HHHHsifiQSSfiBBnoQ iBBB m BBiwBBS ill Hi llliiiiiii khhrhI WW Digitized by the Internet Archive in 2012 with funding from University of Illinois Urbana-Champaign http://archive.org/details/geologyofhardinc41well 3 STATE OF ILLINOIS DEPARTMENT OF REGISTRATION AND EDUCATION DIVISION OF THE STATE GEOLOGICAL SURVEY FRANK W. DE WOLF, Chiet BULLETIN No. 41 THE GEOLOGY OF HARDIN COUNTY and the adjoining part of Pope County By Stuart Weller with the collaboration of Charles Butts. L. W. Currier, and R. D. Salisbury In Cooperation with the United States Geological Survey Printed by Authority of the State of Illinois URBANA, ILLINOIS 1920 STATE OF ILLINOIS DEPARTMENT OF REGISTRATION AND EDUCATION DIVISION OF THE STATE GEOLOGICAL SURVEY FRANK W. DE WOLF, Chief Committee of the Board of Natural Resources and Conservation Francis W. Shepardson, Chairman Director of Registration and Education Kenuric C. Babcock Representing the President of the Uni versity of Illinois Rollin D. Salisbury Geologist Illinois State Journal Co. Springfield, Illinois State Printers 19 2 33465 — 1M TABLES PAGK 1. Classification of rocks in eastern United States and southern Illinois 80 2. Weller's correlation of the Lower and Middle Chester formations of southern Illinois . . . 132 3. Ulrich's correlation of the Lower and Middle Chester formations of southern Illinois . . . 132 4. Production of domestic fluorspar, 1913 to 1917 289 15 LETTER OF TRANSMITTAL State Geological Survey, Ubbana, Apkil 1, 1920. Francis W. Shepardson, Chairman, and Members of the Board of Xatural Resources and Conservation. Gentlemen : I submit herewith a report on the geology of Hardin County, including the adjoining part of Pope County, and recommend its publication as Bulletin 41. Fluorspar mining is the chief mineral industry of the region, and is of national importance, since the output annually is more than two-thirds of the country's total production. During the war period the Hardin County mines were under guard to prevent interruption of work; for the steel industry depends on fluorspar as one of the important fluxes. Naturally, this report has special reference to the fluorspar deposits — their distribution, mode of occurrence, and origin — and presents in the text, and by display on the accompanying map, sug- gestions for prospecting in search of additional veins of commercial importance. The region deserves study, how r ever, for phases of geology and geography ether than those of economic importance. Rock formations of Denovian, Mississippian, and Pennsylvanian age, measur- ing between 3,500 and 4,000 feet, are well exposed along Ohio River or in the dissected upland area of the interior, and offer excellent opportunities for study. This report is a notable contribution to geological knowledge of the Upper Missis- sippian or Chester group, serving to portray with accuracy the stratigraphic suc- cession and the areal distribution of formations which have been established by more than ten years of painstaking study extending from St. Louis to the south- eastern limits of Illinois. The structural geology of Hardin County is also of special interest; the rocks Lave been bowed up into the form of an immense dome, then broken and read- justed by differential settling of the blocks, and intruded by igneous rocks in the form of dikes, sills, and plugs. The rich ores of fluorspar, lead, and zinc occur chiefly along the lines of old fracture planes. On the whole, this report is presented w T ith assurance that the subject matter and the geologic map will be iof great benefit to the economic and cultural develop- ment of the region, and of exceptional interest and value to professional geologists and students. I desire to acknowledge the public service rendered by the several authors, and to thank residents of the district for their constant cooperation. Special credit is due to the U. S. Geological Survey for active participation in the topo- graphic and geologic mapping of the county. Very respectfully, Fraxk W. DeWolf, Chief. PREFACE This report on the geology of Hardin County, Illinois, is based upon, field work which has been carried on by a number of persons. The detailed mapping of the formations has been done by Stuart Weller of the Illinois State Geological Survey, and by Charles Butts and Wallace Lee of the United States Geological Survey, Mr. Weller being responsible for the area south of latitude 37° 30', and Mr. Butts and Mr. Lee for the northern portion of the county. Each of these workers has made repeated visits to the area mapped by the other, and in this work full cooperation has been carried out. Parts of three field seasons have been devoted to the work, the very complex geological structure of the area making it impracticable to complete the mapping more rapidly. In addition to the work by Weller, Butts, and Lee, Mr. L. W. Currier has spent portions of two seasons in the county in a study of the fluorspar deposits which are so characteristic of the area, and the results of his work are incorporated in the chapters on igneous rocks and economic geology. Prof. R. D. Salisbury has contributed a chapter on the geography of the area. The actual writing of the report, except the chapters on igneous rocks, economic geology, and geography, is almost entirely the work of Mr. Weller, although he has been assisted throughout by notes and suggestions supplied by Mr. Butts. In apportioning the credit for the authorship of this portion of the report, certain chapters or parts of chapters are presented under the name of one, and others under the names of both of these men, and where opinions are expressed and conclusions reached in chapters or parts of chapters having the signature of but one author, that author alone is responsible for them. On the geological map that is presented with the report, the area is extended for about five miles westward into Pope County, in order to make more complete the structural area of which Hardin County is a part, and also to make available to the citizens of the State a part of the work which has been completed in Pope County, since the publication of a Pope County map and report must be deferred for several years. The area outside of Hardin County has not been especially described in the text of the report, but the same formations are present that are exposed in Hardin County, and the descriptions are applicable to either county. Stuart Weller CONTENTS PAGE Part I. Introduction, by Stuart Weller, Chas. Butts, and Rollin D. Salisbury. 17 Chapter I.— General introduction, by Stuart Weller and Chas. Butts 19 Location 19 Resources 19 Towns 20 Topography and drainage 20 Chapter II.— Geological introduction, by Stuart Weller and Chas. Butts. 23 Sedimentary rocks 23 Unconformities 26 Geologic time 27 Igneous rocks 29 Geologic structure 29 Dip and strike 31 Topographic development 32 Geological map 36 Chapter III. — General geography, by Rollin D. Salisbury 39 Relief and preA r alent topography 39 The limestone sink area 40 Topography and roads 42 Topography and farming 43 Mantle rock 44 Soil 44 Rock cliffs 46 History of the topography. 47 First uplift 48 Second uplift 49 Third uplift 50 Fourth uplift 51 Part II. Structural geology, by Stuart Weller and Chas. Butts 53 Chapter IV. — Structural geology 55 Faulting 55 Relation of topography to geologic structure . 64 Hicks dome 65 Herod fault 66 Lee fault 66 Wolrab Mill fault 67 Block between Wolrab Mill and Hogthief Creek faults 67 Hogthief Creek fault 68 Rock Creek fault 68 Faults a short distance north of Hogthief Creek fault 69 Peters Creek faults 69 Rosiclare fault 70 Daisy fault 71 Blue Diggings fault 71 7 8 CONTENTS PAGE Big Creek fault 71 Illinois Furnace fault 72 Threemile Creek fault 73 Wallace Branch fault 73 Pell fault 74 Shetlerville fault 74 Minor faults 74 Part III. Stratigraphic geology 77 Chapter V. — Pre-Devonian, by Stuart Weller and Chas. Butts. 79 Chapter VI. — Devonian, by Chas. Butts and Stuart Weller 83 General statement 83 Limestone of Devonian age 84 Areal distribution 84 Lithologic character 84 Stratigraphic relations 85 Thickness 85 Paleontology 85 Chattanooga shale 87 Name and distribution 87 Lithologic character 88 Thickness 88 Paleontology. 89 Correlation 89 Stratigraphic relations ( 89 Chapter VII. — Lower Mississippian, by Chas. Butts and Stuart Weller 91 General statement 91 Osage formation 92 Name and distribution 92 Lithologic character 93 Thickness 94 Paleontology ' 94 Correlation 95 Meramec group, by Stuart Weller 96 Warsaw limestone 97 Name and distribution 97 Lithologic character 98 Thickness 99 Paleontology 99 Correlation 100 St. Louis limestone, by Stuart Weeler 100 Name and distribution 100 Lithologic character 102 Stratigraphic relations 104 Thickness 104 Paleontology 104 Ste. Genevieve limestone, by Stuart Weller 106 Name and distribution 106 Subdivisions of the Ste. Genevieve limestone 109 Lithologic character of the Fredonia limestone member 110 Lithologic character of the Rosiclare sandstone member Ill Lithologic character of the 'Lower Ohara" limestone member. 112 CONTENTS 9 PAGE Detailed sections 112 Thickness 113 Stratigraphic relations 114 Paleontology 115 Chapter VIII. — Upper Mississippian series, Chester group, by Stuart Weller 121 General statement 121 Historical review 121 Shetlerville formation 133 Name and distribution 133 Lithologic character 134 Thickness 134 Stratigraphic relations 135 Paleontology 136 Correlation 142 Renault formation 142 Name and distribution 142 Lithologic character 144 Thickness 145 Stratigraphic relations 145 Paleontology 147 Correlation 150 Bethel sandstone 159 Name and distribution 159 Lithologic character 161 Detailed sections 162 Stratigraphic relations 163 Thickness 163 Paleontology 164 Correlation 164 Paint Creek formation 164 Name and distribution 164 Lithologic character 167 Detailed sections . . . 168 Thickness 169 Stratigraphic relations 169 Paleontology 170 Correlation 171 Cypress sandstone 175 Name and distribution 175 Lithologic character 176 Thickness 176 Stratigraphic relations . 177 Paleontology 177 Correlation 178 Golconda formation 178 Name and distribution 178 Lithologic character 180 Thickness 181 Stratigraphic relations 181 Paleontology .* 182 Correlation 186 10 CONTENTS PAGE Hardinsburg sandstone 187 Name and distribution 187 Lithologic character 189 Thickness 189 Stratigraphic relations 190 Paleontology 190 Correlation 190 Glen Dean limestone 191 Name and distribution 191 Lithologic character 193 Thickness 194 Stratigraphic relations 195 Paleontology 195 Correlation 197 Tar Springs sandstone 198 Name and distribution 198 Lithologic character 199 Thickness 200 Stratigraphic relations 200 Paleontology 201 Correlation 201 Vienna limestone and Waltersburg sandstone 202 Menard limestone 202 Name and distribution 202 Lithologic character 204 Thickness 206 Stratigraphic relations 206 Paleontology 207 Correlation 209 Palestine sandstone 209 Name and distribution 209 Lithologic character 211 Thickness 211 Stratigraphic relations 211 Paleontology 212 Correlation 212 Clore formation 212 Name and distribution 212 Lithologic character 213 Thickness 214 Stratigraphic relations 214 Paleontology 214 Correlation 215 Degonia sandstone 216 Name and distribution 216 Lithologic character 217 Thickness 217 Stratigraphic relations 217 Paleontology 217 Correlation 218 CONTENTS 11 PAGE Kinkaid limestone 218 Name and distribution 218 Lithologic character 219 Thickness 220 Stratigraphic relations 220 Paleontology 221 Correlation 222 Chapter IX. — Pennsylvanian, by Chas. Butts 223 Sub-Pennsylvanian unconformity. 223 Pottsville group 225 Casey ville formation 225 Name and distribution 225 Lithologic character 225 Battery Rock coal 226 Thickness 227 Paleontology 227 Correlation 227 Tradewater formation 228 Name and distribution 228 Lithologic character 228 Coal beds 228 Thickness 229 Paleontology and correlation 229 Chapter X.— Surficial deposits, by Stuart Weller 231 General statement 231 Residual deposits 231 Alluvial deposits 232 Wind blown deposits 233 Part IV. Igneous rocks, by L. W. Currier 235 Chapter XI. — Igneous rocks 237 Part V. Economic geology, by L. W. Currier and Chas. Butts 245 Chapter XII. — Economic geology 247 Mineral resources 247 Fluorspar, lead, and zinc, by L. W. Currier. 247 General statement 247 Brief history of mining development 247 The occurrence of fluorspar 248 Vein deposits 248 Minerals of the veins . 252 Bedding deposits 254 Superficial deposits 256 Paragenesis of minerals 257 Metasomatic relations of minerals 259 Description of photomicrographs and figures illustrating paragehsis 260 Origin of the deposits 275 Views of previous writers 275 Summary of genetic evidence 278 Chemical considerations 279 Geologic history of the deposits 281 12 CONTENTS PAGE Mining and milling methods 282 Mining methods 282 Milling methods 283 Economic considerations 287 Production 289 Suggestions for prospecting 290 Principal mines of Hardin County 293 Iron ore, limestone, road metal, oil and gas, by Chas. Butts 304 Iron ore 304 Limestone 305 Road metal 306 Oil and gas possibilities 307 Structure 307 Possible oil-bearing strata 309 Part VI. Paleontology, by Stuart Weller 311 Chapter XIII.— Paleontology 313 Lower Chester and related species 314 Golconda and Lower Okaw species 355 Miscellaneous Chester species 368 Plates and explanations 379 Index 402 ILLUSTRATIONS Plate page I. Geologic map of Hardin County and the adjoining part of Pope County In pocket II — A to G. Photomicrographs of vein material from the Rosiclare and Fairview mines. . .261 III. Map showing geologic structure in parts of Hardin, Pope, and Saline counties . .In pocket IV-XI. Fossils from the rock strata of Hardin County 383 Figure 1. Diagram to show the movement of rock strata in reverse and in normal faulting. ... 31 2. Diagram to show the application of the terms diy and strike 32 3. Map showing the faults and the structural segments of Hardin County and the adjoining part of Pope County 61 4. Bluff of St. Louis limestone at Tower Rock 101 5. Bluff of St. Louis limestone two miles west of Cave in Rock 103 6. St. Louis limestone with Lithostronon, on the Ohio River bank west of Cave in Rock . . 105 7. Basal Ste. Genevieve (Fredonia) limestone at Elizabethtown Ill 8. Cherty Ste. Genevieve limestone half a mile east of Elizabethtown. 112 9. Unconformity between the "Lower Ohara" and the Shetlerville formations in Fairview bluff below Rosiclare 135 10. Possible unconformity between the Shetlerville and Renault formations in Fairview bluff below Rosiclare 136 11. Unconformable contact of Bethel sandstone on the Renault limestone east of Cave in Rock 146 12. Unconformable contact of Bethel sandstone on the Renault limestone, east of Cave in Rock 147 13. Diagram showing the relations of the Mississippian and Pennsylvanian systems in Hardin County and in the Appalachian coal field . . .224 14. Dike cutting Ste. Genevieve limestone at Orrs Landing 2 miles southwest of Eliza- bethtown 238 15. Renault limestone injected by a sill at Downeys bluff south of Rosiclare 239 16. Panoramic view of the site of the "volcanic plug" south of Sparks Hill 242 17. Exposure of volcanic rock near Sparks Hill 243 18. View of a mass of the volcanic rock exposed in fig. 16 243 19. Sketch map of the Illinois fluorspar district showing locations of principal mines and prospects, igneous rocks, and mineralized faults and veins- 249 20. Diagram illustrating the effects of displacement on the shape of a fault fissure when the fault is undulatory 251 21. Map showing distribution of principal open-cuts of the Pierce mines and the apparent direction of the veins 257 22. Idealized section of a fluorspar vein 258 23. Mill and Good Hope shaft of the Fairview Fluorspar and Lead Company 283 24. Mill and Rosiclare mine of the Rosiclare Lead and Fluorspar Company 284 25. Diagram showing the development of the fluorspar industry in the United States, 1909-1916, and the contributions made by Illinois 291 26. Idealized geologic section through the air shaft of the Rosiclare mine 294 13 14 ILLUSTRATIONS Figure page 27. Idealized geologic section through the new Good Hope shaft 296 28. Idealized geologic section through the Annex shaft 297 29. Map and idealized geologic section to show the vein relationships and stratigraphy at the Blue Diggings shaft 299 30. View of a mass of fluorspar on the weathered apex of a vein 303 PART I INTRODUCTION By Stuart Weller, Chas. Butts, and Rollin D. Salisbury -2 G CHAPTER I— GENERAL INTRODUCTION By Stuart Weller and Chas. Butts Location Near the southern extremity of the State of Illinois, extending from Mississippi River on the west to Ohio River on the east, there is a belt of high, rough land that is commonly referred to as a spur of the Ozarks. This rugged area extends eastward from northern Union County, across Johnson, Pope, and Hardin counties, and spreads northward into the southern portion of Jackson^ Williamson, Saline, and Gallatin counties. The highest point, so far as known, is on Williams Hill in northeastern Pope County, which rises to an elevation of 1065 feet above sea level, but a number of points lie between 900 and 1000 feet. To the north of the Ozark country, the general surface level is near 400 feet above sea level, and south of it the Ohio River flood plain is near 300 feet. The width of the strip is approximately 25 miles. Hardin County is in the extreme southeastern portion of the State and lies wholly within the Ozark country. Its northern boundary, separating it from Gallatin and Saline counties, lies almost exactly along the crest of the ridge. Its surface slopes southward to Ohio River, which bounds the county along both its eastern and southern borders. To the west lies Pope County. Resources Although Hardin County, as one of the most important producing areas of fluorspar in the world, has long been known for its mineral products, the rugged character of its surface long discouraged the building of railways, and indeed, until recently it was one of the two counties in Illinois with no rail- road transportation, 1 and without Ohio River the mineral products of the county could never have been marketed. The lack of adequate transportation facilities has kept the mineral production essentially localized to mines near the river, and numerous promising pro pects have never been fully developed because of the lack of means for bringing the product of the mines to market. The chief industry of the county, and practically the only one aside from mining, is agriculture, and this must always remain the source of livelihood of the greater number of residents. Because of its rugged character much of the land is not tillable, and manv acres which are now cultivated or which 1 The I. C. Railroad has extender! its Golconda branch eastward into Harden County, and trains are now running to Rosiclare. 19 20 GEOLOGY OF HARDIN COUNTY have been tilled hi the past, should never have been plowed. The upland soil, which was originally the soil of nearly the whole county, is loess, a wind blown deposit which covered all the hills. This loess soil is exceedingly liable to be washed and gullied, and when a hill slope once starts to wash it is utterly destroyed for agricultural purposes in a comparatively few years. Such surfaces should be kept in timber, or they should be covered with forage grasses which would prevent the destruction of the soil and would at the same time furnish excellent pasture land. Towns The three most important towns of the county are Elizabethtown, Eosi- clare, and Cave in Eock, all of them situated upon the bank of Ohio Eiver, and all have been entirely dependent upon river transportation for their con- nection with the outer world until the recent extension of the railroad to Eosiclare. Elizabethtown, with 633 inhabitants 1 , is the county seat, and is located some distance west of the middle of the southern boundary of the county. The next most important town is Eosiclare with 609 inhabitants 1 , situated about three miles below Elizabethtown. It is the center of the fluorspar mining industry of the county, and is supported by the Eosiclare and Fairview mines, located in its outskirts. These are the only two mines in the county which have been fully developed by modern mining engineering methods. Cave in Eock, a town smaller than either Elizabethtown or Eosi- clare, is situated about eight miles east of the county seat. Other smaller communities, consisting of only a store or two and a few residences, are Karbers Eidge, Hicks, Eichorn, and Shetlerville, all of them towards the western portion of the county. Elsewhere throughout the county there are a number of country stores, near some of which there are one or two residences. The total population of the county, as shown by the 1910 census, is 7015. Topography and Drainage Hardin County has reached the mature stage of topographic develop- ment. There is but little level land, the whole surface is well drained, and the relief is great. The highest point in the county is at High Knob, the southern spur of which lies nearly on the line between Hardin and Gallatin counties, north of Karbers Eidge. This point rises to an elevation of 900 feet above sea level. Another hill two and one-half miles west of High Knob, and lying just within the northern boundary of the county, rises to an elevation of a little over 860 feet. Both of the high points mentioned are included in a series of high hills which mark the crest of the Ozark belt across southern Illinois. The northern boundary of the county lies almost along this crest of the high land, and the general surface slopes southward to Ohio Eiver, the 1 Census of 1910. GENERAL INTRODUCTION 21 elevation of whose bank is between 300 and 320 feet. The total relief of the county is, therefore, almost 600 feet. The southward sloping surface of the county is deeply dissected by the streams which drain it. The longest stream that is entirely included within the county is Big Creek, which rises within a quaiter of a mile of the northern boundaiy of the county, about two miles northeast of Karbers Eidge. The elevation of the head of the stream is 020 feet, but one tributary very close to the head rises at an elevation of 760 feet. The approximate length of this stream is 16 miles, and in that distance it falls not less than 450 feet. All but 100 feet of this fall is in the upper five miles of the stream's course, and 250 feet of fall takes place in the first mile. The other streams in the county resemble Big Creek in their profiles, the fall being rapid in the upper parts of their courses, and becoming much more gentle below. Along the lower stretches of the larger streams, where they have a gentle gradient, narrow alluvial bottoms have been developed, but they are not continuous along the stream valleys and nowhere are they very wide. The Ohio River bottoms are nowhere extensively developed within the county, the rocky bluffs rising from the river banks along most of the eastern and southern borders. A strip of bottom land is present in the southeastern part of the county, from Gentry Landing extending a little over five miles to a point about two miles east of Cave in Rock. Another stretch of bottom land in the southwestern part of the county extends from Fairview to Shetlerville, about three miles. In neither of these areas do the bottoms attain a maximum width of over one mile. Two distinct types of drainage are exhibited in Hardin County. The greater portion of the county is drained by ordinary surface streams, but in a considerable area northwest of Cave in Rock, and in a small area in the southwestern part of the county, the drainage is underground, through sink- holes. The entire county drains into Ohio River, although much of it does not drain directly into that stream. The larger basins that are wholly within the county and that are directly tributary to the Ohio, in the order of their magnitude, are, Big Creek, Honey Creek, Threemile Creek, Peters Creek, Hosick Creek, and Wallace Branch basins. Other areas drained by shorter streams that are directly tributary to the Ohio lie in the eastern part of the county between Honey Creek and the mouth of Saline River, another along the southern border of the county between Honey Creek and Peters Creek, and still another about Rosiclare. In the northeastern portion of the countv there is a considerable area, drained largely by Goose Creek and Rock Creek, which is tributary to Saline River, and in the western part, especially the northwestern part of the county, a rather large area drains westward into Grand Pierre Creek situated in Pope County, and then into the Ohio. 22 GEOLOGY OF HARDIN COUNTY The larger of the two areas that have no surface outlet for their waters lies northwest of Cave in Hock; it is dotted with sink-holes which must con- nect with underground stream channels which find their outlets through the numerous springs present in the county. The central portion of this large sink area lies near the common corner of sees. 2 and 3, T. 12 S., R. 9 E., and sees. 34 and 35, T. 11 S., E. 9 E. The other and much smaller area with underground drainage is in the southwestern part of the county. It is an elongate, rather narrow basin between Shetlerville and Eichorn, its extreme length north and south being nearly three miles and its width about one mile. The sink through which the drainage of this area passes underground is in the SW. Vi sec. 23, T. 12 S., E. 7 E.^ CHAPTER II— GEOLOGICAL INTRODUCTION By Stuart Weller and Chas. Butts The rocks which constitute the earth's crust in Hardin County and -elsewhere throughout the earth, are of two sorts as regards their origin, sedi- mentary and igneous. By far the greater part of these rocks is sedimentary, this class including the sandstones, limestones, and shales ("slates"), all of which lie in an essentially horizontal position or are locally tilted to a greater or less degree. This is especially true of Hardin County, where the igneous rocks constitute a very minor portion of the crust, and occur in the form of more or less vertical dikes, or as sills which are horizontal expansions from dikes. Sedimentary Rocks The sedimentary rocks of the region have all been formed as accumula- tions of sand or mud at the bottom of basins filled with water, or on the flood plains of ancient and long extinct rivers. The limestones, shales, and a con- siderable portion of the sandstones were accumulated in the bottom of a great embayment of an ancient ocean which covered all of southern Illinois, and parts or the whole of adjacent states. This oceanic embayment was essentially a northward extension of the Gulf of Mexico. These sediments are known to have been deposited in the ocean bottom because of the presence in them of innumerable examples of fossil shells, corals, and other marine organisms which live today and are known always to have lived under conditions fur- nished only by the oceanic waters. Similar deposits are being formed widely at the present time along the shores of the ocean as it now exists. All streams which drain the land surfaces transport quantities of sediment in the form of sand and mud. The muddy waters of the streams which drain Hardin County after heavy rains, the muddy waters of Ohio River at flood times, and of the Mississippi at all times, are due to the suspended particles of sand and clay. Further evidence of the transportation of this material is found in the constant changes in the sand bars of the river. At all times to some extent, and especially during seasons of flood, these materials are carried down stream, and sooner or later are dumped at the mouth of Mississippi River where they help to build up the great delta of that stream. Thence, tbese same materials are reworked over and over again by action of the ocean waves. They are washed alons: the shore line- and assorted into coarser and finer materials; the sand becomes a part of the beach sand of the ocean shore; 24 GEOLOGY OF HARDIN COUNTY the finer particles of mud and clay are held in suspension and are finally allowed to settle in deeper waters off shore where they may not again be disturbed. These processes which are taking place at the present time have been in progress through untold ages. During much of the time while the sedi- mentary rocks of Hardin County were forming, the shore line of the ancient ocean crossed Illinois in a general east-west direction near the center of the State. This shore-line was shifting to the north and to the south, however, from time to time, as is shown by the varying types of sediments that were deposited, and there doubtless were periods when this ocean shore crossed Hardin County itself. The sediments, laid down in this ancient ocean as loose sand and mud, gradually became consolidated by reason of the great weight of the sediments themselves which were piled layer upon layer for several thousands of feet, and through the process of cementation associated with the percolation of waters through the more or less porous mass. As they were originally formed, these sedimentary deposits were essentially horizontal, although there must have been a gentle, almost imperceptible slope or dip from the shore line through the shallow to the deeper waters at some distance from the shore. The type of sediment changed with changing con- ditions. The sands were deposited at such times as the rivers emptying into this part of the ocean were bringing in great quantities of sand, which accumulated along the shore and in the comparatively shallow waters near shore. The shallow water origin of practically all of the sandstones of the county is established by the presence of buried wave and ripple marks upon the rock surfaces in many places. These marks were formed by the undulating motion of the waters of the ancient ocean, just as similar marks are formed today, in the same manner, in the sand along the shores of the present Ohio River. Not all the ripple-marks formed in these ancient seas have been preserved but some of them were buried beneath fresh layers of sand, never to be disturbed until long after the sediments had been consolidated into sandstones, and now these markings are not infrequently exposed when layers of the rock are split apart. The phenomenon of cross-bedding in the sand- stones, or the local development of bedding at an angle to the position of the formation as a whole, is another evidence of shallow water origin of the sandstones. Similar phenomena may be observed today in the sand-liars of Ohio River as well as in oceanic deposits. In general the shales and limestones of the county were deposited in somewhat deeper waters than the sandstones, although the cutting off of the supply of sands by the streams which poured their waters into this part of the ocean permitted the deposition of shales nearer shore. The materials con- stituting the shales were supplied in the same manner as were the sands for GEOLOGICAL INTRODUCTION ^5 the sandstone, namely through the transportation agencies of streams drain- ing the surrounding land surface, but the materials were finely divided particles of clay or mud instead of sand grains. In places there was an accumulation of a mixture of clay and sand from which have arisen some of the sandy (arenaceous) shales of the county. The limestones had a totally different origin from the sandstones and shales. They were formed entirely from the shells and other hard parts of lime-secreting animals which lived in the ancient ocean 8 . In many places in the county the limestones are associated with or are interbedded with shale layers, showing that the depth of water in which they were accumulated was not greater than that in which the shales were formed. In general the lime- stones were deposited in somewhat deeper waters than the sandstones, but if at times no sand was being dumped into the ocean basin by the streams from the surrounding land surface, the accumulation of these calcareous materials might have taken place up to the shore line. That many of the limestones of the region actually were accumulated in shallow waters is shown by the presence of conspicuous cross-bedding in the limestones at many localities. The several layers of sedimentary rocks, limestone, shale ("slate"), sand- stone, and their many lithologic variations, constitute the formations of the geological section of the region. These lie in great sheets or continuous blankets called strata, that vary in thickness up to several hundred feet and are arranged one upon another like the successive boards in a pile of lumber. If it be imagined that a board were laid down each day for 30 days, then the board at the bottom might be regarded as the oldest and that at the top the youngest of the pile. So the strata of rock are of different ages, those above being younger than those next below. What is true for Hardin County is true for the whole earth. The strata are of different ages and each group of strata or formation usually has some mark or character by which its age can be detei mined. It is as if a number of layers of cement were laid down one upon the other on successive days and each layer given a distinctive color, that laid down on the fifth day, for example, being colored red. Then if the layer extended over several counties and could be seen only in a few places where some small pait of it had been exposed by removing the overlying layers by some means, as rock layers are exposed by a stream and are shown on the banks of a river or on the sides of a ravine, or if the strata should be penetrated in boring a well, the red color would show on which day that particular layer was laid down. In other words, if the succession of colors were known, the red color would show at once the relative age of that layer. In like manner each rock formation of the earth's crust has its particular place and age in the general order of suc- cession of layers making the outer crust, and commonly it has some mark by which its age and position in the whole series may be recognized. 26 GEOLOGY OF IIAIiDIN COUNTY The most reliable age marks of a geological formation are the fossil shells and other parts of animals and plants found imbedded in the rocks. These fossils are the remains of organisms that were actually living on the earth at the time when the rock strata which contain them were being formed. The long observations of geologists have shown that the life upon the earth has been continually changing progressively, and no rock formation contains a group of fossils that are exactly like those of any older or younger formation. Because of this established fact the careful gathering and study of collections of fossils is a very important part of a geologist's work. In Hardin County the determinable fossils are mostly confined to the limestone and shale formations, the sandstones in general being poorly adapted for the preservation of such objects. This circumstance, therefore, shows the great importance of the critical examination of the limestone outcrops of the county, for in many cases the age of a particular sandstone can be determined only through observation of its stratigraphic relations with some limestone formation. It has been determined through careful observation and study that each of the limestone formations in the stratigraphic succession of the county have certain distinctive species of fossils or groups of fossils, which are designated as index fossils. These are the fossils commonly to be searched for in the rock outcrops for the purpose of determining the age of the stratum exposed. Unconformities Frequently, during the progress of geological time, and in many places, there have been interruptions in the accumulation of the sediments which constitute the rocks of the earth's crust. Such interruptions are due to the changes in the positions of the ancient shore lines, and consequently to changes in the distribution of land and water. Such interruptions may have been of brief duration, or they may have extended over long periods of time during which greater or less thicknesses of sediments were accumulated elsewhere- than in the region of dry land under consideration. Such breaks in the strati- graphic succession are known as unconformities. Criteria for the recognition of unconformities are of various sorts. Where a period of deformation has intervened between the time of deposition of the older and younger strata,. during which the older beds were tilted or folded, the lack of parallelism between the strata on the two sides of the stratigraphic break shows an unconformity. In a large number of cases, however, and in all cases in Hardin County, the strata below and above the recognized unconformities are essen- tially parallel. Under such circumstances an abrupt change in the character of the sediments, as from a limestone to a sandstone, may indicate uncon- formable relations. Very commonly the surface of the underlying formation is more or less uneven, showing the results of weathering or erosion during GEOLOGICAL INTRODUCTION 27 the time represented by the unconfoimity, and in some cases the surface beneath the unconformity exhibits distinct evidence of oxidization that took place preceding the deposition of the higher formation. Not infrequently the basal portion of the formation overlying an unconformity is characterized by a conglomeratic layer called a basal conglomerate, in which there are pebbles that have been derived from older formations. Geologic Time Geologic time, that is the time during which the earth has been brought into its present form, physically and geographically, from its primitive con- dition, is inconceivably long. The records of the sequence of events in the earth's history have been preserved in the rocks which constitute the earth's crust, and it is the task of the geologist to interpret this history from the available records. Just as our year is divided into the seasons, Spring, Summer, Autumn, and Winter, and just as the history of our country is divided into more or less sharply denned periods, such as the "Colonial period," the "Revolu- tionary period," or the "Civil War period," so geologic time is divided into a series of major periods. Geologic time has continued, of course, without interruption and without break, just as the days of the year follow each other in unbroken succession, and just as the years of our national history have followed uninterruptedly. However, from a study of the rocks, geologic history can be divided into just as distinctly different periods as are the seasons of the year or the epochs of our national history. All the rocks of the outer crust of the earth are divided and subdivided into units of different orders called members; or forum t ion* including one or more members; or groups made up of one or more formations; or systems including one or more groups. The time corresponding to two or more systems is called an era, the time corresponding to a single system is called a period, and that corresponding to a group or to a formation is called an epoch. The eras are named from the stage in the progress of life upon the earth. The Proterozoic was the time of primitive life. The Paleozoic era means the era of ancient life, invertebrate animals predominating; Mesozoic era means the time of middle life, great reptiles predominating; and the Cenozoic era means the era of recent life, the mammals, to which man belongs, predominating. Geologists have commonly recognized fourteen periods pre- ceding recent time, which are grouped under the four larger eras that have been mentioned, as follows: 28 GEOLOGY OF HARDIN COUNTY Table of Geologic Time Divisions Era Cenozoic Period Quaternary Tertiary Mesozoic Cretaceous Jurassic Triassic Permian Pennsylvanian Mississippian Paleozoic Devonian Silurian Ordovician Cambrian Proterozoic Algonkian Archean The geologic record in Proterozoic or pre-Cambrian time is much less definite than is that of the periods from Cambrian to Recent, and it is only in the Cambrian that we find the first fairly complete record of the life of the earth, the pre-Cambrian rocks being so highly metamorphosed or changed that the fossil forms they may have held originally have been destroyed for the most part. Consequently the relative ages of those rocks cannot be so readily determined as can the ages of the later formations. The actual time that has elapsed since the beginning of the Cambrian period can be estimated only in the roughest sort of way. Estimates have been attempted by a number of geologists, and have ranged from thirty to several hundred million years. From seventy-five to one hundred million years may be assumed as a conservative estimate of this time, but the im- portant thought to be held in mind in this connection is not the number of years involved, but that the time is enormously, inconceivably long. In Hardin County all of the hard rocks exposed at the surface are referable to the Devonian, Mississippian, and Pennsylvanian periods. These are undoubtedly underlain by rocks of the older periods, but no deep drilling has ever been done in the county which can give us any information in regard to them. In addition to the hard rock formations, there are the unconsolidated surfieial materials which are for the most part of Pleistocene age, the period immediately preceding Recent time in the Quaternary. GEOLOGICAL INTRODUCTION 2U Igneous Rocks As has already been indicated the igneous rocks of Hardin County con- stitute a very minor portion of its surface. They were originally in the condition of molten rock or lava, that is in a more or less liquid state at a high degree of temperature. A great mass of this molten material was, at one time in the history of the earth, intruded or forced by some great internal pressure into and beneath the sedimentary rocks of the region. It is not im- probable that the bowing up of the rock strata to form the Hicks dome was due to such a deep seated intrusion. In the process of this intrusion the rocks overlying the intruded mass were cracked in places and the liquid or pasty material of the intrusion was squeezed up into the cracks, eventually becoming solidified on cooling and so healing the cracks that had been formed, with material fully as dense and hard as the material constituting the rocks through which the cracks had originally passed. These igneous rocks, which may now be seen at a number of places in the county, are dark in color, in pome places nearly black, elsewhere a little greenish or of a gray tint. In most exposures the igneous rocks occur as nearly vertical bands known as dilces, from one foot or less to one hundred feet or more in width. ^ At one locality the igneous material has spread out horizontally in a bed of limestone about four feet thick, the intruded rock occurring in stringers from a small fraction of an inch in thickness to layers three or four inches thick. These intruded bands for the most part follow the bedding planes in the limestone layer, but at intervals they break across the bedding planes in more or less oblique directions. Such a horizontal intrusion of igneous rock is known as a sill and must somewhere be connected with a dike which served as its feeder. Geologic Structure The original position of the sedimentary rocks in the earth's crust was essentially horizontal, each unit in the section consisting of a continuous blanket-like formation throughout an area of greater or less extent. In many parts of the country the rocks still retain their original horizontal position, but throughout Hardin County they have been greatly disturbed and are now neither entirely horizontal nor continuous. The present lay and arrangement of the formations and the features resulting from their production constitute what is termed the geologic structure. The structures in Hardin County are of two sorts. In one part of the county, with Hicks as a center, the rock strata have at some time in their history been bowed up into a dome, from the center of which they now dip or slope away in all directions. It is not unlikely that the Hicks dome was formed through the agency of a deep seated intrusion of molten igneous material, which lifted the entire thickness of overlying strata into the dome 30 GEOLOGY OF HARDIN COUNTY structure which they now exhibit. Elsewhere in the county, and not directly associated with the Hicks dome, the rock strata exhibit dips of greater or less degree at many localities, these dips being associated with another structural condition known as faulting. The phenomenon of faulting as exhibited in Hardin County and else- where, involves the formation of more or less vertical or oblique cracks in the earth's crust, extending to unknown depths, and the slipping of the rock strata on either side of the cracks either vertically, or horizontally, or in both directions. Such dislocations in the rock strata are known as faults. In all faults, except where the motion is entirely horizontal, the rock strata on one side of a fault are raised relative to the beds on the opposite side, the side which is raised being known as the upthrow side, the opposite as the down- throw side of the fault. The forces in the earth's crust which produce faulting are either in the nature of compression or tension. When a segment of the earth's crust is for some reason squeezed between two adjacent segments, and the compression so exerted is continued beyond the resisting power of the rock formations of the segment, these strata must give way. The beds will either be forced into folds similar to the bending of corrugated sheet iron, or cracks will be formed, the mass on one side of the crack being forced up- ward and over the mass on the opposite side, along a more or less oblique plane. Such a fault is known as a thrust fault or a reverse fault. In many places folds pass laterally into thrust faults, and very commonly thrust fault- ing is associated with very steeply dipping beds which are not infrequently overturned. In the case of thrust faulting the hanging wall is always the upthrow side, as is shown in the accompanying diagram (fig. 1, A) and the segment of the earth's crust involved is shortened. In Hardin County none of the faulting exhibited is of the thrust type. A second type of faulting is produced by the exertion of a tension or stretching force in the earth's crust rather than a compression or squeezing force. When such a force is exerted upon the rocks, many irregular cracks are likely to be former!, the major ones of which will have a general direction at right angles to the direction of the tension force. The rocks will give way in the weakest place, which may be a belt of considerable width, in which the cracks may outline a regular mosaic of blocks of various sizes and shapes. The blocks so formed will slip downward and may tip upon each other and finally come to rest when the tension force has been relieved. Faulting of this sort is called normal faulting, the fault planes are commonly more nearly vertical than in reverse faults, and the hanging wall is always the downthrown side (fig. 1, B). In any given area of faulting the immediate reason for the forces which have produced the faults cannot always be determined. In all cases, however, the development of these forces must be due to internal or surficial changes GEOLOGICAL INTRODUCTION 31 in the earth's crust, some of these changes being in the nature of transfer of materials from one position to another, either on the earth's surface through the processes of erosion, or internally through the movement of masses of molten material. Changes in temperature of parts of the interior of the earth are also effective in producing cumulative forces which may result in faulting, as also will the action of gravity upon certain segments of the crust which are heavier than others. Hanging wall Hanging wall Fig. 1. — A. Diagram to show the movement of rock strata in reverse faulting. B. Diagram to show the movement of rock strata in normal faulting. Dip and Strike The term dip is applied to rock formations to indicate the inclination of the strata (fig. 2). The degree of dip is expressed by the angle included between the surface of the rock stratum and a horizontal surface that intersects it. Thus, if the strata themselves are horizontal, the dip is 0°, or if the strata are vertical the dip is 90° ; any intermediate degree of inclination or dip may be present. There are localities, but not in Hardin County, where strata are inclined even more than 90° and are said to be overturned, but such conditions 32 GEOLOGY OF IIAKDIN COUNTY are associated with thrust faulting which is not the type of faulting exhibited here. The direction of the dip is said to be along the downward slope of the beds. The term strike is applied to the line of intersection between a dipping rock surface and a horizontal plane (fig. 2). This line, then is always at right angles to the dip, and its direction is indicated by reference to the cardinal points of the compass. In situations where the surface of dipping strata is plane, the strike lines are straight, but if the dipping strata are warped, the strike will be along curved lines. Observations upon the dip and strike of rock formations are of great importance in working out the geologic structure of a region. Upon the geologic map (PL I) an appropriate symbol is used to indicate dip and strike, a short, arrow-pointed line being directed i v V V '„: rfff Fig. 2.— Diagram illustrating the structural terms "dip" and "strike." down the dip, and a longer line at right angles to the other being placed in the direction of the strike. Topographic Development Contrary to the common conception of persons without geological ex- perience, the topography of a region, that is the irregularities of the surface, the hills, the valleys, the plains, are not unchangeable. The "everlasting- hills" are a myth, for every season brings about some changes, however small. The most notable changes in topographic form are produced through the agencies of weathering, wind, and stream erosion. The rocks which are exposed at the surface or which are so situated that they come in contact with underground waters, are constantly being subjected to changes of various GEOLOGICAL INTRODUCTION 33 sorts. Limestones are slightly soluble in ordinary ground waters. The "hard water" of springs coming from the limestone possesses that character because of the lime that is held in solution. It is common experience that constant use of such water in cooking utensils where the water is boiled, causes a scale of lime to form upon the inner surface of the vessels. This is the lime residuum from the evaporation of the water by boiling. Springs from sandstone formations have a very different quality of water than limestone springs. Such water is "soft" because no lime is present in solution. Although ground water will hold but a small amount of lime in solution, yet by long continued action of such water upon limestone formations, great quantities of material are removed and all the sink holes, caves, and solution channels of limestone formations are so formed. The greater solubility of limestone as compared with other rocks, is responsible for the fact that limestone boulders are much less commonly met with upon the surface of the ground and in the debris of creek beds, than is sandstone. In many places in Hardin County and elsewhere, where the geologist is certain that limestone is present, the surface of the ground may exhibit nothing but broken masses of sandstone. The solution of rock materials by underground and running waters is not the onty means by which the materials of the earth's crust are removed from one place and carried to another ; in fact this agency is responsible for far less of the work than is erosion. Erosion is the term applied to the processes of wearing away of the rocks of the earth's crust through the agency of running water, ice, or air. Water is and has been by far the most efficient agent of erosion in such a region as Hardin County. In the process of erosion there are three different functions performed by the moving Avater. The first of these is the wearing away or removal of rock fragments from their position as portions of the rock formations in place. This may be accomplished either through the force of the moving water itself in plucking rock fragments from their position along its path, or through the agency of materials held in sus- pension in the water. It is common experience that a hand full of sand, when rubbed forceably upon the surface of a dirty or corroded piece of metal, is efficient in the removal of the dirt or rust. In the same manner sand particles, as well as rock fragments of larger size, when held in suspension in moving water, will remove particles of the rock surface over which the water passes. The second function of the moving water is the transportation of materials. The muddy waters of swollen streams are sufficient evidence of this transportive power. Every grain of sand, and every particle of mud has at some time been a constituent part of a solid rock formation. Swiftly flowing streams are capable of moving large boulders, and rock fragments of all intermediate sizes between such masses and sand grains, and moving water is constantly carrying rock fragments of all sizes down grade. The moving rock masses in a stream bed are constantly bumping against each other as they progress down stream, —3 G 34 GEOLOGY OF HARDIN COUNTY chipping and wearing small particles from each other as they move along, this being the reason for the rounded forms of stream boulders as compared with those lying upon hill sides. The third function of water in the process of erosion is the deposition of materials. The ability of water to transport material is dependent upon the rate of its motion. A swiftly moving stream can carry more and larger rock fragments than a quiet stream. Consequently if the rate of motion of a stream is reduced it must deposit part of its load, and if the water becomes entirely quiet it must eventually drop the whole of its load, although the fine particles of mud may be held in suspension for a long time. When we think of the actual work that is being done by the streams in such a region as Hardin County, exhibited with every hard rain, we realize that there is a constant movement of the materials of the earth's crust down the slopes. Then when we combine this thought with our conception of the inconceivably long geologic time, we can realize that the amount of work which may be accomplished in the sculpturing of the earth's surface, is prac- tically limitless. Geologists are agreed that all of the topographic forms as they now exist in such an area as Hardin County, are due to the work of the streams which are present in the area. Through countless ages the water has been falling as rain and has run off down the slopes, continually shifting material downward, until all the hills and valleys as we know them have been carved out. There are other processes that have served in a notable degree to prepare the materials to be removed from one place to another. These processes come under the general head of weathering. Plants growing upon rock may send their roots into crevices, and by continued growth force off small rock particles which may be carried away by moving water. Decaying vegetation may generate organic acids which act upon the constituent materials of solid rock, loosening the fragments of the rock and causing it to crumble to such a condition that the particles may be removed by moving water. Water may penetrate the cracks in the rock, and on freezing expand sufficiently to be an effective agent in fracturing the solid rock masses. The agencies of weather- ing are acting constantly and tend to break down the solid rock masses into fragments which can be handled by the moving water. The several types of rocks vary in their powers of resistance to the various processes of weathering and erosion. Softer rocks are much more easily broken down and carried away than the harder ones, and limestones, because of their solubility, are less resistant than sandstones. For this reason the less easily decomposed or harder rocks commonly constitute the higher ground; nearly all the hills in Hardin County are capped with sandstone, while the limestone and shales, where present, commonly outcrop on the slopes below the sandstones. GEOLOGICAL INTRODUCTION 35 The accumulations of irregular, broken fragments of rocks, mixed with soil and other finer material, which gather at the base of rocky bluffs or steep hillsides, are termed talus deposits. These talus accumulations are so ex- tensive in places that they effectually obscure the rock strata which constitute the lower portion of the hill or bluff against which they lie. In situations where a bluff consists of a massive sandstone above, and a softer, perhaps shale or limestone formation below, the lower formation furnishes an insecure support for the higher sandstone, and the resulting talus is likely to include great slumped masses of sandstone, some of which may be many cubic yards in extent. In the course of its topographic development every area passes through one or more cycles of erosion, These cycles of erosion may be completed or they may be interrupted in the course of their history in such a manner that a new cycle is initiated before the older one has been finished. The history of a complete cycle of erosion involves the uplift or elevation of a given region, followed by the wearing down of the uplifted surface through the agency of the streams that are developed upon its surface. At the time of their initiation upon a surface newly raised from beneath the sea where no stream channels have been present previously, the paths to be followed by waters which fall as rain, will be along the lines of least resistance. There are .likely to be some unevenesses in the surface which may control the run-off of water falling as rain, and in whatever position the streams begin their work, there in general they will continue to carry it on throughout their history, except as they are influenced by the varying hardness of the rocks and the geologic structure. When first uplifted, either from beneath the sea or from the position of a low-lying plain, the land surface wtll be comparatively level; then the streams will dissect the area into valleys and ridges; and as erosion is con- tinued the hills and ridges themselves will be gradually reduced, provided the region is not re-elevated, until finally an approximately plane surface at or near sea-level is again produced. An area which passes through all these stages completes a full cycle of erosion, and the topographic development is said to be youthful, mature, or old, depending upon whether it represents an earlier or later stage in the accomplishment of the erosion cycle. In the stage of youthful topography the drainage is not yet fully established, parts of the area may be undrained, and the surface relief or the difference between the high and low spots is not great. In the stage of mature topography, drainage lines are fully established; they have cut deeply into the surface to be drained; the hills are narrow ridges which rise to elevations approximating the position of the. original altitude of the surface during youth; and the surface relief is great. In old age topography the surface relief is reduced so that the area approximates a plane ; and a few main drainage lines are present which serve 3G GEOLOGY OE HARDIN COUNTY as trunk lines to carry off the waters which fall upon the surface. Of course, in the topographic history of any region there are no hard and fast lines between the several stages; youthful topography gradually passes into ma- turity, and that into old age, just as a man passes by gradual changes from youth to middle age and then into old age. The topographic development in Hardin County is in the stage of maturity. Geological Map A proper geological map of any region must be drawn upon a topographic base map. The topographic maps constructed by the Illinois State Geological Survey in cooperation with the U. S. Geological Survey, are drawn on the scale of approximately one inch to the mile, that is, one inch on the map represents one mile of actual land surface. Upon this map all the cultural features, such as roads, railroads, houses, and towns are accurately located and printed in black. All the water, creeks, rivers, swamps, ponds, and lakes are accurately mapped and printed in blue. The more or less sinuous brown lines which cover the map are known as contour lines, and their purpose is to represent the topographic form of the area mapped. Each point upon any one contour line crossing the map has the same elevation above sea level, and on the map of Hardin County these lines are drawn to represent vertical intervals of twenty feet. Every fifth line is shown a little heavier than the intervening ones, the heavier lines being the contours with elevations of even hundreds of feet above sea level. The proper elevation of these heavier contours is indi- cated by numbers placed at convenient distances on the lines, and the elevation of the intervening contours can be easily calculated wherever it is desired, from the indicated elevations of the heavier lines. With this system of mapping it will be seen at once that the entire surface of the country, with all its irregularities can be accurately shown. Where the contour lines are close together the hill slope represented is steep. Where the land surface is more nearly level the contours are far apart. Such a map, when accurately constructed, can be made to show the details of surface configuration, the shape and elevation of every hill, the form of every valley and ravine. How- ever, the accuracy of the detail of such a map is directly dependent upon the scale adopted, and on the scale of one inch to the mile less detail can be shown than on a larger scale. Upon the topographic map as a base, the geology of the county is shown by means of different colors or by patterns and shades of color, which indicate the areas in which the several geological formations which have been recognized form the underlying rock. The rock formations are not everywhere exposed; if they were the county would of course be worthless for agricultural purposes. But numerous outcrops are present, and from these observed outcrops it is possible to determine with approximate accuracy the area underlain by the GEOLOGICAL INTRODUCTION 37 formation. The continuous surface existing between two formations lying in contact, one above the other, may be termed the inter-formational surface; and the boundary lines between the formations, as indicated on the map, then represent the intersections of these inter-formational surfaces with the topo- graphic surface. The positions of the inter-formational surfaces cannot every- where be seen, but the geologist can determine them for considerable areas from the observation of comparatively few actual outcrops. If the formations in any region lie in a perfectly horizontal position, the inter-formational surfaces will intersect the topographic surface along or parallel to contour lines, but if the rock formations are inclined, then the intersection between the two surfaces will cut obliquely downward across the contours in the direction of the clip or inclination, while in the direction of the strike, at right angles to the dip, the same line will coincide with or parallel the contours. Wherever the rock formations have been subjected to notable deformation, such as folding and faulting, the intersections between the inter-formational surfaces and the topographic surface become progressively complicated. The faults" which have been detected in Hardin County are indicated upon the geological map by continuous black lines. They are characterized by abrupt changes on opposite sides of the lines, of the positions of the forma- tional boundaries, these changes being due, of course, to the relative dislocation of the rocks on the two sides of the fault. CHAPTER III— GENERAL GEOGRAPHY By Rollin D. Salisbury Relief and Prevalent Topography As has been said in the preceding chapters, the surface of Hardin County is one of strong relief. Its highest point — on the north line of the county where it passes near the crest of High Knob — is 900 feet above the sea. Its lowest point, on Ohio River at the southwest corner of the county, is between 300 and 320 feet above the sea. The total relief therefore is more than 580 but less than 600 feet. Great as this relief is, the mere statement of the range of elevations gives no adequate picture of the real character of the surface. Its most notable feature is its roughness, and the dominant type of roughness is that produced by the erosion of running water. In almost every direction from almost any point, hills and ridges alternate with valleys. Tributary to the Ohio, with its deep valley on the south, and to the Saline, with its almost equally deep valley on the northeast, are numerous creeks. As has been pointed out, eight of these are large enough to receive names on the map, and the number of smaller creeks, some of them intermittent only, is still larger. To the valleys of all these creeks there are tributary valleys, and each of the larger ones has numerous tributaries. Not only this, but five creeks large enough to have names, and several smaller ones, leave the county on the west, joining Big Grand Pierre Creek which flows to the Ohio in Pope County; and as many as seven creeks, large and small, flow across the northern border of the county into Gallatin County. Big Creek and its tribuaries extend almost across the county from north to south, and everywhere in the county the heads of the tributaries of one creek system are near the heads of the tributaries of another system. All told, there are not less than a hundred drainage depressions in the county which might be called valleys. A very much larger number of smaller drainage depressions, which perhaps should be called ravines rather than valleys, are tributary to the larger ones. It will be seen that when these hundreds of drainage lines, large and small, are crowded into an area of 170 square miles, there can be little surface not closely adjacent to them. Most of the surface slopes to some one of them. So thoroughly is the surface of the upland dissected by the valleys that fully 9/10 of the surface of the county has distinct slope. 39 40 GEOLOGY OF HARDIN COUNTY At their lower ends, the valleys of the tributary streams are about as low as the valleys of the Ohio and the Saline respectively. The many tributaries of the creeks which join the rivers have valleys whose depths are in harmony with those of the streams to which they flow. Most of the valleys are 100 to 200 feet deep, but some of them are as much as 300 feet deep. Such relief alone, of course, does not make a mountainous country, but the slopes of many mountain regions are no steeper than many of the slopes of this county. Since most of the valleys are narrow as well as deep, it follows that their slopes are so steep that few roads ascend or descend them at right angles to the axes of the valleys. Not only this, but most of the divides between the valleys are so narrow as to be ridges rather than plateaus. Travelling in a straight course in almost any direction, in most parts of the county, one would descend from a ridge over a steep slope to a narrow valley, and ascend on the other side by an equally steep slope, to another ridge, which is followed, in turn, by another valley, and this by another ridge. In some places, the tributary valleys and ravines dissect the ridges between the larger valleys into hills. There are few parts of the surface from which rain water does not run promptly. After heavy showers, the creeks are swollen promptly, but the waters subside about as rapidly as they rise. The writer has seen a short, insignificant creek rise to the proportion of a raging torrent, and sink back again almost to normal proportions, in four hours. Most of the slopes to the valleys are convex upward. This is one of the distinctive features of the region. A necessary counterpart of the topography just sketched is the general absence of large areas which approach flatness. Narrow flats, less than a mile wide, border the Ohio near the southwest and southeast corners of the county, and there are lesser flats along the lower courses of some of the larger creeks, as Honey Creek in Battery Rock Township, Rock and Harris creeks in Rock Creek Township, and Big Creek in McFarlan Township. The flats along the creeks, even at their lower ends, are in but few cases so much as half a mile wide. Along Hogthief Creek in the western part of McFarlan Township, along Peters Creek in the northwestern ^part of Cave in Rock Township, and along Harris Creek in Rock Creek Township, there are comparable flats for short distances, some four miles back from the main streams. With these and perhaps a few other exceptions, most of the streams are in valleys which have no flats, or flats so narrow that their widths are measured in rods, rather than in fractions of a mile. THE LIMESTONE SINK AREA Aside from the valley flats, there is a larger area in the northern and eastern parts of Cave in Rock Township and in the extreme southwestern part of Battery Creek Township, where the surface has so little relief as to be GENERAL GEOGRAPHY 41 in contrast with its rougher surroundings. The area is shown best in sections 2, 3, 4, and 5, of Cave in Eock Township. On the topographic map, the area stands out because of the fewness of its contour lines. After its relatively slight relief, its most distinctive feature is the existence of numerous small, undrained depressions, marked by depression contours on the topographic map (see "conventional signs" below the map). These depressions are well seen along the east-west road about 2% miles north of the Ohio, and along the north-south road between the village of Cave in Eock and Love School. More than 40 of these depressions (sinks) are represented on the topographic map as containing ponds, most of which are only a few rods across. There are other ponds in sinks of the same type (1) in the southwest corner of Battery Eock Township, the area where they occur being a continuation of the larger area in Cave in Eock Township; (2) in McFarlan Township just northeast of Elizabethtown ; and (3) in Eosiclare Township, north and north- east of the village of Eosiclare for a distance of about two miles. In this last area although the ponds are not shown on the map, there are a few small ones which persist through all but the driest parts of the year. A similar but smaller area is found in sees. 14, 23, and 26 of Eosiclare Township. Unlike the topography of the rest of the county, the topography of these areas was not produced wholly by the mechanical erosion of running water. All the areas where these sinks and ponds appear are underlain by the same formation, the Fredonia limestone. This limestone would seem to be more soluble than most of the other formations of the county, and where the surface was flat enough to allow much of the rain water to enter the ground, instead of running off over the slopes, the entering water either dissolved out openings, or else developed underground caves, the roofs of which have col- lapsed, making the sinks. Probably some of the sinks are of the latter type, and the steepness of their slopes suggest that the collapse was recent; but most of them appear to be depressions developed from the surface, by descend- ing waters. Surface drainage entering the basins forms ponds in some cases, while in others it finds an underground outlet, as the many dry sinks show. This type of topography, characterized particularly by undrained sinks, is known as karst topography, from an area in southwestern Austria where similar features, though on a much larger scale, are pronounced. There is one anomalous minor feature of the topography, perhaps con- nected with the limestone sinks in origin, which deserves brief mention. The valley of Hosick Creek, for example, is much wider two miles above the Ohio than nearer the main stream. A similar condition of things exists along Big Creek, whose valley is much wider four miles above the Ohio, than at most points lower down the stream. The valley of the creek a mile and a half northwest of the village of Eosiclare is another illustration, and Harris Creek in Eock Creek Township affords still another. 42 GEOLOGY OF ITAPDIN COUNTY In all these cases, the wide parts of the valleys are in limestone. It seems possible that these wide places were once basins without outlets, similar to those in some of the limestone areas now, and that they subsequently acquired outlets. If the broad depression a mile and a half northeast of the village of Cave in Bock were to acquire an outlet, as by the headward development of a valley, this valley would be wide where the depression is, and narrow below. An illustration on a much larger scale is afforded by the larger depression a mile northwest of Love School, two miles north of the village of Cave in Bock. If the head of the drainage line in the south part of section 10 were to advance a half mile to the northward, it would connect with this depression. There would then be a valley wider above and narrower below. The wide valley of Hosick Creek at Bassett School conceivably may have originated in some such way. An alternative explanation is suggested on a later page. Topography and Boads With the exceptions noted, the roughness of surface is so great as to have had profound influence on the development of the county. The topo- graphic map shows only the short railway in Ohio Valley very recently con- structed to serve the fluorspar mines in the vicinity of Bosiclare. The existence of only one and that but a local railroad is accounted for by the roughness of the surface. Not only is the uneveness of the surface an obstacle to their construction, but it is so great as to diminish the volume of produce from the land which might otherwise call for ship- ment. The presence of a navigable stream on the east and south borders of the county makes the absence of railways less serious than it would be otherwise. The topographic map shows also that there are few roads in the county which are straight for any considerable distance, and some of those which are straight are over such rough country as to make them hard to travel. The east-west road a mile and a half north of the Ohio in Cave in Bock Township is an example. Most of the roads properly follow winding valleys or the divides between them, their courses being controlled by topography. But few of the roads of the county have been put into first class con- dition. In many places there are no bridges at the crossing of creeks, and after heavy rains the streams are too deep and swift to be forded. This makes travel over some roads impossible for short periods of time. Furthermore the short floods sometimes so wash the stream beds as to make crossing difficult after the flood goes clown. In not a few places, roads follow the beds of small creeks, and such roads are impassable many times a year. In other places, dirt roads, which are reasonably good in dry weather, become very heavy when the ground is wet. Fully half the roads shown on the topographic map are difficult at all times, either because of the steepness of their grades, or GENERAL GEOGRAPHY 43 their stoniness, or both. Most of the roads represented by broken lines on the map, and many others, belong in this category. It follows that getting crops to market is both difficult and expensive. Topography and Farming Most of the surface of the county is too rough for good farming. Some- thing like a third of the area never was cleared of timber, and a large fraction of that which once was cleared and farmed is tilled no longer. The total area actually under cultivation probably is not half of the total acreage of the county. Xot only has the steepness of the slopes made cultivation difficult, but it has favored the washing away of the soil. Xot very fertile at the outset, the soil has been impoverished by the removal of is upper part by surface wash since the timber was cut, and this upper part was the most valuable part, since it contained most of the organic matter needed for the vigorous growth of plants. Xot only has much of the surface part of the soil been washed away, but in places the slopes have been gullied so deeply that tillage is now impossible. While this is not true of a large proportion of the surface, it is true of local slopes in scores, and probably in hundreds, of places. It is not altogether clear whether cultivation of the gullied slopes was given up because of the gullies, or whether their development followed the abandonment of the land. The latter probably is the fact to a large extent, for many lands cleared but not now tilled, are not gullied, showing that gullying was not the cause of their abandonment. Many of the slopes which have been cleared should have been left in forest. The larger timber might have been cut, properly enough, but the small trees should have been allowed to grow. The slopes once cleared but no longer tilled, should be re-forested as promptly as possible. If trees had been planted when tillage ceased, re-forestation would be well under way already, and the value of the land would be much greater than it now is, for not only would the trees themselves be valuable some day, but they would help to prevent the washing away of the soil, and help to restore its depleted content of organic matter. A slope of 1 in 10, which means a slope of 528 feet per mile, or a little more than oy 2 , may, in general, be tilled with safety, if care is taken to prevent erosion by rain wash. In many situations, and with types of soil which do not wash readily, slopes of 1 in 8, 660 feet per mile, or a little more than 7°, may be tilled safely, with proper care. Steeper slopes rarely are tilled profitably for any considerable period of time by the methods in use in this region. But many of the cleared slopes of the region are considerably steeper, in not a few cases 800 feet to the mile, and in some as much as 1000 feet (nearly 11°). For the future ,2'ood of the county, it would be well if all 44 GEOLOGY OF HARDIN COUNTY cleared lands having slopes of more than 600 feet per mile were re-forested. Care should be taken in re-foresting to secure varieties of trees which will grow well in the climate and soil of the region. To this end the advice of an expert forester should be sought. If this is not practicable, it is safe to plant hardy native types of good timber trees. The black locust has been found to be a good tree for re-foresting. 1 It is recommended partly because it flourishes in soils which are poor in nitrogenous compounds, as most of the soils of this county are, and partly because it makes good post and tie timber, for which there is ready market. Mantle Rock soil Hardin County is south of the limit of. glaciation, though its northwest coiner is less than ten miles from the southern border of glacial drift. Though there is no glacial drift in the county, the soil was not derived wholly from the decay of the underlying rock. Over most of the surface except where slopes are very steep, there is a mantle of clayey loam, overlying earthy matter which shows by its composition that it was derived from the rock beneath by processes of decay. The surface loam is absent only on steep slopes and valley fiats. Thicknesses of four to six feet are common, and thicknesses twice as great are not rare. The loam is conspicuously thick on gentle slopes at the bases of steeper ones, as in the northern part of Cave in Rock Township. This obviously suggests that in such situations it has been washed down from the steeper surfaces above and lodged on the flats or low slopes below ; but no such explanation is applicable to the loam on the ridges, where it is as distinct as elsewhere. The surface loam is unlike the obviously rock-derived earth beneath, in that it contains no fragments of the underlying rock. In many places the two types of material differ in color, as well as composition. This is especially true where the underlying rock is limestone. The earthy matter to which the decay of this rock gives rise is dark, reddish-brown clay, and contains bits of residual chert which originally were in the limestone. The loam above is much lighter in color, is less compact, and contains no chert or other rock frag- ments. Where the underlying rock is sandstone, the difference in color between the surface loam and the residual earth below is less conspicuous, but the underlying material derived from the sandstone is more sandy than the loam above. As a rule, however, the upper loam is more sandy where the underlying rock is sandstone than where it is limestone. This suggests that the character of the upper loam is influenced, to some extent, by the character of the underlying rock. 1 Hopkins, Cyril G., and others, Hardin County soils: Agr. Exp. Station Soil Report No. 3, 1912. GENERAL GEOGRAPHY 45 It seems probable that the loam is made up (1) partly of the products of rock decay, modified near the surface by the various processes of weather- ing, and especially by the action of insects, earthworms, and other burrowing animals which bring up the fine material, leaving the coarse below, and (2) partly of dust which has been blown about within the region, and blown in from other areas. The surface portion has been affected much more than the deeper earths by all processes of weathering, such as solution and freezing and thawing, and by the action of roots, insects, earthworms and burrowing animals of all sorts. This surface loam sometimes has been called loess, 1 though but little of it has the texture characteristic of loess, and none of it has been observed to have the other distinctive features of loess, such as fossil land-snail shells, and concretions of calcium carbonate. The characteristic marks of loess may be present here and there, but they are rare, if present at all. Nor is the loam commonly calcareous, as most loess is, and it has not the physical characteristics which permit it to stand in the steep faces so characteristic of typical loess. If loess is defined on the basis of textural features and com- position, most of the surface loam of this county is not loess. If loess is defined as wind-deposited dust, some part of the Hardin County loam might qualify; but the wind-deposited earths probably are so mingled with those derived from the rock beneath, as to make the name loess of doubtful propriety. The soil report of Hardin County just cited indicates that the soil of nearly 90% of the area of the county is "upland timber" soil, of which 70.56% is classed as "yellow silt loam." The same report gives to the county 10.5% of "swamp and bottom" land, and more than 1% of "terrace" soil. The upland soils, especially, are notably deficient in nitrogen, a short- coming that can be overcome best 2 by the growth of clover, alfalfa, cowpeas, and soybeans, the crop being turned under as "green manure." In order to get these crops to grow well, it is important in most parts of the county to add limestone to the soil. Since limestone is abundant in the county and quarries could be opened at numerous points, it should not be difficult to supply the soil with the crushed limestone favoring the growth of the above crops, which, turned under, would greatly increase productivity. In view of the topography of the county, and the resulting lack of rail- roads and good wagon roads, it would seem that more attention should be given to stock-raising, partly because stock can be marketed more readily than grain, and partly because stock-raising would do much to make the soil more productive. In this event, the crops raised for harvest should be those which will serve as winter feed for the stock. In farther support of the above sug- 1 Hopkins, Cyril G., and others, Hardin County soils : Agr. Exp. Station Soil Report No. 3, 1912. 2 Op. cit. 46 GEOLOGY OF HARDIN COUNTY gestion, it may be noted that though much of the cleared land has slopes too steep for continuous cultivation, it might well be used for pasture. Pasture land does not wash as readily as land which is tilled. Even the timber lands are serviceable to a considerable extent for pasture. Few sheep are now raised in the county, although the region would seem to be well adapted to them. There are some lands which probably are well adapted to fruit raising, and this might well receive more attention than it has. Little fruit is at present raised. If undertaken on any considerable scale, attempted fruit culture should be under the advice of someone familiar with the soil- requirements of different fruits, and with the soils and topography of this region. In valleys where there are alluvial flats, the soils of the fiats are made up of detritus washed from the adjacent hills and ridges. They are composed largely of the surface soils once on the uplands, washed down to the valleys below. Such soils contain more organic matter than the upland soils, because the part of the upland soils containing most organic matter is the part which was washed down and deposited on the flats. These alluvial soils are fertile, and when properly tilled, yield good crops, — on the average the best in the county. But the total area of bottom land is small (about 10.5%), and some of the valley flats are so narrow as not to invite cultivation. At a few places there are low terraces along the valleys. These are best developed along the Ohio, as in the vicinity of Eosiclare and farther west, and along the lower ends of its tributaries. In a rather recent stage of geological history (the glacial period), the valley of the Ohio was rilled up somewhat with alluvial material, and this filling of the main valley dammed up its tributaries, and slack water deposits were made in them. The deposits made at this time in the creek valleys are unlike other parts of the mantle rock in the county in that they are highly calcareous, and in some cases are now full of calcareous concretions. A good illustration may be seen where the road crosses the lower end of the valley in the southwest corner of sec. 22, Cave in Eock Township. In a few places there is a little eolian (wind blown) sand along the Ohio valley, as in sec. 33, northeast of Eosiclare, and in sec. 33, Cave in Eock Township. Nowhere was such sand seen to make considerable dunes. In association with it, fairly typical loess in small quantities was seen at two or three points, as at the locality in Cave in Eock Township just mentioned. The sand here overlies the loess. Eock Cliffs Eock cliffs are singularly few considering the roughness of the surface. They exist at some points along Ohio Eiver, and at a few other points in each township. They are mostly at localities where the topographic map shows GENERAL GEOGRAPHY 47 closely crowded contours, as on the south face of Buzzards Point in the northern part of Monroe Township. Considering the steepness of the slopes, it is rather remarkable that the solid rock is so generally concealed by mantle rock. This would hardly be the case if the mantle rock were largely a product of rock decay. History oe the Topography Eocks of diverse sorts, including sandstones, shales, and limestones, underlie the county, and appear at the surface in diverse places, especially on steep slopes; but they appear also in other places, as in the bottoms of many of the small valleys. These rock formations are of different degrees of hardness, and as a consequence have been eroded unequally. Most of the conspicuous elevations are of resistant sandstone, while much of the lower land is underlain by less resistant shale and limestone. The rock strata are nearly horizontal in most places. So generally is this true, that . conspicuous ridges are nowhere formed by the outcropping edges of uptilted resistant beds. The strata of limestone, shale, and sand- stone, originally formed beneath the sea as the marine fossils in them show, were, at some later time, pushed up above their surroundings, and the present topography is the result of the erosion of running water on this uplifted mass; but the history of the topography probably is less simple than this brief statement implies. From this county and its surroundings, it is inferred that there was once a sub-even surface in the region, the remnants of which now have an elevation of about 900 feet. The existing remnants of this surface, if this conclusion is correct, are seen at Buzzards Point at the north edge of Monroe Township 3 miles east of its western border; a mile and a half west of Buzzards Point in sec. 1 ; at High Knob just north of the county line north of the village of Karbers Eidge; as well as at more distant points outside the county. In the southwest corner of the county, the conspicuous Melcher Hills probably is a remnant of the Buzzards Point plain, though lowered considerably below its level. This old plain, whose remnants are so meager that the plain itself is somewhat hypothetical, may be called the Buzzards Point plain- When it was in existence as a continuous plain over the whole region, it did not have an elevation of 900 feet. Probably it was then several hundred feet (500 to 600) lower than its existing remnants. From the topography of this county alone, it would not be clear whether this Buzzards Point plain, the highest represented in the region and the starting point for the decipherable physiographic history of the county, was the surface of the land as it was pushed up out of the sea, or whether it was a plain developed from an older surface by the prolonged erosion of running water. The distribution and the relations of the underlying rock strata, however, make it clear that con- 48 GEOLOGY OF HAEDIN COUNTY siderable thicknesses of rock had been removed from the region before the- Buzzards Point plain was developed. The date of the existence of this old plain is not fixed, even in geologic terms. If certain analogies in other regions may be trusted, it may have been in existence at the close of the Mesozoic Era; but this is at present nothing more than conjecture. After its development, the region was elevated, rela- tive to its surroundings. With increase in its elevation, its streams were quickened, erosion became vigorous, and valleys were developed in the plain, notably roughening its surface. There are strong suggestions that the uplift of the region to its present height took place by successive stages, separated by considerable intervals of time. Eemembering that the succession of uplifts of the old sea bottom which preceded the development of the Buzzards Point plain is unknown, succeeding events may be sketched. If these events are sketched in affirmative terms, it is for the sake of simplicity, rather than because the record is so clear as to make the succession of events unequivocal. FIRST UPLIFT If the Buzzards Point plain, now 860 to 900 feet above the sea, was 300 feet above the sea when the development of present topography began, there is some suggestion that its first elevation was 250 to 300 feet. In this uplifted surface, the quickened streams cut valleys to depths corresponding roughly to the amount of uplift. After the valleys cut in the uplifted plain reached some such depth as 250 to 300 feet, the streams developed flats in their bot- toms after the manner of all normal streams. These flats appear to have become extensive and the divide between them low, so that much of the area of the county was brought down nearly to their level. The result was an imperfectly developed peneplain 250 to 300 feet below the level of the Buzzards Point plain. The peneplain of this stage has been much dissected by later erosion, but remnants of it are seen in the crests of many hills and ridges which have a nearly common elevation, 600 to 640 feet above the sea. Among such- remnants in Monroe Township may be mentioned the ridge on which the village of Karbers Eidge is located. Since this sub-even-crested ridge and its northeastward extension in sections 10 and 3 to the east of the village, and similar crests in the immediate vicinity to the south in sections 7, 8, 9, 16, and 17, are among the most extensive remnants of this plain, if the interpre- tation here given is correct, it may be called the Karbers Ridge plain. There are crests at about the same level in the eastern part of the same township, as at Philadelphia School, at several points in sees. 13, 14, 15, 22, 23, 24, 25, and 26, and crests in the western part in the vicinity of Hicks, in sees. 19, 24, 25, 29, 30, 31, and 32. In these sections many hills and ridge tops come to a GENERAL GEOGRAPHY 49 nearly common level, and many others which are only a little higher. In Rock Creek Township similar crests, similarly interpreted, are in sees. 5, 6, 7, 9, 10, 19, 20, 21, 22, 24, 25, 26, 30, and 35, and many others but a little higher. In Battery Eock Township there are crests correlated with the fore- going in sees. 19, 20, 29, 31, and farther south in sees. 5, 7, 8, and 9. Crests correlated with the above find little representation in the other townships, though in Cave in Rock Township areas in sees. 1 and 6 may be mentioned, and in McFarlan Township, in sec. 17. Quite as significant as the fact that so many crests come to accordant levels, is the fact that these crests are underlain by diverse rock formations of different degrees of resistance. This fact gives support to the view that the hypothetical Karbers Ridge plain was once a fact, for once a region under- lain by rock of diverse resistance is roughened by erosion, it does not again approach planeness until it is reduced nearly to the level of the flats of the rivers which drain it. It is to be noted that most of the remnants of the Karbers Ridge plain are well back from the Ohio, where in later cycles of erosion, the surface should have suffered least. The fact and the process of development of flats along streams is illus- trated along Ohio and Saline rivers now. Both these streams have flats of considerable width. Their tributaries have narrow flats, especially at their lower ends, and most of them none at all a few miles above their mouths. All these flats are being widened all the time. Humanly speaking, the pro- cess is extremely slow, but in time the flats would be so widened, always at the expense of the ridges between them, that they would occupy the larger part of the surface. This is what is conceived to have taken place after the first uplift, while the Karbers Ridge plain was developing. Although much of the surface of the county was perhaps reduced 250 feet or so, there remained above this lower surface, some unreduced remnants of the preceding higher Buzzards Point plain. The highest point in the Melcher Hills in the southwest corner of the county is somewhat above the level of the Karbers Ridge plain, and is probably a lowered remnant of the Buzzards Point plain, lowered somewhat in later time. SECOND UPLIFT After considerable portions of the area of the county had been reduced to the Karbers Ridge peneplain, now represented by hill and ridge tops 600 to 640 feet above the sea, another uplift of 100 to 120 feet appears to have taken place. This uplift terminated the incomplete cycle of erosion just sketched. The streams, quickened by the new uplift, deepened their valleys in keeping with the new position of the surface. After having deepened their -4 G 50 GEOLOGY OF HARDIN COUNTY valleys to grade., the streams developed new flats in their bottoms 100 to 120 feet below the level at which flats were developed in the preceding cycle, at a level which is now 500 to 5-10 feet above the sea, though then much lower. In any valley tributary to the Ohio the flat developed first at its lower end, and was extended slowly up stream at the same time that it was being widened. The valley fiats therefore became wider along main streams and at the lower ends of their tributaries, and remained narrower farther up these streams. While the flats developed at this time appear to have become wide in the lower parts of the valleys of the tributary creeks, they do not appear to have been wide in their middle courses, and do not seem to have appeared at all at their upper ends, when this very incomplete cycle of erosion was interrupted by another uplift. The most extensive remnants of this plain are not so far from the Ohio as the remnants of the Karbers Ridge plain. Illustrations are seen in the crests of hills and ridges which have an elevation of about 520 feet in sees. 1, 6, 7, 9, 10, 11, 13, 15, 16, and 181, in McFarlan Township; in sees. 1, 11, 12, 13, 14, 23, 25, 26, and 35, Eosiclare Township; in sees. 7 and 17, Cave in Rock Township; in sees. 4, 9, and 10 in the southern part of Battery Rock Township, and in sees. 18, 28, 31, and 33 in the northern part; in sees. 2, 3 (near Saline River valley), 23, and 25, Rock Creek Township; and at a num- ber of points in the southwest part of Monroe Township, especially in sees. 35 and 36. No point mentioned which shows distinct remnants of this plain, has a name. If the area in the south part of sec. 10 and the north part of sec. 15, McFarlan Township, had a name, it would be given to this partial plain which came to distinct development only within a few miles of main streams. For want of a better name, it will be called the McFarlan plain. Above it, even within three or four miles of the Ohio, rose remnants of the Karbers Ridge plain, while farther from the Ohio such remnants were extensive. The Mc- Farlan plain therefore was much less extensive and much less well developed than the Karbers Ridge plain. As in the case of the Karbers Ridge plain, the remnants of the McFarlan plain are underlain by rocks of different sorts, and of different degrees of resistance. THIRD UPLIFT After the valley flats of the .McFarlan cycle had reached considerable development, with small remnants of the Buzzards Point plain, and much larger remnants of the Karbers Ridge plain, above them, the region underwent a third upraising, 100 to 120 feet in amount, and the process of valley deepen- ing was repeated. After being deepened by an amount roughly equal to the uplift, the streams in these deepened valleys began to develop flats at a level GENERAL GEOGRAPHY 51 which is now some 400 feet above the sea, and 100 to 120 feet below the Mc- Farlan plain. The flats developed at this time appear to have become somewhat wide along main streams and the lower ends of the larger tributaries, but to have had little development in the upper reaches of the creeks. Remnants of the plains developed at this stage are chiefly seen in Rosi- clare, McFarlan and Cave in Rock townships. Near the Ohio they are the low upland crests, 80 to 100 feet or so above the flood plain of the Ohio. Illustrations are found northeast of Elizabethtown for a mile or more, between Elizabethtown and Rosiclare, north of the latter village, and in the vicinity of the village of Cave in Rock. This plain, but slightly developed except in the vicinity of the Ohio, may be called the Elizabethtown plain, after the area which extends northeast from that village. The wide flatfish area in the northern part of Cave in Rock Township previously described under the heading "The limestone sinks area" probably dates from this time. The con- siderable width of this area probably is connected with the non-resistant nature of the limestone underlying this region. Back from the Ohio, the Elizabethtown plain (400 to 420 feet) is repre- sented by flatfish areas along drainage lines, rather than by crests. The flatfish tract a mile northwest of Elizabethtown is an illustration of the type of which there are many. FOURTH UPLIFT Again elevation ensued, amounting to perhaps 80 feet, and again the streams set about the task of reducing their valleys to levels in keeping with the new stand of the land. Since this uplift, the Ohio has developed a narrow flat at an elevation above sea level, ranging from about 340 feet at the north- eastern corner of the county, to about 320 feet at the southwestern corner, and the tributary streams have developed flats near their lower ends in harmony with the flat of the Ohio. The flats along the tributaries, however, have not been extended far up the valleys, so that in the upper reaches of most of the streams, the deepening of the valleys since the last uplift, has been but a continuation of the deepening which was in progress before. In other words, the last uplift shows itself in the character of the topography near the main streams, but not elsewhere. The fact that several of the creeks have better developed flats some miles above their mouths, than near them, has been mentioned already under the discussion of limestone sinks. In some cases, these flats appear to be the flats of the preceding cycle. The deepening of the valleys since the last uplift has not yet made itself felt back far enough to destroy the flats developed in the preceding cycle. If all the flats some miles above the lower ends of the creeks, and distinctly above the flat of the Ohio corresponded to the level of the Elizabethtown plain, this explanation would seem altogether satisfactory; 52 GEOLOGY OF HARDIN COUNTY but a number of them are at levels intermediate between the Elizabethtown plain level, and the level of the Ohio valley, and the above suggestion as to their origin does not fit these cases. The history sketched above is less decisively marked than could be desired. If the periods of erosion after the second and third uplifts had been longer before being interrupted by renewed uplift, the case would have been clearer. As it is, the areas representing the several plains, especially the later ones, are less extensive than could be wished to make the above conclusions certain; but they are sufficient to suggest the inferences drawn. PART II STRUCTURAL GEOLOGY By Stuart Weller and Chas. Butts CIIAPTEK IV— STEUCTUBAL GEOLOGY Faulting Hardin County, Illinois, is situated upon the margin of a district struc- turally unique for the eastern United States. The unique feature of this district is the great number of faults which crisscross the country, and break the outer crust of strata into blocks and fragments of greatly varying size, and of generally rectangular, rhombohedral, or triangular surface forms. The district so broken up extends over parts of Hardin, Pope, and Johnson counties, Illinois, and Livingston, Crittenden, and Caldwell counties, Ken- tucky, where, so far as known, the faults are even more numerous. These counties, however, probably comprise only that part of the faulted district in which the forces of the deformational stresses which produced the faulting acted with greatest intensity. The region to the south and southwest is known to have some faults, and it is not improbable that the faulted area extends southwestwarcl underneath the cover of more recent deposits to western Tennessee. The New Madrid earthquake in 1811 and the formation of Eeelfoot Lake in western Tennessee possibly were caused by crustal move- ments along a fault belonging to this general region of disturbance. Faults occur as far east, too, as Hopkinsville, Kentucky, and beyond, which shows that region also to be within the faulted district. The Wells Creek basin with associated faults of northwestern Tennessee probably lies within the general area of disturbance. The remarkable fact about this region is, that it is located in the midst of a great region of rocks which are comparatively little disturbed from their originally horizontal attitude and are almost free from faults. Yet in the very heart of this vast region of almost undisturbed rocks, lies this shattered district in extreme contrast to the surrounding regions. There are but few such occurrences in the world. Xot less extraordinary is the common presence within the region, of igneous rocks in the form of dikes, sills, and plugs, as will be described later. While such intrusive occurrences of peridotite are known in other undisturbed areas, they are very rare and are not to be compared in number with those of this region. These faults and igneous intrusions are of further great interest, owing to their close association with the important fluorspar deposits of Hardin County and the western counties of Kentucky. Probably all of these 55 56 GEOLOGY OF HARDIN COUNTY features of the county are traceable to a common origin, and consequently are intimately related to one another. As may be inferred from the foregoing statement, a full discussion of the structural geology of Hardin County involves the consideration of a somewhat larger area, extending westward into Pope and Johnson counties, Illinois, and southward into Kentucky. In general the strike of the rock formations in Hardin County, as indicated by the position of the boundary between the Pottsville sandstone and the top of the Chester group, extends, except as it is interrupted by faulting, from the southeastern portion of the county towards the northwestern corner, having a northwesterly course. About three miles east of the Hardin-Pope county boundary, however, the strike begins to curve to the south, assuming a southwesterly direction, and where these formations are traced into eastern Pope County the southerly direction of the strike is more pronounced and becomes nearly due north and south along the western wall of the valley of Big Grand Pierre Creek. In the southeastern portion of the county, as may be seen upon the accompany- ing geological map (Plate I), the strike of the rock strata again exhibits a slight curvature to the south, a direction that continues as the formations are traced into Kentucky. This area surrounded by the Pottsville-Chester boundary, including Hardin County and the eastern part of Pope, is roughly semicircular in out- line, and is apparently the northwestern border of a dome-like structure of considerable magnitude, stretching in a northwest-southeast direction. The extent of this structure into Kentucky and the completion of the dome structure, if such it be, cannot be determined at the present time, and must await the continuation of detailed geologic mapping in that direction. The explanation of the dome structure indicated by the attitude of the sedimentary formations, is suggested by the presence of the igneous dikes which have been recognized within the area. These dikes are known in Hardin County and in the southeastern part of Pope County, and others are reported from Crittenden County, Kentucky, the actual known outcrops of the dikes being distributed over an area nearly 25 miles in length in a northwest-southeast direction, and nearly 15 miles in width. All these dikes are similar in character, and are undoubtedly connected with a deep seated igneous intrusion, the injection of which into the rock formations has up- warped the overlying strata. As the rocks were elevated in this manner they were subjected to enormous tension or stretching forces which caused them to be fractured along the major lines we now recognize as faults. As the fault blocks lie at the present time, many of them have slipped downward from the position they occupied when the dome was first formed, this downward slipping being nearly or quite 1500 feet in some places. An adequate cause for this settling of the dome must be sought for. The cooling STRUCTURAL GEOLOGY 57 of the igneous material and consequent contraction would occasion a certain amount of reduction, but it would be entirely inadequate to account for the very extensive collapse of the dome. A much more competent reason for the breaking down of the original dome may be found in the gradual spreading out of the material constituting the intrusion. When the igneous matter was first intruded from a still deeper source, it was very hot and was in a more or less liquid condition. It required an exceedingly long period to become cooled and solidified, and during this whole time the enormously heavy load of sedimentary beds resting upon it tended to squeeze it out about its periphery into and between the overlying strata. The result of the action of these forces would be to reduce the height of the dome and increase its diameter. With the transfer in this manner of the position of the material which supported the original dome, the dome itself, or at least sections of it would of necessity break down, and in such a collapse extremely complicated faulting would result. Some basis for the assumption that there has been a spreading out and consequent flattening of the original igneous mass, is the fact that at least one good exhibition of an igneous sill or horizontal intrusion, has been observed in the county. This sill is well exposed (see figure 15) in the upper portion of the first knob-like hill below Eosiclare, on the bank of Ohio Eiver. Another possible explanation for the collapse of the dome supported by the intruded igneous mass, might be the withdrawal of some portion of the molten material into the cavities from which it had been extruded in the first place. Either one of these reasons, or more likely a combination of the two, would be adequate to explain the phenomena as we observe them at the present time. If the explanation of the Hardin County structures that has been suggested is the correct one, then the smaller dome structure exhibited at Hicks is but a minor feature of the larger dome, although at the present time it is in a far better state of preservation than is the major dome. The com- plicated fault pattern exhibited in Hardin County is exactly such as might be expected to be produced under the conditions that have been outlined. The extreme northwestern border of the larger dome structure is quite perfectly exhibited in the northwestern corner of the county and in the adjoining portion of Pope County. A belt passing obliquely across the county from northeast to southwest, from near the mouth of Saline River to the vicinity of Rosiclare, represents a segment of the dome which, taken alone, constitutes a great arch, and is one of the segments that has suffered extensive collapse with the settling away of the underlying igneous material. This arch-like segment has a width of about five miles in a northwest-southeast direction. Its northwestern boundary is a fault line that passes close to Wolrab Mill, and its southeastern boundary is formed bv a fault which 58 GEOLOGY OF HARDIN COUNTY extends from the mouth of Hosick Creek, a mile and one-half east of Eliza- bethtown, to a point a little above the mouth of Saline Eiver. This faulted area is in reality a double arch, or indeed for a part of its distance a three-parted arch, the belt being split longitudinally, the south- eastern and central portions being more deeply depressed and more completely collapsed than the other segments. In this more deeply down-dropped seg- ment, a long tongue-like extension of the Pottsville formations reaches southwest from the northeastern border of the dome, for a distance of nearly eight miles from their normal position of outcrop. A similar tongue-like extension of the same formations reaches in a northeasterly direction towards Eosiclare, but is mostly restricted to the Kentucky side of Ohio Eiver. This segment, therefore, even after its collapse still retains its arch-like form, the higher Pottsville occupying the northeastern and southwestern portions, and the older Chester formations the central region. Most of the faults in the county are confined to the belt indicated above. It is unnecessary to attempt to describe in detail every fault in the county, or to give the evidence upon which it is located, since their locations and dis- placements are shown upon the areal geological map (Plate I) ; but some of the more important ones will be described later. The field criteria which have been used in mapping the faults have been of various sorts. In those situations where the displacement has brought limestone and sandstone forma- tions into juxtaposition on the two sides of the fault, the determination of the fault line is a comparatively simple matter, but where two sandstones are brought together the problem has been more difficult of solution. Very generally, however, the positions of fault lines in sandstone are indicated by the transformation of the sandstone into quartzite, accompanied by notable fracturing of the quartzite. Such hard quartzite ledges in many places stand up conspicuously above the surrounding rocks because of their greater hard- ness, and are commonly designated as "quartzite reefs" by the local pros- pectors. Not infrequently, quartzite is found along some of the fault lines even where limestone is present on both sides, the source of such quartzite doubtless being some overlying sandstone bed, broken blocks of which slipped downward into fractures formed by normal faulting under tension strains. Such blocks, on being crushed between hard limestone walls became quartzitic in character and are now exposed in places along the fault lines. Such an exposure of quartzite is well exhibited along the road on the ridge in A T E. Vi sec. 7, T. 12 S., E. 8 E., a little more than two miles east of Eichorn. When the softer and less resistant formations adjoin a fault on either one or both sides, the fault itself is often not limited to a plane or even a narrow zone, but the break is likely to be a compound fracture with the displacement distributed through a much crushed zone which in places may be one to two hundred feet or more in width. In such crushed zone«, dis- STRUCTURAL GEOLOGY 59 placed masses of formations different from those on either side of the fault, are locally present, or a mass of the formation on the upthrow side may be present well over towards the opposite edge of the zone at a much lower eleva- tion. Such masses occur iri places a little outside the boundaries of the formations as shown upon the map, but they are too small in extent to be indicated upon the scale that is used in the areal map. Associated with- the faults in all sorts of rocks, there are likely to be slickensided surfaces, which are the smoothed, polished, and scored surfaces along which movement between rock masses has taken place. As seen in 'natural exposures the slickensided surfaces are commonly restricted in extent, but in some of the surfaces uncovered in mining operations they are much 'larger. At one point in the Kosiclare mine a finely slickensided surface many square yards in extent is exhibited. The presence of mineralized veins is commonly a good criterion for the location of faults in that part of the county where the rock strata are much fractured; however some of the mineral veins are certainly not along fault jlines but are situated along fractures where there has been no dislocation, or are secondary fillings of solution cavities in limestone, particularly in the Ste. Genevieve limestone. Throughout most of the county the rock strata lie in a nearly horizontal position or exhibit only gentle dips. Steeply clipping strata, however, are locally present, and where they do occur they are commonly, if not always, in close proximity to fault lines, the strike being approximately parallel to the fault. The presence of springs, in some localities, suggests the proximity of fault lines, such fracture lines in many places apparently being open courses through which the ground waters circulate and come to the surface. In places throughout the county, especially in those areas occupied by the denser limestone formations such as the St. Louis and the Ste. Genevieve, the hard limestone exhibits a notable amount of shattering, the fractures being healed with milky white calcite. Such calcite veins commonly occur near fault lines, but in places they are certainly present in locations where the beds have been subjected to strain sufficient to fracture them but not to produce dislocation of any notable amount. If the explanation for the faulting of the rocks in Hardin County that lhas been suggested in the preceding pages is the correct one. then the actual ■fractures that have taken place occurred during two different periods of the same process. The earliest fractures must have been formed during the initial updoming of the region by the intrusion of the deep-seated igneous mass. Produced by the tension strains due to the stretching of the crust above the intrusion, they would doubtless be elongate in character, more or less parallel in direction, or diverging at low angles. The later faults would 60 GEOLOGY OF HARDIN COUNTY be formed during the period of collapse of the dome or of segments of the dome, they would be very much more irregular in their direction, and many of them would be short, bounding the sides of the small, irregularly shaped blocks that would be formed during the progress of the settling of the frag- ments of the collapsing portions of the rock mass. In actual experience it is difficult to differentiate between the faults which may have been formed during these two parts of the process involved, but it is nevertheless possible that recognition of the two sorts of faults may have an important bearing upon the interpretation of the mineral deposits of the county. Certain of the faults in the region have exhibited evidence of mineralization at various points along their courses, and we may assume that they are more or less mineralized throughout their extent. Other faults have nowhere shown any evidence of mineralization, and probably are not at all mineralized. It is possible that the mineralized and the non-mineralized faults may represent on the one hand the initial faults formed with the original updoming of the rock strata, and on the other hand the subsequent fractures formed during the later collapse of the arch. Data are not available at the present time for the demonstration of any such conclusion as that which is suggested, and it probably could be demonstrated only with much addi- tional mining development. If in the future, however, the suggestion proves to be well-grounded, the differentiation of the two sorts of faults may come to have an important application in the direction of further mining development. The diagonal, much faulted zone crossing from northeast to southwest, divides the county into three distinct regions, the faulted zone itself being one of these, and the areas to the northwest and southeast being the other two (see figure 3). In its essential features this central, much faulted zone is a down dropped block or segment of younger formations lying between older rocks. In settling to its present position the dislocation of the strata did not take place simply along two, continuous faults bounding the two sides of the segment, but the fracturing and dislocation was exceedingly complicated in character. In general the central, longitudinal portion of the zone (belt number two, fig. 3) has settled more deeply than the lateral parts, and the total dislocation on each side has been distributed among a number of faults having a northeast-southwest direction. These faults are not all straight and parallel. In many localities a single fault divides and continues as two whose directions diverge at a low angle, bounding a narrow wedge-shaped block which in some places is down-dropped relative to the strata on the two sides, elsewhere on one side only, and in at least one case the rocks on the both sides of such a block are down-dropped relative to the block itself. Some of these wedge-shaped blocks are also much broken, especially near their extremities, by shorter faults whose direction is more or less transverse to STEUCTUEAL GEOLOGY 61 Fig. 3.— Map showing the faults and the structural segments of Hardin County. 62 GEOLOGY OF HARDIN COUNTY the general direction of the longer faults. In some cases two diverging faults, when traced for some distance, again converge and may become joined, in which case they enclose an elongate, narrow block, acutely pointed at each end. The apices of these wedge-shaped blocks are not all uniformly pointed in the same direction, and in some parts of the county there is a distinct intei fingering of the blocks pointing in opposite directions. In the northeastern third of the central faulted zone, the entire area is divided longitudinally into two belts, while in the southwestern two-thirds there are three such belts, the southeasternmost belt at the north becoming the middle belt toward the southwest. For convenience in discussion, these three belts may be designated by the numbers one, two, and three, beginning with the belt on the northwestern border. The accompanying map (tig. 3) outlines the zones and belts. Belt number one is continuous from Saline Eiver to the Hardin-Pope county line. It varies in width from three miles at its northernmost ex- tremity to scarcely more than one mile near Illinois Furnace, and it includes a complete sequence of the strata present in the county from the St. Louis limestone to the Tradewater formation. Like the faulted zone as a whole this belt is in the form of an arch, in which the older formations, the St. Louis and Ste. Genevieve limestones, occupy the greater part of it? central portion. In the northern limb of the arch, a complete section of the Chester and Pottsville formations is represented, but in the southern limb only the lowermost Chester formations are represented, in the extreme southwestern part of the county. The belt as a whole is not so complexly broken up by minor faults as is belt number two, but there is faulting south of Harris Creek, towards the north, and also both east and west of Wallace Branch in the southwestern part of the county. Belt number two (see figure 3) includes the most deeply down-dropped portion of the faulted zone as a whole, being constituted almost entirely of rocks of Chester and Pottsville age. The Ste. Genevieve limestone is present in only one small block east of Stone School, and the St. Louis limestone does not anywhere outcrop at the surface. In the northernmost third of the faulted zone, belt number two becomes the southeastern border of the zone, with the disappearance of belt number three, and is somewhat more than two miles wide. It becomes narrower to the southwest and is but little more than one mile wide at Keelin School. At about the position of Big Creek the direction of this belt deviates slightly and becomes more southerly, and although there is not a great difference in the course of the faulting in the two directions from Big Creek, there is a distinct though slight elbow exhibited, which has had a notable influence upon the complexity of the faulting. The northeasternmost part of this belt consists of the elongate, tongue-like mass of Pottsville sandstone which extends for a distance of approximately eight STRUCTURAL GEOLOGY 63 miles in a southwesterly direction from the position of the margin of the Pottsville outcrop on either side of the zone. Within this Pottsville tongue no extensive faulting has been worked out, although one fault has been mapped northwest of Rock Creek. It is possible that the block is more complexly faulted than is indicated on the map, for in the Pottsville forma- tions it has not been practicable to differentiate the beds with sufficient distinctness to make possible their use in the determination of fault disloca- tions. Southwest from Keelin School, however, beyond which point the Chester formations constitute most of the surface rocks in this belt, it has been possible to work out the complex faulting with a considerable degree of success. In a northeasterly direction from Big Creek, the general direction of the more continuous faults in the belt is X. 50° E., while to the southwest of the same stream the general direction becomes X. 25° E. These is more or less criss-crossing of the faults having these two directions in the region where the change of direction takes place, a circumstance which is sufficient to account for the complexity of the faulting that is exhibited for a distance of two miles along the valley of Big Creek, north from Stone School. In fact, the position of this part of Big Creek valley has doubtless been determined by the faulting, the excessive fracturing of the rocks having constituted a line of weakness along which the stream found it easiest to erode its channel. In the southwestern part of the county the very resistant Pottsville rocks cap two large hills which are included within belt number two, namely, Stone Hill situated in sees. 17, 18, 19, and 20, T. 12 S., R. 8 E., and the hill lying between Threemile Creek and Wallace Branch. A number of small hills east of Threemile Creek, and one east of Wallace Branch, are also capped^ by the same rocks. Xo continuous tongue of Pottsville in this belt, such as that extending northeast from Keelin School, is present in the southwestern part of Hardin County, but the hills capped with Pottsville constitute the terminal portion of a continuous Pottsville tongue which is present in belt number two in its extension to the southwest into Kentucky. Belt number three of the faulted zone (see figure 3) is not continuous entirely across Hardin County. It originates in sec. 33, T. 11 S., R, 9 E., near the headwaters of Peters Creek, and continues in a southwesterly direc- tion to Rosiclare. The southeastern boundary of this belt extends from its point of origin to the mouth of Hosick Creek about one and one-half miles above Elizabethtown. In its narrower northeastern portion, where the rock strata exposed are of Chester age the belt includes a number of narrowly elongate, more or less wedge-shaped blocks. In its extension to the south- west it is less broken up and is largely made up of the Ste. Genevieve lime- stone. At its southwestern extremity in Hardin County, the lower Chester formations are again exposed in the two hills on the Ohio River bank, just' below Rosiclare. 64 GEOLOGY OF HARDIN COUNTY Very little evidence is available concerning the amount of inclination or hade of the fault planes, because extensive mining operations have been con- ducted along but few of them. The Rosiclare vein is practically vertical in position, although at the Good Hope shaft of the Fairview mine it inclines 79° 30' to the northwest. At the Extension shaft of the same mine, which is probably on the same fault, the shaft has been sunk vertically to the depth of 425 feet in 16 feet of vein matter, without encountering either wall. On the other hand the fault along which the workings of the Blue Diggings mine have been conducted, is inclined to the northeast, in places at as low an angle as 45°, and at greater depths than have yet been reached in mining, it may join the Eosiclare fault to the east of it. Relation of Topography to Geologic Structure It has already been pointed out that a much faulted belt crosses Hardin County diagonally from the southwest to the northeast corner. Within this belt the major faults extend in a northeast-southwest direction, and their influence upon the surface configuration is clearly shown upon the topo- graphic map of the county. Such faults constitute lines of weakness in the earth's crust along which drainage courses naturally develop. The most important streams in the county whose direction has been controlled hy fault structure, are Rock Creek, Hogthief Creek and the upper portion of Peters Creek. Many lesser tributaries, both of the creeks mentioned and of other creeks that penetrate the belt, are also conspicuous for their northeast- southwest direction. The tributaries of Big Creek south of the mouth of Hogthief clearly show this structural influence, those from the east all entering the main stream from the northeast, while the tributaries from the west enter the main stream from the southwest. The more or less elongate, sandstone-capped hills or ridges which are conspicuously developed in this faulted belt, all exhibit a northeast-southwest trend, essentially parallel with the major faults. Northwest of Elizabethtown, the much faulted zone exhibits a slight change of direction, the general trend of the fault lines northeastward from this position being approximately N. 50° E., while to the southwest they are more nearly N. 25° E., as has been said. In the elbow where this change of direction takes place, the fault zone is far more broken up by cross faults than elsewhere, and it is through this much fractured part of the belt, which was an especial line of weakness, that the valley of Big 1 Creek has been excavated across the fault zone. No other large creek in the county crosses the faulted belt, and had not the great amount of fracturing of the rocks occasioned this line of weakness, Big Creek doubtless would have excavated its channel in a southwesterly direction from near the mouth of Hogthief Creek, and would have entered Ohio River west of Rosiclare, perhaps where Threemile Creek now joins the river. structural geology 65 Hicks Dome The Hicks Dome is an oval swell or uplift of strata which has its center in sec. 30, T. 11 S., E. 8 E., three-fourths of a mile southeast of Hicks. The longer axis of the uplift extends in a northwest-southeast direction, passing through Wolrab Mill and the SAY. % sec. 14, T. 11 S., E. 7 E. From the center of the dome the rocks dip outward in all directions through a circular belt of country of varying width, to the west and north extending beyond the county boundary. The outward dip of the rocks is exhibited along Hicks Branch from south of the center of sec. 25, T. 11 S., E. 7 E., to Big Grand Pierre Creek. In the southwestern part of sec. 25 the amount of dip varies from 12° to 30° with 15° as a fair average, as shown from observations upon the Osage chert which outcrops along Hicks Branch in that locality. From sec. 25 westward the dip as determined at a great many points does not vaxy greatly from 10°, and a northward dip of 10° prevails around the north- western end of the dome within the county. In the northeast-southwest direction along the shorter axis of the oval, the dip is likewise 15° to 20° outward near the center of the dome, but decreases outward so that at a distance of one and one-half to two miles from the center of the dome, the dip is low and the rocks are nearly horizontal in position. The difference in the dip manifests itself in the differences in width of the outcrop of the Fredonia oolitic limestone. Xorthwest of the Hicks Dome the outcrop of the Fredonia with a dip of 10° is about one-third of a mile wide, while south- west and northwest of the dome the outcrop is fully a mile wide (see Plate I). This distribution illustrates the fact that the less the angle of dip the greater will be the width of outcrop. To the northeast the dip steepens beyond the north margin of the Fredonia limestone outcrop, and is about 10° across a belt about one mile wide, lying mainly in the second tier of sections south of the county line. North of this belt the dip decreases to about 5° as finely displayed in the extensive line of cliffs dipping northward towards the syn- cline of Eagle Valley, which lies a few miles north of the county line. A profile of the structure northeast of the dome, starting at the center, would show a rather steep descent, followed by a gentle descent or nearly horizontal line for a mile or so, then a steeper descent. for three-fourths of a mile, and last a rather gentle descent to the Eagle Valley syncline. Along the northwest- southeast axis the position of the beds and structural profile are exhibited in section A- A on the geological map, Plate I. The structure of the region centering in the Hicks Dome is also exhibited by the map, Plate III, on which the position of the rocks is graphically expressed by contours showing the actual or calculated position of the top of the Chattanooga shale with reference to sea level. Outside of the outcrop of the Chattanooga shale near the center of the dome, this formation lies deep below the surface, and its top 66 GEOLOGY OF HARDIN COUNTY must be approximately determined by the thickness of the overlying strata; hence the contours are only approximately correct, but they represent well enough the general structure of the region, and probably do not vary any- where from the true position by much, if any, more than 100 feet. To the southwest of the Hicks Dome, in the country between Eichorn and St. Joseph School, the strata dip gently westward to the county line. This part of the county is also more or less broken by faults which offset and more or less interrupt the continuity of the outcrops of the formations. Herod Fault The Herod fault crosses the northwestern corner of Hardin County, and takes its name from Herod in Pope County, about one mile west of the Hardin-Pope county line. In Hardin County the fault has a displacement of more than 100 feet downthrow on the southeast, but it increases to the southwestward. The fault, or rather the system of faults into which it breaks up, has been traced to the vicinity of Eeevesville in Johnson County, 20 miles to the southwest. Lee Fault The Lee fault is named from Lees mine in the northwest corner of sec. 14, T. 11 S., E. 8 E. This fault extends northeastward for some distance beyond the county boundary, and has been traced in a southwest direction to the northwest corner of sec. 21, T. 11 S., E, 8 E., beyond which point it probably continues into the limestone mass in which it can not be detected ; and it possibly connects across the Hicks dome with the Hobbs Creek fault which continues into Pope County, although there is no discoverable evidence of such an extension. The displacement along this fault in the vicinity of Lees mine is about 400 feet, the strata on the southeast side being down- dropped so as to bring 1 the Glen Dean limestone opposite the top of the Fredonia oolite in the center of sec. 15, T. 11 S., E. 8 E., as is shown on the map. The throw diminishes to the southwestward rapidly and in the north- east corner of sec. 20, T. 11 S., E. 8 E., it is less than the thickness of, the Fredonia oolite for the Eosiclare sandstone does not appear to have been dropped to the level of the top of the St. Louis limestone. This is a strong indication that the fault dies out in that direction, as indicated by the broken line on the map, and does not connect across the Hicks dome with the Hobbs Creek fault. The effect of the downthrow in offsetting the outcrop southward on the downthrow side, is explained by the north dip of the strata and the erosion of the land to the same general level. An interesting and important feature of the Lee fault is the fluorspar vein associated with it in the vicinity of the Lees mine. So far as has been determined by actual mining operations the more important fluorspar veins STRUCTURAL GEOLOGY 67 in the county are located along fault lines, but this subject will be more fully treated in Chapter XII. Wolrab Mill Fault The Wolrab Mill fault is named from Wolrab Mill, near which it passes. In the middle part of its course this fault deviates from a straight line or an even curve, more than any other fault in the county. It is also the longe'st fault in the county, having been traced from beyond the northern boundary southwest across the western boundary into Pope County. It continues to Grand Pierre. Creek where it apparently breaks up, but is really continued as another fault nearly across Pope County into the hills south of the Bay bottoms, where it is lost a little before reaching the Massac-Pope county line. The downthrow is on the southeast side, about 205 feet in the northern part of the county, but it becomes less to the southwest, and at the Pope County line the throw is not more than 50 or 60 feet. No mineral deposits are known along the fault northeast of the axis of the Hicks dome, but the Stewart mine in the southwestern part of the county is situated upon it. Block Between the Wolrab Mill and Hogthief Creek Faults The Hogthief Creek fault and its connected continuation faults extends to the southwestern corner of the county, and the block between it and the Wolrab Mill fault is a shallow syncline. The St. Louis limestone dips north- west 5° to 7° from the Hogthief Creek fault, and the same limestone dips east at a low angle on the southeast limb of the Hicks dome east of the Wolrab Mill fault. At least one fault, which may be called the Illinois Furnace fault, nearly bisects the syncline just east of Big Creek and Illinois Furnace. This fault is clearly indicated by the relations on the opposite sides of the knob in the XE. % sec. 9, T. 12 S., R. 8 E., where the old iron ore diggings were located. In this hill the Fredonia oolite occupies the east side and the St;. Louis limestone the west side at the same level. The position of the fault is also conspicuously marked at the road intersection in XE. 14 SE. % sec. 4, T. 12 S., R. 8 E. by a thick calcite vein and with abundant calcite scattered about in the road and ditches, On the west side of the calcite vein dark limestone with Lithosiroiion prolif&rum outcrops, while just to the east and only a few feet higher the Rosiclare sandstone caps the ridge showing that the fault is downthrown on the east about 150 feet at least. It is not improbable that the synclinal tract between Hogthief Creek and Wolrab Mill faults, as far to the north as Harris Creek, is more or less cut by fissures along some of which there are slight displacements, but if so the slips are small, for the Rosiclare sandstone has a practically continuous outcrop across the block from northwest to southeast. 68 geology of hardin county Hogthiep Creek Fault The Hogthief Creek fault is named from that creek, the south side of whose valley it follows for three miles. This is one of the long faults of the county, and with possibly one exception it has the greatest displacement of any. Its northern end has not been certainly determined, but no evidence has been found for its extension beyond the east side of sec. 10, T. 11 S., E. 9 E., about one-half mile north of Carrs Store. Like the faults already described, this one is downthrown on the southeast, and to the east of Hogthief Creek the displacement is fully 1200 feet, so that the bottom of the Caseyville formation is dropped down to the level of the St. Louis limestone. The throw decreases to the northeast, so that in the SW. % sec. 10, T. 11 S., R. 9 E., the Caseyville is down only to the Glen Dean limestone, the displacement prob- ably not exceeding 500 feet. Southwestwardly the fault is traceable to Big Creek where it ends in the remarkable fault complex north of Stone Church, although its displacement is continued by a number of shorter, more or less intricately connected faults, to Ohio River at the mouth of Wallace Branch near the southwestern corner of the county. No mineral deposits have been discovered along this fault. Rock Creek Fault A little north of its mid-length, in sec. 30, T. 11 S., R. 9 E., the Hogthief Creek fault divides, the two branches diverging at a low angle for about two miles to the northeast and then becoming nearly parallel. These two branches form the boundaries of one of the characteristic, elongate, narrowly wedge- shaped fault blocks which are so frequent in Hardin County, the block being down dropped relative to the areas on either side of it. The northernmost of the two branches is considered as the continuation of the Hogthief Creek fault because the downthrow along this line is on the southeast side through- out. The southern branch may be known as the Rock Creek fault, and along this line of displacement the downthrow is on the northwest. It extends into sec. 15, T. 11 S., R. 9 E., where it is buried by alluvium and cannot be traced farther. In sec. 21, T. 11 S., R. 9 E., the bottom of the Caseyville formation in the wedge-shaped block between the two faults, is dropped considerably below the Glen Dean limestone so that the displacement is over 500 feet and probably at least 700 feet. Where this fault passes near and south of the road in the sections named, there is an excellent exhibition of the highly inclined quartzite ledges usually associated with the faults where they cut sandstone, and bearing witness of their presence. These quartzite ledges are hard, much fractured and slickensided. At the place described there seem to be two or more parallel ledges of quartzite in a width of 100 feet or so, as if there were more than one break along which, movement has taken place, which was probably the case. structural geology 69 Faults a Short Distance North of the Hogthief Creek Fault Between the Hogthief Creek fault and Goose Creek there are two parallel straight faults connected by a minor oblique fault in sees. 16, 17, 20, and 19, T. 11 S., E. 9 E. The southernmost of the parallel faults seems to branch off westward from the Hogthief Creek fault in the southwest corner of sec. 16. These faults manifest themselves very plainly at several points on ravines and along the banks of streams, by crushed and otherwise disrupted and slickensided quartzite ledges. One of the best exhibitions of these fault phenomena is in the bed of the ravine in the NW. Vi sec. 30, T. 11 S., E. 9 E., where some large slickensided quartzite boulders are lying loose upon the surface, and where the contorted shale and sandstone is exposed for several hundred feet along the bed and banks of the stream. The cross fault extend- ing from sec. 25, T. 11 S., E. 8 E., and into sec. 19, T. 11 S., E. 9 E., is in- ferred from the presence of the Menard limestone in the ravine and hillside just to the east and the Bethel and Cypress sandstones with the Paint Creek shale between on the hilltop and slopes in the eastern part of sec. 24, T. 11 S., E. 8 E. These relations could be brought about only by a fault between the ravine and the hilltop, but its exact location has not been determined. Peters Creek Faults Along the upper two-thirds of the course of Peters Creek there are four subparallel or slightly diverging faults named from the stream. The four separate faults in the group may be designated by the letters A, B, C, and D, beginning with the southernmost one. So far as can be determined all of them unite near the boundary line between sees. 33 and.34 ; T. 11 S., E. 9 E., and continue as a single fault to the northeast which can be traced into the alluvium of sec. 13, T. 11 S., E. 9 E. The downthrow is on the north side of each of these faults and the accumulated effect north of Peters Creek is a doAvnthrow of 1,000 feet or more, nearly the full thickness of the Chester group. The two southern faults of the group, A and B, diverge constantly to the southwestward. Fault A has been continued on the map on rather slender evidence, to Ohio Elver at the mouth of Hosick Creek, and fault B is certainly recognized to the center of sec. 23, T. 12 S., E. 8 E., where it is exhibited in a recently opened prospect hole, beyond which point it probably dies out. The wedge shaped block between these two members of the Peters Creek fault group is broken up by a number of minor faults in sees. 7, 8, and 18, T. 12 S., E, 9 E. Fault C of the group has been assumed to be the boundary line between belts two and three of the central faulted zone of the county. By reason of their divergence from their point of departure in sec. 33, T. 11 S., E. 8 E., and their convergence and union again, a little over six miles to the southwest, faults C and D of the Peters Creek group enclose a 70 GEOLOGY OF HARDIN COUNTY very narrow, elongate fault block in which the Chester formations from the Bethel sandstone to the Tar Springs sandstone are exposed. Where the Peters Creek faults continue to the northeast as a single fault, a small amount of mineralization has occurred, but no profitable mineral deposits have been developed. The Martin and Eureka prospects, where a small amount of lead and zinc have been found, are located on this fault in sec. 23, T. 11 S., E. 8 E. Along the several faults of the group, southwest of the point of breaking up of the single fault, a number of rather deep prospect holes have been sunk, either on the fault lines or in the adjacent, crushed, quartzite zones, but no ore of importance has yet been found. The mineral deposits in the vicinity of Lead Hill, in sec. 4, T. 12 S., R. 9 E., lie one-half mile or so southeast of the Peters Creek fault A, and there may be some con- nection between this mineralization and the faults, but such a relation has not been established. Rosiclare Fault The Rosiclare fault originates near the middle of the line between sees. 16 and 21, T. 12 S., R. 8 E., and continues in a southwesterly direction through the Rosiclare and Fairview mines. Its continuation across- Ohio River into Kentucky has not been traced. It crosses the highway near the top of the hill at the turn in the road a little over one-half mile west of the iron bridge over Big Creek near Elizabethtown. The downthrow along this fault is on the northwest side, the dislocation at Rosiclare being about 200 to 250 feet, bringing the basal portion of the Bethel sandstone on the southeast, against some portion, probably the upper part, of the Cypress sandstone. At the point where the fault crosses the road as indicated above, the amount of throw is greatly reduced, because at that place the structural block lying west of the fault — which is a somewhat irregular triangular mass of moderate size — forms a notable exception to other similar blocks in the region, in that it has remained elevated, relatively, while all of the surrounding blocks have dropped downward during the readjustment which took place at the time of the complex fracturing. In most cases in this region, such triangular blocks have dropped lower than the surrounding masses. An important feature of the Rosiclare fault is its rich mineralization. The Rosiclare and Fairview mines are the only two extensively developed fluorspar mines in the entire southeastern Illinois district, and the main shafts and workings of both of these mines are situated upon this fault. In both of these mines the mineralization has been shown to continue to a depth of 500 feet; how much farther it extends can be determined only by further development and explanation. Other smaller mines have been opened and worked along the same fault from time to time, and although they have shown more or less extensive mineralization, they have not been fully developed. structural geology 71 Daisy Fault The Daisy fault is named from the Daisy mine. This fault is really a branch of the Rosiclare fault, the two diverging to the southwest from a point in sec. 29, T. 12 S., R. 8 E., south of the road from the Big Creek bridge to Stone School. The exact point of departure of the two faults is not clearly established by the surface outcrops, but it must lie somewhere in the northern half of the section mentioned. The block between the two faults is a narrowly elongate, wedge-shaped mass, less than one-fourth mile in width except at its southern extremity; it is occupied by the Cypress sandstone throughout its entire extent at the surface. At the northeastern, pointed extremity of the wedge, where it lies in contact with the relatively elevated triangular block that has been mentioned in the description of the Eosiclare fault, the downthrow of the Daisy fault is to the southeast, but southward from the southern extremity of this triangular block the dislocation is just opposite, with the downthrow on the northwestern side. At the Daisy mine the dislo- cation must be between 150 and 200 feet, becoming somewhat less as the fault is followed to the southwest. This fault continues across Ohio River into Kentucky, where it may be clearly seen in the exposures on the river bank at Carrsville. The fault is richly mineralized at the Daisy mine. Blue Diggings Fault The Blue Diggings mine of the Fairview Fluorspar and Lead Company, is situated upon a fault which cannot be traced from surface outcrops be- cause of the surflcial covering, and because the Cypress sandstone is present on both sides of it. At the shaft it lies only a few rods east of the Daisy fault, and if it were possible to follow it north to its origin, it would doubtless join the Daisy fault, or perhaps, though less probably, the Rosiclare fault. This fault is a fracture in the main block lying between the Rosiclare and Daisy faults, but unlike the nearly vertical Rosiclare fault, it has a dip in places as low as 45° to the east, and if followed to greater depths than have been reached at this time in any mining operations, it would perhaps be found to join the Rosiclare fault. It bounds one side of a wedge-shaped block which has settled a little lower than the adjacent block to the west, between this fault and the Daisy fault. The zone immediately ajacent to the fault plane on the east has been very greatly fractured as is shown in figure 28. Big Creek Fault This fault passes very close to the road corner at Carmens Store, and a few rods west of Stone Church. It is sub-parallel with the Daisy and Rosi- clare faults, and lies a little less than one-half mile west of the Daisy fault. It has a much greater extent longitudinally than either of the two other 72 GEOLOGY OF HARDIN COUNTY faults mentioned, its length from Hogthief Creek in NE. % sec. 9, T. 12 S., E. 8 E., to Ohio River being approximately six miles. It may perhaps be considered as the master fault through the complexly faulted district along Big Creek, north of Stone Church, a circumstance that has suggested its name, and it is the termination of both the Hogthief Creek fault and the main Peters Creek fault from the northeast. In the main the dislocation along the Big Creek fault shows the downthrow on the northwest side, and opposite Rosiclare this dislocation is approximately 350 feet, bringing the top of the Golconda limestone opposite the top of the Menard. In the northern half of its course, where it passes through the complexly faulted district, it separates numerous small, triangular or irregular shaped blocks whose movements have varied in the downward adjustment during the period of deformation, so that between some of the blocks the downthrow is on one side, while elsewhere it is on the opposite side. The fault continues across Ohio River into Kentucky, where it passes between Carrsville and the point of the bluff on the Ohio just west of the town. Comparatively little prospecting lias been done along this fault. One shaft is being sunk near its northern extremity, in SE. *4 SW. % sec. 9, T. 12 S., R. 8 E., where some mineralization is shown, but the work has not yet progressed beyond the prospecting stage. In view of the rich mineralization of the Rosiclare and Daisy faults in the vicinity of Rosiclare, it would seem that the Big Creek fault might also be mineralized towards Ohio River, but the determination of such a condition can be established only by more or less extensive prospecting work. Illinois Furnace Fault The Illinois Furnace fault has already been mentioned in connection with the description of the block between the Wolrab Mill and Hogthief Creek faults earlier in the chapter. It bisects the hill south of Hogthief Creek, opposite the old Illinois Furnace, and continues in a southwesterly direction past the northwest face of Stone Hill, and terminates against the fault that will be described as the Threemile Creek fault, in the SW. % sec. 19, T. 12 S., R. 8 E, Structurally this fault, with the Wallace Branch fault to be described later, is a continuation of the Hogthief Creek fault farther to the northeast, although neither of these faults actually joins this northeastern extension which terminates at the Big Creek fault. The dislo- lation along the Illinois Furnace fault exhibits downthrow on the southeastern side, the line of dislocation like that of the Hogthief Creek fault, being the northwestern boundary of the most deeply downdropped central belt of the faulted zone of the county. The amount of dislocation varies at different points along the fault line because some of the several blocks which are adjacent to it on the southeast side have dropped deeper than others in the STRUCTURAL GEOLOGY 73 period of readjustment of the blocks. The greatest dislocation is at the northern corner of Stone Hill, in the NE. %, sec. 17, T. 12 S., R. 8 E., where the lower beds of the St. Louis limestone lie opposite the lower part of the Pottsville, giving a slip of approximately 1,600 feet, which is probably the maximum dislocation in the county. The amount of dislocation decreases somewhat in a southwestward direction, and at the termination of the fault in that direction amounts to about 1,100 feet, the Clore formation on the southeast being in contact with the Fredonia on the northwestern side of the fault. The Illinois Furnace fault is known to be mineralized at a number of points, but the prospecting work that has been done along this line has not been sufficient to determine the extent of the mineralization. A small mine was opened at one time at the northern corner of Stone Hill in the XE. % sec. 17, T. 12 S., R, 8 E., but it has been abandoned for a number of years. More recently some prospecting has been clone farther southwest along the fault, by the owners of the Eosiclare mine, but this work too, has been abandoned for the present. Threemile Creek Fault The Threemile Creek fault has been so named because some mining operations have been conducted along it at the point where it crosses Three- mile Creek in the SW. % sec. 19, T. 12 S., R. 8 E. It branches off from the Big Creek fault in the west % of sec. 16, T. 12 S., R. 8 E., and continues in a southwesterly direction past the southeastern face of Stone Hill, and past the southern termination of the Illinois Furnace fault, across the Wallace Branch fault to the Pell fault along Wallace Branch, in the west % °f sec - 25, T. 12 S., R, 7 E. Through most of its length the downthrown block lies on the northwestern side of the fault, but beyond the termination of the Illinois Furnace fault, in a southwestward direction, the downthrown block lies on the southeastern side. Whatever mineralization has been recognized along this fault occurs at and near the point where it crosses Threemile Creek, near where a number of faults come together, as is shown on the geological map (Plate I). A number of prospect holes have been sunk at this locality, but no real development work has ever been attempted. Wallace Branch Fault The Wallace Branch fault is really a continuation of the Illinois Furnace fault, offset about one-third of a mile to the west at the Threemile Creek fault. The fault originates in the SW. % sec. 19, T. 12 S., R. 8 EL, and continues in a southwesterly direction to the valley of Wallace Branch, past the southeastern side of Rich Hill, to Ohio River near the mouth of Wallace Branch. Except at its extreme northeastern extremity, where this 74 GEOLOGY OF HARDIN COUNTY fault, the Threemile Creek fault, and a third unnamed fault surround a small, triangular block of sandstone that is probably Bethel, the dislocation along the Wallace Branch fault is represented by the downdropping of the block to the southeast. In the lower portion of Wallace Branch valley the dislocation amounts to 1,100 or 1,200 feet, but it is considerably less than this farther to the northeast. Mineralization has been recognized along this fault just west of Wallace Branch opposite Melcher Hills, very near its junction with a fault extending nearly straight north along Wallace Branch, towards the Pell mine. Con- siderable prospecting has been carried on at this point, but the deposits have not yet been commercially exploited. Pell Fault The Pell fault extends north from the Wallace Branch fault, alon| AVallace Branch valley to the neighborhood of the Pell mine in NW. % sec. 24, T. 12 S., E. 7 E., beyond which point it seems to die out. It seems to be one of those lateral faults which have branched off from one of the more continuous lines of fracture. The downthrow is on the east side of the fault and grows progressively less to the north. The wedge-shaped area lying between the Wallace Branch and Pell faults furnishes a good example of the more or less complicated fracturing which is commonly developed in such situations throughout the fault zone in Hardin County. ■.^' Shetlerville Fault The Shetlerville fault marks a line of slight dislocation which doubtless branches off from the Wallace Branch fault, about as the Pell fault does, but the junction point lies south of the Ohio and so is not shown on the Hardin County map. The fault passes just east of Shetlerville and continues in a direction a little west of north, crossing the Hardin-Pope county line in the northwestern corner of sec. 26, T. 12 S., R. 7 E. It continues northwest into Pope County where it apparently terminates at the Wolrab Mill fault. Minor Faults Besides the faults that seem to be worthy of names, and of somewhat detailed descriptions, there are many others in the district, commonly shorter than those that have been described, and for the most part extending trans- versely between some of the longer ones. Faults of this character which bound blocks that are variable in shape but tend to be triangular or more or less rhomboidal, and w T hich vary in size from only a few acres to nearly a square mile, are particularly characteristic oil a district along Big Creek, between the Peters Creek and the Illinois Furnace faults. As has been STRUCTURAL GEOLOGY 75 ointed out already, this complexly faulted area is situated in a region where le longer and more continuous faults exhibit a slight change in their direc- ; on, and it was undoubtedly due to the complex strains that were developed 1 connection with this change in direction of the major faulting, that the reat intricacy of the minor faults was produced. A not uncommon henomenon in this and other complexly faulted districts in the county is lie development of faults that intersect, like the two lines in the letter X. n some cases the direction of throw along the fault line changes beyond he point of intersection, but such a relation is not exhibited in every case, imong faults generally, throughout the world, intersecting faults of this art are not common phenomena. The suggested interpretation of the area, owever, namely, the up-bulging of the crust by reason of the deep-seated atrusion of an igneous mass, and a later collapse of the dome so formed, T ould seem to afford the conditions that would account for the formation of aults of this sort. PART III STRATIGRAPHIC GEOLOGY CHAPTEE V— PRE-DEVONIAN By Stuart Weller and Chas. Butts The succession of stratified rocks exposed at the surface in Hardin County consists of a series of limestones, sandstones, and shales, extending from the Middle Devonian limestone below, to the Tradewater formation of the Penn- sylvanian system at the summit. The sequence of these formations, and their classification and arrangement into groups and systems, are shown in the accompanying table. The lowest of these rock formations was deposited first, many millions of years ago, and is the oldest, each succeeding higher formation being younger. Even the youngest of the formations, however, is very old, for since its deposition thousands of feet of younger rocks of Mesozoic and Cenozoic ages have been deposited in various parts of the earth, of which there are no representatives in this region. The Rocky Mountains have been formed in western America since the youngest rocks in Hardin County were deposited, and the strata of the Himalaya Mountains in southern Asia have been deposited and then raised to the height of 30,000 feet since he growth of the Rocky Mountains. The Hardin County region has prob- bly been dry land during most or the whole of Mesozoic and Cenozoic time. Although the Middle Devonian limestone is the oldest formation exposed n Hardin County, the region is certainly underlain by rocks of greater age. n some regions deep mine shafts have been sunk, or deep wells have been rilled which give information concerning rock formations which are not xposed at the surface, but in Hardin County no deep excavations or drilling has supplied any such information. The deepest mine shafts do not pene- trate rocks which are older than those exposed at the surface within the ounty. The only evidence available at this time bearing upon the presence oi rocks older than those exposed at the surface, is found in the pebbles of various sorts which are included in a dike which is exposed in the Ohio River J)ank at Jenkins Point, about one mile above Rosiclare. These pebbles are of various sorts, quartzites, limestones, and shales being represented. These fragments must have been derived from formations through which the igneous ntrusion passed in rising to the level now exhibited, and some of them, at least, are of rock types which are not represented in any of the formations now exposed at the surface within the county. One fragment of dark shale crom the dike has preserved a portion of a trilobite, too imperfect for de- ermination, but of a sort not known from any of the surface formations. A 79 80 GEOLOGY OF HARDIN COUNTY Table 1. — Classification of rocks in eastern United States and southern Illinois General time scale Era C en o zoic 'Recent life" (Age of mammals) Mesozoic "Middle life" (Age of reptiles) Paleozoic "Ancient life" (Era of inverte- brate animals and non- flowering plants) Period System Quaternary Tertiary Cretaceous Comanchean Jurassic Triassic Permian Pennsylvanian (Coal Measures) Mississippian Epoch Group Recent Pleistocene Pliocene Miocene Oligocene Eocene Dunkard Pottsvillt PL! J. — — =: £ z - -z Pm Chester So Devonian Silurian Ordovician Meramec Osage Kinderhook Upper Devonian Middle Devonian Lower Devonian Cambrian General section for eastern United States Formations Monongahela Conemangh Allegheny Homewood Mercer Connoquenessing Sharon Lykens coal Nos. ] and 2 Anthracite basin, Pa. Sewell, W. Va. Lykens coal Nos. 4 and 5 Anthracite basin, Pa. Pocahontas, W. Va. Alabama "Coal Measures' Kinkaid Degonia CI ore Palestine Menard Waltersburg Vienna Tar Springs Glen Dean Hardinsburg Golconda CvDress Paint Creek Yankeetown Renault Aux Vases Ste. Genevieve St. Louis Spergen Warsaw Keokuk Burlington Various local formations in Mississippi Valley Catskill Chemung Portage Genesee Hamilton Mareellus Onondaga (Corniferous) Oriskany Helderberg Section for Hardin and Pope Counties, Illinois Formations Surficial deposits Absent Absent Absent Absent Absent Absent Tradewater Caseyville PossiblvCaseyville? Absent? Kinkaid Degonia Clore Palestine Menard Waltersburg Vienna Tar Springs Glen Dean Hardinsburg Golconda Cypress Paint Creek Bethel Renault Shetlerville Absent Ste. Genevieve St. Louis Spergen Warsaw Osage Absent or upper part of Chatta- nooga shale Absent? Chattanooga shale Limestones of Devonian age of Hicks Dome Members Upper Ohara' : Lower Ohara' Rosiclare Fredonia Possibly in lower part of Devon ion limestone of Hicks Dome Not exposed Not exposed Not exposed PKE-DEVONIAN 81 very common type of pebble included in the dike rock is a purple quartzite. The lithologic character of these quartzites is not unlike that of some of the pre-Cambrian quartzites which are known as surface outcrops in various localities in North America far to the north of Hardin County. The quartzite pebbles in the dike are well rounded and have every appearance of being water worn in origin, and they may have been picked up by the lava rising in the dike from some conglomerate formation in which pebbles of this sort were present. In the Sparks Hill plug, limestone pebbles with fossil bryozoa, doubtless of Mississippian age, are not uncommon; but these of course, are not older than rocks exposed in surface outcrops. The oldest sedimentary rocks exposed in the county are restricted in their areal distribution to the central portion of the Hicks dome, around which younger formations exhibit a ring-like distribution. If we judge from our knowledge of the stratigraphic succession in other portions of the Mississippi and Ohio valleys, there are formations representa- tive of all the major geological systems below the Devonian, present in the earth's crust beneath Hardin County. The actual formations that are present are of course unknown, and the total thickness of strata down to the Pro- terozoic cannot be determined, but there are doubtless several thousand feet. -6 G CHAPTER VI— DEVONIAN By Chas. Butts and Stuart Weller General Statement The Devonian sediments in the Ohio Valley have their best development in southeastern Indiana and southwestern Ohio, on the opposite sides of the Cincinnati anticline, and the exposures in Hardin County are undoubtedly continuous with those of southeastern Indiana, beneath the overlying younger sediments. In a westerly direction the Devonian strata outcrop in the Mis- sissippi Valley in Jackson, Union, and Alexander counties, Illinois, and in Ste. Genevieve County, Missouri, and the Hardin County beds are doubtless continuous in that direction as well as to the east, beneath the younger formations. The Hardin County section, therefore, is a sort of connecting link between the sections in southern Indiana and those of southwestern Illinois and southeastern Missouri. In point of distance the Hardin County area is much nearer to southwestern Illinois than to the Indiana area. In all three of the areas mentioned the Devonian section consists of an important black shale formation whose Devonian age has been questioned by some geologists, underlain by limestones. In Indiana and the adjacent parts of Kentucky, the limestone series below the black shale has long been recog- nized as of Middle Devonian age, and many fossils have been described from the "Falls of the Ohio" from both the "Upper Helderberg" or "Corniferous," and the "Hamilton." The first satisfactory differentiation of the Middle Devonian strata of the Indiana section was made by Kindle 1 who applied the name Jeffersonville limestone to the lower or Onondaga, and Sellersburg to the upper or Hamilton part of the series. The limestone beds beneath the black shale in southwestern Illinois also have been shown to include strata equivalent to both the Hamilton and the Onondaga formations of the New York series 2 , and in addition still older formations of Lower Devonian age. Keyes 3 suggested the name Grand Tower limestone for the Middle Devonian strata "below those beds containing the fossils of the western Hamilton." Since there are no "western Hamilton" faunas in the Grand Tower section, the name as originally given would seem to be applicable to the whole of the Middle Devonian section at that locality. iBull. Amer. Pal., vol. 3, p. 8 (1899). 2 Weller, Jour. Geol., vol. 5, pp. 625-635 (1897) ; also Savage, Trans. 111. State Acad. Sci.. vol. 3, pp. 116-132 (1910). 3 Mo. Geol. Surv., vol. 4, p. 42 (1894). 83 84 GEOLOGY OF HARDIN" COUNTY Savage has amended the definition of the formation to include "only that portion of the Devonian strata of southwestern Illinois, and adjacent parts of Missouri, which is the western representative of the Onondaga limestone of New York." The Hardin County exposures are limited in extent, and outcrop in only the one locality, namely immediately southeast of Hicks on the summit of the Hicks dome, where the dome-shaped upward bulge of the rocks has permitted the overlying formations to be worn away so as to expose the Devonian strata in the center. The fossils which have been found in the Devonian limestones of Hardin County, were not collected from the beds in place. The species that have been identified indicate the presence of strata of both Hamilton and Onondaga age, the Sellersburg and Jeffersonville of the Indiana section, and the. Grand Tower and overlying beds in southwestern Illinois. In addi- tion to these there are some forms which have been previously recorded only from the Clear Creek chert of southwestern Illinois, a formation which is commonly considered as late Lower Devonian in age. Since it is not prac- ticable to differentiate these Devonian limestone beds in Hardin County, they are treated together under the head "Limestone of Devonian age." Limestone of Devonian Age areal distribution Outcrops of the Devonian limestone in Hardin County occupy an oval area in sections 30 and 31, T. 11 S., E. 8 E., and in the eastern part of sec, 25, T. 11 S., E. 7 E., the entire area being about one square mile in extent. Exposures of the limestone are very few and it is known almost wholly from its chert which is abundant upon the surface underlain by the limestone. LITHOLOGIC CHARACTER So far as known the Devonian limestone is rather coarsely crystalline and medium dark gray. Limestone of this character, a few feet thick, is exposed at the mouth of a large ravine at the south end of the dome in the NE. % sec. 31, T. 11 S., E. 8 E., and similar limestone was penetrated at a depth of about 50 feet in the shaft of the fluorspar prospect on the Eose farm about one-fourth of a mile southeast of Hicks School. At the Eose prospect, about 20 feet of medium thick-bedded chert is now exposed in the upper part of the pit, dipping about 30° N. A considerable quantity of limestone has been thrown out of the pit and is now mixed with the chert and earth on the dump. The preceding statement gives all the knowledge that has been obtained of the bed rock of the Devonian limestone in the county. In addition the presence: of the limestone below the surface is indicated by much chert scattered upon the ground. The chert is gray and compact, and tends to assume blocky or DEVONIAN 85 rounded forms. From its presence the rocks below have formerly been re- garded as Mississippian and the same as the rocks outcropping on the ridges surrounding the valleys occupied by the Chattanooga (black) shale, since these ridges are also thickly strewn with chert. The cherts of the two areas; can readily be distinguished, however. The Devonian chert takes rounded, blocky, or irregular forms and much of it is highly or fairly fossiliferous, while the Mississippian chert is distinguished by angular, smooth-faced pieces, many of which are striped with narrow, pinkish or yellowish bands, and fossils are very scarce. So far as known at present, this Devonian formation is com- posed of limestone which is commonly silicified to chert near the surface, as indicated by the abundance of loose chert upon the ground, and also by the bedded chert just below the surface at the Rose prospect. It is not improbable that the limestone is prevailingly thus silicified on the outcrop but remains in its original condition as limestone at no great depth. STRATIGRAPIIIC RELATIONS The Devonian limestone underlies the Chattanooga (black) shale but it is not known what is beneath the limestone in this region since the underlying rocks are nowhere exposed. THICKNESS No direct measurement of the thickness of the Devonian limestone of Hardin County is possible owing to the absence of exposures. The thickness can be calculated only by projecting over the Hicks dome the boundary surface between the limestone and the Chattanooga shale as done in section AA. Assuming a regular curvature to this surface, it would pass about 250 feet above the highest knobs near the center of the area of outcrop of the lime- stone, and about 350 feet above the bottoms of the adjacent ravines. As the limestone apparently extends to the bottoms of the ravines its thickness should therefore be in the vicinity of 350 feet. PALEONTOLOGY The Devonian chert scattered over the area of the Hicks dome is fairly fossiliferous but the fossils are mostly in a fragmentary and otherwise poorly preserved condition. A list of the forms either certainly or doubtfully identi- fied is given below : List of fossils from the Devonian limestone Cyathophyllum rugosnm Hall, or Michelinia cf. M. stylopora Eaton? Acervularia davidsoni E. & H. Dolatocrinus sp. Cystiphyllum sp. Hexacrinus leai (Lyon) Heliophyllum sp. Nucleocrinus verneuili (Troost) 86 GEOLOGY OF HARDIN COUNTY List of fossils from the Devonian limestone — Continued Atrypa spinosa Hall? Spirifer duodenarius Hall Amphigenia curta (M. & W.) Spirifer gregarius Clapp? Camarotoechia tethys (Billings)? Spirifer iowensis Owen? Chonetes coronatus (Conrad) Spirifer segmentum Hall Eatonia sp. Spirifer varicosum Hall Leptaena rhomboidalis (Wilck) Spirifer cf. worthenanus Schuchert. Meristella haskinsi Hall? Very close to S. concinnus in form Schuchertella arctistriata (Hall) and number of ribs and exactly Spirifer audaculus (Conrad)? the same in ornamentation Spirifer divaricatus Hall Tentaculites bellulus Hall Of the species listed Hexacrimis leai, Nucleocrvn.ua vemeuili, Camaro- toechia tethys, Spirifer divaricatus, Spirifer duodenarius, Spirifer gregarius, Spirifer varicosus, and Spirifer segmentum are recorded only from the Onon- daga (Corniferons) limestone, (Grand Tower limestone of Union and Jackson counties, Illinois, and Jeffersonville limestone of Kentucky and Indiana) except that Spirifer duodenarius occurs also in the upper 10 feet of beds assigned to the Clear Creek chert (Oriskany) of Union County, Illinois. Amphigenia curta and Spirifer worthenanus are recorded only from the Clear Creek chert of Union County. The genus Eatonia is not elsewhere known above the Oriskany, except a form identified as Eatonia occurs at Grand Tower, Illinois, where it is recorded by Weller in the lower YtV^ feet of beds assigned by him to the Grand Tower (Onondaga) limestone. Pro- fessor Weller now thinks, however, the fossil does not belong to the genus Eatonia. Michelinia sUjlopora is identified from the Clear Creek chert. It would seem reasonable that the Michelinia of the Hardin County Devonian is the same as that of Union County. Michelinia stylopora, however, is a Hamil- ton form originally described from the Hamilton of Erie County, New York. Most of the other species or genera either occur in the Hamilton or at least not below it. Chonetes coronatus especially is considered a good index fossil of the Hamilton. On the basis of the fossil evidence therefore, as usually interpreted in Illinois, the Devonian limestone of the Hicks dome may include representatives of Oriskany time as well as those of Onondagan and Hamilton time, which it surely does include. However, the beds in the upper part of the Clear Creek chert containing Amphigenia. curta, Spirifer worthen- anus, and, in Union County, other brachiopods, probably should be regarded as Onondaga rather than as Oriskany. There would remain only the Eatonia to indicate the presence of the Oriskany, and as the specimen found was near the bottom of a ravine in which the lowest stratigraphic horizon would be exposed, it is not impossible that 50 feet or so of genuine Oriskany outcrops in the deeper ravines on the summit of the Hicks dome. The thickness of 350 feet is ample to include any thickness of beds of Hamilton and Onondagan as well as most of those of Oriskany age known in Illinois. In Union County the Hamilton equivalent is given as 70 feet, the Marcellus as 28 feet, the DEVONIAN 87 Onondaga (Grand Tower limestone) as 156 feet, and the Oriskany (Clear Creek chert) as 237 feet, a total of 490 feet. Assuming thicknesses to be the same on the Hicks dome, the Devonian beds there could extend nearly 150 feet into the Oriskany. As it is clearly impossible under the conditions to separate the several components of the Devonian here, they have been treated together under the head "Limestone or Devonian Age." It is of interest that a vein of high grade fluorspar occurs in the Devonian limestone and probably in the Hamilton part of it, on Goose Creek about one mile southeast of Hicks, where it has been prospected and worked to some extent on the Rose property. The vein is apparently a replacement of the limestone layers and lies parallel to the bedding. The inclosing beds are chert which also is a replacement of the limestone layers. Chattanooga Shale name and distribution A widespread black shale formation has long been recognized by geological workers in Michigan, Ohio, Kentucky, Tennessee, Alabama, Georgia, Indiana, Illinois, southern Missouri, Arkansas, and Oklahoma. The earlier writers commonly referred to this formation as the "Devonian black shale/' but several geographic names have been applied to it by different workers. As early as 1861 Winchell 1 included this black shale as the lower member of his Huron group. In southern Ohio the member was called the "Ohio black shale" by Andrews 2 in 1870. In Indiana Borden 3 applied the name New Albany shale to the formation in 1874. Many of the earlier writers were inclined to correlate the shales with the Genesee shale of the New York De- vonian section, a name which has been used not infrequently in the literature discussing the formation. When the black shale was recognized in northern Arkansas it was named Eureka shale by Simonds 4 in 1888. In 1890 Hayes 5 applied the name Chattanooga to the formation in Tennessee,. In 1904 Adams 6 pointed out the fact that the name Eureka, used by Simonds and others for the black shale in Arkansas, was preoccupied for a formation in Nevada, and thereupon proposed to substitute the name Noel shale for the black shales in northern Arkansas and southern Missouri. From this great array of names it is difficult to select the one which should properly be applied to the formation as it occurs in Hardin County. The IT. S. Geological Survey has consistently used the name Chattanooga in the many folio publications issued, covering areas south of Ohio Eiver, and extending as far west as Oklahoma. This name is not the earliest one pro- posed for the formation, but is rather one of the later ones; however it has *lst Bien. Rep. Prog-., Geol. Surv. Mich., p. 71 (1861). 2 Geol. Surv. Ohio, Rep. of Progress. 1869, p. 62 (1870). 3 5th Ann. Rep. Ind. Geol. Surv., p. 158 (1874). 4 Ark. Geol. Surv., 1888, vol 4, pp. 26-27 (1888). 9 Bull. Geol. Soc. Amer., vol. 2, p. 143 (1890). 6 U. S. Geol. Surv., Prof. Paper No. 24, p. 24 (1904). 88 GEOLOGY OF HARDIN" COUNTY been more widely applied than any other and may be adopted for the Hardin County exposures. In Hardin County the Chattanooga shale outcrops only in the vicinity of Hicks, in a rudely circular belt of country from one-fourth to one-third of a mile in width, surrounding the area of the Devonian limestone of the Hicks dome, and following strictly the headwater valleys of Goose Creek. LITHOLOGIC CHARACTERS The Chattanooga is a black, carbonaceous, fissile shale with occasional non-fissile layers a foot or so in thickness. Such non-fissile layers are in- cluded in the outcrop on Goose Creek in the northeast corner of sec. 31, T. 12 S., E. 8 E. The specimens that have been carefully examined, including specimens of very fissile layers from near the top, and of the thicker, non- fissile layers near or below the middle, are highly siliceous, evidently being composed in large proportion of very fine quartz grains. In some parts of the formation small amounts of pyrite are present, together with occasional, very hard, concretionary bodies that are more or less pyritiferous. The upper- most ten feet or so of the formation is a thin-bedded, green, argillaceous sandstone, the green color being due to the presence of glauconite, some of the layers appearing to be half glauconite in small grains. Upon weathering, the shale loses its deep black color, becoming brownish and finally grayish in color, and splits readily into thin, laminar fragments. thickness y The thickness of the Chattanooga in the region appears to have been underestimated in previous accounts at 50 to 100 feet. The best determina- tion of thickness can be made at the road intersection in the line between sec. 36, T. 12 S., R. 7 E. and sec. 31, T. 12 S-., R. 8 E., one-fourth mile south of the north line of the sections. Here the shale is almost continuously exposed from the top contact with the Osage chert above, for 600 feet across the strike to the northlwest. The dip is 20° to the southwest. This would give a thick- ness of 220 feet to which must be added 50 feet, the difference in elevation between the top and the lowest beds exposed, making the total thickness exposed 270 feet. But the bottom of the shale is not shown and shale is reported to occur on the slopes across the valley to the northwest. Assuming the shale to extend to the foot of this slope the width of the outcrop would be at least 1,000 feet, which with a dip of 20° would make the thickness 364 feet, and by adding the difference in altitude between the top and bottom, which is 80 feet, the total thickness of 444 feet is obtained. It is believed, therefore, that the thickness is reasonably estimated at 400 feet. This would allow for some variation in the dip which might reduce the average below the 20° observed across the exposed part of the shale. DEVONIAN" 89 PALEONTOLOGY The Chattanooga shale in Hardin County is almost devoid of fossils. Mr. Weller has found a single specimen of the Phyllocarid crustacean Spathio- caris, very similar to or identical with Spathiocmis emersoni Clarke. As originally described this species is recorded from the Upper Devonian Portage formation of New York, where it is said to be widely distributed in the Naples beds, and in the upper sandstone of the formation. 1 Examples of a species identical with the form in the Chattanooga shale in Hardin County have been collected by Weller in the Sweetland Creek shale of Iowa, and in a greenish shale overlying the black shale in southwestern Missouri. Mr. Butts has col- lected from the glauconitic sandstone at the top a brachiopod shell, probably a Leptaena or Pliolidostropliia, a fragment of a shell suggesting a Bellerophon, and a plumose, carinated fragment suggestive of a conodont or the fringed tip of a crustacean spine, like the telson spine of Acanthotelson ereni, CORRELATION The Chattanooga shale of Hardin County is supposed to include the equivalent of the Chattanooga of the type locality — Chattanooga, Tennessee — but it must include much more than that for at Chattanooga the shale is only 20 to 30 feet thick. From Chattanooga this characteristic black shale has been traced practically all the way to Hardin County, being of course covered up by overlying formations in extensive areas. The black shale, however, always appears in outcrop wherever its horizon is exposed, so that there is no doubt whatever of its continuity as a great sheet spread over the entire region from Birmingham, Alabama, northwestward to the Mississippi Valley. In Ohio, and at New Albany, Indiana, Ulrich thinks the main body of the black shale is Mississippian and equivalent to the Kinderhook of the Mississippi Valley, while admitting that the lower few feet is of Devonian and probably of Genesee age, that is of the same age as the Genesee black shale of the Genesee Valley in New York. The general scarcity of fossils in the formation everywhere, together with more or less uncertainty as to the stratigraphic range of such as have been found, makes the determination of its age uncer- tain. In Hardin County it lies between limestone of Middle Devonian (Hamilton) age below, and chert of Osage age above, and might, therefore, be either Upper Devonian (Genesee, Portage, Chemung) or lower Mississippian (Kinderhook), or it may be partly one and partly the other. STRATIGRAPHIC RELATIONS If the Kinderhook group of the Mississippian is absent in Hardin County, there is a stratigraphic gap or unconformity between the Chattanooga 1 Hall and Clarke, Pal. N. Y., vol. 7, p. 200 (1888). 90 GEOLOGY OF HARDIN COUNTY shale and the overlying Osage chert. If the Kinderhook is represented by the Chattanooga there is an unconformity between the Chattanooga and the Middle Devonian limestone below it, and this unconformity would mean that the whole of the Upper Devonian, aggregating many thousand feet in thick- ness, is absent in southern Illinois. If the upper and larger part of the Chattanooga is Kinderhook, and the lower part Genesee, then there is a great unconformity in the midst of the formation, in which most of the Upper Devonian is absent. CHAPTEE VII— LOWER MISSISSIPPIAN By Chas. Butts and Stuart Weller General Statement The Mississippian and succeeding Pennsylvania!! systems have been esignated as series by many writers, instead of systems, and have been •rouped together! as the Carboniferous system. There are ample reasons, owever, such as the general character of the rocks, their thickness, their aleontology, and especially the geological history, for elevating each to the ank of a system as is done in this report. The most important reason for his treatment is the fact that hundreds of thousands if not millions of years lapsed between the formation of the highest Mississippian rocks of most of he eastern United States, and the deposition of the lowest Pennsylvanian. )uring this interval there was a great continental revolution. The greater >art of the eastern United States was elevated above the level of the sea in diich most of the Mississippian rocks had been laid down, and dry land con- itions existed where formerly the sea had prevailed. After a very long time he land was worn down nearly to sea level and the Pennsylvanian or "Coal Measures" rocks began to be deposited. Such great changes in the earth's istory are considered as dividing lines between periods, just as great changes q human history serve to divide it into periods and epochs. Further discus- ion of this subject is given on later pages. The Mississippian system has been named from the Mississippi Valley /here it is typically developed. The rocks of the system in this typical region all naturally into a lower and an upper division. The lower division is made ip predominantly of limestones with some shales and a very minor amount f sandstone; the upper Mississippian of the same region is much more hetero- ;enous in its composition, limestones are much less conspicuous than in the ower division, those that are present are commonly much more shaly in haracter, and there are important sandstone formations included in the series, throughout the entire Mississippian the limestones are highly fossiliferous, 'eing filled in many places with a great variety of fossil remains of many ifferent kinds of marine animals. The present chapter will be devoted to a consideration of the Lower ■lississippian formations, the following of which are recognized in Hardin bounty, from the lowest to the highest : Osage formation, Warsaw limestone, \t. Louis limestone, and Ste. Genevieve limestone. 91 92 geology of hardin county Osage Formation name and distribution The Osage formation is widely distributed through the Mississippi Valley, where it is divisible into several members. In southeastern Iowa it is repre- sented by the Burlington and Keokuk limestones, and the same members can be more or less distinctly recognized in a southerly direction in Mississippi River sections, and in the sections which encircle the Ozark region of southern Missouri. In northwestern Arkansas and adjacent portions of Oklahoma and Missouri the beds representing the formation have commonly been designated as the Boone chert, named from the exposures in Boone County, Arkansas. 1 In southeastern Missouri and in the adjacent part of Illinois a basal member of the formation has been called the Fern Glen limestone 2 , and the same mem- ber extends into northern Arkansas where it constitutes a part, at least, of the beds that have been called St. Joe marble by the Arkansas Geological Survey. 3 To the eastward the Osage is represented by the "Waverly group" of Ohio, and by the a Knobstone" of Indiana. In Tennessee and Alabama, beds of the same age have been called the Fort Payne chert by Hayes 4 and others. In Hardin County, Illinois, and in Kentucky, strata of Osage age were called the Tullahoma formation by Ulrich 3 and by Bain , although that name had been used some years ealier by Safford 7 in Tennessee. In Hardin County, Bain used the name Tullahoma to include all the beds between the black shale and the St. Louis limestone, including, therefore, the beds referred to the Warsaw in this report; but as originally applied by Safforcl in Tennessee it is uncertain: whether the Tullahoma includes beds of Warsaw age. In the older literature of Iowa, Illinois, and Missouri, the names Bur- lington and Keokuk limestones were used almost exclusively for these beds, and in most of the area of outcrop in these states these two members can be readily distinguished. As observations were extended, however, it was found to be difficult to establish with certainty in all sections the line of demarcation between these two limestones, and because of this Williams 8 proposed the name Osage for the larger unit which should combine the Burlington and Keokuk limestones and their equivalents throughout the Mississippi Valley. The name Osage was derived from Osage River in Missouri. Keyes, likewise, in his lowaj work, recognized the desirability of uniting these two limestones, and used the name Augusta 9 with the same significance as Osage. The names Boone, Tullahoma, and Fort Payne were all proposed originally as strictly local I 1 Simonds, Ark. Geol. Surv., Ann. Rep. 1888. vol. 4, p. 27 (1891). 2 Weller, Trans. Acad. Sci. St. Louis, vol. 16, p. 438 (1906). 3 Hopkins, Ark. Geol. Surv., Ann. Rep. 1890, vol. 4, p. 253 (1893). 4 Bull. Geol. Soc. Amer., vol. 2, p. 143 (1890). 5 IT. S. Geol. Surv., Prof. Paper 30, p. 26 (1905). C U. S. Geol. Surv., Bull. 255, p. 19 (1905). 7 Elements of Geol. of Tenn., p. 143 (1900). 8 U. S. Geol. Surv., Bull. 80, p. 169 (1891). 9 la. Geol. Surv., vol. 1, p. 59 (1893). LOWER MISSISSIPPIAN 93 )rmation names without any adequate understanding of the broad correlation f the strata, while Osage was proposed as a comprehensive stratigraphic unit aving a definite position in the Mississippian section of the Mississippi Valley >gion, and consequently it is adopted here for use in Hardin County. In Hardin County the Osage formation outcrops only in an oval belt of mntry, from one-half to over one mile wide, encircling the area of the •evonian limestone and the belt of Chattanooga shale of the Hicks dome. The >wn of Hicks is located upon the outcrop of the formation. The chert of le Osage is more resistant to decay and removal than the strata on each side ; msequently its outcrop forms a broad ridge which incloses the valleys in hich the outcrops of the Chattanooga shale are situated. The strata of the sage ridge dip away from the Hicks dome in every direction at angles of ■om 10° to 20°. This ridge is deeply covered with chert debris made up of i-agments generally less than six inches in diameter. The principle exposure f the Osage in place in Hardin County, is along the bed of Hicks Branch in le southwestern part of sec. 25, T. 12 S., E. 7 E., where the stream has oded a deep valley through the ridge. Along the bed of this stream there is nearly continuous exposure of the formation from top to bottom. The bed 'Ck of the Osage is exposed at only two other points in Hardin County, and t each of these only a few feet can be seen. LITHOLOGIC CHARACTER The Osage formation, so far as it is exposed to examination in Hardin bunty, is almost exclusively a chert bedded in layers up to one foot thick, he only exception to this condition that has been observed, is on Goose Creek ; sec. 32, T. 12 S., E. 8 E., about one-fifth mile east of the west line of the etion, where a few feet of black limestone is exposed, passing into chert as |e beds extend upward into the creek bank. As exposed in Hicks Branch e chert is apparently a siliceous replacement of evenly bedded limestone, he chert fragments as they occur strewn over the surface of the ground have me distinctive characters, generally being grayish, yellowish, or bluish in lor. Much of the chert is in elongated fragments with rudely triangular oss sections, and with smooth lateral faces. Much of it is also finely color- .nded in gray, pink, or yellow, the bands being either concentric, as in some the loose pieces, or parallel with the bedding, as in the chert layers in place. texture the chert ranges from compact, brittle, or tough, through somewhat tous and tough to finely porous and somewhat friable. Some of the hard, nse chert seems to be minutely cracked so that it may be shattered easily to small angular fragments. A notable characteristic of these Mississippian erts is the almost total absence of fossils, which is a great contrast to the ert of similar age elsewhere in the Mississippi Valley region where much it is crowded with crinoid plates and other fossils. . Isolated stem plates 94 GEOLOGY OF HARDIN COUNTY of crinoids have been observed rarely in Hardin County, and a very few other fossils, but most of the chert is entirely without fossils of any sort. The finely porous, more friable chert resembles a fine-grained sandstone and may have been originally a siliceous limestone in which the silica occurred in the form of very fine quartz grains. Although the Osage manifests itself on the surface as chert, this chert is quite likely only a superficial development, and the formation is probably a true limestone below the depths to which water and air have had free access and circulation — in other words, below the zone of active weathering. Such superficial change of limestone to chert, by the solution of the carbonate of lime and the deposition of silica in its place, is a common process in nature. The only direct evidence, however, that the Osage is essentially limestone in its original condition, is the outcrop of fine-grained black limestone about ten feet thick, in the bank of Goose Creek, which has been described above. At this locality the change from the limestone to chert upward from the chert bed, is exhibited. It is probable that a boring through the Osage where it is 500 feet below the surface would show little but limestone. There is no evi- dence of any shale in the formation as was asserted by Bain. 1 THICKNESS The only locality in the county where an estimate of the thickness of the Osage can be made, is on Hicks Branch, at the locality described on an earlier page, where the formation is so well exposed. At this place only the edges of the strata are shown in the creek bed, so that the thickness cannot be measured directly, but must be calculated from the width of the outcrop and the average angle of dip. The width of ,the outcrop at this locality is 2,000 feet, and th( average dip is 16° SW., computed from 17 independent measurements well distributed across the outcrop, which makes the thickness 550 feet, this being the best estimate that can be made under the conditions, and being nearly three times as great as that recorded by Bain. 2 PALEONTOLOGY As already stated, the Osage in Hardin County is nearly unfossiliferous, in this respect being in great contrast to its character in the Mississippi Valley and in Alabama and eastern Tennessee. The Hardin County locality occupies a position between the other two regions mentioned, and it is a remarkable circumstance that in this intermediate region fossils should be so scarce. A possible explanation of the fact may be that the crinoids, bryozoa, and brachiopods, so abundant in the Osage and its equivalents elsewhere, were especially adapted to life in shallow, warm water, such as doubtless existed 'U. S. Geol. Surv., Bull. 255, p. 19 (1905). 2 Loc. cit., p. 19. LOWER MISSISSIPPIAN 95 generally throughout the Osage sea. The Hardin County area may have been a deeper depression of the sea floor where the water was too deep for the animals to live upon the bottom, and into which their hard parts were not washed by currents from shallow water. The few fossils that have been col- lected from the formation, as the result of constant vigilance and persistent search are listed below : List of fossils from the Osage formation Productus burlingtonensis Hall? Solenomya ? sp. Productus setigera Hall Sphenotus ? sp. Rhynchopora ? sp. Phanerotrema ? sp. Aviculopecten sp. Griffithides sp. Posidonomya ? sp. Most of the material, as appears from the list, is too poorly preserved and too scanty for even certain generic identification, although there is probably but little ground for doubt on that point. There is nothing especially signi- ficant for stratigraphic determination in the list. All the genera are known to occur in the Osage, and most of them have a longer range, but none of the forms recorded throw any light upon the depth of the water or other conditions in the region during Osage time. CORRELATION Because of the scarcity of fossils the identification of the Hardin County rocks in question as Osage, must rest mainly upon their stratigraphic rela- i tions. They occupy the same position between the Chattanooga shale and the ; Warsaw limestone, as do the Burlington and Keokuk limestones of the Mis- sissippi Valley; also the same position above the Chattanooga shale as the Fort Payne in eastern Tennessee and Alabama. Moreover, by its fossils in Tennessee and Alabama, the Fort Payne is to be correlated with the Keokuk ; at least, and it possibly includes beds equivalent to all or some part of the Burlington limestones, as does the Tullahoma formation in central Tennes- I see. On general grounds, therefore, there seems to be no reason to doubt the ! correctness of the correlation of the Hardin County chert formation with the Fort Payne to the east and the Keokuk-Burlington to the west. Its thick- ness is much greater than that of the Fort Payne, and is about "that of the Osage group of southern Indiana and the Louisville region of Kentucky, 1 where it is composed of the New Providence shale below, the Kenwood sand- stone in the middle, and the Rosewood shale and Holtsclaw sandstone above, aggregating 600 feet. On the other hand the Osage of Hardin County is much thicker than the Keokuk and Burlington limestones of the Missis- j sippi Valley, which have a maximum combined thickness of only 225 feet. From these circumstances it appears reasonable to suppose that the Osage of this region includes the equivalents of both the Keokuk and the Burlington. 96 GEOLOGY OF HARDIN COUNTY However, there is one item of evidence tending to show that the Osage of Hardin County includes only beds of Keokuk age and is nearly equivalent to the Fort Payne chert. Seven miles north of Hardin County and just west of the Saline-Gallatin county line, near the north entrance to Horseshoe Gap three miles southwest of Equality, the Osage outcrops on a conspicuous knoll and is the same in character and thickness as in Hardin County. Below the Osage at this locality the Chattanooga shale also outcrops as re- vealed by fragments of the characteristic, black, fissile shale; but between the Chattanooga and the Osage is a few feet of dark, irregularly fissile shale with peculiar curly markings like worm trails. Now these peculiar markings are characteristic of the Eosewood shale of the Louisville region, Kentucky, of determined Keokuk age. They are widely distributed throughout all the region from Crawfordsville, Indiana, to Overton and Davidson counties, Tennessee, where they invariably occur in shale which is everywhere, by all criteria, to be correlated with the Rosewood. In other words, such evidence as is known indicates that the Osage chert at the locality described above, has beneath it shale of Keokuk age and therefore includes no component of Bur- lington age, and the condition in Hardin County would presumably be the same. Meramec Group By Stuart Weller As originally defined, the Meramec group 1 included in ascending order, the Warsaw limestone, the Spergen (Salem) limestone, and the St. Louis limestone. In the present report the Ste. Genevieve limestone is also included as a fourth formation in the group, above the St. Louis limestone. For many years after the original definition of the Ste. Genevieve limestone by Shumard in 1859, the formation was scarcely or not at all recognized, the beds representing it being commonly included with the St. Louis limestone. LTlrich revived the name of the formation in 1905, but excluded it from the Meramec group, making it the lowest formation of the Chester group, an interpretation that he still holds to at this time. All the evidence, however, that has been gathered during a period of a dozen years, shows the much closer relationship of the Ste. Genevieve to the St. Louis limestone than to any of the Chester formations. In a number of publications during recent years, as a compromise measure, the Ste. Genevieve has been treated as a separate unit in the Mississippian system, being united with neither the Meramec group below, nor the Chester group above, but the time has now come when it is desirable to place the Ste. Genevieve limestone where it properly belongs, with the St. Louis limestone. In order to accomplish this iUlrich, U. S. Geol. Surv., Prof. Paper No. 24, table opposite p. 90 (1904); also, U. S. Geol. Surv., Prof. Paper No. 36, p. 28 (1905). LOWER MISSISSIPPIAN 97 result it is necessary either to modify the original definition of the Meramec group so that it shall include the Ste. Genevieve limestone, or to propose a new group name for the four formations. As it seems more desirable to retain the name already in use with a slightly modified definition, than to add confusion to the nomenclature by introducing a totally new name, such a procedure is consequently followed here. Warsaw Limestone name and distribution Throughout the Mississippi Valley, wherever the Osage is present, it is succeeded by the Warsaw formation. The formation is also present in south- western Indiana and southward into Kentucky, where it has been called the Harrodsburg limestone, and extends southward through Kentucky and Ten- i nessee and probably into northern Alabama. The name of the formation is derived from the town of Warsaw in western Illinois, situated on Missis- sippi Eiver nearly opposite Keokuk, Iowa. As originally used the name was applied only to beds which overlie the so-called "geode beds" of the section at Keokuk and Warsaw, these lower beds being included in the Keokuk formation. More recent studies have shown that the true line of demarcation between the Keokuk and Warsaw is more properly at the base of the "geode .bed" instead of at the summit of it, and such an emended definition of the Warsaw formation has been accepted by the United States Geological Survey. In Hardin County the Warsaw comes to the surface only in an oval belt encircling the Hicks dome, and varying in width from one-fourth of a | mile to one mile. One extremity of this belt, along its longer axis, lies just northwest of Hicks. The opposite extremity crosses Goose Creek at Wolrab Mill. It lies along and east of Big Creek in sees. 28 and 33, T. 12 S., K. 8 E., and passes diagonally through sec. 36 and across the southwest corner of sec. 26, T. 12 S., E. 7 E. The Warsaw in this region is denned as including all the beds lying between the continuous, non-fossiliferous chert of the Osage below, and the lowest limestone beds carrying the fossil coral Lithostrotion above, the beds with Lttliostrotion being assigned to the St. Louis limestone. Ordinarily the bounds between the Warsaw and Osage can not be determined with certainty. On Hicks Branch in the southwest quarter of sec. 25 the boundary is placed between the continuous chert mass of the Osage below and the almost continuous limestone mass, regarded as Warsaw, above. There are no exposures at or near this contact elsewhere in the county, and the position of the boundary has been roughly approximated in the mapping of ;the formation; but it is reasonably certain, from fossil evidence, that the ^boundaries as located are correct to within narrow limits. —7 G 98 GEOLOGY OF HARDIN COUNTY LITHOLOGIC CHARACTER Approximately the lower three-fourths of the Warsaw formation, as exposed on Hicks Branch in the SW. % sec. 25, T. 12 S„ B. 7 E, is a dark to black, fine-grained, somewhat cherty and sparingly fossiliferous limestone. The upper one-fourth contains a considerable proportion of coarse-grained, light-gray limestone in thick layers, intercalated with which are layers of shale and layers of dark, shaly-weathering limestone. Much of the lower main body of the Warsaw is well exposed in the Hicks Branch locality just mentioned. The limestone layers are commonly a foot or two in thickness, the rock is mostly dark to black, fine-grained, and includes a considerable amount of chert in rough, irregularly-sprawling, flatfish masses. This part of the Warsaw is hardly different lithologically from much of the overlying St. Louis limestone. On long-continued weathering near the surface it is largely silicified to brittle, fossiliferous chert, a good exhibition of which is present in the road ditch just north of Hicks. This chert can be distinguished from the similarly bedded Osage chert by its fossils, which are fairly numerous, while the Osage chert is essentially non-fossiliferous. The uppermost fourth of the Warsaw differs from the lower part of the formation in the presence of thick-bedded, coarsely granular, predominantly light-gray limestone, but with some dark gray layers, some dark, fine-grained to almost non-granular layers having a conchoidal fracture, and thick black layers weathering to thin laminae. The general character of this part of the Warsaw is exhibited in an exposure on the bank of Big Creek in the northern part of sec. 33, T. 12 S., B. 8 E. At this locality the following succession of beds is exhibited: Section of part of the upper Warsaw beds on Big Creek in the northern part of sec. S3, T. 12 8., R. 8 E. Feet Limestone, dark gray, coarse-grained (crinoidal) Limestone, thick-bedded, weathering shaly Limestone, light-gray, coarse-grained (crinoidal) Limestone, dark, thick-bedded, weathers shaly, Bryozoa plentiful 3j 40 An exposure of about 40 feet of limestone at this horizon is present on the southern point of the spur in the SW. l/ 4 sec. 5, T. 12 S., R. 8 E., where the road crosses on the line between sees. 5 and 8. It is mainly dark-gray, and coarse-grained, but the dark, non-granular limestone is also present. The limestone exposures of this part of the Warsaw are commonly of the light- gray, coarse-grained sort. The rock is highly fossiliferous and its granular character is largely due to the abundance of crinoidal plates and to fragments LOWEK MISSISSIPPIAN 99 of other fossils which have been altered to crystalline calcite. Layers of this coarse, light-gray, crinoidal limestone are well exposed in the road ditch at Wolrab Mill. The weathered surface is very rough by reason of the pro- jecting fragments of fossils, mostly more or less broken crinoid plates, and shells of various sorts. Much of the limestone, on extreme weathering, has become altered to a coarsely porous, reddish chert resembling a red rubber sponge. Dense, platy, fossiliferous, gray chert also occurs, and is especially abundant on the long, low spurs in the southwest corner of sec. 5, T. 12 S., E. 8 E., and on the slope above the road in the extreme southwest corner of sec. 33, T. 12 S., B. 8 E., about one-fourth of a mile northeast of Wolrab Mill. In some situations, where not; silicified, these coarse-grained layers form, on weathering, a white, chalky crust as much as one-eighth of an inch thick. THICKNESS As deduced from the dip and width of outcrop on Hicks Branch in the SW. % sec. 25, T. 12 S., E. 7 E., the Warsaw is 250 feet thick. PALEONTOLOGY Fossils are fairly abundant in the Warsaw, although they are: not as a rule well preserved, and good specimens are not common. Most of them occur in the upper fourth of the formation which is characterized by the coarse- grained, light-gray limestone, but in addition to these much of the limestone throughout the formation is crowded with fenestellid bryozoans. The fossils are developed from their original limestone matrix and revealed to view only in the chert. A list of the species that have been identified is as follows : List of fossils from the Warsaw limestone Triplophyllum, a small slender species; common Pentremites conoideus Hall Platycrinus, rather large species with oval, spiny stem plates one-half inch in diameter or larger; com- mon; entire bases indicating heads y 2 to 1 inch in diameter occasion- ally found Cystodictya lineata Ulrich Cystodictya pustulosa Ulrich Dichotrypa sp. Fenestella tenax Ulrich Hemitrypa proutana Ulrich Lyropora sp. Polypora simulatrix Ulrich Athyris densa Hall; common Brachythyris suborbicularis (Hall)? Brachythyris subcardiiformis (Hall)? Camarotoechia mutata (Hall)? Cleiothyridina hirsuta (Hall) Eumetria verneuiliana (Hall) Rhipidomella dubia (Hall) Spirifer bifurcatus Hall Spirifer lateralis Hall Spirifer subequalis Hall Spirifer tenuicostatus Hall; common Spiriferina subtexta White? 100 GEOLOGY OF HABDIN COUNTY CORRELATION Several of the species recorded in the foregoing list are known elsewhere only from the Warsaw and Spergen limestones. Snch are Pentremites conoideus, Athyris densa, Brachythyris subcardiiformis, and Spirifer lateralis. Athyris densa has been recorded as a Spergen limestone fossil but the original specimens were collected in Indiana and Kentucky from beds that are now known to be Harrodsburg limestone, of Warsaw age, and there is no authentic record of its occurrence in the Spergen, or in any horizon other than the Harrodsburg limestone or its equivalents. Spirifer lateralis and Spirifer sub- equalis are not known below the Warsaw, nor are they reported from above the Spergen. Poly pom simulatrix is reported from the Keokuk and Warsaw but not from higher beds. Altogether the assemblage seems to be character- istic of the Warsaw and justifies the classification here made. It is possible that the upper fourth of the 250 feet of limestone here classed as Warsaw — the part with the coarse, light-gray, fossiliferous layers — may represent the Spergen, but there is no compelling evidence for such a determination. Athyris densa, a supposed Spergen species, is common in this part of the Warsaw, but as has already been explained the species is actually known elsewhere only in the Harrodsburg limestone of Warsaw age. St. Louis Limestone By Stuart Weller NAME AND DISTRIBUTION The St. Louis limestone has received its name from St. Louis, Missouri, where the formation is typically developed. It is well exposed in the Missis- sippi River bluffs and in quarries in that city, and extensive exposures are present above and below the city upon both the Missouri and Illinois sides of the river. The formation is recognized as far north as southeastern Iowa, and southward along the Mississippi to Ste. Genevieve County, Missouri, and still further south in Illinois. Across southern Illinois the formation out- crops at intervals from Union County to Hardin County, and continues into Indiana, and southward into Kentucky and Tennessee. The same formation extends to Alabama and to other localities in the Appalachians north of Alabama. Throughout this entire area the St. Louis limestone originally existed as a continuous, blanket-like formation, and was deposited as a mass of calcareous sediment in the bottom of a sea which covered the area, and which doubtless spread beyond the limits of the formation as now exposed. The formation is not continuously present at the surface throughout this area. In the larger part of the region it is covered by younger formations, and its presence could only be detected by deep excavations or by deep drilling. Elsewhere within the specified area, rock formations older than LOWER M1SSISSIPPIAN 101 the St. Louis limestone are exposed at the surface, and in such situations the St. Louis formation which was originally present, has been removed through the process of erosion. In Hardin County the St. Louis limestone occurs in each of the three structural areas. In the southeastern, non-faulted portion of the county, the formation occupies a considerable area between Elizabethtown and Cave in Rock. Between these two points it is well exposed in the Ohio Eiver bluffs. Fig. 4. — Bluff of St. Louis limestone at Tower Rock, midway between Elizabethtown and Cave in Rock. and extends northward from the river bank for an extreme distance of about two miles. Farther to the north the formation continues beneath the Ste. Genevieve limestone, so that the northern limit of the area of its outcrop is the sinuous line representing the intersection between the St. Louis-Ste. Genevieve contact surface and the irregular topographic surface. The loca- tion of this line upon the map is liable to some error because of the difficulty 102 GEOLOGY OF HARDIN COUNTY of determining just where the line between these two formations should be placed, and because of the covering of surficial material which so effectually obscures the rock outcrops over much of the area. In the central, much-faulted belt of the county, the St. Louis limestone outcrops in two distinct areas, both of them in the northwestern, less depressed segment of the belt (see figure 3). One of these is situated along the valley of Hogthief Creek, extending nearly two miles northeast of Jacksons Store and about one and one-half miles southwest of the same point. The extreme width of the area does not exceed one mile. On the southeast it is limited by one of the long northeast-southwest faults of the belt, being adjacent to the Pottsville and higher Chester formations. The northwestern border of the area is a sinuous line of normal contact with the overlying Ste. Genevieve limestone. The second area of St. Louis limestone within the faulted belt is in the valley of Big Creek, above and below Illinois Furnace, and in some of its tributaries from the west. It is limited both to the northwest and to the southeast by faults. The extreme length of the area is approximately five miles in a northeast-southwest direction, southwest from a point about one mile northeast of Illinois Furnace. The width of the area increases from about one mile at its northernmost extremity to nearly one and one-half mile at the south. In the northwestern part of the 'county the St. Louis limestone occupies an incomplete belt about the Hicks dome, the width of the belt being approx- imately one-half mile. To the southeast the continuity of the belt is destroyed by the Wolrab Mill fault, so that the area of outcrop is U-shaped rather than elliptical, which is the outline of the outcropping belts of the next three under- lying formations. LITHOLOGIC CHARACTER In its general characters the lithologic features of the St. Louis lime- stone are similar throughout the entire area of its distribution. In Hardin County, where it is not notably different from other localities, the upper 75 to 100 feet is in the main gray, bluish-gray, or blue, while a large part bf the lower 250 to 300 feet is dark or nearly black limestone. For the most part it has been deposited in even beds varying in thickness from a few inches to several feet. In the river sections half a mile west of Cave in Bock, the bluish-gray layers constitute the upper 75 feet. The 200 feet below, exposed to the west along the river as far as Tower Rock, midway between Cave in Bock and Elizabethtown, is predominantly dark gray or black. It is predominantly dense, and fine-grained in texture, a few beds being nearly as fine-grained as lithographic stone. Other beds are more granular and some are coarsely crystalline. No oolitic beds have been recognized with certainty in the St. Louis limestone of Hardin County. Most of the beds are hard and tough, but those of lithographic texture are brittle and break with a LOWER MISSISSIPPIAN 103 conchoidal or splintery fracture. Much of the limestone of the formation contains considerable amounts of chert, occurring commonly as lenticular or irregular masses distributed along horizontal lines parallel with the bedding planes. The chert is still more conspicuous in the residual deposits remaining after the decomposition of the limestone. In the limestone itself the chert is more abundant near the surface, and is clearly of secondar}^ origin being formed by the replacement of limestone by silica; and it is not unlikely that the original source of much of the silica has been the unconsolidated surficial deposits through which the ground waters have percolated before coming in contact with the solid limestone beneath, although some of the silica may have been disseminated originally through the limestone. Fig. -Bluff of St. Louis limestone two miles west of Cave in Rock. In one respect, that is in its color, the St. Louis limestone of Hardin County and the adjacent area, is distinctly different from the exposures of the same formation in the counties bordering Mississippi River. In the region here being described the formation is in general darker in color than it is farther west, the dark color being more marked in the lower part of the formation, where most of the beds are nearly black. In places these dark beds which may be a foot more or less in thickness, are separated by thin, somewhat shaly layers, and the upper and lower parts of the solid layers themselves cleave easily and somewhat irregularly along the bedding planes, the mid-portions of the layers being much more dense and compact. 104 GEOLOGY OF HARDIN COUNTY STKATIGRAPHIC RELATIONS Sedimentation was apparently continuous from the subjacent beds into the St. Louis limestone. The lithologic change is gradual with no sharply defined line of demarcation, but rather with a gradual transition from the lower sediments to the higher, so that the lower boundary line of the forma- tion has been drawn in a somewhat arbitrary manner. The line between the St. Louis limestone and the overlying Ste. Genevieve is also an indefinite boundary/ and no evidence has been found in Hardin County to show that any break in sedimentation occurred in passing from the older to the younger formation. There is, however, a distinct unconformity between these two formations farther west, in Ste. Genevieve County, Missouri, on the flank of the ancient Ozark land, the unconformity representing one of the many uplifts experienced by Ozarkia in Paleozoic time. THICKNESS It has not been practicable to make a complete, measured section of the St. Louis limestone anywhere in Hardin County, and it is commonly im- practicable to recognize the exact line of demarcation between the formation and either the Warsaw below or the Ste. Genevieve above. Because of these circumstances the determination of the thickness of the St. Louis limestone can only be an approximation. The formation must include, however, about 350 feet of strata. This thickness is approximately that which is exhibited around St. Louis, Missouri, where the formation is typically developed, but it is considerably greater than the thickness in Ste. Genevieve County, Missouri, where less than 100 feet is usually exhibited, or in southeastern Indiana. PALEONTOLOGY Good fossils are not commonly abundant in the St. Louis limestone in Hardin County, except locally. Crinoidal fragments, mostly stem segments, are numerous, being most conspicuous in the cherts of the formation where their presence is shown by the cavities from which the fossils themselves have been removed by solution. In the silicification of these limestones the matrix was easily replaced while the crystalline calcite of the crinoid frag- ments and other fossil shells was not acted upon. Later, upon weathering, these crystalline calcite portions of the rock were removed by solution, leaving the residual cherts more or less porous. In places some layers of the St. Louis limestone contain an abundance of bryozoans, largely fenestellid forms, which may be seen in section upon the weathered surfaces of some of the hard limestone layers, but no shaly beds with well preserved bryozoan specimens, such as are present in places in some of the outcrops along Mississippi River, LOWER MLSSISSIPPIAN 105 have been met with in Hardin County. The only fossils of the formation that have been observed anywhere in the county in abundance and in a good state of preservation, are the colonies of the coral genus Lithostrotion. Two species of this coral are present in the fauna. Near the base of the formation there is present, locally at least, and perhaps wherever this part of the forma- tion is exposed, a very conspicuous coral bed in which large colonies of L. proliferum occur in great numbers. One of the best exposed outcrops of this coral zone is along the road running southwest from Illinois Furnace, in the SW. % SE. % sec. 8, T. 12 S., E. 8 E., where it occupies a thickneiss of eight or ten feet of limestone. An entirely similar coral zone, characterized Pig. 6. — Layer of St. Louis limestone, about 75 feet below the top of the forma- tion, with the fossil coral Lithostrotion proliferum; river bank just above low water, three-fourths of a mile west of Cave in Rock, looking east. by the same species, is widely present near the base of the St. Louis limestone in the outcrops of the formation on both sides of Mississippi Eiver south of St. Louis. This coral species also occurs at other horizons in the formation in Hardin County, and one of these higher beds, crowded with large heads of the branching coral, is exposed in the ledges on the river bank at low water, about three-fourths of a mile west of Cave in Eock (fig. 6). Another species of the same genus, Lithostrotion canadensis, is more commonly met with in the higher than in the lower beds. The species has not been observed in any one bed occurring in such numbers, however, as L. proliferum in the coral zones mentioned above. The colonies are more or less scattered, and in "LOG GEOLOGY OF HARDIN COUNTY places considerable thicknesses of limestone over large areas may be wholly barren of them. The most common manner of occurrence of the species is in the form of loose,, silicified colonies in the chert residuum overlying the limestone. These two species of Litliostrotion are commonly considered as good index fossils of the St. Louis limestone, and in Hardin County, at least, neither of them has been observed in any other formation. Ste. Genevieve Limestone By Stuart Weller NAME AND DISTRIBUTION The Ste. Genevieve limestone was first differentiated by Shumard 1 although no adequate description of the formation was given by him. The name of the formation was taken from Ste. Genevieve, Missouri, the type locality specified by Shumard being in the Mississippi Eiver bluffs a mile or tw T o below that town. For many years following the original definition of the formation, both the name and the formation itself were overlooked, and the beds were commonly included in the St. Louis limestone. When Ulrich entered upon the investigation of the lead, zinc, and fluorspar region of western Kentucky, he differentiated a formation lying above the St. Louis limestone which he designated as Princeton limestone. 2 Later, a visit to Ste. Genevieve convinced him that his Princeton limestone in Kentucky, and the Ste. Genevieve limestone of Shumard were the same, and in his final report he adopted the older name of the formation 3 , and gave a much fuller description of it than had been published elsewhere. As the formation is known at the present time it has a wide geographic distribution from Iowa to Alabama. In Iowa it is represented by the so- called Pella beds 4 , originally defined by Bain as a subdivision of the St. Louis limestone of that state. The formation is exposed at many points in the Des Moines Valley from Fort Dodge to the southeastern corner of the state. The Ste. Genevieve limestone is well exposed in the Mississippi Eiver bluffs at Alton, Illinois, and at intervals on both sides of the Mississippi to Ste. Genevieve County, Missouri, and to Union County, Illinois. It is exposed at various localities in the southern counties of the state, from Union to Hardin, and extends into Kentucky. East of the coal fields the formation again appears both in Indiana and in Kentucky. In Hardin County the Ste. Genevieve occupies a larger areal extent than any other formation, the total area being about 35 square miles, the forma- tion most nearly approaching it in extent being the Pottsville. The areas of 1 Trans. St. Louis Acad. Sci., vol. 1, p. 406, (1859) ; Mo. Geol. Surv., Rept. for 1855- 1871, p. 293 (1873). 2 Crittenden Press, Dec. 1890; Bull. U. S. Geol. Surv., No. 213, p. 207 (1903). 3 Prof. Paper, U. S. Geol. Surv., No. 36, p. 39 (1905). 4 Amer. Geol., vol. 15, p. 318, (1895) ; la. Geol. Surv., vol. 5, p. 150, (1896). LOWER MISSISSIPPIAN 10? outcrop of the formation are widely scattered, and occur in each one of the three structural regions that have been described (see figure 3). In the south- eastern part of the county the formation is exposed in the Ohio Eiver bluffs east of Cave in Eock. About one mile east of the town it is present in the upper part of the bluff, but the eastward dip of the strata brings the forma- tion down to river level in a comparatively short distance. The exposures continue eastward for about two miles, to the point of the bluff at about the middle line of sec. 17, T. 12 S., E. 10 E., except where it is interrupted by the rather broad valley from the northwest, the mouth of which is about one and one-half miles east of Cave in Eock. From this outcrop in the Ohio Eiver bluffs the belt occupied by the formation extends in a northwesterly or nearly westerly direction to the first Peters Creek fault, the extreme width of the outcrop being a little over two miles. In a northeasterly direction this area of Ste. Genevieve passes beneath the overlying Chester formations along the line of bluffs that extend southeastwardly from sec. 33, T. 11 S., E, 9 E., to Ohio Eiver two miles east of Cave in Eock. Along its northwestern border this area is limited by the first of the northeast-southwest Peters Creek faults that bounds the central faulted zone of the county. In the central faulted zone the Ste. Genevieve limestone occupies a large portion of belt number three (see figure 3), extending continuously from the gap in the hills on the line between sees. 6 and 7, T. 12 S., E, 9 E., to Eosiclare. In its northern part this area is very narrow, occupying scarcely one-fourth of a mile, but it broadens southwestwardly, becoming one and one- half miles wide where Big Creek crosses the belt. From its northeastern extremity nearly to Big Creek the limestone passes beneath the Chester forma- tions in a northwestern direction, in normal contact, but beyond this point it is in fault contact with Chester formations. At the extreme southwestern extremity of the belt the Ste. Genevieve again passes beneath the Chester formations in normal contact, in the hill on the Ohio Eiver bank just below Eosiclare. A Avedge-shaped block, obliquely truncated at its northeastern, narrower extremity at the line between sees. 7 and 18, T. 12 S., E. 9 E., and extending to Ohio Eiver between the mouth of Hosick Creek and Elizabeth- town, is made up entirely, at the surface, of Ste. Genevieve limestone. The higher Ste. Genevieve limestone of the block is, at its northern extremity, continuous to the east with the lower part of the same formation in the south- eastern, unfaulted portion of the county. In belt number two of the faulted zone the Ste. Genevieve limestone is present as the surface rock in two small areas only, the entire space occupied being less than one-fourth of a square mile. The most extensive of these exposures is east of Stone School, along the road to Elizabethtown. The greater part of the area is occupied by the Eosiclare sandstone which has a broad exposure upon the dip slope, but the Fredonia limestone is also exposed 108 GEOLOGY OF HARDIN COUNTY in the hill slope north of the road. The only other fault block in this belt in which the Ste. Genevieve is exposed, lies one-half mile north of the iron bridge over Big Creek between Elizabethtown and Eosiclare. A large part of the block is occupied by the formation, but it is covered for the most part with the alluvial deposits of the Big Creek valley, but the Fredonia is exposed in the bank of the creek in the southwestern part of the block, and a single outcrop of Eosiclare sandstone has been observed towards the northwestern boundary of the block. In belt number one of the faulted zone the Ste. Genevieve occupies two areas, one on each flank of the arch, with an area of older, St. Louis lime- stone between. The northern of these areas extends from the fault line just northwest of Keelin School, in a northwesterly direction across Hogthief Creek to Illinois Furnace; from here it extends in a northeasterly direction north of Hogthief Creek, beyond Pankeys Store, to the east side of sec. 3C, T. 11 S., E. 9 E. Northward from Pankeys Store the formation extends to the Wolrab Mill fault in sec. 27, T. 11 S., E. 8 E. On the southern limb of this arch the Ste. Genevieve occupies an area lying east, north, and west of St. Joseph School, and continues southward on the west side of Wallace Branch, past Melcher and Eich Hills, to the Ohio Eiver bluffs. In the northwestern part of the county the Ste. Genevieve partly encircles the Hicks dome, being cut off to the southeast by the Wolrab Mill fault. Southwest of the dome towards Eichorn, the width of the Ste. Genevieve belt is approximately two miles, while northwest of Hicks it is less than one-half mile. It broadens out again northeast of the dome, but in that direction there is an offset in the outcrop caused by the Lee fault. Within these areas of outcrop of the Ste. Genevieve formation, by far the greater portion is underlain by the Fredonia limestone which is the thick, lower member of the formation. The best exhibition of the Fredonia member of the Ste. Genevieve formation in Hardin County is in the Ohio Eiver bluffs just above Eosiclare, where the limestone has been extensively quarried. The upper portion of the quarry at Jacks Point,, one mile below Elizabethtown, is also in the Fredonia. The basal, cherty beds of this member are finely exhibited in the river bluff at Elizabethtown in front of the Eose Hotel, and for a mile to the east along the river bank. In the northwestern part of the region mapped, the best exhibition of the Fredonia is on Hicks Branch and Baldwins Eun, just across the line in Pope County, in the eastern part of the NE. % sec. 27, T. 11 S., E. 7 E ; . In the southeastern part of the county good exposures of the Fredonia may be seen in the river bluffs east of Cave in Eock, and a good exposure of a large thickness of the same limestone may be seen at the road corner near Love School, two miles north of Cave in Eock. On Peters Creek, in the NW. % sec. 8, T. 12 S., E. 9 E., twenty feet or more of dark limestone is exposed that in color and texture is very like LOWER MISSISSIPPI AN 109 the St. Louis limestone, but is believed to be a locally developed dark facies of the Fredonia. The upper 100 feet of the Fredonia is well displayed on the west side of Lead Hill in the NW. % sec. 4, T. 12 S., E. 9 E., and the Eosiclare sandstone and overlying "Lower Ohara" are well exposed near and on the top of the same hill at its north end. The Eosiclare sandstone outcrop is commonly a very narrow belt, but in the southwestern part of the county, this thin sandstone member is the surface formation throughout areas of considerable extent in the neighborhood of Eichorn and southward from that place for about three miles, where the dip of the rock strata nearly conforms with the slope of the surface. Another area where this sandstone underlies a dip slope is largely in sec. 18, T. 11 S., E. 8 E., northeast of Hicks. The outcrop of the "Lower Ohara" member of the formation is limited for the most part, to a very narrow belt just beneath the overlying Chester formations. The only part of the county where it occu- pies more extensive area is in the extreme southwestern part, northwest and southwest of Melcher Hills. SUBDIVISIONS OF THE STE. GENEVIEVE LIMESTONE The Ste. G-enevieve limestone in Hardin County is divisible into three persistent members. The great mass of the formation is a limestone consti- tuting the lower member, to which Ulrich has applied the name Fredonia. This lower member is followed by a thin, calcareous sandstone designated Eosiclare sandstone by the same author, from its typical development in the Ohio Eiver bluff just below the town of that name in Hardin County. The upper member of the formation consists of limestone and shale. Ulrich used the name "Ohara" for the upper member of the Ste. Genevieve limestone as he defined it, but the more recent studies have shown that only the lower portion of the original Ohara is properly referable to the Ste. Genevieve, the higher portion being the equivalent of lower Chester formations in the Mis- sissippi Valley section. In order to remove the possibility of misunderstand- ing and at the same time do away with the necessity of introducing a new stratigraphic name, this higher member of the Ste. Genevieve will be desig- nated as "Lower Ohara" in this report, although no "Upper Ohara," as such, will be recognized. In addition to these three members of the Ste. Genevieve formation first recognized and named by Ulrich, a sandstone member occupying a lower position than the Eosiclare has been observed at a number of localities in Hardin County. The best exhibition of this lower sandstone member is near the boundary of the county west of Eichorn, where it occupies a position about fifty feet below the top of the Fredonia limestone. A similar sub- Eosiclare sandstone is exposed about two and one-half miles southeast of Eichorn, and still another localitv is three miles northwest of Cave in Eock, 110 GEOLOGY OF HARDIN COUNTY in the NW. % sec. 3, T. 12 S., E. 9 E., where a cross-bedded, calcareous, fossiliferous sandstone lies 40 feet below the Eosiclare sandstone, both sand- stones being well exposed in the same section. Outside of Hardin County the typical three-fold division of the Ste. Genevieve formation is developed across Ohio Eiver in Kentucky. In the type region of the formation, Ste. Genevieve County, Missouri, and extending into northern Perry County, a median sandstone member, with limestone below and above, is present locally in the Ste. Genevieve forma- tion, while elsewhere such a sandstone member is wholly wanting. The same condition holds in Monroe County, Illinois, where a sandstone member is present in some sections and absent in others, and a similar sandstone mem- ber is present in the section at Alton, Illinois. Whether the sandstone member of the Ste. Genevieve formation in the Mississippi Valley localities is the exact equivalent of the Eosiclare sandstone of Hardin County, or of the sub-Eosiclare sandstone, or indeed whether it is not something different from either of these beds, cannot now be certainly determined. LITHOLOGIC CHARACTERS OF THE FREDONIA LIMESTONE MEMBER The Fredonia member of the Ste. Genevieve formation is a massive lime- stone in which the beds vary considerably in lithologic character. The color ranges from blue-gray, to gray, or even to nearly white, on the whole a distinctly lighter color than that of the St. Louis limestone. The texture of the limestone is also variable, dense, compact beds which exhibit a conchoidal or splintery fracture, not unlike some of the beds of the underlying St. Louis Jimestone, alternating with other more typical strata. The most notable feature of the member is its oolitic character ; in fact Ulrich originally called it the Fredonia oolite member of the Ste. Genevieve. In the lower portion of the member, towards its contact with the St. Louis limestone, the oolitic beds are less conspicuous than in the higher portion, where some thick, apparently persistent, nearly white oolitic beds are present in all the sections of Hardin and other southern Illinois counties, as well as in Kentucky. In the upper portion of the Fredonia, in some sections, there are some thin beds of calcareous shales, which in Hardin County are scarcely more than shaly partings between some of the denser limestone strata. Some of the limestone beds are distinctly cross-bedded in structure. The formation contains a large amount of chert, some of which resembles that of the St. Louis limestone, but in general the Ste. Genevieve chert occurs in plate-like masses, much more regular and smoother in outline than those of the older limestone. The chert is apt to be more conspicuously developed in the lower, less oolitic portion of the formation, but in places persists to the summit of the lime- stone. The chert in the lower part of the Fredonia is best displayed in the LOWER MISSISSIPPIAN 111 Ohio River bluffs, especially at Elizabethtown and vicinity, and a typical expression of the chert of this part of the formation is shown in figures 7 and 8. LITHOLOGIC CHARACTER OF THE ROSICLARE SANDSTONE MEMBER The Rosiclare sandstone member of the Ste. Genevieve formation is a thin but persistent bed throughout the area of outcrop of the formation in Hardin County, and in the adjacent parts of Illinois and Kentucky. Where the sandstone is met with in the excavations of mine shafts it is highly Pig. 7. — Photograph of the limestone bluff in front of the Rose Hotel, Eliza- bethtown, Illinois, showing the dark cherty layers in the lower part of the Fredonia limestone. The chert itself is superficial, being formed by the silicifica- tion of the limestone layers to the depth of only an inch or two. Looking northwest. calcareous, and in places it might be taken for a limestone with an abundance of included sand grains. Where the bed has been subjected to weathering the lime has been entirely leached out, leaving a rather fine grained but very porous sandstone. Where the rock has been unaffected by weathering its color is gray, not uncommonly with a slight greenish or bluish tint, but upon weathering it becomes brown, as in the bluff just below Rosiclare, and where the weathering has proceeded farther and the formation occurs in residual masses, as it does over a considerable area east of Eichorn, it commonly exhibits a distinctly reddish brown color. Wherever seen in section the Rosi- clare exhibits notable cross-bedding. In the NW. % sec. 36, T. 12 S., R. 8 112 GEOLOGY OF HARDIN COUNTY E., the Eosiclare is especially coarse and thick-bed decl, and carries a few small quartz pebbles. On the top of Lead Hill it is rather laminated and cross-bedded. LITHOLOGIC CHARACTERS OF THE "LOWER OHARA^ LIMESTONE MEMBER The "Lower Ohara" limestone member, overlying the Eosiclare sand- stone, is not notably different from the Fredonia below the sandstone. It is commonly oolitic or semi-oolitic in texture, light in color, in places nearly Fig. 8. — Cherty Ste. Genevieve limestone half a mile east of Elizabethtown. white, and it also/ may include some beds which are dense and compact, resembling similar beds in the Fredonia. DETAILED SECTIONS The only considerable continuously exposed section of the Ste. Genevieve formation in Hardin County which has been carefully measured, is the section at Eosiclare, in the Ohio Eiver bluffs. This section includes more than the Ste. Genevieve, however, and continues up into the overlying formations, and as measured it does not continue to the base of the Fredonia member of the formation. The lower part of the section, up to the base of the Eosiclare sandstone member, is exposed in the bluff quarries just above the boat land- ing, the higher portion being exposed below the town at Downeys and Fair- view bluffs. LOWER MISSISSIPPIAN 113 Section of the Ste. Genevieve limestone and Lower Chester formations at Rosiclare Thickness Feet 19. Sandstone, coarse grained, yellow-brown to ferruginous in color, with numerous fragments of plant stems (Bethel sandstone) 50 18. Talus-covered, blocks of Bethel sandstone, no limestone exposed.... 40 17. Limestone with shale partings. Limestone beds mostly gray in color, crystalline in texture, fossiliferous. Pentremites abundant; Lyropora and Talarocrinus 26 16. Talus-covered, but probably shale and limestone interbedded, with shale predominating 30 15. Limestone, gray, crystalline, cross-bedded 4 14. Shale and limestone interbedded, the limestone more or less argil- laceous, gray, crystalline, conspicuously cross-bedded, passing into shale laterally. Shale abundantly fossiliferous 6 (Unconformity) 13. Limestone, oolitic, semi-oolitic, and dense in texture, some beds cross-bedded, others not. Fossils not well preserved, but many of the weathered surfaces covered with crinoid stem segments, among which are those of Platycrinus penicillus, with an occasional base of the same species ("Lower Ohara" limestone member ) 35 12. Sandstone, calcareous, gray-brown in color when fresh, yellow-brown when weathered, conspicuously cross-bedded (Rosiclare sandstone member ) 10 11. Talus-covered, possible interval of 10 10. Limestone, dense, gray, oolitic, with smooth weathered surface 13 9. Limestone, in part oolitic, and in part dense and brittle, shaly part- ings, fossiliferous, with Pugnoides ottumwa 4 8. Limestone, bluish-gray in color, oolitic, fossiliferous 4 7. Limestone, gray, oolitic iy 2 6. Limestone, dark blue to gray in color, evenly bedded, with shaly partings. Fossiliferous 7 5. Limestone, white, oolitic, lower portion with numerous crinoid stems on weathered surface. Fossiliferous 30 4. Limestone, blue-gray to blue-black in color, with inconspicuous bedding planes. Some parts crystalline, others compact in texture with splintery fracture. Fossiliferous, with Pugnoides ottumwa.. 15 3. Limestone, without chert, mostly gray in color, some beds crystalline, others oolitic, certain thin beds buff in color, apparently dolo- mitic. Fossils poorly preserved, but some weathered surfaces covered with crinoid stems, among them Platycrinus penicillus, with an occasional base of the same species 27 2. Unexposed interval 3 1. Limestone, cherty, more or less compact in texture THICKNESS The greater portion of the Ste. Genevieve formation is included in the Fredonia member, which is at least 180 feet in thickness in the Ohio Kiver —8 G 114 GEOLOGY OF HARDIN COUNTY bluffs between Eosiclare and Jacks Point at the mouth of Big Creek. The thickness of the member may be somewhat greater than this, and it is alto- gether probable that a maximum thickness of 200 feet is not excessive. Three to four miles north of Elizabethtown, in the region adjacent to Hogthief Creek, the Fredonia appears to be only 120 to 150 feet thick, and the basal beds in this region differ somewhat in character from those in the river bluff at Elizabethtown. The Eosiclare sandstone member of the formation, in its type exposure below Eosiclare, is about 16 feet thick. In the shaft of the Stewart mine, south of Eichorn, 22 feet of the sandstone was passed through, and in the region east and southeast of Eichorn, where the bed is conspicuously developed, the thickness may be somewhat greater than this, probably 25 to 30 feet. On the contrary, locally, the bed may be thinner than in the type section at Eosiclare, for in one section in Kentucky only two feet is exhibited. Fifteen to twenty feet is doubtless a fair estimate of the average thickness of the bed. The "Lower Ohara" member of the formation varies in thickness from twenty feet and even less, to as much as fifty feet, about 25 feet being the average. The combined thickness of the three members of the Ste. Genevieve formation may be as much as 300 feet, although in places it prob- ably does not exceed 250 feet. STRATIGRAPHIC RELATIONS The stratigraphic relations of the Ste. Genevieve limestone with the underlying St. Louis have already been discussed. Sedimentation from the older to the younger formation was apparently continuous, with no strati- graphic break in Hardin County; consequently the boundary line between the two formations has to be drawn more or less arbitrarily. At one time it was believed that the conspicuously cherty beds were limited to the St. Louis formation, but further observation has shown this criterion to be use- less, for in places there are abundant cherts to the top of the Fredonia. The oolitic character of the limestone is a much safer criterion for the recognition of the Ste. Genevieve than is the absence of chert, and it has been the practice in Hardin County to draw the boundary line between these two formations below the lowest oolite beds observed. Not all the beds of the Fredonia lime- stone are oolitic, however, and the oolitic character of the formation becomes more marked in the higher part of the formation, but in the undoubted St. Louis limestone no oolitic beds of any sort have been noticed in the county. One very characteristic fossil, the stem plates of a crinoid of the genus Platycrinus, characterized by their elliptical outline, and by a series of marginal spine-like processes, has been considered as highly characteristic of the Ste. Genevieve limestone, and wherever these plates are observed in abundance it is safe to consider the limestone as belonging in that formation; but crinoid plates of the same species do occur, though more rarely, in lower LOWER MISSISSIPPIAN 115 beds which can be only St. Louis. On the other hand these plates are entirely unknown above the "Lower Ohara" limestone. The boundary line between the Ste. Genevieve and the next succeeding formation is an unconformable surface in every section observed where the contact can be seen. This is true not only in the counties adjacent to Mis- sissippi River; but also in the southern Illinois counties and in the adjoining part of Kentucky. In Hardin County the unconformity is perhaps best shown in the section above the railroad incline of the Fairview mine, at Fairview bluff. The same unconformity with conglomeratic beds in the basal portion of the higher formation is also exhibited in the sections east of Shetlerville, between that place and Wallace Branch. In the Cedar Bluff section, near Princeton, Kentucky, the same unconformable surface is marked by a distinct basal conglomerate in the limestone overlying the Ste. Genevieve, the size of the limestone pebbles commonly being from two inches to fragments the size of peas. PALEONTOLOGY Much of the limestone of the Ste. Genevieve formation in Hardin County is not of a character to preserve fossils in such a condition that they can be easily collected and identified, although the weathered surfaces commonly exhibit an abundance of organic remains. Crinoid stem segments are par- ticularly abundant upon many of the weathered surfaces, and one of the commonest of these is a small form, elliptical in outline, with a major diameter of an eighth of an inch or less, and with the border ornamented with a row of spines resembling the handles in the rim of a ship's steering wheel. The basal plates of the cup belonging with these stem segments are also met with in many places, but they are much less frequent than the stems; they are characterized by three ribs radiating outward from the stem facet. This crinoid commonly has been identified as Platycrinus huntsvilUe, but it is perhaps more properly referred to P. penicillus M. & W., originally described from Hardin County, from "St. Louis division of the Lower Carboniferous series." In places these remains are represented by innumerable examples, elsewhere they are less common or even wanting, but in no section of the Ste. Genevieve limestone in Hardin County has diligent search failed to discover them. These stem segments are more abundant in the higher portion of the Fredonia member of the formation than in the lower part, but they are present, and in places abundant, to its base, and are also locally met with in the sub- jacent St. Louis limestone. This same crinoid species, represented by the same sort of material upon the weathered surfaces of the limestone, is present in the "Lower Ohara" limestone at the summit of the formation. In some portions of the Fredonia, certain corals occur commonly. The one most often met with is MicheUnia princetonensis, weathered-out speci- 116 GEOLOGY OF HARDIN COUNTY mens of which may be looked for in many localities among the fragments of residual chert. A search near the top of the hill just outside of Elizabeth- town, along the road running north from the town, has rarely failed to supply a number of specimens of this species. Specimens have also been found among the cherts by the roadside on the hill between Elizabethtown and the Big Creek bridge, and at a number of points in similar situations along the road between Elizabethtown and Cave in Eock, and on the river bluff east of Cave in Eock. The only locality in the county where this coral has been observed undisturbed in the limestone ledges, is along the Ohio Elver bank just below Jenkins Point, about one mile above Eosiclare, wherei it must occupy a position fully 180 feet below the Eosiclare sandstone. The range of the species, however, is apparently throughout the entire thickness of the Fredonia, for some of the weathered-out specimens have been collected in such situations that they must have originated very close to the base of the Eosi- clare sandstone. Other Fredonia corals less commonly met with than the Michelinia princetonensis, are Michelinia subramosa and Cystelasma quinque- septata. The only good examples of these species collected in Hardin County are from a bed about 40 feet below the Eosiclare sandstone, at a locality one- half mile north of the bridge across Big Creek, but the same two species have been recognized by Mr. Butts, together with Pugnoides ottumwa and examples of Pentremites, in the sandstone 40 feet below the Eosiclare, at the locality northwest of Cave in Eock, in the NW. % sec. 3, T. 12 S., E. 9 E., already mentioned. Lithostrotion harmodites, a Fredonia coral that occurs rather commonly in this limestone in some Kentucky localities, has not yet been observed in Hardin County, but it may be looked for and doubtless will be found later. Considerable collections of fossils have been made from a number of beds of the Fredonia limestone in the river bluffs near Eosiclare. From a bed approximately 160 feet below the Eosiclare sandstone the following species have been identified : Fossils from the Fredonia limestone member, 160 feet below the Rosiclare sandstone Platycrinus penicillus M. and W. Dizygocrinus sp. undet. Orthotetes sp. Productus ovatus Hall Productus parvus M. and W. Pugnoides ottumwa (White)? Girtyella brevilobata (Swallow) Dielasma formosa (Hall) From a bed 22 feet higher in the section, or at a horizon 138 feet below the Eosiclare, the following species have been collected : Spiriferina transversa (McChesney) Spirifer pellaensis Weller Cliothyridina aff. sublamellosa (Hall) Composita trinuclea (Hall) Myalina sp. Bellerophon sp. Naticopsis sp. LOWER MISSISSIPPIAN 117 Fossils from the Fredonia limestone member, 138 feet below the Rosiclare sandstone Pentremites princetonensis Ulrich Girtyella indianensis (Girty) Spiriferina sp. Eumetria verneuiliana (Hall) Cliothyridina aff. sublamellosa (Hall) Composita trinuclea (Hall) Nucula shumardana Hall Myalina sp. Liopteria n.sp. Sphenotus sp. Cypricardella sp. Still higher, from a horizon 85 feet section on page 113) we have the followin Lepetopsis levettei White Naticopsis sp. Naticopsis sp. Straparollus sp. Eotrochus n.sp. Bellerophon sp. Bulimorpha whitfieldi Weller Bulimorpha sp. Zygopleura sp. Laevidentalium venustus M. and W. Orthoceras sp. below the Rosiclare, (Bed No. 1 in g fauna. Fossils from the Fredonia limestone member, 85 feet below the Rosiclare sandstone Triplophyllum sp. Pentremites sp., cf. P. princetonensis Ulrich Platycrinus penicillus M. and W. Archaeocidaris sp. (spine) Stenopora montifera Ulrich? Orthotetes kaskaskiensis (Mc- Chesney) Productus ovatus Hall Pustula sp. Pugnoides ottumwa (White) Girtyella indianensis (Girty) Dielasma sp. Spirifer pellaensis Weller Spirifer leidyi N. and P. Cliothyridina aff. sublamellosa (Hall) Composita trinuclea (Hall) Bellerophon sp. Productus parvus M. and W. The highest Fredonia collection secured in the Rosiclare section is from a horizon 30 feet below the Rosiclare sandstone, (Bed No. 6, in section on page 113) where the following species have been recognized: Fossils from the Fredonia limestone member, 30 feet below the Rosiclare sandstone Pentremites princetonensis Ulrich Platycrinus penicillus M. and W. Dizygocrinus persculptus Ulrich Fistulipora sp. Stenopora tuberculata (Prout) Fenestella serratula Ulrich Fenestella tenax Ulrich Thamniscus furcillatus Ulrich Septopora n.sp. Rhombopora sp. Orthotetes kaskaskiensis (Mc- Chesney) A complete list of these species, shown in the following table: Diaphragmus n.sp. Rhipidomella dubia (Hall) Pugnoides ottumwa (White) Girtyella indianensis (Girty) Spiriferina transversa (McChesney) Spiriferina cf. spinosa (N. and P.) Spirifer leidyi N. and P. Reticularia setigera (Hall) Cliothyridina aff. sublamellosa (Hall) Composita trinuclea (Hall) with their distribution in the section, is 118 GEOLOGY OF HARDIN COUNTY Fossils from the Fredonia limestone member of the Ste. Genevieve formation, in the river bluffs near Rosiclare Fossils From a From a From a bed 160 bed 138 bed 85 feet below feet below feet below the the the Rosiclare Rosiclare Rosiclare sandstone sandstone sandstone From a bed 30 feet below the Rosiclare sandstone Triplophyllum sp Pentremites princetonensis Ulrich Platycrinus penicillus M. and W Dizygocrinus persculptus Ulrich Archaeocidaris sp. (spine) Fistulipora sp. . . . . . Stenopora tuberculata (Prout) Stenopora montifera Ulrich? Fenestella serratula Ulrich Fenestella tenax Ulrich Thamniscus furcillatus Ulrich Septopora n. sp Rhombopora sp Orthotetes kaskaskiensis^ (McChesney) Orthotetes sp Productus ovatus Hall Productus parvus M. and W Diaphragmus sp Pustula sp Rhipidomella dubia (HalO. Pugnoid«ls ottumwa (White) Girtyella indianensis (Girty) Girtyella brevilobata (Swallow) Dielasma formosa (Hall) Dielasma sp Spiriferina transversa (McChesney) Spiriferina cf. spinosa (N. and P.) Spiriferina sp Spirifer pellaensis Weller Spirifer leidyi N. and P Reticulata setigera (Hall) Eumetria verncuiliana (Hall) Cliothyridina aff . sublamellosa (Hall). . Composita trinuclea (Haii).. Nucula shumardana Hall Myalina n. sp Myalina sp Liopteria n. sp Sphenotus sp Cypricardella sp Lepetopsis levettei White '. Naticopsis sp X X X X X LOWER MISSISSIPPIAN 119 Fossils from the Fredonia limestone member of the Ste. Genevieve formation, in the river bluffs near Rosiclare — Concluded From a bed 30 feet below Fossils the the the the Rosiclare sandstone From a From a From a bed 160 bed 138 bed 85 feet below feet below feet below the the the Rosiclare Rosiclare Rosiclare sandstone sandstone sandstone Naticopsis sp Naticopsis sp Straparollus sp Eotrochus n. sp Bellerophon sp Bellerophon sp Bellerophon sp Bulimorpha whitfieldi Weller Bulimorpha sp Zygopleura sp Laevidentalium venustus (M. and W.) Orthoceras sp X X X X The Rosiclare sandstone is commonly entirely barren of fossils, but specimens of Eumetrva, have been observed in it at a locality one-half mile southwest of Gross, and crinoid plates and a few obscurely preserved brachio- pod shells have been collected by Butts in the center of sec. 30, T. 12 S., R. 9 E. The "Lower Ohara" limestone is abundantly fossiliferous, the weathered surfaces of the formation being covered, in many localities, with crinoid stem fragments and other broken fossils, but the rock is of such a nature that fossils in a proper condition for determination are difficult to obtain. A small col- lection has been secured from the exposures at Fairview bluffs, just below Rosiclare, and the following species have been recognized from Bed 'No. 13 of the section on page 113 : Fossils from the "Lower Ohara'' limestone member of the Ste. Genevieve formation, at Fairview bluff Platycrinus penicillus M. and W. Spirifer bifurcatus Hall Spirifer pellaensis Weller Composita trinuclea (Hall) Nucula illinoiensis Worthen Led a curta M. and W. Cypricardella oblonga Hall Cypricardinia indianensis Hall Laevidentalium venustum (M. and W.) Bellerophon monroensis Weller? Bucanopsis textilis (Hall) Aclisina pygmaea (Weller) Aclisina sp. Orthonychia acutirostra (Hall) Phillipsia sp. 120 GEOLOGY OF HARDIN COUNTY The stem segments of Platycrinus penicillus are common upon the weathered surfaces at this locality. The fossils most commonly found upon breaking up the hard, oolitic limestone, are the diminutive gastropods and pelecypods which are listed. Of the thirteen forms recorded whose specific determination is certain or nearly certain, eight are present in the Ste. Genevieve limestone fauna which has been described by Weller from Monroe County, Illinois. 1 Of the other five species, three, Spirifer bifurcatus, Bu- canopsis textilis and Orthonychia acutirostre, are common forms in the Spergen limestone, and the other two, Platycrinus penicillus and Spirifer pellaensis, are well known Ste. Genevieve limestone forms. Not a single member of the fauna is a distinctive Chester type, although the two species of Spirifer have near relatives in the Lower Chester faunas. Spirifer hifur- catus of the Spergen limestone is similar to S. leidyi from the Chester, but the specimen under consideration is more like the older Spergen limestone form than the Chester S. leidyi. Spirifer pellaensis is a close relative of the early Chester varieties of S. increbescens. 1 Contributions from Walker Mus., vol. 1, No. 10, (1916). CHAPTER VIII— UPPER MISSISSIPPIAN SERIES, CHESTER GROUP By Stuart Weller General Statement The Chester has long been recognized as one of the divisions of the Mississippian system, or the "Subcarboniferous" as it was formerly called, in the geological section of Illinois. Until recent years it has been considered as coordinate in rank with such formations as the Burlington, Keokuk, or St. Louis limestones, no attempt being made by the earlier geologists to sub- divide the "formation" as it was called, although it was known that the Chester was considerably thicker than the older recognized members of the Mississippian. It was observed at an early date that the Chester section of the southern counties of Illinois, viz., Union, Johnson, Pope, and Hardin, is notably different in detail from that of the typical Randolph County area described by Hall and by Worthen, but no serious effort was made to corre- late the several members of the group in the two areas. In order to bring the situation clearly before the reader of this report it seems best to outline briefly the historical development of our knowledge of the Chester group. Historical Review David Dale Owen was the first geologist to make use of the name Archi- medes limestone 1 , in connection with his "Geological Reconnaissance of the State of Indiana, made in the year 1837." The Indiana beds so designated by him were a part of the stratigraphic series now known as the Chester group. The same author, in 1852, used the name again 2 for one of the divisions in his "Section of Subcarboniferous Limestones of Iowa." The Iowa beds to which the name was applied constitute the Keokuk limestone of later authors, Owen having overlooked, apparently, the much more con- spicuous Archimedes-bearing bed of the Warsaw formation. Owen seems to have had the idea that the Archimedes-bearing beds of Iowa were equivalent to those with which he was familiar in Indiana. In 1855 Swallow 3 used the name for beds exposed near the mouth of Des Moines River in northeastern Missouri, and for similar beds in Marion *Geol. Reconn. State of Indiana, (1838), pp. 20-21 of reprint (1859). 2 Geol. Survey Wisconsin, Iowa and Minnesota, p. 92 (1852). 3 Missouri Geol. Surv., First and Second Repts., pt. 1, p. 95, and section 14, opposite p. 92; also pp. 174 and 194 (1855). 121 122 GEOLOGY OF HARDIN COUNTY and Cooper counties, Missouri, which are referable to the Keokuk and Warsaw formations of more recent authors. With this usage the name had been applied to beds at three distinct horizons, as the Mississippian sections have come to be divided in more recent years. Again in the same Missouri report, Swallow 1 designated as "Archimedes Limestone," certain limestones in St. Louis County, Missouri, lying beneath the St. Louis limestone, which are now included in the Spergen formation, this being a still different horizon from any of the three to which the name had been applied by Owen or Shumard. The name was used by Swallow, not because the beds which he so designated were characterized by numerous specimens of Archimedes, for they are rare, but because he believed these limestones of St. Louis County occupied the same stratigraphic position as those to which the name had been applied in northeastern Missouri. The disentanglement of these several "Archimedes limestones" was accomplished by Hall in connection with his work in Iowa and southward along the Mississippi River sections in Illinois and Missouri. In the first statement of the results of Hall's work 2 , read before the tenth meeting of the American Association for the Advancement of Science, in August 1856, the stratigraphic succession of the formations lying between the beds below, sap- posed to be of the age of the Chemung group of New York, but now called Kinderhook and placed in the Mississippian, and the so-called "Coal Measures" above, was as follows : VII. Coal Measures. VI. Kaskaskia limestone or Upper Archimedes limestone. V. Gray, brown or ferruginous sandstone, overlying the limestones of Alton and St. Louis. IV. "St. Louis limestone" or "Concretionary limestone." III. "Arenaceous bed." Warsaw or Second Archimedes limestone. "Magnesian limestone." Beds of passage, soft shaly or marly bed with geodes of quartz, chalce- dony, etc. II. Keokuk limestone, or Lower Archimedes limestone. Beds of passage, cherty beds 60 to 100 feet. I. Burlington limestone. The same section as that recorded above was again described by Hall 3 in a paper read before the Albany Institute in November 1856, although the volume was not published until 1864. In these papers Hall clearly brought out the facts regarding the geological range of Archimedes, and showed that there were in truth three "Archimedes limestones" which he designated as lower, second, and upper, but also applied to them the geographic names iLoc. cit., pt. 2, pp. 170 and 182 (1855). 2 Amer. Jour. Sci., 2nd ser., vol. 23, p. 193; also Proc. Amer. Ass. Adv. Sci., vol. 10, pp. 56-57 (1857). 3 Trans. Albany Inst., voi. 4, pp. 2-36 (1864). UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 123 Keokuk, Warsaw, and Kaskaskia. Hair's observations were repeated in his final report for the Geological Survey of Iowa 1 , published in 1858. Further usage of the name "Archimedes limestone" was made by Owen in the four volumes of his reports on the Geological Survey of Kentucky, published in 1856, 1857, and 1861, and also in his Arkansas reports 2 published in 1858 and 1860. In every instance in these Kentucky and Arkansas reports, as in his earlier Indiana report, the rocks so designated correspond in stratigraphic position with the Upper Archimedes or Kaskaskia limestone of Hall. Hall's further description of the Kaskaskia limestone appears in his Iowa report 3 , where he described the formation as follows: This rock forms the cliff bordering the American bottom from Kaskaskia to Chester, and thence southward, having a continuous exposure of more than ten miles. The lower part is a compact, arenaceous and coarse textured lime- stone, with shaly partings, and containing numerous fossils. In its central and upper portions it includes a thick bed of sandstone, and the limestone beds are separated by shaly partings which often equal the calcareous strata in thickness. Towards the higher part there is a mass of green shale or marl sixty feet thick, and this is again succeeded by some heavier beds of limestone containing Spirifer, Allorisma, and a species of Pinna. The term Chester was first used as a geologic formation name by Swallow 4 in 1858, in a paper entitled "Explanations of the Geological Map of Missouri, and a Section of its Bocks." The name was applied to a sand- stone occurring at the summit of the "Lower Carboniferous" rocks. The entire Mississippian section described by Swallow, with the equivalents in the more modern nomenclature, is as follows : Swallow Modern Feet Chester sandstone 75 Palestine sandstone Upper Archimedes limestone 250 Kaskaskia limestone of Hall Ferruginous sandstone 195 Aux Vases sandstone ( Ste. Genevieve limestone St. Louis limestone 225 J St. Louis limestone ( Spergen limestone Archimedes limestone 350 J Warsaw formation ( Keokuk limestone Encrinital limestone 550 Burlington limestone Most of the thicknesses of the beds as recorded by Swallow in this place are excessive, some of them very much so, but it is quite clear what strati- graphic units he had in mind in his several divisions. The Chester sandstone of Swallow is clearly that formation which has been extensively quarried for building pu rposes in the environs of Chester, and which has furnished the 1 Rept. Geol. Survey Iowa, vol. 1, pt. 1, p. 109, (1858). 2 First and Second Repts. Geol. Reconn. Arkansas, (1858-1860) 3 Rept. Geol. Surv. Iowa, vol. 1. pt. 1, r>. 107, (1858). 4 Proc. Amer. Ass. Adv. Sci., vol. 11, pt. 2, p. ."., (18T.8) 124 GEOLOGY OF HARDIN COUNTY material for the buildings of the Southern Illinois Penitentiary at Menard. The original description of the formation is as follows: "Sixty-five feet of heavy bedded, irregularly stratified, brownish sandstone. It is made up of round and angular pellucid particles, having the interstices filled with a fine, opaque, brown substance, which is often replaced by oxides of iron and manganese." In Hall's interpretation of the Chester section, the Chester sandstone of Swallow was considered as being basal "Coal Measures," since the upper portion of his Kaskaskia limestone was said to be characterized by Spirifer, Allorisma, and a species of Pinna, an association of forms which clearly means Menard limestone. The earliest use of the name Chester for the limestones below the Chester sandstone of Swallow, was by Worthen 1 in 1860. No definition of the forma- tion as such, was given, the name being employed simply as an alternative for the Upper Archimedes limestone, in exactly the same manner as Kaskaskia had been used by Hall at an earlier date. In 1863 Henry Englemann 2 placed on record the results of his observa- tions upon the Mississippian section in Johnson and the adjoining counties in southern Illinois. The name Chester limestone, as used by this author, is the exact equivalent of the Kaskaskia limestone of Hall, and the Chester limestone of Worthen. His general section is divided into four members. "A. — Coal Measures, B. — Millstone-grit, C. — A series of strata which corre- spond to the Chester limestone and Ferruginous sandstone, D. — St. Louis limestone." A detailed section of his division C, which is said to aggregate a thickness of 1,000 feet, is as follows, the successive beds being numbered downward from the summit: 1. Limestone, generally highly siliceous and even flinty, and shades, with an aggregate thickness of about 180 feet. 2. Quartzose sandstone, alternating with shaly beds, and in some places slaty limestones at its upper part, from 120 to 150 feet. 3. Limestones and shales from 80 to 120 feet. 4. Quartzose sandstone with some beds of shale, over 100 feet. 5. Siliceous limestone and shale, in places as much as 140 feet. 6. Quartzose sandstone and shales, from 60 to 100 feet or more. 7. Siliceous limestones and shales, about 150 feet. 8. Quartzose sandstone, with some shaly portions, about 150 feet thick. I have distinguished these by the name Cypress sandstones on account of their prominent development on Cypress Creek. 9. Siliceous limestone and shales, the latter in places arenaceous. These appear as beds of passage between C and D, while at many points No. 8 rests directly on D. 1 Trans, St. Louis Acad. Sci., vol. 1, p. 697, (1860) ; also Proc. Amer. Ass. Adv. Sci., vol. 13, pp. 312-313, (1860). 2 Trans. St. Louis Acad. Sci., vol. 2, pt. 1, p. 189, (1863). UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 125 10. Locally strata of sandstone intervene between No. 9 and D, they generally appear as an important bed of transition, neither thick nor pure, but rather an arenaceous development of the shales of No. 9. The first use of the name Cypress as a formation name occurs, in the description of the section quoted above, and it was Engelmann's belief that "the Cypress sandstone, No. 8, may be regarded as a more fully developed equivalent of the Ferruginous sandstone of the Missouri Geological Report," an incorrect correlation which later came to be generally adopted. Worthen's description of the section at Chester, published in 1866 1 , was based upon more extended field observations than either the sections of Swallow or of Hall, and in this place the use of Chester as the name of a group of beds was applied for the first time, being used to include the sand- stone which had been designated "Ferruginous sandstone" by Hall and by Swallow, and the overlying beds to the summit of the "Sub-Carboniferous" in the Mississippi River section. The name "Ferruginous sandstone," how- ever, was rejected by Worthen in this place, since it had been applied first to the basal sandstone of the "Coal Measures" by the Missouri geologists, and the two sandstones, one Mississippian and the other Pennsylvanian, had been confused. Worthen proposed the name "Lower Sandstone of the Chester Group" for this division of the "Sub-Carboniferous," and consistently used it in his later writings. The original description of the "Chester group" is as follows : "CHESTER GROUP" This group comprises three or more beds of limestone, with intercalated beds of arenaceous and argillaceous shales and sandstones, the whole attaining a maximum thickness of at least six hundred feet. The following section will illustrate the general characters of this group, as it appears in the county above named [Randolph], where its peculiar features were first studied: 10. Hard gray siliceous limestone 25 to 30 feet 9. Shales and shaly sandstones, with fossil plants, Sigillaria, Stiffmaria, Lepidodendron, Knorria, etc. (partly hidden) ... 80 " 90 " 8. Shaly limestone (local) 15 " 18 " 7. Massive brown sandstone 40 " 6. Limestone, partly nodular and argillaceous 45 " 5. Green and blue argillaceous shales, with plates of limestone.. 45 " 70 " 4. Argillaceous and siliceous limestone (local).... 20 " 30 " 3. Massive sandstone and sandy shale 15 " 20 " 2. Compact gray limestone, with intercalations of blue, green and purple clay shales 150 " 1. Brown sandstone 120 " In his supplementary description of the "Chester group," which is the equivalent of Engelmann's subdivision C, Worthen refers to the work of Engelmann in the more southern counties of the State with the statement 1 Geol. Surv. Illinois, vol. 1, p. 77, (1866). 126 GEOLOGY OF HARDIN COUNTY that the beds are considerably increased in thickness in that direction. Engelmann's county reports are published in full in another part of the volume in which Worthen's general discussion appears, but neither Worthen nor Engelmann, in the final report, make any use of the term Cypress as a formation name, although it had been introduced several years earlier. The report on Union County, Illinois 1 , written by Worthen, appeared in 1868, but a footnote 2 states that it is "reported in part from observations of Mr. Henry Engelmann," and in it the same Chester subdivisions are recog- nized which were described at an earlier date by Engelmann in his Johnson County report. It is in this county that the typical exposures of the Cypress sandstone occur, although the name Cypress is not used in the report, this sandstone being referred to in every case as No. 8. A careful reading of the text of the Union and other county reports, in connection with more recent field observations, has shown that three distinct sandstones in the Chester group were confused as No. 8 by these earlier observers. No occurrence of beds 9 and 10 of Engelmann's Johnson County section are recorded in Union County, but in fact sub-Cypress beds are as fully developed in this county as in Johnson, or any of the more eastern counties of the state, and a sub- Cypress sandstone has been referred to as No. 8 in a number of localities enumerated, in such sections the true Cypress being designated as No. 6. Since the publication of these Illinois county reports, 1866-1868, no advance was made in the interpretation of the stratigraphy of the Chester group for nearly forty years. Many species of Chester or Kaskaskia fossils have been described, by many different authors, but rarely has any attempt been made to refer them to closer stratigraphic position than Chester "group' or "limestone," or Kaskaskia "group" or "limestone," and nowhere have any distinct faunal zones in the series been recognized which might be of service in more detailed stratigraphic work than had been attempted . previously. With the constantly increasing demands, during recent years, for greater refinement in stratigraphic work, it finally became apparent that this great series of Chester rocks, a thousand feet or more in thickness, must be divided if any further progress in the interpretation of the series was to be realized. The first attempt to break up the Chester into subordinate formations was made by Ulrich 3 in 1905, in connection with his studies in the fluorspar mining region of western Kentucky, chiefly in Crittenden and Caldwell counties. The subdivisions of the Chester and the relations of the forma- tions recognized by Ulrich at this time are shown in the following table 4 : 1 Geol. Surv. Illinois, vol. 3, pp. 33-57, (1868). 2 Loc. cit., p. 20. 3 U. S. Geol. Surv., Prof. Paper 36, CI 905) 4 Loc. cit., p. 24. UPPER MISSISSIPPI AN SERIES, CHESTER GROUP Ulrich's subdivisions of the Chester, 1905 127 o 1.1 CG Birds ville formation Kaskaskia limestone of Hall Chester limestone of authors Huron formation of recent Indiana reports is same as Birdsville formation Tribune limestone o s- O 02 Cypress sandstone Aux Vases sandstone of Keyes Probably also Big Clifty sandstone of Norwood Ste. Genevieve limestone ' Ohara limestone Rosiclare sandstone Fredonia oolitic limestone Two upper members referred to as lower Chester by Worthen and Englemann. Entire formation referred to as St. Louis by Norwood Ulrich's innovations in this classification consist in the inclusion of the Ste. Genevieve limestone in the "Chester group," and the division of the Kaskaskia limestone into two formations, as well as the revival of the name Cypress for Worthen's "basal sandstone of the Chester group." The present writer entered upon his studies of the Chester rocks and their faunas in the summer of 1906, and these investigations have been con- tinued to the present time. During the earlier years of this period the studies were quite general, but beginning with the summer of 1911 an in- tensive investigation of the Chester stratigraphy and paleontology has been | conducted in connection with the detailed mapping of the areas occupied by the rocks of this age in southwestern and southern Illinois. Early in the progress of this work it was found to be desirable to subdivide the Chester group in its typical area in Randolph and Monroe counties, Illinois, into several distinct formations as follows 1 : 9. Clore limestone 8. Palestine sandstone 7. Menard limestone 6. Okaw limestone 5. Ruma formation 4. Paint Creek formation 3. Yankeetown formation 2. Renault formation 1. Brew'erville sandstone (later referred to Aux Vases sandstone) 1 Trans. Illinois Acad. Sci., vol. 6, pp. 118-129, (1913) ; also Illinois State Geol. Surv. Monog-. I, pp. 23-29, (1914). 128 GEOLOGY OF HARDIN COUNTY During the progress of this work a diligent effort was made to interpret the Illinois section in the light of Ulrich's description of the western Ken- tucky Chester. The basal sandstone of the group was considered as being the equivalent of the "Cypress sandstone" of southwestern Illinois and Kentucky, Engelmann and Worthen as well as Ulrich being followed in this respect. If this correlation were the correct one, and if the two sections were equivalent, then the formation next above this sandstone would be correlated with the Tribune limestone of Ulrich's classification. However, the portion of the southwestern Illinois section which seemed to agree most closely with Ulrich's description of his Tribune limestone was a lower member of the Okaw formation which contains much oolite and is characterized by a fauna which consists largely of diminutive mollusca. With such a correlation the Renault, Yankeetown, Paint Creek, and Ruma formations would seem to have no equivalents in the Kentucky section. If the Okaw or some con- siderable portion of the lower part of that formation was to be correlated with the Tribune, then the formations above the Okaw with perhaps some of the upper portion of the Okaw itself, would represent the Birdsville formation. When all sides of the question were considered there seemed to be many inconsistencies in any such correlation as that suggested. Looked at in one way the lower Okaw seemed to be Tribune, looked at in another way it' seemed as though all the Okaw should be Birdsville. After careful considera- tion of the whole question the decision was reached that it was not possible to fit the western Illinois Chester section into Ulrich's classification of the group in Kentucky. Ulrich's own early interpretation of the western Illinois section was an impossible one. In a communication dated Nov. 18th, 1911, he made the following statement "Regarding your contention that the Birdsville is a synonym of Kaskaskia I am convinced of its error. The latter includes the Tribune, which formation indeed constitutes the lower and most important of the limestones comprised under the term Kaskaskia along the Mississippi. The Birdsville, on the contrary, follows the Tribune and begins locally, as north of St. Marys, with a conglomerate containing pebbles of igneous rocks." In this interpretation by Ulrich the Tribune included the major portion of Okaw limestone. Later observations proved that the conglomerate near St. Marys, which he had considered as marking the base of the Birdsville is situated at the base of the Renault formation and lies immediately on top of the Aux Vases sandstone, the basal sandstone of the group, its stratigraphic position being at least 300 feet beneath the limestone at Chester which he had identified as Tribune. It is needless to say that Mr. Ulrich has recog- nized his error in this correlation, and his present interpretation is a totally different one. UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 129 The differences of opinion between Ulrich and the present writer con- cerning the interpretation of the Chester sections, led to the planning of a field conference in the summer of 1913. This conference was participated in by Messrs. Ulrich, DeWolf, Buehler and the writer. A week was spent in hasty visits to localities in Illinois, Missouri, and Kentucky, but no agreement was reached concerning the questions at issue. During the seasons following this conference the detailed studies and mapping of the Chester have been continued, and the work has been carried into the more southern counties of Illinois, Hardin, Pope, and Johnson, which adjoin that portion of Kentucky in which Ulrich did his work. Upon entering on the intensive study of the Chester group in these southern counties of Illinois in 1915, nearly one entire season was devoted to reconnaissance work in Illinois and in the adjoining portion of Kentucky, including Crittenden and Caldwell counties, which is the heart of the region in which Ulrich prosecuted his studies more than 25 years ago. A second field conference between Messrs. Girty and Butts of the U. S. Geological Survey, and the writer, occurred during this season, at which time Butts received his first introduction to the Chester of the Ohio and Mississippi valleys. The work of this season brought to light a number of serious mis- takes that had been made by Ulrich in his earlier work. In the first place the limestone at Tribune, Kentucky, supposedly the type locality for the formation of that name, was found to occupy a position far up in the Birdsville forma- tion, being indeed the southeastern extension of the Menard limestone of the writer, named from Eandolph County, Illinois. It very soon became evident that Ulrich had included in his Tribune limestone, and had so mapped, lime- stones occupying three distinct horizons in the Chester series. The Tribune limestone at Tribune was found to be well up in the Birdsville formation as that member had been defined. The Tribune east of Princeton, Kentucky, lies above the sandstone that had been mistakenly identified as Cypress by Ulrich, this being the position assigned to the formation by him in the stratigraphic sequence of the Chester. The third limestone referred to the Tribune was that at the mouth of the Fairview mine at Rosiclare, Illinois, which proved to have a position beneath the sandstone which Ulrich had called Cypress, being in fact, the upper part of the limestone he had elsewhere called the Ohara member of the S-te. Genevieve limestone. It was also recognized during this season that Ulrich was mistaken in his interpretation of the Cypress sandstone of Engelmann, the true Cypress being at the base of the Birdsville formation instead of beneath the so-called Tribune. The most serious of Ulrich's mistakes was found to be in his location of the basal boundary of the Chester group. He included the Ste. Genevieve limestones in the Chester, drawing the lower boundary of the group between —9 G 130 GEOLOGY OF HARDIN COUNTY the St. Louis and the Ste. Genevieve limestones. It was found that the upper member of the Ste. Genevieve of Ulrich, the Ohara limestone, included two distinct units, the lower one truly Ste. Genevieve in age, and the upper one characteristically Chester, and a later very careful study of the faunas in the laboratory showed that the upper division of the Ohara was the exact correlate of the Renault limestone of the Randolph County section. This two-fold division of the Ohara was further established by the recognition of a distinct line of unconformity in the midst of the limestone. At a number of places this stratigraphic break is marked by the presence of a basal con- glomerate in the higher bed, and elsewhere by other indisputable evidences of unconformity. Most diligent search at every locality where the Ohara limestone has been met has been made by the writer for the characteristic Ste. Genevieve fossils in the "Upper Ohara" limestone, and not a single such occurrence has been detected during five field seasons. Ulrich has claimed to have collected such typical Ste. Genevieve species as PJatycrinus penicillus in the "Upper Ohara," and he has always referred, in the course of discussion, to a locality near Levias, Crittenden County, Kentucky. This locality was visited by the writer during the past season (1918) in company with Mr. Chas. Butts, and it was established that the sandstone identified and mapped by Ulrich at this locality as Cypress, was really Rosiclare, and consequently his supposed "Upper Ohara" fossils were actually collected from the Fredonia limestone, where they really belong. Since his introduction to the subject, Butts has continued his studies on the Chester. He soon discovered, during his examination of the section in Kentucky, that a cherty limestone member that Ulrich had described as being included in the Cypress sandstone, was in fact the upper portion of the Ohara limestone. With this mistake in interpretation it is seen that Ulrich has identified the "Upper Ohara" in three different localities, as three different units. At Cedar Bluff, near Princeton, Kentucky, the type locality for the Ohara, he placed it where it properly belongs. At a number of Kentucky localities he considered it as a limestone member of the Cypress, and at the mouth of the Fairview mine at Rosiclare, he identified it as Tribune limestone. Butts' most important contribution to our knowledge of the Chester section in southeastern Illinois, was the recognition, for the first time, of an important shale formation lying between the sandstone Ulrich had erro- neously called Cypress and the true Cypress of Engelmann. This shale contains some limestone layers, and on being traced westward into Johnson County is found to be the equivalent of the Paint Creek formation of Ran- dolph and Monroe counties. Extending southward into Kentucky the same unit is represented by at least some portion of the formation having the stratigraphic position assigned to the Tribune limestone by Ulrich, a for- UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 131 mation which has more recently been called Gasper by Butts because of the unfortunate choice of the name Tribune by Ulrich. During the progress of these recent studies on the Chester, the writer's results have been recorded in manuscript from time to time, and copies of all but one of these manuscripts have been transmitted to Ulrich. Most of the mistakes in his earlier work that have been pointed out to him, have been acknowledged by Ulrich, but he has been willing to make no compro- mise in regard to his interpretation of the Ste. Genevieve. He has continued to insist that the Ohara member is not divisible, and that the whole of the Ste. Genevieve formation is properly a member of the Chester group. A final field conference was arranged in the summer of 1916, partici- pated in by Messrs. Ulrich, Butts, Girty, and Ashley, of the U. S. Geolog- ical Survey, with DeWolf and the writer. Nearly two weeks were spent in Illinois, Missouri, and Kentucky. In the course of this conference the present writer's interpretation of the critical section in the Ohio Eiver bluff below Eosiclare was completely accepted by Ulrich. It was agreed that the "Upper Ohara" of this section was the equivalent of the Kenault lime- stone of Randolph County, and that there was a distinct unconformity here separating the Renault from the underlying St. Genevieve limestone. A few days later, on visiting the type locality of the Ohara limestone near Princeton, Kentucky, the writer insisted that the section was the same as that below Eosiclare, but this was disputed by Ulrich. Later in the same season, however, Butts actually traced the beds across country from the Eosi- clare locality to Princeton and demonstrated that the sections were the same. Early in 1918 a volume was issued by the Kentucky Geological Survey, entitled "Mississippian Formation of Western Kentucky." It contains two papers, the first by Chas. Butts on "Descriptions and correlation of the Mississippian formations of western Kentucky," and the second by E. 0. Ulrich on "The Formations of the Cnester series in western Kentucky and their correlates elsewhere". This contribution by Ulrich is essentially a revision of his earlier work published in Professional Paper No. 36, of the U. S. Geological Survey. A number of the mistakes of the earlier report are acknowledged and corrected. Some of his earlier fallacies are adhered to and at least one new error of importance is fallen into. The work is controversial from beginning to end, and is an attempt to establish his own interpretation as opposed to that of the writer, expressed in the several manuscript reports which had been furnished him. The one new mistake of Ulrich, mentioned above, is in the correlation of the Cypress sandstone with the Lower Okaw limestone of the Randolph County section. The results of the writer's studies in Johnson County, Illinois, establish without doubt, the equivalence of the Golconda limestone of southeastern Illinois with the Lower Okaw, the evidence for which will be presented later. 132 GEOLOGY OF HARDIN COUNTY ti s o k. to o CO to 83 CO s *-, o o ^2 s? O o g ■fo 53 o to o to o CD Co « «! O o v. to »t3 to 'CJ 1 ^ DO W pq < H SOUTHEASTERN ILLINOIS faC 2 « .£43 fl 8 03 0) "° S 3 O 1 "8 ° a co § el o £ CO Paint Creek shale and limestone Bethel sandstone or wanting Renault limestone Main Talarocrinus fauna Shetlerville limestone A mplexus geniculatus fauna Lower Ohara and Fredonia limestone Platycrinus penicillus Pugnoides ottumwa xjj^ijQ .iadd^ // \ \ Upper Okaw limestone (Plum Creek beds) w i. <4> to to 3 O CO to O So CQ CO s *> O © •~j S e o s •*o ^ C3 o V i* v. K w o t> CO 60 to l! o to to P t3 1 ^ w J i H SOUTHEASTERN ILLINOIS CO Ji £ -° o CO ^5 C co c3 ep T3 3 3 O St CP CO Paint Creek shale and limestone Bethel sandstone or wanting Renaujt limestone Main Talarocrinus fauna Shetlerville limestone Amplexus geniculatus fauna "Lower Ohara" and Fredonia limestone Platycrinus penicillus Pugnoides ottumwa ~ h ~ /i 1 \ / 1 \ / 1 1 \ / 1 1 \7 1 \|B/ i / 1 ,/ 1 / 1 / - 1/ RANDOLPH COUNTY Upper Okaw limestone (Plum Creek beds) co ,Q c3 So GO n o — 0) g§ II CO Paint Creek limestone TaZarocrinus 3 * s o .2 ^j £ ^^ Renault limestone Main Talarocrinus fauna co a> K* CO Ste. Genevieve limestone Platycrinus penicillus Pugnoides ottumwa UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 133 In the following pages the several Chester formations of southeastern Illinois, especially as exhibited in Hardin Comity, will be described and discussed, and the evidence for the conclusions that have been reached, will be presented. Shetlerville Formation name and general distribution The Shetlerville formation is here named and denned for the first time, it is a constituent part of the upper limestone or Ohara member of the Ste. Genevieve formation as redefined by Ulrich 1 , but the field studies of the past few years have shown that this Ohara limestone is a complex unit which must be broken up in order to properly display the true relations of the stratigraphic succession in southeastern Illinois and the adjacent portion of Kentucky. The "Upper Ohara" is entirely foreign to any por- tion of the true Ste. Genevieve limestone as that formation is typically developed in Ste. Genevieve County, Missouri, and in Hardin County and the adjacent areas it is separated from the "Lower Ohara," which is a part of the Ste. Genevieve formation, by a distinct unconformity. It is the basal portion of the "Upper Ohara" that is here designated as the Shetlerville formation, from its excellent exposures just east of Shetlerville, between that place and Eich and Melcher hills. The formation is well exposed at a number of localities in Hardin County, and in a few localities in the neighboring Pope County. Outside of this area it occurs at numerous localities in the adjacent portion of Kentucky, more especially Crittenden County, but it is known to extend also into Caldwell County. The formation is also believed to be present as far west as Union County, Illinois. In mapping the geology of Hardin County it has not been practicable to separate the Shelterville from the overlying Eenault, so the two forma- tions are included under the same color. The Shetlerville is more or less completely talus-covered in many localities where it might be looked for, but aside from the already mentioned locality east of Shetlerville, the for- mation is well exposed south of Eichorn, and in Downeys and Fairview bluffs below Eosiclare. It also may be looked for in the hills east of Wallace Branch, about two miles from the mouth of that stream, and in the hill southeast of the mouth of Hogthief Creek, three and one-half miles north- west of Elizabethtown, although no outcrops have yet been detected in either of these localities. In belt number three of the faulted portion of the county (see figure 3) the Shetlerville is exposed in at least one locality north- west of Peters Creek, In the southeastern non-faulted portion of the county, the formation should be present beneath the Eenault in the belt extending from 1 U. S. Geol. Survey, Prof. Paper No. 36, p. 39, (1905). 134 3E0L0GY OF HARDIN COUNTY the Peters Creek fault to the Ohio Kiver bluffs ; and at one point, in the fields southwest of the road intersection two and one-half miles north of Cave in Rock, the formation is exposed in sec. 1, T. 12 S., R. 9 E. From the wide distribution of known exposures of the Shetlerville, it may be safely assumed to be uniformly present in Hardin County wherever the proper part of the geological section forms the surface rocks, although it is commonly hidden from view by the surficial covering. LITIIOLOGIC CHARACTER In its lithologic character the Shetlerville formation is in strong con- trast with the subjacent Ste. Genevieve limestone, in that it is much more shaly, much more irregularly bedded and much less uniform in character. The limestone beds of the formation are commonly crystalline in character, and are dark gray or blue gray in color, and when well exposed they exhibit more or less well pronounced cross-bedding in most localities. The lime- stones are interbedded with more or less calcareous shales, but the lime- stonestone and shale succession is not uniform from place to place, and the limestone layers are seen to grade into shales horizontally. In places the limestone may constitute nearly the entire formation, the shale content being reduced to little more than partings between the limestone ledges; elsewhere the shales may constitute the greater part of the formation. Where shales predominate the limestone not infrequently occurs as plate-like, discon- tinuous layers an inch or thereabouts in thickness, which are composed for the most part of fragments of fossils, and whose weathered surfaces exhibit many well preserved fossils. Upon the weathered outcrops, the shales of the Shetlerville formation are commonly yellowish, buff colored, oi ashy gray, but in the unweathered condition they are doubtless of a blue color for the most part. Thin layers of red shale occur in places. The weathered sur- faces of the limestones are commonly a yellowish or buff color. THICKNESS Because of the usual covering of the Shetlerville formation with sur- ficial material, the actual thickness of the formation can be accurately ob- served in but few localities. By far the best section for the accurate de- termination of the thickness is in the northeastern of the two hills between Ivosiclare and the Fairview incline, where both the lower and the upper contacts of the formation can be accurately ol (served. At this point the actual thickness is 30 feet. It is not unlikely that the thickness may vary somewhat from place to place, but it is nowhere a thick formation, and the observed thickness below Rosiclaire mav be safely assumed to be near the average. UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 135 STRATIGRAPHIC RELATIONS It has already been pointed out that the Shetlerville formation rests unconformably upon the underlying Ste. Genevieve limestone. This un- conformity is well exhibited in the Ohio Eiver bluff above the railway incline at Fairview (fig. 9), just below Rosiclare, and it is again shown in the locality just east of Shelterville. At this last locality the lower portion of the formation includes conglomeratic layers, the included pebbles all being limestone. Outside the Hardin County area the base of the Shetlerville is seen to rest unconformably upon the underlying Ste. Gene- vieve in the Cedar Bluff quarry, near Princeton, Kentucky, the original type locality of the Ohara limestone of Ulrich. In the section at this lo- : *zm m 0c ,\ up • * -*^rr" „ j*^ • A, *Si«WHe5»^ • : - - m fSt - -V"' *-i 1' -i Fig. 9. — View showing the unconformity between the Shetlerville and the underlying Ste. Genevieve. The dense, compact bed in the lower part of the picture is "Lower Ohara" limestone; the overlying, thinly-bedded, shaly zone is the base of the Shetlerville. At the old quarry face above the railroad track at Fairview bluff, below Rosiclare, looking northwest. cality the base of the Shetlerville is marked by a distinct limestone con- glomerate a foot or more in thickness, with pebbles ranging in size from two inches down, the basal contact of the conglomerate being the upper limit of the very characteristic Ste. Genevieve limestone fauna. At every locality where the actual basal contact of the Shetlerville has been observed, there are unmistakable evidences of unconformity, and it is safe to assume that this relation is general. The upper contact of the Shetlerville formation has been observed in fewer sections than has the basal contact. In the northeastern of the two hills between Rosiclare and Fairview what may be the line of demarcation 136 GEOLOGY OF HARDIN COUNTY between this formation and the superjacent Eenanlt is very sharp and uneven (see figure 10), clearly indicating an uncomformable relation be- tween the two beds, with an erosion interval intervening between the two periods of sedimentation. Although this contact is obscured in all other sections which have been observed, the fact that the seeming unconformity is so pronounced at one locality suggests that it may be general, although no proof exists that this may be more than a local phenomenon. PALEONTOLOGY The Shetlerville formation is one of the most fossiliferous zones in the entire Chester group in Hardin County. In places where the shale and Fig. 10. — View showing the possible unconformity between the Shetlerville and the overlying Renault limestone, in the Ohio River bluff a third of a mile below Rosiclare, looking northwest. limestone beds of the formation are exposed in open glades upon some of the hill slopes, well-preserved, clean fossils may be collected in quantity. The larger number of the fossils met with belong to a comparatively few species, but associated with these abundant or common forms there are many others which are more or less rare in their occurrence. A combined list of the species from eight different localities in Hardin County, Illinois, and Crittenden County, Kentucky, from which fairly complete collections have been secured, includes 69 species. The combined list is given below, the occurrence of the species being indicated by crosses in the eight columns numbered 1 to 8 for the several localities. The localities from which the collections are recorded are as follows: UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 137 1. Ohio Eiver bluff above the railway incline of the Fairview mine, below Kosiclare (Bed No. 14, in section recorded on page 113). 2. Ravine just east of Shetlerville. 3. Public highway at southwest corner of Melcher Hills, between Shet- lerville and Wallace Branch. 4. Southern slope of hill three-fourths of a mile south of Eichorn. 5. One mile northwest of Peters Creek Store, NE. *4 sec. 7, T. 12 S. ? R. 8 E. 6. Lemon quarry, three miles west of Marion, Kentucky. 7. Thomas Glade, near Bethel School, four miles west of Marion, Crittenden County, Kentucky. 8. Moore Glade, four miles west of Marion, Kentucky. Fossils from the Shetlerville formation collected in Hardin County, Illinois, and Crittenden County, Kentucky FOSSILS Triplophyllum spinulosum (M.-E. and H.) Amplexus geniculates Worthen Cornulites sp Spirorbis sp Eupachycrinus sp Globocrinus unionensis (Worthen) Dichocrinus girtyi n. sp Dichocrinus sp Talarocrinus buttsi Ulrich Talarocrinus trijugis M. and G Crinoid stems and plates Pentremites pinguis Ulrich Pentremites princetonensis Ulrich Pentremites pulchellus Ulrich Pentremites tuscumbiae Ulrich Mesoblastus glaber (M. and W.) Ascodictyon sp Hederella sp Fistulipora excelens Ulrich Eridopora punctifera Ulrich Stenopora tuberculata (Prout) Stenopora cestriensis Ulrich Stenopora montif era Ulrich Stenopora sp Batostomella ? sp Anisotrypa ? sp Fenestella cestriensis Ulrich Fenestella elevatipora Ulrich X X X 138 GEOLOGY OF HAEDIN COUNTY Fossils from the Shetlerville formation collected in Hardin County, Illinois, and Crittenden County Kentucky — Concluded FOSSILS 1 2 3 4 5 6 7 Fenestella serratula Ulrich Fenestella tenax Ulrich Archimedes sp Polypora cestriensis Ulrich Polypora spinulif era Ulrich Lyropora quincuncialis Ulrich Thamniscus furcillatus Ulrich Thamniscus ramulosus Ulrich Septopora subquadrans Ulrich Rhombopora sp Streblotrypa nicklesi Ulrich Cystodictya labiosa n. sp Glyptopora punctipora Ulrich Phractopora sp Crania chesterensis M. and G Orthotetes kaskaskiensis (McChesney) Productus ovatus Hall Productus cf . inflatus McChesney .... Productus sp Diaphragmus elegans (N. and P.) . . . . Girtyella indianensis (Girty) Girtyella brevilobata (Swallow) Dielasma illinoisensis Weller Spiriferina transversa (McChesney) . . . Spiriferina subspinosa n. sp Cyrtina ?? sp Spirif er breckenridgensis Weller Spirifer leidyi N. and P Spirif er increbescens Hall var Reticularia setigera (Hall) Eumetria vera (Hall) Eumetria costata (Hall) Cliothyridina sublamellosa (Hall) Composita trinuclea (Hall) Myalina sp Allorisma sp Bellerophon sp Pleurotomaria ? sp Pleurotomaria ? sp Holopea ? sp Orthonychia cf. chesterensis M. and W. Conularia sp Orthoceras sp Phillipsia sp X X UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 139 Of this list of 72 species eight only are recorded from all of the locali- ties, these eight species being as follows: Amplexus geniculates Spiriferina transversa Stenopora tuberculata Spiriferina subspinosa Fenestella cestriensis Cliothyridina sublamellosa Fenestella serratula Composita trinuciea Most of these species are long-ranging forms, and all but Amplexus geniculatus, Fenestella cestriensis, and Spiriferina subspinosa are known from both older and younger Mississippian faunas. Of these three Fenes- tella cestriensis is widely represented in the Chester formations. Amplexus geniculatus and Spiriferina subspinosa of these eight species are wholly restricted, so far as known, to the Shelterville ; and since they are present in all of the localities recorded they constitute good index fossils and this particular faunal zone may be known as the Amplexus geniculatus zone. Two other species of the fauna, much less commonly met with, which are not known outside of this zone, are Globocrinus unionensis, and Talaro- crinus buttsi. Eleven of the recorded species have been recognized in seven, or all but one of the localities, these species being as follows : Triplopbyllum spinulosum Diaphragmus elegans Pentremites princetonensis Girtyella indianensis Eridopora punctifera Dielasma illinoisensis Fenestella tenax Reticularia setigera Cystodictya labiosa Eumetria vera Glyptopora punctipora Four other species occur in all but two, or in six of the localities, these species being as follows : Mesoblastus glaber Septopora subquadrans Fistulipora excelens Spirifer ieidyi The 25 species enumerated above can be fairly considered as consti- tuting the most important elements in the fauna of the Shelterville forma- tion. Of these species Cliothyridina sublamellosa is the most abundant form in every locality, and in many places where large collections have been made, the examples of this species constitute a bulk as. great as all others together. Composita trinuciea is another abundant form, but it has nowhere been observed to be as numerous as the specimens of Cliothyridina. The ex- amples of Pentremites princetonensis are also very abundant in most locali- ties, although this species, as well as all other blastoids, is strangely lacking in the collection from Lemon quarry, near Marion, Kentucky. In a consideration of the fauna of the Shetlerville formation it is im- portant to make comparisons with the fauna of the preceding Ste. Genevieve limestone. The Ste. Genevieve is particularly characterized by Plahjcrinus penicillus and Pugnoides ottumwa, neither of which has been observed in the 140 GEOLOGY OF HA11DIN COUNTY Shetlerville at any locality. Other species which are of common occurrence in the Ste. Genevieve and absent from the Shetlerville, are Rhipidomella dubia and Dizygocrinus persculptus, an example of which was figured by Worthen as Batocrinus unionensis, but which is specifically and generically distinct from the type of that species. In the following table those members of the Shetlerville fauna which have been specifically identified, have been recorded with their geologic range, the range in formations older than the Shetlerville being shown on the left, and in younger formations on the right. In column 1, occurrences in any pre-Ste. Genevieve formation is recorded; in column 2, Ste. Genevieve; in column 3, Renault; and in column 4, post-Eenault. Where the specific identi- ties in these several formations are the same the record in the proper column is indicated by X, but where the species is represented by a closely allied form, but one not specifically identical, it is indicated by 0. Table showing geologic range of the Shetlerville fauna SHETLERVILLE FAUNA X X X Triplophyllum spinulosum. Amplexus geniculates Globocrinus unionensis. . . . Talarocrinus buttsi Talarocrinus trijugis Pentremites princetonensis Pentremites pulchellus Pentremites tuscumbiae . . . Pentremites pinguis Mesoblastus glaber Fistulipora excelens Eridopora punctifera Stenopora tuberculata . . . . Stenopora cestriensis Stenopora montifera Fenestella cestriensis Fenestella elevatipora Fenestella serratula Fenestella tenax Polypora cestriensis Polypora spinulif era Thamniscus furcillatus Thamniscus ramulosus Septopora subquadrans . . . Streblotrypa nicklesi Cystodictya labiosa X X X UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 141 Table shoiving geologic range of the Shetlerville fauna — Concluded SHETLERVILLE FAUNA Glyptopora punctipora. . . . Crania chesterensis Orthotetes kaskaskiensis . . Productus ovatus Diaphragmus elegans Girtyella indianensis 'Girtyella brevilobata Dielasma illinoisensis Spiriferina transversa Spirifer breckenridgensis . . Spirif er leidyi Spirifer increbescens var . . Reticularia setigera Eumetria vera Eumetria costata Cliothyridina sublamellosa Composita trinuclea X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X The fauna of the Shetlerville horizon is that of the so-called Zone 3 of the Ohara limestone as recorded by Ulrich 1 , and the faunal evidence as pre- sented by him exhibits an association of the characteristic Ste. Genevieve species, Platycrinus penicillus, with the Amplexus geniculates and its asso- ciates. This supposed association is one of the strongest evidences Ulrich has presented for his union of the "Upper Ohara" with the Ste. Genevieve, and consequently his inclusion of the Ste. Genevieve in the Chester group. The published list of this fauna, however, as presented by Ulrich, is grossly misleading as to facts. He has not separated his faunal lists in accordance with the localities from which they were collected, but has lumped together in one composite list all those forms from all of the collections which he has either correctly or incorrectly considered as representing this horizon. The one locality that has always been mentioned by Ulrich in the discussion of this question, where he claims to have collected Platycrinus penicillus and other unquestioned Ste. Genevieve species in the "Upper Ohara," is near Levias in Crittenden County, Kentucky, where the fossils are said to have been secured less than fifty feet beneath the sandstone formerly called Cypress by him, but now renamed Bethel by Butts. A visit to this locality in the summer of 1918 by the writer in company with Mr. Chas. Butts, established the fact that the Eosiclare sandstone had been mistakenly identified and mapped as Cypress by Ulrich, and that consequently these Ste. Genevieve x Ky. Geol. Surv., Miss. Ser. in W. Ky., pp. 139-141, (1918). 142 GEOLOGY OF HAEDIN COUNTY limestone species were actually collected from the Fredonia limestone instead of from the "Upper Ohara." A most diligent search by the writer, at every favorable locality encountered during five years of field work, has failed to disclose a single instance of the occurrence of the index fossils of the Fredonia and "Lower Ohara" passing over into the Shetlerville or any part of the "Upper Ohara/' the higher faunas being characteristically Chester in their relationships. This circumstance, in association with the fact that the Shet- lerville lies unconformably upon the "Lower Ohara" wherever this contact can be clearly seen, has led to the drawing of the basal line of the Chester in Hardin County, at this horizon. OOBEELATION A full discussion of the correlation of the Shetlerville formation will be deferred, and the problem will be considered along with the correlation of the Eenault. The two formations are very closely related faunally. Only four species are restricted to the Shetlerville, these being Amphxus genicuWus, Globo'crmus unionmsis, Talarocrinus buttsi, and Spiriferina subspinosa. Every other Shetlerville species except two forms of Pentremites is present also in the Eenault or in one of the still younger Chester formations, and none of these four species is known in the Ste. Genevieve. The faunal con- nection with the Ste. Genevieve is much less close than with the Eenault, although 16 species are common to the two horizons. Many of these, however, are long-range forms and are either present or are represented by closely allied species in formations still older than the Ste. Genevieve, and the characteristic index fossils of the Ste. Genevieve, Platycrinus penicillus and Pugnoides ottumwa have nowhere been observed in the Shetlerville or Eenault, As will be shown in the later discussion two alternatives are possible in th( correlation of the Shetlerville. From one point of view it may be considered as the time equivalent of the Aux Vases sandstone, while in another way it may be looked upon as the time equivalent of the sub-Eenault unconformity of Eandolph County, representing essentially the time interval between the end of the deposition of the Aux Vases sandstone and the beginning of the Monroe County Eenault. Eenault Foemation name and genebal disteibution The Eenault formation is typically developed in Monroe County, Illinois, and was named from Eenault Township of that county. The formation extends northward into St. Clair County for a short distance, southward into Eandolph, and continues across Mississippi Eiver into Ste. Genevieve County, Missouri. Beyond this area of outcrop, to the east and northeast, the Eenault UPPER MISSISSIPPI^ SERIES, CHESTER GROUP 143 formation is completely buried beneath strata of younger Chester and Penn- sylvanian age. The formation is penetrated in the oil wells of eastern Illinois, and from these beds some of the oil has been derived. Composed as the formation is, of limestones, sandstones, and shales, all of exceedingly variable character, the heterogeneous nature of the constituent beds of the formation in western Illinois, and again in the oil fields of the eastern part of the state, is evidence of the near-shore origin of the Eenault sediments in these areas.. The position of the shore-line indicated by these occurrences must have been very different from- that of the earlier Ste. Genevieve sea which reached far to the north and northwest, at least as far as Fort Dodge, Iowa. Hardin County lies within the area of the embayment whose shores are indicated by the Eenault outcrops in western Illinois and the occurrence of the same formation in the oil wells of the eastern part of the state, but it is a long distance from the shore line indicated and consequently the character of the Eenault sediments in Hardin County is very different from that of those which accumulated in proximity to the ancient shore line. The outcrops of the Eenault formation in Hardin County are much scattered, and they nowhere occupy such broad areas as do the St. Louis and Fredonia limestones. One reason for the more limited extent of the areas underlain by the Eenault than by the older limestones, is its lesser thickness ; but its position immediately beneath a thick, massive, bluff-forming sand- stone, in view of its own somewhat shaly character, is largely responsible for its occurrence in narrow belts along the lower slopes of hill sides whose upper portions rise as more or less precipitous bluffs of sandstone. In the less faulted portions of the county, such narrow, elongate outcrops of the forma- tion have considerable extent longitudinally, but within the central, faulted portion of the county even these narrow, band-like outcrops commonly do not continue for any great distance before being cut off by faulting. In the southeastern, unf aulted portion of the county, the Eenault occupies a narrow belt, beneath the sandstone bluff, beginning at a point in the Ohio Eiver bluff a little over three miles east of Cave in Eock, extending west- ward for about one mile and then in a northwesterly direction along the line of bluffs for a distance of five miles or more to where it is cut off by the northeast-southwest Peters Creek faults which determine the southeastern boundary of the central faulted zone of the county. From this point the formation occurs in more or less disconnected outcrops in belt number three of the central faulted zone (see figure 3), along the line of bluffs running in a southwesterly direction parallel with the major lines of faulting, to a point a little northwest of Elizabethtown. Other outcrops, perhaps among the best in the whole county, are exposed in the isolated hill with two sum- mits, which lies along the Ohio Eiver bank just southwest of Eosiclare. One other group of outcrops occurs a little less than a mile west of Peters Creek 144 GEOLOGY OF HARDIN COUNTY Store, in a narrow, elongate, downdropped fault block which is an off-shoot from the main, much larger, downdropped block to the northwest. Within the median belt, number two, of the central faulted zone (see figure 3), the Eenault outcrops are much scattered. A number of small areas occur in the very complexly faulted region northeast and southwest of Stone Church. A somewhat larger area occurs in the wedge-shaped termination of one of the fault blocks south of Illinois Furnace, and another similar area is present in a downdropped block southwest of St. Josephs School and east of Wallace Branch. The actual limestone outcrops in each of these two latter situations are heavily talus-covered, and the presence of the formation is indicated chiefly by residual fragments of the formation in the surficial covering. In belt number one of the faulted zone, the most extended area where the Eenault constitutes the surface rock, is in the extreme southwestern part of the county including the summit of Rich Hill, surrounding Melcher Hills, and extending northwest and southwest to the Shetlerville fault. In the more northern part of the belt the Eenault occupies a very sinnous strip extending west from Gross to the Hogthief Creek fault west of Eock Creek School. From the Pope-Hardin county boundary just north of the junction of the two forks of Hicks Branch in the northwestern portion of the county, the Eenault occupies its normal position among the formations that partly encircle the Hicks dome. From the county line it extends in a northeasterly and easterly direction to the Lee fault, where the outcrop is offset over a mile to the southwest, and thence continues its easterly course to the Wolrab Mill fault. An excellent outcrop of the formation occurs in the hill a little over one-half mile southwest of Eichorn. LITHOLOGIC CHARACTER Unlike the Eenault of the western counties of Illinois, where the forma- tion includes limestones, shales, and sandstones of various sorts, in Hardin County and adjacent parts of Illinois and Kentucky, the formation is made up mostly of limestone with a subordinate amount of calcareous shales in the form of partings between limestone layers. The limestones are commonly of gray or blue-gray color, and of crystalline or more or less compact texture ; some beds are conspicuously crinoidal, some beds are notably cross-bedded, and locally there are oolitic beds, the oolite grains being commonly much smaller than those of the Ste. Genevieve. In the higher beds are cherty layers, which upon weathering give origin to very characteristic chert fragments of a pale color, which occur in the residual materials as small, subcubical or blocky fragments. On the whole the limestones exhibit a considerable range of variation. The formation as a whole is distinctly more calcareous, with thicker and more massive beds of limestone than the underlying Shetlerville UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 145 formation. In every locality where exposures of the Eenault have been observed, the surface is so heavily covered with talus from the overlying massive sandstone formation that it has been impossible to make detailed sec- tions of the formation, the outcrops being so very patchy and discontinuous that the most that is seen is the rather massive limestone beds. THICKNESS In the measured section at Fairview bluff, described in the previous chapter under the heading "Ste. Genevieve formation," beds 14 to 18 inclusive, which represent the Shetlerville and Eenault, comprise a thickness of 106 feet. The upper part of this interval is talus-covered in the south bluff, where the section was for the most part taken, so that the base of the superjacent sandstone may and probably does continue below the limits indicated, in which case the thickness of the two formations named will be less by that amount. They doubtless are less than 100 feet, but cannot well be less than 80 or 90' feet. In most sections where these two formations occur, the outcrops are more or less obscured by talus deposits, but the interval in which they must be included is commonly somewhat less than 100 feet, and perhaps 75 feet may be taken as a fair average. The thickness of the Shetlerville formation, characterized by the Amplexus geniculatus fauna, is less than half the total thickness of the interval, and probably is generally about 25 feet, which would leave the thickness of the Eenault commonly about 50 feet, although in places the thickness may be somewhat greater or considerably less than this. STRATIGRAPHIC RELATIONS The stratigraphic relations of the Eenault with the subjacent Shetlerville have already been mentioned in connection with the description of the older formation. At one locality below Eosiclare, where the section across the con- tact between the two formations has been observed, apparently there is exhibited a distinct line of unconformity (see figure 10). This phenomenon may be a local feature, however, with no especial stratigraphic significance. The best exhibition in Hardin County of the stratigraphic relations of the Eenault with the overlying sandstone, may be seen in the Ohio Eiver bluffs east of Cave in Eock. In this locality the actual contact of the sand- stone upon the limestone is exposed in the bluff near the line between sees. 16 and 17, T. 12 S., E. 10 E>. (see figures 11 and 12). In a shallow ravine one-fourth mile farther west, not more than six or eight feet of this portion of the section is covered. The exposures are such as to indicate that the sandstone succeeds the limestone abruptly, with no intergradation of sedi- ments. There can be no doubt of the existence of an unconformable surface —10 G 146 GEOLOGY OF HAKDIN COUNTY between the two formations, indicating a withdrawal and succeeding readvance of the sea in the Hardin County area at this time. At the first locality described above, six to eighteen inches of fragmental material is present between the limestone and the overlying sandstone. This layer is composed of flat pebbles, slabs more or less irregularly disposed, much lime-sand, quartz sand of large rounded grains, and many fragments of fossils, some of which are worn and rounded. The layer is plainly a beach formation which has been accumulated by the waves and currents. The exposure is shown in figures 11 and 12. Fig. 11. — Photograph of the unconformable contact of the Renault limestone and the overlying Bethel sandstone, in the bluff 3 miles east of Cave in Rock, on the line between sees. 16 and 17, T. 12 S., R. 10 B. West of Hardin County, in the Cache River bluffs about five and one-half miles south of Vienna, Johnson County, there are other exposures exhibiting unconformity at this same horizon. In this section, which is exposed in the railroad cut on the Chicago, Burlington and Quincy Railway, the Renault limestone is followed by a sandstone formation 12 feet thick, in the base of which there is a calcareous conglomerate bed, while two miles to the east there is no sandstone present, and the Paint Creek shales with thin limestone layers, rest upon the Renault with a conglomerate bed in the base of the upper formation. UPPER MISS1SSIPPIAX SERIES, CHESTER GROUP 147 PALEONTOLOGY Although some of the beds of limestone in the Renault formation are composed largely of more or less complete shells and other fossil forms, they are commonly so firmly imbedded in the matrix that it is difficult to secure specimens in good condition for accurate identification. Upon some of the weathered surfaces good fossils can be collected. Some of the denser and more compact beds have not afforded fossils at all, and but few localities have been met with where the more shaly layers are exposed in such a manner as to afford good collecting. Under these circumstances the Renault fauna from Hardin County is much more meagre than is the Shetlerville fauna, but Fig. 12. — Photograph of the unconformable contact of the Bethel sandstone and the Renault limestone, east of Cave in Rock. Same location as figure 11. sufficient material has been secured from a number of localities to make clear its general character. The two best collections from the Renault limestone of Hardin County have been secured from the outcrops in the Ohio bluffs above the railroad incline of the Fairview mine, below Rosiclare, from the same section that has furnished one of the best collections from the Shetlerville formation; and from the limestone outcrop at the mouth of the Good Hope shaft of the Fair- view mine, on the east side of the vein. A combined list from these two localities is given below, with the occurrence of the species indicated in the two columns at the right, the first column being the first locality mentioned, and the second the last locality: 148 GEOLOGY OF HARDIN COUNTY Fossils from the Renault formation collected in the Fairvieiv bluff and at the Good Hope shaft FOSSILS Triplophyllum spinulosum (M.-E. and H.) Talarocrinus trijugis M. and G Pentremites godoni DeFrance Pentremites princetonensis Ulrich Pentremites pinguis Ulrich Pentremites arctibrachiatus Ulrich Pentremites buttsi Ulrich Mesoblastus glaber (M. and W.) Meekopora eximia Ulrich Batostomella sp Stenopora sp Fenestella cestriensis Ulrich Fenestella serratula Ulrich Fenestella teriax Ulrich Polypora cestriensis Ulrich Lyrcpora sp Septcpora subquadrans Ulrich Streblotrypa nicklesi Ulrich Cystcdictya labiosa Ulrich Phractopora sp Orthotetes kaskaskiensis (McChesney) Productus inflatus McChesney Diaphragmus elegans (N. and P.) Pustula sp Spiriferina transversa (McChesney) Spiriferina spinosa (N. and P.) Spirifer increbescens var. transversa Hall . . Spirifer leidyi N. and P Reticulata setigera (Hall) Cliothyridina sublamellosa (Hall) Composita trinuclea (Hall) Orthcmychia sp Phillipsia sp In 1903 a collection was made by H. F. Bain from the outcrop at the mouth of the Good Hope shaft of the Fairview mine, the same locality from which the collection listed in column 2 of the preceding table was obtained. The following species were identified and referred to the Tribune limestone by Ulrich 1 , some of which have not been recognized in the more recent col- lections : iBull. U. S. Geol. Surv., No. 255, p. 24, (1905). UPPER MISSISSIPPIAN" SERIES, CHESTER GROUP 149 Zaphrentis spinulifera Zeacrinus maniformis Pentremites godoni Lyropora ranosculum Pentremites pyriformis Lyropora subquadrata Pentremites cervinus Spirifer increbescens Stenopora tuberculata Spirifer leidyi In reproducing this list here a number of manifest typographical errors have been eliminated. On translating the names as recorded by Ulrich into the generic and specific terms of the combined list given above, there are four entries which are not included in the larger list, these being Pentremites cervinus, Zeacrinus maniformis, Lyropora ranosculum, and Lyropora- sub- quadrata. There is perhaps some question concerning the correct identifica- tion of Pentremites cervinus which is a species known in recent collections only in much younger Chester formations. Zeacrinus maniformis is really a species of the genus Eupackycrinus and, from its known occurrence elsewhere, this or some closely allied form might well be expected in the fauna. The most important additions to the list are the two species of Lyropora. The genus has been recorded in collections from the river bluffs above the Fair- view incline, but it is represented by imperfect bases only, while the examples submitted by Bain evidently were so preserved that the species themselves were determinable. Full collections have not been made in Hardin County from other Eenault localities than those already recorded, but a few forms have been collected at a number of places. A good exposure of the formation just outside of Hardin County, is exhibited in the bank of Ohio River a short distance above the mouth of Grand Pierre Creek, where some of the weathered surfaces of the limestone are covered with great numbers of Pentremites. A number of well preserved bases of the bryozoan genus Lyropora have also been collected at this locality, as well as good examples of Talarocrinus trijugis of the characteristic Eenault type. In all these faunas from the Eenault limestone of Hardin County and adjacent regions, the most characteristic member is a form of Talarocrinus, although the crinoid is not commonly abundant, nor is it often found in a well-preserved condition. The genus Talarocrinus was present in the Shet- lerville formation, but the species commonly met with there is T. buitsi, a form with a small, smooth, subcorneal base. As in all members of this genus the base is composed of two plates, but instead of forming a subcorneal or narrowly bowl-shaped cup as in T. buttsi, the base in the Eenault form is disk-shaped with the suture between the two plates rather deeply impressed, giving to it a distinctly bilobed appearance. This type of base may be present in a number of different species of Talarocrinus which may either be associated or be characteristic of somewhat different geological horizons, or they may be in part geographic variations of one type. In probably 99 cases out of 100 it is onlv the bilobed base of the crinoid or detached radial plates that are 150 GEOLOGY OP HARDIN COUNTY observed in the fossil condition, so that if they do represent a number of species the distinguishing specific characters can rarely be determined. This type of Talarocrinus has been observed in the Shetlerville formation in only one locality, and there it is represented by a single specimen where it was collected loose from a glade-like surface, and it is quite possible that this one example may have originated in a layer of the Eenault limestone a little higher up. In any event there is a distinct faunal zone in the Hardin County section which is characterized by the species of Talarocrinus with a bilobed base, and this horizon is Eenault. Most of the associates of the bilobed Talarocrinus in the Eenault lime- stone are present also in the Shetlerville formation below, but there is a greater variety of Pentremites, and among them are introduced for the first time those forms of the "piriformis" type with the elongate base. The pecu- liar type of Talarocrinus, the elongate Pentremites, and the common presence of numbers of Lyropora bases constitute the characteristic features of the Eenault fauna. Every species in the fauna occurs elsewhere in the Lower Chester formations, especially in Eandolph and Monroe counties, Illinois, and none of the characteristic Ste. Genevieve limestone species have been found in the formation. The entire fauna is distinctly Chester in all of its features, but it is a Chester fauna in which the genus Archimedes is almost wanting. The spiral axes of this genus are rarely met with in the Eenault and where they are present the}^ are not common; in the two best collections made in Hardin County the genus has not been met with at all, and only one or two examples have anywhere been seen by the writer in the Eenault of the county. CORRELATION The correlation of the Shetlerville and Eenault formatious, the "Upper Ohara" limestone of Ulrich, is a question which constitutes the most funda- mental difference between Ulrich and the writer in the interpretation of the Chester group, and in considering this question a somewhat critical examina- tion into the methods of correlation advocated and used by Ulrich must be made. Ulrich has had much to say in regard to the unsatisfactory results to be attained in correlation by the method of "matching" genera and species. In regard to this method he says in one place 1 , "Under certain conditions, and in so far as basal stratigraphic conceptions are concerned, the idea is correct enough ; but when it comes to the accurate identification of minor units of the time scale it is inadequate and indeed more likely to lead us astray than to the truth." It must be admitted that the method of correlation by matching genera and species has been abused by reason of the too loose practices in the discrimination of fossil species, but to replace it by a method of matching individuals would be equally objectionable. In regard to this new method 1 Bull. Geol. Soc. Amer., vol. 27, p. 472, (1916). UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 151 Ulricli says 1 : "The new criterion is based on the logical belief that combina- tions of biologically unimportant characters can have existed but once, and that they endured for only short periods of time. Accordingly, absolute identification of such minor modifications of species in widely separated localities is regarded as establishing the essential contemporaneity of these occurrences." In the application of this method its author concludes that if he can secure two individuals of a given species that are exactly alike as regards a number of arbitrarily selected minor characteristics, the two speci- mens must be contemporaneous. It is a well recognized fact that all species are variable, some more so and others less. It is also an established fact that each species has lived during a time period of greater or less duration, some species being much longer lived than others, and as our observations are extended we are constantly accumulating new data in regard to the duration in geologic time of fossil species. Now the liability of finding two examples of a given species that are identically alike in a number of characters, is clearly dependent upon the number of examples which may come under observation. If the observer compares a dozen specimens, all from the same bed at one locality, or from different beds at different localities, no two may be exactly alike, but if a thousand individuals be examined two or more that are essentially alike in all the characters selected for comparison are more likely to be met with. From the very nature of the case the longer-lived a particular specific form may be, other factors being equal, the greater number of individuals will be available for comparison, and consequently the greater liklihood will there be that exactly similar individuals will be found; under these circumstances it is apparent that there is no basis for a "logical belief ' that such similar individuals must have been contemporaneous. In his Chester correlations Ulrich has laid great stress upon the im- portance of the members of the genus Pentremites, and in his Kentucky paper he has named without descriptions, at least 23 new species and varities. 2 iBull. Geol. Soc. Amer., vol. 27, p. 456, (1916). 2 This practice of naming species without adequate descriptions cannot be too severely- criticised. Some of the forms may be recognized with a fair degree of certainty from the illustrations, but in some cases there are as great differences shown among the examples figured to represent a single species as are shown between examples that are referred to different species. If strict justice were done none of the species would deserve recognition. The sorts of characters which seem to have been used by Ulrich for the differentiation of his several species are the flatness or degree of concavity of the inter- ambulacral spaces, the straightness or concavity of the base from the lower ends of the ambulacra to the stem attachment, and the pedunculation of the base. Such characters as the flatness or concavity of the interambulacral areas, which may actually vary in the several areas of one individual, are considered by him as having real phylogenetic value. Other characters which have been used in his differentiation of species are to be found in the ambulacral areas, and these are of somewhat higher taxonomic value than the mere variations in actual form. Many internal characters of the Pentremites, con- nected with the hydrospires, are disregarded altogether, although they might be expected to exhibit features of much greater phylogenetic value than any of those used, but such characters require laborious preparation and are difficult to observe, and even under the best of circumstances they would be well exhibited in only occasional individuals. However, any classification or grouping of the members of this genus based upon any- thing less than all the characters, both external and internal, must be considered as artificial. It is undoubtedly safe to state that, with the data Ulrich has supplied, no one would be able to successfully distribute the pyriform examples of Pentremites in a large collection among his species Pentremites pyriformis, girtyi, welleri, lyoni, buttsi, abruptus, and symmetricus. 152 GEOLOGY OF HARDIN COUNTY This genus is certainly an important one in the Chester faunas, many of the forms named by Ulrich are doubtless good species, and the importance of the genus for the purposes of correlation cannot be denied, but by using the method of matching individuals among the species of this genus, the correla- tion of any formation from the St. Louis limestone to the Paint Creek of the Chester, with any other formation in the same series, can be established. Ulrich so regards his determinations of the Pentremites that the correlation of the "Upper Ohara" limestone with the Renault seems to be unreasonable. One of his new species of Pentremites is P. princetonensis. This is said to be the "most common species of the genus in the Ste. Genevieve limestone/' and the specimens which he illustrates are from the Fredonia oolite and the "Upper Ohara" limestone. This occurrence of the species constitutes one of his reasons for including both of these formations in the same stratigraphic unit, the Ste. Genevieve limestone. In his original manuscript he did not recognize this species at all in the Renault of Monroe and Randolph counties, but after being shown an individual from the Renault which agreed exactly in all its details with the type of the species from the "Upper Ohara/' and after admitting his inability to distinguish the two examples although both were well-preserved specimens, he added the following statement regarding the species 1 , "A variety distinguishable only when the specimens preserve the minute characters of the ambulacral areas, is found in both the Renault and Paint Creek formations of Illinois." Of the two common types of Pentre- mites in the Renault fauna of western Illinois, one includes individuals which are perfectly typical of the forms named P. princetonensis and P. pinguis by Ulrich, and which are indistinguishable from examples of the same species from the Shetlerville and Renault formations of Hardin County (see plate IV, figs. 1-12), and the other is a pyriform type which furnishes examples which exactly match examples of P. huttsi, another of Ulrich's newly named species from the "Upper Ohara" (see plate IV, figs. 16-20). By unquali- fiedly using the method of matching individuals, the Renault of Monroe County can be shown to be the equivalent of the Renault or Shetlerville of Hardin County, or even of the Fredonia member of the Ste. Genevieve lime- stone. By matching individuals of P. princetonensis the St. Louis limestone can be made to be the equivalent of the Renault of either Hardin or Monroe counties. Furthermore, by matching individuals of P. godoni (P. planus Ulrich) from the Paint Creek of St. Clair County and from the Renault of Hardin County (see plate IV, figs. 33-36), these two formations can be shown to be contemporaneous. Since things equal to the same thing must be equal to each other, it might be demonstrated by all of these matchings that the St. Louis limestone is the correlate of the Paint Creek. It is to absurdities Form. Chester Ser. in W. Ky., p. 243, (1918). UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 153 of this sort that we are bound to be led if the method of correlation by match- ing individuals is followed systematically. Any student who will take the time to compare large numbers of Pen- tremites from different horizons and from different localities of the same horizon, will find an almost endless variety of individuals. Practically every collection will exhibit some varieties not found elsewhere, and a grouping of individuals which seems to be satisfactory in the collection from one locality is likely to prove to be unsatisfactory in a collection from another locality, perhaps of the same horizon. These forms undoubtedly exhibit a considerable degree of contemporaneous, geographic variation as well as variation with the progress of time, and for such geographic variation there seems to be no place in the system devised by Ulrich. Those species of the genus Talarocriniis with the thickened plates form- ing a bilobecl base, have been considered by the writer as very characteristic of the Lower Chester faunas, more especially of the Eenault, and the com- mon presence of such species in the Eenault of Monroe and Randolph counties and in the "Upper Ohara" of southeastern Illinois, has been considered as one of the lines of evidence for the correlation of these units. Ulrich does not agree with this conclusion for the reason that in his method of matching individuals he has not as yet found two examples, one from the typical Eenault and the other from the "Upper Ohara,^ which he considers to be exactly alike in all particulars. In considering this evidence, however, it must be constantly kept in mind that complete examples of Talarocriniis are exceedingly rare. Many separate radials and plates of the base, and an occasional dorsal cup, are met with at most localities, but probably not one example in a thousand, which can be easily recognized as being a member of the genus, preserves the entire body of the crinoid, which must be available to make the comparison Ulrich demands. It is not reasonable to insist that because exactly similar examples have not been found among the score of individuals that have been available for comparison, that such similarities would not be met with among the thousands of examples that are represented by fragments only. Furthermore, the fact known to every observing pale- ontologist, that the characters of different groups of organisms possess varying time values, has a direct bearing upon the problem before us. Generic char- acters in some groups of organisms may possess as limited time value as specific or varietal characters in other groups. Many brachiopods, for in- stance, are notoriously conservative and slow changing and are consequently of less time value than some other fossil organisms, and in such forms the matching of individuals could not possibly furnish any basis for correlation. The genus Talarocriniis, on the other hand, is one of the camerate crinoid s which are notoriously rapidly changing forms, and furthermore this genus is one of the last representatives of the order and lived at a time when 154 GEOLOGY OF HARDIN COUNTY morphological changes were taking place in the group with extreme rapidity. It is well known that organisms in this stage of their evolution are not only changing rapidly as a whole, but are subject to very great geographic varia- tions, so that species are liable to possess only limited geographic range. In a group of organisms of this sort it is evident that the generic characters may have even greater correlative value than the specific or varietal characters of some other groups. One of the great absurdities of the method of correla- tion by matching individuals is the fact that no recognition is given to the possibility of contemporaneous geographic modifications within a genus, which may be of specific value. The facts in regard to the occurrence of members of the genus Talaro- crinus of the type indicated, are as follows. In Monroe and Eandolph coun- ties examples occur in the Eenault and Paint Creek formations and nowhere else. They are far more common in the Eenault, and no fairly complete fauna has anywhere been collected from the limestone of this formation without producing some examples, although complete specimens are rare.. Many Paint Creek collections have been made with no representative of the genus, but they do occur rarely. In Hardin County and the adjacent portions of Illinois and Kentucky, the "Upper Ohara" is the Talarocrinus horizon, and examples of the characteristic bases or of the radial plates may be found in most localities if sufficient search is made; they have about the same relation to the fauna as a whole as in the Eenault of Monroe and Eandolph counties. Complete specimens are rare, however, and no examples have been observed that are exactly similar to specimens from the Eenault of the more western counties, although individuals sufficiently alike to be considered as coming within the limits of the same species have been studied. In the Paint Creek horizon of Hardin County and the adjacent region, specimens of Talarocrinus have been met with but rarely. In carrying the observations still farther south, this type of Talarocrinus is found to be characteristic of and limited to the Gasper limestone of Butts and Ulrich. In his list of Gasper fossils, Ulrich 1 records four species of the genus, three of which are from the lower Gasper and only one from the upper member of that forma- tion, a distribution which compares with that of the same genus in the Eenault and Paint Creek of western Illinois. Indeed, Ulrich cites much evidence for the correlation of the Gasper with the typical Eenault and Paint Creek. On the other hand Ulrich does not recognize the presence of Talarocrinus in any list of Fredonia species nor in zone 1 of the Ohara, which is the "Lower Ohara" of this report and which is truly of Ste. Genevieve age. Now the results of Ulricrr's correlations are to consider the Talarocrinus-he&vmg Gasper horizons of Kentucky east of Caldwell County as equivalent to the Tatar ocrimis-be&rmg Eenault and Paint Creek of western Illinois, the two >Form. Chester Ser. in W. Ky., pp. 147-148, (1918). UPPER MISSISSIPPI^ SERIES, CHESTER GROUP 155 regions lying at opposite sides of the basin involved. In the intervening region, however, in southeastern Illinois, the Talarocrinus beds are considered by him as lying entirely beneath the Gasper Talarocrinus horizon to the east and the Eenault Talarocrinus horizon to the west, notwithstanding the fact that every single associate of this type of Talarocrinus in the region, is known also from the typical Eenault or from some still younger Chester fauna, and that the index fossils of the Ste. Genevieve are unknown in the fauna. In this place attention may again be called to the fact that Ulrich himself formerly recognized the "Upper Ohara" outcrop at the mouth of the Fairview mine at Rosiclare, as Tribune, which is the name first applied to the Gasper 1 , a correlation which he must necessarily abandon with his present views. In the author's opinion every bit of the paleontologic evidence points to the correlation of the "Upper Ohara" of Hardin County with the Renault of Monroe and Randolph counties; indeed, the correlation may be considered as being so well established that there need be no hesitation in extending the name Renault into southeastern Illinois, as is done in this report. In spite of Ulricas opinion that "the occurrence of a single finely drawn variety in two or more widely separated places — be they in the same province or not — is a more trustworthy indication of the contemporaneity of the beds containing it than would be any quantity of the indefinite testimony afforded by generic alliances," it still seems to the writer that a consideration of all the evidence afforded by a fossil fauna is more likely to lead to its correct interpretation. Generic alliances have their values in correlation, as do specific identities and individual similarities, but for obtaining the most reliable results, all these factors must be considered, and the whole make-up and complexion of the fauna must be scrutinized. Such an examination of the "Upper Ohara" faunas can lead to but one conclusion, that being that these faunas have their closest relationships with the Renault faunas of southwestern Illinois. Ulrich 2 has said "the function of the fossils is to identify horizons." In another place he says 3 "A single species, or preferably two or three constantly associated, rare or common species, may be of greater practical utility and often of more exact value in correlation than all the remainder of a large fauna." Another principle he has enunciated 4 is that "The sequence of minor but well-defined life zones, when found to agree in widely separated localities, is to be regarded as highly significant in establishing the essential contem- poraneity of the respective zones." In Hardin County and the adjacent region, the fossils do serve to identify a number of definite horizons, and these life zones do have a definite sequence throughout the region. In that portion of the geological section under im- J U. S. Geol. Surv., Bull. No. 255, p. 24, (1905). 2 Bull. Geol. Soc. Amer., vol. 27, p. 488, (1916). 3 Bull. Geol. Soc. Amer., vol. 22, p. 509, (1911). 4 Loc. cit., p. 510, (1911). 156 GEOLOGY OF HAUDIN COUNTY mediate consideration there is a well defined lower zone characterized especially by Platycrinus penicillus and Pugnoides ottumwa; this zone is followed by one in which the above mentioned species are wanting and which is character- ized especially by Amplexus geniculatus. This in turn is succeeded by a zone in which there are present one or more species of Talarocrinus with thickened basal plates, this thickening giving to the base a distinctly bilobed appearance. Members of the genus Pentremites are common throughout all three zones, as are brachiopods and other forms. There are a few species which occur in all three zones, but nearly all of those in the second zone are known also in the upper one, while the community between the first and second zones is much less evident. The succession across the same interval in the Eandolph-Monroe county region is as follows : first a Platycrinus penicillus-Pugnoides ottumwa fauna, identical in all essential respects with the same fauna in Hardin County, followed by the Aux Vases sandstone without fossils of any sort, and this succeeded by beds which are characterized by species of Talarocrinus having the same type of bilobed base, some of which are specifically identical with forms present in the third zone in Hardin County. The first and third zones respectively, in the two regions, are considered by the writer to be equivalent, but there may be alternate interpretations of the middle zone. It is an estab- lished fact that an important unconformity exists at the base of the Eenault in Eandolph and Monroe counties, which makes it possible that the beds with Amplexus geniculatus in Hardin County may be a somewhat older member of the Eenault which was not deposited in the more western portion of the basin. Under this interpretation the Aux Vases sandstone of Eandolph County would be represented by the unconformity between the Ste. Genevieve limestone and the Amplexus-heaimg beds in Hardin County, and the Shetler- ville would be represented by the unconformity at the base of the Eenault in Eandolph County. An alternative interpretation would be to consider the Amplexus geniculatus-bearmg beds of the Shetlerville formation, as the time equivalent of the Aux Vases sandstone. The Illinois basin in Chester time was an embayment lying between Ozarkia on the west and Cincinnatia on the east, extending northward to near the center of Illinois. The present loca- tion of the outcrops of the Aux Vases sandstone in the Mississippi Eiver bluffs of Illinois and Missouri, is near the western shore line of the Chester basin, where sand deposits would naturally be accumulated. The Hardin County section lies near the center of this ancient basin, probably fifty or sixty miles from the nearest shore. It was beyond the zone into which the beach sands of the time were conveyed, and although the sea was shallow as is shown by the cross-bedding of the strata, the only sediments being deposited were the calcareous remains of organisms which inhabited the basin, and the finely divided particles of mud which were held in suspension far beyond UPPER AIISSISSIPPIAX SERIES, CHESTER GROUP 157 where the sand was dropped. These conditions might reasonably account for the equivalency of the Shetlerville limestone and shale of Hardin County, with the Aux Vases sandstone of the Mississippi Valley. The correlation of the geologic sections in the two regions, viz., Randolph Count} 7 , Illinois, and Hardin County, Illinois, as interpreted by Ulrich and again as interpreted by the writer, is shown in the two accompanying tables. In the following table are included all those forms in the Renault fauna of Hardin County that have been specifically identified, and their occurrence 'in other regions and horizons is indicated in the columns to the right and left. In the columns to the right the occurrence of the species in equivalent or younger formations is indicated, while in the columns to the left the occur- rence in older beds is shown. When the species are represented elsewhere by closely allied forms, not specifically the same, the fact is indicated by in the proper column ; specific identity is indicated by X. In column 1 a record is made of the occurrence in formations of pre-Ste. Genevieve age; column 2 is Ste. Genevieve, and column 3, Shetlerville. On the right, column 4 is the Renault limestone of Monroe and Randolph counties; column 5 is the Paint Creek; and column 6, any Chester formation younger than the Paint Creek. Table shoicing geologic range of the Renault fauna RENAULT FAUNA X X X ? X X X X X X X X o o X X X Triplophyllum spinulosum. Talarocrinus trijugis Pentremites godoni Pentremites princetonensis Pentremites pinguis Pentremites buttsi Mesoblastus glaber Meekopora eximia Stenopora tuberculata .... Fenestella cestriensis Fenestella serratula Fenestella tenax Polypora cestriensis Lyropora ranosculum Lyropora subquadrata .... Septopora subquadrans . . . Streblotrypa nicklesi Cystodictya labiosa Phractopora sp Orthotetes kaskaskiensis . . . Productus inflatus X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 158 GEOLOGY OF HAKDIN COUNTY Table showing geologic range of the Renault fauna — Concluded RENAULT FAUNA X X X X X X X X X X X X Diaphragmus elegans Spirif erina spinosa Spirif erina transversa Spirifer increbescens var. transversalis Spirif er leidyi Reticularia setigera Cliothyridina sublamellosa Composita trinuclea X X X X X X X X X X X X X X X X The known Renault fauna of Monroe and Randolph counties. Illinois, is considerably larger than that of Hardin County, but the above tabulation shows that every species from Hardin County, except two, is present also in the fauna of the typical Renault of the more western counties. These two exceptional species, although not as yet recognized in the Renault of the west, are both known from Chester formations younger than the Renault in the same region. Most of the additional species in the faunas of the typical Renault are crinoids which even in that region are known only locally, many of the collections being essentially like those that have been recorded from Hardin County. In addition to the identity of species in the Renault faunas of Hardin, and Monroe and Randolph counties, the same faunal peculiarities are exhibited in both regions. Both faunas are characterized by species of Tatar ocrinus with the bilobed base, some species being common to the two regions; in both faunas the bases of Lyropora are present in most collections, and in some localities they are common; it is not always possible to identify the species of this bryozoan because of the lack of preservation of the fenes- trated portion of the colony, but it is quite certain that the same species occur in both regions. Furthermore the Renault association in both regions is a fauna with decided Chester aspect in which the genus Archimedes is usually rare, or is wholly absent, in very strong contrast with the higher Chester faunas. The faunal characteristics of these two limestones in the two regions of Illinois are so uniformly the same, that their correlation on the basis of the paleontologic evidence cannot be gainsaid. The stratigraphic succession in southeastern and in western Illinois affords other evidence supporting that of the fossils in the same correlation. In both regions the Renault is followed by an arenaceous or at least siliceous formation, the Yankeetown of Monroe and Randolph counties, and the Bethel sandstone of southeastern Illinois and Kentucky, which separates it from a higher calcareous horizon which has been called Paint Creek in western Hli- UPPER M1SSISS1PPIAN SERIES, CHESTER GROUP 159 nois, and which is equivalent to some portion of the formation that Butts has named Gasper in Kentucky. In the Paint Creek the pentremites occur locally in enormous numbers and exhibit a great range of variation. The Gasper also is a notable pentremite horizon, and Ulrich has recognized essentially all of the same species in the two formations. Other faunal ele- ments of the Paint Creek and the upper Gasper, and the equivalent of these beds in Hardin, Pope and Johnson counties, Illinois, are in such close con- formity that the equivalence of this horizon from western Illinois to central Kentucky is supported by abundant evidence. The nature of this evidence will be considered in greater detail in the discussion of the Paint Creek form- ation in this report. Bethel Sandstone name and general distribution The Yankeetown formation was originally described from Monroe and Eandolph counties, Illinois, the type locality being in the neighborhood of Yankeetown School near the southeastern corner of the first of these counties. It is a very persistent unit in the Chester succession of western Illinois, although it is not of great thickness. The formation has been traced con- tinuously from northern Perry County, Missouri, through the southeastern corner of Ste. Genevieve County of the same state, northward into Illinois across Eandolph and Monroe counties and into St. Clair County, where it passes beneath the Pennsylvanian formations. Throughout this region the formation probably nowhere exceeds twenty feet in thickness, and in places it is considerably less. It is everywhere a siliceous formation, many of the surface exposures being essentially chert. In most localities, however, the chert is more or less sandy in bands and streaks, and in places it is essentially a quartzite. When Engelmann described the Chester section of southern Illinois 1 , he recognized an alternating succession of sandstones and limestones which were numbered from the highest to the lowest, the odd numbers being limestones and the even ones sandstones. He believed that his two lowest divisions, Nos. 9 and 10, were not widely distributed, but he especially called attention to the lowest sandstone, No. 10, and its superjacent limestone in the Cache Eiver bluff of Johnson County, at Indian Point south of Vienna 2 , resting on what he believed to be St. Louis limestone. Field studies of the last few years have shown that the limestone in this section that Engelmann believed to be St. Louis is really the upper part of the Eenault; his sandstone No. 10 there- fore, occupies the position of the Yankeetown, and his limestone No. 9 is the Paint Creek. In Hardin County, however, the Chester sandstone designated 1 Trans. St. Louis Acad. Sci., vol. 2, pt. 1, p. 189, (1863). 2 Geol. Surv. Illinois, vol. I, p. 380, (1866). 160 GEOLOGY OF HARDIN COUNTY as No. 10 by Engelmann has proven to be the Rosiclare, and limestone No. 9 was the "Lower Ohara/ ? Shetlerville and Eenault together, and the true equivalent of his No. 10 in Johnson County was included in sandstone No. 8, to which he had applied the name Cypress. The recent field work has clearly established the fact that the sandstone referred to by Engelmann as No. 8 or Cypress in Hardin County, includes two sandstone units with a shale formation between, and that the uppermost of these sandstone units alone is the Cypress; that the shale member, with some limestone plates, is the equivalent of Engelmann' s No. 9 in the Johnson County section; and conse- quently that the lower sandstone member occupies the position of No. 10 in the Johnson County section which is the horizon of the Yankeetown. In Hardin and Pope counties and southward into Kentucky, this lower sand- stone is a massive formation at least 100 feet thick in places, very different in this respect from the thin, chert-like typical Yankeetown, and furthermore it is known that the siliceous deposits at this horizon are locally lacking entirely in southeastern Johnson County. Butts has given the name Bethel 1 to this sandstone formation, from a locality in Crittenden County, Kentucky, and this name may be extended to Hardin County. The only reason for not extending the name Yankeetown into Hardin County is the difference in lithologic character and the fact that the deposit is not certainly known to continue uninterruptedly from Randolph to Hardin counties. The Bethel sandstone can everywhere be recognized in Hardin County, from the fact that it is the first of the sandstones of importance coming into the general geological section above the great predominantly limestone mass, including downward the Renault, Shetlerville, Ste. Genevieve, St. Louis, Warsaw, and Osage, in all 1,400 or 1,500 feet, extending from the base of the Bethel down to the top of the Chattanooga shale. In this entire thickness the inconspicuous Rosiclare sandstone, 5 to 20 feet thick, 60 to 100 feet below the Bethel, is the only persistent sandstone formation. The Bethel sandstone is nearly everywhere a bluff-forming bed, rising more or less abruptly above the easily broken clown Renault and Shetlerville formations, which are very commonly covered by Bethel sandstone talus. The distribution of the formation follows closely that of the underlying Renault in all three of the structural areas of the county that have been described (see figure 3). In the southeastern portion of the county the Bethel outcrops in a line of bluffs starting at the Ohio River bottoms a little over two miles east of Cave in Rock, and extending in a northwesterly direction to sec. 33, T. 11 S., R. 9 E., where it is interrupted by the Peters Creek fault, which limits the central faulted zone. Throughout the extent of this outcrop the strata, dip to the northeast, so that from the summit of the sandstone bluff a dip slope underlain by the formation extends for some distance down the dip. 1 Miss. Form, of W. Ky., p. 63, (1918). UPPER MISSISSIPPI AN SERIES, CHESTER GROUP 161 From sec. 33, the Bethel sandstone outcrop continues in a southwesterly direction in the line of bluffs which continue to Big Creek, one and one-half miles northwest of Elizabethtown. Farther to the southwest the Bethel caps the two summits of the isolated hill which rises from the Ohio Eiver bank, just below Eosiclare. In the complexly faulted belt number two of the faulted zone, lying northwest and west of Elizabethtown, the outcrops of the Bethel sandstone are much scattered and occupy comparatively small areas. One such area lies upon each side of Hosick Creek from half a mile to a mile north of Bassett School. A number of small patches of the sandstone are present less than a mile north of the iron bridge over Big Creek near Elizabethtown. A some- what larger area is present in the hills southeast of the mouth of Hogthief Creek, and still another small area occupies the summit of the hill just south of Stone School. In the extreme southwestern part of the county the Bethel occupies a portion of an irregularly outlined fault block east of Wallace Branch and about two miles above its mouth. West of Wallace Branch this sandstone forms the summit of Melcher Hills and of a number of smaller elevations between those hills and the Shetlerville fault. West of the Shetler- ville fault the Bethel formation extends to the Hardin-Pope county line, and in the Ohio Biver bluffs at and on either side of the county line there is a most excellent exposure of the sandstone which exhibits a greater thickness than does any other outcrop in the county. In the more northern part of belt one of the central faulted zone, the Bethel sandstone occupies a belt extending from sec. 26, T. 11 S., B. 8 E., in a general easterly or northeasterly direction to sec. 20, T. 11 S., B. 9 E. Near the middle point of this line of outcrop the formation is slightly offset by a fault extending in a northeast-southwest direction. West of this interrupting fault the formation occupies its normal position in the stratigraphic sequence, but to the east it is limited by a more oblique northeast-southwest fault which brings it into fault contact with considerably younger Chester formations. In the northwestern part of the county the Bethel occupies its normal position in the stratigraphic sequence in a narrow outcrop extending from the Hardin-Pope county line to a point a short distance southwest of Lees Mine where the outcrop is offset for a little over one mile to the southwest by the Lee fault. From this fault the outcrop of the formation continues in a southeasterly direction to the Wolrab Mill fault, although it is slightly inter- rupted by a short fault running a little west of north through sec. 23,. T. 11 S., B. 8 E. LITHOLOGIC CHARACTER The Bethel sandstone is fine-grained, massive, and compact, and is com- monly more or less irregularly cross-bedded. Tt is rather uniform in texture —n G 162 GEOLOGY OF HAEDIN COUNTY although locally a few streaks of quartz pebbles the size of peas, have been observed. Upon the weathered surfaces it is } r ellow or yellow-brown in color, . becoming reddish-brown locally. Unweathered surfaces of the rock, when freshly broken into, are lighter colored than the long exposed surfaces, in some places being nearly white. Not infrequently such surfaces exhibit a strongly mottled appearance, brown spots having diameters up to one-tenth of an inch being thickly scattered through the lighter portion. Neither in this county, nor in the surrounding region in Illinois and elvntucky, does the Bethel exhibit the distinctly cherty phase which is so characteristic of the Yankeetown in the Mississippi River counties. Both formations, however, are essentially siliceous, the western one being a sandy chert, quartzitic in places, while the southeastern is a sandstone. DETAILED SECTIONS No significant detailed sections across the horizon of the Bethel sandstone have been measured in Hardin County, but a number of sections in Pope and Johnson counties which have an important bearing upon the interpretation of the formation in southeastern Illinois, may be recorded here. The first of these is the section already mentioned, in the Cache River bluffs six miles south of Vienna at Indian Point, in southern Johnson County. Section at Indian Point, Johnson County, Illinois, in sec. 32, T. IS 8., R. 3 E. Feet 5. Sandstone, massive, moderately coarse, yellow-brown in color, with fer- ruginous seams, cross-bedded, with numerous fragments of plant stems. Chester bed No. 8, or Cypress sandstone of Engelmann's section 30 4. Limestone and shale interbedded, mostly talus-covered, with slabs of lime- stone, an occasional outcropping limestone ledge. Fossiliferous. Over much of the slope this bed is completely covered by the sandstone talus. Bed No. 9 of Engelmann's section 3. Sandstone, irregularly thin-bedded, cross-bedded in part, with some thin shaly partings. Fine-grained, yellow-brown in color, more or less cal- careous. No. 10 of Engelmann's section 2. Limestone, arenaceous, with thin, discontinuous sandy partings, the lower six inches conglomeratic, with angular limestone pebbles, the contact with the bed below uneven 1 1. Limestone, oolitic or crystalline, much of it strongly cross-bedded, light gray in color. Fossiliferous. Exposed 12 In this section the limestone bed No. 1 carries the characteristic Talaro- crinus fauna of the Renault. Beds 2 and 3 together occupy the horizon of the Yankeetown and Bethel, beds 4 and 5 being respectively the Paint Creek and Cypress. Another section in which the Bethel is well exhibited is in the Ohio River bluff at Birds Point, north of Bay bottoms, about four and one-half miles below Golconda, in Pope County. UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 163 Section at Birds Point, Pope County Feet 6. . Sandstone, fine-grained, massive, yellow-brown in color. The Cypress sand- stone 70 5. Talus-covered, probably a continuation of the shale below 20 4. Shale, bluish to gray in color, thinly laminar and fissile 15 3. Sandstone, moderately fine-grained, yellowish-brown in color, massive be- low, becoming somewhat shaly towards the top 45 2. Talus-covered 80 1. Limestone, hard, bluish-gray in color, with Talarocrinus fauna of the Re- nault. Exposed 5 In this section the sandstone bed No. 3 is the Bethel sandstone. STRATIGRAPHIC RELATIONS The actual contact of the Bethel sandstone upon the subjacent Renault is well exhibited in the bluffs east of Cave in Rock, at the locality that has already been described in the discussion of the stratigraphic relations of the Renault (see figures 11 and 12). Again in the Cache Eiver bluff section in Johnson County, which has already been described in detail, the actual basal contact of the sandstone occupying this horizon is clearly exposed in the Bur- lington railroad cut at that point. This contact line between the Eenault below and the overlying sandstone is irregular and undulating, and the base of the higher formation includes a conglomeratic layer, all of which goes to show that the higher formation lies unconformably upon the Renault, and an unconformity between the two formations is certainly present continuously throughout southeastern Illinois and the adjacent portion of Kentucky. The contact line between the Bethel and the overlying Paint Creek formation has been observed at no point within the region under consideration. In the section at Birds Point in Pope County, four and one-half miles below Gol- conda, described above, the upper portion of the Bethel seems to become shaly in character, which suggests a gradation from the sandstone into the overlying shales without unconformity. Elsewhere, however, conglomeratic beds have been observed at a number of localities at the base of the Paint Creek, which would suggest an unconformity at this horizon. THICKNESS The thickness of the Bethel sandstone in southeastern Illinois is some- what variable. In the Cache River bluff section in southern Johnson County, twelve feet of sandstone equivalent to the Yankeetown and Bethel is exposed, a thickness that is comparable with that in the typical Yankeetown exposures in Monroe and Randolph counties. There are places in southeastern Johnson County where the formation is lacking entirely, but to the east, across Pope and into Hardin County, the formation becomes thicker and takes upon itself 164 GEOLOGY OF HARDIN COUNTY the typical characteristics of the Bethel. In the river bluff below Golconda, in Pope County, an interval of approximately 50 feet is occupied by the Bethel. In the Birds Point section four and one-half miles below Golconda, 45 feet of Bethel sandstone is exposed, but the talus-covered interval of 80 feet between the sandstone and the first outcrop of Renault limestone, makes possible a somewhat greater thickness for the formation. The usual thickness of the formation through Hardin County may be assumed to be approximately 50 or 60 feet, although in places, as in the river bluff below Shetlerville it is certainly greater than this, being at least 100 feet. PALEONTOLOGY Although imperfect fossil remains of various sorts can be detected in the Bethel sandstone in some places in Hardin County, no good fossils have been observed. The fragments which have been met with are brachiopods and bryozoans of common Chester types, among which may be mentioned axes of Archimedes, productoid shells, probably Diaphragmus, and Reticularia setigercb. Besides these invertebrates, fragments of plant stems are commonly met with. CORRELATION The correlation of the Bethel sandstone in southeastern Illinois with the Yankeetown of Monroe and Eandolph counties, is based upon the stratigraphic position of the beds in the two regions. In both of the regions the under- lying and overlying formations are calcareous with many fossils, and their correlation is established by abundant paleontological evidence, which in turn, gives sufficient basis for the establishment of the equivalency of the beds lying between the fossiliferous horizons in the two regions. In the earlier treatment of the Hardin County section by Engelmann 1 , the Rosiclare sandstone of the Ste. Genevieve formation was believed to be the equivalent of Bed No. 10 of the Chester section of Johnson County, and in consequence of this incorrect correlation he included the Bethel sandstone of this county in his sandstone No. 8, to which he had applied the name Cypress. The real importance of the Paint Creek shale in the Hardin County section was not appreciated by Engelmann, and he seems to have believed it to be a shale member in his Cypress sandstone. Paint Creek Formation name and distribution The Paint Creek formation was originally described from Monroe and Randolph counties, Illinois, where it is closely associated with the underlying iGeol. Surv. of 111., vol. 1, pp. 350-375 (1886), also Econ. Geol. 111., vol. 1, PP- 291 " 319, (1882). UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 165 Tankeetown formation. The distribution of the formation in the Mississippi Valley is essentially the same as that of the subjacent formation, extending northward from northern Perry and southeastern Ste. Genevieve counties, Missouri, across Mississippi River into Illinois, where the outcrop of the formation extends across Randolph and Monroe counties and continues into St. Clair County. Beyond the region of its typical development, the Paint Creek formation is well exhibited in southeastern Illinois, and continues southward into Ken- tucky where it is represented by some portion of the Gasper limestone. In the Cache River section in southern Johnson County, Illinois, Engelmann designated the Paint Creek as Bed No. 9 of the Chester group, but his Bed Xo. 9 in Pope and Hardin counties included the Renault, Shetlerville, and "Lower Ohara," because of his mistaken correlation of the Rosiclare sand- stone with Bed Xo. 10 in the Johnson County section. In Hardin County the Paint Creek shale is present in a number of more or less isolated areas within the central faulted region, but it is inadequately exposed in every one of these localities. Perhaps the best exposure in this part of the county is in the XE. 14 NE. % sec. 21, T. 12 S., R. 8 E 1 ., about three-fourths of a mile north of the iron bridge over Big Creek between Eliza- bethtown and Rosiclare. In the hills extending northeast from this locality the Paint Creek is undoubtedly present wherever the proper portion of the geological section constitutes the surface rock, but it is for the most part more or less completely covered. The presence of the formation may be determined in places by fragments of shale in the wash of gullies or small streams, and elsewhere by topographic depressions or sags in the surface indicating the presence of a soft formation much more easily broken clown than the resistant sandstones which occur both below and above the shale. Fragments of Paint Creek shale are met with in the point of the hill nearly clue north of Bassett School (north of Elizabethtown) and the northwest slope of the hill extend- ing from near the middle of the south line of sec. 12, T. 12 $., R. 8 E., to the middle of the north line of sec. 7, T. 12 S'. 3 R. 9 E., is covered, in sec. 12 at least, with the shale, dipping rather steeply to the northwest. The shale is well exposed in the bank of the shallow ravine heading just north of the house on the point of this ridge just north of the south line of sec. 12. Another locality where this shale formation is present within the median faulted zone of the county, is in the hill southeast of the junction of Hogthief Creek with Big Creek, in sec. 9, T. 12 S., R, 8 E., and extending eastward into the edge of the adjoining sec. 10. In this area the shale, along with the other forma- tions, dips to the southwest. Xo good outcrops of the formation have been observed in this area, but the underlying Bethel sandstone is well exposed and at a number of points the presence of the calcareous beds of the shale is sug- gested by the characteristic red clay residuum of the formation. 166 GEOLOGY OF HAKD-IN COUNTY Outside of the central much-faulted zone of the county, the Paint Creel formation is present in the northwestern part, where it occupies its propei position in the stratigraphic sequence. One of the best exposures of the shale in the whole county is in the bed of a small tributary of Pinhook Creek, running northwest across the SW. 14 sec. 7, T. 11 S., K. 8 E., about two miles north of Hicks. The formation is fully exposed in the bed of Buck Creek in the SW. 14 sec. 34, T. 11 S., E. 7 E. At both of these places the position of the shale between the Bethel sandstone below and the Cypress sandstone above is clearly exhibited, both sandstones being exposed at each locality. At the last described locality, the side ravine extending southwest to the corner of the section, and the one on the northeast side of the main ravine running up to Pinhook School, are eroded on the shale, and so also the hollow heading near the road on the line between sections 8 and 9, about three-fourths of a mile southwest of Karbers Eidge and extending northwestward to Pinhool Creek. These ravines are examples of the fact stated above, that the outcro] of the Paint Creek shale is in places marked by low ground or hollows. Th( Paint Creek is well exposed for a long distance along the ravine in the NE. 14 sec. 26, T. 11 S., E. 8 E., one-half mile northeast of Gross, and in the shallow ravine just below the private road in sec. 21, T. 11 S., E. 8 E., and near th( center of its east line. In the southeastern, unfaulted portion of the county, the Paint Creek shale is well exposed in the ravine heading near the center of sec. 34, T. 11 S., E. 9 E., and in the mouth of the north-south ravine nearly opposite the on( last mentioned. The shale is not over twenty feet thick at these last two localities and seems to contain some rather thick sandstone layers which may indicate that the formation is changing into sandstone in this direction. In a southeastward direction from the localities just mentioned the Paint Creel seems to be persistent to the southeast corner of sec. 6, T. 12 S., E. 10 E., beyond which point the formation has not been observed and probably is not present. At this last exposure where the shale has been observed the following section has been studied by Mr. Butts : Section near the center of the SE. y± sec. 6, T. 12 S., R. JO E. Ft. 9. Sandstone, massively bedded ledge-maker, bottom of Cypress sand- stone 10 8. Not exposed, perhaps shale 5 7. Sandstone, laminated, coarse, micaceous, whitish . 2 6 6. Shale, blue 1 5. Sandstone 4. Clay, probably residuum from shale 1 3. Sandstone, thin-bedded 2 2. Not exposed 4 1. Sandstone, flaggy 40 UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 16?' The conditions in a southeastward direction from the section just de- ■ scribed may be interpreted in two ways. Either the Paint Creek thins out entirely, allowing the Cypress sandstone to rest directly upon the Bethel, or the Paint Creek undergoes a change in character from mainly shale to more or less thinly bedded sandstones. At any rate, beyond this point there is a continuous succession of sandstone beds between the Renault and Grolconda limestones, which approximate the united thickness of the Bethel and the Cypress sandstones, both of which formations may be considered as being ! represented in the entire mass. At about the horizon where the Paint Creek shale should be present there are layers of sandstone more thinly bedded than either the beds above or below. In the mapping of this portion of the county it has been considered that the Paint Creek is not present, and the Cypress- sandstone is shown resting upon the Bethel, the boundary between the two- formations being drawn as nearly as possible at the horizon of the thinly bedded sandstone. It must be recognized that this line is placed more or less arbitrarily, and no decision has been reached as to whether the thinly bedded , layers should be included with the lower or with the higher sandstone forma- tion. LITHOLOGIC CHARACTER The typical Paint Creek is a calcareous formation with varying amounts | of shale included with the limestone, in the different sections. In Monroe and Randolph counties where the Paint Creek was originally defined, a very characteristic member of the formation is a bed of deep red, non-laminated clay about fifteen feet thick, in its lower portion. Above this the shales are more or less calcareous and finally pass up into good beds of limestone with shale partings. In the Johnson County section the red clay member of the Paint Creek is wanting, although some variegated shales are present. The lower portion of the formation in Johnson County seems to be mostly shales, passing upward into alternating beds of shale and limestone. Some of the limestone ledges are three or four feet thick, but the whole formation is so heavily talus-covered that its details are not exhibited in any of the outcrops that have come under observation. In Pope and Hardin counties the forma- tion is made up even more largely of shales than in Johnson County. The bulk of the rock is soft, dark, clay shale, and very thinly laminated, very fine- grained sandstone, the latter breaking up characteristically into small chips which are scattered in abundance over the weathered outcrops. In some j exposures, only shales have been observed, but elsewhere there are thin, platy layers or discontinuous lenses of impure, siliceous limestone which contain some imperfectly preserved fossils. At the locality north of the iron bridge between Elizabethtown and Rosiclare, which has already been mentioned, | masses of one or more thin layers of rotten, ferruginous sandstone, with im- perfectly preserved fossils, are scattered over the talus-covered outcrop. These 168 GEOLOGY OF HARDIN COUNTY apparent sandstone layers were doubtless calcareous in their unweathered condition, but the calcareous matter has been leached out where the weathered rocks occur. The clay shale is dark in color, being nearly or quite black where it is exposed along stream courses and is kept moist. It is thinly laminar in structure. Where the shale is exposed in bluff faces it is commonly lighter colored, being bluish-gray or slightly greenish, and on weathering it breaks into flaky bits. At the locality north of the bridge over Big Creek between Elizabethtown and Iiosiclare, there are numerous blocks of limestone con- glomerate in the shale talus, which seem to have come from the base or near the base of the Paint Creek shale. Locally where this formation is present but is more or less completely covered with superficial deposits, its presence is indicated by the characteristic red clay residuum which is commonly derived from such deposits. Southward from Hardin County the shaly character of the Paint Creek persists into Crittenden Count}', Kentucky, but beyond, in Caldwell and Christian counties of the same state the formation becomes more calcareous, becoming a conspicuous limestone formation with oolitic beds, and is the upper member of the formation to which Uirich originally gave the name Tribune limestone. Late observations however, have shown that this name was unfortunately chosen, and that the limestone at Tribune is of quite a different age, and the name Gasper has been substituted. In view of the fact that it is clearly the same horizon from St. Clair County, Illinois, to central Kentucky, and that throughout its extent it is one of the calcareous members of the Chester group, the use of the same formation name for the unit through- out is highly desirable for the sake of uniformity in nomenclature. DETAILED SECTIONS The most significant detailed sections including the Paint Creek forma- tion, and showing its stratigraphic relations in southeastern Illinois, are nol in Hardin County, but in Pope and Johnson counties farther west. The fol- lowing section in the river bluff just below Golconda exhibits the Paint Creel as it is commonly developed in Hardin County : Section in Ohio River bluff below Golconda Fa 5. Sandstone, fine-grained, yellowish-brown in color, only the basal portion well exposed at this point, nearly the entire thickness much better ex- posed at the point of the bluff at the south side of Lusk Creek valley. . . 6( 4. Sandstone and shale, not well exposed, largely talus covered; an arenaceous shale bed about five feet thick exposed about twelve feet below the top. . 3( 3. Shale, greenish in color, argillaceous or somewhat sandy. Near the middle of the bed are some thin, lenticular, arenaceous limestone beds, com- monly one or two inches thick, with fossils 3( 2. Unexposed; talus composed largely of sandstone slabs 1 L 1. Sandstone, massive ledges similar in character to that at summit of sec- tion, exposed at this point 1< UPPER MISSISSIPPI AN SERIES, CHESTER GROUP 1G9 In this section, bed No. 3 is the Paint Creek, exhibiting a thickness of 30 feet. It is certainly thicker than this, however, for the intervals both above and below the shale exposure are talus covered, so that neither of the contacts of the formations can be seen. Below the sandstone in the quarry at the point of the bluff below the boat landing at Golconda, about 40 feet of the shale is exposed, which seems to be essentially the full thickness of the forma- tion at that point. One of the most significant of the southeastern Illinois sections of the Paint Creek is that in the Cache River bluffs in southern Johnson County, which has been described in the discussion of the Bethel sandstone on a previous page of this chapter. The importance of the section is in the fact that the beds at this locality are abundantly fossiliferous, while the exposures in Pope and Hardin counties are rarely fossiliferous, and the fossils which have been collected are commonly poorly preserved. THICKNESS One of the best exposures from which to estimate the thickness of the Paint Creek shale is in the section at Birds Point in the Ohio River bluffs, four and one-half miles below Golconda in Pope County, the section having already been given in the discussion of the Bethel sandstone; 15 feet of shale are exposed with a talus-covered slope of 20 feet above, which is doubtless also underlain by shale, making a total thickness of approximately 35 feet. In the river-bluff section at Golconda described just above, 30 feet of the shale are well exposed, with talus-covered intervals above and below which might increase the thickness to 50 feet. The exposure below the boat landing at Golconda is about 40 feet. The observations in Hardin County, while less definite than those in the Pope County sections, accord with them approxi- mately in regard to the thickness of the formation, and 40 or 50 feet may be assumed to be the normal thickness of the shale where it is fully developed. STRATIGRAPHIC RELATIONS The actual contacts of the Paint Creek shale with the underlying and overlying formations have not been observed commonly, the nearest approach to actual visible contacts being in the Buck Creek section and in the ravine north of Hamp Mine ; the evidence bearing upon the stratigraphic relations of the formation with the adjacent formations is therefore inconclusive. The most significant evidence met with is the presence of the limestone con- glomerate layer, masses of which have been observed in the talus at the base of the shale about three-fourths of a mile north of the iron bridge over Big Creek on the Elizabethtown and Rosiclare road. This conglomerate suggests a condition of unconformity between the shale and the Bethel sandstone 170 GEOLOGY OE HARDIN COUNTY beneath, but no corroborative evidence has been observed at any other locality, and for the present the stratigraphic relations between these two formations must be considered as not certainly determined. The stratigraphic relations between the Paint Creek and the overlying Cypress sandstone are also uncertain. The fact that the shale seems to be wanting in the southeastern part of the county suggests that it may have been removed by erosion subsequent to its deposition and before the initiation of the Cypress sandstone deposition, and such an erosion interval, if it existed, would be evidence to establish the unconformable relations between the two formations. The alternative conclusions are that the shale was never deposited in the area where it is now wanting, or that it is represented by thinly bedded sandstone, which, perhaps, is the most probable explanation. PALEONTOLOGY In Hardin Count}'', fossils have been rarely met in the Paint Creek shales, and those that have been seen are poorly preserved. They occur only in some of the thin, impure limestone layers which are locally present in the forma- tion. In the bluff section at Golconda in Pope County (see page 168) a few fossils have been collected in the siliceous limestone layers that are there present, but the most important fauna that has been secured from the forma- tion anywhere in southeastern Illinois, is from the exposure in the Cache Elver bluffs in southern Johnson County, described on a previous page in the discussion of the Bethel sandstone. At this locality there is much more lime- stone included in the formation than in Pope and Hardin counties, and in consequence the fossils are more numerous. The complete fauna from this locality so far as it has been collected, is as follows : Fossils from the Paint Creek formation collected at Indian Point, Cache River oluffs, Johnson County, Illinois, in sec. 32, T. IS 8., R. 3 E. Triplophyllum spinulosum (M.-E. and H.) Caninia n.sp. Pentremites godoni DeFrance Pentremites pyramidatus Ulrich Pterotocrinus serratus n.sp. Taxocrinus sp. Pachylocrinns cashensis n.sp Pachylocrinus spinuliferus (Worthen) Hydreionocrinus sp. Decadocrinus columbiensis (Worthen) Eupachycrinus maniformis (Y. and' S.) Archimedes meekanus Hall Archimedes proutanus Ulrich Archimedes cf. swallovanus Hall Polypora tuberculata Prout Lyropora ranosculum Ulrich Thamniscus furcillatus Ulrich Septopora subquadrans Ulrich Rhombopora sp. Streblotrypa nicklesi Ulrich Cystodictya labiosa n.sp. Glyptopora punctipora Ulrich Orthotetes kaskaskiensis (McChes- ney) Diaphragmus elegans (N. and P.) Dielasma illinoisensis Weller UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 173 Fistulipora excelens Ulricli Girtyella indianensis (Girty) Batostomella sp. Spiriferina spinosa (N. and P.) Stenopora tuberculata (Prout) Spirifer increbescens var. transvers- Lioclema n.sp. alis Hall Fenestella cestriensis Ulrich Reticularia setigera (Hall) Fenestella serratula Ulricli Eumetria vera (Hall) Fenestella tenax Ulricli Cliothyridina sublamellosa (Hall) Archimedes communis Ulrich Composita trinuclea (Hall) Archimedes compactus Ulrich Orthonychia sp. Archimedes distans Ulrich The impure limestone layers from the section at Golconda have afforded the following species : Fossils from the Paint Creek formation collected in the Ohio River bluff below Golconda Fenestella (several species) Eumetria vera (Hall) Archimedes sp. Leda sp. Cystodictya labiosa n.sp. Aviculopecten sp. Glyptopora cf. punctipora Ulrich Gastropods (several small unde- Chonetes cf. chesterensis Weller termined species) Girtyella indianensis (Girty) Phillipsia sp. Spiriferina spinosa (N. and P.) CORRELATION The one essential point to be determined in connection with the correla- tion of the Paint Creek formation in southeastern Illinois, is the question of its identity with the beds in Monroe and Randolph counties to which the name was originally applied. Such a correlation is established both upon the evi- dence of the fossils and upon the evidence of the stratigraphic succession. In ; both regions the same type of stratigraphic succession is present, that is an alternating series of limy and sandy formations, but the sandy members of 1 the series are attenuated to the. west and are consequently much less con- | spicuous in the Monroe and Randolph county sections. In the sections of the two regions under consideration the beds which have been termed Paint Creek in each, occupy equivalent positions stratigraphieally, and since both are situated within the limits of the same basin of sedimentation the presumption is that they are equivalent in age. A comparison of the fossil faunas of the Paint Creek formation in the region of its typical development, with the fauna of the Cache River bluff section in southern Johnson County, brings out so many common characteris- tics that the evidence, added to that of their equivalent stratigraphic position, is sufficient to fully establish their correlation. The Paint Creek fauna has much in common with that of the Renault, subjacent to the Yankeetown, but there are certain elements which are added 172 GEOLOGY OF HARDIN COUNTY to the assemblage which make it quite different in many ways. The species of Talarocrinus with bilobed base persists into the Paint Creek, and many of the bryozoans and brachiopods are common to the two faunas. Among the bryozoans Cystodictya labiosa is an important one whose range is restricted to the Shetlerville, Renault, and Paint Creek. G-lyptopora punctipora is an- other form having the same range, but it is less common than the Cystodictya. Crinoids other than Talarocrinm are common in the Paint Creek, there being a greater number of known species than in the Renault. The distinguishing paleontological features of the Paint Creek are found in the addition to the fauna of elements which were not present in the Renault. The Pentremites show a very great expansion, the number of individuals and species being vastly greater than in the Renault. Among the crinoids the most notable additions are several members of the genus Pterotocriniis, commonly repre- sented only by the ''wing plates/' and an occasional base of Agassizocrinus. A characteristic form of the Pterotocrinus "wing plates" is a flattened, spatu- late plate, with the outer border serrate. The most notable addition among the bryozoans is the great representation of the genus Archimedes. This genus was unusual in the Renault, being entirely wanting in many f annul es, and always rare wherever found; but in the Paint Creek it is much more com- mon, and at least one species, Archimedes compactus, is rather characteristic of the fauna, although it does occur only rarely in some of the higher forma- tions. Lyropora, represented by several species, continues to be a common form as it was in the Renault. Among the brachiopods, perhaps the most significant feature is the presence of a species of Chonetes. Members of this genus are unusual in the Chester faunas of the Illinois basin, the only species known being restricted to the Paint Creek fauna. Another feature of the fauna is the less common occurrence of Spiriferina transversa, and the great abundance of the typical form of Spiriferina spinosa. In the following list the species which have been identified in the Paint Creek fauna at the Cache River bluff are tabulated. Occurrence in the Paint Creek faunas of Monroe and Randolph counties is indicated by a cross in the first column to the right, their presence in higher Chester faunas being indi- cated in the second column. In the columns at the left the occurrence of the species in the Renault and Shetlerville faunas, is indicated, and also in the Ste. Genevieve limestone. In case the species is represented in any of these horizons by a closely related form, its presence is indicated by instead of X. UPPER MISSISSIPPIAN SERIES, CHESTER GROUP Table showing geologic range of the Paint Creek fauna 173 PAINT CREEK FAUNA X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X Triplophyllum spinulosum Caninia n.sp Pentremites godoni Pentremites pyramidatus Pterotocrinus serratus Taxocrinus sp Pachylocrinus cashensis Pachylocrinus spinuliferus Hydreionocrinus sp Decadocrinus columbiensis Eupachycrinus maniformis Fistulipora excelens Batostomella sp Stenopora tuberculata Lioclema sp Fenestella cestriensis Fenestella serratula Fenestella tenax Archimedes communis Archimedes compactus Archimedes distans Archimedes meekanus Archimedes proutanus Polypora tuberculata Lyropora ranosculum Thamniscus furcillatus Septopora subquadrans Rhombopora sp Streblotrypa nicklesi Cystodictya labiosa Glyptopora punctipora Orthotetes kaskaskiensis Diaphragmus elegans Dielasma illinoisensis Girtyella indianensis Spiriferina spinosa Spirifer increbescens var. transversalis Reticularia setigera Eumetria vera Cliothyridina sublamellosa Composita trinuclea Orthonvchia cf. chesterensis 174 GEOLOGY OF HARDIN COUNTY This tabulation shows that 36 forms have been identified specifically in the Cache River fauna, and of this number all but three are also recorded in the Paint Creek fauna from the Mississippi River counties or in some Chester fauna younger than the Paint Creek. The three species not found in the more western faunas are all crinoids, fossils that are always likely to be local in their occurrence, or at least in their preservation in such a condition as to permit their identification. Twenty-six of the species are also recorded from the Renault and Shetlerville formations. Of the ten species which do not occur in the two subjacent faunas, five are species of bryozoans belonging to the genus Archimedes. An examination of the entire list from Cache River shows the presence of such bryozoa as Cystodictya labiosa and GJyptopora punc- tipora, forms which are very characteristic of the lower Chester faunas; but their association with numerous species of Archimedes, and with the "wing plates" of Pterotocrinus, and especially with a form of these "wing plates" with serrated margins, suggests a higher position than the Renault, this asso- ciation being exactly what is met with in the typical Paint Creek fauna. On the basis of this paleontological evidence, supplemented by the stratigraphic succession, the correlation of the beds in the two regions is made doubly sure. The fauna from the limestone layers in the shale exposed in the river bluff at Golconda also bears out the correlation of the Paint Creek in south- eastern Illinois with that of Monroe and Randolph counties. This fauna, is small and the fossils are not well preserved, but the bryozoans Oystodicfi/a labiosa and Ghjptopora punctipora are quite certainly present, and axes of Archimedes are not uncommon. Perhaps the most significant species in the fauna, and one of the commonest ones, is the species of Chonetes that has been compared with C. chest&rensis. There can be little doubt concerning thp identity of the species which is the only member of the genus known in the Chester faunas of the Illinois basin, and which is wholly restricted, so far as it is known, to the fauna of the Paint Creek formation in the more western counties. The presence of Spiriferina spinosa in the fauna, rather than S. transversa, is another feature to be considered. In itself this occurrence might not be of much importance, but taken in connection with other evidence it adds its weight to the probability of the correctness of the Paint Creek correlation. In the extension of the Chester section southward into Kentucky the interval between the Bethel sandstone and the Cypress sandstone can be traced into Caldwell and Christian counties, but the lithologic character of the beds occupying the interval changes from shale with very subordinate limestone layers, such as is present in the Hardin County section, to an important lime- stone with many fossils. 1 This limestone constitutes the upper member of the formation in Kentucky to which Ulrich gave the name Tribune 2 , but this 1 Butts, Mississippian formations of western Kentucky, pp. 73-75, (1918). 2 U. S. Geol. Surv., Prof. Paper, No. 36, p. 55, (1905). UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 175 name was unfortunately chosen, the limestone at Tribune being an entirely different horizon, and belonging higher up in the Chester group. On this account the name Gasper 1 has been introduced by Butts to take the place of Tribune. For the sake of uniformity of nomenclature, and in view of the undoubted continuity of the Paint Creek formation from the Mississippi River sections through southern and southeastern Illinois and on into Ken- tucky and beyond, the name Paint Creek should be extended to include the upper member of the so-called Gasper, the lower Gasper probably being Renault. Cypress Sandstone name and general distribution When Engelmann elaborated his Chester section in the southern counties of Illinois, he recognized an alternating series of limestones and sandstones which he distinguished by number, beginning with the highest, the odd num- bers being applied to the limestones and the even numbers to the sandstones. He gave formation names to none of the units except sandstone No. 8, which he called the Cypress sandstone 2 from outcrops along Cypress Creek in Union County. Engelmann misinterpreted the formation in Pope and Hardin counties, and included in it the Bethel sandstone as well as the Paint Creek shales which lie between the two formations. Most if not all of the outcrops of the sandstone No. 8 which Engelmann has referred to in his Hardin County report, are in reality exposures of Bethel. The formation is widely distributed across southern Illinois from Union to Hardin counties, and con- tinues southward into Kentucky where Ulrich designated it as bed No. 1 of the Birdsville formation, misapplying the name Cypress to the Bethel sand- stone of this report. In tracing the Chester formations northwest from Union County into the typical Randolph County area, the Cypress becomes much attenuated and is represented in the section by the sandy upper portion of the Euma formation of that region, and in view of the observations made across the state it seems desirable to restrict the name Euma to the sandstone beds alone, or rather drop the name altogether and extend the use of the name Cypress, adding the shaly and somewhat calcareous beds of the formation to the underlying Paint Creek. The areal distribution of the Cypress sandstone in Hardin County follows closely that of the underlying Paint Creek and Bethel formations, except that in some parts of the county where the Bethel is well exhibited, the higher sandstone has been completely removed by erosion. In the southeastern portion of the county the Cypress occupies a belt extending from the Ohio River bluffs three miles east of Cave in Rock, in a northwesterly direction to Miss. Form. West. Ky., p. 64, (1918). Trans. St. Uouis Acad. Sci., vol. 2, pt. 1, p. 189, (1863), 176 GEOLOGY OF HARDIN COUNTY the Peters Creek fault limiting the central faulted zone. The maximum width of this belt is about one mile, but in places it is but a small fraction of a mile. In belt number three of the faulted zone (see figure 3), in the line of bluffs extending south westward from a little northeast of Lead Hill to Big Creek, the Cypress has been eroded, leaving only the Bethel as the bluff- forming bed ; but in both belts one and two of the faulted zone the Cypress is present. In the complexly faulted portion of belt number two the formation is limited in its surface outcrop to the southeastern border, and is well exposed in a number of the fault blocks from a point southeast of the Keelin School to Fair view. For the most part these are rather elongate, narrow areas, although two of the smaller fault blocks about one mile north of the bridge over Big Creek near Elizabethtown, include small areas of Cypress. In belt one of the faulted zone the Cypress occupies a narrow strip extending in a northeasterly direction from Gross to the valley of Goose Creek, with only a slight interruption at one point by faulting. A small patch of the sandstone is also present at the point of a wedge-shaped fault block in sec. 20, T. 11 S., R. 9 E. In the northwestern part of the county, the Cypress occupies its normal position in the stratigraphic succession, the outcrop being a narrow belt in the concentric arrangement of the formations about the Hicks dome, extend- ing from the Pope County line east to the Lee fault. Beyond this fault the formation is offset to the south for about one mile, whence it continues to the Wolrab Mill fault. LITHOLOGIC CHARACTER The Cypress sandstone is a massive formation, similar in many respects to the Bethel where that formation exhibits its greatest development. Much of it, however, especially the upper portion, is more evenly bedded than the Bethel, and in places the regular bedding in some exposures has much the appearance of regular courses of masonry. The sand of which the formation is constituted is fine-grained, and in the fresher exposures is light yellow- brown or buff colored, or in places almost white. The weathered surfaces are commonly darker colored, becoming brownish or reddish-brown. Locally there are numerous ferruginous streaks and bands through the sandstone, but else- where it is free from such markings. THICKNESS The entire thickness of the Cypress sandstone is not commonly exhibited in Hardin County, and no locality has been observed where the underlying and overlying formations are exposed in one continuous section. In most situa- tions the Cypress forms an abrupt bluff with no younger formation retained on the summit, and it is quite likely that the formation in such exposures has UPPER MISSISSIPPI AN SERIES, CHESTER GROUP 177 been reduced in thickness by erosion. One of the best bluff sections where the Cypress is present lying upon the Paint Creek shale is at Birds Bluff in Pope County, four and one-half miles below Golconda. In this section 70 feet of Cypress is exposed, and although the section is in Pope County it is probably representative for Hardin County as well. The total maximum thickness may of course have been somewhat greater than 70 feet, due to erosion from the top of the bluff, but the erosion probably has not been very great. Butts has estimated a thickness of 110 feet for the Cypress sandstone northeast of Cave in Rock, and about 80 feet in the northwestern part of the county. It is probable that the average thickness of the formation in Hardin County is between 80 and 100 feet. STRATIGRAPHIC RELATIONS The stratigraphic relations of the Cypress sandstone with neither the underlying nor overlying formations have been determined from observations upon the actual contacts of the beds, because these contacts are generally, if not everywhere in the region, entirely obscured by surficial deposits. The only fact that seems to have some bearing upon the question is that the Paint Creek shales may be wanting in the section in the eastern part of the county. If the absence of this formation is due to its removal by erosion preceding the period of deposition of the Cypress sandstone, such an erosion interval would indicate the presence of an unconformity at the base of the Cypress. On the other hand it is possible that the Paint Creek was never deposited in the area where it now seems to be wanting. The stratigraphic relations of the Cypress to the overlying Golconda limestone have nowhere been clearly shown in out- j crop because of insufficient exposures, but the presence of somewhat calcareous sandy beds with imperfectly preserved fossils near the horizon of contact between these two formations, suggests that there is a gradual transition from l the, lower formation to the higher. That the sand depositing conditions changed gradually at this time to the lime and shale depositing conditions of I Golconda time, is further shown by the recurrence of the earlier conditions ; exhibited by the sandstone beds which are locally present in the lower part of i the Golconda. The best exhibitions of the gradation from the Cypress sand- ! stone into the Golconda formation are to be seen in Pope County, in the \ neighborhood of Golconda, but the same conditions undoubtedly prevail j throughout Hardin County. PALEONTOLOGY The fossils most commonly met with in the Cypress sandstone are frag- ments of plant remains which are mostly imperfect trunks of Lepidodendron. Aside from these plant remains, imperfectly preserved specimens of inverte- —12 G 178 GEOLOGY OF HARDIN COUNTY brates are present. The forms most commonly met with are bryozoan frag- ments, among others the axes of Archimedes. These are perhaps most com- mon in the lowermost beds of the formation which may originally have been somewhat calcareous. CORRELATION The only question in correlation which need be discussed in the con- sideration of the Cypress sandstone in this report, is in connection with its equivalent in the Chester section exposed in Eandolph and Monroe counties. Because of the absence of an adequate fauna the correlation of the Cypress in this western Illinois section must rest primarily upon the correlation of the underlying and overlying beds which are characterized by prolific faunas. It has already been shown by the faunal evidence that the Paint Creek formation of the two regions in question is a continuous formation. The Cypress is a conspicuous sandstone formation succeeding the Paint Creek in southeastern Illinois, occupying the same stratigraphic position as the much less con- spicuous sandstone bed of the Euma formation in the more western counties. If the faunal evidence sustains the correlation of the limestone formations immediately overlying the sandstone in the two sections, then the correlation of the Cypress with the Euma is apparently established, and the name Euma should be dropped. A discussion of the faunal basis for the correlation of the Golconda limestone of southeastern Illinois with the lower Okaw of the more western region will be left for consideration under discussion of the Golconda formation, but it may be here stated that such a correlation is made, and therefore the Cypress and Euma sandstones are considered to be equivalent formations. It has already been brought out that the Bethel sandstone of Hardin County and the surrounding region becomes much reduced in thickness as it extends into the more western part of the Illinois Chester basin. The same condition is exhibited in the case of the Cypress sandstone, the formation being 80 or more feet thick in Hardin County, while in Eandolph it is reduced to less than one-third of that amount, and in Perry County, Missouri, it is wanting entirely. Golconda Formation NAME AND DISTRIBUTION The Golconda formation has been so named from the excellent exposures of the formation just north of Golconda, in Pope County, in the bluffs of Lusk Creek and Ohio Eiver. The formation has been recognized across Hardin, Pope, and Johnson counties, Illinois, and when detailed work is done it will doubtless be found to extend to the west into Union County. Throughout the whole of this region the formation is certainlv known to be present wherever UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 179 this part of the Chester section forms the surface rocks, and it is also charac- teristically developed in Crittenden and Livingston counties, Kentucky. The complete geographic distribution of the formation can only be determined when more detailed mapping has been done. The formation, however, is essentially a formation of the Ohio Valley region and does not continue as a distinct unit into the Mississippi Valley Chester region, although it is repre- sented in the Eandolph and Monroe county section by the lower division of the Okaw limestone. In Hardin County outcrops of the Golconda formation occur in each of the three regions into which the county is divided structurally. In the south- eastern region the formation outcrops in the point of the river bluff about four miles northeast of Cave in Eock, and about one mile from Fraileys Store. From this point the outcrop continues northwest in a narrow belt to its inter- section with the northeast-southwest faults which bound the central faulted area. Within the central faulted region of the county (see figure 3), the Gol- conda formation is exposed in a number of different blocks, the most extensive exposures being present southeast of the deeply downdropped Pottsville block, and in that part of the region lying southwest of Keelin School. A small area of outcrop is present just south of the middle of sec. 12, T. 12 S., E. 8 E., and another better exposed though smaller outcrop is present in the SE. % sec. 11, just to the west. The largest of the fault blocks exhibiting the Gol- conda formation lies mostly in sees, 15 and 16, T. 12 S., E. 8 E. A large hill which constitutes the major portion of this fault block is made up exclusively of Golconda limestone, capped by Hardinsburg sandstone the strata being essentially horizontal. In this block good outcrops of the formation may be seen just north of the middle of sec. 15, and along the tributary of Big Creek which flows southwest from the northeast corner of sec. 16. Another block within this faulted zone where the Golconda limestone forms the surface rock over a considerable area but is poorly exposed, lies west of Eosiclare. This block is narrowly elongate in outline, its northern extremity being at Stone Church. The Golconda formation is present at the surface in much of the southern half of this block, the greater part of the outcrop being in sec. 32, T. 12 S., E. 8 E., but extending over into the adjacent sections to the north, south, and west. The actual exposures in the block are poor, being heavily covered with surficial material. One or two other smaller fault blocks in the region under consideration also include some Golconda limestone, as is shown upon the map. Northwest of the southern part of the central, more deeply downdropped belt of the central faulted zone, the Golconda limestone is exposed in only one small fault block in sec. 25, T. 12 S., E. 7 E., along a hill slope to the east of Wallace Branch. In the northern part of belt one of the central faulted 180 GEOLOGY OF HABDIN COUNTY zone, however, the Goleonda occupies its normal position as a narrow belt extending in a northeast direction from the Wolrab Mill fault, along the northern walls of Harris Creek and Goose Creek valleys. In the northwestern portion of the county the Goleonda is present as a narrow band, occupying its normal position among the formations arranged concentrically about the Hicks dome. From the Pope-Hardin county line eastward to Karbers Ridge, the formation forms the northern wall of the valley of Pinhook Creek, the easily decomposed shale and the thin bedded limestone of the formation being the cause of the location of the valley in its present position. Eastward from Karbers Ridge the Goleonda formation again outcrops in the north wall of a valley tributary to Big Creek. At the Lee fault the Goleonda, like the other formations, is offset to the southwest, and then continues eastwardly to the Wolrab Mill fault, again forming the north wall of a valley tributary to Big Creek. The location of these valleys upon the outcrop of the Goleonda formation is an excellent example of the operation of a general law, that in regions of tilted rock strata of alternate sandstone, limestone and shale beds, the valleys are invariably worn down in the limestone and shale formations, leaving the sandstone areas as projecting ridges. The reason for this is, of course, that the shale and limestone are softer and are more easily worn away than the sandstone or highly siliceous rocks, which are harder than the shales and less soluble than the limestones, and therefore more resistant to the processes of erosion. LITHOLOGIC CHARACTER The Goleonda formation is essentially a succession of limestones and shales, but the outcrops are so generally talus covered, that the details of the successive beds are obscured. It can be stated, however, that a considerable body of limestone is present, locally at least, in the basal portion of the forma- tion, followed above by a rather thick body of shale with intercalated thin layers of limestone, this again succeeded by another limestone member, or by two limestone members separated by shale. Besides the main shale members there are numerous, thin, shaly partings between beds of limestone. The individual beds o£ limestone in the formation vary somewhat in character, but in the main they are gray or bluish in color, and more or less crystalline. Some minor beds are dense and compact in texture, but such beds are- not conspicuous. The limestone beds in the lower portion of the formation are a mass of fossil fragments which weather with a rough surface, and in the weathered outcrops, at least, are likely to be stained a rusty brown color, either more or less uniformly or in irregular streaks. Oolitic beds are- present, but except locally they constitute a very inconspicuous portion of the formation and have been observed at but a single locah*tv in Hardin County. UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 181 Chert is very rare in the formation, and none whatever has been observed in any Hardin County locality. The shales of the Golconda formation are fully as variable in character as the limestones. Some beds are almost wholly argillaceous and gray, bluish, or essentially black in color. Other beds are calcareous, and gradations may be observed from the moderately calcareous shales to thinly bedded limestones with a considerable content of argillaceous material. At one or two localities in Pope County a thin bed of red shale has been met with in the Golconda formation, and while no such bed has yet been observed in Hardin County its occurrence would not be unexpected. In the lower portion of the formation some of the shale beds are more or less sandy, and they pass locally into sand- stone beds of minor importance. THICKNESS Because of the lithologic character of the formation the outcrops of the Golconda formation are commonly heavily covered with talus, so that the actual detailed succession of beds cannot be elaborated. It is also difficult in many places to determine with accuracy the thickness of the formation. Per- haps the best section in Hardin County for such a determination is that exposed in the NW. % NE. % sec. 16, T. 12 S., E. 8 E. At this point there is at least 140 feet of the limestone and shale, but neither the base nor the summit of the formation is determined by the underlying and overlying form- ations. It is probable, however, that the exposure constitutes almost the whole of the formation. In Pope County, near Golconda, there is an interval of about 150 feet, between the top of the Cypress sandstone and the base of the Hardinsburg, which would make the thickness approximately the same as that observed in Hardin County. STRATIGRAPHIC RELATIONS Wherever the Golconda formation is known it rests upon the Cypress sandstone, and although the contact between the two formations is covered in every locality examined, it is altogether probable that no stratigraphic break occurs between the two formations. The Cypress sandstone becomes somewhat more thinly bedded towards the summit, and in the lower portion of the Golconda there are arenaceous shales and some impure sandstone beds. The lithologic gradation from the lower formation to the higher is probably complete. The stratigraphic relations of the Golconda formation with the overlying Hardinsburg sandstone are well exhibited in Pope County, in the Ohio Elver bluffs above Golconda, and also in Johnson County, but are not so well shown in Hardin County. At the locality above Golconda where the contact between the two formations has been observed the line is sharply defined, the change 182 GEOLOGY OF HARDIN COUNTY from limestone to massive sandstone is abrupt without the slightest inter- gradation, the upper surface of the limestone is uneven and rolling in char- acter, and the topmost portion of the limestone is oxidized to a distinctly red color. A similar condition is exhibited in Johnson County, in Bald Knob, •i miles south of Vienna. These facts point to a stratigraphic break and a plane of unconformity between the two formations. In Hardin County, on the Shawneetown road north of Cave in Eock, and in the bluff to the north of Goose Creek in Eock Creek Township, the Golconda passes into the Hardins burg through 10 feet or more of very thinly laminated sandstone, a condition that is not so distinctive of unconformable relations. PALEONTOLOGY The fauna of the Golconda formation exhibits certain characteristics which are persistent throughout its extent from Johnson County, Illinois, to Crittenden County, Kentucky. The best collections of fossils have not been secured in Hardin County, so that collections from outside of the county will be called upon to some extent in connection with the discussion of the fauna. One of the best collections of Golconda fossils from Hardin County has been secured from the southeastern part of the county, near the middle of the SE. % sec. 9, T. 12 S., E. 10 E., about three and three-fourths miles north- east of Cave in Eock. The following species have been identified in this collection : Fossils from the Golconda formation collected in the SE. 1 i sec. 9, T. 12 S., R. 10 E. Triplophyllum spinulosum (M.-E. Orthotetes kaskaskiensis (McChes* and H.) ney) Pentremites Platybasis n;sp. Productus ovatus Hall Pterotocrinus capitalis Lyon Diaphragmus elegans (N. and P.) Eridopora punctifera Ulrich Rhynchopora perryensis Weller Penestella (several species) Dielasma sp. Archimedes communis Ulrich Girtyella brevilobata (Swallow) Archimedes ccnfertus Ulrich Spiriferina spinosa (N. and P.) Archimedes lativolvis Ulrich Spiriferina transversa (McChesney) Archimedes swallovanus Hall Spirifer leidyi N. and P Archimedes terebriformis L T lrich Martinia sp. Polypora spinulifera Ulrich Reticularia setigeia (Hall) Reteporina flexuosa (Ulrich) Cliothyridina sublamellosa (Hall) Septopora subquadrans Ulrich Composita trinuclea (Hall) Orbiculoidea sp. Phillipsia sp. As already pointed out the best exposures of the Golconda formation in the central faulted zone of Hardin County are in the comparatively large fault block which lies mostly in sees. 15 and 16, T. 12 S., E. 8 E. Fossil col- lections have been made from a number of localities within this fault block, and the following combined list is a record of all the species which have been identified from these localities: UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 183 Fossils from the Golconda formation collected from exposures in the fault block lying in sees. 15 and 10, T. 12 8., R, S E. Triplophyllum spinulosum (M.-E. and H.) Pentremites platybasis n.sp. Pentremites lyoni Ulrich Pentremites sp. Pterotocrinus capitalis Lyon Cornulites sp. Spirorbis sp. Fistulipora excelens Ulrich Eridopora punctifera Ulrich Batostomella ? sp. Stenopcra tuberculata Prout. Fenestella cestriensis Ulrich Fenestella (several species) Archimedes swallovanus Hall Archimedes terebriformis Ulrich Polypora sp. Septopora subquadrans Ulrich Orthotetes kaskaskiensis (McChes- ney) Productus inflatus McChesney Productus ovatus Hall Diaphragmus elegans (N. and P.) Pustula sp. Camarophoria explanata (McChes- ney) Rhynchopora perryensis Weller Dielasma sp. Girtyella brevilobata (Swallow) Spiriferina spinosa (N. and P.) Spiriferina transversa (McChesney) Spirifer increbescens Hall, var. Spirifer leidyi N. and P. Reticularia setigera (Hall) Eumetria vera (Hall) Cliothyridina sublamellosa (Hall) Composita trinuclea (Hall) Orthonychia chesterensis M. and W. Nautilus ? sp. Phillipsia sp. These Hardin County lists do not include all of the species that are known from the Golconda limestone. Fossil collections from Pope and John- son counties and from Kentucky contain a number of additional species, any one of which may be looked for in this county. Xevertheless the already known Hardin County collections include most of the characteristic faunal elements of the horizon. Tin; Go]conda limestone species now known from outside Hardin County, but not yet observed within the county, so far as they have been specifically identified, are as follows. This list, however, does not include a number of species that have been collected in one Johnson County locality where an unusual Golconda fauna has been discovered. This fauna will be discussed separately because of its important bearing on the correlation of the formation. Fossils from the Golconda formation collected at miscellaneous localities outside of Hardin County Pentremites obesus Lyon Agassizocrinus sp. Stenopcra cestriensis Ulrich Anisotrypa solida Ulrich Lioclema ? araneum Ulrich Fenestella elevatipora Ulrich Fenestella serratula Ulrich Fenestella tenax Ulrich Archimedes compactus Ulrich Archimedes proutanus Ulrich Polypora cestriensis Ulrich Polypora tuberculata Prout Lyropora quincuncialis Ulrich Thamniscus furcillatus Ulrich Septopora cestriensis Ulrich Streblotrypa nicklesi Ulrich Crania chesterensis M. and G. Martinia contracta M. and W. Martinia sulcata Weller Euomphalus planidcrsatus M. and W. 184 GEOLOGY OF HARDIN COUNTY Besides the forms here listed there are a number of small species of molluscs,, commonly poorly preserved, that have not been specifically de- termined, and whose genera, also, in some cases are uncertain. Most of the species in these lists are wide ranging Chester forms and would not be out of place in any Chester fauna from the Shetlerville to the summit of the group, but there are a few forms that are highly characteristic, and the combination of species as a whole is known at no other Chester horizon. One of the most characteristic members of the fauna is Pterotocrinus capitals. This crinoid is characterized by five massive, subglobose, "wing- plates" upon the ventral side, quite different from the "wing-plates" of any other member of the genus. No complete example of the species has been collected, but such a one has been illustrated by Lyon. It is the massive "wing-plates" of the species that are commonly met with in the collections, and in places they occur in great numbers. The basal plates are less fre- quently met with. This species, however, has not been observed in the higher beds of the Golconda limestone, and it is most abundant in the basal portion of the formation. The bases of the peculiar crinoid genus Agassizocrinus become common in some of the Golconda formation collections, for the firsi time in any of the Chester faunas. The genus was present, but always rare, in the Paint Creek, but beginning with the Golconda it is a common Chester form. Another characteristic member of the fauna is the gigantic Pentre- mites obesus, but it is much more limited in its range and distribution than the plates of Pterotocrinus capitalis. It has only been found in the lower portion of the formation and as yet has been collected from but few localities, none of them in Hardin County. Another pentremite that is characteristic of the formation, and which is very much more common than P. ohesus is P. platybasis. One of the Golconda species which is introduced for the first time at this horizon in the southeastern Illinois section, is Camarophoria ex- planata, which is present in nearly every collection and which is very common or abundant in many localities. Oliothyridma sublamellosa and Composita trinuclea which were conspicuous species in the lower Chester faunas, continue as common members of the Golconda fauna, but their association with Camarophoria expianata in this horizon is very different from their earlier occurrence. Another important species in the Golconda fauna, one that has been collected in widely different localities from Johnson County, Illinois, to Crittenden County, Kentucky, but which is nowhere a common form, is Rhynchopora perryensis. Rhynchonelloid shells are everywhere unusual in the Chester faunas of the Illinois basin, and this species, originally described from the lower Okaw limestone of western Illinois, is the only one which so far has been recorded from the region, and it has been observed at no other horizon in the southeastern Illinois section. The bryozoans of the fauna are of the usual Chester forms, and most of them have a long geologic range UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 185 within the group. One of the characteristic features exhibited by this ele- ment of the fauna is the abundance of axes of the genus Archimedes belonging to a number of different species. Archimedes swallovanus is the commonest species. A. compactus which was so commonly present in the Paint Creek faunas is rarely met with here. A very unusual Golconcla fauna has been collected in Johnson County, in the NW. % SE. % sec. 22, T. 13 S., R. 3 E., in which there is an associa- tion of species that has not been met with elsewhere in southeastern Illinois, but which is present in the lower portion of the Okaw limestone in Randolph. County. Because of the important bearing that this collection has upon the correlation of the Golconda formation, it may be discussed in this place. The species present in the fauna, as they have been identified, are as follows : Fossils collected from the Golconda formation in the NW. % SE. y± sec. 22, T. 13 S., R. 3 E., Johnson County, Illinois Triplophyllum spinulosum (M. — E. and H.) Pentremites sp. Pterotocrinus capitalis Lyon Pterotocrinus sp. Crinoid plates Stenopora tuberculata Prout Fenestella (several species) Archimedes sp. Polypora sp. Coeloconus sp. Bryozoans (several undet. spec.) Orthotetes kaskaskiensis (McChes- ney) Productus ovatus Hall Diaphragmus elegans (N. and P.) Camarophoria explanata (McChes- ney) Dielasma sp. Spiriferina spinosa (N. and P.) Spirifer increbescens var. americana Swallow Spirifer increbescens var. Martinia contracta (M. and W.) Eumetria vera (H'all) Cliothyridina sublamellosa (Hall) Composita trinuclea (Hall) Sphenotus sp. Nucula randolphensis n.sp. Nucula platynotus n.sp. Nucula sp. Leda sp. Parallelodon sp. Conocardium chesterensis n.sp. Myalina sp. Myalina sp. Schizodus depressus Worthen Schizodus sp. Schizodus (?) sp. Aviculopecten (several species) Lepetopsis sp. Bellerophon chesterensis n.sp. Euphemus randolphensis n.sp. Bucanopsis ornatus n.sp. Ptychomphalus pp. Ptychornphalus sp. Mourlonia sp. Holopea sp. Orthoceras cf Cycloceras cf. Cycloceras sp. Nautilus sp. Discitoceras sp. Stroboceras sp. Phillipsia sp. Phillipsia sp. Cladodus sp. Pish teeth (several species) okawense M. and W. sequoyahensis Snider The really significant members of this fauna are found among the molluscs, essentially all of which are undescribed species. No attempt will be made in this place to describe and name all of the new forms, but descrip- 186 GEOLOGY OF HARDIN COUNTY tions of a few of the more abundant and characteristic species will be pre- J sented in the paleontological chapter of this report. A fauna similar to this one occurs in a number of localities in the lower part of the Okaw limestone of Eandolph County, Illinois, and all of the new species which are here de-: scribed as well as most of the other new forms, are recognized in both Ean- dolph and Johnson counties. In the Randolph County collections no examples of Pterotocrinus capitalis have been observed, but here in the Johnson County collection this best index fossil of the Golconda is found associated with species peculiarly characteristic of the lower portion of the Lower Okaw limestone. Correlation A comparison of the fauna of the Golconda formation with the Chester faunas that have been studied from the Mississippi valley counties in Illinois, establishes a very close relationship between the Golconda and the lower por- tion of the Okaw limestone. Every identified species in the Golconda fauna except Pterotocrinus capitalis and Pentremites obesus has been collected in the Okaw, and although these two forms are among the most diagnostic mem- bers of the fauna, their absence from the Okaw, in view of other close rela- tionships, may be considered as of geographic significance rather than of time value. Among the faunal characteristics that are common to the Golconda and the lower Okaw, are the following : ( 1 ) The first introduction and abund- ant occurrence of Camarophoria expianata. (2) The presence of Rhyncho* pora perryensis in both faunas, and the entire absence of the species from any other horizon in both of the sections. This is a representative of a very unusual type of brachiopod in the Chester faunas of the Illinois basin. (3) The common occurrence of a large group of unclescribed molluscs in both formations which are unknown in any other Chester fauna. (4) The great increase of the genus Archimedes in both of these faunas, A. sivallovanus. being very abundant in both, and A. lativoivis being restricted to them so far as known. (5) In both formations the bases of the crinoid genus Agassizo- crinus become abundant for the first time, although they had been present rarely in the next older Paint Creek fauna in both regions. (6) The com- mon presence of Pentremites platybasis in both faunas, and so far as known the restriction of the species to these faunas. (7) The abundance of the "wing plates" of Pterotocrinus in both faunas. While any one of the above mentioned common characteristics might not be sufficient to establish the correlation of the Golconda with the lower Okaw, the multiplication of evidence makes the case very strong, especially when taken in connection with the stratigraphic succession of the faunas in the two Chester sections. The only really adverse evidence in the case is the absence of the two very characteristic Golconda species, Pterotocrinus UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 187 capitalis and Pentremites obesm, from the Okaw, but these species may yet be found in that formation, or their absence from the faunas of the western side of the basin at that time may have been due to local environmental con- ditions, just as we recognize similar geographic differences in faunas at the present time. In the recent volume published by the Kentucky Geological Survey, devoted largely to the discussion of the Chester formations of the western portion of that state, by Chas. Butts and E. 0. Ulrich, Ulrich 1 has correlated the Cypress sandstone of southeastern Illinois and Kentucky, with the lower division of the Okaw of the Mississippi Eiver section, the Golconda being given no equivalent in the more western section. The reasons for such a cor- relation are nowhere made clear, but the correlation of the Golconda with the Lower Okaw is apparently rejected because of the absence of Pterotocrinus capitalis from the Okaw fauna. A hypothetical "secondary embayment * * * between St. Louis and Cairo 7 ' is assumed to have been in existence during that portion of Chester time in which the limestones of the Okaw formation were being deposited while the Cypress sandstone was being formed in southeastern Illinois. In the writer's opinion there is no basis whatsoever for assuming such a differentiation of the Chester basin. The correlation of the Golconda with the Lower Okaw, which is indicated by the similarity of the usual faunas of the two formations, is rendered doubly sure by the evidence of the fauna listed from Johnson County in which the "wing plates" of Pterotocrinus capitalis, the most characteristic Golconda species, are asso- ciated with a number of very peculiar and unusual Chester species which are known elsewhere only in the lower portion of the Okaw in Eandolph County, and with the establishment of such a correlation the correlation of the Cypress with the Lower Okaw is rendered untenable. Hardinsburg Sandstone name and general distribution The name of this formation has been brought into southeastern Illinois and the adjoining portion of Kentucky from Breckenridge County, Kentucky, east of the western coal fields of that state. In that portion of Kentucky there is an important sandstone formation called by Butts 2 the Hardinburg, which separates two distinct limestone formations that are the equivalents of the upper and lower members of the Okaw formation in Eandolph County, Illinois. In Hardin and the other southeastern counties of Illinois, a similar sandstone, lying above the Golconda formation, occupies the same stratigraphic position, and the same ncme is applied to it. The formation is well developed 1 Ky. Geol. Surv., Miss. Form. West. Ky., p. 50; pi. D, opp. p. 47; also pp. 112-113, (1918). 2 Ky. Geol. Surv., Miss. Form. West. Ky., p. 96 (1918). 188 GEOLOGY OP HARDIN COUNTY in Kentucky, both east and west of the coal fields, and extends westward into Illinois across Hardin, Pope, and Johnson counties. In Eandolph County the formation has not heretofore been differentiated, but there is locally present between the lower and upper divisions of the Okaw, a sandstone mem- ber which is doubtless a westward, much attenuated extension of the Hardins- burg sandstone. In its thinning to the west, the Hardinsburg sandstone resembles both the Cypress and the Bethel. In the southeastern portion of Hardin County the outcrop of Har- dinsburg sandstone occupies a belt extending in a northwesterly direction from the Ohio Eiver bluffs just below where Honey Creek enters the bottom land, parallel with the outcrop of the earlier Chester formations, to the southeastern border of the central faulted zone, the width of the outcrop being approximately one mile. At the southeastern extremity of this belt of outcrop, in the ravine running southeast from McDowell School, the formation is especially massive and forms prominent bluffs along the sides of the valley. In the central faulted zone (see figure 3) the Hardinsburg sandstone outcrops in a number of fault blocks in association with the underlying Gol- conda formation. All of these outcrops lie in belts one and two. In one of the elongate, narrow blocks of belt two, the formation is exposed in a north- easterly direction from near the center of sec. 12, T. 12 S., R. 8E. A small area is underlain by the formation in the NW. 1/4 SW. % of the same sec- tion, and continues westward into the adjoining section 11. One of the largest areas of outcrop of the sandstone is in the fault block lying mostly in' sees. 15 and 16, T. 12 S., E. 8- E., where it caps the hills whose lower slopes are underlain by the Golconda formation. The sandstone exposed in the public highway about three-fourths of a mile north of the Eosiclare mine at Eosiclare, is Hardinsburg, and the same formation caps the elongate hill to the southwest, whose summit is nearly opposite the Blue Diggings shaft. In the southwestern part of belt number one of the faulted zone, the Hardinsburg sandstone is exposed in only one small area, east of Wallace Branch, in sec. 25, T. 12 S., E. 7 E., where it caps an enlongate hill whose lower slope is occupied by the Golconda formation. In the northern portion of this belt the formation outcrops in the upper part of the hill north of Harris Creek, across sec. 24, T. 11 S., E, 8 E., and after crossing this creek the outcrop continues in a northeasterly direction for two and one-half miles to the southeastern border of the belt. In the northwestern third of the county, the Hardinsburg sandstone out- crops continuously on the upper face and brow of the bluff on the north side of Pinhook Creek, and continues to the west into Pope County where it forms the summit of the west bluff of Grand Pierre Creek. It is the first UPPER MISSISSIPPI^ SERIES, CHESTER GROUP 189 sandstone met with on ascending the bluff north and west of these streams, and the east-west road through Karbers Eidge is on the outcrop of this sandstone from one-third of a mile northwest of Philadelphia School, west- ward to the road intersection in the SE. 14 sec. 11, T. 11 S., E. 7 E. To the east the outcrop is offset to the southwest by the Lee fault. Between this fault and the Wolrab Mill fault the formation occupies a narrow belt in its normal position in the Chester group, and is followed by the ridge road to the center of sec. 14, T. 11 S., K. 8 E. LITHOLOGIC CHARACTERS The Hardinsburg is a massive, irregularly cross-bedded sandstone, exhibit- ing much evidence of its shallow water origin in the ripple-marks and other surface markings of the layers. It is moderately fine grained, and in color is yellowish-brown, the weatherbeaten surfaces being gray or reddish-brown. The lower portion of the formation is more massive and in many places rises in nearly vertical bluffs above the underlying, % talus-covered slopes of the Golconda formation. The higher portion of the formation is not commonly so well exposed as the lower, and is apparently less massive, more thinly bedded, and locally at least, there are minor beds of shale. Such a bed of black shale included in the Hardinsburg, is exposed along the road between Sparks Hill and Karbers Eidge, in the ravine crossing the middle of the east half of sec. 10, T. 11 S., E. 8 E. The bed is only a few feet in thick- ness, and should possibly be assigned to the next overlying formation, the Glen Dean limestone. In general the formation is less massive than the Cypress sandstone, but locally there is little difference in this respect between the two formations. On Pinhook Creek, west of Karbers Eidge, the forma- tion includes a considerable thickness of evenly bedded, smooth surfaced flagstones that have been quarried on a small scale in the NW. 1/4 sec. 7, T. 11 S., E. 8 E. THICKNESS The Hardin County outcrops of the Hardinsburg sandstone are not gen- erally so well adapted for determining the thickness of the formation as are some of those in Pope County. At a number of localities in that county where essentially the entire thickness of the formation is exposed, there are approximately 100 feet of the sandstone. At some localities it seems to exceed- this thickness somewhat, and elsewhere it is considerably less, some sections exhibiting not over 30 feet as on the ridge between Goose and Harris Creeks in Eock Creek Township. In one deep well section in northern Pope County, 80 feet of Hardinsburg seems to be present. The thickness of the formation in Hardin County is doubtless essentially the -same as in Pope, and an average thickness of from 75 to 100 feet may be assumed. 190 GEOLOGY OF HARDIN COUNTY STRATIGRAPHIC RELATIONS The stratigraphic relations of the Hardinsburg sandstone with the under- lying Golconda formation have already been considered, and the evidence for the existence of a plane of unconformity between the two formations, as exhibited in Pope County, has been presented. The stratigraphic relations between the Hardinsburg and the overlying Glen Dean limestone are nowhere so clearly shown as is the basal contact of the sandstone. Wherever the sec- tion has been examined the interval between the sandstone and the overlying limestone has been covered, and no actual contact has been observed. It is a fact, however, that the upper portion of the Hardinsburg become shaly, and there is apparently a somewhat gradual transition in character from the lower to the higher formation which would mean continuity of sedimentation and absence of any relation of unconformity. On the other hand, the vary- ing thickness of the Hardinsburg sandstone might be interpreted as evidence of a period erosion subsequent to its deposition and before the deposition of the Glen Dean limestone. PALEONTOLOGY No fossil invertebrates have been collected in the Hardinsburg sandstone, and only indeterminable plant fragments have been observed in Hardin County in any sandstone or shale that is certainly Hardinsburg in age. In Pope County a shale bed near the summit of the formation has afforded plant Temains which are somewhat better preserved, but they have not been critically studied by a paleobotanist ; similar material may be looked for in Hardin County. CORRELATION In the western counties of Illinois the Okaw limestone is a thick for- mation and is quite clearly not a simple stratigraphic unit. It has been shown that the lower Okaw is the time equivalent of the Golconda formation, and it will be shown later that the upper Okaw is to be correlated with the Glen Dean limestone; but in Randolph and the adjoining counties there is no thick sandstone formation comparable with the Hardinsburg, separating the two parts of the Okaw. In that region the exposures are all more or less drift covered in most places, but it is clear that locally there is a thin sand- stone member at the horizon corresponding to the dividing line between the lower and upper Okaw. Elsewhere it is certain that no such sandstone bed separating the two Okaw members is present. However, all the evidence points to the fact that this thin discontinuous sandstone horizon in Randolph County is the much reduced equivalent of the Hardinsburg sandstone of the southeastern part of the state. We have then, in the Hardinsburg, another example of a thick sandstone formation in the southeastern part of the state, like the Cypress and the Bethel, thinning out markedly to the west. upper mississippi ax series, chester group 191 Glex Deax Limestone xame and distribution As with the Hardinsburg sandstone, the name of the Glen Dean lime- stone is brought into southeastern Illinois from Breckenridge County, Ken- tucky, where the formation has been named by Butts. 1 In a "Preliminary Oil Keport on Southern Illinois" by A. D. Brokaw 2 , and later in the final report by the same author 3 and in a report by Stuart St. Clair 4 , this limestone was designated as the Sloans Valley formation, from a locality in Pulaski County, Kentucky, a name which had been previously used as a field desig- nation, but which had not been published. Butts used the name Glen Dean for the formation in Breckenridge County, and it was so designated by Miller in a "Table of Geological Formations for Kentucky," published in March, 191 7. Of the two names Glen Dean is preferable. The Sloans Valley section is less well known than that of southeastern Illinois and the adjacent portions of Kentucky, and it is not unlikely that the section includes much more than the exact equivalent of the strata for which the name has been used in Illinois. The Glen Dean locality, on the other hand, has been definitely selected by Butts as representing the formation under discussion, limited below by the Hardinsburg and above by the Tar Springs sandstone. The general distribution of the Glen Dean limestone is coextensive with the Hardinsburg sandstone, both east and west of the western Kentucky coal fields, and in southern Illinois, where it extends westward through Hardin, Pope, Johnson, and Union counties. The formation has its equivalent in Randolph County in the upper division of the Okaw limestone. In the southeastern portion of Hardin County the Glen Dean limestone is well exposed in the Ohio River bluff at and just west of the point where Honey Creek enters the river bottoms, a little less than one-half mile south- west of Fraileys Store. From this point the formation extends northwestward in a narrow belt, to the southeastern border of the central faulted zone of the county. The upper limestone of the formation is well exposed at a num- ber of points in Rock Creek and Battery Rock townships. One such exposure is on the slope on the southeast side of the ravine crossing the southeastern part of sec. 32, T. 11 S., R. 10 E., in a northeast-southwest direction ; another such exposure is on the south side of the ridge ending eastwardly in the SW. 14 sec. 33, T. 11 S., R. 10 E., where 35 feet of limestone is exposed just beneath the Tar Springs sandstone which caps the ridge and the low knob to the east ; still another outcrop of this limestone is exposed on the southeast angle of the middle of the three spurs in the northeast corner of sec. 4, T. 12 S., R. 10 E., where forty feet of crinoidal limestone is exposed, mostly gray 1 Ky. Geol. Surv.. Miss. Form. West. Ky.. n. 97 ("1918). 2 111. State Geol. Surv., Extr. from Bull. 35. pi. Ill (1916). 3 111. State Geol. Surv., Bull. 35, pi. I (1917). 4 111. State Geol. Surv., Bull. 35. p. 46, pi. IV (1917). 192 GEOLOGY OF HARDIN COUNTY in color but somewhat reddish at the top. A very good exposure of the Glen Dean limestone is exhibited in the southeastern face of the Ohio River bluff, above the road, about one-half mile southwest of Fraileys Store. The outcrop of the Glen Dean limestone in the ridge between Sheridans Branch and Little Creek on the southwest, and Honey Creek and the southeast branch of Bock Creek on the northeast, is conspicuously marked by low ground and saddles at several places. One of the best examples of this topographic rela- tion is in the north part of sec. 26, T. 11 S., E. 9 E., where the three promi- nent knobs are capped by Tar Springs sandstone, the south slopes of the knobs and the low saddles just to the south are occupied by the Glen Dean, while the long slope rising somewhat gently, still to the south, is occupied by the Hardinsburg sandstone dipping beneath the Glen Dean. The same relation is also well shown in the west side of the NE. y± sec. 4, T. 12 S., B. 10 E., where the Tar Springs also caps the knobs and the Glen Dean outcrops on the west slopes and in the saddles. Within the central faulted zone the Glen Dean does not outcrop in belt number three (see figure 3), and is exposed in only a few of the fault blocks of belt number two. The southernmost of these areas is just south of Stone School. The formation is poorly exposed at this place, the only outcrops being shale and of very small extent. Much better exposures of the formation may be seen a little less than one mile north of the iron bridge over Big Creek on the Elizabethtown and Eosiclare road, the exposures being upon the west side of a valley tributary to Big Creek, near its confluence with the main creek. This outcrop is in a comparatively small fault block and consequently does not have any great lateral extent. Farther to the northeast the Glen Dean limestone outcrops in the elongate fault block lying northwest of Peters Creek, as a narrow belt starting in the KE. l/ 4 sec. 12, T. 12 S., E. 8 E,, and continuing four miles to the center of sec. 33, T. 11 S., B. 9 E. In belt number one of the faulted zone, the Glen Dean limestone is ex- posed only towards the northern part of the county, southeast of Sparks Hill. The upper, massive limestone is well exposed on the south bluff of the ridge north of Goose Creek, near its eastern end in sec. 16, T. 11 S., E. 9 E. West from this locality the exposures of the formation are poor, being largely covered. The position of the limestone is in a narrow belt just to the north- west of the outcrop of the Hardinsburg sandstone, which "has been described. In the northwestern part of the county the Glen Dean exposures continue ' to be poor. Between the Wolrab Mill and Lee faults the formation extends in a narrow belt with a general east-west direction nearly ' through the centers of sees. 14 and 15, T. 11 S., E. 8 E. The formation occupies a depression between the, two series of knobs, those to the south being capped by the Hardinsburg and those to the north by the Tar Springs sandstone. West of the Lee fault the Glen Dean is offset about one and three-fourths miles to the UPPER MISSISSIPPIAN SERIES, CPIESTER GROUP 193 northeast, that is, to a point about one-third of a mile north of Philadelphia School, from which point the formation continues westward in a narrow belt to the Hardin-Pope County line. Along this belt the limestone is exposed in the bottom of the ravine about three-fourths of a mile north-northeast of Karbers Eidge, in the center of the NE. % sec. 9, T. 11 S., E. 8 E. In the bead of the ravine in the NE. y± sec - 1^, T. 11 S., E. 7 E., there is an exposure of the limestone which is crowded with one of the most distinctive fossils of the formation, the bryozoan Prismopora serrulata, and another characteristic fossil occurring at the same locality is Pentremites spicatus. Massive lime- stone beds of the Glen Dean are exposed in the bed of Eose Creek in the NE. y± SW. % sec. 11, T. 11 S., B. 7 E., and also near the head of a shallow ravine in the SE. 14 SW. l/^ of the same section. Lithologic Characters In Hardin and the adjoining counties in Illinois, the Glen Dean forma- tion is rather obscure because its lithologic characters are such as to render it liable to be hidden by surficial covering. In favorably situated localities, however, many outcrops of the formation do occur, and in some other locali- ties its presence is indicated by a series of sink-holes in which there may be no actual outcrops of the limestone. Like the Golconda formation, the Glen Dean consists of interbedded limestones and shales. The limestone beds, as observed in different localities, exhibit a great variety of lithologic char- acters, the different beds being less uniform, even, than those of the Golconda. In general the limestone of the Glen Dean formation is crystalline and gray or nearly black in color; but it includes beds of very dense, compact and brittle rock, breaking with a splintery fracture, these dense beds commonly being quite dark, in fact nearly black in color. The crystalline character of the limestones is due to the presence of an abundance of fossil fragments, especially crinoidal plates, and on extreme weathering the beds become crumbly and shaly. In a few localities certain calcareous beds of the formation occur as very hard, brittle shales or shaly limestone. Some of the beds apparently are very pure, while others are highly siliceous. These limestones contain some chert, although the chert is much less conspicuous than in the higher limestones of the Chester group. The chert commonly occurs in plate-like layers one or two inches thick, most or all of which are colored dark chocolate- brown on freshly broken surfaces. On the weathering of the limestone these cherts break into subcubical fragments of the thickness of the plates, and remain in the residuum. Two surfaces of the subcubical fragments, those in contact with the inclosing limestones before weathering, are commonly decom- posed to a depth of about one-half inch, becoming porous and developing a —13 G 194 GEOLOGY OF JIA1JDIN COUNTY yellowish-brown color. In some localities the presence of these chert frag- ments in the residuum is the only evidence of the existence of the limestone. The shales of the Glen Dean also lack uniformity in character. Some beds are argillaceous and break down into a plastic clay, other beds are cal- careous, and still others are siliceous. Between some of the denser limestone layers there are thin shale partings. While the composition of the Glean Dean formation in the western part of the county is not well known, owing to the poor exposures, the formation is better known to the east where exposures are rather better. In Bock Creek and Battery Bock townships a number of fairly good exposures reveal that the upper forty feet of the formation is a coarsely crystalline crinoidal limestone, and that the lower thirty feet or so includes both shales and limestones. A nearly complete section is exposed in the highway in the northwest corner of sec. 15, T. 11 S., B. 9 E., where the following beds may be seen: Section of the Glen Dean limestone in the northwest corner of sec. 15, T. 11 S„ R. 9 E. Fe< 4. Sandstone, basal portion of the Tar Springs 3. Limestone, coarsely crinoidal, weathering shaly and crumbly 40 2. Limestone and shale interbedded 20 1. Limestone, with Lyropora, Prismopora, and Pentremites pyramiclatus 10 The upper part of this section is representative for the region, as show] by a number of exposures at points described under the head of distribution. In some places in this northeastern part of the county a few feet of shale is present between the Glen Dean limestone and the Tar Springs sandstone, as for example in the road on the south face of the low bluff in the center ol sec. 33, T. 11 S., B. 10 E. Another feature of the Glen Dean limestone that is worthy of mentioi is a somewhat widely distributed oolitic limestone bed in the upper part oi the formation. This bed has been observed most commonly in the southen part of Bock Creek Township and is well exposed on the south side of th( knob at the head of Peters Creek, just west of the center of sec. 33, T. 11 S. E, 9 E. ; also at the Martin prospect in the SW. % sec. 23, T. 11 S., B. 9 E, and on the south-facing steep slope in the middle of the east side of sec. 21 T. 11 S., B. 9 E., east of Douglas School. THICKNESS The outcrops of the Glen Dean limestone in Hardin County do not afforc so good an opportunity for determining the thickness as do some of those h Pope County. The accurate determination of the thickness of the formatioi is difficult throughout the area because of the nonresistant shale beds in th( formation and the massive sandstone which overlies it, a set of condition? which make it liable to be more or less completely obscured by the talus froi UPPER MISSISSIPP1AN SERIES, CHESTER GROUP 195 the sandstone. The best estimates of thickness that can be made range from 50 to 70 feet, and about 60 feet may be considered a fair estimate of its thickness in Hardin County. In general it may be said that the Glen Dean is one of the thinner of the Chester limestone formations in the south- eastern part of Illinois. STRATIGRAPHIO RELATIONS The outcrops of the Glen Dean limestone are so commonly obscured by talus that the actual contacts with the subjacent and superjacent formations have nowhere been observed. It is altogether probable, however, that the formation rests conformably upon the underlying Hardinsburg sandstone. As already pointed out there seems to be a gradation from the Hardinsburg sandstone through a shale interval into the typical limestone and shale of the Glen Dean formation. There is perhaps less basis for certainly determining the relations between the Glen Dean and the overlying Tar Springs sandstone, but so far as can be determined the massive sandstones of the higher formation seem to succeed the limestones abruptly, and it is not unlikely that the younger formation rests unconformably upon the older, this condition being essentially the same as between the Hardinsburg and the Golconda. PALEONTOLOGY The unsatisfactory exposure of the Glen Dean limestone in Hardin County accounts for the somewhat meager fossil collections which it has afforded. A number of small collections have been secured, however, from. Hardin and the adjoining Pope County area which together give a good idea of the fauna. The best collection from Hardin County has been secured from the locality already mentioned, situated a little less than one mile north of the iron bridge over Big Creek. From this locality the following species have been identified: Fossils from the Glen Dean limestone collected in the WW. y± WE. % sec. 21, T. 12 8., R. 8 E. Triplophyllum spinulosum (M.-E. Prismopora serrulata Ulrich and H.) Productus inflatus McChesney ? Pentremites brevis Ulrich Diaphragmus elegans (N. and P.) Pentremites okawensis n.sp. Camarophoria explanata (McChes- Pentremites canalis Ulrich ? ney) Zeacrinus sp. Spiriferina spinosa (N. and P.) Eupachycrinus sp. Spiriferina transversa (McChesney) Pterotocrinus sp. (wing plates) Spirifer increbescens Hall, var. Fistulipora sp. Spirifer leidyi N. and P. Fenestella (several species) Reticularia setigera (Hall) Archimedes swallovanus Hall Eumetria vera (Hall) Archimedes distans Ulrich Cliothyridina sublamellosa (Hall) Septopora subquadrans Ulrich Composita trinuclea (Hall) 19G GEOLOGY OF HAnDIN COUNTY In Pope County a number of collections have been made from exposure northwest of Golconda, within a radius of three miles from that town. The combined list of four such collections is as follows : Fossils from the Glen Dean limestone collected northwest of Golcondu Triplophyllum spinulosum (M.-E. and H.) Pentremites spicatus Ulrich Pentremites sp Pterotccrinus sp. (wing-plates) Fistulipora sp. Meekopora exirnia Ulrich Eridopora punctifera Ulrich Batostomella sp. Stenopora tuberculata (Prout) Fenestella serratula Ulrich Fenestella sp. Archimedes communis Ulrich Archimedes distans Ulrich Archimedes lerebriformis Ulrich Polypcra cestriensis Ulrich Polypora sp. Septopora subquadrans Ulrich Reteporina flexuosa (Ulrich) Streblotrypa sp. Rhombopora sp. Prismopora serrulata Ulrich Orbiculoidea sp. Orthotetes kaskaskiensis (McChes- Productus ovatus Hall Productus sp. cf. inflatus McChesney Diaphragmus elegans (N. and P.) Pustula punctata (Martin)? Pustula sp. Camarophoria explanata (McChes- ney) Rhynchopora sp. Dielasma shumardanum (Miller) Girtyella brevilobata (Swallow) Spiriferina spincsa (N. and P.) Spiriferina transversa (McChesney) Spirifer increbescens Hall var. Spirifer leidyi N. and P. Martinia contracta M. and W. Reticularis setigera (Hall) Eumetria vera (Hall) Cliothyridina sublamellosa (Hall) Composita trinuclea (Hall) Conularia sp. Lioptera sp. Platyceras sp. Phillipsia sp. ney) For the most part the species in the Glen Dean fauna, as exhibited in Hardin County and elsewhere, are more or less widely ranging forms; many of them would not be out of place in any Chester fauna. Among the species, however, there are a few which are limited in their geologic range, and which may be considered as being characteristic of this horizon. Among such species the most significant one in the Hardin County faunas is the bryozoan Prismopora serrulata, which may be found in practically every Glen Dean fauna, and in some localities it is a very common form. The species is not wholly limited, however, to the Glen Dean formation, for it has been collected both in the earlier Golconda fauna, and in the later Vienna limestone, and even in the Menard. In the Golconda limestone it is very rare and has been observed in but a single locality, and was there represented by a single speci- men. It has likewise been observed in the Vienna limestone in a few localities, in Pope and Johnson counties, where it is represented by a much larger number of examples than were found in the Golconda. As contrasted with its occurrence in these older and younger faunas, the species is conspicuous UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 197 in the Glen Dean. In places in this formation the surfaces of certain lime- stone layers are literally Prismopora gardens, and careful search brings it to light in practically every outcrop of any size, and it is the first fossil to be sought for when the recognition of the Glen Dean formation is involved. Another fossil species, which so far as known has not been collected outside of the Glen Dean formation, is the large Pentremites spicatus. This species is really much more local in its occurrence than the Prismopora, and has been observed in but few of the Hardin County collections, although it has been found more commonly in Pope County and at Carrsville, Kentucky. Another bryozoan which has been found to be quite characteristic of the Glen Dean horizon in Kentucky and elsewhere, but which has not yet been observed in either Hardin or Pope counties, is Archimedes laxus, and the peculiar "wing- plates" of Pterotocrinus bifurcatus is another form in the same category. Any one or all of these species may be sought for in the Glen Dean of Hardin County. One of the Glen Dean faunas which has been collected from a limestone bed in the upper part of the formation in Pope County is quite different from those in which the Prismopora is commonly met with. In it the bryozoa are practically wanting, and it is especially characterized by numerous examples of Martinm contracta, a species which has a wide range in the Chester faunas, but which is especially abundant in a nearly basal zone of the Okaw limestone in Randolph County. This Martinia fauna has not yet been noticed in Hardin County, but it would not be surprising to find it there. CORRELATION A comparison of the fauna of the Glen Dean formation with the faunas of the Chester group in Randolph County, shows a close similarity between this fauna and that of the upper division of the Okaw limestone in the more western section. In both faunas the bryozoan Prismopora serrulata is a characteristic form ; in the more western region the species has not been met with, at any other horizon than the upper Okaw, and in the southeastern part of the state it has been met with only rarely in any other horizon than the Glen Dean. In both regions the large Pentremites spicatus is known only at this horizon. A number of other characteristic Glen Dean species, such as Archimedes laxus, and Pterotocrinus hifurcatus, which have not been' collected in Hardin County but which are known in Kentucky, are also present in the jpper Okaw of Randolph County. ■ With so many features in common between :he Glen Dean and the upper Okaw faunas the correlation of the two forma- :ions may be considered as established, especially as they occupy corresponding 30sitions in the stratigraphic series of the two regions. The entire assemblage of fossil species so characteristic of the Upper few-Glen Dean horizon, occurs also in the Chester beds at Sloans Valley, 198 GEOLOGY OF HARDIN COUNTY Pulaski County, Kentucky, in the southeastern portion of the state. A num- ber of the characteristic species were first described from the Sloans Valley locality, and there can be no question of the presence of equivalent strata in the Chester section at that locality; but data are not available to determine whether the beds are limited by sandstone formations above and below, as in southeastern Illinois. Tar Springs Sandstone name and distribution In the section of the Chester group in Breckenridge County, Kentucky the limestone and shale beds that are characterized by Prismopom semdaia and Archimedes laxus, are succeeded by a massive sandstone formation. One of the most characteristic features of this sandstone in Breckenridge County, Kentucky, is the presence in it of springs that are highly charged with bitumen. Such springs are locally known as "tar springs/' and the name Tar Springs sandstone was used for the formation as long ago as 1857 by Owen. 1 Owen's name is now being revived and is applied to the massive sandstone formation lying next above the Glen Dean limestone, not only in Breckenridge County, Kentucky, but also in Kentucky west of the coal fields, and in southern Illinois. The formation is practically coextensive with the Glen Dean formation across Hardin, Pope, Johnson, and into Union counties, Illinois, and in the adjoining counties in Kentucky. The distribution of the Tar Springs sandstone in Hardin County is similar to that of the other Chester formations. The sandstone outcrops as one of the outer belts en- circling that portion of the larger dome structure that is included in the county, extending from the Ohio Eiver bluffs near the southeastern corner of the county to the Hardin-Pope County line, interrupted at intervals by xhe longer northeast-southwest faults, the greatest interruption being made by the deeply downdropped block of Potts ville that extends southwestwardly to Keelin School. In addition to this more regular line of outcrop there are a number of smaller areas occupied by the formation in the complexly faulted region northwest of Elizabethtown. In the southeastern portion of the county the Tar Springs sandstone is limited in its distribution to a belt extending from the Ohio Eiver bluffs west of Honey Creek, northwestwardly to the fault bounding the central faulted zone on the southeast. Throughout this belt the formation occupies the north- easterly dipping slopes of the hills lying southwest of Honey Creek and one of the tributaries of Eock Creek. In the complexly faulted portion of belt number two of the central faulted zone, the Tar Springs sandstone occupies parts of a number of the fault blocks. The largest of these areas is in the block lying southwest of Geol. Surv. Ky., vol. 2, pp. 86-87 (1857) UPPER MISSISSIPPI AN SERIES, CHESTER GROUP 199 Keelin School, the outcrops being well shown along the Illinois Furnace road for three-fourths of a mile northwest from the junction with the road leading past Keelin School. These outcrops continue westward from this road for a mile or more to the wedge-shaped western extremity of the block. Adjacent to the valley of Big Creek the Tar Springs is mapped as constituting a part or the whole of a number of small blocks. In those blocks in which no lime- stone is present to assist in the identification of the formation, the Tar Springs is recognized by its more massive character as compared with the other formations. In belt number one of the faulted zone, the Tar Springs occupies its normal position as a belt extending from the Wolrab Mill fault one and one-half miles southwest of Sparks Hill in a northwesterly direction to Harris Creek. In the northwestern portion of the county the belt of the Tar Springs sandstone occupies its normal position from Rose Creek at the Hardin-Pope County boundary, to the Lee fault a little south of Loves Store. Eastward from this fault the formation is offset to the southwest for a mile and one-half, and then continues eastward to the Wolrab Mill fault. LITHOLOGIC CHARACTER The lithologic characters of the major sandstone formations of the Chester group in southeastern Illinois are so much alike that it is difficult or impos- sible to differentiate them in isolated outcrops where the stratigraphic relations with one of the limestone formations of the section are not shown. Like the Hardinsburg and Cypress sandstones, the Tar Springs formation is a yellowish- brown sandstone, becoming distinctly reddish in places. It contains also some very white layers which, however, are likely to be stained brown on the outside by the surficial oxidation of the small amount of iron present. On the whole the color is more like that of the Hardinsburg, being distinctly less pale than much of the Cypress. In texture the sandstone is moderately fine grained, not essentially different from the other two formations that have been mentioned. The basal portion of the Tar Springs sandstone consists of massive beds which locality are remarkably cross-bedded, the individual cross-bedded units being notably saucer-shaped, with the inclination of the beds sloping in all directions. Cross-bedding is conspicuous throughout the formation, and ripple-marked surfaces and other surfaces with markings indicating shallow water conditions are present throughout the whole of it. The upper portion of the formation is notably more thinly and more evenly bedded than the lower part, although it is in places as thickly bedded as the lower part. In the midst of the Tar Springs sandstone, there is locally at least, and perhaps generally throughout Hardin County, a shaly member which may attain a thickness as great as forty feet, but which is commonly less than that. The bed is composed of dark to black, more or less carbonaceous shale, 200 GEOLOGY OF HARDIN COUNTY and thinly laminated to shaly, fine-grained, gray sandstone, coated black on the surfaces of the laminae by carbonaceous matter. Poorly preserved fossil plants, mainly undeterminable stems, occur here and there in this shale, but in a shale outcrop in the road side at Stone Church, two miles west of Eliza- bethtown, which is probably in the Tar Springs, numerous, well preserved fossil plants have been collected. In places a thin bed of impure coal is associated with this shale member of the Tar Springs sandstone, one of its best exhibitions being in the shallow ravine in the NE. % SE. 1,4 sec. 10, T. 12 S., E. 8 E., about three miles north of Elizabethtown. Another exposure of a similar coal bed, having the same stratigraphic position, is in the road midway between Pinhook Creek and Rose Creek, in the NW. % sec. 14, T. 11 S., E. 7 E. THICKNESS In none of the exposures in Hardin County is the Tar Springs sandstone sufficiently well exposed to furnish a basis for a satisfactory determination of its thickness. From the exposures in Pope County, however, the thickness can be estimated with some degree of accuracy, and it seems to exhibit varia- tions between 100 and 150 feet. A similar thickness in Hardin County may be assumed. From this it will be seen that the Tar Springs formation js the thickest of the several Chester sandstone formations in the area. STRATIGRAPHIC RELATIONS The stratigraphic relations of the Tar Springs sandstone with the under- lying Glen Dean limestone can only be inferred, since no actual exposure of the contact between the two formations has been observed. However there seems to be considerable variation in the thickness of the Glen Dean forma- tion as that formation has been studied in several southern Illinois counties and such a variation suggests an erosion period preceding the deposition of the Tar Springs sandstone, in which case the younger formation rests uncon- formable upon the older. The existence of such a relation between these two formations would be in conformity with the relations between the other Chester sandstones and their subjacent limestone formations, since in every case where the actual contact between such beds has been observed, evidences of unconformity are exhibited. Our knowledge from actual observation, of the upper contact of the Tar Springs sandstone is just as unsatisfactory as that of the lower one. Through- out the county and also in the immediately adjoining region the interval between this formation and the overlying Menard limestone is uniformly covered. The absence however, ■ of the Vienna limestone and Waltersburg sandstone throughout Hardin County, as contrasted with their presence in counties to the west, may indicate an unconformity in Hardin County above the Tar Springs sandstone. UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 201 PALEONTOLOGY The only fossils that have been observed in the Tar Springs formation in Hardin County, are plant remains. Butts has found a single determinable fragment of a fossil fern in the median shale member of the formation, about one-half mile east of Eose Creek School, in a ravine south of the road in the SW. % sec. 11, T. 11 S., E. 7 E. This specimen has been identified by David White as a species of Cardiopteris, probably C. pohjmorpha. The plant-bearing shale exposure in the roadside at Stone Church, already mentioned, is in a small fault block in which no limestone formation is exposed, so that the age of the sandstone formation in which the shale is included is not determinable with certainty. However, of all the sandstone formations known to be present in the county, such a shale member conforms best with the Tar Springs. The plants from this locality are not numerous specifically, but the specimens are abundant and well preserved, although fragmentary. The forms that have been identified by David White are as follows: Fossils collected from the middle shale member of the Tar Springs formation Cardiopteris polymorpha Gopp Lepidodendron sp. undet. Sphenopteris sp. undet. It is significant that the species of Cardiopteris which is recognized in this collection, is the same as the one which has been identified from the undoubted Tar Springs east of Eose Creek School. The excellent condition of preservation of these fossil plants, which were of course living upon the land at that time, shows that they could have been transported no great distance, and consequently that the land surface where they grew must have been near at hand. CORRELATION In Breckenridge County, Kentucky, the same species of Cardiopteris that has been identified in Hardin County, is known from the typical Tar Springs, and associated with it is a Lepidodendron comparable with the one recorded from Stone Church. Such evidence helps to corroborate the correlation of this sandstone formation in the two regions. Cardiopteris polymorpha is also known from beds regarded as middle Chester at Abbs Valley, Virginia, and it is found in the Mississippian of Europe. In the Eandolph County, Illinois, section of the Chester group, the upper Okaw limestone, which is the equivalent of the Glen Dean formation of Hardin County, is succeeded by the Menard limestone. In Hardin County these two limestones are separated by the thick Tar Springs sandstone. In the Eandolph County section, therefore, the position of the Tar Springs sandstone is in the unconformity beneath the Menard limestone, the sandstone itself being entirely lacking in the section. A comparison of the western extension of the Tar Springs with the earlier Hardinsburg, Cypress, and 202 GEOLOGY OF HARDIN COUNTY Bethel sandstones, shows that all are similar in regard to their thinning to the west, but that while all three of the earlier formations have at least some representation in Randolph County, the Tar Springs sandstone has thinned out entirely and disappeared. Vienna Limestone and Waltersburg Sandstone 1 In Hardin County the Tar Springs sandstone is succeeded by the Menard limestone, but as this interval is traced westward two important formations, a limestone and a sandstone are found to be intercalated between them. These formations have been designated the Vienna limestone from excellent ex- posures at Vienna, in Johnson County, and the Waltersburg sandstone from the exposures of the beds at Waltersburg, in Pope County. Although neither of these formations has been definitely recognized in Hardin County, both are known to be present in the eastern part of Pope County, in that portion of the county included in the accompanying map (Plate I), and good ex- posures of them may be seen in the lower part of the bluff having a general north-south direction near the middle line of sees. 8 and 17, T. 12 S., R. 7 E., below Grandview School. The limestone is commonly a highly siliceous or cherty bed, and the sandstone resembles the other sandstones of the Chester series. In the ridge whose extremity lies in the NE. % °^ sec - 8 just men- tioned, and which extends northward into the adjoining sec. 5, the Walters- burg sandstone is well exposed in a massive outcrop. In the Pope County portion of the geological map accompanying this report, even where they are known to be present, these two formations have not been mapped separately from the Menard limestone, all three formations being included in the same color. When these units were first observed they were considered as subordi- nate members of the Menard formation, but as the field studies have been carried to the west they have proven to be equal in importance and magnitude with other Chester units, and have come to be recognized as full formations. It is altogether probable that the equivalent of the Vienna limestone does continue into Hardin County, and that the Waltersburg sandstone thins out and disappears to the east, bringing the Vienna and Menard limestones to- gether; and it is not unlikely that a restudy of the region might make possible the separation of the Menard, as mapped, into its probable Vienna and Menard portions. Menard Limestone name and distribution" The Menard limestone has been described from Randolph County, Illi- nois, and so far as it is known in that portion of the state its surface outcrops 1 Written in October, 1919. subsequent to the mapping- of Hardin County and the writing of the major portion of the report. CHESTER GROUP 203 are confined to that county and Jackson County. The typical exposures of the formation are in the Mississippi River bluffs between Menard and Chester. The formation doubtless extends northward from Randolph County, but in that direction it is completely covered by the overlapping Pennsylvanian formations. In the southern counties of Illinois the Menard limestone has been recog- nized in its proper stratigraphic position across Union, Johnson, Pope, and Hardin counties, and it is also known southward across Ohio River in Critten- den and some of the adjacent counties in Kentucky. In Hardin County the Menard limestone has an areal distribution similar to that of the older Chester formations. In the southeastern portion of the county the formation occupies a belt from the Ohio River bluffs above Honey Creek, extending northwestwardly along the northeastern wall of Honey Creek valley and of the valley of a tributary of Rock Creek, to the fault limiting the central faulted zone of the county. In belt number two of the central faulted part of the county (see figure 3), the Menard limestone is exposed in a number of fault blocks. Good ex- posures of the formation are present along the east side of the road past Keelin School, a short distance north of the junction of this road with the Illi- nois Furnace road. This outcrop continues eastward for about half a mile and occupies the southern border of a trapeziform fault block. A little north of the last mentioned outcrop, on the west side of the road, another belt of Menard limestone sets in and continues in a westerly direction across a larger trapeziform fault block. Through most of its course this last belt is heavily covered, so that actual outcrops are not commonly met with. The black shale exposed in the Illinois Furnace road on the line between the sees. 10 and 11, T. 12 S., R. 8 E., is the basal black shale member of the Menard. The best exposure of the formation in this complexly faulted region occurs along the road running north from Stone Church to Illinois Furnace. The extent of this outcrop is about one and one-fourth miles in a north-south direction, along the eastern edge of the larger fault block that extends westward to Wallace Branch. Smaller outcrops of the formation along the eastern margin of this same fault block are present in the valleys marking the fault line in a southwesterly direction from Stone Church. Another Menard limestone outcrop occurs along the road to Illinois Furnace at the northeastern end of Stone Hill, in the NE. % sec. 17, T. 12 S., R. 8 E. A number of smaller patches of the Menard are present in some of the smaller of the fault blocks of the region. In the small Chester area which is mostly included in sec. 21, T. 11 S., R. 9. B., in fault contact with the Pottsville to the northwest and in stratigraphic contact with the same formation on the southeast, the Menard occupies a belt about one and one-half miles in length. 204 GEOLOGY OF HARDIN COUNTY In belt number one of the central faulted zone, the Menard limestone occupies its normal position in the Chester series, from the Wolrab Mill fault one mile south of Sparks Hill, in a northeasterly direction to the Hogthief Creek fault. Other small areas of the Menard are present in a small fault block within this belt, which lies mostly in sees. 19 and 20, T. 11 S., E. 9 E. The Menard limestone in the northwestern portion of the county occupies a narrow belt from one-half to one and one-half miles south of the county line, along the valley of Bose Creek for most of the distance, from the Pope- Hardin County line to the Lee fault, just south of Loves Store. Beyond this to the east, the formation is offset about one mile to the southwest and then continues eastward to the Wolrab Mill fault, along a valley tributary to Big Creek. LITHOLOGICAL CHARACTERS The Menard formation in Hardin County is made up of shale and lime- stone, the limestone predominating. The limestone layers are commonly deposited in beds a foot or so in thickness, which are separated by shaly partings. In the basal part of the formation there is a much thicker body of shale, commonly dark in color and in places black with carbonaceous matter, and locally, at least, with an impure coal bed a few inches thick. This basal shale bed attains a possible thickness of 25 feet, and is well exhibited at the road junction in the SW. y± sec. 11, T. 12 S., E. 8 E., about two and one-half miles north of Elizabethtown, and is again exposed along the Illinois Furnace road about one-half mile west of the same road junction. This shale is also well shown in an exposure on Harris Creek about one mile southeast of Sparks Hill, in the SE. % NW. 1/4 sec. 18, T. 11 S., B. 9 E., and in the SE. y 4 NW. % sec. 21, T. 11 S., E. 9 E. At the latter locality the shale includes an impure coal bed four inches thick, which lies about ten feet above the bottom, and also a thin layer of iron carbonate a little below the coal. The limestone of the Menard formation is mostly a dense, fine-grained, compact rock, bluish, dark gray, or nearly black in color. The weathered surfaces of the outcrops are smooth, and are commonly lighter in color than the freshly broken surfaces, in most places being light gray or bluish. A relatively small proportion of the limestone is rather argillaceous and some- what ferruginous, weathering to a yellowish or buff color. In its lithologic characters the Menard limestone is in rather strong contrast with the Glen Dean and Golconda formations, its smoothly weathering surfaces and dense, compact texture being very different from the rough, even crumbly weathered surfaces, and the granular, or crystalline texture of the two older formations. Crystalline beds, however, are not entirely lacking in the Menard, but they are unusual. Although the Menard is comparatively free from chert in Hardin County, and in many localities none at all is present, locally a small UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 205 amount is found in the form of thin, platy layers which are more or less discontinuous. The limestone beds of the Menard are commonly fossiliferous, but in most places sparsely so, and the fossils are apt to be so firmly imbedded in the rock that it is not easy to collect good specimens. Some of the calcareous, shaly partings between denser limestone layers, however, do have a consider- able number of well-preserved fossils in places. In Hardin County no sandstone member has been observed in the Menard, but in tracing the interval limited by the Tar Springs sandstone below and the Palestine sandstone above, westward into Pope County and still farther into Johnson County, a conspicuous chertv limestone formation is developed beneath the basal shales which have been described, and a sand- stone formation between the shales and the main body of limestone above. This sandstone begins in eastern Pope County as a very thin stratum, but thickens to the west, and in eastern Johnson County it increases to a massive sandstone formation which attains a thickness of fifty feet or more. Whether these two units should be included as members in a larger formation, along with the beds here designated as Menard, or should be defined as distinct formations is still open to question. In connection with the work in Pope and Johnson counties the name Vienna limestone has been applied to the lower, siliceous limestone, because of its excellent exposures at and near that town in Johnson County, and the sandstone has been named Waltersburg from the exposures at that place in Pope County. The main limestone mem- ber, as developed in Hardin County, is continuous throughout Pope and Johnson counties, and is unquestionably the exact equivalent of the limestones to which the name was Originally applied in Eandolph County. No sections through the entire thickness of the Menard in Hardin County are sufficiently well exposed to warrant their detailed description. In one section in eastern Pope County, however, three and one-half miles north of Golconda on the road to Eaum, in the adjoining quarters of sees. 1, 2, and 11, T. 13 S., E. 6 E., the beds occupying the interval between the Tar Springs and Palestine sandstones, are well exhibited, and may be recorded in this place. The lower limestone member in the section is the Vienna limestone, and the shale bed ivTo. 2, with sandstone layers at the top doubtless represents the Waltersburg sandstone, the Menard proper being beds Nos. 3, 4, and 5. Section of the Vienna, Waltersburg, and Menard formations three and one-half miles north of Golconda Feet 5. Limestone, with shale partings. Limestone beds for the most part dense and compact in texture, gray to blue-black in color, in beds one foot more or less in thickness. About 20 feet partly exposed below, 40 feet or more of loess-covered slope above with occasional limestone outcrops and some scattered chert fragments 00 20G GEOLOGY OF HARDIN COUNTY Feet 4. Limestone, compact, dense and hard, blue-gray to blue-black in color, weathering yellowish on surfaces; some beds in the lower portion filled with fossils, among which large examples of Spirifer increbescens are conspicuous 20 3. Limestone, thinly bedded with much calcareous shale, abundantly fossil- iferous, with many bryozoans and brachiopods 15 2. Shale, fissile, dark, blue-black in color in lower portion, becoming some- what arenaceous above, with platy sandstone layers at the top; the aren- aceous beds are olive green in color and break into elongate, sliver-like fragments 50 1. Limestone, dark gray to nearly black in color, in part dense and compact in texture, and in part somewhat crystalline, weathering to a rusty brown color on the surface, deposited in beds one foot or more in thick- ness, with much chert in plate- or sheet-like masses two or three inches thick. The chert is dark, chocolate-brown or black in color, it breaks vertically into subcubical fragments whose upper and lower surfaces, in contact with the inclosing limestone, are commonly decomposed to the depth of about one-half inch, the weathered zone being sharply differ- entiated from the central unmodified portion 20 THICKNESS The thickness of the Menard limestone cannot be directly measured with certainty in Hardin County, owing to conditions of outcrop, but the best esti- mates indicate a thickness of from 80 to 120 feet. In the Pope County section that has been recorded, 165 feet of sediments are included, but the Vienna limestone is well represented in the basal part of the section, and the be- ginnings of the Waltersburg sandstone are also there ; the upper portion of the section that may be strictly considered as Menard has a thickness of 95 feet. STRATIGuAPHIC RELATIONS As in the case of many of the other Chester formations in southeastern Illinois, the contacts between the Menard limestone and the superjacent and subjacent formations are obscured by talus or other surficial deposits in every section that has been observed. The fact that the upper portion of the sub- jacent Tar Springs sandstone becomes more thinly bedded than it is below, suggests the possibility of a gradual transition from the sandstone through arenaceous shales to more calcareous shales and limestone, but the field evi- dence available is altogether inconclusive. There is apparently no stratigraphic break between the several members of the formation — at least that is true in the best and most continuous sections that have been studied — and the con- dition probably prevails throughout the entire region. The transition from the Menard limestone into the superjacent Palestine sandstone is as obscure as is that at the base of the formation. No section has been observed where the contact between these two formations is exposed, but if the conditions governing the beginning of the deposition of the sand of the Palestine forma- UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 20 ? tion resemble those initiating the deposition of the older Chester sandstones, viz., the Bethel and the Hardinsburg, whose lower contacts have been observed, then the Palestine rests unconformably upon the Menard. Perhaps another line of evidence bearing upon this same question is the variation in thickness of the Menard. While this variation is not considerable, it may be due to the erosion of the upper surface of the formation preceding the deposition of the Palestine sandstone. In the southeastern portion of the county a distinct unconformity at the summit of the Menard is certainly present, for the higher Chester, Palestine, and Clore formations are wanting, and the Pennsylvanian Pottsville beds rest upon the Menard, according to the observations of Butts. PALEONTOLOGY The most prolific fossil fauna from the Menard limestone in Hardin or Pope counties, occurs in the basal portion of the formation. Where this bed is well exposed fossils in abundance can be collected. Nowhere in Hardin County has a good exposure of the bed been met with, but in the Pope County section which has been described above, this horizon is exposed in a glade-like surface, and a good- collection has been secured. The following species have been identified from this locality: Fossils from the Menard limestone collected three and one-half miles north of Golconda Triplophyllum spinulosum (M.-E. and H.) Pentremites fohsi Ulrich Hydreionocrinus sp. Agassizocrinus sp. (bases) Pterotocrinus menardensis n.sp. Crinoids (several fragmentary spe- cies) Archaeocidaris (plates and spines) Fistulipora excelens Ulrich Eridopora punctifera Ulrich Meekopora sp. Batostomella ? sp. Stenopora cestriensi? Ulrich Stenopora sp. Stenopora sp. Fenestella serratula Ulrich Fenestella (several undetermined species) Archimedes swallovanus Hall Archimedes distans Ulrich Archimedes meekanus Hall Archimedes proutanus Ulrich, var Lyropora sp. (bases) Septopora cestriensis Ulrich Rhombopora sp. Streblotrypa distincta Ulrich Bryozoan (genus and species unde- termined) Crania chesterensis M. and G. Orthotetes kaskaskiensis (McChes- ney) Productus ovatus Hall Diaphragmus elegans (N. and P.) Camarophoria explanata (McChes- ney) Dielasma shumardanum (Miller) Spiriferina transversa (McChesney) Spiriferina spinosa (N. and P.) Spirifer increbescens Hall Reticularia setigera (Hall) Eumetria costata (Hall) Cliothyridina sublamellosa (Hall) Composita subquadrata (Hall) Myalina sp. Allorisma clavatum McChesney 208 GEOLOGY OF HARDIN COUNTY Archimedes terebriformis Ulrich Sulcatopinna missouriensis (Swal- Polypora corticosa Ulrich low) Polypora tuberculata Proul Conularia sp. Polypora sp. Phillipsia sp. The above fauna may be taken as typical of this horizon of the Menard Jimestone throughout southeastern Illinois. Several of the species in the fauna are characteristic of this zone as it has been recognized, not only in Pope, but in Johnson and Union counties, while others are commonly met with in most Chester faunas. A single species of Pentremites is recorded, P. fohsi, a large form but not so large as P. obesus from the basal Golconda. This species has been met with wherever any considerable collection from this zone has been made in southern Illinois from Union to Hardin counties, and it has not been collected from any other horizon. Among the crinoids a species of Pteroto crimes , represented only by the "wing plates," has been found to be very characteristic. The members of this highly specialized genus have proven to be excellent horizon markers in the Chester group, the small spatulate plates with serrate border in the Paint Creek, the massive, sub- globose plates of P. capitalis in the Golconda, the peculiar bifid plates of P. bifurcatus in the Glen Dean, and now again in the Menard these very large, flattened, subcircular plates of a new species, P. mena/rdensis, which is allied to P. spatulatus. The detached, spinose, summit plates of a species of Hydreionocrinus have been collected in abundance from this horizon at a number of localities, being met with much more commonly at this horizon than anywhere else in the Chester series of southeastern Illinois. Likewise the plates and spines of the echinoid genus Archaeoddaris are more commonly met with in this portion of the Menard than at any other Chester horizon in the region. The bryozoans are represented by the same genera and mostly by the same species that are present in other Chester faunas. Among the brachiopods most of the species are commonly present in other Chester horizons, but the typical Spwifer increbescens, larger and coarser in appear- ance than the representatives of the same species in earlier faunas, is abund- ant, and the smaller Camposita trinwelea is displaced by the larger and broader C. subquadrata. The form of Eumetria that is commonly met with in this horizon is the E. costata, a larger and coarser form than the E. vera which was the usual member of the genus in the older faunas. The presence of the pelecypod Sulcatopinna missouriensis is a very characteristic mark of the Menard formation. The higher beds of the Menard are much less fossiliferons than the lower zone which has afforded the fauna that has been recorded. These dense, compact limestone beds are not well adapted for the preservation of fossils in good condition. Some of the shaly partings are fossiliferons, but nowhere have the specimens been observed to be so well preserved or so abundant as in the basal zone of the member. Some of the less compact limestone beds do UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 209 locally contain numbers of the large Spirifer increbescens and Composita sub- quadrata, and in places the highly characteristic Svlcaio pinna missonriensis is present in considerable numbers. This species is an elongate, wedge-shaped shell, some large examples being six inches or more in length, with a width of two inches at the broader, posterior extremity. Where the species occurs in its usual position it is imbedded vertically in the limestone with the pointed end down, and it doubtless lived in that position in the calcareous mud of the ancient sea bottom. On the weathered surfaces of the limestone ledges the subelliptical cross-sections of the shells may be seen in favorable situations. CORRELATION" The character of the series of sediments occupying the interval between the Tar Springs and the Palestine sandstones in Hardin County, closely resembles the series of beds occupying the interval between the Okaw lime- stone and the Palestine sandstone in Eandolph County. In following this interval westward from Hardin County, the Vienna limestone and the Wal- tersburg sandstone make their appearance below the main body of Menard limestone, but both of these have disappeared from the section, along with the underlying Tar Springs sandstone, in Eandolph County. The main body of limestone, as developed in Hardin County is clearly continuous westward, and is undoubtedly the exact equivalent of the typical Menard limestone in the Mississippi Eiver section. While no such prolific fauna with Pentremites folisi, such as has been recorded from Pope County, and traces of which have been observed in Hardin County, has been observed in Eandolph County, the general fauna of the formation, with Sulcatopinna missonriensis, the large Composita subquadrata and Spirifer increbescens, is present throughout. - The lithologic character of these beds, also, is remarkably similar throughout the whole extent of the formation, the only noticeable difference between the lime- stones of Pope and Hardin counties and those of Eandolph County being the somewhat darker color of the rock from the southeastern part of the state. When all the data are considered there seems to be no question as to the exact correlation of these beds entirely across the state. Palestine Sandstone name and. distribution The Palestine sandstone, like the Menard limestone, was originally named rom Eandolph County, Illinois. The sandstone is well exposed in the Missis- sippi Eiver bluffs in the neighborhood of Chester, where it was evidently •onsidered by some of the earlier workers to be the lower portion of the "Coal Aleasures." It was to this sandstone that the name Chester was originally pplied by Shumard 1 , and if the strict law of priority were followed Chester 1 Proc. Amer. Ass. Adv. Sci., vol. 11, pt. 2, p. 5 (1858). —14 G 210 GEOLOGY OF HARDIN COUNT!' as a group name would be abandoned. The name has become so fully estab- lished in literature, however, that it would be very unwise to attempt to change it at this time, the renaming of the sandstone formation being far more desirable. In the Mississippi Valley counties the Palestine sandstone has been recog- nized only in Eandolph and Jackson. In the more southern counties of the state the formation has been recognized across Union, Johnson, Pope, and Hardin counties. The formation extends southward across Ohio River into Crittenden and adjacent counties in Kentucky, but it is not conspicuously developed east of the western Kentucky coal basin. According to Butts, the Palestine sandstone is not now present in the southeasterly part of Hardin County, and if it and the higher Chester forma- tions were ever deposited there, they were removed by erosion before the Pottsville sediments were laid down. In the central faulted zone of the county (see figure 3) the Palestine sandstone is exposed in belt number two in a number of the fault blocks in the area lying between Keelin School and Ohio River. Just south of Keelin School it is exposed in an east-west belt in a small block, with Menard lime- stone to the south and the Clore formation to the north. The east-west hill lying along the northern border of the SE. % sec. 11, T. 12 S., R. 8 E., is capped with the Palestine sandstone, the south slope of the hill being under- lain by the Menard, although there are few exposures of the limestone. Just north of Stone Church there are two fault blocks whose surface outcrops are entirely of sandstone, that have been mapped as Palestine. With no limestone formations for reference there is of course a possibility of error in such determinations, but the lithologic character of the sandstone resembles the Palestine more closely than any of the lower sandstone formations. The most extensive Palestine outcrop in the region extends from the summit of the hill north of Carmens Store, in a southwesterly direction to the river bluffs about three-fourths of a mile northwest of Fairview. Through the entire length of this block the strata are dipping to the northwest, with the Menard limestone exposed at intervals along the southeastern border. In belt number one of the central faulted region, the Palestine occupies its normal position in the Chester group, between the Menard and Clore. The strip of territory occupied by the formation extends from the Wolrab Mill fault a half mile south of Sparks Hill, to the Hogthief Creek fault just west of Carrs Store. In the northwestern portion of the county the Palestine occupies a narrow belt extending from the Pope-Hardin County line a little north of Rose Creek, to the Lee fault at Loves Store. Eastward from here it is offset about one mile to the southwest, and then continues eastward to the Wolrab Mill fault. The district underlain by the formation is marked by knobs and ridges, in UPPER MISSISSIPPI AN SERIES, CHESTER GROUP 211 contrast with the low ground which in general marks the outcrop of the underlying Menard limestone. LITHOLOGIC CHARACTERS In its general composition the Palestine sandstone is more thinly bedded than the lower Chester sandstone formations, although locally it does con- tain rather thick bedded layers. Some layers are distinctly shaly, but the shales are arenaceous in character, being quite unlike the clay shales which are associated with some of the older sandstones. Some of the surfaces are ripple-marked, and some layers exhibit distinct cross-bedding. The rock is fine grained in texture and yellowish-brown in color, much of it being of a paler tint than most of the Tar Springs and Hardinsburg sandstones. In its appearance the Palestine sandstone of Hardin County more nearly resembles some of the higher sandstone beds of the Chester than it does the older Chester sandstones, and in places, where no limestone is exposed, it is difficult or impossible to know whether certain beds should be referred to the Palestine or to a higher position. THICKNESS The Palestine is one of the thinner sandstone formations of the Chester section in southeastern Illinois, and although its upper and lower limits have not been definitely fixed in any one section, reasonably accurate estimates of its thickness can be made in a number of situations in both Hardin and Pope counties. These estimates give an approximate thickness of GO feet, although in northwestern Hardin County the thickness appears to be fully 100 feet. STRATIGRAPHIC RELATIONS Like the contacts between most of the formations in Hardin County, that between the Palestine and the underlying Menard is commonly obscured. At one locality however, near the top of the hill west of Carmens Store, in the ravine north of the road, the contact between these two formations is fairly well shown. This section shows that the change from the limestone to the sandstone is abrupt, with no intergrading beds, a condition which suggests an interruption in the sedimentary record at this horizon. The exposure of the contact is not sufficiently broad to show clearly whether there are other evi- dences of unconformity here, and there is no development of a basal con- glomerate exhibited. It is not unlikely, however, that the sedimentary record was interrupted at this horizon, and that the Palestine lies unconformable upon the underlying limestone. Further evidence of an unconformitv between these two formations is shown in the Eandolph County section where both formations were originally defined. There is probably no physical break at the top of the Palestine. 212 GEOLOGY OF HARDIN COUNTY In Bandolph County, where both the Palestine and Clore formations were first recognized, there is apparently no stratigraphic break between them, the sandstone passing upward through sandy shales into more calcareous shales, and finally into limestones and shales interbedded. PALEONTOLOGY The only fossils which have been met with in the Palestine are more or less fragmentary tree trunks of the genus Lepidodendron, which has been identified by David White as L. cf. modulatum. CORRELATION The Palestine is the first of the Chester sandstone formations that is equally well developed in the southeastern Illinois section and in the Randolph County section. In both regions the sandstone formation occupies the same position in the section, and it maintains approximately the same thickness throughout. There can be no question as to the continuity of the bed through- out the entire Illinois basin of Chester time. Clore Limestone 1 name and distribution The Clore formation is typically developed in Bandolph County, where it caps some of the higher hills east of Chester. It extends beyond Bandolph into Jackson County to the southeast, and is exposed in its proper strati- graphic position entirely across the southern portion of Illinois, in Union. Johnson, Pope, and Hardin counties, but it becomes thinner and less con- spicuous in its eastward extension. When this unit was originally defined it was believed that the Clore was the highest of the Chester formations, and consequently, during the field studies in Hardin County all those beds between the Palestine sandstone below and the Pottsville above were referred to it. It was recognized, however, that this series of strata in southeastern Illinois was much thicker than in the typical exposures, and that it included an important sandstone member between two limestones, the lower of which is much the thinner and less con- spicuous. With the extension to the south and east into Jackson County, of the studies of the Chester group in the Mississippi Biver section, it has been found that the Clore limestone is succeeded by an important sandstone forma- 1 The sections of this report on the Clore limestone, Degonia sandstone, and Kinkaid limestone have been written after the close of the 1919 field season. Before this time all three of these units have been considered as members of the Clore formation, but later observations in southeastern Randolph and northwestern Jackson counties by Mr. J. Marvin Weller, have established the importance of these higher Chester divisions, ana have led to their recognition as distinct formations coordinate in rank with the other units of the Chester group. Upon the accompanying map all three of these formations are included under a single color with the Clore symbol, but if the area were re-workea all of them could be differentiated in much or all of the area. UPPER MISSISSIPPI AN SERIES, CHESTER GROUP 213 tion 100 feet or more in thickness, and this again by a thick limestone, the two having been named the Degonia sandstone and the Kinkaid limestone respectively. A comparison of this section, and of the fossil faunas of the two limestones, with the uppermost Chester of southeastern Illinois, has estab- lished the fact that the Clore limestone is properly represented in Hardin and Pope counties by the lower limestone member only of the formation as it was originally mapped. In Hardin County the Clore limestone as now restricted, is the thinnest one of all the limestone formations of the Chester group, and comparatively few good outcrops of the formation are exposed. An outcrop of it may be seen, appearing beneath the overlying Degonia sandstone, in the short ravine situated in the NW. % SW. 14 sec. 12, T. 12 S., E. 8 E., three-fourths of a mile southeast of Keelin School. It is again exposed, only a few feet thick, in the roadside less than one-fourth of a mile south of Keelin School, here again being overlain by sandstone. Two and one-half miles northwest of Elizabeth- town an inconspicuous outcrop of limestone by the road-side in the SE. 14 SE. 14 sec. 17, T. 12 S., E. 8 E., is believed to represent the Clore limestone, and from this point it extends in a northeast direction through the extent of the small fault block in which it is situated, although the outcrops are meager. Another exposure may be seen in the road-side a mile west of Carmens Store, north of Eosiclare. One of the best exposures of the formation, and the only one from which a representative fauna has been collected in Hardin County, is situated about half a mile southeast of Sparks Hill, in the xs r W. ^4 sec - 18, T. 11 S., E. 9 E. According to Butts, the Clore and higher Chester formations are not exposed at all in the southeastern portion of Hardin County, the pre-Pennsyl- vanian unconformity being greater in this part of the county than elsewhere, in consequence of which the Pottsville sandstone lies in contact with the Menard limestone. LITHOLOGIC CHARACTERS In its typical exposures in Eandolph County, the Clore limestone is exceedingly variable in character, and includes a considerable amount of shale. In places the limestone ledges are hard, compact, and dense, others are shaly with more or less carbonaceous matter, while still other layers are somewhat crystalline. In color the different layers vary from gray or bluish gray to nearly black. The shale beds commonly constitute a considerable portion of the total thickness of the formation, in places more than one-half; they are completely argillaceous or more or less calcareous in character. In Hardin County the character of the formation is similar to that in its typical exposures. The shaly character of the limestone layers themselves, and the large amount of argillaceous and calcareous shales associated with the limestones is a feature of the formation wherever exposures of it can be seen 214 GEOLOGY OF HARDIN COUNTY Some of the limestone layers are filled with fragmentary fossil brachiopods, bryozoans, which are commonly so broken or so firmly imbedded that it is not possible to identify them with any degree of satisfaction. THICKNESS In Eandolph County the maximum thickness of the Clore limestone is 50 or 60 feet. It becomes thinner in its extension to the southeast, becoming in most of Johnson and Pope counties but 30 or 40 feet. In Hardin County it seems to become still thinner, and probably is nowhere more than 20 or 30 feet, and in places it may be even thinner than this. STRATIGRAPHIC RELATIONS Because of the thinness of the formation and the consequent covering of the outcrop by surficial deposits the actual contacts of the Clore limestone with the underlying and overlying sandstones has nowhere been observed, and the stratigraphic relations of the formation with its bounding sandstones are not known. There are no data to show that any unconformity exists either below or above the formation, and it is quite likely that it constitutes a portion of an entirely conformable series of beds. PALEONTOLOGY All of the limestone beds in the Clore formation contain more or less fragmentary fossils, most of which are too imperfect to be determined. From some of the shaly beds better fossils can be collected. The only important collection that has been made in Hardin County is from the locality about one-half mile southeast of Sparks Hill, in which the following species have been identified : Fossils from the Clore limestone collected near Sparks Hill, Hardin County Pterotocrinus sp. (wing plate) Camarcphoria explanata (McChes- Batostomella nitidula Ulrich ney) Fenestella serratula Ulrich ? Dielasma shumardiannm (Miller) Fenestella tenax Ulrich ? Spiriferina transversa (McChesney) Archimedes sp. Spirifer increbescens Hall Productus arkansanus Girty ? Reticularia setigera (Hall) Diaphragmus ? sp. Eumetria costata (Hall) ; Composita subquadrata (Hall) The presence in this fauna of the large and broad Composita subquadrata marks this fauna as of upper Chester age, that is it could not be older than the Menard, and the large examples of Spirifer increbescens have the same significance. Most of the other species might be present in any one of several other Chester horizons although the rather large Productus that has been referred with a query to P. arl'ansanus has not been observed elsewhere. The one species that is most significant is the -bryozoan Batostomella nitidula UPPER MISSISSIPPI A.N SERIES, CHESTER GROUP 215 This little, branching, cylindrical form occurs rather commonly upon the surfaces of the limestone layers at this locality and elsewhere in the lower portion of the Clore, and has not been collected by the writer from any other horizon in southern Illinois. In places it is very abundant, the broken zoaria being crowded together upon the surface of the limestone layers, and it is perhaps the best index fossil for this horizon. One of the best faunas from the Clore that has been collected in south- eastern Illinois is from a locality in Pop# County about four miles northwest of Golconda. The stratigraphic position of the collection in the formation is between 30 and 40 feet above the top of the Palestine sandstone, and the species that have been identified are as follows : Fossils from the Clore limestone collected northwest of Golconda, Pope County Triplophyllum spinulosum (M.-E. Streblotrypa nicklesi Ulrich and H.) Streblotrypa sp. Pentremites sp. Crania chesterensis M. and G. Pterotocrinus sp. (wing plates) Orthotetes kaskaskiensis (McChes- Agassizoerinus sp. (base) ney) Archaeocidaris sp. (spines and Productus ovatus Hall plates) Productus sp. Batostomella nitidula Ulrich Diaphragmus elegans (N. and P.) Batostomella (several species) Camarophoria explanata (McChes- Stenopora cestriensis Ulrich ? ney) Stenopora sp. Girtyella indianensis (Girty) Lioclema? araneum Ulrich Spiriferina spinosa (N. and P.) Lioclema sp. Spiriferina transversa (McChesney) Fenestella serratula Ulrich ? Spirifer increbescens Hall Fenestella elevatopora Ulrich ? Reticularia setigera (Hall) Fenestella (several species) Eumetria costata (Hall) Archimedes (several species) Cliothyridina sublamellosa (Hall)? Polypora approximata Ulrich Composita subquadrata (Hall) Polypora sp. Bellerophon sp. Septopora cestriensis Ulrich Euomphalus sp. Rhombopora (several species) Pleurotomaria ? sp. The outstanding features of this fauna are the abundance of the delicate, branching, cylindrical bryozoans belonging to the genera Batostomella and Rhombopora. Among these, Batostomella nitidula is a conspicuous form, and it is associated with large numbers of the large and broad Composita sub- quadrata. This faunal combination is characteristic of the Clore faunas throughout southern Illinois. CORRELATION The essential identity of the Clore fauna in Hardin County and elsewhere in southeastern Illinois, with that from the typical exposures of the formation, as well as the fact that the formation can be and has been traced almost con- 216 GEOLOGY OF HARDIN COUNTY tinuously across the state, establishes without any doubt the correlation of these beds in Hardin County with the original Clore limestone of Randolph County. The formation is not so thick towards the east, but nevertheless both its lithologic and faunal characters remain essentially the same throughout the whole extent of the formation. Degonia Sandstone name and distribution The Degonia sandstone has been named from Degonia Township in western Jackson County, Illinois, where the sandstone is a conspicuous cliff- making sandstone in the bluffs of the Mississippi River and especially in the sides of the valleys tributary to the Mississippi. The formation has been traced continuously in the Mississippi River bluffs to the intersection of the valley of the Big Muddy River. Where this sandstone was first observed in Randolph County, some years ago, before it was known to be overlain by a thick limestone with a characteristic Chester fauna, it was believed to be the basal member of the Potts ville formation in that portion of the state, and it has been included in the Pottsville in the mapping of the Murphysboro quadrangle. 1 In the more southern counties of Illinois this sandstone has been recog- nized across Jackson, Union, Johnson, and Pope counties, although in the earlier field work in this area it was considered as a member of a more compre- hensive Clore formation. This and the Palestine are the only two of the Chester sandstones that are present clear across the state from Randolph to Hardin counties. In tracing the formation to the southeast it retains essentially the same characters which it possesses in Randolph and Jackson counties, into Union County. Farther to the east it becomes somewhat more thinly bedded and less massive, and this character persists throughout Hardin County. Much of the Degonia sandstone in Hardin County resembles the Palestine more or less closely, and it is possible that in some parts of the county the Clore has been overlooked where it is thin and has been obscured by surficial covering, and that the Degonia has been included with the Palestine in the mapping. The formation is well exhibited in the hill just west of Iveelin School, three and one-half miles northeast of Elizabethtown, and it is also well exposed in the hills about one-half mile southeast of the same school, in the west half of sec. 12, T. 12 S., R. 8 E., where its relation to the Clore limestone may be seen. Other good exposures of the Degonia sandstone are present in the western part of sec. 20, T. 12 S., R. 8 E., along the road west from Carmens Store, two and one-half miles north of Rcsiclare. At this locality the relation of the sandstone to the underlying Clore limestone may be seen, and the sand- U. S. Geol. Surv., Murphysboro-Herrin Folio (No. 185). UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 21? stone itself is well exposed in the hills south of the road in the same section. The most notable development of the Degonia sandstone in northern Hardin County is a third of a mile south of Sparks Hill, in the northern part of the NE. 14 sec - 13> T. 11 S., R. 8 E., where there is a massive sandstone layer ten feet or more thick which forms a pronounced ridge on which the volcanic plug of that locality outcrops. LITHOLOGIC CHARACTERS In the region of its typical occurrence the Degonia is a very massive, cliff -making sandstone, and is not unlike, in any respect, many of the pebble- less sandstone layers of the Pottsville. In massiveness the formation more nearly resembles the Cypress sandtone of the more southern Illinois counties, than any other Chester sandstone. Locally, however, the formation does con- tain rather thick strata of thinly bedded sandstones which are conspicuously ripple-marked, and which are almost arenaceous shales in places. Much of the formation exhibits notable cross-bedding. The texture of the sandstone in its typical exposures is somewhat coarser than most of the Chester sandstones, and the weathered bluff surfaces not infrequently exhibit more or less conspicuous iron crusts similar to those present in the Pottsville sandstones in many places. The color of the sand- stone is commonly light brown upon freshly broken surfaces, but becomes darker upon weathered surfaces through the oxidation of the iron. THICKNESS In its typical development in Randolph and Jackson counties the Degonia sandstone reaches a maximum thickness of somewhat more than 100 feet, and is probably nowhere less than 100 feet. In the eastward extension of the formation it becomes somewhat thinner, and in Hardin County it probably does not exceed 50 or 60 feet in any section, and in places it may be con- siderably less than this. STRATIGRAPHIC RELATIONS No observations have been made in Hardin County to indicate that the Degonia sandstone lies in other than conformable relations with the underly- ing Clore, or the overlying Kinkaid limestones, although it must be kept in mind that it has not been possible to study the actual contacts between these formations in any section. PALEONTOLOGY Like the other sandstones of the Chester group in Hardin County, the Degonia has furnished very few fossils. Those that have been observed are fragmentary plant remains, the best preserved of which are parts of Lepido- dendron trunks. 218 GEOLOGY OF HARDIN COUNTY CORRELATION Of all the Chester formations in Illinois, the Degonia sandstone is more nearly traceable from the typical Eandolph County area into the southern counties, than any other. It has been recognized in the Mississippi River bluffs southward as far as the valley of Big Muddy River, and it is known to be well developed in northern Union County. When the small intervening region can be examined in detail, it is altogether probable that the formation can be traced continuously. In any event the equivalency of the sandstone from Randolph to Hardin County cannot be questioned, the correlation being based not alone upon its continuity, but also upon the uniformity of the paleontological characteristics of the underlying and overlying limestones. Kinkaid Limestone name and distribution During the past season (1919) it has been shown 1 that an important limestone is present in the Chester section of Randolph and Jackson counties, that has not heretofore been recognized. It is separated from the Clore lime- stone by a thick and massive sandstone formation, the Degonia sandstone which has already been described. The formation is well exposed along Kin- kaid Creek and some of its tributaries in Jackson County. In the NW. % sec. 6, T. 8 S., R. 4 W., the valley of this creek takes the form of a gorge cut through the limestone. Because of these exposures the name Kinkaid has been chosen to designate the formation. This limestone is known to be present in Jackson County nearly as far south as Big Muddy River, and it is again well developed in northern Union County. Excellent exposures of it are to be seen entirely across Johnson and Pope counties, and it continues with diminished thickness into Hardin County. South of Ohio River it is well exposed in the Chester section of Kentucky. With the underlying Degonia sandstone and Clore limestone, the Kinkaid is not present in the southeastern, unfanlted portion of Hardin County, the pre-Pennsylvanian unconformity being greater in this part of the county than elsewhere, in consequence of which the Pottsville sandstone rests upon the Menard limestone, according to Butts. In the central faulted region of the county the outcrops of these three formations are essentially restricted to the deeply down-dropped middle zone (see figure 3), and in all of the fault blocks in which they are exposed, except some of the smaller ones, they are overlain by the Pottsville sandstone which caps all of the higher hills along this belt. In most of the northeastern half of this belt its width is entirely ocupied by the Pottsville beds, but from an east-west line drawn one mile north of the line between townships 11 and 12 Through the field studies of Mr. J. Marvin Weller. UPPER MISSISSIPPI AN SERIES, CHESTER GROUP 219 south, the Kinkaid formation is exposed beneath the Fottsville along both sides of the fault block. Good exposures of the formation are exhibited about Keelin School, and for a distance of about two miles northeast of this school. Farther north the Kinkaid is present in the small Chester area which is mostly confined to sec. 21, T. 11 S., K. 9 E., on the northwestern margin of the same fault block; this outcrop occupies a belt about two miles long. In a south- westerly direction from Keelin School, however, there is an interruption in the outcrops of the Kinkaid and associated formations, but beginning at Big Creek they are again well exposed more or less continuously beneath the Fottsville, to the Ohio Eiver bluffs between the mouth of Wallace Branch and a point half a mile east of the mouth of Threemile Creek. In belt number one of the central faulted zone, these formations extend from the Wolrab Mill fault just south of Sparks Hill, eastwardly to the Hog- thief Creek fault, one-fourth of a mile north of Carrs Store. In the northwestern portion of the county the Kinkaid occupies a belt just south of the Pottsville outcrops. For a short distance between Buzzards Point and High Knob this belt extends northward to the county line and beyond into Gallatin County, but through most of the distance there is an area underlain by Pottsville between the limestone outcrops and the county line. East of the Lee fault this formation, like the others, is offset to the southwest and then continues eastwardly to the Wolrab Mill fault. Northwest of the Herod fault which crosses the extreme northwestern corner of Hardin County there is a small area of limestone which is in fault contact with the Pottsville and Kinkaid on the southeast, and lies in normal stratigraphic contact with the Pottsville on the north. LITHOLOGIC CHARACTERS In Hardin County the Kinkaid limestone resembles the exposures of the same formation farther west. The formation is in the main limestone and shale, but possibly with some thin minor beds of sandstone. The limestone beds are rather variable in character, but they are in the main hard, dense, and compact, breaking with a splintery or conchoidal fracture, being very similar in many respects to the limestones of the Menard formation. In color these limestone beds are gray, yellowish, or black, the yellow layers being a conspicuous feature in many outcrops. The yellow color is rather more common in the Kinkaid than in any other limestone of the Chester group. A notable difference between these beds and those of the Menard is in the much greater amount of silica which they include. Some of the limestone beds themselves are apparently siliceous, and in addition chert beds of greater or less extent are scattered through the formation. One especially notable chert bed present in the lower part of the formation, seems to be continuous through Pope and Johnson counties at least, and extends into the adjoining 220 GEOLOGY OF HARDIN COUNTY part of Kentucky. It is a very massive chert ledge two or three feet in thick- ness, whose presence is commonly indicated by the great amount of chert in subcubical masses, varying in size from less than one foot to as much as two feet or more, which are scattered through the residuum; locally in Pope County this massive chert ledge is well exposed in situ. It is commonly light colored, in places nearly white, elsewhere with a greenish tint. Other chert layers are present in the formation, most of them being dark colored, some bands being not unlike the chocolate-colored chert layers which are commonly present in the Vienna limestone of Pope and Johnson counties. Little or no chert has been observed in the formation in northern Hardin County, and it is nowhere so conspicuous as in many localities in Pope County. The shales of the Kinkaid formation are fully as variable in character as are the limestones. In color they are black, gray, olive-green, and red, one conspicuous red shale bed apparently being a persistent member of the forma- tion. Some of the shales are almost pure clay, some are calcareous, some are siliceous, and some are more or less sandy. The individual beds vary in thick- ness from thin shaly partings between the beds of limestone to strata fifteen or more feet in thickness. THICKNESS The exposures of the Kinkaid formation in Hardin County are not well adapted for determining its total thickness. In those situations where the basal contact of the formation can be approximately fixed, the superjacent Pottsville is not present, and where the upper limits of the formation can be determined the base is not well exposed. The exposures in Pope County arc better adapted for determining the thickness, however, and from field studies made in that county it has been determined that the Kinkaid exceeds 100 feet in thickness as exhibited in surface exposures. As indicated by deep well records it seems that in places the formation is even thicker than this, perhaps 150 feet or more, but the thickness varies because of the extensive post-Chester erosion which preceded the Pottsville sedimentation. Furthermore, from the fact that the Pottsville rests upon the Menard limestone in the south- eastern portion of Hardin County, it is quite likely that the Kinkaid forma- tion becomes thinner in its eastward extension from Pope County, and disappears entirely before the eastern border of Hardin is reached. If this condition is true it is probable that nowhere in Hardin County does the formation attain the maximum thickness exhibited farther west, and it prob- ably does not even reach the thickness shown in the surface outcrops in Pope County. STRATIGRAPIIIC RELATIONS The stratigraphic relations of the Kinkaid formation with the underly- ing Degonia sandstone have already been considered in connection with the UPPER MISSISSIPPIAN SERIES, CHESTER GROUP 221 description of the older formation. No evidence is at hand to indicate that a stratigraphic break separates the two formations. The upper limit of the Kinkaid limestone is everywhere marked by the sub-Pennsylvanian uncon- formity that is present throughout the entire Illinois basin. The actual con- tact between the limestone and the overlying Pottsville is commonly talus covered and has been observed in only one locality near Hardin County. This is in the road north of Karbers Eidge a few hundred feet north of the county line. Here the Kinkaid is immediately overlain by a ferruginous con- glomerate, succeeded above by sandy shales and sandstones as shown in the following section: Section north of Karbers Ridge Feet 6. Sandstone, thin-bedded, white 10 5. Shale, fine, fissile clay, with iron nodules 10 4. Shale and sandstone, white sandy shale and white sandstone layers of pure white quartz sand without mica 10 3. Conglomerate, ferruginous, with quartz pebbles up to one-half inch in diameter 1 2. Clay, red and green, and limestone 15 1. Limestone, buff and gray, fine-grained iy 2 The passage here from the Kinkaid at the top of the Mississippian to the Casey ville formation at the base of the Pennsylvanian is abrupt and com- plete at the bottom of No. 3 of the above section. The highly ferruginous character of the conglomerate indicates a low, swampy condition of the land for a long time before and during the gradual transgression of the Pennsyl- vanian conglomerate upon the old eroded Mississippian surface. In one section in Pope County the base of the Pottsville is marked by a conglomerate which includes numerous limestone pebbles and small boulders that clearly originated from some of the underlying Chester formations, and associated with them are red granite pebbles which vary in size up to the magnitude of a man's fist, which must have been transported a very long distance. PALEONTOLOGY Most of the limestone beds in the Kinkaid formation are more or less fossiliferous, but the fossils are not generally preserved in such a manner as to be properly identified. The shale beds of the formation are commonly very little if at all calcareous, and are likely to be barren of fossils, this being especially true of the red and olive-green shales. From one layer in the lime- stone in the NE. % SW. % sec. 30, T. 12 S., E. 8 E., near Threemile Creek, two miles northwest of Eosiclare, the following species have been collected: 222 GEOLOGY OF HARDIN COUNTY Fossils from the Kinkaid limestone collected in the NE. % SW. 14 sec. 30, T, 12 8., R. 8 E., northwest of Rosiclare Triplophyllum spinulosum (M.-E. Spirifer leidyi N. and P. and H.) Martinia contracta M. and W. Orthotetes kaskaskiensis (McChes- Eumetria costata (Hall) ney) Cliothyridina sublamellosa (Hall) Productus ovatus Hall Composita trinuclea (Hall) Productus sp. • Bellerophon sp. Diaphragmus elegans (N. and P.) Naticopsis sp. Camarophoria explanata (McChes- Allorisma sp. ney) Aviculopecten sp. Girtyella sp. Orthoceras sp. This fauna has nothing strikingly distinctive about it, and were not its horizon definitely fixed by its stratigraphic relations it might be considered as of middle Chester age. The examples of Composita which are present are the the C. trinuclea type, identical with those which occur commonly in the middle and lower Chester faunas, and not at all like the large and broad examples of C. subquadrata of the Menard and Clore. The examples of Martinia con- tracta most resemble those which have been collected from some of the higher beds of the Glen Dean formation near Golconda, in Pope County. A collection from the outcrops of Kinkaid limestone on Kinkaid Creek in Jackson County, exhibits the same features as those just recorded from Hardin County. A species of Martinia, apparently the same as that in the list just given, is one of the common forms, and the Composita present is of the C. trinuclea type rather than the large form so common in the Menard and Clore limestones. A number of pelecypods are present in the fauna, among them the Menard species Sulcatopinna missouriensis. CORRELATION The only question of correlation involved in the consideration of this limestone in Hardin County, is its equivalence to the beds in Jackson and Randolph counties. The almost complete continuity of the formation from one region to the other, with the common paleontological characteristics, sufficiently establishes the identity of the formation entirely across southern Illinois. CHAPTEE IX— PEXXSYLVAXIAX By Chas. Butts Sub-Pexnsylvaxiax Unconformity The Kinkaid formation of the Chester group, the uppermost of the Missis- sippian formations in southern Illinois, is succeeded above by the Caseyville formation, the lowest formation of the Pennsylvanian system. The Kinkaid is probably not the youngest formation laid down in Mississippi an time, and the Caseyville certainly is not the oldest formation of Pennsylvanian time. Between the time represented by the top of the Kinkaid and the time of de- position of the basal beds of the Caseyville in Illinois, there was formed along the eastern margin of the Appalachian coal fields, a great series of rocks that is not represented in Illinois, but which, if it were present, would lie between the Kinkaid and the Caseyville. These beds which are lacking from the Illinois section constitute the lower part of the coal-bearing rocks of the Anthracite fields of eastern Pennsylvania, the rocks of the Pocahontas coal field of Virginia and West Virginia, and the Coal Measures of Alabama, the aggre- gate maximum thickness of which is probably not less than 12,000 feet. These lowest Coal Measures rocks which are not represented in the Illinois section, are classed as lower and middle Pottsville. The oldest Pottsville rocks of Illinois are of middle and upper Pottsville age, and correspond in the time of their origin with the beds which overlie the lower Pottsville and the lower part of the middle Pottsville rocks of the Appalachian coal fields. The strati- graphic relations of these formations are shown in the table giving the general geologic section on page 80, and by the accompanying diagram, fig. 13. In a conformable geological succession one layer is deposited upon another without any considerable lapse of time between the two depositions, and such a relation doubtless exists between some of the formations of the Chester group. Between the deposition of the Kinkaid and the Caseyville formations, however, a very long time, perhaps millions of years intervened — all the time in fact required for the accumulation of the thousands of feet of rocks including many thick coal beds, that are present in the Appalachian coal fields but wanting in the Illinois section. Just what was taking place in Illinois while this great thickness of rocks was accumulating farther east, is not fully known, but it is certain that the region was land during that time or a large part of it. Rocks do not usually accumulate upon a land surface. Instead they are worn away from it and carried by the streams to the ocean 223 224 GEOLOGY OF HARDIN COUNTY NVINVA1ASNN34 NVIddlSSISSIIN cs .°2 Q H t tu *£ (o o * s I s i » o c B d a ft > £ -S 0) o CO W B ctf M a !>» > to 09 Ci .2 O Ph '3 b > CD > >. o CO J CI „ CD* CD Ci B Ph o T3 B Ci cl B Ci a o CD > cl B Ci "ft T3 8 a> ft A 'to £ - 1 - ■f. w >> "tfi ?-. ,0 01 B s 0) Oh -B cd .B X -(-J «W a> o h CM ft O CD a> o F-l to d Ci a> to o to ,ri +j ^2 OS >» B -*-> "3 o CM a; H ,B B +j BO o M) O d CI Ci p V a bJD o HH B to Ci +j B S B -B to Ci T. 11 S., R. 9 E., there is a higher coal bed, apparently over two feet thick, which dips strongly to the northwest so as to lie below the Saline River flats, immediately to the northeast of the road. The yellow limestone layer which has been mentioned, lies just above this coal. In the Tradewater of the mapped area in Pope County, a coal about 18 inches thick outcrops in the road half a mile north of Harts Store. Below it is a siliceous limestone with 1 Glenn, L. C, Coal of the Tradewater River Region; Ky. deol. Surv., Bull. 17, (1912). PENNSYLVANIA^ 229 invertebrate fossils including Marginifera muricata and Chonetes mesolobus. This is probably the Curlew limestone and the coal, the Curlew coal. THICKNESS The thickness of that portion of the Tradewater represented in Hardin County is estimated as 300 feet. The greatest thickness seems to be under the high knob a third of a mile north of Lambert School in sec. 3, T. 11 S., E. 9 E. If the beds were horizontal at this point there would be at least 220 feet of Tradewater represented, but allowing for increased thickness due to a north dip of 5° to 10° the thickness must be not less than 300 feet. In Pope County the thickness probably reaches about 500 feet. The full thickness of the formation in Kentucky is given as 594 feet. 1 PALEONTOLOGY AND CORRELATION The fossils of the Tradewater formation are of Upper Pottsville age, specifically of Mercer, and probably of Conoquenessing age. On combined stratigraphic and paleontologic evidence it is concluded that the Tradewater is of Upper Pottsville age, and probably includes beds of the same age as the Homewood sandstone and Mercer shale of western Pennsylvania. 1 Lee, Wallace, Geology of the Shawneetown quadrangle in Kentucky : Ky. Geol. Surv. (1916). CHAPTER X— SURFICIAL DEPOSITS By Stuart Weller General Statement The hard rocks which underlie the whole of Hardin County are hidden from view over most of the area by unconsolidated materials which constitute the so-called mantle rock or surficial deposits. This material has had a very different origin from that of the hard rocks, all of which were accumulated in the waters of the ocean. These mantle rocks are all terrestrial deposits, that is they were accumulated upon the surface of the dry land, although it is understood that some of the material was actually accumulated through the agency of the fresh waters of Ohio Eiver and its tributary streams. The surficial deposits in Hardin County are of three sorts, the residual deposits, the alluvial deposits, and the wind-blown deposits. Residual Detosits The residual deposits include all of those which have accumulated in place, by the disintegration or weathering of the hard rocks. An important type of such deposits is the red clay beds which overlie the limestone forma- tions throughout the county, wherever such material has not been washed away by the waters that have fallen as rain and have been drained off through the streams. These residual clay deposits are made up of the insoluble ma- terial that is left behind where the limestone has been removed by solution, and imbedded in the clay are any insoluble portions of the limestone, such as the siliceous beds which we call chert or flint. Because some of the limestones of the county are very cherty, some of the clays of this sort are filled with broken fragments of chert which were originally in the form of flint or chert layers in the solid limestone, such as are seen in the Ohio River bluffs at Elizabethtown. Some of the residual deposits may have been moved short distances in the process of their accumulation. Such would include the talus deposits which accumulate along the bases of bluffs or steep hills. These deposits are made up of masses of rock of greater or less size which have been broken off from the outcropping ledges in the upper part of the hills through the action of frost, by the pressure of the growing roots of plants and trees, by the un- dermining of an upper, more insoluble formation like sandstone, by the solution or disintegration of an underlying limestone or shale, or by the slow 231 232 GEOLOGY OF HARDIN COUNTY disintegration of the rocks. Material of this sort rolls down the hill side and accumulates at its base mixed with soil. The size of the rock fragments in the talus may vary in size from small particles to masses of many cubic yards. Alluvial Deposits A second type of surficial material is that which has accumulated in the stream valleys through the agency of the transportive power of the streams. 'The source of the material is the residuum which has accumulated as already described, either here in Hardin County or elsewhere. This material is con- stantly being carried down grade by the movement of water which falls as rain, and will eventually be transported to the ocean, but as the streams have cut down their valleys, level stretches have been formed in some places, in which the slacking of the current has caused the material being carried to be dropped. Such accumulations of material make up in part the alluvial deposits of the so-called bottom lands. In some of the stream valleys of the county, such as Big Creek, whose bottom lands are subject to overflow from backwater of Ohio Eiver, these alluvial deposits are added to from year to year by the settling of part of the fine sediment held in suspension by the overflow waters. There are really three types of alluvial deposits in the county; the ma- terial accumulating in the narrow bottom lands of some of the streams tributary to Ohio Eiver; that accumulating in the Ohio River valley itself at the present time ; and material which has accumulated in the Ohio Eiver Valley at an earlier period when the river level itself was higher than it now is. Geologically speaking all alluvial deposits are only temporary, and if the full cycle of erosion is completed they will finally be deposited in the ocean; but humanly speaking, some of these deposits remain in position for thou- sands of years. If there is any way of determining the length of time that an alluvial deposit in a stream valley remains undisturbed, definite evidence becomes available which establishes within certain definite limits the age of the valley itself where the deposit occurs, and consequently some idea can be gained of the length of time during which the present topography of the region has remained comparatively unchanged. The most satisfactory sort of evidence to establish such facts as these is paleontological, that is, the evidence afforded by the remains of animals or plants which have existed in the past. Fortunately, in Hardin County, evidence of this sort is available, for there has been found buried in the alluvial deposits in the valley of Big Creek, about three and one-half miles from the stream's mouth, the skeleton of a mastodon, an animal related to the living elephant, which formerly lived throughout the interior of North America, but which has long been extinct. While of course it is not possible to determine the exact number of years that SURFICTAL DEPOSITS 233 have elapsed since the mastodan was living in southern Illinois or elsewhere, the evidence which has been accumulated throughout the country, goes to show that these creatures have been extinct for ten thousand years or more, and consequently it can be concluded that the valley of Big Creek has existed in essentially its present condition for that length of time. This evidence from Big Creek carries with it the assumption that the present topography of Hardin County has existed with but comparatively little change for at least ten thousand years, and perhaps considerably longer, and while this is a long time in human experience, it is a very brief period in geology. The alluvial deposits of the Ohio Eiver bottom were accumulated in the same manner as those of the smaller streams, but the source of the material was not local, much of it having been carried down the stream from far to the east, perhaps as far away as Pennsylvania. With every overflow of the river a thin layer of sediment is deposited in places on its flood plain, while elsewhere material may be carried away. On the whole, doubtless, the flood plain is being very slowly built up by the addition of such sedimentary material. If the mouth of Ohio Eiver should be lowered in any manner, the channel would be quickly cut down to a position to conform with the new elevation of the mouth, in which case the present flood plain would be left relatively high above the river level and would not be subject to overflow, and a new flood plain would be built at a lower elevation, the remnants of the present one being left as a terrace. The plain upon which the lower portion of the town of Bosiclare is built is a terrace produced in a comparable manner, and a few other remnants of the same terrace are recognizable elsewhere in the county. Wind Blown Deposits The uplands of Hardin County, and the hill-sides with more gentle slopes, are covered with an exceedingly fine, yellowish or reddish-yellow ma- terial which characteristically constitutes all of the upland soil of the county. This material was not derived from the underlying hard rocks, for its char- acter is the same whether it overlies limestones or sandstones. It could not have been carried to its present position through the action of running water, for it has its best development and is purest in character on the tops of the hills. The wind is the only agent that could have placed it where it now is, and because of its exceedingly fine texture it might easily have been trans- ported in that manner. It is known that in an earlier period in the history of the region the broad Ohio Eiver flood-plain was essentially bare of vegeta- tion, and in consequence during that period an abundance of material was available in the fine river silts of the flood plain, which could be picked up by the winds blowing from the south and be carried up onto the hills. In order to be convinced of the efficacy of the wind to accomplish this sort of 234 GEOLOGY OF HARDIN COUNTY work it is only necessary to watch one of the sand bars of the present river in a low stage of the water, during a brisk wind, when clouds of material are raised into the air. At a time when the whole flood-plain of the stream from valley wall to valley wall was bare of vegetation, one can imagine the great quantity of material which, during a large portion of the year, could be transported in this manner. As the vegetation gradually came to clothe the surface, the wind could no longer pick up such vast quantities of material, and the hills themselves gradually became covered with trees and other forms of vegetation. Since the period of its original deposition much of this wind-blown material which is commonly called loess, has been shifted more or less by slumping and by wash, so that in many places it has now become more or less mingled with the underlying residuum upon which it was originally deposited, but it still remains in its pure condition upon most of the hill tops. It is this loess that is so easily gullied by the run off of the rain water, and upon many slopes that have been cultivated for a considerable period of time, excellent exposures of the formation can be seen in the gullies that have been formed. Not infrequently the strong contrast in character between the upper mantle of loess, and the underlying residuum of clay and chert, which does not wash so easily, can be seen. PART IV IGNEOUS ROCKS By L. W. Currier CHAPTER XI. IGNEOUS ROCKS By L. W. Currier The igneous rocks of Hardin County are of three types: (1) fine- grained, dark-colored trap or, more technically, lamprophyre, (2) medium- grained, dark-colored mica-peridotite, and (3) volcanic breccia (?). All three types are severely altered, so that a close classificaton cannot be made. The lamprophyre occurs as narrow dikes and sheets enclosed in the sediments ; the peridotite is found only in the form of dikes ; and the volcanic breccia is seen as weathered boulders suggesting by their distribution an elliptical in- trusion of volcanic rock. The exposures of igneous rocks thus far identified are mainly along the bluffs of Ohio River, where the prominent and freshly eroded cliffs of sedimentary rocks render the igneous intrusions fairly con- spicuous. North of the river only a few scattered outcrops have been identi- fied, but this does not necessarily indicate an occurrence limited to the outcrops since the igneous material decays readily and becomes obscured by the soil which covers most of the upland. Because of these conditions, igneous activity of greater intensity is to be inferred than the few occurrences of igneous rocks mapped would alone indicate. The lamprophyre and mira-peridotite have been described 1 by Johannsen and thin sections have also been examined by the writer. The lamprophyre is, in general, composed of conspicuous sheets of phlogopite, small crystals of pyroxene partly altered to serpentine, and fresh prisms of apatite, in a dense ground mass almost wholly altered to calcite but still preserving the rem- nants of a pattern suggesting the former presence of feldspar laths. The mica-peridotite is made up of olivine and pyroxene crystals, either partly or largely altered to serpentine, together with phlogopite, iron oxides, and obscure alteration products. The volcanic breccia is composed of angular to rounded pieces of both igneous material and sediments, together with frag- ments of feldspar, grains of quartz, and much alteration material. The separate occurrences will be briefly described, with particular refer- ence to those features which may have a bearing on the question of ore genesis. Orrs Landing Dike.— At Orrs Landing (NE. % sec. 33, T. 12 S., R. 8 E.), a small dike cuts Fredonia limestone in a N. 28° W. direction, with a dip J of 85° NE. (fig. 14). The dike is 12 to 15 inches wide. The wall-rock is affected to a distance of one-half inch by recrystallization. 1 Bain, H. Foster, The fluorspar deposits of southern Illinois: U. S. Geol. Survey Bull. 255, pp. 28-30, 1905. 237 238 GEOLOGY OF HARDIN COUNTY The dike contains many rounded, flattened, and angular pieces of lime- stone and quartzitic sandstone probably representing fragments of the under- lying sediments caught in the fluid magma and incorporated in the intrusive rock. Megascopically, the dike material is dark gray in color, and porphyritic in texture, with a dense grounclmass. The phenocrysts are of dark mica. No other minerals may be distinctly recognized, and the rock shows by its color, hardness, and porosity, the extreme alteration which it has suffered. Under the microscope, the rock is seen to be largely altered to calcite, among the crystals of which may be recognized remnants of augite, grains of magnetite, and prisms of apatite. The latter mineral appears in prominent Fig. 14. — Peridotite dike, with pebbles of different kinds of rock and calcspar, in Ste. Genevieve limestone. At Orrs Landing on Ohio River about 2 miles south- west of Elizabethtown. amount. There are no feldspar grains to be seen, which may be due to the extreme weathering of the whole mass. Probably some of the calcite and other alteration products were derived from feldspar grains. On account of the great uncertainty regarding the original constituents, the rock is given the general name of lamprophyre. The dike is partly bordered and traversed by narrow veinlets of crystal- line calcite carrying a conspicuous quantity of flaky hydrocarbon material, grains of galena and sphalerite, scattered plates of barite, and radiating clusters of aragonite fibers. Both fluorspar and barite are also found lying between the dike and calcite seam, where the latter branches into the wall IGNEOUS ROCKS 239 rock, and these two minerals in places cut across the dike. This mineral association is the same as that characteristic of the fluorspar veins, a fact which seems to signify a close connection between the intrusion of the dike material and the period of mineralization. Rosiclcare Dikes. — In the second quarry in the river bluffs east of Eosi- clare, two dikes, 8 to 10 inches in width, cut the Fredonia limestone. These dikes are about 10 feet apart and trend N. 40° W. with a dip of 80° NE. The limestone walls show no alteration. Parallel with the dikes are veins of calcite, up to 15 inches in width, composed of a very coarse calcite similar to that found in the large fluorspar veins. The material of both dikes is largely altered to calcite and chlorite ; large plates of the latter are the most prominent megascopic feature. Johannsen's microscopical description 1 states that the rock is "almost entirely altered to calcite." He reports the presence of large apatite crystals, iron oxide, chlorite, and serpentine. Fig. 15. — Block of Renault limestone injected by a peridotite sill (black layers) parallel to the bedding planes. Injected layer about 4 feet thick. At Downeys bluff, north knob, just south of Rosiclare. Downey s Bluff Sill and Dike. — Near the top of the eastern peak of the high bluff between Fairview and Eosiclare landings, a sill occurs in the beds of the Eenault limestone (fig. 15). In places it attains a thickness of 2% feet, but elsewhere divides into narrower sheets which cross the limestone beds in several places. This anastomosing and bifurcating effect is very prominent throughout an exposure traceable for nearly two hundred feet. In places the sill contains angular fragments of limestone which usually show strong metamorphic effects. The limestone bordering the igneous material shows no metamorphism other than a thin darkened band about a sixteenth of an inch in thickness, and a few scattered grains of pyrite. Under the microscope, this band shows a recrystallizing and flattening of oolites 2 , and the impregna- tion of some iron oxides. 1 See Bain; op. cit., p. 29. 2 The texture resulting from recrystallization of oolites may be called granospheritic, tUlowing the term "granospherite" as used by Weinschenck on page 187 of Petrographic Methods, Weinschenck-Clark, N. Y., 1912. 240 GEOLOGY OF HARDIN COUNTY The sill rock is aphanitic porphyrinic in texture, the visible phenocrysts being phlogopite or biotite and augite. While the mica grains are scattered throughout, there is a slight concentration of them in bands close to and paralleling the contacts. Microscopically, the greater part of the rock is made up of secondary calcite, but in this mass appears an impressively large amount of fresh apatite crystals which attain unusual sizes for fine- or medium-grained igneous rocks. Phlogopite and biotite appear as very prominent phenocrysts, and their directions of elongation are commonly parallel to the sill walls. Some of the mica grains show a zonal structure of moderately pleochroic material sur- rounding very strongly pleochroic substance. Under crossed nicols the inner zone shows the birefringence colors of biotite while the outer zone shows colors intermediate between those of biotite and muscovite. The contact between the zones is not indistinct or graded as might be the case were the zoning due to progressive alteration of biotite into another mica, but is very sharp, as though the crystals while developing in the magma had been sud- denly deprived of their iron source, changing the mica at this point from an iron-rich to an iron-poor variety. Grains of magnetite or other iron oxide minerals are very abundant. While some show the simple isometric sections so characteristic of magnetite, many others are distinctly skeletal in structure as is said to be commonly true of ilmenite. Triangular and rhombic sections are common. Augite appears as phenocrysts occasionally partly altered to horn- blende. Probably augite was abundant in the fresh rock, but if so it has since been greatly altered. No feldspar grains are to be seen; if ever present — as they may well have been — they have been completely changed. Some of the finer-textured alteration products are distributed in such a fashion as to suggest the former presence of lath-shaped feldspars, but besides this phenomenon no indication is given as to the feldspathic character o:f the unchanged mass. Quartz is present, undoubtedly of secondary origin, and probably very small patches of serpentine. A rather "dusty"-appearing gray alteration product also appears in small irregular patches, commonly asso- ciated with what is probably titaniferous magnetite or ilmenite, and is very likely leucoxene. This substance also seems to mask some small, vaguely outlined areas of brownish mineral with relatively high refraction, which may be titanite. The identification is uncertain, however. Johannsen arranges the minerals in order of decreasing prominence as follows: "mica (chiefly phlogopite), pyroxene, apatite, iron oxide, and, possibly, perofskite." The latter mineral was not seen by the writer. Pyrite grains in small quantity are disseminated through the rock. The rock is termed lamprophi/re as no closer classification seems possible. On the bluff face of the eastern slope of this hill may be found a dike trending a little west of north, and practically vertical in attitude. Its width is about 8 inches. The rock is so completely altered as to give no suggestion IGNEOUS ROCKS 241 of igneous texture. About two hundred feet north by west from this ex- posure a pile of debris, stated to have been removed from a well dug at that point, contains an abundance of fresh igneous rock fragments similar in texture and composition to the material comprising the sill. It probably rep- resents the continuation of the dike at the bluff, and may have been a "feeder" for the sill not far away. Mix Dike. — On the Mix farm about two miles north of Golconda, in NE. % sec. 18, T. 13 S., E. 7 E., a dike of mica-peridotite cuts the sandstone of the river bluff in N. 35° W. direction. The rock is much more coarsely crystalline than "either of those previously described, and has a peculiarly mottled, bronze appearance, due to the large amount of phlogopite present, and due also to its marked poikilitic texture. Besides mica, the hand speci- men shows small scattered grains of olivine, grains of pyroxene, and much white mineral of secondary origin. Microscopically, the rock contains an abundance of phlogopite, serpentin- ized olivine, titanite, and black opaque oxides, probably magnetite and ilmenite. Subordinate in amounts are unaltered olivine, pyroxene (partly altered to serpentine), and biotite. Leucoxene appears as an alteration pro- duct of the titaniferous minerals. Grains closely resembling titanite but possessing no double refraction are probably perovskite. The titanite grains show distinct double refraction. Apatite was not discovered in the section. Golconda Dike. — About three-fourths of a mile southwest of the railroad station at Golconda, in NW. 14 sec. 25, T. 13 S., E. 7 E., a dump pile of an old prospect contains fragments of a medium-grained, greenish-gray igneous rock. The hand specimen displays an even-granular texture. The color of the rock is due to secondary minerals, especially serpentine. The microscope discovers prominent amounts of phlogopite, pyroxene (diopside ?), and serpentine, the latter being secondary after pyroxene and, perhaps, olivine. No olivine grains appear. A considerable quantity of black opaque minerals is disseminated through the rock; with some of these leu- coxene is associated, showing that both ilmenite and magnetite probably make up the black grains. A few very small grains of either titanite or perovskite are found. Feldspars are entirely absent. Johannsen classifies this rock as a "pyroxene lamprophyre," apparently because of the lack of olivine. Olivine in moderate amounts may have been present originally and may now be repre- sented by some of the very abundant serpentine. In other respects, the rock is quite similar to the Mix farm peridotite, with the exception that a small amount of apatite occurs in this rock, while none was found in the Mix dike. In both rocks the mica grains have a poikilitic texture. Soward Intrusion. — On the Soward farm about two miles northwest of Rosiclare, in W. i/ 2 sec. 31, T. 12 S., E. 8 E., many large boulders of a clastic igneous rock were found. Some of the exposures are in place. Their —16 G 242 GEOLOGY OF HARDIN COUNTY distribution suggests an elongated area, possibly 100 feet in width, the long axis trending northwesterly up the hillside. No contacts are exposed. The material is composed of a highly altered fine-grained matrix carrying fragments of igneous minerals and sedimentary rocks. Under the microscope a texture characteristic of volcanic breccia of the explosive type is displayed. Large rounded sub-angular and angular fragments of sedimentary rocks are embedded in the volcanic (?) 'matrix, and many of them show metamorphic effects. The matrix is largely indeterminable, but contains angular fragments of mica, quartz, acid plagioclase (possibly of about the composition of andesine- oligoclase), microcline and, perhaps, orthoclase. Some of the mineral grains contain apatite crystals, and zircon (?). The original magma was evidently Pig. 16 — Panoramic view of the site of the "volcanic plug," half a mile south southwest of Sparks Hill, Hardin County, Illinois, in the NE. % NW. % NE. % sec. 13, T. 11 S., R. 8 E. Known area of plug indicated by line. Looking north- west. of acidic type, and this rock may represent an explosive phase of the region's igneous phenomena. Sparks Hill Intrusion, — South of Sparks Hill, in NE. % sec. 13, T. 11 S., R. 8 E., (figs. 16, 17, and 18) are scattered boulders of volcanic breccia so disposed as to suggest an elliptical area of igneous rock some 200 feet long. 1 The rock contains numerous rounded to angular fragments of sedi- mentary rocks, some of which show metamorphic effects. The ground mass is finely clastic, and under the microscope shows angular and rounded pieces of quartz, microcline, orthoclase, micas (muscovite especially), and plagioclase. Apatite crystals are found in some of the minerals. Some of the included rock fragments are apparently aphanitic igneous types. Limestone inclusions 1 The writer did not observe this igneous exposure, but received his information re- garding it, orally, from Mr. Charles Butts of the U. S. Geological Survey. IGNEOUS ROCKS 243 Fig. 17. — Exposure of volcanic rock near Sparks Hill. Fig. 18. — View of a mass of volcanic rock full of pebbles of many kinds of rock near Sparks Hill. Same exposure as figure 17. 244 GEOLOGY OF HARDIN COUNTY are frequently altered — some being recrystallized, and others partially quartzitized. The rock is apparently similar in composition and origin to the Sowarcl farm intrusion. Section 6, T. 12 S., R. 8 E., 1 'ntrusion.— In the north half of this section occurs an igneous rock of clastic character. This was not seen by the writer, but was brought to his attention by Dr. U. S. Grant, who examined it in the field, and who placed a specimen at his disposal. The hand specimen displays a clastic texture in which large included fragments of sediments are set. Zonal metamorphism has affected these. Large pieces of hornblende and biotite are conspicuous. The hornblende is somewhat rounded in shape and is present in masses up to about one inch in diameter. Much pyrite is disseminated through the larger grains. The biotite appears in irregular plates up to one-half inch across. Microscopically the rock is very much altered, mostly to calcite, which condition nearly masks the original texture. Its clastic nature can be seen, however. In the matrix appear grains of feldspar, pyroxene, and minute crystals of apatite. The latter is found in unusual abundance in some parts of the slides. The feldspar grains are usually so altered as to make exact determination of variety impossible, but a few fresh pieces showing very narrow and closely spaced albite twinning striations gave very low extinction angles, indicating an acidic variety. Some grains of the ground mass of low birifringence may be orthoclase, while a few others give faint suggestions of microcline structure. The original rock may have been intermediate in its composition between the peridotites and the acidic breccias previously described. PART V ECONOMIC GEOLOGY CHAPTER XII— ECONOMIC GEOLOGY Mineral Resources The principal mineral deposits of Hardin County are veins and beds of fluorspar. Of subordinate importance are lead, zinc, iron ore, limestone, and road metal. Soil and water while not usually classed as mineral deposits, still are essentially a mineral resource. In addition to these there is also a possi- bility of oil and gas, although the presence of these substances has not been proven. Fluorspar, Lead and Zinc By L. W. Currier 1 GENERAL STATEMENT Hardin County is noted for its deposits of fluorite, which are the largest and most important in the United States. In addition, small quantities of lead and zinc are recovered in connection with the mining of fluorite, while limestone and clay are resources of local interest. The fluorite region extends into Pope County, adjoining, and across Ohio River into Kentucky, but the major portion of the output is taken from the southwestern part of Hardin County. BRIEF HISTORY OF MINING DEVELOPMENT Although the presence of fluorspar in southern Illinois was known in the early years of the nineteenth century, no mining of the mineral was attempted until 1842. In that year Mr. William Pell discovered fluorspar and galena near the site of the -present Rosiclare mine and development of the deposit was undertaken. 2 From that time on, mining of local importance was prosecuted almost continuously in the district, but not until 1870 or thereabouts were shipments made. Between 1870 and 1905, Hardin County was a small but almost constant producer. Considerable prospecting was done during this time, but most of the mines then established, being at a distance from Ohio River and handicapped by transportation difficulties, found it impossible to compete with the large veins near Rosiclare and were consequently abandoned. 1 The field work, on which this portion of the report is based, was carried on in August and September, 1917, and in June, 1919. To the owners and managers of the various mines the writer is indebted for information and a spirit of ready cooperation. He has also enjoyed the help of Dr. Stuart Weller and Mr. Charles Butts, who have been engaged in a stratigraphic study of this region, and is indebted to Dr. J. E. Pogue, under whose direction the field work was prosecuted and this portion of the report prepared. 2 H. Foster Bain, The fluorspar deposits of southern Illinois: Bull. 255, U. S. Geol. Surv., 1905, p. 12. 247 248 GEOLOGY OF HARDIN COUNTY During the years following 1905, the Rosiclare and Fairview mines and one or two other deposits situated at some distance from the river, among them the Pierce Mines, have been the only important producers. During this period and particularly since 1907, there has been a steady and marked climb in output, with the exception of a slump in 191-1, due to the falling off in production of open-hearth steel in that year. The rise in demand and prices created by the course of the war has not only greatly stimulated the output of the large mines, but has been responsible for the reopening of some of the abandoned mines and the renewed interest in prospecting throughout the country. THE OCCURRENCE OF FLUORSPAR As described in previous chapters Hardin County is underlain by a series of horizontal or slightly inclined beds of limestone, sandstone, and shale of Chester age (Upper Mississippian) which have been extensively faulted. The major faults trend in a northeasterly direction and these are crossed by a large number of small faults running, roughly, northwesterly to westerly. The faults are, almost without exception, of the normal or gravity type (fig. 3). The crustal blocks formed by this system of intersecting faults have been variously displaced and slightly tilted. The sediments are cut in a number of places by dikes of basic igneous rock. (See Plate I and figure 19.) Some of the fault fissures are filled with mineral matter, the minerals commonly present being calcite, fluorspar, galena, sphalerite, and others in subordinate amounts. Calcite is the most widely distributed vein mineral, in many instances comprising the total amount of the vein filling. In the workable veins, the calcite has been partly or wholly replaced by fluorspar, accompanied by galena and sphalerite. In a few places, especially beneath impervious rock layers and in close proximity to fault fissures^ beds of lime- stone have been partly replaced by fluorite, with the formation of flat tabular deposits of this mineral. Many of the vein deposits do not extend to the surface, but terminate beneath a mantle of soil. In such instances, lumps of the relatively insoluble fluorspar are distributed through a zone of clayey soil, forming a superficial or so-called "grave? 7 deposit. Thus the economic deposits of fluorspar may be classified as (1) vein deposits, (2) bedding de- posits, and (3) superficial deposits. VEIN DEPOSITS The fluorspar veins constitute the most important type of deposit in the district. They owe their formation to the circulation of mineralized waters along the porous channels afforded by fissures which are in most cases dis- tinctly fault fissures or zones. Although in a few cases the veins apparently occupy simple fissures, faulted fissures of more or less displacement seem to have been the most accessible channels for the mineralizing waters and the larger veins are found only in such faulted zones. The veins conspicuous ECONOMIC GEOLOGY 249 for their size attain extreme widths of nearly forty feet, and the average vein reaches a width of five to fifteen feet. They are nearly vertical, the most im- portant running in a northerly or northeasterly direction, and lying in the limestone and sandstone members of the Chester group. Scale of Miles 1 2 3 LEGEND ® Settlements X Mines X Prospects * Igneous Rocks 1 Orrs Landing dike 2 Rosiclare quarry dikes 3 Downeys Bluff sill 4 Mix Farm dike *- 5 Golconda dike - e Clastic dike 7 Sparks Hill tuff ^-'Faults and veins 19. — Sketch, map of the Illinois fluorspar district, showing locations of mines and prospects, igneous rocks, and mineralized faults and veins. The most productive veins are near Rosiclare, on the properties of the Rosiclare Lead and Fluorspar Company and the Fairview Fluorspar and Lead Company. The main working of the former company, and the Good Hope, Annex, and Extension shafts of the latter are on the largest of these veins Fig principal 250 GEOLOGY OF HARDIN COUNTY from which the majority of the fluorspar thus far produced in the district has been taken. The vein of the Annex-Extension mine runs about north and south near these shafts, and turns gradually eastward as the Good Hope workings are approached. The drifts of the Annex-Extension mine are not continuous with those of the Good Hope, though a level of each has been recently extended to- within 150 feet of connecting. In the Good Hope mine and other mines to the northeast, the average course of the vein is nearly N. 20° E. The north- south trend of the vein through the Annex-Extension shafts therefore strongly suggested at first that they might have been located on a separate vein inter- secting the Good Hope-Rosiclare vein at a fairly sharp angle, but this sug- gestion appears less strong in view of the recent developments just mentioned above. From a third to a half mile west of the Good Hope-Rosiclare vein is the second largest vein in the district, the Blue Diggings vein. This roughly parallels the Good Hope-Rosiclare vein. A few rods west of the Blue Diggings vein is a fault which converges with the Good Hope-Rosiclare vein to the north, the two probably uniting at a point half a mile east of Stone School and two miles north of Rosiclare just south of the Elizabethtown road. The mineralization at the Daisy mine is believed to be associated with this fault and a short distance northeast of the point of intersection is an outcrop of fluorspar, seven or eight feet wide, bordered by quartzitic sandstone on the west and limestone on the east. The sandstone forms a low ridge, due to its superior resistance to erosion, as is typical of sandstone in the district where bordering a fault. 1 About an eighth of a mile beyond, along the strike of the vein, is an abandoned mine shaft, with a dump showing coarse calcite and fluorspar in a limestone country rock. About a quarter of a mile farther along, the fault is terminated by a cross-fault running X. 70° E., according to Weller. The general trend of most of the veins in the district is northeasterly. The few veins that run in a northwest direction have so far proven to bo of minor commercial importance. The larger veins are not uniform in strike, but display the long sweeping curves characteristic of the general faulting of the district (see Plate I). In dip, the veins are steeply inclined, to vertical. The Good Hope- Rosiclare vein is nearly vertical, here and there changing in inclination from steeply west to steeply east, the average being towards the west. The dip of the Blue Diggings vein is toward the east and very variable, becoming less with depth. On the 500-foot level, dips of 55° and 65° were observed. The mineral bodies of the main veins, as exposed in the workings, are thin lenses joined to one another, both vertically and horizontally, by vein 1 These ridges, due to silicification of sandstone along faults, are called quartzite reefs, by Bain. Bain, H. Fester, The fluorspar deposits of southern Illinois : TJ. S. Geol. Surv. Bull. 255, p. 33, 1905. ECONOMIC GEOLOGY 251 material, so as to form thin tabular masses with a series of pinches and swells. Whether viewed along the course of a drift, or from successive levels along a vertical line, the developed veins show variations in width from almost nothing up to forty feet. Thus in the Fairview Company's workings in the Annex-Extension mine, the 200-foot level extends from near the river north- ward for about 1,700 feet. At the south face the vein width is 7 feet; from here to the Annex shaft, a distance of 900 feet, it fluctuates considerably, nearly pinching out at the shaft for a distance of 25 feet; it then widens Fig. 20. — Diagram illustrating the effects of displacement on the shape of a fault fissure when the fault is undulatory. The arrows show directions of movement. Where the movement is normal to the plane of the fault, namely a simple stretching apart, no variation in width of the fissure results, as shown in A above. But where movement in another direc- tion is combined with the stretching movement as indicated by the two sets of arrows in B, the fissure will vary in width. In Hardin County vertical and hori- zontal displacement acting in this way is responsible for the characteristic pinch- ing and swelling, and the lenticular shape of the mineral matter filling the fissure. considerably attaining a maximum of 30 feet, but pinches again to almost nothing at the north face, 800 feet from the shaft. In the 300-foot level, which is about 1,000 feet in length, similar though less marked conditions of variations are to be seen, indicating pinching and swelling in a vertical direc- tion also. The lenticular shape of the large veins is attributed to the slightly un- dulatory character of the fault fissure and the fact that the walls underwent 252 GEOLOGY OF HARDIN COUNTY a slight horizontal movement in connection with an extensive vertical dis- placement. The resultant zone, which subsequently was filled with mineral matter, was thus given a variable width, as is diagiammatically illustrated in figure 20. Indications of slight horizontal movement are found in nearly horizontal striations of slickensides on the vein walls. Such striatums are particularly well developed in the Annex and Rosiclare mines, and were observed to dip slightly to the north in some instances, and to the south in others. Other sets with nearly vertical striations were observed, and in one place the horizontal set was superimposed on the vertical set. Although the main faulting preceded the formation of the veins, minor faulting movements developed later, involving the vein material. These move- ments were of two types, the more general one along and in the vein matter; the other, across the vein. Movements along the vein are indicated by a persistent plane, or by several nearly parallel planes, near the center of the vein matter. In places, this surface is quite smooth, and at many points shows fine striations. Quan- tities of water are admitted into the lower workings along the "center slip" (see figure 22), as the chief feature of this kind is locally called. On both sides of the "center slip," the fluorspar has been considerably darkened by the deposition of galena and sphalerite. In some places the sulphide dis- semination and minute fracturing have taken place on only one side of the slip, the spar in the other side remaining fairly clear. Cross-faulting within the vein is not pronounced, although minor frac- tures and displacements are noticeable here and there. The most conspicuous example of such movement was found near the south face of the 200-foot level of the Blue Diggings mine, where the vein has been completely cut off by a cross fault. As indicated by the disposition of the wall-rocks, the continuation of the vein may probably be picked up by drifting in a direction running, a few degrees south of west. Cross-faulting appears to have been a distinctly more common occurrence on the Blue Diggings vein than on the Good Hope- Resiclare vein. MINERALS OF THE VEINS Fluorspar. — Other names, fluorite and spar. Composition, calcium fluoride; 51.3% calcium, 48.7% fluorine. Glassy luster; easily scratched by knife blade ; perfect cleavage ; massive or in cubic crystals. Fluorspar occurs in great masses, breaking into angular cleavage pieces upon being mined, and in crystals and crystal groups. In Hardin County the bulk of the mineral is colorless to white or to faintly yellowish, but shades of purple, amber, green, and blue are also common. It is closely associated with white crystalline calcite, which it replaces. Some of the material runs in purity above 98%, although the common intimate association with galena, sphalerite, and calcite, makes it necessary to mill the product before shipment. ECONOMIC GEOLOGY 253 Calcite. — This mineral, calcium carbonate, known locally as calc, is readily recognized by its rhombic cleavage, and occurs in massive form in close association with fluorspar, as well as alone, as a vein-filling material. It is also found as small scalenohedral crystals occupying cavities in the veins. Veinlets of calcite are observed in association with some of the dikes of the region. In the fluorspar veins, it is generally white, coarse-grained, and partly replaced by fluorspar. While it occurs in all portions of the veins, it tends to be gathered near the walls, the fluorspar thus occurring between two bands of calcite, although masses and bands are also included within the fluorspar. Small veinlets of calcite, evidently of a later generation, cut the massive fluorspar, here and there, in the main veins. Quantities of calcite are rejected as waste from the mills. Galena. — This mineral, the sulphide of lead, black, heavy, with perfect cubic cleavage, is the most abundant sulphide, occurring as disseminated grains and as veinlets, in the vein fluorspar. It is nowhere in sufficient abund- ance to be mined directly, but a small quantity is recovered in connection with the milling of fluorspar, and is sold. It is usually associated with sphalerite and chalcopyrite; and in cavities in the veins, small crystals, showing the faces of the cube and octahedron, are often found. In the veins, the mineral tends to be most abundant near the center, and particularly in places where slip planes, or "center slips," are pronounced and numerous. Its general disposition is such as to suggest its introduction toward the close of the period of mineralization, after the fluorspar had been mainly deposited. Its intimate association with sphalerite and chalcopyrite suggests a formation practically simultaneous with these minerals. This galena is more notably argentiferous than the galena of other deposits of the Mississippi Valley. Silver values up to 14 ounces per ton have been reported. 1 Sphalerite. — This mineral, the sulphide of zinc, also known as "blende" or "jack," occurs in close association with galena, though in somewhat less abundance. It is found as small masses, grains, and seamlets enclosed in the fluorspar, and is rarely unaccompanied by grains of galena and chalcopyrite. In no place is it sufficiently concentrated to warrant its exploitation alone. It has been recovered thus far only by hand-picking, in connection with the milling of fluorspar, but experiments have recently been carried on looking to its recovery from the mill flows, where it appears in the table middlings. Since its specific gravity is only slightly greater than that of fluorspar, neither table nor jig concentration is able to make a clean separation of the two. This separation, may, perhaps, be effected by flotation methods, in which event a quantity of otherwise useless middlings will yield a by-product of value. In small cavities in the veins, minute crystals of sphalerite are found showing intricate crystal forms. 1 See Bain, op. cit., p. 38. 254 GEOLOGY OF HABDIN COUNTY Chalcopyrite. — The yellow sulphide of copper and iron is present in sub- ordinate amounts as a common associate of galena and sphalerite. It is found as small grains, both alone in the fluorite and amongst the other sulphides. Its quantity is nowhere sufficient to lend it commercial importance. Hydrocarbon. — A conspicuous quantity of black, flaky, bituminous ma- terial, having an odor resembling that of petroleum, is found here and there in the veins, in cavities, in the calcite, and in the fluorspar. A similar material occurs in the calcite veins associated with the dike at Orrs Landing. Other Minerals. — Other minerals seen or reported in connection with the veins are barite, pyrite, malachite, and stibnite. Barite occurs in small amounts associated with the fluorspar and an occasional, well-formed crystal may be found in cavities in the veins. Pyrite is found as small veins in- conspicuously disseminated in the wall rocks and vein matter. Malachite is noted as a staining in a few places in the veins, and represents a local altera- tion of the chalcopyrite. Stibnite 1 , the sulphide of antimony, has been reported to be present in the Fairview mine, but its occurrence was not verified by the present writer. BEDDING DEPOSITS In places fluorspar occurs in flat-lying tabular masses, forming part of the series of sedimentary rocks. Such occurrences are not so abundant as the veins, and owing to the thinness and impurities of these bodies, they have been subjected to fitful rather than continuous exploitation. Occurrences of this type may be termed bedding deposits. Locally they are called "blanket" formations. These deposits are replacement bodies along the stratification planes of limestone beds. They are found in places where a porous layer of limestone, under a relatively impervious bed, was made accessible to ore-bearing solu- tions through an intersecting fissure or fault. Because of this relationship, they may be expected to be relatively long and narrow. One of these deposits, developed in the west side of Lead Hill apparently continues through the hill, outcropping on the east side at a horizontal distance of about 1,000 feet from the entrance workings. Some of the smaller vein deposits are associated with this type. The most characteristic feature of the ore occurrence in the bedding type of deposit is a conspicuous horizontal banding, which is due to the alternation of layers of pure crystalline fluorspar with layers of finer-grained, sandy- looking and less pure fluorspar, or partially replaced limestone, or crystalline quartz. Layers of quartz are particularly detrimental, because where they form a prominent part of the deposit its successful exploitation is prevented. The bands of the bedding deposits are rather evenly spaced and fairly uniform in thickness in any one deposit. The bands are not continuous through long 1 See Bain, op. cit., p. 39. ECONOMIC GEOLOGY 255 distances, but thin out or widen and bifurcate, finally being terminated abruptly by some minor structural feature, such as a joint crack filled with clay, a thin bed of clay, or a silicified rock ; or else finally thinning grad- ually into disappearance within the limestone bed. The layers are usually V-2 to % of an inch thick, though bands of greater thickness are often found. The clear fluorspar layers are of high purity. The mineral is generally tinted with shades of red, purple, or blue, but in some places much of it is practically colorless. The layers which alternate with the clear fluorspar are more com- monly brownish-gray, porous, and granular, representing in some places iron- stained and silicified limestone only partly replaced by fluorspar; in other cases they are composed almost entirely of "spar/' In a few instances, as mentioned above, the impure granular layers are entirely lacking, their place being occupied by clear and well-crystallized quartz. Galena is present as small grains in the impure bands, and as cubes within and bordering the pure bands. Sphalerite is also present but in much less amount than is the galena. In cavities where crystals of fluorite have formed, the crystal faces are sometimes found with a coating of yellow zinc carbonate, locally known as "turkey fat" ore. In places the galena shows marked alteration to cerussite, the carbonate of lead. The Lead Hill deposit occupies a zone directly beneath a layer of dense limestone. Near the entrance to the northern of the two tunnels running into the hill, the horizontal mass of fluorite is seen to pass downward, funnel- shaped, into a narrow vein, which evidently represents the channel followed by the ore-bearing solutions. The minerals found in the bedding deposits are fluorspar, quartz, galena, smithsonite, and cerussite. Calcite is noticeably lacking except for a few small crystals in cavities. As compared with the amounts found in veins, galena is very unimportant except in a few places — as in parts of the Lead Hill deposit where large cubes are obtainable in sufficient supply to be separated, and at the Miller mine near Melcher Hills, where small veins merge into a bedding deposit. While sphalerite appears to be rare, zinc car- bonate is often found as described above. It is never sufficiently abundant, however, to be of primary economic value. As' contrasted with the massive fluorspar of the vein type deposits, the fluorspar of the bedding deposits is crystalline and transparent. Most of it is colored or tinted as noted previously, and many beautiful specimens of the crystals which would make excellent museum material may be found readily. Such localities give considerable promise as sources of optical spar. The bedding deposits have not been developed extensively, partly because of the lack of promise of continuity for any distance, and probably also because of the difficulty in separating the higher grade fluorspar from the interbanded material of poorer quality. Where the interbands are of quartz, moreover, the deposit is likely to have no industrial value. Also the general 256 GEOLOGY OF HARDIN COUNTY thinness of the deposits would be apt to make mining unprofitable because development similar to the room-and-pillar method of mining would be necessary. SUPERFICIAL DEPOSITS A small part of the fluorspar of the district is won from surface pits and trenches, the deposits thus mined being known locally as "gravel" deposits, because they consist of fragments of fluorspar scattered through a clayey or sandy soil. The material is commonly dug out by pick and shovel and washed in log-washers or in concrete mixers, the product going by the name of "gravel spar." The product is, however, distinctly different from the gravel spar recovered in the Eosiclare and Fairview mills, and the two should not be confused. The matrix of the fluorspar in these deposits is usually the red clay that results from the weathering of the limestone bed-rock, by which process the major portion of tbe limestone, consisting of calcium carbonate, has been dissolved and removed by surface waters, leaving the relatively insoluble clay and fluorspar behind as deposits in the soil mantle. In this way solid veins of fluorspar are converted into zones of weathered material carrying a con- spicuous quantity of fluorspar fragments. The distribution of the surface workings of many of the superficial deposits indicates the position of well-defined veins below. The size of these veins, however, cannot always be judged from the richness of the surface workings, since a given gravel deposit may represent a considerable concen- tration of fluorspar from the extensive erosion of a small vein. Many of the superficial deposits yield a very high-grade product, the processes of formation having served to remove the impurities usually present in the veins. Moreover, this type of deposit may be worked at small initial and running expense, and is therefore adapted to exploitation with little capital. For these reasons, the gravel deposits may be expected to compete in a small way with the larger veins requiring more extensive mining equip- ment. The Pierce mines, in Pope County, about ten miles north of Golconda, are interesting and characteristic examples of this type of deposit. The open- pit development (fig. 21) points to the existence of two veins, one running about N. 75° E., the other about N. 45° E. The deposits have been opened, for the most part, only at the surface, and show the usual dissemination of fluorspar in red clay and in sandy clay. 'Near the bottom in some parts of the pits the spar appears less disseminated and more vein-like in structure. Occasional rounded boulders of limestone are encountered in excavating, some of which are cut by narrow stringers of calcite and fluorspar. Coarse calcite remnants together with fluorspar are also found in these boulders, the asso- ciation pattern resembling that found in the Fairview and Eosiclare veins — that is, the replacement of calcite by fluorspar. The fluorspar is remarkably ECONOMIC GEOLOGY 257 clear and pure, nearly transparent, tinted in shades of red, blue, and purple, thus differing from the fluorspar of the Eosiclare and Fairview mines which is translucent to opaque and mostly white, yellow, or brown. Large cleavage pieces nearly a foot in diameter were seen by the writer. Optical fluorspar is said to have been obtained from the Pierce deposits, which indeed show indications of being a very probable. source of such spar. . MILL k SCALE 100 i 200 i 3 J° FEET Fig. 21. — Map showing the distribution of the principal open-cuts of the Pierce Mines and the apparent directions of the veins. PARAGENESIS OF MINERALS Underground observations and microscopic examinations of polished surfaces of specimens from various parts of the veins, particularly from the Annex, Blue Diggings, and Eosiclare mines, disclose the following relations regarding the order of mineral deposition in the veins. 1. Calcite appears to have been the first mineral deposited in the fissures 1 . It is generally of very coarse texture, milky white, and is present throughout the workings. Very few parts of the veins lack this mineral, 1 It is not at all probable, however, that the calcite was deposited from the same waters, or even waters of similar origin, that deposited the flourspar. -17 G 258 GEOLOGY OF HARDIN COUNTY though the amount to be found is extremely variable. It is found especially along both walls of the veins, with fluorspar occupying the central zone. 2. Fluorspar is the second mineral of deposition, unmistakably replacing calcite. Masses and rhombs of calcite are found in great abundance showing every stage in the process of replacement. 3. The sulphides of lead, zinc, and copper — galena, sphalerite, and chalcopyrite, respectively — were probably deposited nearly simultaneously after the fluorspar, although some initial deposition may have accompanied the formation of the fluorspar, since in many places they are found disseminated through, and entirely surrounded by, the spar. In general, however, they were deposited along fractures in the fluorspar, being introduced along the "slip" planes; and in all such instances they are clearly subsequent to the fluorspar. Fig. 22. — Idealized section of a fluorspar vein, showing general relations be- tween wall rock, minerals, and "center slips." The white areas are calcite, the stippled areas fluorspar, and the blackened areas sulphides. No well-marked stages in the deposition of the individual sulphides are apparent. While some specimens seem to indicate that sphalerite preceded galena, this observation is reversed upon examination of other specimens. In most examples, the intergrowth of the two is such as to imply simultaneous deposition. In some parts of the veins only one of these two minerals occurs, while in other parts both are prominent. The latter condition renders diffi- cult the problem of recovering both sulphides as by-products in the milling operations. 4. Second and later generations of both calcite and fluorspar are present, but in very minor quantities, being developed especially in open cavities of the veins and walls. Quartz is also found in association with these later generations. No indications of the secondary enrichment of the sulphide minerals were seen. ECONOMIC GEOLOGY 259 Figure 22 is an idealized section of the large veins showing the mineral relations. In the bedding deposits the same order of deposition as outlined above seems to have been followed, except that the first generation of calcite is absent. The carbonates of zinc and lead — smithsonite and cerussite — represent the latest deposition, being the result of oxidizing waters of the meteoric circulation. Sphalerite is never prominent in the bedding deposits, but galena is in some places abundant and occurs as well-formed crystals. METASOMATIC RELATIONS OF MINERALS The accompanying photomicrographs (Plate II) of specimens obtained at the Eosiclare and Fairview mines illustrate some of the replacement features. A study of many specimens discloses the same general paragenetic features for all. The fluorite is seen everywhere to replace the original vein calcite or the wall-rock material itself where the latter is predominantly calcareous. No instance was seen of any replacement of sandstone or quartzite wall-rock by fluorite with the exception of cases where the rock appears to be a silicified limestone. '.Whether or not such silicification took place partly or entirely subsequent to the deposition of fluorite is. not evident. Quartz appears prominently in some of the specimens. The major por- tion of it is certainly subsequent to the fluorite, but there are suggestions of contemporaneity of the two in some places. It is quite likely that the precipi- tation of silica began before the deposition of fluorite was completed. Quartz also appears to be subsequent to the sulphides, suggesting a long range of deposition as compared with the other minerals. The quartz is glassy, clear, and crystalline, and not chalcedonic or very fine-grained as is so generally the case with many shallow-depth veins. The sulphides are probably of nearly- contemporaneous deposition, and are closely related to fractures in the calcite and fluorite, replacing both of these minerals. The galena not infrequently appears with strongly idiomorphic outlines. It is often difficult under the microscope, if not impossible in some cases, to determine exact limits between fluorite and calcite. This is partly due to their similarity in appearance, though the larger masses may be easily recognized by their cleavages and the marked dispersion of colors of the fractured fluorite. A very strong and intricate intergrowth with gradual shading from one mineral to the other is common. The following tabulation gives the replacement relationships between the minerals of the veins. In the bedding deposits, only those marked (*) apply. 260 GEOLOGY OF HARDIN COUNTY Calcite replacing limestone *Fluorite replacing limestone Fluorite replacing calcite Sulphides replacing limestone Sulphides replacing calcite Sulphides replacing fluorite * Quartz replacing limestone Quartz replacing calcite The very generalized time relationships between the minerals are dia- grammatically expressed in the lower figure of Plate II-G, where the abscissae measure paragenetic time, and the vertical dimensions of the individual mineral plots indicate relative amounts, not computed, of the respective minerals. DESCRIPTION OF PHOTOMICROGRAPHS AND FIGURES ILLUSTRATING PARAGEJNESIS Some of the most decisive replacement patterns are shown in the thin sections. The features shown by the best one of the sections studied are worthy of brief description. In thin section, fluorite is seen to penetrate calcite irregularly as veinlets which sometimes terminate with idiomorphic outlines. Slight projections of the fluorite along calcite crystallographic directions may be observed, as though the fluorite was controlled by calcite twinning to a small degree, though more commonly the fluorite does not show this directional replacement but rather enlarges irregular calcite boundaries. Galena displays the finest replacement pattern observed, where it is developed in a lattice-like pattern following very definitely the twinning directions of calcite. A photomicrograph of this is shown. Elsewhere galena shows idio- morphic outlines where it is in contact with sphalerite on one side and with calcite on the other. Fluorite appears entirely enclosed by galena, especially along a fracture between fluorspar and calcite grains. In some places galena also completely surrounds small patches of sphalerite. Illinois State Geological Sukvey Bulletin No. 41, Plate II-A A. Quartz (Q), chalcopyrite (Ch), and galena (G), replacing calcite (Ca). No definite order among the sulphides is suggested. The galena is seen to be idio- morphic. The same specimen in other portions shows replacement of calcite by fluorite and of fluorite by sulphides. The transition from fluorite to calcite, while sometimes sharp, generally is indistinct or gradual. Elsewhere, also, the specimen suggests that some quartz deposition accompanied the fluorite; much was closely following and preceded the sulphides, and some may have followed the sulphides. The black spots in the key — corresponding to numerous very small white areas in the photomicrograph — represent blende. Quartz (Q), sphalerite (blende) (B) and calcite (Ca), Magnification 150 diameters. Illinois State Geological Survey Bulletin No. 41, Plate II-C A. Veinlet of fluorite (P) containing later sulphides in silicified limestone (L). A transition zone into wall-rock material is indicated. The rock contains also many specks, apparently of fluorite. B. Section in a small veinlet cutting an argillaceous limestone wall-rock, showing relationships between wall-rock material (W), fluorite (F), galena (G), and chalcopyrite (Ch). The wall-rock contact with fluorite is very irregular and vague, and the fluorite mass contains many minute fragments besides the larger remnants indicated. Illinois State Geological Survey Bulletin No. 41, Plate II-E A. Sphalerite (S) here replaces calcite (Ca) along twinning directions and fluorite (F) cleavage directions. The metasomatic relations between these are considered very strongly indicated. Also to be noted are the various included grains, the idiomorphic quartz (Q) section showing nearly included sphalerite (S), though the inclusion is close enough to the larger sphalerite mass to be a possible tongue of the latter. On the extreme right, the line of contact between calcite and fluorite is very irregular and can not be followed in the photograph, though it appears distinctly outlined in the section. The fluorite areas are portions of a zone filling and replacing along a fracture zone in coarse-grained calcite, in which are also found several idio- morphic quartz grains. Magnification 250 diameters. B. Reduced scale of figure A, showing its relation to larger mass of sphalerite and to fluorite veinlet. Magnification 40 diameters. Illinois State Geological Survey Bulletin No. 41, Plate II-G A. The principal feature of this figure is replacement of calcite (Ca) by quartz (Q) along twinning directions of the former. Galena (G) and sphalerite (S) also appear replacing calcite in the common manner — the galena with partial idiomorphic outlines in places, the sphalerite in irregular masses extending short distances along calcite twinning striations. In this figure quartz includes both sulphides. Fluorite (F) here replaces calcite. In adjacent areas— just out of this field— fluorite is to be seen plainly replaced by the sulphides. Magnification 50 diameters. Calcite Fluorite Chalcopyrite Pyrite Sphalerite Galena Quartz B. Shows the very generalized time relationships between the minerals, as worked out from a study of thin sections and polished surfaces. The abscissae measure paragenetic time, and the vertical dimensions of the individual mineral plots indicate relative amounts, not computed, of the respective minerals. —18 G ECONOMIC GEOLOGY 275 ORIGIN OF THE DEPOSITS VIEWS OF PREVIOUS WRITERS Lead and zinc deposits occur in a number of localities in the Mississippi Valley, notably in southwest Wisconsin and in Missouri, where the weight of evidence is undeniably in favor of an origin from meteoric waters circulating under artesian conditions and leaching their metallic contents from a wide area of sediments. 1 The lead of these regions is notably lacking in silver content. Fluorspar is lacking. The deposits are not of the fissure vein type, but rather are gash veins and bedding replacements in definite horizons. Finally, faulting and igneous rocks are absent, at least as conspicuous structural features. The Hardin County region, however, is seen to present contrasting features throughout, as compared with the Missouri and Wisconsin regions. In the former, faulting is pronounced and the fault fissures contain the most concentrated values in the form of fissure veins. The most abundant mineral is fluorite, a substance known to have a strong igneous affiliation, and the galena is argentiferous, which is commonly true of lead deposits that are formed by deposition from thermal solutions ascending from below and having more or less direct igneous connections. 'Finally, the presence of igneous rocks in the district gives it petrologic distinction. The points of contrast between these two types of lead-zinc regions are of high significance, indicating with little or no uncertainty the origin of these fluorspar-lead-zinc deposits by thermal ascending solutions as opposed to meteoric waters. 2 Emmons, who examined the Illinois deposits in 1892 3 , attributed their formation to material leached from the surrounding limestone. It is difficult to conceive, in this connection, that circulating ground waters obtaining a load in this manner would have selected certain fractures for deposition of minerals leaving many others entirely unaffected. Such water would be of the carbonate type and would dissolve calcium carbonate as well as fluorite, with the probability that low-temperature deposition of both minerals would take place simultaneously when the waters issued into fissures. E. 0. Ulrich and W. S. Tangier Smith reported on the fluorspar deposits of western Kentucky in 1905, and remarked on the general similarity of these with the southern Illinois veins. 4 Smith herein expresses belief in the theory that the fluorspar and ore minerals were leached from deep-lying limestones by rising thermal solutions and that they were deposited in the upper zones 1 For the origin of the zinc-lead deposits of the Joplin region see the clear and con- vincing exposition by C. E. Siebenthal in U. S. Geol. Survey Bull. 606, 1915 ; on the southwest Wisconsin area, by U. S. Grant, Wisconsin Geol. and Nat. Hist. Surv., Bull. 14, 1906, in which he states the meteoric origin and its relation to structural basins having an impervious stratum along which seeping and circulating waters were deflected and directed. 2 Editor's note. Attention is called to a paper describing a Wisconsin occurrence of nuorite in association with calcite, pyrite, and galena, which apparently is related neither to taulting nor to igneous activity. Bagg, Rufus Mather, Fluorspar in the Ordovician iimestone of Wisconsin: Bull, of the Geol. Soc. of America, Vol. 29, pp. 393-398, 1918. 91 10 ™ mo ™' S - F " Fluors Par deposits of southern Illinois: Am. Inst. Min. Eng., Vol. *1, la\)Z, p. 52. 4 Ulrich, E. O., and Tangier Smith, W. S., The lead, zinc, and fluorspar deposits of western Kentucky: Prof. Paper 36, U. S. Geol. Surv.. 1905, p. 15 and p. 150. 276 GEOLOGY OF HARDIN COUNTY of fissures where decreases of temperature and of pressure caused precipita- tion. To quote from pages 150 and 151 of Smith's paper, "The intimate association of the fluorite with other minerals known to have been deposited from solution, the fact that it is known to be somewhat soluble both in pure water and in ordinary surface and underground waters, and to have been actually deposited from such waters, and the fact that the fluorite in the veins is generally sharply limited by the fissure walls, which, except for the effects of recent surface waters, show no corrosion or other action such as might be caused by fluorine gas or molten fluorite, would seem to render it unneces- sary to consider further any other mode of deposition than that from circu- lating underground waters." Regarding these points, the writer feels that he can not remark on the correctness of Smith's statements as applied to the Kentucky deposits, but he can not feel that the Illinois deposits substantiate the ideas quoted above. In the first place, Smith's first point "* * * asso- ciation with other minerals known to have been deposited from solution," has no significance. He does not state that these particular associated minerals are always precipitated from one type of solution. As a matter of fact, the '•'other minerals" have been known to be deposited from solutions of several types and origins. Commenting on the next point — "that it (fluorite) is known to be somewhat soluble both in pure water and in ordinary surface waters and to have been actually deposited from such waters," the writer believes in the truth of these statements, but is unable to agree that this point also "would seem to render it unnecessary to consider further any other mode of deposition than that from circulating underground waters." The thing to be accounted for is not the presence of fluorine — such as might admittedly be precipitated from cold meteoric waters — but the unusual concentration of the substance. For further comment on Smith's statements when applied to the Illinois veins, instead of a lack of corrosive or other strong action being suggested, the writer considers the very strong replacement relationships of the veins and the complexity of boundaries — sometimes almost impossible to define — between the fluorite and the calcite or wall rocks as being of precisely opposite significance, namely, that strong action most certainly is suggested by the textures and structures of the veins. Smith gives several analyses to show that limestones of the region contain an appreciable amount of fluorine, the highest analysis quoted showing 0.10 per cent fluorine. This fluorine may, however, have been subsequently introduced into the limestone as a result of circulating waters obtaining their content from the deposits, or from the same source as the veins. Finally, if circulating ground waters concentrated the fluorspar, why were not more of the more superficial fissures given an appreci- able amount of fluorite ? After a detailed study of the Illinois fluorspar regions in 1904, Bain 1 believed "that the evidence points to heated waters having been the agency 1 Bain, H. Foster, The fluorspar deposits of southern Illinois : U. S. Geol. Surv. Bull. 255, 1905. ECONOMIC GEOLOGY 277 by which the ores were segregated and that they obtained an essential portion of their load from a larger mass of lower-lying intruded rock of which the dikes are now offshoots." As to whether the depositing waters were of magmatic origin, or whether they were meteoric and derived their fluorine content from magmatic emanations at depth, Bain does not make a final decision, but he does not favor the view that they dissolved disseminated fluorite from the limestones. As he states, "* * * it may be pointed out that the normal tendency of heated waters rising along faulting fissures would undoubtedly be toward deposition and dissemination of material rather than toward its segregation.*' This would be especially true if the mineral content of these waters was distinctly less soluble under given conditions than other mineral substances that would be encountered along the line of travel, which is in reality the relationship between fluorite on the one hand and cal- cite on the other. Finally, Bain remarks, "The association of the minerals and the common phenomena of marked silicification of the hanging wall are interpreted as indicating deposition from heated ascending solutions carrying fluosilicates of zinc, lead, copper, iron, barium, and calcium. These are be- lieved to have been broken up and precipitated by descending cold waters, which possibly also furnished the sulphur to combine with the metals, though it is not improbable that sulphur was an original constituent of the rising solutions." In a later study of the Kentucky deposits, Fobs 1 assigns the deposits to the ^'thermo-aqueous'' genetic class, and considers the fluorspar as being derived from silicofluorides of various metals which reacted with calcium bicarbonate to form CaF 2 . Such reaction would also cause the formation of silica, barium sulphate, and metallic sulphides, the sulphur coming from postulated hydrogen sulphide. In this case, however, we might expect to find evidences of contemporaneity among all the vein minerals rather than the definite sequence which has been previously described in this paper. Fohs further speaks of the lack of sufficient fluorine in the wall rocks to form CaF 2 , and suggests a parent magma — of which the dikes are apophyses — as the likely source of the fluorine. This element was "brought up in thermal solutions as an after effect of the eruption." Thus Fohs ascribes a direct- igneous origin for the fluorite and metallic sulphides. Siebenthal 2 alludes to the Kentucky-Illinois district in connection with his study of zinc-lead areas, and calls attention to structural features favoring an origin similar to that imputed to the Joplin ores, apparently seeing no significance in the many localized faults, the igneous rocks, the great develop- ment of fluorite, and the argentiferous quality of the galena. These four 1 Fohs, F. J., Fluorspar deposits of Kentucky: Kentucky Geological Survey Bull. 9, pp. 61-63. 1907. The fluorspar, lead, and zinc deposits of western Kentucky: Econ. Geol.. Vol. V. pp. 377-386. 1910. 2 Siebenthal, C. E., Origin of the zinc and lead deposits of the Joplin region, Missouri, Kansas and Oklahoma: U. S. Geol. Survey Bull. 606, pp. 213-217. 1915. 278 GEOLOGY OF HARDIN COUNTY characteristics are distinctly the reverse of conditions in the Missouri lead- zinc districts which comprise the subject of Siebenthal's paper. Moreover, he apparently regards as mere coincidences the fault-fissure fluorite veins of Durham and Derbyshire, England, which are said to have mineral associations and structures similar to the Illinois veins, and which are, according to Bain, situated in limestones of about the same age, in which very similar basic dikes are present. By no means is this evidence conclusive regarding any mode of origin for the Illinois veins, but it should be considered as suggestive. The veins of fluorspar near Freiburg (according to Bain) occur in gneisses, schists, and slates, but these rocks are likewise traversed by dikes of gabbro besides some dikes of quartz porphyry, and the galena of the veins is argenti- ferous. Commenting on Siebenthal's statement of similarity in the structure of the Joplin and the Kentucky-Illinois regions, it seems to the writer more significant to point out the great dissemilarities in structure of the ore bodies, as well as the extreme faulting of the Illinois region which is certainly not true of the Joplin area. Thus the two regions present great contrasts in some of those geologic phenomena which are among the most important to consider in the solution of an ore genesis problem. SUMMARY OF GENETIC EVIDENCE The evidence bearing upon the genesis of the Hardin County deposits appears throughout the descriptive matter of this paper. It may be sum- marized as follows : 1. The enormous mass and notable concentration of fluorite, a mineral carrying 48.9 per cent of fluorine (when pure), which is a normal product of igneous activity, and in concentrated condition especially the result of mag- matic differentiation. 2. The presence of basic igneous dikes and sills, and of fragmental acidic igneous material which may be explosive products of vulcanism. These opposite chemical types of igneous rock in the same province make possible the postulation of magmatic differentiation at depth, and indicate that the region was one of more or less pronounced igneous activity, that is, it overlies a mass of igneous rock of which the dikes, etc., are offshoots. The proximity of a large igneous intrusion is also suggested by a slight doming of the strata and the complex pattern of mosaic block faulting. 3. The presence in the igneous rock of conspicuous quantities of apatite, phlogopite, and biotite, the first two of which especially are usually fluorine- bearing minerals. These facts indicate that the parent magma held a fluorine content and the point is further sustained by the analysis of the Mix dike near Golconda which shows 0.13 per cent of fluorine. 1 In connection with this investigation, no analyses were made, however. 1 Analysis by George Steiger, U. S. Geol. Surv., quoted from Bain's report, op. cit. ECONOMIC GEOLOGY 279 4. The presence of extensive faults, some of which are evidently deep- reaching, within the largest of which the bulk of the fluorspar occurs and in possible connection with which the fluorspar is almost exclusively found. 5. The absence of fluorite from many faults. This militates strongly against an origin as a result of laterally flowing solutions. 6. Association with sphalerite, argentiferous galena, pvrite, chalco- pyrite, and possibly stibnite, introduced after the chief period of fluorspar deposition, and suggesting a stage in an active mineralizing period. A suc- cession of mineralizing stages and this general association of minerals seem to be very common phenomena of ore deposits formed by solutions which have direct connections with igneous activity. 7. The conspicuous and unmistakable replacement of calcite by fluor- spar in amount and manner suggesting an intense period of mineralizing action. 8. The occurrence of fluorspar-calcite veins within and bordering the igneous dike at Orrs Landing. 9. The relative insolubility of fluorspar as contrasted with calcium carbonate, rendering difficult the gathering of disseminated fluorspar from widespread areas through the agency of circulating ground waters. These points are thought to weigh heavily in favor of an igneous origin as opposed to a concentration by waters from the surrounding sediments. Thus he is in accord, in general, with Fohs' idea of origin, though nowhere in the literature studied does Fohs seem to marshal sufficient evidences to warrant his conclusion. The evidences submitted above are those of the writer's observations and seem to him far more greatly to substantiate the theory that thermo-aqueous solutions which emanated directly from a large igneous intrusion at depths ascended along deep-reaching fissures — probably developed mostly by igneous disturbance — and deposited their solutes in upper zones where decrease of temperature and pressure (aided by the chemical character of the rocks) caused precipitation. CHEMICAL CONSIDERATIONS As to the chemical condition of the fluorine brought up by the ascend- ing waters, the writer feels that sufficient evidence is not at hand to warrant a final decision. Fluorine gas may have been the chief solute. In case, how- ever, silica was also a solute — as it often is in magmatic waters — silicon fluoride would probably have formed according to the reactions : 2F 2 -f2H 2 0=4HF+0 2 (1) Si0 2 +4HF=SiF 4 +2H 2 (2) and the silicon fluoride reacting with CaCO :; ; SiF 4 +2CaC0 3 =Si0 2 +2C0 2 +2CaF 2 (3) 280 GEOLOGY OF HARDIN COUNTY In the absence of silica, the hydrogen fluoride (formed according to equation 1 ) would react directly with calcium carbonate : 2HF+CaC0 8 =CaF 2 +H 2 0+C0 2 (4) The calcium carbonate would be encountered (as limestone) along the path of the ascending waters. If reaction (3) had taken place we should expect to find a contemporaneous deposition of quartz and fluorite. The veins, how- ever, contain very large aggregations of pure fluorite, instead of an inter- growth of that mineral with quartz. Some parts of the bedding deposits (relatively of very subordinate amount) might support this condition. If the magma had differentiated at depth into basic and acidic portions, the fluorine becoming concentrated in the acidic split, magmatic emanations would very likely have carried silica. Differentiation of the magma is sug- gested by the two types of igneous phenomena in the district, namely, the basic dikes and the acidic volcanic breccias, and also by the veinlet of fluorspar which cuts the dike at Orrs Landing. The fluorspar here is plainly subsequent to the igneous rock. The emanation of fluorine from a basic magma would not have been accompanied by much silica, unless the fluorine itself were responsible for an impoverishment of the original magma with respect to silica content. In such a case, we might expect very unusual types of basic rocks to result by subsequent magmatic intrusion. The igneous rocks of Hardin County are indeed somewhat extraordinary in the marked development of apatite and the micas. The latter, especially, occur more commonly in normal acidic or intermediate rocks than in such extremely basic types. In this connection should be considered Fohs' statement that the Kentucky dikes average only 34 per cent Si0 2 , an amount which is sufficient to form only orthosilicate minerals. If, on the other hand, the original magma never had other than a relatively basic composition, fluorine or hydrogen fluoride emanating there- from would react directly with calcium carbonate along the path and wonld have formed fluorite without any contemporaneous deposition of quartz. This condition would be given plausibility if one could prove that the basic dikes and the acidic breccias are not directly consanguineous in origin. Field data neither prove nor disprove this. Another suggestion is that calcium fluoride was a constituent of the magma and emanated as such. This mineral does not appear in any of the igneous rock sections studied, but the rocks are all considerably altered to calcite and thus small amounts of calcium fluoride might have been com- pletely changed in this direction. Fluorite is said to be altered to calcite by the action of percolating bicarbonate waters. 1 Finally, Fohs and others mention the possibility that the ascending waters carried fiuosilicates of the various metals, especially of iron, lead, zinc, 1 F. W. Clark, Data of geochemistry, U. S. Geol. Surv. Bull. 616, 1916, p. 336. ECONOMIC GEOLOGY 281 and copper, and that fluorite was deposited as the result of interaction between these compounds and calcite. This does not seem to be supported to any appreciable extent by the paragenetic relationships between the vein minerals. If such compounds were the source of fluorine in the veins we should expect much contemporaneity of fluorite and sulphides to be indicated, especially in the direction of much dissemination of sulphides of the metals through- out the fluorite masses. We may very likely thus account for the small amounts of disseminated metallic sulphide that we And, but not for the larger sulphide masses. A detailed study of the chemistry of fluoriferous solutions was not possible in connection with the preparation of this paper, but the writer believes that the geochemical problems involved in the genesis of the fluorite veins are worthy of special investigations. GEOLOGIC HISTOEY OF THE DEPOSITS The evidence given throughout the paper points to a sequence of events roughly as follows : The region was subjected, in post-Pottsville, probably post-Carboniferous times, to a period of upbowing followed by normal faulting. The localized geographic situation and character of these deformations suggest a deep- lying adjustment of a magma, the presence of which is manifested in the occurrence of several igneous rock exposures. During the following period of major faulting, calcite was developed as a fissure filling either by direct precipitation from circulating waters, or by some recrystallization of lime- stone fragments caught in the grinding movement along the fault planes. In the latter case, the fault zones loosely filled with calcite would afford easy access to solutions containing fluorine and favorable conditions and material for the deposition of their load. In the former case, the calcite-filled veins would in many cases prove to be the relatively weak zones along which subse- quent Assuring would more easily take place during the later stages of fault- ing, and so also would provide favorable paths for ascending solutions. The fluoriferous thermal solutions, leaving the magma during its differentiation and cooling, followed upward along the deeper-reaching fault fissures, and upon cooling and decrease of pressure attained in the upper sediments, began a process of mineral deposition. Thus great bodies of fluorspar were formed, replacing the calcite of the veins or the limestone of the walls. In the later course of the deposition, though scarcely separated from the earlier by time interval, the uprising solutions changed somewhat in composition, bringing material for the formation of metallic sulphides which were thus introduced into veins already largely filled with fluorspar, and especially along shear zones in the fluorspar mass. This was the terminating main feature of the mineralization period which was probably relatively short and intense. Sub- 282 GEOLOGY OF HARDIN COUNTY sequently, prolonged weathering and erosion, and minor faulting, brought the surface down to the zone of vein formation, thus exposing the' deposits to discovery and exploitation. MINING AND MILLING METHODS MINING Veins. — In exploiting a vein the shaft is generally sunk in the hanging wall if the vein hades appreciably, and in the foot-wall if the vein is prac- tically vertical. Levels are established usually at 100 foot intervals, though in the earlier days the intervals were smaller. Cross-cuts are driven from the shafts to the vein, and drifts are then driven on the vein in both directions. This is the method of opening employed in all the recent developments, but in the case of the old Good Hope shaft, sinking followed the vein matter as a steep incline. Overhand stoping methods are usually employed, except in a few cases where exceptionally rich vein matter makes it desirable to remove material at the drift floor by underhand stoping. Overhand stopes are worked within 15 to 25 feet of the upper level, leaving this amount to supply supporting pillars and tunnel floors. In a few instances all material has been removed between levels for a short distance, necessitating a bridging of the upper drift. The walls are generally very firm and require little timbering. In some places, however, where the faulting has resulted in a wide fracture zone instead of a single break, some difficulty is encountered. One of the greatest mining difficulties met with is in dealing with so- called "mud-runs/ 7 Large solution cavities in the limestone walls, especially near the upper levels, become filled with mud by infiltration. Where these adjoin the veins, they are apt to be tapped in the process of developing the stopes or drifts, with the result that great quantities of a thin clay mud may flow out with disconcerting results. Once a reservoir is thus tapped, nothing may be done but to allow it to drain and to remove the slime at great expense. Probably only at prohibitive expense could these mud reservoirs be located in advance and avoided. Superficial deposits. — The superficial or "gravel spar" deposits are mined by open-cut methods to depths usually less than twenty feet, requiring merely the loosening and removal of the materials by pick and shovel, and washing to remove the adhering clay. This simple form of mining is obviously rela- tively inexpensive, but is applicable only to shallow depths. It may be economically applied to prospects in proving the existence of a vein. The Pierce mines north of Golconda probably afford the best example of successful pit mining in the district, and the distribution of the workings clearly out- lines the courses of the veins (see figure 21). ECONOMIC GEOLOGY 283 Bedding deposits. — In mining the bedding deposits, such as those at Lead Hill, tunnels are driven from one or more points on the hillside and a somewhat irregular room-and-pillar method of development is followed. Since the strata are practically horizontal and usually not over two or three feet in thickness, little difficulty is encountered, providing care is taken to leave adequate pillars for support. Little timbering is used. Apparently no attempt is made to develop rooms and pillars of uniform size and spacing on account of the irregular outlines of the ore-bodies, and their insufficient lateral extent. MILLING METHODS General. — For the concentration of the fluorspar and the metallic minerals mined from veins or bedding deposits, coarse breaking, hand sorting, Fig. 23. — Mill and Good Hope shaft of the Fairview Fluorspar and Lead Company. crushing, and jigging in Joplin jigs, comprise the usual method. Where blende is absent from the galena-fluorspar-calcite ore, very little difficulty is experienced, except that since the jigs make only two products, the tailings will contain all the fluorspar and calcite together. This mixture is sold as low grade "gravel spar." This gravel spar, it will be seen, is an entirely different product from the residual gravel spar of the superficial deposits, and the two should not be confused. If both blende and galena occur in the ore, considerable difficulty is encountered in concentrating. The jig is not applicable to the mixture, for although a galena concentrate of good quality can be obtained, the valuable fluorspar product will be contaminated by the zinc mineral. The practice is 284 GEOLOGY OF HARDIN COUNTY to direct all zinc-bearing materials to a separate part of the mill for finer crushing and treatment on tables. Such material — usually a blende-galena- fluorspar-calcite mixture — also presents difficulties on the tables, it being generally impossible to effect separation of all the minerals. The galena may be separated almost completely, but the similarity of specific gravities of fluorspar and blende, and the dissemination of the latter tend to give a middling product containing both; a clean product is therefore unobtainable except by secondary and expensive installations. The solution of the problem of economic recovery probably depends upon the application of selective flotation methods, along which line experiments have been made. At the time of this investigation the zinciferous products had been thrown out as waste or set aside for experimentation. Fig. 24. — Mill and Rosiclare mine of the Rosiclare Lead and Fluorspar Company. In places where the sulphide minerals do not accompany the fluorspar, jigging with close adjustment and careful operation, effects a fair separation between the spar and calcite or limestone, although the minerals differ but about 0.4 in specific gravity. Where quartz is present it will be found in the calcite tailing, unless it is intimately intergrown with the fluorspar, when it will contaminate this product. The superficial, or gravel-spar, deposits where the fluorspar is embedded in residual clays, contain no other minerals, and require therefore only a thorough washing to remove the clay that adheres to the spar. This washing is done by trommels, log washers, or concrete mixers. ECONOMIC GEOLOGY 285 Fairview Mill. — The output of all the Fairview Fluorspar and Lead Com- pany's mines passes through the Fairview mill located at the Good Hope shaft (fig. 23). In this mill the lumps of pure fluorspar are hand-sorted from the picking belt as first and second grade lump spars. The whitest is the purest, and is known as Keystone, a very high-grade "acid spar." All zinc- bearing material, part, of which contains lead, is crushed by rolls to sufficient fineness for treatment on Deister tables, from which is obtained a marketable lead concentrate, a zinc-bearing middling containing much fluorspar and some lead, and a fluorspar-calcite tailing. All lead-bearing rock containing no zinc is removed from the hand-picking belt to a separate bin, from which it goes to rolls, is screened, and sent to Joplin jigs ; the products of these opera- tions are a lead concentrate, and a "gravel spar" tailing of fluorspar and calcite. A generalized flow-sheet of the mill follows : Ore from mine 1 1 Washing trommel 1 1 Hand picking belt 1 1 Lead-spar-calcite material 1 Zinc-bearing material Fluorspar ! 1 i i 4 grades Rolls toi Joplir and screen ' size i jigs Rolls i" (?) 1 Deister tables —Keystone (lump) — Al ground — A2 lump —gravel 1 Lead concentrate 1 Zinc-fluorspar mixture Fluor-spar tailing 1 Calcite tailing Market 1 (Flotation) 1 (Market) Cleaned 1 Market 1 Lead cone ;entrate 1 Gravel spar Market Market Rosiclare Mill. — The mill of the Rosiclare Lead and Fluorspar Company is located at the main shaft of the Eosiclare mine (fig. 24) and receives all material mined on the company's properties. The operations are funda- mentally similar to those of the Fairview mill but differ in a few details. No attempt is made to save the small amount of zinc. 286 GEOLOGY OF HARDIN COUNTY An interesting feature of the flow-sheet which follows is the use of a small Kichards' pulsator jig for cleaning all the lead concentrates of the Joplin jigs. The flow-sheet given here is abridged and generalized. Ore from mine I Sorting belt Acid spar" (may be ground, or marketed as lump) Second grade spar (lump or ground) Lead-bearing material Gyratory crusher I Disc crusher I Trommel (¥—$") Oversize Middle size G"-i") Jig. Undersize tt"-0) Jig Lead concentrate Lead concentrate Rolls i" Low-grade spar Market Tailings Richards' pulsator jig (for cleaning) Lead concentrate Market TailiDgs .Table Lead concentrate Market Middling I Crusher Spar Market Tailing Waste At the mill of the Lee mine, a simple arrangement of sizing and jigging was practised, as outlined below. This type of mill is suitable for small mines where the product contains but two minerals differing appreciably in specific gravity. Here the separation of calcite (limestone) and fluorspar is quite readily attained. The product of the Lee mine contains neither lead nor zinc minerals. The mill flow-sheet is essentially as follows: ECONOMIC GEOLOGY 287 Ore from mine 1 1 Rolls (25 mm.) Trommel (12 mm.) 1 1 Oversize 1 Undersize 1 Jig 1 1 1 Trommel (4 mm.) 1 1 Concentrate Tailing 1 Oversize (12 mm. to 4 mm.) 1 Jig 2 1 Undersize (under 4 mm.) 1 Trommel (1^ mm.) 1 Concentrate 1 1 Tailing Oversize (4 mm. 1 Jiff 3 1 to U i mm.) 1 Undersize (Under 1| mm.) 1 Jig 4 Concentrate Tailing Concentrate Tailing Other mills closely correspond to one or another of the types described above. ECONOMIC CONSIDERATIONS Fluorspar and the chemical products made from this mineral, possess a wide variety of uses. The value of fluorspar is due to three qualities: to its fluxing properties, which give it a place in certain important metallurgical operations and in the manufacture of glazes; to its fluorine content, which renders it the chief raw material for the manufacture of fluorine compounds employed in various chemical industries, especially for hydrofluoric acid; and to its optical properties, which invest clear specimens with a specialized application in making lenses for microscopes. About four-fifths of the fluorspar produced in the United States is con- sumed as a flux in the production of steel by the basic open-hearth process. For this purpose, the fluorspar should contain less than 10% of impurities, except that calcite which is frequently present is not objectionable and may run higher. Silica should preferably be less than 2 per cent, though the spar may contain up to 4 per cent, 1 This flux is used principally for giving fluidity to slags that are too viscous, but at the same time it facilitates the passage of impurities such as sulphur and phosphorus into the slag. The consumption of fluorspar per ton of steel produced in this way varies from 4 to 17 pounds, with 8 pounds as a fair average. 1 On this basis the 1916 pro- 1 This statement is made on the basis of the purity of commercial gravel spar. 288 GEOLOGY OF HARDIN COUNTY duction of basic open-hearth steel amounting to 29,600,000 tons is estimated to have consumed about 120,000 tons of fluorspar. 1 The bulk of the fluorspar not consumed by the steel industry is devoted to the manufacture of ceramic products, such as enameled and sanitary ware, opalescent glass, facing for bricks, and vitriolite, and to the manufacture of hydrofluoric acid. For both purposes fluorspar of a high degree of purity is required. In the manufacture of hydrofluoric acid the fluorspar is treated with sulphuric acid, and in the process the presence of impurities in the fluor- spar, such as calcium carbonate and silica, are very detrimental. The calcium carbonate neutralizes sulphuric acid, and its presence to the extent of 1 per cent or over causes considerable foaming upon mixing. Silica, if present, forms hydrofluosilicic acid in such proportion that for every part of silica nearly four parts of fluorspar, and more than five parts of sulphuric acid of 66° Baume are wasted, since for practically all applications of hydrofluoric acid the content of hydrofluorsilicic acid is useless. Thus for every per cent of silica in fluorspar, ten per cent should be deducted from the value of the fluorspar. 2 Hydrofluoric acid finds a varied use as a chemical agent and is enjoying a growing application in the production of aluminum fluoride used as a bath material in the electrolytic production of aluminum from bauxite. In addition to its three major fields of application, fluorspar is employed in a great number of subordinate capacities, some of which may be open to considerable development. It thus finds use as a flux in some blast furnace operations, and in iron and brass furnaces; in the smelting of gold, silver, and copper ores; in the refining of copper, antimony, and lead; as a bond for the constituents of emery wheels ; in carbon electrodes ; in the manufacture of sodium fluoride used as a wood preservative; and in connection with the extraction of potash from feldspar and from the flue-dust of Portland cement works. 3 Transparent, flawless, and colorless, or nearly colorless pieces of fluorite are very valuable as material for making apochromatic and semi-apochromatic lenses for use in the lens system of microscopes. Such optical fluorite is very rare, but occurs here and there in Hardin County in association with the common fluorspar, especially as crystals in cavities in the bedding deposits. Pieces suitable for optical purposes need not be large, but must be flawless and of such size as to yield lenses %" in diameter. Optical fluorite has here- tofore been imported into the United States, but Hardin County can probably supply the entire domestic demand for this specialized material. 4 1 The data for this paragraph were largely supplied by J. S. Unger, Carnegie Steel Company. 2 From information communicated by J. E. Foster, General Chemical Co. 3 Burchard, E. F., Our mineral supplies — Fluorspar: U. S. Geol. Surv. Bull. 666-CC, 1917. Flourspar and cryolite is 1916: U. S. Geol. Surv., Mineral resources of the United States for 1916, pp. 309-325. Includes bibliography of the uses of flourite. 4 Pogue, J. E., Optical fluorite in southern Illinois: 111. State Geol. Survey Bull. 38, 1918. ECONOMIC GEOLOGY 289 Commercial fluorspar is sold in several grades according to the purity. The highest grade is known as "No. 1/' or "American lump No. I 99 some- times called "acid spar." This grade contains under 1 per cent of silica and usually not over 2 per cent of all other impurities. The fluorspar is sold in iump form, or is ground, according to the desire of the purchaser. "No. 1" spar is generally white or clear tinted, and sold for use in making enamels, glasses, and hydrofluoric acid, where material of a high degree of purity is necessary. The next grade, "No. 2 lump/ 7 or "American lump No. 2," is used in smelting operations. Such spar may contain up to 4 per cent of silica as well as an appreciable quantity of calcite, but should be free of metallic minerals. Generally this quality comprises the more deeply colored spar that does not contain any metallic sulphides. The lowest grade of fluorspar, called "gravel spar," contains sometimes more than 4 per cent of silica, and large but varying amounts of calcite. This quality comprises 80 to 85 per cent of the domestic fluorspar production, and is used chiefly as a flux in basic metallurgical operations, particularly the open-hearth steel process. Fluorspar for this purpose must be free of sulphide minerals, since sulphur and metals (other than iron) are detrimental to the finished steel. PRODUCTION Hardin County contains the most productive mines in the Illinois- Kentucky fluorspar district and is responsible for more than two-thirds of the fluorspar production in the United States. The position held by the county in respect to the fluorite consumption of the country is shown in figure 25. This diagram also brings out the marked increase in production under the stimulus of the war demand for steel, and the falling off of imports from England in recent years. The total fluorspar production of the United States and the average price, for the five years ending 1917, are shown in the accompanying table: Table //. — Production of domestic fluorspar, 1913 to 1917 (a) 1913 1914 1915 Aver- Aver- Aver- age age age Quantity. price Quantity. price Quantity. price per per per ton. ton. ton. Quantity. Aver- age price per ton. 1917 Quantity. Aver- age price per ton. Ground Lump.. Gravel. Total. Short tons . 8,137 65. 676 6101.767 115.580 $12.31 $6.88 $5.87 Short tons , 68.842 679. 276 95,116 $11.78 $8.45 $5.21 Short tons. 10.757 612.033 6114.151 136,941 $10.80 $7.51 $4.89 Short tons. 7.595 $12.38 Short tons . 10. 136 614,489 $7.94 25.548 6133,651 $5.34 6183.144 155.735 218.828 $17.59 $13.68 $9.61 a Condensed from table given by Burchard, E. F., Fluorspar and cryolite in 1917 : U. S. Geol. Surv., Mineral resources of the United States for 1917, pt. 2, p. 296. b Some lump spar is included with gravel. —19 G 290 GEOLOGY OF HARDIN COUNTY In 1913, the total Illinois-Kentucky output was 91.2 per cent of the total United States output; in 1914 it was 99.6 per cent; in 1915, 99 per cent; in 1916, 93.8 per cent; and m 1917, 91 per cent. In 1913, Hardin County produced about 81 per cent of the total Illinois-Kentucky output; in 1914 about 79 per cent; from 1914 to 1916 inclusive, the statistics published by the U. S. Geological Survey have not differentiated the Illinois production from that of Kentucky, in order to avoid revealing the statistics of individual producers. The diversity of uses to which fluorspar is applied, coupled with the magnitude of the basic open-hearth steel production and the growing im- portance of hydrofluoric acid as a chemical agent, point to a demand for fluorspar which will continue to increase though most likely not at the rate inspired by war demands. Statistics will, indeed, very likely show a marked temporary slump for the period immediately following the termination of the war. Between 1910 and 1914, the domestic source met with considerable foreign competition, chiefly from the English deposits in Derbyshire and Durham, which had been able, owing to cheap production, favorable ocean freight rates, and bunker shipping capacities, to place their product in the eastern centers of steel manufacture at a lesser cost than the domestic deposits could meet. Since 1914, however, as a result of the war, the English shipments have fallen to a very small fraction of our needs (about 9 per cent in 1916), and although the English deposits are reported to be far from exhaustion, it is questionable if they will again assume the ascendancy in the eastern markets, since the American product is of higher grade and the English mining costs are bound to increase. 1 It is believed therefore that fluorspar mining in Hardin County has a good future. Although it is not likely that veins comparable in size to the Good Hope-Eosiclare and the Blue Diggings will be opened up, vet with the establishment of the railroad through the county, a considerable production from smaller, scattered deposits lies in prospect, especially after a period of years when the demand cannot wholly be filled by the large veins now mainly producing. SUGGESTIONS EOR PROSPECTING Fluorspar is so widely distributed throughout Hardin County that its mere presence at any point is not sufficient warrant for prospecting. More- over, probably most of the more promising deposits have already been located during the years of prospecting that the district has enjoyed, so that a new deposit must present many favorable features in the face of a great number of reasonably good known deposits which are unable to produce in competi- tion with the larger mines. Hence the use of the geological map accompany- 1 See Caruthers and others, Fluorspar : special reports on the mineral resources of Great Britain, Geol. Surv. Mem.. Vol. 4, 1916. In Derbyshire the fluorspar has been obtained at small cost from the dump heaps of lead and zinc mines, but this source is approaching exhaustion. ECONOMIC GEOLOGY 291 9 * § 8 § 8 Tr 5 ousands of Short Tons 5 8 8 5 l » U. F a ) * \\ I. i. U \V 2 u l\ i o =s^ \ p? 1 p7^c % i 1 fill .... IP ?» ^1 §| «c vP - v ^% a j^sy m IP * Jo 12 r r y W<% z o w 3 1*1 1 UJ II a 8 fair t- <\ i^' r t£Sv v v 2 to & 53 =£*< *,' \? 1 i ^ -ti *6 & \££\? & w& <© f/l 1 1 1 1 1 1 1 ^ . 00 1 1 1 1 1 1 1 1 1 I 292 GEOLOGY OF HARDIN COUNTY ing this bulletin and the employment of a few simple geological principles will save much time and expense, whether a new deposit is sought or an abandoned prospect is being opened up. The presence of fluorspar either with or without galena and sphalerite, is the first-sought clue to a valuable deposit. Its presence once determined, it is desirable to ascertain if the deposit lies within a fault fissure, since most of the producing deposits show this relation. The faults of the region have been mapped with great care and in great detail, and reference to the geologic map will show whether a deposit in question has a favorable position in respect to the known faults. If it lies upon a fault which has been mapped, preliminary pitting and trenching may be undertaken with greater assurance of success than if its position is not intersected by a fault. But the absence of a mapped fault should not be accepted as final disproof of commercial value, provided other features are promising, since the faulting in the dis- trict is very intricate, and some faults, obscured by the soil mantle, cannot be detected except by excavation; also small deposits may occur in joint planes that connect with faults some little distance away. On the whole, however, the faulting shown on the geological map may be followed with considerable confidence as a guide in prospecting. On the other hand, it is to be remem- bered that the presence of a fault, even if it is mineralized with calcite, is alone not sufficient evidence of promise, for many faults are utterly barren of valuable material. The location of a fault may be recognized in the field where two unlike rock formations, such as sandstone and limestone, lie in juxtaposition; the determination of a line separating the distribution of the two types will indicate the surface trace of the fault plane; in many instances this is also expressed topographically by a ridge of sandstone hardened into quartzite by the mineralizing waters which circulated through the fault fissure. Where the opposing sides of the fault are limestone, the fault cannot readily be determined in the field and recourse must be had to the geologic map or technical advice must be sought. In addition to the presence of fluorspar in association with faulting, other conditions should be borne in mind. The presence of calcite is favor- able, especially if the vein walls are both limestone, or if one is limestone and the other is sandstone. A vein having both walls of sandstone gives less promise, although it should be remembered that the character of the walls changes with depth. The possibility of development of galena or sphalerite unaccompanied by fluorspar should be viewed with considerable question. While these two minerals are present in the large fluorspar veins, and are produced as by-products in the fluorspar mining, there is slight hope that they are sufficiently concentrated at any place in the county to warrant their exploitation alone. Nor can the presence of these minerals be considered as ECONOMIC GEOLOGY 293 an indication of a fluorspar deposit, since they are known to occur here with- out fluorspar in several cases. When a superficial deposit of fluorspar is located, there should be no attempt at sinking a shaft until the size and form of the deposit has been carefully determined by surface pitting and trenching. Too much emphasis cannot be placed upon the necessity for careful preliminary work, for an appreciable body of disseminated fluorspar in the clay may prove in depths of thirty to forty feet to be the disintegrated top of a vein too narrow for successful working. After the general trend of the fluorspar body has been determined by close pitting, a trench should be dug at right angles to the course at what appears to be its widest portion. The trench should be carried deep enough to show up the structural character and width of the vein, after which its length should be explored by means of similar trenches at intervals of a hundred feet or so along its trend. THE PRINCIPAL MINES OF HARDIN COUNTY While there are a great many mines and prospects throughout Hardin County, only a. few are actually producing important quantities of fluorspar. Some are being operated by individuals as lessees and are subject to fitful operations as market conditions change. Since the beginning of the war, and particularly since 1916, the rapid rise in the price of fluorspar has stimulated production, with the result that some of the abandoned properties have been reopened and prospecting for new locations undertaken. This chapter cannot treat separately all the new prospects, a few of which may prove to be valuable deposits, and the majority of which will either prove to be valualess, or perhaps workable only, under the stimulus of highly favorable market con- ditions, or a change in present transportation facilities. A few mines were opened solely as lead or zinc producers, and contain no fluorspar. All such mines have been abandoned, as in no case has a deposit of lead or zinc minerals been found of sufficient size to warrant continuance of operations for the recovery of those minerals alone. Rosiclare mine. — This mine, owned by the Eosiclare Lead and Fluorspar Company, is on the Good Hope-Bosiclare vein and has been mentioned in the discussion of that vein earlier in the chapter. It is located in the north- western side of the settlement of Eosiclare, and is one of the most important mines of the fluorspar district. Its history has been one of almost continuous operation since the earliest days of mining in the region. Originally it formed a part of the Pell property on which fluorspar was discovered in 1842-43, and not long after mined. It has been, therefore, very extensively developed. The main shaft is now 620 feet deep. An older shaft, a few hundred feet southwest of the first, reaches to the 420-foot level. Levels are open at depths of 235, 320, 420, and 520 feet, but all present operation? 294 GEOLOGY OF HARDIN COUNTY 100- 200 300- 400 St. Louis h ig. 26.— Idealized geologic section through the air shaft of the Rosiclare mine. ECONOMIC GEOLOGY 295 are carried on below 235 feet, and mostly below 320 feet. Old workings between the 235-foot level and the surface are no longer accessible. The vein is nearly vertical and its average strike is abont N. 20° E. In places its strike is as much as N. 35° E. In width it varies considerably, reaching a maximum of about 30 feet, though it is more commonly between 6 and 12 feet. Eigure 26 shows the vein relationships and stratigraphy at the mine. Mining is generally carried on in shrinkage stopes, though some under- hand stopes have been developed in the past. Daisy mine. — This mine is owned and operated by the Eosiclare Lead and Fluorspar Company, and is supposed to be on the Blue Diggings vein. This may prove to be true, but until the drifts are carried farther south the identity cannot be verified. The average course of the Blue Diggings vein is about N. 30° E., but in the Blue Diggings mine of the Fairview Company, strikes of N. 23° E. and N. 40° E. were observed. The vein, however, is irregular and the fluorspar bodies appear to be occupying a broad zone of fracture rather than a definite fault fissure. In the Daisy mine, toward the south end of the 200-foot level, the average course of the vein changes from about N. 40° E. to a few degrees west of north. At this bend no indications of a cross fault are apparent, the change in direction appearing to be primary. Xorth of this bend the dip of the vein is toward the northwest, and south of the bend, toward the southwest, in each case about 70°. In the Blue Diggings mine of the Fairview Company, although the dip is variable with depth, in general it is toward the southeast. Such lack of correlation of strikes and dips in the two mines suggests a respective location on different, probably intersecting veins. At the surface, geologic conditions are obscure because of the lack of rock outcrops, and no second fault in the vicinity of the Daisy mine is determinable. According to Weller 1 the fault on which the Blue Diggings mine is located runs through or close to the Daisy. In view of this and of the ill-defined character of the mineral body at the Blue Diggings, the faulting may be expressed by the broad zone of fracture, as suggested above, in which the mineral bodies could wind in serpentine courses, while the general strike of the fault zone can be said to be about N. 30° E. The fluorspar in the Daisy mine is particularly pure and white, with the bluish tinge that characterizes the "Blue Diggings" mineral. The mineral body in the Daisy is 12 feet wide in places, but pinches to 2 feet at the south face. Fairview mines. — The shaft of the Good Hope mine is located at Fairview about three-fourths of a mile west of Eosiclare and one-half mile back from Ohio River. The mine is owned and operated by the Fairview Fluorspar and Lead Company, and is believed to be on the same vein as the Eosiclare mine. The shaft was started about 1862, althongh the existence of ore was discovered 1 Personal communication. 296 GEOLOGY OF HARDIN COUNTY in 1842. The mine was worked in a small way until 1874 without much interruption, but thereafter was shut down for a few years. The operators of the Rosiclare mine worked it under a lease between 1890 and 1895. From Paint Creek .nt)*- Bethel Renault and Shetlerville 'Lower Ohara" Rosiclare •'.*." Fredonia St. Louis Renault and Shetlerville "Lower Ohara" - '.'..'. 'Rosic]are\ '.'.'.'.'.' Fredonia and St. Louis I ! 1 100- 200 300 400 600 Fig. 27. — Idealized geologic section through the new Good Hope shaft. 1905 to 1913 the present company operated the mine through the original Good Hope shaft, but in 1913 this shaft gave trouble and was abandoned. In August, 1917, work on reopening the shaft was commenced. The Good Hope ECONOMIC GEOLOGY 297 shaft was originally sunk on the vein to a depth of 503 feet and in 1913 it had a drift development on 6 levels totalling 5,500 feet. The fluorspar taken from the mine was of high grade. The ore bodies were short but of goo'd width, as wide as 25 feet in places. Figure 27 shows the vein relationships and stratigraphy at the Good Hope mine. Unconsolidated surface materials Fig. 28. — Idealized geologic section through the Annex shaft. The Fairview Company in 1909 opened up an old shaft on the same vein about 1,700 feet north of the Good Hope shaft, and extended it to 300 feet. Operations ceased in 1914, however, most of the fluorspar having been removed. In 1911 work was begun on the Annex and Extension mines, 1,700 feet south of the Good Hope shaft. As stated earlier, this may be on the same 298 GEOLOGY OF HARDIN COUNTY vein as the Good Hope, or it may be on a branch of the main Good Hope- Bosiclare vein. The Annex shaft (fig. 28) is now 340 feet deep and the Extension 440 feet. The mine is in operation. Blue Diggings mine. — The Fairview Fluorspar and Lead Company be- gan operations on this mine in 1910. It is. situated on the Blue Diggings vein heretofore described. The shaft is located about a quarter of a mile west of the main shaft of the Eosiclare mine, in the NE. y± SW. 14 sec. 32. An attempt to operate an old shaft was soon abandoned and a new one started. The shaft, which is now 540 feet deep, was driven in the hanging (east) wall and passed through the vein at the 200 foot level. At the 500 foot level the vein is 150 feet east of the shaft. The general characteristics of the vein are described earlier in the chapter, its dip direction being radically different from that of the Good Hope-Eosiclare vein, as shown in figure 29. Hillside mine. — This mine, opened in the early part of 1919 by the Inland Steel Company, is situated in the NE. % of sec. 32, T. 12 S., E. 8 E., three-quarters of a mile north of Eosiclare. At the time of the writer's visit a shaft had been sunk to a depth of 170 feet and a cross-cut made to the vein at this point. The location of the workings and the character of the vein and wall rocks point to the probability that the mine is located on the Good Hope-Eosiclare vein. Clement mine. — This mine is located about two miles north of Eosi- clare, in the SW. % SW. 14 sec. 21. As the mine is no longer in operation and apparently has not been for several }^ears, the underground workings were not accessible. The dump shows limestone country rock with coarse calcite, and fluorspar replacing the calcite, as at the mines around Eosiclare. Accord- ing to Bain 1 , who examined the workings in 1903, the vein is situated in a fault of slight displacement. Weller 2 considers this fault the continuance of the combined Eosiclare and Blue Diggings faults, which come together prob- ably not more than half a mile south of the Clement shaft. Twitchel mine. — This is a small mine located in the NE. 14 NE. 14 sec. 24, T. 12 S., E. 7 E., just north of St. Josephs school. It has a shaft 55 feet deep and a drift 30 feet long. The vein runs N. 5° W., dips 70° E., and is about three feet wide. Calcite and fluorspar are the vein minerals, and both walls are limestone. It does not appear to be in a fault. Preen mine. — This mine, located in the NW. % NW. % sec - 19 > T - 12 S., E. 8 E., was operated by lessees in 1917, and a small quantify of fluorspar was produced. An old shaft extends to a depth of 75 feet. A new shaft was sunk to 30 feet. Near the surface the fluorspar is somewhat "pockety" in clay, and resembles superficial deposits. At a depth of 75 feet limestone walls were found. Calcite appears as a vein mineral. 1 Op. cit., p. 47. 2 Personal communication. ECONOMIC GEOLOGY 299 Map Soale In Feet inyo 3000 Fredonla Renault and Shetlervllle "Lower Ohara' ■ '■:■ Ro'siqlare ; . • Fredonla and St. Louis Oi 100- 200 300 400 500 Fig. 29. — Map and idealized geologic section to show the vein relationships and stratigraphy at the Blue Diggings shaft. The zone of extensive fracturing is only very diagrammatically shown. 300 GEOLOGY OF HARDIN COUNTY en I Pell mine. — The Pell mine is located about one mile west of the Pre mine, in NE. % NW. % sec. 24. The mine was abandoned for several years until the spring of 1919, when a lessee started to sink a new shaft. The old shaft was 95 feet deep and the dump pile showed that vein materials are coarse calcite, fluorspar, galena, and blende. Pits along the outcrop of the vein show that it is located in a fault with limestone and sandstone as the wall rocks. The fluorspar seen in the dump, and that removed from a very shallow depth by the present lessee appeared to be of good grade. Stewart mine. — This mine is located about one mile northwest of the Pell mine near the center of sec. 14, T. 12 S., R. 7 1 E. It was being operated at the time of the writer's first visit to the fluorspar region in 1917 by open pit operations but a shaft had been sunk to a depth of 75 feet in the hanging (east) wall. The vein runs N. 25° E. and dips about 80° E. In an apen pit 200 feet north of the shaft a vein of fluorspar 2% to 4 feet wide was seen. The open-cut extended for about a hundred feet along the vein. No calcite appeared in the open pit. At a depth of 25 feet the walls were reached, lime- stone on the west and sandstone on the east, showing the existence of a fault. Some galena is present in the deposit. An old dump pile nearby, said to have come from an old shaft driven to a depth of 100 feet, showed much coarse calcite in the wall rock. Martin mine. — At the Martin mine, NW. % NE. % sec. 17, about four miles directly north of Rosiclare, the dump pile shows blue fluorspar in limestone, with a small amount of calcite. The vein is situated in a fault with walls of limestone and quartzitic sandstone. No lead or zinc minerals were seen. The underground workings were abandoned and inaccessible. The property is controlled by the Rosiclare Lead and Fluorspar Company. Miller mine, M etcher Hills. — This mine is located about l 1 /^ miles north- east of Shetlerville, and just east of Melcher Hills, in the SW. *4 SW. % sec. 25. The mine is entered by tunnels in the hillside from which an irregu- lar system of rooms has been developed. The deposit appears to be of the bedding type for the most part but several irregularly trending narrow veins are to be seen. Besides fluorspar, galena is prominent. A shaft was being sunk in August, 1917, and was 75 feet deep at that time, but was inaccessible to the writer. It was stated that a vein was struck running in a northeasterly direction. Limestone wall rocks were found at this depth. Lead Hill mines. — Lead Hill is situated in the western half of sec. 4, T. 12 S., R. 9 E., about four miles northwest of Cave in Rock In the south- western side of this hill are several small mines that represent sporadic operations extending over many years. At the time of the writer's visit some of the workings were being reopened. The deposits of this locality have been described in the section on "bedding deposits." The workings of the Lead Hill Lead and Spar Company consists of a series of chambers entered bv tunnels in the hillside. Since these deposits are ECONOMIC GEOLOGY 301 practically horizontal and never more than two or three feet in thickness, there is only one level of underground workings. An old and now abandoned mill, containing crushers, trommels, and jigs, and a good sized dump pile, testifies to earlier attempts at large scale operations. The deposits are banded and are replacements of the nearly horizontal limestone. At Lead Hill the mineralized beds may continue with more or less thinning and interruption entirely through the hill to the east slope, for outcrops of the banded fluorite are found in several places on that side at nearly the same elevation as that at which the deposits occur on the west side. Some outcrops on the east side of Lead Hill which occur at a somewhat lower level than those mentioned above have recently been slightly worked. Part of the material here has prominent bands of quartz crystals alternating with fluorspar. The general structural relationships appear to be otherwise identical in character with the other bedding deposits. Although a little sphalerite is to be found at Lead Hill, it is not nearly so abundant as galena. Galena appears in pockets and as well-formed cubes along the sides of some of the replacement bands. Cerussite is common as a coating of the galena and of some of the fluorite grains of the finer bands. The drusy yellow variety of smithsonite (zinc carbonate), called "turkey fat," may occasionally be found. Some fluorspar was found here of such character as to warrant careful search for spar of optical quality. Quite flawless and nearly colorless pieces are fairly abundant, and while no very large pieces were found it is alto- gether probable that here is a locality from which mineral suitable for lenses could be obtained. Owing to the banded character of the fluorspar deposit and the relatively poor quality of many of the bands, the product of these mines could probably not compete with the higher grades of vein fluorspar. In some instances however very good material is obtained. Cave in Rock mine. — In the N"W. 14 NE. % sec - 4, about three-quarters of a mile northeast of the Lead Hill mine, another bedding deposit of fluor- spar is being worked. A very irregular bed of the mineral, in places three feet thick, underlies a capping of dense limestone. The deposit is banded similar to the Lead Hill occurrences. This mine is of the open-pit type, although a shaft is present and some drifting has been done. Some exceptionally fine specimens of optical quality have been obtained from pockets in this mine, and the deposit holds considerable promise as a source of optical fluorite. 1 Empire mine. — The Empire mine is situated in Pope County, in the SW. 14 SE. % sec. 27, T. 11 S., E. 7 E., less than half a mile west of the Hardin County line. The original mine is now abandoned. A shaft and Bull* ?8 g iq'l8 J * E *' ° ptical fluorite in southern Illinois: 111. State Geological Survey 302 GEOLOGY OF HARDIN COUNTY th open-cut were seen, and a dump pile in which were found specimens of bo limestone and sandstone, containing fluorspar, blende, and coarse calcite. The relations are similar to those in the mines at Eosiclare. Underground workings were not accessible but Bain 1 describes the vein as "6 to 10 feet wide and with well-developed walls. The vein matter consists of brecciated limestone cemented by fluorspar and calcite intimately intergrown. In this matrix galena, blende, pyrite, and chalcopyrite occur." Zinc carbonate is prominent on some of the pieces in the dump taken, presumably, in the open cut near or at the surface. Pierce mines. — The Pierce mines are situated about half a mile south of the old Empire shaft. There are several open-cuts on the property, the general distribution of which, together with the character of the fluorspar bodies, indicates the presence of at least two veins — one running about N. 75° E, and the other about N. 45° E. Figure 21 is a sketch map of the pitted ground. The fluorspar in the open-cuts is colored, transparent to translucent, and pure. It is somewhat dispersed through red and yellow clays, but sufficiently well localized to indicate the probable existence of very well defined veins at not much greater depths. Some rounded boulders of limestone, remnants of the walls, are encountered during the mining operations. These boulders are traversed by narrow seams of calcite and fluorspar. Some coarse calcite is found on them. At one point rounded calcite fragments were found in the main mineral body. One of the pits shown in figure 21 was dug to a depth of forty feet where limestone was encountered. This pit is not being worked. South and west of these pits are several abandoned shafts and prospect holes. One of these is the Hubbard shaft originally sunk by the Grand Pierre Mining Company. The fluorspar of this locality is of such character as to warrant search for material of optical quality. Other mines near the Empire. — North of the Empire shaft may be seen several abandoned mines, among them the Big Joe, Hutchinson, Eainey, Baldwin, and Hicks, which are described by Bain. 2 All of them are now unimportant. While there are a few shafts on these properties, most of the workings are open-cuts and in the nature of prospects. North of the Empire mine several open-cuts have recently been dug by the operators of the Pierce mines. Hamp mines and adjacent prospects. — In tne NW. V± sec. 18, about four miles northeast of the Empire, are several shafts which have apparently not been operated for a number of years. These shafts comprise the Hamp mines and adjacent prospects. Dump piles displayed limestone with a little fluor- 1 Op. cit., p. 49. 2 Op. cit, pp. 50-52. ECONOMIC GEOLOGY 303 spar, calcite, galena, and blende. The course of the vein at the Hamp mine is said to be X. 60° E. No faulting is discernible at the surface. The vein may be filling a fissure of very little or no displacement connecting at some distance with a more pronounced fault fissure. At a small cut west of the shafts, banded fluorspar of the "bedding" type of deposit was seen. Unlike the fluorspar near Cave in Rock the mineral of these bands was very deeply colored and nearly opaque. When freshly broken, the mineral from this deposit gave a strong odor of petroleum. Several deposits of gravel spar have been dug recently in the vicinity north of the mine. Fig. 30. — View cf a mass of fluorspar on the weathered apex of a vein. Bald- wins Run, Pope County, just west of the county line near the center of the east side of sec. 27, T. 11 S., R. 7 E. Lee mine. — The Lee mine is situated about three and three-quarters of a mile due east of the Hamp mine, in the NW. % NW. % sec. 14, T. 11 S., R. 8 E. A shaft 71 feet deep has been dug, meeting a vein 8 feet in thick- ness. The vein strikes N. 55° E. and dips 80° SE. It can be traced on the surface for nearly half a mile. Faulting is shown by a hanging wall of lime- stone and a footwall of quartzite. Purple and white fluorspar composes the vein matter, neither metallic sulphides nor calcite being present. The mine was being worked in September, 1917. Rose mine. — At this small abandoned mine, situate:! in the SE. 14 sec. 30, T. 11 S., R. 8 E., some pieces of fuorspar of such character as to suggest 304 GEOLOGY OF HARDIN COUNTY a likely source of the mineral for optical purposes were obtained from the clump. It was slightly tinted, crystalline, showing no double refraction, and small pieces were remarkably free of flaws. Much of it, however, contained enclosed grains of pyrite and chalcopyrite. Miscellaneous mines and prospects. — Besides the mines mentioned above, many other abandoned mines and prospects are to be found in Hardin County, among them several mines mentioned by Bain, but not included in this report because of their unimportance and the fact that nothing could be learned about the character of most of them. Also hundreds of new prospect pits dot the ground all through the county, some of which will probably become important later. It would be impracticable to list and describe all of them. Those that have been described comprise the active, and the most important of the abandoned mines, as seen at the time of this investigation. Iron Ore, Limestone, Eoad Metal, Oil, and Gas By Chas. Butts IRON ORE The existence of iron ore in Hardin County has been known ever since its settlement, and the deposits were utilized in early days at the old Illinois Furnace on Big Creek, three and one-half miles northwest of Elizabethtown, and at the Martha Furnace in sec. 2 of T. 12 S., E. 8 E. The Illinois Furnace obtained its ore largely from land immediately to the east in the SW. % sec - 3, and the N*E. % sec. 9, T. 12 S., E. 8 E., on the McKernan farm. The Martha Furnace also secured its ore from its immediate vicinity. As usual with such iron deposits, the ore is limonite, a hydratecl iron oxide, Fe 2 (OH) 3 . As usual, also, it occurs in the form of small pellets or gravel (wash ore), nuggets, and larger irregular chunks and masses scattered through clay and soil, with all of which there is also much chert, and possibly some sandstone. All of this material is residual from the decay of rocks, mainly of limestone. The deepest accumulations of this kind are likely to be located along the sides of the knobs and spurs, as can be seen at the old diggings on the McKernan property which supplied Illinois Furnace. Some of these old pits are said to have been 90 feet deep. On the sides of the pits next to the spurs there are irregular masses of limestone, exposed in digging the ore. These limestone masses now projecting up into the clay and earth which holds the ore, are undisintegrated and undissolved portions of the once continuous rock strata. It is only where the residual clay is deep that much ore can accumulate, and it is useless to look for it elsewhere. Other localities where such iron ore has been dug, or where more or less limonite float occurs in the gullies, are the SW. % NE. % sec. 34, T. 11 S., E.8E.; SW. % sec. 35, T. 11 S., E. 8 E.; sec. 2, T. 12 S., E. 8 E.; NE. 14 sec. 24, T. 11 S., E. 7 E.; SW. % sec. 13, T. 11 S., E. 7 E. ; and on Hicks Branch just west of the church in the SE. Vi ECONOMIC GEOLOGY 305 SW. V 4 sec. 26, T. 11 S., R. 7 E. At the last described place large masses of limonite are exposed over a few square rods of surface on the steep bank on the south side of the road. All the known occurrences of iron ore- deposits in Hardin County are in areas underlain by the Fredonia or St. Louis lime- stones, and in that particular they are like most of the extensive limonite deposits in eastern United States, as in Pennsylvania and Alabama, where the ore occurs in residual material accumulated upon the outcrops of thick limestone formations. It is safe to affirm that there is considerable iron ore in Hardin County, but no definite or reliable estimate of the quantity can be made with our present knowledge, as such estimates can only be made after extensive pros- pecting. According to Owen 1 the limonite ore of the county contains over 50 per cent metallic iron, which is about the usual metallic content of the limonite ores now being used. The composition of a specimen of ore from the bank of the old Martha Furnace in sec. 2, T. 12 S., E. 8 E., is, according to Owen as follows : Composition of iron ore from bank of Martha Furnace Water H=10.8 Siliceous earth 5.0 Peroxide of iron Fe 2 O 3 =80.0=56.02 Fe Alumina Al 3.7 Loss and alkalies not estimated 0.5 100. Owen advanced the theory that these ores are derived from fissure de- posits of magnetite extending to great depths. He thought that this deep seated ore was taken into solution by ascending carbonated waters and carried to the surface and redeposited in the upper part of the fissure or in the earth above, as they now exist. This theory has not been verified, and no such occurrence is known in regions of exclusively sedimentary rocks. The actual deposits of ore, on the other hand, agree fully in their character and manner of occurrence with limonite ore deposits in other parts of the country which are entirely of superficial origin, and these Hardin County deposits are undoubtedly of such an origin. LIMESTONE An abundance of high grade limestone, well suited for the manufacture of lime, cement, and ground rock for fertilizer, exists in Hardin County. The Fredonia limestone carries the best rock for lime and cement. The white oolite beds of this formation, utilized for such purposes elsewhere, are 1 Owen. D. D., Report on Saline, coal and manufacturing companies properties. — 20 G 306 GEOLOGY OF HARDIN COUNTY the most desirable. Such oolite beds constitute a large part of the full thick- ness of the Fredonia limestone, and the formation has large areas of outcrop in the county, with many favorable situations for quarrying. Probably the best site for exploiting the Fredonia is at Lead Hill, four miles northwest of Cave in Eock, in the western part of sec. 4, T. 12 S., R. 9 E. At this locality there is fully 100 feet of the limestone exposed entirely above the surrounding level land, so that a quarry would be naturally drained and the rock could be handled by gravity. The strata are nearly flat, and a very large body of rock is accessible without stripping; only 20 to 50 feet of stripping would be required on the summit of the hill, and most of the material to be moved would be suitable for road metal. In addition to the limestone at this locality, a considerable amount of fluorspar and lead could be recovered in connection with the quarrying. Hardin County is the nearest source of supply of ground limestone and crushed rock, for a large territory to the north in Illinois and Indiana, and with favorable transportation facilities Lead Hill seems to offer an opportunity for profitable enterprise. The other limestone formations of the county should supply good rock for fertilizer and other purposes. The only quarry of importance in active operation in the county, is at Jacks Point, on the bank of Ohio River about one mile below Elizabethtown. The rock quarried at this locality, which is from about the dividing line between the St. Louis and Ste. Genevieve limestones, is transported on barges clown the river for use in work on the river banks. In the past a number of quarries have been operated in the river bluffs above and below Rosiclare in the Fredonia limestone of the Ste. Genevieve, but none of them is in use at the present time. The Menard limestone would be serviceable for fertilizer and other purposes wherever it is well exposed in the county, and the Glen Dean also in the eastern part of the county. Much of the St. Louis limestone would be suitable for agricultural lime and for crushed rock, but most of it is too siliceous for the manufacture of lime or cement. No chemical analyses have been made of the Hardin County limestones, but from analyses of rock from other regions the Fredonia oolite is known to be of a high grade, and as the formation here preserves all of its outward characteristics shown elsewhere where it has been tested, the conclusion is justified that it is of high grade here also. ROAD METAL In addition to its many limestone formations which would afford crushed stone for road metal, the county has a very great reserve of excel- lent road material in the vicinity of Hicks dome, in the form of broken chert, most of the chert being derived from the Osage formation, but with a considerable proportion from the Warsaw, St. Louis, and Ste. Genevieve ECONOMIC GEOLOGY 307 limestones. These chert deposits are of two kinds, first, creek gravel, and second, residual accumulations upon the outcrop of the Osage formation. To a depth of six to ten feet or perhaps more in places, the creek gravel underlies the flood-plain areas along the streams heading near the Hicks dome, as Hicks Branch, its north fork, and Goose Creek. Such material too, is present in abundance along Big Creek and Hogthief Creek in the northern part of the county. It usually has a cover of alluvium two or more feet in thickness. The residual mantle on the outcrop of the Osage is of greatly varying thickness; on the ridge tops it is probably not very deep, but on the sides of the hills, especially towards the bottoms of the slopes, there must be very large bodies of rock waste made up largely of angular fragments of chert mingled with a greater or less proportion of earth. Including both kinds of deposits there seems to be an almost inexhaustible quantity of ma- terial, especially when it is remembered that not improbably the solid chert of the ridges is minutely divided by joints and would easily break up in quarrying into small pieces suitable for road metal. Chert beds of similar character and age in the vicinity of Birmingham, Alabama, do break up thus in quarrying and need hardly any further preparation for road work. All of the creek gravel and residual material on the ridges could be raised by steam shovel and screened to suitable sizes for the different courses of a macadam road, and all would be ready for use with this preparation alone, except that the finer screenings for the top surface might need a small admixture of finely crushed limestone for binding material. No tests of this material have been made, either in actual use or in the laboratory, but in a few places where roads are located upon it on the ridges or for short distances in the valleys near the base of the ridges, the excellent effects of the material in maintaining a hard, smooth surface on the road are amply demonstrated. The excellence of such material is also shown by the roads surfaced with it about Birmingham and Bessemer, Alabama. The building of one of the new state roads which will pass near an easy supply of this material, will present an opportunity for testing it that should not be missed. If it should prove as good as it is believed to be, the material might find a very extensive use in southern Illinois, for the Hicks region is the nearest source of supply of such good road material to a large territory. OIL AND GAS POSSIBILITIES STRUCTURE The structural features in northwestern Hardin County include from southeast to northwest the Hicks anticline, the Eagle Valley syncline, the Horton Hill anticline, the Potato Hill syncline, and the Bald Hill anticline. The Hicks anticline is a strong fold long known as the Hicks dome; farther northwest, mainly in southeast Saline County but extending south- westward into northern Pope County, is a low oval anticline here named 308 GEOLOGY OF HARDIN COUNTY the Horton Hill anticline; and a mile or two still farther west is a strong unsymmetrical anticline, extending from Lusk Creek in sec. 24, T. 11 S., R, 6 E., northeastward through Bald Hill, and hence called the Bald Hill anticline. The form and extent of the Hicks dome and of the Horton Hill anti- cline are shown on the accompanying map (PL III) by structure contours, those on the Hicks dome being drawn on the top of the black Chattanooga shale at vertical intervals of 100 feet, and those on the Horton Hill anti- cline drawn on the top of the Palestine sandstone at vertical intervals of 50 feet. Only the approximate position of the axis of the Bald Hill anticline is indicated, owing to its narrowness, steep northwest limb, and less well-known details. Between the Hicks dome and the Horton Hill anticline is the Herod fault and a syncline provisionally named the Eagle Valley syncline. The Herod fault trends northeast, and the strata are downthrown on the south- east side about 100 to 200 feet. Between the Horton Hill and Bald Hill anticlines is the Potato Hill syncline, the axis of which lies near and just west of Potato Hill in the SW. % sec. 18, T. 11 S., E. 7 E. Two faults cut into the northeast flank of the Hicks Dome and the most southeasterly one may pass across the dome. About 2 miles farther southeast, and outside of the area mapped, another fault cuts across the Hicks anticline in a general northeast-southwest direction like the mapped faults. From the center of the Hicks dome the rocks dip outward strongly to the southwest and northeast and intermediate directions with dips be- tween 10 and 20 degrees; the dip is more gentle to the southeast. In broad areas to the southwest and to the north of the center of the dome, the dip is low or the strata are nearly fiat. This flattening of the structure, producing a structural terrace, is indicated by the spreading of the con- tour lines in sees. 13, 34, and 35, T. 11 S., R. 7 E. ; sees. 2 and 3, T. 12 S., R, 7 E, and sees. 17 and 18, T. 11 S., R. 8 E. Outward from these terraces the dip steepens to about 10 degrees which prevails to the outer margin of the contoured area southeast of the Herod fault. Beyond the contoured area the rocks dip at angles less than 5 degrees as far as the axis of the Eagle j Valley cyncline, but the continuity of the strata is interrupted by the Herod fault. West of the Eagle Valley axis the strata rise westward at a small de- gree to the axis of the Horton Hill anticline. This seems to be an elong- ated dome, the axis of which pitches moderately to the northeast and to the southwest from the center of the dome. 'Northwest of this axis the rocks dip northwestward at a low angle to the axis of the Potato Hill syn- cline. From the last named axis the strata rise westward to the axis of the Bald Hill anticline at a rate gradually increasing to as much as 25 de- ECONOMIC GEOLOGY 309 grees, locally, near the axis. Immediately west of the axis the dip varies from 15 to 60 degrees westward, the steeper dips being observed in sees. 12 and 13, T. 11 S., K 6 E. and in sec. 31, T. 10 S., R. 7 E. A short distance west of the axis the rocks are nearly horizontal. POSSIBLE OIL-BEARING STRATA In the central part of the Hicks dome the possible sources of oil lie below the Devonian limestone. Of these lower strata nothing directly is known in the region, since they are nowhere exposed at the surface and have not been penetrated by deep wells. Rocks that normally occupy this stratigraphic position outcrop in Union and Alexander counties, 111., 50 miles west of Hardin County; and along Tennessee River 70 miles and farther south of the region. From the Devonian and Silurian rocks of these regions must be inferred the possible character of the corresponding part of the general geologic column in Hardin County. According to Savage 1 , in Union and Alexander counties these rocks include limestone of Onondaga age 150 feet thick, sandstones and cherty limestone of Oriskany age 235 feet thick immediately underlying the Onondaga limestone of Illinois reports, limestone of Heldei bergian age 160 feet thick; limestones of Silurian age 120 feet; and sandstone of Richmond age, the Thebes sandstone of Illinois reports, 90 feet thick, lying about 530 feet below the limestone of Onondaga age. If the section of these lower strata in Hardin County is at all like what it is in Union County, it is possible that some of these sandstone beds underlie the Hicks dome at a depth of not over 1,000 feet below the top of the Chattanooga shale. These strata are nowhere known to be oiJ -bearing, but that is not necessarily a condemnation, for in the Colmar field in McDonough County, 111., the oil is in a local sand of Niagaran age, and in Allen County, Ky., the oil is reputed to be in rock of Niagaran age. These are recent discoveries and apparently the only occurrences of the kind. There is always a chance of oil occurring at horizons at which it has not been found elsewhere. Farther out from the center of the Hicks dome, as on the structural terraces previously mentioned, there is, in addition to the chance of oil in the strata below the Devonian limestone, the chance of its presence in that limestone itself, strata of that age being oil-bearing in the Irvine field in eastern Kentucky, in the field in Ohio County, Ky., and in the oil fields of Ontario, Canada. The lower part of the limestone of Onondaga age in Union County, 111., is said to have a strong odor of petroleum. The top of this limestone should be reached at about 1,600 feet below the general level of the country in the area of the structural terraces in sees. 13, 34, and 35, T. 11 S., R. 7 E,; sees. 2 and 3, T. 12 S., R. 7 E. ; and sec. 17, T. 11 S., R. 8 E.; and the possibly oil-bearing strata below the limestone are probably 1 Savage, T. E., Lower Paleozoic stratigraphy of southwestern Illinois: 111. State Geo. Survey, Bull. 8, pp. 103-116, 1907. 310 GEOLOGY OF HARDIN COUNTY within a depth of 2,200 feet below the general level of the ground in these sections. In the region of the Horton Hill anticline there is also a chance that the basal sandstone strata of the Chester group may contain oil. The top of these sandstones is at a depth of 500 to GOO feet along the Horton Hill axis, whereas the limestone of Onondaga age is probably about 2,300 feet deep and the horizon of the Thebes sandstone of the Illinois Survey, the lowest possible sandstone, is probably not over 3,000 feet below the top of the upper Chester sandstone in the center of the anticlinal area. PART VI PALEONTOLOGY CHAPTER XIII— PALEONTOLOGY By Stuart Weller In each of the several chapters of the stratigraphic portion of this report, a section has been devoted to the paleontology of the formations described. These discussions have been confined to a consideration of the general char- acteristics of the faunas of the formations, especially in the consideration of the bearing of the faunas upon their correlation, and no technical descriptions of any of the fossils have been given. In the course of the work, however, it has developed that certain of the fossils need further description and discus- sion. Some of these are undescribed forms which have some direct bearing upon the questions of correlation, and must therefore be defined for the first time. Others are species which are already named but which have some special bearing upon questions which are discussed in the report, and which need further illustration and consideration. No attempt will be made here to include descriptions of all of the fossil species known to occur in the rock strata of Hardin County, for such a treatment would require a large volume. Most of the species that will be discussed in this chapter are members of some one or another of the Chester faunas, although only a small number of the Chester forms as a whole are considered. , In the selection of species for discussion in this chapter a special effort has been made to choose those which have a direct bearing upon certain ques- tions of correlation concerning which some differences of opinion have de- veloped. The differences center around the correlation of the Renault and Golconda limestones. In considering the questions connected with the corre- lation of the Renault, certain Ste. Genevieve and even St. Louis limestone species have to be discussed, as well as a few from the Paint Creek formation, and all of these will be grouped together as "Lower Chester and related species/' The consideration of the correlation of the Golconda limestone involves a comparison of species common to the Golconda and the Lower Okaw limestone of Randolph County, and these forms will be discussed under the heading "Golconda and Lower Okaw species." In addition to these two groups of species which will be treated in the following pages, a number of "Miscellaneous Chester Species" will also be considered. These are species which are particularly characteristic of some one of the Chester formations, and which either have been inadequately described in earlier literature or need some further treatment. 313 314 GEOLOGY OF HARDIN COUNTY LOWES CHESTER AND RELATED SPECIES Amplexus geniculatus Worthen Plate V, Figs. 31-33 .1890. Amplexus geniculatus Worthen, Geol. Surv. 111., vol. 8, p. 82, pi. 10, figs. 7-7a. 1905. Amplexus geniculatus Ulrich, U. S. Geol. Surv. Prof. Paper No. 36, p. 47, pi. 5, figs. 15a-d. This little coral, one of the best index fossils of the Shetlerville forma- tion, is so characteristic that the faunal zone of the Shetlerville has been referred to in the foregoing pages as the Amplexus geniculatus zone. The species is characterized by its slightly tapering, almost cylindrical form, and not infrequently the specimens show abrupt, geniculate bends, as in figure 32 of the accompanying plate. Horizon and locality. — Shetlerville formation, Hardin County, Illinois, and Crittenden County, Kentucky. Lower Chester, Union County, Illinois. Pentremites princetonensis Ulrich Plate IV, Figs. 1-7 1905. Pentremites florealis (part) Ulrich, U. S. Geol. Surv. Prof. Paper No. 36, p. 58, pi. 6, figs. 18 a, c (not figs. 18 b, d-f). 1918. Pentremites princetonensis Ulrich, Ky. Geol. Surv., Miss. Form. West. Ky., p. 243, pi. 2, figs. 8-14. (Name and illustrations published without description.) Description. — Body small, subovate to sub-bipyramidal in form, higher than wide, the greatest width at the bases of the ambulacra. The dimensions of two nearly perfect specimens are: height 11.6 mm., and 9.9 mm., greatest width 9 mm. and 7.2 mm., length of ambulacra 7.8 mm. and 5.8 mm. Dorsal region of the body, below the ends of the ambulacra, obpyramidal in form, the five faces flat or very slightly concave in transverse direction, be- coming slightly subpedunculate at the extreme base, with the apex of the pyramid truncated for the stem attachment; the basal plates occupying about one-half or less than one-half the distance from the stem facet to the ends of the ambulacra; the five angles separating the faces of the dorsal pyramid are narrowly rounded and become much more prominent as they approach the ambulacral extremities. The ventral region, from the ends of the am- bulacra to the summit, with converging profile, the sides becoming increasingly convex towards the summit which is truncate. Ambulacra very gently convex with a distinct median groove, each of the two sides convex with the slope to the median groove short and abrupt, while that to the lateral margins is long and gentle ; the side-plates eleven or twelve in the space of 3 mm. The inter- ambulacral areas gently concave. The deltoid plates one-half or less than one- PALEONTOLOGY 31 5 half the length of the ambulacra, their dorsal margins meeting at the middle line in an angle less than a right angle. Remarks. — Among the specimens of Pentre mites from the Fredonia limestone and the "Upper Ohara" limestone of Ulrich, which might be re- ferred to the species P. princetonensis, there are three rather distinct types which may be separated, although intergrading examples may be found which serve to connect the three forms. Ulrich referred all three of these forms to P. florealis in 1905 1 but in his recent Kentucky report he has distinguished them as three different species, P. princetonensis, P. pulchelhis, and P. pinguis. He has published no descriptions of any of the species, and in the case of P. princetonensis the holotype is not indicated among the specimens illustrated. Considering the first of the figures designated by the name P. prince- tonensis in the Kentucky report, (fig. 8 on plate 2) as the holotype, this name may be applied to those specimens with the deeper dorsal region which is obpyramiclal in form. The more elongate specimens with proportionately shallower dorsal region constitute the species called P. pulchellus, and the shorter and broader specimens with more nearly flat dorsum are P. pinguis The great variability of all these forms is shown in the fact that among the original illustrations of the species there is greater difference shown among specimens that are designated by the same name than between certain of the specimens that are referred to different species. For instance, figures 6 and 7 of Ulrich's Plate 2, called P. pulchelhis, are much more like figures 8 and 9 that are called P. princetonensis, than they are like the specimen designated as the holotype of P. ptdchellus. A critical examination of many hundreds of specimens of these pentre- mites has brought out certain features which are of value for stratigraphic purposes. The typical P. princetonensis, with its obpyramidal dorsum is most widely distributed in the Fredonia limestone of Ohio Valley and in the typical Ste. Genevieve limestone of Mississippi Valley. It is also present in its typical form in the St. Louis limestone, although all pentremites are rare in that horizon, and neither of the other two forms have been recosmized there. It continues upward into the Shetlerville, and is locally present in the Eenault. Pentremites pnxlchellus is common in the Fredonia and in the Shetlerville and may be present in the .Renault, but it is not a conspicuous member of that fauna. P. pinguis occurs in great numbers in the Shetlerville and is common in the Eenault, but is much less common in the Fredonia. Ulrich has used these three forms in his attempt to establish the fauna! connection between the Fredonia and the u Upper Ohara," both of which he places in the Ste. Genevieve limestone, but P. princetonensis connects the Fredonia fauna with that of the St. Louis limestone as closely as or more closely than the connection with the higher beds. The whole situation is 1 Prof. Paper, U. S. Geol. Surv., No. 36, pi. 6, figs. 18a-f. 316 GEOLOGY OF HARDIN COUNTY one which shows the folly of attempting to establish faunal relationships and correlation upon single fossil forms. The true faunal relations must he established through a critical examination of the whole fauna, and when this is done it leaves no basis whatsoever for associating the "Upper Ohara" with the Ste. Genevieve limestone. Formation and locality. — St. Louis limestone, St. Louis, Missouri, and above Little Rock, Ste. Genevieve County, Missouri ; Ste. Genevieve limestone, Fountain Creek, Monroe County, Illinois, and Ste. Genevieve County, Mis- souri; Fredonia limestone, Cedar Bluff quarry, near Princeton, Kentucky, and Rosiclare, Illinois; Shetlerville formation, Fairview bluff, Hardin County, Illinois, and elsewhere; Renault formation, Johnson County, Illinois, and elsewhere. Pentremites pulchellus Ulrich Plate IV, Figs. 14, 28-30 1905. Pentremites florealis (part) Ulrich, U. S. Geol. Surv., Prof. Paper No. 36, p. 58, pi. 6, figs. 18 b, d (not figs. 18a, c, e-f). 1918. Pentremites pulchellus Ulrich, Geol. Surv. Ky., Miss. Form. West. Ky., pi. 2, figs. 1-7. (Name and illustrations published without description.) Description. — Body small, elongate subovate in form, much higher than wide, the greatest width at or a little above the bases of the ambulacra. The dimensions of a practically perfect individual are: height 15 mm., greatest width 10 mm., length of ambulacra 7.7 mm., greatest width of ambulacra 2.9 mm. Dorsal region of the body, below the ends of the ambulacra, with the sides sloping inward to the stem facet, the profile nearly straight distally, becoming concavely curved as it approaches the base; subpedunculate, and truncated for the stem attachment; five shoulder-like elevations pass from the base to the extremities of the ambulacra, becoming more strongly raised distally ; the interambulacral surfaces between gently concave ; the basal plates reaching about half way from the stem facet to the extremities of the am- bulacra. Ventral region of the body, from the ends of the ambulacra to the summit, very much higher than the dorsal region, the sides converging towards the summit, the convergence gentle below, becoming more abrupt towards the summit, giving to the sides a convex profile which becomes more strongly curved above. Ambulacra very gently convex or nearly flat, with a shallow median groove, the lateral surfaces with a very short, rather abrupt slope to the groove and a long gentle slope to the lateral margins; the lateral grooves 9 or 10 in the space of three millimeters, directed transversely at the summit, becoming progressively more oblique towards the distal extremity of the ambulacra where they are directed upward as they approach the median groove. Interambulacral areas gently concave transversely, the concavity PALEONTOLOGY 317 becoming more abrupt ventrally because of the slightly raised lateral margins ; deltoid plates longer than wide, but less than half the length of the ambulacra, the basal margins meeting at the middle of the interambulacral area in an angle of less than 90°. On well-preserved specimens the surface of the radial and deltoid plates is marked by exceedingly fine, regular lines running parallel with the margins, the oblique lines on the radials which are parallel with the radio-deltoid sutures being stronger and somewhat more rugose than the longitudinal lines upon the same plates. Remarks. — The above description of this species has been drawn up from specimens collected from the Fredonia limestone near Princeton, Kentucky, which agree very closely with figures 1, 2, and 3 of Ulrich's illustrations, figure 2 being designated as the holotype. Ulricas illustrations exhibit a considerable amount of variation and include figures of specimens from the "Upper Ohara" limestone as well as from the Fredonia. In some of the figures the basal profile shows much less concavity than the more typical examples, the lines being nearly straight, and his figures 6 and 7, especially 7, really resemble his illustrations of P. princetonensis more closely than they do that of the holotype of the species. The ambulacra are perhaps a little flatter in the typical princetonensis than in pulchellus, but in a large collec- tion of hundreds of specimens from the Shetlerville division of the "Upper Ohara," it is possible to find every gradation between the two forms. The specimens from the Shetlerville formation that are referred to the species have a distinctly shallower dorsal region than do those from the Ste. Genevieve. Horizon and locality. — Fredonia limestone, Cedar Bluff quarry, near Princeton, Kentucky, and Rosiclare, Illinois. Shetlerville formation, Fair- view bluff, Hardin County, Illinois, and elsewhere. Pentremites pingtjis Ulrich Plate IV, Figs. 8-12 1905. Pentremites fiorealis (part) Ulrich, U. S. Geol. Surv., Prof. Paper No. 36, p. 58, pi. 6, figs. 18 e-f, (not figs. 18 a-d). 1918. Pentremites pinguis Ulrich, Ky. Geol. Surv., Miss. Form. West. Ky., p. 244, pi. 2, figs. 16-17. (Fame and illustrations published with- out description.) Description. — Body of medium size or smaller, subovate in form, higher than wide, the greatest width at or very near the bases of the ambulacra. The dimensions of a very complete specimen are: height 11.9 mm., greatest width 9.5 mm., length of ambulacra 9 mm., greatest width of ambulacra 3 mm. Dorsal region of the body, below the ends of the ambulacra, sloping inward nearly to the point of attachment of the stem, with a straight or slightly concave profile, the extreme basal extremity slightly produced or subpedun- culate, the basal plates occupying about one-half the distance from the stem 318 GEOLOGY OF HARDIN" COUNTY facet to the base of the ambulacra; five shoulder-like elevations extending from the base to the extremities of the ambulacra, becoming more strongly raised distally, the interambulacral surfaces flat or slightly concave. Ventral region of the bod)-, from the ends of the ambulacra to the summit, with the sides gently convex and converging to the subtruncate summit. Ambulacra nearly flat, with a shallow median groove, the two sides very gently convex with the curvature a little more abrupt to the median groove ; the side plates about ten in the space of 3 mm. Interambulacral areas transversely a little concave in well preserved specimens, the margins slightly elevated into low, keel-like, bordering ridges in the ventral half, the elevations on the two sides joining at the apex of the deltoid plates, below which there is a slight de- pression. Deltoid plates somewhat variable, more or less than one-half the length of the ambulacra, their dorsal margins slightly curved., and meeting at the dorsal apex in an angle varying from considerably less to more than 90°. Remarks. — Among the six specimens illustrated by Ulrich in 1905 and referred to Pentremites florealis 1 , the two shorter and broader ones with the flatter dorsal region, have been referred to the new species P. pinguis in his re- cent Kentucky report. These two specimens, illustrated in the earlier report, really represent a very abundant pentremite of the Shetlerville fauna of Hardin County and elsewhere, better than does the specimen designated as holotype of the species in the recent Kentucky report. Since the name P. florealis has no standing whatsoever, if the form is to be recognized as a species, Ulricrr's newly proposed name is as good as any, although no descrip- tion of it has been published. Among the mairy hundreds of specimens of pentremites from the Shetlerville formation that have been examined, com- plete gradations may be selected which connect this form with both P. prince- tonensis and P. pidcliellus, and doubtless some observers would hesitate to consider these forms as anything more than varieties of a single species. Some names seem to be desirable, however, for the different forms, especially as they really have some stratigraphic value. P. princetonensis is more com- monly represented in the St. Louis and Ste. Genevieve limestones, but is not restricted to these horizons. P. pttlchellus is best represented in the Ste. Genevieve and Shetlerville, while P. pinguis occurs most abundantly in the Shetlerville and Eenault. On plate IV, examples of this species from the "Upper Ohara" of Hardin County, and the Renault limestone of Monroe County, are illustrated side by side in order to show their identity. Pentremites pinguis is one of the commoner species in the Renault of Hardin County, but it is associated with several other forms, among which is the earliest known of the "pyriform" types of the genus, this association being identical with that found in the typical Renault of Monroe and Randolph counties, and the same "pyrifornr" species occurs in both regions. 1 L T . S. Geol. Surv., Prof. Paper, No. 36, pi. 6, figs. 18a-f. PALEONTOLOGY 319 Horizon and locality. — Shetlerville formation, Fairview bluff, Hardin County, Illinois, and elsewhere. Renault limestone, Hardin, Monroe, and Randolph counties, Illinois. Pentremites godoni (Def ranee) Plate IV, Figs. 31-36, 47 1808. "An asterial fossil," Parkinson, Org. Rem., vol. 2, p. 235, pi. 13, figs. 36-37. 1819. Encrina Godoni Defrance, Diet. Sci. Nat., vol. 1-1, p. 469. 1820. Encrinites florealis von Schlotheim, Petrefactenkunde, vol. 2, p. 38. 1918. Pentremites planus Ulrich, Ky. Geol. Surv., Miss. Form. West. Ky., pi. 5, figs. 1-13. (Name and illustrations published without de- scription.) Description. — Body of medium size, subovate in form, the height and width subequal or higher than wide ; an occasional example wider than high, the greatest width at or a little above the ends of the ambulacra. The dimen- sions of a practically perfect specimen are: height 16.6 mm., greatest width 14.5 mm., length of ambulacra 14 mm., greatest width of ambulacra 4.4 mm. Dorsal region of the body, below the ends of the ambulacra, very short, in some examples nearly flat, with a short, slightly projecting, subpedunculate central region; in profile the slopes from the ambulacral extremities to the central projection nearly straight or gently concave, in some examples the •curvature being slightly interrupted near the middle; between the lower extremities of the ambulacra and the center of the base, the dorsal surface is raised into five shoulder-like ridges which become nearly or quite obsolete about half way to the stem facet, the interambulacral surfaces between these ridges being gently concave towards the outer border; the basal plates reach- ing about half way from the center of the dorsal surface to the lower extremi- ties of the ambulacra. Ventral region of the body, from the ends of the ambulacra to the summit, much higher than the dorsal region, the sides converging towards the summit with a gentle convex curvature which in- creases above. The ambulacra nearly flat transversely towards the summit, becoming gently convex towards the lower extremities, with a shallow median groove; the lateral grooves 7 to 10 in the space of three millimeters, situated transversely towards the summit, becoming progressively more oblique towards the lower end of the ambulacra, where their inner extremities are slightly curved and directed towards the summit. Interambulacral areas gently concave transversely, with their margins somewhat abruptly elevated, this elevation becoming progressively more pronounced in a ventral direction, in some examples being elevated in an almost knife-like edge along the sides of the deltoid plates. The deltoid plates a little less than half the length of the ambulacra, their dorsal margins straight or convexly curved, and meeting in the center of the interambulacral area in an angle of more or less than 90°. 320 GEOLOGY OF HARDIN COUNTY Remarks. — In its immature condition this species is essentially indis- tinguishable from P. pinguis, except that the ambulacra are perhaps slightly flatter towards the summit in most individuals. These immature examples have the dorsal region proportionally deeper than in the mature specimens, and the similarity of this species through its young individuals with P. pinguis, is such as to leave no doubt regarding the genetic relations of the two species. Pentremites godoni has been a much abused species, since almost any pentremite having a subglobular form has been referred to it by authors. The species was originally based upon two figures published by Parkinson in 1808. No name was given to the object figured by this author, but it was referred to in the explanation of the plate as "An asterial fossil from America : probably of the nature of the Encrinus/' In the body of the text the fossil is described as follows : "The nature of the calcareous fossil represented in Plate XIII figs. 36 and 37, is very ambiguous. I was favored with it by Dr. Woodhouse, of Philadelphia, who, avowing his inability to ascertain anything respecting its original mode of existence, informs me that it was obtained from Kentucky, where similar bodies are frequently found of a large size, and that they are there considered as a species of petrified nuts. "This fossil is somewhat of a conical roundish form, the center of its base terminating in a small round projection pierced in its middle, with a little open- ing into the center of the fossil; from this projection the base extends nearly horizontally to five prominent points, between each of which exists a shallow depression. At the apex of the cone five small openings are placed at the angles, formed by the meeting of the lines, which bound five long triangular surfaces, which, commencing at the summit of the fossil, are disposed tapering, down the sides, and terminate in the projecting points which are placed round the base. Along the middle of each of these surfaces, a grooved line passes, from which upwards of forty minute processes on each side pass to the lines which bound these surfaces at their sides. "The opinion which I formed on the first view of this fossil was, that it be- longed to some animal approximating to the encrinus. The central projection at its base has, however, suffered so much by friction as not, I believe, to show its original surface; and hence it is impossible to determine at present anything respecting the substance with which it was connected at this point, or the kind of articulation which was here employed. It, however, very nearly resembles the smaller modiolus, from Ireland, which is represented on Plate XVII." The illustration of this unique fossil form published by Parkinson, at once attracted the attention of other students of fossils in Europe, and in 1918 Defrance gave a name to the illustration, calling it Encrina Godoni. This author seems to have had no specimens of the fossil for study or descrip- tion, but simply gave it a name on the basis of the illustration that had been published by Parkinson. A year later, 1820, von Schlotheim did the same thing that Defrance had done, gave a name to Parkinson's illustrations, with- out having access to specimens for study, the name he applied to it being Encrinites florealis. These two names, therefore, godoni and fiorealis, are PALEONTOLOGY 321 strictly synonyms, and inasmuch as neither of the authors of these names seem to have possessed type specimens upon which to base his species, but depended upon Parkinson's illustrations, the original specimen used by Parkinson must be considered as being the holotype of each of the species, and Parkinson's figures must be the type illustrations. In interpreting the species Pentremites godoni, Parkinson's illustrations must be the guide, and therefore copies of these illustrations are given upon the accompanying plate. Ulrich, however, has not followed this course in his interpretation of the species, but has arbitrarily selected a form which occurs in abundance near Huntsville, Alabama, as representing P. godoni. There is no doubt but that this Huntsville form has been called P. godoni by many authors, and it is not improbable that Defrance himself would have referred such specimens to the species if they had been in his hands, but if we are to follow as great refinement in separating the species of Pentremites as Ulrich has used, then these Huntsville specimens certainly are not con- specific with Parkinson's figures. Among other differences the ambulacra of the form Ulrich has designated as P. godoni are broadly V-shaped trans- versely, while the true P. godoni has gently convex ambulacra. Parkinson's figures represent much more accurately the form for which Ulrich has pro- posed the new name P. planus, and if his figure 3 be compared with Parkin- son's figure 37, due allowance being made for the difference in method of illustration, and the slight difference in position of the specimens, it may be seen how closely alike these forms are, and if the name P. godoni is to be used in a more restricted manner than has been the usual custom, it must be applied to the form which Ulrich has named P. planus. Among the collections from the Chester group of Illinois that have been carefully made during the past decade, this form has been collected in greatest abundance and in the best state of preservation from a locality in the Paint Creek formation on Prairie du Long Creek, in St. Clair County, one and one-half miles northwest of Floraville. Figures 6, 7, 8, and 9 of Ulrich's series of illustrations of P. planus, are from this locality, and it occurs elsewhere in the same formation in Illinois. Individuals occurring together in the same bed at the Floraville locality, exhibit a wide range of variation, and the extremes among these variations would undoubtedly be considered as distinct species if they were found segregated in different locali- ties or in different horizons. At Floraville, however, the variations intergrade into one another and no sharp lines of division can be drawn, and after a long and critical study of large collections, one is forced to the conclusion that they represent a single highly variable species, which the writer prefers to call P. godoni rather than P. planus, the name introduced by Ulrich. As so interpreted, P. godoni is also present in the Renault faunas of Illinois, but the Renault examples of the species are commonly smaller than those in the —21 G 322 GEOLOGY OF HARDIN COUNTY Paint' Creek. Complete gradations can be traced from the typical form of P. godoni in the Paint Creek fauna, through a number of forms which Ulrich has named and illustrated, into P. princetonensis. It is not to be understood that the present writer would advocate the rejection of the names that Ulrich has proposed as specific designations, for they will doubtless be convenient for the reference of certain forms in the remarkably variable assemblage. A study of many hundreds of these specimens leads to the conclusion that we have under consideration an extraordinary plastic genetic group which exhibits a wide range of contemporaneous variations, associated with progressive mutations which change with passing time, neither the varieties nor the mutations being separated by hard and fast lines. The entire assemb- lage of individuals may be divided into innumerable groups, call them varieties or species as may seem best, and a different grouping is suggested from a study of almost every separate collection. The larger, more compre- hensive group may be considered as beginning at least as early as Fredonia time, perhaps earlier, although members of the genus Pentremites are rare in the St. Louis limestone. In the Fredonia limestone the dominant type is the one called P. princetonensis by Ulrich, although characteristic indi- viduals of P. pulchellus are associated with it. Perfectly typical examples of P. princetonensis are also known from the St. Louis limestone. In the Shet- lerville and Eenault formations P. pinguis is one of the leading forms, although perfectly typical examples of P. pulchellus and P. princetonensis are also present, and other specimens which foreshadow or may even be re- ferred to the typical P. godoni (P. planus Ulrich) are also present. In the Paint Creek fauna we find the great development of the typical P. godoni. A continuation of the same genetic line is met with in the Golconda limestone, although in the middle and upper Chester faunas many quite different types of Pentremites make their appearance. For purposes of correlation it is important to take cognizance of the fact that the same stage of development of this group of forms is represented in the "Upper Ohara" limestone of Hardin County, and in the Eenault lime- stone of Monroe and Eandolph counties, and this circumstance, taken to- gether with the fact that every species in the "Upper Ohara" faunas, save about three that are wholly restricted to the Shetlerville fauna, are known also in the Eenault faunas of Monroe County or in the Paint Creek of the same general region, establishes the contemporaneity of these two formations as firmly as such contemporaneity can be established by paleontological evi- dence. Horizon and locality. — Eenault limestone, Monroe, Eandolph, and Hardin counties, Illinois. Paint Creek formation, St. Clair, Monroe, and Eandolph counties, Illinois. PALEONTOLOGY 6X6 Pentremites buttsi Ulrich Plate IV, Figs. 16-20 1918. Pentremites buttsi Ulrich, Ky. Geol. Surv., Miss. Form. West. Ky., pi, 2, figs. 18-19. (Name and illustrations published without de- scription.) Description. — Body of medium size or smaller, subpyriform, much higher than wide, the greatest width at the lower extremities of the ambulacra. The dimensions of a nearly perfect specimen are: height 15 mm., greatest width 10 mm., length of ambulacra 9.3 mm., greatest width of ambulacra 3.2 mm. Dorsal region of the body obpyramidal, the slope from the stem facet to the lower extremities of the ambulacra nearly straight or very slightly concave; the interambulacral surfaces of the dorsal region are flat or very slightly concave transversely, the five surfaces being separated by five rounded, shoulder-like ridges which are best developed just beneath the extremities of the ambulacra and become nearly obsolete towards the stem facet; the basal plates reaching a little less than half way from the stem facet to the bases of the ambulacra. Ventral region a little higher than the dorsal in mature individuals, with the sides nearly straight below and converging towards the summit, curving inward above to the subtruncate summit. Ambulacra flat or very slightly convex transversely, with a shallow median groove; lateral grooves seven to ten in number in a space of three millimeters, directed transversely at the summit, becoming progressively more oblique towards the lower extremities of the ambulacra where they are directed upward as they approach the median groove. Interambulacral areas flat or very slightly concave in a transverse direction; deltoid plates small, their exposed surfaces less than one-third the length of the ambulacra, their dorsal margins straight and meeting at the median line in an angle a little greater than 90 c . Remarks. — The specimen which has been used in drawing up the above description is from the Eenault formation at Fairview, Hardin County, from the bed which Ulrich has called zone 4 of his Ohara, and is, therefore, from exactly the same stratigraphie position from which the holotype of the species is recorded. The holotype is a slightly more elongate specimen than this one, with the dorsal region relatively a little shorter, but the proportionate length of the dorsal region of this specimen is less than in the second example illustrated by Ulrich. There can be no question regarding the inclusion of the specimen here illustrated in the same species with the holotype of P. buttsi, though it is perhaps doubtful whether this name has good standing. Two other specimens that are illustrated here are from the Eenault formation in St. Clair and Monroe counties, and probably would not be identified as P. buttsi by Ulrich, but rather as P. symmetricus or P. ivelleri, because they occur in beds that are assumed by him to be considerably younger than the "Upper Ohara" of Hardin County. No unprejudiced observer, however, 324 GEOLOGY OF HARDIN COUNTY could fail to recognize all three specimens as being conspecific. Still other specimens from the Kenault formation of Union County are illustrated, which are clearly conspecific with those from Hardin and from Monroe and St. Clair counties. This is the earliest representative of the elongate type of pentremite that has been referred commonly to P. pyriformis, and it is really a serious ques- tion whether that name should not be retained for these specimens. Ulrich has arbitrarily restricted the name P. pyriformis for a form that is especially abundant near Huntsville, Alabama, and has at the same time proposed no less than six new names for other pyriform species. These names are not accompanied by descriptions, and from the figures alone it is not possible to determine just how the specific lines have been drawn. Certain figures are associated under one specific name that exhibit less resemblance than do other figures that are included in different species. The original description of P. pyriformis by Say, was not accompanied by figures, and the original definition of the species is too general to make it possible to determine just what form is referred to, if the specimens are to be differentiated along the lines proposed by Ulrich. This form occurs in both the Renault of the Mississippi Valley section, and in the Renault or "Upper Ohara" of the Ohio Valley, and in both places it is associated with P. princetonensis, or P. pinguis. Horizon and locality. — Renault limestone, Monroe, St. Clair, Union, and Hardin counties, Illinois. Pentremites symmetricus Hall Plate IV, Figs. 26-27 1858. Pentremites symmetricus Hall, Geol. Iowa, vol. 1, pt. 2, p. 694, pi. 25, fig. 12. Description. — Body of medium size or smaller, much higher than wide, the ventral region higher than the dorsal, the greatest width at the extremities of the ambulacra. The dimensions of a rather small but very complete specimen are : height 18.5 mm., greatest width 12.8 mm,, length of ambulacra 12.6 mm., width of ambulacra 3.8 mm., distance from center of stem facef to base of ambulacra 9.5 mm. Dorsal region obpyramidal, with the faces of the pyramid well defined towards the line separating the dorsal and ventral regions, and gently concave transversely between the lower extremities of the ambulacra; in profile the sides of the dorsal region are nearly straight or very slightly curved through the most of their length, commonly becoming concave for a short distance as they approach the margin of the stem facet. Ventral region with the sides gently convex and converging to the truncate summit; the ambulacra gently convex transversely with a moderately deep median groove, the convexity becoming somewhat greater towards the lower PALEONTOLOGY 325 extremities; each of the two lateral surfaces gently convex, with the inner margins sloping rather abruptly to the median groove; longitudinally the ambulacra are very gently convex with the curvature increasing a little towards the summit; the side plates from six to nine in number in the space of 3 mm., the lateral grooves nearly transverse in direction towards the summit, becoming quite oblique towards the distal extremities. The inter- ambulacral areas gently concave transversely, with very slightly raised lateral margins, the tips of the deltoids not quite reaching to the summit. The deltoid plates nearly or quite one-half the length of the ambulacra, their dorsal margins straight and meeting at the median line in an angle greater than a right angle. Remarks. — This is a rather common form in the Lower Chester faunas, and may be recognized by its slightly concave interambulacral regions, its dorsum shorter than the ventral region, and by its somewhat coarser am- bulacral side plates and lateral grooves than in most species. In all pentre- mites these grooves vary in coarseness in each ambulacrum, being more closely placed towards the summit than in the distal portions of the ambulacra, and the numbers given in the descriptions presented here are the numbers that may be counted in a unit distance of 3 mm. near the summit and again near the distal extremity of the same ambulacrum. Observations upon many specimens have shown that the size of these plates and grooves is a rather constant character in the various species. Horizon and locality. — Renault and Paint Creek limestones, St. Clair, and Monroe counties, Illinois. Pentremites pykaaeidatus Ulrich Plate IV, Figs. 21-24 1905. Pentremites py ram id at us Ulrich, IT. S. Geol. Surv., Prof. Paper Xo. 36, p. 61, pi. 7, figs. 12-14. (Xame and illustration published without description.) Description. — Body of medium size, bipyramidal in form, much higher than wide, the greatest width at the extremities of the ambulacra. The di- mensions of a nearly perfect specimen are: height 20.8 mm., greatest width 15.1 mm., length of ambulacra 12.6 mm., greatest width of ambulacra 4.5 mm. Dorsal region of the body, below the ends of the ambulacra, sub- pyramidal in form, the faces of the pyramid not sharply defined except towards the ambulacral extremities where the surfaces are nearly flat trans- versely, the profile of the dorsum with nearly straight sides, the basal plates reaching about half way or less from the stem facet to the ends of the ambulacra. Ventral region about equal to or somewhat higher than the dorsal, the sides converging to the truncate summit with a very gentle convex curvature. Surfaces of the ambulacra depressed convex, with the median 326 GEOLOGY OF HARDIN COUNTY groove rather strong. Each of the lateral surfaces unsymmetrically depressed convex, the curvature to the median groove more abrupt than to the lateral margins, side plates eight to ten in the space of 3 mm., the lateral grooves nearly straight and transverse towards the summit, becoming oblique distally. The interambulacral regions fiat or very slightly concave in a transverse direction, with the lateral margins very slightly elevated towards the tips of the deltoid plates. The deltoid plates rather short, their tips not reaching to the summit. Remarks. — In the absence of any definition of this species accompanying the illustrations to which the name was originally applied, the student of these forms is left in some uncertainty. Judging from specimens in the hands of the writer, however, as well as from the original illustrations, one of the more essential characters seems to be the very flat interambulacral regions ventrally, this flatness continuing across the dividing line for some distance into the dorsal region. Accompanying this character are the short deltoid plates, and the very gently convex ambulacra. The specimens that are illus- trated herewith are all from the Paint Creek formation, two from Monroe County and two from Johnson County. The larger of the Johnson County specimens has the interambulacral regions a little less flat than the others, with a slight tendency to a concave profile of the dorsal region, in these characters approaching the forms to which TJlrich has given the name P. lyoni, which apparently differs from P. pyramidatus chiefly by reason of its gently concave interambulacra. None of the specimens here illustrated are as large as the original specimens figured by TJlrich, but this difference can scarcely be considered as justification for any specific separation. Horizon* and locality. — Paint Creek formation, Monroe and Johnson counties, Illinois. Pentremites gemmiformis Hambach Plate X, Figs. 9-10 1884. Pentremites gemmiformis Hambach, Trans. St. Louis Acad. Sci., vol. 4, p. 553, pi. D, fig. 5. This species is characterized by the elongate dorsal region with concave profile, and especially by the broad basal truncation, a short distance above which there is likely to be a slight constriction, giving to the lower portion of the body a slight bulbous appearance. So far as known the species occurs only in the Paint Creek horizon. Horizon and locality. — Paint Creek limestone, St. Clair and Randolph counties, Illinois. PALEONTOLOGY 327 Mesoblastus glabek (M. & W.) Plate VIII, Fig?. 25-28 1869. Granatocrinus glaber Meek and Worthen, Proc. Acad. Nat. Sci. Phil., p. 91. 1873. Granatocrinus glaber Meek and Worth en, Geol. Surv. 111., vol. 5, p. 537, pi. 20, fig. 11. 1905. Mesoblastus glaber ? Ulrich, Prof. Paper, U. S. Geol. Surv., No. 36, p. 57, pi. 6, figs. 13-17. The type specimens of this species are preserved in the collections of the Illinois State Museum of Natural History, at Springfield, and they have been made available to the writer for study by Dr. A. R. Crook, the Director of the Museum. These specimens are from Hardin County, Illinois, and are undoubtedly examples of the species that is commonly met with in the Shetlerville formation. The specific name, however, is a misnomer. The type specimens are somewhat worn so that they are essentially smooth, but perfectly preserved specimens are distinctly marked by rather strong lines of growth upon the deltoid plates and upon the distal portion of the radial plates, the stronger of these lines being parallel with the suture between the deltoid and radial plates. Horizon and locality. — Fredonia limestone, Cedar Bluff, near Princeton, Kentucky; Shetlerville formation, Hardin County, Illinois, and Crittenden County, Kentucky; Renault formation, Monroe County. Illinois. Dizygocbinus peesculptus Ulrich Plate V, Figs. 15-17 1918. Dizygocrinus persculptus Ulrich, Ky. Geol. Surv., Miss. Form. West. Ky., pi. 2, figs. 24-26. (Name and illustration published without description.) Description. — Calyx of medium size, unequally biturbinate in form. The dimensions of a nearly complete body are : greatest diameter at arm bases 19.3 mm., height from stem facet to base of anal tube 18.6 mm., height of dorsal cup 5.9 mm. Dorsal cup depressed subturbinate in form, the sides sloping in nearly straight lines from the basal disk nearly to the arm bases where there is a slight flaring of the sides; the sutures impressed, the plates convex and more or less rugose, those of the radial series slightly elevated medially to form five obscure ridges leading from the radials to the bases of the arms. Basal disc rather large, the stem facet occupying about one-half its diameter; radial plates wider than high, followed by two costals in each ray, both of which are wider than high; distichals two in each series. Arm openings four from each ray, making twenty in all, the arms unknown. Interbrachial areas occupied by one large plate in contact with the radials, followed by two or three smaller plates which in some places, at least, are 328 GEOLOGY OF HARDIN COUNTY slightly joined with the plates of the tegmen. Posterior interradial area with an anal plate somewhat smaller than the radial s, this followed by three smaller plates in the second range, and these again by two or three still smaller ones; there are apparently some still higher plates in the posterior interradial area, but they are not clearly shown in the specimens under observation. Tegmen subcorneal, considerably higher than the dorsal cup, the plates small, convex, or produced into low, thorn-like spines. Anal tube not preserved. Remarks. — In the report on Mississippian Formations of Western Ken- tucky, where Ulrich names and illustrates this species without description, it is reported as coming from the upper part of the "Ohara limestone," two and one-half miles north of Smithland, Kentucky. He also has a number of specimens in his collection, either of this species or I), super sies,, from near Levias, Crittenden County, Kentucky, which are labeled as coming from the "Upper Ohara." The writer's visit to the Levias locality during the past season, in company with Mr. Chas. Butts, showed to the satisfaction of both, that Ulrich had mistaken and mapped the Rosiclare sandstone at this locality for the "Cypress" (Bethel), and that consequently the "Upper Ohara" as he has identified it, is really Fredonia. The locality north of Smithland has not been reexamined, but it seems more than probable that some similar mistake was made in the horizon at that locality. Most diligent search during four years field work, at every locality that has been available, has brought out the fact that every specimen of Bato- crinoid that has been found in the "Upper Ohara," the Shetlerville and. Renault formations of this report, and so far as known they occur only in the Shetlerville, is a member of the species Globo>crinus unionensis. On the other hand, every specimen in the Ste. Genevieve, below the "Upper Ohara," is is a Dizygocrinus, members of which genus have not lost the first costal plates. Ulrich's contention that these crinoid species constitute a bond of union between the "Upper Ohara" and the remainder of his Ste. Genevieve forma- tion, has not been confirmed by these recent field studies. Ulrich's collections were made a score or more years ago, at a time when the stratigraphic prob- lems involved were not clearly defined, and it has been demonstrated in the Levias locality at least, that an error was made in stratigraphic interpretation. Under the circumstances, therefore, none of the evidence which he has pre- sented can be accepted as being conclusive until it has been confirmed by new observations. Dizygocrinus persculptus differs from D. superstes which occurs in the same horizon, in its more depressed dorsal cup, and in the more rugose nature of the surface ornamentation of the plates. Horizon and locality. — Fredonia limestone, Eosiclare, Illinois, and Cedar Bluff quarry, near Princeton, Kentucky. PALEONTOLOGY 329 Dizygocrinus superstes Ulrich Plate V, Figs. 13-14 1918. Dizygocrinus superstes Ulrich, Ky. Geol. Surv., Miss. Form. West. Ky., pi. 2, figs. 21-23. (Name and illustration published without description.) Description. — Calyx small or of medium size, subequally biturbinate ; the dimensions of a nearly complete body are : greatest diameter at arm bases 10.7 mm., height from stem facet to base of anal tube 9.8 mm., height of dorsal cup 4.3 mm. Dorsal cup subturbinate, the sides sloping in nearly straight lines from the basal disc nearly to the arm bases where there is a slight flaring of the sides ; the sutures impressed, the plates convex and more or less nodose, those of the radial series slightly elevated medially to form five obscure ridges leading from the radial plates to the bases of the arms. Basal disc small, the stem facet occupying about half its diameter; radial plates wider than high, followed by two costals in each ray, both of which are wider than high; distichals two in each series, except in the anterior ray where there are three. Arm openings two in the anterior ray, those of the postero-lateral rays not clearly exhibited but on one side there seems to be four openings, on the other not so many; four is probably the normal num- ber, making eighteen in all. The arms unknown. Interbrachial areas occu- pied by one large plate in contact with the radials, followed by three or four smaller plates arranged in two or three series. Posterior inter-radial area with an anal plate somewhat smaller than the radials, this followed by three smaller plates in the second range and these again by three still smaller ones; there are apparently some still higher plates but they are not clearly exhibited in the specimens observed. Tegmen subcorneal, about as high as or a little higher than the dorsal cup, the plates small, convex, or produced into rather prominent, central tubercles. Anal tube not preserved. Remarks. — The specimens of Dizygocrinus that have been met with in the Ste. Genevieve limestone seem to represent two species. In the absence of any descriptions by Ulrich, these two species seem to correspond with those to which he has given the names D. superstes and D. persculptus. The two species differ in the relative height of the dorsal cup, in the ornamentation of the plates, and in the number of arm openings. In D. superstes, the species here described, the dorsal cup is relatively higher than in D. persculptus, it being nearly or quite as high as the tegmen to the base of the anal tube ; the plates too are more smoothly nodose, their surface not being so broken up with rugosities. In the best specimen available for study there are only two arm openings in the anterior ray in D. superstes, while in D. persculptus there are four openings in each ray. It is impossible to determine from Ul rich's illustration, in the absence of any description, the number of arm openings in the anterior ray of the holotype, and it is therefore possible that the two openings observed may be an abnormal character. 330 GEOLOGY OF HARDIN COUNTY When Worthen described Batocrinus unionensis he illustrated two speci- mens, which quite evidently are different species and are here considered as representing different genera. The specimen without arms, which quite cer- tainly was the one used for the most part in the preparation of the specific definition, has here been accepted as the holotype and has been placed in the new genus Globocrinus. The second specimen, the one with the arms, differs from the first in having the two costal plates in each ray which are normal for the genus Dizygocrinus and its close allies. This second specimen is clearly a close relative of the two species of Dizygocrinus that are here described from the Fredonia limestone, and judging from the illustration it is more nearly related to I), sup erst es, and is not unlikely a member of this species. Worthen^s specimens are said to come from the "St. Louis division of the Lower Carboniferous, Union County, Illinois,' 7 but at that time lie included all the strata below the first massive sandstone formation of the section, which was believed to be the basal member of the Chester group, in the St. Louis limestone. This first massive sandstone in Union County is the Cypress, so Worthen's so-called St. Louis would include both the Ste. Genevieve and also the lower Chester formations. In the neighborhood of Anna, in Union County, there are exposures of both the Fredonia limestone and of the Shetlerville and Renault formations, and Worthen's two speci- mens were doubtless collected from these two formations, — at least all field experience farther east would lead to such a supposition, — but both of the species must be collected again in order to confirm their exact stratigraphic positions. Horizon and locality. — Fredonia limestone, Charles Stone Quarry, near Joppa Junction, Johnson County, Illinois, and Eosiclare, Illinois. GLOBOCRINUS n. gen. This genus is a member of the family Batocrinidae,, and although its arms have not been observed it seems to be related to Dizygocrinus. The essential generic character is found in the reduction of the costals to a single plate in each ray, there being no small, quadrangular plate lying between the axillary costal and the radial, as in Dizygocrinus, or in fact, in all closely related members of the family. In the original specimen of the genotype there are only two plates following the anal in the posterior interambulacral space, this arrangement of plates being a characteristic of the family Actino- crinidae, but all other specimens examined possess the three plates that are normally present in the Batocrinidae. The loss of the middle one of these three plates is not an unusual abnormal character among the Batocrinidae, and this is undoubtedly the explanation of the condition in the type of G. unionensis. A loss or reduction in number of plates among the camerate crinoids is well recognized as a progressive character, and the fact that we have in this PALEONTOLOGY 331 species a batocrinid in which the costal s have been reduced in number to a single plate in each ray, would suggest to any student of crinoids, even if he did not know the horizon from which the species was collected, that it occurred late in the history of the family. It is in fact, the last known representative of the family, and the reduced number of brachial plates is just the sort of character that would be looked for here. The loss of these costal plates is such an unusual character that it is certainly worthy of recognition as of generic value. The name Globocrinus is given to the genus because of the subglobular form of the body. Globocrinus unionensis (Worthen) Plate V, Figs. 18-23 1890. Batocrinus unionensis Worthen, Geol. Surv. 111., vol. 8, p. 84, pi. 12, figs. 5-5a, (not pi. 13, fig. 3). (Xot D'izygocrinus unionensis W. and Sp., 1897.) Description. — Calyx small or of moderate size, subglobular or depressed subglobular in form, the dimensions of the holotype being: greatest diameter at arm bases 20 mm., height from stem facet to base of anal tube 16 mm., height of dorsal cup 8.5 mm. The dimensions of a second nearly perfect calyx are: greatest diameter at arm bases 16 mm., height from stem facet to base of anal tube 15 mm., height of dorsal cup 8 mm. Dorsal cup bowl- shaped, the sides convexly curved from the stem facet to the bases of the arms, the sutures impressed, surfaces of the plates convex, commonly flattened on top, either nearly smooth or ornamented by low, irregularly reticulating ridges. Basal disc small, the stem facet concave, occupying half or more than half of the diameter of the base; radials wider than high; one costal only in each ray, wider than high; the higher brachials in the cup commonly wider than high, two disti dials in each division of the rays except in the anterior ray ; interbrachial plates one or two in each interradial area except posteriorly ; in the posterior interradius the anal plate normally followed by three plates, but abnormally in the holotype by only two plates; these in turn followed by a single plate which does not connect with the tegmen. When normally developed there are four arm-openings from each ray, except the anterior one which has only two, making eighteen in all; the arms themselves unknown. Tegmen subcorneal from the arm bases to the base of the anal tube, the plates small, convex, a part or all of them, especially the larger ones, sur- mounted by low spines or thorn-like extensions; anal tube apparently strong, its length unknown. Remarks. — This species was originally described by Worthen as Bato- crinus unionensis, and was said to come from "St. Louis division of the Lower Carboniferous, Union County, Illinois/' Two specimens were illus- trated by him, one lacking the arms (pi. 12, figs. 5-5a) and the other with 332 GEOLOGY OF HARDIN COUNTY the arms in position (pi. 13, fig. 13). The definition was for the most part drawn up from the specimen without arms, this specimen, furthermore, being the one illustrated upon an earlier plate in the report containing the descrip- tion, and it may properly be selected as the holotype of the species. This holotype is now preserved in the collection of the Illinois State Museum at Springfield, and the writer is indebted to Dr. A. E.. Crook, curator of that institution, for the opportunity to study the specimen. The second specimen, with arms, has not been seen by the writer, it is not listed in the collections of the State Museum, and if it is still in existence its whereabouts is unknown. It is certainly specifically distinct from the selected holotype, and doubtless was collected from a different geological horizon. It is quite certain that neither specimen came from the St. Louis limestone. Worthen's definition of the species is misleading. The holotype is peculiar, among the related batocrinid forms, in possessing a single costal plate in each ray, the usual, small, quadrangular costal lying between the radials and the axillary costals being missing. It is this feature doubtless, which led Worthen to misinterpret the specimen. In order to provide for the two costals which are commonly present in these batocrinid forms he inter- preted the radials as first costals, below these he described the radials "as projecting so as to form a prominent rim around the basal cavity." The basals he described as being "very small and concealed in the basal cup." Examination of the specimen shows that the basals and "first radials" to- gether, as described by Worthen, constitute the basal disc of the specimen ; it is divided by three sutures into three equal basal plates and does not differ in any essential feature from the base of other batocrinids. The one essential feature in which the specimen does differ from all other batocrinids is in the loss of the first costal plates. This plate occasionally is lost abnormally in a single ray in certain individuals of Dizygocrinus and perhaps in other genera, but in G. unionensis these plates are all consistently wanting in all specimens, and it is not alone a good specific character, but may be counted as of generic value. This species, differing in no essential features from the type, has been collected at various localities in Hardin County, Illinois, and in Crittenden County, Kentucky, where it is a very characteristic species in the fauna of the Shetlerville formation, the very basal member of the Chester group in that region. This horizon is known to be present in Union County, along Swan Creek near Anna, and it is safe to assume that Worthen's specimen was secured from this same horizon as that which has afforded the specimens from Hardin and Crittenden counties. In the figure of Worthen's second specimen, the one with the arms, the presence of small, quadrangular, first costal plates is clearly shown. It also differs from the holotype in having the sides of the dorsal cup nearly straight between the arm bases and the top of the column, instead of distinctly con- PALEONTOLOGY 333 vex, as in the case of the true G. unionensis. This specimen is, in fact, essentially like a form rather commonly met with in the Ste. Genevieve lime- stone at many localities, which has been named Dizygocrinus superstes by Ulrich. The Ste. Genevieve limestone is well exposed in Union County, and Worthen's arm-bearing specimen doubtless came from that formation, but Worthen did not recognize the Ste. Genevieve limestone as a distinct forma- tion, and both the Shetlerville and Ste. Genevieve formations in Union County lie below the lowest massive sandstone formation of the Chester group, the base of which Worthen apparently assumed to be the upper limit of the "St. Louis division of the Lower Carboniferous.*' It is quite clear that none of the specimens illustrated as Dizygocrinus unionensis by Wachsmuth and Springer in their "North American Crinoidea Camerata" is truly a member of this species, and all of them are much more closely allied to the arm-bearing specimen illustrated by Worthen than to the holotype of G. unionensis. Horizon and locality. — Shetlerville formation, Union and Hardin coun- ties, Illinois; Crittenden County, Kentucky. Platycrinus penicillus (Meek and Worthen) Plate V, Figs. 1-4 . Platycrinus Huntsvillae Troost, unpublished manuscript. 1860. Platycrinus penicillus Meek and Worthen, Proc. Acad. Nat. Sci. Phil., p. 380. 1866. Platycrinus penicillus Meek and W x orthen, Geol. Surv. 111., vol. 2, p. 266, pi. 19, figs. 6a-c. 1897. Platycrinus Huntsvillae Wachsmuth and Springer, N. Amer. Crin. Cam., p. 678, pi. 73, figs. 6, 7a-b, 8-12. 1905. Platycrinus huntsvillae Ulrich, Prof. Paper, U. S. Geol. Surv. No. 36, p. 58, pi. 6, figs. 27a-i. One of the commonest and most characteristic members of the Ste. Genevieve limestone fauna in both the Mississippi and Ohio valleys, is a crinoid which has been referred commonly to Platycrinus huntsvillae. The column joints of this crinoid are the parts most commonly met with. They are characterized by their elliptical outline and by a row of short spines which project radially from the margin like the handles of a ship's steering wheel. The basal plates of the calyx are characterized by the presence of three radiating ridges which coincide with the three anchylosed sutures of the base. The manuscript name P. huntsvillae was given to this crinoid by Troost as early as 1850, but neither description nor illustration of the species was published at that time. In 1860 Meek and Worthen defined the species from Hardin County, Illinois, under the name P. penicillus, and later, in 1866, 33-4 GEOLOGY OF HARDIN COUNTY they republished their definition with illustrations. The name P. huntsvillae was never actually used in connection with a description of the species until 1897, when it was revived by Wachsmuth and Springer in their monograph of North American camerate crinoids. According to this the first actually published name for the species was P. penicillus, given by Meek and Worthen, and this name should be given preference over Troost's unpublished manu- script name which has been in common use since the appearance of Wachs- muth and Springer's monograph. A little careful search will almost invariably bring to light specimens of this species in any Fredonia outcrop, and they can be found in the "Lower Ohara" also in most localities; but the most diligent search on every possible occasion has utterly failed to establish the presence of the species in any exposure of the "Upper Ohara" strata. On the other hand the species does occur in undisputed St. Louis limestone, and in some of the ravines west of Cave in Eock, will preserved bases have been observed in beds which are well beneath strata carrying the characteristic Lithostrotion canadensis of the St. Louis. This species, therefore, constitutes a distinct bond between the Ste. Genevieve limestone and the St. Louis, but the genus becomes entirely extinct, so far as we now know, before the beginning of the Lower Chester "Upper Ohara" beds. Horizon and locality. — St. Louis limestone, Hardin County, Illinois. Ste. Genevieve limestone, wherever the formation is exposed. DlCHOCRINUS GIRTYI n. sp. Plate V, Figs. 5-12 Description. — Calyx subovate or slightly subpyriform, the maximum diameter about three-fourths of the total height. The dimensions of a nearly complete calyx are : greatest diameter at arm bases 8 mm., height from stem facet to summit of tegmen 10.2 mm., height of dorsal cup 7 mm. Dorsal cup obconical, slightly constricted just above the stem facet which is sur- rounded by a flange-like extension, the sides gently convex from just above the. constriction to the arm bases. The basal cup with two plates which extend about one-third the distance from the stem facet to the arm bases. Radial plates subquadrangular in outline, higher than wide, the postero-lateral ones somewhat narrower than the other three. The anal plate nearly or quite as large as the radials, but a little shorter. Surface of the plates of the dorsal cup ornamented with more or less discontinuous and somewhat wavy ridges, on the radial plates the longer of these ridges radiating from the arm bases to the lower angles of the plates, the intervening spaces being occupied by shorter ridges lying in a general direction parallel with the plate margins; on the basal plates "the more continuous ridges extend from the lower angles of the radial plates to the flange surrounding the stem facet; between these PALEONTOLOGY 335 longer ridges the others extend horizontally. Costal plates, pentagonal in outline, apparently one in each ray, very small, their width being about one- fifth or one-sixth of the width of the radial s, and their length about equaling their width; these small plates may be constituted of two closely united, but this is not evident in the specimens studied. The costal plates are followed by at least two small distichals in each series, the succeeding plates being wanting in the specimens examined. The tegmen is gently convex on top curving downward abruptly on the sides between the arm bases, rather dis- tinctly separated from the plates of the dorsal cup by shallow, horizontal constrictions or furrows between the arm bases, this constriction being more profound upon the posterior side; in the depressed interambulacral portions of the tegmen, three or four larger plates ; the posterior interambulacral area broader than the others with a larger number of plates, the anal opening small and circular, directed laterally and situated a little above the junction of the tegmen with the dorsal cup; the ambulacral plates of the tegmen smaller than the interambulacrals ; five somewhat larger interradial plates in the central portion of the tegmen being probably the orals, the posterior one being crowded between the others. The orals and some of the surround- ing plates, especially in the ambulacral directions are distinctly tuberculate, the tubercles growing smaller distally ; the plates of the depressed inter- ambulacral areas are essentially smooth. Remarks. — This species has been established upon a very perfect calyx preserving both the dorsal cup and the tegmen, and a second somewhat larger, nearly perfect dorsal cup, both from the Shetlerville formation in Hardin County. These two examples are the cotypes of the species. Aside from these specimens a number of fragments, separate radial plates especially, have been met with. The most notable of these fragmentary examples are from the Paint Creek formation in St. Clair Country, Illinois, where two examples have been collected, both from the same locality. One of these specimens is a practically complete base which is entirely like those collected in Hardin County except that it is about twice as large. The second example is a crushed calyx exhibiting two of the radials nearly complete and a fragment of the base. Like the base with which it is associated this crushed specimen is considerably larger than the Hardin County examples, and the surface markings are some- what more regular in their arrangement than in either of the cotypes. The difference in the character of the surface markings, however, is scarcely greater than that observed among the examples collected from one locality in Hardin County, and there is certainly not enough difference between the Hardin and St. Clair County examples to warrant separating them speci- fically. The species is quite distinct in form and ornamentation from any of the older members of the genus. It is not evident that it possesses the two small costal plates that are said by Wachsmuth and Springer to characterize the 336 GEOLOGY OF HARDIN COUNTY genus; there appears to be only a single plate in each costal series, although it is possible that two very small plates may be closely united. Horizon and locality. — Shetlerville formation, Fairview bluff below Rosi- clare, and east of Shetlerville, Illinois; Paint Creek formation, Prairie du Long Creek, northwest of Floraville, St. Clair County, Illinois. Genus TALAROCRINUS The genus Talarocrinus is an important one in the interpretation of the Lower Chester faunas. In their monograph on the camerate crinoids, Wachs- muth and Springer 1 state that the genus is "probably restricted to the age of the Warsaw and St. Louis groups," but it is now determined that all of the known species of the genus, except T. simplex which is from the Spergen limestone, occur in the Lower Chester formations and nowhere else. T. sim- plex is quite different from the other members of the genus, and since the characters of its arms and tegmen are entirely unknown, it is not improbable that it may belong to a genus distinct from the true Talarocrinus whose genotype is T. cornigerus. As known now the genus occurs in the Renault and Paint Creek formations in the Mississippi Valley, in the "Upper Ohara" limestone of Hardin County, Illinois, and the adjacent part of Kentucky, and in the two divisions of the "Gasper" limestone elsewhere in Kentucky. No member of the genus is recorded from the Fredonia limestone or the "Lower Ohara" limestone. All of the evidence afforded by the members of the genus goes to show that the Talarocrinus zone is a contemporaneous horizon from St. Clair County, Illinois, to central Kentucky, although Ulrich would ex- clude the Talarocrinus-hearmg beds of Hardin County from this horizon and place them lower in the stratigraphic column. His reason for doing this, so far as Talarocrinus is concerned, is that in the application of his method of correlation by matching individuals, he has not found a specimen from the "Upper Ohara" limestone that according to his standards exactly matches any specimen he has seen from the typical Eenault, and therefore he con- cludes that these two formations are not contemporaneous and he assigns the "Upper Ohara" to a position lower in the stratigraphic column than the Eenault. The species commonly met with in the "Upper Ohara" is T. trijugis, although T. inflatus is certainly present, and out of the thousands of fragmentary individuals that have been observed, scarcely half a dozen preserve the characters of the tegmen which are essential for certain specific determination. In the typical Renault the species commonly met with is T. inflatus, which also is commonly preserved in such a condition that its essential specific characters are not exhibited. The characters of the dorsal cups of the two species are the same, and from the dorsal cup alone it is not possible to separate the two species, nor can either of them be distinguished X N. Amer. Crin. Cam., vol. 2, p. 782, (1897). PALEONTOLOGY 337 from T. cornigerus. The differences between these three species are found in the relative development of spines upon the ambulacral plates of the tegmen. In T. inflatus no spines are present, in T. cornigerus the most prominent spines are developed, and T. trijugis occupies an intermediate position. If all the fragmentary specimens that are met with were complete, it is not un- likely that all three species would be found throughout these Talarocrinus- bearing beds, but as we know them from present collections of complete bodies, we know that T. inflatus and T, cornigerus are both present in the typical Renault of Randolph County, and T. inflatus and T. trijugis are found together in the Renault beds of Union County. No well defined examples of T. cornigerus have been observed in the "Upper Ohara," although T. inflatus is present, but the associations of the species in Randolph, Union, and Hardin counties show that all three lived contemporaneously, and that in all probability they are nothing more than geographic variations of a single form. Neither in the Mississippi Valley nor in the Ohio Valley during a most diligent search for four years, has a single example of Talarocrinus been met with in any true Ste. Genevieve limestone, nor has a single example of Platijcrinus penicillus been found above the top of the true Ste. Genevieve. Furthermore, the associates of Talarocrinus in the "Upper Ohara" are every one found in the typical Renault or the Paint Creek. These two faunal zones are everywhere distinct throughout the region studied, and every bit of evi- dence, both stratigraphic and faunal goes to show that each of these two zones is a contemporaneous horizon from St. Clair County, Illinois, to central Kentucky. Talarocrinus inflatus Ulrich Plate VI, Figs. 1-13 1918. Talarocrinus inflatus Ulrich, Ky. Geol. Surv., Miss. Form. West. Ky., pi. 8, figs. 14-15, pi. 9, figs. 4-6. (Xame and illustrations pub- lished without description.) Description. — Body, when complete without the arms, subellipsoidal or subspherical in form, usually higher than wide, but an occasional specimen has the height and width nearly equal. The dimensions of a nearly perfect body are: height 18 mm., greatest width 15.8 mm., height of dorsal cup 10.2 mm., greatest width of base 10.8 mm. Dorsal cup bowl-shaped, the surface of the plates smooth, their contour rather strongly convex in mature specimens, with the sutures impressed, the base distinctly bilobed by reason of the impressed suture separating the two plates. The radial plates diverging more or less from the base to the top of the dorsal cup, the anal plate similar in form and size to the radials. Tegmen not so high as the dorsal cup, its center occupied by a large plate whose whole surface is elevated into a —22 G 338 GEOLOGY OF HARDIN COUNTY prominent node that is somewhat flattened antero-posteriorly ; from the central nodose plate the ambnlacral regions radiate as rounded shoulders, each including ten or more small plates with depressed convex surfaces; the inter- ambulacral regions depressed and four of them appearing as rather narrowly rounded furrows which curve downward to the margins of the dorsal cup; the posterior interradial area broader than the others, the anal opening situated upon a prominence made up of many small plates, which extends well up towards the large central plate of the tegmen, the opening directed laterally or obliquely upwards. The arms in two pairs from each ray, making twenty altogether; they are uniserial near the base, but above the first five or six plates they become biserial ; their length is commonly a little more than twice the height of the dorsal cup. Remarks. — This species is the common member of the genus in the typical Renault limestone of Monroe and Randolph counties, but it also occurs in Hardin County. The specimen from which the above description has been chiefly drawn up is probably from the same locality as Ulrich's holotype of the species, and in some respects it is a better representative of the species than the one illustrated by Ulrich. The species exhibits a con- siderable degree of variation in the proportional height of the dorsal cup and in the convexity of the plates of the cup. The convexity evidently in- creases with the age of the individual, as is shown in a series of specimens from the type locality. The characteristic features of the species seem to be in the nature of the plates of the tegmen, especially in the large central plate with its flattened node, and the small plates of the ambulacral regions which are not produced into tubercles or spines. Horizon and locality. — Renault limestone, St. Clair, Monroe, Randolph, Union, and Hardin counties, Illinois; also Ste. Genevieve County, Missouri. Talarocrinus trijugis Miller and G-urley Plate VI, Figs. 17-26 1894. Talarocrinus trijugis M. & G., Bull. 111. State Mus. Nat. Hist., No. 3, p. 45, pi. 3, figs. 9-11. 1918 Talarocrinus trijugis Ulrich, Ky. Geol. Surv., Miss. Form. West. Ky., pi. 8, figs. 11-13, pi. 9, figs. 10-11. Description. — Body, when complete without the arms, subellipsoidal in form, higher than wide. The dimensions of the holotype are : height 14 mm., greatest width 13 mm., height of dorsal cup 6.5 mm., greatest width of the base 8 mm. Dorsal cup bowl-shaped, the surface of the plates smooth, their contour rather strongly convex in mature individuals, with the sutures im- pressed; the base bilobed by reason of the impressed suture that separates the two plates; the radial plates diverging more or less from the base to the top of the dorsal cup ; the anal plate similar in form and size to the radials. The tegmen higher than or lower than the dorsal cup, with a large central plate PALEONTOLOGY 339 whose surface is elevated into a prominent node that is somewhat flattened antero-posteriorly ; the ambulacral regions rounded, sloping away from the border of the central plate in the form of five shoulders made up of numerous small plates, from two to five of which in each ray bear low tubercles; the interambulacral regions are depressed and four of them are rounded furrows which curve downwards to the margin of the dorsal cup ; the posterior inter- radius broader than the others, somewhat protuberant medially, the anus situated well up towards the large central plate, and directed laterally or obliquely upward. The arms unknown. Remarks. — So far as the dorsal cup is concerned it is absolutely impos- sible to distinguish this species from T. inflatus, and neither of these species can be separated from T. cornigerus. The only distinguishing character of T. trijugis as compared with T. inflatus, is found in the greater convexity of the plates in the ambulacral regions of the tegmen, a number of these plates in each ray being raised into distinct tubercles. Such characters as these are of very slight value among camerate crinoids, and it is doubtful whether such a difference is of even specific importance. Talarocrinus cornigerus, another related species, has these same plates developed into distinct, short spines, in other respects being entirely like T. inflatus and T. trijugis. Not one specimen in a thousand of any of these species preserves the tegmen in such a condition that its characters can be recognized, but so far as observa- tions have been possible the specimens from Monroe and Randolph counties are mostly referable to T. inflatus, although one good specimen has well developed spines and may be referred to T. cornigerus. In Hardin County T. trijugis seems to be the commoner form, but in Union County a number of good specimens preserving the tegmen have been collected from the same horizon, which represent both T. inflatus and T. trijugis. By his method of matching individuals, Ulrich has come to the conclu- sion that the "Upper Ohara" of Hardin County is not the equivalent of the Renault of Monroe County, because he has not found the same Talarocrinus in both; but the fact that both the Hardin County and Monroe County forms are found together in Union County, and T. inflatus is also present in Hardin County, makes it clear that the individual characteristics of which Ulrich has made use are of less value for correlation purposes than the larger generic characters. Furthermore, it is altogether probable that the characters of all three of these species would be found to merge together in all of the localities if a sufficient number of specimens were available for study. Every student who has given any considerable attention to the study of the camerate crinoids knows how inconsequential are such characters as those which distinguish these species. At the most, they probably represent contemporaneous geo- graphic variations of a single form. Horizon and locality.— -Renault limestone, Hardin. Pope, and Union counties, Illinois. 340 GEOLOGY OF HARDIN COUNTY Talabocbinus coenigeeus (Shumard) Plate VI, Fig. 35 1857. Dichocrinus cormgerus Shumard, Trans. St. Louis Acad. Sci., vol. 1, p. 72, pi. 1, figs. la-c. 1897. Talarocrinus cormgerus Wachsmuth and Springer, N". Amer. Crin. Camer., p. 783, pi. 78, figs. 6a-b, 7a-b. 1918. Talarocrinus cornigerus Ulrich, Ky. Geol. Surv., Miss. Form. West. Ky., pi. 8, figs. 18-19, pi. 9, figs. 7-9. Talarocrinus cornigerus was originally described as coming from the "Mountain limestone/' and Wachsmuth and Springer referred it to the St. Louis limestone. It has been determined by Ulrich, however, that the type specimen is from the Lower Chester, and Wachsmnth and Springer's refer- ence of the species to the St. Louis limestone was erroneous. In the characters of its dorsal cup the species is identical with T. inflatus and T. trijugis. In all of these species the cup varies considerably, in general it is bowl-shaped, but occasionally an individual is met with in which the radial plates flare out in such a manner as to be essentially like that of T. sexlobatus. The differ- ences between these three species, as has already been pointed out, are found in the relative development of spines upon the ambulacral plates of the tegmen, these being best developed in T. cornigerus, and least strongly de- veloped in T. inflatus, with T. trijugis occupying an intermediate position. There is no evidence that these characters have different time values, and if they are of value at all it is as contemporaneous geographic variations. The specimen illustrated is from the Renault limestone in Randolph County, the same horizon in which the typical T. inflatus is known. T. trijugis and T. inflatus occur together in Union County, and this association with the demonstrated contemporaneity of T. inflatus and T. cornigerus in Randolph and Monroe counties, demonstrates the time equivalence of these Talarocrinus beds across the entire state. Horizon and locality. — Renault limestone, Randolph County, Illinois. Talabocbinus ovatus Worthen Plate VI, Figs. 27-33 1883. Talarocrinus ovatus Worthen, Geol. Surv. 111., vol. 7, p. 314, pi. 29, fig. 11. Description. — Body, when complete without the arms, subovoid in form, higher than wide. The dimensions of the holotype are: height 17.3 mm., greatest width 12.6 mm., height of dorsal cup 9.1 mm., greatest width of base 8.3 mm. Dorsal cup bowl-shaped, the surface of the plates smooth and convex with the sutures impressed, the base bilobed by reason of the impres- sion of the sutures between the two plates. The radial plates higher than wide, a little wider at the upper margin than at the junction with the base. PALEONTOLOGY 341 which causes the sides of the cup to diverge slightly from the base to the arm openings. Anal plate similar in form and size to the radials but with the lateral margins more nearly parallel. Tegmen not so high as the dorsal cup, dome-shaped, its center occupied by a plate somewhat larger than those about it, and produced into a short but rather slender thorn-like spine; from the central spinose plate the ambulacra radiate as five rounded, shoulder-like lobes, towards the outer border of the tegmen above the arm openings, the anterior and anterolateral ambulacra each includes a plate which tends to be raised into a short, thorn-like spine smaller than that upon the central plate, some of the other plates are also nodose in some individuals, but the postero- lateral ambulacra are less rugose than the others. The interambulacral spaces scarcely depressed towards the center, but distally they become a little de- pressed below the ambulacra and the surface curves regularly downward into the dorsal cup. The posterior interambulacral area is much broader than the others, and a broadly rounded ridge made up of small plates leads upward from the anal plate towards the central spine-bearing plate of the tegmen, the anal opening is situated well up towards the median plate and is directed obliquely upward and outward. Arms not known. Remarks. — The dorsal cup of this species resembles T. inflatus, T. trijugis, and T. cornig&rus, but in the specimens which have been studied, its sides seem to be somewhat more erect than in those species and proportionally a little higher. The characters of the tegmen exhibit greater differences. The rather massive central plate of the tegmen in those species, with its antero-posteriorly flattened node, is represented by the comparatively much smaller plate surmounted by only a slender, thorn-like spine. In those species, also, all five of the ambulacra tend to be similar in the development of nodes and spines upon the plates, but this species, like T. planus, seems to have the postero-lateral ambulacra less nodose than the three anterior ones, although as in all camerate crinoids, the development of spines and nodes exhibits a considerable degree of variation, and all too few examples preserving these characters are available for study. The specimen from Breckenridge County, Kentucky, illustrated by Ulrich as T. ovatus 1 possesses a massive, flattened, central tegminal plate, and cer- tainly is not correctly identified, it being more nearly related to the T. inflatus and its allies. In the original definition of the species by Worthen, it is recorded as coming from the Chester limestone of Monroe County, Illinois, and Ulrich has assumed that it came from the Renault limestone. No authentic example, however, has been collected from known Renault beds, but a very good speci- men has been collected from the Paint Creek formation of Monroe County, and it is not unlikely that Worthen's holotype also came from this horizon. Horizon and locality. — Paint Creek formation, Monroe County, Illinois. 1 Ulrich, Ky. Geol. Surv., Miss. Form. West. Ky., pi. 8, figs. 16-17. 342 GEOLOGY OF HARDIN COUNTY Talarocrinus planus n. sp. Plate VI, Figs. 14-16 Description. — Body, when complete without the arms, subovoid in form, higher than wide, the dorsal cup higher than the tegmen. The dimensions of the holotype are: height 13.7 mm., greatest width 11.7 mm., height of dorsal cup 7.6 mm. Dorsal cup deep bowl-shaped, the surfaces of the plates smooth, the sutures not impressed. The base bowl-shaped, not bilobate. Radial plates as high or higher than wide, their lateral margins slightly diverging so that the sides of the cup diverge gently from the top of the basal plates to the arm bases. The anal plate similar to the radials in form and size. Costal plates one in each ray, very broadly triangular in form, supporting two series of distichals. Tegmen distinctly pentalobate, the plate occupying the central position is the largest; in the type specimen one ambulacral plate of the anterior ray is produced into a short, thorn-like spine, the corresponding plates in the two anterolateral rays are less produced but are distinctly nodose. the same plates in the postero-lateral rays are not produced at all. Four of the interambulacral areas are distinctly depressed and appear as narrow, rounded furrows leading downward from the center of the tegmen between the arm bases; the posterior interambulacral area is much broader than the others, its surface depressed between the adjacent ambulacra, with a distinct rounded elevation leading upward to the anal opening which is situated at the extremity of a short tube whose upper extremity extends a little above the central plate of the tegmen, the anal opening directed obliquely outward and upward. Arm openings two in each ray, arms unknown, but doubtless four from each ray as in other species of the genus so far as known. Remarks. — This species differs notably from T. cornigerus, T. trijugis and T. infiatus in the absence of the distinctly bilobed base, in the even surface of the dorsal cup, and in the smaller and less protuberant central plate of the tegmen. The form of the body is more like T. butt si from the Shetlerville formation than any other member of the genus, but the tegmen is much more conspicuously lobate, with more prominent arm bases, the tegmen being more like that of T. dewolfi. The species is of especial interest because its relationships are with T. buttsi and T. dewolfi, two species named by Ulrich from the "Upper Ohara" and consequently referred to the Ste. Genevieve by him, while this species is from the Paint Creek formation of Randolph County. Horizon and locality. — Paint Creek formation, Randolph County, Illinois. Pterotocrinus serratus n. sp. Plate V, Figs. 24-26 Description. — This species is based on the "wing plates" only, which are subspatulate in form. The lateral margins are subparallel proximally but PALEONTOLOGY 343 become more or less widely divergent distally, being unevenly concave in outline. The distal margin is more or less arcuate or nearly straight, the plates becoming very thin distally and being divided by incisions of varying depths into a series of serrations. The dimensions of the holotype are: length of "wing plate" 11.2 mm., greatest width distally 7.7 mm. Another specimen with a length of about 12.5 mm., has a maximum width of about 11.5 mm. Remarks. — The genus Pterotocrinus is very common in many of the Chester faunas, but the species are rarely represented by complete specimens, and the detached "wing plates" are the parts commonly met with. For- tunately the variations of these highly specialized plates furnish some of the best of specific characters in the genus, so that it is practicable to define species from these plates alone. The genus is not known in the Eenault faunas, but it makes its appearance in the Paint Creek, and continues through all of the higher fossiliferous horizons of the Chester unless it be the CI ore and Kinkaid. A rather common form in the Paint Creek, both in the typical exposures of the formation in Monroe and St. Clair counties, and in Johnson County, is this form of "wing plate" with the thin, knife-like distal edge which is distinctly but more or less irregularly serrate. Horizon and locality. — Paint Creek formation, St. Clair, Monroe, Ean- dolph, and Johnson counties, Illinois. Pachylocrinus cachensis n. sp. Plate VIII, Fig. 35 Description. — Dorsal cup low basin-shaped, with the base excavated and the sides broadly flaring, the arms of moderate length. The dimensions of the holotype are: diameter of dorsal cup 8.3 mm., depth of dorsal cup 3 mm., height of crown from base to tips of arms, 25.5 mm. The length of the arms would be somewhat greater if they were perfectly preserved. Underbasal plates small, entirely included in the basal concavity, but only partially covered by the column. The basal plates about as wide as long, their proximal portions occupying the sloping sides of the basal concavity. The radial plates wider than high. Eadianal plate much longer than wide, pentagonal in out- line, the longer sides diverging ventrally, the ventral side short and sup- porting the first tube plate. Anal plate smaller than the radianal, resting upon the truncate margin of the posterior basal and lying between the radianal and first tube plate on the right, and the left postero-lateral radial on the left. Each ray with a single costal which is larger than the radial and is spine-bearing, the spine rising from close up to the angular distal extremity of the plate. The distichals, except the first and last in each series, quad- rangular, wider than long, nearly rectangular or slightly wedge-shaped, com- monly six in each series, but one series of thirteen is present in the holotype; the axillary plate in each series is larger and thicker than the others, and is 344 GEOLOGY OF HARDIN COUNTS extended into a short spine. One or two divisions of the rays occur beyond the distichals, the plates are similar to the distichals but smaller, and towards the tips of the arms they become more wedge-shaped, each axillary plate is larger and thicker than the others, and is produced into a node or short spine. Remarks. — This species is allied to P. spinifera (Wetherby), from the Chester of Pulaski County, Kentucky, probably the Glen Dean formation, but it is smaller and much smoother in appearance, with shorter axillary spines in the brachial series, and with the brachial plates much less wedge-shaped. The holotype and one or two other good specimens are from the Cache River bluffs in Johnson County, but the same species is present in the Paint Creek formation in Monroe County, Illinois. Horizon and locality. — Paint Creek formation, Cache River bluffs six miles south of Vienna, Johnson County, and two miles south of Columbia, Monroe County, Illinois. Agassizocrinus dissimilis n. sp. Plate V, Figs. 29-30 Description. — Dorsal cup subhemispherical. The dimensions of the holotype are : greatest width 13.7 mm., height 10.8 mm. Surfaces of the plates convex, with the bounding sutures distinctly impressed. The under- basals anchylosed into a large pentagonal plate with gently convex surface, which is the largest plate in the cup. Basals wider than high, resting upon the nearly straight or slightly concave faces of the pentagonal underbasal. disc. Radial plates short and broad, their width being twice or more than twice their height. Radianal plate quadrangular, rather large. Anal plate smaller than the radianal, its contact with the posterior basal very short. Arms unknown. Remarks. — This species is especially characterized by the shortness of the dorsal cup and by the convexity of the plates and consequently impressed sutures. Horizon and locality. — Paint Creek formation, one and one-fourth miles northwest of Floraville, St. Clair County, Illinois. Genus ARCHIMEDES The genus Archimedes is one of the very characteristic members of most Chester faunas, the main mass of the Chester limestone being originally described as the "Archimedes limestone" by the earlier geologists working in the Mississippi and Ohio valleys. As commonly preserved only the axes of the colonies are found, and these exhibit such a wide range of variation that it is exceedingly difficult to draw lines between specific groups. Ulrich is responsible for most of the named species of the genus, and while many PALEOXTOLOGY 345 examples in the large collections that have been made do resemble more or less closely the illustrations of his several species, and may be identified, there are many other examples that cannot be made to fit satisfactorily in any of the named species. It is altogether likely that the fenestrated expansions from the axes would supply much more satisfactory specific characters if these characters could be observed, but they can be determined in rare cases only. In the arrangement of the zooecia, the fenestrated portion of Archi- medes is identical with Fenestella, but the branches are commonly finer and more numerous, and the fenestrules are shorter, so that the whole net-like expansion is notably finer than in most species of Fenestella. The genus Archimedes is met with in essentially all fossil-bearing Chester horizons in Illinois, but the genus rarely occurs in the Renault of Monroe and Randolph counties, and it is likewise a rare member of the Shetlerville and Renault faunas in Hardin County and the adjacent parts of Illinois and Kentucky. The genus first became a common form in the faunas of the Paint Creek, this being true both in the Mississippi Valley counties in Illinois and in the more southeastern portions of the state. On the accompanying plate VII a number of forms of the genus from the Paint Creek formation of Johnson County are illustrated. The specific identifications are made with some hesitation, as is always the case in this genus, but these specimens seem to be reasonably like the figures of the species they have been identified with, as they have been illustrated by Ulrich. These several forms are introduced in this place in order to show the variety exhibited by members of this genus in the Paint Creek formation in southeastern Illinois, in contrast with the scarcity of the genus in the Renault. This same contrast between the two equivalent faunas is likewise exhibited in Monroe and Randolph counties. Archimedes cf. savallovanus Hall Plate VII, Fig. 26 1857. Fenestella (Archimedes) Sivallovana Hall, Proc. Amer. Ass. Adv. Sci., vol. 10, p. 178. 1890. Archimedes swallovanus Ulrich, Geol. Surv. 111., vol. 8, p. 574, pi. 63, figs. 12-12d. The axes that are referred to this species are stouter forms than the others that are illustrated on the same plate. This species is more common in the Middle Chester formations, and these Lower Chester examples are somewhat more slender, with flanges somewhat more closely arranged than in the more typical representatives of the species. Horizon and locality. — Paint Creek formation, Cache River bluff, six miles south of Vienna, Johnson County, Illinois. 346 GEOLOGY OF HARDIN COUNTY Archimedes compactus Ulrich Plate VII, Fig. 27 1890. Archimedes compactus Ulrich, Geol. Surv. 111., vol. 8, p. 572, pi. 63, figs. 2-2b, 2d-e. The axes of this species of Archimedes may attain a length of 30 to 40 mm., and when well preserved they commonly exhibit a slight taper towards each extremity. There are commonly from nine to twelve flanges present in the space of 2 cm. This type of axis is not uncommonly present in the Paint Creek fauna of Monroe County, as well as in the more southern part of Illinois. Horizon and locality. — Paint Creek formation, Cache River bluff, six miles south of Vienna, Johnson County, Illinois. Archimedes proutanus Ulrich Plate VII, Fig. 28 1890. Archimedes proutanus Ulrich, Geol. Surv. 111., vol. 8, p. 576, pi. 63, figs. 3-3d. This species is characterized by a rather slender axis, with from six to eight flanges in the space of 2 cm. From the border of the flange the slope towards the axis is abrupt above, and more gentle and concave below. Horizon and locality. — Paint Creek formation, Cache River bluff, six miles south of Vienna, Johnson County, Illinois. Archimedes distans Ulrich Plate VII, Fig. 29 1890. Archimedes distans Ulrich, Geol. Surv. 111., vol. 8, p. 578, pi. 63, figs. 9-9b. As its name suggests this species is loosely coiled, with about four volu- tions in the distance of 2 cm. ; the central portion of the axis is rather slender, with the lower slope from the rim of the flange long, gentle and con- cave, and the upper slope abrupt. Horizon and locality. — Paint Creek formation, Cache River bluff, six miles south of Vienna, Johnson County, Illinois. Archimedes meekanus Hall Plate VII, Fig. 30 1857. Fenestella (Archimedes) Meekana Hall, Proc. Amer. Ass. Adv. Sci., vol. 10, p. 178. 1890. Archimedes meekanus Ulrich, Geol. Snrv. 111., vol. 8, p. 578, pi. 63, fig. 4. This is a slender species with from five to six volutions in the space of 2 cm., in which the slope from the rim of the flange to the central portion PALEONTOLOGY 347 of the axis is more nearly symmetrical above and below than in other species with similar proportions. Horizon and locality. — Paint Creek formation, Cache Eiver bluff, six miles south of Vienna, Johnson County, Illinois. Archimedes communis Ulrich . Plate VII, Figs. 31-32 1890. Archimedes communis Ulrich, Geol. Surv. 111., vol. 8, p. 573, pi. 63, figs. 1-ld. This is another one of the rather slender species, with about seven or eight volutions in the space of 2 cm. The slopes from the rim of the flange to the central portion of the axis is nearly symmetrical above and below, but the flange is considerably more extended than in A. meekanus, and the volu- tions are closer together. Examples of the axis of this species are commonly more elongate than those of some other species, and it is possible that this greater length is one of the specific characters. Horizon and locality. — Paint Creek formation, Cache Eiver bluff, six miles south of Vienna, Johnson County, Illinois. Lyropora quincuncialis Hall Plate VII, Figs. 5-6 1857. Fenestella (Lyropora) quincuncialis Hall, Proc. Amer. Ass. Adv. Sci., vol. 10, p. 180. 1890. Lyropom. quincuncialis Ulrich, Geol. Surv. 111., vol. 8, p. 583, pi. 55, ^ngs. 7-7c, pi. 58, figs. 3-3d. Species of Lyropora occur commonly in the higher beds of the "Upper Ohara" in Hardin County and elsewhere, and the genus occurs, though less frequently, in the Shetlerville formation, which is the lower portion of the ''Upper Ohara." The original of figure 5 of the accompanying plate is one of several examples from a Shetlerville exposure about one mile south of Eichorn. Some of the specimens exhibit the characters of the fenestrated portion of the colonies, and they possess about 25 branches in the space* of 1 cm., and 20 fenestrules in the same space. The zooecia are arranged mainly in two series along the branches, and the median line of the branches is marked by a series of low nodes. These characters do not differ in any essen- tial respect from the illustrations of L. quincuncialis given by Ulrich in vol. 8 of the Geological Survey of Illinois, except that there are a few more fenestrules in the unit distance, the number given by Ulrich being 16 or 17, instead of the 20 which our specimens possess. In view of the close agree- ment in all other characters, however, this slight variation cannot be con- sidered as being sufficient to exclude the specimens from this well known Chester species. The original of the larger figure (fig. 6), is from the 348 GEOLOGY OF HARDIN COUNTY Renault portion of the "Upper Ohara" at Cedar Bluff, near Princeton. Kentucky, the type locality of the Ohara. In this specimen the reverse side of the fenestrated part of the colony is well exhibited, and it has 20 fenestrules and 26 branches in the unit distance of 1 cm., being essentially like the Eichorn specimens in this respect. This species of Lyropora is a common one in the typical Renault lime- stone in Monroe and Randolph counties, and although the genus occurs all through the Chester, it is especially common in the iower Chester formations. On the other hand the most diligent search for many years has never brought to light a single example of Lyropora in the Ste. Genevieve limestone, al- though the genus is known in beds as old as the Burlington limestone. This is one of the characters which may be classed among the "complexion" features of the "Upper Ohara," which ties it to the Chester and excludes it from the Ste. Genevieve, and even though a sporadic specimen of Lyropora should be found sometime in the Ste. Genevieve, it would not overthrow the relation- ships of the faunas any more than the finding of Septopora in the St. Louis limestone makes that formation Chester. Furthermore, these "Upper Ohara" specimens of Lyropora are identical with forms in the topical Renault lime- stone elsewhere. Horizon and locality. — Shetlerville formation, one mile south of Eichorn. Hardin County, Illinois. Renault limestone, Monroe, Randolph, and Hardin counties, Illinois, Ste. Genevieve County, Missouri, and near Princeton, Kentucky. Septopora similis n. sp. Plate VII, Figs. 1-4 Description. — Zoarium flabelliform, fiat or undulating, consisting of nearly uniform branches united by dissepiments which are somewhat narrower than the branches themselves; most of the dissepiments are nearly straight, some are more or less oblique, and a few are slightly arched. Branches in- creasing by bifurcation or intercalation .5 mm. or less in width, from 10 to 12 occupying the space of 1 cm. Fenestrules from 10 to 12 in the space of 1 ®ni., more or less nearly rectangular, rhomboidal or crescentic in form, with the length and breadth subequal. The zooecia commonly in two or rarely three rows, from 21 to 22 occupy the space of 5 mm., from 3 to 6 present upon each dissepiment; the apertures subcircular, their distance apart about equal to their diameter, the peristome scarcely elevated above the surface of the branch. The obverse side of the branches with an inconspicuous or essentially obsolete keel, and with low, inconspicuous median tubercles at intervals about equal to the distance between the dissepiments. Reverse side finely striate, the supplementary pores inconspicuous. Remarks. — This species resembles Septopora subquadrans in the dimen- sions of its branches and fenestrules and in the arrangement and number of PALEONTOLOGY 349 the zooecia, but the dissepiments are less commonly arched. It differs from that species especially in the much less conspicuous median keel and tubercles upon the obverse side of the branches, and in the smaller number of zooecia upon the dissepiments. The zooecia are also smaller and have much less prominent peristomes than S. sub quadr arts. The description of this species is introduced here because of its bearing upon one line of evidence that Ulrich has used in his contention for the Chester age of the Ste. Genevieve. Attention is drawn by him 1 to the fact that the bryozoan Coeloconus, which is not uncommon in a number of the Chester formations, is represented by a species in the undoubted Ste. Genevieve limestone of Monroe County, Illinois, and elsewhere, his assumption being that the presence of this particularly characteristic Chester genus in the Ste. Genevieve limestone fauna, is evidence for uniting the Ste. Genevieve with the Chester. As it has been known in the past, the genus Septopora is an especially characteristic genus of the Chester faunas, and nowhere in the literature has any member of the genus been recorded from strata older than the Chester. The species here described, however, is from the St. Louis lime- stone, and was collected by the writer from the lower portion of this forma- tion, in the Mississippi River bluffs above Little Rock, Ste. Genevieve County, Missouri. If the presence of a species of Coeloconus in the Fredonia lime- stone is evidence for the inclusion of that formation in the Chester group, then why is not the presence of a species of Septopora in the lower portion of the St. Louis limestone evidence for placing the base of the Chester still lower in the stratigraphic series, and the inclusion of the St. Louis limestone also, in the Chester? Such an extension of the Chester would, of course, never be advocated. The facts concerning these two genera of bryozoans simply go to show that our knowledge of the geologic range of genera and species of fossils is incomplete. As observations are extended, other genera now known only in the Chester may be found to be present in the Ste. Genevieve or older forma- tions, but such discoveries will not affect the correlation of the formations. The paleontologic basis for the correlation of formations within one basin of sedimentation, must be the sum total of the characters of the faunas involved, in other words the complete faunal association, or the general "complexion" of the life of a formation is of far greater importance than the sporadic occurrence of an unusual genus, or the evidence afforded by the method of matching individuals irrespective of other lines of evidence. Horizon and locality. — St. Louis limestone, lower part, above Little Rock, Ste. Genevieve County, Missouri. *Ky. Geol. Surv., Miss. Form. West. Ky., pi. 11, figs. 11-13. 350 GEOLOGY OF HARDIN COUNTY Cystodictya labiosa n. sp. Plate VII, Figs. 7-24 Description. — Zoarium ramose, with flattened, bifoliate, ribbon-like branches 2.5 to 3 mm. in width, bifurcating or giving off lateral branches at irregular intervals varying from 2 to 9 or 10 mm. apart, with well defined, non-poriferous margins. Zooecia arranged in more or less regular diagonal rows with from 7 to 10 apertures in each row, in places the apertures seem to be in two sets of diagonals starting from the two margins and becoming confused in the central portion of the zoarial surface ; the linear arrangement of the apertures is much less noticeable than the diagonal. The apertures have a raised peristome, giving them the appearance of being situated on low pustules, the distal margins of the peristomes commonly a little more raised and lip-like than the other portions, the zooecia towards the lateral margins somewhat more elevated than those towards the central portion of the branches. The apertures commonly subquadrangular and a little broader in a transverse direction, with a slight indentation upon the side opposite the higher portion of the peristome; in some examples the apertures approach a subcircular outline. The larger diameter of the apertures is commonly about .02 mm. or a little more, and the spaces between the apertures is somewhat greater than the width of the apertures themselves. Remarks. — This species suggests C. pustulosa in some respects, but the linear arrangement of the zooecia is less distinct, the apertures are somewhat more crowded, and the most elevated side of the peristomal border is situated distally, giving the apertures the appearance of being directed backwardly. In C. pustulosa the most elevated portion of the peristome is lateral, upon the outer side in each case, the outer apertures appearing to be directed towards the median line of the branches, while those towards the center have the peristome about equally elevated all around, with no part of it as high as the outer borders of the marginal zooecia. This species is one of the most characteristic forms in the Lower Chester faunas in Illinois. It occurs abundantly and is present in practically every full collection from the Renault and Paint Creek formations in St. Clair, Monroe, and Randolph counties, but has nowhere been observed either above or below these horizons. No species of Cystodictya whatever has been recorded from any true Ste. Genevieve limestone fauna, and none has ever been de- tected by the writer. In southern Illinois the species occurs commonly in the Shetlerville, Renault, and Paint Creek formations, and it is present in the same formations in Crittenden County, Kentucky. The species constitutes one of the many strong bonds between the Renault and the "Upper Ohara." Example of a species of Cystodictya from the Chester of Caldwell County, Kentucky, are preserved in the Walker Museum at the University of Chicago, from the Bassler collection. These specimens are of this common Lower PALEONTOLOGY 351 Chester species, and are recorded under the name of C. labiosa, said to be a manuscript name applied by Ulrich. In describing the species in this place Ulrich's manuscript name is retained in order that no confusion in nomen- clature may be introduced. Horizon and locality. — Shetlerville formation, Hardin County, Illinois, and Crittenden County, Kentucky. Eenault limestone, Monroe, Randolph, and Hardin counties, Illinois, and Crittenden County, Kentucky. Paint Creek formation, St. Clair, Monroe, Randolph, Johnson, and Pope counties, Illinois. GlYPTOPORA PUNCTIPORA Ulrich Plate VII, Fig. 25 1890. Glyptopora punctipora Ulrich, Geol. Surv. 111., vol. 8, p. 519, pi. 78, figs. 9-9a. This bryozoan is a rather common form that has not been found outside of the Lower Chester formations in Monroe and Randolph counties, Illinois, and elsewhere. The specimen figured is from the basal Chester, Shetlerville formation, of Crittenden County, Kentucky. It occurs in southeastern Illi- nois in the same formation and in the Renault. Horizon and locality. — Shetlerville formation, Hardin County, Illinois, and Crittenden County, Kentucky. Renault formation, Monroe, Randolph, and Hardin counties, Illinois. Paint Creek formation, Monroe, Randolph, and Johnson counties, Illinois. Rhipidomella dubta (Hall) Plate VIII, Figs. 5-6 1856. Orthis dubia Hall, Trans. Albany Inst., vol. 4, p. 12. 1914. Rhipidomella dubia Weller, 111. State Geol. Surv., Monog. I, p. 160, pi. 20, figs. 22-26, pi. 83, figs. 9-10. No brachiopod belonging to the family Orthidae has been found in any Chester fauna in Illinois, but Rhipidomella dubia occurs rather commonly in the Ste. Genevieve limestone, and in every preceding fauna down to the Keokuk limestone. Horizon and locality. — Keokuk limestone, Warsaw limestone, Spergen limestone, and St. Louis limestone, many localities in Iowa, Missouri, and Illinois. Ste. Genevieve limestone, Ste. Genevieve County, Missouri, Union and Hardin counties, Illinois. PUGNOIDES OTTUMWA (White) Plate VIII, Figs. 1-4 1862. Rhynchonella Ottumwa White, Proc. Bost. Soc. Nat. Hist., vol. 9, p. 23. 352 GEOLOGY OF HARDIN COUNTY 1914. Pugnoides ottumwa Weller, 111. State Geol. Surv., Monog. I, p. 193, pi. 25, figs. 7-17. Rhynchonelloid shells are unusual in those Mississippian formations that are represented in Hardin County, but Pugnoides ottumwa can usually be discovered if prolonged enough search is made for it in the Fredonia limestone. It is one of the very best index fossils of the Ste. Genevieve, and is a marker of this horizon from Fort Dodge, Iowa, to central Kentucky. It has never been observed in the Chester, and the only rhynchonelloid shell that has been detected in the Chester of Illinois is very different from this one, and is a member of another genus. The Pugnoides ottumwa zone is a well defined horizon in the Mississippian, and all the evidence that has been accumulated goes to show that it is a contemporaneous horizon throughout its whole geographic extent. Horizon and locality. — Ste. Genevieve limestone, numerous localities in Iowa, near Alton, Illinois, St. Clair, Monroe, Randolph, and Union counties, Illinois, Ste. Genevieve County, Missouri. Fredonia limestone, Rosiclare, Illinois, Cedar Bluff, near Princeton, Kentucky, and near Bowling Green, Kentucky. Spiriferina subspinosa n. sp. Plate VIII, Figs. 7-14 Description.— Shell small, wider than long, the hinge-line a little shorter than the greatest width of the shell, the cardinal extremities rounded. The dimensions of a medium sized example are : length from front to umbo of pedicle valve 9 mm., length of brachial valve 6.7 mm., greatest width 10 mm., length of hinge-line 8.5 mm., thickness 8 mm. Pedicle valve with a prominent umbo, the surface curving abruptly to the cardinal margin, more gently to the anterior margin, not compressed towards the cardinal extremities; mesial sinus sharply defined, moderately wide and rather deep, subangular or sharply rounded in the bottom, extending to the beak; the beak small, pointed and incurved; cardinal area moderately high, concave, the curvature increasing towards the beak, the lateral margins defined but interrupted from the surfaces of the lateral slopes by a slight ridge only; height and width of the delthyrium about equal; each lateral slope marked by from four to six subangular, simple plications which originate along the cardinal margin, those bounding the mesial sinus are the strongest, the others being successively smaller to the cardinal extremities. Brachial valve less convex than the pedicle, the greatest depth near the middle; the lateral slopes convex, not distinctly compressed towards the cardinal extremities; mesial fold rounded or subangular, sharply defined, similar to the plications upon the lateral slopes but somewhat broader and higher; beak small; cardinal area very narrow; the plications on the lateral slopes similar to those of the opposite valve and alternate with them. PALEONTOLOGY 353 Surface of both valves, when well preserved, marked by fine, closely crowded, evenly distributed, small tubercles or papillae which exhibit a tendency to be arranged in radiating rows. A number of concentric lines of growth are also commonly present towards the front of the shell. Shell substance coarsely punctate. Remarks. — This shell is allied to S. spinosus and has been referred to as a variety of that species by some authors. The shell differs constantly from the typical S. spinosa in its smaller size and much more rotund form, also in the smaller number of plications and in the much greater fineness of the papillae which cover the surface. These constant differences, associated with the fact that the two forms clo not occur together, but are characteristic of distinct horizons in the Chester series, seem to be sufficient reason for recognizing this form as a distinct species. The size of the shells varies considerably. The specimen whose dimen- sions " have been given is larger than usual ; but the largest example which has been observed has a maximum length of 10.3 mm., and width of 11.5 mm. A usual length for the shell is about 7 mm. Horizon and locality. — Sbetlerville formation, Hardin and Pope counties, Illinois; Crittenden County, Kentucky. The species has been identified in every Shetlerville fauna that has been collected within this area. Spiriferina transversa (McChesney) Plate VIII, Figs. 17-18 1860. Spirifer transversa McChesney, Desc. New Spec. Foss., p. 42. 1914. Spiriferina transversa Weller, 111. State Geol. Surv., Monog. I, p. 297, pi. 35, figs. 41-49. Spiriferina spinosa (N. & P.) Plate VIII, Figs. 15-16 1855. Spirifer spinosus Norwood and Pratten, Jour. Acad. Nat. Sci., Phil., (2), vol. 3, p. 71, pi. 9, figs, la-cl. 1914. Spiriferina spinosa Weller, 111. State Geol. Surv., Monog. I, p. 299, pi. 35, figs. 50-58. In Monroe and Randolph counties, Illinois, Spiriferina transversa is the common member of the genus in the Renault formation, while S. spinosa becomes much more common in the Paint Creek. Both species, however, occur in each formation, and also in still higher formations of the Chester. The same distributional relationship of the two species in the Lower Chester is observed in southeastern Illinois, S. transversa is common in the "Upper Ohara," being especially abundant in the Shetlerville formation, and S. spinosa is much more common in the Paint Creek. While this evidence alone —23 G 354 GEOLOGY OF HARDIN COUNTY would have but little weight in the decision of a 'question of correlation, yet in association with the mass of other evidence which unites the "Upper Ohara" with the typical Kenault, and the beds between the Bethel and Cypress sandstones with the Paint Creek, it serves to strengthen the main contention. Cliothyridina sublamellosa (Hall) Plate VIII, Figs. 21-24 1858. Athyris sublamellosa Hall, Geol. Iowa, vol. 1, pt. 2, p. 702, pi. 27, figs. la-c. .1914. Cliothyridina sublamellosa Weller, 111. State Geol. Surv., Monog. I, p. 482, pi. 80, figs. 31-36. This is one of the very common species in the Lower Chester formations, although it is not limited to these horizons. It is a long range species, being present in the Ste. Genevieve limestone below, and in the Middle Chester above. Composita trinuclea (Hall) Plate VIII, Figs. 19-20 1856. Terebratula trinuclea Hall, Trans. Albany Inst., vol. 4, p. 7. 1914. Composita trinuclea Weller, 111. State Geol. Surv., Monog. I, p. 486, pi. 81, figs. 16-45. The genus Composita is present in practicallv every Mississippian fauna, and in many horizons it is one of the commonest forms. If is one of those genera in which the species exhibit very great variation, and after studying large collections the observer is always at a loss to know whether he should lump all together from all horizons in a single protean species, or to separate them into innumerable, more limited specific groups. As a rule the specimens from one horizon with a limited geographic range, are fairly well circum- scribed in their characters, at least the individuals in such a group vary less among themselves than the group as a whole differs from some) other group from another horizon or locality. However, it is always possible to find numerous examples which are intermediate in character between any of the groups that may be recognized. In the Chester, the examples of Composita found in the Lower and Middle divisions of the group are notably different from those in the Upper division. The earlier type is commonly smaller and more elongate in out- line, the higher one being large and broad, and more or less subquadrate in outline. The earlier form is not notably different from specimens which may be found in the underlying Ste. Genevieve, St. Louis, and Spergen faunas, and all are here referred to the species C. trinuclea. PALEONTOLOGY 355 GOLCONDA AND LOWER OKAW SPECIES Pentremites obesus Lyon Plate X, Figs. 1-3 1851. Pentremites obesus Lyon, Rep. Ky. Geol. Snrv., vol. 3, p. 469, pi. 2, figs. 1-ld. 1905. Pentremites obesus Ulrich, U. S. Geol. Snrv., Prof. Paper, No. 36, p. 64, pi. 7, figs. 1-4. This is the largest species of Pentremites in the Chester faunas of Illinois, and is perhaps the largest known member of the genus. It is highly characteristic of the basal portion of the Golconda limestone, and at this time it has not been found in any other position. No specimens of the species have yet been collected in Hardin County, but they may be looked for in the outcrops of the lower portion of the Golconda limestone in the county. Horizon and locality. — Lower part of the Golconda limestone, Pope County, Illinois, and Crittenden County, Kentucky. Pentremites platybasis n. sp. Plate IV, Figs. 37-42 Description. — Body of medium size or smaller, wider than high in the typical form, the greatest width at or near the lower extremity of the am- bulacra. The dimensions of the holotype, a nearly perfect specimen, are: height 12.8 mm., greatest width 13.5 mm., length of ambulacra 11.9 mm., greatest width of ambulacra 4.1 mm. Dorsal region of the body nearly flat, the surface extended into five shoulders reaching to the lower ends of the ambulacra, the spaces between very gently concave towards the outer margin and curving uninterruptedly into the interambulacral areas of the ventral region; the basal plates reaching about half way from the stem facet to the lower extremities of the ambulacra, marked at their center by a low, triangu- lar callosity the points of which are directed along the three basal sutures, the stem facet in the center of this callosity; the base slightly indented radially along the three basal sutures, the indentation along the suture be- tween the two larger basal plates being somewhat stronger than the others. Ventral region of the body, from the extremities of the ambulacra to the summit, with converging sides, the profile being gently convex with the con- vexity increasing somewhat above, the summit truncated. Ambulacra nearly flat at their lower extremities, becoming very broadly V-shaped towards the summit, each of the two lateral surfaces gently convex below, becoming essentially flat above ; the lateral grooves ten to twelve in the space of 3 mm., directed transversely at the summit, becoming oblique towards the extremities of the ambulacra with their inner ends curving slightly in a ventral direction. Interambulacral areas nearly flat transversely, with the margins slightly 356 GEOLOGY OF HARDIN COUNTY elevated, the marginal elevation becoming a little greater ventrally. The deltoid plates less than one-half the length of the ambulacra, their dorsal margins straight and meeting at the median line in an angle greater than 90°. Remarks. — This species is perhaps a successor of P. godoni, in the Okaw and Golconda limestones. In its typical form it differs from the earlier species in its flatter base and in the shallow Y-shaped cross-section of the ambulacra towards the summit. The lateral furrows of the ambulacra are also a little finer than in P. godoni. Like all species of this genus, this one exhibits a considerable range of variation, the most noticeable variable char- acters being the flatness of the base and the depth of the V-shaped ambulacral groove. Some specimens which are associated with typical examples and are connected with them by all intermediate variations are higher than they are wide and have a base that is less flat, in some cases the profile of the base being not unlike that of P. godoni, The characters of the ambulacra are constantly different from P. godoni, but their depth towards the summit exhibits considerable variation. Horizon and locality. — Lower Okaw limestone, Randolph and St. Clair counties, Illinois; Lower Golconda limestone, Pope and Hardin counties, Illinois. Pentremites nodosus Hambach Plate IV, Fig. 25 1880. Pentremites nodosus Hambach, Trans. St. Louis Acad. Sci., vol. 4, p. 155, pi. B, fig. 2. Description. — Body of medium size or larger, higher than wide, the greatest width at the extremities of the ambulacra, the dorsal and ventral regions subequal in height. The dimensions of a nearly perfect specimen are : height 27 mm., greatest width 22.7 mm., length of ambulacra 17.9 mm., greatest width of ambulacra 5.4 mm., distance from stem facet to ends of ambulacra 17 mm. Dorsal region obpyramidal in form, the lines separating the surfaces of the pyramid well-defined towards the ambulacral extremities, becoming obsolete towards the base, the profile of the dorsum is concavely curved above, becoming straight or even a little convex as it approaches the stem facet; transversely the surfaces are concave between the ambulacral extremities; on three sides, determined by the sutures between the basal plates, there are somewhat flattened surfaces towards the base of the dorsum, so that the pentagonal cross section at the plane separating the dorsum and the ventral regions, becomes somewhat triangular towards the base; the basal plates reaching about one-half or a little less than one-half the distance from the stem facet to the bases of the ambulacra. Ventral region subhemispherical or subhemiellipsoidal in form. The ambulacra with a rather broad median PALEONTOLOGY 357 groove, each of the two lateral surfaces nearly symmetrically depressed convex transversely, in some examples with the slope to the median groove a little more abrupt, these two surfaces together forming a nearly flat ambulacra! surface which may be gently convex; or sloping inward to the median groove to form a very wide and shallow V-shaped cross-section. The side plates eight to ten in number in the space of 3 mm. ; the lateral grooves nearly straight and transverse towards the summit, becoming somewhat oblique distally. The interambulacral regions distinctly concave in a transverse direction in their outer portion; in the upper portions, including the deltoid plates and the extremities of the radials, they are more nearly flat, especially in older individuals. The lateral margins of the interam- bulacral regions somewhat raised and roughened, this roughened region narrow towards the extremities of the ambulacra, but broader above and con- nected entirely across the areas with an arch-like outline below the deltoid plates, and from this position towards the summit occupying much of the area of the deltoid plates. This roughness of the plates most pronounced about opposite the lower extremities of the deltoid plate giving rise to the thickening of the radial plates which doubtless suggested the specific name of the form. The deltoid plates rather small, subrhomboidal in form, the two free margins commonly a little longer than the others, the sutures between the deltoids and the radials straight in the median region, but curving out- ward as they approach the margins. Remarks. — The roughness or callous development upon the interam- bulacral regions of this species, which has given origin to its specific name, is a character that is somewhat inconstant, being nearly obsolete upon some individuals. It is quite likely that the feature increases in prominence with the age of the individuals, and that younger specimens may be quite free from it. Horizon and locality. — Lower Okaw limestone (lower part), Randolph County, Illinois. Pentremites okawensis n. sp. Plate, X, Figs. 5-7 Description: — Body of medium size, much higher than wide, the greatest width at the extremities of the ambulacra, the ventral region shorter than the dorsal. The dimensions of the holotype are: height 25.3 mm., greatest width 15.7 mm., length of ambulacra 12.5 mm., greatest width of ambulacra 5 mm., distance from the stem facet to the extremities of the ambulacra 16 mm. The dorsal region obpyramidal in form, with the lines separating the faces of the pyramid well defined towards the ends of the ambulacra but becoming nearly or quite obsolete towards the stem facet; the profile of the dorsum with straight or nearly straight sides through the greater portion of 358 GEOLOGY OF HARDIN COUNTY their length, but becoming a little concave towards the basal extremities in many specimens; the surfaces of the dorsum flat transversely towards the line separating the dorsal and ventral regions, longitudinally these surfaces curve without interruption into the interambulacral regions of the ventral region; the basal plates reaching less than one-half the distance from the stem facet to the ambulacra. Ventral region having the sides converging with a gently convex curvature to the truncate summit. The ambulacra very slightly convex or nearly flat transversely, with the median groove well defined, each of the two lateral surfaces unsymmetrically depressed convex, the inner margins curving rather abruptly to the median groove, much the larger portion of the surfaces essentially flat or with the extreme outer border having a slight downward curvature; the side plates seven to nine in number in the space of 3 mm., the transverse grooves nearly straight and horizontal near the summit, becoming oblique towards the extremities of the ambulacra. The interambulacral regions flat transversely, the lateral margins only slightly elevated towards the extremities of the deltoid plates. The deltoid plates small, their tips not extending to the summit. Remarks. — In its flat interambulacral surfaces, both in the dorsal and ventral regions, and in its small and short deltoid plates, this species re- sembles P. pyramidatus. It differs from that species in its more elongate and consequently narrower form, with the elongation more marked in the dorsal than in the ventral region, the distance from the stem facet to the bases of the ambulacra being greater than the length of the ambulacra. In the collections that have come under the observations of the writer, P. pyramidatus is more commonly met with in the Paint Creek faunas, while this species occurs commonly in the Middle Chester faunas, both in the Mississippi Valley region and in the Ohio Valley. Horizon and locality. — Okaw limestone, Eandolph County, Illinois. Gol- conda limestone, Pope and Hardin counties, Illinois. Glen Dean limestone, Hardin County, Illinois. Mesoblastus incurvatus n. sp. Plate VIII, Figs. 29-34 Description. — Body small, pentangularly subglobose in form, height and width subequal or the height a little greater than the width, the base im- pressed a little within the distal extremities of the ambulacra. The dimen- sions of a nearly undistorted specimen are : height 7 mm., width 7 mm., width of ambulacra 1 mm. The largest specimen observed, a distorted example is approximately 10 mm. in height. Ambulacra very narrow, the median groove rather faint distally, becoming more pronounced towards the oral extremity where it is nearly equal in strength to the lateral grooves which border the ambulacra, giving to the proximal third or fourth a three grooved appearance ; PALEONTOLOGY 359 the lancet plate nearly or wholly covered by the side plates distally, its central portion exposed proximally by the shortening of the side plates; side plates between 30 and 35 in number, broader distally, becoming narrower towards the summit and consequently the lateral pores of the ambulacra become much more crowded proximally. Interambulacral spaces gently concave trans- versely, becoming rather deeply concave between the distal extremities of the ambulacra; the deltoid plates about one-fourth the length of the body, the base of the plates obtusely angular, the apex acutely angular, the entire plate being unsymmetrically rhomboidal in outline; the lateral margins of the interambulacral spaces becoming progressively more elevated above the am- bulacra towards the summit of the body. Plates nearly smooth, marked only by indistinct lines of growth parallel with the base of the deltoid plates. Remarks. — Two other species of the genus Mesoblastus have been de- scribed from the Chester faunas, ill. glaber and M. sphaeroidalis. The first of these is a common species in the Shetlerville formation and many ex- amples have been collected at Fairview bluff below Eosiclare and elsewhere. ill. incurvatus differs from M. glaber in its larger size, being on the average at least twice as large as the Shetlerville species, by the greater incurving of the distal extremities of the ambulacra, the tips of these areas being relatively much closer to the stem facet; the stem facet is more deeply impressed be- tween the tips of the ambulacra ; the deltoid plates differ in shape, the sutures between the tips of the forked plates and the deltoids being essentially straight and diverging at an angle a little greater than a right angle, while in iH. glaber these two sutures are in more nearly a straight horizontal line and are curved, the base of the deltoid plate being a double curve with the con- vexity towards the oral extremity of the body ; the plates of ill" . incurvatus are smoother, especially in the region near the junction of the forked plates with the deltoids, than in ill. glaber. The specific name of ill. glaber is a misnomer. The type specimens are nearly smooth, but they are apparently somewhat worn examples. Per- fectly fresh, unworn specimens from the Shetlerville shales at Fairview bluff commonly show distinct, ridge-like lines of growth upon the deltoid plates and upon the distal portions of the forked plates, the more distinctly marked of these lines being parallel with the basal sutures of the deltoid plates. Both of the other species are really smoother than ill. glaber. M. sphaeroidalis is less well known than M. glaber, the single type speci- men from Meade County, Kentucky, being the only one observed by the writer, is labeled Kaskaskia group, so its proper position in the Chester is not recorded. In size and in the shape of the deltoid plates this species resembles ill. incurvatus, but in general form and proportions, and in the less incurvature of the extremities of the ambulacra, it is more like ill. glabra. Horizon and locality. — Golconda limestone, NE. V4 sec. 28, T. 12 S.. E. 7 E., Pope County, Illinois. 360 GEOLOGY OF HARDIN COUNTY Pterotocrinus capitalis (Lyon) Plate VIII, Figs. 36-39 1875. Asterocrinus capitalis Lyon, Geol. Rep. Ky., vol. 3, p. 472, pi. 5, figs. la-Ik. 1897. Pterotocrinus capitalis Wachsmuth and Springer, X. Amer. Crin. Cam., p. 794, pi. 79, figs. 6a-b. This species is the genotype of Pterotocrinus, a genus of crinoids which is characterized by the development of a remarkable series of five plates, interradial in position, growing out from the ventral surface. These special- ized plates, commonly referred to as the "wing plates/ 5 exhibit a great variety of forms, and the species are differentiated largely upon the characteristics of these plates. Complete specimens of members of the genus are rarely met with, and even specimens of the dorsal cups are rare, the parts usually found being the separated "wing plates." In P. capitalis these plates are much more massive than in any of the other known species of the genus, their form being well exhibited by the illustrations shown on Plate VIII. This is per- haps the best index fossil of the lower portion of the Golconda limestone, but it has not yet been observed in the equivalent part of the Okaw limestone of Randolph County. Horizon and locality. — Golconda limestone, Johnson, Pope, and Hardin counties, Illinois, Crittenden County, Kentucky. Camarophoria explanata (McChesney) Plate IX, Figs. 35-36 1860. Rhynchonella explanata McChesney, Desc. New Pal. Foss., p. 50. 1914. Camarophoria explanata Weller, 111. State Geol. Surv., Monog. I, p. 173, pi. 23, figs. 46-51. So far as is known this species makes its first appearance in the Chester faunas of Illinois in the Okaw and Golconda limestones, which are equivalent formations. Not a single example has yet come to the notice of the writer from the Lower Chester formations, but it does continue upwards into the higher formations of the group. Scarcely a Middle Chester fauna will be met with in which the species is wanting, and in many places it is one of the common species. Cliotpiyridina sublamellosa (Hall) PJate IX, Figs. 39-41 Examples of this species occur in great numbers in the Golconda and Glen Dean, wherever these formations are well exposed. They are especially abundant in some localities in the lower portion of the Golconda limestone. The Middle Chester examples are not notably different from those found in PALEONTOLOGY 361 the Lower Chester, unless it be in the slightly coarser divisions of the con- centric, lamellar expansions of the shell. COMPOSITA TRINUCLEA (Hall) Plate IX, Figs. 37-38 The examples of Composita found in the Middle Chester faunas are not essentially different from those in the Lower Chester which have already been noticed. The species occurs in abundance in the Golconda and Glen Dean faunas in Hardin County and elsewhere, wherever these formations are well exposed. XtTCULA PLATYNOTUS n. sp. Plate IX, Figs. 11-14 Description. — Shell small, subtrigonal in outline, with very deep valves. The dimensions of the holotype, a complete specimen with valves in articula- tion are: length 11.1 mm., height 9.2 mm., thickness 9.3 mm. Beaks promi- nent, strongly incurved, situated nearly opposite the anterior margin, but not quite terminal. Anterior margin a little concave above, just in front of the beaks, becoming gently convex below and rounding without interruption into the ventral margin which is gently convex or nearly straight in its central portion, curving upward more abruptly in front and behind; postero-ventral margin short, narrowly rounded; postero-dorsal margin long, nearly straight or gently concave, becoming a little convex as it approaches the beaks. Valves very convex, flattened on the umbo, with prominent, narrowly rounded umbonal ridge passing from the beaks to the postero-basal margin; post- umbonal slope very abrupt, these slopes of the two valves when in articulation, forming a broad, nearly flat, postero-dorsal region, becoming concave towards the beaks and slightly keeled along its median line posteriorly, where the cardinal margins of the two valves join. The antero- ventral slope flattened longitudinally for some distance in front of the umbonal ridge, with a very shallow, ill-defined sinus passing obliquely from the beak to near the middle of the ventral margin, strongly convex vertically, rounding abruptly to the antero-dorsal margin in front of the beak. Surface of valves marked by fine, continuous, concentric costae somewhat variable in size, becoming less regular towards the ventral margin in mature shells. An incomplete internal cast shows a pair of strong posterior adductor muscular impressions and well- defined foot muscle scars, and taxodont hinge tooth impressions. Remarks. — This is a unique type of pelecypod shell in the Chester faunas, or in the faunas of any of the Mississipian formations of the Mississippi Valley. In its general form it is suggestive of some species of Schizodus, but it is a thicker shell than is usual in that genus, and moreover it possesses a taxodont hinge dentition which, so far as its characters have been determined, 362 GEOLOGY OF HARDIN COUNTY corresponds with the dentition in Nucula. The species has been placed in the genus Nucula entirely upon the evidence afforded by the hinge characters. The first specimens of this shell that came under observation were col- lected from near the base of the Okaw limestone on Camp Creek in Randolph County, Illinois. Later, the species was collected from near the base of the Golconda limestone in Johnson County, where it has been found to be more abundant and in general somewhat better preserved than in Eandolph County. The mature examples from Johnson County attain a somewhat larger size than any of the Camp Creek specimens, but this difference in size is not a constant character, and all are certainly members of the same species. Horizon and locality. — 'Basal Okaw limestone, Camp Creek, Randolph County, Illinois; and basal Golconda limestone, 3% miles southeast of Vienna, Johnson County, Illinois. Nucula randolphensis n. sp. Plate IX, Figs. 24-30 Description. — Shell of medium size, subovate in outline. The dimen- sions of two mature examples with the valves in articulation are: length 11 mm. and 10.8 mm., height 9.7 mm., and 9.9 mm., thickness 7.5 mm. and 7.5 mm. The beaks strongly incurved, those of the two valves nearly in contact, situated from a fourth to a third of the shell length from the anterior ex- tremity, the umbonal region very prominent in mature shells. The antero- dorsal margin sloping more or less abruptly from in front of the beaks, in some examples passing without interruption to the most anterior point of the shell margin, in others with the margin just above the anteriormost point of the shell slightly truncate; the ventral margin convexly curved through- out, passing without interruption into the postero-ventral margin which is somewhat broadly rounded ; the postero-dorsal margin gently convex or nearly straight, sloping backward from behind the beaks. The valves strongly con- vex, the antero-dorsal and postero-dorsal portions curving somewhat abruptly to the margins, becoming distinctly incurved in front of, and especially for a short distance back from the beaks, the region back of the beaks forming a broad, ill-defined escutcheon; the lunule in front of the beaks less well defined than the escutcheon. The surface of the valves marked by fine, regular, concentric costae, and in some examples by one or more stronger lines of growth. The shell substance thick, with deeply impressed muscle scars; the internal casts of the shell more pointed posteriorly than the shell itself. Remarks. — This species differs from Nucula illinoie?isis in its larger size, and its proportionally lesser length. On comparison of young examples of N. randolphensis with mature individuals of N. illinoiensis of equal size, the valves of N. randolphensis are much less convex so that the shells are PALEONTOLOGY 363 very much thinner, but with growth to maturity the shells of the larger species become as obese as those of the smaller form. Perfect examples of this species exhibit considerable variation in the outline of the shell, especially in the truncation of the antero-dorsal margin, and in the sharpness of the rounding of the posterior extremity. Horizon and locality. — Basal Okaw limestone, Camp Creek, Eandolph County, Illinois; and basal Golconda limestone, 3% miles southeast of Vienna, Johnson County, Illinois. Leda chesteeexsis n. sp. Plate IX, Figs. 31-34 Description. — Shell large for the genus, subovate in outline, contracted posteriorly. The dimensions of a practically perfect right valve are: length 22 mm., height 13.4 mm., convexity 5 mm. A larger individual has a height of 18.8 mm., but it is worn at the posterior extremity so its length cannot be determined accurately. The beaks are directed posteriorly, situated near the mid-length of the shell. The antero-dorsal margin gently convex, with the greatest convexity just in front of the beaks, passing without interruption into the rounded anterior margin; the ventral margin convexly curved, the curvature a little more abrupt in some examples at a point nearly opposite the beaks, becoming straighter both in front and back of this point ; posterior margin very short, abruptly rounded below and subtruncate above, rounding abruptly into the ventral and dorsal margins; postero-dorsal margin con- cave, the greatest concavity commonly just back of the beaks. Valves strongly convex, the greatest convexity along a vertical line through the beaks; along a straight or slightly curved line from the beak to a point near the center of the ventral margin, the regular convex curvature of the surface of the shell in an antero-posterior direction, is interrupted by a slightly more abrupt curvature, which gives the appearance of a low, ill-defined, rounded fold crossing the valves in a nearly vertical direction, in some shells this fold becomes obsolete towards the ventral margin, but in others it continues entirely across the valve. Along the antero-dorsal margin the shell surface is incurved to the hinge margin; postero-dorsally the surface is abruptly inflected to form a broad, flattened escutcheon which is distinctly depressed behind the beaks. Surface of valves marked by fine, regular, concentric costae, five or six of which commonly occupy the space of one millimeter on the body of the shell, becoming finer towards the beak. Remarks. — This species is of the type of Leda nasuta Hall, from the Spergen limestone, and Leda curia M. and W., from the Ste. Genevieve lime- stone, but it attains a very much greater size than either of these species, and it is distinguished also by the indefinite vertical fold which crosses each 364 GEOLOGY OF HARDIN COUNTY valve. Very young specimens agree with the older species more closely than do the mature shells, the mature individuals being perfectly distinct. Horizon and locality. — Basal Okaw limestone, Camp Creek, Randolph County, Illinois; and basal Golconda limestone, 3% miles southeast of Vienna, Johnson County, Illinois. CONOCARDIUM CHESTERENSIS n. Sp. Plate IX, Figs. 21-23 Description. — Shell small, the dimensions of the largest example ob- served are: height along an oblique line from the beak to the ventral margin 10.7 mm., length along an oblique line from the anterior extremity of the hinge-line to the postero-basal extremity 14 mm. Beak situated near the mid-length of the shell, only a little produced beyond the hinge-line. The antero-cardinal extremity of the shell acutely angular; the antero-ventral margin nearly straight or very gently convex; the postero-ventral margin rounded and bending abruptly into the posterior margin which is obliquely truncate; the postero-cardinal extremity not noticeably produced beyond the shell margin. Valves with their greatest convexity along an oblique band extending from the beak to the rounded postero-basal margin, this band gently convex antero-posteriorly, and strongly convex dorso-ventrally. In front of the median band the surface sloping away without interruption into the trian- gular anterior portion of the valve which becomes compressed towards the anterior extremity and incurved rather abruptly to the hinge-margin. The posterior slope of the valves from the median band much more abrupt than the anterior, and nearly flat throughout the greater portion of its width, becoming a little deflected close to the cardinal line. Surface of the valves marked by radiating costae, about five, in some cases six of the stronger ones occupying the oblique median band of the valve; the anterior slope with from twenty to thirty or more costae, according to the size of the shell, which are much finer than those upon the median band; the costae upon the posterior slope ten or twelve in number, and these likewise much finer than those of the median band; the strongest rib of the entire valve is the one along the posterior border of the median band. In addition to the radiating ribs, the shell surface is marked by very fine, raised, concentric lines which are well-defined in the intercostal spaces, and obsolete on the tops of the radiating costae, also in most examples several concentric lines of growth are present. Rmarks. — This form is allied to C. mesialis from the Ste. Genevieve limestone, but it differs from that species in the less abrupt differentiation of the mesial, band from the anterior slope of the valve, and in the absence of a posterior prolongation of the shell at the cardinal margin. It is possible that the postero-cardinal portion of the shell was really somewhat more pro- duced in living shells than seems to be the case in the fossil specimens that PALEONTOLOGY 365 have been examined, all of which have been more or less worn, but in none of these is there any indication of the shell outline itself, or in the direction of the lines of growth, that they ever possessed such an extension as is present in the type of 0. mesialis. Horizon and locality. — Basal Okaw limestone, Camp Creek, Eandolph County, Illinois ; and basal Golconda limestone, 3y 2 miles southeast of Vienna, Johnson County, Illinois. Bellerophon chesterensis n. sp. Plate IX, Figs. 1-6 Description. — Shell of medium size, snbglobose, closely coiled, without umbilicus, the length and width about equal. The dimensions of an essen- tially complete example are: greatest length 20 mm., greatest width across the aperture 20 mm., length of aperture at its mid-width 7 mm. Outer volution broadly rounded dorsally, becoming somewhat flattened in mature shells towards the aperture; center of dorsum marked by a narrow, elevated band, rounded over the top. Aperture transverse, twice as wide as high, with a deep slit dorsally opposite the dorsal band; the inner lip callosed, the cal- losity extending over the umbilical region in mature shells. Surface of the shell marked only by more or less regular lines of growth which are much more conspicuous towards the aperture of the larger shells. Remarks. — This species resembles B. sublaevis from the Spergen lime- stone, but the shell is proportionally broader, and more flattened on the dorsum, with more prominent dorsal band, and commonly with more con- spicuous lines of growth. Horizon and locality. — Basal Okaw limestone, Camp Creek, Eandolph County, Illinois; and basal Golconda limestone, 3% miles southeast of Vienna, Johnson County, Illinois. EUPHEMUS RANDOLPHENSIS n. Sp. Plate IX, Figs. 7-10 Description. — Shell small, subglobose, closely coiled, without umbilicus. The dimensions of a nearly complete but rather small example are: greatest length 10 mm., greatest width across the aperture 8.6 mm., length of aperture at its middle line 3 mm. One of the largest examples observed has a maxi- mum length of about 15 mm. Outer volution with a more or less distinctly rounded ridge medially, from which the surface slopes laterally, becoming more convex a little beyond the middle of each side and curving somewhat abruptly into the umbilical region. Aperture transverse, with a broadly rounded sinus in the outer margin opposite the dorsal ridge; the dorsal band rather broad, rounded transversely, not sharply defined laterally. Surface 366 GEOLOGY OF HARDIN COUNTY marked by fine, regular, raised, revolving costae which are narrower than the intercostal spaces, and which terminate abruptly at about the middle of the outer volution; the outer half of the volution nearly smooth or marked only by more or less inconspicuous lines of growth. In some examples there is a more or less prominent, rounded node upon each lateral slope of the outer volution at about the position of the termination of the revolving costae, and in these same specimens the dorsal ridge is commonly somewhat more promi- nent than in others, giving to the shell a somewhat trilobate form in that portion of the volution. Remarks. — The presence of Bellerophontid shells, even as rare indi- viduals, in the Chester faunas of the Mississippi and Ohio valleys is unusual, and no species of Euphemus has ever been described from the faunas, although Girty has reported the genus from the Batesville sandstone of Arkansas. The occurrence of a species of Euphemus, therefore, in the Okaw limestone on Camp Creek, represented literally by scores of finely preserved individuals, is a remarkable faunal character. Besides this form the same fauna includes a number of other Bellerophontid shells belonging to at least two different genera, some of which are fairly common. This group of species gives to the Camp Creek fauna a very unusual character for the Chester. The same species of Euphemus, along with at least two of the other bellerophontid species from Camp Creek, occur also at one locality in the lower Golconda limestone in Johnson County. The species is fully as abundant in the Johnson County locality as on Camp Creek, and the specimens from the two localities are alike in all essential characters. There are also other notable characteristics that unite the faunas from the two localities. This species resembles the Pennsylvanian Euphemus carbonarius, but it is somewhat less globular in form, with less strong revolving costae, and in most examples with more prominent dorsal ridge upon the outer volution. Horizon mid locality. — Basal Okaw limestone, Camp Creek, Randolph County, Illinois; and basal Golconda limestone, 3% miles southeast of Vienna, Johnson County, Illinois. BUCANOPSIS ORNATUS 11. Sp. Plate IX, Figs. 15-20 Description. — Shell of medium size, with small umbilicus. The dimen- sions of the holotype are: greatest length 21 mm., greatest width across the aperture 21.1 mm., length of aperture at its mid-width, approximately 11 mm. Outer volution rather narrow where it leaves the inner margin of the aperture, expanding regularly from this position until it approaches the aperture where it is distinctly flaring, its surface rather narrowly rounded across the median line, the sides convex and curving rather abruptly into the umbilicus. Dorsal band of medium width, elevated above the surface on PALEONTOLOGY 367 either side, rounded transversely, becoming much more strongly elevated towards the aperture. The aperture about twice as wide as long, with a deep mesial slit in its outer lip, the inner lip with a thin callosity. Surface marked by very fine revolving costae, irregularly alternating in size or nearly uniform from five to nine occupying the space of one millimeter ; crossing the revolving costae there are equally fine but somewhat fainter, transverse markings, the two sets of lines producing an exceedingly fine reticulate ornamentation; the dorsal band marked by longitudinal and transverse lines similar to those on the body of the shell. In addition to the finer markings, the larger shells especially, are marked by more or less conspicuous lines of growth, which in some examples obscure the finer markings. Remarks. — This species is found associated with Euphemus randolphensis in both Randolph and Johnson counties. In the Johnson County locality it is the largest and one of the commonest Bellerophontid shells, but not nearly so abundant as Euphemus randolpliensis. In the Camp Creek locality in Randolph County, only a single example has been met with, and while it is somewhat imperfect, it is clearly the same species as that from Johnson County. Several of the larger shells of the species show evidence of having been injured during their growth, the healed lines of growth being very evident, a condition that is perhaps indicative of the thinness of the shells. The surface markings of the shell suggest B. textilis from the Spergen lime- stone, but this Chester species is very much larger and has a more flaring- aperture. Horizon and locality. — Basal Okaw limestone, Camp Creek, Randolph County, Illinois; and basal Golconda limestone, 3% miles southeast of Vienna, Johnson County, Illinois. 3G8 GEOLOGY OF HARDIN COUNTY MISCELLANEOUS CHESTEE SPECIES Pentremites spicatus Ulrich Plate X, Fig. 8 1918. Pentremites spicatus Ulrich, Ky. Geol. Surv., Miss. Form. West Ky., p. 262, pi. 7, figs. 33-35. (Name and illustrations published with- out description.) Description. — Body large, deeply pentalobate, the ventral region much higher than the dorsal, the greatest width at or a little above the distal ex- tremities of the ambulacra. The dimensions of a nearly perfect specimen are: height 37.9 mm., greatest width 33.1 mm., length of the ambulacra 31.5 mm., distance from the center of the stem facet to the distal extremities of the ambulacra 19.2 mm. Dorsal region with a concave profile from the distal extremities of the ambulacra to the stem facet, the surfaces extending out- ward to the ambulacra in the form of five conspicuous shoulders, the inter- ambulacral spaces becoming deeply concave towards the outer border. Ventral region having sides with a gently convex profile, converging towards the summit. The ambulacra broad, and deeply Y-shaped transversely, each of the two sides flat or slightly concave towards the summit, becoming less concave or slightly convex distally; a narrow lateral border on each side, in which are situated the lateral pores, slightly incurved to the lateral surfaces of the radial and deltoid plates. Side plates eight to ten in the space of 3 mm., the lateral grooves transverse throughout most of the ambulacra, be- coming slightly oblique distally. Interambulacral areas deeply concave below, becoming shallower above; a rather broad band along the lateral borders differentiated from the central region and more or less distinctly marked by oblique irregularities which in some cases give a serrate margin to the plates. The deltoid plates, Avhen uninjured, extending conspicuously above the summit in five points. Remarks. — This is the species that has been referred commonly by authors to P. sidcatus, a species which, as shown in Boemer's original illus- tration, does not possess the projecting deltoid plates which are so character- istic of P. spicatus. The species occurs abundantly in the Glen Dean lime- stone in many localities, from Eandolph County, Illinois, to Breckenridge County, Kentucky, and it seems to be limited to this horizon, but it has not yet been met with in Hardin County. Horizon and locality. — Upper Okaw limestone (Plum Creek beds), Eandolph County, Illinois. Glen Dean limestone, Pope County, Illinois, and Crittenden and Breckenridge counties, Kentucky. PALEONTOLOGY 369 Pentremites brevis III rich Plate IV, Figs. 43-46 1918. Pentremites brevis Ulrich, Butts, Ky. Geol. Surv., Miss. Form. West. Ky., p. 100, pi. 24, fig. 6. (Named and illustrated but without published description.) Description. — Body of medium size, almost as wide as high or a little higher than wide, the greatest width at the lower extremities of the ambu- lacra. The dimensions of a practically perfect individual are: height 17.7 mm., greatest width 17.3 mm., length of ambulacra 15.8 mm., greatest width of ambulacra 5.4 mm. Dorsal region of the body depressed to a somewhat variable degree, in some examples nearly flat ; the stem facet nearly flush with the general surface, or slightly depressed with three, radiating, depressed lines extending outward along the basal sutures, situated in the center of a trifoliate callosity, the lobes of which reach out along the basal sutures ; the ambulacral regions of the dorsal surface a little prominent towards the lower extremities of the ambulacra, the interambulacral surfaces flat or gently concave transversely, curving regularly into the interambulacral surfaces of the ventral region. Ventral region of the body, from the lower extremities of the ambulacra to the summit, with gently convex sides converging towards the summit, which is subtruncate. Ambulacra broadly V-shaped, the median depression extending to the. lower extremities of the areas but increasing in depth towards the summit, each of the two lateral surfaces nearly flat or very gently convex towards the outer margins and flattened towards the median line; the lateral grooves nine to thirteen in the space of three millimeters, directed obliquely inward and upward in the lower portion of the ambulacra, becoming transverse towards the summit. Interambulacral areas nearly flat transversely, becoming slightly concave below in some examples, the lateral margins very slightly elevated as they approach the summit. The deltoid plates less than one-half the length of the ambulacra, the dorsal margins gently convex and meeting at the median line in an angle less than 90°. Remarks. — This species is most closely related to P. platybasis which occurs commonly near the base of the Okaw limestone and in the lower portion of the Golconda limestone. The general form of the two species is essentially alike, and the callosity upon which the stem facet is situated is similar in the two species. The only really important difference between the two forms is found in the depth of the V-shaped ambulacral surfaces, this depression being much deeper in P. brevis than in P. platybasis. The deep- ening of the V-shaped groove is apparently a progressive character, in P. godoni of the Paint Creek formation the ambulacral surfaces are flat or slightly convex; in the lower Okaw and Golconda limestone species, P. platy- habis, it is somewhat V-shaped ; while in the species under consideration from the Glen and Upper Okaw limestones, it is much depressed. —24 G 370 GEOLOGY OF HARDIN COUNTY The specimens that are here illustrated are from the Upper Okaw Plum Creek beds of Randolph County. Typical examples of the species are present in the Glen Dean formation of Hardin County, but no Hardin County examples have been met with which are as large as some of those from Randolph County. Horizon and locality. — Upper Okaw, Plum Creek beds, Randolph County, Illinois; Glen Dean limestone, Hardin County, Illinois. Penteemites fohsi Ulrich Plate X, Fig. 4 1905. Pentremites fohsi Ulrich, U. S. Geol. Surv., Prof. Paper No. 36, p. 64, pi. 7, figs. 5-9. (Name and illustrations published without description.) Description. — Body large, higher than wide, the greatest width at the distal extremities of the ambulacra. The dimensions of an average sized individual are : height 38.8 mm., greatest width 30.4 mm., length of ambu- lacra 28.2 mm., distance from center of stem facet to base of ambulacra 22.3 mm. Dorsal region of the body obpyramidal with the edges between the faces of the pyramid fairly well defined towards the extremities of the ambulacra, but becoming obsolete towards the base; the basal plates covering about one- half or less than one-half the distance from the stem facet to the ends of the ambulacra. Ventral region somewhat higher than the dorsal, the surfaces converging upward with a gently convex curvature, truncate at the summit. Greatest width of the ambulacra about one-fourth of their length, the two lateral surfaces sloping inward to the median groove, making the ambulacra broadly V-shaped in cross-section, each of the lateral surfaces nearly flat in their widest portion towards the summit, becoming gently convex in the narrower, distal portion. Side plates from eight to ten in number in a distance of 3 mm., the lateral grooves nearly transverse in direction towards the summit, becoming a little oblique towards the distal extremities of the ambulacra. The interambulacral areas nearly flat transversely, their lateral margins very slightly or not at all raised above the general surface. Deltoid plates about two-fifths the length of the ambulacra, their upper extremities not extended beyond the summit ventrall t y, their dorsal margins nearly or quite straight, meeting on the median line in an angle of less than 90°. Remarks. — This species may be recognized by its large size, although it is considerably smaller than P. obesus, by its essentially flat interambulacral areas, and the V-shaped cross-sections of the ambulacra. When this name was originally proposed by Ulrich the species was assigned to the "Birdsville" formation, which, of course, means but little in our present interpretation of the Chester. Butts has copied one of the original figures 1 and has assigned the species to the Glen Dean formation. 1 Ky. Geol. Surv., Miss. Form. West. Ky., pi. 24, fig. 21. PALEONTOLOGY 371 The specimen which is illustrated in this report is from the Menard lime- stone, and the species has not been found by the writer in any other formation. It occurs throughout southern Illinois from Union to Pope counties in a restricted horizon near the base of the Menard limestone only a short distance above the Waltersburg sandstone, and it has been found in this zone wherever the formation is well enough exposed to permit of collecting the fossils. Horizon and locality.- — Menard limestone, Union, Johnson, and Pope counties, Illinois. Pentremites cherokeeus Hall Plate X, Figs. 11-13 1858. Pentremites cherokeeus Hall, Geol. Surv. Iowa, vol. 1, pt. 2, p. 691, pi. 24, figs. 12a-c. Description. — Body of medium size, higher than wide, the greatest width at the distal extremities of the ambulacra. The dimensions of a nearly com- plete specimen are : height 14.2 mm., greatest width 13.8 mm., length of ambulacra 10.7 mm., distance from center of stem facet to distal end of the ambulacra 7.9 mm. Dorsal region broadly obpyramidal, the five faces of the pyramid scarcely defined near the base, but becoming depressed and distinctly concave transversely as they approach the line separating the dorsal and ventral regions, those portions of the dorsum beneath the extremities of the ambulacra forming five shoulder-like regions; the basal plates reaching about one-half the distance from the stem facet to the extremities of the ambulacra. Ventral region having the sides converging towards the summit with a gently convex curvature, but with the surfaces a little recurved in the inter- ambulacral regions towards the tips of the deltoid plates when these plates are uninjured; the summit, when the tips of the deltoid plates are disre- garded, is truncate. The ambulacra! areas broad, their width from two-fifths to nearly one-half their length, broadly V-shaped in cross-section above, becoming nearly flat towards the distal extremities, each of the two lateral surfaces nearly flat. Side plates nine to eleven in the space of 3 mm. Inter- ambulacral areas rather deeply concave transversely below, becoming flatter above towards the extremities of the deltoid plates; their lateral margins slightly differentiated along a narrow band adjacent to the ambulacra which is marked in some examples with more or less obscure, oblique irregularities. The deltoid plates, when perfectly preserved, are extended conspicuously above the general level of the summit in five, thin, delicate points which are dis- tinctly recurved towards their extremities. Remarks. — Hall used the name cherokeeus for a species of Pentremites which he described and illustrated in his Iowa report in 1858. The name was taken from Troost's manuscript, but the specimen used for description and illustration was not Troost's type of the species. In the recently published 372 GEOLOGY OP HARDIN COUNTY summary of Troost's early work 1 an illustration is given of the specimen which Troost had selected as the type of his species, and it is quite clearly the form to which Ulrich has recently given the name P. spicatus. Hall's specimen seems to be distinct, and the name P. cherokeeus must stand or fall upon Hair's interpretation. The specimens that are here referred to the species have been collected at a number of localities in the Menard limestone in Randolph and Johnson counties, and have been found in no other horizon. The larger of these is a little smaller than the specimen that Hall has illustrated, but in other re- spects the agreement is comparatively close. The deltoid plates in Hall's figures are somewhat shorter and are not produced above the summit into the delicate points of the specimens under discussion, but these may easily have been broken off, and the illustration does show the beginning of the recurvature of the deltoid plates towards the tip. Most of the recently collected specimens are in about the condition of the example figured by Hall. Under the circumstances it seems to be unwise to propose a new name for this form, although a careful study of Hall's type may make such a procedure necessary. Horizon and locality. — Menard limestone, Randolph and Johnson coun- ties, Illinois. Pterotocrinus bifurcatus Wetherby Plate X, Fig. 14 1879. Pterotocrinus bifurcatus Wetherby, Jour. Cinn. Soc. Nat. Hist., vol. 2, p. 136, pi. 11, fig. 1. 1897. Pterotocrinus acutus var. bifurcatus Wachsmuth and Springer, N. Amer. Crin. Cam., p. 801. pi. 79, figs. 9a-b. This is one of the most peculiar species of this peculiar genus. The "wing plates" which exhibit the most important differential characters of the several species, are thick and heavy, and are divided distal ly into two widely spreading, horn-like processes. The extent of these horns shows a considerable amount of variation in the different specimens of the species, but this feature in some degree of development seems to be constant. The speci- men illustrated on Plate VII is the original type specimen, and shows the five "wing plates" in place, from above. The species has been observed only in the upper or Plum Creek beds of the Okaw limestone in the Mississippi Valley counties, and in the Glen Dean limestone of the Ohio Valley. It has not yet been observed in Hardin County, but it may be looked for. Horizon and locality. — Upper Okaw (Plum Creek beds), Randolph County, Illinois. Glen Dean limestone, Breckenridge County, Kentucky. 'Wood. Bull. U. S. Nat. Mus., No. 64. PALEONTOLOGY 373 Pterotocrinus menardensis n. sp. Plate XI, Fig. 9 Description. — The "wing plates" only of this species are known. These plates are large, thin, and subovate in outline, the margin, except where the plate is attached, being compressed to a knife-like edge. The upper margin of the plate is regularly arcuate in outline, the lower margin being somewhat pointed. The dimensions of the holotype, which is a very perfect plate, are: greatest width vertically 28.2 mm., greatest width horizontally 24.3 mm., length of surface of attachment 14.2 mm., greatest thickness 3 mm. Remarks. — The "wing plates" of this species more nearly resemble those of P. depressus than any other of the described species, in both species the plates being thin and widely expanded. Those of P. menardensis are much the larger and have little or no constriction towards the area of attachment. These plates have been found to be abundant and highly characteristic of the same zone in the Menard limestone of the southern counties of Illinois to which Pentremites fohsi seems to be restricted. Associated with these forms there are numerous bryozoans and some brachiopods, and commonly, many detached, spinose, crown plates of a species of Hydreionocrinus. Neither this species nor the assemblage of which it is a member, has yet been observed in the Menard limestone of Eandolph County. Horizon and locality. — Menard limestone, Union, Johnson, Pope, and Hardin counties, Illinois. Batostomella nitidula Ulrich Plate XI, Fig. 11 1890. Batostomella nitidula Ulrich, Geol. Surv. 111., vol. 8, p. 436, pi. 75, figs. 3-3b. This bryozoan occurs in abundance in some localities in the lower part of the Clore formation. It is characterized by its small, delicate, sub- cylindrical branches, which are commonly much broken. In general the species of this genus require thin sections for examination under the microscope, for satisfactory determination, but experience has shown that the common form in the lower Clore is B. nitidula, and little hesitation may be felt in so identifying examples from this horizon. Horizon and locality. — Clore limestone, Randolph, Union, Johnson, Pope, and Hardin counties, Illinois. Prismopora serrulata Ulrich Plate XI, Figs. 5-8 1884. Prismopora serrulata Ulrich, Jour. Cinn. Soc. Nat. Hist., vol. 7, p. 41, pi. 2, figs. 2-2f. This bryozoan is one of the most easily recognized members of this class of fossils. The triangular cross-section of the branches is different from 374 GEOLOGY OF HARDIN COUNTY any other bryozoan found in the Chester faunas of Illinois, and the species is characterized by the regular scallops on the three thin margins of the branches. The species is the best index fossil of the Glen Dean limestone and of the equivalent horizon in Randolph County, but it is not wholly re- stricted to this zone. During the entire time that has been devoted to the study of the Chester of southern Illinois, one single example of the species has been observed in the Golconda limestone. At one locality a number of poorly preserved specimens of this or an allied species have been collected in the Paint Creek, in Pope County. Above the Glen Dean limestone the species has been observed in a number of localities in the Vienna limestone, a forma- tion which lies between the Glen Dean and Menard, separated from each by a sandstone, in Pope, Johnson, and perhaps also Union counties. Notwith- standing these other occurrences, the species remains one of the most reliable markers of the Glen Dean limestone, and at some localities in the formation there are veritable "Prismopora gardens" exhibited upon the exposed surfaces of the limestone, and the species may be found in practically every exposure of the formation having any extent, if careful search is made, and it not infrequently is found in very limited exposures. Horizon and locality. — Golconda limestone (very rare), Hardin County, Illinois. Glen Dean limestone, Union, Johnson, Pope, and Hardin counties, Illinois, and Breckenridge County, Kentucky. Upper Okaw limestone (Plum Creek beds), Randolph County, Illinois. Vienna limestone, Pope and John- son counties. Spirifer increbescens Hall Plate XI, Fig. 10 1858. Spirifer increbescens Hall, Geol. Surv. Iowa, vol. 1, pt. 2, p. 706, pi. 27, figs. 6a-i. 1914. Spirifer increbescens Weller, 111. State Geol. Surv., Monog. I, p. 343, pi. 46, figs. 1-12. The typical form of S. increbescens is the large, rather strongly convex shell that is commonly met with in the Menard and Clore limestones of the Chester group. A smaller shell is present in the Middle and Lower Chester faunas, but it is more delicate in appearance and is quite distinct from this larger form of the Upper Chester. In much of the past literature the name S. increbescens has been applied to all of these shells in the Chester faunas, but the smaller form should at least be distinguished as a distinct variety and perhaps as a species. Some of those with much-extended hinge-line may be referred to S. increbescens var. transversalis Hall, and others perhaps to S. pellaensis. There is a whole series of Spirifers of this general type, with representatives present in all of the faunas from the early Mississippian to the summit of the series, and even continuing into the Pennsylvanian, in which it is difficult to draw sharp lines of specific demarcation, but which do PALEONTOLOGY 375 have distinct stratigraphic values, and to one who is familiar with all of these the true S. increbescens always indicates Upper Chester. Horizon and locality. — Menard and Clore limestones, Randolph, Union, Johnson, Pope, and Hardin counties, Illinois. COMPOSITA SUBQUADRATA (Hall) Plate XI, Figs. 12-14 1858. Athyris subquadrata Hall, Geol. Iowa, vol. 1, pt. 2, p. 703, pi. 27, figs. 2a-d; p. 708, text fig. 118. 1914. Composita subquadrata Weller, 111. State Geol. Surv., Monog. I, p. 489, pi. 81, figs. 1-15. This species of Composita, which occurs abundantly in the Menard and Clore formations, differs from the common form in the Middle and Lower Chester, which has been referred to C. trinuclea, in its much larger size and its greater proportional breadth. The shells exhibit considerable variation in outline, but the more usual form is that shown in figures 12 and 13, which are distinctly subquadrate in outline, as the name suggests. The mistake has been made in the past by many authors, in referring all Chester Compositae to this species, but the name should be restricted to the large form of the Upper Chester faunas. Horizon and locality. — Menard and Clore limestones, Randolph, Union, Johnson, Pope, and Hardin counties, Illinois. SULCATOPINNA MISSOURIENSIS (Swallow) Plate XI, Figs. 1-2 1863. Pinna Missouriensis Swallow, Trans. St. Louis Acad. Sci., vol. 2, p. 97. 1894. Pinna missouriensis Keyes, Mo. Geol. Surv., vol. 5, p. 116. Description. — Shell elongate-cuneate in outline, gaping behind. The specimens commonly occur in an incomplete condition, but when complete they must have attained a length of at least 15 cm. at maturity, with a maxi- mum height of 4 to 4.5 cm. Beaks terminally situated., the anterior extremity acutely angular; dorsal margin straight, ventral margin straight anteriorly and meeting the dorsal margin at the beak in an angle of about 15 degrees; posteriorly the ventral margin curves dorsally in a gently convex curvature which passes without interruption into the posterior margin, the convexity of the curve increasing gradually as it approaches the dorsal margin which it meets in an acute angle. Towards the beak the convexity of each valve nearly semicircular so that the cross-section of the anterior portion of the shell is subcircular; posteriorly the convexity of the valves relatively a little flattened, the cross-section being subelliptica] but with the shorter ends of the ellipse angular, the ventral side more acutely so than the dorsal because of the slight flattening or in some cases a gentle concavity of the valves in that 376 GEOLOGY OF HARDIN COUNTY direction posteriorly. Along the dorsal margin, just below the hinge-line, a narrow, rounded groove is present, extending the entire length of the shell in examples from which the entire shell substance has been removed, this groove probably being due to the thickening of the inner surface of the valves along the hinge-line. The presence of these grooves in the internal casts, as the species is commonly preserved, gives the shell the appearance of being keeled along the dorsal margin. Surface of each valve marked by from 20 to 25 longitudinal corrugations, some of which may divide at intervals. Postero-ventrally the longitudinal corrugations become obsolete and the concentric growth lines become more conspicuous. These growth lines continue across the longitudinal corruga- tions, but are subordinate to them upon the body of the shell. The approximate dimensions of an incomplete example are: length 160 mm. or more, maximum height about 40 mm., thickness 20 mm. Remarks. — This species is a common one in the Menard limestone, and has not been collected in the Okaw of Golconda limestones, but it has been found rarely in the Vienna and Glen Dean limestones. It commonly occurs standing vertically in the limestone, beaks downward, the position in which it existed during life. The shell itself is very thin and delicate, without articulating processes, and could never have been preserved in so complete a condition as that in which it commonly does occur, with the two valves together, had it been disturbed from its original position. The specific determination of the shell has been attended with some difficulty. Pinna missouriensis was originally described by Swallow without illustrations, and was said to be "abundant in the Ste. Genevieve limestone of Missouri and Illinois." Swallow's definition fits the Menard limestone speci- mens very well, but the shell does not occur in the Ste. Genevieve limestone, nor has it been collected in Missouri, the youngest Chester formation in that state being the Okaw limestone. However, one of the localities cited by Shumard for the Ste. Genevieve limestone, in his original definition of the formation, was the Mississippi River bluff below the mouth of Marys River, in Randolph County, Illinois, where the only limestones exposed are the Menard and Clore formations of the Chester, and he may have had examples of the species from one or the other of these formations, labeled as from the Ste. Genevieve limestone, which might have been used by Swallow in his definition of the species. The original type of the species has been entirely lost sight of. It seems clear that Worthen identified this species, the only one known in the Chester, as Pinna missouriensis, from a statement in his de- scription of Pinna St. Ludovici 1 , which he compares with "P. Missouriensis of the Chester group." Because of the well developed longitudinal corrugations of the shell, this species has been transferred to Hyatt's genus Sulcatopinna, The species is 1 Geol. Surv. 111., vol. 7, p. 326. PALEONTOLOGY 377 very closely related to Pinna maxvillensis Whitfield, from the Maxville lime- stone of Ohio, and it is not unlikely that the two species are identical. Another allied form is Pinna arkansana Weller, from the Batesville sand- stone, but the Arkansas species apparently has much less convex valves than missouriensis. Horizon and locality. — Clore limestone, Kandolph County, Illinois. Menard limestone, Eandolph, Union, Johnson, Pope, and Hardin counties, Illinois. Vienna limestone (very rare), Pope County, Illinois. Glen Dean limestone (very rare), Johnson County, Illinois. Allorisma olavata McChesney Plate XI, Figs. 3-4 1860. Allorisma clavata McChesney, Desc. New Pal. Foss., p. 56. 1868. Allorisma clavata McChesney, Trans. Chicago Acad. Sci., vol. 1, p. 43, pi. 6, figs. 9a-b. This is the common species of Allorisma occurring in the Menard lime- stone. It is easily recognized by its subelliptical outline and well defined concentric corrugations. Horizon and locality. — Menard limestone, Eandolph, Union, and Pope counties, Illinois. PLATES AND EXPLANATIONS EXPLANATION OF PLATE IV Figs. 1-7. Pentbemites pbincetonensis Ulrich p. 314 1. St. Louis limestone, St. Louis, Missouri. 2. St. Louis limestone, above Little Rock, Ste. Genevieve County, Missouri. 3. Fredonia limestone, Rosiclare, Illinois. 4-5. Ste. Genevieve limestone, Fountain Creek, near Waterloo, Illinois. 6. Renault limestone, Fairview bluff, below Rosiclare, Illinois. 7. Renault limestone, Monroe County, Illinois. Figs. 8-12. Pentbemites pinguis Ulrich p. 317 8-9. Renault limestone, Monroe County, Illinois. 10-11. Shetlerville formation, Fairview bluff, Hardin County, Illinois. 12. Renault limestone, Fairview bluff, Hardin County, Illinois. Fig. 13. Pentbemites abctibbachiatus Ulrich Renault limestone, mouth of Good Hope shaft, Fairview mine, Rosiclare, Illinois. Figs. 14, 28-30. Pentbemites pulchellus Ulrich .p. 316 14. Fredonia limestone, Cedar Bluff, near Princeton, Kentucky. 28-30. Shetlerville formation, Fairview bluff, Hardin County, Illinois. Fig. 15. Pentbemites Shetlerville formation, Fairview bluff, Hardin County, Illinois. Figs. 16-20. Pentbemites buttsi Ulrich p. 323 16. Renault limestone, mouth of Good Hope shaft, Fairview mine, Rosiclare, Illinois. 17. Renault limestone, St. Clair County, Illinois. 18. Renault limestone, Monroe County, Illinois. 19-20. Renault limestone, near Anna, Illinois. Figs. 21-24. Pentbemites pybamidatus Ulrich p. 325 21. Paint Creek formation, Cache River bluff, Johnson County, Illinois. 22-23. Paint Creek formation, two miles south of Columbia, Illinois. 24. Paint Creek formation, Cache River bluff, Johnson County, Illinois. Fig. 25. Pentbemites nodosus Hambach p. 356 Near base of Lower Okaw limestone, three-fourths of a mile north of Mari- gold, Randolph County, Illinois. Figs. 26-27. Pentbemites symmetbicus Hall p. 324 Paint Creek formation, one and one-fourth miles northwest of Floraville, St. Clair County, Illinois. Figs. 31-35. Pentbemites godoni DeFrance p. 319 31-32. The holotype of the species, after Parkinson. 33-35. Three examples exhibiting some of the variations, Paint Creek forma- tion, one and one-fourth miles northwest of Floraville, St. Clair County, Illinois. 36. Renault limestone, mouth of Good Hope shaft, Fairview mine, Rosiclare, Illinois. Figs. 37-42. Pentremites platybasis n. sp p. 355 37. The holotype of the species, with very flat base. Near base of Lower Okaw limestone, three-fourths of a mile north of Marigold, Randolph County, Illinois. 38. Another specimen from the same locality with a less nearly flat base. 39-40. Lateral and basal views of a third specimen from the same locality, the basal view showing the callosity at the center. 41-42. Lower Golconda limestone, Hardin County, Illinois. Figs. 43-46. Pentremites brevis Ulrich p. 369 43-44. Lateral and basal views of a specimen from the Upper Okaw limestone, Plum Creek beds, at Menard, Randolph County, Illinois. 45-46. Similar views of another specimen from the same locality. The basal views show the central callosity. Fig. 45 shows an abnormal feature in the division of the base into four plates instead of three as usual, and the consequent four lobes of the callosity. Fig. 47. Pentremites godoni DeFrance p. 319 A much depressed variety from the Paint Creek formation, one and one- fourth miles northwest of Floraville, St. Clair County, Illinois. Illinois State Geological Survey Bulletin No. 41, Plate IV # ^ ) 9 fy ^ $ 3 3 $ 3 S" I 8 » 16 21 36 31 37 9 10 11 12 13 14 ^ 17 1 ^ 22 28 32 38 18 19 20 15 35 39 40 42 43 45 46 47 EXPLANATION OF PLATE V Figs. 1-4. Platycrinus peniciixus Meek and Worthen p. 333 1-2. Basal views of two specimens showing the three radiating ribs. 3-4. Views of two radial plates. Fredonia limestone, Cedar Bluff quarry, near Princeton, Kentucky. Figs. 5-12. DicHocEiNus girtyi n. sp p 334 5-7. Anterior, posterior, and ventral views of a very perfect calyx, natural size. 8. Anterior view of the same specimen enlarged to three and one-half diameters. 9. Anterior view of a second specimen. These two specimens are cotypes. Shetlerville formation, Fairview bluff, Hardin County, Illinois. 10. Lateral view of a crushed specimen. 11-12. Lateral and dorsal views of the basal cup of another specimen. Paint Creek formation, one and one-fourth miles northwest of Floraville, St. Clair County, Illinois. Figs. 13-14. Dizygocrinus superstes Ulrich p. 329 Lateral and dorsal views of a specimen nearly free from matrix. Fredonia oolite, Charles Stone Quarry, near Joppa Junction, Johnson County, Illinois. Figs. 15-17. Dizygocrinus persculptus Ulrich p 327 15. Dorsal view of a nearly complete cup. Fredonia limestone, Cedar Bluff quarry, near Princeton, Kentucky. 16-17. Lateral and dorsal views of a nearly complete calyx. Fredonia limestone, Rosiclare, Illinois. Figs. 18-23. Globocrinus unionensis (Worthen) p 331 18-19. Lateral and dorsal views of the holotype. (111. State Mus. Nat Hist No. 2581.) Near Anna, Union County, Illinois. 20-21. Lateral and dorsal views of a nearly complete calyx. Shetlerville formation, Fairview bluff, Hardin County, Illinois. 22-23. Lateral and dorsal views of a smaller specimen. Shetlerville formation, four miles west of Marion, Crittenden County, Ken- tucky. Figs. 24-26. Pterotocrinus serratus n. sp p 342 24-25. Lateral views of two "wing plates" showing the serrate distal border. Paint Creek formation, Cache River bluff, six miles south of Vienna, John- son County, Illinois. 26. Lateral view of a "wing plate," the holotype. Paint Creek formation, two miles south of Columbia, Monroe County, Illinois. Figs. 27-28. Pterotocrinus sp. Lateral views of two "wing plates." Paint Creek formation, Cache River bluff, six miles south of Vienna, Johnson County, Illinois. Figs. 29-30. Agassizocrinus dissimilis n. sp p 344 Posterior and basal views of the holotype. Paint Creek formation, one and one-fourth miles northwest of Floraville, St. Clair County, Illinois. Figs. 31-33. Amplexus geniculatus Worthen p 314 Views of three specimens. Shetlerville formation, Fairview bluff, Hardin County, Illinois. Illinois State Geological Survey Bulletin No. 41, Plate V S v 3 01 3 4 * 6 9 10 11 12 # 13 14 15 18 20 22 16 19 21 23 *t 25 V V v 29 30 17 26 * 27 * 28 EXPLANATION OF PLATE VI Figs. 1-13. Talarocrinus inflatus Ulrich p. 337 1-3. Posterior, ventral, and dorsal views of a nearly perfect calyx. Renault formation, Monroe County, Illinois, probably from the same locality as Ulrich's holotype. 4-5. Dorsal and lateral views of a very perfect dorsal cup. Renault forma- tion, two miles east of Waterloo, Monroe County, Illinois. 6-7. Anterior and ventral views of a practically perfect calyx. Renault forma- tion, Union County, Illinois. 8. Basal view of a dorsal cup. Renault formation, near Bixby, St. Clair County, Illinois. 9-10. Lateral and ventral views of a nearly complete calyx. Renault forma- tion, Hardin County, Illinois. 11-13. Posterior, basal, and ventral views of a nearly complete calyx. Re- nault formation, near Anna, Union County, Illinois. Figs. 14-16. Talarocrinus planus n. sp p. 342 Posterior, right lateral, and ventral views of the holotype. Paint Creek forma- tion, three-fourths of a mile southeast of Commons School, Randolph County, Illinois. Figs. 17-26. Talarocrinus trijugis Miller and Gurley p. 338 17-19. Posterior, anterior, and ventral views of the holotype. Renault lime- stone, near Rosiclare, Hardin County, Illinois. 20-21. Anterior and ventral views of a large and nearly perfect calyx. Renault limestone, Union County, Illinois. 22-24. Posterior, dorsal and ventral views of a very depressed but nearly complete and undistorted calyx. The basal view suggests T. sexlobatus, but the tegmen is very different. Renault limestone, near Anna, Union County, Illinois. 25. Basal view of a dorsal cup. Renault limestone, Ohio River bank just above the mouth of Grand Pierre Creek, Pope County, Illinois. 26. Basal view of a dorsal cup. Renault limestone, Cache River bluff, six miles south of Vienna, Johnson County, Illinois. The last two figures are indistinguishable from T. inflatus as illustrated in fig. 8. The only reason for identifying them as T. trijugis is that this species seems to be the more common one in southeastern Illinois, while T. inflatus is more common in the Mississippi Valley. Figs. 27-33. Talarocrinus ovatus Worthen p. 340 27-30. Posterior, anterior, dorsal, and ventral views of the holotype. Prob- ably from the Paint Creek formation, Monroe County, Illinois. 31-33. Posterior, ventral, and dorsal views of a nearly complete calyx. Paint Creek formation, five miles east of Waterloo, Monroe County, Illinois. Fig. 35. Talarocrinus cornigerus (Shumard) p. 340 Lateral view of a specimen with stronger ventral spines than is usual in the species. Renault limestone, near Modoc, Randolph County, Illinois. Illinois State Geological Survey £ 10 9 1 1 3 i H & 12 13 14- 15 16 17 18 19 20 21 22 23 24 25 <^ 27 *V 28 31 32 Q 29 30 33 34 EXPLANATION OF PLATE VII Figs. 1-4. Septopora similis n. sp p. 348 1. A specimen showing the reverse surface, natural size. 2. A specimen showing the obverse surface, natural size. 3. The reverse surface enlarged three and one-half diameters. 4. The obverse surface enlarged three and one-half diameters. St. Louis limestone, above Little Rock, Ste. Genevieve County, Missouri. Figs. 5-6. Lyropora quincuncialis Ulrich p. 347 5. A fragment of the median portion of the thickened base, with a bit of the fenestrated portion attached. Shetlerville formation, one mile south of Eichorn, Hardin County, Illinois. 6. A specimen showing the thickened base with a considerable part of the fenestrated portion. Upper part of Renault, "Upper Ohara," Cedar Bluff quarry, near Princeton, Kentucky. Figs. 7-24. Cystodictya labiosa n. sp p. 350 7-18. Views of a number of specimens exhibiting the various styles of branching. Cotypes. Paint Creek formation, one and one-fourth miles northwest of Floraville, St. Clair County, Illinois. 19. The specimen shown in fig. 8 enlarged three and one-half diameters. 20-22. Three specimens similar to those from the Paint Creek formation. Shetlerville formation, four miles west of Marion, Crittenden County, Kentucky. 23-24. Two specimens not free from the matrix. Paint Creek formation, Cache River bluffs, six miles south of Vienna, Johnson County, Illinois. Fig. 25. Glyptopora punctipora Ulrich p. 351 Shetlerville formation, four miles west of Marion, Crittenden County, Kentucky. Fig. 26. Archimedes cf. swallgvanus Hall p. 345 Paint Creek formation, Cache River bluffs, six miles south of Vienna, John- son County, Illinois. Fif. 27. Archimedes compactus Ulrich p. 346 Paint Creek formation, Cache River bluff, six miles south of Vienna, Johnson County, Illinois. Fig. 28. Archimedes proutanus Ulrich p. 346 Paint Creek formation, Cache River bluffs, six miles south of Vienna, Johnson County, Illinois. Fig. 29. Archimedes distans Ulrich p. 346 Paint Creek formation, Cache River bluffs, six miles south of Vienna, Johnson County, Illinois. Fig. 30. Archimedes meekanus Hall p. 346 Paint Creek formation, Cache River bluffs, six miles south of Vienna, Johnson County, Illinois. Figs. 31-o2. Archimedes communis Ulrich p. 347 Paint Creek formation, Cache River bluffs, six miles south of Vienna, Johnson County, Illinois. Illinois State Geological Survey Bulletin No. 41, Plate VII tHVTlff ■i 8 9 10 H 12 13 14 16 IT 1fl 17 18 24 nif ^^O; 20 21 22 19 25 I I 26 27 28 29 30 32 EXPLANATION OF PLATE VIII Figs. 1-4. Pugxoides ottumwa (White) p. 351 1-2. Brachial and pedicle views of one specimen. Ste. Genevieve limestone, near mouth of River aux Vases, Ste. Genevieve County, Missouri. 3-4. Similar views of another specimen, Fredonia limestone, Rosiclare, Hardin County, Illinois. Figs. 5-6. Rhipidomella dubia (Hall) p. 351 Brachial and pedicle views of two specimens. Fredonia limestone, Rosiclare, Illinois. This species is a common one in the Mississippian formations from the Keokuk to the Ste. Genevieve, but neither the species nor the genus has been observed in any of the Chester formations of Illinois. Figs. 7-14. Spiriferixa subspixosa n. sp p. 352 Pedicle and brachial views of four different specimens exhibiting range in size. 7-10. Shetlerville formation, four miles west of Marion, Crittenden County, Kentucky. 11-14. Shetlerville formation, Fairview bluff, Hardin County, Illinois. Figs. 15-16. Spiriferixa spixosa (Norwood and Pratten) p. 353 15. Brachial view, Paint Creek formation, one and one-fourth miles north- west of Floraville, St. Clair County, Illinois. 16. Pedicle view, Paint Creek formation, Cache River bluffs, six miles south of Vienna, Johnson County, Illinois. Figs. 17-18. Spiriferixa transversa (McChesney ) p. 353 Brachial and pedicle views of two individuals. Shetlerville formation, Lemon quarry, three miles west of Marion, Crittenden County, Kentucky. Figs. 19-20. Composita trixuclea (Hall) p. 354 Brachial and pedicle views of an average specimen. Shetlerville formation, Fairview bluff, Hardin County, Illinois. Figs. 21-24. Cliothyridixa sublamellosa (Hall) p. 354 Two pedicle and two brachial views of four different individuals. Shetlerville formation, Fairview bluff, Hardin County, Illinois. Figs. 25-28. Mesoblastus glaber (Meek and Worthen) p. 327 Lateral views of four different individuals, magnified three and one-half diameters. Shetlerville formation, Fairview bluff, Hardin County, Illinois. Figs. 29-34. Mesoblastus incurvatus n. sp p. 358 29-31. Lateral views of three specimens. 32-34. Lateral views of the same specimens magnified three and one-half diameters. Golconda limestone, NE. % sec. 28, T. 12 S., R. 7 E., near Flat Rock ford, Pope County, Illinois. A Fig. 35. Pachylocrixus cachexsis n. sp p. 343 Paint Creek formation, Cache River bluffs, six miles south of Vienna, Johnson County, Illinois. This species is also present in the Paint Creek formation of Monroe County, Illinois. Figs. 36-39. Pterotocrinus capitalis Lyon p 360 36. Lateral view of a cast of the holotype. Golconda limestone, Crittenden County, Kentucky. 37-38. Lateral views of two "wing plates." Lower Golconda limestone, three and one-half miles southeast of Vienna, Johnson County, Illinois. 39. Lateral view of a "wing plate." Lower Golconda limestone, Ohio River bluffs, above Golconda, Pope County, Illinois. Illinois State Geological Survey Bulletin No. 41, Plate VIII & 8 19 21 25 32 m 2 , ^ 3 4 EXPLANATION OF PLATE IX Figs. 1-6. Bellerophon chesterexsis n. sp p. 365 1-3. Three views of a nearly perfect specimen. The holotype. Basal Okaw limestone, Camp Creek, Randolph County, Illinois. 4-6. Three similar views of another specimen. Basal Golconda limestone, three and one-half miles southeast of Vienna, Johnson County, Illinois. Figs. 7-10. Euphemus randolfhensis n. sp p. 365 7. View of the holotype. Basal Okaw limestone, Camp Creek, Randolph County, Illinois. 8. Similar view of another specimen. Basal Golconda limestone, three and one-half miles southeast of Vienna, Johnson County, Illinois. 9-10. Two views of a third specimen, from the same locality as the last. Figs. 11-14. Nucula PLATYNOTUS n. sp p. 361 11-12. Lateral and dorsal views of a practically perfect shell. The holotype. Basal Golconda limestone, three and one-half miles southeast of Vienna, Johnson County, Illinois. 13-14. Lateral and dorsal views of a nearly complete shell. Basal Okaw lime- stone, Camp Creek, Randolph County. Illinois. Figs. 15-20. Bucanopsis ornatus n. sp p. 366 15-16. Two views of an incomplete specimen. Basal Okaw limestone, Camp Creek, Randolph County, Illinois. 17. Dorsal view of the holotype. Basal Golconda limestone, three and one- half miles southeast of Vienna, Johnson County, Illinois. 18. Dorsal view of a somewhat larger specimen from the same locality. 19-20. Lateral and dorsal views of another specimen, from the same locality as the last two. Figs. 21-23. Conocardium chesterensis n. sp p. 364 21. Lateral view of the holotype. Basal Golconda limestone, three and one- half miles southeast of Vienna, Johnson County, Illinois. 22-23. Lateral views of two valves, basal Okaw limestone, Camp Creek, Randolph County, Illinois. Figs. 24-30. Nucula randolphensis n. sp p. 362 24-25. Lateral and dorsal views of the holotype. Basal Okaw limestone, Camp Creek, Randolph County, Illinois. 26-27. Lateral and dorsal views of another nearly perfect specimen. Basal Golconda limestone, three and one-half miles southeast of Vienna, John- son County, Illinois. 28-30. Lateral views of three specimens, from the same locality as the last. Figs. 31-34. Leda chesterensis n. sp p. 363 31-32. Lateral views of two valves. Basal Golconda limestone, three and one-half miles southeast of Vienna, Johnson County, Illinois. 33-34. Lateral views of two valves. Basal Okaw limestone, Camp Creek, Randolph County, Illinois. Figs. 35-36. Camarophoria explanata (McChesney) p. 360 Pedicle and brachial views of two specimens. Golconda limestone, Hardin County, Illinois. Pigs. 37-38. Composita trinuclea (Hall) p. 361 Brachial and pedicle views of two specimens. Golconda limestone, Hardin County, Illinois. Pigs. 39-41. Gliothyridina sublamellosa (Hall) p. 360 39-40. Brachial and pedicle views of two specimens. Golconda limestone. Hardin County, Illinois. 41. View of a pedicle valve upon which the concentric, lamellose expansions of the shell are well shown. Golconda limestone, Hardin County, Illinois. Illinois State Geological Survey 1 1 20 '-J| 23 Bulletin No. 41, Plate IX a a 8 vm mJ 4 2 9 10 6 * O * a 12 13 14 15 16 19 18 17 JP <% J O ^ > 21 22 24 25 ^^ 27 -^^^9 28 29 30 34 > 31 32 33 • 35 36 > 37 38 39 40 41 EXPLANATION OF PLATE X Figs. 1-3. Pentremites obesus Lyon p. 355 1-2. Lateral and ventral views of a nearly complete specimen. Lower Gol conda limestone, quarry of Golconda Portland Cement Company, Pope County, Illinois. 3. Lateral view of a large specimen, incomplete at the base. Lower Golconda limestone, Ohio River bank above Golconda, Pope County, Illinois. Fig. 4. Pentremites fohsi Ulrich p. 370 Lateral view of a nearly complete specimen. Lower part of Menard limestone, three and one-half miles northwest of Golconda, Pope County, Illinois. Figs. 5-7. Pentremites okawensis n. sp p. 357 5. Lateral view of a nearly complete specimen. Lower Okaw limestone, three- fourths of a mile north of Marigold, Randolph County, Illinois. 6. Lateral view of another nearly complete specimen. Golconda limestone, Hardin County, Illinois. 7. Lateral view of a somewhat weathered specimen. Glen Dean limestone, Hardin County, Illinois. Fig. 8. Pentremites spicatus Ulrich p. 368 Lateral view of a nearly complete specimen. Upper Okaw, Plum Creek beds, Menard, Randolph County, Illinois. Figs. 9-10. Pentremites gemmiformis Hambach p. 326 Lateral views of two nearly perfect specimens. Paint Creek formation, three miles east of Prairie du Rocher, Randolph County, Illinois. Figs. 11-13. Pentremites cherokeeus Hall p. 371 Lateral views of three specimens. Fig. 13 shows the very prominent extended points of the deltoid plates, which are commonly destroyed. Upper part of Menard limestone, Mississippi River bluff, five miles below Chester, Randolph County, Illinois. Fig. 14. Pterotocrinus bifurcatus Wetherby p. 372 Ventral view of the holotype, showing the five "wing plates" in place. Glen Dean limestone, Sloans Valley, Pulaski County, Kentucky. Illinois State Geological Survey Bulletin No. 41, Plate X 11 12 13 9 EXPLANATION OF PLATE XI Figs. 1-2. Sulcatopinna missouriensis (Swallow) p. 375 Lateral views of two specimens, neither of which is complete. Menard lime- stone, tributary of Gravel Creek, four miles northeast of Chester, Ran- dolph County, Illinois. Figs. 3-4. Allorisma clavata McChesney p. 377 Lateral views of two specimens. Menard limestone, tributary of Gravel Creek, four miles northwest of Chester, Randolph County, Illinois. Figs. 5-8. Prismopora serrulata Ulrich p. 373 5. View of a well preserved specimen, showing the branching of the colony. Upper Okaw limestone, Plum Creek beds, one mile east of Preston, Ran- dolph County, Illinois. 6-8. Views of three fragments of branches. Glen Dean limestone three miles east of Cloverport, Breckenridge County, Kentucky. Fig. 9. Pterotocrinus menardensis n. sp p. 373 View of a very perfect "wing plate," the holotype. Menard limestone, east of Cobden, Union County, Illinois. Fig. 10. Spirifer increbescens Hall p. 374 Pedicle view of a large specimen. Clore limestone, four miles northwest of Golconda, Pope County, Illinois. Fig. 11. Batostomelea nitidula Ulrich p. 373 View of a portion of a slab containing many examples of this species. Lower Clore limestone, Union County, Illinois. Figs. 12-14. Composita subquadrata (Hall) p. 375 12-13. Pedicle and brachial views of a nearly perfect specimen. Clore lime- stone, four miles northwest of Golconda, Pope County, Illinois. 14. Pedicle view of a larger specimen, from the same locality. Illinois State Geological Survey Bulletin No. 41, Plate XI J I 1 1 INDEX Italics refer to synonyms. Roman numerals refer to plates on which illustrations are given. Bold face numerals in parenthesis indicate special mention, in most cases careful description of the forms, and their geographic and geologic occurrence. PAGE Acanthotelson eveni 89 Acervularia davidsoni E. and H 85 "Acid spar ", definition of 285 Aclisina pvgmaea (Weller) 119 Agassizocrinus 172, 183, 184, 186, 207, 215 A. dissimilis n. sp (344), V Agriculture, importance of 19 relation of, to topography 43-44 Alethopteris decurrens (Artis) 227 A. grandifolia Newberry 227 A. lonchitica (Sch.) (Ohio form) 227 Allorisma 123, 124, 138, 222 A. clavatum McChesney 208, (377), XI Alluvial deposits 232-233 Amplexus geniculatus Worthen 137, 139, 140, 141, 142, 145, 156, (3 1 4), V Amphigenia curta (Meek and Worthen) 86 Anisotrypa solida Ulrich 183 A. ?sp 137 Annex mine, description of 297-298 Annex-Extension vein, description of 250 Archaeocidaris 117, 118, 207, 208, 215 Archimedes 138, 150, 158, 164, 171, 172, 174, 178, 185, 186, 214, 215, (344) A. cf. swallovanus Hall 170, (345), VII A. communis Ulrich 171, 173, 182, 196, (347), VII A. compactus Ulrich 171, 172, 173, 183, 185, (346), VII A. confertus Ulrich 182 A. distans Ulrich 171, 173, 195, 196, 207, (346), VII A. lativolvis Ulrich 182, 186 A. laxus 197, 198 A. meekanus Hall 170, 173, 207, (346), VII A. proutanus Ulrich 170, 173, 183, (346), VII A. proutanus Ulrich, var 207 A. swallovanus Hall 182, 183, 185, 186, 195, 207, (345) A. terebriformis Ulrich 182, 183, 196, 207 "Archimedes limestone ", definition of 122-123 Ascodictyon sp 137 Asterocrinus capitalis Lyon 360 Athyris densa Hall 99, 100 A . sublamellosa Hall 354 A . subquadrata Hall 375 Atrypa spinosa Hall ? : 86 Aviculopectin sp 95, 171, 185, 222 Bain, H. Foster, work of 276-277 Bald Hill anticline, relation of to oil and gas possibilities 308-309 Baldwin mine 302 Barite, occurrence of 254 Basal conglomerate, definition of 27 Batostomella 148, 171, 173, 196, 215 B. nitidula Ulrich 214, 215, (373), XI B. ? sp 137, 183, 207 403 404 INDEX PAGE Battery Rock coal 226-227 Battery Rock Township, limestone sinks in 41 topographic age of hills in 48-49 50 Bedded fluorspar deposits, description of 254-256 methods used in mining of 283 Bell coal 228 Bellerophon 89, 116, 117, 119, 138, 215, 222 B. chesterensis n. sp 185, (365), IX B. monroensis Weller ? 119 Bethel sandstone, description of 159-164 Big Creek, description of 21 Glen Dean fossils collected along 195 outcrop of Warsaw formation on ■ 98 relation of to faults 64 Big Creek fault, description of 71-72 Big Joe mine 302 Birds Point, detailed section at 163 1 Blanket" formations, see Bedded fluorspar deposits. "Blende", see Sphalerite. Blue Diggings fault, description of 71 Blue Diggings mine, description of 298 Blue Diggings vein, description of 250 inclination of 64 Brachythyris subcardiiformis (Hall)? 99, 100 B. suborbicularis (Hall)? 99 Bryozoans (several undet. spec.) 185, 207 Bucanopsis ornatus n. sp 185, (366), IX B. textilis (Hall) 119, 120 Bulimorpha sp 117, 119 B. whitfieldi Weller 117, 119 Buzzards Point plain 47-48 Cache River bluffs, Renault-Bethel contact shown in 146 "Calc", see Calcite. Calcite, deposition of 257-258 description and occurrence of 253 Camarophoria explanata (McChesney) . 183, 184, 185, 186, 195, 196, 207, 214,215, 222, (360), IX Camarotoechia mutata (Hall)? 99 C. tethys (Billings) ? 86 Caninia n. sp 170, 173 Cardiopteris polymorpha Gopp 201 Cardiocarpon sp. ? 227 Carmens store, fault near 71 Carrsville, Daisy fault near 71 Caseyville formation, description of 225-228 Cave in Rock, limestone sinks near 41 Renault-Bethel contact near 145 St. Louis limestone near 102, 105 Cave in Rock mine, description of 301 Cave in Rock Township, topographic age of hills in 48-49, 50, 51 Cement, materials for 305 Cenozoic, definition of 27 ' 'Center slips ", relation of to mineralization 258 Ceramics industry, importance of fluorspar in 288 Cerussite, occurrence of 255 relation of to mineralization 259 Chalcopyrite, deposition of 258 description of 254 Chattanooga shale, correlation of 89 description of 87-90 Cheilanthites cf . macilenta 227 Chemical industries, importance of fluorspar in 288 Chester group, description of 121-131 INDEX 405 PAGE Chonetes cf. chesterensis Weller 171, 172, 174 C. coronatus (Conrad) 86 C. mesolobus 229 Cladodus sp 185 Clement mine, description of 298 Cliffs, distribution of . 46-47 Cliothyridina hirsuta (Hall) ? 99 C. sublamellosa (Hall) 138, 139, 141, 148, 158, 171, 173, 182, 183, 184, 185, 195, 196, 208, 222, (354,360-361), VIII. IX C. sublamellosa (Hall) ? .215 C. aff. sublamellosa (Hall) 116, 117, 118 Clore limestone, description of 212-216 Coeloconus sp 185 Composita subquadrata (Hall) 208, 209, 214, 215, (375), XI C. trinuclea (Hall) 116, 117, 118, 119, 138, 139, 141, 148, 158, 171, 173, 182, 183, 184, 185, 195, 196, 208, 222, (354, 361, 375), VIII, IX Conocardium chesterensis n. sp 185, (364-365), IX Conularia sp 138, 196, 208 Cornulites sp 137, 183 Correlation methods 150-151, 155 Crania chesterensis Miller and Gurley 138, 141, 183, 207, 215 Crinoids, plates and terns 137, 185, 207 Cross-bedding, significance of 24 Curlew coal 228-229 Cyathophyllum rugosum Hall 85 Cycle of erosion, definition of 35 Cycloceras cf. sequoyahensis Snider 185 C. sp 185 Cypress sandstone, description of 175-178 Cypricardinia indianensis Hall 119 Cypricardella oblonga Hall 119 C. sp 117, 118 Cyrtina ?? sp 138 Cystelasma quinqueseptata 116 Cystodictya labiosa n. sp. (Ulrich) . . 138, 139, 140, 148, 157, 170, 171, 172, 173, 174, (350), VII Cystiphyllum lineata Ulrich 99 C. pustulosa Ulrich 99 C. sp 85 Daisy fault, relation of to fluorspar deposits 71 Daisy mine, description of 295 Decadocrinus columbiensis (Worthen) 170, 173 Degonia sandstone, description of 216, 218 Devonian chert, description of 85 Devonian limestone, association of fluorspar with 87 description of 84-87 possibility of oil in 309 Diaphragmus 164 D. elegans (Norwood and Pratten) 138, 139, 141, 148, 158, 170, 173, 182, 183, 185, 195, 196, 207, 215, 222 D. n. sp 117, 118 D. ?sp 214 Dichocrinus girtyi n. sp 137, (334), V Dichotrypa sp 99 Dielasma formosa (Hall) 116, 118 D. illinoisensis Weller 138, 139. 141, 171, 173 D. shumardanum (Miller) '. 196, 207, 214 D. sp 117, 1 18, 182, 183, 185 Dikes, definition of 29 Dip, definition of 31 Discitoceras sp 185 Dizygocrinus persculptus Ulrich 117, (327), V D. sp. undet 116 D. superstes Ulrich (329), V Dolatocrinus sp 85 406 I NDEX PAGE Downeys Bluff sill and dike, description of 239-241 Drainage, types of 21 Eatonia sp 86 Economic geology 247-310 Elizabethtown, Fredonia limestone outcrop near 110-111, 112 limestone sinks near 41 population of 20 topographic age of hills near 51 Elizabethtown plain 50-51 Emmons, S. F., work of 275 Empire mine, description of 301-302 Encrina godoni Def ranee 31 9 Encrinites florealis von Schlotheim 31 9 English fluorspar deposits, competition with 290 Eotrochus n. sp 117, 1 19 Epoch, definition of 27 Era, definition of 27 Eridopora punctifera Ulrich 137, 139, 140, 182, 183, 196, 207 Eumetria costata (Hall) 138, 141, 207, 208, 214, 215, 222 E. vera (Hall) 138, 139, 141, 171, 173, 183, 185, 195, 196 E. verneuiliana (Hall) 99, 117, 118 Euomphalus planidorsatus Meek and Worthen 183 E. sp .215 Eupachycrinus maniformis (Yandell and Shumard) 170, 173 E. sp 137, 195 Euphemus randolphensis n. sp 185, (365-366), IX Eureka prospect, association of with Peters Creek faults 70 Extension mine, description of 297-298 Extension vein, inclination of 64 Fairview, igneous rock near 239 Fairview bluff, Renault fossils from 148 Fairview Mill, description of 285 Fairview mine, occurrence of stibnite in 254 description of 295-298 Farming, relation of to topography 43-44 Faulted zones, description of 55, 58, 60-64 Faulting, association of fluorspar deposits with 55, 275 Faulting, criteria for recognition o 58-59 Faulting, definition of 30 kinds of 30-31 magnitude of 56-57 Fault planes, inclination of 64 Fenestella (Archimedes) meekana Hall 3!fi F. (Archimedes) swallovana Hall 31+5 F. cestriensis Ulrich 137, 139, 140, 148, 157, 171, 173, 183 F. elevatipora Ulrich 137, 140, 183 F. elevatipora Ulrich ? 215 F. (Lyropora) quincuncialis Hall SJfl F. serratula Ulrich 117, 118, 138, 139, 140, 148, 157, 171, 173, 183, 196, 207 F. serratula Ulrich ? 214, 215 F. sp 171, 182, 183, 185, 195, 196, 207, 215 F. tenax Ulrich 99, 117, 118, 138, 139, 140, 148, 157, 171, 173, 183 F. tenax Ulrich ? 214 Fertilizer, materials for 305 Fish teeth (several species) 185 Fistulipora excelens Ulrich 137, 140, 171, 173, 183, 207 F. sp 117, 118, 195, 196 Fluorspar and fluorspar deposits, commercial grades of 289 deposition of 258 description of 252 discovery of 247 INDEX 407 PAGE Fluorspar and fluorspar deposits, economic geology of 247-304 geologic history of 281-282 igneous rocks and faulting in association with 55, 66 milling of 283-289 mining of 282-283 occurrence of 248-257 origin of 275-281 production of 289-290 uses of 287-289 Fohs, F. J., work of 277 Formations, definition of 25 significance of 25-26, 27 Fossils, significance of 26 Fossil lists, Clore limestone 214, 215 Devonian limestone 85-86 Fredonia limestone member 116-118 Glen Dean limestone 195, 196 Golconda formation 182, 183, 185 Kinkaid limestone 222 Menard limestone 207, 208 Paint Creek formation 170-171 Renault formation 148, 149 Ste. Genevieve limestone 116, 117, 118 Shetlerville ormation 137-141 Tar Springs sandstone 201 Fredonia limestone member, lithologic character of 110, 111 relation of sink holes to 41 thickness of 113-114 Fruit raising, desirability of 45-46 Galena, deposition of 258 description of 253 separation of 283 Gastropods 171 Geography 39-52 Genus Globocrinus (330) Genus Talarocrinus (336) Geologic map, description and explanation of 36-37 use of in prospecting 291 Geologic time, definition of 27 ' divisions of 28 Girtyella brevilobata (Swallow) 138, 141, 182, 183, 196 G. indianensis (Girty) 117, 118, 138, 139, 141, 171, 173, 215 G. sp 222 Glacial drift, absence of 44 Glen Dean limestone, description of 191-198 Globocrinus n. genus (330) G. unionensis (Worthen) 137, 140, 142, (33 1 ), V Glyptopora cf . punctipora Ulrich 171 G. punctipora Ulrich. . 138, 139, 141, 170, 172, 173, 174, (35 1 ), VII Golconda and Lower Okaw species, description of 355-367 Golconda dike, description of 241 Golconda, Clore fossils near 215 Paint Creek formation near 168, 171 Glen Dean fossils collected near 196 igneous rock near 241 Menard fossils near 207-208 sections near 168, 205-206 Golconda formation, description of 178-187 Goodhope mine, description of 295-297 Goodhope shaft, Renault fossils from 148, 149 Goodhope vein, description of 250 inclination of - 64 408 INDEX PAGE Goose Creek, fluorspar in Devonian strata near 87 outcrop of Chattanooga shale on 88 of Osage formation on 94 Granatocrinus glaber Meek and Worthen 327 "Gravel" deposits of fluorspar, see Superficial fluorspar deposits. "Gravel spar", definition of 283 Griffithides sp 95 Groups, definition of 27 Hamp mines, description of 302, 303 Hardinsburg sandstone, description of 187-190 Harris Creek, relation of to limestone sinks 41 Harts Store, coal near 228 Hederella sp 137 Heliphyllum sp 85 Hemitrypa proutana Ulrich 99 Herrin fault, description of 66 Hexacrinus leai (Lyon) 85, 86 Hicks, fluorspar in Devonian strata near 87 outcrop of Devonian strata near 84 topographic age of hills near 48-49 Hicks Branch, outcrop of Warsaw formation on 98 Hicks Dome, Chattanooga shale on 88 description of 65 distribution of formations around 81 origin of 29-30 outcrop of Osage formation on 93 of Warsaw formation on 97 relation of to oil and gas possibilities 307-308 High Knob, altitude of . . : 20 topographic age of 47 Hicks mine 302 Hillside mine, description of 298 History of mining development 247-248 Hogthief Creek, relation of to faults 64 Hogthief Creek fault, description of 68 Holopea sp 185 H. ? sp 138 Horton Hill anticline, relation of to oil and gas possibilities 308-309 Hosick Creek, relation of to limestone sinks 41 Hutchinson mine 302 Hydreionocrinus sp 170, 173, 207, 208 Hydrocarbon, occurrence of with fluorspar 254 Igneous rocks 23, 29, 237-244 association of fluorspar deposits with 55, 275 Illinois Furnace, faulting nea 71-73 iron deposits near 304 St. Louis limestone near 105 Indian Point, detailed section at 162 Paint Creek fossils from 170-171 Iron ore deposits 304-305 "Jack", see Sphalerite. Jacks Point quarry, description of 306 Jenkins Point, pre-Denonian pebbles found near 79 Joplin jigs, use of 283 Karbers Ridge, section of Kinkaid limestone near 221 Karbers Ridge plain 48-49 Karst topography, definition of 41 Kentucky, faulting in 55 Kinkaid limestone, description of 218-222 INDEX 409 PAGE Kindle, E. M., work of on Devonian 83 Knorria 125 Laevidentalium venustum (Meek and Worthen) 117, 119 Lamprophyre, description of 237 Lead Hill, bedded fluorspar deposit at 254 Lead Hill mines, description of 300-301 Lead Hill prospect, association of with Peters Creek faults 70 Leda chesterensis n. sp (363-364), IX L. curta Meek and Worthen 119 L. sp 117, 185 Lee fault, description of 66-67 Lee mine, description of 303 Lepetopsis levettei White 117, 118 L. sp 185 Lepidodendron 125, 177, 201 L. cf . modulatum 212 Leptaena 89 L. rhomboidalis (Wilck) 86 Lime, materials for 305 Limestone, economic geology of 305-306 Limestone, origin of '. 25 solution of , 33 Limestone sinks 40-42 Lioclema sp 171, 173, 215 L. ? araneum Ulrich 183, 215 Lithostrotion 97, 105 L. canadensis 105 L. harmodites 116 L. proliferum 105 Loess, definition of 45 Loess deposits 233-234 Lower Chester formations, correlation of 132 Lower Chester and related species, description of 314 Lower Mississippian strata, description of 91-120 "Lower Ohara" limestone member, lithologic character of 112 thickness of 114 Lower Okaw species, description of 355, 367 Lyropora 149, 150, 158, 172, 194 L. quincuncialis Hall (347) L. ranosculum 149, 157, 170, 173 L. sp 99, 148, 207 L. subquadrata 149, 157 McFarland plain 49-50 McFarland Township, limestone sinks in 41 topographic age of hills in 48-49, 50, 51 Malachite, occurrence of 254 Marginif era muricata 229 Martha Furnace, iron deposits near 304 Martin mine, description of 300 Martin prospect, association of with Peters Creek faults 70 Martinia contracta Meek and Worthen 183, 185, 196, 197, 222 M. sp 182 M. sulcata Weller 183 Meekopora eximia Ulrich 148, 157, 196 M . sp 207 Megalopteris sp. ? 227 Melcher Hills, topographic age of 47, 49 Members, definition of 27 Menard limestone, description of 202-209 Meramec group, definition of 96-97 Meristella haskinsi Hall ? 86 410 INDEX PAGE Mesoblastus glaber (Meek and Worthen) 137, 140, 148 157 (327) VIII M . glaber f Ulrich '....' '. . 327 M. incurvatus n. sp (358-359), VIII Mesozoic, definition of ' 27 Metallurgical industries, importance of fluorspar in 288 Metasomatic relations of minerals 259 Mica-peridotite, description of 237 Michelinia 85 M. princetonensis 115 116 M. stylopora Eaton ? 86 M. subramosa 116 Middle Chester formations, correlation of 132 Miller mine, description of 300 Milling methods 283-289 Mineralization along fault planes 60 Minerals, metasomatic relations of 259 paragenesis of 257-274 replacements of 259-260 of the bedding deposits • 255 of the vein deposits 252-254 Mines, descriptions of 293-304 Mining methods 282-283 Mississippian chert, description of 85 Mississippian System, definition and nomenclature of 91 Mix dike, description of 241 Monroe County, early subdivisions of the Chester in 127 Monroe Township, topographic age of hills in 50 Mourlonia sp 185 Myalina sp 116, 117, 118, 138, 185, 208 Naticopsis sp 116, 117, 118, 119, 222 Nautilus sp 185 N. ? sp 183 Neuropteris cf . obliqua 227 New Madrid earthquake, relation of to Hardin County faulting 55 Nuclcocrinus verneuili 85, 86 Nucula illinoiensis Worthen 119 N. platynotusn. sp 185, (361), IX N. randolphensis n. sp 185, (362-363), IX N. shumardana Hall 117, 118 N. sp 185 Ohio River flood plain 19, 51-52 Oil and gas possibilities 307-310 Onondaga limestone, possibility of oil in 309 Optical fluorite 288 Orbiculoidea sp ' '. 182, 196 Orrs Landing dike, description of 237-239 Orthis dubia Hall 351 Orthoceras cf . okawense Meek and Worthen 185 O. sp 116, 117, 118, 119,138,222 Orthonychia 138, 148, 171 O. acutirostra (Hall) 119, 120 O. chesterensis Meek and Worthen 183 O. cf. chesterensis 173 Orthotetes kaskaskicnsis (McChesney).. .117, 118, 138, 148, 170, 182, 183, 185, 196, 207, 215, 222 Osage chert, value of for road material 306 Osage formation, description of 92-96 Ozarks, extent of in Illinois 19 Pachylocrinus cachensis n. sp 170, 173, (343), VIII P. spinuliferus (Worthen 170, 173 Paint Creek fauna, geologic range of 173 Paint Creek formation, description of 164-175 INDEX 411 PAGE Paleontology 85-87, 89, 94-95, 99, 104-106, 115-120, 136-142, 147-150, 164, 170-171, 177- 178, 182-186, 190, 195-197, 201, 207-209, 211-212, 214-215, 217, 221-222, 227-228, 313-401 Paleozoic, definition of 27 Palestine sandstone, description of 209-212 Paragenesis of minerals 257-274 Parallelodon sp 185 Pell fault, description of 74 Pell mine, description of 200 Pennsylvanian System, definition and nomenclature of 91 description of 223-229 Pentremites 116, 150, 151, 152, 153, 156, 172, 208 P. arctibrachiatus Ulrich 148, IV P. brevis Ulrich 195, (369-370), IV P. buttsi Ulrich 148, 152, 157, (323), IV P. canalis Ulrich ? 195 P. cervinus 149 P. cherokeeus Hall, description and location of (371 -372), X P. conoideus Hall 99, 100 P.- florealis (part) Ulrich 314, 816, 817 P. fohsi Ulrich 207, 208, 209, (370-371), X P. gemmiformis Hambach (326), X P. godoni DeFrance 148, 149, 152, 157, 170, 173, (3 1 9), IV P. lyoni Ulrich 183 P. nodosus Hambach (356), IV P. obesus Lyon 183, 184, 186, 187, 208, (355), X P. okawensis n. sp 195, (357), X P. pinguis Ulrich .137, 140, 148, 152, 157, (317), IV P. planus Ulrich 152, 186, 319 P. platybasis n. sp 182, 183, 186, (355), IV P. princetonensis Ulrich 117, 118, 137, 139, 140, 148, 152, 157, (314), IV P. pulchellus Ulrich 137, 140, (316), IV P. pyramidatus Ulrich 170, 173, 194, (325), IV P. pyriformis 149 P. sp 117, 183, 185, 196, 215 P. spicatus Ulrich 193, 196, 197, (368), X P. symmetricus Hall (324), IV P. tuscumbiae Ulrich 137, 140 Period, definition of 27 Peters Creek, relation of to faults 64 Peters Creek faults, description of 69-70 Phanerotrema ? sp 95 Phillipsia sp 119, 138, 148, 171, 182, 183, 185, 196, 208 Pholidostrophia 89 Phractopora sp 138, 148, 157 Pierce mines, description of 302 superficial deposits at 256-257 Pinna 123, 124 P. missouriensis 875 Plant remains, in Casey ville formation 227 Degonia sandstone 217 Hardinsburg sandstone 190 Palestine sandstone 212 Tar Springs sandstone 201 Platyceras sp 196 Platycrinus 99, 114 P. huntsvillae 115, 883 P. penicillus Meek and Worthen . 110, 115, 116, 117, 118, 120, 130, 139, 141, 142, 156, (333), V Pleurotomaria ? sp 138, 215 Polypora approximata Ulrich 215 P. cestriensis Ulrich 138, 140, 148, 157, 183, 196 P. corticosa Ulrich 207 P. simulatrix Ulrich 99 P. sp 100, 183, 185, 196, 207, 215 P. spinulifera Ulrich - 138, 140. 182 412 INDEX PAGE Polypora tuberculata Prout 170, 173, 183, 207 Population 20 Posidonomya ? sp 95 Pottsville group 225-229 Pre-Devonian strata, description of 79-81 Preen mine, description of 298 Prismopora 194 P. serrulata Ulrich 193, 195, 196-197, 198, (373-374), XI Productus 138. 214 P. arkansanus Girty ? 214 P. burlingtonensis Hall ? 95 P. cf . inflatus McChesney 138, 196 P. inflatus McChesney 138, 148, 157, 183 P. inflatus McChesney ? 195 P. ovatus Hall 116, 117, 118, 138, 141, 182, 183, 185, 196, 207, 215, 222 P. parvus Meek and Worthen 116, 117, 118 P. setigera Hall 95 P. sp 138, 215, 222 Prospecting for fluorspar 290, 292-293 Proterozoic, definition of 27 Pterotocrinus 172, 174, 186, 208 P. actus var. bifurcatus 372 P. bifurcatus Wetherby 197, 208. (372), X P. capitalis (Lyon) 182, 183, 184, 185, 186, 187, 208, (360) VIII P. menardensis n. sp 207, 208, (373), XI P. platybasis 184 P. serratus n. sp 170, 173, (342), V P. spatulatus 208 P. sp 185, 195, 196, 214, 215, V Ptychomphalus sp 185 Pugnoides ottumwa (White) 116, 117, 118, 139, 142, 156, (35 1 -352), VIII Pugnoides ottumwa (White) ? 116 Pustula punctata (Martin) ? 196 P. sp 117, 118, 148, 183, 196 Pyrite, occurrence of 254 "Quartzite reefs", association of with faulting 58 Railroad, recent construction of 19 Rainey mine 302 Randolph County, early subdivisions of the Chester in 127 Reelfoot Lake, relation of to Hardin County faulting , 55 Reevesville, Herrin fault near 66 Reforestation, desirability of 43 Relief, mount of 21, 39-40 Renault funa, geologic range of 157-158 Renault formation, description of 142-159 Replacement of minerals 259-260 Residual deposits 231-232 Reteporina flexuosa (Ulrich) 182, 196 Reticularia setigera (Hall) 117, 118, 138, 139, 141, 148, 158, 164, 171, 173, 182, 183, 195, 196, 207, 214, 215 Reverse fault, definition of 30, 31 Rhipidomella dubia (Hall) 99, 117, 118, (351), VIII Rhombopora (several species) 215 R. sp 117, 118, 138, 170, 173, 196, 207 Rhynchonella explanata McChesney 360 R. ottumwa White 351 Rhynchonelloid 184 Rhynchopora perryensis Weller 182, 183, 184, 186 R. sp 196 R. ? sp 95 Ripple marks, significance of 24 INDEX 413 PAGE Roads, character of 42-43 relation of to topography 42-43 material for 306-307 Rock Creek, relation of to faults 64 Rock Creek fault, description of 68 Rock Creek Township, topographic age of hills in 48-49 Rose mine, description of 303-304 Rose prospect, Devonian limestone at 84 Rosiclare, Kinkaid fossils collected near 222 section at 113 topographic age of hills near 51 Rosiclare dikes, description of 239 Rosiclare fault, description of 70 relation of to fluorspar deposits 70 Rosiclare vein, description of 250 inclination of 64 Rosiclare Mill, description of 285-287 Rosiclare mine, description of 293-295 Rosiclare sandstone member, lithologic character of Ill thickness of 114 Rosiclare Township, limestone sinks in 41 topographic age of hills in 50, 51 St. Louis chert as road material 306 St. Louis limestone, description of 100-106 Ste. Genevieve chert as road material 306 Ste. Genevieve limestone, controversy about 129-130 description of 106-120 Sandstones, origin of 24 Schizodus depressus Worthen 185 S. sp 185 S. (?) sp 185 Schuchertella arctistriata (Hall) 86 Section of Bethel sandstone 162-163 Glen Dean limestone 195 Kinkaid limestone 221 Menard limestone 205-206 Paint Creek formation 166, 168 Ste. Genevieve limestone 112-113 Upper Warsaw formation 98 Vienna limestone 205-206 Waltersburg sandstone 205-206 Sedimentar}- rocks, age o 79 formation of 23-26 Septopora cestriensis Ulrich 183, 207, 215 S. n. sp 117, 118 S. similis n. sp (348), VII S. subquadrans Ulrich 138, 140, 148, 157, 170, 173, 182, 183, 195, 196 Shales, origin of 24, 25 Shetlerville fault, description of 74 Shetlerville fauna, geologic range of 140-141 Shetlerville formation, description of 133-142 Siebenthal, C. E., work of 277-278 Sigillaria 125 Sills, definition of 29 Silver, occurrence of 253, 275 Sink holes, drainage by 21-22, 40-42 Slickensides, definition of 59 Smith, W. S. Tangier, work of 275 Smithsonite, occurrence of in bedding deposits 255 relation of to mineralization 259 Soils 20, 44-46 Solenomya ? sp 95 Soward intrusion, description of 241-242 414 INDEX PAGE Sparks Hill, Clore fossils collected near 214 Sparks Hill intrusion, description of 242-244 Mississippian pebbles in 81 Spathiocaris emersoni 89 Sphalerite, deposition of 258 description of 253 separation of 293 Sphenopteris communis Lesquereux 227 S. sp. undet 201 Sphenotus sp 117, 118, 185 S. ?sp 95 Spirifer 123, 124 S. audaculus (Conrad) ? • 86 S. bifurcatus Hall 99, 119, 120 S. breckenridgensis Weller 138, 141 S. concinnus 86 S. divaricatus Hall 86 S. duodenarius Hall 86 S. gregarius Clapp ? 86 S. increbescens Hall 120, 149, 206, 207, 208, 209, 214, 215, (374-375), XI S. increbescens Hall, var 138, 183, 195, 196 S. increbescens var 141, 185 S. increbescens var. americana Swallow 185 S. increbescens var. transversalis Hall 148, 158, 171, 173 S. iowensis Owen? 86 S. lateralis Hall 99, 100 S. leidyi Norwood and Pratten 117, 118, 120, 138, 141, 148, 149, 158, 182, 183, 195, 196, 222 S. pellaensis Weller 116, 117, 118, 119, 120 S. segmentum Hall 85 S. subequalis Hall : 99, 100 S. tenuicostatus Hall 99 S. varicosum Hall 86 S. cf . worthenanus Schuchert 86 Spiriferina sp 117, 118 S. spinosa (Norwood and Pratten) . 148, 158, 171, 172, 173, 174, 182, 183, 185, 195, 196, 207, 215, (353-354), VIII S. subspinosa n. sp 138, 139, 142, (352), VIII S. subtexta White ? 99 S. transversa (McChesney) 116, 117, 118, 138, 139, 141, 148, 158, 172, 174, 182, 183, 195, 196, 207, 214, 215, (353), VIII Spirorbis sp 137, 183 Steel industry, importance of fluorspar deposits to 287 Stenopora cestriensis Ulrich 137, 140, 183, 207 S. cestriensis Ulrich ? 215 S. montifera Ulrich 137, 140 S. montifera Ulrich ? 117, 118 S. sp 137, 148, 207, 215 S. tuberculata (Prout) 117, 118, 137, 139, 140, 149, 157, 171, 173, 183, 185, 196 Stewart mine, description of 300 relation of to Wolrab Mill fault 67 Stibnite, occurrence of 254 Stigmaria 125 Stock raising, desirability of 45-46 Stone Hill, fossil collections from 227 maximum faulting near 73 Stone for roads 306-307 Straparollus sp 117, 119 Strata, definition of 25 Stratigraphic geology 77-145 Streams 21, 33-34, 39, 40 Streblotrypa distincta Ulrich 207 S. nicklesi Ulrich 138, 140, 148, 157, 170, 173, 183, 215 S. p 196,215 Strike, definition of 32 INDEX 415 PAGE Stroboceras sp 185 Structure 29-32, 56, 64 Structural geologj^ 55-75 Sulcatopinna missouriensis (Swallow) 208, 209, 222, (375-376), XI Superficial fluorspar deposits, methods in mining of 282 description of 256-^57 Surficial deposits 231-234 Systems, definition of 27 Talarocrinus 149, 150, 153, 154, 155, 156, 158, 162, 172, (336) T. buttsi Ulrich 137, 140, 142, 149 T. cornigerus (Shumard) (340), VI T. inflatus Ulrich (337), VI T. ovatus Worthen (340), VI T. planus n. sp (342), VI T. trijugis Miller and Gurley 137, 140, 148, 149, 157, (338), IV Talus deposits, definition of 35 Tar Springs sandstone, description of 198, 202 Taxocrinus sp 170, 173 Tentaculites bellulus Hall 86 Terebratula trinuclea Hall 854 Thamniscus furcillatus Ulrich 118, 138, 140, 170, 173, 183 T. ramulosus Ulrich 138, 140 Threemile Creek fault, description of 73 Thrust fault, definition of 30, 31 Topography 20, 39-40 development of 32-36 history of 47-52 relation of to farming 43-44 geologic structure 64 roads 42-43 Tower Rock, St. Louis limestone at 102 Towns 20 Tradewater formation, coal in 228 description of 228-229 Triplophyllum 99 T. sp 117,118 T. spinulosum (Milne-Edwards and Haime) 137, 139, 140, 148, 157, 170, 173, 182, 183, 185, 195, 196, 207, 215, 222 Twitchell mine, description of 298 Ulrich, E. 0., opinions of concerning Renault formation 150-159 work of 126-132, 275 Unconformities, definition of 26-27 Upper Mississippian series, se Chester group. Vein deposits, description of 248-254 methods used in mining of 282 minerals in 252-254 shapes of 251-252 Vienna limestone, description of 202 section of 205-206 Volcanic breccia, description of 237 Wallace Branch fault, description of 74 Waltersburg sandstone, description of 202 section of 205-206 Warsaw chert, value of for road material 306 Warsaw formation, description of 97-100 Weathering, definition of 34 Weller, J. Marvin, work of 218 Whittleseya microphylla Lesquereux 227 Williams Hill, altitude of 19 416 INDEX PAGE Wind-blown deposits 223-234 Wolrab Mill, outcrop of Warsaw formation near 99 Wolrab Mill fault, description of 67 Zaphrentis spinulif era 149 Zeacrinus maniformis 149 Z. sp 195 Zygopleura sp 117, 119 R.8E. ¥ 4 / WSSsm iHHIBli mSSsSmSSm Sou MftWMmmi mfim % 1 JMBFtBB HnHnHDONOnsfH