UC-NRLF QE <m C? B 3 bt,7 ERK, iBRARY VY OF :ALIFORNIA __ i SAKIfl SCIENCES LIBRARY MAR SOME PHASES OF THE PLEISTOCENE OF IOWA BY EMMETT J. fcABLE JGAl BERKELEY UNIVERSITY OF CALIFORNIA EARTH SC/ENCES LIBRARY EXCHANGE THE UNIVERSITY OF IOWA SOME PHASES OF THE PLEISTOCENE OF IOWA WITH SPECIAL REFERENCE TO THE PEORIAN INTERGLACIAL EPOCH A DISSERTATION SUBMITTED TO THE FACULTY OF THE GRADUATE COLLEGE IN CANDIDACY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY DEPARTMENT OF GEOLOGY BY EMMETT J. CABLE EARTH SCIENCES LIBRARY CONTENTS. INTRODUCTION. CHAPTER I. THE HISTORY OF INVESTIGATION AND CLASS- IFICATION OF THE PLEISTOCENE OF IOWA. CHAPTER II. THE DRIFT SHEETS OF IOWA. CHAPTER III. THE INTERGLACIAL DEPOSITS OF IOWA. CHAPTER IV. THE LOESS. CHAPTER V. THE PEORIAN INTERGLACIAL EPOCH. INTRODUCTION. Purpose of this report. Preliminary preparation. Area of investigation. Acknowledgements. CHAPTER I. The History of Investigation and Classification of the Pleistocene of Iowa. Early students of the drift. D. D. Owen. J. H. Hall. A. W. Worthen. C. A. White. T. C. Chamberlin. W J McGee. Chamberlin's classification of the Pleistocene. Work of Iowa geologists. Discovery of the Illinoian drift by Leverett. Chamberlin's revised classification of the Pleistocene. Introduction of the terms Yarmouth, Sangamon and Peorian. Introduction of the term Nebraskan. The lowan drift questioned. Calvin's reply to critics of the lowan. Work of Alden and Leighton. Geological investigations in northeast Iowa by A. C. Trowbridge. Carman's work in northwest Iowa. Work of Dr. G. F. Kay. Table showing history of the classification of the early Pleis- tocene of Iowa. Classification of Pleistocene deposits as used recently in a report of the Iowa Geological Survey prepared in cooper- ation with the United States Geological Survey. Bibliography. 4657V-J CHAPTER II. The Drift-Sheets of Iowa. Classification of Pleistocene deposits as used recently in a report of the Iowa Geological Survey prepared in cooper- ation with the United States Geological Survey. A study of the various drifts. Nebraskan Drift. Kansan Drift. Illinoian Drift, lowan Drift. Wisconsin Drift. CHAPTER III. The Interglacial Deposits of Iowa. Classification of Pleistocene deposits as used recently in a report of the Iowa Geological Survey prepared in cooper- ation with the United States Geological Survey. Basis for making divisions in the Pleistocene Period. Criteria used in establishing interglacial epochs. The interglacial epochs. Aftonian interglacial epoch. Yarmouth interglacial epoch. Sangamon interglacial epoch. Peorian interglacial epoch. CHAPTER IV. The Loess. Early discovery of the loess and introduction of term. First studied in the United States by Sir Chas. Lyell. Early workers on the loess in Iowa. D. D. Owen. W. H. Pratt. C. A. White. J. E, Todd. F. M. Witter. W J McGee. Work of T. C. Chamberlin and R. D. Salisbury in the "Driftless Area." The study of the loess by Professor Shimek. The study of the loess by other Iowa geologists. Theories advanced for the genesis of the loess. The age of the loess as interpreted by geologists. Geographical distribution of loess within the state. Stratigraphic relation of the loess to the various drift-sheets. Oxidation and leaching of the loess, criteria for determining its age. The fossil life of the loess, its value. Summary. CHAPTER V. The Peorian Interglacial Epoch. Work of Leverett in Illinois. The value of the Toronto beds. A comparison of the life of the Mahomet beds with that of the Scarboro beds. The topographic features of the Wisconsin drift compared with the topographic features of the lowan drift. Leaching and oxidation of the Wisconsin and lowan drifts. State of erosion of lowan and Wisconsin drifts contrasted. Weathering of the Wisconsin and lowan gravels. Stratigraphic relation of the Wisconsin drift to the lowan drift. The Peorian interglacial deposits and their significance. Summary. ILLUSTRATIONS. 1. Buchanan gravels, Independence, Buchanan county. 2. Mature erosion of Kansan drift, Washington county. 3. Young valleys in Illinoian drift plain. 4. Granite bowlders on lowan drift plain. 5. Typical undissected lowan drift plain, Howard county. 6. Mantled lowan topography contrasted with Kansan topography. 7. Typical Wisconsin drift plain, Kossuth county. 8. Deposits of Aftonian gravels near Afton Junction. 9. Soil and vegetable zone between Kansan and Wisconsin drifts. 10. Kansan loess topography near lowan border. 11. Loess bordering the Kansan areas, near Osage, Mitchell county. 12. lowan drift capped by loess. 13. Relation of the loess to the Kansan and Wisconsin drifts. 14. Section in brick yard, Iowa City. 15. Fossils of the loess. 16. Loess showing loess kindchen and iron concretions. 17. Drift cut west of Rhodes, Marshall county. 18. Loess section, Winneshiek county. 19. Fossil life of the loess. 20. Peorian cut showing" fossiliferous loess on Sangamon soil. INTRODUCTION. The purpose of this paper is to present some of the chief facts regarding the Pleistocene of Iowa, in connection with which special reference will be given to the Peorian Interglacial Epoch. It will be shown that, just as in the case of the earlier interglacial epochs, there are definite criteria by which the Peorian Interglacial Epoch can be recognized. Moreover, evidence of a distinctive character will be presented with regard to the duration of the Peorian epoch. Before field investigations were begun, the literature bearing on the subject was reviewed carefully. Much helpful information was obtained from personal interviews with Dr. G. F. Kay, who has made a careful study of the Pleistocene, particularly in southern Iowa; Prof. H. F. Wickham, Department of Entomology, State Uni- versity of Iowa, who has studied fossil insects from interglacial deposits; Dr. M. M. Leighton, who has worked on the lowan drift; W. H. Schoewe, Assistant in the Department of Geology, State Uni- versity of Iowa, and others. Parts of two seasons were devoted to field investigation during which time the regions most closely related to the problem were examined carefully. Thorough exam- inations were made of cuts, well logs, borings and surface deposits. Sixty auger borings were made along the border of the Wisconsfn and lowan drifts, to determine the relative amounts of leaching of the two drift sheets. Trips were made to different parts of the state to study the loess. Field work was done chiefly along the margin of the lowan and Wisconsin drifts from Worth county to Hardin county a^d where the loess which is related to the lowan drift pas es beneath the Wisconsin drift. SOME PHASES OF THE PLEISTOCENE OF IOWA 7 CHAPTER I. THE HISTORY OF INVESTIGATION AND CLASSIFICATION OF THE PLEISTOCENE OF IOWA. Dr. (1) D. D. Owen was the first person to give careful attention to the drift of Iowa. Early in 1848, after having traveled over the northeastern part of the state, he became interested in the presence of numerous (granitic bowlders scattered here and there over the surface. It became his firm belief that these enormous granite masses could have been transported to the position in wnich they are now found only by floating ice, which must surely have been drifted in from the north before the land of this area emerged from the sea. During the years that James Hall served as state geologist, much of his time was devoted to the study of the indurated rocks. He was more interested in the problems associated with Stratig- raphy and Paleontology than in the problems of the Pleistocene. His assistant, A. W. Worthen, had, however, observed frequently the surface drift with its numerous erractics, but made no attempt to explain their distribution. (2) C. A. White, in 1867, was the first state geologist to give any detailed consideration to the Pleistocene. He recognized the glacial origin of the surface deposits and referred some of the erratics found in them to their true sources in the granitic and quartzitic ledges lying in regions to the north. Further than this he made no attempt to separate the various drifts into their respective ages. (3) T. C. Chamberlin was the first geologist in America to advance definitely the idea of the duality of the Pleistocene. As early as 1876, he recognized a first and second glacial epoch with a distinct interglacial epoch between them. Later he published his views, but at the time, gave no names to the deposits. (4) W J McGee introduced methods of investigation in Iowa which furnished finally the key to the interpretation of thedrift deposits found within the state. After having traveled extensively over northeastern Iowa, he became convinced that there were two drifts within tne region, and to these he gave the names "Lower" and "Upper" tiU, both of which are earlier than Chamberlin's second glacial epoch. He believed his "Upper" till to be, in general, a continuous ground moraine, composed, for the most part, of a sheet of bowldery clay, yellow, buff, or brown in color. The entire mass showed itself to be interspersed with subangular or rounded and sometimes striated pebbles and bowlders, the greater portion of which he believed to have come from beyond the borders of the state. He made note of the fact that the structure of the drift, like that of ice-formed deposits elsewhere observed by him, was homogeneous in a large way, though heterogeneous in detail. He observed further, that the thickness of the drift ranged from a thin veneer in places to possibly a hundred feet or more in others and that toward its periphery, especially southward, the "Upper" drift- sheet underwent great modification. This modification was revealed in the gradual disappearance of the surface bowlders, the finer disintegration of the drift materials and a characteristic longitu- dinal ridging of the surface topography. It was found that the superficial portion of the drift-sheet passed insensibly into loam, and that the glacial deposit either 8 EARLY STUDENTS OF THE DRIFT graded up into a superimposed mantle of loess or was overlain by the loess unconformably, though frequently an intermediate bed of material was present. It was his firm belief that the "Lower" till was analogous to the "Upper" till in composition and distribution, though the continuity of the "Lower" till was more broken. The materials are much the same though he noted the prevailing color was blue and the alternative color brown. It was observed also that the pebbles were more numerous than in the "Upper" till-sheet, that the granite bowlders were almost absent and that throughout the mass greenstones predominated. In thickness the "Lower" till was found to vary from a thin veneer up to more than 200 feet, so that in spite of its frequent breaks, its average thickness was about twice as great as that of the "Upper" till. It was found that it possessed a more modified periphery than the "Upper" till. To- ward its margin it differentiated into a lower member, composed for the most part of clay, often laminated, and showing frequently evidences of water deposition. Since the "Lower" till was not a surface exposure in northeastern Iowa, it could best be studied only in cuts and well sections. Wherever examined carefully, it was found to grade upward into a soil or bed of humus charged with remains of an ancient forest. The identifiable remains are chiefly coniferous woods, though fragments of other hard woods, impres- sions of leaves and traces of other vegetation are frequently found. These vestiges of an ancient soil and its products are so widespread as to prove beyond doubt, that the older drift was covered with soil and clothed largely with forests before the advance of the "Upper" till ice-sheet. McGee believed that his "Lower" drift might be Miocene-Tertiary in age, and that the interval represented by the forest bed, which he reasoned must have been of great length, might include all Pliocene time. As a result of McGee's work there was made for the first time in northeast Iowa a separation of the drift into a "Lower" and "Upper" till. McGee had read Chamber- lin's report of Wisconsin wherein he had divided the Pleistocene into two distinct glacial epochs with an intervening epoch of deglacation. It was with this classification of Chamberlin in mind that he named his lower dfift "Lower" till and his upper drift "Upper" till. McGee thus looked upon his forest bed as a plane of division between his "Lower" and "Upper" till-sheets and the equivalent in time of Chamberlin's interglacial epoch. He was not aware at this time that his "Lower" till contained not one but two distinct drift-sheets, and that it was between these two that the forest bed he described so well was located. In 1894, after further study of the Pleistocene, (5) T. C. Chamber- lin in his discussion of the "Glacial Phenomena in North America," advanced a threefold classification for the Pleistocene of Iowa, wherein he called the outermost drift Kansan, and in writing of it says "this drift-sheet reaches its maximum southerly extension in Illinois, and it would seem most appropriate to select a name for this formation from this apex. There are, however, in some places, two distinct drift-sheets separated by a soil horizon, and since it is not yet satisfactorily determined whether this division is widespread and important, or merely local, it is not thought advisable to entertain a name for the formation from this locality. Perhaps the extension of the ice southwesterly in the direction of the great arid plains, is equally worthy of consideration. While the formation in Kansas is subordinately divisible, it appears to be an essential unity, and for this reason the name Kansan has been selected for SOME PHASES OF THE PLEISTOCENE OF IOWA. 9 present use as a convenient designation for the outermost drift- sheet." Overlapping- the Kansan, but separated from it by a well defined soil line lies a second and similar drift-sheet which he termed East-Iowan since it had been most carefully worked out by McGee in northeastern Iowa and there displayed its most distinct- ive topographical and lithological characteristics. Overlapping the East-Iowan, and separated from it by a well marked soil line and vegetable accumulation, oxidized and ferruginated zones and ero- sional surface is a third drift which he termed East-Wisconsin. The name East-Wisconsin was chosen since this drift-sheet assumed some of its most prominent surface topography and development in eastern Wisconsin. Later at the suggestion of Upham, the terms East-Iowan and East-Wisconsin were changed to "lowan" and "Wisconsin." McGee and Chamberlin spent some time inves- tigating the drift in Iowa, giving especial attention to the deposits in the vicinity of Afton Junction. After having studied carefully the type sections here, the following interpretation was reached: The upper surface drift exposed directly beneath the loess was taken to be the equivalent of McGee's "Upper" till of northeast Iowa, while the lower drift beneath the gravel deposits was interpreted to be McGee's "Lower" till of northeast Iowa. It was their belief, at this time, that the upper drift at Afton Junction, East-Iowan of Chamberlin, did not extend southward much beyond Afton Junction, while the lower till-sheet here ex- tended much further to the south, and hence into Kansas where it became the surface drift and exhibited its most striking surface characteristics, hence the name Kansan. The interglacial deposits of sand and gravel found at Afton Junction were taken to be the equivalent of McGee's "forest bed" in northeastern Iowa, thus the name Aftonian. It (6) should be noted here that in the original paper by Chamberlin, the term Aftonian was not applied to the gravels which form so conspicuous a feature of the Afton-Thayer cuts. These gravel deposits were considered to represent rather kame-like accumulations upon the surface of the lower and older drift-sheet. As a result of the work of Chamberlin and McGee-the following classification of the Pleistocene may be presented to represent their views: Chamberlin's Earliest Classification of the Glacial Period. Second Glacial Epoch. First Interglacial Epoch. First Glacial Epoch. McGee's Classification McGee and Chamberlin of Northeast Iowa. at Afton Junction. "Upper" till (Chamberlin's East-Iowan) lowan. Forest bed Aftonian. "Lower" till Kansan. After the work of Chamberlin and McGee at Afton Junction, a number of the Iowa geologists, more particularly Bain and Calvin, after much detailed work had been done on the drifts of the state, reached the conclusion that the upper of the two drift-sheets, at the Afton Junction section, was equivalent to McGee's "Lower" till of northeastern Iowa, that the "Upper" till of McGee's area, East- Iowan of Chamberlin, was not present at Afton Junction, and that the lower drift at Afton Junction was distinct from any of the drift-sheets recognized in northeastern Iowa. The Afton Junction 10 WORK OF IOWA GEOLOGISTS region was la^er visited by T. C. Chamberlin in company with other Iowa geologists, and Bain's conclusions were confirmed. After the work of the Iowa geologists at Afton Junction, they applied the term lowan to the "Upper" till of McGee's area, which they agreed did not extend to the Afton Junction region. Professor Calvin then interpreted McGee's area as follows: The "Lower" till of McGee's area is separated from the "Upper" till by a distinct forest bed, but the Iowa geologists had shown that McGee's forest bed, instead of lying between the "Upper" and "Lower" till-sheets, lies between two drift-sheets, both of which are to be found below McGee's "Upper" till. There are therefore, in the northeastern area of McGee, three CO drift-sheets instead of two as reported by McGee, and his forest bed lies between the two lower till-sheets instead of between the "Upper" till and the "Lower" till. (6) Calvin, in his study in McG :e's area, had noted extensive gravel deposits lying between McGee's "Upper" till and "Lower" till-sheets. These gravels were first recognized as a distinct interglacial form- ation at the Illinois Central gravel pit, located in the northwest quarter of section 32, Byron Township, Buchanan county, Iowa. Plate I. Buchanan Gravels (Calvin). When these gravels were first studied, in 1896, it was the current belief that there had been but two ice invasions, except in the region occupied by the Wisconsin lobe. As the Aftonian gravels were thought to hold an intermediate position between the "Upper" and "Lower" till-sheets, and since the Buchanan gravels occupied plainly what seemed to be a similar position, they were first referred to the Aftonian stage. After the work of the Iowa geologists who had proven that the till overlying the Aftonian beds SOME PHASES OF THE PLEISTOCENE OF IOWA n at Afton Junction was Kansan, the "Lower" till of McGee's area and not the lowan or "Upper" till, as had been assumed, a readjustment became necessary. A new adjustment in terms was made and the Aftonian deposits were assigned to their true position beneath the Kansan. The Buchanan .gravels were then placed as representative deposits of the second interglacial epoch. (5) T. C. Chamberlin, in 1895, had published his classification of American glacial deposits wherein he recognized the Kansan, lowan, and Wisconsin ice-sheets. It was in these publications that the Aftonian beds were referred to the interval between the Kansan and the lowan. After Bain's work, and the adjustment that followed, the Buchanan gravels were left as the only recognized deposit of this interval in northeast Iowa. It should be noted here that the use of the term Buchanan for a second interglacial epoch is open to question, since at the time of its introduction, the Illinoian ice-sheet had not been estab- lished. After a detailed study of the drift in southeastern Iowa and southern Illinois, (9) Leverett discovered a till-sheet older than the lowan but younger than the Kansan. The term Illinoian was first used by him in 1894, in his correspondence with his colleagues, it being his desire not to introduce the term into literature until some of the leading geologists had seen the drift and verified his conclusions. In August, 1896, T. C. Chamberlin and H. F. Bain, in, company with Mr. Leverett, visited some of the exposures where the drift was best shown and each arrived at the conclusion that it was necessary to introduce a distinctive name for the newly discov- ered drift-sheet. The term Illinoian was first introduced into liter- ature by T. C. Chamberlin, in 1896, though he gives Mr. Leverett credit for suggesting the term. Because of the more recent discoveries in the Pleistocene, (10) T. C. Chamberlin, in 1896, revised his classification, recognizing five drift- sheets instead of three. Shortly after the discovery of the Illinoian drift, Mr. (H) Leverett reported the finding of a soil zone containing small bones underlying immediately this drift-sheet, to which he applied the term Yarmouth, since the well section in winch the soil zone was discovered, was located near the village of Yarmouth, Des Moines county, Iowa, and because the name of the village is less likely to be confusing than names which are common. The term Sangamon, representing the third interglacial epoch, was first introduced into literature by Mr. (12) Leverett, in 1896. Since the earliest recognition of the occurrence of a definite soil horizon, between what is assumed to be lowan loess and Illinoian till, was first discovered in Sangamon county, Illinois, it was agreed by T. C. Chamberlin and Mr. Leverett, that the most fitting term by which to designate the deposit should be Sangamon. With this discovery Mr. Leverett suggested that since there was no Illinoian glacial stage to break the continuity of interglacial conditions in McGee's area, that the term Buchanan gravels covered but a small time between the retreat of the Kansan ice and the lowan advance. But since there was an interglacial interruption between the Kansan and the lowan ice-invasions, there is need for names which will stand for the weathered zones above and below the Illinoian till-sheet, hence the name Sangamon. The term Peorian was first introduced into literature by T. C. Chamberlin to designate those deposits of interglacial age lying between lowan loess and deposits of early Wisconsin age, though the discovery of the deposits was made by Mr. (13) Leverett. The type 14 BIBLIOGRAPHY 3. Kansan drift (of Iowa geologists) 2. Aftonian gravels, vegetal deposits, soil and weathered zone (of Chamberlin) (including super-Nebraskan "gumbo," or "gumbotil" of Kay) at top of Nebras- kan drift i. Nebraskan drift (of Iowa geologists) (pre-Kansan or sub- Aftonian of Chamberlin) BIBLIOGRAPHY. 1. D. D. Owen. Report of a Geological Survey of Wisconsin and Minnesota, p. 144, 1852. S. Calvin. Proc. la. Acad. Sciences, Vol. 5, 1898, p. 64. 2. C. A. White. Report of the Geological Survey of Iowa, by J. H. Hall and J. D 1 . Whitney, 1858., pp. 187, 200, 210, 221. 3. T. C. Chamberlin. Geology of Wisconsin, Vol. I. 4. W J McGee. Eleventh Annual Report U. S. Geological Sur- vey, 1893, p. 192. 5. T. C. Chamberlin. Amer. Jour. Science, Vol. 45 third aeries, p. 197. 6. H. F. Bain. Proc. la. Acad. Sciences, Vol. 5, 1898, p. 87. 7. S. Beyer. Proc. la. Acad. Sciences, Vol. 4, 1897, p. 58. 8. S. Calvin. Proc. la. Acad. Sciences, Vol. 5, 1898, p. 66. S. Calvin. Buchanan gravels and interglacial deposit in Buchanan county, Amer. Geologist, Vol. 17, 1898, pp. 76-78. 9. F. Leverett. Introduction of the term Illinoian. Proc. la. Acad. Sciences, Vol. 5, 1898, p. 73. 10. T. C. Chamberlin. Classification of the Pleistocene. Jour, of Geology, Vol. 4, 1896, p. 874. 11. F. Leverett. The weathered zone (Yarmouth) between the Illinoian and Kansan till-sheets. Proc. la. Acad. Sciences, ^ Vol. 5, 1898, p. 81. Jour, of Geology, Vol. 6, 1898, p. 238. 12. F. Leverett. The weathered zone (Sangamon) between the lowan loess and the Illinoian till-sheet. Proc. . \a. Acad. Sciences, Vol. 5, 1898, p. 71. 13. F. Leverett. The Peorian interglacial epoch. , Monograph U. S. Geol. Survey, Vol. 38, 1898, pp. 185-190. Jour, of Geology, Vol. 6, 1899, pp. 244-249. SOME PHASES OF THE PLEISTOCENE OF IOWA . 15 14. F. Leverett. "Weathering- and Erosion as Time Measurers," Amer. Jour. Science, third series, Vol. 27, 1909. "Comparison of North American and European Glacial Deposits," 1910. 15. S. Calvin. The lowan drift. Jour, of Geology, Vol. 19, 1911, PP- 577-602. 16. Alden & Leighton. The lowan drift, a review of the evidence of the lowan stage of glaciation. la. Geol. Surv. Reports, Vol. 26, 1915, pp. 55-213. 17. A. C. Trowbridge. Preliminary report on geological work in northeastern Iowa. Proc. la. Acad. Sciences, Vol. 21, 1914, pp. 205-211. 18. J. E. Carman. The Pleistocene geology of northwestern Iowa. la. Geol. Surv. Reports, Vol. 26, 1915, pp. 239-445. 19. G. F. Kay. Gumbotil a new term in Pleistocene geology. Science new series, Vol. 44, 1916, p. 637. 20. G. F. Kay. Some features of the Kansan drift in southern Iowa. Bull. Geol. Soc. America, Vol. 27, 1916, pp. 115-117. 21. Classification of the Pleistocene used recently in a report of the Iowa Geological Survey prepared in cooperation with the United States Geological Survey. la. Geol. Surv. Reports, Vol. 26, 1915, pp. 56-57. 14 BIBLIOGRAPHY 3. Kansan drift (of Iowa geologists) 2. Aftonian gravels, vegetal deposits, soil and weathered zone (of Chamberlin) (including super-Nebraskan "gumbo," or "gumbotil" of Kay) at top of Nebras- kan drift i. Nebraskan drift (of Iowa geologists) (pre-Kansan or sub- Aftonian of Chamberlin) BIBLIOGRAPHY. 1. D. D. Owen. Report of a Geological Survey of Wisconsin and Minnesota, p. 144, 1852. S. Calvin. Proc. la. Acad. Sciences, Vol. 5, 1898, p. 64. 2. C. A. White. Report of the Geological Survey of Iowa, by J. H. Hall and J. D. Whitney, 1858, pp. 187, 200, 210, 221. 3. T. C. Chamberlin. Geology of Wisconsin, Vol. I. 4. W J McGee. Eleventh Annual Report U. S. Geological Sur- vey, 1893, P- T 9 2 - 5. T. C. Chamberlin. Amer. Jour. Science, Vol. 45 third .series, p. 197. 6. H. F. Bain. Proc. la. Acad. Sciences, Vol. 5, 1898, p. 87. 7. S. Beyer. Proc. la. Acad. Sciences, Vol. 4, 1897, p. 58. 8. S. Calvin. Proc. la. Acad. Sciences, Vol. 5, 1898, p. 66. S. Calvin. Buchanan gravels and interglacial deposit in Buchanan county, Amer. Geologist, Vol. 17, 1898, pp. 76-78. 9. F. Leverett. Introduction of the term Illinoian. Proc. la. Acad. Sciences, Vol. 5, 1898, p. 73. 10. T. C. Chamberlin. Classification of the Pleistocene. Jour, of Geology, Vol. 4, 1896, p. 874. 11. F. Leverett. The weathered zone (Yarmouth) between the Illinoian and Kansan till-sheets. Proc. la. Acad. Sciences, M Vol. 5, 1898, p. 81. Jour, of Geology, Vol. 6, 1898, p. 238. 12. F. Leverett. The weathered zone (Sangamon) between the lowan loess and the Illinoian till-sheet. Proc. . la. Acad. Sciences, Vol. 5, 1898, p. 71. 13. F. Leverett. The Peorian interglacial epoch. .', Monograph U. S. Geol. Survey, Vol. 38, 1898, pp. 185-190. Jour, of Geology, Vol. 6, 1899, pp. 244-249. SOME PHASES OF THE PLEISTOCENE OF IOWA . 15 14. F. Leverett. "Weathering- and Erosion as Time Measurers," Amer. Jour. Science, third series, Vol. 27, 1909. "Comparison of North American and European Glacial Deposits," 1910. 15. S. Calvin. The lowan drift. Jour, of Geology, Vol. 19, 1911, PP- 577-602. 16. Alden & Leighton. The lowan drift, a review of the evidence of the lowan stage of glaciation. la. Geol. Surv. Reports, Vol. 26, 1915, pp. 55-213. 17. A. C. Trowbridge. Preliminary report on geological work in northeastern Iowa. Proc. la. Acad. Sciences, Vol. 21, 1914, pp. 205-211. 18. J. E. Carman. The Pleistocene geology of northwestern Iowa. la. Geol. Surv. Reports, Vol. 26, 1915, pp. 239-445. 19. G. F. Kay. Gumbotil a new term in Pleistocene geology. Science new series, Vol. 44, 1916, p. 637. 20. G. F. Kay. Some features of the Kansan drift in southern Iowa. Bull. Geol. Soc. America, Vol. 27, 1916, pp. 115-117. 21. Classification of the Pleistocene used recently in a report of the Iowa Geological Survey prepared in cooperation with the United States Geological Survey. la. Geol. Surv. Reports, Vol. 26, 1915, pp. 56-57. 16 A STUDY OF THE VARIOUS DRIFTS CHAPTER II. THE DRIFT SHEETS OF IOWA. CLASSIFICATION. It is quite generally agreed that Iowa was invaded by five dis- tinct ice-sheets. In no other area are all the drifts known to be represented and in such a way that it is possible to study one drift in relation to other drifts. (^As shown in Chapter I the following is -the classification of the Pleistocene deposits that was used recently in a report of the Iowa Geological Survey prepared in cooperation with the United States Geological Survey. 9. Wisconsin drift (of the Des Moines lobe) 8. (b) Peorian soil and weathered zone (of Leverett) at top of loess and beneath Wisconsin drift (a) Main deposit of loess 7. lowan drift (of Iowa geologists) 6. Sangamon soil, vegetal deposits, and weathered zone (of Leverett) (including super-Illinoian "gumbo," or "gumbotil" of Kay) at top of Illinoian drift and beneath loess 5. Illinoian drift (of Leverett) 4. Yarmouth soil, vegetal deposits, and weathered zone (of Leverett) (including super-Kansan "gumbo," or "gumbotil" of Kay) at top of Kansan drift; also Buchanan gravel (of Iowa geologists) beneath Iowa drift and loess 3. Kansan drift (of Iowa geologists) 2. Aftonian gravels, vegetal deposits, soil and weathered zone (of Chamberlin) (including super-Nebraskan "gumbo," or "gumbotil" of Kay) at top of Nebras- kan drift i. Nebraskan drift (of Iowa geologists) (pre-Kansan or sub- Aftonian of Chamberlin) The oldest drift, as shown by the above classification, is the Nebraskan. Being the (l) oldest, the Nebraskan is for the most part covered by younger drifts. That the drift is distributed widely over the state is shown by data obtained from deep cuts and well logs. Stratigraphically, the drift lies unconformably beneath the Kansan and upon an erosional bed rock surface/ Recent work within the "Driftless Area" of Iowa, by (2) A. C. Trowbridge, of the State Univer- SOME PHASES OF THE, PLEISTOCENE OF IOWA 17 sity of Iowa, has revealed two distinct peneplains, the upper of which has been called the Dodgeville peneplain, and the lower the Lancaster peneplain. The region seems to have been invaded by the older ice-sheets, possibly the Nebraskan and the Kansan. O'n the lower plain Professor Trowbridge has found a highly weathered and oxidized drift which he seems to think may prove to be Nebraskan drift. Because of its high state of weathering and oxidation, it becomes very difficult to distinguish the drift from the residual material. The best exposures are to be seen along the valley walls. The relation of the drift to the two peneplains is shown in figure (i). Figure 1. A is the Dodgeville plain, B the Lancaster, and C sections of the drift. That the drift had not been discovered sooner may have been due to its high state of weathering. From its relation to the Lan- caster plain, it seems quite conclusive that the drift must have been deposited before the cutting of the valleys of the lower plain. Lithologically and petrologically there is a wide variation in this drift from place to place, as is true of the other drifts. In many sections, where the Nebraskan has been studied, (3) it is a dark bluish-black, tough, jointed clay containing a few pebbles and bowlders. When thoroughly dry it breaks with a starch-like frac- ture into small angular fragments, and becomes exceedingly hard. J More recent investigations by (4). J. E. Carman, in Cherokee county reveal this same drift with a color that is gray, rather' than black, and containing tints of various colors such as chocolate, brown, purple, and blue. That the surface of this old drift has undergone marked physical change, as well as chemical change, is shown by the fact that it is covered by gumbotil, except where the gumbotil is eroded away. As shown by Dr. (5) G. F. Kay, the thickness of the 3 i8 A STUDY OF THE VARIOUS DRIFTS gumbotil varies somewhat, but a maximum thickness of about thirteen feet is known. The drift deposited by the second ice-invasion is known as the Kansan. It was one of the most widespread invasions experienced by the North American Continent, and it left its record in a very widely distributed till-sheet. While possibly no more widespread than the Nebraskan in Iowa, it has a much wider areal distribution, since over considerable areas it has not been covered by later drifts. Lithologically, (6) it is composed of a compact, tenacious clay, which when dry, shows, like the Nebraskan, a highly jointed struc- ture. On weathering it breaks down into angular "blocks or frag- ments. Distributed through the body of the till are numerous greenstones and granite bowlders varying in size from a few milli- meters up to a foot or more. A drift, exposed at the surface for long periods of time, should show a high state of oxidation and leaching.} That this is true of the Kansan is shown in figure (2). The lower dotted line shows the average depth of leaching while the upper continuous line, shows the depth of oxidation. 1 i XfF, 10 Figure 2. Curves showing depth of leaching and oxidation of the drift-sheets. It should be noted that averages only are here considered. The great depth of oxidation and leaching shown by the Kansan, is in SOME PHASES OF THE PLEISTOCENE OF IOWA i-; harmony with its great age. The prevailing color, as stated previ- ously, is blue where unoxidized. Where it is oxidized thoroughly the color becomes a dark yellowish brown or brownish red. Early students of the drift thought it possible to distinguish one drift from another by their color and lithological characters. Recent investigation has proven that this is not true always. It is probably true, that all the drifts originally, in their fresh state, were more or less bluish in color, and that the yellowish to yellow- ish brown to brownish red colors now so prevalent on the surface of the various drift-sheets have resulted from long periods of con- tinued oxidation and leaching. The great age of the Kansan drift is shown, by its high state of oxidation and leaching, and by its advanced stage of mature erosional topogn phy. | Plate II. Mature erosion of Kansan drift plain, Washington county, la. (Alden) The main Kansan area is a region in which there are well developed valley systems, each valley system having primary, sec- ondary, and tertiary tributaries. Many of the major streams have incised their channels into the drift to a depth of 100 to 150 feet below the accordant upland level, and having reached grade, they are flowing in^ wide, sinuous courses across the old Kansan drift- plain. Except for local, wide tabular divides found in the southern part of the state, the surface is reduced wholly to slope and valley bottom in about equal proportions, the dissection being more marked in the vicinity of the major streams. The undissected tabular divides found between the streams, as for example, in Washing-ton county, are, as shown by Dr. G. F. Kay, but remnants of the Kansan gumbotil-plain. These tabular divides owe their existence to the fact that streams have, as yet, not reached them. The divides are . conspicuous topographic features, being- very flat and poorly drained. The ice-invasion subsequent to the Kansan, is known as the Illinoian. This ice-sheet advancing from the Labradorean center, 20 A STUDY OF THE VARIOUS DRIFTS crossed the state of Illinois and pushed beyond the Mississippi river, into the southeastern part of Iowa. The lithological characters of the Illinoian drift are much the same as those of the Kansan drift. It is composed largely of cl.ay, with some sand and gravel. The prevailing color of the weathered drift is yellowish brown near the surface but beneath it passes into the blue gray till. It will be seen from figure (2), that the drift shows an average leaching and oxidation of 8 and 15 feet respectively. As to. the amount of erosion shown by the Illinoian drift w r hen contrasted with the Kansan and lowan drift areas, it may be said that in Iowa the Illinoian drift plain presents bro^d upland areas quite flat and of greater extent than similar divides in the adjoining Kansan area. That portion of the Illinoian plain located in southeastern Iowa and across the Mississippi river in Illinois, being near to the major drainage line in the Mississippi basin, is dissected by pronounced valleys of great depth. Away from the major drainage line, where the streams have gentle gradients, the surface of the Illinoian drift is relatively flat with slight undulations. Few drainage lines are pres- ent, and when present, they have always a youthful appearance. Near the Mississippi river, however, in both Illinois and Iowa, the surface of the Illinoian drift is characterized by -broad, flat upland divides and deep vaPeys with numerous deep incised tributary valleys. The valleys of the major streams have a depth of 100 to 150 feet, with well pronounced flood-plains. Contrasted with the Plate III. Young valleys on Illinoian drift, Fairview Sheet, Illinois. (Alden) lowan drift topography, the Illinoian drift-sheet would seem to show considerably more erosion. The lowan drift is not dissected SOME PHASES OF THE PLEISTOCENE OF IOWA 21 by such sharp cut valleys, the slopes are more gentle, and the entire area gives the impression of a drift younger than the Illinoian. Subsequent to the Illinoian, came the fourth ice-invasion, known as the lowan. It had its origin in the Keewatin center, entered the northeastern part of the state, and extended approx- imately as far south as the Iowa river. The total area covered by this sheet was about 9000 square miles, or one seventh the entire area of the state. On the west, this drift sheet is bounded by the Wisconsin, while on the east it in no place reaches the Illinoian drift, except in Clinton county, where a tongue of the lowan reached possibly to the Mississippi river. Since recent investigation into the lithological composition of the drift-sheets has shown them to be very similar, it will be rather difficult to determine accurate- ly the eastern and southern margins of the lowan drift. Its margin, no douBt, is lobate, rather distinct in places, while in others it is difficult to outline. All 'evidence points to a thin ice-sheet, which was devoid of any marked terminal moraine. The drift is largely clay, with locally a mixture of sand and gravel. The common pebbles are greenstones, limestones, schists, quartzites, cherts and sandstones. One of the outstanding differences between this drift and all others, is the presence of many immense granite bowlders on the surface. Plate IV. lowan bowlder west of Cedar Falls. They are prevailingly coarse grained, both dark and light col- ored, though the light colors predominate, and show little effect of 22 A STUDY OF THE VARIOUS DRIFTS weathering. Topographically, the bowlders are found to lie on the surface without respect to specific location. They may be found on the .crests of gentle swells, on the slopes of flat swales, and in the bottom of shallow depressions. The color of the drift is dominantly a light yellowish brown, varying to a buff where highly weathered. Since the drift is nowhere deep as compared with some of the other drifts, it is in many places, leached and oxidized to the base. In auger borings made it was often found that the drift was leached and oxidized to the bottom. The depth of leaching and oxidation, compared with the other drifts, is shown in figure (2). The late Professor Calvin in speaking of the surface topography of this drift, emphasized the importance of a surface drift much as the ice had left it. So slight has been the erosion that very little trenching is Plate V. Typical undissected lowan drift-plain, Howard county. (Calvin) visible even along the major water courses. Where streams have gotten started, their valleys are but mere ditches. It has been suggested that since the lowan ice-sheet was not a thick mass of ice, it left but a veneer of drift which was not sufficient to oblit- erate all the irregularities of the o 1 d mature Kansan surface over w r hich it moved. There are within the lowan area exceptional phases of topography, where the surface presents a marked undu- latory effect. Examples of such areas may be found in Butler, Bremer, Black Hawk, Howard and Mitchell counties. Examination of these areas shows them to be hills of Kansan drift, capped with loess. If the lowan ice-sheet was not a vigorous ice-s'ieet, it is reasonable to assume that there may have been pronounced rough areas on the Kansan surface, over which the lowan ice was not able to pass. These areas are characterized generally by elliptical hills standing prominentlv above the level of the surrounding ground moraine plain. With belter drainage than the ground moraine plain, after the immediate withdrawal of the ice, the-e hills would furnish favorable sites for the growth of vegetation, mollusks would find here a suitable environment and become buried finally in the loess. . The recent work of Alden and Leighton on the lowan drift calls attention to the fact that, while the general features show an erosional topography, they lack many of the details characteristic of such a surface. The valleys do not have sharp definition, but are broad, round-bottomed draws of dendritic pattern having long, smooth slopes. In many cases the heads and even the slopes of the SOME PHASES OF THE PLEISTOCENE OF IOWA 23 valleys contain small basins or ponds. The distribution of the bowlders suggests a promiscuous mantling of the pre-Iowan surface, in some places sufficient to obscure all evidences of the old Kansan topography, while in others, the mantling had the effect of soften- ing the old Kansan surface. Plate VI. Upper part of plate shows lowan topography compared with the Kansan topography beneath. (Alden) The last ice-sheet which invaded the state was the Wisconsin. This ice lobe, which had its origin in the centers east and west of Hudson Bay, entered the north part of the state between Clear Lake and Spirit Lake, and extended southward to and slightly beyond the city of Des Moines. Stratigraphically, this drift-sheet lies un- conformably upon the Kansan and Nebraskan, except at its eastern margin, where it is thought to overlap the lowan. Detailed inves- tigation by the writer along this eastern margin, failed to show a single instance where lowan was to be found beneath Wisconsin. Numerous places were found where it rests upon Kansan drift. It is composed of clay, sand, gravel and bowlders. The color differs from all the other drifts. This is due to the fact that there has been little leaching and oxidation so that the yellowish brown and reddish brown colors of the older drifts are not to be seen in the Wisconsin area. The prevailing color is light yellow where oxi- dized, and blue where unoxidized. Granite bowlders are present but are of darker color than the lowan bowlders and are, in general, not so large. While oxidized to an average depth of three feet, it is common to find it calcareous to the very surface. With the exception of the pronounced terminal moraine which borders the drift, and local recessional moraines within the outer border, the Wisconsin drift-plain presents a level surface with but slight undulations. A STUDY OF THE VARIOUS DRIFTS Plate VII. Typical Wisconsin drift-plain, Kossuth county, Iowa. (Calvin) The plain is composed of interlocking and disconnected, shal- low saucer-like depressions with intervening" swells of slight convex outlines, hence it has been designated frequently as the (?) "saucer type topography." The numerous basins vary greatly in shape, size and depth. Few have outlets and where present, these outlets are but shallow ditches through which the water flows very sluggishly. Few streams are present and where they are found they are always of the extremely youthful type. The surface, as a whole, is little dissected, is poorly drained, and is much as the ice left it. The most marked topographic feature of the Wisconsin plain is the massive terminal moraine which borders it. This moraine stands as a witness to a vigorous ice-sheet that was heavily loaded. Border- ing the moraine are many excellent examples of kames and kame- like eskers, and valley trains, all witnessing to a vigorous outwash from the edge of the ice. BIBLIOGRAPHY. 1. B. Shimek. The Nebraskan drift. Science new series. Vol. 34, 1911, p. 28. H. F. Bain. Aftonian and pre-Kansan deposits in southern Iowa. American Geologist, Vol. 21, 1898, p. 256. 2. A. C. Trowbridge. Geological work in northeast Iowa. Proc. la. Acad. Sciences, Vol. 21, 1914, p. 205. SOME PHASES OF THE PLEISTOCENE OF IOWA 25 3. B. Shimek. Geology of Harrison and Monona counties. Iowa Geol. Surv. Reports, Vol. 20, 1909, p. 307. 4. J. E, Carman. Notes on the Nebraskan drift of the Little Sioux Valley, Cherokee county, Proc. la. Acad. Sciences, Vol. 20, 1913, p. 231. 5. G. F. Kay. Pleistocene deposits between Coon Rapids in Car- roll county and Manilla in Crawford county. Iowa Geol. Surv. Reports, Vol. 26, 1917, pp. 230-231. 6. S. Calvin. Present phase of the Pleistocene Problem in Iowa. Bull. Geol. Soc. America, Vol. 20, 1909, pp. 133-152. , 7. S. Beyer. Geology of Story county. Iowa Geol. Surv. Reports, Vol. 9, 1898, p. 199. 26 CRITERIA FOR FIXING INTERGLACIAL EPOCHS CHAPTER III. THE INTERGLACIAL DEPOSITS OF IOWA. CLASSIFICATION. There are recognized in the state five ice-invasions and four interglacial epochs. The first two interglacial epochs are well established stratigraphically, while the other three are not so defi- nitely fixed. The relation of the drifts to the interglacial deposits is shown in the following table. The following (D classification was used recently as a report of the Iowa Geological Survey prepared in cooperation with the United States Geological Survey. 9. Wisconsin drift (of the Des Moines lobe) 8. (b) Peorian soil and weathered zone (of Leverett) at top of loess and beneath Wisconsin drift (a) Main deposit of loess 7. lowan drift (of Iowa geologists) 5. Sangamon soil, vegetal deposits, and weathered zone (of Leverett) (including super-Illinoian "gumbo," or "gumbotil" of Kay) at top of Illinoian drift and beneath loess 5. Illinoian drift (of Leverett) 4. Yarmouth soil, vegetal deposits, and weathered zone (of Leverett) (including super-Kansan "gumbo," or "gumbotil" of Kay) at top of Kansan drift; also Buchanan gravel (of Iowa geologists) beneath lowan drift and loess 3. Kansan drift (of Iowa geologists) 2. Aftonian gravels, vegetal deposits, soil and weathered f zone (of Chamberlin) (including super-Nebraskan "gumbo," or "gumbotil" of Kay) at top of Nebras- kan drift i. Nebraskan drift (of Iowa geologists) (pre-Kansan or sub- (\ J^. Aftonian of Chamberlin) If the Pleistocene is to be regarded as a period it should be treated and analyzed upon the same basis as any other division of geological time. In dividing the rocks of the earth into groups, systems and series we have used, (i) erosional unconformities, (2) faunal and floral changes, (3) physical changes and, (4) petrological and lithological differences. It is known that the Pleistocene is set off from the underlying formations by a great unconformity, and is thus recognized as a period. If it can be shown that these same factors can be found within the period itself, it would seem that there are good reasons for dividing the Pleistocene period into Epochs, the Epochs into Stages and the Stages into sub-Stages. In attempting to apply the above criteria to the Pleistocene great care should be exercised as there are places where deposits of interglacial age cannot be interpreted with any degree of certainty. SOME PHASES OF THE PLEISTOCENE OF IOWA 27 It is not easy to correlate, in age, deposits widely separated. The first interglacial epoch, the Aftonian, is well established. The factors which have been used in determining it are, (i) the great erosional unconformity between what is known to be older drift and the underlying till-sheet, (2) the presence of gumbotil, (3) thick deposits of peat and soil and, (4) the presence of faunal and vegetal remains. Recent investigations by Dr. G. F. Kay have thrown much light upon the length of the Aftonian interval. The presence of a zone of gumbotil with a maximum thickness of about thirteen feet on the Nebraskan drift, where not eroded away, points certainly to a long time. Gumbotil, as defined by Doctor (2) Kay. who introduced the term, is a gray to dark colored, thoroughly leached, non-laminated clay, deoxidized, very sticky, and which breaks with a starch-like fracture when wet, which is very hard and tenacious when dry, and which is chiefly the result of a long period of weathering of drift. If gumbotil is the result of thorough weathering of glacial till it must be assumed that the land remained in a position such that there was little erosion during the time of the formation of the gumbotil. The leaching of drift is a slow process. While the pro- cess of leaching and oxidation may be rapid at the surface, if all conditions are favorable, these same processes must be much slower as depth from the surface increases. Were it possible to analyze all the factors that affect the above processes, it would be quite probable that we could get a more definite estimate as to the time element. Some of the more important factors that must be consid- ered when an attempt is made to interpret the problem of leaching are, (i) the porosity of the drift, (2) the height of ground water, (3) the amount of soluble matter in the drift and, (4) the effect of plant and animal life. Any attempt to make statements as to the time involved in the leaching of a certain depth of drift must always be made with the above factors in mind. It would seem fair to conclude, however, that if it should take thousands of years to leach the upper two or three feet of drift, it would require many times this length of time to leach and oxidize the next three or four feet. Leaching of drift is brought about chiefly by the perco- lating of meteoric waters, but this is a slow process, especially where drift is clayey, since entering water soon becomes saturated with mineral matter, especially if the drift is highly calcareous. Water once saturated will continue to leach but very slowly, indeed. To leach drift to the depth of 15 to 20 feet must require then the gradual descension of the solution zone, a process which can never be rapid. As has been shown bv (3) Leighton, the rate of descension of the solution zone is probably greatest from the surface down to ground water level, for wet seasons, and less rapid from this horizon to the permanent ground water level where solu- tion becomes very nearly zero. So important is this factor in deter- mining the depth of leaching and oxidation of drifts, that to over- look it. is to give an incorrect interpretation to the fundamental facts of the problem of oxidation and leaching. That the Aftonian was an epoch of great duration seems war- ranted from, (i) the thick layer of gumbotil, an average of 13 feet, formed on the surface of the old Nebraskan drift-plain and the great depth of oxidized and leached drift found beneath the gum- botil and, (2) the pronounced erosional surface produced upon the 28 THE INTERGLACIAL EPOCHS gumbotil-plain before the deposiMon of the subsequent Kansan drift. That the Aftonian was a true interglacial (.]) epoch is further shown by the occurrence of widespread accumulation of peat and soil zones. Large deposits of sand and gravel, which have been called Aftonian, are found to occupy a position between the Nebraskan and Kansan drift-sheets. Plate VIII. Deposits of Aftonian gravels near Afton Junction, Iowa. (Calvin) (5) The age and mode of deposition of these gravels are not yet definitely known. Recent work on these gravels by Kay would seem to indicate that they are not interglacial as thought formerly by Shimek and others, but inclusions in the drift at the time of its deposition. The Aftonian interglacial ep^ch would seem to w r arrant the following conclusions: (i) a long time interval during which the Nebraskan drift plain suffered little erosion but was undergoing marked chemical changes which resulted in the formation of a thick zone of gumbotil, (2) a period of diastrophism, as suggested by Kay, during which there was an elevation sufficient to inaugu- rate erosion and, (3) a long period of erosion on the above plain. The (6) Yarmouth deposits, representing the second interglacial epoch, were first discovered by Mr. Leverett. A distinct peat and soil zone was found to occupy an intermediate position between what was known to be Kansan and a younger overlying drift, called Illinoian. That the Yarmouth was a true interglacial epoch of great duration has been determined positively by, (i) marked SOME PHASES OF THE PLEISTOCENE OF IOWA 29 physical changes which have taken plac3 in the Kansan drift. The surface of the Kansan CO drifc, like that of the Nebraskan, is now known to have had developed upon it gumbotil to a depth of 20 feet. To produce such a great thickness of gumbotil must have involved a long time; (2) no interglacial epoch has a more marked forest horizon than does the Yarmouth. In the investigation car- Tied on by the writer, scores of places were found in Marshall, Hardin, Franklin, Cerro Gordo and Worth counties, where vestiges of this ancient forest were found to c ccupy a horizon that is possi- bly Yarmouth. An excellent exposure of this old forest and soil zone w r as found in Worth county, where the Wisconsin drift lies, not on lowan, as might be expected, but upon Kansan drifc. Plate IX. (i) Wisconsin drift; (2) soil and vegetable zone. Soil and vegeta'ble zone between Kansan and Wisconsin drifts, \\ orth county. The following is a section of the new coal shaft located in the N. W. l /4 Sec. 6, Eldora township,. Hardin county, about one-half mile northwest of the city. Feet. Inches. Black loamy soil i Yellowish clay streaked with ferruginous markings, highly calcareous . 6 Blue clay (probably Wisconsin) grades into (2)-- 12 Bluish gray silt, non-gritty; varied in thickness across the shaft; calcareous 2 30 THE INTERGLACIAL EPOCHS 5. Compact blue clay, highly calcareous containing greenstones, pebbles and pieces of limestone 20 6. Sand and gravel, mostly sand, greenish blue in color 10 7. Fine grained, flour-like clay, greenish blue 5 8. Forest zone; well preserved limbs, tree trunks, and t branches were found some of which were in situ, and as large as six inches in diameter. In every case the vegetal remains were found beneath horizon (7) 4 9. Blue clay, highly calcareous, the bottom part of which is black, as it rests upon the shale over- lying the coal _ Base. From the relation of the Wisconsin drift here to the lowan, it would seem that (4) is loess which is abundant on the east side of the Iowa river bordering the lowan drift. The layer of bluish green sand (6) is possibly Yarmouth, and (8) may be Aftonian. The following is a type section in Story county, Section 12, T. 82 N. Range, 21, W. Feet. Inches. 1. Black soil i 2. Muck and peat 2 3 3. Kansan gumbotil 3 Here Wisconsin drift rests upon a peat and muck zone which represents the time between the deposition of the Kansan, the for- mation of the gumbotil, the deposition of the peat and muck, and the coming of the Wisconsin ice-sheet. A study of the amount of erosion shown by the Kansan drift leads to the conclusion that this interval must have been long, for in many cases the streams have removed the entire thickness of (8) gumbotil. The term Sangamon was first used by (9) Leverett, and was applied to a peat and soil formation lying above Illinoian drift and beneath loess which was thought to be of lowan age. That the Sangamon epoch was a true interglacial epoch is shown by, (T) the presence of a layer of gumbotil covering the Illinoian to a depth of 5 feet or more, (2) marked erosion of the gumbotil-plain, (3) the presence of thick peat and soil zones and, (4) the presence of fossil insect life. The last interglacial epoch, the (10) Peorian, represents the time between the retreat of the lowan ice and the coming of the Wis- consin. In no place has Wisconsin drift been found to overlie the uneroded lowan. Such a place would be ideal for the study of the Peorian interglacial epoch. While peat and soil zones have been reported to underlie the Wisconsin in McHenry, Kane, DeKalb and Bureau counties, Illinois, by Mr. (U) Leverett, there is no evidence that the drift underlying the peat and muck is lowan drift. More recent work on the drift in (12) Illinois, since the writing of Mr. Lev- erett's Monograph, has reduced greatly the area of distribution of the I ! owan drift. The only place where the lowan drift appears to have crossed the Mississippi (13) river, and this is probably question- SOME PHASES OF THE PLEISTOCENE OF IOWA 31 able, was the northwestern and west central part of Rock Island and Whiteside counties, Illinois. The Toronto (14) beds of Ontario, Canada, were thought by Pro- fessor Chamberlin to be Peorian in age, but there is some doubt as to whether these beds should be classed as Peorian or Sangamon. Recent investigation would seem to favor placing these beds as Sangamon in age. The only deposits that are now known to be definitely of Peorian age are the loess deposits, which will be dis- cussed in another connection in this report. BIBLIOGRAPHY. 1. Classification of the Pleistocene of Iowa made in cooperation with the United States Geological Survey, la. Geol. Surv. Reports, Vol. 26, 1915, pp. 56-57. 2. G. F. Kay. Gumbotil, a new term in Pleistocene Geology. Science Vol. 44, 1916, p. 637. The Origin of Gumbotil. Jour, of Geology, Vol. 28, No. 2, 1920. 3. M. M. Leighton. Leaching of the Pleistocene drifts of eastern Iowa. Proc. la. Acad. Sciences, Vol. 22, 1915, p. 19. 4. G. E. Finch. Drift section at Oelwein, Iowa. Proc. la. Acad. Sciences, Vol. 4, 1897, p. 54. S. Calvin. Geology of Chickasaw county, la. Geol. Survey Reports, Vol. 13, 1902, p. 281. 5. H. F. Bain. Aftonian and pre-Kansan deposits in southwestern Iowa. American Geologist, Vol. 21, 1898, p. 255. G. F. Kay. The Aftonian gravels near Afton Junction are they interglacial? Proc. la. Acad. Sciences, Vol. 26, 1919. S. Calvin. Aftonian gravels and their relations to drift-sheets in -regions about Afton Junction and Thayer. Proc. Daven- port Acad. Sciences, Vol. 10, 1907, p. 18. B. Shimek. Aftonian sands and gravels in western Iowa. Sci- ence new series, Vol. 28, 1908, p. 923. 6. F. Leverett. Weathered zone (Yarmouth) between the Illi- noian and the Kansan till-sheets. Proc. la. Acad. Sciences, Vol. 5, ,1898, p. 81. 7. G. F. Kay. Some features of the Kansan drift in southern Iowa. Bulletin Geol. Soc. America, Vol. 27, 1916, pp. 115-117. 8. G. F. Kay. Some evidences regarding the duration of the Yarmouth interglacial epoch. Science, Vol. 43, 1916, p. 398. 9. F. Leverett. Weathered zone (Sangamon) between lowan loess and Illinoian till-sheet. Proc. la. Acad. Sciences, Vol. 5, 1898, pp. 71-80. Jour, of Geology, Vol. 6, 1898, pp. 171-181. 32 BIBLIOGRAPHY 10. F. Leverett. The Peorian soil and weathered zone (Toronto formation). Monograph U. S. Geol. Survey, Vol. 38, 1899, p. 185. 11. F. Leverett. The Peorian soil and weathered zone (Toronto formation). Monograph U. S. Geol. Survey, Vol. 38, 1899, p. 185. 12. H. H. Barrows. Geography of the middle Illinois valley. 111. Geol. Survey Bulletin No. 15, 1910, p. 31. 13. J. E. Carman. The Mississippi valley between Savannah and Davenport. 111. Geol. Survey Bulletin, No. 13, 1909, pp. 40-41. 14. A. P. Coleman. Glacial and interglacial beds near Toronto. Jour, of Geology, Vol. 9, 1901, p. 285. 'f SOME PHASES OF THE PLEISTOCENE O'F IOWA 33 CHAPTER IV. THE LOESS. Since the loess has such an important bearing upon the prob- lem of the Peorian interglacial epoch it will be necessary to review briefly the early history of investigation of loess, especially in Iowa. Loess was recognized and brought into prominence first by German geologists. It w r as in the valley of the Rhine, where for a long time the term loess had 'been in common usage by the peasants, that the term as it is now known, had its origin. After its discovery in Germany, it was not long until it was recognized in other localities of Europe and even in Asia and America. It was observed in this country first by Sir Chas. Lyell. During his second visit to this country, in 1846, he recognized along the lower portion of the Mississippi river valley, in the vicinity of New Orleans, a deposit of loam-like material mantling the bluffs, which he believed compared very favorably with the loess found in the valleys of the Rhine and Danube. After having made a careful examination and analysis of its fossil life, he pronounced it loess, and gave as his theory of genesis, "the flood-theory of the Mississippi river," which theory was accepted by many American geologists until the time of Richthofen's work on the loess in China. In Iowa, loess was first discovered along the Des Moines and Missouri rivers by (1) D. D. Owen who characterized it as siliceous marl. He explained its presence along the Missouri river by the fact that the waters of the Missouri river had at some time, conse- quent to the drift period, been dammed making a vast inland lake, into which was carried a fine, lacustrine-like sediment which en- tombed fresh water and terrestrial forms of shell life. As early as 1868, (2) W. H, Pratt of Davenport, made a very careful study of the loess within the vicinity of the city. He described it as a yellowish clay in which the lines of stratification consisted principally of iron stained streaks and occasional thin layers of sand. The whole presents a distinctly laminated structure. In certain portions of the deposit many calcareous concretions are to be seen, which, in almost every respect, correspond to those described by Lyell from the Rhine valley loess. So firmly did he believe the Mississippi loess to be analogous to the loess of the Rhine valley that he spoke of it as having the same characters. From the presence of so many fragile, unbroken fossil remains, and the apparently horizontal laminations, he reasoned that the entire body pf the loess must have been deposited in comparatively quiet bodies of water. In 1878, (3) C. A. White re-examined the loess along the Missouri river and gave to it the name "Bluff Loess," a term which had been adopted by the Missouri geologists. He interpreted its genesis somewhat differently from those who had preceded him. He be- lieved the material to have been obtained from the Tertiary and Cretacecus formations of the upper Missouri river. This material was washed down and deposited in large inland lakes which were formed subsequent to the close of the glacial period. In 1878, (4) J. E. Todd again investigated the loess deposits of southwestern Iowa and contrasted them in an analytical way with the loess of China, which had been described by Baron Richthofen. 34 EARLY WORKERS ON THE LOESS OF IOWA In writing of the loess he says "the description of the Chinese loess might readily be applied to pur southwestern deposits. The same language might be used word for word, except that a few diminu- tives must be thrown in when altitudes are referred to. We have the same surface features, the same minute vertical tubes inter- joining at acute angles, the same absence of stratification, the same vertical underdrainage and cleavage, the same line of calcareous concretions, which line is usually horizontal, the same yellow earth, and lastly the Missouri, or "Muddy river," which is the close counterpart of the Hoang-Ho, or "Yellow river." Professor Todd found upon closer investigation serious objections to the Richthofen theory in the presence of, (i) aquatic shell life, as he then inter- preted it, (2) evidence of distinct lamination and, (3) the constant, almost uniform, thickness of the loess on the hill tops and hillsides. This peculiar distribution is explained in that, subsequent to the glacial period, the Missouri river became dammed somewhere to the south, forming a large inland lake. Into this lake, sediment was carried and deposited uniformly over the entire bottom. Owing to a complete drainage of the lake, before the inequalities of the bottom had been levelled up by the complete filling of the lake to water level, the material was left at the various altitudes at which it is now found. (5) R. E. Call, writing of the loess in 1882, referred to it as being limited, for the most part, to the extreme western and southwestern part of the state. In a later article, he speaks of finding loess in the vicinity of Des Moines, situated beneath the drift of the last ice- sheet, and makes the following interesting remark: "It has been thought best to record simply the finding of it here and leave to others the forming of any theory as to its mode of origin." After studying critically the fossil life of the loess in this vicinity, he classified it as either aquatic, semi-aquatic or lacustrine, and in so doing, held analogous views to those of Professor Todd as to its genesis. The small size of the fossil shell life he attributed to a pauperitic fauna which must have resulted from the severe temper- ature conditions of the glacial waters. (6) F. N. Witter, Superintendent of Muscatine Public Schools, early became interested in the loess deposits found on the bluffs border- ing the city. He had observed that the loess rested on glacial drift, and that a part of it was stratified. In the underlying drift he found bowlders, coarse gravel, sand and clay, and above, without any sharp line of demarcation, rested the loess which expended to the top of the highest hill. It was observed that the color was yellowish to ashy, that stratification lines, though present in places, were in other places absent, and that scattered throughout the mass, were stony concretions of very irregular form. Large num- bers of land snails, most perfectly preserved, were found in all parts of the loess, except near the base. Since stratification lines were generally absent, Mr. Witter was convinced that the material never could have been deposited by currents of water. It was his belief that deposition took place in a large lake, the bed of which must have been subjected to little or no change during the time of its deposition. (8) W J McGee, during the time of his association with the Archaeological Reconnaisance work, made his headquarters at Farley, Iowa. It was while a member of this organization that he became interested in the drift deposits of northeastern Iowa. Hav- SOME PHASES OF THE PLEISTOCENE OF IOWA 35 ing made a careful study of the loess, he divided the northeastern region into three distinct areas, namely, (i) the northern loess, (2) the bluff and paha area and, (3) the southern loess. The northern area included the driftless region, as then known, and its southern flanks, comprising roughly the area from the Mississippi river westward to the head waters of the Turkey, Volga and North Maquoketa, in eastern Jackson and northern Clin- ton counties, and thence to the eastern part of the territory. With- in this area the loess was noted for, (i) its striking inability to stand for any length of time in steep slopes as shown by the loess further to the south, (2) its approximate homogeneity from summit to base where it was found to pass almost universally into a sheet of sand and gravel and bowlder-charged clay, (3) its universal mantling of the "Driftless Area" though it became very thin and rather inconspicuous, especially on the divides, where it is made up largely of the local material, so that it becomes difficult often to tell it from the residuary clays, (4) its overlapping of the drift margin for great distances, (5) its marked hypsometric distribution, being as great as the relief of the region in which it is found, (6) its variability in thickness from the north to the south and from the Mississippi river to its southern margin, (7) its prevailing light drab color above, with a grading into a bluish gray below, (8) the wide, universal distribution of fossil shells, and more commonly in the bluish loess, (9) the presence of irregular cylindrical-like con- cretions in the upper part of the loess, (10) the lack of stratification lines, except in the basal portions of a few localities and high level terraces, (n) the presence of a pebble bed at the base into which the loess grades, sometimes slowly, at other times rapidly, (12) the greater abundance of loess shells in the southern portion of the area than in the northern and, (13) the predominance of a more or less pauperitic fauna in the south. The loess of the river and paha phase extends over a larger area which lies somewhat to the west of the first area above described, and extends to the Iowa-Cedar divide in Marshall county. From here the line crosses the Iowa river in northern Johnson county and then extends on east to a line connecting the great elbow of the Cedar river with the Wapsipinicon river. Within this area the loess was noted for, (i) its tendency to arrange it c elf in lines or belts which are more or less roughly parallel with the water courses, (2) its increasing tendency, especially in the south- east, to become more homogeneous and fossiliferous, (3) its rather close blending, toward its southern margin, with the underlying drift material, (4) its universal capping of the summits of the pahas and ridges through which the streams had often cut gorges, (5) the tendency to mantle high northern banks of aberrant stream tributaries and, (6) its marked continuity in passing from higher to lower levels. The region of the southern loess described by McGee lay to the south of the river and paha phase and extended from the southeast corner of the above area to the Mississippi river in Muscatine and Scott counties. The loess here was noted to contain abundant fos- sils, loess kindchen, iron tubules and to be continuous from one level to another. Unlike the northern loess, however, it thinned away from the streams and passed into the underlying drift so gradually that it becomes very difficult to tell just where the loess ceases and the drift ' commences. Because of this latter striking peculiarity, McGee thought his southern loess a sort of connecting 36 EARLY WORKERS ON THE LOESS OF IOWA link, not only between divergent phases of the deposit, but between the deposit itself and one commonly regarded as distinct in genesis and period of formation, hence his name of "hybrid loess." He was strongly convinced that the origin of the loess was associated with glacio-fluvial waters. The most noted and detailed early work on the loess was the investigation by (8) T. C. Chamberlin and R. S. Salisbury. Their report of the loess of the "Driftless Area" is one of the classics in geological literature. An attempt is made to analyze the loess as to its composition and physical properties. Their earlier inter- pretation regarding the genesis of the loess has within recent times been modified so as to include the work of the winds as well as the work of glacial waters. In the more recent investigations of the loess, the work of no one man stands out more prominently than that of Professor Shimek, of the State University of Iowa. His study of the loess has extended over a period of more than a quarter of a century. He first became interested in the loess through the study of soils in relation to plant ecology. Soils have much to do with providing a suitable environment for plant growth, as well as providing a feed- ing ground for molluscan life. It was from this viewpoint that Professor Shimek began his intensive study of the loess. After having made a most careful investigation of the loess within the state, he began the study of the loess deposits of other localities without the state. After having made a careful study of all the various theories that have been advanced to account for the origin of loess and its distribution, he was convinced that no other theory could account for all the difficulties so well as the eolian theory. The facts that any theory must answer are, (i) the presence of terrestrial mollusks, (2) the hypsometric and geographic distribu- tion, (3) the absence of stratification lines but the presence of lamination, (4) the absence of aquatic shell life and, (5) the rela- tion to the drift-sheets. As early as 1888, Professor Shimek published a list of the fossils which he found in the loess around Iowa City, and showed most conclusively, that the fossils of the loess are not aquatic, but on the other hand, are terrestrial almost without exception. He has called attention further to the fact, that there is a marsh fauna and a pond fauna which lives in or near small pools of water, but this fauna is not to be considered as aquatic, since by its organiza- tion it is proven that it could not live in the water. He has shown also that habitat determined largely the grouping of the mollusks, each species when living, demanding different environment from other species. The differences are due not so much to temperature as to the moisture conditions. This explains easily the difficulty that R. E. Call and W J McGee found in trying to account for the small shell life of the loess in the vicinity of Des Moines. By making a comparison of the present life, in any region investigated, with th^e fossil life of the same locality today, Professor Shimek has proven beyond doubt, that the climate during which the shell life of the loess was entombed, was much the same as the climate of those same regions today, and that there are very few, if any, differences in the species that were living in these same regions when they were buried in the loess. It may be said with safety that the eolian theory has now become the generally accepted theory for the loess distribution. The names of many other men who have made some study of the loess might be mentioned. Some of the more prominent are SOME PHASES OF THE PLEISTOCENE OF IOWA 37 S. Calvin, T. E, Savage, H. F. Bain, A. G. Leonard, W. H. Norton, and S. Beyer. The most important theories that have been advanced to account for the loess and its distribution are, (i) the lacustrine theory which ran through many forms, from a semi-marine or estu- arine on the one hand, to the forming' of large inland lakes by the damming of rivers on the other, (2) the fluviatile theory which assumed that the loess is more or less associated with running water, (3) the glacio-fluvial theory which assumed that at the maximum extent of the ice, numerous shallow lakes were formed on the surface immediately in front of the ice into which the streams, issuing from the glacier front, deposited their loads of fine material, (4) glacio-eolian which differed from (3) in that, after deposition of the material by glacial streams, the wind played an important part in its distribution and, (5) the eolian theory which makes the agent of distribution the wind. While there is quite general agreement regarding the genesis and mode of distribution of the loess, there are still differences of opinion as to its age. There are those who would make the main body of loess lowan or immediately post-Iowan, while still others would make the loess of two distinct ages, or even more. Since the age of the loess is important in relation to the Peorian interglacial epoch the views of some of the more noted workers regarding the age of the loess will be given. B. Shimek in speaking of the age of the loess says (9) "there is at present no warrant whatever for the reference of all loess, and especially that of the Missouri river, to lowan age. A thin layer of loess is found over a part of the lowan and Wisconsin and no connection has yet been established between the lowan and the loess of the Missouri and Big Sioux rivers. It is known that loess in northwestern Iowa belongs probably to more than one glacial epoch, and Professor Macbride's work in Humboldt county, makes it conclusive that the lowan did not cover the region immediately north of Carroll county as heretofore believed. The correlation of the loess in Carroll county with the lowan drift is accordingly open to considerable doubt. Differences in composition and texture may frequently be observed in the loess which make it evident that not all loess is of the same age as measured with reference to the sev- eral drift-sheets which have extended southward into the latitude of Iowa. The (10) deposition of the loess has continued through all the intervals of the ice age, though in more northerly regions, over which the several ice-sheets passed, there are more or less sharply defined differences between different portions of the loess. The reference of the loess to any one drift epoch must be objected to, and not the, least, on account of frequent presence of numerous land snails which indicate also the existence of a vigorous vegeta- tion. The several loesses represent interglacial and post-glacial epochs during which conditions could not have been materially different from now." Prof. (ID S. Calvin in discussing the loess of northeastern Iowa, contended that the main portion of it was a silt derived from the finer materials of the lowan drift. This was shown to be true by its color, composition, its geographical relation to the margin of the lowan drift, and by its superposition, sometimes on the lowan till and as often on an eroded and oxidized surface of the Kansan. In his work in (12) Howard county, Professor Calvin calls attention to the fact that here, as well as in Mitchell and other counties, there 38 THE AGE OF THE LOESS is a thin veneer of loess that is younger than the lowan and is probably of Wisconsin age. In Winneshiek county, Professor Calvin calls attention to two distinct loesses which are quite widely distributed throughout the county. He makes the older loess close- ly related to the Kansan drift-sheet, while the upper and younger he calls lowan. Observations on the loess near Peoria, Illinois, led to the conclusion that here the loess is as strongly developed as it is on the lowan in Howard county. In his (13) Presidential Address before the Geological Society of America, in 1909, Professor Calvin states that the loess of certain parts of the Mississippi valley be- longs to the interglacial epoch between the lowan and the Wiscon- sin. Prof. (14) H. W. Norton in discussing the loess of the Kansan areas that are found within the lowan in Bremer county, reaches the conclusion, though both gray and yellow loess are present, that the loess is decidedly younger than the Kansan and apparently is lowan in age. Prof. (15) J. L. Tilton in his report on the geology of Warren county, suggests the possibility of more than one loess. He takes as his type section the exposure at the Indianola Brick and Tile Company's plant, situated to the north of the city. Here the upper loess, about 5 feet thick, is fossiliferous, appears darker and is more clayey. The line of separation between the two is more or less sinuous, suggesting a possible downward limit to oxidation, or a disturbed surface upon which the loess was deposited. The lower and darker loess may have been derived from the Kansan drift. Prof. (16) R. S. Salisbury reports loess of Wisconsin age in the vicinity of Green Lake county, Wisconsin, to the east of the south end of Devil's Lake, and again at Ableman's in the same vicinity. In the former exposure, loess is found at an elevation of 150 to 200 feet above the level of the lake and lying on Wisconsin till. Here the loess is not fossiliferous nor does it contain the loess kindchen. A second exposure in the vicinity of the same lake was found at a much lower level, was thicker, fossiliferous, more normal in color and texture and contained concretions. At Devil's Lake the loess was found directly associated with the Wisconsin terminal moraine. Here the moraine is at an elevation of 60 to 80 feet above the lake level, and on the side facing the lake, the loess is associated with fine clay which suggests that it was accumulated in the expanded lake which occupied the present lake and its surroundings at the time of the ice occupancy. The loess here is, no doubt, the fine sediment' that settled from the muddy waters of the lake which stood here for some time after the ice had withdrawn some distance to the east. The exposure at Ableman's is not associated with any of the older drift-sheets. It occurs at a very low elevation on a lacustrine flat made by a lake which formerly occupied the valley of the Baraboo river just 'above the city of Baraboo, during the time that the ice blocked the eastern drainage of the valley. This lake was maintained sometime after the retreat of the ice by a moraine dam. On this flat the loess was deposited probably by glacial waters. It is reported also that loess-like loam is present on the low plain about Camp Douglas. It is possible that the loam- like material was deposited by the flood waters associated with the Wisconsin ice-sheet. In describing the deposits of Crowley's Ridge, Professor (17) Salis- bury refers the loess to two distinct epochs; the first deposition followed the first glacial episode, which is to be interpreted as SOME PHASES OF THE PLEISTOCENE, OF IOWA 39 Kansan; the second epoch of deposition followed the episode which corresponds to the lowan. Prof. (18) J. A. Udden reported a fossiliferous loess-like clay from a well, located in the city of Moline, Illinois. The well is located on the Mississippi bottom, and the loess-like silt was penetrated at a depth of 30 feet. The upper clay is interpreted to be Illinoian drift. Whether these deposits are extra-glacial deposits, made in a lake in front of an advancing ice-sheet, or are material like the surface loess, remains to be determined. Prof. (19) C. W. Wilcox of Ames, Iowa, in describing some loess deposits of southwest Iowa, calls attention to the occurrence of a red, a white and a yellow loess, and suggests that they cannot be different phases of one original loess which has undergone second- ary modification produced by weathering and interstitial deposition of material by infiltration. It has been suggested that McGee had in mind two distinct loess-making periods. A careful study of the literature which has to do with the work of McGee on the loess, fails to reveal any specific statement to more than one loess. It is of interest to note that, after having studied the loess in the vicinity of Des Moines, he made note of the fact that the loess was overlain by drift, while beneath, it graded into drift, and that midway between Des Moines and the Mississippi river, especially in the southern part of the state, it merged into the underlying drift so gradually that no sharp line of division between the two could be drawn. Mr. (20) W. H. Pratt, in his report of the loess of Davenport and vicinity, makes reference to a loess that in the lower part was bluish gray, and ^in the upper part was yellowish, but makes no definite reference to the age. Mr. (21) H. F. Bain, in his report of the Geology of Carroll county, was of the opinion that the age of the loess could not be positively fixed. In eastern Iowa and Illinois the bulk of the loess seems to have been deposited contemporaneously with the maximum advance of the lowan ice, but the loess in northwestern Iowa belongs to more than one ice epoch. Since the lowan ice did not reach this far west any attempt to associate it with the lowan drift is open to some doubt. Prof. (22) T. O. Maybray, in writing of the loess said "the loess of the north has been distinguished as belonging to separate epochs, and a two-fold division of the same in the south has been men- tioned. The yellow loam is here considered as the interfluvial equivalent of the loess, but there seems little reason for making such a division." (23) M. S. Fuller and E, G. Clapp in their study of the loess of the Wabash river, arrived at the conclusion that the marl-like loess is a deposit associated with the glacial waters of the lowan ice-sheet, and is not different in age from the northern loess. Mr. (24) O'. H. Hershey would make all the upland loess of south- eastern Minnesota, a large portion of central and southern Illinois, southern and southeastern Iowa, northeastern Kansas and eastern Nebraska, with a possible greater extension up the Ohio-Missouri rivers, extensive extra-glacial lowan silt. In so doing, Mr. Hershey makes his upland loess and his lowland loess of the same age. It has been suggested that Mr. (25) Leverett, in his Monograph on the Illinoian Glacial Lobe, was undecided as to the age of the loess. After having questioned the existence of a distinct lowan /jo THE AGE OF THE LOESS ice-sheet, and after having- made a study of the drift deposits of Europe, Mr. Leverett wrote an article entitled "Comparison of North American and European Glacial Deposits," in which he refers to a loess covering the earlier Wisconsin terminal moraine of central and eastern Illinois. He notes that the loess is about a meter in thickness, that it is not separated from the Wisconsin drift by any perceptible weathering or erosional interval, and that it displays all the appearance of having been deposited as the Wisconsin ice retreated. A study of the foregoing evidence seems to warrant the follow- ing conclusion that, in most cases, the loess has been assigned to different epochs upon the basis of, (i) difference in color, (2) an apparent difference in texture, (3) an apparent difference in physical aspect, (4) the presence of a distinct ferruginous band which usually separates the yellow loess from the gray loess and, (5) the stratigraphical position of the loess. Within the state there are several well denned geographical areas where the loess occurs. These areas are, (i) around the border of the lowan, (2) along the margin of the major streams which are tributary to the Mississippi river and which may be found immediately south of the border of the lowan drift, (3) along the western border of the state from Sioux City to Omaha, (4) on restricted areas within the lowan border, (5) along the Mississippi river, (6) over much of the Kansan and Illinoian drift-sheets. Along the lowan border the loess presents several marked char- acteristics. It is found, (i) that the loess has here pronounced thickness, and is little thicker, if any, where the streams cross the lowan margin onto the Kansan drift, (2) that while true loess is present in the upper part, beneath it, it becomes ftore or less sandy, Plate X. Kansan Loess Topography near lowan Border, Oakdale, Johnson county. (Leighton) (3) that there is a uniform tendency for the loess to thin out as the distance from the lowan border increases and to become less sandy, more clayey and less porous. Since the edge of the lowan ice expressed itself in long, narrow, marginal-like lobes, the thick- ened belt of loess is very irregular, reaching much farther south in SOME PHASES OF THE PLEISTOCENE OF IOWA 41 some places than in others. In every case where it mantles the Kansan drift, it is observed to rest upon an eroded surface of hig-hly oxidized and leached drift, or on Buchanan gravel. Fre- quently it is possible to find the loess lapping upon the lowan drift some distance back from the margin. While the loess reaches its greatest thickness along the Missouri and Mississippi rivers, deposits of considerable depth occur along the Wapsipinicon, Cedar, Maquo- keta and Iowa rivers, especially where these streams pass from the lowan area to the Kansan. It is quite generally agreed that the source of the loess is from the glacial drift, having been separated from it by the work of the rain, wind, plants, birds, insects and all forms of burrowing life. A large part of the finely comminuted material would, as soon as dry, be picked up by the winds, while still other portions would be carried by the run-off to the major streams. Streams heavily loaded in times of floods, would deposit much of the fine material over their flood-plains in the form of bars or mud flats. When the water receded, and the sediment be- came dry, it would be picked up' by the winds and deposited wherever lodgment was afforded. The agents affecting the deposition of the loess are vegetation and topography. It is well known that forest vegetation is earliest and best developed along streams. This being true, it is easy to understand why the loess might be thicker along the streams. Not only does the vegetation act as a barrier to the winds, but the valley walls act also as an obstruction. Strong currents of air sweeping up from the valley bottoms and impinging on the steep bluffs, produce a rarefaction of the air immediately back from the bluff, and thus a deposition of the wind's load. With strong pre- vailing westerly winds in this latitude, and favorable conditions for lodgment of the loess, there should be a thicker mantling of the eastern valley slopes than the western. This accounts, possibly, for the pronounced ridging of the loess so common near the above mentioned streams. The greatest thickness of the loess is found along the eastern border of the Missouri river valley, where it reaches, in places, a depth of 50 to 100 feet. There are differences of opinion as to the age of the loess here. There are those who would make it the deposition, largely, of one interglacial epoch, while there are still others who would make it the result of three different interglacial (26) epochs. When studied, however, in relation to the drift further to the east, the Missouri river loess shows a very striking and interest- ing stratigraphic relation. Along the Missouri it is found super- imposed upon Kansan drift, but when traced eastward to the margin of the Wisconsin terminal moraine border, it is found to disappear suddenly beneath the younger drift. In no instance is it found to mantle the Wisconsin drift for any great distance back from the Wisconsin drift border. The specific instance referred to, is in the vicinity of Carroll (27) county, where the loess is found to overlap occasionally for short distances, the Wisconsin drift border. This overlapping is due no doubt to the later work of the winds blowing the loess from the Kansan onto the Wisconsin drift. A careful search by the writer for other places where loess might be found on the Wisconsin drift, failed to reveal any. It was found that, wherever the loess occurs over the Kansan drift between the Missouri river and the margin of the Wisconsin drift, it is fossil- iferous whenever it reaches a depth of five feet or more. Where the depth is less than five feet it is found to be non-fossiliferous. 42 DISTRIBUTION OF LOESS WITHIN THE STATE This fact would seem to indicate quite strongly, that the loess deposit is of more recent origin than immediately post-Kansan. When studied with regard to the underlying stratigraphic relations, it is found to rest quite frequently, upon a loess-like clay much like the loess-like clay found mantling the Kansan drift in the southern part of the state. Dr. (28) George F. Kay, in speaking of the loess-like clay of southern Iowa says "while there is, in places, loess of eolian origin on the Kansan drift of southern Iowa, much of the material which has been de- scribed as loess is thought to be not of eolian origin, but to be related more or less closely to the Kansan gumbo. The upper few feet of the Kansan gumbo, which is now limited to the tabular divides, is a fine grained, loess-like, joint clay, in which if diligent search is made, it is possible to find a very few very small pebbles similar to those in the normal gumbo, and it is thought that this loess-like clay is the result of changes that have been going on at and near the surface of the gumbo during the great length of time since the normal gumbo was formed." It is now known that loess- like clay is in places unconformable w r ith loess where loess overlies it. (29) This would seem to establish clearly the fact that the loess in this southern area, is not immediately post-Kansan. If the loess here were of Kansan age, it would seem difficult to account for the fossiliferous state. Loess exposed at the surface for such a great length of time as would be involved, w r ere it Kansan in age, would be thoroughly leached and oxidized, thus removing all traces of shell life. There are, within the lowan drift margin, restricted areas where loess forms the surface covering as, for example, in Bremer, Butler, Black Hawk, Delaware, Benton, Clinton, and Mitchell coun- ties. In all of the above mentioned areas the loess is found to lie upon Kansan drift, or upon rock hills. The former relief features are called pahas. Plate XI. Loess bordering the Kansan Areas southeast of Osage, Mitchell county. (Calvin) SOME PHASES OF THE PLEISTOCENE OF IOWA 43 A careful study of these areas revealed a fossiliferous loess in almost every case, where the loess was of sufficient depth. Most of the places examined show the loess to be thickest on the lee side. These hills are to be interpreted as areas where the pre-Kansan topography was so rough that the thin lowan ice was unable to pass over them, or if it did glaciate them, little drift was left on them. During the loess deposition these prominences would form obstacles to the wind currents and cause deposition. If, as suggested by Alden and Leighton, these areas, after the retreat of the lowan ice, were the first to become clothed with vegetation, a further obstacle to dust transportation by the winds would be offered. Outside the areas above described, loess has been reported to rest on lowan drift in the vicinity of Denver Junction, Bremer county. Plate XII. lowan drift capped by loess-like soil, west of Denver Junction, Bremer county. (Calvin) South of Latitude 410 50' the Kansan drift, except in regions bordering the Missouri and Mississippi rivers, is mantled with a loess-like clay which differs from true loess in that it is more ashy in color, more clay-like in character, is non-calcareous, and is not sharply set off from the underlying drift material. As has already been mentioned, the loess-like clay is possibly related to the Kansan gumbotil and not true loess. Doctor (30) Kay holds that the surface of the Kansan drift, after the Kansan ice withdrew, was a ground moraine plain, which, from the main divide between the Mississippi and Missouri rivers, sloped gently southeast to the Mississippi river 44 RELATION OF THE LOESS TO THE DRIFT and southwestward to the Missouri river. This drift plain was so situated topographically that the weathering agents were effective, but erosion slight. As a result of the weathering during an exceed- ingly long time a grayish, tenacious, thoroughly leached and non- laminated joint clay was formed, which Doctor Kay has called gumbotil. Figure 3. Undissected Kansan Gumbotil-Plain. After a long period of quiescence followed by diastrophism, the streams became active in the dissection of the surface. As a result of a long erosional period, the land was reduced to a mature topography, the gumbotil being removed everywhere except on the interstream divides of the old upland plain. Figure 4. Dissected Kansan Gumbotil-plain. The stratigraphic relation of the loess-like clay to the gumbotil is of importance in determining its age. If the loess-like clay be the result of the further weathering of the gumbotil, its date of origin must be post formation of the gumbotil-plain. SOME PHASES OF THE PLEISTOCENE OF IOWA 45 Figure 5. Dissected Kansan Gumbotil-plain showing the relation of the loess- like clay to the gumbotil and Kansan drift. From figure (5) it is clearly seen that the loess-like clay has been deposited after the erosion of the gumbotil-plain, since it mantles not only the gumbotil, but the lower spurs from which the gumbotil has been removed. This distribution from the higher interstream divides has been affected by rain-wash, wind work, slumping, and other transporting agents. This clearly makes the loess-like material post-erosion of the old Kansan gumbotil-plain in age and not loess of immediate post-Kansan age. The relation of the loess-like clay to the Kansan drift is better seen by a study of the following type section in Lee county as worked out by A. H. Dewey. Atchison, Topeka & Santa Fe R. R. cut, east of Nixon, Iowa, Sec. 31, T. 67, R. 5 W. The top of this cut is at the level of the Kansan upland. Feet. Inches. 1. Loess-like clay. The top two feet very light gray; below yellow to light brown on dry surface; when freshly cut into a chocolate color is seen; the whole is jointed clay and grades into (2). 12 2. Typical gumbotil (Kansan). Gray on a dry surface with a checked appearance; when freshly cut into it shows a drab color; very sticky; contains here and there spots of brown. Pebbles are present; the whole is leached and grades into (3) , 12 3. Oxidized and leached drift (Kansan). Contains pebbles and small bowlders and lime concre- tions. To the bottom of the cut. Pebble counts from the above section. Loess-like clay '__6 pebbles of quartz and chert. Gumbotil (Kansan) 100 pebbles. Quartz 51% Feldspar 3% Chert 34% Felsite i% Greenstones 3% Sandstone 2% Granite 6% If this material covering the gumbotil is loess how shall the siliceous pebbles be accounted for? In a study by several workers of many sections from the Kansan drift area, in the southern part of the state, where the loess-like clay mantles this drift, it was 46 RELATION OF THE LOESS TO THE DRIFT found that the quartz pebbles were present. If the material is related to the gumbotil, then the presence of the quartz pebbles may be explained. Mr. (31) F. Leverett, in his work on the Illinoian drift in Iowa, reports the entire surface covered with a mantle of loess. In writing of the loess he says "the deposit is found not only along the streams, but on the divides between the streams. It is much thicker along the borders of the Mississippi and Illinois rivers than on the divides between them. Over much of the southern part of the Illinoian drift the thickness is from 3 to 5 feet." In Iowa, the reports that have to do with the Illinoian drift, speak of it as being mantled with loess. This interpretation is correct. However, recent investigation of the Illinoian drift in Iowa, shows that in places the Illinoian gumbotil is covered with loess-like clay similar to loess- like clay covering the Kansan in southern Iowa. If the main body of the loess is Peorian in age, it should hold a very definite stratigraphic relation to the Wisconsin, the youngest drift-sheet. This is found to be true. From the edge of the lowan border the loess was traced westward and was found to pass beneath the Wisconsin drift. The eastern edge of the Wisconsin drift was traced from Eldora, Hardin county, south to the southern edge of the Wisconsin, in Marshall county, and data from sections and wells were obtained. Many sections might be given where loess was found to be present beneath the Wisconsin. One of the best localities to show the stratigraphic relations of the loess to the Wisconsin and underlying drift is found in the vicinity of Rhodes, in the southwest part of Marshall county. Plate XIII. Cut showing the relation of the loess to the Wisconsin and Kansan drifts. (Alden) SOME PHASES OF THE PLEISTOCENE OF IOWA 47 It will be seen that the loess here rests upon Kansan gumbotil and is overlain by Wisconsin drift. Many other sections might be given, all showing this same stratigraphic relation of the loess with respect to the Wisconsin, when traced westward from the lowan border. A study of the loess on the western margin of the Wis- consin drift reveals the same stratigraphic relations as are found on the eastern Wisconsin margin. The age of the loess may be determined by, (i) its strati- graphic position with relation to other drift-sheets, (2) unconform- ities in the loess body itself and, (3) the physical and chemical changes which the loess has undergone. If the loess had its origin in fresh unweathered glacial drift, it should show many of the same elements contained in the drift. An analysis of the loess shows it to be composed largely of silica, feldspar, hornblende, pyroxene and mica. These are the dominant minerals found in the glacial drift which is strong evidence that the loess must have come from the igneous and sedimentary rocks which, decomposable on prolonged weathering, would yield the minerals of the loess. The high percentage of magnesium and calcium carbonates independent of the presence of shell life, indicates that the loess coujd never have been derived from the residual clays. If the loess was originally calcareous, it must have been derived from a drift that was fresh and unleached, and accumulated suffi- ciently rapid that leaching and oxidation could not keep pace with the rate of accumulation. Unconformities, such as erosional sur- faces and soil bands within the body of the loess, would be positive evidence of more than one loess making epoch. Since no such evidence has been obtained, the division of the loess has been made upon, (i) the difference in color, (2) the presence of a ferruginous band located generally between the yellow and gray loess and, (3) textural differences. Earlier investigators of loess gave little atten- tion to the problem of leaching and oxidation, but it now seems that this factor has become one of the most important factors in determining the age of loess. As has been shown already, leaching and oxidation of the loess are the result largely of the action of meteoric waters. The factors affecting the depth of leaching and oxidation of the loess are, (i) the porosity of loess, (2) the height of ground water, which in turn is affected by several factors and, (3) the chemical composition of the loess. In making a study of the depth of leaching and oxidation all of the above mentioned factors need to be taken into consideration. In many of the areas studied, where the loess was of sufficient depth, two colors were universally seen; the gray loess below and the yellow loess above. Because of the presence of these two colors there are those who would make the gray loess immediately post- Kansan in age, and the yellow loess, immediately post-Iowan. A careful investigation has been made of many exposures where the two colors may be seen with the result that color cannot be used to separate loess into different ages. There is no zone of leaching or any other weathering effects in the top of the gray loess. The following is a type section which shows the average relations of the gray loess to the yellow loess. OXIDATION AND LEACHING OF THE LOESS Plate XIV. Section in Brick Yard east of Lucas and north of Ronald streets, Iowa City, Johnson county. Feet. Inches. i. Loess yellow, streaked with iron stains; upper 2 feet leached and non-fossiliferous; the lower part highly fossiliferous 5 2. Ferruginous band containing fossils and iron tubules passing from zone (i) into zone (3) 3. Loess gray, unleached and highly fossiliferous 2 2-4 In this specific case iron tubules, shown by the vertical lines on the above plate, were found to pass from the portion of the yellow loess through the ferruginous zone into the gray loess below. Shells were found also to pass from the lower portion of the yellow zone through the ferruginous band into the gray loess below. The fossils from these three zones were compared with the fossils taken from these same three zones from loess obtained at Des Moines. The following plate shows the results obtained. SOME PHASES OF THE PLEISTOCENE OF tO'WA 49 Plate XV. Fossils of the loess. The upper row consists of shells taken from the yellow loess at Iowa City and Des Moines; the middle row shows shells obtained from the ferruginous band from the same localities as mentioned above; the lower row shows shells secured from the gray loess at Iowa City and at Des Moines. The only difference noted is a slight difference in size, but this variation may be found in any one of the above mentioned horizons. THE LOESS FOSSILS AND THEIR VALUE Plate XVI. Loess section showing zone of loess-kindchen and iron connections. In the lower part of the yellow loess and the upper of the gray there may be seen iron pipe stems and loess kindchen. An exam- ination of the above plate, taken in the west end of a cut in an Iowa City brick yard, shows the loess filled with loess kindchen, the white irregular pebbles shown in horizon (2). In studying the problem of leaching and oxidation of loess, where it was found associated with the Wisconsin border, the cuts west of Rhodes and in the vicinity of Carroll are taken as types. Rhodes Cut located on the C. & Milwaukee R. R. two miles west of Rhodes. Feet. Inches. 1. Soil ashy gray 10 2. Drift yellow, the upper two feet leached; contains numerous limestone pebbles (Wisconsin) 22 3. Loess yellowish, calcareous and fossiliferous; the whole streaked with gray; in places large pockets of gray loess may be seen 31 4. Ferruginous zone of varying width and sinuous, also fossiliferous 2 4 5. Loess gray, unleached and fossiliferous resting on Kansan gumbotil 24 6. Kansan gumbotil, grayish and streaked with spots of brown 12 7. Leached Kansan drift Base. SOME PHASES OF THE PLEISTOCENE OF IOWA 5 i Plate XVII. Cut two miles west of the town of Rhodes, Marshall county. Here the loess is calcareous throughout. It is sharply set off from the Kansan gumbotil and the line between the Wisconsin drift and the loess above is sharply drawn. This may be seen from an examination of the above plate. The upper six feet of the yellow loess contains the loess kindchen. Cut north of Carroll on C. & G. W. R. R. Feet. Inches. 1. Soil, grayish black 10 2. Wisconsin drift, grayish to yellowish buff, upper four feet leached and oxidized 6 3. Loess, yellowish gray to buff, the upper two feet leached, the lower unleached and fossiliferous; patches of gray present 10 4. Oxidized layer of loess with iron 2-3 5. Loess, bluish gray and fossiliferous 6 6. Soil band containing carbonaceous matter i 7. Kansan gumbotil 18 The two loesses reported from Winneshiek county by Professor Calvin show similar conditions. OXIDATION AND LEACHING OF THE LOESS Plate XVIII. Loess cut, Sec. 3, Decorah township, Winneshiek county. (Shimek) The dark line in the plate is the ferruginous band separating the two colors. This was the dividing" line between the Kansan loess below and the lowan loess above. Many other sections might be given showing the same relations as already indicated. The foregoing evidence would seem to warrant the conclusion that the loess must have come from a drift that was highly calcareous, and must have been deposited so fast that oxidation and leaching had little effect during the time of accumulation. The loess was origin- ally gray in color, and the two colors now seen so commonly are the result of the oxidation of the upper portion of the loess since its deposition. The iron band so persistent beween the yellow and the gray loess represents the zone of redeposition of the iron material leached out from the overlying loess. The iron concre- tions and loess kindchen represent also material leached from the loess and redeposited. As to textural differences, there is great variation from place to place and even in the same deposit at any one place. Some areas will show loess that is very sandy, while others will show a more clayey composition. A study was made of the loess from the yellow zone, the ferruginous zone and the gray loess from numerous cuts. Slides were made and examined under the microscope to see if the difference in the shapes of the particles and percentage of different minerals varied from the yellow to the gray in any one section, where both were present, with the result that there was practically SOME PHASES OF THE PLEISTOCENE OF IOWA 53 no difference in either shape of particles or percent of mineral content in any of the three zones, the yellow band, the ferruginous band or the gray band. As previously mentioned (see Plate XV.), the fossil life of the loess is of little value in determining the age of the loess, since there is little change in species of the Molluscan life. Plate XIX. Fossil Life of the Loess. The upper row represents shells obtained from loess near the border of the lowan in Marshall county. The middle row are shells from the ferruginous zone at Rhodes, while the lower are shells collected from the gray loess at Rhodes. If there is any value in the shell life at all in helping to determine the age of the loess, it might be said that there is no difference in age between the gray and the yellow loess. The greatest difference noted in the study of the shell life, was the variation in size, which according to Professor Shimek, is due to surface conditions during the time of habitation and not to difference in time. The study of the loess has led to the following conclusions: ;(i) the source of the loess was chiefly the fresh, unweathered gla- I cial drift, some of which may have come from the silts of glacio- ' fluvial waters and silts from dried-up lakes and river-flats; (2) the difference in texture does not mean necessarily a difference in the age of deposition of the loess, but is what might be expected if winds are the distributing agent; (3) the mass of the loess was originally gray in color, the differences in the color seen now are the result of oxidation and leaching. Where the loess is thick enough, so that oxidation and leaching have not reached to the base, the yellow loess is found above and gray loess below. The presence of grayish streaks and patches throughout the yellow lends weight to this conception; (4) the ferruginous zone, which has been thought to represent the old erosional surface of the gray loess is but the horizon where meteoric waters, highly charged with 54 CONCLUSIONS AS TO' THE AGE OF THE LOESS the iron content leached from the overlying loess, have become saturated and have redeposited their load. That this is true is further strengthened by the presence of fossil life in the zone and the iron pipestems which have been found to pass from the yellow zone into the gray loess beneath; (5) some of the loess-like ma- terial, found mantling the drift in the southern part of the state, may not be loess but be related to the Kansan gumbotil; (6) the overlapping of the lowan drift border, and the mantling of local areas within the lowan margin, do not prove the loesses to be of different ages. Withdrawal of the lowan ice edge, if loess deposi- tion was in progress, would offer conditions favorable to lodgment where vegetation may have gotten started on the newly glaciated surface; (7) the study of the fossil life while of little value in de- termining the mode of genesis and age of the loess, is of great value in working out the manner of its distribution and certain climatic conditions during its deposition; (8) physical analyses of the gray and yellow loesses of many sections, fail to show any great differences in the shapes and sizes of the particles, or any great variation in the kind and number of minerals present; (9) that while several loesses have been reported, there seems to have been one epoch more favorable for loess accumulation than any other. This epoch followed the retreat of the lowan ice and is known as the Peorian interglacial epoch. During the Peorian epoch the rate of accumulation of loess must have exceeded the rate of leaching of the lime carbonate. The epoch may be said to have come to a close when the rate of leaching and oxidation ex- ceeded the rate of accumulation. BIBLIOGRAPHY. 1. D. D. Owen. The loess of the Des Moines and the Missouri rivers. Senate Document 26th Congress, second series, Vol. 5, Pt. II. 237, Washington 1871. 2. W. H. Pratt. Loess near Davenport. Proc. Davenport Acad. Sciences, Vol. i, 1867-1868, p. 56. 3. C. A. White. The bluff deposit. la. Geol. Survey Reports, Vol. i, 1870, pp. 103-109. Proc. Amer. Association for the Ad- vancement of Science, Vol. 27, 1878, p. 231. 4. J. E. Todd. Relation of the loess to drift in southern Iowa. Proc. la. Acad. Sciences, 1875-1880. More light on the origin of the Missouri river loess. Amer. Jour. Science, Vol. 16, 1906, pp. 181-194. 5. R. E. Call and W J McGee. On the loess and associated deposits of Des Moines. Proc. Amer. Association for the Advancement of Science, Vol. 27, pp. 21-22. The loess in central Iowa. American Naturalist, Vol. 15, 1881, p. 72. 6. F. N. Witter. Loess of Muscatine. Proc. la. Acad. Sciences, 1875-1880, p. 16. SOME PHASES OF THE PLEISTOCENE OF IOWA 55 7. W J McGee. The Pleistocene history of northeastern Iowa. Eleventh Annual Report U. S. Geological Survey, Vol. n, 1890, pp. 291-303. Philosophical Society of Washington, Nov. 10, 1883. 8. T. C. Chamberlin and R. S. Salisbury. The loess of the "Drift- less Area." Sixth Annual Report U. S. Geological Survey, 1884-1885, pp. 278-306. 9. B. Shimek. The loess. Bull. Natural History State University of Iowa, Vol. 5, 1904, pp. 341, 359, 360. A theory of loess. Proc. la. Acad. Sciences, Vol. 3, 1895, p. 82. The genesis of the loess, a problem in plant ecology. Proc. la. Acad. Sciences, Vol. 15, 1908, p. 57. The loess of the Missouri river. Proc. la. Acad. Sciences, Vol. 14, 1907, p. 237. The distribution of loess fossils. Jour, of Geology, Vol. 7, 1899, p. 122. 10. B. Shimek. Geology of Harrison and Monona counties. la. Geol. Surv. Reports, Vol. 20, 1909, pp. 376-404. Loess and the Lansing man. Bull. Lab. Natural History State Univer- sity of Iowa, Vol. 5, 1904, pp. 337-340. IT. S. Calvin. Geology of Johnson county. la. Geol. Surv. Reports, Vol. 7, 1896, pp. 88-90. 12. S. Calvin. Geology of Howard county. la. Geol. Surv. Reports, Vol. 13, 1902, p. 70. 13. S. Calvin. Present phase of Pleistocene problems in Iowa. Bull. Geol. Soc. America, Vol. 20, 1910, pp. 133-152. 14. H. W. Norton. Geology of Bremer county. la. Geol. Surv. Reports, Vol. 16, 1905, p. 376. 15. J. L. Tilton. Geology of Warren county. la. Geol. Surv. Reports, Vol. 5, 1895, p. 318. 16. R. S. Salisbury. Loess in Wisconsin drift formation. Jour, of Geology, Vol. 4, 1898, p. 929. 17. R. S. Salisbury. Geology of Crowley's Ridge. Ark. Geol. Survey, Vol. 2, 1889, pp. 230-231. 18. J. A. Udden. Silveria formation and other silt deposits. Mono- graph U. S. Geol. Survey, Vol. 38, 1899, pp. 114-115. 19. O. W. Wilcox. On certain aspects of the loess of southwest Iowa. Jour, of Geol. Vol. 12, 1904, p. 716. 20. W. H. Pratt. Loess near Davenport. Proc. Davenport Acad. Sciences, Vol. i, 1876, pp. 96-99. 21. H. F. Bain. Geology of Carroll county. la. Geol. Surv. Reports, Vol. 9, 1898, p. 91. 22. T. O. Mabry. The brown or yellow loam of north Mississippi, and its relation to the northern drift. Jour, of Geology, Vol. 6, 1898, p. 273. 56 BIBLIOGRAPHY 23. 1VL S. Fuller and E, G. Clapp. The marl-loess of the lower Wabash valley. Jour, of Geology, Vol. n, 1903, p. 116. 24. O. H. Hershey. Derivation of the loess of the common type. Bull. Geol. Soc. America, Vol. 14, 1903, p. 169. 25. F. Leverett. The lowan drift-sheet and associated deposits. Monograph U. S. Geological Survey, Vol. 38, 1898, pp. 131, *37> T 38, 143, 153. Comparisons of North American and European glacial drifts, 1910, p. 299. 26. B. Shimek. Geology of Harrison and Monoria counties, la. Geol. Surv. Reports, Vol. 20, 1909, pp. 376, 378, 386. 27. H. F. Bain. Geology of Carroll county. la. Geol. Surv. Re- ports, Vol. 9, 1898, p. 91. 28 G. F. Kay. Gumbotil, a new term, in Pleistocene geology. Science new series, Vol. 44, 1916, p. 637. 29. G. F. Kay. Personal communication. 30. G. F. Kay. Some features of the Kansan drift of southern Iowa. Bull. Geol. Soc. America, Vol. 27, 1918, p. 115. 31. F. Leverett. The Illinoian glacial lobe. General aspects of Illinoian drift-sheet. Monograph U. S. Geological Survey, Vol. 38, 1899, pp. 32, 42, 183, 184. Comparison of North American and European glacial deposits, 1910, p. 299. 32. S. Calvin. Geology of Winneshiek county. la. Geol. Surv. Reports, Vol. 16, 1905, pp. 126-127. SOME PHASES OF THE PLEISTOCENE OF IOWA 57 CHAPTER V. THE PEORIAN INTERGLACIAL EPOCH. The evidence of a Peorian interglacial epoch was based first upon the presence of fossiliferous loess beneath the drift of the Shelbyville lobe of the early Wisconsin and superimposed upon a peat and muck horizon of known Sangamon age. Plate XX. Cut east of Peoria, Illinois, showing fossiliferous loess resting on Sangamon soil. (Leverett) As stated previously, Mr. Leverett has reported rather extensive deposits of peat and muck at the base of the Wisconsin drift in McHenry, Karte, DeKalb and LaSalle counties, Illinois. More recent work on the drifts in Illinois would seem to make it very doubtful whether Towan drift underlies the Wisconsin in the above areas cited by Leverett. Since the above evidence was obtained from well data, it should not be taken as proving that the soil horizon here referred to is of Peorian age. It is just as likely to be Sangamon or Yarmouth in age as Peorian. The Toronto beds have been assigned by T. C. Chamberlin to the (!) Peorian. Professor Chamberlin remarked when he introduced the term "that the grounds for correlating these deposits with the Peorian interglacial deposits of Illinois, are not strong and that further investigation may show these gounds to be erroneous." He further states "that whether the Don beds belong to the Peorian or not, it is certain 58 THE TORONTO BEDS that vegetal beds were formed in the interval of retreat between the formation of the lowan till and the Wisconsin till, and that some of these less well developed and less well known deposits must be looked to as a type of this interglacial horizon if the Toronto beds prove unreliable. Mr. Leverett in speaking of the Toronto beds questioned seriously whether they should be assigned to this horizon, as it is his opinion that there is no lowan drift represented here. The drift underlying these beds is held by Mr. Leverett to be Illinoian and not lowan. If this is true, the deposits represented by the Toronto beds must include the time allotted to both the Sangamon and Peorian interglacial epochs. It would be very difficult to determine the exact time of deposition of these beds, since they do not hold a definite stratigraphic position where the Wisconsin drift overlaps the lowan, which would be the ideal con- dition. To better understand the relation of the Toronto beds to the drift-sheets, the following theoretical diagram is here given. Figure 6. Theoretical diagram of the Toronto beds. 5. Sand, stratified and containing trees like those in number (4) Feet. 55 4. Peaty clays containing trees and other plants of a cool temperate climate 94 3. Stratified clays and sands containing leaves and trunks of trees as well as unio shells, which are now found in middle latitudes of the United States 41 2. Old bowlder clay; deposits resting on preglacial surface of the Hudson river shale (probably Illinoian) ? i. Hudson river shale Base. Beds marked (3) in the above diagram are the Don beds, and those marked (4) and (5) are called the Scarboro beds. After the deposition of the Illinoian till (2) of the above figure, a long period of erosion ensued during which time a deep valley was cut into the drift, and sixteen feet into the underlying Hudson river shale. Upon this old surface there developed a vegetation in which grew SOME PHASES OF THE PLEISTOCENE OF IOWA 59 forest trees, which, according to Professor Penhallow, were much like those found in southern latitudes, in the vicinity of Pennsyl- vania. In this old valley was deposited coarse shingle with which was mixed trunks and branches of red-cedar, elm, oak and pawpaws. Later, owing to rise of water in Lake Ontario, which was possibly due to the damming of the lake's eastern outlet by ice, a deposit of sand and gravel to the depth of sixty feet was made in the lower part of the valley. It is in these beds that A. P. Coleman found leaves, woods, and unio shells. An examination of the fossil remains by Professor Penhallow leads him to conclude that the time represented by the Don deposits was one of moderate climate for this northern region. After the formation of the Don beds, a further deeping took place, until the water level stood 150 feet higher than at the present time. Into the lake at this point flowed a large river, which, according to Professor Coleman, was the outlet of the upper lake-region of interglacial time. The rise of the lake water produced a wide, deep bay into which the river deposited its clay and sand in the form of a large delta. That the deposit was in the form of a delta is shown by the fine lamination and stratifi- cation lines throughout the deposit. In these clays (4) of figure (6) peaty material was obtained. In the sands and gravels near the top (5) of figure (6) were found fossil unio shells and beetle wings. A study of this fauna and flora by Professor Scudder, revealed a climate much cooler than is indicated by the fauna and flora of the Don beds, and one which was more like the present climate of the Ontario region of to-day. It was because of this interpretation placed upon the fossil life that Professor Coleman has interpreted these beds as representing the outwash of the advancing Wisconsin ice-sheet. During the last two years a study of the insect life of the Sangamon beds has been made by Professor Wickham of Iowa State University. The Mahomet beds are located near Mahomet, Illinois, and belong to the Sangamon interglacial epoch. By a study of the in c ect life of this horizon it becomes possible to make a comparison of two widely separated American faunas which are thought to belong to the same interglacial stage, the Sangamon. Professor Scudder recognized in the Scarboro beds 76 species of 33 genera and 8 families, and reached the conclusion that the climate of Ontario, at the time of the deposition of the beds, was similar to that found in the same region to-day, or perhaps slightly cooler, since a num- ber of the recent allies of the fossils have now a more northern habitat. Professor Wickham, in his study of the Mahomet collec- tion, found 10 determinable species belonging to seven genera and four families, these families containing as well the bulk of Scudder' s Scarboro species in the proportion of 36, 8, 19 and 12, respectively; that is, 65 species out of the 76 which Scudder de- scribes, and five of the genera, are common to both of the collec- tions but all the species appear to be quite different. Professor Wickham contends that the basis for deductions as to climate is not very broad, but is of the opinion that the presence of Carabus meander sangamon and Chlaenius plicatipennis, the general north- ern flavor of the remaining species and the entire absence of any without fairly close recent boreal allies, suggests a more rigorous climate than that found in southern Illinois at the present time. Possibfy the climatic conditions were as severe here as they were in Ontario at the date of the formation of the Scarboro beds. It is 60 LIFE OF MAHOMET AND SCARBORO BEDS known that the same genera are living in Illinois to-day, but they occur also very far to the north, extending in part, to the very shores of the Arctic ocean. It would thus seem that an account must be taken of the entire absence of anything characteristically southern. While the species described by Professor Scudder are all different from those described by Professor Wickham, they are very closely allied. The differences are not great enough to indicate any wide dissimilarity in ecological conditions nor a sep- aration by a long period of time, neither do the likenesses prove that the beds in question were synchronous. If the Sangamon was 20,000 to 100,000 years in length, and granting both beds in question were deposited during the same stage, their formation still might be separated by thousands of years, certainly long enough to bring about a differentiation in species. In making comparisons of species, when attempting to correlate widely separated geological deposits widely different lithologically and stratigraphically, we must consider also the wide separation in space of the two places in question. Professor Wickham questions whether 10 species of the Carabiddae, Dytiseidae, Staphylinidae and Chrysomelidae, taken at random in a recent Illinois bog, would all be different from 65 species of the same families collected during the same year and in similar surroundings at Toronto. Seasons may have little to do with the divergence of the character of the fossils, since the peat deposits in which the fossils are found, are forming throughout the warmer part of the year as well as the colder, so that insect remains might readily be preserved at any time. The fact that there is no evidence of intrusion of more southern types in the Mahomet beds would suggest that the deposits perhaps formed when the Illinoian glacial lobe was well advanced on its southern route, previous to a far northern recession. The conclusion reach- ed, so far as the study of the Mahomet species is concerned, is that there is no definite way of determining whether the Mahomet fauna is more closely related to the Scarboro fauna or to those of the present day. All three, Professor Wickham thinks, would be similar to a general facies. The conclusions reached from a comparative study of the faunas of two widely separated interglacial deposits, is that it is not possible to determine positively the exact time relations of the Scarboro beds, but indications point to a nearer relationship with the Sangamon interglacial epoch than with the Peorian. In attempting to arrive at a definite conclusion as to the Peorian interglacial epoch the following lines of investigation have been undertaken: (i) the amount of erosion shown by the two last drift-sheets, (2) the degree of leaching and oxidation of the Wisconsin and lowan till-sheets, (3) the amount of erosion affected on the lowan drift before the advance of the Wisconsin, (4) the outwash products of the Wisconsin border and their state of preservation as compared with similar deposits of the lowan and, (5) the study of the interglacial deposits. One of the evidences as to whether little or much time has elapsed since the deposition of a drift-sheet, is the amount of erosion that has been accomplished since the deposition of the drift. The Wisconsin drift area may be divided into two distinct regions: (i) the terminal moraine area and, (2) the interior or plain-phase area. Since the Wisconsin was a vigorous ice-sheet it built along its margin a prominent terminal moraine. This moraine is more conspicuous for marked relief in some places than SOME PHASES OF THE PLEISTOCENE OF IOWA 61 others. This terminal area may be characterized as a surface broken by knobs and depressions with no apparent attempt at any order of assemblage. Many of the depressions are large, deep, and without outlets. In these deep kettles are to be found Iowa's largest lakes. So recent has been the retreat of this ice-sheet that the carving effect of running water has in no way changed the contour of these hills. The interior plain-phase is characterized by a level surface which is broken by gentle undulating swells and compensating convexities, which give rise to numerous, shallow, undrained saucer- like depressions. So slight is the surface relief that drainage is difficult. In this area the only evidence of erosion is to be found along some of the major streams, which are still flowing in courses they occupied before the advance of the Wisconsin ice-sheet. All the larger valleys have the appearance of immaturity. The only exceptional phase of topography found within the plain-phase is the prominent hills, sometimes solitary, sometimes grouped in irregular clusters, which seem to rise out of the level plain. At first, these hills give the appearance of hills that have been formed by erosional agents, but upon closer examination they are observed to be constructional and of glacial origin. They represent the halting attitude of the ice edge in its northward recession. Excellent examples of these morainic hills may be seen in Hamilton and Wright counties. The numerous small ponds and marshes, the larger lakes associated with the terminal moraine hills, the undissected surface all testify to a surface much the same as it was after the retreat of the Wisconsin ice-sheet. In passing from the Wisconsin border to the lowan drift but little disparity in topographic evidence is noticeable (see Plate VI.). Much of the same undulatory surface effects are present only slightly exaggerated. Drainage lines are a little more pronounced, there being a tendency in places to take on the dendritic pattern. The relief in some portions of the lowan plain is as great as 50 feet. The surface of the lowan, in many places, resembles a pre- glacial topography that has been softened by a thin mantling of drift. This is quite noticeable in passing from some of the larger streams that head in the Wisconsin drift and continue their courses across the lowan. In the lowan area the valleys are large and out of all proportion to the sizes of the streams found in them. Upon closer examination it is discovered that these channels are pre- lowan in age, this fact being shown by the superposition of lowan drift over Buchanan gravels in valley bottoms. In many instances so little erosion has been accomplished by the streams since the deposition of the lowan drift in these valley bottoms that the streams have failed to cut through the thin veneer. The numerous small ponds and marshes found upon the gentle slopes of runs and draws, and in the valley bottoms and on the crests of gentle swells, the small amount of dissection of the drift plain indicate clearly no great period of erosion between the deposition of the lowan drift and the advance of the Wisconsin ice. While it is not possible always to distinguish the different drift-sheets by their color or lithological composition, it is .gener- ally more possible to tell the Wisconsin from the other till-sheets by its color. Except in restricted areas, where the Wisconsin is 62 OXIDATION AND LEACHING OF THE DRIFT more or less sandy, the drift is of a light yellow color, while the older drifts, due to a more advanced stage of weathering, are more of a brownish yellow to a reddish brown color. Its freshness is further shown by the predominance of limestone pebbles and bowlders, many of the latter showing striae as fresh as if deposited but yesterday. Granite bowlders are present and differ from those of the lowan area in size and general color, the color of the Wisconsin bowlders being prevailingly darker than those of the lowan area. Very little difference in the degree of weathering of the Wisconsin and lowan bowlders can be detected, the bowlders of the lowan showing as fresh a surface as do those of the Wisconsin. The difference in age of the lowan and Wisconsin drift-sheets is shown best by the degree of weathering. It will be seen that the depth of oxidation, figure (i) is greater for the lowan. To deter- mine the depth of leaching, auger borings were made on the two drifts, the locations for the borings being chosen where topographic conditions were similar for both drifts. The average depth of leaching shown on the Wisconsin was 1.5 feet, while the lowan showed a depth of leaching of 3.9 feet. A study of the morainal border of the Wisconsin drift where it is in proximity to the lowan drift for outwash plain evidence, failed to reveal any worthy of note. Locally there are subdued phases, but topographically they are of little importance. A type section of such outwash plains is found in sections 5, 8 and 9 of Reeve township, Franklin county. Feet. Inches. 1. Peat, muck and black soil, highly fossiliferous 2-4 2. Clay, fine textured and white in color 6-12 3. Gravel, stratified, containing pebbles which vary in size from one-fourth to one-half inches 2-4 4. Sand and clay, greenish in color, fine grained and uniform in texture 2 5. White sand, even grained and pebbleless 2 During the time that the Wisconsin ice front occupied a position to the west of the above outwash plain, an extensive lake was formed immediately in front of the ice-edge into which was washed the fine rock flour, sand and gravel. The most important outwash deposits found along the eastern border of the Wisconsin terminal moraine are kames, kame-like eskers and the valley trains. All the larger streams which issue from the morainal border are fringed with more or less continuous gravel deposits which are usually in the form of terraces in the larger valleys of the lowan drift area. In some cases these deposits may be traced for great distances beyond the morainal border, and not infrequently, they are found extending back across the terminal moraine onto the Wisconsin drift. Some of the more important deposits are to be found near Gifford, Hardin county, where the South Fork of the Iowa river joins the parent stream; at Sheffield, along Bailey's creek, Franklin county, and at Thornton, on Beaver's creek, Cerro Gordo county. The material is largely sand and gravel, and varies greatly from place to place. The gravels are largely composed of pebbles of limestone, granite, quartzite, greenstones, and quartz. As a whole SOME PHASES OF THE PLEISTOCENE O'F IOWA 63 the gravels are fresh in appearance, except where they are poorly drained. In such positions the gravel is highly iron stained. Kame-like eskers were seen in the valley of Elk Creek, sections 21, 22 and 23, Brookfield township, Worth county, and in section 24, Green township, Cerro Gordo county. The lowan gravels when compared with the Wisconsin gravels show a higher state of weathering, though not nearly so great as the difference between the Kansan and the lowan. This would seem to clearly indicate that the time between the deposition of the lowan gravels and the coming of the Wisconsin ice, was a shorter interglacial epoch than any of the former interglacial epochs. The shorter time is further emphasized by the degree of erosion shown by the lowan drift when contrasted with the erosion of the Wisconsin. That little erosion has taken place in the valleys of the lowan, bordering the Wisconsin, is shown by the fact, that in no instance have valleys of any depth been excavated in the lowan drift. On the other hand, it is found that where streams cross from the Wisconsin onto the lowan, their channels on the latter drift are for the most part in bed-rock, and are out of all propor- tion to the size of the streams found in them. It can be demon- strated clearly that the major streams are flowing in valleys that are pre-Iowan. One of the best examples is that of Lime creek. In the S.E. 1 A Sect. 35 and S. l / 2 of Sect. 36, Fertile township, Worth county, where the river flows along the northern margin of the Wisconsin drift, there is found the following interesting relations of the drift-sheets to the valley. Figure 7. Type section of Lime creek valley east of Fertile. To the north of the river is the broad flood-plain of the river with its distinct terraces covered with lowan drift. Here the Buchanan gravels can be seen, covered with lowan bowlders. To the south of the river and immediately bordering it may be seen the steep bluff reaching to a height of 90 feet. This bluff is com- posed of Kansan drift beneath, and Wisconsin drift above. A section of this bluff was taken which shows the following inter- esting facts. 64 RELATION OF THE WISCONSIN DRIFT TO THE 1OWAN Feet. 1. Sand and gravel, largely sand in the upper part, and yellowish brown color 18 2. Clay, yellowish gray, highly stained with iron along the joints and smaller fissures 2 3. Clay, grayish yellow, highly calcareous; limestone pebbles numerous , 22 4. Clay, bluish, compact, highly calcareous 40 5. Soil and muck on a bluish, compact jointed clay; contains pieces of wood and fossil shells 2 This would seem to indicate clearly that the channel of Lime creek is pre-Iowan in age and probably of post-Kansan origin. Many other streams in the lowan area bordering the Wisconsin, show this same relationship to the lowan drift. A careful and detailed investigation was made along the east- ern margin of the Wisconsin drift, where this drift is thought to overlap the lowan, to see if it were possible to find any evidence of an overlap and to discover, if possible, evidence of interglacial deposits. This border extends from the northern boundary line of Iowa, in Worth county, to Eldora, in Hardin county. Examinations of well logs, sections of the drift, and a new coal shaft, failed to reveal the presence of anything that could be positively called lowan drift, immediately underlying the Wisconsin, along this border. There were no evidences found of interglacial deposits that could be said to be Peorian in age. A further investigation of the Wisconsin drift to the west was made to determine, if possible, whether the lowan drift extended westward beneath the Wisconsin drift. Sections examined in Hardin, Hamilton, Boone and Webster counties failed to show any positive evidence of lowan drift. The result of this investigation has led to the 'following conclusions: (i) if the Peorian was of sufficient length to permit of the develop- ment of a soil and forest zone, the vigor and strength of the Wisconsin ice, when it overrode the lowan area, destroyed all evi- dence of such deposits; (2) If the Wisconsin overlapped the lowan for any great distance, all such evidence has been obliterated. This may be due to the fact that the lowan, being a thin drift-sheet, may have been incorporated with the Wisconsin drift and hence has lost its identity; (3) the only deposit that is so far recognizable as Peorian in age, is the loess. The result of the investigation has led to the following judg- ments regarding the Peorian Interglacial Epoch; I. Geologically, the Peorian was the shortest of the- inter- glacial epochs. The evidence which seems to justify such a conclu- sion is, (i) the small amount of erosion of the lowan drift between its deposition and the advance of the Wisconsin ice, (2) the great similarity of topographic evidence between the lowan and the Wisconsin, (3) the small degree of difference shown in the state of weathering of the two drift-sheets, (4) the depth of leaching of the two drifts is such as to suggest that the two drifts are not con- SOME PHASES OF THE PLEISTOCENE OF IOWA 65 temporaneous, and a comparison of the weathered state of the Wisconsin and lowan gravels shows the Wisconsin gravels to be younger than the lowan gravels, but of no such great difference as is shown between the weathered state of the Kansan and lowan gravels, (5) the time may be largely represented by the loess deposition and the small amount of oxidation and leaching that has taken place in the loess. II. The Peorian interglacial epoch is best represented by the loess deposits. Evidence seems to justify the assumption that the Peorian was a period during which conditions were favorable for the making of thick deposits of loess. That the loess is of Peorian age has been shown by, (i) its stratigraphic relations to the lowan and Wisconsin drift-sheets, (2) by the fact that where it is found to occupy a position between the Kansan and Wisconsin drift- sheets, it shows a sharp line of demarcation from the Kansan drift, often resting on Kansan gumbotil, and is found calcareous and fossiliferous from base to top. This would seem to warrant the assumption that the loess must have been deposited shortly after the withdrawal of the lowan ice and sufficiently rapid not to permit leaching and oxidation to keep pace with the accumulation, and that the Wisconsin ice must have advanced shortly after the deposition of the loess, as the zone of leaching and oxidation be- neath the Wisconsin drift is very shallow. BIBLIOGRAPHY. T. C. Chamberlin. Glacial and interglacial beds near Toronto. Jour, of Geology, Vol. 9, 1901, p. 285. ' r 14 DAY USE RETURN TO DESK FROM WHICH BORROWED EARTH SCIENCES LIBRARY This book is due on the last date stamped below, or on the date to which renewed. Renewed books are subject to immediate recall. LD 21-40m-5,'65 (F4308slO)476 General Library University of California Berkeley <\<\ J