos Eaton). Am. Jour. Sci., 1st series, vol. 5, p. 41. 1822. Prof. Amos Eaton. Geological and agricultural report of the region adjoining the Brie Canal (with profile of the rocks across Massachusetts from Boston to Northfield, by E, Hitchcock). 1822. Prof. C. Dewey. Notice of crystallized steatite (Middlefield); 1 page. Am. Jour. Sci., 1st series, vol. 5, p. 249. See Hampshirite in Mineral Lexicon, Bull. U, S. Geol. Survey No. 126, p. 91. 1822. E. Hitchcock. Fluate of lime and noble agates in Deferfleld. Ibid., p. 407. 1823. H. J. Brook. Cleavelandite. Annals of Philosophy, p. 381. 1823. E. Hitchcock. A sketch of the geology, mineralogy, and scenery of the regions contiguous to the Eiver Connecticut; with a geological map and drawings of organic remains, and occasional botanical notices. (Bead before the Amer- ican Geological Society at their sitting, September 11, 1822. Part I, 86 pages; section. Mount Toby, pi. 8; fossils, pi. 9; map, pi. 10.) Am. Jour. Sci., 1st series, vol. 6, p. 1. 1823. E. Hitchcock. Same. Part II, Simple minerals ; 35 pages. Ibid., p. 201. 1824. E. Hitchcock. Same. Part III, Scenery. Ibid., vol. 7, p. 1. 1824. E. Hitchcock. Same. Part IV, Miscellanies; 1 plate. Ibid., p. 16. [The same published separately. New Haven. S. Converse, publisher, 1823.] 764 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. 1824. Anon. Hogtootli spar from Williamsburg. Hampshire Gazette, July 14. 1824. Prof. Chester Dewey. A sketch of the geology and mineralogy of the -western part of Massachusetts and a small part of the adjoining States; with geolog- ical map; 60 pages [contains many notes from Emmons]. Ibid., p. 1. 1824. E. Emmons. Article on minerals of Chester, etc., in Lyceum of Nat. Hist, of Berkshire Medical Institute; published ("?). See Am. Jour. Sci., 1st series, vol. 8, p. 32. 1824. Dr. William Meade. Localities of minerals; Siliceous oxide of manganese, Ches- terfield. Ibid., p. 54. 1824. Dr. Jacob Porter. Localities of minerals; Red oxide of titanium, Cummington; Sulphuret of molybdena, Chesterfield. Ibid., p. 58. 1824. George T. Bowen. Analysis of spodumene from the vicinity of Conway, Mass. Jour. Acad. Nat. Sci., Phila., vol. 3, p. 284; Am. Jour. Sci., 1st series, vol. 8, p. 121. 1824. J. Porter. Localities of minerals; 1 page. Am. Jour. Sci., 1st series, vol. 8, p. 233. 1824. C. U. Shepard. Localities of minerals; Pelham, etc. [First notice of Shay's flint.] Ibid., p. 235, 1824. Prof. Chester Dewey. Additional remarks on the geology of a part of Mas- sachusetts, etc. ; 5 pages. [First notice of spodumene, before called white augite.] Ibid., p. 240. 1824. E. Emmons. Notice of the granite veins and beds in Chester; 3 pages, 1 plate. Ibid., p. 250. 1824. Professor Dewey. Additional notice of argentine [in Williamsburg]. Ibid,, p. 248. 1824. C. U. Shepard. Green feldsijar associated with sappare and siliceous oxide of manganese, in Chesterfield. Ibid., p. 251. 1824. Jacob Porter, Localities of minerals. Ibid., p. 252. 1824. Dr. Eben Emmons. Miscellaneous localities. Ibid., p. 254. 1825. S. Robinson, M. D. Catalogue of American minerals, with their localities, Boston, 8°. 1825. E. Hitchcock. Notice of several localities of minerals in Massachusetts; spodu- mene corrected; pyrophysalite in Goshen; 3 pages. Am. Jour. Sci., 1st series, vol. 9, p. 20. 1825. C. U. Shepard. Localities of minerals. Ibid., p. 47. 1825, J. Porter. Localities of minerals. Ibid., p. 54. 1825. A.O. Hubbard. Remarks oti the lead veins of Massachusetts and glacial lakes; 2 pages. Ibid., p. 166. 1825. B. Silliman. Notice of a mineral supposed to be phosphate of lime from Wil- liamsburg and of the localities of several other minerals. Ibid., p. 174, 1825. J. W. Webster. Determination of chlorophteite from Turners Falls, from speci- mens sent by Dr. E. Hitchcock. Boston Jour, of Phil, and Arts, vol, 2, p. 610, 1825. E. Hitchcock. Mineral localities. Ibid., p. 610. LIST OF PUBLICATIONS. 765 1825. C. U. Shepard. Mineral localities, witli description of aiitbophyllite (=actino- lite), iolito (=beryl). Boston Jour, of Phil, and Arts, vol. 2, p. .395. 1825. O. U. Shepard. Mineral localities, with description of spodumene, beryl, schiller- spar, heulandito, hematite, anthophyllite, and zoisite from the Berkshire Hills. Ibid., p. (i07. 1825. E. Hitchcock. Geological sketch of the country on the Connecticut Eiver;- map and engravings. Noticed. Am. Jour. Sci., vol. 9, p. 179. 1825. E. Ilitchiiock. Topaz* (in Goshen); 1 page. Ibid., p. 180. 1825. C. U. Shepard. Locahties of minerals. Ibid., p. 248. 1825. F. .Mohs. Jlineralogy. English edition; W. Haidinger, Edinburgh. Cites American localities fully. 1825. E. Emmons. Carbonate of manganese, Cummington, Mass. Am. Jour. Sci., 1st series, vol. 9, p. 249. 1S25. Emerson Davis. Localities of minerals (West Springfield). Ibid., p. 252. 1826. E. Emmons. Localities of minerals (Chester and vicinity). Ibid., vol. 10, p. 11. 1826. Simeon Coltou. Localities of minerals (Monson and vicinity). Ibid., p. 12. 1826. J. Porter. Localities of minerals (Plainfleld and vicinity); correction of D wight's Travels; house on Eouud Hill of soapstone from Middlefleld, not Plainfleld. Ibid., p. 18. 1826. J. Finch. Memoir of the new or variegated sandstone of the United States. [First suggestion that the Connecticut Eiver sandstone was the New Eed.] Ibid., p. 209. 1826. Emerson Davis. Notice of rocks and minerals in Westfleld. Ibid., p. 213. 1826. Prof. J. W. Webster. On chlorophaeite from Gill, Mass. Boston Jour, of Phil. and Arts, vol. 4. 1826. E. Hitchcock. Chlorophseite (Gill). Am. Jour. Sci., 1st series, vol. 10, p. 393. 1826. Anon. Tabular quartz at Palmer, Mass. The Chemist and Meteorological Journal, John E. Cutting, editor, Amherst Mass., vol. 1, p. 78. 1826. Eeport of the commissioners of the State of Massachusetts on the routes of canals from Boston Harbor to Connecticut and Hudson rivers. Letters from Prof. Edward Hitchcock and others, with geological details; 248 pages; large folding map. 8°. 1827. Alanson Nash. Notices of the lead mines and veins of Hampshire County, Mass., and of the geology and mineralogy of that region ; figures in text and engraved map; 33 pages. Am. Jour. Sci., 1st series, vol. 12, p. 238. 1827. Jacob Porter. Localities of minerals. Ibid., p. 378. 1828. E. Hitchcock. Miscellaneous notices of minerals, with geological remarks; 16 pages. Am. Jour. Sci., 1st series, vol. 14, p. 215. 1828. Editor's note. Chesterfield tourmalines. "Mr. Clark designs to explore his locality and will be better prepared to furnish collectors of cabinets who may visit him." Ibid., p. 400. 1829. Editor's notice. Analysis of tourmaline (green, Chesterfield), by Gmelin. Ibid., vol. 16, p. 389. 1829. E. Hitchcock. Tin at Goshen. Ibid., vol. 16, p. 188. 766 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. 1829. 0. U. Shepard. Discovery of stanniferous columbite in Massachusetts (Chester- field) ; 8 pages. Am. Jour. Sci., 1st series, vol. 16, p. 218. 1829. A History of Berkshire, Mass. ; in two parts. By "Gentlemen in the county." Pittsfteld, Samuel W. Bush. 1830. A. Eaton. All primitive general strata below granular quartz are contempo- raneous and schistose; 2 pages. Am. Jour. Sci., 1st series, vol. 17, p. 334. 1831. J. Porter. Localities of minerals. Ibid., vol. 20, p. 170. 1831. John G. Hales. Plan of the town of Northampton, in the county of Hampshire, surveyed under direction of the selectmen, January, 1831. Pendleton's Lithography, Boston. 1832. E. Hitchcock. Report on the Geology of Massachusetts, examined under the direction of the Government of that State during the years 1830 and 1831 ; Part I, The economic geology of the State, with a geological map ; 70 pages. Am. Jour. Sci., 1st series, vol. 22, p. 1. 1832. Alfred Smith. On the water courses and the alluvial and rock formations of the Connecticut Eiver Valley. Ibid., p. 204. 1832. E. Emmons. Manual of mineralogy and geology. 2d edition; Albany; 12°; 299 pages. Am. Jour. Sci., 1st series, vol. 24, p. 397. 1833. Editor's note. Professor Hitchcock's report of the geology of Massachusetts (chromate iron, Blandford; rotten stone. West Springfield). Ibid., p. 396. 1833. E. Hitchcock. Report on the geology, zoology, and botany of Massachusetts; 692 pages; atlas, 19 plates. (See 1832 above.) 1835. C. U. Shepard, Microlite, a new mineral species; 2 pages. Am. Jour. Sci., 1st series, vol. 27, p. 361. 1835. C. U. Shepard. Treatise on mineralogy; (Part I, containing terminology and characteristics, 1832, 256 pages) ; Part II, in 2 vols.. Description of species, 1835; vol. 1, 300 pages; vol. 2, 331 pages; 12°; New Haven. 1835. E. Hitchcock. Report on geology, etc. ; 2d edition ; 702 pages, 18 plates, 60 cuts. (See 1833 above.) 1835. Editor's note. Soapstone or steatite of Middlefleld. Am. Jour. Sci., 1st series, vol. 27, p. 382. 1835. Editor's note. Uranite at Chesterfield, Mass., described by Professor Shepard. Ibid., vol. 28, p. 382. 1836. E. Hitchcock. Ornithichnology, or description of the footmarks of birds (ornith- ichnites) on New Red sandstone in Massachusetts; 34 pages, 3 plates. Ibid., vol. 29, p. 307. January, 1836. 1836. E. Hitchcock. Controversy with Rev. Mr. Chapin, of Connecticut, on foot- marks. Knickerbocker, vol. 8, p. 289. September, 1836. 1836. J. H. Redfield. Fossil fishes of Connecticut and Massachusetts, with a notice of an undescribed genus. Annals of the Lyceum of Nat. Hist., N. Y., vol. 4, p. 35. 1836. Editor's note. Albite of Chesterfield. Analysis by MM. Aug. Laurent and Ch. Holms. Am. Jour. Sci., 1st series, vol. 30, p. 381. LIST OF PUBLICATIONS. 767 18.H7. 1"]. ITitclicock. Fossil footsteps in sandstone and graywacke; .'5 pages. Am. Jour. Sci., 1st series, vol. 32, p. 174. 1837. 0. U. Shepard. Chemical examination of microlite. Ibid., p. 338. 183S. J. W. Foster. New locality of iolite, with other minerals associated (Brimfleld); 2 pages. Ibid., vol. 33, p. 399. 1838. C. IJ. Shepard. Notice of a second locality of topaz in Connecticut, and of the ])heiiakite in Massachusetts; 3 jjages. Ibid., vol. 34, p. 329. 1S3S. F. Hitchcock. Eeport on a reexamination of the economical geology of Mas- sachusetts; 139 pages; Boston, Button & Wentworth, State printers. 1839. Editor's note. Solid impressions and casts of di'ops of rain. Am. Jour. Sci., 1st series, vol. 37, p. 371. 1840. E. Hitchcock. Elementary geology. Amherst. J. S. & 0. Adams; 12°; 329 pages. 1841. W. C. Eedfleld. Short notices of American fossil fishes; 5 pages. Am. Jour. Sci., 1st series, vol. 41, p. 24. 1841. Mr, Teschemacher. On the occurrence of phosphate of uranium in the tourma- line in Chesterfield; abstract. Proc. Boston Soc. Nat. Hist., April, 1841, vol. 1, p. 15. 1841. E. Hitchcock. Final report on tlie geology of Massachusetts ; in four parts : 1, Economical geology; 2, Scenographical geology; 3, Scientific geology; 4, Elementary geology, with an appended catalogue of the minerals and rocks in the State collection ; by E. Hitchcock, LL. D. ; 4° ; 831 pages, 55 plates, geological map. 1841. H. I). Eogers, L. Vauuxem,R. C. Taylor, E. Emmons, T. A. Conrad. Eeport on the ornithichnites or footmarks of extinct birds in the New Eed sandstone of Massachusetts and Connecticut, observed and described by Professor Hitch- cock, of Amherst. Am. Jour. Sci., October, 1841, 1st series, vol. 41, p. 165. The report of a committee appointed by the Association of American Geol- ogists to determine if the tracks described by Hitchcock were really tracks and not imitative forms. 1841. S.Borden. Account of a trigonometrical survey of Massachusetts. Trans. Am. Philos. Soc, Phila., new series, vol. 9, p. 33. 1842. J. E. Teschemacher and A. A. Hayes. On the identity of pyrochlore with the microlite of Professor Shepard; 3 pages. Am. Jour. Sci., 1st series, vol. 43, p. 33. 1842. C. U. Shepard. On the want of identity between microlite and pyrochlore; 6 pages. Ibid., p. 116. 1842. C. TJ. Shepard. Washingtonite and phenacite (Goshen). Additional notices of the supposed phenacite of Goshen; 2 pages. Ibid., p. 364. 1842. E. Hitchcock. On a new species of ornithichnite from the valley of the Connect- icut Eiver, and on the raindrop impressions from the same locality (title only). Trans. Assoc. Am. Geol. Nat., vol. 1, p. 63. 1842. William C. Eedfield. Eemarks on Sunderland Triassic fishes. Ibid., p. 65. 768 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. 1842. James G. Percival. Rej)ort on the geology of tlie State of Counecticut; map. New Haveu. 8°. 1843. 0. Lyell. On the fossil footprints of birds and impressions of raindrops in the valley of the Connecticut (abstract); 4 pages. Proc. Geol. Soc. London, vol. 3, p. 274; also Am. Jour. Sci., 1st series, vol. 45, p. 394. 1843. B. Silliman. Ornithichuites of the Connecticut Eiver sandstones and the dinor- nis of New Zealand (correspondence of Dr. Deane, Dr. Mantell, and E. Owen) ; 12 pages. Am. Jour. Sci., 1st series, vol. 45, p. 177. 1843. F. Alger. Identity of lincolnite and heulandite; abstract. Proc. Boston Soc. Nat. Hist., vol. 1, p. 145. 1843. J. E. Teschemacher. On the occurrence of the phosphate of uranium in the tourmaline locality at Chesterfield. Jour. Boston Soc. Nat. Hist., vol. 4, 1844, p. 35. 1843. J. E. Teschemacher. Description of the oxide of tin at the tourmaline locality, Chesterfield. Trans. Assoc. Am. Geol. Nat., vol. 1, p. 296. 1843. E. Hitchcock. Descriptionof five new species of fossil footmarks; 10 pp. Ibid., p. 254. 1844. James Deane. On the fossil footmarks of Turners Falls, Mass., Avith 2 plates; 6 pages. Am. Jour. Sci., 1st series, vol. 46, p. 73. 1844. A. A. Hayes. Eeexamination of microlite and pyrochlore; 8 pages. Ibid., p. 158. 1844. Francis Alger. Beaumontite and lincolnite identical with heulandite ; 4 pages. Ibid., p. 233; also in Jour. Boston Soc. Nat. Hist., vol. 4, p. 422. 1844. J. E. Teschemacher. Mineralogical notices; Pyrochlore (microlite). Jour. Bos- ton Soc. Nat. Hist., vol. 4, p. 501. Lon. Edin. and Dub. Philos. Mag., 1844. 1844. Simeon Borden. Topograiahical map of Massachusetts, compiled from astro- nomical, trigonometrical, and various local surveys, made by order of the legislature. 1844. E. Hitchcock. Geological map of Massachusetts, made by order of the legis- lature; scale 5 miles to the inch (on same sheet with maj) above). 1844. E. Hitchcock. Geological map of Massachusetts. Explanation of the newly colored geological maj) of Massachusets; 22 pages; 12°. 1844. E. Hitchcock. Eeport on ichnolithology or fossil footmarks, with a description of several new species, and the coprolites of birds from the valley of Con- necticut Eiver, and of a supposed footmark from the valley of Hudson Eiver; 31 pages, 2 plates (read before the Association of American Geologists and Naturalists at Washingtoji, May 11, 1844). Am. Jour. Sci., 1st series, vol. 47, p. 292. 1844. O.U. Shepard. A treatise on mineralogy; 2d edition; 12°; pp.168. New Haven. (See first edition, 1835.) 1844. E. Hitchcock. Discovery of more native copper in the town of Whately, la Massachusets, in the valley of the Connecticut Eiver, with remarks upon its origin; 2 pages. Am. Jour. Sci., 1st series, vol. 47, p. 322. LIST OF PUBLICATIONS. 769 18-14. J. Uoaiie. On the discovery of fossil footmarks j 9 pages. Am. Jour. Sci., 1st scries, vol. IT, \^. 381. 1844. I'j. Jliti'licock. Itejoimler to the preceding article of Dr. Deaiie; 10 pages. Ibid., p. 390. 1841. J. Deane. Answer to the "Rejoinder" of Professor Hitchcock; 2 pages. Ibid., p. 399. 1S44. E. Ilitchcock. Extract from letter respecting the liucolnite. Ibid,, p. 416. 1844. Francis Alger. An elementary treatise on mineralogy. By William Phillips; Stlied.; 602 pages; Boston. 1845. K. Hitt-hcockl Mount Holyoke. Eeport on celebration of the opening of a road onto the mountain. Franklin Express, Vol. I, March, 1845. 1845. S. L. Dana, M. D. Analysis of coprolites from the New Eed sandstone formation of New England, with remarks by Professor Hitchcock. Am. Jonr. Sci., 1st series, vol. 48, p. 46. 1845. E. Hitchcock. Extract from a letter * * * on fossil footmarks, lincolnite, and letter from E. Owen on great birds' nests of New Holland. Ibid., p. 61. 1845. J. Deane. Description of fossil footprints in the New Eed sandstone of the Connecticut Yalley; plate. Ibid., p. 158. 1845. C. U. Shepard. Eeply to notice of mineralogy with notice of microlite, goshenite, urauite. Ibid., p. 168. 1845. J. E. Teschemacher. Eemarks on uranium and pyrochlore; reply to above on microlite and urauite. Ibid., p. 395. 1845. J. Deane. Notice of new sijecies of batrachian footmarks; 3 pages and cut. Ibid., vol. 49, p. 79. 1845. J. Deane. Fossil footmarks and raindrops (letter). Ibid., p. 213. 1845, J. Deane. Illustrations of fossil footmarks; 8 pages, 1 cut. Jour. Boston Soc. Nat. Hist., vol. 5, p. 277. 1845. J. Barratt. On fossil footmarks iu the red sandstone of the Connecticut Valley, Proc. Assoc. Am. Geol. Nat., p. 23. 1845. J. Barratt. On the evidences of congelation in the N"ew Eed sandstone. Ibid,, p. 26. 1845. C. Lyell. On fossil footsteps of birds on Connecticut Eiver. Hist. Travels in U. S., vol. 1, p. 200. 1846. Note. Washingtonite of Shepard = ilmenite ; analysis. Am. Jour. Sci., 2d series, vol. 1, p. 122. 1846, F. Alger. Eeaffirms his opinion that lincolnite is heulandite (abstract). Proc. Boston Soc. Nat. Hist., vol. 2, p. 89. 1846. F. Alger. Notices of new localities of rare minerals and reasons for uniting several supposed distinct species; lincolnite is heulandite. Washingtonite analysis. Jour. Boston Soc. Nat. Hist., vol. 5, p. 297. 1846, J. Barratt. Sentinel and Witness extra ; Middletown, Conn., July 3, 1846. Geology of Middletown and vicinity (reprint of). On the tracks of large birds found at Middletown, Conn.; by Joseph Barratt, M. D. MON XXIX 49 770 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. 1846. J. E. Tescliemaclier. Damourite in Chesterfield. Proc. Boston Soc. Nat. Hist., vol. 2, p. 107; Am. Jour. Sci., 2d series, vol. 2, p. 119. 1847. James Deane. Notice of new fossil footprints. Am. Jour. Sci., 2d series, vol. 3? p. 74. 1847. E. Hitclicock. Elementary geology; 8tli edition, enlarged; 361 pages. 1847. Editor's note. Ornithichnites. Am. Jour. Sci., 2d series, vol. 3, p. 276, 1847. E. Hitclicock. Description of two new species of fossil footmarks found in Massachusetts and Connecticut, or of the animals that made them; cuts. Ibid., vol. 4, p. 46. 1847. J. Deane. Eossil footprints. Ibid., vol. 4, p. 448. 1847. E. Hitchcock. On the trap tuff or volcanic grit of the Connecticut Yalley, with bearings of its history upon the age of the trap rock and sandstone generally in the valley. Ibid., p. 199. 1848. J. Deane. Eossil footprints of a new species of quadruped. Ibid., vol. 5, p. 40. 1848. B. Hitchcock. Eossil footmarks of United States; 128 pages, 24 plates. Mem. Am. Acad. Arts Sci., new series, vol. 3, p. 129. 1848. Dexter Marsh. Fossil footprints; 3 pages, cut. Am. Jour. Sci., 2d series, vol. 6, p. 252. 1849. J. Deane. Illustrations of fossil footprints of the valley of the Connecticut; 16 pages, 9 plates. Mem. Am. Acad. Arts Sci., new series, vol. 4, p. 209. 1850. J.D.Dana. Spodumene, Norwich; mouoclinic. Am. Jour. Sci., 2d series, vol. 10, p. 119. 1850. J. D. Dana. Staurotide, Norwich. Ibid., p. 121. Was triphylite. 1850. J. G. Brush. On American spodumene. Ibid., p. 370. 1850, E. Hitchcock. On the river terraces of the Co,nnecticut Eiver, and on the ero- sions of the earth's surface. Proc. Am. Assoc. Adv. Sci., vol. 2, p. 148. 1850. C. Hartwell and E. Hitchcock, jr. Description of certain mineral localities, chiefly in the northern part of Worcester and Franklin counties in Massa- chusetts (title). Ibid., j). 159. 1850. J. Deane. Fossil footprints of Connecticut Eiver. Jour. Acad. Nat. Sci. Phila., 2d series, vol. 2, p. 71. 1850. E. Hitchcock. On terraces and ancient sea beaches, especially those of the Connecticut Eiver and its tributaries. Eep. Brit. Assoc, 1850, p. 87. Com- munications. "l850. E. Hitchcock. On the erosions of the earth's surface, especially by rivers. Ibid., p. 85. 1850. William H. Gibbs. An address delivered before the literary association. Bland- ford, Mass., September 21, 1850. Springfield, George O. Wilson. Contains notes on minerals in Blandford, furnished by Dr. Shurtleff, of Westfleld. 1851. W. J. Craw. Chemical examination of a phosphate of iron, manganese, and lithia from Norwich, Mass. ; 2 pages. Am. Jour. Sci., 2d series, vol. 11, p. 99. 1851. J. D. Dana. Physical and crystallographical characters of the phosphate of iron, manganese, and lithia of Norwich, Mass. ; 2 pages. Ibid., p. 100. LIST OF PUBLICATIONS. 771 1851. J.I). I)aiiii(!). Miiieralogical notices No. III. Mineral species described by Prof. C. U. Shepard, New Haven. Proc. Am. Assoc. Adv. Sci., New Haven. Enmauiti'. Am. Jour. Sci., 2d series, vol. 12, p. 211. 1851. J. D. Dana (?). On the crystallogiaphic identity of eiimauite and brookite. Am. Jour. Sci., 2d series, vol. 12, p. 397. 1851. William 0. Red field. On tlie post-Permian date of the red sandstone rocks of New Jersey and the Connecticut Valley, as sliown by their organic remains. Proc. Am. Assoc. Adv. Sci., vol. 5, p. 45. 1852. J. E. Teschemacher. On tlie angles of eumanite. Am. Jour. Sci., 2d series, vol. i:\ p. 117. 1852. E. Hitchcock. On the terraces and sea beaches that have been formed since the drift period, especially those along the Connecticut Eiver. Proc. Am. Assoc. Adv. Sci., vol. 6 (1851), p. 261. 1852. C. U. Shepard. On the triplite (allaudite) of Norwich, Mass. Ibid., p. 234. 1852. C. U. Shepard. A treatise on mineralogy; 3d edition; 451 pages, 488 illus- trations. 1852. E. Hitchcock. On the geological age of the clay slate of the Connecticut Val- ley in Massachusetts and Vermont. Proc. Am. Assoc. Adv. Sci., vol. 6 (1851), p. 299. 1853. Jules Marcou. Geological map of the United States and the British Provinces of North America. Text and profiles. 1853. J. L. Smith and C T. Brush. Eeexamination of American minerals ; Spodumene ; Norwich. Am. Jour. Sci., 2d series, vol. 16, p. 471. 1853. E. Hitchcock. Eeport on certain points in the geology of Massachusetts ; Coals, ancient glaciers. 44 pages, 3 plates. 1853. E. Hitchcock. Eeport on soapstone of Middlefield, Mass.. to Metropolitan Soap- stone Co., of New York; 4 pp. 1853. E. Hitchcock. Eemarks on sandstones and fossil footmarks (abstract). Proc. Boston Soc. Nat. Hist., vol. 4, p. 378. 1853. Scientific intelligence; Tryphyline, Norwich. Am. Jour. Sci., 2d series, vol. 15, p. 445. 1853. Scientific intelligence. Triplite of Norwich, Mass. : reference to Shepard and editor's note. Ibid., p. 445. 1853. C. U. Shepard. Triplite, Norwich. Proc. Am. Assoc. Adv. Sci., vol. 6, p. 234. 1854. J. C. Warren, M. D. Eemarks on fossil impressions in the sandstone rocks of Connecticut Eiver ; 54 pages, 1 plate. Boston. 1854. J. W. Mallet. Analysis of beryl from Goshen. Am. Jour. Sci., 2d series, vol. 17, p. 180. 1854. J. W. Mallet. On phosphate of iron and manganese, from Norwich, Mass. Ibid., vol. 18, p. 33. 1854. W. B. Eogers. Fossils of the New Bed sandstone and its relations to the rocks of Virginia and North Carolina. Proc. Boston Soc. Nat. Hist., vol. 5, p. 18; also Am. Jour. Sci., 2d series, vol. 19, p. 123. 772 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. 1854. T. T. Bouv4. Note on Portland society's fossil footprints. Am. Jour. Sci., 2d series, vol. 19, p. 37. 1854. J. Wyman. On impressions of a doubtful sixth toe in some batracliian foot- prints. Proc. Boston Soc. Nat. Hist., vol. 5, p. 84. 1854. J. C. Warren. Note on ripple-marked slabs from Turners Falls. Ibid., p. 84. 1854. J. C. Warren. Note on slabs with impressions from Connecticut Elver sand- stone. Ibid., p. 209. 1854. W. B. Eogers. Note on Olathropteris. Ibid., 212. 1854. C. T. Jackson. Note on tail traces in Connecticut Eiver sandstone. Ibid., p. 309. 1854. T. T. Bouve and W. B. Eogers. Note on plates prepared by Mr. J. Deane for a proposed work on the fossil imjiressions of the Connecticut Valley. Ibid., p. 348. 1855. C. H. Hitchcock. Impressions (chiefly tracks) on alluvial clay in Hadley, Mass. Am. Jour. Sci., 2d series, vol. 19, p. 391. 1855. E. Hitchcock, jr. Description of a new species of Olathropteris discovered in the Connecticut Valley sandstone. Ibid., vol. 20, p. 22. 1855. J. Wyman. Notice of fossil bones from the red sandstone of the Connecticut Eiver Valley, from East Windsor, Conn. ; Eeptiliau sauroid, but with hollow bones. Ibid., p. 394. 1855. E. Hitchcock. Bones and tracks from Connecticut Eiver sandstone. Ibid., p. 416. 1855. W. B. Eogers. On the age of the so-called New Red sandstones of the United States. Proc. Am. Assoc. Adv. Sci., vol. 8, p. 290. 1855. James Hall. Eed sandstone of the Connecticut Eiver Valley and the proofs of its Oolitic or Liassic age. Ibid., p. 290. 1855. E. Hitchcock. Description of several sections measured across the sandstone and trap of the Connecticut Eiver Valley in Massachusetts. Ibid., vol. 9, p. 225. 1855. E. Hitchcock. Additional facts respecting the tracks of Otozoum moodii on the Liassic sandstone of the Connecticut Valley. Ibid., p. 228. 1856. E. Hitchcock. Discovery of a new species of fossil fish and fossil footmarks from the sandstone of Turners Falls. Am. Jour. Sci., 2d series, vol. 21, p. 97. 1856. E. Hitchcock, jr. A new fossil shell in Connecticut Eiver sandstone. Ibid., vol. 22, p. 239. 1856. W. C. Eedfleld. On the relations of the fossil fishes of the sandstone of Connect- icut and other Atlantic States to the Liassic and Oolite periods (name " Newark Group " proposed for the Triassic sandstones of Alleghany slope). Ibid., p. 357; also Proc. Am. Assoc. Adv. Sci., Part II, vol. 10, j). 180. 1856. B. Hitchcock. Description of a large bowlder in the drift of Amherst, Mass., with parallel striae on four sides. Am. Jour. Sci., 2d series^ vol. 22, p. 397. 1856. E. Hitchcock. Illustrations of surface geology; 155 pages, 12 plates. Smith- sonian Contributions, vol. 9. Second edition published in 1860. LIST OF PUBLICATIONS. 773 1856. Roswell Field. Note on the new web-footed species of track. Proc. Boston Soc. Nat. Hist., vol. 6, p. 10. 1856. J. C. AViii-ren, M. 1), On jiew and renuirkable gigantic fossils and footmarlvS (read before Boston Society of Natural History; vol. 5, j). 298). Daily Traveller, January 24. 1856. E. Ililclicock. Additional facts concerning tracks of Otozoum moodii on Liassic sandstone of Connecticut Valley. Proc. Am. Assoc. Adv. Sci., vol. 9, p. 228. 1856. J. Deaue. On sandstone fossils of Connecticut Eiver; 6 pages, 3 i>lates. Jour. Acad. Nat. Sci. Pbila., 2d series, vol 3, p. 173. 1857. J. W. Mallet. On the rose-colored mica of Goshen. Am. Jour. Sci., 2d series, vol. 23, p. 180. 1S57. E.Hitchcock. Tadpoles' nests; argument from number of jihalanges as to bird nature of Triassic animals. Proc. Boston Soc. Nat. Hist., vol. G, p. 111. 1857. H. F. Walling. Map of Hampden County, Mass., based ui^on the trigonometrical survey of the State; 240 rods to the incb. 1857. E. Hitchcock. Geological map of Hampden County (with the above). 1858. E. Hitchcock. Ichnology of New England (list of works on ichnology to date); 232 pages, 60 plates; 4°. 1858. C. H. Hitchcock. Geological section from Greenfield to Cbarlemont, Mass. Proc. Boston Soc. Nat. Hist., vol. 6, p. 330. 1858. H. F. Walling. Map of Franklin County, Mass., based upon the trigonomet- rical survey of the State; 240 rods to the inch. 1858. E. Hitchcock. Geological map of Franklin County (with the above map). 1859. Dr. Henry T. Bowditch. Life and character of Dr. J. Deane. Boston Medical and Surgical Journal, February 7, 1859. 1859. Who described the bird tracks? Controversy between Dr. Deane and Professor Hitchcock. Four-column article from life and character of Dr. Deane, of Greenfield, by Dr. Henry T. Bowditch. Springfield Eepublican, May 7, 1859. 1859. E. Hitchcock. A half column letter of Mr. W. W. Draper, of Greenfield, who claims to be the first discoverer of footmarks. Ibid., May 21, 1859. 1859. E. Hitchcock. Reply to Mr. Bowditch and defense of claims to priority in dis- covery of footmarks. Four columns, Springfield Eepublican, May 14, 1859; 6 pages; 8°. See Eeminiscences of Amherst College, p. 388. 1859. E. Hitchcock. Catalogue of geological specimens in the State House; 69 pages, Appendix to Sixth Annual Eeport of the State Board of Agriculture. 1860. Eoswell Field. Ornithichnites, or tracks resembling those of birds. Am. Jour. Sci., 2d series, vol. 29, p. 361. 1860. E. Hitchcock. Illustrations of surface geology. 155 pages, 14 plates ; 4° ; 2d edition, Amherst, J. S. & C. Adams; 1st edition in Smithsonian Contribu- tions, vol. 9, 1856. 1860. E. Hitchcock and C. H. Hitchcock. Elementary geology; 31st edition; rewrit- ten ; 430 pages. 774 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. 1860. Eoswell Field. Note on reptilian nature of tracks. Proc. Boston Soc. ISTat. Hist., vol. 7, p. 316. 1860. Eoswell Field. Ornithiclinites. Proc. Am. Assoc. Adv. Sci., vol, 13, p. 337. . 1860. Henry F. Walling. Map of Hampshire County, Mass., based upon the trigo- nometrical survey of the State, the details from actual surveys ; 240 rods to the inch. 1860. E. Hitchcock. Geological map of Hampshire County (with above map). (I doubt if President Hitchcock prepared this map. See page 408.) 1861. J. Deane. Ichnographs from the sandstone of Connecticut Eiver. Boston: Little, Brown & Co. 1861. E. Hitchcock, C. H. Hitchcock, etc. Eeport on geology of Vermont; 2 vols., 4°. (Connecticut Eiver terraces, Bernardston, Shelburne Falls section.) 1861. E. Hitchcock. Eemarks upon certain points in ichnology. Proc. Am. Assoc. Adv. Sci., vol. 14, p. 144. 1861. E. Hitchcock. Additional facts respecting the Clathropteris of Easthampton, Mass. Ibid., p. 158. 1862. J. D. Dana. Fossil larvtB in Connecticut Eiver sandstone. Am. Jour. Sci., 2d series, vol. 33, p. 451. 1862. G. J. Brush. On the occurrence of tryphyline at Norwich, Mass. Ibid., vol. 34, p. 402. 1862. E. Hitchcock. Supplement to the ichnology of New England. Proc. Am. Acad. Arts and Sci., vol. 6, p. 85. 1862. E. Hitchcock. Postscript to above. Ibid., p. 104. 1863. E. Hitchcock. New facts and conclusions respecting the fossil footmarks in the Connecticut Eiver Valley. Am. Jour. Sci., 2d series, vol. 36, p. 46. 1863. E.Hitchcock. Eeminiscences of Amherst College; 412 pages; 8°. Geological map of Amherst and vicinity. 1863. C. T. Jackson and Charles S. Eichardson. Manhan Silver-Ijead Mining Com- pany, Hampshire County, Mass. * * * Geological surveys and reports, March, 1863. Boston: Alfred Mudge & Co., printers, 34 School street; with map. The copy in the library of Amherst College came from Dr. Jackson's library, and has his notes in pencil. 1864. C. IT. Shepard. Miueralogical notices; Tungsten in Chesterfield. Am. Jour. Sci., 2d series, vol. 37, p. 407. 1864. C. T. Jackson. Discovery of emery in Chester, Mass. Proc. Boston Soc. Nat. Hist., vol. 10, p. 84. 1864. Council of American Academy of Arts and Sciences. Obituary of President E. Hitchcock. Proc. Am. Acad. Arts and Sci., vol. G, p. 291. 1865. C. T, Jackson. Discovery of emery in Chester. Am. Jour. Sci., 2d series, vol. 39, p. 87. 1865. 0. U. Shepard. Miueralogical notices (Chester, Whately). Ibid., vol. 40, p. 112. 1865. C. TJ. Shepard. Miueralogical notices (addition to Chester, Whately). Ibid., p. 123. LIST OF PUBLICATIONS. ^75 18«5. E. mtcUcock aud 0. H. Hitchcock. Supplement to the ichnolo.y of New Kno-laiid: 96 pages, 20 plates; 4°. 1805. C. U. Shepard. East Wl.ately mine of yellow ocher and s.enna; 8 page.; 1. , Amherst, Mass., January IS, 1805. ■, ^ ,. U. Shepar.l. A description of the emery mine of Chester, Hampden County, 1865. C. U. Slu'paril Mass., U. S . U. Shepari cfij'ipg vol. "All ]}* -'^^* i..^,,.,'; nornndonhilite of Shepard. Ibid., p. 394. (Analysis.) Mass U S. A.; 10 pages; London. 1866. C. U Shep;'i. Scheelefine at Southampton lead mine. Am. Jour. Sc, 2d 1866 F.Pisani. Corundophilite of Shepard. ^^^^'^ ^'- ^''-^f^^'i .. 860 J Lawrence Smith. On the emery mine of Chester. Ibid vol. 42, p 83. SfiO C T Jackson Analysis of some minerals from Chester. Ibid., p. 107 ISOe' C U sCr"i. Mine'ralogical notices; Cotunnite in Southampton lead mine; Colnmbite, Northfield, Mass. Ibid., p. 240. i«rfi T n THin Corundophilite of C. U. Shepard. Ibid., p. 269. 1800". C. U. She^nd KoteUncerning the minerals of the emery mine of Chester. 1866 C Tji^l Chemical analyses of minerals associated with the emery of Chester Proc. Boston Soc. Nat. Hist., voL 10, p. 320. 1866. C. nlS^hcoc.. Description of a new reptilian ^^f^-^^'^^^Zl^^^^^^^ sachusetts. Annals New York Lyceum, vol. 8, p. 301. (Tarsodactylus 1867 T 7ZZL Facts about peat as an article of fuel (cites Hitchcock and Dr. H." N Lucas's experiments on mixing coal and peat). Private P^bUcation 1867 C U Shepard. On the supposed tadpole nests or imprints made by the Batra- clfdes nidiiicans (Hitchcock) in the red shale of the New Eed sandstone of South Hadley. Am. Jour. Sci., 2d series, vol. 43, p. 99. 1867 J P Cooke, jr. Clinochlore from Chester, Mass. Ibid., vol. 44, p. 20b 1867'. j'. D. dI.' ' Note on the corundophilite of Chester. Ibid pp 258 and 283. 1868 C.U. Shepard. Corundophilite (analysis by Eaton). Ibid , voL 46, p. .ob. 1868 J D Dana. A system of mineralogy; Descriptive mmeralogy 87 : J. H. Adams (a'member of the class of 1870, Amherst Od ege). Jotice of asbestos and corundum with other minerals at Pelham. Am. Jour. Sci., 2d 1870. E.i"c:;r'o'!\LMlgadactyluspolyzelusof Hitchcock; abs^^^^^^^^^^^^ in Transactions of American Philosophical Society. Ibid., p. 390. 1870. C.U. Shepard. Mineralogical contributions (microlite, vermicuhte). Ibid., vol. 1870 C.H.' Hitchcock. The geology of Vermont; 5 pages; 4°; privately printed as part of a proposed geological atlas, but without map. 1871. J.D.Dana. On the Connecticut Eiver Valley glacier. Am. Jour. Sci., 3d series, 1871 W^H.Nnef Peculiar phenomena observed in quarrying (Monson, Mass.). ■ Proc. Boston Soc. Nat. Hist., VOL 14, p. 80; also separate publication. 776 GEOLOGY OP OLD HAMPSHIEE COUNTY, MASS. 1871. A. A. Julieii. Analysis of cymatolite. Am. Chemist, vol. 1, p. 300. 1871. H. F. Walling and O. W. Grey. OfScial topographical atlas of Massachusetts. Stedman, Brown & Lyon, Boston. 1871. C.H.Hitchcock. "Geological description" and geological map of Massachu- setts, in above atlas, p. 17. The map was so incorrectly colored by the pub- lishers that it was publicly repudiated by Professor Hitchcock at the Boston meeting of the American Association for the 'Advancement of Science. 1872. E. Hitchcock. Discovery of the tooth of a mastodon in Massachusetts (Cole- raine). Am. Jour. Sci., 3d series, vol. 3, p. 146. 1873. J. D. Dana. On rocks of the Helderberg era in the valley of the Connecticut, the kinds including staurolitic slate, hornblendic rocks, gneiss, mica-schist, etc., besides fossiliferous limestone. Ibid., vol. 6, p. 339. 1873. J.D.Dana. On the Glacial and Champlaiu eras in New England. Ibid., vol. 5, p. 198. 1873. J. D. Dana. Additional note to above. Ibid., p. 217. 1873. W. H. Niles. Note on movements of rocks in Monson. Proc. Boston Soc. Nat. Hist., vol. 16, p. 41. 1873. H. P. Walling. List of lakes, ponds, and reservoirs in Massachusetts. Appen- dix B, Eeport Massachusetts State Board of Health. 1873. Adolph Knop. Studien liber Stoffwandlungen in Miueralrelche. Leipzig. Serpentine, p. 50. 1874. W. H. Niles. On some expansions, movements, and fractures of rocks observed at Monson, Mass. Proc. Am. Assoc. Adv. Sci., vol. 22, Part II B, p. 156. 1875. E. S. Dana. Trap rocks of Connecticut Valley. Ibid., vol. 23, Part II B, p. 45; also Am, Jour. Sci., 3d series, vol. 8, p. 390. 1875. G. W. Hawes. Trap rocks of Connecticut Yalley (analysis of Holyoke trap). Am. Jour. Sci., 3d series, vol. 9, p. 185. 1875. J. D. Dana. On southern New England during the melting of the great glacier; No. 1, ibid., vol. 10, p. 168; No. 2, ibid., p. 280; No. 3, ibid., p. 353; Sup., ibid., p. 497. 1875. J. P. Cooke, jr. On two new varieties of vermiculites (pelhamite). Proc. Am. Acad. Arts Sci., vol. 10, p. 453. 1875. Report of the water commissioners of the town of Westfleld on the construc- tion of the waterworks, including engineer's (L. F. Root) report; 50 pages; Westfield, Mass. 1875. J. Hall, T. Sterry Hunt, Thomas Doane. Reports upon Hoosac Tunnel, with profile, in report of the corporators of the tunnel. House document No. 9, Mass. legislature. 1876. C. U. Shepard. Hermannolite, a new species of the columbium group. Am. Jour. Sci., 3d series, vol. 11, p. 140. 1876. J. D. Dana. On southern New England during the melting of the great glacier. Appendix Am. Jour. Sci., 3d series, vol. 12, p. 125. 1876. W. O. Crosby. Report on geological map of Massachusetts prepared for Cen- tennial Exposition; 52 pages; 8°; Boston. Map not published. LIST OF PUBLICATIONS. 777 187(J Iteview of above. Am. .lour. ScL, 3d series, vol. 12, p. 459. . ^ ^ , , isic U SbepanL Oontribatious to n,ineval.,,y (private pal.licat.on.) Amherst 'Coll.-e Mavll;5pages. Chester and I'elliam minerals. 1876 C U. Sheiard. Catalogue of minerals found within about 7.5 miles of Amherst ■ Colle-e. Amherst College, May 20; 8 pages; private publication. 1870 W. H. Niles. The geological agency of lateral pressure exhibited by certain movements of rocks. Proc. l^oston Soc. Nat. Hist , vol. 18, p^ 272 ^7.5 C H Hitchcock. Lenticular hills of glacial drift. Ibid., vol. 19, p. 63. 18;6: F. ris^.r Notices mineralogiques (amesite, euchlorite). Comptes Eendus Acad. Sci. Paris, vol. 83, p. 100. 1877. J. D. Dana. Note ou the Helderberg formation of Bernardstou. Am. Jour. Sci., 3d series, vol. 14, p. 379. i iq ^ or.^. «pp 1877. J. S. DiUer. Westfield during the Champlain period. Ibid., vol. 13, p. ^02, see map. ^^ ,. , 1877. A. A. Julien. On aglaite. Bug. and Min. Journ., New York. 1877. C.U.Shepard. Contributions to mineralogy. Amherst ; 8 pages ; private pub- 1877. Cn'^^ffitchcock. Note upon the Connecticut Yalley Helderberg. Am. Jour. Sci., 3d series, vol. 13, p. 313. 1877. C. H. Hitchcock. The geology of New Hampshire, vol. 2; 428 pages; Con- 1877 J rDilfer^ Geology of Westfield and vicinity. Westfield Times and News- ''' Letter vol 36, /ebruary 21, 28; vol. 37, March 7, 14, 21, 28, September 19. 1878. I. C. Eussell. The physical history of the Triassic formation of New Jersey and the Connecticut Valley. Annals New York Acad. Sci., vol. 1, pp. 220-254; published separately. • 1878. C. Doelter. Ueber Spodumen und Petalit. Teschermaks Mm. Mit., n. s., vol. 1, 1879. A. A^ Julien. On spodumene and its alterations, from the granite veins of Hampshire County, Mass. Annals New York Acad. Sci., vol. 1, p. 318; 1 plate. 1879. Above reviewed. Am. Jour. Sci., 3d series, vol. 19, p. 237. 1879. A. A. Julien. Composition of cymatolite from Goshen. Ibid., vol. 17, p. 39». 1879 S L Penfield. Chemical composition of triphylite. Ibid., p. 226. 1879 H F Walling. Some indications of recent sensitiveness to unequal pressures in the earth's crust. (Contour map of Mount Toby and Sugar Loaf.) 1 roc. Am. Assoc. Adv. Sci., vol. 27, p. 190. 1879. Majority and minority reports of committee on permanent protection of town from future floods, with report of Hiram F. Mills, civil engineer. Westfield, Mass. ; 30 pages. , 1879. Louis H. Everts. History of the Connecticut Valley in Massachusetts. 2 vols, Philadelphia. Many geological notes under the town histories. 1880. G. J. Brush and E. S. Dana. Spodumene and the results of its alteration. Am. Jour. Sci., 3d series, vol. 20, p. 257. 778 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. 1881. C.H.Hitchcock. Geological map of the United States, with pamphlet; map 13 by 8 feet. J. Bien, 'Sevf York. 1881. G. W. Hawes. On the mineralogical composition of the normal Mesozoic diabase npon the Atlantic border. Proc. U. S. 'Sat. Mus., vol. 4, p. 129. 1881. Chauncey Stephenson. Local geology, West Worthington. Hampshire Ga- zette, November 22, 1881. 1881. Chauncey Stephenson. Local geology No. 2 (emery and iron). West Worthing- ton. Hampshire Gazette, November 29, 1881. 1882. J. D. Dana. The flood of the Connecticut River Valley from the melting of the Quaternary glacier. Am. Jour. Sci., 3d series, vol. 23, pp. 87 and 179. 1882. B. K. Emerson. The Deerfleld dike and its minerals. Ibid., vol. 24, pp. 195, 270, 349. 1882. W. M. Davis. Triassic trap rocks of Massachusetts, Connecticut, and New- Jersey. Ibid., p. 345. 1882. W. M. Davis. The structural value of the trap ridges of the Connecticut Val- ley. Proc. Boston Soc. Nat. Hist., vol. 22, p. 116. 1883. Albert Williams, jr. Mineral resources of the United States, vol. 1. U. S. Geol. Survey. 1883. G. P. Kunz. American gems and precious stones. Separate publication from above work. (Mineral resources, vol. 1.) 1883. W. M. Davis. On the relation of the Triassic traps and sandstones of the east- ern United States. Bull. Mus. Comp. Zool. Harvard Coll., vol. 7 (Geol. Series 1), p. 251; 3 plates of sections. 1883. R. P. Whitfield. Observations on the fossils of the metamorphic rocks of Ber- nardston, Mass. Am. Jour. Sci., 3d series, vol. 25, p. 368. 1883. G. H. Cook. Annual Report of the State Geologist of New Jersey for 1882. Trenton, N. J. 1883. J. D. Dana. Review of above, including notes on " The origin of the Jura- Trias of eastern North America." Am. Jour. Sci., 3d series, vol. 25, p. 383. 1883. J. D. Dana. Western discharge of the flooded Connecticut. Ibid., p. 440. 1883. J. D. Dana. Phenomena of the Glacial and Champlain periods about the mouth of the Connecticut Valley; that is, in the New Haven region. Ibid., vol. 26, p. 341. 1884. J. D. Dana. Drift and terraces. Papers on the Quaternary of New England, from Am. Jour. Sci., 1871 to 1884, and Mem. Conn. Acad., 1870. (A small edition of the papers cited above, bound in a single volume.) 1884. M. E. Wadsworth. Lithological studies. Mem. Mus. Comp. Zool. Harvard Coll., vol. 11, p. 1; 4°; 8 plates; serpentine, Westfleld, Mass. (pi. 7, fig. 3). 1885. Albert Williams, jr. Mineral resources of the United States, vol. 2. U. S. Geol. Survey. 1885. G. F. Kunz. Precious stones. Separate publication from above work (Mineral resources, vol. 2). 1885. A. G. Dana. On the gahnite of Rowe, Mass. Am. Jour. Sci., 3d series, vol. 29, p. 455. LIST OP ITBLIOATIONS. 779 1885. G. P. MeiTill. The collection of biiilcling and ornamental stones in the Uuited States National Museum; a handbook and catalogue. Report Smith- sonian Institution, 1S85-C, Part II, pp. 277-648; Plates I-IX. 1885. Dwiglit Porter. Keport ou the water power of the region tributary to Long Island Sound. Tenth Census U. S., vol. 10. 1S86. A. Williams, jr. Mineral resources of the Uuited States, vol. 3. U. S. Geol. Survey. 18S6. G. V. Kunz. Precious stones. Separate publication from above work (Mineral resources, vol. 3). 188G. Chauncej^ Stephensou. The talcose slate ledge in western Hampshire (native gold on farm of Austin Geer, in West Worthington). Hampshire Gazette, February 22, 1886. 1886. Samuel H. Scudder. The oldest known insect larva, Mormolucoides articulatus, from the Connecticut Eiver rocks. Mem. Boston Soc. ISTat. Hist., vol. 3, No. 13, p. 431; 40. 1886. W. M. Davis. The structure of the Triassic formation of the Connecticut Valley. Am. Jour. Sci., 3d series, vol. 32, p. 342. 1886. B. S. Dana. Mineralogical notes; Columbite, Northfleld; Diaspore, Chester. Ibid., p. 386. 1886. David T. Day. Mineral resources of the United States for 1885; vol. 3. U. S. Geol. Survey. 1886. Eaphael Pumpelly. Mining industries of the United States ; Corundum, mica, feldspar, quartz. Tenth Census U. S., vol. 15. 1887. B. K. Emerson. Preliminary notes on the succession of the crystalline rocks and their various degrees of metamorphism in the Connecticut Eiver region (abstract). Proc. Am. Assoc. Adv. Sci., 35th meeting, Buffalo, p. 231 ; also Am. Jour. Sci., 3d series, vol. 32, pp. 323 and 324. 1887. B. K. Emerson. The age and cause of the gorges cut through the trap ridges by the Connecticut and its tributaries; Prelim. Notes. Proc. Am. Assoc. Adv. Sci., p. 232. 1887. B. K. Emerson. The Holyoke range of the Connecticut. Ibid., p. 233. 1887. B. K. Emerson. The geology of Hampshire County. W. B. Gay, Syracuse, N. Y. Chapter II in Gazetteer of Hampshire County; also Am. Jour. Sci., 3d series, vol. 32, j). 223. Section on the Glacial lake copied. Am. Jour. Sci., 3d series, vol. 34, p. 404. 1887. N. S. Shaler. Fluviatile swamps of New England. Ibid., vol. 33, p. 210. 1887. J. 'D. Dana. Taconic rocks and stratigraphy, with a geological map of the Taconic region. Ibid., p. 393. 1887. David T. Day. Mineral resources of the United States, vol. 4. U. S. Geol. Survey. 1887, G. F. Kunz. Precious stones. Separate publication from above work (Mineral resources, vol. 4), 780 GEOLOGY OF OLD HAMPSHIRE COUNTY. MASS. 1887. W. O. Crosby. The elevated potholes near Shelburae Falls, Mass. Technology Quarterly, Boston, vol. 1, p. 36. 1888. J. D. Whitney. Names and places, studies in geological and topographical nomenclature (100 copies printed), p. 117. Sugar Loaf an "Eddy-Peak," of Triassic sandstone. 1888. J. D. Dana. On the crystalline limestone and the conformably associated Taconic and other schists of the Green Mountains region. A separate publica- tion of the author's papers upon this subject dated 1873-1882, with separate title and preface. 1888. J. S. Newberry. Fauna and flora of the Trias of New Jersey and the Connec- ticut Valley. Trans. N. Y. Acad. Adv. ScL, vol. 6; Eeview, Am. Jour. ScL, 3d series, vol. 36, p. 70. 1888. W. O. Crosby and Charles L. Brown. Gahnite from Rowe, Mass. Technology Quarterly, vol. 1, pp. 407, 408; also Am. Jour. Sci., 3d series, vol. 36, p. 167. 1888. John S. Newberry. Fossil fishes and fossil plants of New Jersey and the Con- necticut Valley. Mon. U. S. Geol. Survey, vol. 14, 1888; 4°; 96 pages, 26 plates. 1888. O. Luedecke. Ueber Datholit. Zeitschrift fiir Naturwissenschaften, vol. 61, p. 235. Halle. 1889. I. C. Eussell. The Newark system. Am. Geol., vol. 3, p. 178. 1889. C. H. Hitchcock. Eecent progress in ichnology. Proc. Boston Soc. Nat. Hist., vol. 24, p. 117. 1889. W. M. Davis. The structure of the Triassic formation of the Connecticut Valley. Seventh Ann. Eept. U. S. Geol. Survey, p. 461. 1889. W. M. Davis. Topographic development of the Triassic formation of the Con- necticut Valley. Am. Jour. Sci., 3d series, vol. 37, p. 423. 1889. W. M. Davis and Chas. L. Whittle. The intrusive and extrusive Triassic trap sheets of the Connecticut Valley. Bull. Mus. Comp. Zool. Harvard Coll., vol. 16, p. 99. 1890. B. K. Emerson. Porphyritic and gneissoid granites iu Massachusetts (abstract). Bull. Geol. Soc. Am., vol. 1, p. 599. 1890. G. F. Kuuz. Gems and precious stones of the United States. New York, Scientific Pub. Co. 1890. B. K. Emerson. A description of the "Beruardston Series" of metamorphic Upper Devonian rocks. Am. Jour. Sci., 3d series, vol. 40, pp. 263 and 362; map and sections. Eeview in N. Y. Independent, December 4. 1890. E. Nason and G. F. Varney. A Gazetteer of the State of Massachusetts. • Boston. 1890. David T. Day. Mineral resources of the United States for 1888; vol. 6. U. S. Geol. Survey. 1890. W. O. Crosby. The kaolin in Blandford, Mass.; 9 pages. Technology Quar- terly, vol. 3, p. 228. 1890. Samuel H. Scudder. The fossil insects of North America, with notes on some European species. 2 vols; 4°; illustrated. Macmillan «& Co., New York. LIST OF PUBLICATIONS. 781 1891. Jules Miircou. Biotiiapliical iiotieo of Ebenezer Eniiiious. Ami'iician Geolof^ist, vol. 7, !>. 1. 1891. r.. K. Kmersou. On the Trias of Massachusetts, with map; pages. Bull. (ieol. Soc. Am., vol. 2, p. 4')1. 1S91. Ci. P. JMui'iill. Eouks in buildiug and decoration. New York: J. Wiley & Sons. 1891. H. S. Williams. Correlation papers. Devonian and Carboniferous. Bull. SO, U. S. (!eol. Survey. 1892. B. K. Emerson. Proofs that the Holyoke and Deerfield trap sheets are con- temporaneous flows and not later intrusions. Am. Jour. Sci., 3d series, vol. iS, p. 140. 1892. E. S. Tarr. Central Massachusetts moraine. Ibid., p. 141. 1892. I. C. Eussell. Correlation papers. The Newark system. Bull. 85, U. S. Geol. Survey. 1892. C. E. Van Hise. Correlation papers; Archean and Algonkian. Bull. 86, U. S. Geol. Survey. 1892. David T. Day. Mineral resources of the United States for 1889 and 1890; vol. 7. U. S. Geol. Survey. 1892. Lester F. Ward. The plant-bearing deposits of the American Trias. Bull. Geol. Soc. Am., vol. 3, p. 21. 1892. J. D. Dana. Additional observations on the Jura-Trias trap of the New Haven region. Am. Jour. Sci., 3d series, vol. 44, j). 165. 1892. M. M. Mitivier. New footprints from the.Connecticut Valley. Proc. Am. Assoc. Adv. Sci. for 1891, p. 286. 1892. J. F. Kemp. Notes on a granite from Chester, Mass. [should be Becket]. Trans. N. Y. Acad. Sci., vol. 11, p. 129. 1893. C. H. Hitchcock. The Green Mountains anticlinal. Science, vol. 20, p. 328. 1894. B. S. Lyman. Some New Eed horizons. Proc. Am. Phil. Soc. Phila., vol. 33, p. 192. Contains an extremely incorrect geological map of the Massachusetts Trias. 1894. E. Pumpelly, J. E. Wolff, and T. Nelson Dale. Geology of the Green Moun- tains in Massachusetts. Mon. U. S. Geol. Survey, vol. 23; 203 pp., 4o, 23 plates. 1895. C. H. Hitchcock. The Connecticut sandstone group. Sustaining the above name for the Triassic sandstones of the Atlantic coast. Science, n. s., vol. 1, p. 74. 1895. W. O. Crosby. Eeport on serpentinic or verd antique marble in Westfleld. Cited in correspondence of Springfield Eepublican, Feb. 3, 1895. 1895. B. K. Emerson. Serpentine j)seudomorphs after olivine, formerly called salt pseudomorphs (from Middlefleld). Bull. Geol. Soc. Am., vol. 6, p. 473. 1895. B. K. Emerson. Calcite pseudomorphs after salt in Triassic shale. Ibid., p. 473. 1895. B. K. Emerson. Puckering of corundum crystals around allauite (from Pel- ham). Ibid., p. 47. 782 GEOLOGY OF OLD HAMPSHIEE COUNTY, MASS. 1895. B. K. Emersou. The geology of Old Hampshire in Massachusetts. Abstract. Bull. Geol. Soc. Am., vol. 6, p. 473; American Geologist, vol. 16, p. 238. 1896. B. K. Emerson. The Archean and Cambrian rocks of the Green Mountain range in southern Massachusetts. Title, Proc. Am. Assoc. Adv. Sci., vol. 44, p. 149; abstract, American Geologist, vol. 1(5, p. 247. 1895. Anon. Another vein of corundum discovered by a Chester man. Announces the discovery of corundum by Dr. H. S. Lucas a mile east of Middlefleld, and repeats the history of the original discovery at Chester. Springfield Repub- lican, December 12, 1895. 1895. J. Volney Lewis. Corundum of the Appalachian crystalline belt. Trans. Am. Inst. Mining Eug., Atlanta meeting, October, 1895. 1896. E. H. Forbes. On the epidote from Huntington, Mass., and the optical jiroper- ties of epidote. Am. Jour. Sci., 4th series, vol. 1, p. 26. 1896. Dwight Porter. The flow of the Connecticut Eiver. Science, u. s., vol. 3, p. 579. 1896. C. H. Hitchcock. The geology of New Hampshire. Journal of Geology, vol. 4, p. 44. Review of the report of the Second Geological Survey of New Hampshire, and statement of the changes In the classiflcation of the rocks there made, dependent on later study of New England geology; with refer- ence to the Bernardstou sei'iea, Leyden argillite, etc. 1896. New topographical atlas of Hampden County; 33 maps; J. Richards & Co., Main street, Springfield, Mass. 1896. The building stones of Pennsylvania. Appendix Ann. Rept. Pennsylvania State College. Reports building stones of Massachusetts. 1897. B. K. Emerson. Diabase pitchstone and mud inclosures of the Triassic trap of New England. Bull. Geol. Soc. Am., vol. 8, pp. 59-86, pis. 3-9. 1897. Gilbert H. Montague. Fossil bird-track discoveries in the Connecticut Valley. Springfield Republican, November 14, 1897. 1897. Anon. Further finds of tracks at Mount Tom. Ibid., November 15, 1897. 1897. William Orr, jr. Studies in local geology. 1. The trap ridges of Holyoke. Ibid., November 28, 1897. 1897. J. C. Rand. Minerals of Massachusetts. The Mineral Collector, vol. 4, p. 161 &. 24 U.S. GEOLOGICAL SURVEY. MONOGRAPH XXIX. PL. XXIV. MONROE. ROWE. HEATH. ROWE SCHIST. SAVC'Y AMPHIBOLITE. HAWKEY SCHtST OOSHtN SCHIST. AMPHIBOUTE. GOSHEN SCHIST SECTIONS ALONG LINES I TO IV ON GEOLOGIC MAP. 25 U. S. GEOLOGICAL SURVEY. MONOGRAPH XXIX. PL. XXV. WINDSOR ASHFIELD. CONWAY SCHIST. Sea Lgvel ROWE CHESTER SCHIST. AMPHIBOLITE. WORTHINGTON. CHESTERFIELD. VI. PEOMATITE. ROWE SAVOr HAWLEY GOSHEN ^"'9T. „..^^^^„ SCHIST. AMPHrBOLITE. CONWAY SCHrST. LIMESTONE. AMPHIBOLITE. MIDDLEFIELD CHESTER. CHESTERFIELD. WESTHAMPTON. VII. SERPENTINE, Se a Leveli \ \ \ CHESTER AMPHIBOLITE, LIMESTONE. CHESTER. HUNTINGTON. WESTHAMPTON. Sea ROWE CHESTER SAVOY SCHIST. AMPHIBOLITE. SCHIST. GRANITE. CONWAY SCHIST. SECTIONS ALONG LINES V TO VIII ON GEOLOGIC MAP. 26 U. S. GEOLOGICAL SURVEY. BLANDFORD. MONTGOMERY. MONOGRAPH XXIX. PL. XXVI. SOUTHAMPTON. Sea Leval MUSCOVITE GRANITE. SUQARLOAF ARKOSE. CHESTER AMPHiaOLITE. BLANDFORD. RUSSELL. WESTFIELD. X. iTTVn'^^N:^! Sea .■By«yil5£§l^^>^;^^iill PEGMATITE. A\l\\\l; BECKET QNEISS. HOOSAC SCHIST. BECKET QNEISS. HOOSAC SCHIST. . ; "Li.".' ■feJgJJ--. Atr TOLLAND. CHESTER AMPHIBOLITE. GRANVILLE. SAXONITE IN ,.' CHESTER AMPHIBOLITE''. CONWAY SCHIST. SOUTHWICK. XI. '"f "///,'/.'! :::: Lever////// //„- •• ,' :• ■: Sea Levef. rTTrrrr^;,^. PEGMATITE. BECKET QNEISS. WASHINGTON QNEISS. i:!!;!;'ii\i\\\vvxy^a \\v; HOOSAC SCHIST. SAVOY SCHIST. CHESTER AMPHIBOLITE. HOOSAC SCHIST. i Alill HOOSAC SCHIST. PHIBOLITE. SAVOY SCHIST. BECKET QNEISS. HOOSAC SCHIST. SAVOY SCHIST. SAXONITE. MUSCOVITE ORANITE. SUQARLOAF ARKOSE. TOLLAND. GRANVILLE. SOUTHWICK. XII. till til Sea Level, ^" 'Jill i.i XxxMihitinuy. WASHINGTON QNEISS. BECKET QNEtSS. HOOSAC SCHIST. H005 CHESTER AMPHIBOLITE. ■lOVITE GRANITE. BECKET QNEISS. PEGMATITE IN HOOSAC SCHIST. SUQARLOAF ARKOSE, UONGMEADOW SANDSTONE. SECTIONS ALONG LINES IX TO XII GEOLOGICAL MAP. 27 U. S. GEOLOGICAL SURVEY. MONOGRAPH XXIX. PL. XXVII. HALIFAX. GUILFORD. VERNON. CONWAY SCHIST WITH LIMESTONE BANDG. HALIFAX. AMPHIBOLITE. CONWAY SCHIST WITH LIMeSTONE BANDS. LEYDEN ARQILLITE, GUILFORD. LEYDEN. BERNARDSTON. VERNON 0NEIS8. ( DEVONIAN.) GILL, CONWAV SCHIST WITH LIMESTONE BAND AMPHtBOLITE. CONWAY SCHIST, WITH LIMESTONE BANDS. LEYDEN ARQILLITC QtJARTZITE. I QUAHTZITE. ; t ; QUART2ITE. LIMESTONE. MICA SCHIST ; j AMPHIBOLITE.' BERNARDSTON (DEVONIAN,) MICA SCHIST AND AMPHIBOLITE. SHELBURNE. GREENFIELD. MONTAGUE. .^ .^ MON80N GNEISS. HAWLEY AMPHIBOLITE. CONWAY AMPHIBOLITE, QUARTZITE WITH BANDS OF LIMESTONE. SUQARLOAF ARKOSE. SCHIST DIABASE. LO NQM E A DOW ^^^^^511 SANDSTONE. BECKET QNEISS. LEYDEN ARQILLITE BERNARDSTON QUARTZITE, CONWAY DEERFIELD. MONTAGUE. CONWAY SCHIST, WITH LIMESTONE BANDS. AMPHIBOLITE. SUQARLOAF ARKOSE. BECKET QNEISS. BERNARDSTON QUARTZITE. SECTIONS ALONG LINES XIII TO XVI ON GEOLOGICAL MAP. ^ 28 U. S. GEOLOGICAL SURVEY. MONOGRAPH XXIX. PL. XXVMI. DEERFIELD. XVII. r SUNDERLAND. , o HOLYOKEMT. >; i DIABASE. 1^^ ^ .t ! MT. TOBY CONGLOMERATE. LEVERETT. CONWAy SCHIST. Sea Level. BHIMFIELD SCHIST. ROWE BfllMFlELD SCHIST. SCHIST. CHESTER AMPHIBOUTE. 'H'J' HATFIELD. AMHERST. fJ^^ i»->X CONWAY SCHIST. AMPHIBOLITE. LIMESTONE. BRIMFIELD FIBROLlTiC RUSTY SCHIST FULL OF PEGMATITE. WILLIAMSBURG. SOUTH HADLEY. GRANBY. BLACK ROCK DIABASE. MT. HOLVOKE aRKOSE. DIABASE. LONGMEADOW SANDSTONE. . ?__ SECTIONS ALONG LINES XVII TO XIX ON GEOLOGICAL MAP. 29 U. S. GEOLOGICAL SURVEY. MONOGRAPH XXIX. PL. XXIX. SOUTHAMPTON. SOUTH HADLEY. GRANBY. MT. TOM. -^ Saa LeyaLilVjiyli WESTFIELD. SUQARLOAF ARKOSE. WEST SPRINGFIELD. LONQMEAOOW SANDSTONE. CHICOPEE. CRYSTALLINE SCHISTS. XXI. Sea LeveT CRYSTALLINE SCHISTS. WESTFIELD SUQAHLOAF ARI?OS£ LONQMeADOW SANDSTONE. CRYSTALLINE SCHISTS. WEST SPRINGFIELD. SPRINGFIELD. XXII Sea LavefXlI ^1 -v /''l^/ CRYSTALLINE SCHISTS. SUFFIELD, CONN. BUCK HILL. ENFIELD. XXMl; Sea Level' i 1 LONQMEADOW SANDSTONE. 1 A 5 Miles SECTIONS ALONG LINES XX TO XXIII ON GEOLOGIC MAP. 30 U.S. GEOLOGICAL SURVEY. NORTHFIELD. WARWICK. RICHMOND. N. LEYOEN AROILLITE BERNARD8TON MICA 9CHI8T WITH HORNBLENDE BANDS. 6EflNAR0ST0N QUARTZtTE CONWAV SCHIST. fiOWE SiCHIST. BECKET ORANITE. QNEtSS, ROWE ; SAVOY SCHIST.i SCHIST. CHESTER AMPHIBOLITE. CONWAY SCHIST. AMPHIBOLITE : BECKET QNEISS. CHESTER; AMPHIBOLITE, SAVOY SCHIST. MONOGRAPH XXIX. PL. XXX. XXIV. CONWAY CHESTER SCHIST. AMPHIBOLITE. SAVOY SCHIST. NORTH FIELD ORANGE. BERNARD3T0N MICA SCHIST WITH HORNBLENDE BANDS. GILL. ROWE S(i"4''' CHESTER AMPHIBpLITE. S MMJ^P^WL^^^ BECKET QNEISS. ROWEScHIST. ; ■; ^XXVI. AMPHIBOLITE. ROWE Schist.; CHESTER AMPHlfiOLITE. BRIMFIELD SCHIST WITH AMPHIBOLITE, RBROLITIC. LEVERETT. SHUTESBURY. NEW SALEM. XXVII. .^ N ^ v\ \ \ TOV\\ \?Ss^g<^^^ ROWE SCHIST. CONWAY SCHIST. SAVOY BECH SAVOY SCHIST. SCHIST. CHESTER AMPHOBOLtTE. CHESTER AMPHIBOLITE. -- t_- SERPENTINE. SECTIONS ALONG LINES XXIV TO XXVII ON GEOLOGIC MAP. 31 U.S. GEOLOGICAL SURVEY. MONOGRAPH XXIX. PL. XXXI. LEVERETT. SHUTESBURY. NEW SALEM. DANA. XXVI S aa Level SAVOY SCHIST. GRANITE, CHESTER AMPHIBOLITE. CONWAY SCHIST. SAVOY SCHIST. CHESTER 'AMPHIBOLITE. SAVOY sChIST. xxvin. BECKET QNEISS. AMHERST. XXIX. AMHERST SCHIST PELHAM GREENWICH. DANA. Sea CONWAY SAVOY SCHIST. SCHIST. AMHERST. AMHERST SCHIST Sea Level, SAVOY SCHIST PEUHAM QUARTZITE. CHESTER AMPHIBOLITE. CONWAV SCHIST. PELHAM. ENFIELD. GREENWICH. CHESTER AMPHIBOLITE. XXX. S^O.A' W Vj A\ , , ■ 1 1 ■ » . ■ ■ ■ 1 » . \ \ . \ ■ ■ < > ■ 1 . . . ■ . 1 ^_ CONWAY SCHIST. TONALITE. BECKET QNEISS. PELHAM QUARTZITE, PELHAM qUaRTZITE. schiST. CONWAY 3AV0V SCHIST. SAVOY ; SCHIST. CHESTER AMPHIBOLITE. BECKET GNEISS. CHESTER AMPHIBOLITE. LUDLOW. Sea LeveU';:;->';;.Vt/aj?.' SUQARLOAF SAVOY AHKOSE. SCHIST. T^i— tf^^TT^ii (,'' ". I mm??^, BELCHERTOWN. BRIMFtELO SCHIST. PALMER. XXXI BECKET QNEISS. ROWE SCHIST. CHESTER AMPHIBOLITE. SAVOY BHIMFIELD SCHIST. SCHIST FIBROLITIC. ROWE feCHIST. CONWAY SCHIST FIBROLITIC, AMHERST GRANBY. BELCHERTOWN. XXXII. GRANBY TUFF SUGARLOAF. Wmm!mmf^Smm;^rn:tl HOLYOKE DIABASE SECTIONS ALONG XXVIII TO XXXII ON GEOLOGICAL MAP. 32 U. S. GEOLOGICAL SURVEY. MONOGRAPH XXIX. PL.XXXII. SUOARLOAF AflKOSE. CHESTER AHPHlfiOUTE. ROWE 'schist. SAVOY SCHipT. AMPHlBQUTEj QNGI88. BRIMFIECD SCHIST.! CONWAY pCHIST. HARDWICK GNEISS. CHESTER' AMPHieOUTE. SPRINGFIELD. WILBRAHAM. PALMER. XXXIV. Sea Levei;xr-:- '.-"----:: -^-~.--r-r-:'--r-£'-^:->-^'cT^rr^ SUOARLOAF ARK08E. MT. TOBY CONGLOMERATE. CONWAY Ct^STER AMPHIBOLITE. BECKET GNEISS. SCHIST. { SAVOY SCHIST. sikVOY fee HIST. QXiwUK'i SCHIST. R'OWE SCHIST. CHESTER AMPHIBOLITE. GRANITE. CONWAY SCHtST. BRIMFIELD SCHIST. HARDWICK GNEISS. SPRINGFIELD. ^ XXXV WILBRAHAM. MONSON. BRIMFIELD. XXXV. S ea LevflL ■ LONQMEADOW SANDSTONE ■;uqarlo.:f rJGLCiMERATE. CONWAY SAVOY SCHIST. SCHIST. i CHESTER A^S|PHIBOLITE. BECKET GNEISS. LONGMEADOW. XXXVI. HAMPDEN. CONWAY SAVOY SCHIST. I CHESTJER AMPHIBOLITE". SAVOY SCHIST. ROW^ SCHIST. MONSON. SCHIST. DIABASE. BECKET GNEISS. I BRIMFIELD SCHIST HARDWICK CHESTER GNEISS. AMPHIBOLITE. BRIMFIELD WALES. XXXVI ^^:^^^r^\ LONGMEADOW SANDSTONE. SUQARLOAF ARKOSE. MT. TOBY CONGLOMERATE. CONWAY SCHIST. SAVOY SCHIST. CHESTER AMPHIBOLITE. BECKET QNEISS. CHESTER AMPHteOlilTE. \ CHESTER AMfJHIBOLITE. SAVOY SCHIST. SAVOY SCHIST. ; CONWAY'SCHIST. ROWE SCHIST. BECKET GNEISS. CHESTER i HAHDWICK AMPHIBOLITE.; QNEISS. BRIMFIELD SCHIST. BRIMFIELD SCHIST SECTIONS ALONG LINES XXIII TO XXXVI ON GEOLOGICAL MAP. INDEX A. Page. Actinolik'-quurtzito. Pelham and Wilbraham 45-47 Actinolite-tremolite-gneiaa, occurrence of 46 Adams, C. B., cited ou Green Mountain gneiea 67 Albitc, occurrence ol" '^^'^ Albitic granite, occurrence and character of 323-331 crushing of minerals in 329 liydrotbenual cbauges in veins of 32£>-330 Albitic mica schist, areas of 66-76 75 .. 19-30 754 133 6 133 221 liornblendic bands in Algonkian rocks, description of Allauite, occurrence of Alliu,E.S., report on Chester emery Ames, James T., raineralogic Tvork of report on Cheater emery by Amherst, analyses of hornblende schist from . . . Conway schist in 222-225 Amherst Eidge, terrace along 644-649 Amherst schist, correlation of 224 minerals in 224-225 Ami)hibolites, occurrence of 66-177 descriptions of 96-97 Chester series 147-155 derivation from limestones 153, 154 analyses and sections of 167, 168, 195-196, 300-306 Conway schist 189-196 "Whately 192-194 Leverett and Amherst 218-220 Warwick 227-228 Orange 228 metamorphism of 236-237 pyroxenic 243-246 Bernardston series 293-294 porphy ritic character of 304 Amygdaloid al sandstone, description of 435-436 Analyses, amphibolite 167, 168, 195-196, 303 andesine 140 claystones 717 Coles Brook limestone 27 cortlandite 347 diorite 345 emery 125 gneiss - 62 granite 37, 316 Hinadale limestone 26 hornblende schist 221 indianite 140 limestones 26,27,189 Longm eadow sandstone 369 mineral spring waters 750-752 Monson gneiss 62 pitchstone 437 sandstone 369 Page. Analyses, serpentine 84,88, 116-117 tonalite , 336 trap rock ■>' 464 waters of mineral springs J 750, 752 Andesine, analyses 140 Ankerite, occurreuce of f 754 Anorthite, South Hadley \ 485 Anthophyllito, description of 52 occurrence of 754^755 Anthracite, occurrence of 755 Antigorite-serpentine , 98 Apatite, occurrence of 755 Aplite, occurrence of 331 Aragonite, occurrence of 755 Argillite, description of 201-210 quartzite in , 202 Bernardston series 261-262 areas of 272-273 pseudo-glacial strite on 531-532 Artesian wells, records of 380-389 Asbestiform anthophyllit.e 52 Asbestos c[uarry at Pelham, description of 47-54 figures of walls of - - 48, 49 Ashfieldlake, deposits of 601-602 Athol, eastern synclin e in 234-236 metamorphism of amphibolite at 236-237 biotite-muscovit'j-sranite from 316-317 section in 572 Augite, South Had4ey 486 B. Barite, occurrence of 755 Bastite, occurrence of 755 Bastite-serpentlne, occurreuce of 98 Batterson's quarry, South Hadley, dike at 489 Bear Eiver lake, deposits of 600-601 Becket, conglomerate-gneiss at 31-38 gran itoid. gneiss from 36 crushinjr tests of granite from 36-38 Becket gneiss, contact with "Washington gneiss 31-32 Belchertoypn, contact zone in 243-248 section of schists near 244 record of artesian-well boring in 245 descriptioi.s of rocks from 246-248 cortilandito at 346-347 dit/es in 481^82 de,8criptJoji of former lake in 575-577 sfe,ction :a 670 Belc'hertown tonaUte, contact zone around 243-248 Bernardston, table showing succession of rocks near. 258 /Upper Devonian fossils of 259-260 I description of range from South Vernon to 272-282 HemardsJon gneiss, Montague 362-363 / 783 784 INDEX. Page. Bemardsi \ ^ series of DcTonian rocks, discussion of. 253-300 descri^\''ioii of region of 260-261 relation to argillite 261-262 fault in... 265 limestone '.of 265-267 magnetite I'ed iii 267-268 quartzite bed in 268-269 mica-scliist and liovnblendic beds in 270-271, 276-282 feldspathic quartzite of 282-283 beds of J 285 original cbaraclter and metamorpbism of 285-287 petrograpbical ^description of 287-295 Beryl, occurrence oA- 755-756 Eiotite-gneiss, descri5)tion of 44-45,182-183 Eiotite-granite.Cheste-'field 318-322 Biotite-muscovite-granite, occurrence and cbaracter of 314-318 Biotite-quartz-scbiat, Ee\ uardston series 289-290 Black Rock core, Mount iHolyoke, description of 489-404 contact -^vith diabase /figured) 490 Blandford, Hoosac acliist in 73-75 py roxenite in ii 85-90 Blandford, serpentines in. A 85-90, 102-104, 104-108, 111 ampbibolites at .V 96-97 Savoy schist in -V 159 description of biotite-gneW from 182-183 dikes in \ 327 Blue-qnartz gneiss -V 28 Bolton limestone, metamorpbisiw of 155 Boston and Albany Railroad, sed^ion along 71-72 Bo-\vlders, description of \ 559-561 Bowlder trains \j 549-550 Brirafield, cordierite granitite at -i 321-322 garnet-biotite-uorite at \ 345-346 description of former lake in. -A 565-566 Brimfleld station, section at .V 566 Brown, M. A., mineralogic work of . . \^ 6 Brookite, occurrence of i 131 Buckland lake, deposits of 602-603 C. Calcite, occurrence of 756 Calcite and dolomite, pseudomorphs of 383-391 Cambrian (Lower) gneisses 31-65 Camp Meeting cutting, sections at 677-691, 694 junction of clays and sands at 705 Ceruasite, occurrence of 756 Cbalcopyrite, occurrence of , 131 Chamberlain, "W. G., report on Chester etaiery by 133-134 Cbamplain clays, description of 697-721 junction of sands with -, 705-706 structure of 706-707 time occupied in deposition of 707 action of ice on .\ --- 707-709 joints in \. -- 709-711 concretions in \ -- 711-718 fossils of 1 - 718-721 Champlain period, phenomena of .V 562-592 Chandler, C.F., analyses by A 369 Charlemont, mica-schist from iV 162 dikes in quartz veins in I.... 169 Chemical analyses, (^ee Aualj^ses.) \ . Chester, amphibolite and serpentine in i... 7A-156 sections at emery mine in \. . 85,\1-11 history and description of emery bed in \ . 117-147 Savoy schist in i 159,3^0 Page. Chester, sericite-scbist from 162 dikes in 327 Chester amphibolite and serpentine, occurrence and character of 78-156 Chester amphibolite series, description and correla- tion of 147-155 sedimentary origin of 155 Chester emery, mode of formation of 154-155 Chester emery bed, history and description of 117-147 Chester Emery Company, organization andVork of. 121 Chester emery mine, association and paragenesis of minerals at 143-147 Chester Granite Company, quarry stones of. 36 Chester series, extent and character of 149 Chesterfield, hiotite gneiss from 183 oopiier mine at 504 Chiastolite-schist... 209-210 Chicopee shale, occurrence and character of. 370 Chlorite-schist, dikes and quartz veins in 169 Chloritoid, description of 129 Clax>p, O. M., Ttiineralogic work of 7 Clark, J. D., mineralogic work of 7 Clark Hill quarries, Middlefield, granitoid gneiss from 34-36 Clarke, John Mason, cited on character and age of fossils from Bernardston 259-260 Clay and marl deijosits, origin of 459-460 Clays and till, contacts of 701-703 Claystones, analj-ses of 717 Clinochlor, occurrence of 756 Coles Brook, sections at 22, 23 Coles Brook anticline, description of 21-24 Coles Brook limestone, analysis of 27 occurrence and character of 27, 28 College Hill, Amherst, section at 557 Connecticut, origin of name ' 2 Connecticut River, old course of 513-515, 627 terraces of 722-738 oscillations of 733 oxbows of 734 deflection of 734-735 Connecticut River lakes, description of 609-696 Connecticut River sandstone, area of 351-354 summary of history of 495-500 Connecticut River tributaries, deflection of 735 terraces of 736 repulsion of 746-747 Connecticut River "Valley, general geology of 13-14 general description of 9-10 Conglomerate-gneiss, Becket 31-38 Conway, mineral vein at 504 deposits of old lake in 598-600 Conway schists, occurrence and character of 183-201 gneiss beds in 185 subordinate beds in 185-199 limestone beds in 188, 189 ampbibolites in 189-196 protrusion through Leyden argillite in Whately 196-197 cleavage in 199-200 .fossils of 200-201 ■ age of 204-205 Leverett 222 Amherst 222-225 Cook, Helen P., analyses by 84 Copper ores, Hawley schist 171 Cordierite, figured 208 INDEX. 785 Page. CordiorltCKninito, Brimfiohl 321-322 Correlation of rtu-kti, hi-rtiou showing 16-18 Cortlandito, Holchurtown :t4fi-347 iiiiiilyHi'H of 347 Coruuiloplitlito, (loacription of 130 t'oruiuhioi, oi'cnroiu'u of 128, 750 Cotiriilo, fij^uioof luyer of 174 Coys Ilillporpliyriticgniiiiie 319-320 Coucretionti, Cham plain rlnya 711-718 Crosby, W. 0., cited on niarhlo of Westfield 92-93 cited on niica-j^ranitoa 312, 314 Crushiugtosta, granite 36-38 Cuniiuingtonite (rhodonite), occurrence of . - 171, 172, 756, 757 Cuahmana Brook, delta of 640, 641 D. Dana, E. S., cited on Triaasic diabases 408-409 Dana, J. D., cited on origin of limestone fragments in trap 460 Dana, J. D., titles and abstracts of papers on TTpper Devonian rooks 253, 254, 256, 257, 259 Dana, J. D., titles of papers on Pleistocene 508 Datolite, occurrence of 757 Davia, W. M., cited on Cretaceous degradation 8 cited on trap rocks 409-410 Deerfield bed, description of 476 Deertield Kiver and tributaries, description of 597-598 Deerlield River, delta of 634-635 Deertield River lakes, deposits of 595-597 Deerfield sheet of eruptive rock, description of 418-446 normal diabase of 441-443 diopside-diabase of 443-444 Delaney 's quarry, Northampton, section at 470 rocks at 470-473 hollow bomb from 480 Deltas at high level, traces of 605-606 Dennis, L. M., analysis of granite by 36-37 Devonian argillites, pseudo-glacial striae on 531,532 Devonian rocks, "Williams Farm, map and sections. 263-264 Dewey, C, cited on mica-granites 312,313 Diaspore, nature of 129 Diabase, Deerfield and Holyoke 372 dikes of 411-418 alteration of 419-439 contact of sandstone with -439,452,455-456 description of 441-443, 461-464 Deerfield sheet 441-443 limestone inclusions in 452-455 granitic inclusions in 483-488 Diabase amygdaloid, contact witli clayey limestone (figured) 208 Diabase-pitchstone, description of 432-433 Diabase-tuff, occurrence and character of , 369 Dike rocks, description of 324-328 Dikes, Charlemont I(i9 pegmatite 216 diabase 411-418 Diller, J. S., titles of papers on Pleistocene 508 quoted on geology of TTcstfleld and vicinity 654^656 Diorite, Prescott (figured) 208 North Prescott and New Salem 342-345 Leverett Center 34^345 analyses of 345 Diopside-diabase, Deerfield sheet 443-444 Dolomite changing to serpentine, (figured) 106 Dolomite and calcite, pseudomorphs of 389-391 MON XXIS 50 l»age. Drift, upland 535-537 valley 537-543 . Dry Jirook Hill, gorge terrace of 601-662 Druniliiis, (leseriptioii of 543-549 Dunes and wind loess, occurrence of 748-749 Dwigbt, Timothy, (luoted 6U9 Dwight station, sections near 609,671 E. Eakins, L. G., analyses by 167, 168, 196, 221, 303, 316, 336, 345, 347 Eastern syncliue, description of *. 234-242 East Greenwich-Enfield syucline 251 Eaton, Amos, cited on occurrence of serpentine at Loudville 190 cited on mica-granites 312 Eights, James, cited on plants of Champlain clays . . 718 Emery, analyses of 125 varieties of 126-127 map of veins of 136 mode of formation 154-155 Emery bed, Chester, description of 117-147 section 141 association and paragenesis of minerals" at 143-147 Emmons, Ebenezer, early mineralogic work 4 cited on mica-granites 312 Enfield, rocks in 232-233 Enfield- Greenwich basin .- 9 Enstatite, formation of 148-153 altered to serpentine (figured) 106 occurrence of 757-758 Enstatite-serpentines, Granville and Russell 90-92 Russell 111-112 Enstatite-serpentine and limestone complex at "West- field marble quarry 147-155 Enstatite-serpentine pseudomorphs in white marble (figured) 152 Epidote, description of 130 occurrence of 758 Epidote-fibrolite, Northfield 328 Epidote-gneiss, Pelham 54 Erratics, description of 559-561 Eruptive rocks, enumeration of 14 description of 307-350, 407-501 contact effects of 349-350 Eruptions, epochs of 410,411 P. Fall River, fault at mouth of , 439, 440 old course of 621, 622 Faults, descriptions of 95-96 Feldspathic quartzite, Eernardston series 282-283 Fibrolite, occurrence of 229-758 Fibrolite-schist, Belchertown 246-248 Field, Roswell, early geologic work 6 Fishes, Triassic 398-400 Fitts, F. H., analyses by 26,336 Florence, analyses of granite from 316 Flynt's quarry, Monson, gneiss at 59-65 Foot tracks and trap sheets, possible connection between 379 Fort River, old oxbow of 737-738 Pleistocene beetles of 740-746 Fossils of the Terrace period 738-740 Fox Brook, Triassic sandstone outcrop along 271 786 INDEX. Page. Franklin County, Eowe schist in 76 K-owe serpentine in 79-80 G. Gadrite, occurrence of 758 Galena, occurrence of 758 Garnet, Northfield 106,328 Garnet-biotitc-norite, Brirufield 345-346 Garnetiferous quartzite, figure of 174 Glacial action in Triassic time 363-364 Glacial grooves and strise 522-531 Glacial notclies 529-531 Glacial period, erosion during 515-517 topography during 518-521 Glacial and Triassic periods, interval between 508-517 Glass in trap, origin of 437-439 Glass-breccia, description of 433-435 Gneiss, Monson 15, 41-45, 56-65 Hinsdale 20,24^25 Lee 20,29-30 Washington 20 blue-quartz 28 Lower Cambrian 31-65 Middlefield 34-36 Becket 36 Shelburne 38^1 Pelham 43-44 Orange 56-65 Goessmann, C. A., analysis by ^. 750 Goshen, limestone at 191 dikes in 326-327 galena at 504 Goshen anticline 175-176 Goshen schists 177-183 Granby, cores and dikes in 482-483 Granhy Plain, moraine across 664 Granby Eoad Lake, description of deposits of 587 Granby tuff, occurrence and character of 369 Granby tuff bed, description of 476-479 source of material of 480 Granite, Becket, crushing tests of 36-38 analysis of 37, 316 Hardwick 317-318 Huntington, crush ing of minerals in 329 age of 348 genetic relations of 348-349 included in diabase 483-488 Granitite, occurrence of 317-322 Granitoid gneiss, Middlefield 34-36 Becket - 36 Pelham 43-44 Granville, Hoosac schist in 73-75 enstatite-serpentines in 90-92 serpentines at 108-111 deposits of former lake in 593 Graphitic mica-schist series 177-210 Greenfield, altered diabase in 419-439 exposures in quarry at 424-431 details of trap ridge east of 426 thin sections from ' ' ash bed " at 430 mineral vein at 505 terrace In 632-634 Greenwich -Enfield basin 9 Green River glacier, deposits of 630-631 Gulf road, sections on 213-215 H. Page. Hadley Lake, deposits of 629-657, 673-677 drainage of 584-586 sections of beds of 646, 647 clays in 698-701 Hadley Lake basin, terraces in 726-729 Hampden Emery Company, organization and work of 121 Hampden County , am pliibolite and serpentine in 85 Rowc schist in 76-78 Hampshire County, Kowe schist in 76-77 former area of 1 serpentine in 81-85 Hardwick gneiss 239-241 Hardwick gneissoid granite and grauitite 317-318 Hassam Brothers, report on Chester emery 1 33 Hatfield, mineral vein at 505-506 section of clays in 691-692 Hausmann, early description of kieselspath (albite) by 5 Hawes, G. "W"., cited on Triassic diabases 408-409 analyses by 463-464 Hawley , great fault in 172 Hawley schist, occurrence and character of 163-171 possible igneous origin of 169 mineral deposits in 170-171 copper ores in 171 Hayes, S. D., analysis by 750 Heath, pyroxene-schist from 163 ampbiboli te from 168 Hematite, South Hadley 486,487 Hillebrand, W. F., analyses by 88 Hinsdale, rocks in 19-24 Hinsdale gneiss, occurrence and character of 20, 24-25 Hinsdale limestone, occurrence and character of. . . 20, 25-27 analysis of 26 Hitchcock, C. H., titles and abstracts of papers on Upper Devonian rocks, by 253,254, 255-257 cited on relation of limestone to quartzite in Ber- nardston series 286 cited on recent progress in ichuology 400-404 Hitchcock, Edward, early geological work of 3-6 cited on metamorphism of mica-schist 67-68 analyses by 188-189,463,464 titles and abstracts of papers on Upper Devonian rocks by 253-255 cited on mica-granites 312-313 cited on occurrence of syenite 331-334 cited on Triassic fossils 394-398 cited on trap rocks 407-408 titles of papers on Pleistocene 508 cited on topography of Connecticut Valley 510-511 cited on glacial notches 530 cited on notable bowlders 559-560 cited on muck sand of Sunderland 728 Holyoke, record of artesian well at 383-385 high terrace near 662-663 Holyoke dam. crushed baud at 370-372 section of 371 Holyoke Eange, description of 10-11 trap rocks in 365-367 Holyoke trap sheet 446-460 diabase of 461^64 fissured quartz deposit in 752-753 Hoosac fault, notes on 95-96 Hoosac schist, occurrence and character of 66-76 INDEX. 787 llooeac ai'liist, hcirnlili'nilic banila in 75 Huiisai' Tnmicl, I'luill iit c-nst i«ivM lit 80-81 lIcMiililimlo sihi.st, uiiiilj 80S 221 l!()nililiinlio Imiuls in llooano Bi^lii.it 75 lliinililoiulii' liwln. licrnarilstou series 270-282 IIiisriMil, l!.,inim'ni1o;;ii' worlc of 5 Uul.liara, A.o., cilrd ^ lluilson iiml Chi'sUTGraiiito Conirany, lieokot, tests of t;raiiite of 36-38 Hunt, David, early niincralogic wiirk of 4 Uiintingtim, mineral vi'in at 51'7 IIyilr(nnica-8c:lii8t 7G-78 I. Ice, Triasslc 363-364 Ice barriers, positions of 565 Ii-lmoli.gy, Triassio rocks 400^04 Inilianite, description of 130 analyses of l*" Insects, Triassic 398 Intrusive rocks in Savoy schist 163 Irving station, section near 217 J. Jackson, C. T., cited on Chester emery bed 119-120 cited on character of emery at Iforth Mountain. 138 cited on occurrence of emery at South Moun- tain 138-139 cited on occurrence of andesine at South Moun- tain , 140 analyses by 140 cited on width of emery bed at South Mountain . 141 Julien, A. A., cited on mica-granites 312,314 cited on tourmaline-spodumene dikes 324,325,326 cited on meteoric alteration of rocks 330 K. Kemp, J.F., cited on Becket granite 36-38 Kettle-holes, occurrence of 664-672 Kibbe quarry. East LongmeadOTi', analyses of rock of 369 King, "W., cited on joints in clays 710 L. Labradorite, South Hadley 485 Laidley, T.T.S.. report on Chester emery 134 Lake bottoms and terraces, descriptions and sec- tions of 672-096 Lazulite, occurrence of 758 Lee gneiss, occurrence and character of 20, 29-30 Leverett, Conway schist in 222 mineral vein at 506 description of former lake in 584-586 Leverett Center, amphibolite and mica-schist series in 220 diorite 344-845 Leverett- Amherst area, description of 218-225 Leyden argillite, protrusion of limestone of Conway schist through 196-197 description of 201-210 stratigraphy of 203 boundary on Conway schists 203-204 age of 204-205 contact metamorphism of 205-210 change tochiaatolite-achist 208 Lily Pond, GUI, terrace at 724-725 Paga Limestone, Ilinsdah' 20,25-27 aniilyses ol 26,27,188 Cidos Brook 27-28 jiyroxi-nit^ 1G3 Conway schists 188-180 Wliateiy 191 mclaniorphosed by granitito 197-199 fossiliferous, description , 362-271 Hernardston series 265-267,289-200 I'MU-nurdston 294-295 included in diabase 452-455 included in trap 456-459 Lithophysfl3, occurrence of 436 Little Mountain, Northampton, trap sheet at 466 Locks Pond Lake, description of 556 Loess, occurrence of 748-749 Longmeadow sandstone, occurrence and character of 364-369 analyses 369 Loudville, mineral vein at 502-504 Lower Cambrian gneisses 31-65 Lower Silurian sericite schists and amphibolites, discussion of 66-177 Lucas, H. S.,mineralogic work of 6 cited on Chester emery bed 118, 1 20-121 Lyman, Benjamin Smith, cited on New Eed horizons . 446 M. Magnet (The), a notable bowlder, description 559 560 Magnetite, occurrence of 127-128 deposits of 172-174,175 Bernardston series 267-268 South Hadley 487 Magnetite-emery bed, Chester 117-147 ICarble, Westfleld 92-95 stellate (dgnred) 152 Margarite, description of - 129 Marl and clay deposits, origin of 459-460 Masonite, description of 129 Maynard quarry, analyses of rock from 369 Mayr, Charles, analysis by 752 Meade, William, early mineralogio paper by 3 Meriden, thin sections of * ' ash bed "at 430 analysis of pitchstone from 437 Merrill, G-eorge P., cited on cost and strength of Tri- assic sandstone 394 Mesozoic time, erosion in 515-517 Metamorphism, Bernardston series 285-287 Meteoric alteration of rocks 330 Mica and amphibolite, Bernardston series 291-293 Micaceous quartzite, description of 46 Mica-granites, historical notes on 312-314 Mica-schist, relation of Becket gneiss to 72-73 description of 162 South Orange and New Salem 231-232 Ware 238-239 Bernardston series 270-271 , 276-282, 291-293 Northfleld 285 Mica-schist series, description of 177-210, 218-220 Hiddlefield,Ho03ao schist in 70-73 serpentine in 81-85 porpbyritic grauitite in 318-319 Mill Ei ver, Northampton, section of drift at 540 delta of 637-639 Millers Falls, dikes near 412,413 Millers Falls station, section near 666-668 Millers Eiver, rocks at mouth of ^95-299 788 IKDBX. Page. Millers Kiver delta, description - 625-629 Mineral deposits, Hawloy schist 170, 171 Mineral lexicon of Franklin, Hampsliire, and Hamp- den counties, aiipplement to 754-761 Mineral springs, locations of 749-752 analyses of waters of 750-752 Mineral veins, description of 502-507 Monroe, Hoosac schist in r 67-70 Monson, gneiss in 15, 41-45, 56-65 granite quarries at 60-65 rocks in 241-242 analyses of gneiss from 316 dikes in 414-415 glass-hearing dikes in 616-418 description of esker in 566-567 lake deposits in 567-569 Monson gneiss, description of 15,41-45 analyses of.. 62 strengt h of 63 expansion in quarrying 63-65 conglomerate structure in 63-65 mineral vein in 65 Monson syncline 249-250 Montague, Bernardston gneiss at 362-363 lake deposits in 615-629 clays in 697-698 terraces in 725-726 Moore's quarry, Florence, analj'ses of rock from 316 Moraines and bowlder trains 549, 550 Mount Holyoke, analyses of trajj from 464 lakehonchon 649-650 Mount Toby conglomerate, occurrence and character of 358-363 Mount Tom, faults at 449-451 lake bench on 640-650 Mount "Warner, bench around 648-649 Muscovite-granite, occurrence and character of 322-323 Muscovite-schist, description of 181-182 Nash, A., mineralogic work of 4-5 cited on mica-granites 312,313 Kewberry, J. S., cited on Triassic fishes 398-399 Newell, "William, mineralogic work of 5 New Salem, serpentine in 55,56 great central syncline in 230 diorite 342-345 Niles, "W. H., cited on expansion of Monson gneiss.. 64-65 North Amherst, granite at 323 breccia at 363 Northampton, record of artesian well at 385-388 trap sheet in 466 dikes in 494-495 mineral vein at 506-507 section of diift at 540 Northfield, description of semi-syncline in 212-216 quart'zite in 284 mica-schist in 285 North. Granville lake, deposits of 593-594 North Leverett, rocks in 219-220 North Prescott, diorite 342-345 Northerner {a notable bowlder), description of 559 O. Olivine, occurrence of 758 Olivine-enstatite rock 52 Page. Orange, gneiss in 56-65 great central syncline in 227-230 eastern syncline in 234-236 Ordway, John M., letter on Monson granite 62-63 Orr, William, jr., analyses by 336 Osborn soapstone quarry, Blandford, section at 87 analysis of serpentine from 88 rocks at 102-104 Ottrelite, description of 129 Owen, Richard, cited on Triassic reptiles 405 P. Paleontology, Triassic rocks 394^406 Palmer, rocks in 241 former lake in...' 569 Paragenesis, secondary minerals, Deerfield sheet 44 4 44 5 Peaked Mountain, section near 249-25U Pegmatite, occurrence and character of. 322-323, 328 Pegmatite dikes and minerals 216, 323-331 Pelham, gneiss in 42-45 asbestos quarry at 47-54 figures of walls of asbestos quarry at 48, 49 serpentine from 55 diabase in 413 microscopic diabase dike from 416-417 section in 578 Pelham and Wilbrahani, actinolite-quartzite of 45-47 Pelham lake and esker, description of 578-584 Pelham-Shutesbury syncline, description of 225-227 Peru, blue quartz of 28 Pitchstone, analysis of 437 Pitchstone breccia, alteration of diabase to 419-439 Plagioclase-feldspars - 52-54 Plainfield, limestone at 192 Plant remains, Champlain clays 718-720 Plants, Triassic 394-398 Pleistocene period, phenomena of 508-517 Pleistocene heetles of Fort Eiver, description of 740-746 Porphyritic granite, occurrence of 319-320 Porphyritic grauitite, occurrence of 318-319 Posterior trap sheet, description of 464-476 Pot-holes, occurrence of 532-533 Prescolt, rocks in 232-233 Prochlorite, occurrence of 759 Publications on geology and mineralogy of Franklin, Hampshire, and Hampden counties, list of . . 762-782 Pyrite, occurrence of 170-171, 759 Pyrolusite, occurrence of 759 Pyroxene, occurrence of 759 Pyroxene- schist, description of 163 Pyroxene-serpentine, Blandford 104-108 Pyroxenic limestone, description of 163 Pyroxenic amphibolites 243-245 Pyroxenite, Blandford 85-90 Q- Quartz, occnrrence of 169, 752-753, 759 Quartz-diorite, occurrence and character of 331-342 Quartz-gab bro, occurrence of 331-342 Quartz-garnet rock, figure of 174 Quartz veins, Charlemont 169 Quartzite, Shutesbury 46 "Warwick 227-228 Orange 228 Bernardston series 268-269,287-290 areasof 273-276 Northfield 284 INDEX. 789 Page. 405-400 7-10-747 -172, 759 500-501 473-474 559 559 560 98-101 76-78 77 158 79-80 364 90-92 111-113 163 507 E. Ecptiloa, Trinsaio Jtcpulaion uf trilpiitiii-ioa, illustnit iini.s of inicMloiiiti- (ciiiimiinstouiti'), iiciuricncu of 171- Eoiul iiiiitorlal, use of triip aa ]!o:i rill;; lirook, fiiult :lt l;ork Elnni (n notiililo bci-wlilor), (Icsci'ilitioii of Km^U Ori'li, iloscription of ItiK-Uin;; Stoiio (The), doscviption of liowi'. aerpoiitinoa at Howo schist, occurreuce and cliaraitoi- of section •_ - compared with Savoy soliiat Rowo serpcDtine, section of Eusaoll. I. C, cited on action of iie in Triassic time. Eusaell, enBtatitc-aerpentiiica in sorpentinea iu pyroxenic limestone in mineral vein at S. Salilite clianging to ti-emolite, figured 106 Salt, occurrence of '"" Sands, interglacial 550-558 Sandstone, Connecticut Eiver 351-354 amygdaloidal 435-436 contact of diabase w ith 439, 452, 455-456 sills intruded in - 469-470 Savoy scliist, occurrence and cliaracter of . . 156-163, 220-221 intrusive rocks in 163 Saxonite, Monson gneiss 47-56 Schists, enumeration of 15,16 Silurian 211-252 Schlierengiinge, E. Hitchcock's suggestion of the- ory of 334 Scudder, S. H., description of Pleistocene beetles liy. 740-746 Sentinel (The) (a notable ho-wlder), description of.. 500 Sericite-gneiss, Whately 206-209 Sericite-schists and amphibolites, description of 66-177 Sericite-schist, description of 162 Serpentine, Monson gneiss 47-56 Pelham 54,55 ohutesbury 55 Middlefleld 81-85 analysis 34,88,116-117 Chester ' 85 Blandford 85-90 petrographic descriptions of 97-117 Westfteld 92-95,113-114 derivation of 115 occurrence of...: 760 Serpentinization, exauiples of 95-96, 147-148 Shearing, Weatfield marble quarry 148 Shepard, C. U., mineralogic work of ■ 5,7 cited on Chester emery mine 122-135 cited on occurrence of indianiteat Soutli Moun- tain 140 Shellmrne, gneiss at 38-41 rocks of anticline in 162 mineral vein at 505 Shelburne Falls anticline, rocks of 75 Shutesbury, serpentine from 55 section in 230 Siderite, occurrence of 760 Silliman, Benjamin, report by, on Southampton lead mine * 3 Silliman's .Journal, cited on lead mines and veins Silla intruded in sandstone Silurian (Lower) sericite scliists and amphibolites.. Silurian schists, east aide of valley Smith, J. Lawrence, cited on minerals accouipanying emery Smitlis Ferry, dike at Sodom Mountain, serpentine from Savoy scliist at pyroxenic limestone from • • South Hadlcy, record of artesian well at dike rocks in South Orange, great central ayncline in South Vernon, description of range from Bernards- ton to '. Spodumene, occurrence of Springs (mineral) , occurrence of analyses — Springfield basin, terraces in Springfield lake, deposits of Springfield lake bottom, description of clays in Springfield Republican, cited on Monson granite quarry cited on Cheater emery bed quoted on use of Triaaaic sandstone for building purposes State Line fault, Holyoke dam Steatization, "Weatfield marble quarry Stellate marble, figure of Stokes, H. N., analysis by Stria;, glacial Sneas, E., cited Sngarloaf arkoae, occurrence and character of Sunny Valley lake, deacription of Swift Eiver lake, description of Pago. i 409-470 66-177 211-252 131 495 114 159 163 381-382 483-494 230-232 272-282 760 749-752 750-752 720-731 657-665 077 . 701 60-62 120 391-394 370-372 147 152 437 522-531 13 354-358 592 569-575 Talc, occurrence of 760 Taft, John B., reports on Chester emery made to 133-134 Terraces, Connecticut Eiver 722-738 Terraces and lake bottoms, detailed sections of 677-696 Terrace period, Connecticut Eiver, fossila of 738-740 "The Crater," North Blanford, rocks at 86, 101-102 Thomas, Judson, acknowledgments to , 141 Till, description of 533-543 contacts of clays and 701-703 Titanite, occurrence of 761 Tolland, rocks in 24,73-75 Tonalite, occurrence and character of 331-342 analyses of 336 crushing and alteration of 339-342 Topograijhy of the region -. 8-11 Tourmaline, description of 130 occurrence of 761 Tourmaline dendrite, Leeds 316 Tourmaline-spodumene dikes, Cheaterfield 324-326 Trap, limestone fragments iu 456-459 underrolling of 460-461 Trap rocka, Holyoke range 365-367 origin of glass in 437-439 analyses of 464 road-making use of 500-501 Trap sheets and foot track, possible connection be- tween 379 Tremolito changing to serpentine, figured 106 790 INDEX. Tremolite rock, occurrence of 108-110 Treniolltization, Westfieltl marTjle qiiJirry 148 Triassic basin, mode of formation of 373-379 Triassic beds, summary of history of 495-500 Triassic eruptions, three epochs of 410-411 Triassic eruptives, occurrence and character of 407-501 Ti iassic glaciers 363-364 Triassic fossils 394-406 Triassic incks. description of 351-406 general section of 354 sumniary of history of 495-500 Triassic sandstones, thicknesses of 375 architectural use of 391-394 Triassic and Glacial periods, interval between 508-517 Tuir, occurrence of 47G-481 Tutfaceous agglomerate, occurrence of 476-481 Turners Falls, record, of artesian well at 380-381 mineral vein at 505 y. Vernon limestone, areas of 276 Villarsite, occurrence of 758 Wapping, section at railroad cutting in 694-696 Ware, rocks in 237-239 diabase dikes in 414 Ware River Lake, description of 569-573 Warwick, great central aynclinein 227-230 Warwick road, sections near 215-216 Washington gneiss, occurrence and character of 20 Washington gneiss, contact with Becket gneiss 31-32 AVashingtonite, occurrence of 131 Wells, artesian, records of 380-389 Wendell Branch syncline 217-218 West Brook, delta of 635-637 Westfield-Holyoke Railroad, trap filled with lime- stone fragments along 456-459 Page. Westfield, serpentine and marble in 92-95, 1 13-114 WestBeld, sections in 92-94 ^ Westfield Little River, artesian well on 389 Westfield plain, deposits of 650-657 Westfield Paver, deposits of 607-608 Westfield marble quarry, eustatite-serpentino and limestone complex at 147-155 Westfield marble quarry, alteration of rocks at 147-155 Westhampton, mineral veins at 502-504 lead mine at 503-504 deposits of former lake in 594-595 West Hawley, section in 173 Whately, ampliibolite bed at 190, 192-194 carbonaceous limestone at 191 protrusion of Conway schist through Leyden argillitoat 196-197 mineral veins at 504 Whetstone schist, occurrence and cliaracter of 186-187, 220-221 Whitfield, E. P., titles and abstracts of paper on metamorphic rocks 254, 256 Whitmores Ferry, Sunderland, amphibolite at 1£10 -191, 104-195 hornblende-schists at 361-362 Wilbrabam, gneiss in 42-45 Wilbraham syucliue, description of 248-249 Wilbrabam and Pelham, actinoliie-quartzite of 45-47 Williamsburg, mineral veins at 505 deposits of glacial lake in 595 Williams farm, map and sections of rocks at 263, 264. 266 Williams farm section, description of 262-271 Wortliington, amphibolite from 167 Worcester, analyses of rock from 369 Z. Zoisite, occurrence of ■ 761 Zoisite-hematite, Northfield 328 [Monograiih XXIX.] The statute approved March 3, 1879, establishing the United States Geological Survey, contains the following provisions: "The publications of the Geological Survey shall consist of the annual report of operations, geo- locieal and economic maps illustrating the resources and classification of the lands, and reports upon general' and economic geology and paleontology. The annual report of operations of the Geological Survey shall accompany the annual report of the Secretary of the Interior. All special memoirs and reports of said Survey shall be issued in uniform quarto series if deemed necessary by the Director, but otherwise in ordinary octavos. Three thousand copies of each shall be published for scientific exchanges and for sale at the price of publication ; and all literary and cartographic materials received in exchange shall be the property of the United States and form a part of the library of the organization : And the money resulting froin the sale of such x)ublicatious shall be covered into the Treasury of the United States." " .,., . ^ ^ ■ z ■ ^.u z Except in those cases in which an extra number of any special memoir or report has been sup- plied to the Survey by special resolution of Congress or has been ordered by the Secretary of the Interior, this office has no copies for gratuitous distribution. ANNUAL REPORTS. I. First Annual Report of the United States Geological Survey, by Clarence King. 1880. 8°. 79 pp 1 map. — A preliminary report describing plan of organization and publications. II. Second Annual Report of the United States Geological Survey, 1880-81, by J. W. Powell. 1882. 8°. Iv, 588 pp. 62 pi. 1 map. III. Third Annual Report of the United States Geological Survey, 1881-'82, by J. W. Powell. 1883. 8°. xviii, 564 pp. 67 pi. and maps. IV. Fourth Annual Report of the United States Geological Survey, 1882-'83, by J. W. Powell. 1884. 8°. xxxii, 473 pp. 85 pi. and maps. V. Fifth Annual Report of the United States Geological Survey, 1883-'84, by J. W. Powell. 1885. 8°. xxxvi, 469 pp. 58 pi. and maps. VI. Sixth Annual Report of the United States Geological Survey, 1884-'85, by J. "W. Powell. 1885. 8°. xxix, 570 pp. 65 pi. and maps. VII. Seventh Annual Report of the United States Geological Survey, 1885-'86, by J. W. Powell. 1888. ,8°. XX, 656 pp. 71 pi. and maps. VIII. Eighth Annual Report of the United States Geological Survey, 1886-'87, by J. W. Powell. 1889. 8°. 2pt. xix, 474, xii pp., 53 pi. and maps; 1 p. 1., 47.5-1063 pp., 54-76 pi. .and maps. IX. Ninth Annual Report of the United States Geological Survey, 1887-'88, by J. W. Powell. 1889. 8°. xiii, 717 pp. 88 pi. and maps. X. Tenth Annual Report of the United States Geological Survey, 1888-89, by J. W. Powell. 1890. 8°. 2pt. XV, 774pp.,98pl. andm.aps; viii, 123 pp. XI. Eleventh Annual Report of the United States Geological Survey, 1889-90, by J. W. Powell. 1891. 8"^. 2 pt. XV, 757 pp., 66 pi. and maps; ix, 351 pp., 30 pi. and maps. XII. Twelfth Annual Report of the United States Geological Survey, 1890-'91, by J. W. Powell. 1891. 8*^. 2 pt., xiii, 675 pp., 53 pi. and maps ; xviii, 576 pp., 146 pi. and maps. XIII. Thirteenth Annual Report of the United States Geological Survey, 1891-'92, by J. W. Powell. 1893. 8°. 3 pt. vii, 240 pp., 2 maps; x, 372 pp., 105 pi. and maps; xi, 486 pp., 77 pi. and maps. XIV. Fourteenth Annual Report of the United States Geological Survey, 1892-'93, by J. W. Powell. 1893. 8°. 2 pt. vi, 321 pp., 1 pi. ; xs, 597 pp., 74 pi. and maps. XV. Fifteenth Annual Report of the United States Geological Survey, 1893-'94, by J. W. Powell. 1895. 8°. xiv, 755 pp., 48 pi. and maps. XVI. Sixteenth Annual Report of the United States Geologicnl Survey, 1894-'95, Charles D. Walcott, Director. 1895. (Part I, 1896.) 8°. 4 pt. xxii, 910 pp., 117 pi. and maps; xix, 598 pp., 43 pi. and maps; xv, 646 pp., 23 pi. ; xix, 735 pp., 6 pi. XVII. Seventeenth Annual Report of the United States Geological Survey, 1895-'96, Charles D. Walcott, Director. 1896. 8°. 3 pt. in 4 vol. xxii, 1076 pp., 67 pi. and maps; xxv, 864 pp., 113 pi. and maps; xxiii, 542 pp., 8 pi. and maps; iii, .543-1058 pp., 9-13 pi. XVIII. Eighteenth Animal Report of the United States Geological Survey, 1896-'97, Charles D. 1 II ADVEETISEMENT. Walcott, Director. 1897. (Parts II and III, 1898.) 8°. 5 pt. in 6 vol. 1-440 pp., 4 pi. and maps ; i-v, 1-653 pp., 105 pi. and maps: i-v, 1-861 pp., 118 pi, and maps: i-x, 1-756 pp., 102 pi. and maps; i-Xii, 1-642 pp., 1 pi. ; 643-1400 pp. MONOGEAPHS. I. Lake Bonneville, by Grove Karl Gilbert. 1890. 4°. xx, 438 pp. 51 pi. Imap. Price $1.50. II. Tertiary History of the Grand Canon District, with Atlas, by Clarence E. Dutton, Capt., U. S. A. 1882. 4°. xiv, 264 pp. 42 pi. and atlas of 24 sheets folio. Price $10.00. III. Geology of the Comstook Lode and the Washoe District, with Atlas, by George F. Becker. 1882. 4°. XV, 422 pp. 7 pi. and atlas of 21 sheets folio. Price $11.00. IV. Comstock Mining and Miners, by Eliot Lord. 1883. 4°. xiv, 451 pp. 3 pi. ' Price $1.50. V. The Copper-Bearing Eocks of Lake Superior, by Eolaud Dner Irving. 1883. 4°. xvi, 464 pp. 151. 29 pi. and maps. Price $1.85. VI. Contributions to the Knowledge of the Older Mesozoic Flora of Virginia, by William Morris Fontaine. 1883. 4°. xi, 144 pp. 54 1. 54 pi. Price $1.05. VII. Silver-Lead Deposits of Eureka, Nevada, by Joseph Storv Curtis. 1884. 4°. xiii, 200 pp. 16 pi. Price $1.20. VIII. Paleontology of the Eureka District, by Charles Doolittle Walcott. 1884. 4°. xiii, 298 pp. 24 1. 24 pi. Price $1.10. IX. Brachiopoda and Lamellibranchiata of the Earitau Clays and Greensand Marls of New Jersey, by Eobert P. AVhitfield. 1885. 4^^. xx, 338 pp. 35 pi. 1 map. Price $1.15. X. Dinocerata. A Monograph of an Extinct Order of Gigaiitic Mammals, by Othniel Charles Marsh. 1886. 4°. xviii, 243 pp. 56 1. 56 pi. Price $2.70. XI. Geological History of Lake Lahontau, a Quaternary Lake of Northwestern Nevada, by Israel Cook Eussell. 1885. 4'-". xiv, 288 pp. 46 pi. and maps. Price $1.75. XII. Geology and Mining Industry of Leadville, Colorado, with Atlas, by Samuel Franklin Emmons. 1886. i^. xxix, 770 pp. 45 pi. and atlas of 35 sheets folio. Price $8.40. XIII. Geology of the Quicksilver Deposits of the Pacific Slope, with Atlas, by George F. Becker. 1888. 4°. xix, 486 pp. 7 pi. and atlas of 14 sheets folio. Price $2.00. XIV. Fossil Fishes and Fossil Plants of the Triassic Eocks of New Jersey and the Connecticut Valley, by John S. Newberry. 1888. 4°. xiv, 152 pp. 26 pi. Price $1.00. XV. The Potomac or Younger Mesozoic Flora, by William Morris Fontaine. 1889. 4'^, xiv, 377 pp. 180 pi. Text and plates bound separately. Price $2.50. XVI. The Paleozoic Fishes of North America, by John Strong Newberry. 1889. 4°. 340 pp. 53 pi. Price $1.00. XVII. The Flora of the Dakota Group, a Posthumous Work, by Leo Lesquereux. Edited by' F. H. Knowlton. 1891. 4°. 400 pp. 66 pi. Price $1.10. XVIII. Gasteropoda and Cephaloiioda of the Earitan Clays and Greensand Marls of New Jersey, by Eobert P. Whitfield. 1891. 4°. 402 pp. 50 pi. Price $1.00. XIX. The Penokee Iron-Bearing Series of Northern Wisconsin and Michigan, by Eoland D. Irving and C. E. Van Hise. 1892. 4°. xix, 534 pp. Price $1.70. XX. Geology of the Eureka District, Nevada, with an Atlas, by Arnold Hague. 1892. 4° xvii, 419 pp. 8 pi. Price $5.25. XXI. The Tertiarv Ehynchophorous Coleoptera of the United States, by Samuel Hubbard Scud- der. 1893. 4°. xi, 206 pp. 12 pi. Price 90 cents. XXII. A Manual of Topographic Methods, by Henry Gannett, Chief Topographer. 1893. 4°. xiv, 300 pp. 18 pi. Price $1.00. XXIII. Geology of the Green Mountains in Massachusetts, by Eapbael Pumpelly, T. Nelson Dale, and J. E. Wolff. 1894. 4°. xiv, 206 pp. 23 pi. Price $1.30. XXIV. Mollusea and Crustacea of the Miocene Formations of New Jersey, by Eobert Parr Whit- field. 1894. 4°. 193 pp. 24 pi. Price 90 cents. XXV. TheGlacialLakeAgassiz, by Warren Upham. 1895. 4°. xxiv,658pp. 38 pi. Price $1.70. XXVI. Flora of the Amboy Clays, by John Strong Newberry; a Pusthumous Work, edited by Arthur Hollick. 1895. 4°. 260 pp. 58 pi. Price $1.00. XXVII. Geology of the Denver Basin in Colorado, by Samuel Franklin Emmons, Whitman Cross, and George Homans Eldridge. 1896. 4'^. 556 pp. 31 pi. Price $1.50. XXVIII. The Marquette Iron-Bearing District of Michigan, with Atlas, by C. E. Van Hise and W. S. Bayley, including a Chapter on the Eepublic Through, by H. L. Smyth. 1895. 4°. 608 pp. 35 pi. Price $5.75. XXIX. Geology of Old Hampshire County, Massachusetts, comprising Franklin, Hampshire, and Hampden Counties, by Benjamin Kendall Emerson. 1898. 4°. xxi, 790 pp. 35 pi. Price $1.90. XXX. Fossil Medusa;, by Charles Doolittle Walcott. 1898. 4o. ix,201pp. 47 pi. Price $1.50. In preparation: XXXI. Geology of the Aspen Mining District, Colorado, with Atlas, by Josiah Edward Spurr. XXXII. Geology of the Yellowstone National Park, Part II, Descriptive Geology, Pe'trograph y , and Paleontology, by Arnold Hague, J. P. Iddings, W. Harvey Weed, Charles D. Walcott, G. H. Girty, T. W. Stanton, and F. H. Knowlton. XXXIII. Geology of the Narragansett Basin, by N. S. Shaler, J, B. Woodworth, and August F. Foerste. ADVERTISEMENT. Ill XXXIV Tho Glacial Gravols of Maine and their Associated Deposits, Ijy George H. Stone. —The l.atfi- Kxtinct Floras of North America, by .John Strong Newberry; edited by Arthur — Flora of the Lower Coal Measures of Missonri, by David White. —The Crystal Falls Iron-l!earing District of Michigan, by .1. Morgan Clements and Henry Lloyd Snivth ; with a Chapter on th(^ Sturgeon Hiver Tongue, by William Shirley Bayley. — Sanropcxla, by (). C. Marsh. — Steiiusaiiria, iiv O. C. Marsh. —Mrnntollieriida', by O. C. Marsh. ,,,„,. —Flora of tho Laramie and Allied Formations, by 1' rank Hall Kuowlton. BULLETINS. 1 On Hypersthene-Andesite and on Triclinic Pyroxene in Augitic Rocks, by Whitman Cross, with a (Joolog'lcal Sketch of Buffalo Peaks, Colorado, by S. P. Emmons. 1883. 8°. 42 pp. 2 pL 2 (iolcl and Silver Conversion Tables, giving the Coining Values of Troy Ounces of Fine Metal, etc., computed by Albert Williams, jr. 1883. S'^. 8 pp. Price .5 cents. 3 On the Fossil Faunas of the Upper Devonian, along the Meridian of 7b° 30', from Tompkins County', N. Y., to Bradford County, Pa., by Henry S. Williams. 1884. 8°. 36 pp. Price 5 cents. 4. On Mesozoic Fossils, by Charles A. White. 1884. 8". 36 pp. 9 pi. Price 5 cents. 5. A Dictionary of Altitudes iu the United States, compiled by Henry Gannett. 1884. 8^. 325 pp. Price 20 cents. ,„ „ . _ 6 Eleyations in the Dominion of Canada, by J. W. Spencer. 1884. 8°. 43 pp. Price 5 cents. 7 Mapoteca Geologica Americana. A Catalogue of Geological Maps of America (North and South), 1752-1881, in Geographic and Chronologic Order, by Jules Marcou and John Belknap Marcou. 1884. 8'^. 184 pp. Price 10 cents. 8. On Secondary Enlargements of Mineral Fragments in Certain Rocks, by R. D. Irving and C. R. VauHise. 1884. 8'^. 58 pp. 6 pi. Price 10 cents. 9 A Report of Work done in tho Washington Laboratory during the Fiscal Year 1883-84. F. W. Clarke, Chief Chemist; T.M.Chatard, Assistant Chemist. 1884. 8°. 40 pp. Price 5 cents. 10. On the Cambrian Faunas of North America. Preliminary Studies, by Charles Doolittle Waleott. 1884. 8°. 74 pp. 10 pi. Price 5 cents. 11. On the Quaternary and Recent Mollusca of the Great Basin; with Description of New Forms, by R. Ellsworth Call! Introduced by a Sketch of the Quaternary Lakes of the Great Basin, by G. K. Gilbert. 1884. 8°. 66 pp. 6 pi. Price 5 cents. 12. A Crystallographic Study of the Thinolite of Lake Lahontan, by Edward S.Dana. 1884. 8°. 34 pp. 3 pi. Price 5 cents. 13. Boundaries of tho United States and of the Several States and Territories, with a Historical Sketch of the Territorial Changes, by Henry Gannett. 1885. 8°. 135 pp. Price 10 cents. 14. The Electrical and Magnetic Properties of the Iron-Carburets, by Carl Barus and Vincent Strouhai. 1885. 8°. 238 pp. Price 15 cents. 15. On the Mesozoic and Cenozoic Paleontology of California, by Charles A. White. 1885. 8°. 33 pp. Price 5 cents. 16. OntheHigherDevoniauFaunasofOntarioCounty, New York, by John M.Clarke. 1885. 8°. 86 pp. 3 pi. Price 5 cents. 17. On the Development of Crystallization in the Igneous Rocks of Washoe, Nevada, with Notes on the Geology of the District, by Arnold Hague and Joseph P. Iddings. 1885. 8°. 44 pp. Price 5 cents. 18. On Marine Eocene, Fresh- Water Miocene, and other Fossil Mollusca of Western North America, by Charles A. White. 1885. 8". 26 pp. 3 pi. Price 5 cents. 19. Notes on the Stratigraphy of California, by George F.Becker. 188.5. 8°. 28 pp. Price 5 cents. 20. Contributions to the Mineralogy of the Rocky Mountains, by Whitman Cross and W. F. Hille- brand. 1885. 8*^. 114 pp. 1 pi. Price 10 cents. 21. The Lignites of the Great Sioux Reservation; a Report on tho Region between the Grand and Moreau Rivers, Dakota, by Bailey Willis. 1885. 8°. 16 pp. 5 pi. Price 5 cents. 22. On New Cretaceous Fossils from California, by Charles A. White. 1885. 8°. 25 pp. 5 pL Price 5 cents. 23. Observations on the Junction between the Eastern Sandstone and the Keweenaw Series on KeweenawPoint, Lake Superior, by R. D. Irving and T. C. Chamberlin. 1885. 8=. 124 pp. 17 pi. Price 15 cents. 24. List of Marine Mollusca, comprising the Quaternary Fossils and Recent Forms from American Localities between Cape Hatteras and Cape Roque, including the Bermudas, by AVilliam Healey Dall. 1885. 8°. 336 pp. Price 25 cents. 25. The Present Technical Condition of the Steel Industry of the United States, by Phmeas Barnes. 1885. 8°. 85 pp. Price 10 cents. 26. Copper Smelting, by Henry M. Howe. 1885. 8°. 107 pp. Price 10 cents. 27. Report of Work done in the Division of Chemistry and Physics, mainly during the Fiscal Year 1884-'85. 1886. 8°. 80 pp. Price 10 cents. 28. The Gabbros and Associated Hornblende Rocks occurring iu the Neighborhood of Baltimore, Maryland, by George Huntington Williams. 1886. 8°. 78 pp. 4 pi. Price 10 cents. IV ADVERTISEMENT. 29. On tlie Fresh- water Invertebrates of tie North American Jurassic, hyCharlea A. White. 1886. 8*^. 41 pp. 4 pi. Price 5 cents. 30. Second Contribution to the Studies on the Cambrian Faunas of North America, by Charles Doolittle Walcott. 1886. 8". 369 pp. 33 pi. Price 25 cents. 31. Systematic Review of our Present Knowledge of Fossil Insects, including Myriapods and Arachnids, by Samuel Hubbard Scudder. 1886. 8°. 128 pp. Price 15 cents. 32. Lists and Analyses of the Mineral Springs of the United States ; a Preliminary Study, by Albert C. Peale. 1886. 8°. 235 pp. Price 20 cents. 33. Notes on the Geology of Northern California, by J. S.Diller. 1886. 8°. 23 pp. Price 5 cents. 34. On the Relation of the Laramie Molluscau Fauna to that of the Succeeding Fresh-Water Eocene and Other Groups, by Charles A. White. 1886. 8°. 54 pp. 5 pi. Price 10 cents. 35. Physical Properties of the Iron-Carburets, by Carl Barus and Vincent Strouhal. 1886. 8'^\ 62 pp. Price 10 cents. 36. SubsidenceofFineSolidParticlesiuLiquidSjbyCarlBarus. 1886. 8°. 58pp. Price 10 cents. 37. Types of the Laramie Flora, by Lester F. Ward. 1887. 8°. 354 pp. 57 pi. Price 25 cents. 38. PeridotiteofElliottCounty,Kentucky,byJ.S.DiUer. 1887. 8-^. 31pp. Ipl. Price5cents. 39. The Upper Beaches and Deltas of the Glacial Lake Agassiz, by Warren Upham. 1887. 8°. 84 pp. 1 pi. Price 10 cents. 40. Changes in River Courses in Washington Territory due to Glaciation, by Bailey Willis. 1887. 8°. 10 pp. 4 pi. Price 5 cents. 41. On the Fossil Faunas of the Upper Devonian — the Genesee Section, New York, by Henry S. WiUiams. 1887. 8°. 121 pp. 4 pi. Price 15 cents. 42. Reportof Work done in the Division of Chemistry and Physics, mainly during the Fiscal Year 1885-'86. F. W. Clarke, Chief Chemist. 1887. 8". 152 pp. 1 pi. Price 15 cents. 43. Tertiary and Cretaceous Strata of the Tuscaloosa, Tombigbee, and Alabama Rivers, by Eugene A. Smith and Lawrence C. Johnson. 1887. 8°. 189 pp. 21 pi. Price 15 cents. 44. Bibliography of North American Geology for 1886, by Nelson H. Darton. 1887. 8°. 35 pp. Price 5 cents. 45. The Present Condition of Knowledge of the Geology of Texas, by Robert T. Hill. 1887. 8°. 94 pp. Price 10 cents. 46. Nature and Origin of Deposits of Phosphate of Lime, by R. A. F. Penrose, jr., with an Intro- duction by N. S. Shaler. 1888. 8°. 143 pp. Price 15 cents. 47. Analyses of Waters of the Yellowstone National Park, with an Account of the Methods of Analysis employed, by Frank Austin Gooch and James Edward Whitiield. 1888. 8°. 84 pp. Price 10 cents. 48. On the Form and Position of the Sea Level, by Robert Simpson Woodward. 1888. 8°. 88 pp. Price 10 cents. 49. Latitudes and Longitudes of Certain Points in Missouri, Kansas, and New Mexico, by Robert Simpson Woodward. 1889. 8°. 133 pp. Price 15 cents. 50. Formulas and Tables to Facilitate the Construction and Use of Maps, by Robert Simpson Woodward. 1889. 8°. 124 pp. Price 15 cents. 51. On Invertebrate Fossils from the Pacific Coast, by Charles Abiathar White. 1889. 8°. 102 pp. 14 pi. Price 15 cents. 52. Subaerial Decay of Rocks and Origin of the Red Color of Certain Formations, by Israel Cook Russell. 1889. 8°. 65 pp. 5 pi. Price 10 cents. 53. The Geology of Nantucket, by Nathaniel Southgate Shaler. 1889. 8°. 55 pp. 10 pi. Price 10 cents. 54. On the Thermo-Electric Measurement of High Temperatures, by Carl Barus. 1889. 8°. 313 pp., incl. 1 pi. 11 pi. Price 25 cents. 55. Report of Work done in the Division of Chemistry and Physics, mainly during the Fiscal Year 1886-'87. Frank Wigglesworth Clarke, Chief Chemist. 1889. 8°. 96 pp. Price 10 cents. 56. Fossil Wood and Lignite of the Potomac Formation, by Frank Hall Knowlton. 1889. 8°. 72 pp. 7 pi. Price 10 cents. 57. A Geological Reconnoissance in Southwestern Kansas, by Robert Hay. 1890. 8°. 49 pp. 2 pi. Price 5 cents. 58. The Glacial Boundary in Western Pennsylvania, Ohio, Kentucky, Indiana, and Illinois, by George Frederick Wright, with an Introduction by Thomas Chrowder Chamberlin. 1890. 8°. 112 pp., incl. 1 pi. 8 pi. Price 15 cents. 59. The Gabbros and Associated Rocks in Delaware, by Frederick D. Chester. 1890. 8°. 45 pp. 1 pi. Price 10 cents. 60. Report of Work done in the Division of Chemistry and Physics, mainly during the Fiscal Year 1887-'88. F. W. Clarke, Chief Chemist. 1890. 8°. 174 pp. Price 15 cents. 61. Contributions to the Mineralogy of the Pacific Coast, by William Harlow Melville and Wal- demar Lindgron. 1890. 8°. 40 pp. 3 pi. Price 5 cents. 62. The Greenstone Schist Areas of the Menominee and Marquette Regions of Michigan, a Con- tribution to the Subject of Dynamic Metamorphism in Eruptive Rocks, by George Huntington Williams, with .an Introduction by Roland Duer Irving. 1890. 8°. 241 pp. 16 pi. Price 30 cents. 63. A Bibliography of Paleozoic Crustacea from 1698 to 1889, including a List of North Amer- ican Species and a Systematic Arrangement of Genera, by Anthony W. Vogdes. 1890. 8°. 177 pp. Price 15 cents. ADVERTISEMENT. V 6-1. A Rtiport of Work dono in tho Division of Chemistry and Pliysirs, mainly during the Fiscal Year IHSH-'Sil. !•'. \V. ClarUo, CliicrCilicuiist. IWIO. S' \ CO ]i]i. I'l-ico U) cents. t!"). Stiali;;ra])hy of tlie liitMUiinons Coal Field of Pennsylvania, Ohio, and West Virginia, by Israel C. Wliit<>r Wfl. «". 212 p)!. U pi. Price 20 cents. {)(i. On a (Sroiip of Volcanic Kocks from the Tcwau Mountains, New Mexico, and on the Occur- rence of Primary Quartz in Certain Basalts, hy Joseph Paxsou Iddiugs. 1890. 8°. 34 pp. Price 5 cents. 67. The Relations of the Traps of tho Newark System in the New Jersey Region, by Nelson Horatio Darton. 1«I0. 8^'. 82 pji. Price 10 cents. 68. Earthquakes in California in 1889, by James Edward Kceler. 1890. 8°. 25 pp. Price 5 cents. 69. A Classed and Annotated Biography of Fossil Insects, by Samuel Howard Scudder. 1890. 8°. 101pp. Price 15 cents. 70. A Report on Astronomical AVork of 1889 and 1890, by Robert Simpson Woodward. 1890. 8°. 79 pp. Price 10 cents. 71. Index to the Known Fossil Insects of the World, including Myriapods and Arachnids, by Samncl Hubbard Scudder. 1891. 8". 744 i)p. Price 50 cents. 72. Altitudes between Lake Superior and the Rocky Mountains, by Warren Upham. 1891. 8°. 229 pp. Pri^ ;i^^^;^??Maryiand, and Virginia, .,WiJufn...,n..UC.ari.^^^l^O. J'.^ Northwestern Louisiana, hy ■'■'^''l^i^^^.^^'oi&.'^r^^e^'^^^^^ 1B06. 8o. lU pp. Price l^^'^euts^ moraines of the Missouri Coteau and their Attendant Deposits, by James Ed^vard Todd. '^'- ^.'^lj'l:ot^^Fo^^^n^t.Sua.,X>j^.U.Voni^ne. 1896. 8°. 149 pp. 2 pi. Price ^^ °*"u6 Biblio-vraphy and Index of North American Geology, Paleontology, Petrology, and Miner- -- ^'^X^i l^^lH^Sn iS^by^ha^'S; pS l^J^t Astronomer iu Charge "^ -S^^'t:ii^fS,^^U,^3^o;;1^t^:f Method ^^.tod S^tes Geological ^&', 1880^^^^^^^ ^^^'- '• '^'^ l^^'' ^'"'^ ^^ '""u9 Biblio-raphy and Index of North American Geology, Paleontology, Petrology, and Miner- ^Wvff.vX Ye rr 1896 by Fred Bough ton Weeks. 1897. 8°. 152 pp. Price 15 cents. alogy ("^t]}!^,^«^»ional Series of^Eock Specimens Collected and Distributed by the United States GeoloiealSurveJ; by Joseph Silas Diller. 1898. 8-. 398 pp. 47 pi. Price 2o cents. Inpress^^ The Lower Cretaceous Gryphicas of the Texas Kegion, by R. T. Hill and T. Wayland ^^'^. A cftalogue ol^t^'be Scerutrnd Te^xtary Plants of North America, by F. H. Knowlton. ^^^^' 153. A Bil.liogmphio Ind;x.of North American Carboniferous InYortebrates, by Stuart Weller. 1898. 8°. pp. Price cents. WATER-SUPPLY AND IRRIGATION PAPERS. By act of Congress approved June 11, 1896, the following provision was inade: „^„ .„„ . ^Provided, That hereafter the reports of the Geolosical Survey in relation to the gauging of streams and to the methods of utilizing the water resources may be prin' ed m octavo form not to exceed onrhundred pa^es in length and live thousand copies m number; ouethousana copies of ^ ycli shall be for the official'use of the Geological Survey, one thousand five hundred copie^ shall he ddiv- ered to the Senate, and two thousand five hundred copies shall be delivered to the House of Repre- sentatives, for distribution." . Under this law the following papers have been issued: 1. Pumping Water for Irrigation, by Herbert M. V ilson. 1896. 8^. 57 pp. 9 pi. 2. Irrigation near Phceuix, Arizona, by Arthur P. Davis. 1897. 8-. 9/ pp. 31 pi. 3. Sewage Irrigation, by George W. Rafter. 1897. f. 100pp. 4pl 4. ARec^onnoissancein Southeastern Washington, by Israel Cook Russell 1897. 8°. 96 pp. 7 p . 5. Irrigation Practice on the Great Plains, by Elias Jranson Cowgill. 1897. 8°. 39 pp. l.^p . 6. Underground Waters of Southwestern Kansas, by Erasmus Haworth. 1897. 8 . 65 pp. l^ pi. 7. Seepage Waters of Northern Utah, by Samuel Fortier. 1897 8^-. 50 pp. dpi. 8. Windmills for Irrigation, by Edward Charles Murphy. 1897 S'^. 49 pp. 8 pi. 9. Irrigation near Greeley, Colorado, by David Boyd. 1897. 8^-90 pp. 21 pi. 10. Irrigation in Mesilla Valley, New Mexico, by F. C. Barker. 1898. 8°. ol pp. 11 pi. 11. River Heights for 1896, by Arthur P. Davis. 1897. 8°. 100 pp. ^"^"'l2.'Water Resources of Southeastern Nebraska, by Nelson H.Darton. 1898. 8°. 55 pp. 21 pi. 13. Irrigation Systems in Texas, by William Ferguson Hutson. 1898. 8. .67 pp. 10 pi. 14. New Tests of Certain Pumps and Water-Lifts used m Irrigation, by Ozni P. Hood. 1889. 8 . ^^15. Operations at River Stations, 1897, Part I. 1898. 8^. 100 pp. 16. Operations at River Stations, 1897, Part II. 1898. 8°. 101-200 pp. TOPOGRAPHIC MAP OF THE UNITED STATES. When, in 1882, the Geological Survey was directed by law to make a geologic map of the United States there was in existence no suitable topographic map to serve a^s a base for the geologic map. The preparation of such a topographic map was therefore immediately begun. About one-hth ot the area of the countrv, excluding Alaska, has now been thus mapped. The map is published in atlas sheets, each sheet representing a small quadrangular district, as explained under the following head- ine The separate sheets are sold at 5 cents each when fewer than 100 copies are purchased, but whmi thev are ordered in lots of 100 or more copies, whether of the same sheet or of difterent sheets, the priie is 2 cents each. The mapped areas are widely scattered, nearly every State being represented. More than 800 sheets have been engraved and printed; they are tabulated by States m tbe burvey s "List of Publications," a pamphlet which may be had on application. vni ADVEETISEMENT. GEOLOGIC ATLAS OF THE UNITED STATES. The Geologic Atlas of die United States is the final form of publication of the topographic and geologic maps. The atlas is issued in parts, progressively as the surveys are extended, and is designed ultimately to cover the entire country. Under the plan adopted the entire area of the country is divided into small rectangular districts (designated quadrangles), bounded by certain meridians and parallels. The unit of survey is also the unit of publication, and the maps and descriptions of each rectangular district are issued as a folio of the Geologic Atlas. Each folio contains topographic, geologic, economic, and structural maps, together with textual descriptions and explanations, and is designated by the name of a principal town or of a prominent natural feature within the district. Two forms of issue have been adopted, a "library edition" and a "field edition." In both the sheets are bound between heavy paper covers, but the library copies are permanently bound, while the sheets and covers of the field copies are only temporarily wired together. Under the law a copy of each folio is sent to certain public libraries and educational institu- tions. The remainder are sold at 25 cents each, except such as contain an unusual amount of matter, which are priced accordingly. Prepayment is obligatory. The folios ready for distribution are listed below. No. Name of sheet. State. Limiting meridians. Limiting parallels. Area, in [Price, square miles, cents. 11 29 1 Livingston 2 Einggold 3 Plaoerville 4 Kingston 5 Sacramento 6 Chattanooga 7 Pikes Peak (out of stock) . . 8 Sewanee 9 Anthracite-Crested Butte . I 10 Harpers Ferry Montana. 'Georgia. Jackson . Eatillville . Fredericksburg . Staunton Lassen Peak Knoxville Marys ville.. Smartsville . Stevenson . Tennessee California Tennessee California Tennessee Colorado Tennessee Colorado (Virginia West Virginia . Maryland Caliibrnia {Virginia ^ Kentucky > Tennessee J /Maryland \ \Virginia J /Virginia \ tWest Virginia . . / California (Tennessee \ \North Carolina . J California California [Alabama ] ^Georgia > iTennessee :J Tennessee Tennessee Tennessee Cleveland Pikeville McMinnviUe ^r ■ • /Marvl.and l^o™"' {Virginia Three Forks Loudon Pocahontas . . Morristown.. Piedmont. Nevada City. /Xellowstone tional Park ity-1 ley- ) ill .) Pyramid Peak . Franklin Brieeville Buckbannon... Gadsden Pueblo Downieville ... Truckee [Nevada City ..< Grass Valley iBanner Hili {Gallatin Canyon... Shoshone. Lake Montana Tennessee Virginia AVest Virginia - . Tennessee (Virginia Maryland "West Virginia.. California Wyoming , California (Virginia tWest Virginia - Tennessee West Virginia Alabama Colorado California California 121° 00' 121° 01' 120° 57' 110°-111° 850-85° 30' 120° 30'-121o 84° 3ll'-850 1210-121° ;io' io°-Ki° 30' 1050-105° 30' 85° 30'-86o 106° 45'-107° 15' 77° 30'-78° 120° 30'-121o 82° 30'-83° 770.770 30' 79°-79° 30' 1210-1220 830 30'-84o 1210 30'-122° 1210-121° 30' 84° 30'-85o 850-850 30' 85° 30'-86° 76° 30'-77° 1110-1120 84°-81° 30' 81°-81° 30' 830-83° 30' 790-79° 30' 25"-121o 03' 45" 35"-121o 05' 04" 05"-121° 00' 25" 120°-120o 30' 79°-79° 30' 84°-84° 30' 80°-80° 30' 86°-86o 30' 104° 30'-105o 1200 30'-121° 120°-120o 30' 450-46° 340 30'-35o 38° 30'-39o 350 30'-36° 38° 30'-39° 35°-3.->o 30' 38° 30'-39° 350-35° 30' 38° 45'-39° 39°-390 30' 38°-38° 30' 36° 30'-37o 380-380 30' 380-380 30' 400-41° 350 30'-36o 390-390 30' 39°-39° 30' 35°-35o 30' 350 30'-36° 36° 30'-36° 38°-38° 30' 450-46° 350 30'-36° 370.370 30' 360-36° 30' 39° 13' 50"-39-) 17' 16" 39° 10' 22"-390 13' 50" 39° 13' 50"-39o 17' 16" 440-45° 38° 30'- -39° 36°-36° 30' 38° 30' -390 340-340 30' 1 38°-38° 30' 39° 30'-40° 39°-39° 30' ,354 980 932 969 932 975 932 975 465 957 938 938 3,634 925 925 925 975 969 969 938 3.354 969 951 963 925 11.65 12.09 11.65 3,412 932 932 963 932 986 938 919 925 25 25 25 25 25 25 25 25 50 25 25 25 25 25 25 25 25 25 25 25 60 25 25 25 50 75 25 25 25 25 25 60 2.-) 25 ADVERTISEMENT. IX STATISTICAL PAPERS. Mineral Rosourcos of the United States [1882], liy Albert WillianiB, jr. 1883. 8'^. xvii, 813 pp. Price 5(1 conts. Mineral KosourccH of tlio Uiutcd States, 1883 and 1884, by Albert WiDiams, jr. 1885. 8". xiv, 1016 pp. Trice (!() cents. Mineral Ixosonrccs (if the United States, 1885. Division of Mining Statistii-s and Technology. 1886. y--'. vii, 576 jip. Trice 40 cents. Mineral Resources of the United States, 1886, by David T. Day. 1887. 8°. viii, 813 pp. Price (50 cents. Mineral Resources of the United States, 1887, by David T. Day. 1888. 8°. vii, 832 pp. Price 50 cents. aiineral Resources of the United States, 1888, by David T. Day. 1890. 8°. vii, 652 pp. Price 50 conts. Mineral Resources of the United States, 1889 and 1890, by David T. Day. 1892. 8°. viii, 671 pp. Price 50 cents. Mineral Resources of the United States, 1891, by David T. Day. 1893. 8°. vii, 630 pp. Price 50 cents. Mineral Resources of the United States, 1892, by David T. Day. 1893. 8°. vii, 850 pp. Price 50 cents. Mineral Resources of the United States, 1893, by David T. Day. 1894. 8°. viii, 810 pp. Price 50 cents. On March 2, 189.5, the following provision was included in an act of Congress: "Provided, That hereafter the report of the mineral resources of the United States shall be issued as a part of the report of the Director of the Geological Survey." In compliance with this legislation the following reports have been published: Mineral Resources of the United States, 1894, David T. Day, Chief of Division. 1895. 8°. xv, 646 pp., 23 pi. ; six, 735 pp.. 6 pi. Being Parts III and IV of the Sixteenth Annual Report. Mineral Resources of the United States, 1895, David T. Day, Chief of Division. 1896. 8°. xxiii, .542 pp., 8 pi. and maps ; ill, 543-1058 pp., 9-13 pi. Being Part III (in 2 vols.) of the Seventeenth Annual Report. Mineral Resources of the United States, 1896, David T. Day, Chief of Division. 1897. 8°. xii, 642 pp., Ipl.; 643-1400 pp. Being Part V (in 2 vols.) of the Eighteenth Annual Report. The report on the mineral resources for the calendar year 1897 will form a part of the Nineteenth Annual Report of the Survey. The money received from the sale of the Survey publications is deposited in the Treasury, and the Secretary of that Department declines to receive bank cheeks, drafts, or postage stamps; all remit- tances, therefore, must be by money oedee, made payable to the Director of the United States Geological Survey, or in currency— the exact amount. Correspondence relating to the publications of the Survey should be addressed to The Director, United States Geological Survey, Washington, D. C, May, 1S98. V^^ashlngton, D. C. (Take this loaf out and paste the separated titles upon throe of your cata- logue cards. The iirst and second titles ueed no additioo ; over the third write that subject under wliioli you would place the book in your library.] LIBRARY CATALOGUE SLIPS. United States. Dejmrtmetit of the interior. ( XJ. S. geological survey.) Department of the interior | — | Monograplis | of tlie | United States geological survey | Volume XXIX | [Seal of the depart- ment] I Washington | government printing office | 1898 Second title: United States geological survey | Charles D. Waloott, director | — | Geology | of | OH Hampshire Coirnty, Massachusetts | comprising | Franklin, Hampshire, and Hamp- den counties | by | Benjamin Kendall Emerson | [Vignette] | Washington | government printing office | 1898 4°. xxi,790pp. 35 pi. C Emeison (Benjamin Kendall). 5 United States geological survey | Charles D. Walcott, di- -< rector | — | Geology | of | Old Hampshire County, Massachusetts I comprising | Franklin, Hampshire, and Hampden counties | hy I Benjamin Kendall Emerson | [Vignette] | . Washington | government printing office | 1898 4°. xxi, 790 pp. 35 pi. [tTNrrED States. Department of the interior. (V. S. geological survey.) Monograph XXIX.] United States geological survey | Charles D. Waloott, di- rector I — I Geology | of | Old Hampshire County, Massachusetts I comprising | Franklin, Hampshire, and Hampden counties | by I Benjamin Kendall Emerson | [Vignette] | Washington | government printing office | 1898 4°. xxi, 790 pp. 35 pi. [TTXTTED States. Department of the interior. (U. S. geological survey.) Mouograpb XXIX.] MOlJ-gnUXSNOp JO S30 VMM3X am JTuittoi iql1|Snoi NOISOdS JO 3DVUH3J. OS Z ^ ; *r Q. !d se i -sfl! ^1 S Mo - : ;^H! ° 'I , o j| IE i-s n EiED^ r=-T'i IMTMix:^ i^fiUiinnjjst^ '-ii" m m^m ml 8 iLlll <-J '"il^Q ^;1 QO' ■~~ ) o ^t idk 1 f J. n t < zS jiCl I? HliiiElD I V) S 4 % 3:^ (1) ^ z s 3 E <0 i^- ir.