«•** ■ ■ •• ' ]' GEOLOGICAL SURVEY OF MICHIGAN ALFRED C. LANE, STATE GEOLOGIST VOL VIII PART 111 MARL (BOG LIME! AND ITS APPLICATION TO .THE MANUFACTURE OF PORTLAND CEMENT A BY DAVID J. HALE AND OTHERS u ACCOMPANIED BY TWENTY-THREE PLATES AND FORTY-THREE FIGURES PUBLISHED BY AUTHORITY OF THE LAWS OF MICHIGAN UNDER THE DIRECTION OF THE BOARD OF GEOLOGICAL SURVEY LANSING ROBERT SMITH PRINTING CO., STATE PRINTERS AND BINDERS I 903 Digitized by the Internet Archive in 2015 / https://archive.org/details/geologicalsurvey83mich / LIST OF PUBLICATIONS. DOUGLASS HOUGHTON, State Geologist. Reports from 1838-1846 were published with Legislative documents as follows: S. D. means Senate documents; II. D., House documents; J. D., joint documents. State Geologist is abbreviated S. G., and State Geological Survey, S. G. S. 1838. Report of a select committee of the Board of Regents of the University on the collection of the S. G. H. D. Vol. I, p. 1-2; S. D. No. 1, p. 1. H. D. No. 55 is duplicate of No. 1. Statement of the expeditures on account of the S. G. S. for the year 1837. H. D. No. 8, pp. 115-118; S. D. No. 21 (First annual account of the S. G.), pp. 315-318. Report of the S. G. (first annual). H. D. No. 24, pp. 276-317; S. D. No. 16; separately, No. 14, pp. 1-39. Communication from the S. G. H. D. No. 46. pp. 457-460. 1839. Report of the S. G. in relation to the improvement of State Salt Springs. H. D. No. 2, pp. 39-45; S. D. No. 2, pp. 1-8. Report of the committee on the S. G.’s report in relation to the improvement of the State Salt Springs. H. D. No. 4. pp. 123. Report of the S. G. in relation to the iron ore. etc., on the school section in town five south, range seven west, in Branch county. II. D. No. 21, pp. 342-344. Second annual report of the State Geologist. H. D. No. 23, pp. 380-507; S. D. No. 12, pp. 264-391; also separately H. R. No. 23, and S. R. sometimes misprinted No. 13 and No. 23, pp. 39 and appendix of sub- reports 123 pp. Report of the Committee of the Senate on Manufactures, to whom was referred the communication of the S. G. relative to salt springs and the salines of the State. S. D. No. 3, pp. 85-86 (parallel to H. D. No. 4). Communication from the S. G. relative to the G. S. S. D. No. 25, pp. 463-466; J. D. No. 3, app. 1840. Report of S. G. relative to the improvement of the Salt Springs. H. D. No. 2, Yol. I, pp. 18-23; S. D. No. 8. Yol. II, pp. 153-158. Annual report of the State Geologist (third, map of Wayne county). II. D. No. 27, Vol. II, pp. 206-293; S. D. No. 7, Vol. 2, pp. 66-153; separately H. R. No. 8, pp. 1-124. Report of the select committee to whom was referred the several reports of the S. G. % % % No. H. D. No. 46, Vol. IT, pp. 455-461. Report of the majority of the Committee of Finance on the communication and accounts of the S. G. for 1839. Report of the minority of the Committee on Finance on the same subject. Report of the select committee on S. G.’s report and accounts relative to improve- ment of Salt Springs, etc. S. G.’s account for the year 1839, the same being the subject matter of the three preceding reports. S. D. No. 15, 16, 17, 18, pp. 209-224. 1841. Special message concerning State Salt Spring^. H. S. and J. D. No. 5, pp. 235-254. Annual report of the S. G. (fourth). H. S. and J. D. No. 11, pp. 472-607; separately H. D. No. 27, pp. 1-184; S, 16, pp. 1-184. Report of the S. G. relative to county state maps. II. D. No. 35, pp. 94-98. 1842. Report of the S. G. relative to the State Salt Springs. H. D. No. 2, pp. 15-21; S. D. No. 1, pp. 1-9. Report of the select committee in relation to the report of the S. G. H. D. No. 19, pp. 77-79. Annual report of the S. G. (fifth.). H. D. No. 14, p. 6; J. D. No. 9, pp. 436-441. 1843. Annual report of S. G. (sixth). H. D., S. D., and J. D. No. 8, pp. 398-402. Report of the S. G. relative to the State Salt Springs. S. D. No. 9, pp. 402-408. 1844. Annual report of the S. G. (seventh). S. D. No. 11 (three pages). Maps of Washtenaw. Calhoun, Jackson and Lenawee counties were published separately. 1846. Report from. Geological Department by S. W. Higgins, principal assistant. J. D. No. 12, 22 pp. Report of the joint committee relative to the Geological Survey. .7. D. No. 15, 8 pages. ///- // T A. WINCHELL, State Geologist. 1861. First biennial report of the progress of the G. S. of M. Embracing observations of the Geology, Zoology and Botany of the Lower Peninsula. Made to the Governor, Dec. 31, 1860. The Walling Tackabury State Atlas contains a paper with geological and topo- graphic maps by A. Winchell, reprinted separately under the title “Michigan.” 1869. Report of the Join Committees on Geological Survey, made to the Legislature of Michigan, Lansing, W. S. George & Co., Printers to the State, pp. 1-15. 1871. Report of the progress of the S. G. S. of M., pamphlet, pp. 1-64. 1873. Vol. 1. Upper Peninsula. 1869-1873. Accompanied by an Atlas of maps. Edition 2 , 000 . Part I. Iron Bearing Rocks (Economic), T. B. Brooks. Of this an extra edition of 500 with thirteen accompanying atlas plates (1 to 13, No. 2 is misnumbered 11) was issued. Part II. Copper Bearing Rocks, Raphael Pumpelly (Plates 14, 14a 15-23 of the atlas accompanying; Chapters IV, VII. VIII are by A. R. Marvine). Part III. Palaeozoic Rocks, Dr. C. Rominger. ( Plate 24 of the Atlas accompanies. There was an extra edition of 500 dated 1872. without map or index, differing slightly in title page, introduction and paging). 1873. Yol. II. Upper Peninsula. 1869-1873, appendices to Part I, Vol. 1. A. Lithology by A. A. Julien, B. Lithology by T. B. Brooks and A. A. Julien. C. Lithology by Charles E. Wright, D. Ore deposits, E. Lithology (Notes by D. Houghton), F. Iron ore dock (by Jacob Houghton and Chas. H. Palmer, with Plate 20), G. Census statistics (1870), H. Magnetic Analyses (by F. B. Jenney), I. Mining laws (by C. D. Lawton), J. Metallurgical qualities by H. B. Tuttle, K. Con- tortions of Laminae (by T. B. Brooks). C. ROMINGER, State Geologist. 1876. Vol. III. Lower Peninsula. 1873-1876, accompanied by a geological mab. Edition 2 , 000 . Part I. Geology of the Lower Peninsula, by C. Rominger. Appendix A. Observations on the Ontonagon Silver Mining District and the State Quarries of Huron Bay, by C. Rominger. B. Report on the Salt Manu- facture of Michigan, by S. S. Garrigues, Ph. D., State Salt Inspector. Part II. Paleontology. Fossil corals, by C. Rominger (with 55 plates). 1881. Vol. IV. Upper Peninsula. 1878-1880, accompanied by a Geological map. (Edition 2,000.) Part I. Marquette Iron Region. Part II. Menominee Iron Region, by C. Rominger. See also Vol. V, Part I. See also reports by Brooks, Pumpelly and Wright in the reports of the Wisconsin Geological Survey. C. E. WRIGHT AND M. E. WADSWORTH, State Geologists. See Vol. II and Vol. V, also the reports of the Commissioners of Mineral Statistics and the following entry: 1893. Report of the State Board of Geological Survey for the years 1891 and 1892, to which are appended exhibits setting forth the Expenses of the Survey from its Inception to November, 1892, Exclusive of the Cost of Publication. Also the Reports of Dr. Carl Rominger for the years 1881-2; of Mr. Charles E. Wright for the years g 1885-8, ; of Dr. M. E. Wadsworth for the years 1889. 1890, 1891, 1892, made to the State Board of Geological Survey for the years named; also a Provisional Report by Dr. M. E. Wadsworth. State Geologist, upon the Geology of the Iron, Gold and Copper Districts of Michigan. L. L. HUBBARD, State Geologist. 1895.' Vol. V. Upper Peninsula, 1881-1884; Lower Peninsula, 1885-1893. (Edition 2,500.) Prefatory Historical Note by L. L. Hubbard. Part I. Geological Report on the Upper Peninsula of Michigan, exhibiting the progress of work from 1881-JS84. Iron and Copper Regions, by C. Rominger, accompanied by a map and two geological cross-sections. Part II. The geology of lower Michigan, with reference to deep borings. Edited from notes of C. E. Wright, late State Geologist, by Alfred C. Lane, Assistant State Geologist, with an introduction on the origin of salt, gypsum and petro- leum, by Lucius L. Hubbard, and accompanied by seventy-three plates and a map. 1899. Extracts from the annual reports of the State Geologist of Michigan, Lucius L. Hubbard, for the years 1897-1898. (By an error in Lansing this report really contains only the report for 1S98. Edition 500.) Vol. VI. Upper Peninsula, 1893-1897 (edition 954 and 200 of each part privately printed). Part I. Geological Report on Isle Royale, Michigan, by Alfred C. Lane. Assist- ant State Geologist. Accompanied by 16 plates and 29 figures, including map in cover. Part II. Keweenaw Point, with particular reference to the felsites and their associated rocks, by Lucius L. Hubbard, State Geologist. Accompanied by 10 plates and 11 figures. Part II. Appendix. The crystallization of the calcite from the copper mines of Lake Superior, by Charles Palache. Accompanied by six plates (100 extra printed separately). ALFRED C. LANE, State Geologist. Coal in Lower Michigan, by Alfred C. Lane, published serially in the Michigan Miner, Vol. I, Nos. 3 to 10, February to September, 1899, (500 reprints). 1900. Annual Report for the year 1899. Michigan Miner, Vol. II, No. 3, February, 1900 (500 reprints stitched in with the following No.). The Origin, Properties and Uses of Shale, by H. Ries, Special Agent for the State Geological Survey. Preliminary, inofficial, see Vol. VIII, Part I. Published in the Michigan Miner. Vol. 1, No. 12, Vol. 2, Nos. 1 and 3 (500 reprints). Vol. VII. Lower Peninsula. 1893-1899. (Edition 1,500 and 500 of each part issued separately). Part I. Geological Report on Monroe County, Michigan, by W. H. Sherzer. Accompanied by 17 plates and 8 figures, including three colored maps. Part II. Geological Report on Huron County, Michigan, by Alfred C. Lane, accompanied by 11 plates, 12 figures and one inserted table, including two colored maps. (100 extras of Chapter IX, X §2. and X §3). Part III. Geological Report on Sanilac County. Michigan, by C. H. Gordon, accompanied by 5 plates and 2 figures, including one colored map. Vol. VIII. Economic Geology, 1899. (edition 1,500, 500 of each part bound sepa- rately). Part I. Clays and Shales of Michigan, their Properties and Uses, by H. Ries, Accompanied by four plates and six figure 1901. Annual Report for the year 1900. Michigan Miner Vol. Ill, Nos. 2 and 3 (Reprints furnished and issued by State Board.) 1902. Vol. VIII. Part II. Coal in Michigan, its Mode of Occurrence and Quality, by Alfred C. Lane, accompanied by nine plates and nine figures, including one colored map. Report of the State B. of G S. for the year 1901. seven figures, fifteen plates and maps. (Edition 1,500, reprints 200 each, of numerous papers.) 1903. Report of the State B. of G. S. for the year 1902. (Edition 500, reprint from Michigan Miner, Vol. V. No. 2.) Vol. VIII. Part 111, Marl (Bog lime) and its Application to the Manufacture of Portland Cement, by David J. Hale and others, accompanied by 23 plates and 43 figures, including one colored map. GEOLOGICAL SURVEY OF MICHIGAN LOWER PENINSULA 1 900 - 1 903 VOL. VIII PART I. CLAYS AND SHALES, H. RIES PART II. COAL, A. C. LANE PART III. MARL [BOG LIME], D. J. HALE GEOLOGICAL SURVEY OF MICHIGAN ALFRED C. LANE, STATE GEOLOGIST VOL. VIII PART 111 MARL (BOG LIME! AND ITS APPLICATION TO THE MANUFACTURE OF PORTLAND CEMENT BY DAVID J. HALE AND OTHERS ACCOMPANIED BY TWENTY-THREE PLATES AND FORTY-THREE FIGURES PUBLISHED BY AUTHORITY OF THE LAWS OF MICHIGAN UNDER THE DIRECTION OF THE BOARD OF GEOLOGICAL SURVEY LANSING ROBERT SMITH PRINTING CO., STATE PRINTERS AND BINDERS I 903 Entered according to Act of Congress in the year 1903, by Governor A. T. Bliss for the State of Michigan, in the Office of the Librarian of Congress, Washington, D. C. Office of the State Geological Survey, Lansing Mich., March 31 , 1903 . To the Honorable , the Board of Geological Survey of Michigan: Hon. A. T. Bliss, Governor and President of the Board. Hon L. L. Wright, President of the Board of Education. Hon. Delos Fall, Superintendent of Public Instruction and Secretary of the Board. Gentlemen — Herewith I transmit as Part III, the concluding part, of Vol. VIII, a report containing the results of examination of the raw materials of the Portland Cement industry, more particu- larly the beds commonly known as marl, but more properly known as bog-lime, for the more nearly pure calcium carbonate a bed is the more valuable it is. My original plan was for a brief report something upon the order of that by H. Ries in Part I of this volume, arrangements for which were made about the same time, to be prepared wholly by Mr. Hale. But the subject grew upon him, and he obtained the promise of co- operation from Messrs. Lathbury and Spackman and R. L. Hum- phrey, whom we have to thank for their valuable papers. I had also expressed to C. A. Davis my feeling that, for reasons which I have elsewhere given, none of the theories then current were competent to account for the origin of these very extensive and pure deposits of calcium carbonate. He suggested the agency of the algae, and at my request worked the matter out, with the results herein in- corporated, and I believe his contribution is a most valuable addi- tion to science. In the meantime , facts of one sort and another kept accumulating, and so the present report was built up. I trust that its lack of unity may be atoned for by its value. If it trespasses rather far into the field of manufacturing for the economic geologist, I can only say that Mr. Hale thought that this would be useful, and that some description of the methods of manufacture were needed to understand those properties of the raw material which were most valuable. This volume is already too large, or I should have been tempted to IV MICHIGAN GEOLOGICAL SUBVEY. add to the treatment of the three materials for cement considered herein, clay, coal and bog-lime, a fourth part on limestone. The State contains much limestone suited for the manufacture of Portland cement, and the question between it and bog-lime is a business one, whether it is cheaper to grind up the limestone or evaporate the water out of the marl. The output of a plant will ordinarily be in- creased by using ground limestone. Nothing in science is final, and this report is not the last word on the subject. Prof. E. D. Campbell of the University at Ann Arbor is even now at work on a very important series of papers, affecting, however, more especially the theory of manufacture. With great respect I am your obedient servant, ALFRED C. LANE, State Geologist. TABLE OF CONTENTS. CHAPTER I. INTRODUCTION. CHAPTER II. USES OF MARL. Page Sec. 1. Quicklime - •> 3 2. Fertilizer 3 3. Minor uses 4 CHAPTER III. THE USE OF MARL FOR CEMENT MANUFACTURE. 1. Description • • ^ 5 2. General distribution 9 3. Prospecting tools 9 Method of operating 12 4. Location of marl 13 5. The distribution of marl in a single bed 16 6. Surroundings of marl 23 (a) Shore wash 4 (b) Streams 25 (c) Surface 25 (d) Silt under water 26 (e) Lining of marsh growth or decayed plant life 27 (f) Organic matter permeating deposits 27 (A) Organic matter of the marl deposit 27 (B) Organic matter of drainage 27 (g) Materials underlying marl 28 (h) Materials overlying marl 29 7. Method of prospecting a given area 29 8. Commercial importance of composition 30 (1) Appearance 31 (2) Composition 32 (3) Interpretation 34 Calcium carbonate 34 Magnesium carbonate Ferric oxideand alumina . . Insoluble and soluble silica. Soluble silica 36 Organic matter 37 Sulphuric and phosphoric acids, chlorine, etc 37 9. Location and size of bed 38 CHAPTER IV. THEORIES OF ORIGIN OF BOG LIME OR MARL. 1. Introduction— the various theories 41 (1) Shell theory 41 (2) Sedimentary theory 42 (3) Chemical theory 42 8 ? $ 8 ? VI CONTENTS. Introduction - Continued: Page S EC. 2. Shells 43 3. Sedimentary theory 44 4. Chemical theory 44 5. Indications by circumstances of occurrence 47 CHAPTER V. A CONTRIBUTION TO THE NATURAE HISTORY OF MARL. BY C. A. DAVIS. 1. Historical introduction 65 2. Ultimate sources 66 3. Alternative methods of deposition 66 4. Cause of deposition upon aquatic plants 69 5. Relative importance of Chara (Stone wort) 70 Analytical tests 71 References in literature 77 Sources of thick crust 79 6. Marl beds without Chara 81 7. Association of marl and peat 82 8. Turbidity due to marl 83 9. Conclusions 86 10. Method of concentration by Chara 87 11. Blue-green algae and their work 90 12. Littlefield Lake, Isabella county 92 Appendix, on the shells of marls by Bryant Walker 97 Notes 98 Localities 99 CHAPTER VI. RECORD OF FIELD WORK. 1. Lansing— Summer, 1899 103 White Pigeon 103 Bronson, Quincy, Cold water 104 Jonesville 106 Kalamazoo 106 2. Cloverdale 107 Cloverdale Region— Summary 128 3. Pierson Lakes 131 4. Lime Lake and vicinity 133 Lime Lake 134 Twin Lakes 134 5. Fremont district 135 6. Muskegon district 137 7. Benzie county 137 8. Harrietta 138 9. Escanaba 138 10. Munising 139 11. Wetmore 139 12. Manistique 140 13. Corinne 140 14. Grand Traverse Region 141 15. Central Lake 142 16. East Jordan and vicinity 148 17. Manistee Junction 150 18. Rice Lake 151 19. St. Joseph River and tributaries 1 d4 20. Onekama 154 CONTENTS. vii CHAPTER VII. MANUFACTURE OF PORTLAND CEMENT FROM MARL. Page Sec. 1. Introduction r 153 2. Definition of terms 158 3. Historical 159 4. Materials for cement 160 5. Kiln process of cement manufacture 162 6. The rotary process 163 7. Preliminaries 165 1. Digging 165 2. Draining 166 3. Dredging 166 8. Estimates on raw material 167 9. Requisites for marl deposit 169 Surfacing 169 Necessary composition — ........ 169 Depth 169 Sulphuric acid A 170 Magnesia 170 Grain 170 10. Clay 170 11. Admixture of raw materials 171 12. Mixing and raw grinding 173 13. Burning 174 14. Clinker grinding 179 15. Motive power 184 16. Storage and packing 185 17. Specifications for cement 186 18. Buildings 188 19. Review 189 APPENDIX TO CHAPTER VIII. THE DEVELOPMENT OF MARL AND CLAY PROPERTIES FOR THE MANUFACTURE OF PORTLAND CEMENT. BY B. B. LATHBURT. CHAPTER VIII. NOTES ON THE ORIGIN OF MICHIGAN BOG LIMES. BY A. C. LANE. 1. Introduction 199 2. Origin of bog lime, chemical considerations 199 Abstract of Treadwell and Reuter’s article 201 1. Calcium bicarbonate 206 2. Magnesium bicarbonate 213 3. Calcium bicarbonate in solution with NaCl 213 4. Sodium bicarbonate 215 3. Microscopic investigations 218 (a) Microscopic precipitate by loss of CO 2 and heating 218 (b) Precipitate by evaporation 220 (c) Chara fragments 220 ( Blue green algae .h 221 (e) Shell structure 221 (f) Limestone flour 222 4. Conclusions 223 Vlll CONTENTS. CHAPTER IX. LIST OF LOCALITIES AND MILLS. COMPILED BY A. C. LANE. Page Sec. 1. Introduction 224 Alpena Portland Cement Co 224 Omega Portland Cement Co 227 Peninsular Portland Cement Co 233 Peerless Portland Cement Co 237 Bronson Portland Cement Co 239 Newaygo Portland Cement Co. (Gibraltar Brand) 240 Elk Rapids Portland Cement Co 244 Wolverine Portland Cement Co 246 Michigan Alkali Co., Wyandotte (J. B. Ford) 248 Hecla Cement and Coal Co 251 George Lake 252 Edwards Lake 253 Chapman Lake 253 Plummer Lake 254 CrapoLake 254 Mills Lake 255 The Great Northern Portland Cement Co 268 Detroit Portland Cement Co 270 Egyptian Portland Cement Co 272 Twentieth Century Portland Cement Co 281 Zenith Portland Cement Co 282 Standard Portland Cement Co 288 Wayne Portland Cement Co 291 Pyramid Portland Cement Co 291 German Portland Cement Co 291 Three Rivers Cement Co 292 Farwell Portland Cement Co 292 Clare Portland Cement Co 293 Watervale Portland Cement Co 297 Lupton Portland Cement Co 297 Standiford Portland Cement Co 301 Bellaire Portland Cement Co 306 West German Portland Cement Co 306 Locations reported by Douglass Houghton Survey 306 Marl 309 Local details of Marl, Jackson County 309 Eaton and Kalamazoo Counties 310 Calhoun, Kent and Ionia Counties 311 Locations arranged by counties 312 Monroe and Lenawee counties 312 Hillsdale, Branch, St. Joseph and Cass counties 313 Berrien, Van Buren, Kalamazoo, Calhoun counties 314 Jackson, Washtenaw, Wayne, Macomb counties 315 Oakland, Livingston, Ingham counties 316 Eaton, Barry counties 317 Ottawa, Allegan, Kent counties 318 Ionia county 319 Clinton, Shiawassee. Genesee counties. 320 Lapeer, St. Clair, Sanilac, Huron, Tuscola counties 321 Saginaw, Gratiot, Montcalm counties 322 Muskegon, Oceana, Newaygo counties 325 Mecosta, Isabella, Midland, Bay, Arenac, Gladwin counties 326 Clare, Osceola, Lake, Manistee, Wexford, Benzie, Grand Traverse, Leelanau counties 327 Ogemaw, Iosco, Alcona counties 334 Oscoda and Crawford counties 337 CONTENTS. IX Sec. 1. Introduction— Locations by counties— Continued: Page Kalkaska, Grand Traverse. Benzie, Leelanau, Antrim, Otsego, Mont- morency, Alpena counties 338 Presque Isle county 339 Cheboygan, Emmet counties and Upper Peninsula 340 Houghton county 341 Marls and Clays in Michigan by Delos Fall 343 Marl 343 Michigan Clays 345 Discussion 317 CHAPTER X. METHODS OF AND COMMENTS ON TESTING CEMENT. BY RICHARD L. HUMPHREY. Sampling 359 Chemical analysis 360 Specific gravity 362 Fineness 363 Normal Consistency 364 Time of setting 366 Tensile strength 368 Constancy of volume 374 Conclusion 377 X ILLUSTRATIONS. LIST OF ILLUSTRATIONS. PLATES. Opposite page Plate 1, Marl soundings 1, 2, 3, 4, 11a, lib, by D. J. Hale 16 Plate 2, Horseshoe lake and soundings 48 Plate 3, Union City (Peerless plant) and Coldwater (Wolverine plant) 104 Plate 4, General exterior view of an eleven kiln plant 160 Plate 5, General plan of four kiln plant, with place for expansion 168 Plate 6, General interior view of slurry department 170 Plate 7, View of rotary 174 Plate 8, Front hoods of rotary kilns and clinker elevators 184 Plate 9, General plan of a three kiln plant, with elevations 184 Plate 10, Battery of Griffin mills, grinding clinker 184 Plate 11, Cross section of a Griffin mill 184 Plate 12, View in Newaygo Cement plant 190 Plate 13, General interior showing tube mills 198 Plate 14, A. Office building, with laboratories, etc 198 B. Stockhouse, with self-discharging bins under construction C. Bottom of concrete slurry pits under construction D. Dry marl deposit with hauling arrangement . Plate 15, General exterior view of four kiln plant 198 Plate 16, Microscopically enlarged fragments and sections of Chara 220 Plate 17, Plan and view of Newaygo plant 240 Plate 18, Dam and raceway for Newaygo plant 240 Plate 19, Property and borings of Farwell P. C. Co. at Littlefield Lake 292 Plate a), Map of Marl beds of the S. P. Cement Co., Athens, Mich 304 Plate 21, Silver Lake marl beds 320 Plate 22, General view of Newaygo Cement plant 328 Dredge excavating marl, Newaygo Plate 23, Index map End. FIGURES. Page Fjg. 1. Liquid marl sampler 11 “ 2. Robert W. Hunt &Co., sampler 13 “ 3. Sketch map of Hope Township, Cloverdale district 14 4. Soundings 28. 29, 31, 32 Cloverdale 115 “ 5. Soundings 33, 34, 36, 37, Cloverdale 120 “ 6. Soundings 36, 37, 38, 39, 40, 42, Cloverdale 121 “ 7. Soundings 3 to 8 Pine Lake 127 8. Fremont Lake 135 9. Soundings 1 to 4, Duck Lake 142 “ 10. Soundings at Central Lake 143 “ 11. Section across North Island, Central Lake. 144 “ 12. Section across South Island. Central Lake 145 “ 13. Rice Lake 152 “ 14. Portage Lake, Onekama 155 15. Tube mill 173 “ 16. Apparatus for filtering bicarbonate soluiion 201 17. Treadwell and Reuter’s apparatus 203 18. Bottle 204 19. Isotherms and ground water temperatures of Michigan 216 “ 20. Precipitated crystals 219 • 21. Plat of Raffelee Lake 273 “ 22. Plat of Runyan Lake 274 “ 23. Plat of Mud Lake 275 “ 24. Plat of Warren and adjacent lakes 276 “ 25. Plat of Bush Lake 277 ILLUSTRATIONS. xi Page . 26 Sketch map of Grass Lake, Zenith P. C. Co 284 27. Sketch map of Lakelands, Standard P. C. Co 289 28. Table of soundings of Standiford P. C. Co 303 29. Table of analyses of Standiford P. C. Co 304 30. Table of analyses continued 305 31. Sketch map of Cedar Lake and adjacent marl beds 323 32. Section of marl deposit near Houghton 341 33. Apparatus for determining the strength of mortars 355 34. Figures illustrating cement tests 356 35. Vieat needle, as originally designed 356 36. Modern form of Vieat needle and other testing apparatus 365 37. Tanks for the preservation of briquettes 370 38. Olsen Testing Machine, hand driven 371 39. The same, power driven 372 40. Fairbanks testing machine 373 41. Riehle testing machine 374 42. Result of tests of constancy of volume 376 43. The same, “pat tests” 376 44. Diagram illustrating the relative strength of cement at various epochs 383 ERRATA. Page 190, line 18, for Cederburg read Cederberg. Page 277, the figure 25 is inverted. CHAPTEE I. INTRODUCTION. The grayish mud underlying our lakes and marshes has but very recently become one of the greatest resources of our state. On account of its position, being covered in most part by water or muck, it is not often seen and few people are familiar with its name or appearance. Factory men have, however, after having become aware of its presence in such quantities in the state, made good use of it as a raw material for the manufacture of the best Portland Cement. A factory was started at Kalamazoo in 1872 (a description of its marl bed is found in Ch. Y, Sec. 1). Here the old set or dry kiln process proved too costly and the site was abandoned. The first successful factories were started at Bronson and Union City. At the former place the marl was discovered by a section foreman who was sinking piles to support a railroad bridge which was to span the creek draining the deposit. The Bronson works use the Ransome rotary kiln wet process and the Union City factory, which first used the older style set kiln, are also adopting the wet process. These plants have proved very successful and the interest among capitalists and landowners throughout the State has been intense to know more about the industry and how to gauge the true value of marl lands. It will not be possible in the following pages to describe the raw material marl and its factory requisites so that any one may at once identify his marl bed as either worthless or specially fitted for cement manufacture. This comes only with the examination of many beds and the correct summing up of numberless possibili- ties all of which cannot be so minutely described as to be foreseen. The work of deciding on the final merits of a bed should be left where it belongs, with a specialist. The writer will then be satis- fied if, from reading the following pages, landowners and amateur o MABL. prospectors can form a clear idea of what commercial marl is, how to go about prospecting for it, and how to decide correctly whether a given bed warrants a thorough examination for factory purposes. Chapter II touches lightly upon other uses of marl. Much may be found in the early State and United States reports concerning these uses. Chapter III discusses the adaptability of marl to cement manu- facture. In Chapter IY it is intended to give a description of as many views as possible of the origin of marl in the hope that there may be something of truth in one or all. Aside from its prime interest from a scientific point of view this chapter should afford some clue as to the location of marl beds and assist in their discovery by the explorer. Chapter VII is intended to show both the magnitude of the cost and the numberless details to be calculated to a nicety by any individual or company embarking in the enterprise of cement manu- facture. Chapter VI gives many details which it is hoped will be useful to any one interested in the subject and shows somewhat the varia- tion in mode of occurrence. Credit is due to A. C. Lane, State Geologist, for his advice and assistance in the work throughout, also to Lathbury & Spackman of Philadelphia for their article and cuts of machinery. I also wish to tender thanks to the many men throughout the State who have assisted me in sounding beds and aided me with timely information. Assistance was given to Prof. I. C. Russell in the preparation of his report on the Portland Cement Industry of the State, in the Twenty-second Annual Report of the U. S. Geological Survey, which he has therein acknowledged, but his report did not come to hand until this report was being read in page proof, so that we are not able to incorporate all the additional valuable information therein contained. CHAPTER II. USES OF MAKE. § 1. Quicklime. Marl has long been known in this State for its use in many dif- ferent ways.* On the shore of many marl lakes there are to be found the remains of old lime kilns. These were erected for the purpose of burning the marl to lime. By a slow fire from beneath the organic matter was partly burned out and the carbon dioxide was driven off, leaving a fairly pure calcium oxide or the ordinary quicklime. Many log houses are still standing which were built with mortar of this kind or even with the unburned marl itself. But on a large scale this proved too costly a process compared with that later employed, wdiich is the burning of limestone for lime. The reason for the greater costliness of the marl method is that the marl is really too bulky to handle with profit, for after the water is driven off there remains but little over half the original bulk as dry marl. From ten to fifty per cent of what remained after drying would then be burned as organic matter, implying a further shrinkage. On the other hand the limestone is more com- pact, has as a rule less organic matter, and is drier so that there is not the immense waste of fuel in driving off the water in the form of steam before the actual work of burning takes place. For these sufficient reasons limestone has taken the place entirely of marl as a raw material for the production of commercial lime. § 2. Fertilizer. Marl is used widely as a fertilizer. New Jersey marl is very much more useful than ours on account of its valuable content of phosphorus. As the marl of Michigan contains little besides cal- cium and magnesium carbonates it has scarcely a commercial value for this purpose as the cost of transportation to any distance would easily exceed the value of the benefit derived from it as a fertilizer. *Winchell, 1860, p. 131. See also Houghton’s reports, 1838, p. 34; 1839, 1840, p. 94, etc. 4 MARL. Its real value, however, when in close proximity to the land upon which it is to be used, is often underestimated. Many beds of marl in this State were visited which lay very near to land which they would enrich, upon a judicious application, and the benefit to be derived from such application would have been greater than that from application to factory purposes. If marl is dug and allowed to lie over winter till it has been exposed to freezing and thawing, its lumpy tendency will be overcome and if then spread on a tough clay it will break it up and make it more easily cultivated. On the other hand, if it is to be applied to a coarse sand it will fill up the interstices of the coarser soil, rendering it able better to hold moisture and retaining humus which would, if allowed, accumulate,, as well as other fertilizers which may be added. The chemical effect of marl is not described minutely, as much may be found written elsewhere on the subject. The effect, though slow in making itself felt, is very beneficial, as the lime of the marl gradually makes soluble for the plant the otherwise insoluble con- stituents of the soil. It must not, therefore, be taken for granted that, because a marl bed does not prove fit for the manufacture of Portland Cement, it is altogether useless to an agricultural community. Despite the amount of time and trouble so far devoted to the explanation of its value as a fertilizer its use for this purpose is not fully understood or taken advantage of. § 3. Minor uses. There are several other uses for marl which cause but little de- mand. It is often used in tooth and scouring powder and as adul- terant for paints. As these uses on account of the very small de- mand they could create for marl are of scarcely any commercial importance it is proper to pass on to its prime use in the manufac- ture of Portland Cement. CHAPTER III. THE USE OF MARL FOR CEMENT MANUFACTURE. § 1. Description. The name “marl” is often heard but not with the precise mean- ing in which it is used in Michigan. It is a somewhat general name applied in different parts of the country to substances which differ in appearance and characteristics. Descriptions are given in the United States Geological Reports of extensive deposits of “marl” or “green sand” in New Jersey. These deposits occur in a distinct geological formation and contain the remains of animals and hence are rich in phosphates. They are called “green sands” from their color and are much prized on account of their phosphorus as fertilizers. The marls in North and South Carolina cover some two thousand miles area and like the New Jersey marl belong to a different geological era from ours. Another meaning of marl which more easily fits the term as used in Michigan is the name marl as applied to calcareous clays. In this sense of the word, however, half of Michigan could be called marl, for the light colored clays which form half our clay banks are calcareous or very rich in calcium carbonate. The indefinite or uncertain meaning of the term “marl” is very well illustrated by the definition as given in our dictionaries. “A deposit of amorphous calcium carbonate, clay, and sand in various proportions characterized usually by the most prominent ingredient; as clay-marl; shell-marl, a valuable fertilizer; green sand marl, a valuable mixture of green sand and clay.” The first step in the study of Michigan “marl”* should be to dis- tinguish it carefully from the marls of other localities and from other formations closely allied to it in appearance and chemical composition. First of all our marl is nearly a pure “amorphous calcium carbonate.” This is likewise true of several other similar ♦More properly bog lime. L. 6 MARL. compounds. An amorphous calcium carbonate is a mineral com- pound, calcium carbonate, the particles of which appear not to exist in a crystalline form.* Chalk is an amorphous carbonate as well as limestone. The composition of pure marl, chalk, and lime- stone agree very closely, but they differ much in the tenacity with which the individual particles cohere and in their con- tent of moisture. Our marl as now considered is much like the other two in color and grain, but is more bulky and usually contains more organic matter. On the other hand a very good example of a calcium carbonate which is not amorphous, but is crystalline, is marble. This has undergone changes which have made its molecules very tenacious of one another so that if would be too expensive to grind it into powder for the manufacture of cement as in the case of the materials before mentioned. The marl then, closely resembles in composition chalk and limestone and lacks with them the crystalline formation of marble,* although the last is a calcium carbonate. The four materials of like com- position decrease in the tenacity with which their particles cohere in the following order; marble, limestone, chalk, marl.. The last named, our own raw material, is then the most easily ground and, in that respect at least, much the easiest to pulverize for intimate mixture with clay in the manufacture of Portland Cement. The marl of our State should also be distinguished clearly, not only from kindred materials, but also from other materials bearing the same name. It was above mentioned that the New Jersey and Carolina marls belonged to a distinct former geological period. Our own deposits as far as can be ascertained are distinctly of the present time and occur in an area limited by the former extent of the ice-sheet. They extend about the Great Lakes, being found in Wisconsin and both peninsulas of Michigan, extending northward into Canada and southward into Indiana and Illinois. It is not a continuous bed, but lies only in the deep pockets or holes and old drainage valleys left by the glaciers. As so far seen it has never been covered by over thirty or forty feet of modern drift. Before it is studied further as definite a description as possible should be given of its appearance and composition with variations carefully noted so that it may be easily and certainly identified. *But see notes on microstructure in the last chapter. THE USE OF MARL FOR CEMENT MANUFACTURE. 7 It is often mixed with clay and the combination, a calcareous clay, is termed “marl.” This usage does not give the meaning of marl as it is now used in Michigan in the cement industry,* but con- fuses it with clay with which it should be sharply contrasted. Again marl is found either mixed with sand, organic matter, or shells, to such an extent that its own characteristics are not clearly shown. It will therefore here be described as it exists in a fairly pure condition. First it is found under lakes or swamps in the form of a mud consisting of from 25$ to 50$ moisture. In this condition it may appear dark gray, about the color of wood ashes, or nearly white. Upon drying it becomes much lighter in color. It coheres slightly and upon drying lumps much as does clay, but upon weathering breaks down into a friable mass. A very pure marl tastes much like chalk and often has a more granular appearance than the darker samples. As compared with the clay which is often found in its neighborhood it is much lighter bulk for bulk, and if each is stirred up in water the marl water clears much more quickly as its granular nature causes it to deposit first, while on the other hand, the particles of clay remain suspended in the water for some time before complete sedimentation takes place and the water becomes clear. Also upon the addition of an acid to two samples, one of marl and one of clay, the former will effervesce with formation of gas much more freely than the latter. The easiest way to dis- tinguish marl from sand is by detecting the presence of grit. The particles of marl crumble easily upon compressing between the thumb and finger while fine sand feels hard. Shells, or their re- mains, are easily distinguished by their form and usually though not always form a greater or lesser portion of the marl. The greatest adulterant of marl, always forming at any rate a part of it, is organic matter. Its proportion can be roughly estimated by color of the mixture, — the darker the sample, the greater the per- centage of organic matter. This may be sometimes so large that the marl becomes practically a muck or so small that it scarcely affects the pure white of the calcium carbonate. As the contamination and consequent variation in appearance of marl is important to both manufacturer and scientist, its cause should be thoroughly understood. As stated in the definition, an ♦Though correct enough in itself. The Michigan “marl” is more properly bog lime. L. 8 MARL. impure marl derives its name from the impurity which predomi- nates. It has been stated briefly how to distinguish the true marl from each of its impurities when the marl and its adulterant exist as separate samples. Sand, clay and organic matter are not only found near the marl, but intimately mixed with it. The following analyses are those of three samples of so called marl taken from the same chain of lakes. Insoluble. Aluminum and Iron Oxides. Calcium Carbonate. Magnesium Carbonate. Organic matter. <1) 75.04 1.90 14.02 6.05 2.99 (2) 57.04 4.30 22.06 12.45 4.15 <3) 15.14 13.73 43.13 1.66 26.34 The measure of purity in each of the above samples must be found in the column under calcium carbonate. It is readily seen that all are very low and that each sample is very impure. The impurity in each case is, however, due to a different cause. No. 1 was largely sand, and in confirmation, notice the high per cent of “insoluble.” Though of a marly nature it is full of grit, as could easily be detected by the touch. No. 2 is largely clay and has also a high “insoluble.” It has besides nearly twice the magnesium carbonate of No. 1. The reason for this is that clays laid down at the same level as marls nearly always have a high per cent of magnesium carbonate as well as calcium carbonate, which increases the proportion of the former as compared with the percentage in true marl. No. 3 shows a more even distribution of the different impurities, but organic matter predominates. This appeared as a dark grayish muck and resembled but slightly a pure marl. It contains also 13.73$ of iron and aluminum oxides so that it in- clines somewhat toward a bog iron. It was found fifty feet under water. In all of the above samples the marl has partly lost its identity, becoming in the several instances, a marly sand, a marly clay, and a marly muck. A careful examination of the color, grittiness, weight and effect of acid will soon reveal the true nature of the mixture and to what ingredient the contamination is due. Fortunately for the factory interests of the State, marl is not often subject to such great variation in appearance and composi- tion, but has somewhat definite characteristics of its own. Its THE USE OF MAIiL FOE CEMENT MANUFACTURE. 9 exact chemical nature together with its factory requisites will be considered in the final section in this chapter. All that will help the prospector to identify it on the ground is to know that it is generally somewhat granular in appearance, in color varying from that of dark ashes to dirty flour, is sticky and sometimes even soapy and greasy to the touch, and is distinguished from clay by its greater bulk and granular nature, from sand by the absence of grit (it usually contains a trace at least of quartz sand or diatom- aceous silica), and from organic matter by its lighter color. § 2. General distribution. The physical appearance of Michigan is necessarily of much inter- est to the prospector. Glacial action in past ages diversified the surface of the State and has left it ridged and hollowed thoroughly. Whatever may be taken as the agent of marl deposition, it is certain that these glacial valleys furnish the most favorable conditions for its existence and are its most usual resting place and that the present drainage furnishes the direct cause for its impurities. § 3. Prospecting tools. The first thing necessary in prospecting is to get tools to work with. Several machines have been patented for the purpose, but as an owner usually wishes to sound only his own locality, the simpler and less costly the apparatus, the better. The following is the description of a very simple outfit which is all that is necessary in the majority of beds. It must, however, be manipulated with care to obtain strictly trustworthy results. 1. Weld an ordinary two inch augur on a three-eighths inch gas pipe two feet long. 2. Thread the unwelded end of the pipe for coupling. 3. Cut three lengths of pipe each in half, or cut each into four equal lengths if it is desired to carry the outfit long distances. Thread the ends of the pipe for coupling. 4. Get couplings enough to couple all together making a con- tinuous rod with an augur attached. 5. A “T” coupling must be inserted on the rod farthest from the augur and through this a rod or stick can be passed to turn the rod. A better way is to screw into each free end of the “T,” a rod or a piece of gas pipe eighteen inches long. This makes a handle to the augur that can be inserted at any distance from the end. Usually a pair of Stillson wrenches are needed to untwist the pipe, which becomes very tightly connected during the boring. 2-Pt. Ill 10 MARL. Three-eighths inch pipe will be found to lift out much easier than half-inch, but will not stand boring to a great depth. If three- sixteenth inch is used it is liable to kink badly when sunk to any depth. On the other hand inch pipe cannot be thought of for the purpose as it would take a jack screw to lift the rod out. In the use of any size pipe or any style of sounding implement it must always be borne in mind that the quicker the work of sinking the rod, securing the specimen, and raising it is performed, the easier the work can be done. The reason for this is that the marl con- sists of finely divided particles partly suspended in water, making a mud. When the rod shoves aside these particles it takes them but a short time to pack around it. If it is withdrawn quickly before the particles assume their new position, about half the friction of marl against pipe is avoided and the work of withdrawal much lessened. This is the simplest and most easily prepared and also the cheap- est means of reaching the marl. Care must be taken to bore, twist- ing the handle as the rod is shoved down. It can generally be shoved through the mud with application of but little force, but if this is done the pod of the augur will remain filled with the surface marl which is first encountered in its descent and will bring that same marl to the surface again instead of filling with that at the bottom. Also the couplings must be very firmly started when each new length of pipe is added as the rod penetrates the marl. Many outfits are lost by the neglect of this little precaution. There is no reason why this simple apparatus should not do good work in most of the marl beds of the State. It can be made to penetrate a marl of medium consistency with considerable ease, requiring two or three men to run it. When the augur strikes sand at the bottom of the bed or in its course downward it can generally be detected by the peculiar grating sound and jar of the pipe in the hands of the operator. When it strikes clay the increased difficulty of bor- ing is at once made manifest and it is well to immediately hoist the rod, as after boring a short time in the clay beneath the marl the apparatus will be freed with great difficulty. In deep borings care must be taken to keep the rod moving if possible, either up or down, as its recovery is easier. This apparatus suffices for a fairly dense marl because the augur will clear itself of the surface drift on the way down and will retain fairly well the clean sample taken at the bottom. It will not, THE USE OF MARL FOR CEMENT MANUFACTURE. 11 however, take true samples where the grass or roots are very thick at the top and the marl is so fluid as not to be retained readily on the pod of the augur. In beds of this nature a different device will be required to obtain samples which will give a trustworthy idea of the center of the bed. A rather clumsy but efficient device has been used (Fig. 1), which is a remodelling of that used by Mr. Farr of Onekama. ; i ; ; k- ir 9 »t -.«.***« - -V ^'4 '£ - Fig. 1.— Farr’s Liquid Marl sampler. For description see p. 11. 1. Cut a piece of inch gas pipe two feet length. 2. Thread one end of the same. 3. Screw reducers on the threaded end till the last reducer can take a half or three-eighths inch pipe. 4. If no time and materials are at hand to make a disk to close one end of the large pipe the following effective but clumsy device may be used: Upon the end threaded according to direction, screw three-eighths inch pipe of any desired length to form the rod. 5. Sharpen the open edge of the inch pipe and fit into it a plug with a shoulder that fits against the rim, allowing the plug to penetrate a half inch into the open end of the inch pipe. 6. Sharpen the end of the plug opposite the shoulder and bore a hole lengthwise through the plug. 7. Pass a three-sixteenths inch iron rod through the plug from the shoulder end and bolt it by screwing a nut upon the end opposite the shoulder, which end should be sharpened so as to more easily penetrate the marl. The end of the rod may be threaded for several inches and a nut first screwed on, then the end of the rod passed through the plug and the nut on the end screwed tight against the plug. This will hold the plug from being shoved up the rod by the force of the thrust against the marl, and the nut on the end will prevent the plug being pulled from the rod. The rod with the plug securely fastened on the end is then in- serted in the open end of the cylinder formed by the inch pipe and is passed up through that and the three-eighths inch pipe which has already been screwed to the upper end of the inch pipe. The free end of the rod may project through the pipe at the upper end. 12 MARL. When placed in the water the apparatus is in the form of a long rod of three-eighths inch piping, at the lower end of which is a cylinder of inch pipe. The lower end of this is closed by the plug which fits easily against the lower end of the cylinder by the shoulder already described. This plug is manipulated by means of the iron rod to which it is firmly bolted, which runs up through the hollow rod to the operator above. Method of Operating. The plug is first held firmly against the mouth of the cylinder by means of the rod. The whole apparatus is then shoved down the desired length. The pipe is then raised, the rod being held stationary and after raising the rod is then shoved down to its former level, being shoved tightly against the shoulder of the plug. In this position both are then raised to the surface, the plug shoved out by means of the rod, and the sample taken from the cylinder. This takes a perfect sample to a depth of about 18 feet and can be rigged in a short time at any good hardware store. It is cumber- some on account of handling the long iron rod, but is perfect and very trustworthy for any marl not too solid to be penetrated by this means. The plug keeps all grass, roots, silt and foreign matter from the cylinder while it travels downward, and after the sample is taken, the plug being again shoved against the mouth of the cylinder, excludes all foreign matter during the ascent of the sample. A slot could be devised to close the mouth of the cylinder, and divide it into two halves. This could be made to rotate when the cylinder was at the desired depth and allow the marl to enter, and then being rotated half around again, could close the orifice while the rod ascended. This is not a contrivance that could be fitted out in a few minutes, but when once made would be much less cumber- some as dispensing with the iron rod. This apparatus is easily made and can be relied upon to give perfectly satisfactory results. One other must be mentioned and that is one invented and manu- factured by Robert G. Hunt & Co., Chicago, 111. It consists of a piece of steel about 18 feet long and much the shape of the half of a long gun barrel slit longitudinally. The end which first enters the marl is capped and pointed with steel so that it will penetrate more easily, and the other is surmounted with a handle for raising. The two edges running lengthwise are sharp so as to cut the marl. THE USE OF MARL FOR CEMENT MANUFACTURE . 13 When the instrument has been shoved to the depth desired it is turned half around, filling it with a clean swath of marl its whole length. When it is withdrawn there is a perfect sample of the bed from top to bottom and any portion of it can be sampled if desired. It is not suitable for liquid marl as the sample would run out be- fore the apparatus could be raised to the surface. A device for very fluid marls will be found described in the account of the operations at Cloverdale. § 4. Location of marl. With a general idea of its location and the means at hand for sounding, the question next presents itself, “Where is it most likely to be found?” As has been said, marl is found in the hollows or glacial valleys that scar all parts of our State. Its more definite location is a puzzling and interesting study. The facts so far ascertained will here be given, but the theory of its origin which they seem to sustain will be given in Chapter IV. 1. Marl is always found in some place that was originally covered with water.* The water level of Michigan has fallen within recent years so that the old water lines of lakes can be easily traced even by the casual observer. The marl therefore is not confined to the immediate vicinity of present existing bodies of water. It under- lies dried up swamps sometimes a thousand acres in extent and the banks of what now appear small streams are solid marl. However, ^Somewhat similar subaerial deposits are known as calcareous tufa or traver- tine. L. i , , 14 MARL. upon noticing the comparative depth and altered course of such a stream, and its source when the low lands through which it flowed were all covered with water, its banks often prove to have once been the bottom of a large channel or lagoon. Often, also, a large swamp can be easily identified as the bottom of a large dried up lake. 2. The next point of interest is that the water above marl is Fig. 3. — Map of Cloverdale district. By a mistake the r in Guernsey is omitted. Sec. 18. The lake on Sec. 22, is properly Balker or Horseshoe Lake, the name given being an error in the county atlas. usually hard, containing first of all calcium and magnesium car- bonates in fairly large proportions.* In the Cloverdale region (Fig. 3, Chap. VI, Sec. 2) the general observation of people living about the lakes w r as that the rharl is found in a hard water lake, but not in one containing soft water. A half day’s sounding in Mud or Round Lake yielded but one sample of very organic marl a few feet in depth and lying in mid-lake under ten feet of silt. This was a lake with very soft water while the water of one a few^ hun- *See analysis of water in description of Peninsular plant. L. THE USE OF MAUL FOR CEMENT MANUFACTURE. 15 dred feet from it, which contained 20 to 30 feet of marl, was hard. Little Lake (Chap. V, Sec. 7) contained nothing but silt and did not even respond to the hard water test. 3. A fact closely connected with the foregoing is that hard water springs are everywhere found in close connection with marl lakes. One striking example of the converse of this fact was noticed at Escanaba. There both springs and marl were said to be absent, and flowing wells were tapped only at great depths, though the district was solid limestone. 4. The presence of w r ater and its hardness both being somewhat related to the presence or absence of marl another closely related and interesting study is the comparative level of marl lakes and those lakes or depressions in which marl is absent. As there could be found no reliable contour maps showing the levels of different points in Michigan an aneroid barometer was tried, but it was found that only those lakes contiguous to each other could be at all accurately compared. The results of these comparisons agreed very well and served in the end to establish a somewhat general rule, that of two depressions, the one most deeply indenting the surface of the land, will contain the marl. It is also true that the deeper depression will contain the harder water, provided it cuts the deeper water bearing strata of subsoil. This conclusion was very often verified in the hilly country where the surface is deeply cut by streams and lakes. It is also quite generally the rule in comparing adjacent marshes for the presence of marl. Still it must be considered dangerous to conclude that a deep depression always forms the basin for the hard water bearing strata about it, as these same strata may slant away from, rather than toward such a basin. No general rule can be formed which will guide the prospector unerringly to the presence of marl. As the marl is in nearly every case covered by finely deposited sediment, muck, other marsh growth, or water, its exact location and depth can be determined only by actual soundings made through its covering. Still the guides here given have proved rather useful in the absence of any other helps whatever, and as simple results of experience must not be taken as fixed rules. 5. In a chain of lakes the marl is generally deeper and of better quality in the lakes toward the head of the chain. Where the head lake lias had no large body of water or stream other than a spring 16 MAUL. stream opening into it, the marl appears of purer quality and with less of foreign matter overlying it than do any of the lakes below it. In two of the chains of lakes so noticed the head lake formed the first of the series and so lay that there never could have been any other natural drainage than the one in action at the present time. This rule works well in the case of a series of two lakes, the upper one of which is fed by springs. The marl lies bare and of greater depth to the upper end of the upper lake, and sediment above the marl, if it occurs in large quantity, is liable to be in evidence tow r ard the lower end. 6. In a large lake or one unevenly and thinly underlain wdth marl the deepest marl is often found in bayous or indentations of the shore-line. In such cases the marl generally thins very rapidly to the deeper portions of the lake.* § 5. The distribution of marl in a single bed. In the consideration of this subject much depends upon the stage of the deposition in which the bed to be sounded exists at the time. For marl beds exist in two different states. The first is the dried up lake or marsh, the second, the hard water lake wdiere the marl is still depositing. In the old lake bed or marsh the marl deposit is basin shaped. It generally has one or more centers toward which the marl deepens regularly. The marl has evidently deposited as long as water remained. As the marl reached the surface or the w r ater dried down to the surface of the marl or both, vegetation started upon the shallow r s and sealed the deposit over very evenly. Where the lake bottom proper was even in the first place, the marl deposit is very regular. This w 7 as shown in some old dried up lake beds of the Upper Peninsula where the deposit was laid down evenly, in- creasing from the outside edge twm feet in depth, to the center twenty-five feet. It is the rule and not the exception in marshes entirely covered by vegetation and containing no open w r ater, but underlain w T ith marl. It must be remembered that many of our inland lakes and marshes have their bottoms, uneven in their na- ture, cut and seamed with terraces, kettles, and holes left by reced- ing glaciers. In the evening or blanketing process of marl deposit these holes are leveled over. In such cases the depth of marl can be calculated with only general accuracy and the above rule can scarcely be verified. A fair illustration of even deposit w r ould be that at Central Lake in Antrim County, Plate I. See also Chapter *See description of Onekama Lake. Geological Survey of Micliigan. Vol. VIII. Part III. Plate] L MARL SOUNDINGS, 1, 2, 3, 4, 11A, 11C. y?' ! ' ■ ■ THE USE OF MAUL FOE CEMENT MANUFACTURE. IT VI, §16. An illustration of uneven deposit or better uneven depth caused by sudden variation of contour of lake bottom would be the lakes sounded at Cloverdale (Plate I, Diagrams 1 to 4). It is very often the case, however, that the marl bed does not cover the whole depression formed by the original lake bed or by the marsh as it appears at the present day. In such a case the main body of the marl forms a basin of its own which is liable to lie as at Portage Lake, Onekama (Fig. 14), in an indentation or nook of the greater basin forming the marsh. It may or may not lie near the deepest portion of the original basin. The sealed marl bed is on the whole the more regular in its increase and decrease in depth, and is, excepting in the case or exception of an uneven original bottom, regularly deepest toward the center of the deposit. In the lake where the deposit is still continuing or just being discontinued, the variation in depth is markedly the opposite. In this condition the lake is nearly always surrounded by a fringe of shallows containing the deepest and purest marl. In deep water the marl may be much shallower, may cease entirely or may be a marly muck, the first and third named conditions prevailing in nearly all cases in deep water. In studying this second condition it is found that marl forms most rapidly in shallows or about points. This was strikingly illus- trated at Long Lake, near Cloverdale. This was being cut into two different bodies of water, by decrease in depth of water and at same time by rapid growth of marl, which was 33 feet in depth in the narrows at Ackers Point. Thus Horseshoe or Balker Lake was being cut into two lobes or basins and the marl was very deep at the narrows connecting and on the points of marl growing out to sep- arate the lobes. Nearly every actively growing marl lake represents three stages or steps of growth, the shore or marsh of marl bed grown to water level and sealed over by marsh growth, the actively depositing marl of the fringing shallows, and the deeper parts which are more slowly filling up with a cruder and more impure marl. Eventually the fringe of shallows will grow to the surface or far enough for rushes to catch organic matter to form a solid cover- ing of growth. As the center of the lake grows shallower with in- creased depth of marl the marl becomes whiter and deposition more active, the marl fills to water level and is sealed like the former fringe of shallows. We have then from the second condition a growth to the first condition, a completed and preserved marl bed. 3-Pt. Ill 18 MARL. When the water was higher during the deposit of the shallows marl* the shore marl will have deposited to a higher level than that in mid-lake. When sounded w r e often say that the “surface” is deepest at the center, when in reality the marl was deposited to a second lower water level and then filled in with marsh growth to nearly the level of the shore fringe of shallows. In such a case it is noted that the shallow marl is of much finer quality because it was deposited in shallow water while the marl in mid-lake w T as de- posited in deep water, and this latter was suddenly brought to the surface by a fall of water level and covered with an organic blanket preventing a finer deposit. In deposits of the second or uncompleted condition the gradation in quality, due to the variation in content of organic matter, is often very marked and seldom absent. The shore shallows unless very deep deposits, are the purest, then as soundings are made toward the center the bed decreases in thickness and the marl de- creases in quality, organic matter steadily increasing at the expense of the calcium carbonate. Below are given a table of soundings in lakes about Cloverdale, pages 78, 79, 80 and table of analyses made of samples taken, page 80. On page 82 is a list or key to all the samples of marl elsewhere taken, of which analyses were made. On page 83 are the partial analyses of these samples. It will be seen by consulting the table that several samples are marked A and B. The samples marked A were taken near the bottom of the bed and those marked B were taken at the surface of the bed directly over the first sample marked A. Owing to the fact that the deposits were very heavily adulterated with clay and sand it is difficult to compare for increase of organic matter. TABLE OF SOUNDINGS, CLOVERDALE DISTRICT. No. Analy- sis. Location of Sounding. Depth of water. Depth of marl. Bottom. Long Lake. Tamaracklogon bottom. 1 1A.... In Narrows at Ackers Point 2 30 Gravelly sand. 2 IB.... Surface of above 3 2A lOo yards east of No. 1 4 17 4 Same as above 10 3 ft. into fine sharp sand. 5 2B .... Same as above. Sampled at surface. 6 200 yards east of No. 3 24 10 7 Sand of No. 6. Not preserved 8 3A Shallows 200 yards southeast of No. 6. 4 30 Very fine sand. 9 3B Surface of No. 8 *As for instance at Cedar Lake in Montcalm County. L. No. 10 11 12 13 14 15 16 17 18 21 22 23 24 25 26 27 28 .29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 2 3 4 5 6 7 8 USE OF MABL FOR CEMENT MANUFACTURE. 19 TABLE OF SOUNDINGS, CLOVERDALE DISURICT —Continued. Location of Sounding. About center of lake Surface of above At lower narrows East of rocky islet Beyond No. 12 toward outlet Surface of No. 14 To side toward springs. Marl sandy . In narrows Surfaceof No. 17 not preserved Muck sample of No. 6 taken near here — Northwest of rocky islet Fishing hole Just outside of narrows north of Ack- ers Point Half way between Nos. 22 and 23 200 yards south of No. 23 At point of lake opposite Cloverdale. Opposite springs issuing beneath blue clay at end Just below boxed spring (x) 20 feet out from No. 28 Just west of Cloverdale, south side. . . Opposite Beechwood Point Shallows in toward Beechwood Point Balker or Horseshoe Lake , Lobe Next to Outlet. In front of narrows at lower end In narrows Shore opposite landing Up lake on slight point Just outside of No. 36 in deep water. . In straight line across slight neck to south shore Sample can No. 5 of water taken over No. 40 Deepest sounding made. Brought up trailing water plant which had pow- erful odor of polecat* At mouth of large boiling spring 200 feet toward outlet from point on south shore marking previous line of soundings. No. 4 collected at this spring 200 feet west of 41 at the end of series of soundings across lake Toward upper end of lake from No. 42 At outlet of lake. Jar No. 9 taken at surface Center of basin. Jar of water No. 6. . In inlet from other lobe, forming nar- rows Guernsey Lake. Blue clay flats at narrows In west channel or arm of south lobe. West shore of shallows Surface of same 30 feet out from 49 Surface of No. 51 100 yards south or up from No. 51 Surface of No. 53 Bottom of Mud Lake Pine Lake. Cove of landing at lower end In front of boiling spring on opposite side In outlet First line across narrows Out from No. 4 In line across In line across At farther side Depth of water. Depth of marl. Bottom. 25 20 Heavy gravel. 6 12 4 9 3 33 Heavy gravel. 2 2 2 10 Pepper and salt sand. 4 23 16 31 2 y 2 17 Fine sand. 2 17 4 25 3 25 3 25 3 18 18 27 4 23 3 23 2 17 2 23 2 30 3 27 Gravel. 3 29 13 15 50 10 muck. 0 32 Marl dark blue. 2 37 2 27 y z 32 10 30 2 4 muck. 8 30 6 in. 27 5 24 Sand bottom. 4 20 Sand. 33 5 3 20 6 in. 10 6 in. 10 3 15 9 9 2 6 6 in. 12 (?). L. 20 MARL. ANALYSES OF CLOVERDALE SAMPLES. Number. Insoluble in HC1. ALO3. Fe2C>3. CaC0 3 . MgCOg- Organic Matter. 1A 3.34 2.55 84.30 3.18 6.63 IB 2.35 2.94 82.11 2.64 9.96 2A ....... . 2.95 .04 85.00 4.62 6.39 2B 1.84 2.00 81.00 10.21 4.95 3A 20.54 2.30 67.53 3.48 6.15 3B 75.04 1.90 14.02 6.05 2.99 4 11.70 3.92 69.30 3.17 11.91 5B 14.15 1 18 75.15 6 32.32 4.62 42.14 2.91 18.01 7 13.04 2.70 40.00 9 15.14 13.73 43.13 1.66 26.34 10 4.64 2.00 65.09 3.28 24.99 11A 57.04 4.30 22.06 12.45 4.15 12B 7.20 1.25 64.12 2.38 24.05 13 A 55.10 6.80 25.28 3.10 9.74 13B 13.70 2.24 65.00 2.72 16.34 14 A 61.10 4.50 21.00 11.76 1.64 14B 1.44 .90 84.50 4.19 8.97 15A 41.94 3.80 47.23 3.79 3.24 15B 65.64 5.35 16.60 9.53 2.88 16 11.97 1.45 75.62 3.44 7.52 17 1.80 .80 83.00 2.38 12.02 18 A 5.04 1.84 74.46 2.31 16.35 18B 3.06 .95 85.04 4.20 6.75 19A D.34 2.00 86.43 2.42 7.81 19B 2.04 1.84 84.46 2.83 2.83 20A 3.55 3.50 80.18 3.33 9.44 20B ....... . 1.24 1.60 88.30 3.03 5.83 21 9.70 10.90 71.00 1.92 6.48 Kent 16 .14 .73 90.30 3.21 4.82 Remarks. Long Lake. Sand and gravel. Mg. precipitated. Balker Lake. Balker Lake, t Mostly clay. | Guernsey Lake. Clay and sand. Clay. Checked volumetrically Pine Lake. For position and depth of samples above analyzed see preceding table of soundings of Cloverdale region. LOCATIONS OF SAMPLES COLLECTED FROM DIFFERENT PARTS OF THE STATE BY D. J. HALE AND ANALYZED BY A. N. CLARK. No. 4. Marl from Big White Fish Lake. Springs emptying near contain iron and sulphur. 6. Marl of Lime Lake, 17 feet below surface of bed. 7. Shell marl at surface of same bed (Lime Lake). At first pure white, it turns brownish red upon exposure to the air. 12. Marl near spring at Corinne. Very hard and difficult to bore in with ordi- nary augur. 13. Marl at Wetmore in bottom of boiling spring. 18. Central Lake. Head of lake. Deep sounding. 19. South end of lake, 27 feet deep. Below level of bed. 21. In channel S. E. of S. Island, Central Lake. (Intermediate Lake.) 25. Center of Mound Spring. 26. N. side of Mound Spring. 28. 10 feet below the surface of Mound Spring, Central Lake. 29. Clay on Clout’s farm, Central Lake. 30. Low clay west side of Central Lake. 31. Mixed strata of clay in brickyard at Central Lake. 32. 33 and 34 are three depths of clay on a side hill near Central Lake. 32. The highest layer consisting of broken down shale or clay. 33. Shale below 32. 34. Lowest shale. 37. Black shale from a sixty foot shaft west of E. Jordan 4 or 5 miles. Shaft was mined without success for coal. 38. Green shale lower in level than black shale of sample 37. 27. Iron from north side of Mcund Spring, Central Lake. THE USE OF MARL FOR CEMENT MANUFACTURE. 21 PARTIAL ANALYSES OF SOME OF THE SAMPLES COLLECTED FROM DIFFERENT PARTS OF THE STATE BY D. J. HALE AND ANALYZED BY A. N. CLARK. Number. CciCOg. MgCOn. Fe2C>3. ai 2 o 3 . Insoluble. Remarks. 4 23.57 1.89 6.75 56.95 Red sandy marl. 6 92.00 0.57 Very white. Ferrous iron, 70%. 7 90.00 .30 Brown on exposure to air. Fe 0.72. 12 76.07 1.59 1.00 19.00 White. 13 90.71 1.51 1.60 4.50 Cream color. 18 42.32 2.04 19 57.32 1.51 21 90.90 1.59 25 32.76 1.89 26 27.32 0.53 28 1.09 1.73 29 .18 1.05 30 4.82 1.51 31 20.18 1.40 32 .36 1.66 33 .44 .98 34 1.25 1.89 37 3.21 1.96 38 2.00 2.42 27 85.00 1.13 5.65 Red shale. Above samples were analyzed by acid solution the same as for marls and limestones. The method gives too low results for CaO where samples consist mostly of clay.— A N. Clark. Upon comparison of the eight double samples marked A and B, page 20, it will be found that five of them show an increase of organic matter of the deep soundings over the sur- face soundings.* If the surface samples are taken directly at the surface they are liable to show greater organic matter due to the presence of roots of grass and rushes in shallow water, as these generally form a thick somewhat impervious mat over the surface of a bed of shallows. If soundings or samples 6 and 7 above are compared, it will be seen that both are high in calcium carbon- ate and that the surface marl is more impure than that at a depth of 17 feet. This was a shell marl bed and may form an exception to the rule that the deepest marl has less of carbonate and more of organic matter. This rule cannot be too much emphasized as it forms one of the few guides in the examination of a typical marl bed. If the marl all lies under deep water it will vary little in con- tent of organic matter being nearly the same at the bottom as at the top. Such was the case in the bed at Bice Lake, which con- tained a bed 35 feet in depth and yet the marl did not vary as much as in many smaller lakes, remaining about the same at the bottom as at the top. On the other hand a chain of small lakes was examined which had marl in the shore shallows 30 feet in depth. It *As also in analyses furnished by Michigan Portland Cement Co. L. 22 MAUL. was nearly pure on top, but at bottom was scarcely more than a muck. In lakes where the deposit of marl is continuing at present or has only recently ceased, the conditions governing its location are highly interesting. In such cases it appears to have covered the lake bottom evenly like a sediment, but with this difference, that it is a sediment that fills in and helps very much to do away with inequalities in an uneven lake bottom. This was very strikingly illustrated in the series of soundings at Cloverdale given above. (See also Chap. V, Sec. 2.) In a marl lake which is depositing at the present time there will be seen little if any black sediment. The common river or lake alluvium or sediment that will naturally accumulate is sur- rounded by the white particles of marl and forms a part of the marl bed, but of course loses its dark color, becoming light in color like the remainder of the bed. Twigs, limbs of trees that fall into the water, the water plants themselves that die and would naturally become black and so color the bottom, are surrounded by the white marl particles and are transformed into a part of the bed. When this process is in active operation the bottom of the lake shallows is perfectly white from the transforming power of the forming marl. The prospector can readily trace out the point at which this process has ceased by the presence again of sediment on the lake bottom, giving it its customary black color. And this symp- tom is a satisfactory and sure index to the variation of the marl bed. Where sediment has begun to form, instead of being coated by marl, the marl will decrease in depth beneath as the sediment increases in depth above, and where there is any great depth of sediment above there will be found little marl beneath it.* The position of this lake sediment must, however, be thoroughly understood. It lies under the water and above the marl and when it begins to cover the marl, it is pretty good evidence that the bed has ceased growing. When the bed on the other hand, from any cause such as the fall bf the water level to the surface of the bed or the growth of the bed up near to the surface of the water, gives marsh growths, etc., a chance to form on the surface of the bed, growth will stop and the bed will become sealed over and forever afterward will be a part of a marsh or dry lake bed, assuming at once the condition spoken of under dry lake beds. •This is illustrated also by soundings near Riverdale, Gratiot Co. L. THE USE OF MARL FOR CEMENT MANUFACTURE. 23 It lias been already stated that the edges of lakes where marl is at present forming contain the deepest marl. It is true that the rule in regard to these lakes is decrease of depth toward the center, for the marl is not at the present day forming as well in deep water as in shallow. Its quality toward deep water decreases by virtue of increase in per cent of organic matter. This seems a reliable rule with few exceptions and has been found so true as to be depended on in almost every case. The marl rapidly deteriorates till in very deep water it becomes little more than a mucky marl or perhaps a bog iron. The marl at this depth exists in a fine state of suspension and could be taken only with an instrument so tight as to hold water as well as marl. A sample was taken in fifty feet of water while at the sides of the lake a few hundred feet distant there were twenty-five to thirty feet of fair marl. In this short distance with sudden increase to a great depth the marl has be- come almost a muck losing the characteristic light color of marl. The presence of springs, while characteristic of a marl region,, has nothing to do with the depth of marl at a given spot. Though the presence of hard water in and about a marl lake is expected as a rule and may generally be calculated upon, a spring is no guide whatever to the location of the deepest marl. One spring will be found to bubble through marl many feet deep while another spring in the same lake and containing water fully as hard is as likely to be surrounded for any distance by pure sand, or to issue from the ground through pure lake silt or through muck. If anything, the balance of instances is against marl near springs as they, if not in the lake issuing from marl, start small rivulets of water which, as the outlets of the springs, bring down a slight drift of sand or other foreign matter. In highly charged hot water mineral springs such as may be easily found in the Rocky Mountains or in Europe, the minerals contained in the water are upon its arrival at the sur- face immediately released from solution and thrown down as a deposit at the mouth of the spring. The method of deposition and the location of such deposit is obvious. Our springs certainly do not discharge their burden of lime immediately and therefore give no sure clue to the manner in which the deposit is brought about or to the whereabouts of the marl deposit. § 6. Surroundings of marl. It is of the greatest practical importance to the prospector to note carefully the surroundings of marl. In the definition and 24 MARL. identifications of marl attention was called to the immense varia- tion in appearance and chemical constitution of marl brought about by the impurities with which it often becomes contaminated. In the surroundings of marl the prospector must always seek the direct source of these impurities and it is for this reason that the location of marl in relation to its surroundings must always be carefully noted. (a) Shore wash. Marl can never be considered as a deposit oc- cupying very large single areas as do many other minerals. It is confined to those depressions which have once formed lakes and are now lakes or marshes. It then fills a pocket or hollow of the above description and is directly subject to the natural forces that act upon the hills and banks forming the rim of the depression which is nearly always the shore line of the lake. If the indenta- tion is deep the shore line will be a bluff of clay or sand. If the marl and the water which must have originally covered it extend up under or close to the commanding bluff, the action of rain or run- ning water can be nearly always traced in surface wash upon the marl, for gravity will then bring down upon the marl which is in process of formation, large quantities of sand or clay, depending upon whether the bluff is sand or clay. The presence of sand may still be expected even when the banks of the lake are very low, pro- viding the deposit of marl is very deep. The reason is this. If we consider that a marl bed 30 feet in depth is stripped from a lake we have a valley originally thirty feet deeper than the one which lies before us filled with 30 feet of marl. Still in case of very low banks the deep marl bed was always covered by water and the slant of the bank alone will do much to govern the amount of sand or clay washed down upon the bed. In such cases, if a deep marl bed ter- minates abruptly at the foot of its bank or bluff, the deposit will be found to be thoroughly mixed with the wash of the overhanging bank. Soundings in such cases reveal a layering of sand then a layer of marl and then of sand and so on to the bottom. Or it may appear from the shore to some little distance out that there is nothing but sand. Upon sounding it is found that the sand and gravel from the shore have swept down and over the marl com- pletely covering it for some distance out, the marl in some cases being found to terminate very abruptly against a steep bank and underneath a covering sheath of sand. This is the immediate and very local effect of the banks or shores of the lake upon the appear- THE USE OF MARL FOR CEMENT MANUFACTURE. 25 ance and constitution of the marl bed about its edge, but in such cases the marl is rather thoroughly mixed with sand or gravel some distance out and is entirely unfit for manufacture. A lake with steep banks or with marl lying close under low banks must be watched closely for local traces of mixing. Long Lake at Cloverdale is an example of this. (b) Streams. The next contaminating agent is running water. A stream running through a sandy or clay ravine upon a bed of marl in a lake can generally be traced for some distance by the presence of sand, muck, silt and other foreign substance in the marl. In many cases the formation of marl seems to have been prevented entirely. But on the other hand the course of streams changes rapidly, as does the drainage of many lakes, so that a stream is often found flowing over marl which has been already formed, very likely before the stream existed at that point. If a stream coming from another lake flows over marl all the way and comes from a marl lake its evil effects as a sand bearer are very slight. If it comes with considerable force from a sandy region and has been rather permanent it produces a large patch of sand for some distance about the inlet, and there is an entire absence of marl. Small rivulets and ditches formed or dug in recent years across a marl bed, carry in their path large amounts of sand and even gravel, which sometimes render the marl unfit for use. They should be watched with great care by the prospector to see that they do not bring impurities in quantities sufficient to destroy the value of the marl. Marl when once formed in a rather solid bed is not easily penetrated by sand bearing waters. Springs w r hich bub- ble up through marl beds from a sandy bottom beneath do not often cause the sand to permeate the bed in large amounts. The sand brought by streams flowing over established beds does not penetrate the bed to any great depth provided such a bed is rather solid. If, however, the sand bearing agent, such as a stream or wash from hills, has been at work layering or steadily mixing with the bed at all depths during its formation, the bed will be found to be mixed with the adulterating sand or gravel for long distances, sometimes completely destroying the commercial value of the deposit. (c) Surface. This is a name used to designate the covering of the marl, whatever that may be. The first covering of marl has always been water. It is formed under water and it is necessary 4-Pt. Ill 26 MARL. as long as it grows that it be destitute of all other covering. Ex- ception or explanation must accompany this statement. The nat- ural clothing of a marl which is in active growth is generally a characteristic water plant* growing on the marl. This covers large areas in the usual marl shallows, and is seldom found lacking in an actively growing bed. It is small, lying close to the bed, reclines and almost trails and has very bare branches which issue from the stem in whorls or circles completely surrounding the par- ent stem. These plants, together with all other objects not posses- sing the power of motion, are thickly covered with a whitish coat- ing of the marl. This condition of active formation of marl ceases when the shallows approach the surface of the water so closely that rushes and marsh growth of all kinds can obtain a foothold on the marl as a soil. The marl then becomes coated with a surface of muck and marl deposit ceases. When the marl rapidly dries out there remains but a thin coating of marsh growth which may re- main as only a few inches of soil surmounted by ordinary marsh grass. There is then practically no surface or one which may be easily removed by the dredge. If, on the other hand, the surface next to the marl remains very wet, it is conducive to a very luxuri- ant marsh growth which may sometimes consist of from two to seven feet of loose roots and rushes. Sometimes a thick wood has sprung up on the bed consisting of trees of large size or a very thick tangle of underbrush. This means a tough surface of roots to be removed before the marl can be used and its nature should be carefully noted by the prospector. This is one way in which a marl bed is covered and gets its surface. (d) Silt under water. When the conditions have become un- favorable for further formation of marl, the silt which is constantly deposited from lake water, ceases to be enveloped by the particles of marl an ,5. falls upon the bed, making a dark covering. This sometimes forms over a bed or a part of it and the growth then ceases. In time the deposit of silt reaches the surface of the water or the water sinks to that of the silt. In either case the silt is exposed so that marsh growth gets foothold and seals the deposit as before. One marl bed was found where there were layers of this silt with its attendant marsh growth intervening between layers of marl. The rule is in nearly every instance, however, that when sediment of the nature of silt or marsh growth is found be- ♦Chara, see chapter by C. A. Davis. L. THE USE OF MABL FOB CEMENT MANUFACTUBE. 27 neath the marl such a layer is an indication that the bottom of the bed has been reached. (e) Lining of marsh growth or decayed plant life. It is true that pure lake sediment often smothers and seals the growth of a marl, even when the bed is covered with many feet of water. There is another very interesting phenomenon which the prospector notices when he has penetrated often to the bottom of some beds. Just before the sounding apparatus penetrates the sand or clay under- lying the bed, it passes through a thin layer of nearly pure organic matter which seems to be the finely compressed and decomposed residue of plant life. It is green or blue in color, fine in texture, and it forms a very sharply defined layer a few inches thick. It lies just under the marl between the marl and sand forming an organic lining. It contains some lime and does not effervesce very freely with acids. This layer was noticed in several rather deep deposits.* (f) Organic matter permeating deposits. Remains of plant life always form a characteristic part of a marl bed, but the prospector will find a more or less sharp distinction between two kinds associa- ted with the bed: (A) . Organic matter of the marl deposit. This organic content of the marl bed varies with the depth of the bed and the depth of the water above the marl. It is as much a part of the marl bed as is the content of lime. It can be depended upon that this content of plant remains will increase in two ways; first, from the shallows toward the center of a lake or marsh, and second, from the surface of a thick deposit toward the bottom of that deposit. This is one of the rules with fewest exceptions and will serve as one of the best practical guides to the prospector. This rule works, of course, only in the absence of outside influences; i. e., when the composition of the bed is not interfered with by drainage, water streams, etc. The consideration of this leads di- rectly to that of (B) . Organic matter of drainage. When a stream brings in much silt or drift of any kind the condi- tions favoring the deposit of marl cease to exist, then a heavy ad- mixture of organic matter follows with no fixed rule by which to judge it except perhaps direction and force of the water which may empty upon the deposit. In the majority of such cases the dividing line between marl and foreign matter is sharp enough so ♦Compare what is said about Schizothrix. L. 28 MABL. that the area of the marl can be fairly outlined. Yet this some- times varies and the influence of the foreign organic matter is felt for a varying distance into the body of the marl deposit. (g) Materials underlying marl. The various soils which surround and influence the quality of the marl bed have been described and it now remains to describe the substratum or foundation upon which the marl lies. The thin layer of organic matter which often forms the lining of the bed has already been described. In Rice Lake, at the bottom of thirty-five feet of marl, a thin layer of pure organic matter lay under the marl and rested upon sand. This layer was pierced in nearly all soundings in the lake, where bottom was struck. Marl never lies on muck or organic matter of any great depth. The usual foundation is sand or clay. The majority of beds lying upon clay seem to indicate that the marl is a distinct deposit differ- ing from clay and that if the clay is mixed with marl to any extent it is due to a sedimentary deposit of the clay by water flowing off of some adjacent clay bed. Instances were seen where, in this way, clay of a highly magnesian composition was freely mixed with the marl deposit. In most cases there is, however, a sharp line of division between marl and clay. Perhaps the most characteristic material which forms the final basis of marl deposits is sand. This is in nearly every case a fine quartz sand which may be mixed with fine grains of mica, forming a pepper and salt sand.” If the deposit of marl is lined with the above described layer of organic matter the material, whether it be sand or clay forming the basis or foundation for the marl, does not work into the deposit affecting the uniformity of its quality. If coarse gravel takes the place of the fine sand bottom it indicates the former presence of flowing water and foreign matter of all kinds so far described must be watched for by the prospector. A fact in this connection is noteworthy. This is that an amount running from 1$ to 3$ of fine quartz sand is fairly well distributed through most deposits of marl. This seems to be strictly separate from the ordinary surface washings of coarse sand. In one case, Onekama Lake, the sand of this special kind was nearly absent. But on the bottom of the deposit and in some cases at intervals to- ward the surface, there were thin layers of half decayed organic matter. In some cases the wood, at 10 or 15 feet beneath the marl, was well preserved so that the fibre could be split. In one instance THE USE OF MARL FOR CEMENT MANUFACTURE. 29 at Ackers Point, Cloverdale, a well preserved tamarack log was struck at the depth of thirty feet. It lay on the bottom of the true lake bed as a sounding near by showed sand at the same depth. (h) Materials overlying marl. It is difficult to judge of the age of a marl bed by its covering. Large areas of marl are covered by marsh. This generally is in a semi-fluid condition so that it jars with the tread for yards around. It will remain in such a condi- tion as long as the water level allows the water to stand within a foot or so of the surface of the marl. In such a condition the marsh growth of rushes and their roots grow rapidly and surface soil, etc., is caught making a spongy growth of sometimes five to eight feet or more in thickness. Solid ground may form over it and the pres- ence of marl be unsuspected. When, however, a shaft is sunk or railroad spiles are driven, or a grading put on the ground, the pres- ence of marl is attested. The spiles sink suddenly or the grading sinks, or the shaft is suddenly filled with mud. All these things have occurred, showing that the marl is often buried deeply. In such cases the water at the level of the marl kept it fluid all the time. A surfacing of marsh growth develops rapidly and leads one to think the marl very old on account of the thickness of the sur- face upon it. On the other hand, if water level sinks, the marl dries out and no luxuriant vegetation grows excepting the ordinary marsh grass. There are hundreds of acres of this that may be very much older than that covered by marsh growth. At Rice Lake there was a marsh growth of from two to six feet which must have been largely grown since the lake bottom was drained but a few years ago. § 7. Method of prospecting a given area. The general rules for the location of marl beds by the prospector have been given. Also those more particular laws which will as- sist in judging of the probable effect on the bed of foreign materials surrounding, above, around, and beneath the bed. After a marl bed is located in a valley or depression, either as a lake or marsh or combination of both, the next step is to estimate the area and depth. If a surveyor’s outfit is to be had the work can of course be performed with unquestioned accuracy. Lines may be run measured distances and at right angles to these, another set of measured lines making of the marl a checker board of squares the intersecting lines of which should be fifty to one hundred feet apart. At these intersections soundings could be made and the 30 MABL. depth noted upon the plat of the bed. As it very often happens the prospector does not carry surveying or measuring instruments, a practical and at the same time accurate method of testing the bed must be devised, as follows: (1) A record of everything done must be kept and this record must be made as soon as each fact is ascertained, not trusting to the memory any detail. (2) Soundings must be made as nearly as possible in straight lines with the lines parallel to each other. (3) The best and most permanent marks by which to locate the soundings made and the work done are the section lines and bound- ary lines of landowners. A bed can usually be located as included within boundaries of a quarter section or of a forty or eighty acre plat of ground. (4) To measure the distance between soundings, the woodman’s method of pacing the ground can be resorted to. Soundings should be made at first not over fifty feet apart, but if the deposit, after many soundings, is found to be very regular in both depth and quality, the distance may be increased to one hundred or two hun- dred feet, care being taken to at once decrease the distance between soundings upon the slightest signs of change of quality or sudden unevenness in depth. (5) If soundings are entirely upon land the distance is more easily calculated, but if on water and in summer it is more difficult to determine accurately. In making deep soundings in shallows on water it is safest to use boats. A rough frame or planking will serve to bind the boats together and the soundings may be made between the boats, the operators standing upon the cross planks as nearly as possible at the center. It is often found possible, where the bed is not very thick, for the soundings to be made with augur and pipe from one boat, the boat being rocked by the persons in it to exert a leverage on the side in raising the pipe. This sometimes fails, resulting in inability to raise the pipe, which becomes stuck in the marl. It must always be remembered that small pipe and quick handling make light work. § 8. Commercial importance of composition. It will now be well to consider marl in regard to the manner in which its chemical composition affects its usefulness for factory purposes. In the following treatment the impurities of marl will not be THE USE OF MAEL FOE CEMENT MANUFACTURE. 31 considered but the fairly pure marl only, leaving out sand, clay, and extraneous organic matter. The best marl and that which should most nearly typify marl as an economic deposit lies, we will say, in a small inland lake. It is covered by but a few feet of water and by no silt or foreign matter whatever. It is growing at the present time. It rests on a bed of tine quartz sand, which does not affect its composition to any great extent. The influx of surface waters and drainage streams has not interfered with the purity of the deposit. This lake is fed mainly by hard water springs. Such are the surroundings of a very pure marl when it is in the process of deposition. (1) Appearance. The marl on the shoals or marl flats of such a lake is very white and somewhat granular. The marl near or not quite at the surface is very much the purer and will generally give the higher analysis if it is not mixed with roots of water plants in gathering the sample. We see such high analyses of marl quoted frequently as 95$ to 98$ calcium carbonate. This is all true enough, but repre- sents usually but a small portion of the bed which in reality would average much below such a percentage if the analysis of a sample from the bottom of a thirty foot bed were to be given, even if the deeper sample were to be taken over the same spot as the shallow. Marl of the above location, at the surface of an actively depositing bed is often very granular and even gritty to the touch. Upon a careful examination it will be found that the grit is composed en- tirely of marl and not of sand, as might at first be supposed. The marl is seen gathered into pebbles and has often formed about roots and small objects of every kind. When the root has died and rotted down it often leaves a hollow pebble around which the marl continues to form.* Toward the bottom of a thirty foot deposit, at the same place, few if any of such accretions will be found pres- ent. The marl is at the same time finer grained, more adulterated with organic matter and darker in color. Toward the center of such a bed, in deeper water, the marl is also darker in color at the surface, the concretions disappearing also. It is this reason that single chemical analyses reveal little of the nature of a bed. As far, however, as the exact nature of marl and its identity as distinguished from every other calcareous deposit are concerned, a pure bed as above described serves as the best illustration of one ♦Produced by Schizothrix. See C. A. Davis’ paper. L. 32 MARL . very important fact to the manufacturer, which is that marl as a distinct deposit and free from all contamination varies very much in its own composition, in the same bed at the same spot. (2) Composition. With the foregoing understanding an endeavor will be made to explain the composition of marl. Marl is certainly due to one clearly defined agency. (1) It derives its composition from the carbonates contained in the hard water of the springs. It is not deposited immediately around the springs. A secondary agent in the deposition is the growth of shells, snail shells, bivalves, etc., which have died leaving their shells to more or less increase the depth of the calcareous deposit of marl. In places favorable to their growth, or where they have been sifted from the surrounding marl by wave motion they form a nearly solid bed, while in other places they and their broken down forms are nearly if not entirely absent. The following are the analyses of fairly pure samples of marl from different parts of the State. MARL ANALYSES. Calcium Carbon- ate. Magne- sium Carbon- ate. Ferric Oxide. Alum- ina. Insolu- ble Silica. Soluble Silica. Mois- ture. Organic Matter. Sul- phuric Acid (S0 3 ). Phos- phorius.. 1 74.480 0.50 2.36 0.54 7.20 1.25 12.88 0.89 2 82.142 4.620 0.9775 1.151 16.27 11.173 0.00 0.037 3 89.965 1.672 0.999 0.158 1.222 9.750 5.984 0.00 0.03 4 83.045 1.201 Undete rmined 3.569 11.700 0.485 0.00 5 87.000 0.910 1.30 0.070 0.780 0.130 0.600 9.800 0.270 6 97.000 1.010 1.260 0.08 .30 7 94.496 1.250 0.432 2.528 0.235 0.790 0.00 0.150- 8 92.91 1.89 0.53 0.21 1.54 2.01 0.80 trace. 9 77.1 3.28 ^1.92 9.64 7o.99 10 11 92.1 60.00 3.2 3.00 - .76 0.62 .30 .22 34.60 ^3.7 1 .50 12 .10 .14 1.90 0.64 5.69 .56 .01 13 93.8 0.73 differen cel .17 14 92.79 2.27 A). 52 3.25 15 93.75 2.42 .25 .55 1.01 .18 differen cel.84 trace. 16 91.34 .77 .40 .55 .78 .42 5.79 .26 THE USE OF MAUL FOB CEMENT MANUFACTURE. 33 KEY TO PRECEDING TABLE. 1. Marl from Alpena, Mich., W. E. Courtis, analyst. 2. Marl, Cass City, Michigan, same analyst. 3. Marl, Cass City, Michigan, same analyst. 4. Marl, Grass Lake, Michigan. This sample was dried at 100 degrees centigrade, dry residue 42.11$. Undetermined 3.569. Same analyst. 5. Near Grayling, Michigan. Average sample when dried lost 61$ of its weight. Same analyst. 6. Same sample, but figured without organic matter. 7. From lake shore near Grand Rapids. This sample loses 6.376$ of water and volatile hydrocarbons when heated to 100 degrees centigrade. The silica is not sand. The 0.235 moisture is com- bined water. Phosphorus as tricalcic phosphate. It also contains chlorine as sodium chloride, 0.119$. Same analyst. 8. Marl from Alpena, Michigan. Total 99.87. 9. Grass Lake, Michigan. Was collected by A. C. Lane and analyzed by F. S. Kedzie. 10. Marl at Peninsular Plant, Cement City, Goose Lake, Mich. Total 99.98. 11. Marl at Cedar Lake, Montcalm County. Not dried. Ana- lyzed by F. S. Kedzie. 12. Marl near Grayling, Michigan. M. A. C. bulletin 99; CaO 45.16, MgO 0.32; K 2 0 0.37. Dried marl is 49$ of original weight of sample. 13. This sample is the same marl as 11, but a different sample and figured dry. The marls as taken contained 9.95$ water. F. S. Kedzie analyst. 14. Naubinway marl. World's Fair report, p. 132. 15. Light marl, Michigan Portland Cement Co., H. E. Brown, chemist. 16. Blue marl, Michigan Portland Cement Co., H. E. Brown, chemist. Additional analyses will be found elsewhere in the report under the descriptions of individual deposits, by reference to index. See also table of tests. Nearly all of these samples are very high in calcium carbonate and are well fitted for the manufacture of cement. 5-Pt. Ill 34 MARL. (3) Interpretation. In the first place the percentages as here seen represent but a small portion of the bed as it is gathered in sampling. On the other hand it does not represent the true proportion of compounds which enter into the composition of the finished cement. The sample, as shown in many of the remarks in the key given above, is, when received at the laboratory, evaporated to dryness, so that water evaporation during analysis will not affect the final percent- ages by the steady loss of weight of the sample which would con- tinue to dry. In evaporating to dryness a sample generally loses from 40$ to 60$ of its weight, or in other words, a bed of marl as it lies ready for prospecting is at least half water, which must be lost in the process of manufacture. With this understanding each compound above named will be considered separately. Calcium Carbonate. This is the one necessary compound to be considered in the manu- facture of the cement. It should be at least 90$ of the dried sample. The calcium carbonate is derived by some agent from the hard w r ater of the lake above or at one time above it. It is pure wdiite and largely influences the color of the whole sample of marl depending upon the percentage of it contained. In the process of analysis the calcium carbonate is separated into two most ordinary compounds. In analyses given out from a labora- tory these are often stated separately, a percentage of calcium oxide and one of carbon dioxide being given, part of the carbon dioxide belonging originally to the magnesium carbonate. In such a case the easiest way to get from the stated analysis of the sample the percentage of calcium carbonate is to add to the stated per- centage of calcium oxide 78.577 of itself. We will have, within a very small fraction of a per cent, the amount of calcium carbonate which the calcium oxide represents. In the process of manufacture as well as in that of analysis, the calcium carbonate is broken up into calcium oxide and carbon dioxide. The carbon dioxide, which is a gas, passes off as a smoke and is not of any use in the finished cement, which should be free from it. It follows from this that after the 50$ or 60$ of water is driven off, 44$ of the percentage of calcium carbonate is. lost as gas and does not enter into combination in the finished cement. The all important compound CaC0 3 enters the factory in a wet, finely divided state, best fitted for mixing it most easily w-ith clay, THE USE OF MAUL FOB CEMENT MANUFACTURE. 35 is dried, then heated, expelling the carbon dioxide and leaving it as calcium oxide surrounding the other finely divided particles of clay which contain the silica to be made soluble by the action of the intense heat of the rotary kiln. Magnesium Carbonate. This is a compound analogous in many ways to the calcium carbonate. As seen in the above analyses it exists in the marl in very small percentages. This is the case when the marl is pure. A large percentage of magnesium carbonate in marl as pure as the above would not be characteristic of marl as a deposit. It would show generally that some clay had become mixed with the marl. For in such cases, when clay is laid down at the level of marl or during the deposit of marl, it generally contains a larger per cent of magnesium carbonate than does the marl,* and so influences markedly its composition. In such cases the percentage of insolu- ble matter, or silicates and aluminates is much higher than in the marls given above, on account of the increase in per cent of silica in clay over that in the natural marl, which of itself would contain a low per cent of silica. The magnesium carbonate has not been found to add to the real value of the marl, and it is certain that if it is present in any large amounts it will be a positive detriment to the finished cement. As the marl will vary from day to day in its content of organic matter and other components it is well to have the dangerous elements as much as possible absent. It must also be remembered that one of the greatest troubles is too much carbonates in clay. For this reason if for no other the marl should be low in magnesia. As seen, however, in the above analyses the purer marls are nearly free of magnesium carbonate and it seldom causes trouble in samples with a very high calcium carbonate con- tent. Ferric oxide and alumina. These are very likely deposited as ferrous hydrates in the marl bed. As such they are nearly colorless. When, however, a deep sample of very white marl is brought to the surface and exposed to the air for some time it may turn to a red or brownish red tinge from the oxidation of iron. These are seldom deposited in the marl in amounts to cause trouble. A case has before been pointed out where a marl with high content of organic matter showed also a very large percentage of iron and aluminum oxides. This fact is *See analyses of clay given elsewhere in this report and in Part I. 36 MARL. remarkable, that an intensely iron spring may discharge its highly mineral waters at the edge of a very pure marl bed. The grass about the spring will be covered with oxidized iron showing a red slime or even a bog iron effect, but the marl itself is not influenced in the slightest. It will generally be noticed that marls with the highest organic content also contain the highest percentage of iron and alumina. Insoluble and soluble silica. The per cent of insoluble silica is traceable to several sources. First of all, nearly all beds contain fine quartz sand independent of the ordinary coarse drainage sand and pebbles that may be washed into the bed as already explained. This sand can some- times be found to permeate the bed from top to bottom, even when the bed is thirty feet deep. If, however, there is a very even layer of the organic lining above referred to, the sand does not seem to penetrate as well, if at all. The sand will be found in the purest beds to vary from a fraction of a per cent to several per cent in amount. This is in the case of a comparatively pure marl. In case a clay has at any time mixed with the bed the content of insol- uble silica will vary, but will remain larger together with other disturbing features, such as the increase in magnesia before men- tioned. Sometimes such a condition will produce an increase in content of magnesia toward the bottom of the bed, while in a pure bed little if any regular variation of magnesia has been discoverable with increase in depth from which sample may have been taken. Soluble silica. The marl is intimately associated with the remains of plants, no matter how pure it may be or at what depth it may be sampled. The same may be said in regard to shells although samples have been found where the shell formation could not be traced. It is certain that plants, especially diatoms in the course of their growth, render a very small amount of silica soluble. This of course would remain in the body of the marl after the death of the plant. Certain shells are said to have the same power. The amount of silica in a good marl is very small. The soluble silica will not be in amount to help or hinder greatly, for, as may be seen in the analyses cited, it is but a fraction of a per cent. The insoluble silica is, however, higher in per cent and it is that which must be watched closely. It ought not to exceed three or four per cent for the reason that it interferes with the balancing of the silica and THE USE OF MAIiL FOB CEMENT MANUFACTURE. 37 calcium content of the slurry and prevents the best burning of the mixture. Insoluble silica as sand is one of the most refractory substances known. It is not as finely divided as clay silica, does not make as intimate a mixture with the lime of the marl and does not flux so easily. Although sand marl cement can be made, sand is entirely out of place in the process used in Michigan, and should be guarded against carefully in the selection of raw material. Organic matter. Organic matter is a necessary evil in relation to marl. It is of no positive harm except that it increases the weight and bulk of marl without adding to it its usefulness. It is burned out as nearly as possible in the manufacture of cement and all that remains is ash. As noticed in the sample above given, where the calcium carbonate content falls suddenly it is nearly balanced by increase in organic matter. It has already been explained how profoundly organic matter influences the character of a bed. The law of its own variation can be depended upon to hold true where outside agents have not also contaminated the bed. It will also be noticed that in the very pure samples where organic matter is nearly ab- sent the marl has but a trace of other compounds beside calcium carbonate and that where it increases in a large degree, all the elements already mentioned spring into prominence again. If, then, marl is very free from organic matter it is liable also to be free from dangerous compounds. If it is high in organic- matter it will be bulky to handle, will not yield a large percentage of calcium oxide for the production of cement, and will necessitate continual watching for fear of dangerous compounds. Sulphuric and phosphoric acids, chlorine, etc. These compounds, if present in large quantities, would be dangerous. They cause little trouble unless the marl is highly organic when, as before explained, it is of little use anyway. Sul- phuric acid is often present in dangerous amounts in otherwise com- mercial marls. In the above samples some have been given in full and then figured without the organic matter. This is not a true representation of the real value of the sample as it exists in the marl bed and is not intended as such. Care should be taken to dis- count the high and flattering percentage of calcium carbonate shown by such an analysis. This reconstruction of the real analysis is made to determine whether or not the dangerous elements would, in the burned marl, exist in sufficient quantity to forbid its use. The per- 38 MAUL . centages exist in such reconstructed analyses as they would enter into the formation of the cement and directly influence its forma- tion. Perhaps, this one fact should be borne in mind, that in the new proportion of compounds brought about by burning, the carbon dioxide derived from the carbonates of calcium and magnesium is driven off also by the heat before the marl has reached the propor- tions which it possesses upon incipient vitrification. In order then to give the truest percentage estimate of the marl as its component parts would exist when ready for use, the analyses should be figured with both organic matter and carbon dioxide absent. Having noticed the various ingredients and their variation, the final question to the prospector is fitness of marl as shown by analyses. The sample, not the bed, is suitable if it contains 90$ or over of calcium carbonate and no dangerous element in large proportions. If the marl runs over 90$ of calcium carbonate it is not liable to have other ingredients in dangerous proportions, pro- vided the bed is a characteristic deposit, not mixed with any of the adulterating foreign matters before mentioned. As a matter of fact it would be hard to find a bed, all samples of which are above 90$ calcium carbonate of depth or extent suitable for manufactur- ing purposes. The reason for this is the steady variation of organic matter before mentioned. § 9. Location and size of bed. Besides quality of the marl there are other points worthy of notice. Is the bed located on a railroad or one of the Great Lakes? If it is found necessary to build a railroad to the deposit, the extra cost must be reckoned, compared with competing raw material more favorably located. If the bed is located where vessels can easily reach it from the Great Lakes, it has one of the best natural ad- vantages. The expression has often been heard, upon the sounding of a small bed of marl to the depth of 15 or 20 feet, “Oh, here is marl to last for years.” Besides quality and location, the size is the third vital point always under consideration, and one which the owner should be able to determine himself. To illustrate the point clearly, the changes which the marl undergoes up to the time of partial vitrification, will be reviewed as nearly as possible. The marl as it lies in ordinary swamp consists of from 40$ to 60$ by weight of water. First this water must be evaporated from THE USE OF MARL FOR CEMENT MANUFACTURE . 39 the slurry and then whatever organic matter is contained in the marl must be oxidized, burned out, passing away to remain func- tionless in the finished cement. Still another important shrinkage in volume and weight must take place. The remaining useful calcium carbonate is also oxidized, losing 44$ of its weight in the form of carbon dioxide, which passes off as gas with the smoke of the kiln. Shrinkage or gain in weight of the other ingredients is slight on account of their small percentage. Take for example, sample No. 5 of the foregoing analyses. (1) 100$ less 61$ equals 39$ dry marl. (2) 87$ of 39$ equals 33.93$ of original wet marl as calcium car- bonate. (3) Calcium oxide is always 56$ of a given weight of calcium carbonate. (4) 56$ of 33.93 equals 19$ of original weight of wet marl as calcium oxide. Of sample No. 5, but 19$ therefore of the weight of the sample as it was taken from the marsh, enters into the final weight of finished cement as an active cementing agent. Nearly all of the remainder passes off as useless gas or as water requiring great expense in heat to evaporate it. While this sample is rather low in calcium carbonate, probably the very best samples of fairly wet marl could not show above 25$ calcium oxide available after burn- ing. This has a direct bearing upon the question of area and depth of marl necessary for cement manufacture. The estimates given by the factories in active operation in the State, figure 1J to 2 cubic yards of marl as equal to one barrel of Portland Cement. This would vary according to the purity of the marl and the amount of water contained. The water is a necessary evil; by the wet or slurry process there must be enough water so that the marl will mix and pump readily, though, after mixing all the water must be evaporated in the burning which requires expense in fuel. Tak- ing 1J cubic yards as the equivalent of one barrel of cement it will be well to calculate the consumption of an ordinary fourteen rotary mill. (1) 1| cu. yds. equal 40.5 square feet of marl one foot deep. (2) 14 rotary mill produces 1000 barrels cement per day. (3) 1000 times 40.5 equals 40,500 cubic feet of marl per day consumed. 40 MARL. (4) 43,560 square feet equal to one acre. (5) Dividing 43,560 by 40,500 we find there are 1.0755 days work in an acre one foot thick. (6) In 200 acres 200 times 1.0755 equal 215.1 days work. This is about the number of days the factory would run out of a year. If the deposit were 25 feet thick, such a deposit 100 acres in area would run a 14 rotary factory 25 years. Such a rate of consump- tion of raw material seems enormous and according to this esti- mate there are few single beds of marl that would furnish raw material for a length of time to guarantee the erection of a plant. Certainly at strip of marl 75 to 100 acres in area would scarcely warrant the erection of a large mill. The largest cement corpora- tions in the State buy all the marl in a given vicinity comprising several lakes. In considering the largest area of bed for one factory it must be remembered that marl cannot be transported any distance to a factory. The factory must be located on or very near the bed. The immense shrinkage in volume of marl during process of manufacture has already been shown. The expense of carrying marl any distance cannot be met when competing with other factories located on their beds and with the immense lime- stone districts of other parts of the country.* In conclusion, a marl must be of the best and the most uniform quality, it must be free from its natural adulterants, it must be located near some waterway or on a railroad, must be 15 or 25 feet in thickness over an area of several hundred acres. Such qualifica- tions all included in one vicinity, are very difficult to find. High quality throughout, unlimited quantity, and fine natural location are necessary, for a good article must be manufactured and shipped easily and cheaply and upon an enormous scale to make the manu- facture of marl a very paying and useful industry in the State. •Compare, however, what is said concerning the Hecla Portland Cement Co. CHAPTER IV. THEORIES OF ORIGIN OF BOG LIME OR MARL. § 1. Introduction — the various theories. An effort will be made to give all the ideas obtainable upon this phase of the subject. It is fact and no more than natural, that every one who has examined marl deposits has some one view as to the origin of so peculiar a resource. With the knowledge the prospector now has of the nature of marl it would be very helpful to arrive at a correct conclusion as to the origin of marl deposits. This would rapidly aid in pointing out the most probable location of the marl and would prepare the explorer somewhat beforehand as to the exact quality of the marl and would inform him as to the necessity of a more or less minute examination of different parts of the bed to pass upon its fitness for practical purposes. A scientific conclusion in regard to the origin of marl is also a small contribution to the exact knowledge of the geology of the State of Michigan and as such should be of permanent scientific value. In the hope that out of many opinions the truth will finally come, space is given in this place to all views obtainable upon the origin of marl. Prof. Davis’s work on the subject is given a sep- arate chapter (Chap. Y), while the others will be stated as clearly as possible under this heading.* (1) Shell theory. The idea has often been expressed by those who examine a bed that shells are the origin of marl. There are certainly beds that verify this statement. Some are beds of nearly solid shells, and shells too that are well preserved to a depth of fifteen or twenty feet. In such cases, no doubt, the location of the bed has been especially favorable to the formation of shells. Samples of shell formation from Florida have been seen where the shells formed a *Some farther suggestions, and observations, microscopic and otherwise, by me, will be found in the last chapter. L. 6-Pt. Ill 42 MARL. calcareous mass of shells and their broken down remains, very similar to the purely shell marl of our own State. (2) Sedimentary theory. This theory is that the lime existing as it does in our State in fine particles distributed through the soil, was washed by the ac- tion of the water from the pebbles and limestone rock of the State during the glacial period. That after the ice had melted the lime- stone sediment of finely ground rock was washed into the drainage valleys left by the ice in melting, and formed a fine sediment much like a clay, but being of a different density than clay, was deposited separately, forming the beds we now have.* This idea was sug- gested by H. P. Parmelee. (3) Chemical theory. This theory, one that was found to be held by many chemists of the State, is at least, a very plausible solution of the cause of the formation of marl. It is based on this fact or principle in chem- istry. Carbon dioxide by its presence in water aids it in holding in solution a greater amount of calcium and magnesium carbonates. The minerals are held in the form of double carbonates of calcium and magnesium. When a water containing carbon dioxide under pressure and a larger amount of the carbonates than it could other- wise hold in solution without the presence of the carbon dioxide escapes from confinement underground, and is exposed to the air, the carbon dioxide as a gas, escapes and the carbon- ates, no longer held in solution by the presence of the gas, are precipitated as simple carbonates of calcium and magnesium. There is no doubt whatever that such a reaction takes place in many instances which can be cited in nature. The idea here held is that all the conditions are correct for such reaction. The water of our springs is confined in underground waterways, or better, reservoirs. The gas cannot escape and is under pressure. The carbonates washed from the soil and lime rock are in the water and in solution as evinced by its clearness. When the spring flows out from beneath a hill the water spreads out in the calm inland lake, is released from pressure and perhaps warmed, and the gas escapes, and the carbonates are precipitated to the bottom in the form of a marl. Such is this popular and striking theory. It has much to recommend it. Some if not all the conditions named are * Or possibly where the rock was practically all limestone, the glacial rock flour might be almost wholly composed of calcium carbonate. L. THEORIES OF ORIGIN OF BOG LIME OR MARL. 43 present. The cases in nature where such change undoubtedly takes place are to be found in our mineral springs of the West and Europe, and in calcareous tufa of our own State. The heavy mineral springs are surrounded at their very openings by the minerals precipitated from them as the waters issue. In most of these cases the process of precipitation has been aided by the cooling of the waters which are very hot. Hot water such as these contain will also take into solution a much greater percentage of minerals. From the above comparison it will be seen that though minerals are somehow precipitated in both cases, the conditions are not exactly identical and it would be dangerous, therefore, to reason from one to the others. The conclusions reached in the search for an origin for marl deposits are much the same as those reached by Prof. Davis in his report, which is given in full in Chapter V. The endeavor will be made in the following pages to show wherein the several theories above named point to the real causes of the formation of marl, and also to record the steps taken to test the relative value of the same, as made in the special survey of the State requested of me by the State Geologist. § 2. Shells. Shells form a greater or less part of a marl bed. Their presence is sure evidence that they are an agent in the origination of a bed. An analysis of pure shells from a marl bed shows that they help to form the purest part of the bed and that the proportion of their compounds as compared to that of a very pure marl without shells is very nearly the same. They are, however, but a minor agent in the formation of most beds. Their existence and plentiful growth depend upon much the same causes which are responsible for the principal agent of cement formation. They are therefore plentiful in most beds in the marl, because they are produced at the same time and under the same conditions as the marl. Many marl beds may be seen on the other hand, which contain few if any shells. They are not broken down so that their identity is lost, as many would have us believe, for where shells exist in a bed, they may be seen at some depth, delicate and frail but perfect in outline, so that if they are the sole cause of marl their fellows should have remained in great numbers and many partly broken down, instead of here and there a perfect shell at fifteen and twenty feet below 44 MARL. the surface. In soundings of twenty to fifty feet beneath the sur face of the water under that many feet of marl or in the center of a lake, they are nearly and often entirely absent. They could scarcely be held responsible for the presence of marl in such quan- tity at that depth. § 3. Sedimentary theory. Among all reasoners upon the subject there is no difference of opinion as to the ultimate source of marl. It certainly came from limestone through erosion and the carrying power of water. An- other basis point of this theory is also true. Marl is deposited much like a sediment. It lies very evenly unless disturbed by sudden jumps in the outline of the lake bottom. Further proof of the theory does not appear to exist. Marl deposits do not seem to occur regularly in given districts, they do not appear to extend in a given direction and so far this theory has not assisted in the location or accounted for the peculiar facts which hold good in this formation.* § 4. Chemical theory. According to the theory of simple chemical precipitation of marl from spring waters, the marl should be deepest, piled or crusted about the mouths of these springs and stopping by its accumula- tion their outlets. Such is not the case as the marl does not con- fine itself to the immediate neighborhood of these springs which are in most cases surrounded by sand or muck. According to this same theory, if the water managed to escape and mingle in the lake beyond, the marl should then deposit evenly all over the bottom of the lake as it does in depositing in a kettle or basin. This is also contrary to fact as marl is very intermittent, in its deposit, is often not deepest in the deepest portions of the lake, and does seldom form a layer continuous and even over an entire lake bottom. Another question of importance is this: Is there with the relative proportions of carbon dioxide and carbon- ates existing in our inland lakes to-day more than enough of the latter to exhaust the power of the water at its ordinary tempera- ture and pressure to hold in solution the percentages of carbonates existing in these waters, or will the spring water not be able to easily hold in solution the small amount of carbonates with or without the free carbon dioxide? If the calcium carbonate is not *It does apply to some of the fine grained calcareous clays, such as those used for white brick at various points. But in no case is the separation of calcium carbonate mud from other mud anywhere near as perfect as in the bog lime. L. THEORIES OF ORIGIN OF BOG LIME OR MARL. 45 great enough to overburden the water it can be held in solution in the lake with or without the presence of the free carbon dioxide. In this case the carbon dioxide can escape from the water or re- main with it, but the water can yet hold in solution its salts of calcium and magnesium and carry them out of the lake without depositing them as marl. Two facts are to be ascertained before this theory can show the necessary conditions under which it may be possible to operate. (1) What is the point of solubility of our spring waters, or the per- centage of those salts necessary to produce over saturated solution? (2) Is the percentage of calcium and magnesium salts in the ground water below or above the percentage? If below the theory must be groundless for it must be above in all cases supplying a cause for all phenomena in regard to the formation of marl. (1) The point of saturation of spring waters and the influence of carbon dioxide upon the same. After search the carefully conducted series of experiments of Treadwell and Reuter upon the solubility of carbonates was found by the State Geologist and an abstract, translated from the Ger- man by him, will be found elsewhere. (2) We have to compare with the results of these experiments the actual proportions of calcium carbonates existing in the spring waters of lakes and springs of Michigan, as given below. (See page 46 and also the hardness tests of Cloverdale on page 131.) According to Treadwell and Reuter’s carefully made experiments, water at ordinary temperature and pressure containing no free C0 2 may yet contain permanently 0.38509 grams of calcium bicar- bonate or .238 CaC0 3 per liter, while the authorities quoted below estimate it at differing temperatures and pressures from .7003 to 3. per liter. Now the analyses of waters from Michigan show a content of calcium carbonate from .175 to .250 grams per liter or 175 to 250 parts in a million. With this in mind it can easily be 46 MARL. ANALYSES OF JARS OF WATER FROM CLOVERDALE DISTRICT BY A. N. CLARK FOR STATE GEOLOGIST AT MICHIGAN AGRICULTURAL COLLEGE LAB- ORATORY. INTENDED TO BE TIGHTLY SEALED AND PROMPTLY ANALYZED BUT NOT THOROUGHLY SATISFACTORY, C0 2 NOT ENTIRELY RELIABLE. RESULTS STATED IN PARTS IN 1,000,000. Number of Sample. 1 . 2 . 3. 4. 5. 6 . 7. 8 . 9. 10 . 11 . 12 . 13. 14. 15 16 17. 18. ?9 20 21 . Location. Boiling spring at the head of Long Lake Water at outlet of Long Lake, running creek Well in vicinity Large boiling spring near sounding, No. 41 Horseshoe Lake, between sounding 39 and 40, in 50 feet of water From bottom of marl basin in 10 feet of water, sound- ing 45 *. Surface of basin. Marl 6 inches beneath water sur- face Spring at side of Horseshoe Lake (N.lobe, not sounded) Spring at head of Guernsey Lake, surrrounded by gravel Surface water at Guernsey Lake Water at bottom of Mud Lake (35 feet) with 18 parts per million dissolved Fe 2 0 3 & A 1 2 0 3 Water from surface of Mud Lake Seepage spring on Mud Lake Water from well on divide between Long and Mud Lakes, 35 feet. Quicksand at 18 feet, with 12 parts dissolved, Fe 2 0 3 & A1 2 0 3 Water from drive well 25 feet into spring, which for- merly emptied into Long Lake, between Long and Mud Lakes: with 21 parts dissolved Fe 2 0 3 & A1 2 0 3 Stagnant water in ditch which formerly connected Mud Lake and Long Lake. Metallic scum and red bottom; with 16 parts dissolved Fe 2 0 3 & Al 2 0 3 Water from well on high divide between Long and Twenty-One Lakes; with 50 parts of Fe 2 0 3 & A1 2 0 3 . .. Spring above level of Pine Lake Outlet from Pine Lake, Pine Creek Large boiling spring near outlet of Pine Lake In 25 feet of water, center of Pine Lake irbon oxide. Calcium Carbon- ate. Magne- sium Car- bonate. 0.00 100.00 67.80 44.00 217.00 100.00 3.30 160.00 69.7 19.3 200.00 75.3 19.8 117.00 85.6 6. 100.00 75.6 70. 58. 15.4 160. 81.7 66. 130. 61.3 0 . 40. 71.1 6.6 53.6 trace 0 . 30. trace. 0 . 80. 62. 60. 80. 28. 39.6 240. 116. 44. 48. 16.6 39.6 156. 203.4 22. 170. 80.2 6.6 80. 76.6 11. 136. 81.7 6.7 80. 69.6 4.4 87. 75. WATER OF OTHER REGIONS ANALYZED FOR CARBONATES.* Location. Free C0 2 . CaC0 3 MgC0 3 . Fremont flowing well 0 100 Water, spring in marl bottoms, Corrinne 26.40 210 Water at Straits of Mackinac 0 130 Traverse Bay at Traverse City — 0 150 Duck Lake near Green Lake. Spring at head of lake Marl Outlet of Duck Lake. Sandy bottom 0 165 W ater south end of Central Lake 24.2 0 '200 Clear water of Mound Spring near Central Lake 190 Water in spring 100 feet above Central Lake 0 190 Flowing woll in F.a.st, .Torda.n o 210 Kettle near Manistee Junction, is 15 or 20 feet lower than Round Lake near by; it is supposed to discharge by an underground channel in the Pere Marquette River 0 110 T.ong T,n.lfp Ma.nist.oo .Tnnotinn rod wa.t.or 0 205 Water, Round Lake, Manistee Junction 0 185 Outlet of Long and Round Lakes Manistee Junction 6.60 175 * The content of C0 2 here given cannot be relied on as the bottles were stoppered with cork, permitting the escape of the gas. Analyzed by A. N. Clark at the M. A. C. Laboratory. THEORIES OF ORIGIN OF BOG LIME OR MARL. 47 seen that the carbonated waters of our springs and marl lakes are generally far below the point of precipitation. The water of these springs and lakes could take into solution 100 or more parts in a million of calcium carbonate instead of being over burdened and precipitating them in marl.* It appears very clearly that Tread- well and Reuter’s experiments are carried on under artificially produced conditions which tally closely with those found in our springs and lakes. They have decided for us carefully the precipi- tating point at which carbonated waters with various pressures of free C0 2 , and temperatures usually 59° F. cease to bear in solu- tion carbonates in the form of the bicarbonate. They find that point above that of Michigan spring waters, or in other words show very clearly that our spring water can generally take more salts into solution instead of being ready with slight changes of temperature and pressure to precipitate those which they already carry. In other words our spring waters cannot precipitate their bicarbonate as marl by the simple chemical process of precipitation from a saturated solution, because they lack considerable of being saturated. § 5. Indications by circumstances of occurrence. In discussing the origin of marl to form as perfect a chain of evidence as possible the conditions obtaining must be determined as accurately as possible. If aualogies are used as proofs the conditions in both analogies must be alike. If this is not followed fatal mistakes are likely to occur. An agent has produced an effect which is before us in the form of marl beds. The bearing of the facts concerning position, com- position, variation in composition, location, variation in depth, foundation or basis and covering, which we have described, should be studied with this in mind: The marl beds lie upon the surface or in the present geologic stratum, and since they are not covered by any great thickness of earth and are clearly produced since the *Compare also the analyses in U. S. G. S., Water Supply Paper No. 31. 48 MARL. glacial period from the fact that they lie in the hollows left by the glaciers, the agencies producing them must be modern as well. Marl has been described elsewhere as a complex compound. In the impure marl there are large percentages of insoluble matter which can readily be traced to the presence of foreign clay, sand and organic matter. It can be readily seen that these have nothing to do with the production of marl and therefore the very purest samples of marl must be considered in order to arrive at a conclu- sion in regard to its origin. As marl is analyzed in the laboratory it consists of calcium carbonate, forming nearly the entire percentage, magnesium carbonate (always in very small percentages, that is, in the very pure sample), iron and alumina, organic matter and traces of sulphuric acid and sometimes phosphates. The following is the analysis of such marl: Calcium carbonate 95.281 Magnesium carbonate .946 Ferric oxide .536 Alumina .159 Silica insoluble 1.205 Silica soluble 1.316 Organic matter 1.510 Water 300 Phosphoric acid traces. Sulphuric acid slight traces. Chlorine slight traces. Alkalies traces. Total 101.203 Very likely such a marl as the foregoing is as nearly as possible to purity as can be obtained. The content of organic matter is too low to be typical, while the content of soluble and insoluble silicates is a trifle high. There is always the marked difference in percentage between magnesium and calcium carbonates above, excepting when a clay forms a part of the deposit, when the percentage of magnesium carbonate may increase to large percentages. In very pure marls or in those containing 90$ and over of calcium carbonate, the magnesia does not form any large proportion. It is noticeable that in the study of deposits for factory purposes, it is found that where other impurities increase the magnesia increases as well. The only direct source of carbonates about to be studied is the water which in all cases lies or has at one time lain above the Geological Survey of Michigan. Vol. VIII. Part III. Plate II. ■J* 3 o 3 $ » 4 V.O it No. 6. HORSESHOE LAKE, CLOVERDALE DISTRICT, T. 2 N., R. 9 WEST, WITH DIAGRAMS OF SOUNDINGS. THEORIES OF ORIGIN OF BOG LIME OR MARL. 4<> deposit. Now if all the salts contained in the hard water were precipitated, the proportions between the calcium and magnesium in the water should be the same as the proportion of calcium and magnesium in the marl. Such is not at all the case. Notice in the foregoing analyses of waters from springs and lakes about Cloverdale (pp. 20 and 21), that the proportion of calcium carbon- ate to magnesium carbonate is about 2 : 1. No analysis of marl was ever seen in which the proportion was anywhere nearly equal,* the proportion of 90 : 3 being the most typical. This brings to light a very important principle or lack of principle ruling the formation of marl and as it occurs with other compounds besides magnesia it will be well to notice it in the outset, to wit, the lack of the rela- tionship as established between the compounds in the water and the compounds in the marl. This is most easily illustrated by the wide distance between percentages of calcium carbonate and mag- nesium carbonate in the marl deposit. As found in water CaC0 3 : MgC0 3 ::2 : 1, but in marl :: 90 : 3. In relationship of iron and alumina it cannot be shown as well that they differ because in both marl and water they are found in much smaller amounts. They are always very low in the purest marls. Especial search has been made for bog iron in the presence of marls. Here we meet a very interesting fact; marl does not occur in admixture or in the immediate presence of bog iron ore. One locality was noticed where a marl lake was drained by a creek that had bog iron ore along its course, but no bog iron could be found in, or immediately surrounding the marl. There is only one case in which the iron may increase to any appreciable extent. This is in deep water soundings where the marl has been displaced by a mucky marl.' Such was the case in the following sample of muck-marl found in 54 feet of water at center of Horseshoe Lake, Cloverdale region: per cent. Insoluble 15.14^ Fe 2 0 3 (A1 o 0 3 ) 13.73 CaC0 3 43.13 MgC0 3 1.66 Organic matter 26.34 The surprising features of this analysis are the high per cent of iron and aluminum oxides and organic matter. ♦This may perhaps be accounted for upon the chemical theory by greater solubility of magnesium salts, for we have as yet no exact data as to the relative solubility of the calcium and magnesium carbonates, and yet it is not likely that so great a difference would exist. L. 7-PT. Ill 50 MARL. The chemical agency in the deposit of iron oxide must therefore be different in the case of springs from that working in the marl beds into which these same springs empty. For example, an in- tensely hard water spring is seen to empty into a pure marl lake. The growth of water plants along from the spring is coated with a thick deposit of iron oxide. No marl is deposited. On the other hand, in the lake immediately below there are but traces of iron and nearly all the deposit is calcium' carbonate. This leads us to conclude that the agencies most active in the precipitation of iron and calcium are different at the spring and in the marl bed as the same water furnishes material for both. There is plenty of iron left for precipitation in the lake as the waters emptying out of the same show about the same percentage of iron. Sulphuric and phosphoric acid are usually estimated as salts. In the purest marls they are scarcely ever far above 0.30$. In deep specimens where there are large proportions of organic matter they sometimes run higher. The organic matter is a component part of every marl which plays a very important part in its history. As we speak now of the purest marls only, it is here found in small percentages. It can never really be said to be absent and is that compound or con- stituent part of the marl which is the most widely fluctuating. There are found certain exceptions (see Lime Lake, p. 133) where the marl is a nearly solid shell bed. In such a case, the conditions having been always favorable to the growth of shells, the quality remains constant even at a great depth. The ordinary marl bed varies in composition. It is very much higher in content of organic matter at the bottom than at the top. It often happens in a bed thirty feet in depth that at the top it is 95$ CaC0 3 and at the bottom 65$ to 80$. The change is generally due to increase in organic matter at the expense of the content of calcium carbonate. In a lake, of which the bottom is entirely covered with marl and the shallows around the shores consist of deep marl covered with but a few feet of water, the marl toward the center of the lake, as the water deepens, becomes much higher in content of organic matter and of course suffers in its percentage of calcium carbonate. The content of magnesium carbonate does not increase with depth of the sounding, but may vary, either becoming slightly greater or less. If clay has sifted in with the marl it usually shows in a higher percentage of magnesium carbonate. THEORIES OF ORIGIN OF BOG LIME OR MARL. 51 Marl beds are not seen to show any variation in the content of iron, chlorine or other such foreign substances where springs di- rectly above such beds contain the same. Having reviewed the extensive variation in composition and depth, together with the condition of surface and basis for deposit, the next important consideration is the water above the marl. The water of our springs and lakes as shown by our analyses on pages 46 and 131, runs as follows : Free C0 2 , 0 to 44 parts in a million. CaC0 3 , 80 to 217 parts in a million. MgCO s , 62 to 100 parts in a million. This excludes the soft water lake and the well on the divide which seem to be extremes on either side. The water so laden flow r s from the springs into the lakes by the springs upon high land and by the water holes or living springs which empty under water in the lake and are indicated by open water in the dead of winter and are avoided by skaters as air holes. In either case the cold water will at once flow down till it reaches the deeper parts of the lake, being naturally heavier than the somewhat heated water about it. The water of the spring holes must carry all its C0 2 with it as there is no open air for it to escape to. The running water of the upland springs must lose some of its C0 2 , but not all as it is a gas, heavier than air and does not escape easily. There has been some careful research into the behavior of lake waters,* and an instrument called the thermophone has been in- vented to trace accurately the changes of temperature at great depths, not easily reached by an ordinary thermometer. It was found by a study of Lake Cochituate, near Boston, that in very deep water the bottom temperature remained the same and the water stagnant throughout half the year and that in the fall and spring a general vertical circulation of the water took place. “The diatoms and some of the infusoria are most abundant in spring and fall, or during the two seasons of the year when the water circulates freely from the top to the bottom.” The temperature of our lake waters controls their density and their density their power to move by the law of convection, the warmer water rising, the colder sinking. ♦Warren and Whipple, Meteorological Journal, June. 1895. Technology Quarterly, July, 1895, VIII, 2, pp. 125 to 152. 52 MARL. The position of the water again controls its power of getting to the air and losing its carbon dioxide. It will be seen by careful perusal of the experiments of Messrs. Warren and Whipple that our deeper lake waters must have a systematic movement each year. In the deeper portions of 30 to 50 feet depth or more, the water- remains at or near the point of 39.2° F. or that of greatest density. It is a little above that point in winter while the surface water next the ice is of course nearer freezing point. The water in the deeper portions of the lake already referred to moves in spring and fall changing places with the surface waters. It then acts as a reservoir of cold heavily laden carbonated waters which replenish the surface waters, and the carbonates and C0 2 are car- ried to the surface where any free C0 2 may escape. It is then clear that the very cold water of our springs may not in summer flow at once to surface of the lake and is not at once thoroughly aerated by contact with the air at the surface of the water, but on the contrary flows to the deeper parts of the lake and is buried for a season till convection brings it to the surface when it naturally spreads out, being the warmer, and has free access to the air. Now we find that to no great extent is the marl precipitated in deep water. In soundings made in 40 and 50 feet of water, the marl nearly lost its nature, becoming marly muck. When we allow a basin with spring water to stand, the C0 2 collects in bubbles on the bottom and sides and little rises to the surface. In the same way we can tell that it collects on the bottom of a lake, for if we stir the bottom small bubbles of gas find their way to the surface. This is the condition in which the water remains as it lies deep in mid lake. It is difficult to tell from a comparison of the analyses of lake waters and those of the springs that flow into them whether any carbonates and C0 2 are lost by precipitation from the fact that the lake waters must of necessity be diluted by surface drainage waters and rain water containing no carbonates. It will there- fore be necessary to compare the springs with each other, the wells with each other, and the lakes, taking each sounding that corre- sponds in position with the other. The Cloverdale Lakes may be graded in the intensity of their deposit, the first named having the deepest marl and most active deposition, the last having but traces, in the following order: Horseshoe, Long, Price, Guernsey, Mud Lakes. Upon comparing Nos. 1, 8, 9, 13, 18, 20, which are samples THEORIES OF ORIGIN OF BOG LIME OR MARL. 53 of spring water, analyses of which are given on page 46, it will be found that the most intensely marly lakes have the springs with the greatest content of carbonates and grade down to the soft water lake, which in turn has a spring with the smallest content of carbonates. Upon comparison of well samples 3, 14, 35, 16, 17, the same rule applies, but not with equal clearness as the well near Mud Lake was scarcely more than a surface well. Nos. 7, 10 and 12 are the surface samples of Horseshoe, Guernsey, and Mud lakes. They again show the same order of the marl deposit. They all lack free C0 2 and contain carbonates in the order of the marl deposit. Horseshoe is the greatest and Mud Lake is again least. Nos. 5, 11, 21 form a comparative set of the deep waters of Horse- shoe, Mud and Pine lakes. The relation is again maintained with- out break although the free C0 2 in Mud and Guernsey are nearly the same. For comparison of water in the lake itself we have Nos. 5, 6, 7, of Horseshoe Lake. These are named in the order of their depth, No. 5 being taken in 50 feet *of w r ater in mid lake, No. 6 in a marl basin and at the bottom next to the marl, and No. 7 at the very surface. It will be seen that according to these analyses, the water is steadily and rapidly losing its content of C0 2 and car- bonates as it approaches the surface. At the bottom it had the highest (19.88 parts C0 2 ), at ten feet it has but 6 parts, and at the surface nothing. The carbonates are lost much in the same pro- portion, less of the magnesium carbonate being lost than of the calcium carbonate. This tallies very well with the marl which gains more calcium than magnesium. The above comparisons deduced from the table of analyses would point to the following conclusions. The deep springs furnish the hard waters for the marl lakes. The cold water sinks to the deeper parts of the lake, which con- tain a supply of carbonates and C0 2 . When this water reaches the surface by aid of convection, it loses its C0 2 entirely or in part and its proportion of carbonates suffers as well. It must be borne in mind in this consideration, that water and C0 2 must differ in volume as the temperature rises. The water as a liquid would not have a great change of volume in rising from its temperature of greatest density in mid lake to luke- warmness at the surface under a summer sun. On the other hand 54 MAUL. carbon dioxide would be almost entirely lost and would expand greatly. While its content per liter of water at the depth of 10 feet would be much less than at the bottom of the lake, one thing is cer- tain that at the surface it is lost entirely, not being contained in any of the samples taken from the surface of any of the lakes. Having discussed the composition of marl itself we find it in- fluenced by the depth of water over it and by its own depth. Upon the study of water and its content of carbonates we find the opposite. The deep water contains the greatest amount of carbonates, but does not release them till shallow water or the surface are reached. Heat and the seasons play an important part in renovating the deep water, bringing it to the surface where it loses its carbonates by some agency. The conditions of marl formation have been discovered as nearly as possible. It is found that the carbonated waters even if at first rendered stagnant are brought twice a year to the surface, to light and heat, but that according to carefully conducted experiments, they cannot lose their carbonates by simple precipitation of the carbonates upon withdrawal of C0 2 *because none of the compounds in question are in great enough proportion to form a saturated solution. For a pure analogy and not as a proof, let us look at other paral- lel cases in nature where chemical compounds exist in such mild proportions that it does not seem possible for change to take place, but nevertheless such change is going on upon a large scale. The nitrates or compounds of nitrogen can not readily be formed and made soluble from the compounds existing in the soil and plants would suffer without them. They are formed, however, by the interposition of an outside agent. This is a minute living organ- ism that forms upon the roots of the plant at the same time, form- ing a large amount of soluble nitrates for the use of the plant. This is a case of chemical recombination impossible without the aid of this living organism. The process of biochemical down- tearing is so varied and frequent that it need hardly be pointed out The process of rotting so necessary to the destruction of plant and animal life and its recombination in simpler forms fit for plant food, is accomplished by millions of bacteria. Acids, alkalies and numbers of new 7 compounds are formed where if chemical action alone were depended upon, plant life w r ould starve in need of less complex food. THEORIES OF ORIGIN OF BOG LIME OR MARL. 55 It is here in the discussion of precipitation of calcium carbon- ate in the form of marl, that a new set of phenomena or conditions must be duly represented and described. There are clearly live marl lakes, i. e., lakes that are depositing at the present time. The deposit is carried on in shallows in intensely marly lakes. It is not confined to plant organisms that can be seen with the naked eye. The reason for this is clearly proven. All live and dead plants or all inanimate objects on the bottom are covered with the white deposit of calcium carbonate. The objects covered need not necessarily have grown in the water. Many trees may dip half decayed branches into the water, yet these twigs are covered with a thick coating of the marly substance. The numerous water plants upon the bottom in the shallows are also thickly coated with white. One plant especially thrives in these shallows. It is to be easily distinguished bv its whorls or leaves at each joint.* It would seem probable that these plants, espe- cially in shallow water would act as distributors for their coating of marl, as the ice of winter must certainly tear them out, in being floated to different parts of the lake as the ice breaks up in the spring. The marl in the shallows of such a lake forms around everything, forming pebbles around rushes and roots that extend above the surface of the bed. The pebbles are somewhat hard and in boring they sometimes seem like stones. The roots die away leaving a hollow nearly enclosed pebble. The marl in these cases forms fine accretions and is very granular, seeming at first exactly like sand, but yielding to repeated efforts to crush it with the finger. Upon closer examination of plants upon which the marl is depositing it is found that they are coated with a fine slime which is more or less whitened by the presence of the particles of marl. When a lake or portion of the same lake is examined where the deposit is not so active, the same slime is found, but it is not so thick and it is trans- parent rather than white, on account of the absence of the white particles of marl. Such was the case in a chain of lakes near Colon, the difference between the lower and the upper of the lakes being very marked in this respect. The active precipitation of marl in this manner was first remarked in notes on Horseshoe Lake near Cloverdale. It is an interesting fact that, while the shores of this lake were thickly encrusted with thick marl in the ♦This is the Chara referred to by Davis, Chapter V. L. 56 MABL. process of precipitation, the marl at the center was of the poorest, though not over a few hundred feet removed. Several actively depositing lakes have been noted since. Such a lake was usually the upper lake of a chain. It received little drainage water and had the first of the spring water. The precipitation, while it is taking place must be very rapid. Stakes stuck in the marl as anchors for fishermen are whitened by the deposit of marl; branches, twigs, etc., sometimes have an incrustation of a quarter of an inch or more in thickness. Even in such lakes the marl when traced out into deep water, becomes darker and heavier in organic matter, and if sounded to the bottom, shows much the same in- crease in organic matter. It appears the only feasible and true explanation of the origin or exact method of precipitation of marl that minute water organisms absorb the C0 2 from the water in building up their life and leave the calcium carbonate to precipi- tate upon the twigs, plants, or bottom, or anything available. That the visible water plants serve mainly to precipitate the marl I can hardly believe as it clings to the dead twig as thickly as to the live. Moreover it fastens to wood that has not had life while in the water and could not have evolved carbon dioxide. There is every reason, however, to believe that these plants aid in increasing the content of calcium carbonate in the marl deposits even in the deepest water. In 50 feet of ^ater at the center of Horseshoe Lake, a long trailing vine was brought up from the bottom, these vines often winding about the augur. The vine had the distinct and very strong odor of pole cat. It was without doubt some one species of the Characem. The family are well knowrn for their high content of calcium salts. The Chora foetida as analyzed by Gustav Bischof is as follows: per cent. Ash of dried plant 54.84 Of this ash calcium oxide 54.73 Carbon dioxide 42.60 Such a plant dying would add a considerable portion of its sub- stance to the formation of a marl bed. It is not difficult to believe that the Characese are responsible for the growth of the marl bed when the actively depositing marl beds are seen to be covered thickly with a luxuriant growth of this plant. As they have stems and finest branches thickly coated with THEORIES OF ORIGIN OF BOG LIME OR MARL. 57 tlie calcium carbonate also, every crop each year forms an addition to the bulk of the marl bed. This luxuriant growth is often in water so shallow that in winter the ice must freeze down nearly to the bottom, enclosing the plants, stem and branch. In the spring when the ice loosens and is shifted into deeper water by the winds, a large number of these plants must be carried and deposited in mid lake. This will account in part for the distribution of marl in deep water. It is hardly deemed possible, however, that Char- acese are the sole cause of the growth of our vast beds of marl, for the following reasons: Actively depositing marl is found in the absence of these plants. In the absence of these plants the marl encrusts all objects around, dead or alive. Fully as thick an incrustation has been found upon dead twigs and old stubs stuck up in water by fishermen, in a lake nearly devoid of Characese, as in those the bottoms of which are covered with plants. Another very significant fact is that in cases where the plants themselves are taken from the water, they are found surrounded by a gelatinous scum. Where the marl is not depositing thickly this scum is nearly transparent, while on thickly depositing beds its surface is whitened by the presence of the particles of calcium carbonate. We must notice in connection with this another important fact. Where the marl is depositing upon a bare bottom, upon rocks and pebbles as noticed at Long Lake, Cloverdale, the accretions as deposited, have a pronounced inner lining of chlorophyl.* This green color does not show upon the outside of the incrustation which shows the white or gray color of marl. Such a deposit is very soft and breaks apart easily when in the water. When, how- ever, it is exposed to the air for some time it hardens so that it is difficult to tear apart. The pebbly concretions formed in some lakes are rather hard and gritty even under water and were even found in two cases at the depth of 6 to 10 feet in the marl bed, feeling like pebbles when struck in boring, yet the beds were al- most entirely free from silica in any form. Free sand was entirely absent. The very hard pebble like accretions seen in both in- stances were on the south side of the lake in question. It seems possible that other forms of plant life, invisible to the naked eye, also assist in the precipitation of the salts from the •Showing the presence of the blue green algae, referred to in Davis’ paper. I have noticed in Higgins Lake, the sand of the bottom continuously cemented in a thin layer about 1-10 of an inch thick, brown above and green below. L. 8-Pt. HI 58 MARL. water. While the Characem in shallow water are coated thickly with marl, in deep water there is no sign of the scummy or gelatin- ous covering, nor is the marl of anywhere near as limy a composi- tion, showing that the precipitation process is largely if not en- tirely inoperative in deep water. Yet in water 20 to 25 feet deep there is often if not always a fair marl. Sample No. 4 (sounding 10) at the center of Long Lake, Cloverdale, shows at the bottom of a 20 foot marl bed in 25 feet of water, 69.30$ calcium carbonate and 11.91$ organic matter. There is another very remarkable feature about very intensely hard water lakes. The waters are often as clear as crystal. Every dark particle of organic matter not only settles to the bottom, but is covered as well with the marly precipitate. The plants and debris of mid lake are buried by the marl as well as those nearer the shallows, but the deposit of marl must be much more rapid in the latter because of the greater content of calcium over organic matter which it always contains. There can be little doubt that purely chemical precipitation of marl from our dilute spring and lake waters would be impossible. The analyses of the Characese and their presence in such large numbers proves them to be surely responsible for a part of the com- position of the marl bed, especially in deep water. For in deep water the organic content is always very high and the forms of the water plants can always be distinctly traced, embedded and preserved in the impure marl. The analyses always show a great proportion of organic matter in deep water marl, or in most marls taken at great depths, whether in deep water or at the bottom of a deep bed or both. On the other hand a local precipitation takes place and that very actively. Moreover it takes place at or near the surface and very little in deep water. That was well shown by samples 5, 6 and 7 of the waters at Horseshoe Lake, Cloverdale district. These were in their order, analyses of water at 50 feet in mid lake, water on bottom at 10 feet in depth, and water at the very surface. From deep water to the surface the C0 2 escapes entirely and the carbonates are least at the surface also. The manner in which the marl is laid down also favors a precipi- tation process. Where the regularity of the bottom will allow it the marl is deposited so evenly that it is sometimes impossible to note any such variation in depth, the marl remaining very even over an extended area and then increasing or decreasing gradually. THEORIES OF ORIGIN OF BOG LIME OR MARL. 59 Of course in mam- cases, our lake bottoms being full of sudden jogs, the marl must vary also. As a rule it behaves much like an even deposit or sheet, leveling hollows and decreasing the abrupt- ness of sudden rises in the original bottom. See for measurements taken, Cloverdale, Central Lake, Rice Lake. Another point of significance is that near the surface, there are many samples of marl taken which have a content of 95^ calcium carbonate and sometimes but a fractional per cent of organic mat- ter, the latter indicating the proportion of plant tissue used in building up such a portion of the deposit. According to such analyses (see commercial analyses of marl in the appendix) the plant life remaining as organic matter would not be sufficient to account for the production of such very pure marl, being some- times but a fraction of a per cent. The following would then appear as the most plausible explana- tion of the manner of precipitation of marl. The mineral is washed from the soil and finds its way to our deep underground springs as a bicarbonated salt. These springs issue from the deep cuts and clefts left by the glaciers and called by us lake valleys. Analyses of the water and parallel experiment prove that the solution of carbonates and free carbon dioxide are in too small quantity to form a saturated solution and therefore cannot from purely chemical laws precipitate as marl on the bottom of the lakes. The very dense cold waters of the springs, whether they issue from bottom or sides of the lake, seek by their greater weight the deeper portions of the lake. They remain there with their burden of salts and C0 2 till the semi-annual overturning of the still water, when they approach the surface. When the water reaches the sur- face, it is warmed by the direct rays of the sun. If the place is shel- tered and the water is shallow, the bottom reflects the rays of the sun still further heating it. If in deeper water not all the rays of the sun are stopped as they are wasted in heating a greater depth of water to a less temperature. The warmer the water the better all plant life thrives in it. These plants are of two kinds, the larger fixed plants that may be seen without the aid of micro- scope and those invisible to the naked eye. The former or larger fixed plants live on the bottom and absorb a large percentage of the carbonates in their growth and also give 60 MARL. off free oxygen. The smaller plants are movable, being carried slowly through those portions of the water where they have suffici- ent sunlight and warmth to multiply rapidly. They must also give forth oxygen in large quantities, but as they must live toward the surface or warmer part of the water and be capable of reaching every particle of it, they form a more perfect oxygen carrier and serve to ‘furnish oxygen in a very thorough manner for precipita- tion of the salts from their weak solution. They could thrive only at or near the surface in deep water on account of the lack of heat and while supplying an even distribution of oxygen, would not thoroughly do so as in shallow water. The result of this is that in very shallow water the precipitating process is rapid wherever sunlight and warmth have made the very best conditions possible for the growth of plant life. Here the rate of formation of marl is more rapid and while plant remains are always found, the proportion of precipitated salts is greater than that remaining from the breaking down of gross plant tissue. The method of precipitation is a process of accretion. That is to say, every particle of organic matter, silt, etc., that finds its way into these waters where the process of marl making is very rapid, is surrounded by a coating of marl and sinks to the bottom, form- ing forever a portion of the deposit. The Characem and larger water plants containing lime in their formation are torn from their places by ice in winter, or perhaps to some degree by the action of the wind and disintegrate in the deeper parts of the lake, help- ing to form the deposit. The plants which themselves form in deep water are not encrusted to any extent with the marl and then when they disintegrate do not make as marly a formation. Moreover the dead drift of silt and other matter which must always fall into a lake is not in mid lake coated as thickly as in shallow water where the process is much more rapid. These particles sink to the bottom of the lake and form a part of the bed as they do in the shallows, but on account of the lack of precipitation upon them they add a higher amount of organic matter to the growing bed. As the water in turn throughout a depth of 50 feet is none of it heated so warm as on shallows where nearly all the heat is re- flected from the bottom, the finer more minute water plants do not multiply so fast or furnish as much oxygen. Where the heat is greatest at the surface they must, however, cause a deposit to some extent. It follows from these conditions that the marl must THEORIES OF ORIGIN OF BOG LIME OR MARL. 61 form more slowly in deep water and that its organic content must be greater.* As many of our lakes are filled with marl varying from twenty feet in mid lake to thirty or forty feet in depth on shallows or points near shore, or in marshes at that depth at the center, the following must have once been the condition : Our lakes were originally thirty or forty feet deeper when their basins contained no marl. The marl first deposited was deposited in much deeper water than this as the water level of our lakes has sunk greatly in the last few years. All the soundings made in deep water or to the bottom of deep deposits show them to be, one and all, of a more impure character than those in our shallows at the surface. The only exception found was a nearly pure shell deposit which at surface contained 90$ calcium carbonate and at a depth of 17 feet 92$. See Nos. 6 and 7, page 83. So universally does this rule apply that in one case a sudden change of former water level could be traced by a like sudden variation in the quality of the marl overlying the bottom. A broad glacial chain of dried lake beds extended from east to west. There were lakes above which emptied through a narrow stream which flowed over a bed. A line of soundings from north to south at right angles to the length of the system showed an extensive shal- low which originally lay on the north side. At about the center of the depression was the deep channel of the former body of water. Then at the south side was another area of shallows. Nearly all was dry land excepting the small stream named. Instead of the valley being filled nearly even with the deposit of marl and enclos- ing marsh growth, the deep channel was marked upon the surface as a sharp depression. The shallows were of finest marl ever seen, being nearly pure calcium carbonate with but a trace of organic matter and other salts. It formed a shallow deposit some six or seven feet in depth. The channel was very impure and formed a rather sharp line of contrast with the pure shoal marl, following the rule above mentioned regarding the quality of marl as ac- counted for by depth of water over it. The whole basin was once covered with water, the area of pure marl consisting of a terrace of shallows on either side. This de- posited pure marl in shallow water till it reached the surface and marsh growth sealed the deposit on the shallows, stopping its growth. The marl in deep water formed in a much more impure *See paper by Wesenberg-Lund, p. 68. 62 MABL. state on account of the greater depth of water over it. It would have filled to the surface, becoming purer with shallower water if the drainage had not in some way altered so that the water level sunk to the surface or near the surface of the marl and organic matter in the shape of marsh growth choked and sealed the de- posit. This peculiar suddenness of change in quality and forma- tion was traced for a half mile to the first chain of upper lakes. The channel broadened in places, but its surface never showed other than a silt formation, the marsh at the present time being in process of sealing the deposit. The upper lake of the chain, how- ever, while it showed a gradual decrease in quality with increase of depth was actively depositing in the shallows at the upper end. We would conclude that in former times the process of marl for- mation was much slower on account of the greater depth of water and that our fall of water level of late years has hastened the pro- cess, bringing deeper water nearer the surface and heat and sun- light. We also conclude that the great clearness of our lakes is due to the fact that every particle of floating silt and dust and matter no matter how large or small, is surrounded by the fast depositing marl and buried in the deposit. It is noticeable that many lakes where the process has ceased and the marl is being covered with silt, show a very dirty reddish water due to particles of deteriorating organic matter. Yet these lakes are fed by springs and have outlets. It is difficult always to account for the presence of marl in one lake and its absence in another. In most cases there is found a difference in water supply. Mud Lake and Long Lake, Cloverdale district, were one soft water, and the other hard. The former was fed by surface soft water springs and the latter by deep water springs. The wells near each showed the same difference in hard- ness of water. In portions of the State where there are no hard water springs no marl is found. Such were said to be the condi- tions surrounding the limestone district about Escanaba. A pros- pector who had explored carefully said that there were no marl lakes within thirty miles. The hard water was tapped only by' the deepest artesian wells. It was noticed where two lakes were near enough together to be compared that the one indenting the general outline of the country deepest and tapping the most hard water springs, contained the deepest deposit. THEORIES OF ORIGIN OF BOG LIME OR MARL. 63 There is yet another circumstance which must be accounted for. This is the presence in one part of the lake of a marl deposit which may taper off to a sandy or clay bottom not covering the whole lake bed. In the first place it will be noticed that in a lake not covered entirely with marl, it favors with its presence the bayous, points, and shallow water and in most instances, though not al- ways, avoids deep water. As light and heat are always necessaries of plant life the facts of the location of the deposit in shallow water in the presence of the same is very good argument for the theory of vegetable origin. But there is still a further fact to account for. Even in shallows a bed may end or taper to the orig- inal bottom, generally becoming toward the edges much more highly organic in its nature. This is illustrated by the fact that sometimes at one end of a lake, generally though not always the upper end, the marl is bare or has not ceased depositing and at the lower end becomes a deposit of lake silt, or in another case the marl is bare of silt at one end, though covered with water and is at the other end covered by a few inches to a few feet of silt. In other words the marl has changed its position for depositing or has continued to deposit at one end and has ceased- entirely for some time to deposit at the other end and the deposit there is sealed to some depth with silt, over which there lies several feet of water. A good illustration of this was seen at Central Lake. The depth at both ends to the original bottom was nearly the same. The quality of the marl was about the same for the same depth, but the marl had ceased depositing at one end and had con- tinued actively at the other. We must conclude from this that the conditions for successful growth of the marl producing plants of our lakes change in different parts of the lake, causing a more or less permanent cessation of the process of marl making. At Portage Lake, Onekama, this process seems to have been inter- rupted at intervals and continued again according to the layers of marl and organic marsh growth alternating. However, as this was the only instance seen of the kind, it would not be safe to assume from the instance of one lake, that such was the rule. It can scarcely be argued that marl is the result entirely of the breaking down of the structure of gross plant growth for the same reason that shells cannot be said to account for the formation of all marl. At a depth of thirty-five feet stems and branches of 64 MARL. small size may be well defined in samples taken as can the forms of small shells. Wood of a more fibrile texture is preserved in a nearly fresh state and the grain can be clearly separated. It can hardly be said that different parts of the same plant have deterior- ated at such a different rate as to leave in one portion a nearly perfect branch or shell and right beside it a marl formation that cannot be found to resemble plant tissue or anything else except- ing an amorphous form of mineral. The lack of any finely pre- served lime formation of the tests of minute animals or the forms df fresh water plants,* also discourages the idea that the bodies of the same have died and formed the deposit. The clearest ex- planation would therefore seem to be that of a chemical precipitate brought about by plant life both great and small, abstracting C0 2 , and acting where conditions for its existence are most favorable. One of the strongest of reasons why the purely chemical theory is not true is lack of marl in some shallows and its presence in others. The lime bearing water must be distributed evenly to all shallows and should precipitate upon all at an equal depth. This is often con- trary to fact, while on the other hand it would be possible for a local precipitation to be brought about in the presence and only in the presence of water plants producing oxygen. As these views of the subject are nearly if not exactlyf the same as those of Prof. Davis, given in another portion of this work, it has not been thought necessary to repeat his chain of evidence or any of his ideas except to bring out those points in the constitution and location of marl beds which would seem to prove the same idea from different facts of observation. *But see Davis’ observations, pp. 74 to 80. L. tThe main difference between Mr. Hale and Prof. Davis is that the former is more inclined to look to microscopic plants and to the abstraction of C0 2 by plant life generally as inducing a chemical precipitation favored by light and heat. L. CHAPTER V. A CONTRIBUTION TO THE NATURAL HISTORY OF MARL. BY C. A. DAVIS. § 1. Historical introduction. Botanists have long been familiar with the fact that, in some regions, aquatic plants of all, or nearly all, types are covered with a more or less copious coating of mineral matter, while in other localities the same types of plant life are free from any trace of such covering. In New England, for example, plants growing in the water are generally without such coating, while in Michigan and adjoining states it is generally present. In many lakes and streams the mineral deposit on the stems and leaves of the higher plants is very noticeable, and nearly all vegetation growing in the water is manifestly an agent of precipitation of mineral matter, Various writers in Europe* and Americaf have called attention to the influence of the low types of plants growing in and around hot springs and mineral springs, on the formation of silicious sinter, calcareous tipa, and other characteristic deposits of such springs, and the connection between the beds of calcareous tufa which are sometimes formed about ordinary seepage springs whose waters carry considerable calcareous matter in solution and certain species of moss has been suggested, but so far as the writer knows, no one has given attention to the possible relation of vegetation to the more or less extensive beds of the so called marl, found about, and in, many of the small lakes in Michigan and the adjacent states. As has been pointed out elsewhere, this “marl,” more properly lake lime, is made up principally of nearly pure calcium carbonate, “carbonate of lime,” with greater or less admixture of impurities. When dry and pure it is white or slightly cream colored, nodular, coarsely granular to finely powdery, very loosely *Cohn: Die Algen des Karlsbader Sprudels, mit Rucksicht auf die Bildung des Sprudel Sinters: Abhandl. der Schles. Gessell., pt. 2, Nat. .1862, p. 35. fWeed: Formation of Travertine and Silicious Sinter by the Vegetation of Hot Springs. U. S. Geol. Surv., IX, Ann. Rept., p. 619, 1889. 9-Pt. Ill 66 MARL. coherent and effervescing freely in acids. On dissolving it, particles of vegetable and other organic and insoluble matter are found scattered through the solution. § 2. Ultimate sources. The ultimate source of this material, except the vegetable mat- ter, is, undoubtedly, the clays of glacial deposits and like disin- tegrated rock masses. These clays are rich in finely divided lime- stone and in the softer rock-forming minerals, some of which con- tain calcium compounds. Percolating water, containing dissolved carbon dioxide, the so called carbonic acid gas, readily dissolves the calcium and other metallic salts up to a certain limit. The water with the dissolved matter in it runs along underground until an outlet is reached and issues in the form of a spring. This, in turn, uniting with other springs forms a stream which runs into a lake, carrying along with it the greater part of its mineral load. If the amount of carbon dioxide contained in the water is consider- able, some of it will escape on reaching the surface, because of decrease of pressure, and with its escape, if the saturation point for the dissolved mineral matter has been reached, a part of this matter must be dropped in the form of a fine powder, as the water runs along over the surface. Theoretically, then, some, if not a great part of the dissolved matter, should be thrown down along the courses of the streams which connect the original outlets of the water from calcareous clays and lakes where marl occurs, and we should find the marl occurring in small .deposits along these streams wherever there is slack water. Moreover, we should ex- pect the waters of these springs and streams to show more or less milkiness on standing exposed to the normal pressure of the atmosphere at usual temperatures. Actually, however, none of these phenomena have been noted and we infer that there is not a large amount of carbon dioxide, and not an approach to the satur- ation point for calcium bicarbonate, in the springs and streams feeding marly lakes.* § 3. Alternative methods of deposition. We are then left, among others, the following alternatives, ex- planatory of marl formation: (1) The marl is not being formed under existing conditions, but has been formed in some previous time when conditions were not the same as now. (2) The amount of dissolved salts is so small that the saturation point is not ap- *This point is considered more extensively later. L. CONTRIBUTION TO THE NATURAL HISTORY OF MARL. 67 proached until after the lakes are reached and the slow evapora- tion added to the reduction of the amount of dissolved carbon dioxide in the water brings about deposition of the mineral salts. (3) Some other cause, or causes, than the simple release from the water of the solvent carbon dioxide must be sought. The first of these suggestions is met by the fact that marl is found in lakes at and below the present level of the water, and that it extends in most of them to, or even beyond, the very edge of the marshes around the lakes, and over the bottom in shallow parts of living lakes, even coating pebbles and living shells. (2) The water of lakes with swift flowing and extensive outlets, such as most of our marly lakes have, is changed so rapidly that little if any concentration of a given volume of water would occur while it was in the lake, and there is no probability that any of the lakes visited by the writer have ever been without an outlet. Indeed many of them have outlets which occupy valleys which have been the chan- nels of much larger streams than the present ones. Moreover, definite measurements which, however, are subject to further in- vestigation, have been made, which show that the volume of water flowing out of these lakes is practically the same as that flowing into them, i. e., the loss by evaporation is too small a factor to be taken into account. Farther, recent investigations* have shown that calcium, as the bicarbonate, is soluble to the extent of 238 parts in a million, in water containing no carbon dioxide. As most of our natural waters, even from limy clays, contain no more than this amount of this salt, even when they carry com siderable free carbon dioxide, and many analyses show a less, amount of it, the fact becomes plain that even if the carbon dioxide were all lost there would be no precipitation from this cause. (3) Considering these objections as valid it seems fitting to ex- amine into the possibility of the plant and animal organisms living in the waters of the lakes being the agents which bring about the reduction of the soluble calcium bicarbonate to the insoluble car- bonate even in waters low in the amount of dissolved mineral matter, and containing considerable carbon dioxide. That mollusks can do this is shown by the fact, which has fre- quently come under the writer’s notice, that the relatively thick and heavy shells of species living in fresh water are partly dis- *Treadwell and Reuter: Ueber die Loeslichkeit der Bikarbonate des Calciums und Magnesiums. Zeitschrift fur Anorganish-Chemie, Vol. 17, p. 170. Summarized elsewhere in this report. 68 MARL. solved and deeply etched by the action of carbonic acid after the animals have, by their processes of selection, fixed the calcium car- bonate in their tissues, precipitating it from water so strongly acid and so free from the salt that re-solution begins almost im- mediately. No natural water seems so free from calcium salts that some species of mollusks are not able to find enough of the neces- sary mineral matter to build their characteristic shells. While some limited and rather small deposits of marl are pos- sibly built up, or at least largely contributed to, by molluscan and other invertebrate shells,* the deposits which are proving commercially valuable in the region under consideration, do not contain recognizable shell fragments in any preponderance, al- though numerous nearly entire fragile shells may be readily washed or sifted from the marl. The average of quantitative de- terminations of the shells and shell debris in three samples of marl from widely separated localities was less than one per cent of the entire weight of the marl and of these the highest contained but a trifle over one per cent, 1.04$. The conditions under which marl are found are such that the grinding of shells into impalpable powder, or fine mud, by strong wave action is improbable, if not impossible, for exposed shores and shallow water of considerable extent are necessary to secure such grinding action, and these are not generally found in connection with marl. We are, then, reduced to the alternative of considering the action of plants as precipitating agents for the calcium salts. It has been shown already that plants generally become incrusted with mineral matter in our marly lakes, and it is easy to demon- strate that the greater part of the material in the incrustation is calcium carbonate. It is also easy for a casual observer to see that in many cases the deposit is not a true secretion of the plants, for it is purely external, and is easily rubbed off, or jarred off from the outside of the plants in flakes, while the tissues be- neath show no injury from being deprived of it, and again as has already been pointed out, the same species of plants in some sections of the country do not have any mineral matter upon them. It has also been remarked in a recent important paper, f that the amount of the incrustation varies with the depth of water in which the plants grow, i. e., the amount of light they receive, *C. Wesenberg-Lund: Lake-lime, pea ore, lake-gytje, Medd. fra Danskgel Forening U. Copenhagen, 1&01, p. 140. tC. Wesenberg-Lund, p. 156. CONTRIBUTION TO THE NATURAL HISTORY OF MARL. 69 the season, and the roughness of the surface water, waves causing the incrustation to break up and fall off. The deposit is formed incidentally by chemical precipitation upon the surface of the plants, probably only upon the green parts, and in performance of usual processes of assimilation of the plant organism. § 4. Cause of deposition upon aquatic plants. All green plants, whether aquatic or terrestrial, take in the gas, carbon dioxide, through their leaves and stems, and build the carbon atoms and part of the oxygen atoms of which the gas is composed into the new compounds of their own tissues, in the pro- cess releasing the remainder of the oxygen atoms. Admitting these facts, which are easily demonstrated by any student of plant physiology, we have two possible general causes for the formation of the incrustation upon all aquatic plants. If the calcium and other salts are in excess in the water, and are held in solution by free carbon dioxide, then the more or less complete abstraction of the gas from the water in direct contact with plants causes precipitation of the salts upon the parts ab- stracting the gas, namely, stems and leaves. But in water con- taining amounts of the salts, especially of the calcium bicarbonate, so small that they would not be precipitated if there were no free carbon dioxide present in the water at all, the precipitation may be considered a purely chemical problem, a solution of which may be looked for in the action upon the bicarbonates, of the oxygen set free by the plants. Of these, calcium bicarbonate is the most abundant, and the reaction upon it may be taken as typical and expressed by the following chemical equation : CaH 2 ( 003)2 “f" O = H 2 0 -j- CaC0 3 -j- 0O 2 -(- O calcium ) . , , \ calcium . \ carbon , bicarbonate [ + oxygen = water+ j carbonate + j dioxide + oxygen in which the calcium bicarbonate is converted into the normal car- bonate* by the oxygen liberated by the plants, and both carbon dioxide and oxygen set free, the free oxygen possibly acting still farther to precipitate calcium monocarbonate. It is probable that the plants actually do precipitate calcium carbonate, both by abstracting carbon dioxide from the water and freeing oxygen, which in turn acts, while in the nascent state, upon the calcium salt and precipitates it, but in water containing rela- tively small amounts of calcium bicarbonate the latter would seem *Which is only very slightly soluble, 100 parts to the million. 70 MABL. to be the probable method. In all likelihood these methods for accounting for the precipitation of calcium carbonate will suffi- ciently explain the ordinary thin and relatively insignificant in- crustation which is found on the higher plants, but for the algae it is doubtful, or even improbable that they account for all the facts, as will be shown further on. The calcium salt is deposited in minute crystals, and by the aggregation of these crystals the incrustation is formed on the plants. The crystals are distinguishable as such only for a short time on the newer growth of plants, but the incrustations are said to show a recognizable and characteristic crystalline structure when examined in thin section under a compound microscope with polarized light. § 5. Relative importance of Chara (Stonewort). Not all aquatic plants in the same lake seem equally active in the precipitation of mineral matter. Not even all species of the same genera, although growing side by side, will be coated equally, a fact which seems to indicate some selective metabolic processes not understood. Considering the precipitation of calcium car- bonate by plants as established, even if the exact physiological and chemical processes by which this precipitation is brought about, are not yet worked out fully, it is still necessary to consider the constancy of the action and the sufficiency of the agency to produce the extensive deposits of marl which are known. If one confines his studies simply to the seed-producing plants and other large vegetable forms which are conspicuous in lakes during the summer season, while he will find them covered with a thin coating of manifestly calcareous matter, he will at once be convinced that such work as these plants are doing is but a small factor in the total sedimentation of the lake. On the other hand, if a visit be made to a lake in early spring or late fall, all plants of the higher types will not be found, so that it becomes apparent that this agency is merely a seasonal one and works intermittently. Farther study of the plants of the same body of water, however, shows that the algae, the less conspicuous and entirely submerged plant organisms must be taken into account before we finally abandon plants as the agents of precipitation. Of these, two groups, differing widely in structure, habits and method of precipi- tation, will be found. The first and most conspicuous, and probably the most important as well, is the Cliaraceae or Stoneworts. These CONTRIBUTION TO THE NATURAL HISTORY OF MARL. 71 plants are well known to botanists, and may readily be recognized by their jointed stems, which have at each joint a whorl of radiat- ing 'leaves and branches, which are also jointed. Both stems and branches are made up of long tubular cells,* extending the length of the internodes or spaces between the joints. There is a large cell in the middle and a series of smaller ones around it, their walls touching but not usually compressing each other, so that the cylindrical shape of each cell is generally maintained and the cross section of the stem appears like a relatively large ring surrounded by a single row of small ones tangent to each other, and to the central large one. The outer, or cortical cells, are usually more or less spirally twisted around the large central one, and all the cells are thin walled and delicate, the plants containing no thick walled tissue or cells of any sort. The structure of the plant is so well marked and peculiar, that it cannot well be mistaken for that of any other, and so makes it easy to identify even small fragments of it. In some species the stems and branches are covered with a thick coating of mineral matter, are almost white, and very brittle because of this covering. These plants not only grow near the sur- face in shallow water of our ponds and lakes where the bottom is unoccupied by other plants, but in the deeper parts as well, and, as they thrive where light is feeble, they continue to grow throughout the year, although in winter they must grow less rapidly than in summer, because ice and snow on the surface of the lakes make less favorable light conditions. Analytical Tests. The sufficiency of these plants alone to fix and deposit calcium carbonate in large quantities is indicated by the following: In November, 1899, the writer collected a large mass of plants of Chara sp. ?, from which five stems, with a few branches, were taken at random and without any particular care being taken to prevent the brittle branches from breaking off. The stems were each about 60 cm. long, and after being dried for some days, they were roughly ground in a mortar and dried for one-half hour at 100 degrees C., dried and weighed until the w T eight was constant. The weight of the total solid matter obtained in this way from five plants was 3.6504 grams, 0.73 grams per plant. This was treated with cold hydrochloric acid diluted, twenty parts of water to one *See Plate XVI. 72 MARL. of acid, filtered, washed, and the residue dried at 100 degrees C., on a weighed filter paper, until weight was constant. The weight of insoluble matter was 0.5986 grams; of the total soluble matter 3.0518 grams, or .6103 grams per plant. In the lake from which the material analyzed was derived >from 50 to 80 plants were counted to the square decimeter of surface in the Ohara beds. A partial quantitative analysis of material from the same source, but using stronger acid to effect solution (hydrochloric acid, diluted with four parts of water,) gave the following results: Insoluble residue 11.19$ Iron and aluminum oxides 0.722 Calcium carbonate 76.00 Magnesium carbonate 2.359 Soluble organic matter obtained by difference 9.279 The composition of the insoluble residue was obtained by heating the residue to redness in a platinum crucible for one-half hour, and the 11.19 per cent of this matter was found to consist of: Combustible and volatile matter 9.243$=82.6$ Mineral matter 1.947 =17.4 The mineral matter was found to be: Silica 1.787$=92.4$ Not determined 160 = 7.6 Microscopic examination showed the silica to be largely com- posed of whole and broken tests of diatoms, minute plants which secrete silicious shells and attach themselves to the Chara stems and branches. The mineral matter obtained in this analysis, reduced to parts per hundred, gives the following: Calcium carbonate 93.76 Magnesium carbonate 2.93 Silica and undetermined mineral matter 2.40 Iron and aluminum oxides 89 This, with a small decrease in the mineral matter and a small amount of organic matter added, would be the composition of CONTRIBUTION TO THE NATURAL HISTORY OF MARL. 73 ordinary marls, and would be a suitable sample to consider in con- nection with Portland Cement manufacture. The large amount of silica may be explained by the fact that the material analyzed was collected at a season when diatoms are especially abundant. The following is a copy of an analysis of the marl from the beds lying about the lake from which the Chara plants were taken. This analysis was made by L. G. Leltz, chemist for the Alma Sugar Company, season of 1900-1901: H 2 0 and organic matter 7.438$ Sand (insoluble silica) 0.104 Carbon dioxide 38.48 Calcium oxide 52.28 Iron and aluminum oxides 0.61 Magnesium oxide 0.455 Sulphur trioxide 0.32 Soluble silica 0.0532 Chloralkalies 0.07 Phosphorus pentoxide 0.12 99.9302 It may be well to call attention to the fact that in many marls, especially those of large deposits, which the writer has examined chemically, the silica has been found to be mainly in the form of diatom shells, and hence, because of the small size and great delicacy of structure, it is available as a source of silica for calcium silicate in cement making. If such deposits as are made up largely of diatom shells were adjacent to marl beds, it is possible they might be considered as clay and be used in cement making. Some of the silica in marl was found by mechanical analysis to consist of grains of white quartz of rather large size for sand. These may have been carried into the lake by winds, by drifting ice, by fish or by birds. The fact that these sand grains were white and of a rounded character, would point to the fish or to birds, which use such matter in their gizzards, as most probable agents of transportation, especially as no dark colored grains were found. From the above considerations, it is evident that both because of the quality and quantity of its works, Chara may be considered an important agent in marl production, and it only becomes necessary to account for the chalky structure of the deposits to make the chain of evidence complete. 10-Pt. Ill 74 MARL. All algse are plants of very simple structure, without tough or complicated tissues. Chara stems and branches are made up of aggregations of thin walled cells, and when the plants die the cell walls must rapidly decay and the residue of lime be left. In a laboratory experiment to determine this factor, it was found that a mass of the broken-up plants in the bottom of a tall glass vessel filled with water became decomposed very quickly, giving the char- acteristic odor of decaying vegetable matter, and after a few weeks all organic matter had disappeared, leaving the incrustations in tubular, very brittle fragments. In studying the structure of marl, the writer fias found that near the top of the beds there is usually a “sandy,” or even a coarsely granular structure. This is noticeable at times, at all depths from which the samples are taken, i. e., in some cases it extends through the bed. Close examination of such marl shows that this coarse- ness is due to the remains of the characteristic Chara incrustations, and that the “sand” and other coarse material is made up of easily identifiable fragments of the coatings of stems and branches of the plant. The presence of such coarse matter near the top of the beds may be considered due to sorting action of the waves, and such surface currents as may be caused in ponds and small lakes, in shallow water, by wind action. If these agents are effective in producing the coarser parts of the deposits they may also be con- sidered so in connection with the finer parts as well, for the matter produced by the breaking and grinding up of fragments is held in suspension for a longer or shorter time, carried about by currents, and finally sinks to the bottom in the quieter and deeper parts of the lakes. This has not been left, however, as mere conjecture, but a series of mechanical analyses of typical white marl from different localities was made. The method of analysis used was a simple one, a modification of the beaker method, used in soil analysis. The samples, chosen at random from large average specimens from the deposits under investigation, were dried in an air bath at 110° C., for sufficient time to remove any included moisture, and weighed. Each sample was then mixed with distilled water in a large beaker and thoroughly stirred with a rubber tipped glass rod, care being taken to keep up the stirring until all lumps caused by the adhesion of the finer particles to the coarser had been broken up. Care was CONTRIBUTION TO THE NATURAL HISTORY OF MARL. 75 also taken that no more crushing should take place than was absolutely necessary. After disintegration of all lumps was accomplished, the water with the finer particles in suspension, was poured off into another beaker, and fresh water added to the first and the material again stirred. This was continued until the water was nearly free from the finer matter and became clear on standing a short time. The coarse material left in the bottom of the beaker was dried, sorted into various grades by a series of sieves and each grade weighed. The finer material was also sorted by stirring, settling and decanta- tion, and that of different degrees of fineness dried and weighed. The finest matter was separated from the water by filtering through a weighed filter and the water concentrated by evaporation and again filtered to remove any of the calcium carbonate dissolved in the various processes, and the final residue of water was evap- orated in a watch glass and weighed. An exceedingly interesting feature of this latter experiment was the finding of a water soluble calcium salt, in small quantity it is true, but still easily weighable, and not to be neglected. The results of such an analysis of a sample from the Cedar Lake marl beds gave the following results. The sample was the one of which a chemical analysis is given above, and was taken from a hole made with a spade by cutting away the turf over the marl, then taking out sufficient marl to be rea- sonably sure that there was no peat or other surface matter pres- ent, and the sample used from a spadeful thrown out from two or three feet below this. From this sample about thirty grams were taken and treated as described above, and after the coarser material had been separated from the finer by washing and drying, it was passed through a set of standard guage sieves 20, 40, 60, 80 and 100 meshes to the linear inch, after which all shells and recog- nizable shell fragments, sand grains and vegetable fragments up to the 60-mesh siftings were removed and weighed separately. The following grades of material were obtained by this sorting: (1) That too coarse to pass through the 20-mesh sieve, (2) that held by the 40-mesh sieve, (3) that held by the 60-mesh, (4) that held by the 80-mesh, (5) that held by the 100-mesh, (6) that which passed through 100-mesh, (7) that which was filtered out, (8) water soluble salts, (9) shells, shell fragments, etc. 76 MARL. Analysis (1) is the result of the analysis made and the material graded as described: Cedar Lake Marl. Littlefield Lake Marl. Coldwater Marl. Residue from dead Chara. 1. 2. 3. 4. Grade (1) 32.25% 31.52% 0.36% 1.12% “ (2) 6.06 14.48 3.53 24.43 “ (3) 7.58 12.76 6.51 14.63 “ (4) 2.90 2.56 3.34 8.26 “ (5) 4.81 6.74 6.44 7.81 “ (6) 15.64(1) j- 30.42 | 28.99 j 49.12 | 33.83 “ (7) “ (8) j 30.52 0.27 1.02 0.39 “ (9) 0.28 1.04 0.69 0.12 100.04 99.89 100.00 90.59 (i)In this case determined by drying down the residue and weighing. A second analysis was made from a specimen made up of twenty samples taken by boring with an augur over about one-half the deposit at Littlefield Lake, Isabella County, most of the samples coming from a depth of at least twenty feet below the surface of the deposit. This analysis is given as 2: Grade cc cc cc cc CC Cl Cl Cl ( 1 ) ( 2 ) (3) (4) (5) ( 6 ) (7) ( 8 ) (9) = 31.52# = 14.48 = 12.76 = 2.56 = 6.74 = 30.42 = 0.27* = 1.04 99.89 A third sample from the holdings of the Michigan Portland Cement Company, at Coldwater, a fine high grade white marl, very powdery, gave Analysis 3: Grade (1) Cl (2) Cl (3) 1C (4) Cl (5) Cl f6) Cl (7) Cl (8) Cl (9) Soluble matter and loss by difference = 0.36# ss 3.53 = 6.51 = 3.34 = 6.44 — 28.99 =49.12 not determined. = 0.69 = 1.02 100.00 *The soluble matter contains a certain undetermined amount sodium and potassium salts as well as soluble calcium compound. CONTRIBUTION TO THE NATURAL HISTORY OF MARL. 77 These samples represent (1) the central, (2) the north central and (3) the southern parts of the Lower Peninsula of Michigan, respec- tively, and may be taken as typical of the marl deposits of Mich- igan. When it is stated that, in general, it is easily possible to recognize with a simple microscope, particles which are held by the 100-mesh sieve or even those which pass through it, if the finer matter has been carefully separated by washing, as characteristic Chara incrustation, or Schizothrix concretions, it will be seen that these results show conclusively that a large part of the marl from these three samples is identifiable as of algal origin and studies of the marl from other localities give similar results. The Cold- water sample (3) was exceedingly fine in texture, and it was diffi- cult to avoid loss in sorting and weighing, as every current of air carried away some of the particles, and some also adhered to the sieves and weighing dishes, in spite of all the usual precautions against such loss. Even this sample shows nearly fifty per cent of easily identifiable Chara incrustation. The fineness of the particles in a given marl bed varies much in different parts of the bed and the degree of fineness is probably largely dependent upon the conditions of current and wave action under which the bed was formed as noted in another place. This fact was noted at Little- field Lake, where samples of marl were collected along exposed shores near the wave line, which were ninety-five per cent coarse fragments of Chara incrustation and Schizothrix nodules, while in other parts of the shore line the marl was of such fineness that it was like fine white clay. References in Literature. Fragments of the Chara incrustation are generally easily rec- ognized, even when of minute size, because they preserve, usually very perfectly, but sometimes less so, the peculiar form of the stem, branches, leaves and fruits of the plant. This fact has led various authors, both geologists and botanists, to note the occur- rence of “fossil” Chara stems and fruits in the beds of lakes and even in marl beds. Sir Charles Lyell* as early as 1829 described a marl bed in Forfarshire, mentioning as especially interesting the finding of “fossil” Chara fruits and stems. In two editions of the “Principles of Geology, ”f which have been consulted, the same *Lyell “On a recent formation of fresh water Limestone in Forfarshire.” Trans- actions Geolog-. Society, 2, p. 241, 1829. f6th Ed., Vol. 3, p. 350, 1842 ; 9th Ed., p. 766-7, 1853. 78 MARL. writer points out the importance of the remains of Chara to the geologist in characterizing entire groups of strata, and describe^ and figures the fruits and stems of recent species Chara hispida from Bakie Loch, Forfarshire. He also mentions the occurrence of Chara in abundance, in several lakes in New York State. Geikie* mentions the occurrence of Chara as a true fossil in the beds of “a form of travertine from which fresh water shells and a rich assemb- lage of plants have been obtained.” These beds are “lower Eocene, the limestones of Rilly and Sizanne, Basin of Paris.” Chara Lyelli fruits are figured. Kernerf says, “The spore fruits of Stoneworts (Characese) have been found over and over again inclosed in these formations of lime.” He points out also that it is possible for calcareous strata of great depth to be produced by plants in fresh water. SchimperJ, Solms-Laubach§, Seward|| and Wesenberg-Lundfl all mentioned these plants as agents of deposition of lime formations, the latter especially showing the plants able to produce extensive deposits of what he terms “Characee-lime” in the lakes of Denmark. Mosely**, in speaking of the deposits of “tufa” about the remark- able springs near Castalia, Ohio, mentions the fact that the deposit “is composed mostly of petrified Chara.” Even when this structure is destroyed, as may be the case with the thin and incomplete incrustations, it is frequently possible to recognize fragments of the tubes with the compound microscope. Finally in Chara as in other plants the incrustation is distinctly crystalline in the ultimate form of the constituent particles, and when it has disintegrated, the crystals and their fragments are found to constitute a large per cent of the finer particles of the resulting marl. On the growing tips of the younger branches and leaves of Chara, numbers of isolated crystals of calcium carbonate may be seen, and farther back the crystals become more numerous, then coalesce into a thin fragile covering, and finally on the lower part of the plant the covering becomes dense and thick. It is evi- dent, therefore, that the decay of the younger parts of the plants would furnish a mass of more or less free or loosely aggregated *A. Geikie; Text-book of Geology, 2nd Edition, p. 853, p. 859, 1859. fKerner and Oliver; Nat. Hist, of Plants, Vol. 1, p. 261. ISchimper, W. Ph. “Traite de Paleontologie Vegetale,” Vol. 1, p. 216, 1869. §Solms-Laubach, Fossil Botany, pp. 36-37, 1S91. HSeward, A. C., Fossil Plants, V, p. 69, pp. 222-228, 1898. tiWesenberg-Lund. C., loc. cit., pp. 155-156. **Moselv, E. L.. Sandusky Flora, p 87, Ohio State Academy of Science, special papers, No. 1, 1899. CONTRIBUTION TO THE NATURAL HISTORY OF MARL. 79 crystals of microscopic size which would retain their crystalline form, in some degree at least, for an indefinite time and be recog- nizable, hence the presence of these micro-crystals in marl is an- other indication of the origin of the deposits. Source of Thick Crusts. The larger fragments of Chara incrustation as found in marl are frequently much thicker and heavier than those which occur among the fragments of recent origin, namely those obtained from any part of living, vigorously growing Chara from beds of the plant existing in the ponds from which the marl may have been obtained. While the subject needs further investigation, it is probable that such thickened incrustations have originated in several ways, the principal ones being, if the writer’s notes have any bearing on the subject, as follows: First, On short, stunted plants that grow for a long time on unfavorable soil, such as sand, or pure marl. Such plants have relatively very short internodes, and generally thick incrustations. Second, From the growth of the lime secreting blue-green algse, such as Schizothrix, Zonotrichia, etc., either upon living Chara, or upon fragments of broken incrustation as a nucleus. Third, From the inclusion of the fragments within the nodules formed by the growth of the blue-green incrusting algse in shal- low water and the subsequent destruction of the nodules by wave or other disintegrating action, in which case, the thickened frag- ments may be left either free, or attached to other material. In this way, several fragments may be cemented together and such aggregations have been observed by the writer. Fourth, By the deposition of calcium carbonate on fragments of incrustations, a deposition caused by the decomposition of soluble organic calcium salts, left free in the water by the decay of dead Chara plants, through the reducing action of chemical compounds derived from the decay of organic matter, or the growth of bacteria, or both. Fifth, By the deposition in more or less coarsely crystalline form of the calcium carbonate which is dissolved by water percolating through the marl. This is probably considerable in amount and takes place in a manner analogous to, if not identical with, the formation of concretions in clays and shales. It is probable that in this way, the crystals may be formed w T hich rather rarely are 80 MAUL. found filling the cavities left by the large axial cells in Chara incrustations. The fact that in the great majority of cases these cell cavities are entirely empty, or simply mechanically filled with fine particles of marl, is the most serious objection to considering that this form of chemical precipitation is an important one in the history of marl, but that it is occasionally operative is most prob- able. Sixth, It is possible that the thick incrustations may have been formed at some earlier period in the history of the lakes when con- ditions were more favorable for the development of Chara and its activities were greater. This is not probable, however, for the thick incrustations are frequently found from the surface of the marl beds throughout the deposits. A check analysis was made of a specimen of material made up from the washings and fragments of a mass of Chara plants col- lected from Cedar Lake, and allowed to die slowly, and break up in water kept cold and fresh by conducting a small stream from the hydrant through it. The plants gradually died, broke up and set- tled to the bottom of the containing vessel, and seemed to undergo farther disintegration there, eventually forming a relatively finely divided deposit which was of rather dark color when wet. A quantity of this was dried at 100 degrees C., some of the larger and longer fragments of stems were removed and the residue weighed and subjected to the same treatment as the marl samples. Ten grams was the amount taken, and the analysis yielded the re- sults given by No. 4, page 76. It will be seen that nearly as much fine matter was present in this material as in the finest of the marls analyzed and that the finer grades of sifted material are quite as well represented, as in the finer marl. The material is somewhat more bulky for a given weight and is perhaps slightly darker in color, but not much more so, than many samples of marl. Grade for grade it is identical in appearance and structure to the marl samples, and the only pos- sible difference that can be detected is the slightly greenish tint due to the organic matter present in the plant residue. It is also noticeable that the larger pieces do not show as thick an incrusta- tion as do larger pieces from the marl samples, and, of course, Schizothrix and other coarse matter is not present. It will be seen by inspecting the analyses that shells and recog- nizable shell fragments are but a very insignificant part of the CONTRIBUTION TO I HE NATURAL HISTORY OF MARL. 81 total quantity of the marl. It is surprisingly small when all things are taken into account. While it is probably true that not all the minute shell fragments have been separated in any of these analyses, it is also true that the weight of such particles as were overlooked, is more than counterbalanced by marl fragments, which are included within the cavities of the whole shells and adhere to both broken and whole shells, in crevices and sculpturings, in such a way as to refuse to become separated in the processes of washing out the marl. The whole shells are mainly small fragile forms, many of them immature, and it is evident that they would be broken by any action that would crush the Chara incrustation. § 6. Marl beds without Chara. As is easily observed in many marl lakes, and as has been pointed out to the writer by several students of lake life, marl beds are often found in parts of lakes in which there are no well developed beds of Chara. At least two explanations of this may be offered without appeal- ing to other modes of marl formation, both of which may be ap- plicable, either independently or together, in individual cases. The first of these is that such beds of marl were formerly occupied by Chara, but for some reason or reasons, the conditions for growth became unfavorable and the colonies disappeared, or became insig- nificant and escaped notice. The second explanation takes into account the action of waves and currents upon the deposits near thriving growths of Chara and assumes that the more remote marl deposits may be the result of such action combined with the trans- porting power of the surface and other currents which may exist in the lakes. In support of the former consideration are the notes of Dr. Henry B. Ward on Pine Lake in the Traverse Bay Region of Mich- igan. He says:* “Pine Lake has undoubtedly undergone some considerable modification, within recent geological times. The old outlet to the northward is easily traced through a line of tamarack swamp to Susan Lake; — thence to Lake Michigan, it follows a small stream which is at present the outlet of Susan Lake. The marl bottom which underlies a very considerable part of Pine Lake can by borings be found not far from the surface at various points around the lake. The gravel and glacial drift are evidently at present being washed out into the lake over the marl and the thick- *H. B. Ward, “A Biological Examination of Lake Michigan, in the Traverse Bay Region.” Bull. Mich. Fish Com. No. 6, p. 65. 11-PT. Ill 82 MARL. ness of the latter decreases gradually as one recedes from the shore. Mollusca are not very abundant and while the species recorded by Mr. Walker are recent and most of them at least found in this locality at present, the existing conditions are inadequate to ac- count for such a bed of marl, and I am inclined to believe it to be the bed of an older lake now gradually disappearing.” He also says: “On the marl one finds no living thing save here and there scanty tufts of dwarfed Chara, which was never found in fruit: it was uniformly encrusted by heavy calcareous coating.” Here, as the author sti clearly points out, there has been a change in the lake within recent geological time and with this it is possible that the agencies producing the marls have become less active. The fact that Dr. Ward mentioned Chara as growing abundantly on the south arm of the lake would point to that plant having been more abundant formerly than now, but as the lake has not been visited, nor specimens of the marl seen by me, no claim is made that the beds described were formed by Chara. § 7. Association of marl and peat. Chara may also be looked upon as an important agent in giving the peculiar distribution to marl which has been noticed by every- one who has “prospected” beds of this material. The fact is fre- quently noticed that beds of several, and even as much as twenty or more, feet in thickness will “run out” abruptly into beds of “muck,” or pure vegetable debris (peat), of equal thickness. This distribution may show that up to a certain time conditions un- favorable to the growth of Chara and favorable to other plants obtained, until a depth of water was reached at which Chara was able to occupy the bed of muck, covering it from the bottom up, and holding the steep slope of the muck in place by mechanically binding it there by its stems and the root-like bodies by which it is connected with the mud. From the time when the Chara began its occupation of the muck the amount of organic matter left would decrease, and the amount of calcareous deposit would increase, until the latter predominated. The disturbing factors of currents and waves can be disregarded, for these abrupt unions of marl and muck are found, so far as the observations of the writer go, in most sheltered places, and not where either currents or w r aves could ever have operated with any force or effectiveness. Moreover, in a lake where the marl is evidently now actively extending, the slope was observed to be nearly perpendicular, and the steep banks thus CONTRIBUTION TO THE NATURAL HISTORY OF MARL. 83 formed were thickly covered with growing Ohara to the exclusion of other large forms of plant life, and the lower parts of the grow- ing stems were buried in mud which was mainly pure marl. § 8. Turbidity due to marl. That the finer parts of marl deposits may readily be moved from place to place in lakes in which they exist and where any part of the deposit is exposed to wave action, seems demonstrated by a series of studies, suggested by the milky appearance of the waters of some marl lakes. This has been considered by some investi- gators as possibly due to the presence of calcium carbonate, pre- cipitated from the water, either by liberation of dissolved carbon dioxide or by a change of temperature of the water after it has reached the lakes.* The writer has not found among the marl lakes of Central Michigan, that those with turbid water were com- mon, even where marl banks were apparently forming with con- siderable rapidity. “Merl” or Marl Lake in Montcalm County, situated on the same stream as Cedar Lake, and a mile or more below it, is, however, one of the lakes in which the water is usually of almost milky whiteness and has sufficient suspended matter in it to render it nearly opaque for depths over a meter or a meter and a half. The conditions in this lake are widely different from those at Cedar Lake and other marl lakes in the vicinity and are sug- gestive of the cause of the turbidity. At Cedar Lake, there is a border of grassy and sedgy marsh extending around the lake on three sides, that is generally underlain by marl, and the lake bot- tom slopes sharply and abruptly from the edge of the marsh to a depth of at least 10 meters. In other words the lake is simply a deep hole, with steep sides, and, perhaps, represents the deepest part of the more extensive lake which formerly occupied the area included by the marsh and marl beds. This marsh covering is generally found upon all the marl beds of the region and the lake may be said to be typical, for the locality in which it lies, for there are several others near by, which are practically identical in essen- tial points of structure. At Marl Lake, however, the filling of the lake has not reached the same stage. There is practically no open marsh, but the lake is shallow for seventy-five or a hundred meters from the shore, then abruptly deepens to an undetermined depth over a relatively small area. The bottom over the shallow area is of pure white marl and *25th Annual Report, State Geologist of Indiana. 84 MARL. the water is apparently not more than sixty or seventy centimeters deep at the margin of the central hole, while near the shore it is scarcely a third as deep. In brief, here is a lake in which there is a broad platform of marl surrounding a deep hole, which again, is all that remains of the deep water of a lake which is filling ^ith marl. Boring shows that the bed of the lake is nearly as far below the surface under the marl platform as where the marl has not yet been deposited. Upon the shoreward margin of the platform, and in small areas farther out upon it, the turf-forming plants are beginning to establish themselves, but as yet they have not made any marked impression, seeming to have a hard struggle to get a foothold. The conditions are then, a broad area of shallow water, overlying a wide platform of marl, which, if a strong wind should reach it, would be stirred to its depths, and with it, the lighter parts of the marl upon which it rests. The marl thus stirred up, in turn is carried to all parts of the lake by surface and other currents and makes the water turbid. These facts led to an investigation into the rapidity with which marl once stirred up would settle out of perfectly still water and some interesting results were obtained. The experiments were made as follows: (1) A glass tube 1.58 m. long and 2.5 cm. wide was filled with dis- tilled water, and a quantity of finely divided marl was added and thoroughly mixed by shaking. The tube was then clamped in a vertical position and left perfectly still until the marl had settled out, record being kept of the rate of settling. At first the heavier particles settled rapidly, forming as does clay in settling out from water with which it is mixed, distinct stratification planes, which after a few days disappeared, and only the lighter parts of the marl remained in suspension. These were distinctly visible for five weeks on looking through the tube towards a good light, and at the end of six weeks a black object lowered into the tube, in a well lighted room, was not visible beyond 90 cm. from the surface of the water. (2) A glass cylinder with a foot, 88 cm. high and 7 cm. wide, having a capacity of a little more than a litre was nearly filled with distilled water and the residue from the washings of a sample of marl from which the coarser matter had been separated, was thoroughly ^ shaken up in it. This was left to subside as in (JO NT BIB U TION TO THE NATURAL HISTORY OF MARL. 85 the first experiment, and at the end of ten weeks the bottom of the vessel was barely visible. The results obtained by Barus,* in his work on the subsidence of solid matter in suspension in liquids, show that settling is much more rapid in water containing dis- solved salts even in small proportion than in distilled water, so check experiments were made as follows: (1) A cylinder approxi- mately the size of the one used in the second experiment above, was filled with water in which a small amount of calcium chloride had been dissolved, and ammonium carbonate was added until a precipitate was formed. The contents of the jar were then stirred thoroughly and left to settle. In three days the entire pre- cipitate had settled out and the liquid was clear. In this case, however, it was deemed probable that the conditions were not at all like those occurring in nature and a second experiment in which the marl was shaken up with the ordinary natural water of the region, obtained from a river partly fed by marl lakes. In com- parison with distilled w r ater the subsidence w r as notably more rapid than from distilled w r ater for the finer part of the marl, but for fifteen days there was distinct turbidity noticeable. These results indicate that if for any cause the marl in one of the marl lakes is stirred up effectually, as it may be where the beds are exposed to wave action, that the water will remain turbid for some time, and the chances are that even in summer time there will be sufficiently frequent high winds to keep the water always turbid. It may be stated that in some of the lakes which have been studied by the writer the marl has filled the entire lake to within a meter or less of the surface of the water, with some parts even shallower. Until such shallows are occupied by plants and turfed over, the water is likely to be turbid from the mechanical action of waves upon the deposits. At Littlefield Lake, described else- where, f the water is only slightly turbid, although there are exten- sive marly shallows and exposed banks, but there the body of the water is extensive and of considerable depth, while the greater part of the exposed marl is granular and the particles too coarse to be held long in suspension, and the finer parts too small and too well protected to be reached by effective waves, so that the amount of suspended marl is not great enough to produce marked turbidity in the entire body of water. It is worthy of note that the residue *Subsidence of fine solid particles in liquids, Carl Barus, Bull. U. S. Geol. Survey, No. 36. fJournal of Geology, VIII, No. 6, and this report, p. 92. 86 MARL. filtered out from the sample of Chara fragments (Analysis 4) was sufficiently fine to give a marked turbidity to distilled water for several days and at the time of filtering had not subsided. It is difficult to account for the fact that the deeper parts of marl lakes are generally free from any thick deposits of a calcareous nature. Lack of records of sufficient exploration makes any statement purely tentative, but about 7-9 meters seems to be limit of depth of the recorded occurrence of Chara plants.* The remains of the plants then would only accumulate in place, over bottoms above that depth, and the material reaching greater depths would have to be that held in suspension in the water, hence be relatively small in quantity and accumulate slowly. A probable additional cause is that in the greater depths (i. e., over 9 meters) a greater abun- dance of dissolved carbon dioxide, due to the decomposition of organic matter in relatively cold water under pressure, dissolves the fine particles of marl which reach these depths, but at present no data are at hand on which to base a conclusion as to the exact efficiency of this cause. f § 9. Conclusions. From these investigations it seems: First, that marl, even of the very white pulverulent type, is nearly made up of a mixture of coarse and finer matter, covered up and concealed by the finer particles, which act as the binding material. Second, that the coarser material is present in proportion of from 50$ to 95$ of the entire mass. Third, that this coarser material is easily recog- nizable with the unaided eye and hand lens as the incrustation pro- duced on Schizothrix and Chara, principally the latter, to particles less than one one-hundredth of an inch in diameter. Fourth, that the finer matter is largely recognizable under the compound micro- scope as crystalline in structure and derived from the algal in- crustations by the breaking up of the thinner and more fragile parts or by disintegration of the younger parts not fully covered. Fifth, that some of this finer matter is capable of remaining sus- pended in water a sufficient time, after being shaken up with it, to make it unnecessary to advance any other hypothesis to explain the turbidity of the water of some marl lakes, than that it is caused by mechanical stirring up of the marl by waves or other *C. Wesenberg-Lund: Loc. cit.. p. 156. A. J. Pieters: Plants of Lake St. Clair, Bull. Mich. Fish Commission, No. 2, p. 6. Compare, however, reports of the Indiana Survey. fBut see tests 11 and 12, etc., of water in the Cloverdale district, p. 46. L. CO NT RIB V TION TO THE NATURAL HISTORY OF MARL. 87 agency. Sixth, that shells and shell remains are not important factors in the production of the marl beds which are of the largest extent. Seventh, there is in marl, a small amount of a water sol- uble calcium salt, possibly calcium succinate, readily soluble in dis- tilled water after complete evaporation of the water in which it was first dissolved. § 10. Method of concentration by Chara. After these facts were developed studies were undertaken to de- termine the method of concentration and precipitation of the cal- cium carbonate by Chara. Some such studies have already been reported upon by various authors, but none of these have appar- ently been exhaustive, and the original papers are not at hand at the present writing, although abstracts of the more important ones have been seen. As has been already indicated elsewhere, the calcium carbonate is present on the outside of the plant as an incrustation and this is made up of crystals, which are rather remote and scattered on the growing parts of the plants and form complete covering on the older parts, which is uniformily thicker on the basal joints of the stems than it is on the upper ones. Considering the hypothesis that the deposition of the salt was the result of purely external chemical action, as not fully capable of satisfying all the existing conditions, the formation of the incrustation was taken up as a biological problem and investigation was made upon the cell con- tents, at first, microscopically by the study of thin sections. Vari- ous parts were sectioned while still living and the attempt was made to find out if the calcium carbonate were present, as part of the cell contents in recognizable crystalline form. In no case were such crystals found, although reported by other observers. Next an attempt was made to determine the presence of the cal- cium in soluble form in the cell contents by the use of a dilute neutral solution of ammonium oxalate. An immediate response to the test was received by the formation of great numbers of min- ute characteristic octahedral crystals of calcium oxalate on the surface and embedded in the contracted protoplasmic contents of the cells. The number of these crystals was so large and they were so evenly distributed through the cell contents, that it was evident that a large amount of some soluble calcium salt was diffused through the cell sap of the plant. The next step was to isolate this compound and to determine its composition. A considerable 88 MARL. quantity of the growing tips of Chara were rubbed up in a mortar and the pulp was thoroughly extracted with distilled water. This water extract was filtered, concentrated by evaporation on a water bath, and tested to determine the presence of calcium. An abun- dant precipitate was again obtained by using ammonium oxalate, which on being separated and tested proved to be calcium oxalate. It was evident that the calcium salt in the plant was stable and readily soluble in water. This latter fact was farther demon- strated by evaporating some of the extract to dryness and again taking it up with water. Almost the entire amount of the calcium salt was redissolved, only a small portion of it becoming insoluble, precipitating as the carbonate. This ready solubility demonstrated that the salt was not derived from the incrustation on the portions of the plant used, and the same fact excluded from the list of pos- sible compounds, salts of the more common organic acids found in plant juices. Qualitative chemical tests were, however, made to determine, if possible, whether any of these acids were present, with negative results, and it was demonstrated by this means that there was but a single salt present and not a mixture. Search was then made to determine the acid present and a result was obtained which was so unexpected that it was seriously questioned, and the work was gone over again. The second result confirmed the first, and the work of ascertaining the correctness of these two results was turned over to Mr. F. E. West, Instructor in Chemistry in Alma College, who had special training and much practice in organic analysis. His work was done entirely independently with material gathered at a different season, and by another method of analysis, but his results were identical with my own and show that calcium exists in the water extract of Chara as calcium succinate, Ca (C 4 H 4 0 4 ). The fact that the succinate is one of the few water soluble salts of calcium and that there is a soluble salt of the metal in the cell sap of the plant, makes it probable that this is the com- pound which the plant accumulates in its cells. It is not yet pos- sible, from actual investigation, to explain the method by which the calcium salt is abstracted from the lake water, where it exists as the acid or bicarbonate, or as the sulphate,* in small per cent, and concentrated in the cells of the plant as the calcium succinate and later deposited upon the outsides of the small cells as the *The formation of CaC0 3 incrustation by Chara in water impregnated with CaSO* accompanied by the liberation of H 2 S is reported in a book called the “Universe.” CONTBIB TJTION TO THE NATUBAL HISTOBY OF MABL. 89 normal or monocarbonate in considerable quantities. Culture experiments which were undertaken by the writer to determine under what conditions of soil, light and temperature Chara thrives best, incidentally demonstrated that the plant actually gets its lime from the water about it and not from the soil. One of the soils which was used as a substratum in which to grow plants was pure quartz sea-sand, which had been thoroughly washed and tested with acid to be certain that no calcium salt was present in it. The plants grew in this medium readily, and on the newer parts, devel- oped nearly if not quite as many calcium carbonate crystals as plants growing on pure marl. It should be apparent, however, to even the casual observer, that the plants cannot take all the lime they use in forming incrustations from the soil, for if they did the marl beds, being made up principally of Chara remains, would never have accumulated, for the material would have been used over and over again and could not increase in amount. In the present state of our knowledge of the life processes of aquatic plants it seems hardly possible to state the probable method of formation of the calcium succinate or even the probable use of it to the plant and no attempt will be made by the writer just here to do so. It does seem probable, however, that this com- pound accumulates in the cells until it reaches sufficient density to begin to diffuse through the cell walls by osmosis. Outside the cells it is decomposed directly into the carbonate, possibly by oxida tion of the succinic acid by free oxygen given off by the plants, possibly, by the decomposition of the acid by some of the organic compounds in the water due to bacterial growth in the organic debris at the bottom of the mass of growing Chara. The water extract of Chara rapidly changes on standing, undergoes putrefac- tive decomposition, becomes exceedingly offensive in odors devel- oped, and a considerable quantity of calcium carbonate crystallizes out on the bottom and sides of the containing vessel, while the succinic acid disappears, gas being given off during the process more or less abundantly. Whether these changes take place on the outside of living plants has not yet been determined. In regard to the species of Chara which seems to be the active agent in precipitation in the lakes of Central Michigan, it is the form commonly known as Chara fragilis, but it is probable that careful study of the species throughout the range of the marl will reveal, not a single form, but a number of allied species, engaged in 12-Pt. Ill 90 MABL. the same work. It may be well to suggest that in lakes to which silt is brought by inflowing streams, or which have exposed shores where the waves are constantly cutting and stirring up rock debris, the more slowly accumulating marls will be either so impure as to be worthless, or so obscured as to escape notice altogether, even where Chara is abundant. It may also be pointed out that shallow water, strong light, and a bottom of either clay, sand, or muck, present conditions favorable for the growth of the higher vascular plants, and that these cause such rapid accumulation of vegetable debris that the calcareous matter may be hidden by it, even when Chara is a well marked feature of the life of a given lake. This view is amply supported by the presence of large accumula- tions of Chara plants heavily incrusted with calcium carbonate, at the storm-wave line along the shore of Saginaw Bay, in Huron County. These windrows, however, soon disappear, leaving noth- ing more than a limy layer in the sand, scarcely to be distinguished from the rest of the wave-washed shore, and ultimately all trace of them is lost. § 11. Blue-green algrn and their work. Another plant form, like Chara, an alga, but of a much lower type, which is concerned in the formation of marl, is one of the filamentous blue-green algae, determined by Dr. Julia W. Snow, of Smith College, to be a species of Zonotrichia, or some closely related genus. The work of this species is entirely different in its appearance from that of Chara, and at first glance would not be attributed to plants at all. It seems to have been nearly over- looked in this country, at least, by botanists and geologists alike, as but three references to it have been found in American litera- ture.* Curiously enough, however, material very similar, if not identical, to that under consideration has been described from Michigan in an English periodical devoted to algse.f In this the alga is identified as Schizotlirix fasciculata Goment. Mr. F. S. Collins of Malden, Mass., has identified Schizotlirix fasciculata as present in the concretions from Littlefield Lake, but does not spec- ify it as the form which has the calcareous covering. The plant grows in relatively long filaments formed by cells growing end to end, and as they grow, the filaments become incased in calcareous *McMillan: Minn. Plant Life, 1899, p. 41. Penhallow: Botanical Journal, 1896. p. 215: J. M. Clarke. “The Water Biscuit of Squaw Island, Canandaigua Lake, N. Y.” Bull, of the N. Y. State Museum, No. 39, Vol. 8, p. 195, 1900. fG\ Murray: Phycological Memoirs No. XIII, 1895, p. 9, PI. XIX. CONTRIBUTION TO THE NATURAL HISTORY OF MARL. 91 sheaths. The feature of the plant which makes it important in this discussion, however, is its habit of growing in masses or col- onies. The colony seems to start at some point of attachment, or on some object like a shell, and to grow outward radially in all di rections, each filament independent of all others and all precipi- tating calcium carbonate tubules. The tubules are strong enough to serve as points of attachment for other plants, and these add themselves to the little spheroid, and entangle particles of solid matter, which in turn are held by new growths of the lime-precipitat- ing Zonotrichia, and thus a pebble of greater or less size is formed which to the casual observer is in no wise different from an ordi- nary water rounded pebble. These algal calcareous pebbles show both radial and concentric structure and might well be taken for concretions formed by rolling some sticky substance over and over in the wet marl on which they occur but for the fact that a con- siderable number of them show eccentric radial arrangement, and that the shells of accretion are likewise much thicker on one side than on the other, and finally, because the side which rests on the bottom is usually imperfect and much less compact than the others. The pebbles are characteristically ellipsoidal in shape. The radial lines, noticeable in cross sections of the pebbles, are considered by the writer to be formed by the growth of the filaments while the concentric lines probably represent periods of growth of the plants, either seasonal or annual. Included within the structure are great numbers of plants, besides the calcareous Zonotrichia, among them considerable numbers of diatoms, and it is probable that a large part of the algal flora of a given lake would be represented by in- dividuals found in one of these pebbles. It is probable that to a certain extent they disintegrate after the plants cease to grow, for they ^e never very hard when wet. It is possible to recognize them, as lumps of coarser matter, even in very old marl, and the writer has identified them in marl from Cedar Lake, which was taken from a bed a foot or more above, and several rods away from, the lake at its present level. From the fact that these pebbles have been found, by the writer in four typical marl lakes in different parts of Michigan (in Zukey Lake and Higgins Lake by Dr. A. C. Lane, who was struck with their peculiar character) and have been reported from a number of others by Mr. Hale and other marl hunters, it is probable that they have a wide distribution in the State and are constant if not 92 MARL. important contributors to marl beds. It may be said in passing that the limy incrustations which are found upon twigs, branches, shells, and other objects in lakes and streams, and called generally “calcareous tufas,” are of similar origin and are formed by nearly related, if not by the same plants that form the pebbles. Studies have been begun by the writer to solve, if possible, some of the questions which have, arisen in connection with the state- ments embodied in this paper, but enough has already been done to show that these forms of fresh-water algse are important lime- precipitating agents now, and to suggest the possibility that in all likelihood they have been more active in former geological times, and that, as has been suggested again and again by botanists, the formation of certain structureless limestones, and tufa deposits may have been due to their work. § 12. Littlefield Lake, Isabella County. Early in June, 1900, the writer visited this interesting body of water, and from its peculiar form, and the deposits about it, it seemed worthy of special description.* The country about the lake is of a well-marked morainal structure, the till, however, being sandy in places, and noticeably gravelly and bouldery throughout, and was formerly heavily covered with pine. The lake occupies a deep depression in a trough-like valley, sur- rounded by moderately high morainal hills, and from its apparent connection with a series of swamp valleys, suggests a glacial drain- age valley, but as it was not followed for any distance, its origin was not determined. The lake itself is about one and one-half miles long, by three- fourths of a mile broad in the widest part, which is near the middle of the long axis and the shape is that of an irregular blunt ended crescent. It was said to be over eighty feet deep in the deepest part, but no soundings were made by the writer. Its greatest length is from northwest to southeast, with the outlet at the southern end. There are no considerable streams entering it, but at least three small brooks fed by springs from the surrounding hills were noted flowing in, and the outlet is of such size that a boat may be easily floated on it at high water, although its level is maintained during the summer by a dam about two miles below the lake. The main inlet was not seen by the writer. 'See Plate XIX. OONTB1BUTION TO THE NA TUBAL HIS TO BY OF MABL. 93 The shore lines are relatively regular, especially on the east and north sides, the convex side of the crescent, with banks twenty or more feet high close to the water on the east, while on the west side are two rather deeply indented bays. At either end are three small ponds, parasite, or daughter lakes, and surrounding the entire shore except on the eastern side, and the northeastern, or inlet, end is a cedar swamp which is underlaid by marl. The outlet is through the most southerly of the daughter lakes, and the entire shore of the lake is formed by beautifully white marl, the exposures vary- ing in width from a few feet to three or four rods in width, so that as one overlooks the lake from one of the surrounding hills it seems to lie in a basin of white marble. There are three small islands in the lake, two relatively near together at the northern end, and one quite near the shore at the south end. These islands are also of marl, covered partly with a thin layer of vegetable matter and a scanty growth of grass, bushes and cedar. There is a visible connection, under water, be- tween at least one of the islands and the shore, and it is probable that all of them are thus connected by submerged banks. The marl on the islands is from twenty-five to thirty feet deep, with sand below. Explorations in the swampy border of the lake, show that the shore was formerly more irregular than now, and that the marl extends back from the water in some places for at least one-fourth of a mile, gradually becoming more and more shallow until the solid gravel or clay is reached. The marl is frequently thirty feet deep along the shore and at no place was it found to be less than fifteen feet deep at the present shore line, the shallowest places being along the shore where the high bank comes down near the water. The deepest vegetable deposit, or peat, found in one hun- dred and fifty borings in all parts of the deposit was three feet. The main deposits of marl are about the southeast end and along the western side of the lake, with a body of considerable size, under- lying a swampy area at the north end. Of the six daughter lakes, four are very small, an acre or two in extent and entirely sur- rounded by deep marl, the connection between three of them and the mother lake being shallow and narrow, a few inches deep, and a few feet wide, and only existing at high water, while two of the other three are of much larger size, with marl points extending out from either side of the straits which are still relatively wide and deep. 94 MABL. Of the two bays on the west side of the lake, one is much nar- rower than the other and at the mouths of both, marl points are extending towards each other to a noticeable degree. At all points along the shore, the slope of the marl is very abrupt from the shallow water to the bottom, always more than forty-five degrees, and frequently nearly ninety, this steepness being notice- able in the small as well as in the parent lakes, while on the east side of the island, at the south end of the lake, the wall of marl seemed positively to overhang, although this appearance was prob- ably due to refraction. The texture of the deepest part of this marl deposit is apparently that of soft putty, a sounding rod passed through it with com- parative ease, and samples brought up have a yellowish or creamy color, which disappears as they dry, leaving the color almost pure white. At the surface the marl is coarser, slightly yellowish and more compact. Where it lies above the water line it is distinctly made up of granular and irregular angular fragments, resembling coarse sand, but the fragments are very brittle, soft and friable, and may be converted into powder by rubbing between the thumb and fingers. On the parts of the shores where apparently the wave action is chiefly exerted, there are small rounded calcareous pebbles, mixed with molluscan shells, drift material and considerable quantities of stems, branches and more or less broken fragments of the alga, Chara, all parts of which are heavily incrusted with calcareous matter. This Ohara material was often piled up in wdndrows of considerable extent at the high water mark. The marl banks of the lake, from a little below the water’s edge down as far as could be seen, were generally thickly covered with growing Chara, at the time of the writer’s visit and wherever a plant of it was examined it had a heavy coating of limy matter, which was so closely adherent to the plant, as to seem a part of it, and because of this covering, the plants were inconspicuous, and would easily escape notice. Little if any other vegetation of any character w r as growing in the lakes at this season. Indeed, from the steep slope of the banks of marl, it would be hardly possible for any considerable amount of vegetation of higher types than algae, to flourish here, because of the lack of light at the depth at which it would have to grow to establish itself. CONTRIBUTION TO THE NATURAL HISTORY OF MARL. 95 As Chara of several species, is known to occur within our limits, at depths as great as thirty feet, and probably grows at even greater depths, where the water is clear and the bottom soil is of the right character, i. e., of clay, finely divided alluvial matter, marl, etc v it is apparent that there must be an immense growth of this type of plants in such a lake as the one under discussion. That there is an abundance of Chara in Littlefield Lake is shown by the amount of drift material, composed of the plant, which had accumulated in heaps at the high water wave marks along the shore at various places. From even a casual inspection of this drift accumulation, it is evident that it is the source of much of the granular and sandlike marl on the beaches, and in the coarse upper layers of the deposit. This wind and wave accumulated material was dry and bleached, and was very brittle, so fragile indeed, that a mere touch was generally sufficient to break it into fragments and it passed by insensible gradation from the perfect, unbroken, dried plant form at the high water mark, in which every detail, even the fruit, is preserved, to inpalpable powder at, and below the water’s edge. In other words we have in Chara, a plant of relatively simple organization, and one able to grow in abundance under most con- ditions of light and soil which are unfavorable to more highly de- veloped types, a chief agent in gathering, and rendering insoluble, or precipitating, calcium and other mineral salts brought into the lake from the clays of the moraine around it by the stream, spring and seepage waters. After precipitation is accomplished and the plant is dislodged, or dies, it drifts ashore, where after decomposing and drying out the small amount of vegetable matter, the various erosive agents at work along shore break up the incrusting chalky matter, and the finer fragments are carried into deeper water, the coarser are left along the lines of wave action. The pebbles mentioned above as occurring on parts of the shore, are also the result of the development and growth of an alga, Zono- trichia or a nearly related genus, a much lower type than Chara, having a filamentous form. The vegetable origin of these pebbles would not be suspected, until one is broken open when recently taken from the water, when it is found to show a radiating struc- ture of bluish green lines, the color indicating the presence of the plants, as it is characteristic of the group to which Zonotrichia belongs. 96 MARL. The relation of the deposits about Littlefield Lake to the direc- tion of the prevailing strong winds of the region, is probably signi- ficant. The area of deposition is at the southeast end and along the whole western side of the lake. The winds which would be most effective in the valley of the lake would be those from the north and north- west, which would drive the surface waters down the lake towards the southern end, and, striking the shore on the eastern side, cur- rents formed thus would be turned across the lake to the west, de- positing sediment at the turning area and in slack water beyond. The daughter lakes are not easily accounted for, except in a general way, that they were formerly deep bays, which, by the building out of points of marl on either side of their mouths, were finally enclosed. The tendency, already noted, for existing bays to have points of marl, of spit-form, extend from either side of the mouth would seem to indicate this as a probable method of formation. On the island at the south end of the lake there was manifestly a strong current, which was running southeasterly and depositing fine marl on the east side of the island, the wind at the time the observation was made, blowing gently from a few points north of west. As has been already noted, the islands consist of marl from twenty-five to thirty feet deep, the bottom on which they are built up being, to judge from soundings, made with an iron rod, of rather fine sand. These foundations of sand have deeper water all around them, if soundings, said to have been made by local fishermen, can be relied upon, so it is possible they represent shal- lows in the original lake bottom, upon which after Ohara had estab- lished itself, the marl accumulated, both by direct growth of the plants and by sedimentation. It may be worthy of mention, that the Chara growing on the steep banks, may in part, account for their steepness, by acting as holding agents, bind the particles of sediment in place by stems and the rootlike organs which the plant sends into the mud. It is probable that but a small part of the Chara that grows in the lake, ever reaches the shore wave line, and much must break up by the purely chemical processes, resulting from the organic decay in relatively deep water. CONTRIBUTION TO THE NATURAL HISTORY OF MARL. 97 APPENDIX, ON THE SHELLS OF MARLS. BY BRYANT WALKER. Detroit, Michigan, Nov. 25th, 1901. A. C. Lane, Esq., Lansing, Michigan: My Dear Sir. — I enclose my report on the mollusks fqund in the seventeen lots of marl material received from yourself and Prof. Davis during the last two years. I have not included the recent species, of which several lots were received from Prof. Davis, as their determination was not particularly pertinent to the marl fauna. I can send you a list of them if you desire. There is, however, noth- ing of special interest in them and the list of Saginaw Valley shells, which you made use of in your former report,* will include them all. Taken as a whole the fauna of the marl deposits does not differ from the present fauna of that portion of the State from which they come. Nor have I found in the specimens from any particular locality any special peculiarities, which would indicate peculiar local conditions of environment. Individual variations occur more or less frequently, but no more than is often found in similar col- lections of recent species. The inference is, therefore, that the marl fauna lived under substantially the same environmental conditions as the present fauna does or at least not sufficiently different to produce any special or characteristic variations. The one species peculiar to the marl deposits of this State is Pisidium contortum Prime. It was originally described from the Post-glacial formation at Pittsfield, Mass. It occurs abundantly in the marl deposits both in Michigan and Maine. It has recently been found living in one locality in the latter State and it is quite possible that it may yet be found alive in this State. But so far as our present knowledge extends it is extinct in Michigan. Why this one species out of the fifteen, to say nothing of the other genera represented in the marl, included in our list, should have failed to survive, while all the others are still abundantly repre- sented in our present fauna is very curious. I have been entirely unable to imagine any adequate explanation. The characteristic feature of the marl fauna is the great relative abundance of certain of the smaller species. This is especially noticeable in Planorbis parvus Say, Yalvata tricarinata Say and * Vol. VII, Part III. 13-Pt. Ill 98 MARL. Amnicola limosa Say and lustrica Pils. Tlie larger Planorbis bi- carinatus Say and campanulatus Say occur in nearly every lot of material, but the number of individuals is comparatively small. Pisidium both in the number of species and individuals is also a characteristic feature of the marl as it is indeed of our prest nt fauna. There is probably no district in the United States, in which this genus abounds to a greater extent, both in species and individ- uals than in the inland waters of this State. The terrestrial species represented in the marl are few both in number and individuals. This is what would naturally be expected, as those that do occur are the occasional examples that have been washed into the water from the adjacent land. Such as have been found present no peculiarities as compared with recent specimens from the same region. The almost complete absence of the Unionidce from the collec- tions is also noticeable. The peculiar variations noted in Yalvata tricarinata Say from Cement City are of considerable interest. A similar tendency to unusual variation, although in another direction, has been noticed in the same species from a Post-glacial deposit near Niles in this State (Nautilus XI, p. 121). In both instances, however, the variation was not common to the whole colony, but was limited to a very few individuals. It cannot therefore be attributed to any peculiar conditions in the environment for in that case it would undoubtedly be more general in effect. Yours very truly, (Signed) BRYANT WALKER. N. B. — Please don’t forget to give Dr. V. Sterki the credit for identifying the Pupidce and Pisidia. NOTES. Numbers refer to numerals in table. 1. Fragment or fragments only. 2. Young shells, just hatched, undoubtedly recent. 3. Apparently recent. 4. Fragment, possibly 8. avara Say. 5. Peculiar form. 6. Peculiar form, probably L. Tiumilis Say. 7. Peculiar form. CONTRIBUTION TO THE NATURAL HISTORY OF MARL. 99 8. Young. 9. Undoubtedly recent. 10. One left valve with teeth wholly reserved, one right valve with anterior laterals and cardinals reversed. 11. One valve with posterior laterals reversed. 12. Two samples with the apertural portion of the last whorl separated from the body whorl. One example with the superior and peripheral caringe present, the umbilical carina wanting, its position however is represented by a slight angulation of the whorl. This remarkable variety has never been seen before among hun- dreds of examples examined. So far as I know there is no previous record of its occurrence. Should other examples be found it would be entitled to rank with the varieties already described. But as only single specimens from two different localities have been noticed it may be only an individual variation or “sport.” 13. One example with the superior and peripheral caringe pres- ent, basal one obsolete. See Note 12. 14. Deformed. LOCALITIES. 1. Shell-bearing deposits in digging a well about 100 feet north- east of Sec. 36—13—5 E. A. C. Lane, Coll. No. 1. 2. Marl from A. F. Gorton. Lake near Howell. A. C. Lane, Coll. No. 2. 3. E. i S. E. i Sec. 3—11 N.— 5 E. A. C. Lane, Coll. No. 3. 4. Cascade near Grand Rapids. A. C. Lane, Coll. No. 4. 5. Cedar Springs. A. C. Lane, Coll. No. 5. 6. Sec. 22, T. 10 N., R. 11 W. A. C. Lane, Coll. No. 9. 7. T. 11 N., R. 11 W. A. C. Lane, Coll. No. 10. 8. Pickerel Lake, Newaygo County. A. C. Lane, Coll. No. 11. 9. Indian Lake. A. C. Lane, Coll. No. 12. 10. Fremont Lake (12 N., 14 W.), Newaygo County, 150 to 200 feet above lake. A. C. Lane, Coll. No. 14. 11. Cut between Sec. 24 and 25, Spaulding Township, Saginaw County. A. C. Lane, Coll. No. 15. This is a sand deposit of Lake Algonquin? 12. Marsh north side of Cedar Lake, Cedar Lake Station, Mont- calm County. Lane and Davis, Coll. 100 MARL. 13. Dry marl bed mile east of Cedar Lake Station, Montcalm County. Lane and Davis, Coll. 14. Marsh on east side of Mud Lake, N. W. \ S. W. J, Sec. 3, T. 12 N., R. 4 W. Lane and Davis, Coll. 15. From bank of ditch, N. W. J S. W. i, Sec. 3, T. 12 N., R. 4 W. Gratiot County. Lane and Davis, Coll. 16. Bottom of ditch from Mud Lake. N. W. J S. W. J, sand sec- tion, Gratiot County, “possibly washed from marl.” Lane and Davis, Coll. 17. Goose Lake. Cement City. J. G. Dean. This is the marl of the Peninsular P. C. Co. LIST OF SPECIES. CONTRIBUTION TO THE NATURAL HISTORY OF MARL 101 r- x : : x x- : xL : : x x x : x ; ; X ; M ; ; ; i : x X X i x xti x : x x x x : : x ^ . . x so : : : : x : : x : X : :xx : rxxxxx : :xx i : ; « 1 ; ; ; ; X * LC X X ~ |x X i : x : : :xxxxx : : l x x • : ■ • x ... . . x .... so ::::::: x : x 1 X : : ^ x : : : x : : x x i : : : : : x • "M : : x :::::::: x : :£* ix::x::xxx:::: - X ; : : : :- x x- : : .... x X X : : : x x : : x x : : : : : x : o :::::::*: X : :^xx :xx : : : x i x x ; : : x 55 : : : x i:::x::xx::::: GC : : : : - : : x ' a • • . . x • • X : x : : : : * x x x : x : :xxxx : :^x I> :x ; : : : x : : : : x : : : x SO : : : x j ::::::: xj x : x : : :::xxx:x::xxx::::x .O :::.:•;: x ::xxx::x:::xx::::x ■sf ::::::: ^ : ix 1* i : : x x : : : x !::::: x » : : X OJ x : .::xx::x::xx::::: ’-I : : x :*x • j i j j 1 X : : x i j jx jx j 3 CQ * w 33,2 2 H3 o -el © . :m 3 ;ccQ SS ® o, S3' C/3 3 3 3 m > - 131!* e3 > -9 >s O a * .2 w >1 3 IP 3 3 .►,.3 3 . 3^M f> >> S-^aS * _b£ 3 3 p o .W ~ © ||S3= 3 2 1 ■*■' 3 33 ,P y. 3 “ - '' " " =3 i™ !S S© ,,- '3P;©p©5< 1> 3;a._, H H cfl CQ > cfl CMi-I • c3 *3 2 ^ 3 M CC 3-r; 2 32,5 ©2 f-i+a n .^7 © <« © >j33 3 .2 &£© 3 - H2 c3 Si ^ VI ©‘E a >■.£ 3«na“ £j 3 3 CZ2 ce £©S a‘> S 4 © © > o Sep 3 c > 3 © X 3‘C ©pi © ft3 © © 2Q 2 3 3CC 3.2 © 3 3 3 2>3c.2 2 P O 3 S|.^ ^ ^ 2 t< ^C/2 3 «3 W & > © 3 ' ! 32 P 3 © tf 5 a 3 :a£ 7? © 3 IL> p O^ftg j. f) 2 pm S 3 ”© © „ _ a - ‘2 ' 1" a o <1 LIST OF SPECIES-Continued. 102 MAliL x Present. x ’Identification doubtful. CHAPTER VI. RECORD OF FIELD WORK BY D. J. HALE. § 1. Lansing — Summer, 1899. Before starting on a longer tour of inspection a short trip was made from St. Joseph through White Pigeon, Bronson, Coldwater, Quincy, and several other towns near which marl had been re- ported. The surroundings of the marl, its location and manufac- ture and any other points needing investigation were to be noted. White Pigeon . — The first bed visited was that of Mr. Theodore E. Clapp on Section 17, St. Joseph County, two miles southeast from White Tigeon, and one and a quarter miles from the Lake Shore & Michigan Southern railroad. The following are his figures on the bed. The depth is from six to thirty feet with average depth twenty feet, area 100 acres. The marl at the center of the lake is about twenty-two feet deep, at the edges thirty feet deep, water in shallows two to four feet in depth. The marsh land about the lake is underlain with the deepest marl, and this greatest depth is between the lobes of the lake. The marl is not overgrown suffi- ciently with marsh grass for cattle to graze upon it safely. For sounding, the deposit he used two inch drive well pipe cut into six foot lengths, and fitted with couplings so. that they could form a continuous rod. Upon one length an augur was welded. This was the apparatus used to bring up the specimens. Mr. Clapp made fifteen soundings, requiring a force of five men. Tests were carried on during the winter through ice. The analy- sis of the marl made at Purdue University was as follows: Moisture 81$ Insoluble in Hydrochloric acid 1.46$ Silica 37$ Iron and Alumina 56$ Calcium oxide 51.00$ Magnesium oxide 1.02$ Potash 17$ Soda 52$ Carbonic Acid 41.10$ Organic matter combined with water 4.01$. Sulphuric acid trace. Phosphoric acid trace. 101.02 104 MAUL. Tlie above analysis would indicate a first class marl. The marl between the lobes of the lake, which was before remarked as deep- est, in this instance probably marked the center of the lake. Ac- cording to Mr. Clapp's soundings the deepest water did not contain the deepest marl, but rather the shallows at the edge of the lake toward the center of the whole lake basin. The lake basin would be the whole depression including the two lobes of the lake, and the low marsh surrounding it. Bronson , Quincy, Coldwater. After leaving this lake the old glacial valley or chain of marl lakes extending irregularly through Bronson, Quincy, and Cold- water was examined. It was near Bronson, while sinking piles for a bridge over a creek that a section foreman discovered a marl bed. The whitish or greyish soft mud which he found there, proved upon analysis to be a marl suitable for cement. A thriving factory was started upon this same land, also one at Union City some fourteen miles distant. One is built at Coldwater and another completed at Quincy, these two belonging to the Michigan Port- land Cement Co. This constituted the district which was at the time (1899) actively devoted to the manufacture of marl, although factories were in the process of building in many parts of the State. The bed and factory at Bronson were first examined. The factory itself is located on a sandy island a few acres in extent. These islands are sprinkled through the marl bed, and upon some of them good sized trees are growing. The deposit is one of the old lake valleys above mentioned. In reply to questions asked Mr. Wheeler, the chemist of the factory, the following facts were given. The area of marl is estimated at 1,300 acres. It follows the bed of Swan Creek, and two or three other streams from Spring Lake. The depth varies from one to fifty feet according to measurements made with solid iron rods. Beneath the marl there is a white quartz sand, the outline of which is regularly undulating, which Mr. Wheeler accounts for by the former action of waves. The marl is about thirty to forty per cent water. The lake basin is in the form of an oblong one mile wide and several miles long. The factory contains seven rotaries and six tanks with an output of 1,000 barrels per day. The occurrence of marl under one part of the marsh does not signify that it will be found under the whole marsh. The bed is thickest at the center. It contains no bog iron ore, Geological Survey of Michigan. Vol. VIII Part III Plate III. COLD WATER PLANT OF WOLVERTON P. C. CO. UNION CITY PLANT OF PEERLESS P. C. CO. RECORD OF FIELD WORK BY D. J. HALE. > 105 and few shells. The well water in the vicinity is rather soft. Ilis analysis of the marl is as follows: Volatile matter 45.64$ Insoluble matter 1.72$ Iron and aluminum oxides 1.17$ Calcium oxide 49.21$ Organic matter 7.07$ 104.81 Further analyses and descriptive notes will be found in the last chapter. The marl is dug by an ordinary dipper dredge which scoops out the marl to a depth of twenty-two feet and empties it into small cars in which it is hauled to the factory. The dredge first removes the surface from one to six feet of tough marsh grass, roots, etc., and piles it up at one side or dumps it in place where the marl has already been removed. As the water stands at within from one to two feet of the surface, after a small channel is cleared the dredge has water room to float over the marl which is to be re- moved. The marl when first dug is much darker on account of being nearly half moisture, but after drying, it becomes about the color of light wood ashes. The next point of interest visited was the clay pit from w 7 hich the supply of clay for this factory is de- rived. The pit is on a siding about two miles south of the factory. It is in this vicinity that the great stratum of Coldwater shales is uncovered. In this case the shale does not quite reach the surface, and a shaft seventy-five feet deep has been sunk to penetrate the surface soil, and from the vertical shaft a tunnel with several smal- ler branches has been dug through the solid shale. A regular min- ing hoist is used to reach shale and hoist it to the surface. Clay is transferred from the head of the tunnel to the shaft by small cars run on a wooden track. The clay, which is a shale compressed until it shows lines of cleavage, is hard like a rock and is blasted by giant powder as coal is mined.* The next point visited was Coldwater. Between Bronson and this city the land is rolling and very stony. It does not present the sudden contrast in outlines which characterizes the marl regions further north. The Coldwater mill is located near several small lakes. The manager of the works who was present during the prospecting *For analysis see Part I, p. 41 (Clays and Shales by H. Ries). 14 Pt. Ill 106 MARL. could not see that there was any regularity in the depth of marl. It showed no greater thickness in the center. The soundings were a succession of sudden changes in depth. Compare the soundings at the lower end of Long Lake in the Cloverdale region. It must be remembered that these lakes were lined with clear marl at the bottom as at Cloyerdale and not a completely leveled marsh filled in with vegetation as at Bronson. It is well to notice how the dif- ferent lakes compare with swamps in increase of depth toward the center of the marl deposit. The marl lands at this point avail- able for cement manufacture were said to aggregate two thousand acres. The beds in this chain of lakes are to be worked by two fourteen-rotary mills, one at Coldwater, and the other at Quincy. The clay used at Coldwater differed somewhat in appearance from that used at Bronson. It is a surface clay mixed blue and grey in color. Its advantage lies in its easy access and cheap grinding. Janesville. At Jonesville, Mr. Chase Wade "was interviewed. A factory was completed at this point. The bed to be utilized has an area of from seventy-five to eighty acres with an average depth of twenty-five feet. The analysis showed from ninety-three to ninety- five per cent of calcium carbonate.* Kalamazoo. The return trip from Lansing to St. Joseph was made by way of Hastings and Kalamazoo. At the town of Cloverdale the Chicago, Kalamazoo & Saginaw railroad passes through a cluster of lakes, and on account of the promising outlook it was deemed advisable to make a thorough investigation later, the result of which is given in the description of the Cloverdale district. Kalamazoo was next visited, and the site of the former cement plant was examined. A chain of three small lakes form a deep val- ley with a rate of fall so great that a small water flume bringing water about a half a mile from the creek at the headwaters of the lake furnished ample water power for a large mill. The lower of the three lakes was nearly dry and the marl exposed was very light colored with many shells. In this lake there was little or no sur- face muck. In the upper, however, the depth of marsh surface was *See report by W. M. Gregory, upton the plant of the Omega P. C. Co. RECORD OF FIELD WORK BY D. J. HALE. 107 so great as to render tlie marl scarcely available for manufacturing purposes. One of the first factories started in the State was built on this marl bed, but with the old kiln process and with the expen- sive method of handling raw materials, it did not pay.* The next marl bed reported was in the vicinity of Niles. It was five and a half miles east of the town, and covered about forty acres. Deep wells in the vicinity were said to have very hard water, and the hills surrounding terminated abruptly at the edge of the marsh and were of gravel. § 2. Cloverdale. The peculiar formation of the region about Cloverdale makes a very interesting locality for the study of the formation and oc- currence of marls. By consulting a map of Michigan it is seen that the townships of Hope, Barry, and Prairieville of Barry County con- tain an unusually large number of inland waterways and lakes. (Fig. 3.) The country is a network of deep depressions forming dry channels, gullies, water courses and lake beds. Between chan- nels are high gravel and clay hills. The soil is very heavy but forms a greatly varying mixture. At one place it may be a tough till of clay, gravel and boulders which may be traced a short dis- tance and then may be replaced by fine sand, clay or gravel. A cross section of the land as seen in cuts in side hills, washouts or wells shows as much if not more variation. The bottoms of gullies and kettles left by the receding water generally have a blue, black or red clay bottom hidden by a few inches to as many feet of loam or sand. • These dense clays formed the bottoms of numerous lakes and channels, many of which have dried out with the fall of water level, but the largest and deepest of which form the present lakes of the township above mentioned. Within a radius of three or four miles of Cloverdale, Hope Township, on the C. K. & S. there are five lakes and several other holes not entirely dry, a fair sample of the latter class being “Twenty-one Lake” west of Cloverdale. The five lakes examined, all of which contained marl, were Long, Round, Balker or Horseshoe, Guernsey and Pine. The purpose of the investigation was to study the mode of forma- tion, extent and quality of the marls and clay in and about the lakes, so as to ascertain if possible their origin and their adapta- bility to cement manufacture. As the marl is supposed to originate *The quality was very good, as is shown in many places in Kalamazoo, where 20 years has made little impression on the cement. L. 108 MABL. in one of several possible ways from the salts contained in under- ground waters, the relative hardness of spring and well waters surrounding the lake to the hardness of the surface of the lake and its deep water together with its outlet, was determined. This required the collection of samples of water in small fruit jars, which after filling were shipped to the Michigan Agricultural Col- lege for analyses. On page 46 will be found a table and key to analyses with a brief enumeration of results obtained. The sur- roundings of the beds themselves, the nature of the soil, and gen- eral impressions as to the formation of the whole lake may throw light upon the changes which may have brought about these curious deposits. These were therefore noted where possible and the con- clusions drawn from these facts have been noted in Chapter IV. To determine depth and outline of marl beds and to obtain samples at any depth the following apparatus was made. It con- sisted of fifty-four feet of inch pipe (three 18-feet lengths each cut in tw'o making six pieces each nine feet long). Each piece w T as threaded on both ends and when a coupling had been screwed upon one end of each pipe the whole could be united into a continuous tube fifty-four feet long. Fifty-four feet of one-half inch pipe w r as cut, threaded and coupled as above except that the couplings were turned down slightly in a lathe so that when coupled with the half inch pipe, they would allow it to pass freely within the inch pipe. Two shorter pieces (each four feet) of one-half inch pipe were provided, threaded as the others, but each shod to suit solidity of the material to be penetrated. The lake bottoms investigated in this region varied from a fine almost impalpable mud suspended in very deep water to very sandy or very dense clay carbonate. The very sandy and very muddy bottom would be washed off the worm of an ordinary augur. To obviate this difficulty and to preserve the specimens while being hauled to the surface, one of the short pipes was shod with a device which is somewhat of a miniature of a well driver’s sand pump. It consists of a cylinder of iron just the diameter of a half inch coupling hollowed out and chisel pointed. Upon one side of the chisel surface a hole is drilled up the center to the hollow, which hollow is the exact size of the inside diameter of the one-half inch pipe. The hole is stopped on the inside by a ball valve, the ball being retained within the cylinder by a wire passing through the cylinder at right angles to its length three- IiECOBD OF FIELD WORK BY D. J. HALE. 109 eights of an inch from the bottom of the hollow. A thread is cut on the inside of the upper end of the cylinder making the end with threading just the size of a half inch pipe when threaded. It must, therefore, screw inside a coupling which joins it to the short piece of the pipe. When chugged down the valve allows the soft mud to spurt up into the cavity but when lifted the ball drops down into the hole drilled through the bottom, stopping the egress of the contents through the hole by which it entered. At each fresh downward thrust of the chisel the content of the cylinder increases, rising in the hollow half inch cylinder to the top where elbow or one-way coupling may be screwed on to direct the outflow which may be received for examination. The other short pipe was shod with an augur, the worm of which was similar to a ship augur, but the stock of which was hollow so as to allow whatever ascended through the worm to pass up into the half inch pipe as in the previous case. Wlien the marl was somewhat solid as was the case when the chisel was used, an iron poker one-fourth inch in diameter was used to shove the speci- mens out of the pipes. These are the only means so far seen which serve to bring to the surface a correct specimen of lake bottom from any depth. Specimens of lake marl were brought to the sur- face from beneath several feet of mud and fifty feet of water. The outer pipe serves solely as a protection and support to the inner pipe which is liable to break loose from the couplings when forced to great depths. This outfit while absolutely necessary for scientific research was not used by me in later soundings. Where the marl becomes nearly as dense as a limestone, as in the several instances in the Northern Peninsula, the chisel of the sand pump with a double tube, the outer being shoved down as the inner cuts its way through, is the best outfit that can be used. But as the marl in two-thirds of the cases seen lies on top about like “butter in sum- mer,” and at the bottom like “butter in winter,” an ordinary inch augur welded on to inch pipe will retain the marl and stand the strain necessary for numberless soundings. If one man is sounding alone he may use f inch or J inch pipe, but is liable to bury the lower half of his rod out of reach in some marl bed. With the outfit first described, which was fitted up in half a day at a machine shop in Kalamazoo, five lakes were examined in five days with a crew of four section men. A raft was made by slightly fastening two boats together with a framework of boards, the two 110 MARL. heaviest boards lying parallel to each other across the boat amid- ship. These furnished a footing and prevented the tip of the boats from pressure of lifting on their inner gunwhales. Four men rigged a boat raft from a pair of boats and old lumber in about half an hour. They then rowed to any desired position, anchored at bow and stern and made soundings. Specimens were generally taken from bottom and surface of the marl bed at the same spot. Boats and men were then taken to the next lake by team, about a day’s work being expended on each lake. The first lake examined was Long Lake (Fig. 3, p. 14) . It is about three and one-half miles long and very narrow, being nearly cut in two by Ackers Point. The C. K. & S. R. R. runs parallel to it and bounds it nearly the whole length on the southeast side. The town of Cloverdale lies nearly all south of the railroad and at the south- west corner of the lake. The surroundings of the lake are worthy of notice as perhaps having a remote bearing upon the origin of the lake and its con- tents. The southwest or upper end of the lake is bounded by an abrupt hill or bluff about seventy-five feet high, consisting of a dense till or mixture of tough clay, gravel, and boulders, and crowned by hard wood timber. This hill is flanked upon the south by lower land than on the north, the only land touching the lake being a heavy blue clay, which has flowing beneath it several springs. That on the south forms a narrow isthmus between Long Lake and Round Lake lying to the southwest. A canal or ditch had at one time been dug through this neck of land to a distance of two to three hundred feet, and the fall of water from Round into Long Lake was 16 feet, furnishing water to drive a mill. The sur- face of the neck of land, beneath which is clay and quicksand, is sand. The banks of Long Lake are flanked on the northwest and southeast sides by high, rolling, gravelly clay hills, whic'h end abruptly at the shore and through which several cuts have been made by the railroad. The lake rapidly narrows at the northeast, and to its outlet, which is a small creek flowing through a narrow low land into other holes which have once been lakes but could not be reached in any way with sounding apparatus. When the water was higher the whole must have looked like a large river without low lands, with little current, and abrupt shores. The first sounding was made in the narrow channel connecting RECORD OF FIELD WORK BY D. J. HALE. Ill the two halves of the lake at Ackers Point. From here soundings w r ere made at short intervals circling the shore to the right and south side toward the outlet, from thence returning on the west side to place of beginning, and from there on the north side of upper half around the upper end past Cloverdale, and back on south side to place of beginning. The bottom immediately about Ackers Point was of heavy sand and gravel for some little distance out, probably having been washed down from the point over the bed. The first sounding, 40 feet out from heavy gravel shallows, showed a depth of 30 feet of marl, and at the bottom a fairly solid tamarack log, sample of which was bored and torn out, being brought to the surface by the augur. See pages 18 to 21 for a list of soundings taken, showing depth of water, depth of marl, nature of bottom and analysis number where a sample was preserved for analysis. This number, upon reference to the accompanying table of analysis, will give the chem- ical constituents of the sample as far as determined. The sounding No. 1 at Ackers Point was one of the deepest made and the sample taken was among the purest. As the lake widens from the narrows the shallows spread out and divide, following the north and south shores. The shallows extend out from the lowlands on shore perhaps 200 feet, gradually deepening, when there is a sudden jump into deep water, making a shelf much like a sand bar in a river, but not to be expected in a lake. Where oppor- tunity offered, soundings were made on the edge of the shelf and in the deep water outside to determine exactly what was the relation- ship of depth of water, marl, bottom and true bottom. For the sake of clearness this relationship is pictured crudely by diagrams, wTiich will be referred to by numbers. Diagram 1, Plate I, shows the shelf as found by soundings Nos. 8 and 10. It will here be seen that the fall of level of the true bot- tom is more gradual than that of the marl or false bottom as the layer of marl decreases 10 feet. It is not well to form an opinion upon this one relationship, but to watch if it holds true in further comparisons. It is also noticeable that samples 3A and 3B, or speci- mens taken from Nos. 8 and 9 on the shelf show more sand than No. 10 (4) taken off in deep water. This comparison was made about half way down the lower lobe of the lake. The shallows finally again covered the bottom and joined, making an extensive flat which 112 MARL. continued to the outlet. At the head of this flat, and about the center of the lake, was the next object of interest. This was a rocky islet about 40 feet long and 10 feet wide, formerly a cigar- shaped, stony shallow along the center of the lake, The largest boulders are just above w T ater. All are covered with a thick, very soft, white coating of lime, which is fastened to glacial pebbles, covering them all much like a snow storm, i. e., thickest on top and scarcely at all upon the under side, though the stone may be free from others and exposed to the water. The white coating of lime hardens quickly when exposed and dried in air. A cross section shows two layers of granular friable lime, between which is a layer of green organic matter or chlorophyl revealing the presence of liv- ing organisms. Soundings 13 and 21 were made in the mid channel, 13 to the south and 21 to the north of the island, showing conditions on each side of it. With the depth of water the depth of marl is, respectively, 9 and 23 feet, showing that the north channel was originally much deeper, the marl now filling both and making them very shallow. The conditions thus shown immediately at the beginning of the large shallows at the foot of the lake are interesting. A rocky islet just reaches the surface of the water. From this islet the depth of marl increases from a coating a fraction of an inch thick to 23 feet thick on the north, 9 feet thick on the south, with a shallow channel 4 feet of water. Soundings Nos. 12 to 14 show the conditions in a line down the lake, 12 before the island is reached, 14 after passing around the island in a line toward the outlet. These soundings show again that the island is surrounded in two other directions by 12 and 33 feet of marl. The increase is not, as the soundings would indicate, sudden, but gradual, the island seeming like a bouldery outcrop of the bottom, which is at No. 12 heavy gravel, at the island bouldery, and at No. 3 at 33 again fine lake sand. Taken as a whole, soundings 12-22, inclusive, show somewhat the shape of the lake bottom under the shallows to the foot of the lake and as far into its outlet as the raft could be propelled. In no case is the water over 6 feet deep, except in the swimming hole near the north bank. Soundings 14 and 17, taken in nearly a straight line, show a deep channel which narrows and runs into the shallow outlet. No. 16, taken to the north and left of these, shows but a trace of sandy marl with a gravelly bottom. No. 16 is more RECORD OF FIELD WORK BY D. J. DALE. 113 notable, as it was taken from the foot of a hill from which several springs issue. Prodding 50 feet to the south of 16 shows about the same condition, proving that the bed rapidly narrows, but the sud- den jump downward in No. 17 shows that the outlet still remains the old channel, though nearly choked up with marl, with no surface muck. Proddings not recorded as soundings show that southwest of sounding 16, returning to Ackers Point along the north side of the lake, the muck and gravel from steep hills encroach upon the bed. The sudden contrast in the nature of the bottom is shown by comparing sample 6 of table, which is a muck from the narrowest outlet, with Nos. 5 A and 5B, fair samples of marl in deep or old channel. Nos. 21 and 22 (Plate I) are again parallel to Nos. 8 and 10. No. 21, the same referred to as north of Rocky Island, was taken just out- side the swimming hole. Diagram 2 shows the relative change in depths of water, marl and true bottom. Here the relation in fall of marl and true bottom is exactly reversed as compared with Dia- gram 1. The marl bottom or shelf is less pronounced than the orig- inal shelf made by the true bottom before marl was deposited be- cause the marl bottom is like a thick bottom before marl was depos- ited, is like a thick blanket taking away the sharpness of the edge and by its own increase in thickness of 8 feet, making the fall less sudden and the lake bottom more nearly level. Still the increase of water from 4 to 16 feet is so immediate that outline of the white bottom seems to sink suddenly out of sight. The original bottom with an almost immediate fall of (47-23) 24 feet must once have formed a bold precipitous terrace or more likely in this case a small deep kettle.. By the above soundings, together with many proddings and examination of bottom in shallow water by the eye, the following general idea of the broad shallows at the foot of the lake and merg- ing into its outlet is given: The bottom of the deep mid lake sud- denly rises to form an extensive shallow. It even shows above the water’s surface in the stony islet, but slopes down on either side of the islet to form deep channels, the one on the north being deeper (27 ft.), the one on the south 13 feet. The bottom is some- what uneven and pebbly where it is shallow. On the other hand there are many holes, the largest of which, the swimming hole, is 47 feet deep below water with the bottom surrounding it 27 feet. Besides holes there is a deep middle channel north of the stony 15-Pt. Ill 114 MAUL. islet and running into the outlet. The bottom rises on each side to a pebbly shore covered toward the outlet with muck or sandy marl in very thin layers. The shore on the north side has the steep hiil's and springs back of it. The marl lies upon this original bottom covering it, nearly filling up the old channel and hiding all but the deepest hole, which it helps to fill. It, however, forms but a thin incrustation on the rocky islet, but in the channel thickens again to natural depth. It merges into sand and mucky marl (Analysis No. 6) toward shore, but shows admixture of sand even in the deep channel. Upon continuing up the lake on the north side, leaving the broad shoal, a layer of sand is found between Nos. 22 and 23. But the shoaling marl again thickens on approaching Ackers Point on north side, showing no unusual features excepting that it can be easily seen that the old channel past Ackers Point has been filled to a depth of 30 feet with marl like the channel described leading out of the lake, and also that the marl is much thicker immediately in the narrows and about the point than along the shore down the lake. From the point opposite Cloverdale the lake widens with a slight bend reaching out to the north toward the only low land. No. 27 was taken to find if the depth were any greater below the springs which emerge from the heavy clay lowland at the north corner of the lake, but no great difference in depth between that and many other soundings taken in the absence of the springs could be noted. The depth and quality of marl here are just the opposite to sound- ing 16 at the foot of the lake. In this case a fairly deep layer (25 ft.) of marl, with fine sand bottom, was found, while No. 16 showed 2 feet of sandy marl with gravel bottom. At the foot of the steep till bluff before referred to as forming the boundary of the head of the lake there was a boxed spring (Sample 1, p. 46, taken here). Below this spring the water was shallow and appeared to be a sand bar, but upon investigation from boat with sounding apparatus the marl was found to run almost up to the bluff at a good depth, but the sand has washed over it to such a depth that it was reached with difficujty by the pipes. In coming down the northwest shore, which was described as mostly sand in places, where marl was struck, it was found that the marl was interlayed with sand, the augur first sinking through soft marl, then grinding in sand. This would seem to point toward a washing BE COED OF FIELD WORK BY D. J. HALE. 115 action of the sand over the marl during the period of deposit of the marl. Nos. 28 and 29 (Diagram 3, Fig. 4) were another parallel set (Plate I), showing the position of the marl overlying the shelf, being made close to shore under the bluff and in shoal water. The soundings were taken as closely as possible to each other and the changes in depth are very sudden. The marl layer again tends to break the abruptness of the descent of the true bottom. The differ- ence in depth of water on and off the shelf being greatest before the deposit of marl, for before deposit the shelf was 24 feet high, after deposit 15.* Another test opposite Beechwood Point, a short distance below Cloverdale, showed the gradual increase in depth of marl, from deep shoal water in as far as possible toward Beechwood Point, at ]ce Jour)dL*.t( 46 oft. Gre i tafeff Oe///L ''fT- soundings 28, 29, 31, 32, Long Lake. right angles to the length of the lake. In 50 feet the increase of depth of marl is 5 feet to an increase in depth of water of 1 foot (see Fig. 4, Soundings 31 and 32). The general idea of the lake as given by the foregoing examina- tion is that of a very long, narrow body of water. It consists of two quite distinct parts, the deep water and the surrounding exten- sive terrace or bar. Over the whole the marl lies as a thick and more or less even deposit which thins toward the shore edges w T here it is pretty thoroughly mixed with sand, clay or muck. The sudden changes in thickness of the marl layer seem due in greater part to the inequalities in the bottom, which is full of jogs, chan- nels and holes. In all cases excepting Diagram I, the marl in cover- *(45-21) instead of (18-3), see Fig. 4 . 116 MAUL. ing the terrace or shelf made by the bottom always lessens its very abrupt descent, being thicker just outside the shelf in deep water than in the shoal water upon the shelf. In this lake variation in the composition of the marl is very marked. In close proximity to the shore the marl is quite thoroughly mixed with sand. This condition extends out one or two hundred feet, as in samples from Sounding 8. Other instances before alluded to show 7 the marl to be layered with sand and next to the very steep hill at the south- west end the sand has washed completely over, hiding the marl. In this lake the marl layer seems to lie heaviest on the south side of the lake. It covers the whole lake bed, including the bottom 45 feet deep at the center, but lies heaviest over the terrace on the south side and has choked and completely filled holes and channels as deep as the mid lake 47 feet (Diag. 2, Plate I). Compare with sounding 10 mid lake, also No. 14 deep channel. A comparison of their soundings shows the former capacity of the lake. On account of the repeated admixtures of sand and muck the duplicate analyses furnish little data for consideration of difference in depth excepting in the deepest sounding, as 1A and B, 2A and B, 3A and B. These, the most nearly pure samples taken show, if anything, an increase of organic matter with increase of depth. There is no doubt that within a short distance of the bottom sand has worked up into the bed so that a sample, though taken with the greatest care, will show high in sand when taken within tw 7 o or three feet of the true bottom. Here as in nearly all soundings taken during my experi- ence the deeper soundings and the surface samples differ con- siderably in appearance, the deeper being fine grained, compact and of a steel-blue tinge, which w ith a high per cent of organic matter, becomes darker.* The surface samples Avere generally wdiiter, more flaky in appearance and lighter. No. 4 (Sounding 10) is of in- terest on account of its position 45 feet beloAv the surface in Mud Lake. Like the other deep soundings it is high in “organic matter” and matter insoluble in HC1, No. 6 is a fair example of the mucky marl of the lake, little of w T hich was found and that at the narrow ed outlet. Notice the increase in organic matter and insolubles which far exceeds all but 3B, w 7 hich was mostly sand. With this increase of organic matter there is an increase of iron and aluminum as there *See pp. 16 and 18. RECORD OF FIELD WORK BY D. J. HALE. 117 is also in No. 4, the mid lake sounding. This is natural, as organic matter is supposed to aid in the deposit of iron. All in all, sand and organic matter have penetrated this bed from beneath and from the edges. Only in mid lake in the thickest part of the deposit for some distance from the surface down is the marl free from foreign matter. The bold shores and the manner in which the sand is found constantly washed over and against the beds are perhaps good explanations of this condition. Organic matter as a constituent of the marl is found in largest percentages in the bottom of the deepest parts of the lake. Mud or Round Lake, as before described, lies southwest of Long Lake, the two being separated by the high clay and gravel hill. This lake continues southwest, paralleling the railroad for a short distance, then winding to the north. The lobe at Cloverdale and nearest Long Lake was examined for marl. The water of its out- let could not be sampled as it was at the other end of the lake, its waters emptying in a nearly opposite direction from those of Long Lake, the hill forming a divide. The hardness of the water as com- pared with Long Lake, was as 1 to 16, being nearly as soft as rain water. The bottom was heavy gravel or muck with finer sand. Of all the soundings made but one revealed the presence of marl. This marl of poor quality was found 38 feet below surface beneath several feet of silt and by the deepest sounding made in the lake. Standing at the divide between the lakes the general contour of the bluffs or shores of the two lakes would show Mud Lake to be much higher, about 15 feet according to the fall of water at the mill. The hills about it are not as bold and upon the whole its waters do not so deeply indent the surface of the country. The springs which do not flow from the hills slip out at the shore line, are softer and probably are not from as low a level as those of Long Lake, being mostly surface drainage. The wells in Cloverdale and those near the two lakes and on the divide were tested. The deep drive wells of Cloverdale were of the hardest water found. The deeper one on the divide was hard, the surface one soft. . As the people’s idea of hardness and softness of waters in a given vicinity are very conflicting some method was sought to obtain a definite comparison of waters upon the field. A standardized soap solution was made in the laboratory by titrating a known volume against a known weight of crystalized 118 MARL. CaC0 3 or marble, so that every cubic centimeter of the solution needed to make a suds with 50 cc. of water, would imply one degree of hardness,- — one grain per U. S. gallon of calcium carbonate or its equivalent. The soap solution was carried in the field and measured against 50 cc. of spring or well water tested. The figures below, opposite the well or spring located, are the number of cc. of the solution re- quired to neutralize 50 cc. of the water and form a comparative test of the hardness of the water in question: 1. Well, Hotel at Cloverdale 20.00 2. Water of Mud Lake 1.00? 3. Water of Long Lake 16.00 4. Deep well on divide between lakes 16.6 5. Ludwigs (box spring at foot of hill) 12.2 6. J. L. Chamberlain’s well west of hotel 16.6 7. Simon Dayton shallower well on divide 8.0 8. Deep drive well Southwestern Michigan. ... 13. From this it will be seen that the lakes contrast sharply. The deep wells (Nos. 1 and 6) are hard, shallow wells on divide, No. 7 medium, and Mud Lake very soft. No. 5, the deep spring, is quite hard. No. 8, from non-marl region, is softer than deep well waters of this locality. In comparison the waters of the two lakes form a sharp contrast. It is the settled idea in this part of the country that a hard water lake means marl and a soft water lake the absence of it. Several instances besides this under my direct observation were given me and I have never in my own experience found a lake which tested very soft water to show anything but traces of marl.* In the case in question Mud Lake is not cut so deeply into the glacial drift as Long Lake. While there is sand and gravel on the edges, deeper there is a clay hard-pan, while Long Lake is in fine sand bottom. On the divide between the two in the wells driven there is said to be a heavy clay layer. Under these circumstances the only explanation to be seen is that Kound Lake receives the surface drainage of soft water and is withheld from seepage into Long Lake by a clay hard-pan. Long Lake cuts deeper into the drift and receives the hard water springs and drainage from the same layer as the deeper wells. *See analysis of Goose Lake water, of Peninsular P. C. Plant. BECOItD OF FIELD WOBK BY D. J. HALE. U9 The next lake tested (Pl. II) was Balker or Horseshoe Lake. It lies about two miles east of Cloverdale and a mile in direct line at right angles to the C. K. & S. in the N. E. \ of Sec. 22 of Hope Township. It draws one of its names from its shape. It has two lobes or arms and a basin into which both empty and from which issue its outlet. All the attention was devoted to the south lobe and basin as a raft and tools could not be propelled into the north arm on account of the shallowness of the channel which was filled with marl, covered with a few inches of water. The two arms, like the sides of a horseshoe, are surrounded by a low marsh covered with tamarack, a good part of which must have recently been covered with water as it is but little higher than the lake surface. The south arm as it now exists is nearly round or elliptical in form. The east end consists of a large and very shallow flat upon which the first soundings were made. This flat leads into the basin by a narrows almost choked with marl. Here it is well to remark that the marsh vegetation characteristic of marl flats in general is a long cylindrical reed without leaves or branch, which shoots up many feet from a marl bottom or grows in very shallow water, as in this case, where it almost blocks passage of a boat. It is true that this reed* is found to greater or less degree on sandy or mucky bottoms, but it is one of the few practical guides to the loca- tion of marl, though like all others never entirely trustworthy. Except for the shallow flat mentioned the rest of the lake has the shelf-like bottom already noted, the shallows forming a ring but 20 or 30 feet wide about the abrupt descent into deep water. Soundings were made on the edge of the shallows and across the lake from two sight points to determine if possible the profile of the bed or its cross section as cut across the lake. Before describ- ing the various soundings it will be well to notice that the lake proper, which so far as determined is underlaid with a deep deposit of marl does not cover anywhere near all the depression lying be- tween the steep bluffs. The lake as a whole more deeply indents the surface of the country than does Long Lake. The bluffs are steeper and more abrupt, the springs are noticeably larger and more numerous especially near the lake proper, which lies horse- shoe shaped, curving around the south and west side of the valley, the remainder of which is covered with low tamarack marsh. The springs are also of harder water. *Scirpus lacustris? L. 120 MARL. The soundings were begun at the approach to the narrows in the south arm. The bottom as at Ackers Point, Long Lake, rises at the mouth of the narrows into a flat shallows. Soundings 33 and 34 (Diag. No. 5) were taken approaching from the center of the lake toward the shallowest place in the narrows leading’ into the basin. The distance between soundings is about 50 feet, and while the depth of water remains the same, original bottom sinks 7 feet, i. e., the depth of marl increases that much. The real bottom of the lake is the opposite in incline to false bottom. This is paralleled in Long Lake where the narrows at Ackers Point, though choked with marl, were nearly as deep as the remainder of the lake, as the false bot- tom has a gradual incline, not terraced like the sides, but built up by marl. This is true in the east shallows of the lake, but not true of terraces on north shore. (See Diagram No. 6.) The next surprise is the relation of 36 and 37. No 36 is taken on the usual terrace and 37 just outside (see for slopes of bottom No. 6. Fig. 5. Soundings 33, 34, 36 and 37, Horseshoe Lake. T. 2 N., R. 9 W. Diagram No. 6). Here the depth of original bottom is less by 3 feet toward the center of the lake than on the shore terrace. As this shore was lined with marsh it is hardly possible that the marl extends in a perpendicular bank against an opposite solid bank or shore, but in all probability the marl layer extends out a great dis- tance under the marsh. This could not be determined, but this must be inferred from a comparison of the soundings of the other terraces before made. I know of no possible explanation of the almost immediate drop of level (29-15), 14 feet in thickness of marl bed unless currents of long ago where different water level and di- rection of drainage may have cut marl out in some places and filled in others. (See Fig. 5, Diag. No. 6.) From this short point, upon which No. 36 was taken, the line of the soundings was continued RE COED OF FIELD WORK BY D. J. HALE. 121 straight across a slight neck in the lake to the neighborhood of springs on slightly higher ground. No. 38 showed increase in depth of marl again. At No. 39 a sample of water was taken by lowering a corked jug to the bottom, pulling the string allowing it to fill and at once raising to the surface and putting the water into the fruit jar which was sealed as usual. (Analysis 5, page 46.) No. 40, the deepest sounding anywhere made, was interesting both from what it revealed and left buried in obscurity. All the pipe in the apparatus was used without touching the original fine sand bottom of the lake. At the depth of 60 feet the sample which was almost fluid was retained by the sand pump and is shown in Analysis 9 of the table on p. 20. This analysis shows the high- est per cent of Fe 2 0 3 and organic matter of any taken in the lakes. There was no clay and comparatively little sand as shown by the low per cent insoluble. It is also lacking in MgC0 3 , showing a de- cidedly low r er per cent than the rest of Horseshoe Lake. This, as Fig. 6. Section showing Soundings 36, 37, 38, 39, 40 and 42 of Horseshoe Lake. may be noticed in later soundings, is not the only lake in which the marl of the deeper portions gains greatly in organic matter. But such an increase in iron has not been elsewhere noticed. Sounding 41 was a little to the east of the foregoing series, at the mouth of a very large spring. This spring emptied from be- neath a bank at some distance back from the water’s edge and by a small rill into the lake. The boats were shoved in as far as pos- sible and a sounding taken in a few inches of water. The pipes sank with little effort to a depth of 32 feet. The sample from the very bottom was like that at the top, a fine silt with a trifle of lime which could be faintly detected by acid. The spring formed a large reservoir 8 feet across and 5 feet deep. At the bottom was its fountain a foot across and boiling up through black silt. The analysis of this sample of water is No. 4 of page 46. The peculiar phenomenon here witnessed was that one of the largest and liard- 16-Pt. Ill 122 MAUL. est springs should show no trace of marl immediately in or at its outlet. But next comes Sounding 42, made perhaps 50 feet to the west and completing the outline series, the whole of which are set forth, making a cross section of the lake bottom as shown in Fig. 6. Sounding 42, but a short distance from the spring and within 25 feet of solid ground, a bank about 15 feet high, showed marl to 37 feet depth, the deepest sounding anywhere on the lakes. And here it is well to remark that Horseshoe or Balker Lake had the uniformly thickest layering of marl of any of the five. It in fact was so thick that its nature was difficult to discover on ac- count of the slowness and labor in making deep soundings. What- ever the agents were by which such a bed was laid down they should be apparent in so thick a bed. The springs were large and their water hard, but no visible connection between the water of the springs and the marl of the lake could be discovered. The largest spring and its immediate vicinity were free from all but traces of lime. A very deep layering, about same depth as marl, of silt replaced the marl in and about the spring and at its outlet. The interesting phenomena apparent on the Bock Islet in Long Lake, namely the thick lime coating of the pebbles, was again mani- fested in a part of the lake at the shallows at the foot of the lake next to the narrows leading into the basin. This shallow. area covered several acres and was from 1 to 2 feet in depth. The marl layer as shown by the first tw T o soundings varied from the center in toward the narrows from 23 to 30 feet in depth. In an ordinary marsh, especially in the reeds or rushes, the bottom is black or dark-brown from dead rush, twigs, silt, and other marsh accumula- tions, but the bottom here, even in the reeds which ought to catch and hold everything that came to them, was gleaming white marl. In fact it was very much lighter in color than the specimens at the bottom which were in almost every case steel-blue in color. This color with a lack of a trace of organic matter at the surface was in this particular case perhaps explained by a more minute examin- ation of the bottom. A branch of a dead tree leaned over and where it touched the water disappeared from sight. Upon follow- ing it beneath the water’s surface it was found to have become coated with white lime covering, essentially the same in structure and appearance as that of the pebbles in Long Lake. There was the same triple coating of green or chlorophyl between the layers RECORD OF FIELD WORK BY D. J. HALE. 123 of granular lime. In the distribution the lime reminded one of the limbs of a tree after a snowstorm, the greatest thickness of lime being on top and scarcely any underneath. This coating was not confined to twigs, but included anything that had fallen into the water, all being covered so that they lost their identity and blended closely with the brownish white bottom. The last portion of the lake investigated was the basin. This basin is nearly circular in form, is shallow and overgrown with round rushes at the margin and increases gradually to about 10 feet depth at center. Its waters, clear as crystal, lie over a very deep bed of marl. It has three arms, one leading from the north arm of the Horseshoe Lake* one from the south arm and lastly the outlet or creek. All are so overgrown with rushes and choked with marl that boats are forced through with difficulty. The sound- ings made and marked in the list make the average uniform depth of marl about 30 feet. The clearness of the water can perhaps be accounted for by the fact that every particle of foreign matter, organic or otherwise which might find its way into the pool, seems to be surrounded and buried by the lime as described in the case of twigs. Whether the lime or marl be precipitated carrying down the organic matter with the marl or whether the particles attract the lime by the assimilating action of minute animal or plant organ- isms one result is here obtained. The water is left so pure and clear and free from foreign matter that fish or water plants can be seen entirely across the basin. Here it is well to remark that the bottom was overgrown with a plant much in appearance like a small pine tree. In the middle of the lake sound at 40 feet, a deep water plant was brought us, smelling exactly like a pole cat.* The best samples of Balker Lake were not analyzed. The very deep samples were tough and steel-blue, were evidently high in clay and organic matter, but on the whole not so sandy as those of Long Lake. As will be seen by descriptions on page 46, samples of water were taken from two springs, from the deepest part at sounding 41, from the surface and outlet of the basin and it can be easily seen that on account of the intensely marly nature of the lake its waters should reveal something of the marl’s origin. It is impossible to reconstruct the lake as it once existed. Its bold shores and large marsh hint at a far greater depth and volume 'See pp. 56, 89. 124 MABL. of water with currents which may have done something toward disturbing the evenness of so thick a layering of marl. As in reality a small portion of the whole bed was examined the rest lying under the adjoining marsh, the cross section (Fig. 6) is rather incomplete and the individual soundings do not show the pronounced relations between true and false bottom. Attention is especially called to the sounding mid lake, which shows the re- markable difference in quality of the marl in the deep water, as it contains much iron and organic matter and only about half calcium carbonate. It has been suggested as an explanation that the or- ganic matter of the lake upon account of the dish-like shape of the lake tends to slide into the central or deeper portions, giving them a more highly organic character. It was especially noticeable that Long Lake contained a more caustic marl than Horseshoe Lake. In Long Lake the hands of the operators were severely chapped and seamed, while this was scarcely noticeable in Horseshoe Lake. The marl did not seem to bite. A review of the springs of Horseshoe Lake hardly seemed to justify the theory of immediate precipitation of lime. There was no trace of marl in or around them although at a distance of a few hundred feet the deepest marl wds found. L T pon the whole this lake is very deeply indented in the surface of the country, having high, steep bluffs. The portion covered by water has a steep ter- race or shelf, less shallows than Long Lake, with a deeper and larger lake center. It has a thicker, more homogeneous marl with considerable organic matter distributed most largely toward a somewhat clay bottom. The next lake visited was Guernsey. This lake lies northwest of Cloverdale about l\ miles in Secs. 17, 18, 19, Hope Township. Its two long lobes form like Long Lake what might have once been an old river valley. This is continued by a rather narrow marsh and creek forming an outlet. This marsh, several miles away, is said to contain bog iron. The lower lobe only could be examined, as it was impossible to get the raft through the narrows between the lobes. The lobe examined appeared something like a mitten. The wrist forms the extension, shallows and narrows leading to the north arm the hand. The main body of deep water is fringed with shallows. The thumb to the west was a long lagoon lying in marsh. The south end was 125 4 BECOBD OF FIELD WOIiK BY D. J. BALE. all sandy bottom destitute of marl. Yet the usu^l terrace was there and so close to shore that teams must be careful not to drive in far for fear of suddenly slipping off the shelf into deep water. A spring was found near the south end, of which the water was sampled in jar 9. (See page 46, Chap. IV.) A small deposit of iron was on the vegetation, but no trace of lime could be seen in the vicinity of the spring. As proddings were made from time to time up the east side of the lake a sandy marl was found which increased to a depth of several feet as usual at the approach to the narrows. There were broad flats or shallows which, being covered with marl, gave the neighboring fishermen the idea that there must be an extensive deposit of marl. Upon actual sounding it was found that the flats were covered by 1 to 3 feet of water, beneath which was 3 to 4 feet of marl and below this a tough, almost im- penetrable blue clay bottom. The lagoon opening on the west side, described as the thumb, contained nothing but fine silt to a depth of 25 to 30 feet. It seems queer, but is a fact, that upon the west side of the narrow tongue of marsh dividing off the lagoon there should be pure silt of the ordinary marsh or river formation, while upon the east side in lake proper there were 20 to 25 feet of the best marl in the lake, the bottom also in the latter case show- ing strict terrace formation, which was tested in the usual way by Soundings 49 and 51. In this case the bottom was found nearly level and about the same depth beneath water level as that in the lagoon. West of it the difference in the terrace was, in this, the first instance cited, caused by difference in thickness of marl layer. But this is a very slight terrace. Compared with real ones previ- ously examined there is but a four foot fall. This could have easily been displaced or washed over the sand, which is further south and to which it sinks. An examination of analyses 12A and B, 13A and B, and 14A and B shows a very interesting condition of the bed. The surface samples, 12B, 13B and 14B show by far the higher lime and in every case a much smaller percentage MgC0 3 , but far the higher percentage organic matter and lower percentage insolubles. In other words the marl is at the surface fair marl but with considerable organic matter, but at the bottom it merges into a blue clay which of course is higher in insolubles, higher in MgCCK and much lower in organic matter, except in case of 14A. The MgC0 3 is not very high, and as the clay is very fine 126 MARL. grained, if not too deeply buried, it could be used mixed with the marl for factory purposes. 14B is one of the best samples found in the lakes and was taken in Sounding 32. To recapitulate the important features of this lake. It is long and river-like, undoubtedly one of the old glacial valleys like Long Lake. The layering of marl lies toward the west side of the south lobe, is underlain by blue clay, is from 2 or 3 to 28 feet deep, is not as uniformly thick as Horseshoe Lake, does not cover the whole lake, is flanked upon the west side by a deep lagoon filled with silt. Its springs show no unusual trace of marl. It does not indent the surrounding hills very deeply, being the shallowest placed lake so far visited. The next lake examined was Tine Lake. This lake, north of Cloverdale, is in Sections 8 and 9 of Hope Township. The portion covered by water when the lake was examined rendered its out- line very different from that given on the county atlas. It con- sists of three large lobes, the narrows of which were larger and less obstructed than any so far visited. Time permitted only the exam- ination of the south lobe and its connecting narrows. The first sounding was made at the cove or landing where stock and teams are driven and row-boats usually land. • The surface of the marl is muddy, which is an unusual occurrence not found elsewhere in the lake. It may be due to the constant roiling at the water’s edge. The next sounding was made across that end of the lake at a large boiling spring. This spring was about a yard across and its loca- tion was marked by a large number of bubbling fountains which boiled up through the marl 10 feet thick. This is the first case where marl was found in or about a spring. The analysis of this marl (No. 17) shows it to be remarkably free from sand or clay, but quite high in organic matter. Although the bottom from which the spring came was fine sand like the rest of the lake, and al- though the water was washed up through it and the marl, the ascending stream seems to have no power left to mix the sand with the overlying marl. As the remainder of the south lobe presented no unusual ap- pearance, a series of soundings were made across the first narrows, which were perhaps 100 feet wide. These soundings are numbered from 3 to 8 on the record sheet. Figure 7 shows the cross section of the bottom as platted from the soundings. RECORD OF FIELD WORK BY D. J. HALE. 127 By this it is seen that from Sounding 3 to Sounding 7 there was a deep original channel nearly filled with marl except where gouged out in the center of the modern narrows. On the west side Sound- ing 8 shows another channel almost entirely filled with marl. As the true bottom shows no sudden terrace or shelf so the marl or false bottom, though it slopes to form the deep depression of mid- channel, does so gradually without the sudden step or terrace for- mations. To appreciate this compare true and false bottom here and in Diagram 3, Plate I. From the way the marl lies it would appear worn away in mid-channel. It would be unfair to establish this as a fact as the marl might have formed more easily about the side or points forming the narrows and so have built out into the channel. The samples taken from this lake are analyses Nos. 16, 17 and 18 A and B, 20A and B. They average better than those of other lakes of the group. The first, No. 16 (Sounding No. 1), is the poor- est. Though taken about 30 feet from shore and at a depth of 20 feet, the sample contains considerable sand which has evidently worked out from the shore. This is shown by a high per cent of “insoluble in HC1.” The surface was before described as being covered with organic matter, the only black bottoms on the lake and probably due to the landing. No. 7, taken in front of the boiling spring at 10 feet depth, shoves a very high per cent of organic matter though otherwise light in A1 2 0 3 , Fe 2 0 3 , insolubles, and MgC0 3 . The especially low percent- age of insolubles and A1 2 0 3 , Fe 2 0 3 are interesting, as the sounding showed the spring boiled up through a 10-foot bed of marl. At the bottom was fine sand. This sand was not mixed with marl as would appear natural, but the sample taken was unusually free from insolubles as the first column indicates. Again, this sample is the freest from Fe 2 0 3 , A1 2 0 3 of any taken. The spring then left none of its iron in passing through clear marl, but carried it away in solution. Near by there is an outlet to this lake and this out- 128 MARL. let, several miles away, contains a large deposit of bog iron ore though within the immediate vicinity there v is no trace of it and the samples are free from all but slight amounts of iron and alumina; .8$ to 3J$. In 19, 20 and 21 both surface (B) and deep (A) samples were taken. These samples belong to Soundings 5, 6 and 9, respectively. (See Diagram No. 8.) These soundings form part of the cross section of the narrows and are about 20 feet apart. Some investigators have thought that deep samples show higher percentage of magnesia than do shallow, so it was thought advis- able to compare analyses of surface and deep samples in order, if possible, to arrive at a conclusion as to the increase in percentage of magnesia. Such a conclusion might assist in tracing the origin of marl. In the three pairs of analyses, 18, 19, 20, the first two show the highest magnesia at the surface while 20 is a little in favor of the deep samples. In two cases out of three, 18 and 20 against 19, the organic content is the greater with the increased depth. In all three instances Fe 2 0 3 , A1 2 0 3 is highest in deep samples. In 19, where the organic matter varies least with depth, Fe 2 0 3 , A1 2 O s varies least. This sample, Sounding 6, is, however, but 9 feet in depth, giving the least distance of any of the three soundings sampled, between surface and deep sample. It is notice- able that there is less variation in any of these components than in the soundings where distance between samples is greater. In two out of three the insoluble matter is highest in the lower sounding. In comparison of future samples from different depths it will be well to keep in mind the mutual relation with varying depth of the samples in order to find if possible the constant variation in com* position of a marl bed. This would be of little aid to the factory chemist as the dredge makes a clean cut from bottom to top, but may assist in our scientific research for the origin of marl. For the sake of clearness and to give some system to the perusal of further descriptions it is thought best to review the work upon the five lakes so far discussed. CLOVERDALE REGION SUMMARY. Long Lake is covered with a sheet of marl varying from 20 to 30 feet in depth. The bottom of the lake is not level and even, but has a more or less regular terrace on the south side, a deep channel which runs from mid lake under the marsh at the present outlet, BECOBD OF FIELD WOBK BY D. J. HALE. 129 narrowing at the same time to a width of thirty or forty feet. This channel, which forms the deeper portions of the lake, is choked at Ackers Point, about mid-way and the lakes outlet, with a depth of marl of about thirty feet. At a depth of twenty-five feet of water in mid lake there is twenty feet of marl, showing that the bed thins in water of that depth. Besides the main channel there are many sudden holes in the outline of the original sand bottom, and also a sandy islet where pebbles and stones crop out at the surface. To each side of this islet the channel, while it is not as deep as toward the outlet of mid lake, is filled evenly with marl. The depth from surface of water to original bottom is, on the north side of the island, 27 feet, on the south side 13 feet, while the depth of water is four feet in both cases. The accompanying map of the lake and cross sections of the bed are made to show the manner in which the marl is deposited upon the terraces. The effect of the marl in all cases is to round over and fill up holes. It deposits sparingly upon the rocky islet and fills the channels to each side. It thins toward the center, but produces a less sudden descent from the terraces than would have been found on the original bottom, before the deposit of marl. The deposit lies evenly at both ends, and along the southeast shore, but is thin and persistent only at points which project from the northwest shore. The lake being three miles long and but a few hundred feet wide, and having high gravel and clay hills, is very subject to washings of surface soil. Its composition is heavily influenced by sand and clay rendering it of little use for factory purposes. The waters flowing into the lake by its springs are very hard, as were also the deep drive wells of the immediate vicinity. The lake adjoining, called Mud or Round Lake was remarkable for its contrast. It apparently received the soft waters of surface seepage, was clearly of higher level, with sand, clay and mud bottom. A trace of marl under several feet of muck was found in thirty-five feet of water. The saying that “hard water makes hard marl” was very well exemplified in these two lakes. From a view of the two so close together, yet so different in their content of marl and the hardness of their waters, it would appear that Mud Lake indented the surface of the country less and did not receive the drainage of 17-Pt. Ill 130 MARL . the springs from the deeper strata of soil. Its surface is about fifteen feet higher than that of Long Lake and the ditch connecting the two lakes had furnished water fall sufficient to run a mill. Horseshoe Lake (Plate II) contains the deepest and most actively depositing bed of marl and the deepest of any of the lakes investi- gated in this region. The lake as it now exists encircles a portion of the whole basin in the form of a horseshoe, the remainder being covered by marsh. The largest and most intensely carbonated springs and lake water were found here.* This lake, running from 20 to 37 feet of marl on shallows. It also shows the same tendency to fill the sudden step made by the greatly increasing depth from the shallow terraces to deep water. In this deposit the greater variation in composition resulting from increase in organic matter, is seen every time a deep and a shallow sounding are taken in the same spot for comparison. The great coldness of the deep water of mid lake is sharply contrasted with the luke-warm water of the shallows. The great abundance of plant life in shallow water and the thick incrustation of every object covered by shallow water are very striking, as are the absence of incrustation plants from deep water.f This is the remainder of a very large deeply indented lake basin, which has held the hard waters of its deep springs for many centuries. Nearly all the basin is sealed by marsh growth. The portion remaining consists of the waters of Horseshoe Lake, which are actively depositing the best grade of marl at the surface of its shallows. The portion of Guernsey Lake examined is remarkable for its strictly local deposit of marl. The thumb described contains very good marl on its east side and a corresponding depth of loose lake silt on its west side in the lagoon. On the one side the particles of silt are surrounded by the deposit of marl, making a marl bed with 22$ calcium carbonate at bottom and 64$ calcium carbonate at surface, while on the other side of the tongue of land fifty feet away there is a deposit of twenty to thirty feet of pure silt. At the head of the lake there is no marl at all, though there is a ter- race and a spring of water containing 130 parts in the million of calcium carbonate, which is a fair average. It appears from this that conditions are not always favorable for the growth of marl, given the same kind of bottom and the same water. True, the con- *See Nos. 4, 8 and 5, page 46, Chapter IV. tWesenberg-Lund. BE CORD OF FIELD WORK BY D. J. IIALE. 131 ditions are not exactly identical with those of the deep deposit at Horseshoe Lake. The springs are not so plentiful or of such hard water. The sandy spot alluded to is bare and unsheltered. Pine Lake shows fairly hard water, a good deposit of marl over the entire lake and not as great difference in content of organic matter as Horseshoe Lake or Guernsey Lake. This v^as a case where a spring bubbled up through ten feet of marl without bring- ing sand into its composition or otherwise affecting its quality. We must conclude that the immediate locality of springs has no effect upon the position of the marl either in regard to depth or quality. The samples of water taken are interesting only from one point of comparison. For the whole list of samples and analyses of some, see page 46, Chap. IV. CaC0 3 COMPARED IN PARTS PER MILLION. Springs. Wells Surface. Water medium deep. Horseshoe 200. 160 70 100 117 Long Lake 100 160, 156 40 O-uernsey 130 40 Pine Lake 170, 136 80 Mud Lake 80 30 53.6 From these comparisons and those made with soap solution in the field, it appears: that the most intensely marl lakes have the most heavily carbonated waters, the soft water lake showing much poorer in all cases ; that in the lake itself, the deep water contains the most gas and carbonates and that they uniformly disappear in every lake at the surface, the gas being lost entirely and the car- bonates in a fairly even proportion. These well, spring and lake waters substantiate the idea that the water’s hardness is respon- sible for the presence of the marl in a somewhat direct ratio to the strength of the carbonates it contains. § 3. Pierson Lakes. I visited Big and Little Whitefish Lake, southwest of Pierson three or four miles, Pierson Township, Mecosta County. The general outline of the land is a rather monotonous level, but in the neighborhood of the lakes it is considerably broken, but not as much as at Cloverdale. Big Whitefish Lake is about three miles long by a mile wide. Its shore level sinks into extensive shallows consisting of somewhere between 20 and 30 feet of marl. 132 MAUL. At near the center a “blind island” rises from the very deep water and is covered by about 25 feet of marl. Blind islands are met with often in these lakes. They are small shallows in the deep water of mid lake. There are large flowing springs along the shores of the lake. These springs deposit iron upon the stones and vegetation at their borders, but the marl in the lake below them appears to be unaffected by iron coloring. One spring at the south end gave marked smell and taste of sulphur and was valued highly for its medicinal properties. At its southeast corner the lake is bounded by a sandy ridge containing gravel with fossils and granite boulders. Beyond this ridge, perhaps 200 yards to the east, is a deep hole or smaller lake, about 200 feet across. This is fed by intensely irony springs and empties by a deeply cut creek into the larger lake. The sudden fall gives about ten feet of water fall for turning light machinery. The creek is very interesting. Its bottom is composed of marl which continues up its steep bank 20 or 30 feet. About half way to the top of the ridge upon the sides the marl is shown on the up- rooted stumps of large forest trees. Between the two bodies of water mentioned is a kettle not as deep, but with sides so steep that there was some speculation as to whether the Indians had not dug it out to make their mound which was on the ridge to the east. Upon examination a crude marl was found in the bottom of this kettle under a few feet of loam, showing that it, with the low ground adjoining, had been under water. It looked as though the three, the larger lake, the hole and the kettle between had once been one and that the creek bed was once but a connecting channel. A bed of clay was examined on the farm of Mr. Shanklin some little distance from the lake. The clay bed was covered by 2 or 3 feet of red and yellow ochre, which had at one time been dug for paint. An augur was used and the ochre and clay bed beneath penetrated to the depth of 10 or 12 feet. The samples brought up showed a fine clay which reacted feebly with acid, but was in most cases mixed with sand, which seemed to run through the bed some- what in layers, there being found several samples entirely free from grit. Little Whitefish Lake, two or three miles from Whitefish Lake, was visited briefly and a few soundings made at the south end. Here there was a swamp at the southeast corner which was RECORD OF FIELD WORK BY D. J. HALE. 133 probed to a depth of 15 feet without striking anything but silt. The marl upon this side seemed slightly red or brownish in cast, but at the west side it was much whiter. The marl was (28 ft.) deeper upon the points or shallows running out from the shores and of the prevailing consistency. North of the marsh and jutting almost into the lake was a bluff showing 25 to 30 feet of clay which was nearly like rock, of light color and was calcareous. § 4. Lime Lake and vicinity. The lakes about to be described are near Cedar Springs in the northern part of Kent County. The country through which our guide led us showed very distinctly the effects of the glacial action. Steep hills, waterways, creeks and small lakes produced a very undulating surface. The first fact worthy of notice was very strik- ingly illustrated in the examination of road cuts in several side hills. These hills were generally coarse sand which was thor- oughly seeded with small pebbles and boulders. At varying dis- tances up their sides, clay strata projected slightly, or their exposed surfaces were worn down and hidden by sand and gravel from above. These clay banks are typical of half the clay in Michigan. In color it is light or ashy gray. Its texture or grain is ruined by the admixture of fine sand. Upon addition of acid it effervesces more freely than many samples of marl because it contains so high a percentage of carbonates of calcium and magnesium. Upon a further examination of the bank or hill the carbonated condition of the soil is found to continue not only in the clay, but also in the loose and apparently pure sand as, upon contact with acid, the surfaces of the sand grains freely effervesce. Parallel with the stratum of clay are often found small ledges or boulders of a matrix of coarse sand in which are cemented small pebbles. The upper surface is even as if smoothed by the leveling action of water, although the rock, as it has now become, is fifty feet above the level of a stream and buried in a hill. The lower surface of this rock or tufa is uneven and jagged. Upon the addition of acid to this rock it also, as in the case of the sand and clay, bubbles with escape of gas, and the particles of sand and the pebbles fall apart showing that the matrix or binding element is not the insolu- ble sand, but the very soluble carbonates.* A comparative test for hardness was made upon the springs and creeks of this region during the trip and all were found to be very *A similar recent sandstone occurs beneath the clay bed at Harrietta. L. 134 MARL. hard. Lime as a carbonate was found to permeate very thoroughly the soils of the whole district, and the soil mixing effect of glacial action was very marked. Lime Lake. The first lake visited in this region was Lime Lake. An old kiln was still to be seen marking the place where marl from the lake had once been burned for lime. The lake as a whole made a very sharp and deep indentation or circular hole in the plain of the sur- rounding country. The shallows on its shores formed a white but narrow margin ending in an abrupt terrace and very deep water toward the center of the lake. The shallows, the dry land of the valley, and the broad entering valley of a small creek, formed a solid body of very white marl from fifteen to twenty-seven feet in thickness. Shells, large and small, constituted nearly the entire body of marl even at the greatest depth and they preserved their form perfectly. This is certainly one instance, at least, in which shells can furnish nearly if not all the excuse for the origin of marl. Several samples of marl taken a few feet below the surface, upon drying, turned from nearly white to a pronounced red. This was very likely due to the oxidation, upon exposure to air, of the fer- rous or nearly colorless iron to the ferric state in which the color is red. The valley opening into the lake from above was very large and probably once formed an old glacial valley. It connects Lime Lake with several higher ones and is a pure marl bed with but slight covering of surface soil. Twin Lakes. These lakes were remarkable for their great contrasts with each other. They had no visible union, but they were said to connect with each other by an underground channel. The lower one was shallow and sandy, the upper one was a hole between huge banks which, in cuts made by washouts, were almost identical in nature with the sand hills before described. Its banks or bluffs, fifty feet high, descended with but a step for a shore line, directly into water, making no shallows whatever. So abrupt was the descent to the bottom that one standing on shore could shove a pole out of sight in the water without touching bottom. The lake is said to be one hundred and seventy feet deep, according to the measurements of the Fish Commission. The springs emptying into .the lake formed a glistening scum of iron. This lake is a very good example of the hundreds of holes made by glacial action and without which we could not have the conditions necessary for the formation of marl. RECORD OF FIELD WORK BY D. J. HALE. 135 § 5. Fremont District. Fremont Lake and the town of Fremont are situated on the Pere Marquette railroad in the northern part of Sheridan Township, Newaygo County. The country surrounding the lake is rather level and the lake makes but a slight indentation in the surface. In sharp contrast to this the lakes in the hilly country before examined seemed to depend for the depth and extensiveness of Fig. 8. Fremont Lake, Newaygo County. their deposits upon the comparative depth at which their basins were sunk below the level of the surrounding country. Fremont Lake has a very shallow basin and it therefore differs entirely from the regions before mentioned. The lake is to be the site of a fourteen-rotary cement plant to be run by power transmitted from the other factory to be built at Newaygo. 136 MAUL. The map of Fig. 8 represents Fremont Lake or the portion of its basin covered by water. The dotted line encloses that portion most available for cement purposes. The depths of marl sound- ings are shown by the figures. The drawings together with the accompanying analyses were kindly loaned us by Mr. John Cole.. The lake examined closely on the side toward the town shows marl shores covered partly with sand. The shallows which are very extensive extend out toward the center of the lake as long, parallel peninsular shallows. The change from shallows to deep water is very abrupt, even between the peninsulas. These abrupt changes are very similar to those at Cloverdale. The soundings in this lake show the shallow marl toward deep water. Soundings of eighteen and twenty feet are found toward the center as con- trasted with thirty-four feet at the inside edge. There is a blind is- land in the center of the lake which brings within reach much valu- able marl. The peninsulas above mentioned are covered with from one to three feet of water, the marl has no covering of organic matter and supports a thick growth of the cylindrical rush known as marsh-rice which is so prevalent as to be almost characteristic of marl beds. The analysis of the sample of marl by Prof. Delos Fall of Albion was as follows: Silica 2.28$ Aluminum and iron oxide 1.60 Lime 88.25 Carbonate of magnesium 1.40 Organic matter and undetermined 6.47 Carbonate of lime after the removal of organic matter ... : 94.85 Beneath this marl lies a blue clay which was analyzed to deter- mine whether it would be of proper composition to mix with clay in the manufacture of cement. It was found that the clay directly underlying the marl contained over seven per cent magnesium oxide, which was considered unsafe. The startling fact from a scientific point of view is the sudden variation in content of mag- nesia in the marl and in the clay immediately beneath it. 1.40$ magnesium carbonate equals .7$ magnesium oxide and the propor- tion of the oxide in the marl to the oxide in the clay beneath is then, as .7 to 7. For this reason either a totally different agent or RECORD OF FIELD WORK BY D. J. HALE. 137 the same agent, with greatly varying power, must have controlled the deposit of magnesia in the marl and that beneath it in the clay of Fremont Lake. Mr. Cole showed me another clay from a different part of the country, which was to take the place of that just mentioned. It appears as a dense blue shale and the following is the analysis as given to me by Mr. Cole (chemist not known) : Silica 42.94$ * Alumina (A1 2 0 3 ) 12.94 Oxide of iron 5.73 Calcium oxide (CaO) 12.93 Magnesia 2.97 Loss by ignition 18.94 Alkalies (sulphuric acid, etc.) 4.07 There were said to be numerous hard water springs in the vicin- ity. A drive well near the station was examined and showed very hard water. A well bored near the lake failed to strike anything but marl till at a depth of thirty feet it penetrated a limy clay. The clays of this region are very calcareous. § 6. Muskegon District. Bear Lake just north of the mouth of the Muskegon River, was visited and probed for marl. It appeared after investigation that Bear Lake was but the old mouth of a river. Muck and silt to the depth of 35 feet was found, but no marl, excepting at one place. Near its outlet was a streak of clay at right angles to the outlet and to the mouth of the Muskegon River. This clay was found to run under the lake and above it and beneath the silt was a foot or two of genuine marl. Several soundings were made at the mouth of a creek emptying into the lake and also in the rushes at the head of the lake. The bottom was in every case a foundation of fine sand covered by many feet of silt. The Muskegon River flats were said to contain marl and several samples were submitted to me by Prof. McClouth of the Muskegon High School, but nearly all of them showed an intimate mixture of clay, sand or muck with the marl demonstrating nicely what has usually appeared, that marl generally loses its individuality and becomes an admixture of marl with sand, clay or muck in the neighborhood of running water. § 7. Benzie County. Benzie County contains a number of marl lakes, several of which are drained by the Platte River. 18-Pt. Ill 138 MARL. Also a company was formed to work the marl in Herring Lakes, five miles south of Frankfort. These two lakes are connected by a stream which has a waterfall of 15 feet. This is to be obviated by cutting a canal through a bend in the creek partly draining the upper lake. This lake contains a deposit of marl about 30 feet in depth. It is fed by numerous springs, which form a network of creeks. To the east the lower lake is very deep and is connected jvith Lake Michigan by a short channel which is to be deepened for the entrance of large lake boats. The bluffs of clay about Frankfort were examined for a cement clay, but none was found. Some clay was quite free from grit, but all was highly calcareous. On a farm north of Frankfort there was a sink hole, some 300 feet from the lake. There was no visible drainage, but upon the bluff opposite the hole there was a seepage of water from between the clay and the sand lying above, showing that the water might in part be held in and turned lakeward by a dense underlying stratum of clay. This may also explain the drain- age of some marl lakes which have perfectly fresh water but no visible outlet. Crystal Lake was examined but showed no signs of marl. It had a very gradually increasing depth and pebbly beach like Lake Michigan and was unlike most marl lakes in formation and slope. The Lake Michigan bluffs, which are here very high opposite the lake, suddenly sink to within 15 or 20 feet of its level. The sharp well-defined channel with abrupt banks on each side seemed to show a former connection between the two lakes. A comparative test of hardness of water showed them about alike. In Frankfort River, south of the town, there is a large elbow or basin formed by the river bottoms which broadens into a large marsh to the south. This marsh looks as if it could easily have been a shallow basin or lagoon. It is said that several thou- sand acres are underlain with 2 to 3 feet of marl. That examined was under 2 to 3 feet of muck and was very white. § 8. Harrietta. In the trip from Frankfort to Cadillac the clay banks of Harrietta were passed.* It was near this point, the highest in that part of the State, that marl was reported as lying in a creek and upon its banks. Marl is deposited everywhere regardless of elevation. § 9. Escanaba. The country about Menominee is largely limestone. A lake in Sec. 6, T. 24 N., R. 26 W., is said to be marly. No marl lakes were popularly known around Escanaba. See Part I, p. 53. EE COBB OF FIELD WORK BY B. J. HALE. 139 At Escanaba a light prospecting outfit was made consisting of the following: 40 feet of f-in. pipe, cut in 4-foot lengths. Couplings for the above. lj-in. common wood augur bit welded to short piece of f-in. pipe. 1 alligator wrench. This could be loaded into a sack, strapped up and checked from one station to another. It is found that for soundings in deep water for marl J-inch pipe is the safest, although the smaller f-inch pipe is lighter and can be raised or lowered in the marl easier, but is liable to bend out of shape and tear out at the couplings. A 2-inch bit is a good medium size. A larger one requires too much work to raise it through the marl. A smaller one does not hold the marl in its coil. For practical purposes this is the most serviceable outfit for the average marls of Michigan. But in the Upper Peninsula marl was found too hard to penetrate by this means and in the Lower Peninsula marl and mud are sometimes too soft to be held in the worm of the augur. For all round sampling we find the outfit at Cloverdale very good though bulky. At Little Lake, the junction of the Chicago and Northwestern and the Mun- ising R. R., Marquette County, Upper Peninsula, several lakes were examined. Their water showed in comparison as 2.7 to 11-13 as contrasted with that of Lake Michigan. This was quite soft and bore out the result of investigations at Cloverdale. The lakes were in a low, level country themselves, had very low banks, and noth- ing but seepage springs. Upon sounding they gave depths of marsh silt varying from 12 to 25 feet upon a fine sand and gravel bottom. § 10. Munising. At Munising no marl lake was found in the immediate vicinity. I was informed by the superintendent of the road of a marl lake once discovered in sinking a shaft. The boring was carried through 20 or 30 feet of muck, when the drills passed through about 30 feet of marl and then into sand and rock again. The well filled and the liquid marl was pumped up as it constantly filled the hole and prevented progress. Finally the upper layer of denser muck sank like a flap till it shut out the liquid marl and the boring was com- pleted, no ore being found. § 11. Wetmore. At this village there was a large creek fed by a mass of boiling 140 MARL. springs in its bottom. This bottom consisted of a very dense white marl covered by a few inches of silt. When the augur penetrated with the greatest difficulty and was pulled out with a specimen, a new spring boiled up in the puncture of the crust made. This point is near the divide of the Upper Peninsula upon the side which drains into Lake Superior. The creek is bounded on either side by somewhat low hills. The marl obtained was half way in con- sistency between marl and limestone. It was rather granular, though the particles themselves examined under microscope cannot be distinguished from those of dried marl. The marl is an almost pure white and very heavy considering its volume in comparison with ordinary marl. The creek is said to drain several lakes nearly upon the divide. § 12. Manistique. Here lime kilns were visited. The whole country is limestone and there are no lakes or flowing springs or wells in the immediate vicinity. The limestone itself is over 30$ MgCO s but burns well, making a good lime. A sample of marl from a dried up lake-bed some 30 miles distant was shown me. Its analyses revealed 95$ CaC0 3 and it seemed one of the purest samples seen in my trip, being, with the exception of a little darkening organic matter, pure white. The analyses showed but traces (slight) of MgO and this too in a country noticeably abounding in magnesian limestone. The lake-bed from which this came showed a basin-shaped depres- sion of about 37 acres filled with purest marl from a shore depth of 1 to 2 feet constantly increasing to a center depth of 29 feet. This is a good example of a completed lake. § 13. Corinne. A spring creek was examined and a small bottle of water taken.* Most of the bottom of the creek was underlain as at Wetmore with a very hard granular marl 2 to 3 feet deep with clay beneath and one or two feet of muck on top. There were no indications that this had been a lake bed in very recent times, though the ground which formed a small swamp had very likely been under water for differing lengths of time. A lake about three miles farther south was visited and had a peculiar history. It was said that it in- creased in size during the spring months, but in summer, July and August, it suddenly disappeared and it was wondered if it found some crevice in the marl and suddenly emptied itself. It is prob- *Sample of water taken from large spring. See (No. 2). EE COED OF FIELD WOEK BY D. J. HALE. 141 able that the lake filled up from spring rains and then gradually dried and by summer it had got so shallow that when steady hot weather came the thin sheet of water left evaporated quickly and the shore line advanced very suddenly. When the lake was visited in late July it had 3 or 4 feet of water upon it and was one great shallow of marl one mile or more long and a quarter mile wide. The marl was the purest seen, but was so dense and granular that the augur did not penetrate over five feet. The marl, however, formed rather dense layers. As the augur penetrated it it would sink easily for a foot or two and then strike an almost impenetrable layer which seemed like sand, but the specimen obtained would be very hard marl. It is probable that this stratified condition or layering of different density is caused by the sudden drying and baking given the crust during the annual drying of the lake. The lake was in an extensive forest bottom and I was informed that it was fed only from the surface waters which collected in the wetter portions of the year. I could hear of no marl in the region of Trout Lake, but near St. Ignace there were deposits of marl and dolomite. No marl could be located in the neighborhood of Mackinaw City. Little Traverse Bay had marl underneath the sand as shown by driving of spiles for piers. At the straits and for many miles inland the immense area covered by limestone scraped bare of glacial drift perhaps shows where the lime of our marl once originated no matter how subse- quently deposited in the lakes. The immense number of small, smoothly rounded limestone pebbles show that a great body of lime must have at one-time been washed away by the action of water, being removed in the form of either a fine sediment or a solution. § 14. Grand Traverse region. The district about Traverse City was next visited. The marl upon a low basin on the asylum grounds was examined and found to contain an underlining of marl about 2 or 3 feet deep. Here it became evident as at Frankfort that an originally greater depth of water lying over the marl did not imply a greater depth of marl, but rather the opposite. This is considering the water level of Michigan as a whole. This, together with the thin bed at Frank- fort and others very near the water level of Lake Michigan, showed unusual thinness. In this case a large basin had been grown over with muck and covered with debris to the depth of 4 to 5 feet and 142 MARL. only showed where a large ditch had been dug. The marl was very hard and dense, very white. The lakes about Interlochen were next visited. Duck Lake had been dammed to allow the floating of logs so that the marl being under greater depth of water was more difficult to examine. The effects of sand washing over the marl were very noticeable in this lake, probably on account of the increase of water depth. No marl was found in the immediate vicinity of the opening of any spring or creek into the lakes. In the bay or lagoon on the east side, made by the peninsula, water and marl were shallow enough to permit of sounding with the result of a gradual increase in depth of marl from shore to center. (Fig. 9.) Hard sand prevented a like test upon the opposite shore. The marl was in no case exposed close to shore excepting on the shore of the peninsula, which was very low and overgrown with trees and may have been but recently lake bottom, though upturned stumps showed no traces of marl. Upon the point itself there was a shallow coating of marl which increased but slowly at greater depths of water. The deep marl extended to a greater depth than 20 feet of sounding pipes and lay to the east of the peninsula where it joined mainland. Upon the west shore of the lake there was a coating of sand with no marl in or ftbout the outlet. There is marl in the deepest parts of the lake. Lake rice reeds in 6 feet of water were coated with marl deposit. No Characeae were visible. This was long known by the logging men, who in raising the weights let down to pull rafts, brought to the surface the gray mud. In general this lake as viewed seems to be a very marked case of the washing of sand from shores onto marl. The marl was undisturbed, was at the center and upon the peninsula where it thinned toward shore. The lake opposite Duck Lake, into which Duck Lake emptied, showed nothing but sand, shallow sand shores and seepage springs. § 15. Central Lake. Central Lake, also called the Intermediate Lake, being one of a series or chain, extends for some distance along the coast and RECORD OF FIELD WORK BY D. J. HALE. 143 very nearly at a level with Lake Michigan. The bold terraced shores and the sharply contoured hills that run down at right angles to the course of the lake are very striking to the eye. These contain a mixture of every kind of soil from pebble to fine clay. The lake itself can scarcely be called more than a river, though the ratio of fall is so slight that it has no perceptible current. Below the village of Central Lake sand is in some cases washed over the marl. But its presence underneath the lowland was shown in a startling manner. It was desired to raise the bed of the railroad which passed within 30 or 40 feet of the water’s edge, and to do this a heavy grading of sand and gravel was loaded into the low- land. One night this suddenly sank with the land supporting it about 20 feet. There is no doubt the semi-liquid marl beneath the lowland was forced by the greatly increased weight of the grading out into the lake, when the land above with railroad and all sank to solid bottom. The lake was examined mostly about midway and from there to the south end. The first series of soundings was made on the Fig. 10. Central Lake, Antrim County. See also Plate I. east side, out a hundred feet in shallow water, from a steam launch and from there in, and then along a slight creek about 200 feet in. The accompanying diagram (Fig. 10) will show relation of water, marl and land level. It will be seen that the large woods which had but a few inches of muck covering gave a very steady depth of marl. This marl while it contained no real sand or grit was very intensely granular and of a very brownish tinge. I should say that it had little organic matter, much iron and was decidedly differ- ent in grain from that usually met. The marl also showed well at points and beyond one of these to the south we examined the marl islands in the south lobe of the lake. These are very interesting as they are islands of solid marl nearly in a center line and about \ mile apart. 144 MARL. North Island* is very small, has upon it but few trees, is 30 or 40 feet long, 20 feet wide, and has its longest axis from east to west. Soundings were made first from the south, approaching from the median line of the lake to the small strip of dry land forming North Island. The following is a table of soundings on and about North Island (see Fig. 11 and Diagrams 11, 11 A and 11B of Plate I) : Fig. 11. Section across North Island, Central Lake. See also Plate I. No. of sounding. Depth of water. Depth of marl. Location. 1 5 5 50 ft. S. 2 2 14 40 ft. S. 3 2 14 30 ft. S. 4 6 in. 17 10 ft. S. 5 Dry land 19 On S. shore. 6 Water’s edge.. Dry land 20 North shore. 7 21 East end. 8 Dry land 2l W est end. 9 Water, 4 ft 15 25 ft. N. The regularity of increase and decrease in depth of marl, the steady variation in the relation of true bottom, marl, and water depths, is very striking and has been met with in but few other lakes. The island was about 40 feet long by 10 feet wide at the center and was oval shaped. A shallow of weeds extended north and south about 50 feet at right angles to the greatest length of the island, making an oval-shaped island, surrounded by an oval-shaped belt of shallows, the ovals being at right angles to each other. The island itself was barely above the water’s edge and was solid marl except for an inch or two of loam on top. Several trees grew on the dry ground which could be crossed easily on foot as the marl island seemed quite solid, which is not usual with marsh islands of this kind. Upon the shores, especially the north shore, the light shells had been sifted at the water line by the action of the waves, so that the *See Plate I, Diagram 11B. RECORD OF FIELD WORK BY D. J. BALE. 145 shore was a mass of shells of all sizes from a pin-head to an inch in diameter, and also intermixed with the shells were pebbly accre- tions something like those forming the coarse grained marl be- neath the woods to the north, before mentioned. Here it is very easy to see how shells could be broken into fine particles and merge with a bed of marl, losing their former identity. Several large clam shells were noticed lying just under the water’s surface, upon the marl. These crumbled when grasped between the fingers though they had once been very strong hard shells. It is very easy to see that if this is the condition of a large clam shell, Fig. 12 Section across South Island, Central Lake. the more delicate shells would be crushed by a much smaller pres- sure. South Island, — considerably larger and lying J mile south — was next examined. It was one or two hundred feet long, oval-shaped and with its axis north to south, or at right angles to that of North Island, but in a line with the axis of the shallows of North Island. From the south end of South Island, shallows run to the south end of the lake. These shallows, the deeper channels on each side and the approaches to South Island on all sides, are all solid marl of fair quality with no surfacing of muck or silt whatever. The soundings taken about South Island and on to the south end of the lake are as follows: Depth of water. Depth of marl. Location. 1 9 Close on S. E. shore S. Is. 1 19 10 ft. S. No. 1. Away from shore. 2 19 10 ft. S. E. No. 2. 2 22 70 ft. S. E. of 3. 5 ft. 20 Mid-channel S. E. of lake. 1 27 At extreme S. end of lake. 3 21 West side of island. Light muck 2 in. 21 Center of South Is. 19-Pt. Ill 146 MARL. By consulting this table and the figure (12) accompanying it, it will be seen that the outcropping of the island of marl is produced not by an added depth of marl, but by a rise in the true sand bot- tom of the lake. This rise in level is sharp but uniform, the depth of marl being greater if anything away from the island. It helps to solve no problem concerning the formation of marl excepting to show that the marl behaves like any thick layer either of chemical deposit or sediment. It lies like a thick sheet over projections such as this, making them less pronounced than they would be without the covering. Also notice that the soundings of South Island are deeper than those of North Island, not on the island, but immediately around it. The island is larger and judging from the size of the trees and thick- ness of the muck, has been first exposed by the slowly receding waters and has probably had somewhat more of the marl washed off of the higher parts on account of its longer exposure to wave action of the lake and leveling influence of water. The two series of sound- ings show a gradual increase of depth of marl from the north to the south end of the lake, where in tall pipe-stem reeds and one foot of water, the deepest sounding of marl (27 feet) in the lake was made. These two series as compared with that made in the woods and immediately west are deeper and show a whiter, more finely divided marl toward the south wher° in the deeper parts it loses almost entirely its granular character am* brownish trace of oxides of iron. The north end of the lake a mile and a half or more north of Central Lake was visited and a brief attempt made to examine the conditions there. They were strikingly different. While at the south end there was no sign of muck or any organic covering, here there was found everywhere 2 to 8 feet of very fine river silt. Be- neath this was 6 to 10 feet of marl, the deepest having a bluish tinge. The striking features of the lake were the granular appearance of a few beds, the gradual change in depth of marl and lack of sudden irregularities in bottom. Perhaps the whole can be traced to the slight fall of level of the lake, there never having been any current to disturb the original bottom of the marl deposited upon it. There is but 3 feet fall in eight miles in this chain of lakes. There are many good springs of hard watOr flowing into the lake. The samples, 8 to 11 inclusive, represent the waters of this region (p. 46). RECORD OF FIELD WORK BY D. J. HALE. 147 A mound spring examined, of which there are said to be several, is very peculiar, seeming to be formed by the water issuing from a side hill with a sloping clay bank, down which the water finds its way to pile up for itself a mound, and to boil from the top of this. The mound, four or five feet high by six broad, consisted to the depth sounded, about 10 feet, of a sandy bog iron ore mixed with clay below, making it withstand the seepage of water, and above with muck. The water itself carried up with it as a fine sediment a marly clay. (See Analysis 28, page 21.) This mound spring may be of interest perhaps as showing what a limy water will form upon being stopped and allowing to deposit upon issuing from a spring. No pure marl was found in the vicin- ity. A very interesting fact was its absence. The clays of the vicinity were next examined. The hills west of the town were in some cases strewed with glacial boulders and were more largely of sand and gravel than those on the east. On the east side was a brickyard that showed very nicely the thorough mixture that the clays of the vicinity had undergone. The clay as dug for use was somewhat moist and capable of being picked. A lump upon drying and examination showed a very fine grain, and was full of carbonates. On slicing a section of clay the different color of the fine layers gave it a highly streaked or stratified appearance, and these layers were rumpled and bent almost like the grain of curly maple, showing that the bed must have undergone great disturbance after being laid down. An inspection of the whole cut showed an upper layer of sand, a fine much rumpled layer of fine dark clay, then beneath this a fine bluish sand, hardly distinguishable from clay at first sight.* The whole hill looked as if it had been scraped together by some great power, and just before the mixture of layers became intimate, and they lost their identity, the movement ceased. This was the one of the lowest of a series of low hills, the highest of which was at an elevation of between 100 and 200 feet above the village. The clay hills above Mr. Crow’s farm were next visited. It was found that the clay anywhere near the level of the lake showed a strong heavy admixture of carbonates, but the shales higher up in the hills were freer from them. On the farm in question clay en- *The same contortion of the clay laminae may be noted at Clippert and Spauld- ing’s yard in Lansing (Part I, page 56), and at the location described by Dr. C. H. Gordon, in the Annual Report for 1902. It appears to be due to a readvance of the ice sheet, after the clay had been laid down just in front of it. L. 148 MABL. tirely free from carbonates was found, but mixed with shaly peb- bles, which were very heavy in iron and somewhat gritty. There is no doubt that in the hills about the town a genuine shale of fair quality for cement manufacture could be found. § 16. East Jordan and vicinity. The marl lying at the head of the south arm of Pine Lake* and about the mouth of the East Jordan, in the large valley once form- ing a continuation of the lake, was next examined. The bed was reached by steamer from Charlevoix. At Charlevoix, where the railroad crossed the outlet of the lake, marl was noticed nearly worn away by washing of the water and in most cases buried by sand, but is seen in streaks where it is exposed on the bottom. Its only significance is its presence in this part of the lake in a very thin layer. Along the shores of the south arm of the lake layers of marl of a few feet in thickness were seen cropping out under the banks washed away sharply by the action of the waves. The general appearance of the valley examined was very similar to that of Central Lake. Sharp hog-back ridges from high hills ran down somewhat parallel to each other and at right angles to the length of the valley, which is clearly the result of glacial action. A series of soundings were made to determine the manner in which the marl was deposited. So far as found the marl lay in the form of a basin showing 2 feet at the edges to 20 feet in the center, the center of the basin corresponding somewhat to the center of the valley. Upon the whole the marl lacked the granular appearance found at Central Lake, but was not of as uniformly great thickness and was covered with from 5 to 10 and 15 feet of muck and swamp growth and in most places with heavy forest growth or its remains. The disturbing action of a current of water was here noticed, for at the mouth of the river and at the head of the lake the marl was covered with many feet of silt and mixed more or less with sand. As a rule the whole of this bed was underlain with blue clay. A large area of land suitable for tillage, forming a rather dry table- land with the old deeper channel of the lake surrounding it, was covered with a light muck, 1 to 3 feet of marl and then very tough reddish or blue clay. In the above instance if marl has any value as fertilizer it should, upon the admixture of muck, marl and clay, produce splendid crops, as was already shown in one or two in- : See reference in Davis’ paper. RECOBD OF FIELD WORK BY D. J . HALE. 149 stances where the land was utilized. It is a significant fact worthy of notice for its bearing upon the origin of marl that these clays at or below the level of marl beds are nearly always heavily impreg- nated with carbonates. One sounding that showed well the con- dition upon the table-land was as follows: One or two inches of surface soil, 2 feet marl, 3 feet tough red clay, 15 feet black clay, water and gravel bottom. Upon the whole the marl of this section lay in a basin shaped depression nearest the water, but spread in a thin layer over nearly the whole valley. It is covered with forest or thick layers of muck in most instances and in others mixed with the debris of silt and sand brought down by the rapidly flowing river. It is, therefore, scarcely available for cement manufacture, but may some day be used to mix with and make more tillable the tough clays of the valley. The clays of this region, however, were of greatest interest. They were of two somewhat distinct types. Those before men- tioned were a fine-grained sediment laid down at or below the level of the marl. Black, blue and red were distinct colors noticed. They were all very tough and dense. In a drive taken 8 or 10 miles south and on the west side of the valley the clays of every color and condition from fresh sediment to a heavy shale in place were examined. Those examined rather low down the bluff nearer the valley,, always showed carbonates and more or less admixture of grit, prob- ably brought down by water. As we ascended in the cuts, clay in various stages of weathering from an almost compact shale to com- pletely disintegrated soil could be seen. The color also varied, being of a yellowish or greenish tinge. These were in a number of places quite free from carbonates, but the shales were always coarser grained and would, while being more compact, dig and grind hard. Finally an old mine shaft, where an attempt had been made to find coal, was visited. A heavy black, coarse-grained shale had been pierced by a shaft to a depth of 75 feet. The shale was nearly like rick and cropped out at the surface, breaking up and seaming where exposed to the weather. It reminded me very strongly of the Coldwmter shale visited in the southern part of the State.* In the seams the shale was reddened by oxidized iron and it was said *It is, however, the Devonian black shale. L. 150 MARL. that occasionally pockets of iron pyrites were found in it, although none could be seen at the time. A brickyard was then visited on the east side of town which con- tained strata of different colored clay and sand, much as at Central Lake, except that the level of the layers was not disturbed. All this clay, however, showed the presence of carbonates in very large quantity. § 17. Manistee Junction. Lakes about Manistee Junction were next visited. This part of the country shows very well the condition of the lakes and the out- line of country prevailing in a large part of the marl districts of the north part of the State, to-wit: an almost level pine plain in which suddenly occur drops below the level of the surrounding country without the slightest warning, much like a hole upon a level plain. A small lake was examined. This was circular in form and con- tained the deepest marl (20 feet) at upper end. This marl was bare of muck and covered only by water. Upon the west side a test was made of a shelf like those found in other lakes, these shelves prob- ably corresponding to the bare shelves found about Central Lake and East Jordan, which marked the recession of the ice. Near the shore were ten feet of marl and shallow water, while out 12 feet there were 12 feet of water and muck. At the lower end the con- ditions were the opposite of those at the upper. Here were found 26 feet of muck, beneath which were 4 feet of marl. Fine sand was the bottom in every case where soundings were made. The quality of marl was poor, being much mixed with organic material and sand. Notice the parallel case of Central Lake where silt and fine muck deposit shifted toward the outlet, marl being thinner. The water of this lake and others visited in this vicinity was tinged deep brownish red, lacking the remarkable clearness of most intensely marly lakes. Long Lake next visited was cut deeper into the surrounding country. The marl did not show upon the water’s margin which was of compact sand, but farther out, away from the shore, was a fair quality of marl and a good extent of shallows. Soundings were not made as no way could be devised to get upon the water. Calhoun Lake could not be reached in its deep parts where the marl, if any, was located. A marsh near here drained by a creek and practically dry showed good marl 15 feet thick, below one foot of muck. It was as usual in the form of a basin. No distinctive RECORD OF FIELD WORK BY D. J. HALE. 151 marks could be found to separate it from numerous marshes in which marl has not been found. Marl was also reported in hills between Reed City and Clare. This appeared to be a high clay country and rolling, quite distinct from the sand plains. § 18. Rice Lake. This lake is situated in Newaygo County, Town 11 North, Range 12 West. The greater part of the marl examined belongs to Messrs. John H. Kleinheksel, Henry Beers, and Dr. M. Veenboer. The above map was prepared by Prof. Kleinheksel, who accom- panied me during the survey of the bed. The lake abruptly breaks the level of the pine barrens and on account of the present condition of its bottom affords especial advantages for the study of its marl. From the appearance of the large lowland or marsh the water has not for some time occupied the whole depression indicated on the map by the double traced outer line “A,” and a later limit is well defined by the presence of a thick growth of scrub oak which encloses the area covered by the recent lake. This older water line is shown by the outer single line marked “B,” and the more recent one by the inner line “C.” Between these two shore lines, the old and the new, is found a con- siderable thickness or accumulation of marl ranging from two to twelve feet and lying under an overgrowth or accumulation of solid land some six or seven feet in thickness. This land is covered as before mentioned by a thick growth of scrub oak. By a system of large ditches the lake is still further drained till it has shrunk within the inside shore line to its present limits as marked on the map* (Fig. 13). With the above understanding, the first notable fact revealed by the numerous and carefully located soundings is that as far as could be discovered, the center of the marl basin is not exactly under the center of the water basin, nor the present lake. The center of the marl basin is very clearly shown to be in the northeast quarter of the southeast quarter of Section 10. Around this deep- est portion the gradually decreasing depths group themselves. The sounding of twenty feet in the above-mentioned section forbids an increase of depth toward the water as do the soundings of twelve feet and twenty-two feet in the quarter section adjoining it on the right. The soundings made up the center of the present lake still *The original U. S. Land-office Survey in 1838, was made in January, and no lake was recognized. The re-examination in 1854, sketched in a rather small lake. In hardly any two maps has the lake the same shape. 152 MAUL. further deny the presence in mid lake of any marl center. The fact is then established that in the case of this lake the greatest body of marl does not lie beneath the deepest portions of the lake. Furthermore the marl as sounded extends in its location toward the large north and east lobes of the lake as marked by the line of bluffs forming the original depression. The next point of interest is the rapidity with which the surface has overgrown and sealed up the present deposit. This deposit is covered by some three to five feet of tangled roots of marsh growth, forming a layer which jars for yards around with the weight of one’s tread. This growth, though light and easily pene- trated, is thick and must have nearly all formed in the few years BE COED OF FIELD WORK BY D. J. HALE. 153 since the lake has been drained. It leads us to beware how we judge of the age of a marl bed or the length of time since it has ceased depositing by the depth of its covering or surface. The material underlying the marl was in all cases found to be a fine lake or quartz sand, such as has been met with in the major- ity of lake soundings. One peculiar feature here was that just as the augur passed from the marl into the sand, it brought up a greenish layer which contained little sand, and was a grade be- tween organic matter and marl. This is the foundation upon which the marl started its growth, and should be of the utmost im- portance in the study of the method by which it is laid down. Having noted the surroundings of the marl, the final matter of consideration is the marl itself. The marl which was studied most closely in regard to quality was taken toward the center of the marl basin in deeper soundings. Though the examination care- fully covered two quarter sections, the quality of the marl through- out remained surprisingly uniform. From the accompanying anal- yses by Prof. Frank Kedzie of the Michigan Agricultural College, and those selected by Prof. John Kleinheksel and analyzed by Prof. Delos Fall of Albion, it will be seen that the marl is rather high in insoluble matter and low in carbonates. Of this insoluble mat- ter a small and constant part is quartz sand met with in many marl lakes and seemingly independent of the sand washed in by drainage. The organic matter though high is steadier than in most lakes, re- maining the same through all the deep soundings. The analyses by Prof. Kedzie, Nos. 1 and 2, are at surface and 35 feet deep respec- tively. The variation in organic matter and magnesia is slight. These analyses illustrate the fact that the deeper parts of the bed vary but slightly, probably owing to the distance from hills and surface washings of all kinds. Following are the results of analyses : Agricultural College, Nov. 25, 1899. No. 1 Insoluble matter 6.66 Oxides of iron and aluminum 1.34 Calcium oxide (equivalent to 71 .66$ Ca C0 3 ) .... 40.12 Magnesium oxide 1.10 Carbonic acid gas 32.50 Organic and undetermined 18.28 (Signed) FRANK S. KEDZIE. 20-Pt. Ill 154 MARL. Agricultural College, Nov. 14, 1899. Insoluble matter Oxides of iron and aluminum . . Calcium oxide (equivalent to carbonate) Magnesium oxide Carbonic acid gas Organic and undetermined No. 2 4.36 2.36 76.82$ calcium 43.01 97 15.05 100.00$ (Signed) FRANK S. KEDZIE. Albion, Mich., June 22, 1900. Prof. J. Kleinheksel, Holland, Mich.: No. 3 No. 4 Silica, SiO 2.84* 8.67f Alumina, A1 2 0 3 2.76 3.55 Iron oxide, Fe 2 0 3 none trace. Carbonate of lime, CaC0 3 79.55 65.67 Carbonate of magnesium, MgCO s none none. Sulphuric acid, as S0 3 3.15 2.50 Organic matter, etc 11.70 19.58 100.00 99.97 DELOS FALL. § 19. St. Joseph River and tributaries. In and about the mouth of the St. Joseph River there are beds of marl. Very small creeks have in the meadows surrounding them, small beds 1 to 3 feet thick of marl. Hickory Creek and Paw Paw River, which has a large marsh near its outlet, have marl along their course. § 20. Onekama. Portage Lake (Fig. 14), on which is situated the Town of Onek- ama, is about eight miles north of Manistee and opens by a short but very navigable channel into Lake Michigan. It is surrounded by high hills on all sides and on account of the deep depression made by the lake the springs which issue from beneath the hills are numerous and large. One spring contained a considerable per- centage of sodium carbonate, but the marl in the immediate vicin- *This is. a marl containing over 5# of clay and running rather low in carbonate of lime. After the organic matter is excluded the percentage of carbonate of lime amounts to 90.09. tThis is a sandy marl. Excluding the organic matter there is 81.65# of carbon- ate of lime. RECORD OF FIELD WORK BY D. J. HALE. 155 ity in which the spring emptied showed no unusual trace of alkaline salts. This is only another illustration of the fact that the agency at work in the deposit of the marl has a power of discrimination, refusing certain salts from the water and depositing others. Such was shown to be the case of the sulphur and iron springs mentioned in the description of Big Whitefish Lake. Portage Lake is fed entirely by a network of springs and spring creeks and the water is very clear. The shallows of the lake are not all marl. Strips of marl from three to four feet in thickness alternate with sandy beach around Fig. 14. Portage Lake, Manistee County. the whole lake, but the deepest marl and that which engaged our attention lay to the north and northwest shores toward Lake Mich- igan. It lay in a lobe of the smaller lake forming its northwest corner and reaching beyond the water up to the low sand hills bord- ering on the Great Lake. It is only thought necessary to mention certain features which will illustrate the general ideas already gath- ered as to the location of marls. As before stated the deeper marl was confined to the large lobe or lagoon of the lake. Lines of soundings were run in different directions and the following conclusions reached: The marl de- creased in depth with increase in depth of water toward the center of the lake. There was also an increase of content of organic mat- 156 MARL. ter as the center of the lake was approached. In the lobe above mentioned, considered as a whole, the marl deepened rather evenly toward the center of the whole basin. Here the marl was 21 to 22 feet thick with little or no surface covering. As the parallel lines of soundings approached the borders of the basin the depth of marl decreased rather evenly to 19, 17, 15 and 13 feet. The last named depth remained nearly constant as far as followed through the thick undergrowth to the northwest toward Lake Michigan. The marl on the whole was little contaminated with foreign matter such as sand and clay. There was a very small content of the fine quartz sand often found in deposits. The ditches and small creeks emptying into the lake over the bed had carried down sand which had mixed with the marl in their immediate vicinity. The marl was deposited upon a fine pepper and salt sand which formed the lake bottom. In the marl basin or lagoon before mentioned soundings revealed a peculiar condition. The bed contained a small layer of intervening organic matter alternating with the marl. The material was well preserved and seemed to consist of driftwood and marsh growth pressed firmly into a layer a foot or so in thickness. The layer was about fifteen feet below the sur- face of a firm pure marl deposit.* Its presence might indicate that the marl had ceased to deposit for a time, and with the return of favorable conditions had again deposited, burying the layer of drift and marsh growth which had accumulated. A large part of the lobe examined was not under water at the time. A part of it had recently been covered by water before a new outlet into Lake Michigan had been dredged for the lake. When it was drained the surface of the marl had been left dry. This left the marl more or less dense and dry and as a consequence there was nothing but a slight growth of grass and the consequent “surface” was only a few inches to a foot in thickness. This was a great contrast to Rice Lake which had been drained about the same length of time, but was left very wet. The marsh growth had be- come luxuriant and the “surface” is from three to five feet of marsh growth. Beyond the dry portions of Portage Lake and forming the fringe of real marsh was the portion which had always remained wet. Here the growth of soil and roots reached five feet or more. From these comparisons it can readily be seen that it is impossible *This may indicate a lower level for Lake Michigan at one time. L. RECORD OF FIELD WORK BY D. J. HALE. 157 to judge tlie age of a marl bed from the depth of surface growth covering it. Sixty-nine soundings were made varying from 13 to 22 feet. The bed covered about 125 acres, not including a large area along the shores containing marl from seven to ten feet. The marl is of fair quality and its variation in quality is shown by the following analyses: No. 1. Si0 2 2.81 A1 2 0 3 and Fe 0 0, 65 CaO as CaCO,“ (47:89) 85.63 MgO as MgC0 3 (1.47) 3.08 Phosphorus 014 Total organic matter 6.96$ No. 2. Silica 3.64 Oxides of iron and aluminum 1.35 Calcium oxide 45.37 Magnesium oxide . 63 Carbon dioxide 35.86 Organic and undetermined 11.85 Submitted by A. W. Farr. Samples No. 3 and No. 16, or Nos. 1 and 2 above, were taken at the respective depths of three and sixteen feet by the owner, Mr. Farr, and the latter was evidently mixed with sand as a careful examination of the whole basin showed no such amounts of sand, the sand being in all cases, excepting in the presence of flowing water, fine quartz sand and in small quantity. The remaining samples show a fair marl with no harmful compounds in propor- tions too large for manufacture. CHAPTER VII. MANUFACTURE OF PORTLAND CEMENT FROM MARL. § 1. Introduction. The purpose of this chapter is not to give a full technical descrip- tion of the process of cement manufacture. This may be found in any one of a number of large volumes written upon the subject.* In a later chapter, prospeeti from various cement plants in the State are cited and will furnish farther information. In order to connect these details and to give compact description of the process and to emphasize important points, this chapter is written. § 2. Definition of terms. The name “Portland” was derived from the name of a popular building stone used in England at the time that our present cement was given its name. The cement was thought by some to some- what resemble this natural rock, hence was named after it. Portland cement is an artificial mixture of calcareous matter with silicious (generally clayey) matter, which is properly pro- portioned and burned to a point just short of vitrification or melting. The resultant slag will, upon grinding, set with the addition of water to form a cement. Natural cement differs only from Portland cement in that nature has mixed the calcareous and argillaceous ingredients in nearly the proper relations. Slurry is the properly ground and mixed clay and marl or lime- stone suspended in enough water so that the mixture can be pumped from one reservoir to another. *See also 25th Annual Report of the State Geologist of Indiana; 22nd Ann. Proc. Ohio Soc. of Surveyors and Civil Eng., p. 18; 21st Proc. Indiana Eng. Soc., several papers; American Engineering Practice in the Construction of Portland Cement Plants, by B. B. Lathbury, 1902; A Rotary Cement Kiln for use in the Laboratory, by E. D. Campbell; Jour. Am. Chem. Soc., March, 1902; various papers, especially those by the Newberries in the Cement and Engineering News, and other pamphlets issued by the same press. Beside Lathbury and Spackman, Robert W. Hunt & Co., of Chicago. The Osborn Co., of Cleveland, and Hassan, Tagge and Dean of Detroit, may he mentioned as de-ipners of cement nlants. See •Us rep >rr bv Prof. I. C. Russell in the 21st Annual Report of the Director of the U. S. Geological Survey. MANUFACTURE OF PORTLAND CEMENT FROM MARL. 159 The gradual perfection of Portland cement to-day is owing to the application of raw material and high grade machinery to the development of one chemical principle which is and will remain at the foundation of the cement industry, that two elements or groups, lime on the one hand, and silica or alumina on the other, when properly proportioned and intensely heated, have the power to combine with each other and then later with water such strength that after the combination once occurs, fire, water, acid or salts, have little power to disturb them or weaken their hold upon each other. The first group is calcareous. We see it purest in lime or calcium oxide, in an amorphous rock as calcium carbon- ate of limestone, as calcium carbonate in chalk, and in the purest state as crystallized rock or marble. The second group is silicious. This forms a large part of the earth and is found purest in quartz sand and fire clay. When these two groups, the calcium carbonate of the marl on the one hand, and the silica and alumina on the other, are finely ground and mixed and subjected to a temperature of between 2,000 and 3,000 degrees Fahrenheit, the calcium car- bonate loses its carbon dioxide becoming calcium oxide and the silica becomes soluble. When the slag is ground to powder and mixed with water the nascent compounds recombine to form a tricalcic silicate and aluminate, an insoluble, non-combustible rock which becomes harder if anything, with age. § 3. Historical. Before the discovery of the principles which govern the setting of cement, the Romans and they who followed them used slaked lime and a volcanic dust called pozzuolana. It contained the above mentioned substances in the right proportions to form a fair cement. About the year 1756, Smeaton, an English engineer, made experiments to find a mortar which could be used under water in the construction of the Eddystone lighthouse. About the year 1818 Pasley in England and Vicat in France began experimenting upon cement materials to ascertain the proportions necessary to produce an artificial cement. “In 1824 Joseph Aspedin, a bricklayer of Leeds, discovered and patented a method of making a hydraulic cement and named it ‘Portland,’ from its fancied resemblance in color and texture to 160 MAIiL. the oolitic limestone of the Island of Portland, well known and in great favor in England as a building stone.” § 4. Materials for Cement. It will not be of profit to follow the development of natural cements. They are produced by grinding and burning a rock hav- ing cement materials in nearly the right proportions mixed to some extent by nature. As the proportions always vary the product is more or less unstable, is liable to crack and warp more than Portland cement and never stands as high tensile tests. It is therefore untrustworthy in those portions of great engineering works where great soundness and durability are required. It is produced more cheaply than the Portland cement and is very satisfactory for low grades of work. It is manufactured in large quantities in the United States and supplements largely the use of the more costly Portland cement. As the State of Michigan is supplied with extensive deposits of marl and clay suitable to the production of the highest grade of artificial or Portland cement it is well to notice the different ways in which material and machinery are manipulated to produce the same result. It is a common idea that but one or two materials may be used for the production of cement, but this is erroneous. Anywhere where lime and clay constituents can be found sufficiently near each other and pure enough, Portland cement may be manufac- tured. In England, much of the cement is made from chalk, which is a calcareous formation similar in composition to marl, but dry, and a dense blue clay. In the United States chalk is used so far only at Yankton, S. D. Limestone and clay are the favorite materials in most parts of the world and it is only within the past few years that marl and clay as raw materials have come into any great prominence. Several methods of burning and mixing the raw materials are used in Europe and the United States, adapting themselves some- what to the nature of the raw materials used. In England, in the Medway district on the Thames, marl and clay are ground and mixed with 120 per cent to 140 per cent of water. The finer particles are then flushed off by water passed into “settle backs,” where the mud settles and the clear water is drawn off. When this sediment has dried to the consistency of a paste, it is gathered up and deposited on floors where it is dried still more rapidly by waste heat from the kilns. It is then mixed with Geological Survey of Michigan. general exterior view of an eleven kiln plant. MANUFACTUBE OF POBTLAND CEMENT FBOM MABL. 161 charcoal and burned in kilns. The remarkable feature about this method is the thorough manner of mixing. It is the most thorough method known as the particles of clay and calcium carbonate are suspended together in water and allowed to settle somewhat as in a natural sedimentary deposit. It requires a month to dry the material, and the method is therefore too costly in time and is giving way to more rapid methods. Mixing by the semi-wet process is probably most widely used throughout the world. When limestone and clay are used they are mixed with about 30 per cent or 40 per cent water, by means of sludge mills or similar contrivances. The mixture is then ground, passed to a drying floor, subjected to waste heat from kilns and burned as before. The mechanical mixture of the particles is not as perfect as in the first method but the drying occupies but 20 hours. Sometimes the materials are mixed nearly dry and formed into bricks which are burned as before, with coke in a kiln. This is done in the dome kiln or dry method, which has been used to some extent in our own State at Union City and Kalamazoo, and will be described later. The method of burning differs somewhat. Where a dome kiln is used the layers of mixed cement material alternate with layers of charcoal. In the continuous kiln, charcoal and unburned and partly dried cement materials are fed in at the same time at the top and the whole ignited at the bottom. The portion of heat not used in burning the cement at the bottom escapes upward and helps to raise the temperature of the half wet material above. In this way much more heat is said to be saved than in the dome and rotary kiln process. Cement could be manufactured using sand to furnish the silicious elements instead of clay. Briquettes of cement made in this manner seem to have stood very good tests. Yet in practice, sand in any form is dreaded in cement manufacture, from the fact that it is so expensive to grind it to a sufficient degree of fineness for the purpose. The clay* is preferred instead because it is divided much more finely than sand, being already ground on account of the breaking down processes of nature. It can be readily seen that the materials used and the processes relied upon vary widely in different districts although the finished *What is commonly known as clay is often very largely only extremely fine particles of quartz, mineralogically the same as common sand. L. 21-Pt. Ill 162 3IABL. product must be almost the same in all cases, as Portland cement has a narrow range of standard composition, which must be approximated in all methods of manufacture. The process used in Michigan depends mostly upon the materials at hand. The silic- ious element used is either a surface sedimentary, or a shale clay, depending upon which one having the best composition is at hand. The method of burning in nearly all cases is the rotary kiln pro- cess. There are few lakes or marshes that can be sufficiently drained so that the marl can be treated by the dry or semi-wet process and for this reason a more detailed description of the rotary or wet process will be given. From the foregoing it must be clearly understood that the factories of Michigan have not only to compete with those using their own process, but also with the remainder of the manufacturies by the limestone process, which alone furnishes more than half the cement produced in the United States. It must always be borne in mind that 40 to 60 per cent of the marl is water and nearly a half of the remainder carbon dioxide, a gas which is driven off in burning. The cost of handling and drying this great bulk of material must never exceed the cost of quarrying and grinding the limestone. When this happens, Michigan factories will be undersold by those of the limestone district. Besides this competitor there is the natural cement. This will take the place of Portland cement in many cases where the price of the better cement rises too high. § 5. Kiln process of cement manufacture. The two methods so far employed in this State are the dome kiln and the rotary process. The former of these two processes is fast going out of use in this part of the country as it does not seem to fit the materials used as well as the rotary process. In 1872 a plant of this kind, the first cement plant in Michigan, was started at Kalamazoo. The marl beds which were used are described in Chapter VI. Another plant of this kind is erected at South Bend, Indiana, but was not in operation when visited. The process which was employed at Union City, Michigan (See Plate III), may be briefly described, as follows: The marl was scooped up w r et from the marsh and is thoroughly mixed with dry clay. The mixture, now of a doughy consistency, is pressed through a square orifice and is cut about the form and size of building bricks. These marl clay bricks are laid on flooring MANUFACTURE OF PORTLAND CEMENT FROM MARL. 163 of T rails to thoroughly dry, when they are then ready for the burning, which is accomplished by kilns. A kiln resembles a mammoth hollow cigar, cut off at both ends. It is built of fire-brick, and is about forty feet high by eight to ten feet in diameter. Beneath is a fireplace of about five feet to furnish a thorough circulation of air. At the base of the kiln, above the open air space, an arched layer of these bricks is packed, a layer of lumps of charcoal, then another layer of bricks till the kiln is one-third or one-half full. The mass is then fired, and burns for about forty-eight hours; the bricks fuse into lumps of heavy, black slag, perforated by the exit of the carbon dioxide, which is expelled by the fierce heat. The whole mass shrinks and collapses and cools, and is then raked out. Then the slag is sorted by hand into two grades of cement, and is ground by mills into fine, dark brown powder which we know as Portland cement. This is an extensive process, requiring large buildings for drying, many kilns for burning ( for a kiln burns only seventy-five barrels at a time ) and many men to transfer and sort. This process is not as exact as it should be. Part of the bricks are overburned and part underburned and must be sorted by hand, requiring great expense in time and labor. It has been displaced at Union City by the rotary process, and all the new factories in the State are employing the latter. § 6. The rotary process. The rotary process, in order to be successful, should be carried on upon a large scale. The buildings which protect such a factory generally cover several acres (Plate IV). The prevalence of dis- astrous fires which have wiped out several large factories in the past year, causing great delay in the work as well as the financial loss, should emphasize the construction of durable and fire-proof buildings. The latest are being built largely of steel and cement. The machinery is so grouped (Plate V), that the raw material is transferred by machinery from one step of the process to the next, till it enters the storing bin a finished cement. The following is a brief description of the whole process, as seen at Bronson, Michigan. The marl is scooped up by an ordinary dipper dredge and is drawn a few hundred feet to the factory on small dump cars, where 164 MARL. it is stirred and screened and then pumped into a large funnel to measure it. Meanwhile, the clay, which is mined several miles away and drawn to the works by rail, is elevated to the second story by machinery, is weighed by the wheelbarrowful and dumped into a hopper which drops it to a cluster of revolving millstones, which at the same time receives the semi-liquid contents of the huge marl funnel. When both have been ground and mixed with each other, this mixture drops into a second reservoir, where it is thoroughly stirred and mixed for some time by revolving paddles. From this reservoir the mixture, now termed slurry, is pumped into huge tanks, where it awaits the burning process. Grouped with each tank is a huge cylinder about 40 feet long and four or five feet in diameter. The cylinder lies with the end that is farthest from the tank a little beloAV the horizontal. The end opposite the tank is closed by a cupola. The cold, wet slurry flows in at the tank end, the whole cylinder revolves, and the liquid mixture, caught on its inner surface, runs slowly towards the cupola at the further end. Here a falling stream of crude petroleum is ignited and blown by air blhsts into the end of the cylinder. The solid sheet of flame penetrates six or seven feet, being under control. The slurry, slowly approach- ing, is first dried, then heated, and by the time it reaches the end of the cylinder is fused into liquid nodules about the size of peb- bles, and falls through a slit at the base of the cylinder. Here it is received by an endless chain of wheeled trays, and, having cooled, is borne to the mills. These mills are very efficient, grind- ing it to a powder, ninety-nine per cent of which will pass through a sieve with 10,000 meshes to the square inch. It is then finished cement and is stored in bins. In this process, as in the kiln pro- cess, the fine powdering and mixing of the crude material is care- fully accomplished, and, by burning, the carbon dioxide is driven off and the mass thoroughly fused. This process is almost entirely accomplished by machinery. The machinery is expensive, but only requires the labor of fifty men to run the whole plant. It is economical, as the burning is performed with exactness, and there is no charcoal to fuse with, and impair the strength of, the cement, nor is any hand sorting necessary. In the latter case the quality of the cement and cheapness of manufacture is unrivaled. MANUFACTURE OF PORTLAND CEMENT FROM MARL. 165 § 7. Preliminaries. 1. Digging. The raw material, marl, is in nearly all cases found in a lake or a swamp or both. In this condition it may be covered by a few inches to several feet of water, may be bare, or be covered with a surface of grass or rushes which must be “stripped” before the marl can be dug. In many parts of the State there are extensive marshes covered with a growth of timber which must be cleared and grubbed before the marl becomes available. In such cases it is noticed that nearly all the roots follow the moist surface of upper soil which has been deposited on the marl, and do not penetrate deeply. This renders clearing much less expensive- and the clearing can nearly all be done by burning. It is best in selecting sites for factories to avoid as much as possible the marl lands covered with a thick surface of swamp growth or forest. There is much marl land available in the State that does not require expensive sur- facing, which should be chosen in preference to the less available territory. The content of moisture will often vary much according to the position of the marl below or above the water line of the lake or marsh. In the same lake basin there may be marl in mid-lake containing 60 per cent to 75 per cent moisture, and at the same time marl on higher marsh land at the sides, which will not contain over 25 per cent. The expense of surfacing is of course somewhat governed by the thickness of the bed and the depth to which it may be dug or dredged. A bed ten feet thick will be much more wasteful in digging than one thirty feet thick, for in each case the surface soil or growth is mixed with the marl in dredging, and must be burned out in the rotaries, involving cost of fuel in drying and time in handling the surface material, which is in the end burned, forming only an ash. Not only is there expense in handling and burning the material that becomes mixed with the marl from the surface, but also there must be a certain margin or remainder of the marl at the bottom of the bed which cannot be dredged on account of admixture with sand or unsuitable clay, of which the true bottom may be composed. It can then be easily seen that there is a greater proportion of the marl in a thick bed, available for use, than there is in a thin deposit, for the waste must be the same in both cases. With the present large supply of deep beds 166 MAUL. the shallow deposits will not be immediately used. If there are but a few inches of grass and loam above the marl, no appreciable cost will be incurred, excepting to increase the organic content of the upper scoopings of the dredge. If there are several feet of dense marsh growth, sometimes as high as six, it may cost $75 an acre for surfacing, — quite a handicap. 2. Draining. In many lakes it is found expedient to drain by a short channel and thereby lower the water level, bringing the deeper parts of the lake within working depth of the dredge. Not every lake is located so as to be easily drained. Also it will be found, if attempt is made to so utilize marl that has laid at any great depth under water, that the quality of such marl will be much poorer, being higher in organic matter and lower in the essential calcium carbonate. When the lake or marsh has a level of several feet above the stream or lake which empties it, it may be possible to drain it so that the semi-wet or even dry method of mixing may be used. This was to have been done at Watervale and was contemplated by the Hecla Cement Co. 3. Dredging. On account of the semi-fluid condition of most of the marl of the State, and its location partly in or adjacent to water, the easiest method of digging the marl has been by the ordinary steam “dipper” dredge. This is a barge or scow floating on the water and operating a large scoop or dipper, which can work to a depth of about twenty-two feet, as was claimed at Bron- son. The rubbish or surface growth of a marsh is piled to one side, and the dredge makes a channel for itself as it digs the marl. It can be seen that this method is best adapted to the greater part of the marl in Michigan, which lies either under water in shallows or flats or in a marsh which is at or near water level. Another proposed machine may come into general use. It is also built on a scow and consists of a movable crane carrying an end- less chain of buckets. This chain can be lowered to greater depths by the crane and will perhaps be able to dredge to a depth of thirty feet, though, as the quality of marl decreases and the expense of power in digging will rapidly increase with great depths, it will not be found economical to dredge to the bottom of deep beds. In many factories in the State the marl is dredged and then piped to the* factory. MANUFACTURE OF PORTLAND CEMENT FROM MARL. 167 There is one more method of transporting the marl. This is by digging or dredging and then pumping to the factory.* When the marl is pumped it must be mixed with slightly more water, which must in turn, be dried out in burning in the rotary. The increased expenditure for fuel will likely offset the cheaper trans- portation. The pumping plan is only considered where the marl is adjacent to the factory. Where the marl is several miles away,f a railroad must, of course, be employed, as hauling by wagon is entirely out of the question as being too expensive. Where it is near, an overhead trolley bearing cars or a narrow guage road, in which steam or horses are the motive power, can be used. Clay must be quarried or dug according as it is a shale or a clay. For quarrying see the account of Bronson in Part I, and elsewhere in this report. Out of 14 factories where the raw material could be located, all but one had the marl deposit within two or three miles of, or directly on the site of the factory, while but four had their clay near the factory, most of them getting it long distances away, in some cases in Ohio$ or Indiana. The estimated cost of putting the material at the factory therefore, varied from eight cents to seventeen and one-fifth cents per barrel of finished cement, being greatest in the case of the Hecla works, who were to transfer their marl about thirty miles from bed to factory site. § 8. Estimates on raw material. One factory in the State was running about 2,000 pounds of marl to the cubic yard, while it was said to require one and one- half cubic yards of liquid marl to a barrel of cement. Now a barrel of Portland cement weighs 380 pounds. An aver- age of 65 $ of this is calcium oxide; 65$ of 380 equals 247 pounds of calcium oxide required for a barrel of cement. Taking a wet marl, which has 40$ moisture, and 90.3$ calcium carbonate in the dry residue the available calcium oxide would figure as follows: 100$ less 40$ equals 60$ dry matter. 90.3$ of 60 equals 54.18$ calcium carbonate. 100$ calcium carbonate less 44$ carbon dioxide equals 56$ cal- cium oxide. 56$ of 54.18 equals 30.3$ of original weight as available calcium oxide. *As at the Woodstock and other plants. fAs is the case in the Hecla Plant at Bay City, 30 miles from the bed. jMillbury. 168 MABL. At the above factory estimate of raw material 1^ cubic yards of marl would weigh: 1^ times 2,000 equals 3,000 pounds. 30.3$ of 3,000 equals 909 pounds available calcium oxide, whereas it really furnishes but the 247 pounds necessary for the barrel of cement. This means that the deposit which was worked must have had a higher content of organic matter and moisture than we have assumed. Notice the effect of increased per cent of moisture and decreased percentage of calcium carbonate on the percentage of available calcium oxide. Take for instance a marl 60$ moisture and 75$ calcium carbonate. 100$ less 60$ equals 40$ dry matter. 75$ of 40 equals 30 of original weight as calcium carbonate. 56$ of 30 equals 16.8$ of original weight as available calcium oxide. In this case but 16$ of the original weight of the marl as dredged and transported to the factory, contributes to the active elements of the cement. It can thus be seen that the actual supply of raw material is greater or less per acre, according to the condition in which it may be found. No exact volume of marl to the barrel of cement can then be given, as it varies in each bed, but for a high grade marl of medium moisture, probably 10 cubic feet to the barrel would be an average. It is estimated in the Clare bed with 94$ to 96$ calcium carbonate, and 50$ to 70$ moisture to be from 7.5 to 12.5 cubic feet. At Zukey Lake, the Standard Portland Cement Company, with calcium car- bonate 93.92$, estimates 9 cubic feet of marl to the barrel of cement. The clay is much more compact and free from moisture. The volume of clay of the Omega Cement Company required for one barrel of cement was estimated to be 1.12 to 2.12 cubic feet, accord- ing to the per cent of calcium carbonate contained in the clay. The question as to the requisite acreage of marl is discussed by several cement plants, and the estimated acreage varies widely, being from 262 acres to 2,000 acres. The favorite plan is to show an acreage which will run a factory of the desired size for 100 years. Several factories have been projected upon 75 or 100 acres, but have evidently given up from lack of material. Geological Survey of Michigan. Vol. viii Part III Piate V. GENERAL PLAN OF FOUR KILN PLANT WITH PLACE FOR EXPANSION MANUFACTURE OF PORTLAND CEMENT FROM MARL. 169 § 9. Requisites for marl deposit. Taking the consensus of opinion as laid down in the prospec- tuses of the different factories built or building in the State, and the relative merits of beds as viewed in various parts of the State, the requisites of marl are as follows: Surfacing. There should be little or no surfacing and the water covering the marl should be as shallow as possible, not over six or eight feet. The amount of raw material in the State does not necessi- tate the use of beds covered with any great depth of muck or other useless matter which requires surfacing. The marl must be located on or near railroads, but better than all, on the Great Lakes. See freight rates, under shipping. Necessary composition. The prospectuses so far examined do not give any analyses of marl lower than 90$ calcium carbonate. They vary all the way from this to 96$. It is doubtful in some cases, whether this is the highest sample found, or the average of samples in the bed. One prospectus which gave a sample analysis in its prospectus of 95.73$ calcium carbonate gave in two samples taken and analyzed by two reliable chemists, when its bed was sampled as fairly as possible, 83.04 and 77.05$ calcium carbonate respectively. In the majority of beds the marl varies with the depth, and when it is 90$ CaC0 3 near the surface it is likely at 20 or 30 feet to be only 75 or 80$ calcium carbonate, as explained in previous chapters. It is very safe to say that if an average of all samples taken, whether deep or shallow, and irrespective of the choicest location, reaches 90$, the bed is safe as regards calcium carbonate. This will imply unless the bed is exceptional, that many samples will run as high as 95$. Depth. The depth of marl used or counted upon in the State varies from as low as 15 feet to depths which no scow of the present kind in use could possibly reach. It is fair to say that marl seems to be used anywhere from 15 to 25 feet below water level, with the re- strictions as to water mentioned in the paragraph on surfacing. Low calcium carbonate means high organic matter, which is un- desirable from the greater bulk of useless matter transferred to the factory to be burned. The dangerous constituents are sulphuric acid and magnesia. 22-Pt. Ill 170 MARL. Sulphuric acid. This does not appear to be troublesome according to the analyses seen in the various prospectuses, being given from .08$ to .58$. It could go considerably above this, depending upon the amount in the clay. It is not often very troublesome in pure marls, but should be watched. Magnesia. This is very much more troublesome, as a strain of magnesian clay in the marl may cause it to vary dangerously. The cement prospectuses giving analyses, show from 1.41$ to 1.79$ magnesium oxide, which is a very safe limit. Grain. Some of the marls of our State are very fine and rival the finest grinding of any material by machinery. One case was noted where there was but 4$ left on a 200x200 sieve, or 40,000 meshes to the square inch. This is certainly wherein marl excels all other raw materials for cement manufacture. It need hardly be said that an excess of shells or pebbly accretions somewhat increase the power necessary to grind finely and are a drawback. A marl with above 3$ or 4$ coarse or fine sand, must be ruled out. Effects will be noticed further on. For analyses of marls for factory purposes, see p. 32 and the descriptions of different plants. § 10. Clay. We have in this State two kinds of clays, one being shale, which is often very hard to grind, but is steady in composition, and generally most free from carbonates. The other class are not of the nature of rock, but have been more recently laid down by the action of water and are not compressed. The grains are more easily separated, and grinding is effected with less cost in power. A good cement clay analysis is that of Millbury, O., being the average of 50, as given by J. G. Dean; Si0 2 61.06, A1 2 0 3 18.10, Fe 2 0 3 6.65, CaO 1.25, MgO .53, S0 3 1.05, organic matter and water 9.20.* The principal points about clays are the relations of silica and alumina and the proportions of lime, magnesia and sulphuric acid. If there is much lime the clay will not go as far with the same amount of marl. Hence, if it is to be carried by railroad any distance, there is the resulting dis- advantage of increased cost of transportation. Organic matter and moisture are of course a dead weight. The above clay is a *See also Prof. Fall’s paper. Geological Survey of Michigan. Vol. VIII Part III Plate VI. GENERAL INTERIOR VIEW OF SLURRY DEPARTMENT. MANUFACTURE OF PORTLAND CEMENT FROM MARL. 171 fairly good sample of surface clays used for cement manufacture, the same bed from which this was taken, being used by two fac- tories in this siate. A surface clay, if of the right composition, is much better because easier to dig and grind than shales. Often in the neighborhood of shale outcrops there is found a good surface clay, which is the broken down and decomposed shale, and makes a very suitable clay. The great body of Michigan clays are too high in magnesia and in alumina in proportion to silica.* An average result from six factories giving their clay analyses was the following analysis: Silica, 59.90. Alumina, 22.76. Magnesia, 1.47. Sulphuric acid, 1.04 (but two out of six stated). § 11. Admixture of raw materials. This of course depends upon the exact amount of moisture and the percentage of calcium oxide in the marl, on the one hand, and the percentage of silica, iron and alumina in the clay, on the other. It can never be correctly determined without a careful analysis of both raw materials. A good clay is less variable than the marl. At Bronson, it was said that the clay was analyzed once a week and the marl was analyzed every day. The slurry is analyzed frequently to see if it continues in the right proportions, showing at once whether the measurement of the raw materials is carried on exactly and whether the raw material is varying much from the last analysis. If it does, one raw material or the other must be added to preserve the correct balance for the production of a cement of even composition. Lathbury and Spackman, who write the article on cement mak- ing given below, say in their magnificent triglot on American Engineering Practice in the constructing of Rotary Portland Cement Plants :f “A glance at the analyses of the standard brands of cements, both American and Foreign, will show a great uniformity, and it can be stated that in a good cement, the amount of the different ingredients will only vary within very narrow limits, as shown in the accompanying table. *See Part I of this volume, i. e., Hies’ report on shales and clays of Michigan, and the analyses of shales in the descriptions of various plants. fPublished by G. M. S. Armstrong, Harrison Building, Philadelphia. 172 MARL. Silica Minimum. 19* Maximum. 260 Alumina 4 10 Iron 2 5 Lime 58 67 Magnesia 0 5 Sulphuric Acid . . . 0 2.5 Alkalies 0 2.8 Le Chatelier, after long study of the composition of cements, concluded that the two important compounds existing in the clinker were a tri-calcic silicate (3CaO . Si0 2 ), and a tri-calcic aluminate (3Ca0.Al 2 0 3 ). The hardened cement consists of hexa- gonal plates of calcium hydrate Ca(OH) 2 imbedded in a white mass of interlaced crystals of hydrated calcium mono-silicate (CaO. Si0 2 2f HoO). The chief reaction which takes place during the setting of cement, according to Le Chatelier may, therefore, be represented as follows: 3 Ca0.Si0 2 +xH 2 0=Ca(0H) 2 +Ca0.Si0 2 .2y 2 H 2 0. Assuming that three equivalents of lime and no more can enter into the combination with silica and alumina in a cement, then as- suming magnesia to act the same as lime, the proportion of lime should not be less than that required by the formula ^ CaO+MgO or greater than ' — S 1 O 0 — AL O a CaO-)-MgO Si0 2 -f A1 2 0 3 -Fe 2 0 3 “The Messrs. Newberry in a series of researches as to the con- stitution of cement, determined by synthesis: “First, that lime can combine with silica in the proportion of three molecules of lime to one of silica (3Ca0.Si0 2 ) and give a product of practically constant volume and good hardening properties. With more than this proportion of lime the product is not sound. “Second, that lime can combine with alumina in the proportions of two molecules of lime to one of alumina (2CaO . A1 2 0 3 ) giving a product which sets quickly, but shows constant volume and good hardening properties. With more than two molecules of lime the product is not sound. Thus Newberry gives as the formula for a cement with the maximum amount of lime, x(3CaO . Si0 2 ) +y(2CaO. A1 2 0 3 ) x and y being variable factors, dependent on the relative proportions of the silica and alumina in the clay. MANUFACTURE OF PORTLAND CEMENT FROM MARL. 173 “In practice, cements contain a slightly less quantity of lime than the above formula requires, because of the difficulty of secur- ing perfect mixing and burning and the danger of over liming if the formula is exceeded.” § 12. Mixing and raw grinding.* The marl is dumped into a large tank or vat and is generally screened to relieve it of gross organic and foreign matter, useless to the process. As before mentioned it may arrive at the factory in little dump cars, by means of an overhead trolley or cable work- ing from factory to bed, by horse or mule power, by scow towed in the lake, or by pumping from the dredge where it is scooped up directly to the factory by pipe. In all but the last method the marl becomes somewhat dried during transportation. The marl may be pumped into a large hopper and estimated by volume, while the clay is weighed directly, the right weight of it being added to each hopper of marl, when the two are then mixed and ground together. Sometimes the clay and marl are said to be ground separately. At Bronson, millstones were used to grind the raw materials in the wet, and at Omega they were ground as a slurry in tube mills (Fig. 15). The devices used to handle the raw Fig. 15. Tube mill. materials at this stage of the process vary much. The idea should always be to handle the resulting slurry with as low a percentage of water as possible and yet make a perfect mixture of the two materials. Screw conveyors and sludge mills are used for mixing and conveying from vat to vat and to the tanks which supply the rotaries. The slurry in the tanks must be kept in motion as it is *See Plates V, VI, and IX. 174 MABL. fed out, because the more solid material settles by gravity to the bottom and would, if allowed, disturb the equality of the mixture. The expense of the raw grinding department was estimated for a 2,400 barrel plant at Lupton, as follows: Raw grinding department (two shifts). 2 millers at $2.00 $4 00 4 scalemen at $1.50 6 00 1 electrician 1 75 Oil and grease 3 00 Total $14 75 125$ repair account 18 44 Grand total $33 19 Cost per barrel 1.4 cents. Other plants, planned to manufacture from 600 to 1,000 barrels, show four to six cents cost per barrel for this step. No expense should be spared to do this step thoroughly. The whole suc- cess of the process depends upon the fineness of grinding and intimate mixing of every particle of clay and marl so that each particle of silica and alumina shall have its portion of calcium oxide ready to satisfy it. The larger the lumps of raw material left unground, the more unsatisfied and harmful material remains. § 13. Burning. Every factory now going or projected in this State uses the Ransome rotary kiln process (Plate VII of Lathbury and Spack- man’s illustrations). It was invented by F. Ransome, an English engineer. “It consists essentially of a revolving furnace (cylindrical in form), constructed of an outer casing of steel boiler plate lined with good refractory fire brick, so arranged that certain courses are set forward in order to form three or more longitudinal pro- jections, films or ledges. The cylinder is rotated slowly by means of a worm gear and wheel driven by a pulley upon the shaft carry- ing the worm. The cylindrical casing is surrounded by tw T o cir- cular rails or pathways, turned perfectly true, to revolve upon steel rollers, mounted upon suitable foundations. Gas, oil or pulverized coal may be used for fuel.”* The kilns are usually arranged in a row (Plate VIII), with the *Cement and Engineering News. Geological SurAey of Michigan. Vo i. vm Part m piate VII % VIEW OF ROTARY. MANUFACTURE OF PORTLAND CEMENT FROM MARL. 175 supply tanks or reservoirs back of them. The kilns lie side by side with their longest axes parallel so that the motive power may be applied over as small space as possible. In a fourteen rotary plant as at Coldwater or Quincy, there are two rows of rotaries, seven kilns each with the rows facing each other. Petroleum, gas or pulverized coal is used as fuel. This depends somewhat upon which can be delivered most cheaply at the factory. The price of petroleum, of course, is in the hands of a few and is liable to vary more or less, while coal may be had on the grounds in many parts of the State. It is therefore coming into use more generally. According to Stanger and Blount, its ultimate success is depend- ent upon the method of injecting the stream of coal dust into the rotaries.* The Ransome kiln has been modified much to get around some of the difficulties encountered, and has been used with success in America, though it proved unprofitable in England. The chief trouble in the wet process, as employed in nearly all the factories in this State, is the cost of fuel. This is considerably greater than it should be when the actual heat is figured out theoretically. The weight of coal necessary to be consumed to produce clinker has been estimated as 23.28$ of the weight of the clinker produced. If a portion of the heat of the waste gases is used and they are allowed to escape at 200 degrees C., the percentage is reduced to 17.1$ of the weight of clinker in coal. In wet process, 40$ moisture, with escaped gases at 200 degrees O., 49.3$ of the heat is required to dry the mixture.* The upper end of the kiln is metal while the lower end toward the flame is lined with magnesia or aluminum brick, to withstand the great heat. While the bricks are as nearly pure as possible, the lime of the slurry acts upon them, producing fusion to such an extent that it has been estimatedf that three kilns did about the work of two, because of the break downs and delays caused from the fusing of the lining. A way is suggested and looks very feasible, of lining the fire brick with a coating of cement, packing it down so as to afford a protection to the brick below. This method is employed at the Atlas Cement works as described by Stanger and Blount. ♦Engineering News, October 24, 1901. tA. H. Cederberg. 176 MARL. Analysis of kiln brick, Stanger & Blount. Silica 55.82$ Alumina 37.98 Ferric oxide 4.02 Calcium oxide Magnesia .78 Soda 88 Potash .37 In the furnace the slurry is first dried, then as it travels further toward the flame the different materials become oxidized. The 50 or more per cent of water is driven off in the form of steam. The organic matter is reduced to ash, the carbon being driven off in the form of carbon dioxide. The calcium carbonate loses 46 per cent of its weight as carbon dioxide driven off as a gas. The silica and alumina are made soluble and brought into a nascent condi- tion with the calcium oxide. If there is much sand in the slurry, it is not as easy to grind nor as likely to be ground fine, and the sand, resisting the heat, delays the point of semi-vitrification and in- creases the cost of burning besides being hard to grind at any stage of the process. As the heat necessary to clinker cement material is between 2,000 and 3,000 degrees F. the blast of air coming in with the coal or petroleum and the gases driven off, must carry with them an immense amount of heat. The amount of heat necessary to produce clinker for one barrel of cement is estimated by S. B. Newberry as follows: Intermittent or vertical kiln (coke) 76 to 95 lbs. Continuous vertical kiln 42 to 46 lbs. Rotary kiln, dry material 110 to 120 lbs. Rotary kiln, wet material (50$water) 150 to 160 lbs. It is also estimated by Fred W. Brown, E. M.* that an additional 3 gallons of oil or 301bs. of coal is consumed where wet material is used in a rotary kiln instead of dry. These figures tally rather closely and show the increased expense at this stage of wet marls over dry limestone as a raw material. In case the marl contains a large per cent of organic matter this is nearly as expensive as water because it calls for a large draft of cold air which must be heated to the furnace temperature in oxidizing the useless organic matter. The question is then, how to utilize the immense amount of heat which is wasted. This is roughly estimated as of 175 horse ! Cement and Engineering News. MANUFACTURE OF PORTLAND CEMENT FROM MARL. 177 power intensity when but about 100 horse power of energy is used in clinkering the material. Mr. Brown makes the following suggestions for improvement. 1. Recovery of heat from clinker produced. 2. Reduction of radiation of heat to a minimum. 3. Reduction of surplus air over that used in combustion to a minimum. 4. Reduction of temperature of escaping gas to a minimum. 5. Development of the efficiency of the melting chamber to a maximum. He further recommends an induced draft to control the rate of combustion and the removal and cooling of the gases engendered in burning. There is no doubt that this could be done and also that the hot clinker could help to heat the air entering the rotary. The idea also of using the super heated air and gases to generate steam to furnish motive power, and packing or lining the surface of the rotary to prevent undue radiation of heat is promising, but their application must hinge on the ingenuity of inventors. There is no doubt that in many parts of the State the waste heat could be used to aid in evaporating the brine of salt wells so that salt could be produced in connection with cement. The two weak features of wet marl as a raw material come out in the portion of the process employing the rotaries. High organic matter is said to “clog” the rotaries and if not that, it must be dried and then oxidized so that there is another expense added to the extra cost of conveying it and handling it as slurry. The increased amount of fuel necessary to accomplish this and to drive off the moisture of about 50 $ in the form of steam is one thing that makes the process expensive as compared with handling dry and compact limestone. It is of course counterbalanced by the extra cost of grinding limestone because the marl is already finely divided by nature. 23-Pt. Ill 178 MARL. ESTIMATES OF COST.* A. 2,400 barrels per day. Coal Grinding. 4 feeders at $1.50 $6 00 2 firemen at $1.50 3 00 2 general men at $1.40 2 80 8 tons coal at $1.50 12 00 Oil and grease $25 80 Burning Department. 2 electricians at $2.00 $4 00 2 headburners at $3.33 6 66 24 underburners at $1.80 43 20 100 tons slack at $1.60 256 00 2 oilers at $2.00 4 00 8 general men at $1.30 4 50 $328 76 $354 56 10$ repair account 35 45 $390 01 B. 1,200 barrels per day. Coal Grinding (one shift). 2 feeders at $1.50 $3 00 '2 firemen at $1.50 3 00 2 general men at $1.40 2 80 4 tons of coal at $1.50 6 00 Total $16 30 4 feeders at $1.50 * . . . . $6 00 2 firemen at $1.50 3 00 2 general men at $1.40 2 80 8 tons coal at $1.50 12 00 Oil and grease 2 00 $25 40 *For some of these detailed estimates Mr. Hale is indebted to Mr. Cederberg. MANUFACTURE OF PORTLAND CEMENT FROM MARL. 179 Burning. 2 electricians at $1.75 $3 50 2 headburners at $3.00 6 00 12 underburners at $1.80 21 60 80 tons slack coal at $1.60 *128 00 2 oilers at $1.50 3 00 2 general men at $1.50 3 00 $171 14 $206 14 10 3 By titration above 1158 Indicating 0025 impurity. A fair correspondence. t9.98$ COo in the air at t=12.50 and 726.1 mm. pressure. NOTES ON THE ORIGIN OF MICHIGAN BOGLIMES. 207 surest, for a slight decrease in calcium in solution corresponds to a relatively great change of partial pressure, and if two succes- sively performed gas analyses are alike, the solubility of the bicarbonate is alike, as test 3 showed. At the beginning of the test the atmosphere above the water contained 8.94$ C0 2 . Air was sucked in and the gas above the water at once tested. The C0 2 was but 3.47$. After a while a second sample of the gas as well as one of the water was investigated. The C0 2 had risen to 6.23$, nearly double, while the bicarbonate has dropped off about one part per liter. After this point the C0 2 and lime remained constant.* CSi. Per 1000. % co 2 Pressure . Ca. Bicarbonate. \ .462 1.872 8.94 Experiment 2 •< .463 1.876 3.47 .439 1.776 6.23 67.9 Experiment 3 \ .433 .433 1.755 1.755 6.04 6.02 Treadwell and Reuter give the following other observations on the solubility of CaC0 3 in carbonated waters. Grams CaC 03 in liter. Tern. Authority. .7003 0° Lassaigne Journ. p. Chem. 44, 84. Lassaigne Journ. p. Chem. 44, 84. Bergmann Arch. Pharm. (1874) [3] 4 : 145. Bischoff Jahr. Chem. phys. geol. (quicklime). Bischoff Jahr. Chem. phys. geol. (quicklime). Marchand, 2.64, (pure) from Caro. Struve, 2.64, (pure) from Caro. Caro inaugural dissertation. Warrington p = 7.483 mm. .8803 10 .6700 1.8000 2.8000 2.5000 1 .0 to 1.5 3.0 .9852 21° ^Details of figures are as follows: Partial pressure P — p (percentage reduced to normal) X 7.60 Experiment 2. Zero point 2.2. Height of mer- cury in barometer tube of apparatus. Temperature in degrees centigrade. Barometer in mm. Initial volume 58 0 12.5 726.1 After absorption of C0 2 127.7 12.5 726.1 Tension of water vapor at 12.50 5' = 10.8. Initial volume stood under the pressure 726.1— (55. 8 + 10.8) = 659.5. Gas — COj stood under the pressure 726.1 — (121.5 + 108) = 593.8. Therefore percent air = 659.5 divided by 593.8 X 100 = 90.02 and percent CO 2 = 9.98. 208 MABL. According to these data the solubility of calcium carbonate in carbonated waters varies from 0.7003 to 3.0 grams per liter. The statements of Bischoff that the solubility of CaC0 3 is dependent on the purity of the material which furnishes the CaO or C0 2 can not be confirmed, but at 15°C, saturated calcium bicarbonate solu- tions gave, whether made of pure or impure limestone, from 1.13 to 1.17 grams per liter of CaC0 3 . At 13.2° C. the solubility was 1.31 grams per liter for the CaC0 3 from common quicklime, and 1.30 grams per liter for the pure material. At a temperature of 2.8°C there was 1.45 CaC0 3 in the liter, showing a greater solubility at the cooler temperature. Long standing produced no increase in calcium. A study of the solubility of calcium carbonate from an analo- gous point of view is presented by a work of Schloesing Compt. Bend. 74:1552. His table is as follows, but he does not describe how the partial pressure was computed: Pressure of CO 2 in atmospheres, t = 16°. Total CaCC >3 per thousand. Total 0O 2 . 0.000504 .0746 .06096 0.000808 .0850 .07211 0.00333 .1372 .1230 0.01387 .2231 .2184 0.0282 .2965 .3104 0.05008 .3600 .40863 0.1422 .537 0.2538 .6334 1.0720 0.4167 .7875 1.500 0.5533 .8855 1.8460 0.7297 . .9720 2.2700 0.9841 1.0860 2.8640 We may also add as of interest to us in this connection the fol- lowing extracts from Koth’s Chemical Geology. Vol. 1, p. 44, solubility of gases and other substances in water. Baumert found in the air absorbed by rain water (t=11.4°C; after a long rain) 1.77 volumes C0 2 , 33.76 O, 64.47 N, while in atmos- pheric air there is but 1 vol. C0 2 to 628 of O. Bunsen says that 1 volume water absorbs at 760 mm (atmos- pheric) pressure (i. e. about 1 atmosphere) : At 10° C. At 15° C. At 20° C. 1.03250 or 1 0.02989 0.02838 1.1847 36.4 1.0020 or 33.5 0.9014 or 31.8 0.01607 or 0.50 0.01478 0.49 0.11403 or 0.4 NOTES ON THE ORIGIN OF MICHIGAN BOGLIMES. 209 Thus more C0 2 is absorbed at low temperatures. The air free from C0 2 absorbed at 23°C consists of 34.91 volumes N and 65.09 O. Bunsen estimates from the power of absorption the ratio of the gasts in rain water, supposing atmospheric air to be 20.951 O and 79.007 N and 0.042 C0 2 , as follows: 5° C. 10° C (50° F:) so cu o o 20° C. 2.68 2.46 2.26 2.14 o..': 33.97 34.05 34.12 34.17 N 63.35 63.49 63.62 63.69 Under otherwise similar relations the amount of absorbed gas is proportioned to the pressure. Peligot found in 1857 2.4 per cent C0 2 by volume in the air absorbed by rain water. P. 45. According to Boussingault and Levy 100 volumes of air from a soil not rich in humus and not manured for a long time, con- tain at least 25 times, that from humus rich soil 90 times, and that from recently manured soils as much as 250 times, as much C0 2 as atmospheric air, — the maximum in 100 volumes of air 9.74. Pettenkofer found in the ground air of Munich down to 4 meters depth a maximum of 1.838 per cent C0 2 . P. 48. Solubility of Ca C0 3 . Fresenius: 1 part in 10,600 cold or 8,834 boiling water; Graham, 0.0343; Bineau, 0.016 to 0.02; Cruse, 0.036; Peligot, 0.020; Schloes- sing, 0.0131 in 1,000. If at 15°C water takes up 1 volume of C0 2 (i. e. about 0.2^ by increase of pressure and lower temperature more), the amount of carbonate dissolved increases. In water saturated with C0 2 (which does not occur in nature) is dissolved in 1,000 parts of water, according to : Bischof of chalk 9 to 10 Cossa of chalk of Luneburg (18°, 740 mm) .835 Carrara marble (7.5° to 9.5°, 753 mm) .... 1.181 Cossa Carrara marble (20.5 — 22°, 741 — 746 mm) .9487 Cossa Carrara marble (26 — 28°, 737 — 742 mm .855 Calcite 12° 754.2 mm. 1.217 Iceland spar 18°, 735.1 mm pressure .970 Precipitated CaC0 3 at 18° and 735.1 mm. . .950 Boutron and Boudet (several atmos- pheres pressure of C0 2 ) 1.16 27-Pt. Ill 210 MARL. According to Warrington at 13° and 747.3 mm. pressure water with Vf> ammonium chloride dissolves 1.050 CaC0 3 . If to a solution of CaC0 3 in C0 2 water MgCl 2 is added the solu- tion will stand weeks and can even be boiled without clouding. By continued evaporation magnesium carbonate is precipitated. According to T. S. Hunt the solubility of CaC0 3 is increased also by addition of sodic or magnesic sulphate, because bicarbonates of soda respectively magnesia form. According to Northcote, 1,000 parts of saturated salt solution contain 1.77 CaC0 3 . P. 50. Magnesia carbonate is somewhat more soluble in carbon- ated water than calcium carbonate. Merkell’s results are: In 1,000 parts at 50° C. under a pressure of C0 2 of : 1 atmosphere 2 3 4-5 6 1.31 1.34 7.5 9.0 13.2 Cossa 18°, 750 mm pressure, from magnesite 115 Bischof, 750 mm, pressure from magnesite 049 from pure magnesia 135 P. 51. Fresh precipitated magnesia carbonate is quite soluble in a solution of the sulphate and precipitates Ca C0 3 from solution in carbonated waters. Vol. Ill, p. 417 Engel and Ville, under pressure of 1 atmos phere C0 2 , the solubility varies with the temperatures, 19.° 5' C. 29°.3 C. 82° C. as follows: 257.9, 219.95, 49.0; the presence of alkaline chlorides, sulphates, and carbonates and magnesia salt increases the solubility of magnesium carbonate. Dolomite. Vol. I, p. 52 Cossa at 18° C. 750 mm. pressure, 1,000 parts of water dissolve : of dolomite (CaMg (C0 3 )) 2 .310 Of mesitine FeMg (C0 3 ) 2 075 At 16° C, 758 mm carbonated water dissolves of (Mg Fe 3 C 4 0 12 ) 115 A. Kupffer, of dolomite 2967 Siderite FeCO 3 . Wagner, at 4 or 5 Atm. pres- sure of C0 2 FeCO s 725 Cossa 18°, 760 mm 720 Bischof 60755 K. von Hauer, usual pressure, iron dust of precipitate 91 Carbonates of alkalies lessened solubility. NOTES ON THE OBIGIN OF MICHIGAN BOGLIMES. 211 Iron Carbonate. Ill, p. 417, J. Ville found in carbonated water. . 1.39 E. Ludwig Wilhelm’s quelle water 0.9648 Schloesing worked thus: crystallized pure calcium carbonate was suspended in water and through the fluid air charged with C0 2 passed until gravimetrically no increase in calcium carbonate could be detected. But there is probably an error in calculation, for Schloesing in the strongly carbonated solution, assumes car- bonate together with bicarbonate to be present. For instance, he computes: 1. CaC0 3 (neutral) according to special tests of solubility 0131 g in liter All calcium as carbonate 360 Difference calcium carbonate existing as bicar- bonate 3469 Accordingly he refers to the three following parts of C0 2 : C0 2 in the neutral calcium carbonate 00576 C0 2 in the bicarbonate 30530 C0 2 free 09757 .40863 Caro denies the existence of calcium in carbonate, from the fol- lowing test. A solution of calcium bicarbonate with excess of C0 2 was allowed to stay exposed to the air until calcium carbonate began to form at the surface, and then the CaO and C0 2 of the clear solution determined. Caro gives the following figures : 5 cm. solution contains : 0.00270 CaC0 3 = 0.0015176 + 0.0011924 C0 2 , or in grams per kilo- gram : 0.540 CaC0 3 = 0.35352 plus 0.23848 C0 2 . “Total C0 2 was determined by precipitation of 5 ccm. with am- monical BaCl 2 . The computed C0 2 is 0.0142 gr. (.142 gr. per 100 cm.) . Caro’s result is thus: Combined C0 2 = 0.0011924 g., half com- bined and free = 0.0142 g. The ratio of two numbers is 1 : 10, which certainly points to the presence of calcium bicarbonate and much free C0 2 . 212 MARL. A series of tests, Nos. 3 to 12, showed that the three values, partial pressure in per cent at 0° C. and 760 mm. pressure, amount of calcium bicarbonate and free C0 2 decrease together so that when the first and last become zero the total C0 2 is just equal to the amount needed for calcium bicarbonate. From this the conclusion is justified that calcium bicarbonate is a perma- nent salt in solution, whose solubility is for the mean bar- ometric pressure at Zurich and the temperature of 15° 0.38509 per liter. Two tables and curves are given, showing the solubility of this salt, first as a function of a partial pressure, and second, as a function of the amount of free CO dissolved. We do not repeat the curves, which can be constructed from the table below, of the original figures 5 and 6. SOLUBILITY OF CaO IN BICARBONATED WATER AT 15° C.=59° F. AND 760 mm. PRESSURE. In Air. In Water (parts per thousand). Test. % co 2 . Pressure of C0 2 . Free co 2 . Calcium bicar- bonate. 1 100.00 760 1.574 1.872 2 8.94 67.9 1.574 1.872 3 6.04 45.9 .863 1 . 755 4 5.45 41.4 .528 1.597 5 2.18 16.6 .485 1.540 6 1.89 14.4 .347 1.492 7...... 1.7 1 13.1 .243 1.331 8 0.79 6.0 .145 1.249 9 0.41 3.1 .047 .821 10 0.25 1.9 .029 .595 11 0.-08 0.6 .402 12 .385 13 .385 14 .385 Fixed C0 2 . CaCOg. Ca. Total. co 2 . CaO. .509 1.156 .462 2.587 .647 .509 1 . 156 .462 2.587 .647 .477 1.083 .433 1.817 .606 .434 .986 .394 1.396 .552 .418 .951 .380 1.321 .533 .405 .921 .368 1.157 .516 .362 .822 .329 .967 .440 .339 .771 .308 .823 .432 .223 .507 .203 .493 .284 .162 .368 .147 .353 .206 .109 .248 .099 .214 .139 .105 .238 .095 .211 .133 .105 .238 .095 .211 .133 .105 .238 .095 .210 .133 The data given above lead to the inference that calcium bicar- bonate may exist in very dilute solution. In consequence, it was of interest to determine the electric con- ductivity of this salt, for Kuster says* that the bicarbonate in very dilute solutions is hydrolytically separated since its solution colors phenolphthalein feebly red. This was found true, but the result of electric tests was that “the conductivity reached no max- imum, even in the greatest dilution, as is usually the case with salts that are hydrolytically broken up,” and bicarbonate of potash behaved in the same way. *Z. Anorg. Chem. 13, 127. NOTES ON THE ORIGIN OF MICHIGAN BOGLIMES. 213 Calcium ’bicarbonate in solution with NaCl. From Kippenberger’s tests it appears that calcium carbonate is about three times more soluble in concentrated salt solutions than in water. This greater solubility is probably dependent on the formation of double salts. Therefore it was to be expected that these double salts, like Karnallite, would be fully decomposed in dilute solution, so that the solution of calcium carbonate in dilute solutions of salt would be similar to that in pure water, and a similar behavior should be found for the bicarbonate. Tests performed as for pure water on dilute saline solutions charged with C0 2 , which contained 5 grams per liter NaCl, result as follows: ABSTRACT OF TABLES 3 AND 4. • 0.41 0.50 3.16 6.07 11.47 16.95 %0O 2 at 00 and 760 mm in gas .082 .081 .083 .086 3.4 .121 3.8 .182 24.0 .292 46.1 .368 87.2 .529 128.8 .539 pressure of CO 2 Ca .090 .089 .092 .095 .133 .201 .321 .405 .582 .593 CO 2 corresponding to .205 .203 .208 .216 .303 .456 .730 .921 .1323 1.348 CaC0 3 .CaC0 3 .332 .329 .337 .349 .409 .739 .1183 .1492 .2143 2.184 Bicarbonate .003 .027 .135 .235 .1101 1.325 Free CO 2 From this (comparing with tables 1 and 2), it is apparent that the solubility of calcium bicarbonate is but little influenced by the salt. Figures 7 and 8 of the original paper showed the solubility as function of partial pressure, and as function of percentage of C0 2 . II. Magnesia bicarbonate. No new principles involved. (Details of experiments omitted.) 1. Without alteration of the partial pressure (of 1 atmosphere C0 2 ) the C0 2 and MgO remained constant down to experiment 4, when the amount of C0 2 ceased to be enough to form bicarbonate of magnesia with the MgO present. . Fig. 9 and Fig. 10 (should be 8 and 9) showed the solubility of the bicarbonate as function of partial pressure and total C0 2 8 in mg. The result is that magnesium bicarbonate does not exist by itself without a marked excess of free C0 2 dissolved in the water. The partial pressure needful thereto, corresponds to between 2° and 4° C0 2 . If the partial pressure is less, the solution loses all of the free CO 2 with a part of the half combined and a mixture of carbonate and bicarbonate results. When the partial pressure sinks to 0 at ABSTRACT OF TABLES III. AND IV. (SHOULD BE V. AND VI.).* 214 MAUL. co 2 pressure. Mg. MgO. C0 2 combined. CO 2 half combined. CO -2 free. Total CO 2 Mg bicarbonate. Mg carbonate. 18.86 143.3 2.016 3.339 3.639 3.639 1*. 190 8.468 12.105 .547 41.6 2.016 3.339 3.639 3.639 .866 8.144 12.105 4.45 33 8 2.016 3.339 3.639 3.639 .035 7.3 3 12.105 1.54 11.7 2.016 3.339 3.639 3 236 6.875 10.766 .773 1.35 10.3 1.492 2.471 2.692 2.293 4.985 7.629 .765 1.07 8.2 1.224 2.028 2.210 1.789 3.999 5.952 / .807 .062 4.7 .865 1.433 1.561 1.101 2.662 3.663 .701 .060 4.6 .788 1.305 1.4.2 1.027 2.449 3.417 .758 .033 2.5 .655 1.084 1.181 .791 1.972 2.632 .748 .021 1.6 .594 .983 1.072 .670 1.742 2.229 .771 .014 1.1 .566 .938 1.022 .652 1.674 2.169 .710 CO 1C C? CO Gi • • r>> i> *0 00 • 1C 00 05 1ft 1.544 1.960 .702 .520 .861 .939 .613 1.552 2.036 .625 1 t'* 00 o> — 1 ^ O • • lO 00 05 to 1.525 1.954 .616 .518 .859 .935 .601 1.536 1.954 .641 NOTES ON THE ORIGIN OF MICHIGAN BOGLIMES. 215 average pressure and at 15° 0, we have 0.6410 grs. magnesium car- bonate and 1.9540 grs. magnesium bicarbonate per liter. References to the solubility of magnesium bicarbonate are very rare. Oossa and Kippenberger assume its presence only when there is much free C0 2 . Merkel is cited in Roth. IV. Sodium bicarbonate. To close the investigation, the presence of sodium bicarbonate in dilute solution was tested. The phenolphthalein test shows that the bicarbonate little by little gives off C0 2 and the solution be- comes stronger in carbonate. Referring to Kuster’s work indicating that sodium bicarbonate, by its effect in turning phenolphthalein red, is decomposed at mod- erate temperatures, the effect vanishing at 0° F, it is to be remarked that solutions of bicarbonate left long standing do the same, and the effect does not disappear at 0°, which leads to the inference that it has lost C0 2 , and a series of four tests show this to be true. We have given above, all the data we have been able to find on the solubilities of the carbonates for different temperatures and pressures. Now for the actual temperatures and pressures, the map figured herewith (Fig. 19), gives some data as to the mean annual temperatures by the isotherms or lines which have the same annual temperature. Upon the map are also placed the tempera- tures of certain flowing wells, in degrees Fahrenheit. It appears that the temperature of ground water is usually not far from 49°, increasing according to the depth of the source quite irregularly, but at times as much as 1° in 40 feet. The farther north a place is, other things being equal, the lower the tempera- ture. But it probably goes hardly below 45°, being more or less above that according to the amount of blanketing effect that the snow exerts, and the depth of the source. The water of all our deep lakes is cool, and in the bottoms of the deeper lakes it will often tie permanently cooler than the ground- water temperature. Hence chemical precipitation can never occur in the lake more than half the year, and it will not occur at great depths. Boglime, however, occurs more in lakes originally deep, than in lakes originally shallow. Still it appears to be generally true that in Michigan the marl is thicker in the shallow water at 216 MARL. the margin, and Wesenberg-Lund reports the same to be true in Denmark. This can, however, be easily explained under either theory, that of organic or chemical precipitation. But it is curious to remark that in Indiana the geologist reports* not only a deepen- ing of the marl towards the deeper water, but a more widespread Fig. 19. Reproduced from Water Supply paper No. 30, Fig. 4 , with some observations on the temperatures of flowing wells. distribution, a fact which hardly agrees with their theory of the origin of the lime. Not only will the spring water that enters the lake be cool and under pressure at the bottom so that it will not lose its carbon dioxide nor calcium carbonate there, but as it approaches the sur- *25 th Annual Report, 1900, p. 45. NOTES ON THE ORIGIN OF MICHIGAN BOGLIMES. 217 face it is liable to be diluted with rain and surface water, which as all tests and* analyses show, are far from saturated with bicarbon- atesf and a dilution with only 10^ of rain-water would keep in solution all the calcium carbonate of any ground water of which we know. Mr. Hale has pointed out that marl is liable to occur most in the uppermost of a series of lakes, into which presumably less surface water would enter, and this is a distinct point in favor of the theory of chemical precipitation. But it is by no means confined to such lakes. Davis’ observation, that in certain lime depositing lakes, the outflow is practically equal to the inflow, does not necessarily mean that the evaporation is too small to be noticed, J but merely that it is nearly balanced by subterranean springs and direct rainfall. It does, however, make it almost cer- tain that there is enough dilution of ground water springs to pre- vent direct chemical precipitation. All winter again, the water under the ice is colder than the ground water, and the escape of C0 2 is prevented. There can be no direct chemical precipitation. In the spring the influx of snow water must dilute the spring water and prevent precipitation. Only after the hot dry weather of summer has evaporated and heated the lake to saturation point could, if ever, precipitation begin, but it seems doubtful if it could get that far.§ Considerations like the above had made the origin of the bog- limes by chemical precipitation very doubtful to me, even before Messrs. Davis and Hale made it so clear that organic life was the precipitating agent in some cases, at any rate. In fact such doubts led me to suggest to them their lines of work. Mr., Davis’ discovery of calcium succinate Ca==0 2 =(C 4 H 4 0 2 ), in Chara, and its lime secretions yield a new test of the origin of the fresh water limes. Until this very peculiar salt is shown to be formed in some other way, it is a safe presumption that chara, or at least plant life has contributed largely to lime deposits containing it. It is *See also Water Supply Paper No. 31, analyses 35 to 45, and ante pp. 46 and 118. tWhile as shown above rain-water selectively absorbs considerable C0 2 from the air. tlf we look at the figures given in the “Meteorological Chart of the Great Lakes for the Season of 1899,” Vol. II, No. 9, of the Weather Bureau publications, p. 21, we see that the evaporation must be between 20 and 36 inches, and the precipitation is from 4 inches to 20 inches more. §Yet the number of facts that must be known, accurately, evaporation, ground water supply, surface water supply, temperatures, and co-solubilities under a large range of conditions, prevent our saying absolutely that it could not occur. In fact, in such a case as the marl referred to by Mr. Hale at Corrinne, where the whole lake dries up, it must. 28-Pt. Ill 2L8 MABL. also found that, as Hale has remarked, organic matter always accompanies even the purest marls. Moreover, it seems to be true that in a marl analysis, in which the CaO, MgO, and C0 2 are sepa- rately and independently determined, there is never enough car- bonic acid to satisfy the caustic lime and magnesia,* even after making all allowance for the presence of calcium sulphate. While in clayey marls it might be supposed that calcium and magnesium silicates were present, in many of the purer ones the effect is too great to be thus explained, and we are forced to believe that we have the lime united to an Organic acid, probably this succinic acid. It is not uncommon in commercial marl analyses to figure from the CaO and MgO the amount of carbonates, and for many pur- poses this is sufficient, but in such cases the chances are that the amount of carbonates is overestimated and the amount of organic matter underestimated some 2f c . § 3. Microscopic investigations. Although it might seem that the subject of the origin of boglime had been pretty thoroughly threshed out, it must be kept in mind that, in view of the number of causes that are competent under proper conditions to throw down lime, no available light should be neglected. It seemed possible that a study of the microstructure of the lime with the petrographic microscope might be an aid. For comparison with them, some artifical precipitates were made for study. (a) Microscopic precipitate by loss of C0 2 and heating. I took a sample of water from the flowing well at the end of Hazel street, Lansing, close to the bank of Cedar river.f This well flows into the air about six feet above the usual river level and has about two feet free jet. The depth is 340 feet, but the water doubtless comes in mainly at much less depth. Within half an hour of the time of taking the water, it was heated to the simmering point, when of course the C0 2 was practically lost. A film was seen floating on the top, — not a continuous coating, but a lot of calcite crystals. With an enlargement of 150 diameters their crystalline character wlls very apparent. Hexagonal outlines were plain. They were not all simple forms, nor always the same form. Rhomb faces and hexagonal outlines were common (Fig. 20), but ♦For instance, the average amount of C0 2 which Prof. F. S. Kedzie found by analysis in thirty marl analyses in which C0 2 ranged from 27.13$ to 44.60$ was 36.28$ while the amount of COo required by the weights of CaO and MgO in the marl was in each case higher, the average being 38.30$, a good 2$ more. tTemperature 50.8° F. NOTES ON THE ORIGIN OF MICHIGAN BOGLIMES. 219 simple rhombohedra were not the prevalent form. In relative dimensions and habit they resemble often Fig. 21 of the Appendix to Part II of Vol. VI of our reports, or figures 11 and 13 of the cal- cite illustrations in Dana’s System of Mineralogy. Though they are too small (about 0.02 nim) to be exactly determined, the prism or a very long scalenohedron, and the terminal rhombohedron — \ are quite probably present. The optical properties leave no doubt that they are calcite. When transmitting the ordinary ray they *1 00 1 Fig. 20. Crystals produced by evaporation. have much higher refraction than that of the balsam used for mounting (n=1.521), while with the extraordinary ray their index is very close to that of the balsam, — just a shade less, as it should be. (1.49) The directions of + and — extinction parallel to the diagonals of the rhomb faces are characteristic (Fig. 20). One twin with the twinning face probably — J was observed. In mount- ing these crystals a second crop was formed as the water around them evaporated, considerably smaller, being half or quarter the size, and spindle-shaped, like dog-tooth spar (Fig. 20f), and the 220 MABL. forms illustrated on Plate XI of the Appendix to Yol. VI, Part II, and Dana’s figures 15 to 20. Xo marl seen consists to any considerable degree of similar material. Had it been present in quantity, I do not think I could have failed to recognize it. It must be said, however, that every marl had been more or less dried, and therefore a certain amount of secondary chemical precipitation from the hard water of the lakes was to be expected. As a matter of fact, I noticed no mater- ial that need necessarily be ascribed even to this source. (b) Precipitate by evaporation. I also allowed drops of water of the artesian well used in the Hollister block (which is 150 feet deep in white sandstone of the coal measures and probably similar in chemical character to the previous well and analyses Nos. 238 and 239 of U. S. G-eol. Sur. Paper No. 31), to evaporate. In one drop the dimensions of the larger crystals are about 0.005 mm and less, about a three hun- dredth of a millimeter. In so minute crystals it is hard to measure angles accurately, but it appeared that a termination of the funda- mental rhombohedron was combined with prismatic or acute scalen- ohedral faces. In Figure 20, groups a to j were drawn by C. A. Davis from some Alma well water. I think that b and g, h, i, and j, are groups of gypsum crystals, while a, c, d, e, and k, appear to be mainly rhom- bohedral calcite, and f is plainly a scalenohedron. In 1 we have a crystal drawn by myself with the cameralucida with some pains to get the angles right, and the optical orientation indicated. The contrast in relief brought out by rotating the crystals im- mersed in balsam above a single nicol is very striking and charac- teristic. (c) Chara fragments. The calcareous Chara stems have a hollow core surrounded by a single slightly twisted row of elongate cells. The diameter is com- monly about half a millimeter. The lines between the cells are continuous and produce the effect of slightly spiral ribs. This is shown in Fig. d of Plate XVI. Fragments of chara therefore, like figures b and c appear ribbed. Close to the ribs the granulation of the calcareous aggregates is very fine, while between it grows coarser, — up to 0.02 mm. The boundary between the various areas or patches of uniform polarization color appears vague or crenu- lated. This interdigiting effect or crenulation is especially shown Geological Survey of Michigan. Vol. VIII. Part III. Plate XVI. MICROSCOPICALLY ENLARGED FRAGMENTS AND SECTIONS OF CHARA. -- NOTES ON THE OIUGIN OE MICHIGAN BOGLIMES. 221 in c, and sometimes needs the use of crossed nicols to bring it out. Even very Ismail fragments show patches of different polarization colors. Sections a and e are cross-sections of chara stems drawn by Mr. B. O. Longyear and myself. Such sections are hard to prepare and seem never to occur accidentally. The other sections b to d are such as will ordinarily be found in looking at a sample of marl. (d) Blue green algae. The apparently calcareous pebbles which are really concretions of calcium carbonate thrown down or out by Schizothrix, have already been described by Davis. Similar pebbles are noted as occurring in the marl in certain horizons -at Goose Lake, from which the Peninsular plant take their marl. The pebbles in marl referred to in the discussion of Prof. Fall’s paper before the En- gineering Society are probably similar, and the calcareous coating on dead branches and shells also. The “pebbles” on the southeast side of Zukey Lake at Lakelands, which turn brown on the side ex- posed to the light, are of the same nature. A cross-section of such a “pebble” or concretion shows a faintly radiating structure. Under the microscope I have not been able to discern this, but instead, there appears to be a cloudy aggregate of irregular calcite, not sharply crystalline nor coarse grained, not over a hundredth of a millimeter at the outside. There is not very much that is characteristic about it, and very much of the commercial boglime deposits is precisely similar. Near the Cottage Grove Higgins Lake resort, not only are the upper sides of pebbles overgrown with warty deposits of these algae, but the bottom sand is cemented in a layer about 3 mm thick, brown on the upper side and greenish on the lower. (e) Shell structure. The shells which occur in the boglimes are as Walker’s list shows, mainly (bivalves) pelecypods or gastropods (snails). What- ever the genus, and whether the structure be foliated or prismatic, aragonite or calcite, the ground up shells should show, and do as a matter of fact show a fibrous structure under the microscope. Larger pieces are commonly composed of bundles of fibres, more or less opaque, owing to the interlamination of material of different refraction. The direction of extinction is usually either parallel or varies according to some law, and there is a pronounced organic structure which can hardly be mistaken, but which varies of course, 222 MARL. according to the species. I do not think, however, that any con- siderable amount of such material could escape detection under the microscope. It generally forms an unimportant part of the com- mercial marl or boglime deposits. (f) Limestone flour. What is called clay in Michigan, is, so far as the glacial clays are concerned, more properly rock flour, and contains a great deal of finely divided quartz and other minerals, being by no means merely a hydrous alumina silicate. Inasmuch as the limestones and dolo- mites form a large part of the subsurface or bedrock of the State, and of Canada to the northeast, the almost universal presence of lime in the clays, wdiicli are thus rendered properly marls, is quite natural. Now it is not inconceivable that, as Mr. Parmelee has suggested, in a region of limestone rocks sedimentary clay-like deposits might form, aided perhaps by the greater weight of the carbonates, in which lime would predominate to almost any extent. Such clays as the following analyzed by Prof. Fall, from Alcona county are very largely limestone, though in them magnesia is present in quantity, and this we should expect would be generally true. A typical till clay from the old brickyard southeast of Harrisville, Alcona county, containing some small limestone frag- ments, is composed as follows: Free sand 11.53 11.53 Sand. Combined silica 25.71 Oxide of aluminum Oxide of iron 7.08 3.99 Organic matter, basic water Difference chiefly alkalies 3.46 2.60 42.84 Clayey matter. Calcium oxide 1 .70 Magnesium oxide 6.52 Carbon dioxide 21.00 45.22 Limestone. Sulphur anhydride 0.41 0.41 In pyrite or gypsum. 100.00 Examined under the microscope, such clays, which as we see are nearly half limestone flour, show a good deal of material which is almost indistinguishable from the alga deposits or the commercial marls. But the material is in general, more brown and opaque, contains more or less angular quartz, and almost always fragments of limestone which are over 0.01 mm in diameter, — frequently 0.05 to 0.08 and larger. Such fragments are absent in the marl, and this is the best distinction I can at present make. This is not very satisfactory. It seems quite possible that there might be quite an amount of sedimentary lime material in a bog lime, before we could NOTES ON THE ORIGIN OF MICHIGAN BOGLIMES. 223 separate it microscopically from materials of another origin, organic or otherwise. It seems likely that the rise in magnesia to which Hale refers is fully as sensitive a test of the admixture of clay marl in a boglime as microscopic examination. The calcareous clays become much harder when dry, and even when wet again, only very slowly break down. § 4. Conclusions. It appears, therefore, that any appreciable mixture of lime sediment does not produce the quality of bog lime which is desired for Portland cement manufacture. While a continual accumula- tion of boglime or marl requires a continual supply of lime which is furnished in the hard water of the springs, yet the animal and vegetable life of the lakes never allows this to accumulate to the point of chemical precipitation of the bicarbonate, but it is de- posited through organic processes. In this the Characeae play a con- spicuous part, especially in the purer marls. More minute algae may have, collectively, greater and more widespread importance. In fact, microscopic examination seems to indicate this. The role of animal life is usually quite subordinate. Notes on the microscopic examination of the different marls will be found in connection with the description of the different deposits in the next chapter. CHAPTER IX. LIST OF LOCALITIES AND MILLS. § 1. Introduction. The object of this chapter is to give as full a list as conveniently could be made both of the Portland Cement mills actually at work and the materials that they use, and also of the plants that have been planned and materials prospected as well. In description of the manufacturing plants, it must be remem- bered that the interest of the geologist is in the first place in the raw materials, and for farther discussion of the details of manufacture we must refer to Lathburv and Spackman of Phila- delphia on engineering practice, the Detroit Journal of Wednes day, April 16, 1902, the nineteenth annual report of the Labor Commissioner, and professional journals like Cement, Cement and Engineering News, Stone, etc. Still we cannot thoroughly treat the raw materials without also considering the processes of manu- facture to which they are adapted. We have also thought it would add considerably to the value of the report, to add a few statistics, for which we have to thank the Secretary of State and the Commissioner of Labor, to whose department such matters belong. A large amount of material is derived from the printed circulars and prospectuses of the various companies, or private corre- spondence with the same. It is, however, almost wholly from signed reports of reputable engineers, and is duly credited. Occa- sional comments which may be helpful by way of comparison are added. Alpena Portland Cement Co. Organized Aug. 9, 1899; capital, $500,000. The officers and direc- tors of the company are: F. W. Gilchrist, president; William B. Com- stock, vice-president; George J. Robinson, secretary; A. M. Flet- cher, treasurer; W. H. Johnson, auditor; John Monaghan, C. H. Reynolds, Water S. Russell, J. H. Cobb, attorney; F. M. Haldeman, LIST OF LOCALITIES AND MILLS . 225 superintendent. Mill located just east of Alpena on the shores of Thunder Bay. A thousand foot pier gives water transportation, and the Detroit and Mackinac R. R. also runs to the mill. This was the first mill using limestone for the calcium instead of bog lime. It is the Alpena limestone close to the mill and belongs in the Traverse group, and corresponds somewhere nearly to the Encrinal limestone of the Hamilton of New York State. Not all of the bed is equally pure, however, and Dr. A. W. Grabau, who has made a par- ticular study of the conditions for us, reports that the old coral reefs which occur in the bed furnish the purest calcium carbonate.* This is a very important result, for these coralline parts are easily recognized. The limestone is also used, especially the purer part, for the purification of sugar, and the report of the beet sugar chemists confirms the analyses of the local chemists, that at times the limestone is practically pure CaC0 3 . This is one of the plants which have the advantage of using a local shale clay. “The raw materials are very economically handled. The clay brought from the beds to the north is piled in great bins in the clay storage house. This house is 225 by 60 feet in dimensions and will hold clay sufficient for 60,000 barrels of cement. From the quarries to the plant, a distance of 800 feet, tracks on which are run cable cars are laid through the clay shed. Here, each car of rock, as it passes through, is weighed, analysis having been made, and the correct amount of clay is added to make a perfect cement mixture. The cars then run to the mill and their contents are dumped into the crushers. “The materials then pass through the crushers, rolls, ball mills and tube mills automatically, being ground finer and more thor- oughly mixed during each process. During the wet grinding pro- cess, water is added in the ball mills and the final and finishing mixing is done by the tube mills, which contain imported flint peb- bles. . The action of these against the wet mixture, produced by the revolutions of the mill, reduces it to a slurry. Then the mix- ture passes into correction tanks, from which samples are taken by the chemists and tests are made to guard against error. The contents of each tank are corrected before the slurry is allowed to leave it. There are 12 mixing tanks, each 14 by 16 feet high and with sufficient capacity to each hold enough slurry for 250 barrels of cement. “This mixture carries about 33 per cent of water which makes the resultant process better. Marl and clay mixture must neces- sarily carry a higher degree of moisture than with the dry pro- cess. The kiln capacity is much greater where lime rock is used, ♦Annual report for 1901, pp. 174 to 191, especially page 178. 29-Pt. Ill 226 MARL. as there is less water to drive out of the material before it is cal- cined. At Alpena the rotaries each have a daily capacity of from 140 to 150 barrels of cement as against 100 by those using marl. “From the storage tanks the slurry is fed into the rotary kilns. The fuel used in these kilns is powdered coal, prepared by drying and grinding, and is fed into the kilns by an air blast. The kilns are taken care of by experienced burners. From the kilns the cement clinker is discharged into conveyors and carried to the clinker room to cool. Six rotaries are in constant operation and the daily capacity reaches 1,000 barrels of cement. From the clinker room the material passes to the grinding machinery, con- sisting of rolls, ball mills and tube mills — the chemist takes the ground cement at this point and tests it for fineness, after which it is conveyed to the stock house, which has a capacity of 50,000 barrels, where it is allowed to season and then packed for ship- ment. Before shipment the chemical department makes a thor- ough test of the finished product for specific gravity, constancy of volume, soundness, tensile strength and setting time, and the cement is shipped only as certified by them to be in proper condi- tion for immediate use. “The dimensions of the various buildings are as follows: Stock house, 240 by 100 feet, making about 24,000 square feet of floor space; mixing and kiln building, 259 by 105; cement grinding room, 190 by 105. In addition to these there is a thoroughly equipped cooper shop, machine shop and round house.” Although they are not at present using the bog lime it may be of interest to give the analysis of it as well as of their shale. It comes from Middle Lake, on Sec. 18, T. 32 N., R. 9 E., about seven miles north of the mill, where the company own a thousand acres tract including also their shale clay beds. BOGLIME. Calcium carbonate 92.91 Magnesium carbonate 1.89 Silica tr. Ferric oxide 0.53 Alumina 0.21 Sulphuric anhydride tr. Organic matter .80 Water, etc 2.01 99.87 LIST OF LOCALITIES AND MILLS. 227 CLAY SHALE. CaO (CaC0 3 = 4.48) 2.51 MgO 65 Silica 61.09 Ferric oxide 6.78 Alumina 19.19 Sulphuric anhydride 1.42 Water and C0 2 5.13 Potassium oxide 1.80 Sodium oxide 1.36 99.93 Some of these shales and clays around Alpena will doubtless make good face and even paving brick. The rocks around Alpena have been quite fully described by A. W. Grabau* in the Annual Report for 1901, and will before long be subject of a monograph by him. Omega Portland Cement Co., Organized Feb. 18, 1899; capital, $300,000; Jonesville, Hillsdale Co. The officers were: Frank M. Stewart, president; Israel Wickes, vice-president; Chas. F. Wade, secretary-treasurer; George H. Sharp, superintendent; Homer C. Lash, chemist. The following report was prepared for us by W. M. Gregory, and was printed in the Michigan Miner for May 8, 1901, Vol. 3, No. 6. The Omega Portland Cement Company, of Jonesville, Hillsdale County, Michigan, with a plant at Mosherville, owns extensive land tracts in Section 15, Township 5 south, 3 west. The plant stands near Cobb’s Lake, on the Fort Wayne branch of the L. S. & M. S. Railway. This lake is one of a series of small lakes at ♦See especially pages 175 to 190. 228 MARL. the headwater of the Kalamazoo River. In the near vicinity are Hastings, Johnson, and Mosher’s Lakes, and many large areas of marsh. All by actual tests and explorations have been found rich in marl deposits of an excellent quality. The immediate topography of the land in this region is rolling and hilly, this being in the locality of parallel morainal ridges deposited by the ice fronts as it retreated to the north. The clay loam forms a storage basin for the lakes — three to four miles is the average distance from the crest to crest of the valley which holds this lake chain. The valley sides are gently sloping and in places covered with sand. This region of meandering creeks, sluggish rivers and plant choked lakes was formerly considered valueless and even hindered farming interests. The discovery of marl has been the means of making a busy little village in the midst of what was once worth- less soil. This is only a type of what is occurring in many places in our State. Northeast of Jonesville there are many lakes of this same character. Already at Woodstock a 600 barrel plant has been erected, and a prospective plant is in consideration at Grass Lake. Near Hanover, Moscow, Duck Lake and Addison Lake are lands rich in marl deposits. At Coldwater, Quincy, Bronson, Union City and Sand Lake extensive deposits occur and four of these places have successful plants in operation. Spencer Lake, some miles east, has also a marl bed. This region within a radius of less than fifty miles is especially favorable for extensive cement manufacture because of the abundance of marl and clay. A few words concerning the lakes in this region: They are all elliptical in shape, and on the southern shore are low morainal ridges which extend northeast to north; in many cases partly en- closing the lake. A mile is the greatest length and one-half mile is the average width. The most valuable marl deposits occur in the deepest lakes, and in fact no extensive amount occurs in any of the shallow lakes, and in such cases the sand renders the marl valueless, as some of our manufacturers have found by experience. No large inlets are known to exist in lakes with an abundance of the deposit and as a rule the outlet is plant choked. The water is in the greatest part derived from the underlying Marshall sand- stone. The plant of the Omega Company has a daily capacity of 700 barrels. The buildings are of brick and steel, and the storage LIST OF LOCALITIES AND MILLS. 229 house of cement concrete. The power house is 80x160 feet, con- taining a 750 horse power engine, air compressors, pumps, dynamo, etc. The largest building or wet end department is 80x200 feet, and here the handling of marl and clay takes place. The marl is taken from the lake, which is 400 feet east of the mill, by a large steam dredge, and the beds of marl run to an average depth of 50 feet. The lake is one-half mile in length and one-quarter mile wide, being filled along the shore with much plant material. ■ At the center of this lake a depth of 40 to 60 feet is found. For con- venience in handling the marl the lake has been slightly lowered by dredging the outlet. The slimy marl taken from the lake bottom by the dredge is deposited in horse cars, skips or buckets, with a capacity of one cubic yard, and drawn to the conveyor shed, 30x130 feet, where the marl is elevated and conveyed by trolley and by automatic dump in skips dropped into the stone separator, which disinte- grates the marl and relieves it of all sticks, grass and stones. In this building the clay, which is shipped in from Millbury, Ohio, is pulverized by passing through a dry pan, dried and weighed, and elevated to the mixing floor, where with the marl coming from the stone separator is mixed with the clay, forming a mud or slurry and passes to the pug mills. The Omega Company also have a clay bed one and one-half miles northeast of the works, which matches the clay brought from Mill- bury, and which they use at times when weather and roads will permit of transportation economically. ANALYSIS OP MILLBURY CLAY.* SiOo Al 0 O s Fe.,0., CaO MgO Volatile matter 64.85 17.98 5.92 2.24 1.40 4.98 It would seem quite possible that some of the shale outcrops near Reading, or south of Jackson, would furnish suitable clay. Clay taken from the surface, if free from sand, is more apt to *Compare other analyses of this clay elsewhere given, e. g., those made by J. G. Dean at the Peninsular plant at Cement City, and those given by Prol I. C. Russeli in his report in the 21st Annual of the United States Geological Survey. 230 MARL. prove satisfactory for cement manufacture, as it is easier to mine and freer from lime than a lower strata. The manufacturer wishes a clay low in lime. Slurry is carefully watched and tested at the Omega plant, and no trouble has been encountered through presence of sand in the marl or clay. After leaving the disintegrating pug mills, the slurry passes into vats and is pumped up to an elevated tank, where it is again screened and runs by gravity to the mixing and grinding wet tube mills, these mills being lined with wood and one-half filled with Greenland flint pebbles. After the material has passed through these mills it will all pass a sieve of 10,000 meshes without residue. The object in very fine grinding is the attainment of the most intimate admixture possible of the clay and marl, so that the heat will quickly produce incipient vitrifica- tion. Slurry is then passed into large storage tanks and from these passes by gravity as needed into vats at rear of the five 60-ton rotary kilns; these are 60 feet long and six feet in diameter: the shell being made of extra heavy boiler iron, lined with alumin- ate brick. The slurry is pumped from vats and forced into the end of the rotary, the rotaries being set on an incline of one-half inch to the foot. The department containing the rotaries is 80' x 100'. The rotaries are heated to about 2,900° F. by means of a gas flame generated by a continuous blast of powdered coal; the slurry while in the kilns is subjected to temperatures varying from 1,290° F., at which CaO, Si0 2 is formed, to 3,000° F., where CaO, A1 o 0 3 is formed. The calcined product of the kilns is termed clinker and has the following analysis: SiO, 22.24 A1 2 0 8 7.26 Fe,C 3 2.54 CaO 64.96 MgO 2.26 SO, 41 H 2 0 and C0 2 33 The clinkers, if good, have a lava-like texture, being somewhat porous and with a greenish black bronzed color. Too much clay is shown by a tendency to give a flaky powder on cooling. An excess of lime gives a clinker of great hardness with a glassy black luster, or a fractured surface may show white specks of free lime. The excess of lime is very injurious to cement, because LIST OF LOCALITIES AND MILLS. 231 caustic lime expands in slaking and will disintegrate the cement mortar and produce “blowing.” Too much silica will cause the clinker to crumble. Iron imparts a bluish black color and tends to produce fusion in the presence of heat. The building where the coal is prepared for use in the kilns is 52 ft. x 68 ft. in size, and contains a preliminary crusher, dryer, pre- liminary pulverizer and two German tube mills for finishing the product. The coal is pulverized to pass sieve of 10,000 meshes with not over two per cent residue. On an average three cars of coal are used per day and all is prepared in this special way for use and con- veyed by blast from fans into the kilns. The composition of the coal is an important factor, as an abundance of sulphur or iron pyrites is a damage to the quality of the cement, and the percent- age of ash in the coal is also an important factor, and coal must be analyzed each day. PITTSBURG COAL. Moisture 1.00 Vol. matter 39.37 Fixed car 55.82 Ash 3.81 Sulphur 92 The coal item in the expense of manufacture is a large one, and if Michigan coal could be used it might lessen the cost, but as yet its use has not been successful in this plant. After the clinkers are properly burned they pass from the rotaries to the cooling and grinding department. The grinding mills are of the ball and tube mill patterns of German manufacture; the fine grinding of the clinker is one of the essential elements of cement manufacture. The following are some of the tests of the Omega brand: THE OSBORN ENGINEERING CO. (INCORPORATED). CLEVELAND, OHIO. CEMENT TESTING DEPARTMENT. Report No. 2. Records, p. 89. Reports of Tests of Omega Portland Cement. Samples received from John Laylin, City Engineer, Norwalk, Ohio. Reported to John Laylin, Sept. 6th, 1900. Fineness, Activity, Constancy of Volume. 232 MARL, LIST OF LOCALITIES ANI) MILLS. 233 It is said that 98^ will pass a sieve of 10,000 meshes to the square inch, and that briquettes possess a tensile strength of 400 to 700 pounds when one week old and 500 to 800 pounds when one month old. In the season of 1900 it produced 54,500 barrels, in 1902, 120,000 barrels. The following are three analyses by Mr. W. H. Hess from the Cement and Engineering News for February, 1900: 1 2 3 Silica 39.53 58.24 68.21 Alumina 11.46 20.56 18.64 Iron oxide 4.59 5.68 5.32 Calcium oxide 13.78 0.61 0.22 Magnesium oxide 5.19 9.24 9.16 Sulphur anhydride .... Difference, carbon diox- 1.62 9.91 9.12 ide, organic matter, water, etc 23.83 15.49 7.33 100.00 100.74 100.00 In No. 1, which is of the surface clay type, the calcium oxide would mean 24.63 per cent of the carbonate, and similarly 19.84 magnesium carbonate, or 35.47 carbonates, which would leave from the “difference” about 7.33 for organic matter, basic water, alka- lies, etc. Peninsular Portland Cement Co. Organized June 24, 1899; capital, $875,000. The office of the com- pany is at Jackson, and Jackson capital is largely interested, but the plant is in the northwest corner of Lenawee County at Wood- stock, on the L. S. & M. S. R. R., and Cement City on the Cincin- nati Northern, the latter town site having been platted by the company. The officers were: W. R. Reynolds of Jackson, presi- dent; C. A. Newcomb of Detroit, vice-president; W. F. Cowham, secretary and manager; N. S. Potter of Jackson, treasurer. The capitalization was half 7^ preferred stock to be returned in 5 years. The net cost of manufacture was estimated at 80 cents. The output when I visited it in the fall of 1901 was about 700 barrels a day. The following notes are from my visit and informa- tion kindly furnished by Mr. J. G. Dean, then chemist. : The plant is located on the borders of Goose Lake not far from the northwest corner of Lenawee County. 30-Pt. Ill 234 MARL. The marl is dredged from the water of the lake, and forced through a pipe line into the marl tanks, where it is stirred and allowed to flow slowly into the mixers. While in transit the clay, which comes from Millbury, Ohio, is incorporated in the right pro- portions by an Archimedean screw. From the mixers it passes into the slurry tank, thence into the rotary roasters, in which a coal dust blast gives the heat. The coal is high in volatile matter and low in sulphur. At the lower end of the rotary it drops as a clinker and then passes into the grinder where it is reduced to powder, by being rattled with flint pebbles brought from abroad (France and Greenland). Experiments with Michigan pebbles have proven entirely unsatisfactory. The plant is producing about 700 bbls. a day with six kilns, and is beginning enlargement. The company owns a number of marl lakes in the region besides, but Goose Lake is the one which they are now using. It lies in an east and west deep trough, 60 feet or more, below the adjacent county. This trough to the east crosses the line of the Cincinnati Northern in a wide valley or open swamp, probably largely under- lain by marl, and is said to extend up into Jackson County to the northeast. The outlet of the lake is to the west and the trough extends there also. This trough appears to be not merely superficial, but to extend to the rock surface also, for in the village of Cement City half a mile north of the plant one has to put down a well but 8 to 11 feet to encounter sandstone and shale, while in the lake 60 feet below, soundings even 80 feet deep are said to be sometimes still in marl. Elk horns are said to have been found 30 feet down. Not far north on Sec. 19, T. 4 N., R. 1 E., in a low swamp about the same distance below the high flat-topped hills as Goose Lake there is a drilled well (t. 50° F.) flowing. It is quite likely, there- fore, that as seems often the case around marl lakes there is an upward artesian pressure of the ground water. It will be noticed that this lake conforms to Haleys rule that the marl lakes tend to lie in deep depressions. LIST OF LOCALITIES AND MILLS. 235 The water is also hard, as is shown by the following analysis by Dr. Hodge: Grains per U. S. Gallon. Parts per thousand. CaO 6.410 .110 . 101 C aO+ . 079CO, = . 180C aCOcj MgO 2.562 .044 + .048CO 2 = .092 MgC0 3 Fe 203 Alj 03 tr tr Si0 2 .203 .003 S0 3 .800 .014 + .009CaO=.025CaS0 4 C0 2 6.275 .107 Alkalies etc .220 .004 Residue 16.470 .282 Now in this analysis it is noteworthy that after supposing that all S0 3 is combined with CaO and that and the MgO is combined as carbonate, there is not enough C0 2 (.107 — .048=.059 instead of .079) to satisfy the calcium oxide. This perhaps indicates some organic salt of lime, for instance the calcium succinate discovered by Davis. Another point is that the water, as will be seen by reference to Treadwell & Reuter’s paper, is almost or quite saturated with lime and magnesia. The sample was taken over the marl bed, directly at the mouth of the intake ditch. A third point of interest is the higher ratio of MgO to CaO than in marl or clay. This indicates that the water is of a residual nature, left after the deposition of the marl. The company own a clay bank about two miles west of the plant, on the north side of the hollow in which the marl lies. They have not used it for cement manufacture, preferring to use Millbury Ohio clay of the composition of Analysis 1. 236 MARL. (Average of 50) 1 2 3 4 5 Si0 2 61.06 67.06 55.26 58.85 45.27 ai 2 o 3 18.10 20.50 23.34 18.36 8.33 Fe 2 0 3 6.65 2.52 2.52 7.16 4.84 CaO 1.29 .94 4.15 1.18 15.99 MgO .53 tr. tr. 1.98 so 3 1.05 tr. 2.00 .38 HoO and organic mat- ter 9.20 8.01 11.40 8.13 Difference, alkalies, etc 2.12 0.97 1.33 2.96 100.00 100.00 Analyses 2 to 5 are of local clays, — No. 2 from under the marsh, No. 3 a surface yellow clay probably leached of much of its lime, — No. 4, also from the top of the bank, with a low per cent of lime, while No. 5 is a partial analysis of the clay 8 feet down in the bank. A similar relation of clay analyses is very widespread and will be noticed in many other sets given in this volume. From such analyses it is probable that the ground water leaches out the carbonates unequally, preferring the magnesia, and by comparison with the following marl analysis we see that the agent which throws down the marl decidedly prefers the lime, so that there must tend to be a concentration of the magnesia in the water : ANALYSIS OF GOOSE LAKE MARL. CaO 51.56 MgO 1.26 SiO, 0.22 Fe 2 0 3 Al 2 0 3 0.76 Volatile matter, etc 46.20 100.00 The marl abounds in shells which have been determined by Mr. Bryant Walker in his paper elsewhere given, but Chara is also found in the marl. As bearing on the origin of the marl it is worth noting that at times streaks of material which dredgers would call sand or gravel are struck. This proves, however, to be pure calcium carbonate and is prob- ably largely composed of the Schizotlirix aggregates which are LIST OF LOCALITIES AN1) MILLS. 237 elsewhere described. They tend, however, to settle in the slurry and cause trouble. The marl is said to range from 10 to 42 feet thick, and the lake is half a mile wide and a mile and a half long, ,the deposit of marl being over 300 acres. Besides this lake other lakes in the neighborhood are owned by the same company. It is said that the average of over 200 borings ran 96.12^ CaC0 3 and less than lf 0 Mg. Peerless Portland Cement Co. Organized Aug. 23, 1896; capital, $250,000. Oldest of the recent plants. First operated as a vertical kiln plant, as when visited by Hale and Ries, and when the view given in Plate III was taken; it has recently been remodelled to the rotary kiln. The officers of the company are: A. W. Wright, of Alma, presi- dent; S. O. Bush, Battle Creek, vice president; J. R. Patterson, general manager; Wm. H. Hatch, secretary and treasurer; direc- tors, the above officers, with W. T. Knowlton, Saginaw. It is a close corporation, and the stockholders are few. The plant is located at Turtle Lake near Union City close to the line between Branch and Calhoun Counties. Six hundred and seventy-five acres of marl land are owned by the company, and is reached by means of a little railroad. The marl is found upon the surface, and is so dry that water has to be added when it reaches the plant. The marl is almost entirely free from organic matter and is very readily worked. By means of a bucket dredge, operated on a track, the marl is dug and lifted into the dump cart. To obtain the marl thus dry the level of Turtle Lake, “which had been twice lowered before, the last time in 1873/’ but still stood 22-J feet above the St. Joseph River, was lowered some 14 feet. From the Detroit Journal of April 16, 1902, we cite the follow- ing account of the changes in the manufacturing plant: “Intermittent vertical kilns were first installed by the company. These kilns were charged, then lighted and burned out like a lime- kiln. From a distance of three miles the marl was first hauled to the plant in wagons, then it was mixed with clay in a pug mill and made into bricks. These bricks were first dried in a drying kiln, then piled in the burning kilns with alternate layers of coke. After being burned the clinkers were drawn off and ground. The process was necessarily slow, as compared with that in use the present day. Two years ago another change was made in the mill and Dietch Continuous Vertical Kilns installed. In these kilns the mixture was charged at the top and the clinker drawn off at 238 MARL. the bottom. Still progressing the company decided last fall to construct a modern cement mill and to that end hundreds of work- men have been engaged all winter in the erection of a model cement plant. Many entirely new features have been introduced into this mill, and right from the start an output of 1,200 barrels per day is confidently expected from the eight 70-foot rotaries. Two hun- dred thousand dollars is being expended upon this plant. “Beds of both plastic and clay shale owned by the company are located within a mile of the mills. The shales belong to the Cold- water formation. “The cars of marl are pulled up an elevated tramway on the track scales where the marl is weighed, and then the clay is added before being dumped into the stone separator. From there it goes to the pug mill and then into a large tank where through return pipes the mass is kept running continuously in order to obtain a uniform mixture. It is corrected at this point by the addition of the proper amount of clay or marl determined by the chemist. From these tanks the mixture is pumped into the wet grinding tube mills and then falls into great floor tanks of concrete. In the bottom of these tanks a continuous screw conveyor forces the slurry into mammoth concrete correction tanks. These tanks are the source of just pride to the engineering force of the company. They are constructed entirely of concrete and are 22 feet deep by 22 feet wide and 22 feet long. The slurry in these tanks will be agitated by compressed air. The clay is prepared by being first dumped into a dryer and then ground in a Williams mill. “The great rotary room is undoubtedly the most interesting part of the plant. Some innovations are here introduced that will materially increase the output of each rotary. The inventions are the product of advanced thought and the broadest of experiments. The rotaries are seventy feet long, being ten feet longer than the largest rotaries in any Michigan mill. The pulverized coal, to feed the rotaries, is prepared in a separate building where the most im- proved coal grinding machinery has been erected. The Peerless company has placed devices on the rotaries from which the waste heat from the kilns is utilized in drying the slurry before it enters the kilns. This is automatic and is said to increase the capacity of each kiln to a marked degree. The rotary room was con- structed on a side hill and this has proven especially advantageous, as it saves the handling of the clinker as it leaves the kilns. Under the clinker end of the kilns has been constructed a retaining wall and in this room, 21 feet below the kilns, are the foundations for the eight automatic Wentz clinker coolers, which are being erected so that the hot clinker falls directly into them. By this device the hot air is fanned off of the clinker and driven back to aid in reducing more slurry to a calcined state. “As the clinker drops from the coolers it is conveyed along the floor to the rolls and from there into eight Griffin mills and then into two large tube mills for the finishing process. As the cement leaves these mills it is elevated by belt and tripper arrangement to the top of the three-story warehouse and there dumped into hopper bins. These bins are two stories in height and are con- LIST OF LOCALITIES AND MILLS. 239 structed of the best Kentucky oak, the huge pillars not depending upon the walls of the building, the construction being entirely within itself. As the cement drops from the third to the second story bins it is turned over and from there goes to the packer. The old and the new warehouses, which extend along the Michigan Central tracks, have a capacity of 100,000 barrels. It will be seen that the company is amply provided for winter storage. At the track, coal can be unloaded and elevated to the boiler room of the plant. “The power plant promises to be one of the finest in the state. Four Scotch Marine internally fired boilers will furnish steam for driving a 500 horse power Hamilton Corliss engine, a Fitchburg Tandem Compound 450 horse power, and a 300 horse power simple engine. Rope drives will be used in part of the plant. Twenty electric motors are being installed and electrical transmission used to advantage in driving the gear of many of the machines.” Bronson Portland Cement Co. Organized March 3, 1897 ; capital, $500,000. It is said that there is a mortgage of $100,000 on the plant which is said to have cost about $250,000. This is one of the well established plants of the State, and has been visited both by Dr. Ries* and Mr. Halef and tests and a description of the process of manufacture are elsewhere given. In materials and location it is like and not far from those of the Wolverine Co. The following are additional analyses of the Bronson clays, be- side that given in Part I of this report. REPORT OF ANALYSIS. Date of receipt, Dec. 5th, 1900. Composition 592 593 594 595 65 66 67 68 Silica . . 61.94 56.64 61.10 59.36 Alumina . . 11.58 12.18 13.91 12.38 Iron oxide (ferric) 3.49 3.59 3.62 3.62 Oxide of calcium . . 5.92 8.17 6.32 5.63 Oxide of magnesium . . 4.85 4.29 3.91 4.62 Sulphuric acid (anhydrid) . . . Organic matter .18 .31 .31 .30 Respectfully submitted, (Signed) W. H. SIMMONS. Bronson, Mich., Dec. 17th, 1900. ♦This volume (VIII), Part I, pp. 42 and 43. tChap. VI, p. 104. 240 MABL. Newaygo Portland Cement Co. (Gibraltar Brand.) Capital, $2,000,000; organized May 24, 1899. Cornerstone laid June 29, started June 5, 1901. The officers of the Newaygo Portland Cement Company are Daniel McCool, member of American Society of Civil Engineers, president; Wm. Wright, vice president; B. T. Becker, secretary and treasurer. Directors: F. G. Bigelow, Milwaukee; H. D. Hig- inbotham, Chicago; George Barrie, Philadelphia, and W. North- rup, St. Louis ; Clay H. Hollister, Grand Rapids, Mich. Description by Richard L. Humphrey.* The Newaygo Portland Cement Company’s plant is located at Newaygo, on the banks of the Muskegon river, thirty-six miles north of Grand Rapids, Michigan. It is one of the finest designed and equipped plants in the State of Michigan. The plant is electrically operated, the power being furnished by two 500 H. P. 3-phase generators, driven by eight Lombard water wheels acting under a 15-foot head. The water is furnished by the Muskegon river. The accompany- ing views, Plates Nil and XVIII, show the darn, race way and in- terior of power house. The slurry is agitated and handled entirely by compressed air. The efficiency of this system cannot be overestimated. The centri- fugal pumps usually in use are very expensive to maintain as they wear out very rapidly. The absence of line shafting is noticeable, each machine being equipped with an individual motor, in some cases two, which enables the mill to continue in service in case of break down of one of the motors. The automatic system for controlling the com- pressed air is admirable. The marl is found in a series of lakes owned by the company in Newaygo county and about five miles from the plant, known as Little Marl, Great Marl, Pickerel, Kimball, Fremont and Hess lakes. The following is an analysis of marl taken from the Great Marl lake: Silica 1.24 Iron and Alumina 80 Calcium carbonate 90.90 Magnesium carbonate 2.97 Organic matter by difference 4.09 400.00 "“Consulting Engineer, Philadelphia, Pa. GENERAL VIEW OE PLANT. Geological Survey of Michigan. Vol. VIII. Part III. Plate XVII. PLAN OF NEWAYGO PLANT. Geological Survey of Michigan. Vol. vm Part m piate x vm I DAM AND RACE-WAY FOR NEWAYGO PLANT. LIST OF LOCALITIES AND MILLS. 241 The composition of the marl in calcium carbonate ranges from 65 to 95 per cent. Olay is found on the company’s property along the Muskegon river opposite the plant; the following is a representative analy- sis of this clay :* Silica 55.84 Iron oxide 3.02 Alumina 8.90 Lime 9.98 Magnesia 5.16 Loss 13.68 96.58 The cement produced by this plant is first-class in every partic- ular, and the machinery is the best of its kind. The following is a brief description of the plant : The mill is on the line of the Pere Marquette railroad over which road for a mile and one-quarter the marl is hauled to the mill; the remaining three and one-half miles is over the cement company’s siding. The dredge and plant used in excavating the marl is shown on Plate XXII. The marl is dumped into a bin [(1) on Plate XVII]. There is also storage provided, under the trestle (400 feet long), to supply the mill during the winter months. From the bin the marl flows through a gate in the bottom, operated by a slide valve, into a machine called a separator, which drives the marl out through a perforated head in the machine, separating from the marl all foreign matter, such as sticks, stones, etc. Water is introduced at this point in quantity (about 55$) sufficient to reduce it to a slurry. The pure marl flows through a pipe into pump (2) which pumps it by compressed air into three storage tanks (3) connected together by pipe. These tanks hold about 90 cubic yards each. From these tanks the chemist takes his samples for analysis, to determine the proportion of clay to be added. From the tanks the material, now in form of slurry, flows by gravity in a pump marked (4) which pumps it into two measuring tanks (5), these being used alternately. The *At present, however, the company is using a clay found in connection with the gypsum at Grand Rapids whose analysis is more like that given in Part I, pp. 40 and 41. L. 31-Pt.III 242 MARL. clay is brought from storage and fed into a pair of rolls, then into a pug mill where water is added and it is reduced to a thin slurry. From this mill it passes into two Gates’ tube mills in which it is made impalpably fine. This in turn is forced or pumped by air into a measuring tank. The marl and clay are fed separately from the bottom of the measuring bins into a measuring hopper. From this hopper it is pumped into three 90 cubic yard tanks (11). The number of hoppers of marl and clay pumped into each of these tanks will depend on the composition of the marl. When the tank is full it is thoroughly agitated by air. The chemist then takes another sample. These are called correction tanks. Should the composition not be correct clay or marl is added until the desired mixture is obtained. From these tanks the slurry or syrupy mixture of clay and marl flows by gravity into a pump (12) which forces it into the automatic feeders into the three tube mills (13), in which the material is reduced to an impalpably fine state. The tube mills discharge it into a trough running to a pump (14), which forces it into the 90 cubic yard tanks (15) back of the kilns; there being a tank for each kiln. All tanks are continuously agitated by means of compressed air. From the tanks it is pumped into automatic feeders from which it is fed into the rotary kilns marked (17), in which it is clinkered and is discharged into the McCasslin conveyor marked (19), which forms a continuous belt around all the rotary kilns passing in a trench underneath, then up a tower at the side of the building, over- head through the ventilator or louvre of the building and down the opposite side, where it is discharged into a cooling tower (33), and delivered by this tower onto a conveyor belt (34), which takes it to the dry grinding building and delivers it to elevator (35), by which it is elevated and deposited on conveyor belt (36) which in turn delivers it to clinker storage bins marked (37), there being one for each Griffin mill. From these bins it is fed by gravity into the Griffin mills marked (38) and pulverized to an impalpable powder; flowing from them by gravity again into a screw conveyor marked (39), by which it is delivered to elevator (40), and delivered by this elevator to either screw conveyor (41) or belt conveyor (42), either one being in reserve in case of a break down. These conveyors take the finished cement and deposit it again into a screw conveyor (43), LIST OF LOCALITIES AND MILLS. 243 which carries it overhead, through the cement warehouse, emptying it into any bin desired. When the cement is shipped, it is drawn from the bottom of any one of these bins into screw conveyor (44) of which there are two, one on either side of the alleyway, conveyed by the screw conveyor to a second screw conveyor (46), which delivers it into the packing bins in the packing house, where it is either barreled or sacked by machinery, and if cars are not at hand to take it to market, it is piled in the warehouse adjoining the packing house. The coal is either shoveled direct from a car standing on the trestle onto the conveyor belt (20), or is wheeled from storage under the trestle and dumped onto this same belt, which carries it to a coal cracker (21). From there it is elevated into a Cummer dryer (22), passes from the Cummer dryer into a second elevator, which carries it up and dumps into small bins over Griffin mills (23), where it is pulverized and then passed by a screw conveyor (26) into elevator (27), which elevates it into screw conveyor (28), by which it is carried and deposited in coal storage bins (29). From there it is fed into the rotaries by a blast of air from fan (31), driven by motor (30). These rotaries are all driven by motor (18) of which there is a duplicate kept in reserve. In the coal grinding building the machinery is driven by motor (25), belted to a jack shaft (24), which drives both the Griffin mills. Each of the Griffin mills in the dry grinding building is driven by a separate motor, as in each of the tube mills in the wet grinding building. The agitators of each set of tanks, Nos. 3, 11 and 15, are also driven by separate motors. The plant has been in continuous service for over one year during which time it has proved to be one of the most successful and economical in the State. The story of the discovery of the deposits of marl is as follows:* “A year ago Charles E. Greening, of the firm of Greening Bros., extensive nurserymen at Monroe, was on a business trip through the northern part of the lower peninsula. On May 23 he delivered an address at Newaygo, and the day following joined a fishing party at Pickerel Lake, near that village. While sitting on the trunk of a fallen tree, Mr. Greening observed that the roots of the tree were covered with a white substance resembling snow. His curiosity prompted him to taste it and he detected in it a strong flavor of lime. He sent a sample to the Agricultural College for analysis, but never heard from it.' *Grand Rapids Herald, May 9, 1899. 244 MAUL. “A few months later Mr. Greening met Prof. Fred H. Borradaile, State analyst, and gave him a sample for analysis. When the lat- ter reported he startled Mr. Greening by urging him to go to New- aygo at once and buy up all the land containing the deposit that he could get his hands on, explaining that the substance was a most valuable specimen of marl. “Mr. Greening hastened to Newaygo and immediately purchased about 1,000 acres of land surrounding four little lakes, the shores and bottoms of which contain unlimited deposits of marl which is said to be of a finer quality than any heretofore discovered in this country, the analysis showing 96 per cent of carbonate of lime, with little or no trace of iron. Within a short distance of the marl beds there is to be had an abundance of clay, which is an essential in the manufacture of cement. Numerous other deposits of marl at Pine Lake, Fremont Lake, etc., exist not far off, elsewhere referred to. That of Fremont Lake is described by Mr. Hale on p. 135. Elk Rapids Portland Cement Co. Organized March 3, 1900 ; capital, f 400,000. Bonds issued in 1902 to improve machinery, etc., $100,000. Original actual cost of plant about $225,000, the balance of stock being issued for land or unsold and issued as bonus with bonds, which were floated at par. Officers: Schuyler S. Olds, president and general manager; Fitch R. Williams, vice president; Frank B. Moore, secretary and treasurer. Directors: Fitch R. Williams, attorney, Elk Rapids; M. B. Lang, merchant, Elk Rapids; Frank B. Moore, president Elk Rapids Savings Bank; Schuyler S. Olds, railroad counsel, Lansing, Mich.; Thomas A Wilson, attorney, Jackson, Mich.; C. A. Whyland, Chicago, 111. ; H. B. Lewis, manager Elk Rapids Iron Co. Within the limits of the village of Elk Rapids the plant of the Elk Rapids Portland Cement Co. has been erected. The company own a frontage of 80 rods on the shores of Grand Traverse bay, Sec. 20, T. 9 N., R. 9 W., and the plant was built at the water’s edge. The surroundings are far more picturesque than usually found accompanying a large industrial institution. In a grove of pine trees the various buildings were erected, and thrusting its arm out into the waters of the bay, a distance of 1,200 feet, is a substantial pier. At the end 16 feet of water is found, which al- lows the largest boats on the lakes to discharge and load. This dock is equipped with clam shell, hoisting engine, boiler, etc., and the cable dock car system for loading and conveying cargoes to and from the plant. As lake transportation is generally cheaper than rail, the company possess a decided advantage in this particular. Tracks of the Pere Marquette also run to the mills and the company uses both methods of transportation. Thirteen and one-half acres comprise the land owned by the company, upon which the plant has been erected. Two and one- LIST OF LOCALITIES AND MILLS. 245 half miles south of the plant site, in the extreme northern end of Grand Traverse county, is situated the marl lands of the company. This tract comprises 350 acres of solid marl. It was formerly a shallow lake (Petobago Lake, sometimes called Tobacco Lake, Sections 5 and 8, T. 28 N., R. 9 W.), about 20 feet above Grand Traverse Bay, but the company drained off the water and the marl is now very easy to raise and put into the dump cars of the com- pany. This great body of marl averages, in depth, throughout its extent about 18 feet. Very little muck or organic matter lies on top of this marl bed and it goes to the mill in a very pure state. They have also recently bought some limestone lands. Within a stone’s throw of the plant, clay of fine quality has been discovered. Besides this clay the company own a fine bed of shale clay (Antrim shale) on the east half of Sec. 3, T. 33 N., R. 7 W., on Pine Lake in Charlevoix County, also I am told in Lake Susan, Charlevoix County, and if needed the Watervale lands, No. 23, could be acquired. “The buildings of the company are quite extensive and are ar- ranged with the view of economically handling the materials as they pass from one process to another. The buildings comprise, frame coal storage building with cement floors, 50x175 feet, equip- ped with coal crushers, two elevators, two screw conveyors, rope drives. Concrete storage and packing buildings, 98x118 feet, con- crete floors and conveyors for handling the cement. The capacity of this building is about 30,000 barrels of cement. Machine and blacksmith shop of brick, 30x50 feet; this room is very essential in a cement plant as all necessary repairs can be made in a short space of time. This shop is equipped with all of the tools and machines necessary to perform a high class of work. “The engine and grinding rooms are in one building. This is of brick, 80x160 feet, with steel trusses, iron roof and cement floors. Steam for power is generated in two Sterling water tube boilers of 500 horse power and the motive power consists of a 500 horse- power Russel engine, with rope drives, also a Westinghouse dynamo and engine for the lighting plant. These are separated from the clinker room by thick walls. Four Griffin mills are re- quired to grind the clinker and in this room are clinker car con- veyors, cement conveyors and elevators. The rotary building is of brick set in cement and is 80x200 feet, with steel trusses, iron roofs, and cement floors. Here are found two pug mills, four tube mills, clay grinder, six large cement vats, ten steel slurry storage tanks, 12x16 feet each, and five Bonnet steel rotary kilns, 6x60 feet, lined with fire brick for burning cement clinker. The foundations of all machinery and all ground vats are constructed of solid concrete, resting on clay strata about eight feet below the surface of the ground. Besides these build- 246 MARL. ings there are laboratory, office, barns, boarding house and resi- dences on the ground and owned by the company. “To reach the marl beds the company have built a standard gauge railroad and the cars are propelled by a 35-ton locomotive. Economy in getting the raw materials to the mills has been sought and the operation expenses are very low. The road extends over the marl bed about half a mile and improved dredging apparatus is in use there. “All of the machinery was installed with a view of increasing the plant to 10 rotaries as soon as the occasion demands. As the marl is carried to the separating machine ‘it is weighed and then goes into the separator where all foreign matter is extracted; then the clay which has been finely pulverized is added in the quantity desired by the chemist and it goes into the pug mills and the mix- ing machines. After the most thorough grinding and mixing the correction tanks are reached and here the mixture is again ana- lyzed and corrected to the proper mixture desired to make a fine grade of cement. Through the rotaries the slurry rolls and as it leaves the far end it has been transformed into a small clinker. These clinkers are of irregular size. By means of an automatic conveyor this clinker goes to the mill to be ground into a fine powder. Test sheets are sent out with each shipment and the party receiving them knows just what he has purchased. The corps of cement makers and chemists have been carefully selected and every process of manufacture is carefully watched. A splendid system of tests has been inaugurated and any hour of the day test sheets will show just what results are being accomplished.” Wolverine Portland Cement Co. Coldwater Portland Cement Co., organized May 25, 1898 ; capital, $300,000. American Construction Company, Michigan Portland Cement Company, capital $2,500,000; organized June 30, 1898. This group of companies has had a somewhat varied financial history, but this has not prevented the steady production of cement under the “Wolverine” brand. The first company planned, the Coldwater, was a relatively modest affair with a capital stock of $300,000, $150,000 paid in with 640 acres of marsh land, a building to cost $100,000 and to have but 500 barrels capacity. Soon the plants and the capital were enlarged and the original company under the name of the American Construction Company took the contract of preparing the plant, turning in what it had done to the larger company, the Michigan P. C. Co., which issued $1,000,- 000 of bonds, covering the plant and lands, mainly in Coldwater and Bettrel Townships. In recapitalizing $100 in 6$ bonds were offered with every $100 of stock for $100 cash. When, therefore, in the fall of 1901 the interest failed to be paid on these bonds foreclosure proceedings LIST OF LOCALITIES AND MILLS. 247 began, and as a result of a compromise between the bondholders, which may be taken to represent the subscribing public, and the other creditors, prominent among which was the Construction Company, representing the promoters, the present company was formed. The officers of the Coldwater Company were: John T. Holmes of Detroit, president; L. W. Hoch of Adrian, vice president; George M. Conner of Detroit, secretary and treasurer. In the Michigan Portland Cement Co. W. L. Holmes became president, H. H. Hatch vice president, and John T. Holmes secretary, while Mr. Hoch remained manager for a while. The officers were later changed. This company, perhaps more than any other, brought the cement industry of Michigan into prominence by the thorough advertis- ing they gave it in placing the large amount of stock. They have two plants which were visited by Mr. Hale (page 105). One is at Coldwater, on the margin of Coldwater Lake, — a four- teen rotary plant, said to have cost $500,000, with a capacity of 1,500 barrels a day. The other is at Quincy, and the total capacity is said to be some 3,000 barrels. Some of the marl is said to run as high as 99$ CaC0 3 . The following are typical analyses of the raw materials fur- nished by the chemist, Mr. H. E. Brown: Light marl (dried). Blue marl (dried). Calcium carbonate 93.75 91.34 .77 .42 Magnesium “ 2.42 Soluble silica .18 Insoluble “ 1.01 .78 .55 Aluminum oxide .55 Ferric “ .25 .40 Sulphur trioxide tr. 1.84 .26 *5.79 Alkalies and rest (by difference,) Clay, light brown upper layer. Blue, 12ft. from surface. Silica 60.59 57.26 Titanium oxide .40 .82 Aluminum oxide 17.70 20.77 6.53 Ferric oxide 7.54 Magnesium oxide 1.46 2.31 Sulphur trioxide .41 1.34 Calcium oxide 1.50 Loss on ignition 8.74 6.19 Alkalies 3.16 3.28 • ♦Determined. 248 MARL . It is to be noted that all these samples are dried, the loss on ignition being to a slight extent C0 2 , but mainly combined water. It will be noticed that while neither specimen of clay contains much carbonate of lime, being derived from the Coldwater shales, which are practically free from it, the superficial clay has the least, and much less sulphur, showing that the calcium carbonate and pyrite have been leached out. Tests of this cement have been satisfactory, and are elsewhere given. “At various times while the dredges have been at work in the marl beds near the cement works, bones of prehistoric animals — or what are supposed to be the bones of extinct animals — have been exhumed. Chemist Brown, of the cement works, could not deter- mine to what age they belonged and has sent a box of them to Chicago for classification. The bones of elk and deer are fre- quently found, and one workman on the dredge has a beautiful pair of antlers that are as perfect and sound as though just taken from the animal .” — Coldwater Courier. Michigan Alkali Co., Wyandotte (J. B. Ford). This establishment originally came into Michigan in order to make soda for glass and other use, out of the vast beds of rocksalt which extend between Trenton, Michigan, all the way along the Detroit River and the Saint Clair River to Goderich in Canada, and Alpena, Mich. The limestone is obtained from the company^ extensive quarries at Bellevue but having yielded its carbonic oxide was of no farther use. To utilize this waste was a problem placed in the hands of their chemists. The plant was designed by the engineering firm of Lathbury and Spackman.* After several trials a hard burned, dark green clinker was pro- duced from the mixture of 100 parts of clay to 260 parts of waste, by weight, and tests showed the quality to be equal to the best of American or imported cement. The clay is dredged a few hundred feet from the company’s plant, and is conveyed to a stock shed; the clay is first run through a dryer, freeing it from water, then it is pulverized very fine and put into bins, then analyzed and properly proportioned by weight with the lime which has also undergone a purifying process. The raw materials are weighed exactly and the mixture is therefore abso- lutely correct. Afterward it passes through both pug mill and agitator and then ground. Proper proportioning of the raw materials is the most important factor in manufacturing a perfect cement and the product of the Wyandotte mill passes all chemical cement tests. ^American Engineering Practice, p. 110 to 118, with view plan and profile from which we take the description below. LIST OF LOCALITIES AND MILLS. 249 After leaving the rotaries, the clinkers pass through the finest of grinding machinery and all cement is passed through a mechanic- ally agitated screen, before passing into the bins, thus insuring a uniformly ground product. Here the cement is thoroughly sea- soned and none of it leaves the bins for the consumer until it is at least two months old. During 1900 and 1901, the city of Detroit used Wyandotte Port- land cement exclusively for all public work, which is in itself a fitting testimonial as to the efficacy of this superior product. To show the standing of Wyandotte cement in the market it is but necessary to mention a few of the buildings in which it has been exclusively used. In the mosaic floors and artificial stone walks of the new Wayne county court house 3,500 barrels of the Wyandotte Portland cement were used; 1,200 barrels in Brown Bros/ tobacco factory; 4,000 barrels in the engine foundations of the Detroit City Water Works building; 2,000 barrels in the Detroit Sugar Co.’s plant at Rochester, Michigan; hundreds of barrels in Wonderland Temple theater. Ten thousand barrels of Wyandotte Portland cement will be required in the construction of the great bridge across the Maumee at Maumee, Ohio, now being built by the Toledo Terminal railway company. The engineers, after careful tests of imported and American brands of cement, selected Wyan- dotte cement. “The plant was erected on the low lands bordering the Detroit river. High grade materials were used throughout and the process made practically automatic. The buildings, constructed of steel with brick sides, have clear spans, the trusses being carried on brick pilasters. As water and quicksand were discovered at two feet, the walls were built on brick arches which transferred the entire weight to concrete piers extending to solid ground. The mill building and stock house are parallel twin buildings, each with a 42-foot clear span; at the north end the roof is raised and 84-foot trusses span the width of both buildings, giving room for a second story. Adjoining the mill room but separated by brick partitions are the coal grinding, the engine and the boiler rooms. The clay building, with a 30-foot span, is of steel and corrugated iron, and runs toward the river at right angles to the mill building. The waste material is transported to the mill by a travelling crane, which, securing a charge in the soda plant, transports it into the second story of the cement plant. The clay, after excava- tion, is stored in a clay building, from which it is conveyed to an elevator, discharging into a rotary dryer, where it is subjected to the direct heat of a coal fire and afterwards passed through a disintegrator from which it is elevated to the second floor, and discharged into steel bins, ready to be added to the lime waste. The raw mix passes through a pug mill on the second floor, which discharges into a storage tank directly underneath. This tank is provided with agitators which prevent any separation by settling. From this tank the slurry flows to wet-grinding tube mills for a final reduction. These discharge into concrete pits, so arranged that a high lime or clay slurry can be discharged into 32-Pt. Ill 250 MARL. any of them to correct the chemical composition. After being analyzed and corrected if necessary, the material is pumped to steel storage tanks located in the second story. Agitators keep the slurry in motion in all pits until pumped into the rotary kilns. The material is fed to the kilns through water-jacketed chutes with pulverized coal; all three kilns discharge into a concrete pit, from which it is elevated to the cooling towers. Air is forced in at the bottom of these steel cooling towers, 12 feet in diameter and 22 feet high, arranged with a succession of metal floors, having radial openings, through which the clinker is swept by a scraper fitted to a central shaft. The clinker is moved 350 degrees on each floor, before falling to the next. Arriving at the bottom it is elevated into steel bins over the ball mills, from where it is raised and conveyed to bins over tube mills which finish the cement. From here the cement is elevated and conveyed by an overhead conveyor, through the mill room wall, into the stock house, and dis- charged into two lines of conveyors resting on the top of the storage bins, thus delivering into any bin desired. These bins have hoppered bottoms and are arranged in two rows with a passageway between containing two lines of screw conveyors; these carry the cement drawn from the bins to an elevator at the packing room, which discharges it into the bins supplying the packing machinery. The power plant consists of one 600 H. P. tandem compound con- densing engine, and three water tube boilers. The river water passes from jet condensers to hot well from which feed water for the boilers is taken. The engine is belted directly to the main line shaft which passes through the engine room walls in stuffing boxes, thus cutting out the dust from the mill; the engine room projecting beyond the walls of the mill so as to give clearance for main shaft. The shafting is so arranged that the power can be cut out from any department by the use of clutch couplings. A notable feature of the plant is the relatively small area covered by the buildings when compared with the total capacity, making it one of the most complete plants in operation. Including the stock house with a capacity of 40,000 barrels, all the buildings cover an area of only 25,000 square feet, and the plant has a daily average of 450 barrels. The above plants are those which were actually in operation in 1901. We take up next proposed mills which will in all probability be running before this report is out. The most extensive in plans will be the Hecla, which will be a group of allied industries, more like the Michigan Alkali Company last mentioned. The remaining three are Portland cement propositions pure and simple. LIST OF LOCALITIES AND MILLS. 251 Hecla Cement and Coal Co. Organized April 6, 1901. Capital $5,000,000, in shares of $100. A West Virginia corporation, but with offices in Detroit and busi- ness centering around Bay City, consisting of marl lands in Oge- maw county, and coal and clay shale lands in Bay county. The officers and directors of the company are: Julius Stroh, presi- dent; Cameron Currie, first vice president; Waldo Avery, second vice president; Edward H. Parker, treasurer; U. B. Loranger, sec- retary; Lem W. Bowen, Theodore D. Buhl, James N. W T right, M. M. Green. Briefly the plans of the company can be outlined as follows : The manufacture of Portland cement from dry marl and clay shale; the mining of coal, of which the lump will be marketed and the slack used in the manufacture of cement and the creation of power to run the great mills; the evaporation of salt in large quantities with the exhaust steam and hot gases escaping from the rotaries; the by-products of salt and limestone to be used in the running of a large chemical plant; the erection of coke ovens, also used as an auxiliary to the plants; the operation of a standard gauge rail- road to be utilized for hauling the coal to the dock of the company for lake shipment as well as the raw materials to the cement plant. The novel features of their plans are, — the transportation of the marl to the clay and shipping point, instead of building the factory at the marl bed; the use of waste coal and slack, and especially of Michigan coal, as well as clay and marl; the utilization of by-pro- ducts and waste heat, and the employment of a dry process. Ordi- narily the marl being the most bulky raw material, does not pay to ship. In this case, however, we have to counterbalance it a saving- on shipping coal, clay and cement, while the marl comes down grade. In the planning of the plant, marl analyses have been made by *R. E. Doolittle, State Analyst, Lathbury and Spackman, and others. To the courtesy of TJ. R. Loranger we owe details of the company’s analyses which cover a range of materials and have a scientific value in showing how analyses of such material run in the State. We append extracts from the reports of some of their experts. Beside the draining of the lakes and handling of the marl or boglime dry,* another important feature of this plant is the proposed utilization of shales of the coal measures. *Edwards Lake has been lowered. 252 MABL. One mile of river front on the Saginaw river, near the mouth, and only a short distance below West Bay City, is owned by the Hecla company, where the cement plant has been erected. The erection of this plant will shortly be followed by the other mills included in the general plan of development. The company owns about 6,000 acres of coal lands, about 800 acres of marl land, 2,000 acres of lime rock, and a mill site with nearly a mile of river front on the mouth of the Saginaw river, and is incorporated to manufacture and sell Portland cement, alkali, salt, paving and fire brick, coal, fire clay, etc. Experts who have looked over the property say that by reason of the fuel situation, with coal deposits under the company's mill site, it will possess a great advantage over those who are obliged to buy their coal in the open market and pay freight on it. The company will sell the lump coal and use its slack coal. The four marl lakes, known as George, Edwards, Chapman and Plummer, are located on the headwaters of the Tittabawassee River, and all within the radius of five miles in the township of Edwards, Ogemaw county, Michigan, Plummer being on the Hamp- ton branch of the M. C. R. R., and the others lying two, three and four miles respectively, from the same. There is a roadbed already constructed and in very fair condition, extending from Plummer Lake to Edwards Lake. Your next deposit, known as Crapo Lake, lies a little less than two and one-half miles northeasterly from the village of West Branch, on the Michigan Central railroad, in Ogemaw county, Michigan, and about six miles north of George Lake. Your Mills Lake deposit is located about four and one-half miles from the village of Prescott, on the Prescott branch of the Detroit & Mackinac railway, in Mills township, Ogemaw county. George Lake. The property at George Lake was found to consist of 380 acres, of which 200 acres are covered with marl. The marl is high quality, as shown by the following analysis, which is an average of samples taken from borings over the entire lake. Lab. No. 662 (see p.260). The chemical composition of the clay is shown by two average samples taken from the deposits as follows:* Lab. No. 712 (see p. 266 ). Lab. No. 713 (see p. 266). This lake presents probably the deepest deposit of marl of any owned by you, many borings showing a depth of from 27 to 34 feet, but the dry marl is thickly covered with a growth of small trees and brushwood, and a large portion of the deepest marl is *It will be noticed that these and all the other clays which are surface clays in connection with the marl deposits are about one-fourth to one-fifth carbonates, with generally 5 $ MgO. The company is depending not on these, but on shale clays of the coal measures. LIST OF LOCALITIES AND MILLS. 253 under water of considerable depth. The water in this lake could, however, be reduced by deepening the channel at the outlet, but it is a question whether a sufficient change of water level could be made without an expenditure of a considerable sum of money. Edwards Lake. Edwards Lake lies in a southwesterly direction from George Lake and is about three miles distant. It contains the largest acreage of marl of any of your deposits. The land owned by you here aggregates about 400 acres, of which 240 acres are covered with marl, of an average depth of 20 feet. The lands of this prop- erty are situated in sections 21, 22 and 27, and a second body is located about one mile eastward on the stream formed by the outlet. The clay deposits immediately at the outlet of the lake extends under the surface, and has been found by careful examin- ation to cover a tract one-half mile square, and is of good depth, though overlaid to some extent with sand and gravel. A second deposit further down the creek has been explored for about 40 acres, and shows a depth of 20 feet, at which point the bottom was not reached. The analyses of these two clays are as follows : Lab. No. 658 Edwards Lake, No. 1 (see p. 266). Lab. No. 676. Edwards Lake, No. 2 (see p. 266). The marls are also of most excellent quality, as shown by the following analysis, which represents an average of some thirty samples taken in various parts of the lake. Lab. No. 659. Edwards Lake Marl (see p. 260). Chapman Lake. Chapman Lake is in the extreme southwest corner of Edwards township, sections 31 and 32. The property owned by you here con- sists of some 230 acres, of which 160 are marl. Chapman Lake is fully equal in the quality of the marl to the preceding lakes, and partakes equally with Edwards Lake in the advantages resulting from being readily drained. An average analysis of Chapman Lake is as follows : Lab. No. 663. Chapman Lake Marl (see p. 260). The clay deposits examined in connection with this lake are located in section 7, Clement township, Gladwin county, about three miles distant from the lake. The bed is over 40 feet thick, and has been explored for a distance of over one-half a mile. An average analysis of this clay is as follows: Lab. No. 660. (Sec. 7, see p.266). Another clay deposit in section 3, same township and county was examined, an average analysis of which is as follows: Lab. No. 661. (Sec. 3, see p/266). 254 MAUL. Plummer Lake. Plummer Lake is the smallest of the group, but is advantage- ously located with regard to railroad transportation. The Haupt- man branch of the Michigan Central railroad passes through your property at this point. The lake is situated about seven miles west from the main line of the Michigan Central railroad. At this point the property owned by you comprises some 120 acres of land, of which about 40 acres are marl, which is of exceptional purity, and only a small portion of it covered with water. The clay on this deposit lies in direct conjunction with the marl at the east end of the lake, and runs down under the marl at the southern side. The clay deposit is covered with about three feet of surface earth and is 40 acres in extent. The clay average is eight feet in depth. An average sample of the marl shows the following analysis: Lab. No. 623. Plummer Lake marl (see p. 260). The analysis of the clay shows the best chemical composition for the manufacture of cement of any deposits examined in con- nection with the marl deposits, and is as follows: Lab. No. 675. Plummer Lake clay (see p. 266).* Crapo Lake. Crapo Lake is located on the east side of the main line of the Michigan Central railroad, about two miles northeast of the village of West Branch in West Branch township, sections 7, 8 and 16. The property comprises 340 acres, of which about 240 acres are marl. This deposit has an average depth of about 12 feet, and is covered with a light growth of grass and brushwood, with a top coat of muck six inches deep. The brushwood can very readily be burned off, while the level of the lake can no doubt be lowered con- siderably by deepening the channel at the creek outlet, and thereby exposing nearly all the deposit. About two-thirds of this body of marl occurs in the low swampy basin which was formerly covered with water. At the present time several narrow channels pass through the deposit with here and there a small lake, all of which drains into the west branch of the Rifle River. The marl in the small lakes shows a depth of at least 15 feet, while the water ranges in depth from two to fifteen feet. This deposit is entirely free from grit; analysis of samples shows it to be of uniform quality and containing a high percentage of carbonate of lime. Average analyses of samples taken from this lake give the following results: Lab. No. 891. Crapo Lake marl, No. 1 (seep. 260). Lab. No. 896. Crapo Lake marl, No. 2 (see p. 260). *See, as regards the availability of surface clays, pp. 267 and 268. LIST OF LOCALITIES AND MILLS. 255 The clay lands of the deposit are located along the bank of the lake, in sections 9 and 16, and cover about 40 acres, while the depth is about 30 feet. Analyses of samples of these clays give the following' results: Lab. No. 822. Section 9 (see p. 265',. Lab. No. 823. Average of section 10. Mills Lake. Mills Lake deposit, located in Mills township, in sections 24 and 25, on the east side of the main line of the Michigan Central rail- road, and about four miles from Prescott on the D. & M. railroad. This property covering 360 acres of land contains about 160 acres of marl. The main body of marl occurs in the lake under water, whose depth ranges from three to fifteen feet. The marl itself, has an average depth in the lakes of about 20 feet. At the north end of the lake a considerable part of the deposit of the marl is covered by water whose depth does not exceed three feet, and the entire lake level can be readily lowered by deepening the creek, and removing the log obstructions at the outlet. This will expose about three-fourths of the deposit. The marl in the lake is very uniform in quality, but in several spots is covered with a slight growth of vegetable matter; below this, however, the marl is of very great purity, having no topping or muck. An average analyis of samples taken from this gives the follow- ing results: Lab. No. 895. Mills Lake marl, No. 4 (see p. 260). The clay deposits in connection with this lake are located about one-half mile below the lake outlet, bordering both sides of the creek draining same and covering about 80 acres. It is over 30 feet in thickness, and an average analysis gives the following results: Lab. No. 904. Clay marked No. — 04 (see p. 263). Lab. No. 905. Clay marked No. 2 — 05. Samples of shale were taken from borings in four different loca- tions on your coal field which show an extensive acreage, running from five to fifty feet. The analyses of four samples of these shales are as follows: Lab. No. 727. Light shale (see p. 265). Lab. No. 728. Dark shale (see p. 265). Lab. No. 863. Goetz shale, No. 1 (see p. 265). Lab. No. 906. Clay marked 06 (see p. 263). These shales are all suitable for combining directly with your marl in the manufacture of the Portland cement; Lab. No. 727 and Lab. No. 906, being especially good. 256 MARL. The coal properties are taken up and discussed in detail by the report of Mr. Brown, superintendent of the N. A. Chemical Com- pany’s coal mines, and the report of Lippencott & McNeil, mining engineers. The raw material after being mixed, ground and burned in a set kiln, was ground, and the cement showed the following results : Lab. No. 921. Silica (Si0 2 ) 19.71* Alumina and iron oxide (A1 2 0 3 — Fe 2 0 3 ) 11.03 Lime CaO 64.25 Magnesia (MgO) 2.20 Sulphuric acid (S0 3 ) 1.42 The marl was taken from your property at West Branch, and the shale from your property at Bay City. In addition to the above analysis the finished cement was sub- jected to physical tests with the following results: CEMENT TEST. Fineness. Setting Test. No. 100 sieve 99.01$ Initial set 1 hr. 40 mim No. 200 sieve 84.70# Final set 6 hrs. 15 min. Gold Water Test Good. Hot Water Test Good. Tensile Tests. Neat 4-8 Hours. Briquette No. 14,870 257 lbs. Briquette 1 235 lbs. Briquette 2 225 lbs. Average 239 lbs. LIST OF LOCALITIES AND MILLS. 257 cement test. — Continued. Neat. Briquette No. 14,880 Briquette 1 Briquette 2 Briquette 3 Briquette 4 585 lbs. 675 lbs. 520 lbs. 640 lbs. 675 lbs. 7 Days. 3 to 1. Briquette No. 14,885 190 lbs. Briquette 6 155 lbs. Briquette 7 140 lbs. Briquette 8 225 lbs. Average Average Neat. Briquette No. 14,890 Briquette 1 Briquette 2 177 lbs. 619 lbs. 28 Days. 3 to 1. 7 80 lbs. Briquette No. 14, 895 236 lbs. 742 lbs. 755 lbs. Briquette Briquette 6 270 lbs. 7 242 lbs. Average 759 lbs. Average 249 lbs. Respectfully submitted, (Signed) LATHBURY & SPACKMAN. The following extensive suites of boglime analyses, which we owe to Mr. U. R. Loranger, are of especial value, as not select, but showing much better how an average deposit runs, than select analyses which are published in prospectuses. In regard to these analyses, however, as to many others, it must be remarked that probably what was directly determined was: calcium; magnesium; residue insoluble in Hpl, which is called sand and clay, or silica; iron oxide and aluminum oxide precipitated together; and sul- phuric anhydride. The calcium and magnesium are estimated as carbonates, and the difference between the total then and 100 per cent is called organic matter. But direct determination of the carbon dioxide shows as we have elsewhere mentioned, that it falls short of the amount calculated as sufficient to turn the calcium and magnesium into oxides by some two per cent. This is due to the fact that part of the calcium is combined with the sulphuric anhydride, and somewhat more with an organic acid (succinic acid). In practice, however, the calcium succinate would probably be soon broken up on heating into calcium carbonate and organic matter, so that it does not make much practical difference. 33-Pt. Ill 258 MARL. BOGLIME ANALYSES BY R. E. DOOLITTLE. Sample No. * 2 3 4 6 8 X 8 ft 9 E Calcium as Car- bonate 84.45 83.42 77.70 90.64 91.14 92.50 88.61 90.69 91.31 87.39 Magnesium as carbonate 2.76 2.03 3.43 2.30 2.62 0.39 2.25 1.58 0.34 1.22 Sand and clay (insol.) 7.37 7.89 14.25 2.04 2.25 1.01 5.33 3.87 1.63 4.99 Iron and alumi- num oxide 0.82 1.69 1.13 0.64 0.95 0.90 0.58 0.36 1.00 0.92 Sulphuric anhy- dride 0.60 0.83 0.48 0.47 1.00 1.97 0.45 0.43 1.91 1.70 Difference (or- ganic) 4.00 4.14 3.01 3.91 2.04 3.23 2.78 3.07 3.81 3.78 Totals 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 Sample No. or mark. Chap- man. Dun- ham. No. 10 No. 12 No. 20 No. 19 No. 21 Camp- bell Lake E.Side. Frost Dam. Plum- mer Lake. Calcium as car- bonate 89.86 85.17 86.95 90.01 89.44 89.92 89.66 87.24 63.14 87 25 Magnesium car- bonate 0.54 1 02 0.80 2.78 0.82 0.80 2.18 3.56 3.08 1.57 Silica 3.45 7.66 4.17 1.25 2.36 3.50 3.16 3.46 5 18.93 ( 6.28 3.54 Alumina 1.46 1.38 1.76 0.44 0.54 0.72 0.44 0.80 3.14 1.94 Iron oxide 0.30 0.32 0.40 0.36 0.34 0.30 0.42 0.40 0.62 1.08 Sulphur anhy- dride 1.78 2.34 3.52 2.08 2.98 2.22 1.92 2.30 1.76 2.26 Organic matter by difference. . 2.61 2.11 2.40 3.08 3.52 2.54 2.22 2.24 3.05 2.36 Totals 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 Sample No. 1 is from Edwards Lake, T. 21 N., R. 1 E., 11 feet thick. 2 is from Edwards Lake, “a little high in clay.” 3 is from Edwards Lake, “very high in clay with some sand, not a good sample.” 4 is from Plummer Lake, “a very good sample.” 5 is from Plummer Lake, “fair; rather high in sulphuric acid.” 8 is from Plummer Lake, “fair; rather high in sulphuric acid.” X is from Plummer Lake, “ a good, fair sample.” 8 feet, Chapman Lake, “a very good sample.” 9 Chapman Lake, “fairly good marl though a little high in sul- phuric acid.” E. Chapman Lake, “good, fair sample, a little high in sulphuric acid.” LIST OF LOCALITIES AND MILLS . 259 On the whole, the above set of analyses run high in silica for bog- limes, and especially in sulphur anhydride, — gypsum. The first analysis of the second set comes from Chapman Lake, like No. 9, of the previous set. The next comes from Dunham Lake, Sec. 19, T. 21 N., R. 1 E. No. 10 comes from Campbell Lake in the same township. Nos. 12, 20, 19, 21, and the rest of this set are all from this town- ship. All the above samples were analyzed by R. E. Doolittle, State Analyst, and as regards the amount of sulphuric anhydride, which is high, it will be noticed that the lakes are in a region just south of that where the Michigan series is bedrock, in which gypsum occurs frequently in the drift. Some of the limes highest in sul- phates are not high in iron, and the sulphates are probably not largely derived from pyrite. Silica is also often high. Of the sample at the First Dam, 18.93$ is soluble silica and fine sand, 6.28$ coarser sand. Calcium as carbonate. . . . 80.89 80.78 85.46 97.09 Magnesium as carbonate. 0.43 3.20 3.74 1.44 Silica 7.96 7.96 3.74 0 Alumina 3.74 1.76 1.88 Iron oxide 0.62 1.16 0.40 trace Sulphur anhydride Organic matter by differ- 1.94 2.51 1.28 0 ence 4.42 2.63 3.50 1.47 Totals, 100.00. Of the set above, the first is from Plummer Lake, the second is also (“A”). The third is from Campbell’s Lake, west side; the fourth from Plummer Lake, and all are by R. E. Doolittle. The first three have too much sand for cement, — better analyses are to be found in the other sets. There is more iron in the marls with sand and clay. In the pure boglimes it is only a fraction of a per cent. 260 MARL. BOGLIME ANALYSES BY LATHBURY AND SPACKMAN. Sample No. 662 659 663 623 891 896 892 893 894 895 Calcium oxide. . 58.28 51.44 50.83 52.38 49.45 50.75 48.74 49.47 49.37 50.43 Magnesium oxide. 1.22 1.23 0. 89 1.49 1.33 1.46 1.46 1.44 1.28 1.26 Loss on ignition 46.34 43.32 45.05 44.31 46.06 45.02 46.51 47.30 47.29 47.08 Calcium as car- bonate 93.35 91.85 90.76 93.53 88.30 90.62 87.04 88.34 88.16 90.05 Magnesium as carbonate — 2.56 2.58 1.86 3.13 2.78 3.06 3.05 3.01 2.68 2.64 Organic matter, (loss on ignition less C0 2 ) 3.93 1.56 4.15 1.52 5.76 3.55 6.62 6.87 7.10 6.08 Silica 0.72 3.14 2.23 1.78 1.64 1.46 2.45 1.08 0.97 0.70 Iron and alumi- num oxide 0.57 0.75 0.64 0.61 0.61 0.36 0.56 0.68 0.46 0.46 Difference +1.13 0.12 0.36 0.57 .91 0.95 .28 .04 .63 .07 Totals 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100-.00 Lab. No. 662 is an analysis of an average of samples taken all over George Lake, Sec. 13, T. 21 N., R. 1 E. ; of a property of 380 acres, 200 are covered with boglime. Lab. No. 659 is an analysis of the average of 30 samples from Edwards Lake, Sections 21, 22, 27, 28, of the same township. There is said to be 240 acres of marl that averages 20 feet thick. The outlet from the lake to the Tittabawassee has been cleaned out, draining the boglime. Lab. No. 663 is an average analysis from Chapman Lake in Sec- tions 31 and 32 on the same township. There is said to be some 160 acres of boglime here, also readily drained. Lab. No. 623 is an analysis of an average sample from Plummer Lake, near Sec. 8 of the same township, close to the Hauptman branch of the Michigan Central. There is only about 40 acres, but of very nice lime, and in conjunction with a clay deposit elsewhere analyzed. Lab. No. 891 is from Crapo Lake, Sections 7, 8, and 16, T. 22 N., R. 2 E. Here there are said to be about 240 acres of lime with an average depth of about 12 feet, covered with a top coat of peatmuck about six inches thick, and a growth of grass and brushwood. It is said that about two-thirds of the body of marl occurs in a low swampy basin, an old filled lake, now traversed by several narrow LIST OF LOCALITIES AND MILLS. 261 channels, with a small remnant lake here and there, the whole draining into Rifle River in such a way that much of it can be easily drained. Lab. No. 896 is from the same lake. Nos. 892 to 895 are Mills Lake marls, Nos. 1 to 4, Sections 24 and 25, T. 21 N., R. 3 E. Here there is about 160 acres of marl, mainly in the lake under from three to fifteen feet of water, but the out- let, it is said, can be readily deepened. There is a slight growth of vegetable matter over in some spots, but there is said to be no muck topping. The somewhat larger amount of organic matter in the Mills Lake and Crapo Lake analyses is attributed by Lathbury & Spackman, who collected and analyzed them, to the relatively superficial char- acter of the samples, and consequent larger amount of vegetable matter. ANALYSES BY R. C. KEDZIE. Clays, Sample No. 7B 36 Sand 1.10 0.95 Silicate of alumina 46.70 45.66 Carbonate of calcium 15.30 18.40 Carbonate of magnesium 2.63 1.50 Oxide of iron 8.15 6.90 Water . . . . . 25.00 24.00 Difference Totals, 100.00. 1.12 2.69 These analyses were of clay near & Spackman’s analysis. Plummer Lake; see Lathbury Marls sample. Mills. Crapo. Calcium as carbonate 84.50 88.57 Magnesium as carbonate 2.20 1.50 Insoluble (as sand, etc.) 1.00 2.00 Oxide of iron, etc Difference (water and organic .50 1.00 matter) Totals, 100.00. . . . . . 11.80 6.93 These two samples from Mills Lake and Crapo Lake respectively, should be compared with Lathbury’s analyses from the same place (891 to 896). The Crapo Lake analyses agree quite closely. The Mills Lake lime either contains more organic matter or is less dry. 262 MARL. CLAY ANALYSES BY F. S. KEDZIE. Sample No. Bay. 2. Plummer. I. Standish. 3. Silica 58.95 54.88 44.27 42.53 36.52 39.10 Aluminum oxide 14.45 6.80 12.86 11.12 8.93 12.38 Iron oxide 7.60 5.52 5.76 5.96 2.80 3.36 Calcium oxide 2.94 15.42 16.20 16.16 19.03 17.00 Magnesium oxide .86 5.50 6.62 5.97 7.26 3.10 so 3 1.73 2.62 ) 3.68 3.32 ) 2.92 ) Alkalies as K 2 0 2.54 3.46+ Comb, water Organic matter and loss 7.50 3.43 V 9.26+ [■ 10.61+ V 14.60+ 2.74 19.26* V 22.14 Total 100.00 100.00 100.00 101.40 105.69 100.00 t Difference. * C0 2 . % Manganese tr. With the exception of the first analysis, which is of a Bay County shale, and is a “good clay; it is entirely free from calcium carbon- ate, and is to be recommended for its content of silica and freedom from grit,” the rest are surface clays with the usual large amounts of carbonates, and considerable percentages of magnesia. A num- ber are, I believe, near Standish. No. 3 is from Plummer’s Lake. Compare Lathbury & Spackman's analysis 675, which runs much higher in silica. The percentages of calcium and magnesium as carbonates are as follows: Calcium carbonate 5.25 27.58 28.96 28.90 33.99 32.00 Magnesium carbonate . . . 1.80 11.51 13.84 12.50 15.25 6.49 7.05 39.09 42.80 41.40 49.24 38.49 In the Standish analysis in which the C0 2 is determined, it will be noted that the sum of the calcium oxide, magnesium oxide and carbon dioxide is but 45.55 per cent, while the sum of the carbon- ates as above given, is 49.24, which shows that not all the calcium and magnesium oxide are combined as carbonates, but some, espe- cially of the magnesia, probably are present as silicate. LIST OF LOCALITIES AND MILLS 263 CLAY ANALYSES BY R. E. DOOLITTLE. Sample No. 5 11 C “lift” Coarse sand 2.00 11.60 1.00 14.70 Silica 42.56 40.76 44.02 44.29 Alumina 9.47 10.05 13.36 9.00 Iron oxide 3.56 2.70 1.82 2.60 Calcium oxide . . . 15.15 14.80 17.28 14.45 Magnesium oxide. 5.95 7.45 2.60 6.26 Sulphur anhydride 1.06 1.73 2.36 1.50 Difference Totals, 100.00. 20.25 10.76 17.56 7.20 No. 5 is from Edwards township, No. 11 the same , but contains too much sand and gravel for cement making. 0 is the same in location. The other is from Chapman Lake. These clays are all surface clays, with 35 to 40 per cent carbon- ates, and a high but variable percentage of magnesia. Owing to the large amount of carbonates it would be necessary to use a large amount of clay, and it would be hard to keep the magnesium as low as desirable, or, I fear, the composition uniform. It is not intended to use any of these clays for cement manu- facture, though similar clays have been sometimes endorsed. CLAY ANALYSES BY LATHBURY AND SPACEMAN. Sample No. 1-04 2-05 06 02 03 01 870 867 849 Silica 39.34 35.12 65.24 44.60 40.76 48.88 48.52 54.06 51.40 Iron and al. oxide 15.93 1351 23.56 13.11 15.39 22.17 20.67 24.01 29.30 Lime 14.76 16.46 0.00 11.47 12.83 6.65 6.63 .12 .15 Magnesia 6.13 7.52 1.11 7.09 6.83 4.50 2.58 2.85 2.23 Loss on ignition 19.58 22.08 6.72 17.91 18.35 12.51 14.03 9.56 11.84 Difference (alkalies) 4.26 5.31 3.37 5.82 5.04 5.29 7.57 9.40 *5.08 Totals 100.00 100.00 104.99 100.00 99.20 100.00 100.00 100.00 99.21 including .79 sulphur. Lab. No. 904, field No. 1 — 04 is from Mills Lake, about half a mile below the outlet. The lime as carbonate would be 26.35 and the magnesia 12.86. The area is about 80 acres, the depth over 30 feet. Lab. No. 905, field No. 2 — 05, from the same place,. the lime car- bonate 29.39, and the magnesia 14.78. Lab. No. 906, field mark 06, from hole No. 11, on the Leinberger land, Frankenlust township, Bay County, Sec. 2, T. 13 N., R. 4 E. 264 MARL. The first two clays are like those analyzed by Doolittle, surface clays, about 40 per cent carbonates. Large quantities would have to be used of them, the amount of magnesia would be undesirably large, and it would probably be difficult to keep a uniform compo- sition. The next is a regular coal measure shale clay, and would probably be a valuable paving brick clay, as well as suitable for cement. Lab. No. 897, field No. 02, from Michigan Clay Co., Frankenlust township, Bay County, in the northeast part. This is a surface cal- careous clay, properly called marl, the lime would be 20.48 as car- bonate, and the magnesia 14.82, or over a third carbonates, and the remarks above upon surface clays apply. Lab. No. 898, marked 03, is from the Williams Clay Co., just north, and is an entirely similar surface clay, with 22.91 per cent calcium carbonate and 14.30 per cent magnesia carbonate. Lab. No. 887, marked 01, is from Everett’s at Corunna, and is also a surface clay with a considerable amount of carbonates, although perhaps because it is farther from the outcrop of the Eocarbon- iferous Limestones, decidedly less, namely, 11.87 of calcium carbon- ate and 9.50 of magnesium carbonate. It is also said to have no sulphates, which is rather remarkable! It is probably derived largely and not very remotely from a coal measure shale clay, like the following analysis. Lab. No. 870, also a Corunna clay, but with much less of carbon- ates, so much so, that it can hardly be a surface clay. Both this and the previous analyses are remarkably high in iron and alumina, but Prof. Campbell of the University of Michigan got similar results for clays of this district, which are 15 to 20 feet thick, have little sand and occur on high ground directly over shale, to wit: about 48 per cent silica, 16 of alumina, and 5 of ferric oxide. In some cases of very fusible shale there was as much as 25 per cent alumina. Lab. No. 867 is a clay from south of Tawas City, in Iosco County. The form of the analysis indicates that like the analyses of pp. 40 and 41, in Part 8, it is practically of a shale of the Michigan series. The high per cent of difference undetermined is probably sulphates (gypsum) as well as alkalies. There is, however, some uncertainty about this sample. LIST OF LOCALITIES AND MILLS. 265 Lab. No. 849 is from a boring one mile north of Goetz farm, Sec. 36, Monitor township, Bay County, T. 14 N., B. 4 E. The lime is remarkably low in proportion to the magnesia. There is some pyrite (0.79 sulphur) and the large loss on ignition and large amount of alumina and iron are noteworthy. It should be readily fusible. This is not at all of the fire clay type. CLAY ANALYSES BY L. AND S.— CONTINUED. Sample No. 814 815 725 726 727 728 863 816 822 823 Silica 55.06 47.83 37.75 42.71 61.13 54.93 41.38 39.81 43.53 41.00 Iron and al. ox- ide 30.53 35.21 13.13 14.92 26.90 31.43 27.02 18.57 14.71 17.19 Calcium oxide. . 0.12 0.14 17.04 13.72 .12 .22 .52 3.74 12.69 12.79 Magnesium ox- ide 1.47 1.19 6.88 6.36 6.47 1.58 .90 5.20 5.65 5.68 Loss on ign. (or- ganic matter and C0 2 , etc.) 7.47 10.09 29.20 22.29 6.47 7.41 23.11 18.22 17.89 18.39 Difference ( al- kalies, etc. ) . . 5.35 5.54 4.42 5.43 7.07 4.45 5.53 4.95 No. 814, St. Charles shale, No. 1, is a coal measure shale of the fusible variety. No. 815, St. Charles shale, No. 2, is similar but even lower in silica. The lime, it will be noticed, is extremely low. No. 725 is another surface clay with similarly high per cent of carbonates (30.46 calcium carbonate + 13.07 magnesium carbonate) and low silica. No. 726 is a similar surface clay; it is from Sterling, not Standish, No. 2, S. W. Arenac County, not far off. Calcium as carbonate is 24.56 and magnesia 12.09. No. 727 is a light shale from the Bay County coal field. No. 728 is said to be a dark shale from the same field. The iron must contribute with the organic matter to the darker color. No. 863 is a shale from the Goetz land, Sec. 36, Monitor township, a coal measure shale. With the low amount of lime and magnesia characteristic of these shales, the large loss on ignition shows much organic matter (black shale), and it will be readily fusible. No. 816 is from the Prairie farm, and I think the same deposit as No. 18 of Part I, though I cannot account for the discrepancy in silica. The lime as carbonate would be 24.38 and the magnesia 9.90, — about the usual 35 per cent carbonates of the surface clays. No. 822 is a surface clay from Crapo Lake, T. 22 N., R. 2 E., an average of Section 9. 34- Pt. Ill 266 MARL. No. 823 is from the same locality, an average of Section 10. The former has about 22.66 per cent calcium carbonate and 11.65 magnesium carbonate, and the latter has 22.83 per cent calcium carbonate and 11.70 magnesium carbonate, or as usual, about one- third carbonates. CLAY ANALYSES BY L. AND S.— CONTINUED. Sample No. 643 (?) 712 713 658 676 660 661 675 Silica 41.54 41.58 43.35 42.95 44.69 47.59 44.40 39.26 52.75 Alumina 13.15 9.96 14.43 14.98 9.90 10.20 9.54 13.32 11.34 Ferric oxide 4 83 4.20 4.29 2.71 2.71 4.57 Calcium oxide 13.93 15.02 13.58 143.73 12.96 13.75 14.04 14.21 12.68 Magnesium oxide 5.45 6.36 5.71 5.84 5.98 5.15 6.09 0.94 4.75 Loss on ignition (water and organic) 17.92 19.03 17.71 16.38 6.09 18.83 14.83 Difference (alkalies. etc.) 3.18 3.85 22.93 22.50 4.47 6.93 23.22 8.87 3.67 Totals 100.00 100.00 100.00 230.00 99.91 102.71 106.09 100.00 100.02 Lab. No. 643, clay 3 — B, is a surface Bay County clay, with 24.87 calcium carbonate and 11.39 magnesia carbonate. The next sample is similar, but has even more carbonates, — 26.82 of calcium carbonate and 13.30 of magnesia, over 40 * in all. Lab. No. 712 is a clay from near George Lake, T 21 N., B. 1 E., — also a surface clay with 24.25 calcium carbonate and 11.96 mag- nesium carbonate. Lab. Nos. 658 and 676 are both near Edwards Lake in the same township, the former from a clay bed “directly at the outlet of the lake in Sections 25 and 27, — and has been found to cover a tract one half mile square, and is of good depth, though underlaid to some extent with sand and gravel.” This has 23.14 calcium car- bonate and 12.54 magnesium carbonate. No. 676 comes from a deposit about one mile east and down the outlet stream, of over 40 acres area and over 20 feet depth. This has 24.57 calcium carbonate and 10.80 magnesium carbonate. Lab. No. 660 comes from near Chapman Lake, Sec. 7, Clement township, Gladwin County, T. 20 N., R. 1 E., — a bed “over 40 feet thick, explored for over half a mile.” Here again we have 25.07 calcium carbonate and 12.78 magnesium carbonate. Lab. No. 661 comes from Sec. e, near by and has 25.97 calcium carbonate and only 1.92 magnesium carbonate. This is the lowest in magnesia of any of the surface clays. LIST OF LOCALITIES AND MILLS. 267 Lab. No. 675 is of a clay from Plummer Lake, “in direct con- junction with the marl at the east end of the lake, and runs down under the marl at the southern side. The clay deposit is covered with about three feet of surface earth and is 40 acres in extent,” averaging eight feet in depth. Taking the set all together, we see that these surface clays rarely run less than 35 per cent or over 45 per cent of carbonates, but that the amount of magnesia varies materially, though it is usually over a third of the carbonates. The uncertainty as to whether the percentages will remain uni- form through a deposit, and the difficulty in getting a satisfactory analysis from the resulting cement, are what make the surface clays not, except possibly in rare instances, to be recommended for the making of cement. All these clays will fuse readily. 12. The Great Northern Portland Cement Co. Organized 1899, capital, $5,000,000, in 50,000 shares; there was $2,000,000 preferred stock bearing Tf> interest, the balance common. In selling the preferred, a bonus of one-half share of common was given. Located at the company’s village of Marlborough, two miles south of Baldwin, on the Pere Marquette R. R., Secs. 14 and 15, T. 17 N., R. 13 W., Lake County. They also own or control about 6,200 acres in the neighborhood, about 3,500 of them with bog lime, to wit : Sec. 13, T. 16 N., R. 13 W., lakes and lime; Sec. 15, and most of 21 and 10, T. 17 N., R. 12 W., also the N. % of the N. E. %, Sec. 18, sur- face clay, especially on 10, and marl; N. y 2 of N. W. 14 of Sec. 36; N. % of Sec. 34; Sec. 27; Sec. 14; E. % of Sec. 15 and part of Sec. 10 all in T. 17 N., R. 13 W. Also parts of sec- tions 23, 26, 27, 34, 35, T. 18 N., R. 13 W. ; Sec. 27 and parts of 26, 34, 35, T. 19 N., R. 13 W., all marl lands. There are in all some 17 lakes, generally with low shores, surrounded with a rim of marsh or tamarack growth underlain with marl. One of these lakes is said to have water from one to six feet deep with marl from 20 to 58 feet deep. In general the deposits are 18 to 20 feet deep. The first unit, now nearly complete, to which one or two more may be added later, has a capacity of 4,000 barrels a day, with 24 rotaries. 268 MARL . The following is Booth, Garrett & Blair’s report: Great Northern Portland Cement Co., 82 and 84 Griswold street, Detroit, Michigan. Gentlemen : — Following your visit to this city in September, we received in due course your letter of instructions to make a thorough investi- gation of your cement property in Lake County, Michigan, and agreeably therewith, our Mr. Whitfield made an extended examina- tion of the property, taking a large number of samples, gauging and locating the deposits and establishing their quantity and acces- sibility. Since his return, we have made analyses of the samples selected, and have burned three lots of cement from suitable mixtures of these samples and subjected these cements to analyses and to numerous physical tests. All this data is now in your hands in a series of preliminary reports. We are now prepared to render final report on the broad project which you have in view, to wit: the construction of a modern plant of large capacity for the manufacture of Portland cement, near Baldwin, and this report follows: Raw materials. — Regardless of large tracts of land which you have since purchased, we find that the property examined by Mr. Whitfield contains deposits of raw materials suitable for high grade Portland cement, and in sufficient quantity to supply a large plant for many years. These raw materials are white shell marl and blue clay. The clay “D,” used by us so successfully in making cements, is found in immense quantities on Sec. 10, T. 17 N., R. 12 W. This locality is shown in relation to the marl tracts in the map attached to this report. Cement. — An expression of opinion on the quality of raw mater- ials for cement may in some cases be quite sufficient, but will never be so convincing to practical minds as an actual test. For this reason we have burned from your raw materials three successive lots of clinker with increasing percentages of lime, and have tested the cements with results as follows: LIST OF LOCALITIES AND MILLS. 269 Samples. No. 1. No. 2. No. 3. 64% 660 lbs. 614 “ 670 “ 636 “ 628 “ 65% 710 lbs. 790 “ 672 “ 820 “ 890 “ 66% 930 lbs. 974 “ 924 “ 952 “ 946 “ 7 day tests of neat briquettes Average 641 lbs. 776 lbs. 945 lbs. 7 day tests 1 cement to 3 sand 265 lbs. 282 “ 270 “ 262 “ 250 “ 396 lbs. 370 “ 370 410 “ 350 “ 460 lbs. 424 “ 432 “ 448 “ 420 “ Average 266 lbs. 379 lbs. 437 lbs. Samples. No. 1. No. 2. No. 3. Nominal lime content 64% 3 03 1 °50' 3°40' 65% 3.04 2° 10' 4° 40' 66% 3.08 1°25' 4°0' Specific gravity Initial set Final set ANALYSIS. Silica Alumina Iron oxide — Lime Magnesia ». 24.01% 24.84% 23.87% 5.51 % 4.51% 4.82% 2.38% 1.74% 2.30% 63.91% 65.69% 66.01% 3.40% 3.10% 3.03% 28 day tests were made from lot No. 1 with results as follows: Neat. 1 cement to 3 sand. 28 day tests, sample No. 1 952 lbs. 918 “ 470 lbs. 450 “ 980 “ 486 Average 950 lbs. 469 lbs. Very truly yours, BOOTH, GARRETT & BLAIR. The clay bank in Sec. 10, T. 17 N., R. 12 W., covers some 400 acres. From the amount of magnesia in the finished cement, it would seem that there is as usual, some six or seven per cent of magnesia in the clay, though I have seen no series of analyses of it. This rises as a hill of clay, a bit of glacial deposit, some 90 270 MARL. to 150 feet high, in the midst of prevailing sand and gravel. It is said to be free from grit. Gravel for concrete was found in the excavations for the plant. In the beginning, the lime of North Lake, right by the mill, will be used, and the clay shoveled by steam shovel and transported in special cars, and the marl in scows. Prof. R. C. Carpenter reports in part, as follows: “I find that the marl exists as is represented, and is found in a great number of lakes and surrounding marshes, occurring to a depth varying from 20 to 70 fee.t. The marl in every case is of excellent quality and free from any material which would inter- fere in the manufacture of cement. The clay deposit is located a short distance from the center of the marl deposits and the site of the w r orks. The clay has been thoroughly tested by reputable chemists, and is found to possess all desirable qualities required, both as shown by analysis and by actual trial in the manufacture of cement. The clay deposit is of almost unlimited magnitude and would supply the plant for more than a century, even when work- ing on a scale of 12,000 barrels per day. The examinations of the deposit have convinced me that the materials are all that has been claimed, both as to quality and quantity.” 13. Detroit Portland Cement Co. Organized March 7, 1900. Capital, $1,000,000. From the Fenton Independent of March 31, 1900, comes the fol- lowing item (see Plate XXI and Fig. 23) : Deals for marl land on which the Becker Bros, hold options in Fenton township are being closed up. The lands embrace the marl on 110 acres of the McKugh farm at Mud and Silver Lakes, and the marl on 74 acres on the Beals farm at Silver Lake, and the marl on 89 acres on the Latourette farm on Mud Lake. Only the marl rights are purchased, the price paid Beals being $2,500, and the price paid Latourette being $800. The marl rights to 30 acres of the Tunison farm at Silver Lake were purchased for $900. This factory is built on the line of the Grand Trunk. It is to be an eight rotary plant, with provision for enlargement (com- pare Plate Y). The plant is designed by Lathbury & Spackman, and illustrated in their work, “Engineering Practice,” already referred to so often. Their description is as follows : The plant of this company, located at Fenton, Michigan, is now nearing completion. The mill is designed to manufacture Portland cement from a mixture of marl and clay by the wet process, and possesses some distinctive features not embodied in the marl plants heretofore erected. The buildings substantially constructed of brick and steel, are fire-proof, and so designed that the material in process of manu- facture will move in one direction from the time the raw materials LIST OF LOCALITIES AND MILLS. 271 are brought in at one end, until the cement is shipped out from the packing house at the farther end. All the buildings have clear spans. The mill is located on a slight elevation overlooking the large marl deposits of Mud and Silver Lakes, owned by the com- pany. The clay is obtained from pits a few miles distant from the plant. The marl is dredged from the lakes and loaded into cars of two cubic yards capacity which run on a track along the edge of the laks. The cars are then drawn by cable hoist up an inclined trestle into the mill and dumped into the hopper over the stone sep- arator. The clay is brought into the plant by rail. The two ingre- dients after passing through separate preliminary preparation are mixed together in the proper proportions and ground in tube mills. Large concrete storage pits contain the marl and clay before mix- ing, and similar pits are provided for the mix and ground slurry. It has heretofore been the practice to pass the clay through some suitable dryer, then after it has been ground to an impalpable pow- der, to mix this powdered clay with the marl. In this plant, how- ever, the clay is unloaded directly from the cars into a disintegrator, from which it discharges into a pugging conveyor which carries it to a wash mill, where it is reduced to a thin sludge. The marl passes, first, through a stone separator which reduces it to a smooth plastic state and removes any roots, grass or stones which may have been brought up by the dredge bucket. It is then stored in the marl pits. Each pit is provided with an agitator to prevent settling. The marl and clay are pumped to the mixing pits in proper proportions and thoroughly agitated. From these pits the raw mix is pumped to iron tanks above the tube mills, from which it is fed to the mills by gravity. After being ground in the tube mills, the slurry is discharged into concrete storage pits which supply the kilns, the slurry being pumped to a stand pipe from which it is fed at a constant pressure directly into the kilns. After passing through the kilns, of which there are eight, the clinker falls into air-tight, self-emptying concrete cooling vaults, located below the kiln room floor and directly under the discharge from the kilns, two vaults being provided for each kiln; the Lathbury & Spackman patent regenerative clinker cooling apparatus being used. Cold air is drawn in through openings in the bottom of these vaults, and passing upward through the clinker cools it. The hot air being exhausted from the top is forced into kilns mixed with pulverized coal, thus utilizing the heat contained in the clinker for burning. The clinker is drawn out at the bottom of the vaults into Cars which run on tracks located in the tunnel below the clinker cooling vaults. These cars are run out of the tunnels and raised by an electric lift to the level of the top of the bins feeding the clinker ball mills, and the clinker is discharged from the cars into these bins. After passing through the ball mills, the partially ground clinker is elevated and conveyed to the bins supplying the tube mills. From these mills it is elevated and conveyed to the stock house and distributed in the bins. The stock house is equipped with Lathbury & Spackman self-discharging bins, described else- where in detail. 272 MAUL. Conveyors in the tunnels of the stock house carry the cement to the packing room, located at the extreme end of the building, and deliver it to the bins over the packing machine. The packing department, fully equipped with both barrel and bag packing machinery, has a capacity of 1,500 barrels of cement per day. The power house, located close to the main building, is equipped with four 200-horse power vertical water tube boilers. Two 500 horse power compound condensing engines, direct connected to two 300 K. W. direct current generators are located in the engine room. An auxiliary 150 K. W. direct connected dynamo and engine is provided to furnish current for lighting and power when the plant is operating under light loads. The power plant is completely equipped with the usual accessories, such as switchboards, pumps, condensers, etc., and special attention has been paid to securing economy in the generating of power. The entire plant is electric- ally driven, the motors being distributed throughout the plant, each machine being belted direct to its own motor. ANALYSIS OF THE CLAY AND THE MARL. Marl. Clay. Silica, SiCL .96 54.70 18 80 Alumina, ALO3 \ -44 52.43 Iron, Fe20:} .". Lime, CaO 7.17 Magnesia, MgO , 1.66 3.87 Carbon Ioxide CO-> 42.99 1.52 9.80 .96 Difference Total 100.00 100.00 14. Egyptian Portland Cement Co. Organized June 30, 1900. Capital, $1, 050, 000, in $10 shares. Also bonds, $350,000. The officers are, George A. Foster, president ; J. Fletcher Williams, vice president and general manager; C. B. Shotwell, secretary, and E. D. Kennedy, treasurer. The factories are at Fenton and Holly. Robert W. Hunt & Co., are engineers and W. H. Hess, chemist. We reprint many of the careful surveys which were made of the company’s lime lakes. One (Plate XXI) is of Silver Lake, the Fen- ton property, and another, (Fig. 21) is Raffelee Lake, the Holly property of the same company. In Plate XXI the bluffs which mark the original margin of the lake are shown as in Fig. 13, and if we compare the outline of the lake with that shown in the county atlases from the original land office surveys, we find it entirely different. Apparently a good deal LIST OF LOCALITIES AND MILLS. 273 of this is due to the filling* up of the lake by the deposits of boglime, isolating “daughter lakes,” as Davis has described them, from Littlefield Lake. It is possible, however, that a change of lake level -* DF < s ~ MAflL LA/VJJ As Sun/eyeMSanykc) Robert W Hun t& Co CB/cayo /m JcaZe. '/cn /ty>rov*B may also have been an important factor. Finally, but not least important, the surveyors in meandering these marsh bordered lakes, which are often full of rushes, find it very difficult to deter- mine where marsh ends and lake begins. We also reproduce reduc- 35-Pt. Ill 274 MARL. tions of careful surveys of Runyan Lake, Sections 9 and 10, T. 4 N., R. 6 E. (Fig. 22), and of Mud Lake, just north of Silver (Fig. 23). Also of lakes on sections 27, 28 and 30 and 29 of Holly township (Figs. 24 and 25). There is peat in connection with these deposits “partially over- lying and directly contiguous, which it has been proposed to use as fuel, though it is not at present seriously planned. The coal and very probably the shale will come from the neighborhood of Cor- unna. The Grand Trunk and the Pere Marquette system cross at Holly. A resurvey after some years, of such of these properties as may not have been seriously touched, will give important light on the growth of the deposits. Extracts from the prospectus, Robert W. Hunt & Co.’s report, are as follows: Report dated Jan. 30, 1900. We beg to submit the following report in full on the survey and investigation of the marl lands situated near the cities of Fenton and Holly, Michigan. io B orsnfs /or-AlCi lyiii Fig. 22 . Runyan Lake. T. 4 N., R. 6 E., near Fenton. LIST OF LOCALITIES AND MILLS. 275 The marl land surveyed and sampled consisted of four separate deposits. The first and largest, is in the southeast corner of Gene- see County, two miles west of the town of Fenton, and extends south into the northern part of Livingston county (Plate XXI). The second is in Oakland County, two miles east of Fenton, and about midway between Fenton and Holly. The third deposit is in and north of the town of Holly. The fourth deposit is about two miles southeast of Holly on Raffelee Lake (Fig. 21). The first tract consists of Runyan Lake (Fig. 22), Marl Lake, Upper and Lower Silver Lake, a part of Mud Lake (Fig. 23), Squaw Lake, and the low swamp land contiguous to these lakes, together with a strip of land in the town of Fenton. As a rule the hills sur- rounding these lakes are high and steep, and the slope of the marl 276 MARL . deposit is quite abrupt, which latter is also true of the lake bottoms. Many bars of marl, covered with only a few inches of water, extend into the lakes, but just off these bars the water is deep. The second tract (Fig. 24) consists of marsh land around Warren Lake and several small ponds near by, Dickson Lake and the two Mineral Lakes. The hills around these are also high and steep and the shores are abrupt. Fig. 24. Warren, Dickson, Mineral and adjacent lakes and marl beds. Sections 29 and 30, T. 5 N., R. 7 E. The third tract (Fig. 25) is in and around Bevin Lake and Bush Lake. There are no hills around these lakes, and the marl deposit is shelving, and the shores are not abrupt. A large part of Bush Lake is only a few feet deep. There is no tamarack or underbrush. The fourth tract is along the south edge and west end of Raffelee Lake, including the swamp lands just west and northwest of Raffe- lee. Part of this swamp land is heavily timbered, and the average stripping is about two feet. The first tract is cut by three highways and the Detroit, Grand Haven & Milwaukee railroad track, together with the public road which lies between Silver and Mud Lakes. Another road is just south of Silver Lake, and still another south of Marl Lake. LIST OF LOCALITIES ANI) MILLS. 277 There are no public highways crossing the second tract, but the main highway between Fenton and Holly runs very close to it. Between Bevin and Bush Lakes are the tracks of the Pere Mar- quette railroad, a public highway and some meadow land. The Detroit, Grand Haven & Milwaukee railway runs alongside of Baffelee Lake. There are no highways crossing this tract, but it will probably be easy to secure one on the section line. The maps which we send you will show the location of these different tracts. There are eight detail maps, which show all lands surveyed and sampled, except where the results were not good enough to justify mapping the properties out. These maps show location of property, name of original owner, and location of test holes from which marl samples were taken. The numbering of these test holes is the same as the sample numbers in the com- plete analysis. In determining the extent of the deposits, about four hundred additional test holes were sunk, from which no samples were taken. 278 MARL. The following statement shows total acreage: Examined. Tract No. 1 Tract No. 2 Tract No. 3 Tract No. 4 1569.7 acres Considering the results obtained from the chemical analysis of the marl, lots or deposits of marl have been located wherein the marl, as shown by the analysis, is of such composition as is re- quired to make good cement. The total amount of marl in the foregoing lots, upon which we report favorably, is 14,350,720 cubic yards, which is enough to manufacture about 28,700,000 barrels of cement. The following tables show the maximum, minimum, and average determination of the samples from the accepted lots, together with the average depth of marl, quantity of stripping, and quantity of marl in each lot. Sampled and Mapped. 976.0 acres 190.6 acres 163.6 acres 239.5 acres j LIST OF LOCALITIES AND MILLS. 279 Lot 11. 84.51 J 92.86 1 78.21 3.60 j 4.54 1 3.18 1.035 j 1.51 ) .70 2.014 j 3.02 1 .70 20. 54.710 812, 300 Lot 20. 89.82 j 92.18 1 86.63 3.23 j 3.99 1 2.56 1.57 j 3.12 1 .58 1.60 j 3.04 1 .28 15. 51,000 588. 220 Lot 10. lO CO • o o TfiOiT* ^ o to to SO O Cl CCCH i>CCC0<0 ^coi^ CO^Oi • CO ^^H*^COOi 05 Oi CX) OJ CO Lot 19. 89.586 j 92.99 1 84.40 3.038 j 3.85 ) 2.07 .93 5 1.86 \ .04 0.88 j 2.06 ) .22 27. 330. 000 4, 513. 250 Lot 9. 88.57 j 92.86 j 83.57 3.276 J 3.69 1 2.82 1.22 j 1.44 1 1.04 2.926 ) 5.44 i 1.28 16.4 None. 511,200 Lot 18. 85.342 J 88.09 1 82.34 3.765 j 4.34 ) 3.25 .860 j 1.5(1 1 .40 1.924 j 2.44 1 1.08 12.5 None 325, 460 00 o J 87.57 J 90.71 | 80.89 3.84 j 4.27 j 3.57 .92 J 2.00 | .54 1.411 j 2.54 1 .84 11.8 61 ,38) 533, 300 Lot 17. 84.686 5 85.47 l 83.24 3.546 j 4.22 ) 3.14 1.992 J 3.84 1 .54 2.44 j 3.22 | 1.18 11.8 4,360 285, 830 Lot 7. 89.633 j 92.1 j 86.61 3.902 j 4.35 ) 2.70 1.049 J 2.72 1 .56 .966 j 1.92 1 .64 13. 7. 600 1,280, 340 Lot 16. 85.266 j 87.00 1 82.43 3.440 j 3.90 1 3.64 2.57 j 4.54 1 .84 1.695 ) 2.38 1 1.01 11. 32, 750 105, 930 Lot 6. 87.92 J 92.86 | 83.75 3.16 j 3.48 j 2.66 .828 j 1.18 | ' .58 1.415 j 1.96 1 .72 23.4 4,000 880, 420 Lot 15. O CO '£> O O • ?ocoo t- co os m • i> ooin o-^eo a'tinwooic • OS 50 5Q TP ~ -s’ ' -H * 00 eo to ; 00 00 ^ t- Lot 5. 90.31 ( 94.64 | 85.89 3.072 j{ .765 j 1.46 1 .30 .959 j 2.82 1 .48 16.9 ft. 77, 280 849, 600 Lot 14. 83.976 ( 88.14 ) 79.48 2.676 \ 3.56 1 1.66 1.446 j 2.30 1 1.00 1.29 j 1.59 | 1.04 17. 32, 980 195, 060 Lot 3. 89.03 j 95.00 1 85.71 34.303 j 3.74 1 2.82 1.008 j 1.24 1 .82 .994 J 1.62 1 .40 13.6 ft. 70, 340 183. 780 Lot 13. 86.54 ) 88.13 | 85.20 3.39 j 3.82 ( 3.02 1.38 ( 2.48 \ .90 1.760 j 2.94 1 .98 11.8 47, 780 225, 500 Lot 1. 89.717 j 92.19 | 85.19 3.408 5 4.76 1 1.97 .895 < 3.66 1 2.24 .708 ) 1.20 1 .24 22 6 ft. 153, 920 1,879, 260 Lot 12. 88.48 J 90.71 1 86.79 3.13 j 3.63 | 2.24 .57 1 .78 < 16 1 000 j 1.54 1 .68 15. 75, 000 365, 100 Calcium carbonate. Average Range Magnesium carbonate. Average Range Iron and Alumina. Average Range Silica. Average Range Average depth feet Stripping cu. yds Lime cu. yds Calcium carbonate. Average Range Magnesium carbonate. Average Range Iron and Alumina. Average Range Silica. Average Range ’.... Average depth Stripping cu. yds Lime cu. yds Total stripping cu. yards, 999,290. Total lime cu. yards, 14,350,720. 280 MABL. The best locations for cement plants are upon the Grand Trunk railway, between Silver and Mud Lakes at Fenton, and upon the same road at Raffelee Lake, just east of Holly. At the latter point the Pere Marquette system would doubtless be glad to build a switch into the plant, giving it the benefit of junction point rates, which could probably be extended to include the Fenton plant as well. From the chemical analysis of marl, its desirability for the man- ufacture of cement is determined. The analysis also gives data for determining the amount of clay that should be mixed in order to give good results. A large percentage of silica is not desirable, but four to five per cent is not prohibitive, providing it does not vary to too great an extent. The amount of iron and alumina oxide that is detrimental depends upon the analysis of clay with which the marl is to be mixed. The magnesium carbonate should not be over four to five per cent, which, of course, will be reduced in the finished cement between two and three per cent. If the amounts of silica, iron and alumina, and magnesia in a body of marl are small, a comparatively large variation in the calcium carbonate can be allowed, because its percentage will vary almost directly as the amount of organic matter. We would respectfully recommend that all material possible be conveyed by mechanical means, and that the labor account be reduced as low as possible. (Signed) ROBT. W. HUNT & CO. Lansing, October 1, 1900. Egyptian Portland Cement Company, Detroit, Michigan. Gentlemen — I beg leave to make the following report of tests of cement made from clay and marl received from you from Fenton, Michigan : FINENESS. Passing No. 50 mesh sieve 100$ Passing No. 100 mesh sieve 98 SETTING TIME OF NEAT CEMENT. Initial set 2 hrs. 10 min. Final set 4 hrs. 40 min. CONSTANCY OF VOLUME TESTS. Cold water pats . . . Boiling water pats Sound and hard. Sound and hard. LIST OF LOCALITIES AND MILLS. 281 TENSILE TESTS OF STANDARD NEAT BRIQUETTES. (1 square inch section.) Serial No. Hardening Period. Neat Briquettes. Sand Briquettes, 1:3. In Air. In W ater. Total Days. Strength in lbs. Strength in lbs. 1165 1 0 1 270 50 1165 1 1 2 440 82 1165 1 2 3 545 135 1165 1 3 4 610 168 1165 1 4 5 680 190 1165 1 5 6 755 212 1165 1 6 7 815 246 Government Standard. . 1 400 160 Very respectfully, (Signed) R. E. DOOLITTLE, Chemist. Lansing, Michigan, Oct. 1, 1900. Egyptian Portland Cement Company, Detroit, Michigan. Gentlemen — I have been investigating the peat question, and submit for your information the following table: Carbon. Hydrogen. Oxygen. Calorific or heat unit value. Capacity of high heat, or calorific in- tensity Centigrade. Wood 50.18 6.08 43.74 4212 2380° Peat 61.53 5.64 32.82 5654 2547° Lignite coal 67.86 5.75 23.39 6569 2628° Bituminous coal 79.38 5.34 13.01 7544 2694° Charcoal. 90.44 2.91 6.63 8003 2760° Anthracite 91/86 3.33 3.02 8337 2779° Coke 97.34 0.49 8009 2761° In examining this table, note the column designated “Calorific Intensity,” and notice you can get as high heat with peat as you can with bituminous coal, lacking 150 degrees Centigrade, and the conclusion is therefore warranted that you can burn Portland cement with dried peat as rotary fuel. It would not cost over twenty cents per ton to prepare peat for rotary work, using waste heat as a drier. The grinding would be very easy. Yours truly, (Signed) W. H. HESS, Chemist. Twentieth Century Portland Cement Company. Organized March 2, 1901. Capital, $750,000.00. Office at Fenton and plant about four miles from the village, and stock said to be 36-Pt. Ill 282 MARL. mainly held there. It is said that marl options are held on Runyan Lake, mainly in Sec. 9 (see Fig. 23), and elsewhere, amounting to 526 acres, and 9,500,000 cubic yards. This is not a very large supply and so far as I know, this and the following companies and loca- tions referred to are not very near production. Zenith Portland Cement Company. Organized July 17, 1900. Capital $700,000. Bonds $300,000. The board of directors were Marshall H. Godfrey, B. H. Rothwell, G. Johnston, E. T. Allen, Stowe, Fuller & Co., R. H. Evans, E. J. Foster. The following are extracts from reports of engineers: Extract from prospectus of the Zenith Portland Cement Co.: I have spent six months in Michigan in the examination of marl deposits, and have no hesitancy in stating that the Grass and Tims Lake deposits are far superior, both in quality and quantity, to any deposit I have examined. I estimate that there is enough marl in this deposit to make 30,000,000 barrels of high grade Port- land cement, or enough to supply a factory of 1,000 barrels per day for over 100 years. The banks of this phenomenal deposit are adjacent to the M. C. R. R., and well adapted by nature for a solid foundation and favor- able location of the plant. Close at hand is found a very fine deposit of clay, which was originally used in the manufacture of brick, but will now be used in the manufacture of cement. Having both of these raw materials so close at hand, a high grade cement can be made here cheaper than any other place I know of. T. C. BEEBE, C. E. Cleveland, Ohio, July 23, 1900. The Zenith Portland Cement Co., Detroit, Michigan. Gentlemen — In answer to your letter of inquiry in regard to the marl bed at Grass Lake, Michigan, I would say that I have twice made an examination of this bed, and have had thorough analysis made from different sections. I have been over most of the marl beds in Michigan, and consider the Grass Lake bed equal in quan- tity of any in the State. As to chemical analysis, it runs about the same as the Brownson and Coldwater beds, but has the advantage of being much finer in texture. Ninety-eight per cent of this marl in its natural state will pass 20,000 mesh sieve, leaving only a very small residue, which is mostly organic matter, and will burn out in the rotaries. This fineness would save considerable wet grinding machinery. This marl is finer in its texture naturally, than any marl I know of now being used, even after grinding. This would insure a very fine mixture, and the very highest grade of Portland cement, as fineness of mix is one of the most important items in the manufacture. The marl bed itself is nearer the railroad than any I know of in the State. It needs no stripping, which will save much expense in handling. A factory can be located at this point LSIT OF LOCALITIES AND MILLS. 283 to handle material both to and from the factory, of fine grade and cheaper than any place in this country. Yours very truly, C. B. STOWE. The analysis of the Grass Lake clay is entirely satisfactory and the quantity is abundant. Our marl has been repeatedly and carefully analyzed, and follow- ing results were universally obtained: Silica (Si0 2 ) 1.22 Iron and aluminum (Fe 2 0 3 +Al 2 0 3 ) 61 Carbonate of lime (CaC0 3 ) *. 95.13 Magnesium carbonate (MgCO s ) 2.04 Sulphuric acid 26 Organic and water, etc .74 100.00 It has a residue of less than two per cent on a sieve of 40,000 meshes to the square inch, thereby saving considerable expense in grinding the raw material; and as there is no muck or organic matter overlaying it, it can be qxcavated and conveyed to the works at a minimum cost. The company’s property virtually includes all of both Grass and Tims Lake, on which the original owners guarantee an average depth of 20 feet of marl on 400 acres. On this basis Grass Lake alone contains enough marl to supply a factory of 1,000 barrels per day capacity for 75 years, and Tims Lake enough more to supply the same demand 37 years. This marl requires no stripping. There is ample water to float our dredges, on each of which will be placed a pug mill. The marl beds of this company lie in Sections 20, 29 and 30 of Grass Lake township, T. 2 S., R. 2 E. (Fig. 26), on the east side of Jackson County. Portage Lake and other lakes of this region are said to contain some marl, but this bed has the advantage of being close to the Michigan Central railroad, so that but a few hundred feet of siding will be necessary. At first, in the prospectus, the factory site was placed at the point marked A in the map, but now the foundations are at the point marked B. Grass Lake is prevailingly shallow. The deeper holes do not appear to be over five to ten feet deep, and large areas are less than three feet deep. Over most of the lake bulrushes ( Scirpus lacus- tris) are growing more or less scattered. In a general way they are most thinly scattered over the deeper holes, and these are points where the marl is covered by most water and appears to have most 284 MARL. organic matter. On the figure their distribution, i. e., that of the marl which comes close to the surface and appears to be better is Fig. 26. Sketch of Grass Lake, T. 2 S., R. 2 E. Property of Zenith Portland Cement Company. The numbers are references to tables of soundings, not of depths. indicated. When the marl surface comes within a foot or so of the surface, Sagittaria and other plants join and soon the marl becomes LIST OF LOCALITIES AND MILLS. 285 covered with a peaty layer extending over the marl, ending abruptly in a vertical wall a foot or two high. Between localities 10 and 14 extends a tamarack swamp. At 14 good yellow marl is found beneath four feet of muck and sand, and at 10, which was at the inside edge of the belt of rushes and pond lilies, and at the beginning of that of tussocks, ferns, and ordinary swamp vegeta- tion, there was marl close to the surface and over eight feet deep, so that the point 25 feet high with steep gravel banks, and shores terraced on the west side, which cuts off the northernmost bay of Grass Lake was once an island but is now joined to the shore on the east by this marl bottomed tamarack swamp. A similar marginal bog, a hundred feet wide, Soundings 15, 16, 17, and 18, lines the north shore, covering marl which is quite thick, but it does not extend up to Tims Lake, as might seem probable from the connecting marsh, and sluggish stream which joins the lakes, because at 18 there is eight feet of peat, and at 19 there is only a trace of marl under the peat at six feet, — below which is sand. The marl is quite extensively covered with the creeping vine- like stems of Chara, which are brittle with coats of lime. The deeper holes are more likely to be covered with a darker green plant (Potamogeton). The west shore is sandy or gravelly where dotted on the map. The land rises gently and the lake bottom is not marly. The water along the edge made a suds, showing an abundance of organic matter. In general the water of the lake seemed full of organic matter and was green rather than blue. Shells did not appear re- markably abundant on the marl beds. On the east side of the lake a point projects with steep bluffs, near which the marl appears to be thinner, poorer and mixed with sand (soundings 1 to 4), and it is said that a shoal streak extends across the lake. North of this point the lake deepens to four feet of water, then rises to a heavy bed of marl (sounding 7), then deepens very slightly. It is not at all likely that this lake was originally abnormally shallow, and there is every indication that its present shoal charac- ter is due to its being filled up with lime, mainly deposited by the Chara growth from variable depths, — over a large part of the lake doubtless over 13 feet deep. There are about 560 acres of marl or more. 286 MARL. The following is a tabulation of the results of the soundings: Water. Peat or muck. Marl. Bottom. Samples. Remarks. 1. 4 ft. 2 2 6' sand 6' 2. 4 “ 7+ 9' 3. 3 “ 5-|- 5 Sandy, shells. 4. 2 “ crust 2 gravel 5. 4 “ 4+ 6. 4 “ 4+ 7. 2 “ 6+ 8. 3 “ 5-j- 9. 1 ‘ 5 6 3' and 5' 10. 1 “ 7+ 8' 11. 2 “ 11+ 12. 4 “ 4+ 13. 4 + Mucky marl. 14. 15. 6 in. 6+ 6' 16. 1 ft. 1 6+ 17. ? 18. 8 19. 6 Trace at 6 6 sand 20. 5 ft. gravel 21. 5 “ 4- Sludge. 22 2 6+ Very good. 23. 2 “ 2 3* 5 Vi Sandy. The auger reached only 8 feet. In Tims Lake (we had no boat there), the marshes surrounding the lake seemed very extensive and it appeared as though they connected the islands shown, — in fact the shores appeared some- what like the dotted line of Fig. 26. The general aspect of the lake, however, is like that of Grass Lake. The temperature of the marl sample at eight feet at sounding 10 was 58°, while the water a foot or less deep was 71° F. and the air 83° F. At sounding 22 the temperature of the marl sample at eight feet was 66° F. During the day the water temperature warmed up from 79° to 83°. No material difference could be noted in the water at the surface and five or ten feet deep, for there was a fair southwest breeze. It is said that the company have clay lands in Ohio. There are brick clay pits to the south of this lake in the village of Grass Lake, and in the flat immediately adjoining the lake to the south, soundings 25 to 28, there are some smooth pebbleless clays, an analysis of a sample of which is given below, from the grass roots down, though in sounding 25 at six to seven feet down, a streak of very fine-grained quicksand was found. The clay at the lake is the ordinary surface calcareous clay of Lower Michigan, the finer part of a rock flour derived from almost all kinds of rocks settled by itself, and its availability for Portland LIST OF LOCALITIES AND MILLS. 287 cement manufacture on a large scale is rather doubtful. For ex- ample, it is doubtful whether it will remain of the composition shown by analysis. The surface, where soundings 25 to 28 were, is less than eight feet above the lake. The analysis of the marl cited in the prospectus is given in column (1). An analysis by W. M. Courtis of Detroit, is given in column (2), and one by Prof. F. S. Kedzie in column (3). No. (1) is evidently of a sample of dried marl, and I think that more or less organic matter must have been removed with the water. No. (2) is of a sample dried at 100° C. and only 42.11$ of the orig- inal sample. Analyst. Silica Si0 2 Alumina and iron Calcium oxide CaO as carbonate Magnesia as carbonate Sulphuric acid SO s Carbon dioxide C0 2 Organic matter and water Difference Prospectus av. W. M. Courtis. F. S. Kedzie. 1.22 .61 74 See diff. 83.045 1.201 0.485 11.700 3.569 9.64 1.92 43.15 (77.2) 1.50 (3.72) 32.80 10.99 100.000 100.00 The character of the deposit is distinctly that of Chara lime and it will be noticed in analysis No. 2 that there is but 32.80$ of C0 2 whereas to turn the calcium and magnesium oxides into carbon- ates 36.27$ would be needed, so that probably quite a little of the lime is united either with sulphuric, or more likely an organic (succinic) acid. The supply of marl is said to be equivalent to 400 acres 20 feet deep. As we could not sound over 13 feet, we have no means of checking the statement exactly. There is certainly a large supply of marl over most of which no stripping will be necessary. The plan is to dredge the marl, and transport by a lakeside entrance to the factory, and pump out. The plan is to have a rotary pump of the latest design and the cost is figured not to exceed 80 cents per barrel. The prospectus figures selling price at $1.40 a barrel, which was probably right then, but later, September, 1901, cement was de- livered in Lansing at from $1.40 to $1.50 per barrel, and even at times perhaps $1.25 for new brands, and I am told that it has been sold in 288 MARL. Michigan f. o. b. at factory at 90 cents to $1.00. Cement advanced, however, during the printing of this report, to over $2.00. The prospectus also says that “coal for power can be obtained in abundance within ten miles of the plant.” There is very likely some coal at that distance, but hardly an abundant supply. An analysis by Prof. F. S. Kedzie of an average marl from loca- tions 5, 7, 9, 15, 16, at 8, 6, 3 and 5, 6 and 8 feet, respectively, is given in column (3). His analysis of the apparently best sample of clay at the south end of the lake, location 28 at eight feet depth, is as follows : Si0 2 49.86 (Al 2 Fe 2 )Oa 21.22 CaO 6.32 MgO 2.75 C0 2 5.44 Organic matter and water 7.14 Undetermined 7.27 100.00 Carbonates are unusually low for a surface clay, which is a good point, but the alumina is high. Standard Portland Cement Co. Organized November 15, 1900. Capital $1,000,000. Office at Detroit. This company will develop the lime which exists in Zukey and adjacent lakes, at Lakelands, where the Ann Arbor R. R. and the Air Line of the Grand Trunk R. R. cross. Prof. I. C. Russell was employed to test the marl beds, but some time before I made a cursory examination. His report was published, — in part, — in the prospectus. Referring to the map, Fig. 27, we see a group of lakes, which evidently were once much more continuous, and have been sepa- rated by marsh growth, while the 15 to 20-foot bluffs which mark the original borders of the lakes are plain. Zukey Lake was the one which I studied myself more carefully. The west side is lined with a thick and pure bed of bog lime which is capped by a growth of marsh plants and peat, a foot or two thick and coming up to the lake in a perpendicular wall. The marl bed projects out white be- neath, and upon it there is Chara, and occasionally dead shells of LIST OF LOCALITIES AND MILLS. 289 37-Pt. Ill Fig. 27. Lakes near Lakelands. T. IN., R. 5 E. Location of Standard Portland Cement Co. 290 MAUL. Unio, etc., and twigs are heavily coated with lime. The north part of the lake has a sandy shore and the bluffs are of gravel, and the shells are not so coated. On the southeast side of the lake the marl seems to he covered with pebbles, brown above and green below, which prove, however, to be Scliizothrix concretions. The cuts through by the marl bottomed Round Lake to Strawberry Lake are artificial, through a marsh covering a bed of boglime. In Strawberry Lake itself, which is merely an enlargement of Huron River, the lime does not seem to he so continuous. In this lake however, at the place which I have called Blind Island, is an atoll- like formation which is significant of the origin of the lime in general. There is a small, nearly circular area of uniformly shallow water, beneath which is boglime, around the margin of which there 1 is a mat of vegetation of rushes and other aquatic forms, in a ring. Outside the ring the water drops off suddenly to a depth beyond my sounding pole. I should say that the whole region is one of irregular topography, of kames and gravel knolls, and the explana- tion of this island seems to be that in the original bottom of the lake there was a knoll which rose near enough to the surface of the water to make a good seat for the lime secreting plants, which built up the deposit to near the surface, thereafter building out slowly in all directions on the debris which forms and slides down the slopes, whereupon the other plants came in, but possibly the spring* ice has checked the formation of a permanent bog mat of vegetation. If the explanation is correct, they are like the coral islands in origin as w r ell as looks. The cause of the distribution of the boglime is not altogether clear. It does not seem generally to prefer to run up against a gravel shore, but possibly that may be due to gravel washed down upon it. Silver Lake, for instance, which lies in quite a deep hollow, does not appear to have boglime, while the marshy hollows next east appear to be underlain with it. This property is said to have been sold to Cincinnati capitalists recently. * Prof. E. D. Campbell of Ann Arbor tested the materials. The analvsis of the raw material “gathered bv Frof. Russell during his examination and a composite sample 7 ’ is Ho. 1. Ho. 2 is from Lime Lake, Ho. 3 from Zukey. * Detroit Today, 12: 6: 1902. LIST OF LOCALITIES AND MILLS. 291 No. 1. No. 2. No. 3. Silica .96 1.30 1.30 Ferric oxide .62 ) .70 .58 Alumina .00 j Calcium carbonate 93.92 94.98 94.52 Magnesium oxide 1.79 1.44 1.44 Sulphuric anhydride Difference, carbon dioxide .58 tr. tr. and organic matter 2.13 1.58 2.16 100.00 100.00 100.00 Clay. 3.7G 62.55 17.40 5.08 1.67 tr. 5.55 2.30 1.69 100.00 Wayne Portland Cement Co. Organized March 18, 1903. Capital $800,000, Office in Detroit. Dr. G. Duffield Stewart says that they own 470 acres of marl land within six miles of Brighton on the T. & A. A. R. R. Pyramid Portland Cement Co. Organized January 17, 1901. Corporation office at Detroit. Cap- ital $525,000,000. This plant is to be located at Spring Arbor, where abundant ma- terial is said to be near. It is planned to be a 1,200 barrel a day plant. The lime deposits here were noted by the Douglas Houghton Survey, and there are exposures of coal measure shales not far off. An average analysis of the marl is given among Prof. Fall’s analyses on p. 352, — and also an analysis of Jackson clay, — a little high in alumina. German Portland Cement Co. Organized March 29, 1901. Capital $300,000.00. Office in Detroit. This company was organized to develop the beds around White Pigeon, T. 8 S., R. 11 W. They are now building tlieir plant near the village on the Lake Shore road, hoping to be ready by July, 1902. Sand . . Silica Alumina Ferric oxide Magnesium oxide Sulphuric anhydride Combined water and organic matter Calcium oxide Difference, alkalies, etc 292 MARL. They expect to use water power for grinding and electricity. The lime comes from Marl Lake, two miles southeast of the town, which has been described by Mr. Hale on p. 103. Three Rivers Cement Co. Organized August 10, 1900. Capital $20,000.00. Office at Three Rivers, and intended to develop the beds of boglime in that region, at Pleasant and Fisher’s Lakes, where it is said to be all over the lakes and 14 feet deep in some places. These plants are geologically in the same region as the already established Branch County plants, and in a general way similarly located, though there are no out- crops of shale clay at hand. Farwell Portland Cement Co. Organized June 29, 1901. Capital $350,000.00 ; $10 shares. Bonds $175,000.00, 6^ twenty-year gold bonds. Officers: J. L. Littlefield. Geo. W. Graham, T. F. Bingham, W. C. Hull, W. C. Fuller. This is the company organized to develop the Littlefield Lake marl deposits, elsewhere described by Prof. G. A. Davis,* and illustrated in Plate XIX. It will be noticed that the marsh cover- ing is rarely as much as three feet, usually from two feet down. The deposit, while not as accessible as some, is not far from the junction of the Ann Arbor and Pere Marquette systems, which will give good shipping facilities at Farwell, where the plant will be. The analysis from samples collected by Prof. Davis personally, and analyzed by Prof. F. S. Kedzie, is as follows: Calcium as oxide 51.00 51,67 51.04 51.23 Magnesium as oxide 1.75 1.22 1.61 1.20 Carbon dioxide 42.94 42.41 42.96 42.80 95.69 95.30 95.68 95.23 Calcium as carbonate. . . . 91.1 92.2 91.2 91.6 Magnesium as carbonate. 3.67 2.55 3.36 2.50 Shortage of C0 2 .92 .55 1.05 1.13 Insoluble (silica) 0.31 0.34 0.30 0.63 Iron and al. oxides 0.33 0.16 0.24 0.10 Difference (organic) .... 3.67 3.60 3.65 3.99 Total 100.00 Geological Survey of ^Michigan. Vol. VIII. Part III. Plate XIX. NORTH SOUTH PROPERTY OF FARWELL PORTLAND CEMENT COMPANY. RECORD OF BORINGS. V — ■ LIST OF LOCALITIES AND MILLS. 293 No. 1 is a mixed sample from the large islands. No. 2 is from hole 24, unmixed. No. 3 is a mixed sample from fifteen holes, the northwest half of the lake. No. 4 is a mixed sample from five holes, the southwest half of the lake. Farwell is only about 50 miles from the Saginaw coal fields by the Pere Marquette, and the Ann Arbor runs direct to Ohio, in case the clay should be drawn thence, and also passes close to the shale clays around Corunna, already mentioned. In the Littlefield Lake marl the calcite is in lumps of all sizes, but even when no larger than 0.001 mm. often showing aggregate polarization. I was not able to discover any sharply crystalline grains like those in precipitates. Clare Portland Cement Co. Incorporated in New Jersey. Capital $1,000,000.00, with 100,000 shares. The company owns 1,905.61 acres of land in Grant and Hatton townships, Clare County, T. 17 and 18 N., R. 4 W. It is mainly located at Five Lakes, Sections 5, 8, 9 and 16. The report of the consulting engineer, Prof. R. C. Carpenter, follows. It will be noticed that the clays are the ordinary surface clays with a large amount of carbonates, except one, which is probably only a relatively thin superficial layer in which the lime has been leached out. A production of 1,000 barrels a day is planned. The officers are, H. Robinson of Akron, president; C. W. Somers of Cleveland, and the J. II. Somers Coal Co., of St. Charles, vice president; C. W. Perry of Clare, secretary; F. G. Benham of Sag- inaw, treasurer. Extract from Prof. Carpenter’s report to the Clare Portland Ce- ment Company. In September last I made an examination of the Portland cement lands owned by W. H. Shepard and partners, of Saginaw, Michigan, and would respectfully report the following results of the examina- tion: Location. These lands are located in township 17 north, range 4 west, known as the township of Grant. They comprise altogether 1,905.61 acres, and are principally located in Sections 5, 8, 9, and 16. They 294 MARL. are situated at an average distance of about five miles from the city of Clare, and at a distance of about three and one-half miles from the village of Farwell. The lands are located about one mile from the Harrison branch of the Pere Marquette railroad, and about three miles from the Ann Arbor railroad. No less than five switches for logging railroads were at one time graded through the property, and these grades are now all in good condition for rail- road service by simply laying of ties and track. The property is all owned by Mr. W. H. Shepard and partners, who claim to have a perfect title. The country surrounding this property is a highly developed farming region with a clay loam or clay soil, and is quite rolling in character. Amount of Cement Material. The cement material which is found on this tract of land consists of marl and clay of very excellent quality. The marl is found in the bed of five lakes, where it is covered with water, which varies in depth from a few inches to several feet; it is also found in several swamps which surround the lakes or lie adjacent to them, where it is covered with muck, having a depth which varies from a few inches to one or two feet. The total amount of the marl land as measured by a planimeter from an accurate map submitted, is 754 acres, of which 233 acres are lake and 521 marsh. The average depth of the marl over this entire tract would seem, from such information as I can obtain, which was checked from actual measurement, in a large number of places to exceed twenty feet in depth, but in order to make a safe estimate, I have assumed that it was but fifteen feet in depth. To determine* the amount of Portland cement which could be manufactured from this amount of material, we will consider the following data refer- ring to the composition of Portland cement. One barrel of Portland cement contains 380 pounds, of which, under usual conditions, 64 per cent would be lime (CaO) and the remainder part clay. Koughly speaking, two-tliirds of the Port- land cement is lime and one-third clay. The marl is carbonate of lime (CaCO). The weight of the carbonate of lime for a given bulk is in excess of that of the lime as 100 is to 56. Calculating from data thus submitted, it will be found that for one barrel of Port- land cement would be required 570 pounds of carbonate of lime, which is about the equivalent of marl when perfectly dry. In order to account for impurities of various kinds, and to make the esti- mate doubly safe, it is assumed that 600 pounds of dry marl will be required for each barrel of cement. The marl as found at the bottom of the lakes usually contains 70 per cent of water, and that from swamps usually contains 50 per cent of water. This would indicate that for each cubic foot taken from the bottom of the lake would contain 48 pounds and that from the marsh would contain 80 pounds of carbonate of lime. ♦Compare calculations on page 39, and pages 167 to 168. LIST OF LOCALITIES AND MILLS. 295 Consequently, it would require for each barrel of cement made, 12.5 cubic feet of lake marl, or 7.5 cubic feet of marsh marl. It is seen from this that the marsh marl is preferable, for the reason that it contains less water, which must be evaporated during the process of manufacture. One acre equals 43,560 square feet, and if worked fifteen deep would make 93,100 barrels of cement from the marsh marl and 52,150 barrels from the lake marl. The total capacity of the deposit by this calculation would be from the marsh marl 48,505,100, and from the lake marl 12,150,950 barrels, making a total of 60,650,050 barrels. If the deposit were worked at the rate of 1,000 barrels per day for 365 days each year, it would furnish a supply for 166 years. Clay of very excellent quality, as shown by the analysis accom- panying the report, is found in large quantities immediately ad- jacent to the marl beds. The clay covers an area exceeding 160 acres and has a depth varying from 20 to 60 feet. About one and one-half cubic feet of clay are required per barrel of cement, al- though when carbonate of lime is mixed with the clay, as is found in this deposit, the amount required will be more, and may average two and one-half cubic feet per barrel. This condition, of course, implies the use of less marl, which is not taken into account in estimating our quantity. Taking the clay as averaging 30 feet in depth, one acre would supply enough for 493,000 barrels. This calculation indicates that the amount of clay available is much in excess of that required to manufacture the marl into Portland cement. In addition to the clay in the upland adjacent to the marl, an in- vestigation shoAvs that it lies underneath the marl, and consequently the amount available is much in excess of Avhat the calculation indi- cates. Roughly speaking, there is enough material to operate a cement plant making a thousand barrels per day, for a period exceeding 166 years. Character of Cement Material. An analysis of the dry sample of marl shows as follows: No. 1. No. 2. Clay, i. e., .silica, alumina, iron 3.65 2.56 Calcium carbonate 94.15 96.04 Magnesium carbonate 2.20 1.40 An examination of various samples has as yet 4 shown no free sand. As the surrounding country is largely clay, it is very im- probable that any is found in the marl. 296 MARL. An analysis of dried samples of clay shows the following results: Silica Alumina Iron Calcium Carbonate — Magnesium Carbonate. Loss No. 1. No. 2. No. 3. No. 4. Top of About 20 About 30 Beneath bank. feet down. feet down. marl. 65.05 47.60 45.60 50.40 25.00 1 5.80 f 15.00 15.85 22.10 2.05 28.29 28.82 24.00 0.40 6.00 8.60 0.52 1.64 2.91 3.13 2.98 These analyses differ from each other in the percentage of cal- cium carbonate present. This is a very desirable addition* to the clay, but for the purpose of comparison, the following table is presented, which is calculated on the basis of no carbonate of lime being present, and magnesia is reduced from carbonate to oxide. No. 1. SiO, 67.60 MgO 0.19 A1A 26.00 Fe 3 0 3 6.00 No. 2. No. 3. No. 4. 69.45 68.95 69.02 4.20 4.65 0.35 21.80 23.35 30.26 These clays are all high in silica, which is the most desirable element in the manufacture of Portland cement, and they are low in any elements which are undesirable. In fact, these clays are hardly to be surpassed in chemical composition, and so far as the writer knows, are fully equal to those found in any locality. In connection with the manufacture of Portland cement expe- rience has shown much more difficulty in securing deposits of good clay than in obtaining carbonate of lime, and most of the difficulties which have been experienced in the manufacture of Port- land cement have been due to the fact that the clays obtainable contain less than 50 per cent of silica. As showing the fact that a good cement is made with clay of a similar composition, I submit analysis of the Sandusky clay, used by the Sandusky Cement Com- pany, and the Warner clay, used by the Empire Cement Company, two of the oldest companies using marl and clay, reduced to a similar basis as the table given above. SiO, Al 2 6., Fe 2 0, MgO Sandusky. . 72.2 . 18.30 ) . 6.65 1 2.08 Empire. 65.50 33.50 0.82 99.23 99.82 ♦In this statement, Prof. Carpenter differs from the prevailing opinion, as may be noted by comparing statements elsewhere in the report. This is not because the calcium carbonate itself is deleterious, but because as the analyses show it is liable to be very variable, and to be associated with much magnesia, so that it is more difficult to make the cement of a constant composition. L. LIST OF LOCALITIES ANIJ MILLS. 297 I have made in the laboratory of Sibley College, a small quantity of Portland Cement from this material. This was of excellent qual- ity, giving a tensile strength as follows : Age. Strength (lbs.) Remarks. Age. Strength (lbs.) Remarks. 2 days 285 Neat. 1 month.. 1022 Neat. “ 328 Neat. 1 “ 1112 Neat. 2 “ 375 Neat. 1 “ 962 Neat. 8 “ 868 Neat. 7 days — 240 3 pts. sand. 8 “ 860 Neat. 7 “ .... 220 3 pts. sand. 1 month. . . 1056 Neat. 7 “ .... 212 3 pts. sand. All samples left one day in air and remainder of time in water. Watervale Portland Cement Co. Capital $1,000,000.00, of which $600,000 common, $400,000 pre- ferred, a share of the common being given with every share of the preferred. The parties interested in this project were also interested in the Omega and Elk Rapids, and apparently are letting it lie dormant, until the latter are better established. The company was said to own 800 acres of marl bed, on the Willow Brook farm and about the Herring Lakes, in Sections 13, 14, 15, 22, 23, 24, T. 25 N., R. 16 W., and Sections 18 and 19, T. 25 N., R. 15 W. The average depth is said to be 20 feet. There is said to be also 60 acres of clay banks three feet deep and over, and 175 acres of other lands with houses, etc. A feature of this proposition is the nearness to the Great Lakes, so that the lower lake can easily be made a harbor, being from 60 to 80 feet deep, while the upper lake, which is said to be shallow, is said to be underlain by over 25 feet of marl, which also extends beneath the sw T amps around its margin, where it is covered by not over three feet of vegetable matter. One reason for the comparative scarcity of marl near the Great Lakes may be that owing to the relatively recent fluctuations of level, there has been not enough time for its accumulation, and it is worth noting that these lakes are close to an axial line of tilting, passing through Port Huron, along which the lake level must have been relatively permanent. The property was reported upon by Lathbury & Spackman, Prof. Delos Fall, and C. B. Stowe. Lupton Portland Cement Co. Organized under the laws of New Jersey, January, 1901. Office in Chicago. G. T. Stanley, president; E. A. Worthington, vice pres- 38-Pt. Ill 298 MARL. ident and treasurer; W. Higgs, secretary; W. C. Edgar, assistant secretary; A. H. Cederberg, superintendent. Capital $1,250,000, of which $600,000 was to be placed on the market, two-thirds at two- thirds of par, the balance at par. The size of the plant planned may be 1,200 barrels daily output. The photographs in the prospectus show very well the swampy outbuilt margin to the lakes. Mr. Stanley is we believe the prime mover in this company, having been engaged in lumbering around Lupton. Extracts from the report of A. H. Cederberg : Messrs. Lathbury & Spackman of Philadelphia, have analyzed our marl and clay, and the following is their analysis in full : Lab. No. 942. — Marl from North Lake No. 1. Silica (Si0 2 ) 25$ Alumina and Iron Oxide (A1 2 0 3 Fe o 0 3 ). .19$ Lime (CaO) 52.38$ 93.53$ Magnesia (MgO) 1.14$ CaC0 3 Sulphuric Acid (S0 3 ) 18$ Loss on Ignition 46.05$ 100.19 Lab. No. 943. — Marl from North Lake No. 2. Silica (Si0 2 ) 24$ Alumina and Iron Oxide (A1 2 0 3 Fe 2 0 3 ). .08$ Lime (CaO) 52.97$ 94.58$ Magnesia (MgO) 1.13$ CaC0 3 Sulphuric Acid (S0 3 ) 08$ Loss on Ignition 45.49$ 99.99 The analyses of marls show them to be very uniform and high in lime, while they are low in injurious ingredients, such as mag- nesia and sulphuric acid, hence are well adapted to the manu- facture of Portland cement. LIST OF LOCALITIES AND MILLS. 299 Lab. No. 957.— Blue Clay No. 2.* Silica (Si0 2 ). . . Alumina & Iron Oxide (A1 2 0 3 & Fe 2 0 3 ). Lime (CaO) Magnesia (MgO) Sulphuric Acid (S0 3 ). Loss on ignition Difference (alkalies) 56.0<)$ 28.89$ Very 00.00$ little 0.58$ Iron 0.41$ Oxide 7.58^ 93.55 6.45 This clay is well adapted for the manufacture of Portland cement. A mixture of equal parts of blue clay No. 2, and the Spring Lake clayf would give a clay with a better ratio between the alumina and silica than the blue clay alone; also the percent- age of magnesia in this mixture would be low. The village of Lupton is on the Rose City branch of the Detroit and Mackinac railway, about 27 miles from the Emery Junction. The conditions for a mill site are more than satisfactory. A mill could be erected, as suggested, right on the outskirts of the village of Lupton, or down by the lakes, which are situated about a mile from the village itself. The marl deposits are, I may say, inexhaustible; and from samples taken, not only by myself, but that have also been for- warded to me from Lupton, and to Messrs. Lathbury & Spackman of Philadelphia, I can state that the marl is superior in its quality to most of the marl beds that have come to my notice. The con- ditions for getting at this marl are very easy. One of the lakes could be drained at an expense of a few hundred dollars to such an extent that steam shovels would be entirely unnecessary, thus reducing the cost of putting the marl on cars to the mill, to a minimum. The clay that has been found on the property is not very well adapted to the mixture for first-class Portland cement; but only a few miles away from there another clay deposit is of superior quality, and which clay, I am informed, can be deposited at the mill at a cost of thirty cents per ton. The ratio between marl and clay necessary for a mixture would lie between the figures of four and five to one. After having made thorough chemical analyses of the raw mater- ial, and which analyses correspond entirely to those made by Mes- srs. Lathbury & Spackman, I went to work and ran through a rotary kiln enough raw material to produce a sufficient quantity of Portland cement, in order to make necessary physical tests. The chemical mixture in the raw was made so as to conform to the standard requirements of Portland cement by the American Soci- ety of Civil Engineers, which requirements, as you well know, are * Apparently a Michigan series shale clay. L. f A brown clay used to reduce the ratio of alumina which is rather too high. 300 MARL. higher even than those of the United States government. The fine- ness of the cement produced was satisfactory, the color was pure Portland shade and tensile strength on the briquettes made was as follows: After 24 hours: 444 and 486 lbs. Average 460 lbs. (Final set in air and balance in water.) After 3 days : 618 and 643 lbs. Average 633 lbs. (1 day in air ; 2 days in water.) After 7 days: 702 and 829 lbs. Average 815J lbs. (1 day in air and 6 days in water.) After 30 days: 891 and 916 lbs. Average 903J lbs. (1 day in air, 29 days in water.) Only the regular methods adopted in cement making were used in the making of these briquettes. The initial setting was 172 min- utes; the final setting 360 minutes in all instances. The amount of sulphuric acid in the finished product was 1.53 per cent, thus insur- ing a cement of superior quality and of which there can be no doubt * as to its durability. In using the above figures, I have taken into consideration the fact that you must produce 1,200 barrels of Portland cement every 24 hours. That means that a plant must be constructed that has a capacity of turning out 1,500 barrels in the same time — 24 hours. It may astonish you to hear of this, but it remains as an absolute fact. We very often hear that parties go to work and construct a cement plant of 1,200 barrels capacity, and they will figure in this 1,200 barrels on no repair account whatever. As a result, there are in the United States to day only two mills to my knowl- edge that have a capacity of standing by the figures agreed upon for a 24 hour output. These two mills are the Lehigh Portland Cement Company’s mill “B,” at West Coplay, Pennsylvania, and the Whitehall Portland Cement Company’s mill at Cement Town, Pennsylvania. It must be very evident to you that if an engineer guarantees to turn out cement at a certain figure, he must rely upon it that his output is never decreased, but if anything, increased during the 24 hour run. We very often hear some engineers that guarantee 200 barrels of clinkers per day in a 60-foot rotary kiln. That guarantee is all nonsense. It is an absolute impossibility. In the first instance, every rotary kiln must stand idle from three to four hours every day, owing to what is in burning practice called “patching.” Hence, if a cement mill has got, say for in- stance, twelve rotary kilns, it means to say that two rotary kilns will practically lie idle during the 24 hours. Furthermore, patching is not the only cause of shutting down a kiln. The brick lining in a rotary kiln is subjected to an intensely hard wear and tear, and it is nothing unusual to see rotary kilns relined partly every two months. That means another shut down of the rotary kilns for two whole days. In my experience, I have found that it is a safe guarantee to figure on a rotary kiln having an average monthly capacity of 130 barrels of clinkers in 24 hours. So much in the rotary kiln process. LIST OF LOCALITIES AND MILLS. 301 If we go to the grinding process, the conditions there are just the same. A maker of machinery has been invited to figure on an out- fit to produce so many barrels. He gives you a figure. You go to another manufacturer and he turns in .another figure. The manu- facturer that turned in the first figure has probably been a trifle too high. He wants to reduce his first figure in order to obtain the bid, and the first thing he overrates the grinding capacity of the machine he proposes to furnish. Knowing this full well, it is an absolute necessity to have a surplus of grinding capacity in both raw and clinker departments of not less than twenty-five per cent, thus doing away with any danger whatever tending to reduce your 24 hour output and increase the manufactured cost per barrel. It is also a very noticeable fact that repairs cost as a rule a mere trifle as far as cost is concerned, and that the largest expense in connection with the cement mill is the shut downs , thus decreasing the output and increasing the cost to a considerable degree. Standiford Portland Cement Company. In the northwest corner of Branch County, not far from Union City but nearer Athens, are some extensive beds of marl which have been investigated for the Standiford Portland Cement Company by Prof. Delos Fall and H. K. Whitney. The results of a thorough sur- vey and a large series of analyses are shown in the map and tables herewith given. Mr. Whitney says that over one thousand sound- ings of same were taken, 508 on land, 96 on the shore line and 445 on the lakes when covered with ice. “The bottom of lakes, except at mouths of cracks and in the deep- est portions of the line of flow is clear clean marl. At the mouths of creeks and in the deepest portions of Kynion and Lehr Lakes in line of flow it is overlain with sediment as indicated on the large map. It is a soft dark sediment, apparently organic material, or the same mixed with marl and is almost entirely in deep water and outside the estimated 240 acres.” “A large part of the surface material (peat, muck, etc.) on the marl lands could readily be burned off in time of low water. Points when it is four feet deep are in general two feet above ordinary water level ; some points four or five feet above. It is, however, quite possible that it might have value for fuel. “The difference in water level between the lakes (between Lehr and Kynion Lakes, 5 meters between Kynion and Clayton Lakes 1 inch), could be readily eliminated by dredging for marl. On May 9, 1900, Clayton Lake was 2 feet 4% inches above the river at mouth of outlet from lakes at time of medium high water. At ordinary or low water it would be as much as 3 feet. The distance is miles 302 MARL. direct (about 1% miles on the line of the creek). For about $1,000 the water in lakes could be lowered about 1 % to 2 feet, with a benefit of about $1,000 to the adjoining property, aside from marl beds. Then at low water nearly all of the surface material could be burned off from the marl lands, with little trouble and at nominal expense. Similar large marshes in the vicinity have been burned off below the ordinary water level, accidentally or intentionally.” The estimated acreage of marl with an average depth of 20 feet is computed as follows by Mr. Whitney : Acres. Acres. Area of lakes more than 20 feet deep 13G Area of lakes less than 20 feet deep 104 Total area of lakes 240 Area of land 10 feet or more marl, 4 feet or less of surface . 90 Area of land 10 feet or more marl, 4 feet or more surface 52 10 to 30 feet of marl (average 20 feet) 142 5 to 10 feet of marl ( average 7 feet) 33 175 Area of land and water 415 Less allowance (possible error) 15 400 For commercially available marl we shall have, however, within 30 feet of surface : Acres. Area of land, 10 to 30 feet thick, averaging 20 if we neglect thickness above 30 feet 142 Area in lakes to 20 feet of water 104 246 Less allowance for error 6 Total acreage 20 feet thick 240 The 33 acres of marl between 5 and 10 feet thick would be equivalent to 10 Total 250 Which would be equivalent to over 8,000,000 cubic yards. Prof. Fall went over the bed and took at proper intervals a series of 84 samples, and the analyses given in the table annexed are of the entire series not omitting any. IASI OF LOCALITIES ANI) MILLS. 303 5AMPLE5 Marl Beds of t he STANDIFORD PORTLAND CEMENT CO. •fSlTHENS.ffilCH. TABLE S^ou/mg:- beprh of Sample.*. T)epth of Marl, bepth and Ki nd of Surface Cootr/ny , efc., at all point j ,n said Marl Beds from which Samples lire-e taken for Chemical A nalyjii ■ To- Location of Same.. See Map [ S+aJion Mos- Correspond mef to Sample A/oi.] Samples taken by Delos Fall. So D, Analytical Chem.tt, tlarlan K Whitne,,. Surueito'- A/b'oh . Mich. Qotf-fe Cn&ck, M ch £,*|5lanatioc • — — — M -Muck P = Pent G ■" Grass-roots, and loose, rsart/u decaued UeoitaTion W - Water f /> 7 0 ■ Bottom of Mori fat depth Sounded 1. + = No Boltom tl w Quite Hard [• qiuen] 28. H ~ Hard at 28 below Surface Q ~ C/auej apnearante at bottom. Sam file, dumber. Surface Material Pip* of Surface mot and D*P f A Sampler D*P*K of Clear Marl . Of Sounding. Qo*«-' efMorl etc Scirnpl e Number Surface ; Mali r ml 5e,r* <( Surfoci Dtp* 1 ' of Sample Vtpr A of C'«a- beptu of Sound m q Botham oj dart, etc _laA P T l f IS to If 2*T zs g [Bo*o~. .rji] 3/ 95 F 9- 16 22 <• 26 H. L. i b VI ft >1 96 a I 18 13 2 9 13 ■ X a 4+ 31 ■f- [- N 0 Bottom] T 3. 97 p X If 10 12 73 * l b. xo-19- >> 9-2 p AT 7 54 8 73 r 3 P 3 ? 101 13\ 8. ST 9-9 w '/? 17 /7 'T IS B 9- a / ? 19 30 3- 50 p 3 n / A 15 B * 5 w ts zii * 31 4- J~7 S’ w ,y 2. m 2^4 X6 B 8 6 w 5 19 10 IS s '57. 52 V A -8 /s 10 B r 7 w 9 /o 34 s. r? S3 a J/ l i8 27T 2.8 B •0 8 A a&w' / It ji * 31 + CO 59- vV / 16 A3 19 B •• 8 b zx 6/ 55 p 3 10 '3 :6 B 'i 9 Mi ( P Z zo zxi 1 91 3. oA 56 W '4 if J/T* 31 + n 10 M &P f 19 16 2/ B 57 w ! if 31* 31 -t II MS P 6 10 19 30 d. (>« fS vV 9 19-lt XI 30 8. •5 IZ MkP It m- 3 X 9 is-. 59 p X n -SS 19* 3 ' + [28. A], •S ns ;i7 zo 19 31 3 u 60 p X 5-/7 22 4 2V-1 u. •7 iv- p as It- 10 AST 26 B. 17 6i p 1 T t-9 8 >0'X 8. /8 it VO i X XI z 3 13. 6? 62 V 3 9-n 9'ri. nh. 13 /?. 16 IV i to 1*1 A 51 B. u 63 w / '2-/7 19 zo B. 10 17 p 9 18 18' 31 + C 30 91 7 0 69 VI Vi f- 6 8 S'f B. ~TT 18 p 3 10 Zf 18 s >■ 65 vV 7 6 -10 19 * 3 / + u. 19 p 9 is-/* 13 / 7 5 77 66 w / il ■ 15 '6i 17% 3. i-3. XO F 3 10 /8 11 B. 'J. 67 p 1 ll n 19 3. IV XI w8,& i If 17 i s B 7V. 68 w i i9 30* 31 + xs. XX p l*L IS 23 19'^ 13- 7 5. 69 w 2 17 17* 19 H [Dsrk Ofi/or], u>. XI wild / 15-16 xz\ AJ*T 13 76 70 tv i7 30-3/ 1 9 31 ■hi if. Hi 17. XV- p £ It -/l /8 13. 77 7/ £ V- * 22 16 B ■li- zs w / 10 29 25 3. 7 S' 7A w i i 19 If Z5X a 19 16 p 3 IS 19 t 32 + t?. 73 tv / // 10 1/ B. 30- 17 p 9 '5 18 1 31 + 2o. 79 w 10 XO 18 18 s. 3/ UU p A il 30 + 32 + f. 75 w 8 18 13* 3/ + [19. *] 3a. 18 b. ZO Si 76 w 6 /9 10 2b Bllh, Hi 33 18 c. 31 C S3. 77 tv 1 IS X! 13 13. 39. Z9 M 3 5-7 Zf ay 13 3+ 78 V A IX XI 13 3 J5-. 30 9 19 19 4 234 B. ss-. 79 a 4 / 3 28T X9 13. 3*. -SI £ X y 9 '6 17 \ 30 3. 26 80 IV / 4 It A5-S 17 B- 37. 31 p 3 IS IS i8 3. 87 3/ tv / 19- A** 15 4 13. 3& 33 w / X 17 18 13. 22- 8A p S’ HO ? ?? 3* p 34 IX II ^ /s 13. 90. 35-« p A-4 10 IS /7>i B. V/ tsb .. ! 6 >i [Samf/e , C J . 92 36 w / t- 6 18 2? 13- *3- 37 p 10 17 2/ B 38 p 5 16 IS * 30 H. ,5- 39 M 6 /0 1/ A 7 13 *6 90 p 3" 10 19* A? H V 7 91 w 0 9 HI. A/'t B. - WA ken a t 1 [See Map], 91 p 5 a 9 3. **A? • 18 2>, 1 , 28 c , all lube 7 or - It. 9.3 w #/ 2. n 10 13. 3ou> '(/erf it ip^e ? urattr uraS on ■F h, l 99 p 3 10 19 AA 13. H.K. Whitney, Suru-e. Or Fig. 28. Table of data relative to the samples, the analyses of which are given on subsequent pages. The numbers in the second column correspond to the Roman numerals in the second columns of Figs. 29 and 30. Soundings of Standiford Portland Cement Company on Kynion, Lehr and Clay ter lakes. Sections 4, 5, 8, 9, 16 and 17, T. 5 S., R. 8 W. 304 MABL. Artalv^ses of K^r\lot\,LeKT,a.ndCLa.^tc5>a Lo_Ke /\Acxrls,T 5 S., R.8 W. , b'j Delos F~cx 11 PK-D.^cxt tKe AAcMLllQ.wCKeYm.cal Louboya-t o>r^,^lVavo>n. L CL b MLLYwber Fl eld N\o.r K Si. Os. Fe^O^ Ala 03 ColCO, AAqCO* 50, by^antt Mutter "Cept W mVeet. 500 I Ik .6 3 .99 8 88.13 20 'TKjO-nJL 9-45 501 IB 72 12 86.85 23 YVffya 1 V. 11 501 II A \.\6 3 06 39.05 .31 VLffvx G. 38 50 3> mB \.-9\ 4 . 83 a 35.58 V\>o-rAJL V%-CTV\X 1 . 68 5o4 u 3 78 5.l9 a 82.97 rvmt YNCru 8.06 50 5 3 78 4.29°“ 80.48 'r\-0>UL. \ \.45 506 m 2.99 \.9\ 19.36 y\ar\L f>_onrvL 1 l .4 4 507 m 2.01 4.W 8143 trv-cucx YOOTUL \ 2.4 5 508 m .65 2.4o°- 38. S5 rv-o-yvx vvcmjL 1.40 50&B sub \ .60 4.05 8415 "trooucjl. .31 9.23 5oS 1 5 .13 2.oo 11.6o tn tvo. 1 . 54 \8.\3 5 lO X 3 52 2 1 \ 93.16 tn^oocu. b y o3-ojl xm 2.98 1.60 88.03 •n~0-y^A. .55 6.01 514 Ti? 1.89 3.4 3 90.21 XruOucjL 4.41 515 XX 2.20 4 i 5 88.49 ■tnuex-ca. 5. v 1 St 516 TVT 5.98 2. 14 86 89 tn-OUCX YVOVUL 2.58 nf«f r 517 xm 2.68 3.26 85.15 TVO-V\JL VocnruL 8.91 t ? 518 xm 1.64 3.13 89.11 vvo^r-u. VoCTrUL 3.92 5I3> nr 3.25 CL 355 85.51 Vfcj-oOoCX J .38 5.25 i i i 520 • XX 2. \o 1.2 5 88.36 Itr^jcvxu. J 2.43‘ ? \ 4 d-0 o 52 1 XX \.2\ 2.oo 89.83 bvOvOi. 4.26 522 ZKB \.\1 5 31 91.38 tr\^oucx ■wcnnjL 3.14 5 23 Ml \. \ 5 .11 91.51 •troo.-c.jL \ .1.5 3.38 524 vox 3. \ 8 \ .5 5 90.6 \ ItN-CUOl J \.19 2.93 525 YYY \. 33 .1 \ 91.91 .11 90 2.13 — W 526 MI 2 .\ 8 \. i4 9 \.\5 42 \.v9 3.o6 527 XX7U 3.62 5.21 81 03 .39 392 528 xxxni 3.34 4.33 9o.vo \ru3uca. VLCrv\X \. 13 12 52SB xssm’B 4.84 514 85.42 trooucL WCnrx*. 4oo 2o 528C XXV ITT C 1.41 1.31 82.89 .25 -rsjaryyjL 2.02 32 529 ffll 2. \ \ 5. \3 81 4 \ 2 . 1 \ VvOT-UL 5.35 5-1 52o .65 \. 21 92.83 "troo^cx .21 5.25 li-^o 5 3 \ m \ \ \ \ 38 9395 ,\9 A9 3.\9 532 8 34 5.22 83.26 trvXXCX i.34 \ .48 5 33 l . 56 \ 36 93 2o XnxMU. 66 3.22 534 mFi 2 62 2.93 92.61 •OOvu \ .94 i \3 535 6 92 C 3. \ v 85.53 .66 3.18 536 XXEH i 08 \.\5 9\ v \ trooucji 1 o4 5 62 537 XSX3H 88 no 9\ .5 1 2 .Ol \ 28 2.52 538 smm 2.\ 5 2.1 3°^ 9o 84 AS 4.43 Fig. 29. Analyses of the Standiford P. C. Co. marl by Delos Fall. See Fig. 28. ATHENS, M!GH. Vol. VIII. P art III. Plate XX, \ / LIST OF LOCALITIES AND MILLS. 305 Av\o.l\jS es of K ^ mon.LeKr ani Cla. ^ton LcxK e/Wa'fU.T^S, R.8 vs/., b Vj Pg\ 05 F a.Vl Pw.p. t at the /V\c/V\ ill. a. y\ Cw e iCcJl Laborat Of ^,Albvon. Lab K v-«.YY\her Field Mar K. SlOi A\ 2 Oj ColCOj /V\qC0 3 SO, Or^a nit Matter Oep-t K vnf eet 539 v m 2.30 2.18 90.83 .30 1.11 3.02 540 XL \ 20 2.80 88.82 troOuCJL 1. 26 5.92 541 XLl S3 11 92 29 tr\JOuCA S5 592 542 xm l IS 96 33.98 truOuOL 54 2.14 542 xl in 22 11 91 .40 \ .33 .5 l 5.12 544 XLET .6 8 1 vS 31.85 2.oo l 10 2.59 545 Xuv 2.iO l o4 90.21 troojafi- .63 602 5.46 XLVL .41 90 88 82 1.44 91 5.35 541 nm 2.o5 2.65^ 81.33 trvo-cn. .32 1.65 548 JlTE 5.59 2 .6 5 82.05 tn-ouct .31 9.36 549 XLU V.43 i.8l 89.49 2 41 \.o9 3.11 550 L V. \2 1-2 1 89 39 39 61 1.22 551 LI .45 1.43 89.60 i 11 14 6.03 552 lh 1 V.45 4.83 11.94 Lrwouet 2 28 1.12 553 lie .60 91 90 84 1.39 lo 4.18 £ 554 let 58 l 24 90 50 1 i3 81 569 555 LY io.63 b 3.41 6S.o2 1.11 2.18 3.8o 556 l m .51 l 21 91.41 .12 329 u 551 LTH \ 21 \ lO 89.33 1.39 6\ 3.96 u 558 LIT 1.23 9o 8106 t>VCMCj». .69 9.12 1 » 559 LIX 2.93 4.22 81 06 .3 19 4.10 d -0 560 LX 2.2o 2 32. 30.01 .5 13 4.24 56\ LH 4.31 242 8844 trvouca. 'oora 2.33 562 lsi 3.91 2.60 86.25 1 5o 91 4.11 562 Lxm 2.3o 2.95°“ 8188 Tjuojca l-\3 514 564 LW 3.14 2.85°’ 81.33 -TLCnru. V.22 4.36 565 LX3. 2.38 2 36 81.25 lo 13.3V 566 LHI 6.21 1 oo 246 85.68 tlrvojcai 19 2 86 56 7 lxei 3.54 l 8o 91.19 truo-t*. 52 2.9 S 569 lot 2.oo CL 3.o3 82.58 19 \ 1 66 5H lie 6.16 .81 88.57 V\joucjl 5o 475 512 LTTU 1.11 1 36 9224 .5 .61 4 06 513 lxm \.3o 99 9 1 21 tn-ouCA G So 514 L£5£? V.39 \.9V 1 1 93 1 5 y ^v-O nr 'A 1 44 2 6 515 LOT \ 25 1 ss* 9o.A2 \9 6 29 5 16 1 XXVI 1.15 2.85 °‘ z 'm > 90 65 11 4.04 512 l mu 2 H 4.21 85.68 30 61 3 69 519 Lim 250 2 48 90 54 trvjo-ca. 8 2 5o 5 So LUX 4.42 \ 04 88 83 Xruoutii. 89 4 81 SSI L 033. \ 55 93 92 89 'trv.«=>g» 34 4 29 582 i_xxsr 9i \ 13 81 33 52 9 51 -S&3 L_ X_S3.TR ... l^- -3,g>L ... 90 \0 lO-Onr-vx \ i9 3v9 Fig. 29. Analyses of Fig. 28 continued; where ever in column 4, there are are two sets of figures, the upper is for the total iron and alumina, the lower the alumina alone. 39-Pt. ! IT. 306 MARL. The material in every ease was dried at 100 C. to expel moisture and sampled. The tables give the amount of each substance found in 100 parts of the dried samples. The samples were taken with apparatus especially prepared for the purpose by which it was possible to know accurately the depth from which the samples came. The average analysis is given among the other analyses by Prof. Fall, page 352, and “exclusive of organic matter the carbonate of lime in these eightv-four samples would average 93.10$, a showing which is remarkably' high when it is taken into account the nearly five per cent of clay which the marl contains.” The marl is also very finely and evenly divided. One peculiarity of this bed to which he calls attention is that a small percentage of clay, ranging from two to six per cent, is found as an admixture with the marl. This of course would be no detriment to the quality of the cement, for as he remarks, the per- centage of magnesia and sulphuric acid are insignificant. Bellaire Portland Cement Co. This company has been but recently organized to operate near Bellaire, Antrim County. The conditions will be not unlike those at Elk Rapids. It is, I presume, a successor of the Lake Shore com- pany. West German Portland Cement Co. Articles filed at Ann Arbor Aug. 13, 1902. Capital $1,000,000, half preferred. Object to manufacture cement, coke and peat in Lima Township. Linus E. Leach of Detroit appears to be chief stockholder. The marl beds are around Four Mile Lake. Locations reported by Douglas Houghton Survey. The first geological survey of the State back in the forties, paid considerable attention to the location of marl, not, however, for its value for cement, but as a fertilizer, though at that time it was also extensively used for making quick lime. A brief summary of the locations which they noticed, should be given here, since the reports in question are not only out of print, but not easy to obtain second-hand. I have added foot-notes calling attention to the fact, when the locations have since been utilized. That marls were generally attributed t^ shells may be indicated by the fact that the symbols used on the maps to denote the location of marls was a small figure of a shell. First Annual Report, 1838, H. D. No. 4, pp. 276-317, No. 14, pp. 1-39, p. 13 or 287, description of Ohara deposits pear Grand Rapids, LIST OF LOCALITIES AND MILLS. 307 in Saline Springs; p. 34 or 306, northern part of St. Joseph,* Monroe near Monroe, f and Jackson County.! Second Annual Report, 1839, H. D. No. 23, p. 393, White River, p. 451, p. 464, Sec. 15, T. 1 N., R. 1 W. 33 Leslie township.§ 35, T, 2 N., R.l W. p. 479, Sections 2, 4, 5, 8, 22, 27, Plymouth Township, Oakland County. Sec. 9, Canton. 495, Monroe, Sections 7 and 9, T. 6 S., R 9 E. From Third Annual Report of State Geologist, p. 94 of the separate edition. MARL OR BOG LIME AND TUFA. “That variety of the mineral which is here designated by the name of marl, is chiefly a carbonate of lime, or lime combined with carbonic acid. It is frequently argillaceous, and mixed with earthy and carbonaceous matters. Throughout the counties enumerated, this mineral is found only in the gravels, sands and clays which overlie the rocks, and may be defined as an alluvial deposit from the waters which have percolated soils charged with lime. On reaching the surface, the water parts with a portion of its carbonic acid, arid becomes no longer capable of holding lime in solution, which is then deposited in the form of a pulverulent, chalky sub- stance, in the beds of lakes or beneath the peat marshes. “As carbonate of lime is a constituent of the covering of mollus- cous animals, these circumstances are favorable to the collection of great numbers of shells, so that these not unfrequently constitute even the main portion of the bed itself, which may then receive the name of ‘shell marl.’ “That form of lime which is called tufa, has a similar origin. It differs in external character, being hard, light and porous, and is that which is familiarly known as ‘lioney-comb lime.’ This char- acteristic difference is the result of circumstances, not of composi- tion. Tufa is formed in situations which allow access of air, when a strong union of the particles takes place. Marl being always deposited under water, or beneath the peat of bogs, the surrounding *Seep. 312. 1 See p. 312. % See pp. 291, 309, 315. §See p. 316. 308 MAUL. fluid prevents cohesion. This condition is that which is very com- monly designated as bog lime.” p. 95. “Thus, according to circumstances, we find a variety of forms assumed by these deposits, from a ‘tufaceous marl,’ in which the particles have but partially cohered, to a hard ‘tufa’ or tra- vertin rock, appearing as ledges in exposed hillsides. “All these recent fresh water limes exist in great abundance in most of the counties enumerated, as well as throughout the interior of the State. In the northern part of Hillsdale,* and the counties of Washtenaw and Oakland, in particular, so extensive and uni- versally distributed are the beds of this useful mineral, that an attempt to ascertain and enumerate all the places in which it exists, is unnecessary, if not impossible. “But notwithstanding its wide distribution, the uses, and even the existence of this mineral are so little known or heeded, even by those who have most reason to appreciate its value, that I shall adventure some remarks upon its application to practical purposes, and the method of ascertaining its presence. “For making quicklime, the value of marl and tufa is already appreciated in those parts of our State which, like the counties under review, are nearly destitute of lime rock. Consequently these have supplied the deficiency, and been applied to all the pur- poses of the best rock lime. Though somewhat inferior in strength, the lime thus obtained is even preferred for particular purposes. It is said, for instance, to be preferable as a wash, owing to its superior whiteness. Its real value is frequently underrated from its not being sufficiently burned; marl being erroneously supposed to require a less degree of heat than limestone. “Some of the largest deposits of tufa I have met with are formed along the banks of the Huron Valley, between Ypsilanti and Dexter, at several of which, large quantities of lime are manufactured. “The circumstances which may give rise to the formation of either tufa or shell-marl, where the same source of supply exists, are hereby exemplified. Ledges of tufa occupy the elevated sides of the valley, while copious springs discharging from its foot, occasion a peat morass between it and the river, beneath which is a body of soft marl several feet in thickness. “Impressions of leaves and branches of trees, and even bones of animals, are numerous in some portions of the tufa, these sub- stances have evidently served as nuclei around which the particles of lime were deposited from the water of the springs, thus both giving an interesting character to the bed and illustrating its formation.” p. 98. “After this recommendation of marl, it may be expected that I advise under what circumstances to look for it. Marl is fre- quently to be recognized by its light ash color, about the margin and occupying the shallows of lakes. In general, the marl which is most obtainable, will be found to be overlaid by peat or muck of the marshes, often at a depth of several feet. Sometimes its presence, under these circumstances, is indicated by a slight coat- ♦Location of Omega plant. LIST OF LOCALITIES AND MILLS. 309 ing of lime visible upon the vegetation on the surface. The growth of the marl bed often causes the overlying bog to swell up into a protuberant form. But such indications are not always visible, and then trial may be made by thrusting down a pole or rod through the peat, when sufficient of the marl, if there be any, will adhere, usually, to make known its presence. “Every farmer ought to examine well his marshes with this view, and if there is any reason to believe that marl exists there, to test the question fully by digging. “It may be advisable to raise the marl in the fall and subject it to the action of the winter’s frost, in order to bring it to a pulver- ized state previous to use upon land.” p. 55. “Marl, which is more universally distributed than any other of the calcareous manures of this district, and which will, in consequence of this fact, admit of a universal application, is in itself more valuable for this purpose than limestone, since it gen- erally contains vegetable and animal matter in combination, and its effects are more immediate. It exists in a state of minute sub- division, and is in a condition prepared to become directly a con- stituent of the soil, while it is necessary that limestone, as well as gypsum, should first be reduced to powder.” Marl . “Deposits of marl were found in nearly every town in the coun- ties under consideration, occurring in beds and banks of lakes and streams, in marshes, as well as occasionally, on the more elevated and dry lands, at a considerable distance from water. “This latter position is not unfrequent, but marls found in this situation, invariably show that they occupy what has heretofore been the bed of of some lake or pool. Thus the marl does not seem to be confined to any particular soil or geological position. “For further particulars, respecting the origin and formation of marl, I refer you to Mr. Hubbard’s report.” Local Details of Marl. Jackson County. “Shell-marl occurs more or less abundantly in the town of Napo- leon, on Sections 12, 14, 15, and 19, and other deposits of minor importance were also noticed in this town. “In the town of Columbus, marl occurs, forming very extensive deposits in the vicinity of Clarkes Lake. It also occurs abundantly on Sections 8, 9, 13, 19, 28, and 29, in the same town. Several of these deposits have an area of more than one hundred acres.* “Several very extensive beds of marl were noticed in the (P. 56) town of Liberty, on Sections 11, 13, 23, 24, and 27, as well as in the bed of Powell’s Lake and its vicinity. ♦These deposits are not far from the Peninsular plant. 310 MARL. “The town of Spring Arbor abounds in extensive beds of marl, which were more particularly noticed on Sections 21, 28, and 29.* “ Hanover. — A bed of marl having an area of more than one hun- dred acres, was noticed, forming a portion of the bed and banks of FarwelPs Lake. Inexhaustible deposits of shell and tufaceous marl occur near a lake which forms the head of Kalamazoo River.f “Town of Sandstone. — Marl is not unfrequently met with in mak- ing excavations in the marshes in this town. It was noticed near the village of Barry, and also on the farm of Hon. Mr. Gridley. “Pulaski. — Marl occurs in abundance in many of the lakes and marshes of this town. A very extensive bed of shell and tufaceous marl was noticed on the farm of Isaac N. Swain, Sec. 2, occupying an area of more than 60 acres, and having a thickness exceeding six feet. An extensive bed was also noticed on Section 25. “Rives. — A somewhat extensive deposit of marl occurs on Section 9. “Leoni. — Marl, in inexhaustible quantities, occurs near the outlet of Wolf Lake, and also upon Sections 4, 11, 12, 22, and 23. “Town of Jackson. — Marl occurs in this town, in abundance, on Sections 20, 21, 26, 27, and 31 (town 3 south, range 7 west), and also on Section 31 (town 2 south, range 1 west). “Concord . — Several extensive beds of marl occur in this town which were more particularly examined on Sections 8 and 9. Also in the bed and banks of the Kalamazoo River. “Grass Lake . — On Sections 13 and 29 , % in this town, extensive beds of shell-marl were examined. “Springport. — An extensive bed of marl occurs on Section 15. “Tompkins. — An extensive bed of shell-marl was examined on Sec- tion 17, in this town.” Eaton County. p. 57. “Kalamo. — Several very extensive beds of marl were ob- served on Sections 22 and 25 (town 2 north, range 6 west), and on Sections 19 (range 5 west).” Kalamazoo County. “Texas. — Shell and tufaceous marl occur in beds of several lakes in this town. Also on Sections 31 and 32 of the same town, is an extensive deposit of this mineral. “Alamo. — On Sections 1, 9, 12, and 24, extensive beds of marl were examined. “Cooper . — Marl is not unfrequently met with in the alluvial lands in the vicinity of the Kalamazoo River. “Ross. — Marl was noticed in several of the lakes and marshes of this town. *This is the location of the Pyramid Portland Cement Co.; the Peerless have recently bought beds here. tThese deposits are not far from the Omega plant. ^Location of the Zenith plant. LIST OF LOCALITIES AND MILLS. 311 “Kalamazoo . — Tufaceous and shell-marls occur in a large marsh and in the valley of a small stream northwest from the village of Kalamazoo.* “Chester . — Extensive deposits of marl abound in this town, on Sec- tions 4, 9, 10, 11, 12, and 24.” Calhoun County. “Marl occurs at intervalsf through this county in the alluvial lands of the Kalamazoo River, and pebbles and boulders are not unfrequentlv seen in the bed of the stream, incrusted with a thick coat of tufaceous marl. “Milton . — Marl was observed in this town on the farm of Hon. S. McCamly. It also occurs in several of the small lakes and streams. “Marengo . — Marl is not of very frequent occurrence in this town. An extensive bed was observed on Sections 1 and 2. “Marl was observed in the town of Marshall , near the Hon. Mr. Pierce’s mills. Also, in comparatively small quantity, in the low- lands between the village of Marshall and the Kalamazoo River.” Kent County. p. 58. “Town 0 North, Range 2 West. Tufa occurs in this town- ship in the bed of the Flat River, on Section 26, in a very extensive deposit.^ “Marl was observed on Sections 3 and 8, Township 6 north, Range 12 west. “Extensive deposits of sliell-marl occur on Sections 22 and 23, township 7 north, range 10 west. “Marl was examined in township 8 north, range 11 west, on Sec- tions 13 and 14, in a dry burr oak plain.” Ionia County. il Tufaceous J marl occurs in inexhaustible quantities in the vicinity of Lyon, town, Maple P. O. Incrusted in some portions of the tufa, are quantities of leaves, recent shells, and in one instance have been found the vertebra and other remains of a large snake. “Marl occurs on Section 1, township 6 north, range 5 west; its extent unknown. “Extensive beds of shell and tufaceous marl exist near Mr. Dexter’s mill in the village of Ionia. Also, in the bed and banks of several of the small streams west of Ionia village. “Extensive beds of marl occur on Sections 10, 11 and 22, township 8 north, range 8 wesL “This abstract of the locations of this valuable mineral only in- cludes some of the most extensive deposits. It is sufficient, how- *Probably the site of the original Portland cement plant. +Six places are noted on the map issued by the Douglass Houghton Survey. iThis is tufa and not the bog lime desired by cement plants. 312 MARL. ever, to render it apparent that marl is distributed in sufficient abundance to afford a ready supply for use as a manure, as also for the manufacture of quick lime. It is within the reach of every man to obtain this restorative for his soils or a lime for economical pur- poses; an article of which otherwise much of the country would be nearly destitute.” p. 73 (Report of C. C. Douglass). The great profusion in which the deposit is distributed through the counties of VanBuren, Allegan, and Ottawa, is deemed a sufficient reason for noticing a few of the most extensive deposits. On Sections 20 and 21, half a mile norther st from Mr. Newell’s steam mill, on Maskego Lake, is a very extensive deposit of shell- marl that may be profitably used as a manure on the sandy lands of that vicinity. Extensive deposits of shell and tufaceous marl occur in the valley of the Kalamazoo River, on Sections 9, 10, 16 and 17, township 3 north, range 15 west, of more than one hundred acres. Also on Sections 16 and 17, township 4 north, range 16 west, there is a de- posit of shell and tufaceous marl occupying the area of more than seventy-five acres. A very extensive deposit of marl was examined on Sections 16 and 17, township 3 north, range 13 w T est. Some of the portions of this marl are found to contain too much iron ore to make good quick- lime. Care should therefore be had in selecting that portion of the marl which is free from this mineral. On Sections 13 and 14, township 2 south, range 13 west, marl of a good quality occurs. Fourth Annual Report, 1811, p. 104, marl and peat, p. 122, marl and peat. The remaining references to minor, or at least not extensively exploited deposits, will be arranged according to counties, proceed- ing east and north from Monroe County. Locations arranged by comities. 1. (68) Monroe County . — The marl deposits referred to by the Douglass Houghton Survey above have been also described by Sher- zer in his recent report on the county.* He lists : Claim 422, the largest; claim 161; claim 520 ( ?) ; S. E. quarter of Sec. 24,* Summer- field; Sec. 7 of Exeter; Sec. 9 of Ash; Sec. 9 of London. A good sample of marl taken from along the line of the Detroit and Lima Northern was sent in by C. A. Chambers, in September, 1898. 2. (67) Lenawee County . — Marl has been reported two miles from Britton, Deerfield township, which is probably calcareous clay and not bog lime. *Vol. VII, Part I, p. 200. “No large beds of this substance are known to occur.” LIST OF LOCALITIES AND MILLS. 313 In the extreme northwest corner we have the Peninsular plant, and there is a good deal of bog lime in this region, e. g., Lowe’s Lake, T. 5S.,R. 3E. 3. (66) Hillsdale County . — Besides the deposits of the Omega plant there is said to be a bed of bog lime 11 to 28 feet thick at Nettle Lake, just south of Camden. It is said to have been examined by Mr. Hunter of Philadelphia. Another bed is near Reading. It belongs to the Monolith Port- land Cement Co., whose officers are L. McCoy of Battle Creek, presi- dent ; M. H. Lane of Kalamazoo and I. P. Baldwin, vice-presidents ; H. T. Harvey of Battle Creek, secretary; G. B. Tompkins of Sturgis is treasurer. Another deposit is at Sand Lake, three and one-half miles west of Hillsdale, Sec. 1, T. 6 S 1 ., R. 3 W. 4. (65) Branch County . — The plants of this county have already been described by Mr. Hale, and in connection with factories. The county is very rich in bog lime. See the descriptions of the plants at, and visits to, Quincy, Coldwater, Bronson, Union City. Still other factories are planned at Helmer on the Fort Wayne branch of the L. S. & M. S. R. R. and just west of Coldwater on the river. 5. (64) St. Joseph County . — Marl beds near Sturgis have been tested by the same people interested in Bristol and Turkey Lakes, Indiana. 6. (63) Cass County . — Near Yandalia at Donald’s Lake, Sections 31 and 32, T. 6 S., R. 1 W, is said to be a large deposit, in some places over 25 feet deep. Near Dowagiac, just north of town, in the lowlands of this old glacial drainage valley, is bog lime. It is said that there are 600 acres, running from 18 to 28 feet in depth, with a percentage of from 75 to 84$ calcium carbonate. Near Niles, beside the bed described by Mr. Hale,* on the farm of R. A. Walton, within half a mile of the C. C. C. and St. L. R. R. is said to be a large bed of marl of excellent quality, along a small stream. Marl was also found with mastodon bones near here. Harwood Lake on the St. Joseph county line, about 10 miles from Constantine, is said to be surrounded by bog lime. The circumfer- ence is about 2 miles and the depth in one place over 50 feet. The owner is W. W. Harvey of Constantine. He also owns a 200 acre bed near Bair Lake, Sec. 5, T. 6 S., R. 13 W. *Page 107. 40-Pt. Ill 314 MARL . 7. (62) Berrien County . — As mentioned on p. 154 by Mr. Hale, bog lime occurs in the marshes near Benton Harbor. 8. (61) Van Buren County . — There is marl in Secs. 13 and 14, T. 2 S., R. 13 W. 9. (60) Kalamazoo County . — This is the county where cement was first manufactured, as already described by Mr. Hale,* and Kalamazoo has been quite a headquarters for such enterprises. (Indiana Portland Cement Company, Kalamazoo Portland Cement Company.) Beside the old site and the Hope township regions described by Mr. Hale there are : Sugar Loaf Lake, southeast of Schoolcraft, T. 4 S., R. 12 W. Mud Lake north of Schoolcraft. ‘ Vicksburg, T. 4 S., R. 11 AY. Around Vicksburg in Kalamazoo County there are lakes with abundant marl, and directly west is a shaking bog (bog lime) where the Grand Trunk has had much difficulty in maintaining grade. The neighborhood is heavily covered with drift with a good deal of sand. Near Climax, “100 acres with 20 feet of marl.” 10. (59) Calhoun County . — Homer Lake on the farm of H. O. Cook, and under the lake and under 120 acres of the marsh is bog lime varying from 10 to 30 feet in depth. This bed lies west of the town, and there is said to be a bed of greater area around Kesslar’s Lakes north of town. As mentioned in the early reports there are also marl beds of size in Eckford township. Near the north line of the county, close to Bellevue at Mud Lake are large deposits. See Eaton County. Also in Convis township, on Kinyon Lake (the Creed farm), is a bed. On the Torrey farm one and one-half miles west of Albion is a small bed of marl, said to be about 25 acres, at the center over 17 feet deep, and shallowing gradually to the edges, and covered by a foot or two of earth (muck). This is an ideal of a complete and as Mr. Hale calls it sealed bed. LIST OF LOCALITIES AND MILLS. 315 Tlie analysis by W. H. Simmons was as follows: Silica 00.70 00.82 Iron and Aluminum oxides 1.71 1.86 Calcium as carbonate 87.57 95.18 Magnesia no trace. Sulphurous anhydride .20 .22 Organic matter 7.91 Difference 1.85 1.92 Total 100.00 100.00 It is suggested that this was taken so near the surface as to con- tain an unusual amount of organic matter. The second set of figures is referred to marl free from organic matter. 11. (58) Jackson County . — There is said to be a bed, 2.5 feet thick, of white marl underlain by one of blue marl, on the farm of J. Dooley and neighbors, almost a mile long, about six miles northeast of Albion. The bed in Rives township mentioned by the early survey is ex- tensive, and has recently been tested somewhat. The Michigan Portland Cement Co., a forerunner of the Wolverine is said to have contracted for about 4,000 acres around Portage Lake, which lies in an old glacial drainage channel, surrounded by extensive swamps, and to have planned a private railroad striking the Michigan Central at Munith. A bed of bog lime is reported near Kelley’s Corners. Also four miles from Jackson. 12. (57) Washtenaw County.- — Four-mile Lake, between Chelsea and Dexter, on the line between T. 1 and 2 S., and in Range 4 E. is surrounded by bog lime. It is said to average 36 feet deep of bog lime, 96^ calcium carbonate. About 1,000 acres are said to have been secured, $20,000 having been paid for 200 acres of land. Marl is also reported from Mill Lake, J. H. Runciman, owner. Near Ypsilanti is said to be a bed of 75 acres, 12 feet thick, under 2 feet of stripping. On Sec. 12, T. 2 S., R. 6 E., is a small bed of 25 acres and others similar near by. Such beds are very common. 13. (56) Wayne County . — Wayne County is not likely to contain much bog lime except perhaps in the extreme northwest corner. 14. (55) Macomb County . — The same remarks apply to Macomb County. 316 MARL. 15. (54) Oakland County . — This county being high, and early uncovered by the ice, and full of old lines of glacial drainage, and deep holes occupied by glacial lakes, probably contains much bog lime, — more than has been reported. W. A. Brotherton says that there is marl 40 feet above the stream on Stony Creek, near Rochester. Large deposits are reported near Clarkston. I. J. Hiller is said to have a bed of 75 to 100 acres. A small bed of 20 acres and 6 feet deep is reported at Bloomfield Center, with others near by. The beds near Holly have already been referred to in connection with the Egyptian P. C. Co. 16. (53) Livingston County . — Near Brighton and only a mile from the R. R. are marl beds of about 100 acres and average depth of about 12 feet. Lime Lake, Sec. 36, T. 2 N., R. 5 E. West of South Lyon in Green Oak township are also some bog lime deposits. Two miles north of Oak Grove is a bed of bog lime, on land belonging to Pierce Elwell and others, on the line of the Ann Arbor R. R. There are other beds on the Ann Arbor line north of Howell. The deposits around Hamburg and Lakelands, have been men- tioned in connection with the Standard Portland Cement Co. There are said to be over 1,600 acres of land underlain by bog lime, which in places is 60 feet deep. 17. (52) Ingham County . — There are beds of bog lime near Leslie, one located by H. C. Barden. Analysis of a Leslie marl is as follows: Silica and alumina, oxides 2.60 Ferric oxide 2.25 Calcium as carbonate 72.20 Magnesium as carbonate 85 Organic matter 12. Water 10. Total 09.90 LIST OF LOCALITIES AND MILLS. 317 There is also a small bed covering about 15 acres, in the north- west quarter of Sec. 24, Vevay, T. 2 N., R. 1 W., belonging to Chester Dolbee. It is now cut through by a stream. On the north side Mr. W. F. Cooper found 6 inches muck, 18 inches marl to sand and gravel; then 20 steps south-southwest, near center of marsh, and gravel ; then twenty steps south-southwest, near center of marsh, 9 inches muck ; and over 5 feet marl, without reaching bottom. Here the marl was quite white and pure: northwest of this hole a sample full of shells was taken. In general immediately beneath the muck it was full of shells, and deeper down, became a sludge. Under the microscope Chara material appears to be abundant. 18. (51) Eaton County . — On the line between Eaton and Cal- houn Counties, three miles from Bellevue, there is said to be a bed of bog lime of over four hundred acres, and an average depth of 20 feet. In places it is 37 feet deep and there is but one to two feet of water above it. There is said to be a suitable clay immediately adjacent. Around Lacey’s Lake, Kalamo township, T. 2K, R. 6 W., is said to - be a large bed of bog lime in places 20 feet thick. 19. (50) Barry County. — Mr. Hale has described quite fully the deposits at Cloverdale in Hope township. Near Prairieville, the township southwest, the bog lime deposits exist, but are said not to be enough to start a cement plant upon. To the west at Fish Lake, near Orangeville, there are extensive deposits of very good quality as shown by Prof. Fall’s analyses. There are 200 acres or more of an average depth of perhaps 20 feet. There is from 0 to 2.5 feet of peat stripping on top. The deepest marl appears to be often close to the water’s edge. Though a good bed it is over four miles from the railroad. Just north in Gun Lake, between towns 2 and 3 N., K. 11 W., the relation between the thickness of the marl and the depth of water is shown by the following table of soundings. ‘Chapter VI, pp. 107 to 131. 318 MARL. Depth of water. Of marl. Remarks. 4 10 3.5 4.5 500 feet from the previous sounding. 5 27 No bottom, edge of deeper water. 3.5 5 20 25 feet from shore. 5 1 to 4 North end of the west lake, bottom gravel. 4 9 3 2 3 3 4 2 to 4 7 12 8 17 8 Sand at 31 feet. 7.5 22.5 3 3 Gravel bottom. e r It is obvious that the deeper water does not always have the deeper marl. A little farther north. Cobb Lake, Sections 5 and 8, and Barlow Lake, Sections 7 and 18 contain bog lime (said to be 93$ calcium carbonate). 21. (48) Ottawa County . — A sample of marl from the Lake Shore west of Grand Rapids gave W. M. Courtis, M. E. the following results: (The sample loses 6.376$ of water and volatile hydrocarbon when dried at 100°.) The dried marl contains : i Organic matter 0.790$ Combined with water less organic matter as above 0.235$ Silica (no sand) 2.528$ Tricalcic phosphate 0.150$ Chlorine as sodium chloride. 0.119$ Alumina and a little iron 0.432$ Carbonate of magnesia 1.250$ Carbonate of lime to balance 94.496$ Sulphur none 100.000$ 20. (49) Allegan County. 22.. (47) Kent County . — Beside a number of marl beds in the north part of the county around Cedar Springs, which have been in part described by Mr. Hale, and the insignificant deposits de- scribed in the Douglass Houghton reports, there are other large deposits. Mr. Nellist reports 13 deposits many of them first rate, LIST OF LOCALITIES AND MILLS . 319 though not convenient. The largest he says is in Wabsis Lake, which is in some places over 100 feet deep (T. 9 N., R. 9 W.). An analysis by A. N. Clark gave: Calcium as carbonate 90.30 Magnesia as carbonate 3.21 Alumina and ferric oxide 0.73 Insoluble in HC1, mainly sand. . . 0.94 Difference, organic matter and water 4.82 Total 100.00 Lamberton Lake, Grass Lake, in Cannon township, and Crooked Lake all contain marl. There is also marl in a lake on Section 21, Grattan Township, in two little lakes on Section 8, Cascade Township, and in small quan- tities on Sections 9 and 15, Wyoming Township. The following is an analysis of a Kent County marl from near Cedar Springs controlled by Stewart and Barker of Grand Rapids. It will be noticed that No. 1 is an analysis of the fresh Avet marl, of A\diich \\ T e have very few, almost all the analyses being figured to dry marl, or being made on specimens already pretty well air dried. 1 2 Insolubles (silica or sand) 00.30 00.42 Iron and Alumina .30 .42 Calcium as carbonate 67.66 95.04 Ma gnesia as carbonate . . 1.67 2.34 Soda and loss of dry material 1.27 1.78 Water and organic matter 28.80 Total 100.00 100.00 The second set of figures are as bog lime analyses are often given, and show that this, except for possibly too much organic matter is a A 1 * * * * * 7 ery good specimen indeed. The stripping or top layer contains much more organic matter, — up to 38^. See also Mr. Hale’s de- scriptions in Chapter VI, § 4. 23. (46) Ionia County . — Jordan Lake near Lake Odessa, is sur- rounded by extensive beds of bog lime, which extend into Barry County. The lake lies on the line. The} T are convenient to the P. M. R. R. 320 MARL. Mud Lake just west of South Lyons also contains bog lime. There is also said to be some near Muir. 24. (45) Clinton County . — In the Chandler marsh three and a half miles north of Lansing, T. 5 N., R. 2 W., there is marl, i. e., bog lime, as reported by F. R. Singlehurst. Merle Beach, significantly named, on Muskrat Lake, T. 6 N., R. 2 W., has also deposits of bog lime over quite a large area under about a foot of peat muck. 25. (44) Shiawassee County . — Marl is reported right at Owosso in the Abley Addition, in the southwest corner of Owosso town- ship. Also in the Maple River flats marl is reported in some places over 1G feet thick. Over in Gratiot County around Bannister, Mr. Davis and I did not find any marl along the Ann Arbor R. R. in the extensive marshes there. A deposit is also reported on the farms of M. Carey, R. F. Kay and J. G. Marsh, in Woodhull township. 26. (43) Genesee County . — A number of deposits in this county have already been described in connection with the Detroit, Egyptian, and Twentieth Century Portland Cement plants. Holden and Buell Lakes, Thetford township, T. 9 N., R. 7 E. Mud Lake, Arbela township, T. 10 N., R. 7 E. Marl properties around the above lakes were gathered together by Fred C. Zimmerman and R. Adams of Saginaw. “It appears that the depth of the deposit is anywhere from 10 to 30 feet and deeper. In Holden (Sec. 3) and Buell (Sec. 2) Lakes, where the water is shalloAv, it can be seen in large quantities. All these lake beds consist of extraordinarily pure marl beds of unusual depth and such consistency that it can be pumped from the bottom. “Analysis of the material taken from these deposits made at the Ohio State University, gives the following: Carbonate of lime 89.39 Carbonate of magnesia 1.95 Silica 6.29 Iron and alumina .99 Organic matter 1.00 99.62” The silica is too high if this is a fair sample of the marls, which I doubt. Geological Survey ^of Michigan. Vol. VIII. Part IIL Plate XXL SILVER LAKE MARL BEDS, LIST OF LOCALITIES AND MILLS. 321 27. (42) Lapeer County . — In the Annual Report for 1901, Mr. F. B. Taylor says : “There is a considerable quantity of marl in the county, and localities so far determined are shown upon the map. None of them, so far as learned, are of large enough extent to form a basis of cement works in the present stage of this industry. No beds have yet been found having an extent of over 100 acres. Because of their present unavailability mainly, the marls found have not been tested thoroughly to determine their suitableness for cement. The largest swamps in the county in the eastern and northeastern parts, appear not to yield marl. Marl was formerly burned for lime in several parts of the county, most notably in southwestern Had- ley, and southeastern North Branch townships. There is marl near Orion.” 28. (41) St. Clair County . — This county may have a few small beds of bog lime in the western part but none have been reported. 29. (40) Sanilac County . — This county was reported in Yol. VII of our reports. Marl probably underlies a good many of the swamps, such as the “Stone wall swamp,” in the western part of the county. Some beds have been tested by Cass City parties and are believed to be extensive enough to work. Their proximity to the fine shale exposures of the Lake Huron shore in Huron and Sanilac Counties might be a point in their favor. 30. (39) Huron County . — Deposits of marl are described in Vol. VII, and may be sought from Mud Lake northeast to Bad Axe and east to Parisville. It is not likely, however, that there are any very extensive deposits. 31. (38) Tuscola County .- — Mud Lake in Arbela township has been mentioned in connection with the deposits in Thetford town- ship, Genesee County, just south. Near Cass City there are said to be big beds of marl 10 feet or more thick. Shale clays can probably also be obtained in this region, as there are exposures of the Michigan series. An an- alysis of the marl by Prof. Kedzie runs : Insolubles (silica) 24 Oxides of iron and alumina .14 Calcium oxide (as carbonate 94.32) 52.82 Magnesium oxide (as carbonate 2.56) 1.25 Carbon dioxide 39.16 Difference, organic, etc., (2.72) 6.39 Total : 100.00 41-Pt. Ill 322 MAUL. The variation in the items of difference shows how much of the lime is combined as sulphate or with an organic acid. 32. (37) Saginaw County .—*- There is probably no bog lime, i. e., what the cement companies call marl, in Saginaw County, beyond possibly a few inches in swampy hollows. What has been reported as marl, like that on the Prairie farm, is clay marl, — and only an ordinary surface clay free from grit. For instance, E. Wetzel opened a bed of clay at ZilwaUkee, eight feet down and about twelve feet thick, which gave H. &. W. Heim on analysis: Top. Bottom. Silica and alumina 65.1 63.75 Calcium carbonate 20.4 20.9 Difference 14.5 15.35 Total 100.00 100.00 33. (36) Gratiot County . — Five miles west of Alma is a small bed of 30 acres from 4 to 16 feet deep. Cedar Lake District. 34. (35) Montcalm County . — Montcalm County is full of lakes, many of them containing more or less marl. The neighborhood of Cedar Lake, a few miles east of Alma, shows an interesting variety of occurrence of marl and has been somewhat investigated, as shown by Fig. 31 and the following description : Bass Lake occupies a hollow in the sands. The shore is sandy and there is no outlet. Rock Lake is similar but the surroundings are somewhat more gravelly. Marl Lake has already been re- ferred to by Mr. Davis.* The water is pure, milky white, and this he attributed to the fact that there is a bench about 100 yards wide of pure marl around the lake over which the water is shallow, from two feet down. There is no peat covering over this bed. This we notice in the diagram of soundings (Fig. 31), and we notice too that soundings 5, 9 and 13 outside the shore of the lake show no marl. Mr. Jno. Webster’s estimate of marl on this lake is of marl 12 feet thick, 100 yards wide, over a circumference of 5,186 feet, i. e., 6,220,800 cubic feet. It is quite likely that there is tw T o or three times as much as this. On the other hand, about Cedar Lake, which is slightly higher, the conditions are entirely different. The marl is covered with peat *Page 83. LIST OF LOCALITIES AND MILLS. 323 and at the edge, which is the margin of the lake, drops off to 30 feet depth very rapidly. The peat varies, as we see in the diagram of soundings, from one to five feet thick, being generally about three feet thick. The land south of the railroad rises rapidly 175 feet or more, and the lake lies in a valley in the till. As bearing on the origin of the marl, it is worth noting that an upward pressure of the ground water is shown by three ffowing wells on the road from the station to the lake, which penetrate the drift to a depth of 48 feet. The Fig. 31. Sketch map of the lakes near Cedar Lake Station of the Pere Marquette R. R., T. 10 N., R. 5 and 6 W. temperature of these wells is for the first one 49° F., and for the one nearest the lake 48.3°F. It is quite likely that Cedar Lake is deep enough to come pretty near to the artesian stratum and allow considerable upward seepage. This is of interest in consider- ing the origin of the marl. An analysis by R. C. Kedzie of clay brought to the surface by one of these wells, is as follows : Silica, soluble 60.00 Sand 3.00 Calcium as carbonate* (CaCO.,) 18.31 Magnesium as carbonate (MgC0 3 ) 3.00 Alumina and oxide of iron 14.80 Difference (water or losses) .89 100 . 00 ^ *It is probable that the lime and magnesia are not all carbonate, hence the dif- ference is too small. 324 MARL. This would not be a bad Portland cement clay so far as the anal- ysis goes, but it is one of the common surface calcareous clays and it is not likely that the lime would come twice alike. Cedar Lake is quite a little lower than the railroad station and the water level seems to have been falling. This may have helped the peat to spread more rapidly over the marl, and helps to account for the marl extending considerably above the present level of the lake, a meadow north of the station showing a considerable thickness of marl, analyses B and C. In this latter place it is cut into by the stream. When the material for analyses A and B was taken the marl was 22 feet thick with about one-half foot of muck on top. Of B, which is near the railroad, a sample barrel was shipped to the H. S. Mould Co., for briquetting, and they report that it can be successfully handled at a cost not to exceed 50c a ton. ANALYSES. Analyst. F. S. Kedzie. F. S. Kedzie. R. C. Kedzie. No. A. No. B. No. C. Wet. Dry. Wet. Dry. Clay. Insoluble m titter . . . . 73 \ 26 52.54 1.24 1.58 \ 10 52.36 .97 Alumina j .30 | .60 Iron oxide CaO as carbonate. 60.00 3.00 MgO as carbonate no.. 42.00 3.23 41.12 3.86 Organic, mat.t.ftr .... 1.50 ^ 34 . 60 Ty a.t.flr 29.95 70.05 34.81 65.19 Difference 100.00 100.00 100.00 Around Cedar Lake, Mr. Webster estimates, see Fig. 31, 13 feet of marl on 20 acres, i. e., 11,325,600 cubic feet, and tributary two acres 11,174,400 cubic feet beside. Around Geiger Lake (Fig. 31), he finds six and one-half acres 12 feet thick, or 3,136,320 acres. These estimates are certainly very conservative, and as Mr. Webster himself says, a larger amount of time in testing would have materially increased the amount upon which he could surely estimate. Altogether in the region there is said to be some 700 acres of marl lands, options upon which were held by W. S. Nelson and George Reed. Cedar Lake has furnished some of the material for microscopic tests, and for Prof. Davis’s experiments. LIST OF LOCALITIES AND MILLS. 325 The Cedar Lake marl is very pure Chara lime, and there is just a small residue, in which angular quartz grains, rarely as large as 0.07 mm occur. There is one well terminated crystal of tourmaline, prismatic with blunt rhombohedral terminations, the two ends slightly different in tint, with appropriate refraction, bi-refraction, and pleochroism. Riverdale. Just east of the Cedar Lake district in Gratiot County is River- dale, — where the Pere Marquette crosses the Pine River. The valley here is much too large for the Pine, being an old gravel filled drainage channel of the ice. Southwest of the village, toward a small pond known as Mud Lake, is a swamp covered with peat, underlain with marl toward the center. The peat cover is thicker toward the margin of the swamp. This also is in a region where flowing wells occur at a shallow depth (38 feet). In Mud Lake marl, from near Riverdale, the cell walls appear as dark lines, with more coarsely polarizing matter between as the calcite radiates from them. There are also bodies, fruit or spores (?) with a little greater index than balsam, spheroidal in shape, with yellow polarization colors and a distorted black cross of + character, and a diameter of about 0.04 mm. Olson Lake and a number of other lakes near Howard City con- tain bog lime, and some of them have been already described by Mr. Hale. 35. (34) Muskegon County . — A sample of marl has been sent by Mr. Keating of the Muskegon Board of Trade, but no extensive de- posits have been made public. Mr. Hale found nothing pure. 36. (33) Oceana County. — H. Kennedy is said to have on his farm in Rothbury, a lake whose bottom is rich in fertilizer (?) of a marly nature. Marl is also said to have been found near Pentwater. 37. (32) Newaygo County . — Some of the deposits of Newaygo County have been quite fully described in connection with the Newaygo plant and by Mr. Hale. 326 MABL. A sample of marl from Fremont Lake gave : Prof. Fall. Prof. : F. S. Kedzie. Silica 2.28 3.93 Iron and alumina 1.60 Calcium as carbonate 88.25 (44.48CaO) 79.44 Magnesia as carbonate . . . 1.40 ( 1 . 39MgO) 2.91 Difference, organic and loss 6.47 13.33 100.00 100.00 Direct estimate gave Prof. Kedzie 34.54 C0 2 and 15.36 organic matter and loss. Thus the indications are that in this marl the CaO is combined in part with an organic acid. An adjacent surface clay gave Prof. Kedzie : Silica 38.36 Alumina (and ferric oxide) 22.18 Calcium oxide 13.96 as carbonate 24.96 Magnesium oxide 8.19 as carbonate 17.12 Carbonic oxide 16.45 Difference, organic matter, alkalies and loss .86 100.00 102.62 38. (31) Mecosta County . — In the neighborhood of Barryton are said to be marl beds from 20 to 300 acres in extent (T. 16 N., R. 7 W0. The beds around Pierson are described in Sec. 3 of Chapter VII. 39. (30) Isabella County . — The Littlefield Lake deposit has already been described by Mr. Davis, and in connection with the Farwell P. C. Co. It is probably the best in the county. 40. (29) Midland County . — This county probably contains no large deposits of bog lime. 41. (28) Bay County. — The shale clays of this county, developed by the coal mines, have been mentioned in connection with the Hecla P. C. and C. Co. There are probably no large deposits of bog lime, unless possibly in the extreme north. 42. (27) Arenac County . — There is not much bog lime in this county, though there is some valuable limestone. There may be a little under the marshes of the west end. 43. (26) Gladivin County . — Extensive beds are reported within a mile of the county seat, and there may be others northwest. LIST OF LOCALITIES AND MILLS . 327 44. (25) Clare County. — See description of the beds of the Clare Portland Cement Co. 45. (24) Osceola County. 46. (23) Lake County. — The principal beds here are probably in the possession of the Great Northern P. C. Co., and have been previously described. A bed is reported on Section 3, T. 20 N., R. 14 W. 47. (22) Mason County. — Large beds are reported near Miller- ton on the line of the Manistee and Grand Rapids R. R., near the west line of Lake County. 48. (21) Manistee County , and region of the Manistee and North- eastern R. R. Through the courtesy of J. J. Hubbell, chief engineer of the Manistee and Northeastern R. R. we have the result of their work of investigation which he has superintended, to add to the notes of Mr. Hale on his trip. This region includes: 49. (20) Wexford , and also Benzie, Grand Traverse and Lee- lanau Counties. The collection embraces some 52 clays and marls of which samples were taken, and turned over to us in their case. Many of the surface clays, though calcareous as usual, run lower in carbonates than those from other parts of the State as the fol- lowing group of analyses shows, in which, however, the magnesia is so high relative to the lime as to show that the material is dolo- mitic. Such clays will hardly effervesce w 7 ith cold acid, and may thus pass for better cement clays than they really are. Lab. No 592 593 594 595 Mark 65 66 67 68 Silica 61.94 56.64 61.10 59.36 Alumina 11.58 12.18 13.91 12.38 Iron (ferric) oxide. . . . 3.49 3.59 3.62 3.62 Calcium oxide 5.92 8.17 6.32 5.63 Magnesium oxide .... 4.85 4.29 3.91 4.62 Sulphuric anhydride . . .18 .31 .31 .30 Difference, C0 2 , organic, alkalies, etc 12.04 14.82 10.83 14.09 Total 100.00 100.00 100.00 100.00 Calcium as carbonate* . 10.6 14.6 11.3 10.1 Magnesium as carbonate 10.1 9.0 8.2 9.7 Total dolomitic matter 20.7 23.6 19.5 20.2 Difference to balance, organic, alkalies, etc 2.11 3.48 1.56 4.14 ♦Calculations below totals are by A. C. Lane and not by analyst. 328 MARL. W. H. Simmons of Bronson, Chemist, Dec. 17, 1900. The following are some of the clays: No. 1A. East Lake, Sec. 4, 5, T. 21 N., R, 17 W., 4 feet thick ; 1 to 2 feet of sand stripping, a smooth, pink brick clay, surface, high in lime with a trace of sand. No. IB. Same location, 5 to 10 acres of it, 2 feet thick; 2 feet muck stripping, with no sand and but a trace of lime. A smooth blue clay. No. 2. Arendale Hill, Sec. 15, T. 22 N., R.16 W.; extensive quan- tity, with 2 to 4 feet stripping; very high in sand and high in lime. A pink brick clay, but probably with some small pebbles. No. 4. Onekama, Sec. 25, T.. 23 N., R.16 W.; extensive quantity, with light stripping, no sand, but a moderate amount of lime; in appearance like 1A. No. 5. Manistee, Sec. 31. T. 22 N., R. 16 W. ; moderate quantity, with 3 feet stripping, a trace of sand and high in lime ; a laminated, pinkish, fine grained clay. No. 6. Copemish, Sec. 7, T. 24 N., R. 13 W.; small quantity, with 2 to 4 feet stripping; no sand, and a moderate amount of lime; a smooth, pink plastic clay. No. 7. Betsey River (“Aux Bees Scies”), Sec. 10, T. 24 N., R. 14 W. ; quantity large and stripping light; sand high, and moderate amount of lime; a massive, pink clay. No. 8. Horicon, Sec. 9, T. 25 N., R. 12 W.; a limited quantity of clay with heavy stripping; with no sand, but much lime; a blue clay, till clay, with small limestone fragments. No. 9. Carp Lake, Sec. 1, T. 28 N., R. 12 W.; large quantity with light stripping; only a trace of sand, and a moderate amount of lime; a tough, plastic, pinkish clay. No. 10. Duck Lake, Sec. 26, T. 26 N., R. 12 W. ; limited amount with heavy stripping; no sand but high in lime; a pink clay. No. 11. Cedar Run, Sec. 6, T. 27 N., R. 12 W.; moderate amount with 4 to 6 feet of stripping; no sand, and a moderate amount of lime; looks much like 1A; a smooth, pink clay. No. 12. Carp Lake, Sec. 15, T. 28 N., R. 12 W. ; quantity limited and stripping 5 to 19 feet; with no sand but high in lime; compared with No. 9 it appears smoother and more uniform. Geological Survey of Michigan. Vol. VIII Part III Plates XXII. : GENERAL VIEW NEWAYGO PORTLAND CEMENT CO’S PLANT. METHOD OF EXCAVATING MARL. LIST OF LOCALITIES AND MILLS. 329 No. 13. Traverse City, Sec. 28, T. 28 N., R. 11 W. ; large amount, with no stripping; a smooth, pink clay. Analysis: Sand 00.00 Silica 29.06 Magnesia 2.07 Lime (carbonate?) 50.02 Alumina 12.03 Iron (oxide?) 5.02 Difference 1.80 100.00 No. 14. Sherman, Sec. 31, T. 24 N., R. 11 W. ; limited amount; too much stripping; sand but a trace, but lime heavy; a cream-colored clay. No. 16. Platte River, Sec. 14, T. 27 N., R. 14 W. ; amount limited, and too much stripping; with no sand, but a trace of lime; a deep red, plastic clay; would be a valuable clay if favorably located. No. 17. Platte River, Sec. 29, T. 27 N., R. 14 W. ; amount limited, but stripping light; only a trace of sand and lime; a red clay, sim- ilar to No. 16; a valuable clay, but I suspect that the lack of lime is due to leaching, and would not be found persistent. No. 20. North and east of Wexford, Sec. 23, T. 25 N., R. 11 W.; amount small, and amount of stripping unknown; high in sand and lime ; a buff, brick clay. No. 22A. Dean's Mill, Sec. 30, T. 24 N., R. 11 W.; one hundred acres of it, with 1 foot of stripping; no sand and only a trace of lime; a buff clay with roots in it; valuable if not merely a super- ficial layer. No. 22B. Dean’s, Sherman, same section; a large amount di- rectly under 22A; analysis as follows: Sand trace Silica 41.76 Alumina and iron oxides 17.16 Lime (oxide?) 16.00 Magnesia 5.62 Loss (on ignition?) 19.02 Difference 0.44 Total 100.00 42-Pt. Ill 330 MARL. This analysis shows that the clay 22A is probably due merely to superficial leaching, and will be irregular in thickness. No. 22B is a smooth, pink clay with a resemblance to No. 1A, etc. No. 22C. Dean’s, Sherman, same locality and directly under 22B; large amount; analysis as follows: Sand trace Silica 47.68 Aluminum and iron oxides 17.70 Lime (oxide?) 12.50 Magnesia . 5.38 Loss 16.42 Difference 0.32 Total : 100.00 A drab, plastic clay, a shade bluer than 22B. No. 22D. Dean’s Mill, same section; .stripping of 22C; amount extensive; with no sand, and only a trace of lime, according to report, but the sample with the survey effervesces freely; appar- ently the same as 220. No. 23. Northport, Leelanau County, Sec. 10, T. 31 N., R. 11 W. ; large amount, but 3 feet of stripping; no sand and a moderate amount of lime; a good deal like the clay at Manistee, No. 5. No. 24. Meesick, Sec. 11, T. 23 N., R. 12 W.; large quantity with light stripping; no sand, and a moderate amount of lime, much like No. 23. No. 25. Eddington farm, Sec. 25, T. 27 N., R. 13 W. ; extensive quantity with light stripping; high in sand, and moderate in lime; a dark red clay, with some small pebbles. Apparently a till clay of little value. No. 26 A. Stanley’s clay, Harrietta, Part I, p. 3, this report, prob- ably, large amount with 2 or 3 feet of stripping; analysis (compare analysis 24 of Part I) : Sand 0.00 Silica 55.60 Magnesia 12.00 Loss on ignition (?) 15.60 Difference (calcium oxide?) 16.80 No. 27. Carp Lake, Sec. 12, T. 28 N., R. 12 W. ; amount limited and stripping heavy; no sand but much lime; much like 26A. LIST OF LOCALITIES AN1) MILLS. 331 No. 28. Carp Lake, same section as No. 27 ; large amount and no stripping; no sand and moderate amount of lime; one of the common, smooth red clays. No. 29. Manistee, Sec. 30, T. 22 N., R. 18 W. ; large amount and no stripping; much sand and a moderate amount of lime; contains some small pebbles, — a till clay. No. 30. Monroe Center, Sec. 6, T. 25 N., R. 11 W. ; large amount, with 1 to 3 feet of stripping; no sand and no lime; a greenish, somewhat vari colored clay, and if the sample is a fair one, it ought to be a very valuable clay. No. 31A. Russell’s farm, Sec. 22, T. 21 N., R. 18 W.; a very large amount with no stripping; only a trace of sand and a moderate amount of lime; one of the common type of smooth, pink plastic clay. No. 31B. RusselFs farm, same location; 2 to 3 feet thick with No. 31 A as a stripping; with much sand and a moderate amount of lime; these two clays together could be nicely combined in making brick. No. 32. Sherman, Sec. 31, T. 24 N., R. 11 W. ; large amount with light stripping; no sand and no lime; smooth, with a greenish tinge; apparently a valuable clay for cement or other purposes. No. 33. Near Sherman, Sec. 25, T. 24 N., R. 12 W.; large amount with light stripping; no sand and a moderate amount of lime; a smooth, pink plastic clay. No. 34. On Manistee River, Sec. 10, T. 23 N., R. 12 W.; amount unknown, but no stripping; no sand and a moderate amount of lime; a smooth, red clay. No. 35. Bear Creek, Sec. 6, T. 22 N., R. 14 W. ; very large amount with very light stripping; no sand and a moderate amount of lime; a very smooth clay, in color between blue and purplish drab. No. 36. Platte township, Sec. 29, T. 27 N., R. 14 W.; limited amount with light stripping; only a trace of sand or lime; the sample, however, is a common pink clay, probably a till clay, with small pebbles, and free effervescence in acids. No. 37. State Lumber Co., S. W. quarter of N. E. quarter, Sec. 22, T. 26 N., R. 14 W.; high in sand with a trace of lime; a pink clay. No. 38. Jas. Case, Homestead P. O.; southwest quarter of north- west quarter, Sec. 22, T. 26 N., R. 14 W. ; no sand and moderate in lime. 332 MAUL. No. 39. Carp Lake, Sec. 12, T. 28 N., R. 12 W.; see No. 9; large amount, with moderate amount of stripping; a trace of sand and moderate amount of lime; a reddish clay. No. 40. Carp Lake, same location as No. 39; a large amount with moderate stripping; similar in sand, lime and appearance. No. 42A. Brosch Estate, Traverse City, Sec. 1, T. 27 N., R. 11 W. ; 80 acres of it, 2 feet thick; stripping light; analysis: Sand 00.00 Calcium oxide 3.15 Magnesium oxide 0.31 Aluminum and iron oxides 32.79 Silica 60.62 Difference 3.13 Total 100.00 No. 42B. Same location as 42A; 80 acres of unknown depth,, with 2 feet of stripping; no sand and a moderate amount of lime. This appears to be the main bed of reddish clay, of which A is a superficial layer, produced by leaching. They are very similar in looks. No. 43. B. Hoke, Sherman, southwest quarter of the south- west quarter, of Sec. 29, T. 24 N., R. 11 W. ; moderate amount with light stripping; high in sand, and moderate in lime; a blue clay. No. 44. Southeast quarter of the southeast quarter, of Sec. 18, T. 24 N., R.12 W.; with a trace of sand and moderate in lime; a pink clay. No. 45. Corey, Wexford Corners, northeast quarter of the north- west quarter, Sec. 18, T. 24 N., R. 11 W.; with a trace of sand, and moderate in lime ; a pink clay. No. 47A. Lake Bluff, north of Leland, Sec. 4, T. 30 N., R, 12 W.; bluff 200 feet high, with no stripping; free from sand with a mod- erate amount of lime; a typical pink till clay; the sample is not free from pebbles and sand, though they are not abundant. No. 48A. Lake Leelanau, Sec. 11, T. 30 N., R. 12 W.; large amount with light stripping; no sand but high in lime; a smooth, pink clay. No. 48B. Lake Leelanau, layer below 48A; similar in character; very smooth. No. 48C. Lake Leelanau, layer below 48A; similar in character. LIST OF LOCALITIES AND MILLS. 333 No. 52. Sec. 19, T. 24 N., R. 11 W. ; small amount with heavy stripping; a large amount of sand, but low in lime. No. 53. Clay to be used at Baldwin; a smooth, pink clay, freely effervescing; one of the common surface clays. No. 60 is a calcar- eous shale from the bay shore near Petoskey. No. 15 was a clay from the foundation of the new pumping station at Detroit, of the water-works; a green sand clay, freely effervescing. The remain- ing numbers are of marls, many of them not along the line of the road. No. 54 was a very fine specimen of almost pure calcium carbonate, bog lime, or marl, from Baldwin. No. 55 was supposed to be an average sample, perhaps not quite so fine, showing some root marks and some shells. No. 56. Bronson Lake, Sec. 4, T. 26 N., R. 13 W.; a widening of the Platte River; dark blue with quite a number of shells. No. 57 is from Newaygo. No. 58. This is from the “Bigmarsh/’ a filled lake connected with the Betsey (“Aux Bees Scies”) River, Sections 1 and 2, T. 25 N.,R. 13 W. ; said to cover 988 acres, varying from 6 to 35 feet in depth, and (as a result of 600 soundings) to contain about 23,000,000 cubic yards. The sample has a slight bluish tinge, and a good many shells, and appears quite as good as most commercial marls, that of the Peninsular plant, for instance. No. 59. From Carp Lake, Sec. 24, T. 30 N., R. 12 W.; at the nar- rows; a good-looking, bluish marl. Besides these deposits of bog lime, there are in this district : Mr. Farr’s deposit at Portage Lake, Onekama, already described by Mr. Hale. A deposit in the south end of the lake at Arcadia. A deposit in Little Platte Lake, Sec. 36, T. 26 N., R. 15 W. The deposits in Upper Herring Lake, controlled by the Water- vale P. C. Co., already described. A deposit at the head of Carp Lake, Sec. 10, T. 28 N., R. 12 W., as well as that at the narrows, already mentioned. On the north side of Gflen Lake, Sec. 26, T. 29 N., R. 14 W. In Traverse Lake and perhaps Lime Lake, T. 29 N., R. 13 W. 50. (19) Missaukee County. 51. (18) Roscommon County. 334 MARL. 52. (17) Ogemaw County . — Beside the extensive deposits of bog lime described in connection with the Hecla cement plant, and the projected plant at Lupton, the same is reported from : Gamble Lake, Sec. 11, T. 23 N., R. 3 E. Devore Lake, Sec. 11; same township. Sage Lake, a large lake in the south central part of T. 23 N., R. 4 E. As there are probably outcrops on the Rifle River of the Michigan and coal series, it is likely that good shale clays may be found in the county, although as already remarked, the Hecla plant plans to go to the coal measure shales of Bay County for theirs. 53. (16) Iosco County . — Valuable marl beds are said to have been found near East Tawas by H. C. Bristol. The Michigan series of outcrops in this county at Alabaster, and up the An Gres River, and exposures of argillaceous limestone, suitable for rock cement and probably of shale clays for mixing in Portland cement occur. The following analysis, is I think, from one of these clays, taken from near Alabaster and Sherman, and analyzed for R. A. McKay of Bay City by F. S. Ivedzie : Silica 58.95 Aluminum oxide 14.45 Iron oxide 7.60 Calcium oxide 2.94 Magnesium oxide .86 Sulphuric anhydride (S0 3 ) 1.73 Alkalies as K 2 0 t 2.54 Water of combination 7.50 Difference, organic matter and loss 3.43 Total 100.00 Its freedom from lime or grit, and high per cent of silica are valuable qualities. 54. (15) Alcona County . — In the annual report for 1901 some details are given regarding the bog lime deposits of this county (pp. 60 to 65) as follows. The deposit north of Harrisville about three miles, near what is known as Ludington’s Spring, received some attention in the summer of 1899. “Marl deposits in Alcona County, so far as examined, were found to be too thin to be utilized in the cement industry. There is from 6 to 10 feet of rather impure marl in the lake on the township line a mile west of Lincoln. About 6 feet of marl, apparently of good LIST OF LOCALITIES AND MILLS. 335 quality, was found on the border of Tubb’s Lake in Sec. 31, T. 26 X., R. 7 E., forming a platform 10 or 12 rods wide. Marl deposits 1.5 to 2 feet thick are exposed in a railway ditch in a swamp one-half mile north of Harrisville station, and a similar depth in railroad ditches south of Greenbush on the west side of Cedar Lake. “Deposits of what is popularly known in Michigan as marl, but is nearly a pure calcium carbonate, occur at a number of points, at Springport (South Harrisville), Kirk Ludington’s, Sliabno’s, four miles south of the Presbyterian church in T. 27 X., R. 9 E., etc. “A very interesting deposit in some ways is one that is crossing the lake road about three miles and a half north of Harrisville. “It covers probably not less than 20 acres. 1 do not know how deep it is, though it has been tested. “The interesting thing, however, is its mode of occurrence, which is directly in front of the bluffs worn by the lake at a higher level, with no ridge or barrier between it and the present lake. It is wasting away, but the upper part is redissolved and precipitated and passes into a firm and hard calcareous tufa, while as one goes down, it becomes granular and then soft. It appears to be a gen- uine Char a lime formed by the precipitation of lime by the lake weed known as Chara, but it can hardly be supposed to have been found in such purity directly on the beach of a great lake, and we are forced to assume that it is the relic of a small lake, the rest of which has been eroded away. “There is enough, perhaps, for lime kilns, but hardly, I think, for a cement plant; and, beside, it is so hard and granular on top that the advantage of marl, its fine sludgy character suitable for mix- ing, would be lost. “In general, clay deposits of fine and uniform texture are rather rare in Alcona County, but in the southeast part there is con- siderable clay that carries but few pebbles, an area of several square miles being found around Mikado and northward from there between Gustin and Killmaster. Although an unsuccessful attempt has been made to burn a kiln of brick at Mikado, it seems probable that the surface clay will in many cases prove suitable for brick or tile. It will at least be worth while to experiment further with the clay, for that part of the county would be greatly improved by underdraining with tile, and it will be an advantage to manufacture the tile where it is to be used. “At South Harrisville, Sec. 32, T. 26 X., R. 9 E., some brick has been made of glacial clay. It is not entirely free from pebbles, effervesces somewhat and makes cream-colored brick. Some brick has been made also at Mikado (West Greenbush) of a similar quality. All the clays of the county are Pleistocene or surface clays, and it is the almost universal rule that such clays have more or less calcium and magnesium carbonate. Generally the top of the bed is free from carbonates, which have been leached out. “Just at Harrisville the stream falls over a smooth, well-bedded clay, apparently an old lake clay free from pebbles. If not, it could easily be washed free by the stream.* Back of Sturgeon *As at Sebewaing, see Vol. VIII, Part I. 336 MARL . Point, on Sec. 25 ; T. 27 N., R. 9 E., are fields where a similar clay appears to be present, but it is particularly well exposed where the Black River opens out from the hilly moraine country to the swampy land of the old lake bottom on Sec. 3 of the same township and Sec. 34, just north. Here a calcareous clay is extremely well exposed in the bed of the stream, appearing almost as though it were bed rock. But probably the same bed of clay also appears in the bluffs of the outer valley above the flood plain and at least 10 feet above the river. Here it is light reddish in color, does not effervesce with acid, lies close to the abandoned track of an old logging rail- road, and could be readily worked. I should think it would make unusually good brick and tile (see analysis below), though it is quite possible that farther working and testing with auger would show that in going deeper more lime was encountered. Still it is quite likely that an important top layer may have been leached free. “There are indications that similar clays occur all the way along, but somewhat below and nearer the shore, the highest former shore line of Lake Huron (645 to 655 A. T.). “The An Sable River also flows at several points over firm, well- bedded pink clays, apparently free from pebbles but full of lime. A good place to observe them, however, is in a little side stream at Bamfields, Sec. 11, T. 25 N., R. 5 E., where they are well exposed. “Three typical samples of the clays were sent to the McMillan Chemical Laboratory, Albion, for analysis, and the following re- ports were received from Prof. Delos Fall : 910. 911. 912. Millbury. Free sand 13.98 11.53 38.55 Combined silica 27.60 25.71 22.73 41.58 37.24 61.28 61.03 Alumina 12.58 7.08 16.37 Oxide of iron 3.59 3.99 5.59 18.10 6.65 Calcium oxide 13.04 17.70 2.33 1.29 Carbonic oxide 17.26 21.00 Calcium carbonate 23.28 31.60 Sulphur anhydride 0.41 0.41 0.67 1.55 Magnesia* 6.44 6.52 1.21 .53 Org. Matter 3 72 3.46 9.14 9.20 98.62 97.40 96.59 Difference, principally alkalies 1.38 2.60 3.41 100.00 100.00 100.00 “No. 910 is the ordinary calcareous clay or marl of the district from Black River near the water level, and is free from pebbles or grit. It will be seen that it is composed of about one-third very fine sand or rock flour, one-third clay proper, and one-third dolo- mite. It may be used for making brick, but will yield a light brick that will not stand hard burning. *There is not enough C0 2 to combine with all the lime and magnesia, hence there is probably some hydrous magnesian silicate present. L. LIST OF LOCALITIES AND MILLS. 337 “Samples of clay from the Au Sable valley appear to have similar composition. “No. 911, from the old brickyard southeast of Harrisville, is a typical tile clay and contained some small limestone pebbles. It will be seen that it contains even more lime, nearly half. Neither of the clays appear to be suited to the higher uses for clay. “No. 912, the third clay, which lies over No. 910 and may be de- rived from it by solution of the calcareous material, is of an entirely different character. If the silica is finely divided enough, and I think it is, it would make an excellent clay to mix with Portland cement for the manufacture of marl. I have given an analysis of the Millbury, Ohio clay, which is largely used in the State for cement manufacture, for comparison. “It would also make an excellent grade of red brick, and very probably also paving brick. It remains to be seen by a series of borings how much of this clay there is, but in all probability field tests showing whether it effervesces with muriatic acid will be sufficient to show this.” 55. (14) Oscoda County . — There are probably considerable beds of bog lime, but little is as yet known of them. Some is reported about 5 miles west of Luzerne, near Tyrrell, T. 26 N., R 1 E. 56. (13) Crawford . — There is a deposit of bog lime close to Grayling. The analysis given in the Agricultural Bulletin, No. 99, is in error. An analysis of the Grayling marl by W. M. Courtis, M. E., is as follows: Water lost at 100°C 61. Dried marl 49. Moisture 0.60 Organic matter 9.80 Insoluble silica 0.78 Soluble silica 0.13 Ferric oxide 1.13 Alumina 0.07 Calcium carbonate 87.00 Magnesium 0.91 Sulphuric acid 0.27 100.69 43-Pt. Ill 338 MARL. Same sample figured without the organic matter is: Calcium carbonate 97.00 Silica 1.01 Ferric oxide 1.26 Alumina 0.08 Magnesium carbonate 1.01 Sulphuric acid 0.30 100.66 57. (12) Kalla ska County. 58. (11) Grand Traverse County. — Some of the beds of this county have been described in connection with the Elk Rapids Portland Cement Co., and the work of Mr. Hubbell for the Manistee and Northeastern R. R. 59. (10) Benzie County . — The beds of this county have been largely described in connection with the Watervale plant, by Mr. Hale, and in connection with the Manistee and Northeastern R. R. A hed of 100 acres is reported near Aral. 60. (9) Leelanau County . — Some beds of this county have been referred to in connection with the explorations of the Manistee and Northeastern R. R. 61. (8) Antrim County. — The Elk Rapids factory is located in this county. Beside this company the “Lake Shore Cement Com- pany” (G. W. Davis of Mt. Pleasant, S. B. Daboll of St. Johns, and others) , have obtained options on the lime deposits of Intermediate (which Mr. Hale has described, p. 142, as Central Lake), Grass and Clam Lakes, some 600 acres in all it is said. The shores of this county contain, as I believe, valuable exposures of shale clay. 63. (6) Otsego County. 64. (5) Montmorency County. 65. (4) Alpena County. — There is marl, limestone and clay, abundant in this county. See the description of the Alpena Port- land Cement Co. and the Annual Report for 1901. The limestones I believe to be especially valuable. LIST OF LOCALITIES AND MILLS. 339 We have also an analysis of an Alpena County marl, by W. M. Courtis: Carbonate of lime 74.48 Carbonate of magnesia 0.50 Silica 7.20 Alumina 0.54 Ferric oxide 2.36 Sulphuric acid 0.89 Organic matter 12.88 Water 1.25 100.10 66. (3) Presque Isle County . — The conditions of Alpena County are repeated here. The following are analyses of a limestone and yellow clay shale south of Rogers City, which I owe to Mr. J. G. Dean of Hassan, Tagge, and Dean of Detroit : Limestone. Silica 0.62 Alumina and ferric oxide 0.20 Calcium carbonate 98.34 Magnesium carbonate 0.45 Sulphur anhydride tr. Organic 0.09 99.70 Shale Clay, Yelloic. Silica 66.39 Alumina 13.60 Ferric oxide 5.87 Calcium oxide .99 Magnesium oxide .50 Sulphur anhydride 1.00 Organic, etc 10.32 98.67 There are said to be marl beds half a mile from the court house. Only about two feet of stripping are said to be needed. 340 MAUL. 67. (2) Cheboygan County . — Very extensive deposits of marl are reported on Black (otherwise called Cheboygan) lake. The little steamer Eva of the Onawav-Oheboygan mail line is said to plough through acres of it in the bed of Lower Black River, be- tween Stony Point and Taylor’s Landing. Limestone also occurs frequently. Beds are reported 5 miles from Mullett Lake, and at other places. A sample has been sent me, said to come 7 miles from Wolverine, close to the county line, west and a little south, T. 33 N. ; R. 3 E., probably near the Cobb and Mitchell lumber R. R., extending over 160 acres, and said to be enough to make 16,000 barrels of cement. This deposit or another near by has been described as occur- ring in a dried up lake which has a trout stream flowing from it, but no distinct inlet in an area perhaps 40 rods by 160 rods, and in thickness 7 feet or so. It is bluish, effervesces freely, but not as rapidly as many marls, and appears to be clayey or magnesian. Turns the acid somewhat amber, mainly from a little organic matter. 68. Emmet County . 69. JJyyer Peninsula . — The marls of the Upper Peninsula have been relatively little investigated. Beside the account given by Mr. Hale of beds near Munising, Wetmore, Manistique, Corrine, we have the following notes: At the World’s Fair in Chicago Dr. W. H. Tucker made an ex- hibit of the marls from Naubinway, Mackinac County. He reports them to exist in large quantities, lying in a bed some 10 feet in thickness, which is overlain by but 6 inches of a mixture of marl and vegetable mould. The analysis is reported by him to be as follows ; Insoluble matter 3.25 Alumina and iron 0.52 Carbonate of lime 92.79 Carbonate of magnesia 2.27 Organic matter by difference 1.17 100 . 00 * Compare the deposit at Corinne near by described by Hale, p. 140. A large deposit of marl is said to exist close to St. Ignace, dis- covered by John Prophet. ♦Report of the Board of World’s Fair Managers. LIST OF LOCALITIES AND MILLS. 341 Near Manistique large deposits of marl are said to be controlled by the White Marble Lime Co. North of Menominee, T. 34 N., R. 26 E., there are said to be lakes with marl, and near Stephenson. A very good sample has been sent in by W. B. Rosevear from Drummond’s Island. Houghton , Houghton- Count g . — A small amount of marl was found in making some excavations at West Houghton. Fig. 32. Section of marl deposit at Houghton. I. Ordinary forest swamp surface, ground covered with grass. II. About 2 feet of old rotten timbers fairly well compressed. III. About 6 feet of very fine peat, pretty solid. IV. Marl full of small shells. V. Fine clay. VI. Glacial drift. Mr. W. Y. Savicki gives the section of a deposit shown in Fig. 32, perhaps the same, not far from the old Atlantic Mill, but in the thickest place exposed only about a foot thick. It was exposed in a ditch dug in 1899 for the Houghton water supply. Mr. Geo. L. Heath made the following analysis : The marl was first dried at 105°C. As the organic matter is determined by difference and as it is not altogether certain that the lime is combined as carbonate, or the potash and soda as carbonate instead of silicate the difference 6.81 may be less than the real amount of organic matter by a trifle. 342 MAUL. Silica Iron oxide and trace of alumina Potassium oxide Sodium oxide Calcium oxide Magnesium oxide Calcium sulphate Loss on ignition, C0 2 organic matter, etc .... 1.85 1.18 0.27 as carbonate 0.40 0.19 as carbonate 0.32 48.88 as carbonate 87.28 0.78 as carbonate 1.64 1.02 46.33 difference 6.81 Total 100.50 Concluding Remarks . It will be easily seen that the foregoing notes are a very uneven and imperfect account of the deposits of bog lime in the State. Yet they are enough to show why they are so imperfect. Deposits of bog lime are everywhere present in the State, though most often probably in the higher parts. They are often covered by swamp, and often difficult of access, being neither water nor yet land. The resources of the State Survey are entirely inadequate to make a systematic study of them, and we have depended largely upon the investigations of private parties, incidental observations, and the hasty summer’s work of Mr. Hale. Enough has been learned to bring out some salient points, how- ever, and show that there is no lack of marl which should be more properly called bog lime, in most of the State. It is much higher in lime, and lias but few transitions to the calcareous clays, which are abundant in the State, and have just as good title to the name marl, but usually run from 30 to 40 $ of carbonates. Deposits of surface clays which run low in carbonates are rare, and there is usually a fair percentage of magnesia. An important exception are the clays which are but weathered shales. I have concluded to append together instead of scattered through the text according to locations a group of analyses which are due to Prof. Delos Fall of the Board, and at the same time reprint a valuable paper which he presented to the Michigan Society of En- gineers, with thq permission of the Society. Other* analyses by Prof. Fall will be found scattered through the report, having been received at various times from the various parties for which he has executed them. It is apparent that he and his pupils have been con- nected with many of the successful enterprises in the State. Namely: pp. 136, 154, 304 , 305, 336. MARLS AND CLAYS IN MICHIGAN. 343 MARLS AND CLAYS IN MICHIGAN.* BY DELOS FALL, SC. D., ALBION COLLEGE. Portland cement is a chemical compound resulting from the burning at a temperature of about 3,000 degrees Fahrenheit, of an intimate mixture of a certain definite proportion of pure limestone and clay of a definite condition and pure quality. The limestone may be of a solid and crystalline form or in a finely divided condi- tion, in which it is found in the so-called marl beds of Michigan. The clay may be solid shale or the plastic variety that has resulted from the disintegration of the parent rock. What constitutes a pure quality for clays and marls, and what the proportion in which they are to be mixed is a question which the chemist alone can deter- mine. Marl. The term “marl” from the mineralogical standpoint, is a mixture in any proportion of limestone and clay. In certain parts of our country, notably in the southern states, but not to my knowledge in Michigan, this mixture approximates very closely to the proportion required for a high grade Portland cement. As the term is used with us, it applies to a comparatively pure calcium carbonate with a certain very small proportion of clay, and it may contain a small percentage of magnesia and sulphuric acid. Deposits of marl are found in the beds of those lakes in Michigan, which, owing to their past histories, are now surrounded by marsh lands, the marl being found in the bottom of such lakes, and ex- tending out under the overlying muck or peat. These beds vary in depth from a few inches at the edge of the deposit to 30 or 40 feet in the center. In places they are almost entirely uncovered, and are exposed to view in such a way that they can be immediately utilized without any expense for the stripping process which must be em- ployed in all cases where there is a covering of muck or peat. Occa- sionally a layer of greater or less thickness is found, consisting of decayed organic matter lying in the center of the depth of the bed, indicating that in the process of the deposition of the marl, the level *From Michigan Engineer, 1901, pp. 124, 133. 344 MARL. has alternately risen and fallen. In some cases it has been found pos- sible, by draining off the water of the lake, to uncover rich and extensive beds of marl with no other expense attached to this stage of the work. The composition of the marl in the various beds in Michigan varies to a considerable extent. One bed, which has been exhaustively examined, gives, on analysis, average of fifty samples, a composition as follows: Silica, Si0 2 .53$ Alumina, A1 2 0 3 .75 Iron oxide, Fe 2 0 3 Trace. Calcium carbonate, CaC0 3 . . . 96. Magnesium carbonate, MgCO s .09 Sulphuric anhydride, S0 3 .02 Organic matter 3.09 99.90 This is extremely pure. Marl is rarely found running so high in calcium carbonate, and so low in clay, magnesia and sulphuric acid. An average of eighty-four samples from another bed resulted in the following composition, which may be taken as fairly to represent the marls of Michigan : Silica, SiO., QO © c4 Iron oxide, Fe 2 O s ) 2.59 Alumina, A1 2 0 3 j Calcium carbonate, CaC0 3 88.06 Magnesium carbonate, MgC0 3 Sulphuric anhvdride, S0 3 32 78 Organic matter 5.30 99.13 Exclusive of the organic matter, the 84 samples average 93.10$ of carbonate of lime. The above bed is characterized by a strain of blue clay accom- panying the marl. This is not a serious adulteration of the marl, except that it will require more constant attention from the chemist in order to produce a mixture of constant composition. The main points of interest concerning prospective value of marl for manufacturing purposes are the proportions existing in the raw material of carbonate of lime, magnesia, sulphuric acid, and organic MAULS AND CLAYS IN MICHIGAN 345 matter. It is desirable that the carbonate of lime should run as high as possible, in order that there may be the largest percentage available of this, which is the most important contribution to the final composition of Portland Cement. Too much organic matter will lower the percentage of carbonate of lime, and clog the rotaries in the process of burning, and, because of this fact, will diminish the amount of the finished product which the rotary furnace is capable of producing per day. With excessive amount of organic matter present in the marl, the total output of a sixty by six rotary furnace might be as low as seventy-five or eight barrels per day, when, with the marl containing less organic matter, say from two to not more than five per cent, the product should be from 120 to 130 barrels per day. The presence of magnesia in the cement must be considered dele- terious to the quality of the cement, from the fact that it refuses to unite with the clay at the temperature required for the burning of the cement, and is left at the end of the process in the form of caustic magnesia MgO. When water is added it takes up that water to an extent which produces a hard product of increased volume, and hence produces a cracking or disintegration of the proposed structure. In general, it may be said that a percentage of not more than two per cent is not considered harmful. The presence of sulphuric acid in the marl and clay, and its effect upon the finished product, does not seem to be appreciated as it ought to be, for it is noticeable that most of the reports of chemists as to their analytical findings give no mention of sulphuric acid in the marl. The chemi- cal analyses should always state the presense or absence of this in- gredient, and the proportion in which it occurs. Sulphuric acid gen- erally occurs in the form of calcium sulphate or gypsum, and it is well known that the presence of it in considerable percentage, say more than two per cent, retards the setting quality of the cement. At the same time, it possesses no hydraulic qualities, but will in the presence of water partially dissolve and thus lead to the final dis- integration of the proposed structure. Michigan Clays. Originally it was supposed that the difficult problem for the initiation of a Portland Cement industry was to discover sufficiently large beds of pure marl ; indeed, it is true that that feature of the problem is not so easy a task as many suppose, the number of beds 44-Pt. Ill ! 34G MAUL. in Michigan being somewhat limited as to quantity, accessibility, and quality. On the other hand, it was supposed that an inex- haustible supply of clay of proper quality could be found adjacent to any marl bed. Farmers and others would point to large deposits of clay which they were sure would prove of sufficient purity and quality for the purpose. By this general tradition, promoters and investors have been led into the establishment of large plants, only to find that they must seek long and sometimes unsuccessfully for clay of the proper material. Not all clay will make a good Port- land Cement.. Clay is essentially a silicate of aluminum but rarely occurs with- out the admixture of iron oxide, calcium carbonate, or sulphate, and sometimes magnesia. While calcium carbonate must be used in the mixture for cement making, its presence in the clay com- plicates to a large degree the problem of the chemist in making that mixture, and more especially maintaining the mixture in a uniform and constant composition. Many of our clays run too high in alumina, making, upon burning, a quick setting cement, not so dur- able and permanent as that produced from a clay containing a less amount of alumina. The following analysis of a Michigan clay will aptly illustrate this point: Silica, Si0 2 G0.1$ Alumina, A1 2 0 3 20.73 Iron oxide, Fe 2 0 3 5.18 Lime, CaO 1.19 Magnesia, MgO .44 Sulphuric anhydride, SO... 3.35 Loss on ignition S.1G 99.15$ This analysis illustrates another bad feature existing in some Michigan clays, namely, the too large percentage of sulphuric acid. The 3.35 per cent of sulphuric anhydride present in this clay repre- sents 5.G per cent of calcium sulphate, and this in the mixture with marl would bring the percentage very near to two per cent, which might be considered to be the limit permissible for that ingredient. The analysis above given would be almost ideal if the alumina ran at from six to ten per cent, and the SO.. was lower or altogether absent. Clays should contain very little free sand, iron oxide, or organic matter. It should have a tendency to gelatinize when treated with MAULS AND CLAYS IN MICHIGAN. 347 acids. The silica must be combined and not free, for the reason that at the temperature at the command of the cement-maker, free silica will not combine to form a silicate of lime which is the essen- tial ingredient in Portland cement. About sixty per cent of the clay should be silica. Three classes of clay found in Michigan are illustrated by the following analyses from my- laboratory note book : No. 1. No. 2. No. 3. Silica, SiO., . . . 49.3G 60.70 71.84 Alumina, A1 9 0.> . . . . ... 10.30 20.92 15.53 Iron oxide, Fe o 0« . . . 3.90 7.06 3.57 Cal. car., CaC0 3 . . . 31.01 .73 .75 Mag. car., MgC0 3 . . 1.77 None. Trace. Sul. triox., SOo 3.15 .60 1.24 Organic matter 1.00 9.89 5.68 100.49 99.90 98.61 No. 1 possesses the great disadvantage of a variable quantity of calcium carbonate, the fact that the relation of this to the clay itself is that of a mere mixture growing out of accidental and there- fore varying conditions making it very certain that no two samples of the same bed would show the same composition. The clay is also imperfect from the presence of a high percentage of calcium sulphate. The temperature at which proper calcination would take place can scarcely be inferred, but it would probably be high. Clay ]S T o. 2 is a good clay; it wi # ll burn at low temperature and be economi- cal of fuel. It is rather high in alumina and would make a quick setting cement. Its setting quality could be retarded by the addition of a small percentage of gypsum. Clay No. 3 is too high in silica; the temperature required for the calcination would necessarily be very high, the excessive tempera- ture being hard to acquire and very disastrous to the life of the rotary. Discussion. Mr. Lane. — There are many kinds of marl in the State, — one kind and another, and the question of clay is an important one. One advice I should give — if I were to give advice — would be to call in the services of a chemist; but before starting to get your chemist, it might be well to make a few preliminary tests yourself. Now 348 MAUL there are two simple tests, which, if applied might prevent two- thirds of the samples sent to my office from ever being sent. In the first test, if yon can feel grit when you chew the clay, feel the clay in your teeth, probably it is not worth investigating. In the second place testing it with hot acid, if there is a good deal of lime, it is likely to mean more or less magnesia, and certainly a good deal of trouble for the chemist. Ordinary vinegar or anything hot and sour, is not bad to test it with, if you happen to be in. the woods. These two tests will rule out a good many clays. Another point is the question of coal to be used, which certainly is important. I think there is Michigan coal that ought to be good enough to answer the purpose. I have seen samples of Michigan coal, taken from near Saginaw, which show very little ash and very little sulphur, and I think ought to be quite good. Professor Campbell. — I will say that in our own work here we have undertaken in the past couple of years to do a little work on cement, though our experiments have not taken us far enough along to draw any reliable conclusions. We are trying to study a few of the conditions which exist. In the course of that we have analyzed a number of samples of clay and marl, and in the synthetic work that we are trying to do, we are trying to control the different factors which determine the property of the cement. The chemical composition is only one. It is well known that the proportions are very close between the different elements* which have to be main- tained in the mixture, but this is only one of the various factors which we are working on. One factor is the time limit, the time for which the material is subjected to a given heat, say 2,300 degrees Fahrenheit, will give the same result as a shorter time at 2,700 or 2,900 degrees.* These are some of the problems on which we are try- ing to throw a little light; but it will be necessary to make long- time tests. One of the great troubles that we find in studying the tests that are made, is the short-time tests that are used in making cement tests. They are usually from 24 hours to 28 days, most of them not exceeding 28 days in time, which seems like nothing to me to determine the property of cement. We have had a few from three to six months, and a year, and some that we had kept longer. Cement will change very much after three months, and after six months. We have had cement that would gain steadily up to three months, then drop off at six months, after that gain again, so it is almost impossible to draw conclusions from a short-time test. Now *See “Some preliminary experiments upon the clinkering of Portland Cement,” by E. D. Campbell, Journal Am. Chemical Society. Vol. XXIV, No. 10, Oct. 1902. MARLS AND CLAYS IN MICHIGAN. 349 this question of magnesia is a vital one. In one case, for instance, we have made cement with as high as 6.90 per cent and after six months’ test, the cement has been gaining with no signs of deteriora- tions as yet. Whether this will continue, or whether it will com- mence to deteriorate after a year or two, time will tell. I do not feel like expressing much of an opinion on the clay that is best adapted to cement work, because the longer we work on it, the less I feel we know or are able to pass an opinion on. I have had occas- ion two or three times to change my own opinion after working awhile. While some experiments will lead to the idea that clay should have a certain proportion of aluminum, silica, etc., other clays that at first are thought to be not at all satisfactory will give equally good results. So I do not feel like expressing my opinion very strongly as to what an ideal clay or marl should be. Mr. Lane. — I would like to ask if you have made any experiments as to the influence of the fineness of grinding. Professor Campbell. — We always try to grind to the same degree of fineness and test the fineness. To give an idea of a single property of the cement, the time of setting, for instance. I think there are not less than six different factors that determine the time of setting, and every one of these may vary, so that it is hard to get at the exact benefits of a single factor. It will undoubtedly be years before we can get at the true nature of what cements are, and the influence of the different factors on the properties of cement. Portland cement is extremely sensitive to water, and quite a difference will be produced by the addition of a little water. Mr. Russell. — By the addition of a little water when you have reached the turning point, it is remarkable how the strength runs down. The man who works for a dollar a day says, '‘Turn on more water,” and “That is enough” and it is impossible to induce people to see that their structure would be very much better if they would understand that there were quantitative relations between the water and the cement, and that they might learn that lesson from those who are able to give it, and their work would be very much bettered by it. Mr. Brigden. — What is the least amount of water possible to make a good mixture? Mr. Russell. — Well, you can work with a trowel a neat cement with something like 22 per cent, can you not, Professor Campbell? 350 MABL. Prof. Campbell.— That is as low as you often get. If you go much beyond that, say 23 per cent, it will run the tensile strength down to a remarkable degree and impair the structure. Mr. Brigden. — I think the man with the hoe has the better end of the argument in almost every case, whether the work is done by the contract or by the day; you can stand and watch him, and you will occasionally hear from the bottom of the trench, “This is too stiff,” or “This is too wet,” etc. I know but little about the matter of cement or its development. I can hardly understand what Mr. Greene meant when he said there seemed to be a large opening for the manufacture of water pipe from cement. I had supposed, and I think I have good authority for that supposition, that the use of cement-lined pipe (if that is what he means) was going entirely out of date, and that cast iron was used for water mains, and I think that is true of every portion of the United States east of the Mississippi River. Mr. Rogers. — I would like to ask Prof. Campbell something about the practical methods for the engineer to determine whether there is too much magnesia in the cement or not, if he has only a few days or a month or so to test the cement before he has to use it. Prof. Campbell. — I cannot give a method that would be entirely satisfactory, because a great many of the difficulties that are often attributed to magnesia in the way of expansion, I think are not due to magnesia at all ; so I do not as yet think there is any entirely satisfactory test. Of course, the chemist’s analysis Avill show the per cent of magnesia in the cement; but then the old question comes up again as to what per cent is allowable. That is the question that we are working on at the present time. Mr. Whitney. — In regard to this question of mixing cements, I think one feature that is often overlooked, and it may be because of the contractor’s haste or tendency to save, is thaU long mixing has a good deal to do with the strength of the cement, that is, after the water is put on to the cement and mixed with the sand, the mixture grinds the cement finer and makes the water appear milky, and when that condition prevails, you will find the strength pf the cement is a great deal more. I want to say a word about marl beds. I have had some experience the last two years, and have probably made something like fifteen hundred soundings. In the question of sounding marl beds, there are two or three points of interest that may be brought out. They vary as to their depth where sometimes we would least expect it, and the bottom of them is so irregular MARLS AND CLAYS IN MICHIGAN. 351 that sometimes it is necessary to take soundings quite close to- gether. Another thing is that there is a good deal of difference in the appearance of marls. Some will be yellow as ordinary corn meal, and very mealy and be very poor marl, and there will be some that is nearly white, and some that you will first feel like throwing out, when, upon a careful examination, you will find that it is pure marl, but not so finely disintegrated. Then again you will run across sand that is a little closer to the surface than you hoped to find it, and you have got to be rather particular to know when you strike it. Often you will find marl in the condition of nodules*, and the person sounding will be almost sure he has struck gravel; he can hear it grate, and it is almost impossible to turn an augur through it ; it takes a pretty good job of well-driving, sometimes, to get through a small layer of it. There is one other fact I might mention, that of course would be easily observed, and that is that around the mouth of streams flowing into lakes where there are marl deposits, there is apt to be a layer of organic matter over the marl, sometimes very deep, sometimes quite deep, and sometimes mixed with the marl and extending quite a way. I would also say that in sounding with augurs, it is sometimes quite desirable to have two sizes of augurs, as there is often a good deal of suction which pulls off the material at the bottom, and gives you, when you pull it up, the material from the top. Of course the use of a go- devil is something by which you can take up samples from any depth, and is quite valuable. Where there is a mixture of clay it is quite apt to be found toward the bottom, and one can rapidly detect it by the color or by the feeling or the appearance of the augur when it is pulled up, and of course the per cent of clay can be actually determined by a chemical analysis. * Probably Schizothrix. elsewhere described, p. 90. 352 MAUL. CLAY ANALYSES. Analyst. Delos Fall. A. N. Clark. Fall. No 601 to 604 822 635 SiC >2 Silica 62.65 69.72 69.00 75.10 12.35 Alumina 23.06 18.96 15.16 Iron oxide 6.82 1.29 5.00 8.21 1.13 Calcium as Cxide 1.02 .40 .80 Magnesium as oxide .11 tr. 3.36 Sulphuric anhydride 4.23 1.13 2.42 Organic matter 2.22 Difference 7.76 6.68 100.11 99.26 100.00 99.21 Nos. 601 to 604 represents the average of 4 clay samples, from near Athens. No. 822. Jackson, probably a coal measure shale clay, rather too high in silica. An analysis by Mr. Clark. 635. Kalamazoo county clay. MARL ANALYSES. Analyst. Delos Fall. No 327 to 424 819 820 Si02 .53 .58 60 Alumina .754 .76 70 Calcium as oxide 52.61 as carbonate (93.91) 20.90 94.75 75.06 Magnesium as oxide .09 tr as carbonate (1.88) Sulphuric anhydride .62 1.24 as calcium sulphate .... P 2 05 Organic matter and carbon dioxide 42.28 W ater Difference 3.12 Nos. 327 to 424 is an analysis representing the average of 14 samples from Lime Lake. Nos. 819 and 820 represent the average of 25 borings at Spring Arbor, the Pyramid Portland Cement Co. location. MARL ANALYSES. Analyst. Delos Fall. No 425 426 427 428 506 to 886 740 743 Silica 2.658 .371 .332 .452 2.08 .645 .61 Alumina 2.658 .721 .729 1.563 2.59 2.22 1.90 flaloinm n.ci f'.a.rhnnsit.ft 86.373 84.973 86.439 89.675 88.06 94.18 93.81 Magnesium as carbonate tr. tr. tr. .50 .32 .234 .19 Sulphuric anhydride .78 .201 .22 Organic ) vy at 8.351 13.935 12.001 8.31 5.30 nifff>i , pnr>p j 100.040 100.000 99.501 100.500 99.13 97.480 96.73 Nos. 425 to 428 are from Goose Lake. See p. 233. Nos. 506 to 886 represent the average of analyses of 84 samples from Athens, T. 4 S., R. 8 E.; the iron included with the alumina. MARLS AND CLAYS IN MICHIGAN. 353 FISH LAKE MARL ANALYSES. Analyst. Delos Fall. No 688 689 690 691 692 693 694 695 Insoluble Si0 2 Alumina Iron .54 .82 tr. 94.33 none 1.12 .77 .64 tr. 949.4 none .87 .41 .38 tr. 93.35 none .44 .38 .60 tr, 95.42 none .87 .26 .38 tr. 94.21 none .80 .22 .55 tr. 98.86 none .75 .06 .5 .24 .54 Calcium as carbonate Magnesia Sulphuric anhydride 92.12 none 1.10 90.52 none .82 KALAMAZOO COUNTY MARL ANALYSES.— Nos. 622 to 638, except 635. Analyst. Delos Fall. No. 622 624 625 926 627 628 629 630 Insoluble Silica Alumina Calcium carbonate in Ing marl, as carbonate MgO SO3 3.47 3.32 90.22 80.93 tr. 1.98 .82 .34 92.11 A .8 1.68 1.95 92.47 .0 1.39 2.38 1.25 92.52 88.16 .0 .75 1.70 3.04 92.55 87.33 .07 1-29 2.40 2.46 91.19 86.84 .0 3.52 1.86 2.60 94.15 92.21 .0 1 26 1.33 3.52 9L38 tr. 1.82 Analyst. Delos Fall. No 631 632 633 634 636 6.37 6.38 Insoluble Silica 1.36 .58 3.4 clay 1.64 1.16 3.21 3.61 Soluble Silica Alumina 2.10 .90 4.05 1.95 2.40 4.07 4.04 Iron oxide Calcium as oxide tr. as sulphate as carbonate 91.26 92 34 88.98 88.28 91.86 88.18 90.77 Magnesium as oxide as carbonate 0 . mere 0 . 1.36 0 . 0 . 0 . Sulphuric anhydride .96 2.11 trace 1.40 1.53 4.49 2.02 45-Pt. Ill CHAPTEB X. METHODS OF, AND COMMENTS ON TESTING CEMENT. BY RICHARD L. HUMPHREY. Structures of masonry or concrete owe their stability almost en- tirely to the character of the substance which binds or cements to- gether the brick, stone, and other materials used in their construc- tion. From the earliest times, therefore, there has been an almost constant endeavor to obtain some material which would attain great strength in a very short period of time and which would re- sist the forces which tend to disintegrate or decompose it. Such a material must harden rapidly, equally well in air or water and have great adhesive qualities. The material used in the earlier structures consisted of a mix- ture of sulphate of lime (gypsum) and sand (the latter usually of a volcanic origin) or a mixture of lime and volcanic ash or trass and sand. Such mortars required considerable time to harden and also protection during the initial stage of hardening from rain and frost, which readily dissolved and disintegrated them. It was necessary therefore to frequently renew the mortar in masonry by pointing, unless sufficient carbonic acid had been ab- sorbed from the air to convert the lime into a carbonate, in which form it offered greater resistance to the weather. This material proved very unsatisfactory even w r hen carefully used and protected from the weather during the early stages of hardening, and at best was only meagrely hydraulic. The most satisfactory results were obtained with Roman Cement, a mixture of fat lime and a volcanic ash. With the downfall of the Roman Empire the art of making this cement was lost and subse- quent experimenters endeavored to recover and to equal or excel this Roman Cement. As these efforts became fruitful of results and the quality of the mortar improved, it became necessary to de- TESTING CEMENT. 355 vise some means by which the relative value of different mortars could be determined. The present system of testing may be said to have begun with the experiments of John Smeaton in 1756, in connection with the rebuilding of the Eddy stone Lighthouse. Smeaton in his endeav- ors to obtain a cement which would harden under water made cements from various materials and tested their hydraulic quali- ties by immersing small pats or cakes, made of the cement, under water. F,g + Pig. 33. Apparatus for determining the adhesive strength of mortars. Later Pasley measured the adhesive qualities of mortars by sticking two bricks together and determining the force required to pull them apart. See Fig. 33 (2). He also determined this same property by building out from a wall, horizontally, as many bricks as possible in a given time. See Fig. 34 (1). *In accordance with our custom all illustrations printed with the text are figures. In this paper a number of the figures are reduced from plates containing a number of figures, the numbers referring to which are placed in parentheses 356 MAUL. This test was more properly a test for determining the rate of setting. Fig. 34. Illustrations of apparatus in cement tests. Vicat gauged the relative hardness of his mortars by measuring the penetration of a weighted needle falling from a given height. Fig. 35. Vicat needle as originally designed. TESTING CEMENT. 357 The apparatus which he devised for this purpose is shown in Fig. 35. It was not, however, until 1858 that a definite system for testing was evolved. In that year John Grant, the Engineer in charge of the London Main Drainage System proposed the tests by which the cement used in this work was inspected. This marked the beginning of systematic tests of cement. The evolution from these few simple tests has been rapid; at present there are numerous tests in use, all more or less rational, many impractical and none entirely satisfactory. It is to be noted, however, that the extreme methods formerly in vogue are becoming less used, and the better informed engineers are adopting less radical and more simple tests. During this period in the development in the methods for mak- ing tests, the manufacturer has been forced to meet tests of con- stantly increasing severity. As a result the quality of the cement, particularly of the American Portland, has been so greatly im- proved that today the manufacturer is able to produce a material, capable of attaining great hardness in a few hours and exceeding in a few months the strength attained by ancient mortars after 2,000 or more years. Indeed the quality of the modern Portland Cement has improved so considerably that it has engendered a greater confidence on the part of the Engineer, resulting in a rapid extension of its field of usefulness. So great is the varied application of cement in con- struction that some one has truly said, “We are on the threshold of the Cement Age.” In Fig. 44 is shown the relative strengths of the modern high grade Portland Cement, cement of the time of Grant, common lime mortar etc., which illustrates the marked superiority of the modern cement. While it is true that the quality of cement has been vastly im- proved, the methods for making the tests are still crude and leave much to be desired. Nor are the tests sufficiently defined to enable the novice to follow them with satisfactory results. It is only after considerable experience that sufficient skill is acquired which permits of even approximately satisfactory results. To define a system of testing which will serve as a reliable guide for novice and expert alike in determining the qualities of cement is a problem of no little difficulty. 358 MARL. The object in testing cement is first to ascertain whether the quality is up to a certain prescribed standard (the specifications), and second for purposes of research. The inspection and testing of cement is an art requiring consid- erable experience and much skill. The difficulty in making the tests lies almost wholly in the “personal equation” of the person who makes the tests, a variable which renders the results of such tests not only relative but inaccurate. Another difficulty in the inspection of cement is the fact, that a cement having passed satisfactory tests at the place of manu- facture is no guarantee that the cement will yield the same or even as satisfactory tests at the place of consumption, even should the same person make the tests. From the moment the clinker is re- duced to an impalpable powder until it is made into a mortar or concrete and becomes a part of the structure, its physical and chemical properties are constantly undergoing changes which af- fect its value as a building material. It is doubtful whether these changes ever cease. The cement being to a greater or less extent affected by external influences tending to decompose the mass or by internal influences tending to disintegrate it. The selection of methods for testing is not so easy as it would at first appear. The system should not depend on cumbersome methods or expensive apparatus. The number of tests should be. few and simple in execution. The inspection of cement may be divided into two classes (1) the mill tests or those made at the place of manufacture, and (2) tests of acceptance or those made at the place of consumption. The lat- ter can be further subdivided into field and laboratory tests. In the first class are those made by the manufacturer to check the quality of his product and are usually as severe as it is possible to make them, especially as regards constancy of volume. This is due to a desire on the part of the manufacturer to thor- oughly test the quality of his cement before it is shipped. The methods for making the tests in both classes are, however, the same. The tests in general use are for the determination of fineness, time of setting, tensile strength, neat and with a stand- ard sand, for 24 hours, 7 and 28 days, together with the cold water pat test, and some form of accelerated test, usually the so called “Boiling test.” In addition to these, the determination of specific gravity and a chemical analysis of the finished product are made TESTING CEMENT. 359 at the mill at regular intervals; most mills make at least one com- plete analysis of the product each day in order to check the com- position. Where analyses are required on work not possessing the requisite facilities they should be made by some well established chemical laboratory. While the methods used by the consumer or manufacturer are the same the number of tests made are modified to suit the time available for the purpose and the facilities. The standard for gauging the results of the tests of acceptance for determining the value of a cement for the purposes for which it is to be used is the specifications. The requirements of this specification should be based on the results obtained by the per- sons who make the tests. Before fixing these requirements it should be first ascertained what results can be obtained from well known brands of cement by the persons making the tests. Upon these results should be based the requirements of the specifications. The scope of the tests to be made will depend on the facilities and the importance of the work. In permanent laboratories the testing should be systematic and thorough. Such a system will now be described in more or less detail, indicating where it may be modi- fied to suit other conditions. SAMPLING. The selection of the sample from which the tests are to be made while apparently a very simple matter is one of considerable im- portance and should therefore be carefully done. At the time of sampling a note should be made of the condition of the cement, i. e., whether cement is lumpy, caked or otherwise damaged. The sample should be taken from the heart of the package as the outer portion is sometimes more or less impaired. About one barrel in every ten should be sampled. Where the cement is delivered in barrels the sample can be drawn through a hole made in one of the staves midway between the heads by means of an auger or sampling iron similar to the ones used by sugar inspectors, Fig. 36 (9. ) if the shipment is in bags, the sample is taken from the heart of the package with the hand, or a scoop. When the sample is taken at the place of manufacture it should be done regularly as it comes from the mill and goes into the bin. 360 MABL. Where cement is held in storage pending the result of the tests, it should be protected from the weather, in order to prevent its being damaged. The samples should be passed through a sieve having twenty or thirty meshes per lineal inch in order to remove the lumps and foreign matter. This is also a very efficient means of mixing the individual samples in case an average sample is desired; where time will permit, the individual samples should be tested separately in addition to the test on the averaged samples. CHEMICAL ANALYSIS. Systematic chemical analyses of cement should be made in all permanent laboratories, not with a view of eventually introducing into specifications chemical requirements (other than those for sulphuric acid and possibly magnesia) but in order that we may have some data pertaining to the composition of the cement when studying the results of the long time tests. Chemical analyses are chiefly valuable to the manufacturer. The determination of silica, of iron and alumina and of lime is of little value as an indication of quality. They furnish valuable aid in detecting adulterations with inert material in considerable quan- tity and in determining the quantity of certain deleterious con- stituents as magnesia and sulphuric anhydride. The following scheme of chemical analysis is recommended: One half gram of the finely pulverized sample, dried at 100° C., is thoroughly mixed with four or five times its weight in sodium carbonate, and fused in a platinum crucible until carbon dioxide, C0 2 , no longer escapes; the crucible and its contents is placed in a beaker and twenty or thirty times its quantity of water, and about 10 cubic centimeters of dilute hydrochloric acid (HOI) is added; when complete solution is effected, it is transferred to a casserole and placed on a water bath, and evaporated to dryness several times. The mass is taken up with dilute hydrochloric acid, HC1, and water, heated for a short time and filtered, washing the resi- due on the filter thoroughly with hot water. The filter is dried, ignited and weighed. This weight (less ash) gives the amount of silica, Si0 2 . The filtrate is brought to boiling, and ammonia is added in slight excess; the boiling is continued until the odor of ammonia is no TESTING CEMENT. 361 longer perceptible. Filter and wash, re -dissolve in hot dilute HC1, again precipitate with ammonia and filter through the previous filter and wash with boiling water. The precipitate dried, ignited and weighed, less ash gives the amount of alumina, A1 2 0 3 , and ferric oxide, Fe 2 0 3 . The iron is determined volumetrically by fusing the ignited pre- cipitates of alumina and iron with de hydrated potassium sulphate in the platinum crucible, it is then dissolved in sulphuric acid and titrated with potassium permanganate. The filtrate from the iron and alumina is heated to boiling, and boiling ammonium oxalate is added until a precipitate is no longer formed. After boiling for a few minutes it is set aside for a short time, when the precipitate has settled perfectly, decant the clear liquid through a filter, wash by decantation, dissolve the precipi- tate in hot dilute hydrochloric acid, HC1, using as small a quantity as possible to effect a complete solution; heat to boiling and add ammonia, heat on a water bath for a few minutes; when the solu- tion clears filter through the previous filter, wash thoroughly with hot water. Dry the precipitate, ignite to constant weight, and weigh as CaO; or dissolve with sulphuric acid and determine the lime volumetrically by titration with potassium permanganate of a known strength. The thoroughly washed precipitate of calcium oxalate is dis- solved in hot dilute sulphuric acid and the solution is titrated with standardized potassium permanganate. The filtrate from the calcium oxalate is made alkaline with ammonia and 30 cubic centimeters of solution of hydro-disodium phosphate is added; the whole is set aside in a cool place for twen- ty-four hours; it is then filtered and washed about fifteen times with ammonia water solution (1:5). Dry the precipitate on the filter, brush on to a large watch glass, burn filter on the lid of the weighed crucible. When the carbon is consumed transfer the pre- cipitate to the crucible and ignite to dull redness, keeping the crucible covered. If the precipitate is not perfectly white on cool- ing, moisten with a few drops of nitric acid, evaporate and ignite to dryness; weigh as magnesium pyrophosphate and calculate to MgO. Sulphuric acid, — This is determined in a separate portion. Weigh out about five grams and treat as in the regular analysis, separating the silica; the filtrate is heated to boiling, acidulated 46-Pt. Ill 362 MARL . with hydrochloric acid, and boiling barium chloride is added; the boiling is continued for ten minutes; when the precipitate has sub- sided, filter. The precipitate is thoroughly washed in hot water, dried, ignited and weighed as barium sulphate and calculated to sulphur trioxide, S0 3 . Carbonic acid . — This can be determined with sufficient accuracy by means of the ordinary extraction apparatus. For routine work where quick results are desired the above scheme may be shortened in the follownig manner: The first solution may be effected by treating the finely pulver- ized sample with concentrated HC1 diluted with an equal portion of water to which a few drops of concentrated HN0 3 has been added. Evaporate to dryness on the sand bath until all odor of HOI has disappeared. The residue is then treated with concentrated HC1 boiled a few minutes diluted with water and filtered. The silica is separated by filtration and ignition as above, or the residue after taking up and boiling with concentrated HC1 can be treated with sodium carbonate and the solution effected with concentrated HC1 and water as above described. While other short cuts could be suggested, it is not deemed ad- visable since the saving in time is not commensurate with the ac- curacy. SPECIFIC GRAVITY. The determination of specific gravity or true density is of ques- tionable value except in the hands of an experienced operator. In as much as the differences in the results are very small, con- siderable care must be exercised to obtain accurate determinations. It is perhaps useful in detecting underburning or adulteration with material of low specific gravity. The adulteration must, how- ever, be in considerable quantity in order to materially effect the results. A better means of detecting adulteration is through the use of a liquid of heavy gravity and not capable of affecting the cement. Le Chatelier’ s apparatus is the best means for making determination of specific gravity. This apparatus consists of a flask D Fig. 36 (3) of 120 cubic centi- meters capacity, the neck of which is about 20 centimeters long. In the middle of this neck is a bulb C, above and below which are two marks engraved on the neck, the volume between these marks, TESTING CEMENT. 363 E and F, being exactly 20 cubic centimeters. Above the bulb the neck is graduated into 1-10 cubic centimeters. The neck has a diameter of 9 millimeters. Benzine free from water is used in making the determinations. The specific gravity can be determined in two ways: (1) The flask is filled with benzine to the lower mark E, and 64 grams of powder are weighed out; the powder is carefully introduced into the flask by the aid of the funnel B. The stem of this funnel de- scends into the neck of the flask to a point a short distance below the upper mark. As the level of the benzine approaches the upper mark, the powder is introduced carefully and in small quantities at a time until the upper mark is reached. The difference between the weight of the cement remaining and the weight of the original quantity (64 grams) is that which has displaced 20 cubic centimeters. (2) The whole quantity of cement is introduced, and the level of the benzine rises to some division of the graduated neck. This reading -f- 20 cubic centimeters is the volume displaced by 64 grams of cement. The specific gravity is then obtained by dividing the weight in air by the displaced volume. The flask, during the operation is kept immersed in water in a jar A, in order to avoid any possible error due to variations in the temperature of the benzine. The cement in falling through the long tube completely frees itself from all air bubbles. The re- sults obtained agree within .02. FINENESS. The degree of final pulverization which the cement receives is exceedingly important. It has been found that the coarser par- ticles in cement are inert and have no hardening qualities. The more finely a cement is pulverized, all other conditions being the same, the greater will be its cementing properties or what is usually known as its “sand carrying” capacity. The test for fineness consists in determining the percentages of grains of certain sizes. By our present methods this is accom- plished by separating the particles with standard sieves. These sieves are of brass wire cloth having a circular frame 6 to 10 inches in diameter about 2-J inches high and usually provided with a top cover and bottom pan, figure 36 (8.) What are known as the No. 100 and No. 200 sieves are generally used. These sieves should have theoretically 100 and 200 meshes 364 MARL. per lineal inch and the wire should have diameters of .0045 inch and .0023 inch respectively. As it is impossible to obtain sieves having exactly this number of meshes on account of the impossibil- ity of weaving the wire cloth with sufficient uniformity by hand methods, the specifications should state the approximate number of meshes and the size of the wire of the sieves to be used in mak- ing the tests. The sample for sieving should be thoroughly dried at a tempera- ture of about 212° F., since in this condition the cement sieves much more readily. One hundred grams make a very convenient quan- tity to sieve. The manner in which the sieving is done determines to a large extent the time required for the operation. After the fine flour has passed through the sieve the coarser particles pass through very slowly; and since the final operation determines the fineness, it is important that it should be done thoroughly. The cement' is best sieved by moving the sieve forward and back- ward with one hand in a slightly inclined position and striking the side of the sieve gently with the palm of the other hand at the rate of about 200 strokes per minute. The cloth of the sieve should be carefully watched, as it is liable to break and produce abnormal results. The introduction of large pebbles or gravel, retained on a screen having ten meshes per lineal inch, into the sieve, accelerates the oper- ation of sieving. The sieving can be considered complete when not more than one tenth of one per cent passes through the sieve after one minute of continuous sieving. NORMAL CONSISTENCY. The percentage of water to be used in making tests of setting, briquettes and pats is of the greatest importance, for upon this de- pends the results obtained. The paste used in these tests should be of definite or Avhat is called a standard consistency. The same consistency should be used for all tests. The best consistency is one so wet that mortar cannot be com- pressed in molding and not so wet as to make a sloppy test piece which would shrink. The best method for estimating the proper percentage of water to be used is by means of the Yicat apparatus. TESTING CEMENT. 365 This apparatus illustrated in Fig. 36 (1), consists of a frame K, bearing the movable rod L, having the cap A at one end, and the piston B, having a circular cross-section of 1 centimeter diameter at the other. The screw F holds the needle in any desired position. The rod carries an indicator which moves over a scale (graduated to centimeters) attached to the frame K. The rod with the piston and cap weighs 300 grams; the paste is held by a conical hard rubber ring, I, 7 centimeters in diameter at base, 4 centimeters high, rest- ing on the glass plate J, 15 centimeters square. Trial pastes are made with varying percentages of water. The paste is of proper consistency when the piston gently applied to the surface of the paste (confined in the hard rubber ring) sinks to a point a given distance above the upper surface of the glass plate J. (about 28 mm.). 366 MARL. Having determined the requisite percentage of water for neat pastes the percentages required for sand mixtures can be deter- mined from the following table: PERCENTAGES OF WATER FOR STANDARD MIXTURES. Neat. 1 to 1. 1 to 2. 1 to 3. 1 to 4. 1 to 5. 15% 11.0 9.3 8.5 8.0 7.7 16% 11.3 9.6 8.7 8.1 7.8 17% 11.7 9.8 8.8 8.3 7.9 18% 12.0 10.0 9.0 8.4 8.0 19% 12.3 10.2 9.2 8.5 8.1 20% 12.7 10.4 9.3 8.7 8.2 21% 13.0 10.7 9.5 8.8 8.3 22% : 13.3 10.9 9.7 8.9 8.4 23%., 13.7 11.1 9.8 9.1 8.5 24% 14.0 11.3 10.0 9.2 8.6 25% 14.3 11.6 10.2 9.3 8.8 26% 14.7 11.8 10.3 9.5 8.9 27% 15.0 12.0 10.5 9.6 9.0 28% 15.3 12.2 10.7 9.7 9.1 29% 15.7 12.4 10.8 9.9 9.2 30% 16.0 12.7 11.0 10.0 9.3 31% 16.3 12.9 11.2 10.1 9.4 32% 16.7 13.1 11.3 10 3 9.5 33% «. 17.0 13.3 11.5 10.4 9.6 34% 17.3 13.6 11.7 10.5 9.7 35% 17.7 13.8 11.8 10.7 9.9 36% 18.0 14.0 12.0 10.8 10.0 37% :.... 18.3 14.2 12.2 10.9 10.1 38% 18.7 14.4 12.3 11.1 10.2 39% 19.0 14.7 12.5 11.2 10.3 40% 19.3 14.9 12.7 11.3 10.4 41% 19.7 15.1 12.8 11.5 10.5 42% 20.0 15.3 13.0 11.6 10.6 43% 20.3 15.6 13.2 11.7 10.7 44% 20.7 15.8 13.3 11.9 10.8 45% 21.0 * 16.0 13.5 12.0 11.0 46% ■ 21.3 16.1 13.7 12.1 11.1 Cement 500 333 250 200 167 Sand 500 666 750 800 833 E=2-3 N A x 60 where N=weight of water (in grams) required for 1,000 grams of neat cement. A = weight of cement (in kilograms) in 1,000 grams of sand mix- ture. E = weight of water (in grams) required for sand mixture. TIME OF SETTING. The determination of the time required for a cement to set or the time w r hich elapses before the paste ceases to be fluid and plastic is of considerable practical importance. The beginning of this state is called the initial set” and the moment when the paste TESTING CEMENT. 367 offers a given resistance to change of form is called the “hard set.” After the cement has set the process of crystallization or hardening begins. To add water and again mix a cement which has set is called “retempering.” As a cement loses a great deal of its initial strength by “retempering” it is necessary to determine the length of time re- quired for the cement to set in order to avoid “retempering” the mortar on the work. Tests for the time of setting are made on pastes of neat cement only, as in sand mortars the grains of sand impede the free pene- tration of the needle. Yicat devised the original apparatus (Fig. 35) for determining the rate of hardening of lime mortars. In the tests as recommended by Vicat, the weighted needle was allowed to fall into the mater- ial under test. In the test as now used, the needle is applied care- fully to the surface and allowed to sink into the mass under a given weight. Fig. 36 (1). This apparatus has been described under Fig. 36, two pages before. In this test the cap A is replaced by the cap D, and the piston B is replaced by the needle H. The rod L then weighs 300 grams. The hard rubber ring containing the paste of normal consistency is placed under the needle which is gently brought in contact with the surface and allowed to sink into the mass under the load of 300 grams. For neat pastes the setting is said to have commenced when the polished steel needle weighing 300 grams, does not completely trav- erse the mass of normal consistency confined in the rubber ring, and the setting is said to be terminated, when the same needle gently applied to the upper surface of the mass does not sink visibly into it. A thermometer C graduated to 1-5° C. is stuck into the mass and the increase of temperature of mass during setting can be thus ob- served. Care should be taken to keep the sides of the needle clean as the collection of cement on the needle retards the penetration of the needle, while cement on the point of the needle reduces the area of needle and tends to increase the penetration. The test specimens should be kept in moist air during the test. This is best accomplished by placing the specimens on a rack over water contained in a pan covered with a damp cloth kept away from the specimen by a wire screen. The specimens can also be kept in a moist closet. 368 MARL. TENSILE STRENGTH. The setting of cement is the change from a condition of fluidity to a solid state. When cement has set, the process of hardening is said to commence. The relative degree of hardening at any age is measured by determining its transverse, compressive, adhesive or tensile strength in pounds per square inch. Of these tests the tensile test is universally used and has met with great favor on account of the convenience with which the test is made and the cheapness of the apparatus required. The test piece is of one inch section and is shown in Fig. 36 (5). For convenience in molding and removing the briquettes from the molds, the sharp corners should be rounded off with curves of one half inch radius, the briquettes to be of the form shown in Figure 36 (6). Molds . — The molds should be made of brass or some equally non- corrosive material and can be either of the single or gang-type, the latter is preferable since the convenience and facility for molding several briquettes at one time is greater than in the case of the single mold. The greater quantity of material which can be mixed at a time tends to produce more uniform results. The convenience in cleaning, compactness and facility with which they can be handled are also arguments in favor of the gang type. There should be sufficient metal in the sides of the mold so as to prevent spreading of the mold when in use. The molds should be wiped with an oily cloth before using, this prevents the cement sticking to the mold and damaging the bri- quette during the removal from the mold. Mixing . — About one thousand grams of cement makes a very con- venient quantity of material to mix at a time and will make about eight or ten briquettes. The French system of weights and measures because of the re- lation between the gram and the cubic centimeter is the most con- venient to use. The proportions should be stated by weight. The mixing should be done on some non-absorbing, non-corroding surface, preferably plate glass, although marble or slate would do. If the mixing be done on a surface of marble or of slate it will be advisable to keep this surface covered with a w r et cloth when not in use, or thoroughly wet the surface previous to being used. A surface of this character when not in use, becomes quite dry, TESTING CEMENT. 369 and absorbs some of the water from the first few batches mixed on it; this renders the mortar much dryer and materially affects the results, especially with sand mixtures. The cement is weighed out and placed on the mixing slab and formed into a crater into which the proper percentage of clean water is added. The material is turned into the crater with a trowel and when the water is absorbed, the mixing is completed by thoroughly kneading with the hands; the latter process being similar to that used in kneading dough. The duration of the kneading should be about one minute. During this operation the hands should be pro- tected with rubber gloves. An inexperienced operator should mix for a definite length of time; a one or more minute sand glass is a very convenient guide. If the person making the test is very inexperienced, it may be necessary to use one half the quantity of material. When the moisture begins to disappear from the surface and the paste becomes meally and does not stick together, the cement has begun to set and should be thrown away. Paste which appears to be of the proper consistency at first, be- comes quite wet after thorough kneading, while pastes which ap- pear at first quite dry become plastic. In sand mixtures the mixing should be thorough in order to in- sure coating each grain of sand with cement. This is a very im- portant feature in sand tests and is often the reason why one per- son obtains so much higher results than others. The temperature of the room and of the water used in mixing should be kept as near 70° F. as practical. The air of the room should be kept moist. A high temperature and dry air in the room in which the tests are made, tends to dry out the test pieces, thereby checking the proces of hardening, resulting in low strengths and often in cracking of the test pieces and in some instances disintegration. Molding . — The mortar having been mixed to the proper consistency is placed at once into the molds with the hands. The molds are filled at once, the paste is pressed in with the fingers and smoothed off with a trowel on both sides. This should take about two or three minutes for 8 or 10 briquettes. The mortar should be heaped upon each mold and then pressed in by drawing the trowel over the surface of the mold, holding the blade of the trowel at an 47-Pt. Ill 370 MARL. angle of about 5°. The mold is turned over and the operation repeated. The briquettes are marked in the head with steel dies while still soft, or with a large soft lead pencil just before removing from the molds. An excellent idea of the uniformity of the mixing and molding is afforded by weighing the briquettes upon removing from the molds. The variations in the weight of briquettes should not exceed 3 per cent. Preservation of briquettes . — After the completion of the molding, care should be taken to keep the briquettes in moist air; this prevents them drying out thus checking the process of hardening, and prevents the production of checks and shrinkage cracks. The most convenient way to preserve the briquettes prior to immersion in water, is by means of a moist box or closet. (Fig. 37.) TES TING CEMENT. 371 This may be of soap-stone or slate, or a metal lined wooden box, covered with felt on the inside; the closet should hold water in the bottom and be provided with shelves on which to place the briquettes or the molds containing the briquettes. For the twenty-four hour tests the briquettes should be placed in the moist closet immediately after molding and kept there until broken; briquettes to be broken at longer periods should be im- mersed after 24 hours in moist air, in water maintained as near 70° F. as practical. For preserving the briquettes in water either pans or large tanks are used. The former should be of the agate ware type, since they do not corrode and are easily cleaned. A very convenient arrangement for tanks is shown in (Fig. 37). The tanks are in tiers, the supports can be framed of angle iron or 372 MAUL. wrought iron pipe. Each tank is provided with a hot and cold water supply pipe and a waste pipe; the inlet being at the bottom and the overflow at the top of the tank. These tanks can be built of soapstone or slate, or they can be enamelled iron sinks. Where pans are used the water should be renewed once each week. Care should be observed to keep the briquettes covered with water. When running water is used, care should be observed to main- tain the water as near 70° F. as possible. Fig. 39. Olsen testing machine, power driven. Breaking briquettes . — Briquettes should be broken as soon as they are removed from the water. Care should be exercised in centering the briquettes in the testing machine, as cross strains are liable to result from improper adjustment, producing cross strains which lower the results of the tests. The breaking load should not be ap- TES TING CEMENT. 373 plied too suddenly, as the vibration produced, often snaps the bri- quettes apart before the full strength is developed. Figs. 38 and 39 show one form of testing machine. Fig. 40. Fairbanks testing machine. The clips should be kept clean, and the briquettes free from grains of sand or dirt which would tend to prevent a good bear- ing. Care should also be observed in applying the initial load; this is particularly the case with the Fairbanks machine Fig. 40 and consti- tutes the chief objection to this machine. In long time tests the initial load must be very great, and as there is no way of regula- ting this load satisfactorily, the variations in the results are often largely due to variations in the amount of the initial load applied. In order to regulate the application of this initial strain, it is the practice in some laboratories to place weights in the shot pan at the commencement of the test, the amount of the weight being de- pendent upon the ag-e and character (natural or Portland) of the cement under test, the weight increasing with age — it being great- er for Portland than for natural cement, and also greater for neat than for sand tests. It often happens that the last molded and usually the densest briquettes are broken at twenty-four hours or seven days and the first molded or less dense at twenty-eight days or longer. This difference in density may be considerable, in which case the tests may show an apparent falling off in strength. Again, the cement 374 MARL . may begin to set before the last briquettes is molded, and should these briquettes be broken at the long time period, a loss of strength might be again apparent, or even indications of disintegra- tions appear. All these facts tend to emphasize the necessity of uniformity in mixing and molding in order to secure uniform dens- ity in the briquettes, and thus avoid the resultant apparent losses in the tensile strength. Fig. 41. Riehle testing machine. CONSTANCY OF VOLUME. One of the most important tests to which. cement is subjected, and one which is the most difficult to make, is that which pertains to the soundness. The methods that have been suggested are legion. This test cannot be used by a novice with safety, and even in the hands of an expert all tests for soundness must be made with extreme care. The object of the test is to determine whether the cement will maintain a constant volume, and develop no evidence of unsound- ness or loss of strength. It is exceedingly important that cement TESTING CEMENT. 375 should not only develop strength but it should also maintain this strength. Tests of this character can be divided into two classes: (1) nor- mal pat tests and (2) accelerated tests. The former consists in immersing a pat of neat cement after hard set in water maintained at a temperature as near 70° F. as possible. To successfully meet this requirement it should remain firm and hard and should not check, become distorted or show other evidence of unsoundness. Months and even years are requisite to develop evidences of unsoundness by this method unless the cement be of very poor quality. The accelerated tests are for this reason in greater favor, because results can be obtained in considerable less time — in a few hours as a matter of fact. Of the latter class of tests one best adapted to general use is to immerse the pat (after twenty-four hours in moist air), for three hours in an atmosphere of steam coming from boiling water contained in a loosely covered vessel. The pat to satisfactorily pass this test should remain firm and hard and show no signs of checking, cracking, distortion or disintegration. A more reliable test, but one which is more expensive and which requires consider- able care in maintaining the water at a fixed temperature, is to im- merse the pat (after twenty-four hours in moist air) in water main- tained at a constant temperature of 170° F. One of the difficulties encountered in these tests is in making the pats; these are usually made of neat cement, about three or four inches in diameter, from one-quarter to one-half of an inch thick at the center and tapering to thin edges at the circumference. The pats should be made with the same percentage of water as in the case of the other tests. Simple as the making of these pats may appear to be, it is extremely difficult for inexperienced persons to make them correctly. Pats may be so trowelled as to give initial strains which develop cracks during the test. A good plan is to strike the glass on which the pat is made after molding; this re- arranges the mass, drives the moisture through the pat and makes the density of the pat more uniform. Care should be taken that the pats do not dry out — this produces shrinkage cracks, which give a false impression of unsoundness. Most pats leave the glass, and unless this is accompanied by swelling, curvature of the pat, or cracking at the edges, it should not be taken as evidence of unsound- ness. In some cases the cement may set before the pat is finished, 376 MAUL. and when placed in steam or hot water, the outer edge may lift off. This to the inexperienced is also misleading. Fig. 42. Result of tests of constancy of volume. Cements should not be condemned on the results of the acceler- ated tests alone, nor should a cement be considered sound because it has passed such tests. The results of such tests are shown in Figs. 42 and 43. Fig. 43. Result of tests of constancy of volume. TESTING CEMENT. 377 CONCLUSION. The tests just described constitute those most essential for gen- eral purposes in determining the value of cement delivered for use. Tests for determining the compressive, transverse, adhesive or abrasive strength, together with those for determining the effect of frost, action of sea water and the porosity, furnish information having a value for the purposes of research, or where the condi- tions render such data desirable. Permanent laboratories where work of this kind can be carried on, should be equipped for such tests. Tests of still greater importance, which cannot be used as tests of reception, are those made on the work. These consist in tests of briquettes made from mortar taken from the mixing box or cubes of concrete. Data obtained from such tests is valuable, inasmuch as it fur- nishes information concerning the strength of the concrete or mortar taken from the mixing of the mortar or concrete. There should be some system under which the tests are made, that is, there should be a regular number of briquettes made from each sample, and they should be broken at regular intervals; when- ever possible these tests should be extended beyond the regular twenty-eight day period, as it is very desirable to know what the strength is at the end of several years. In addition to the tensile tests, each sample should be submitted to all the tests usually employed. The data obtained from these tests should be carefully recorded in a book kept for the purpose. Having made the above tests, the interpretations of the results obtained is the next and most serious difficulty which confronts the inspector. It is impossible always to insist on a rigid com- pliance with the requirements of the specifications, since the fail- ure to meet these requirements .may be due to faults in the testing. It often happens that the person who makes the tests does not use the same amount of energy in each test; this is particularly the case where the number of tests made is large, or the test pieces may dry out or they may be effected by the conditions under which they are preserved. In cases where the cement fails to meet the requirements, it should be given a re-test before condemning it. It may be well at this point to call attention to the falling off in tensile strength which occurs at the end of one, two or more months. 48-Pt. Ill 378 MARL . Just what causes this action has not as yet been satisfactorily explained. All cement as it acquires hardness becomes brittle, the length of time required varying from a few months to several years. In the early stages of the process of hardening, the mass is tough and in a more or less amorphous condition; but as the crystalliza- tion proceeds, the mass becomes brittle. It would seem that the loss in tensile strength can be attributed to crystallization. The modern rotary kiln process is such that we can obtain arti- ficially, in a very short space of time, a result that nature requires centuries to accomplish. We are required to make tests of a material, which for all prac- tical purposes can be considered a stone; it would seem logical therefore to apply those tests usually applied to tests of stone, i. e., compressive tests. This would seem to be a proper method for ascertaining the real strength of cement especially for long periods of time. Tension tests should be used for the purpose of determining the relative value of shipments of cement, and should be confined to tests not extending over 28 days. When small compression machines, capable of crushing one inch or one and a half inch test pieces, can be built to compete with the present tensile machine, then we will be able to retire the tension tests. Passing judgment on the quality of a shipment of cement, is one of the most difficult problems that confronts an engineer. You are dealing with a material subject to numerous conditions, any one of which may affect its value as a material of construc- tion. It should be borne in mind that cement is manufactured in one form, tested in another and used in a third. Abnormal behavior in the tests does not necessarily indicate its probable action in actual use. When we consider the ancient structures which were built with materials of inferior quality (when gauged by our present stand- ards) we are impressed with the hardness and durability of the mortars. Again it is very rare that we see cases of failure that can be ascribed to the bad quality of the cement. Our facts are not suffi- ciently established to enable us to state just what qualities or ingredients are requisite for a good cement. TESTING CEMENT. 379 We know, however, as far as our knowledge extends, that the modern rotary kiln product possesses the property of acquiring great strength and hardness in a very short period of time and has thus far been able to resist all normal forces tending to destroy it. What the future will develop only time will tell. Our system of testing under the best conditions is very imper- fect and leaves much to be desired. Without positive information as to what is required of a good cement, and under an imperfect system of testing it does not seem fair to be too rigid in our requirements. Testing cement and the interpretation of the results obtained, requires the liberal application of common sense and good judg- ment, mellowed by practical experience. No better rule can be observed by the person acquiring his first experience in testing cement than, “When in doubt re-test the cement.” The future alone can prove the correctness of our present the- ories, and in the meanwhile, in lieu of something better, we must accept our present cements with faith in their high excellence as a building material. 380 MAUL. An excellent form of record book is shown in the following form: No. sample. Collection. Fineness in per cent residue. Specific gravity. Time of setting, in minutes. Initial. Hard. Temperature. Brand. Date. Place. Bags or barrels No. 100. No. 200. Air. Water. Rise. Strength in pounds per e© ©jere©i©j concoco stop 83 C©COire©l t-OO-HCO cot- — co o ere ire ire osoocooo coi9<©t- C© ©} ©1 ©J ' ©l ©1 ©t ©J ©1 C© ©} stop i NMON OO TT 00 00 TPCOOJW ' CO CM »— < ^ ^ CM CM C* CM •sanoq fz ooocooi ire ere t- 1 - ©j ire co CO O CM i> Ci 00 O) i> CO C- {> Ci 00 00 00 •stop 83 ere ire © -^ T»ooere©i 230 i>i< ire i- ©} oo t- c© — eo co co oo t— t— ire ire co ocirecooo •stop/, <31 CO 05 — C© 05 05 0001-00 o ire o — ■ ere t- oo — o^om lft^t-t- ^TjiTjure oo ire co c- sanoq fz ■crereoe© t-oireo — os-^co cfOOtO-H -i®oo coooooo ©t — e© if — — ere ere ereere-fe© Setting. Temperature. Water. ° F. CO0000C5 OOOOOOOi 00 00 00 00 cocococo co co co co co co co £ fe ^3 o 71-70 71-73 71-70 71-73 71-70 71-70' 71-71 71-73 71-70 71-71 71-73 71-71 Time in minutes. •pjUH CMkCCO^ ^HOOt^iC CM O O CO iCTfNrH kO CM i> CM lC CO Tf vr lOrf CO Tfi CO ^ qniqmi OCi--CO OO^kCCM OiOO^fO ^CikftkO CM CO 1-1 1— ^-1 CM •pasn aoiuAi jo aSn^naojOti ©J © CO CO OMON 000©100 ©1©J — — « ©J bi) S 384 MARL. The above table shows results on samples of Portland cement col- lected by D. J. Hale and sent to Richard L. Humphrey for testing (not samples which were submitted by interested parties) taken from warehouses where cement was sold every day by retail dealers. Each sample and each duplicate of a sample were packed separately, first in a paper sack then in a cloth sack, then in a small oblong wooden box just containing the package. Each box then contained but one sample and the sacks could not possibly mix by breaking or sifting. About 10 or 12 pouuds of cement was taken at a time. It was taken as nearly as possible at the center of a sack or barrel, no sack being sampled which by caking showed the effect of moisture, a leaking roof or a situation exposing it to moist draughts of air as between doorways. The duplicate sample was selected in as nearly the same spot as possible to the one in which the first sample was taken. That is, it was selected from the center of the same barrel or sack. Numbers 3, 4, 7, 8, 10 and 11 were collected at Meecham & Wright’s warehouse, 98 Market St., Chicago, 111. Numbers 1 and 6 Peerless, was sampled from Stevens, Hobbs & Co., Benton Harbor, Mich. Numbers 2 and 5 Bronson, was sampled from Jno. Wallace & Co., St. Joseph, Mich. The following are the numbers and names of the samples taken : No. 1. Peerless. No. 2. Bronson. No. 3. Atlas limestone cement, made at Coplav, Pa. No. 4. Atlas — Duplicate of No. 3. No. 5. Bronson — Duplicate of No. 2. No. 6. Peerless — Duplicate of No. 1. No. 7. Dvkerkoff. No. 8. Alsen’s Portland Cement, from Jtzehoe, Germany. No. 9. Wolverine, taken at Thomas Moulding’s warehouse, 40th and Wentworth Sts., Chicago, 111. No. 10. Dykerhoff — Duplicate of No. 7. No. 11. Alsen — Duplicate of No. 8. No. 12. Wolverine — Duplicate of No. 9. The samples were submitted to Mr. Humphrey with the number only and not the name of the manufacturer or statement of which were duplicates. TESTING CEMENT. 385 NOTES BY D. J. HALE. Upon consideration of these results the following comments ap- pear to be suggested. The specific gravity checks the closest of all the tests and would serve to identify the duplicates as such. The specific gravity of well made cements does not vary greatly, and in the case of these cements the total variation is only 14 per cent. The duplicates, however, vary from each other very slightly indeed, three sets agreeing and three varying, one duplicate from the other, one per cent (Nos. 1 and 6, 3 and 4, 9 and 12) . Fineness seems to have agreed best with the 50 mesh sieve. In this test three out of six pairs gave identical results. This was not paralleled in the tests on the 100 and 200 mesh sieves, the diver- gence between brands in several instances not being as great as that between duplicates. Duplicates only can be compared as to setting, since different per- centages of water were used, duplicates, however, receiving the same. It will be noticed that the greatest divergence in initial set between duplicates 2 and 5 occurred when the air temperature varied. How- ever, in 3 and 4 there was the same difference in air temperature with a difference of only seven minutes in the initial set. It can be seen that even with the careful effort here made to keep tempera- tures as nearly as possible equable the time of setting is very diffi- cult to keep even, and as a source of comparison between respective brands would scarcely be reliable, as the divergence between samples is in many cases not as great as that between duplicates. The tensile strength appears from a comparative point of view the most unsatisfactory of the tests. In the 96 tests made there were but two instances in which duplicates gave the same number of pounds breaking test, 3 in which they varied 2 pounds from each other, 3 in which they varied 3 pounds from each other. The great- est variation between duplicates was 159 pounds. On the other hand so often do the duplicates differ more from each other than from other brands that it does not seem as if this test could show which was the sounder of two brands. For example take 1 and 6, 24 hours neat. The difference is 80 pounds; between 6 and 5 which are not duplicates but rival brands, 57 pounds; between 6 and 2, rival brands, 23 pounds. 49-Pt. Ill 386 MARL. There can be no doubt that this set of tests was made as care- fully as they could be made by our present methods of testing. In a general way some test higher than others but in such an uncertain manner that excepting for the specific gravity test it would scarcely be possible to pick out by means of the record here shown, those samples which were duplicates from those which were different brands. It can scarcely be fair to allow one set of experiments no matter how carefully carried out to settle the question of whether or not our present methods are an actual test or not of the quality of our cements. This series seems to accentuate the emphatic dec- larations of Mr. Humphrey and many who are called upon to in- vestigate the merits of different cements by present methods, that these methods are of small value as an actual test. They cannot, however, be totally condemned until a better system is devised. It is also not to be forgotten that the water and steam test showed all to be first-class cements. While this test does not serve to dis- tinguish between good cements, it should certainly not be dis- carded because it is valuable in detecting a worthless brand. INDEX TO PART III. A. Absorption, rates of 209 Acids, chlorine, etc 37 Acid, sulphuric 37, 50, 345, 361 phosphoric 37, 50 See names of bases and also Analyses. Ackers Point, 17, 110, 111, 113, 114, 120, 129 Addison Lake 228 Adhesive strength of mortars, ap- paratus for determining 355 Adulterants for paints 4 Agricultural Bulletin No. 99... . 337 Alabaster 334 Alamo, Kalamazoo county 310 Albion 314 Alcona County 334 Algae, fresh water, lime precipita- ting agents 90, 92, 221 See also Chara. Alkalies, effect of, on solubility 210 Alpena County 33, 179, 388, 339 Alpena Portland Cement Company 180, 224, 225, 338 Alsen’s Portland Cement, test. 382, 384 Alma Sugar Company 73 American engine practice . . . 158 American Portland. , 383 American Society of Civil Eng ! ’rs.. 299 Amnicola limosa 98 Amnicola lustrica. 98 Analyses 5 , 46, 121 Analyses by Gustav Bischof 56 by C. H. Hess by Frank S. Kedzie. . . 153 by L. S. Leltz 73 by Delos Fall 153 See Fall. of Bronson clays 239 marls 105 of Cedar Lake marl .75, 76 Analyses.— Con. of cement 360 of clay 137, 171, 222, 241, 288, 291, 296, 322, 323, 327, 329, 330, 332, 334, 336, 337, 339, 346, 347, 352 Bronson 239 Millbury 229 of Cloverdale samples 20 of Coldwater marl 76 of Fremont Lake marl 136 of gas 203 Partial, of samples collected by D. .T. Hale, and analyzed by A. N. Clark 21 of kiln brick 176 of limestone 339 of Littlefield Lake marl 76 of marl 8, 32, 48, 75, 76, 103, 136, 218, 226, 236, 240, 279, 283, 287, 291, 292, 295, 298, 315, 316, 318, 319, 320, 321, 324, 337, 338, 339, 342, 344, 353 methods of 218 Millbury clay 229 Pittsburg coal 231 Portage Lake marl 157 Quantitative 72 of raw material used by Wolverine Portland Ce- ment Co 247 of waters 46, 99 of Goose Lake 234 White Pigeon 103 Antrim County 16, 338 Apparatus to determine depth acd outline of marl beds 108 for cement tests 356 Appearance of marl. 31 See also Marl. Aral, bed of marl near 338 Arbela Twp., Tuscola Co 321 388 GENERAL INDEX. Arenac Co Arendale Hill Armstrong, G. M. S Artesian stratum Asli Aspedin, J Athens. 301, Atlas Co. cement 175, 179, An Gres River. An Sable 336, B. Balker (or Backus) Lake 17, 19, 107, 119, 122, Bad Axe Bair Lake Bakie Loch, Forfarshire Baldwin Ball principle of grinding Barlow Lake Barns Barry 310, Barrytown Bass Lake 322, Bay City Bay County Bayous, marl in Bear Creek Bear Lake Beebe, C. E Beechwood Point Beers, Henry Bellaire Portland Cement Co Bellevue Benliam, F. G. Benzie County 137, 327, Berrien County Betsey River Bevin Lake 276, Bicarbonated salt Bicarbonates 202, See also Calcium, Magne- sium and Analyses. “ Big Marsh ” Big White Fish Lake 20, Bineau Bischof, Gustav 56, Black Lake Blind Island Bog iron Bog iron ore. 19 Bog lime 307, 313, 342 See also Marl. Analyses of 226 Deposit of, near Fish Lake . . 317 General description of 307 near Eaton County 317 near Kelly’s Corners. . 315 near Lacey’s lake 317 lakes 325 near Leslie 316 near Oak Grove 316 origin of 41, 199 Bones 234, 249 Boussingault and Levy 209 Boutron and Boudet 209 Brickyard 147, 150 Britton, Deerfield township 312 Branch County. 313 Brigden, Mr. W. W 349, 350 Briquettes of cement 161, 370, 372 Bristol, H. C 334 Bristol Lake 313 Bronson 1, 103, 104, 163, 167, 171, 173, 182, 228, 384 Bronson Lake 333 Bronson clays. Analyses of 239 Brosch estate 332 Brotherton, W. A. 316 Brown, F. W 176, 177 Buell 320 Buildings 188 Bulrushes 283 Bunsen 209 Burning 174, 179 cost of 178 Burning Dept 178 Bush Lakes 277 C. Calcareous clays 5 Calcareous tufa 335 Calcium 205 Calcium bicarbonate 206, 207 Calcium carbonate. . .34, 47, 50, 56, 313, 318 See also Marl, Bog Lime and Analyses. Solution of 200, 210 Effect of, on marl 296 . 326 . 328 . 171 , 323 , 312 . 159 , 352 , 382 334 , 337 123 321 313 78 333 181 318 85 317 326 333 167 326 16 331 137 282 115 151 306 317 293 338 314 328 277 59 211 333 131 209 209 340 290 124 GENERAL INDEX. 389 Calcium carbonate.— Con. Compared in parts per million, Horseshoe, Long, Guernsey, Pine and Mud Lakes Precipitation of Calcium hydrate Calcium oxide Calcium succinate 87, 88, Calhoun County 311, Calhoun Lake Campbell, E. 1).. .158, 291, 348, 349, Canada, extension of marl in Canton Carbonates See also Analyses. Carbonate of iron. ............... Carbonates, magnesium Solubility of 45, Carbon dioxide, dissolved 86, Caro Carp Lake 328, 330, 331, 332, Carpenter, Prof, R. C Case, James Cascade Township.. 99, Cass County 33, 313, Castalia, Ohio Cedar Lake, in Montcalm County, 18, 33, 80, 91, 99, 100, 322, 323, 324, Cedar Run. Cedar Springs Cederberg, A. H. 190, Cement 160, Adulterated with coarse mat- erials Amount of material for Chemical analysis of Cooling of Constitution of Curing of grinding industry Making of manufacture, Adaptibility of marl to. Mill tests of “Silica” specifications 186, Tensile strength of 297, testing, Humphrey’s report on Tests of .280, Cement and Engineering News, 224, 233 Cement City 33 Cement factories, Burning of 189 Central Lake. 16, 20, 46, 59, 63, 142, 145, 146, 148, 150, 338 Central Lake, Antrim County 143 Chalk 6, 160 Chambers, C. A 312 Chapman Lake 252, 253 Chamberlain’s well 118 Chara, agent in marl production, 73, 78 deposits 306 foetida, analyzed by Gustav Bischof 56 fragilis 89 fragments, Analysis 4, 86, 220 hispida. 78 in lakes of Denmark 78 lyelli 78 material 317 Method of concentration by. 87 plants, Limit of depth ... 86 Source of thick crusts on 79 (Stonewort), 56, 57, 58, 60, 70, 71, 73, 74, 77, 78, 82, 83, 88, 89, 95, 223, 287, 335 Cheboygan County 340 Chemical analyses 282 See Analyses. Chemical effect of marl on fertilizers 4 theory of precipitation of marl 42, 44, 58, 64 Chester.. 311 Chlorophyl, 57, 112, 122 Clam Lakes 338 Clapp, Theodore E 103 Clare 151, 168, 294 Clare County 293, 327 Clare Portland Cement Company. 293 Clarke, J. M 90 Clark, A. N., Analyses 20, 21, 319 Clark’s Lake 309 Clarkston 316 Clays, Michigan 345 Clay.... 105, 147, 149, 167, 170, 192, 296, 299, 322, 324, 327, 329, 331, 333, 336, 338, 339, 346 Analyses of 137, 171, 222, 241, 288, 291, 296, 322, 323, 327, 329, 330, 332, 334, 336, 337, 339, 346, 347, 352 131 55 172 192 89 314 150 350 6 307 342 211 50 200 362 211 333 293 331 314 321 78 335 328 99 298 224 181 294 360 179 172 180 180 189 171 2 358 181 188 300 188 357 390 GENERAL INDEX. Clay- Con. on Clout’s farm Calcareous for cement Hubbell collection of fifty-two 328, Effect of on marl. land Marly Clay shale Clayton Lakes 301, 303, 304, Clinker grinding 177, 179, 183, Clinton County Clippert & Spalding’s brick yard. . Cloverdale Lakes.. 13, 14, 17, 18, 25, 49, 52, 107, 114, 115, 117, Coal 175, dust series Cobb Lake Cohn Coldwater shale Coldwater . . . .77, 103, 104, 105, 106, 175. 228, Cole, John 136, Collins, Mr. F. S., Walden, Mass. . Colon Columbus Township Composition of raw material Commercial importance of. . . See Analyses. 20 5 331 333 230 286 8 227 305 184 320 147 118 185 175 334 318 65 149 313 137 90 55 309 40 30 Crystal Lake 138 Crystals 219 “ Curing” 185 D. Davis, C. A 43, 64, 65,. 97, 100, 199, 217, 273, 292, 322 Dayton, Simon 118 Dean’s clay, Sherman 329, 330 Dean, J. G 170, 200, 233, 332 Detroit Journal 224, 237 Detroit Portland Cement Company 320 Devonian black shale 149 Devore Lake 334 Diatoms 36, 73, 91 Dickson Lake 276 Digging 165 Dipper dredge 166 Dolbee, Chester 317 Dolomite, solubility 210 in clay 222, 327, 336 Dooley, J., farm N. E. of Albion. . . 315 Doolittle, R. E 251, 281 Douglass, C. C.. 312 Douglass Houghton Survey 306 Dowagiac 313 Draining 166, 344 Dredging 166 Dome kiln 161 Drummond Island 341 Duck Lake 46, 142, 228, 328 Consistency 364 Convis Township 314 Cooper Township 310 Cooper, W. F 317 Copemish 328 Corey . . . 332 Corinne ..20, 46, 140, 340 Corunna 293 Cossa .209, 215 Cost, Estimates of .178, 186 of construction of cement plant .194, 196 of manufacture. . 197 of burning 178 Courtis, W. M., Analyses by. . . .287, 318, 337, 339 Crane, use of 166 Crapo Lake. .252, 254 Crow’s Farm 147 Cruse on solubility — 209 E. East Jordan 46, 148, 150 East Lake 328 East Tawas 334 Eaton County 310, 314, 317 Eddystone lighthouse . . 159 Edwards Lake 251, 252, 253 Egyptian Portland Cement Co.... 316 Electrical installation 195 Elk 248 Elk horns 234 Elk Rapids Portland Cement Com- pany Plant 189, 244 Elwell, Pierce 316 Emmet County 340 Empire Cement Company 296 Engel and Ville 210 England, cement materials of 160 Escanaba 15, 62, 138 GENERAL INDEX. 391 Evaporation Exeter Expense of raw grinding at Lupton F. Faiji, Henry Fairbanks Machine Fall, Delos, Analyses by 136, 154, 301, 302, 304," 305, 306, 326,336, 342, 352, paper by Farr, A. W 157, Farr’s Liquid Marl Sampler Farwell Farwell’s Lake Farwell Portland Cement Co Ferric oxide and alumina, ........ Fertilizer, (marl) Fineness of cement Fineness of grinding, effect of on cement *. 181, Five Lakes Fish Lake. 317, Flint pebbles for grinding. ... — Ford, J. B, Forfarshire. Frankfort 138, Freight rates Fremont Lake 99, 135, 240, Fremont Lake marl, Analysis of 136, Fremont flowing well. Fresenius Fuel G. Gamble Lake Gases in rain water Geikie, A Geiger Lake 323, Genesee County 275, 320, George Lake German Portland Cement Co Gladwin County Goose Lake 33, 100, 233, Gordon, C. H Grabau, A, W. 225, Grand Rapids 33, Grand Traverse County — Grand Traverse Region. 141, Grant, John. Grass Lake 33, 228, 286, 310, 338 quantity of marl in 283 Grattan Township 319 Gratiot County 100, 322 Grayling 33, 337 Great Marl Lake 240 Great Northern P. C. Company. . . 327 Green bush 335 Greening, Chas. E 243 Greenland chert pebbles 181 Greensand Marls 5 Green Oak Township 316 Gregory, W. M 227 Gridley 310 Griffin mill.... 180, 181 Grinding, Fineness of 181, 349 Guernsey Lake 19, 46, 52, 53, 107, 124, 130 Gypsum 345 H. Hale, D. J. 20, 21, 91, 199, 292, 313, 325, 384 record of field work by 102 Hanover 228, 310 Hard water lakes. . . 58 Hardness of Water 118 Harrietta 138, 330 Harrisville . 334, 335 Harwood Lake. 313 Hassan, Tagge and Dean 158 Hastings Lake 228 Hauer, K. von. 210 Heath, Geo, L 342 Heat necessary to produce clinker, Amount of 176 Hecla Portland Cement and Coal Company 40, 166, 179, 185, 250, 251, 326, 334 Heim, H, and W 322 Helmer 313 Herring Lakes 138, 297 Hess Lake 240 Hess, W. H,. .. 233, 281 Hickory Creek 154 Higgins Lake 91 Hillsdale 308, 313 Hodge, Dr 235 Hoke, B 332 Holden 320 Holly 316 220 312 174 180 373 353 343 333 11 294 310 292 35 3,4 363 349 293 353 234 248 77 141 193 326 326 46 209 174 334 209 78 324 321 252 291 326 352 147 227 206 338 327 357 392 GENERAL INDEX. Homestead P. 0 Hood, A. P Hope Township See Cloverdale. Horicon Houghton Horseshoe Lake 19, 46, 49, 52, 53, 55, 56, 58, 119, 122, 123, 124, 126, soundings of Howard City Howell Hubbard, Bela Hubbell, J. J 327, Humphrey, R. L..188, 240, 354, 382, Hunt, T. S Hunt, Robt. (4. & Co.. .13, 158, 274, Huron Huron Valley I. Identification 9, Ignition, Loss on Illinois, Extension of marl in Indiana Lakes, Extension of marl in 6, 99, Engineers, Proceedings Ingham County lnterlochen Intermediate Lake 142, See Central Lake. Ionia County Iosco County Iron, bog Iron, effect of on marl Iron oxide, deposit of Isabella County J. Jackson County 309, Jackson Johnson Lake Jonesville Jordan Lake. K. Kalamo Kalamazoo 1, 106, 161, 162, 310, Kalamazoo County 310, marl, analyses Kalamazoo River 310, Kalkaska County MtettB Keating, Mr 325 Kedzie, Frank S„ Analyses — 154, 288, 292, 321, 323, 326, 334 Kent County 311, 318 Kerner 78 Kimball 240 Kiln brick, Analyses of 176 Kiln, continuous 161 Dietsch and Schofer 196 Kiln process of cement manufac- ture 162 Kiln, Rotary 162 Kiln, set 196 Kippenberger 215 Kirk, Ludington 335 Kleinheksel, John H 151 Kynion Lake 301, 303, 304, 305, 314 Kupffer, A 210 Kuster’s work 215 L. Labor Commission 224 Lacey’s Lake 317 Lagendorfer 182 Lake Bluff 332 Lake Cochituate 51 Lake County 327 Lakelands 288, 289 Lake Leelanaw 332 Lake Spring and water 51 Lane, A, C.. . .2, 91, 97, 99, 100, 199, 347, 349 Lansing 218 Lapeer County 321 Lathbury, B. B 158, 191 Latlibury and Spackman 2, 158, 171, 174, 179, 186, 198, 199, 224, 251, 298, 299 LeChatelier, tests by 172, 362 Leelanau County 327, 328 Lehr Lake 301, 303, 304, 305 Leland 333 Leltz, L. G 73 Lenawee County 233, 312, 327, 338 Leoni 310 Leslie Township, marl 307, 316 Liberty 309 Lime kilns ... 3, 140 Lime 345 Lime, see Calcium and Marl. Effect of, on clinker 230 Lime Lake 20, 50, 133, 134, 233 331 181 119 328 341 130 121 325 99 309 338 384 210 280 231 308 308 192 6 167 158 316 142 338 311 334 124 231 50 326 315 310 228 106 319 310 311 314 353 312 338 GENERAL INDEX. 393 Limestone, Analyses Limestone flour Limmea humilis. Lincoln Little Lake 15, Littlefield Lake. .77, 85. 92, 95, 273, 292, Little Marl Lake Little Traverse Bay Little Whiteflsli Lake 131, Livingston County London Long Lake. .17, 18, 25, 46, 52, 58, 62, 106, 107, 110, 117, 118, 119, 120, 123, 124, Caustic marl of Soundings of Loranger, U. R. ..... . ........ Lower Black River Ludington’s Spring Ludwig Lupton 185, 189, 298, Expense of raw grinding at. . Lupton Portland Cement Company Lyell, Sir Charles M. Mackinaw City Macomb County Magnesia. . . . .128, 169, 170, 192, 223, 342, 345, See also Analyses. Magnesia or alumina brick Magnesium carbonate 35, bicarbonate oxide Maine Manistee and Northeastern rail- road... Manistee. 154, 327, 328, Manistee Junction.. 46, Manistee River Manistique 140, Manufacture of Portland Cement from marl. ... Marble. Marengo Marshall. Materials for cement Marl, Acreage of. 50 Pt. Ill Marl— Con. Adaptability of, for cement manufacture 2 analyses.. 8, 32, 48, 75, 76, 103, 136, 153, 157, 218, 236, 240, 279, 283, 287, 291, 292, 295, 298, 315, 318, 319, 320, 321, 324, 337, 338, 339, 342, 344, 353 Appearance of 31 Average of volume to barrel. 168 in bayous 16 beds, without chara 81 bed, sealed 17 bed, increase and decrease in depth of 17 Caustic 124 Chemical precipitation of . . . 58 claims ! 312 and clay in Michigan 343 and clay in properties, De- velopment of, for Mfg. of Portland cement 191 Suitable time for investi- gation of 191 Tests to be made of 192 Necessary composition of . . . . 169 and origin of 32 Contamination of 7 covered with water 13 deposits, Relation of to direc- tion of prevailing strong winds 96 Requisites for 169 Deposition of 22, 63, 66 upon aquatic plants 69 Depth of 169 Distribution of 5, 6, 9 in a single bed, 16 as fertilizer 309 formation 54, 62 found in hard water lakes ... 14 Gradation in quality of 18 Grain of 170 Granular structure of 74 Impure 48 Indication of, by circum- stances of occurrence 47 Interpretation of 34 Lake 275, 276, 292, 322 lakes, Level of 15 Location of, according to counties 312 339 222 98 334 139 293 240 141 132 316 312 126 115 124 251 340 334 118 299 174 297 77 141 315 349 175 210 213 192 97 338 331 150 331 341 158 6 311 311 160 302 394 GENERAL INDEX. Marl— Con. Location of 13, 15 Location of, Conditions gov- erning 22 Location and size of bed 38 manufacture, magnitude of cost of 2 Materials, overlying 29 underlying 28 Natural history of, Contribu- tion to 66 Organic matter in 21 See also Succinic Acid. Origin of 2, 323 and peat 82 precipitate 58 Precipitation of 59 Precipitated in deep water. . 52 Pure 61, 83 Quality of 61, 193 Quality and formation of, Change in 62 Quantity of 278, 283 Settling of 84 shells 97 stages or steps or growth of . . 17 Surroundings of 23 Tufaceous 311 Ultimate sources of 66 Uses of 3 Underlying cedar swamp 93 Wet 177 Marshgrowtli, Lining of 27 Maskego Lake 312 Mason County 327 Mastodon 313 Material for cement, Amount of.. . 294 McLouth, C. D. 137 McMillan 90 Medway 160 Merkell’s 210, 215 “ Merl 99 or Marl Lake 83 Mecosta County . 326 Michigan 65 Michigan Alkali Co., Wyandotte (J. B. Ford) 248 Description of plant 248 Michigan clays 171 Extension of marl in 6 Michigan Miner 227 Michigan Portlaud Cement Co.. 33, 315 Midland County 326 Mikado 335 Milton 311 Milwaukee 182 Millbury 170, 229, 336, 337 Mineral Lake v . 276 Mill tests of cement 358 Mills Lake 252, 255 Millstone 180 Missaukee County 333 Mixing cement 368 Mixing materials, Methods of .... . 161 Mixing and raw grinding 173 Molds, cement 368, 369 Mollusca 68, 101 Monroe 307 Monroe Center 331 Monroe County 312 Montcalm County 322 Montmorency 338 Mortars, Adhesive qualities of 355 Moscow 228 Mosher’s Lake 228 Mosely 78 Motive power 184 Mound Springs 20, 147 Muck, Marly 8 Muck or peat 343 See Peat. Mud or Round Lake 14, 46, 117, 129 Mud Lake, Hope Township. .52, 53, 62, 100, 116, 118 Oakland County 275, 276, 280 Ionia County 320 Genesee County 320, 321 Muir, Ionia County 320 Mullett Lake 340 Munising 139 Murray, G 90 Muskegon County 137, 325 N. Napoleon 309 Naubinw ay marl 33, 340 Nellist, J. F 318 Nelson, W. S 324 Nettle Lake 313 Newaygo County — 325 Newaygo Portland Cement Co., livTi iljgU JTUI lldUU LCUlvUl vvif (Gibraltar) 189, 240, 241 Newberry 172, 176 GENERAL INDEX . 395 Newell’s steam mill 312 New England 65 New Jersey marl 3, 5 New York . 78 Niles 98, 107 Northbrancli Township 321 Northcote 210 North Island. 144, 145, 146 North and South Carolina marls. . 5 O. Oak Grove, Bog lime near 316 Oakland County 275, 308, 316 Oceana County.. 325 Ogemaw County,.... 334 Ohio ...'.167, 337 Engineers’ proceedings 158 Olsen testing machine 371, 372 Olson Lake 325 Omega Portland Cement Company 168, 173, 179, 227, 232, 308, 313 Onekama Lake 28, 154, 328 Organic matter 27, 37, 345 See also Analyses. Origin of marl Orion Osborn Company Osceola County Oscoda County Ottawa Otsego County Overburning Owosso Oxides of iron and alumina P. Paints, Adulterants for 4 Parmelee, H. P. 42 Pasley 355 Pat tests 375 Paw Paw River 154 Peat.. 93, 281, 303, 314, 317, 320, 322, 325, 343 Peerless Portland Cement Co.. 237, 310, 384 Penliallow 90 Peninsular Portland Cement Co. 233, 313 Pentwater 325 .... 321 158, 232 .... 327 .... 337 312, 318 .... 338 .... 179 .... 200 .... 192 Per cent of water used in mixing cement 182, 349, 366 Perry, C. YV., of Clare 293 Petobago Lake 245 Petroleum 175 Pettenkoper 209 Phosphoric acid 37, 50 Pickerel Lake, Newaygo County, 99, 240 Pierson 326 Pierson Lakes 131 Pine Creek 46 Pine Lake. . .19, 46, 53, 81, 107, 126, 131, 148 Deep waters of 53 Soundings at 127 Pisidium 98 Pisidium contortum 97 Pittsfield, Mass 97 Planorbis parvus 97 Planorbis bicarinatus 98 Plant life, Decayed 27 Plants, Fixed 59, 60 Platte Lake, Little 333 Platte River 137, 329 Platte Township. 331 Plummer Lake. 252, 254 Plymouth Township 307 Portage Lake, Onekama. . . .17, 63, 154, 155, 333 “ Portland,” Origin of name 158 Portland cement. Manufacture of. 158 Power, Cost of 185 Prairieville 317 Precipitate of crystals — 217, 218, 220 Presque Isle County 339 Price Lake 52 Proceedings, Ohio Society of Sur- veyors and Civil Engineers, 158 Proceedings, Indiana Eng. Soc 158 Prophet, John. 340 Prospecting, Methods of 29 Prospecting tools 9, 139 Pulaski 310 Pupidse. 98 Pyramid Portland Cement Com- pany 291, 352 Q. Quicklime 3, 308 Quincy 103, 104, 106, 175, 228 396 GENERAL INDEX. R. Raftelee Lake 273, 275, Ransome rotary kiln 1 , 174, Ranr material, Admixture of Record book, Form of Reading- Reed City Reed, George Rice Lake 21, 29, 59, 151, 152, Rice reeds Riley and Sizanne, chara lime- stones Rim roller Riverdale Rives Robinson, H Rock cement Rock Islet Rock Lake 322, Rocky Island. Roman cement Roscommon County Rosevear, W. B. Ross, Kalamazoo Rotary kiln 1, Rotary process Round Lake 46, 107, 110, Rnnciman, J. H., Mill lake Runyan Lake 274, 275, Russell’s farm Russell, I. C 158, 180, 288, 291, S. Sage Lake Saginaw Bay. Saginaw County Sagittaria St. Clair County St. Ignace St. Joseph 307, St. Joseph’s River Saline Springs Salts Bicarbonated 59, See Bicarbonates. Salt (sodic chloride), effect on sol. . Salt wells Sampling cement Sand 28, 161, Sand lake 228, Sand marl cement Sand, Marly Sandstone 310 Sandusky Cement Co 296 Sanilac County 231 Savicki, W. Y 341 Schimper 78 Scirpus lacustris 119 Schizothrix 77, 90, 351 aggregates 236 Schloessing 208, 209, 211 Sedimentary theory 42, 44 Setting, Time of 187 “ Settle backs” 160 Seward 78 Shabno’s 335 Shale 149, 192, 321, 338, 342. Shells 43 Shell deposit 61 Shell theory 41 Shell structure 221 Shepard, W. H 293, 294 Sherman 329, 330, 331, 332 Sherzer, Will H, 312 Shiawassee County 320 Shore wash, spoiling marl 24 Siderite 210 Silica, Effect of, on marl. . . .36, 73, 192, 231 See also Analyses. Silt, under water 26 Silver Lakes 276, 280, 290 Simons, W. H 328 “Slag” 187 Slurry 158, 176 Smeaton 159, 355 Smidth, F. L,,