INFORMATION TO USERS

This material was produced from a microfilm oopy of the original document Whila n mooxm vom m w in o io p wii w n 10 iwmiuynw n rapronKm vw uuwwiin i have baan used, fha quality Is heavily dapandant upon tha quality of the original submitted.

Tha following explanation of tachniquas is providad to halp you understand markings or patterns which may appear on this reproduction.

1v 1. Thaa rfv sionar^p a or vara "taraet" a wr for ini^^^a oaaas annaienttv ar ■ • g taekina vh rv lwrr from ^ *a the waa*» rinmwnant a vw ■ a ohotoaraohad is "Miwlrxi Paaelt)". If it was possible to**—I a tha mlseino paga(s) or section, they are spliced into tha film along with ad|aoant pages. This may have necessitated cutting thru an image and duplicating adfacant pages to insure you complete oontinuity.

2. Whan an image on the film is obliterated with a large round black mark, it •n tnmeraon to i m pnoioinpnvr sw pm o um to oopy nwy iwv mowed during exposure and thus causa a blurred image. You will find a q o o g m o po t t o pipm vnMyaomirvnw*

3 • |aMk^MnINn mI fvMPi QvHrlll|Of OnVlf UBif aAb VMS pii«MAae AS Oi PSaM Oil OlOIOolW OVUip pnoioprOpMQ w o pnorofraprar t o h o w o q i qvtiniw iiisuiuu m rr^AAelnmQiiinp am lAAtr 011 skA fnoomtm* la10 I10a ou n o v M y a*so luAiiM N p n poovoini ma C 0sA 11m, uppir NfiIa J a ksikdrano oornoraaaa Aa .or i 1a Bwpim_M ^k^ASiw i inaaia J 00Aa conimuoaam Ma ii A pnososnimL a Aa Ihm ^iramaa ^m m|Ads i A «dm n_l^ke p n S^Am opuosaai Ia I aa ikqovn^AIAmA MiMMvvmi Ai omam mB oam v^^J m pa ^. * 11 Bm nsssiioYi ^^BA^MAAMi AAAelMMiAMnoooiiinp o| a oononmoAAMelMaU^d Aaa flpmnIm — k^^lMMlMA oopnnsnp Ld oa I k jam jw Stoma Mftfiim A AAMirow 010 ^^^elMaiLAA oononuinj AM oni^A llunoi oompleta.

d VL a M A l A a l k i jhd A tMAaa I m m I a a ^ A i L a A s L a A a m A i M M M C ^ M e M m a w b ^ a a ^ A a C u a M i i a 4 * in i rnofonry orumts h q w i i v t o o t o i h o i i oqviomi 10 or p v n n v M if however, a somewhat higher quality reproduction oould be made from rfMkAAAAA^MkANpiiOvopnpns MW if ^^^AMelAfn w iim aa0 0 o A^a i aaMjAaSteAAJiMM unomroomosnp ovm^ A^Ao n ^ onowoiwon#a oim< AILa Aa pM^ rm oA wa mor2 SfMMkAAAMAAIA^A pnototropno ff ^ irtoyaab * d o ^ou v|J qaam w iJ o akA looroonmAaaieiAAAl J^Ai mam pA koy u .mMlalAAwrninj the Order Department ^ving dw catalog number, dtfla, author and specific pages you sM i reproduced.

8. PLEASE NOTE: Soma pages may have indistinct print. Filmed as received.

Xsraa Unhrarslty Mon N m set Na*ei Sm* hn * Am awst , mm*sm sates 75- 11,415 QUINN. NIchMl John. 1946- THE GLACIAL GEOLOGY OF ROSS COUNTY. OHIO. Tho Ohio Stat* Uhlvtrslty* Ph.D.. 1974 Otology

Xsrox University Microfilms, Ann Mmr. Michiganw o t

THIS DISSERTATION HAS K E N MICROFILMED EXACTLY AS RECEIVED. THE GLACIAL GEOLOGY 0? ROSS COUNTY, OHIO

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

By Michael John Quinn, B.S., M.S.

******

The Ohio State University 1974

Reading Committee: Approved By Richard P. Goldthwait Sidney E. White Robert L. Bates

Advise? Department of Geology and Mineralogy ACKNOWLEDGMENTS

I wish to express my gratitude to the Ohio Division of Geological Survey for financial assistance during the field investigation. Thanhs are also due Mr. James Petro of the Soil Conservation Service, U. S. Department of Agriculture, for discussing characteristics of soils in Doss County and providing unpublished information on the pedology of the region. My sincerest thanks go to Dr. Richard P. Goldthwait, my adviser, for his continual guidance through all facets of this study. The manuscript was appreciably improved through the critical reading and suggestions made by Dr. Goldthwait, Dr. Sidney E. White, and Dr. Robert L. Bates. I want to thank my fellow students John McKeon, Steve Derksen, Marc Hoyer, and John Muskopf for helpful suggestions during the laboratory and writing phases of thi3 study. Special appreciation goes to my wife, Anne, for her help in preparation of the manuscript and glacial map. VITA

August 14, 1946 . • • Born - Plattsburgh, New York 1968 ...... • B.S. Geology, St. Lawrence University, Canton, New York 1968-1970 ...... United State3 Army 1970-1971 ...... Teaching Assistant, Department of Geology, The Ohio State University, Columbus, Ohio 1971-1972 ...... Research Assistant, Department of Geology, The Ohio State University, Columbus, Ohio 1972 • •••..»• M.Sc. Geology, The Ohio State University, Columbus, Ohio 1972-1974 ...... Teaching Associate, Department of Geology and Mineralogy, The Ohio State University, Columbus, Ohio

PUBLICATIONS

"Late Glacial History of the Cedar Bog Area"; Ohio Biol* Survey, Info. Circ. no. 4, pp. 7-12, 1974. "The Glacial Geology of Champaign County, Ohio"; in press, with R. P. Goldthwait; Ohio Div. Geol. Surv. Kept, of Investigation.

FIELDS OF STUDY

Ma.jor Field; Geology Studies in Glacial Geology. Professor Richard P. Goldthwait

iii VITA (CONTINUED)

FIELDS OF STUDY (CONTINUED)

Studies in Geomorphology. Professor Sidney E. White Studies in Glaciology. Professors Colin Bull and Walter Schytt Studies in Agronomy. Professor Larry P. Wilding. Studies in Hydrogeology. Professor Wayne A. Pettyjohn TABLE OP CONTENTS

Page ACKNOWLEDGMENTS ...... ii VITA ...... iii LIST OP T A B L E S ...... viii LIST OP ILLUSTRATIONS...... x Plates Figures INTRODUCTION...... 1 Regional Setting Purpose of Investigation Methods of Investigation Economic Aspects of Glacial Deposits Chapter I. PREGLACIAL ENVIRONMENT...... 13 Bedrock Geology Preglacial Surface and Topography II. DRAINAGE H I S T O R Y ...... 18 Preglacial Drainage - Teays System Deep Stage Glacial Modifications of Drainage III. GLACIAL STRATIGRAPHY - METHODS, IDENTIFICATION, AND CORRELATION...... 35 Stratigraphy Rainsboro Till Boston Till Caesar Till Darby I Till TABLE OF CONTENTS (CONTINUED)

Pedology Till-Soil Associations Depth of Carbonate Leaching Loess Faleosols Granulometric Analyses Calcite-Dolomite Analyses Clay-Mineral Analyses Pebble Counts Heavy-Mineral Analyses Till-Fabric Analyses Radiocarbon Dating IV. DESCRIPTION OF GLACIAL DEPOSITS AND FEATURES ...... 126 Illinoian Margin and Glacial Boundary Illinoian Ground Moraine Late Wisconsin Moraines Knockemstiff Moraine Lattaville Moraine Reeseville Moraine Yellowbud Moraine Ice-Contact Deposits Outwash Illinoian Stage Higby Outwash Early Wisconsin Substage Late Wisconsin Substage Bainbridge Outwash Kingston Outwash Circleville Outwash Worthington Outwash Lacustrine Deposits Minford Silt Deposits in the D-ring Glacial Lake Humboldt Glacial Lake Massieville Glacial Lake Boumeville "The Prairie" Beech Flats Boulder Concentrations Directional Indicators V. GLACIAL HISTORY ...... 202

vi TABLE OF CONTENTS (CONTINUED)

Nebraskan Glacial Stage Kansan Glacial Stage Illinoian Glacial Stage Wisconsin Glacial Stage Early Wisconsin Substage Middle Wisconsin Substage Late Wisconsin Substage APPENDIX...... * ...... 229 Section A •••••••*•••••••••• 229 Sample Locations Pebble-Count Locations Section B •••••••••••#•••••• 240 Results of Laboratory Analyses Section C ••*»••*•••••• ...... 257 Sections With Significant Stratigraphy BIBLIOGRAPHY ...... 260

vii LIST OF TABLES

Page Generalized bedrock section of P.oss County ...... 15 Characteristics of till-soil associations in Ross County ...... 50 Soil characteristics associated with the till units in Ross County ...... 56 Confidence levels for significance of difference between mean depths of leaching in Ross County till units »••••••• 57 Comparison of rates of carbonate leaching in Ross and Highland counties ...... 61 Heavy mineral data of the Massieville and Seymourville loess sections • ...... 70 Particle-size distribution of Ross County till units •••••••••••••••• 85 Confidence levels for significance of differences between particle-size analyses for Ross County till units ...... * 85 Calcite and dolomite content in the less than 2mm fraction of Ross County tills • . 87 Confidence levels for significance of differences between mean carbonate content for Ross County till units • ...... 89 Clay mineralogy of Ross County till and loess units ...... 91 Pebble lithologies of Ross County tills » 99 Pebble lithologies of Ross County outwash units 103 v* m • * LIST OF TABLES (CONTINUED)

Table Page 14 Heavy-mineral data from Ross County tills 108 15 Radiocarbon dates from Caesar Till in the southern Scioto Sublobe • • • ...... 121 16 Radiocarbon dates from Ross County • . . 123 17 Pebble lithologies of Ross County ice- contact deposits ••••••••••*. 158 18 Characteristics of Ross County outwash deposits 168 19 Results of laboratory analyses from the Lickskillet lacustrine section • • • * * 188 20 General stratigraphic section in the Buckskin Creek valley • ••**••*** 192 21 Relative percent of clay minerals in Ross County stratigraphic units ••••••• 243 22 Particle-size distribution of bulk samples from Ross County 243 23 Chittick analyses data from Ross County stratigraphic units •••.«•••••• 248 24 Heavy-mineral data of Ross County stratigraphic units ••••••••••• 231 23 Pebble lithologies of Ross County stratigraphic units 233

ix LIST OF ILLUSTRATIONS

Plates

Plate Page I Map of the Glacial Geology of Ross County, Ohio ...... • . pocket II Reconstructed profiles of Ross County outwash terraces •»•••• ...... pocket

Figures

Figure 1 Map of Ohio showing the location of Ross County...... 2 2 Map of the bedrock geology of Ross County 14 3 Generalized bedrock contour nap of Ross County •••••••••• ...... ••• 16 4 Map of the D-ring showing drainage trends 19 5 Photograph of the northern portion of the D-ring...... 21 6 Map of Teays System drainage in Ross County region...... 23 7 Map of Deep Stage drainage in Ross County region ...... 23 8 Map of drainage trends in the Slate Mills- Alum Cliffs area ...... 28 9 Map of Beech Flats area showing Illinoian position during maximum advance ...... 31

x LIST OF ILLUSTRATIONS (CONTINUED)

Figure Page 10 Map of Beech Flats area showing Illinoian ice margin at retreatal position .... 31 11 Laboratory analyses of tills at Anderson Run till section •••••••»•••• 41 12 Granulometric analyses of Dry Run section 43 13 Chittick analyses of Dry Run section • • 44 14 Heavy-mineral analyses of Dry Run section 44 13 Clay mineralogy of the Dry Run section • 43 16 Map of southwestern Ohio showing end moraines and till-soil associations • * • 47 17 Map of distribution of till-soil associations in Ross County...... 49 18 Map of Ross County showing area3 that are differentiated on mean loess thickness and depth of carbonate leaching . . . . * 34 19 Histograms showing distribution of depths of carbonate leaching versus frequency • 35 20 Graph showing curve which represents the changing rate of leaching with time . . * 59 21 Histograms showing the thickness of loess cover in relation to frequency...... 64 22 Laboratory analyses at the Massieville loess section ...... 69 23 Photograph of Higby Outwash at Seymour- ville loess section ...... 72 24 Photograph of excavation at Seymourville loess section . 73 23 Laboratory analyses of Seymourville loess section ...... 76 xi LIST OP ILLUSTRATIONS (CONTINUED)

Page Diagram of the Massieville section . . . 80 Three-component diagram showing relation­ ship of particle-size distribution of Ross County till units to other tills • 84 Bar graphs of calcite and dolomite content of Boss County till units , • • 88 Three-component diagram showing clay- mineral composition of Boss County tills 94 Three-component diagram contrasting clay- mineral compositions of tills • . • • • 95 Three-component diagram showing the mean pebble lithology for Ross County tills • 100 Three-component diagram showing the mean pebble lithology for Ross County outwash 105 Bar graphs of heavy mineralogy of Ross County till units ••••* ...... 110 Rose diagrams of till fabric of Darby I T i l l ...... 113 Rose diagrams of till fabric of Caesar T i l l ...... 114 Rose diagrams of till fabric of Boston and Rainsboro Tills ...... 115 Map of southern Scioto Sublobe showing end moraines and radiocarbon dates , » • 119 Photograph of Lattaville Moraine near Hallsville ...... ••••••• 142 Map showing relationship between the Yellowbud Moraine, Circleville Outwash, and later end moraines ...... 149 Photograph of western Paint Creek valley 155 xii LIST OF ILLUSTRATIONS (CONTINUED)

Figure Page 41 Photograph of western Paint Creek valley 156 42 Photograph of Higby Outwash levels near H i g b y ...... 171 43 Photograph of Kingston Outwash exposure 179 44 Diagram of Lickskillet section * * . • - 188 45 Distribution of boulders and orientation of striae in Ross County •.•••••* 198 46 Nap showing Illinoian ice margin during formation of Beech Flats plain • • * • • 208 47 Hap showing Illinoian ice margin during deposition of ice-contact features in the Paint Creek valley ....»**•• 209 48 Map showing Illinoian ice margin during formation of Glacial Lake Boumeville . 212 49 Correlation chart of glacial deposits of Erie Lobe 215 50 Hap showing maximum position of Late Wisconsin ice in Ross County .*•••• 217 51 Hap showing Late Wisconsin ice margin at Lattaville Moraine •*•*•••*•*• 220 52 Hap showing Late Wisconsin ice margin at Reeseville Moraine •••••••»*•• 223 53 Map showing Late Wisconsin ice margin during Circleville Outwash deposition • 225 54 Hap showing Late Wisconsin ice margin at Powell Moraine during Worthington Outwash deposition •••••*••••• 227

xiii INTRODUCTION

Regional Setting

Ross County is located in south-central Ohio (Figure 1)* Chillicothe, the county seat and larges- city, is approximately 45 miles south of Columbus ari 45 miles north of Portsmouth, Ross County is the second largest county in Ohio, having an area of 687 square miles which is divided into seventeen townships. Two distinct topographic provinces occur in Ross County. The northern one-third of the county consists of rolling plains and low hills of the glaciated portion of the Central Lowlands Province (Thombury, 19S5). Streams in the lowland area flow in broad shallow valleys. The central and southern portions of Ross County rise 200 to 300 feet above the lowlands and are include! in the Appalachian Plateau Province (Thombury, 19c5)* This region is hilly, with streams flowing in deep narrow valleys. The north-facing bedrock escarpment which separates these two physiographic provinces trends north­ east-southwest across central Ross County. The county is drained by the southward-flowing Scioto ■ ■ p m i c c c ;

f«Ve]y*»w» r‘"u T\ ■T"T-

¥ «T> *T*V«t : ..

t M n «»— —

Figure “I. Location of Ross County on a portion of the map of the glacial deposits of Ohio* Rivor and its tributaries. The principal tributaries draining western Ross County are Deer Creek, Paint Creek (with its tributary Worth Fork), Indian Creek, and Stoney Creek, Major streams draining eastern portions of Ro3s County are Kinnikinnick Creek, Walnut Creek, and Salt Creek. The maximum elevation in Ross County is 134-2 feet on Horseback Knob, 1.3 miles northeast of Sumaithill in Huntington Township. The lowest elevation, 359 feet, is along the Scioto River on the Ross County-Fike County boundary.

Purpose of Investigation

The glacial geology of Ross County has previously been studied on a reconnaissance basis, with detailed investigations limited to small areas of the county. These preliminary studies indicated a complex history of multiple glaciations that produced a variety of erosional and depositions! features. The distribution of glacial deposits in Ross County indicates pronounced topographic control on the glaciations and deglaciaticns. This bedrock control caused the well-definea, arcuate end noraine3 of the southwestern portion of the Scioto Sub­ lobe (Greene, Fayette, Clinton, and Highland counties) to coalesce into a mor&icic belt in Ross County near the margin of tha north-facing bedrock escarpment. Only by 4 detailed mapping and laboratory analyses could the various morainic elements, the associated till units, and the glacial history of the county be firmly established. Thus a thorough study of the glacial geology of Ross County promised to provide substantive information on the extent and nature of Illinoian and Wisconsin glaciations and deglaciations along the glacial boundary. Additional information on the sequence of glaciations, related drainage modifications, and erosional and depositional features would significantly add to the glacial history of south-central Ohio.

Previous Investigations

The earliest geological descriptions of Ross County were subsidiary reconnaissance investigations produced during land surveys or during travel by visiting scientists. From observations taken during a canal trip through southern Ohio, Hildreth (1854) published the first geological report on the Ross County region. He noted water quality, deposits of iron ore ("globular pyrites"), "high banks of gravel" (terraces) bordering the major valleys, and the general regional topography. The first report on the general geomorphology of south-central Ohio was made by Whittlesey (1853), the Topographer of the Geological Survey of Ohio. Ancillary to his preparation of base maps of the entire state, Whittlesey briefly 5 described the topographic situations he encsundered during surveying. He produced a cross-section of central Ohio and indicated that the surface material was "Tertiary with primitive bowlders". Orton (1874) published the first geological report on HoS3 County. His comprehensive publication included thorough discussions of the bedrock geology, economic geology, and glacial history of the entire county. In 18?4 Newberry published a lengthy description of the glacial materials, glacial theory, the glaciated portion of Ohio, and glacial morphology. In Ross County he noted the presence of a "forest bed" beneath 30 feet of glacial material, bowlders ("erratic blocks"), and loess which he considered to be "... the sediment precipitated from the waters of our great inland sea in its shallow and more quiet portions, to which icebergs, with their gravel and bowlders, had no access." In his classic report on the glacial deposits of the midwestern and eastern United States, T. C. Chamberlin (1883) observed and mapped the moraines of tha "Scioto Glacier". He noted relative ages, topographic relation­ ships, and composition of the moraine segments in Ross County. The morphology of the glacial limit was described in detail by G. F. Wright (1890) in his report on the glacial boundary from Pennsylvania to Illinois. Glacially induced drainage modifications in south- central Ohio were discussed by Powke (1895) and Tight (1395a, 1895b). William Morris Davis (1884) briefly described the Alum Cliffs diversion of Paint Creek in his summary of the major gorges and waterfalls in the United States. In his classic monograph, Leverett (1902) described the Scioto River drainage system and its modification by repeated continental glaciation. He described the structure, thickness, and composition of "Early" and "Late" Wisconsin drift and associated glacial features in Ross County. Campbell (1918) and Hyde (1921) reported on the bed­ rock geology and general glacial geology of that portion of Ross County surrounding the now defunct Camp Sherman. The general geomorphology of south-central Ohio and the influence of glaciations on the topography and drainage history was discussed by Stout and Lamb (1938). Goldthwait (19^7), Poster (1950), and Reynolds (1959) each studied a small portion of Ross County and determined age relationships between various glacial deposits and erosional features. Hubbard (1954-) discussed the sand and gravel terraces along the present day Scioto River valley. He described terrace morphology, composition of the terrace materials, and general correlation of terrace levels with episodes of continental glaciation. Kempton and Goldthwait (1959) described the outi/ash terraces of the Scioto and Paint Valleys and indicated time correlations with various end moraines of “he Scioto Sub- Iocs. The relationships between glacial deposits, buried valley systems, and ground water resources have been described in various publications (Stout et al., 194-3; Van Tuyl and Bernhagen, 194-7; Schmidt, 1954-, 1961-62; and Y/alker et al., 1965). Selected exposures of glacial deposits were described and discussed in a series of guidebooks for the Friends of the Pleistocene and Geological Society of America (Goldthwait, 1955, 1962; Goldthwait and Rosengreen, 1969). The glacial history of the Scioto Sublobe is summarized in Goldthwait et al. (1965) and Dreiraanis and Goldthwait

(1973). In addition to Orton (1874-), Campbell (1918), and Hyde (1921); various aspects of the bedrock geology of Ross County were thoroughly studied by Stauffer (1909), Melvin (1933), Stout (194-1), Carman (1947, 1955), and Hyde (1953).

Methods of Investigation

Field Procedures Field investigations were carried out during the summers of 1972 and 1973- Nearly every road- and stream- cut in Ro s 3 County was examined and sampled. Areas not readily accessible by car were visited on foot. The 8 field area was revisited in late fall and early spring to plot boulder occurrences and map moraine topography. Crops and foliage, present during summer and early fall, hinder a comprehensive survey of these features at those times. Numerous pits were dug in areas of limited natural exposures to examine and sample the glacial materials. Several hundred auger borings were made to delineate thickness of loess cover, depth of leaching, soil parent material, and the soil profile. Sample for laboratory analyses were obtained from pits, augerings, and natural exposures* Pebble counts were made at 70 exposures in till and sand and gravel. One hundred 1- to 3-inch pebbles were carefully collected at each exposure. In the lab each pebble was washed, broken, and identified as to rock type. This procedure allows consistent lithologic identificaton, especially in separating limestone from dolomite. Till fabric was determined at 30 locations by measurement of the preferred long-axis orientation of 30 to 30 pebbles greater than one inch long. Only pebbles with a length-to-breadth ratio of greater than 3.0 were measured. Drake (19&8) has shown that an A:B axial ratio *1.7 coupled with 50 to 100 azimuth measurements gives a high degree of reliability in till fabric determinations. Young (1969) indicated that in tills with a "strong" fabric, the variation in calculate! prsferrei orientation is minor even when the number of measured orientations ranges from 25 to 100. Mapping was done on 7*5-minute quadrangle naps (U. S. Geological Survey, 1960-61). Plate I (in pocket) is a composite of these quadrangles reduced to half-scale (1:4-5000). All or portions of the following ?.5-minute quadrangles were used in mapping: Andersonville, Bain- bridge, Boumeville, Chillicothe Bast, Chillicothe Vest, Circleville, Clarksburg, Frankfort, Good Hope, Green­ field, Hallsville, Kingston, Laurelville, Londonderry, Morgantown, New Holland, Rainsboro, Ratcliffburg, Richmondale, South Salem, Summithill, V/averly North, and Williamsport. The Soil Survey of Ross County, Ohio (Petro et al., 1967) was very useful in determining parent materials associated with specific soil groups.

Office and Laboratory Procedures Over 750 water-well logs and several bedrock test- hole records, which are on file with the Ohio Division of Water, were analyzed for information concerning the thickness and nature of the glacial deposits in Ross County. Particle-size distribution was determined for 92 till and loess samples. The total sand fraction was 10 removed by wet sieving and then after drying was separated into the various sand fractions by dry sieving. Hydro­ meter analysis was used to determine quantitatively the silt (0.062 - 0.002mm) and clay (<0.002mm) fractions according to procedures described in the American Society for Testing Materials (1964, p. 95-106). The calcite, dolomite, and total carbonate content of 92 till and loes3 samples were determined by Chittick gasometric analysis (Dreimanis, 1962). Samples were ground to 0,074mm prior to analysis to diminish the effect of grain size on the gasometric reaction and thus to promote internal analytical consistency. Air photos of Ross County were examined to aid in delineation of outwash terraces, channels, moraine topography, and ice-contact features. Clay mineralogy of 42 till, loess, and lacustrine samples was semiquantitatively determined by a modified Johns et al. (1954, p. 242-251) x-ray diffraction method developed by Dr. L. P. Wilding and L. R. Drees of the Department of Agronomy, The Ohio State University. Peak areas were measured with a planimeter and utilized in conjunction with relative diffraction intensity ratios of various clay minerals that were determined from analysis of laboratory standards. The criteria by which the various clay minerals were identified and the calculated percentages are given in the Appendix, Section B. 11 Economic Aspects of Glacial Deposits

Glacial outwash and ice-contact d9po3i*rs are the primary source of aggregate for construction in Ross County. Wisconsin ice-contact sand and gravel is extracted in extensive pit operations in Section 30 of Green Township, south of Kinnikinnick. Late Wisconsin (Worthington) outwash is extracted from large-scale sand and gravel operations in the Scioto River valley south of Chillicothe (Sections 31 &n& 32, Jefferson Township)* Locally, borrow pits in Wisconsin and Illinoian sand and gravel provide road fill and minor amounts of concrete aggregate* t In addition to serving as an aggregate source, the thick Wisconsin and Illinoian sand and gravel deposits form excellent aquifers which yield large quantities of high-quality groundwater* The excellent water-bearing properties of these aquifer materials provide a strong economic foundation on which the present rapid industrial expansion of the lower Scioto River valley is based* During the late 19th century high-quality laminated lacustrine clay was mined near Vigo (Section 23, Jefferson Township) for use in local brick and tile manufacture* Gold has been derived from the drift in south- central Ohio (Orton, 187*0• However, Orton pessimistic­ ally viewed the chances of anyone becoming wealthy from gold mining in Ohio by noting that the best results thus far known to have been obtained in gold-mining in Ohio are reported from Warren County where in one day gold to the value of six dollars was obtained by an outlay of ten dollars; a half-dozen days' work being also thrown in." Chapter I

PREGLACIAL ENVIRONMENT

Bedrock Geology

Exposed bedrock in Roas County ranges from Silurian to Pennsylvanian in age (Figure 2, Table 1). Exposures predominate in the southern, unglaciated portion of the county, although some stream cuts, road cuts, and uplands in the glaciated area display the bedrock stratigraphy* The upper Paleozoic strata generally dip east-southeast at approximately 30 feet per mile, forming part of the east limb of the Cincinnati Arch* The oldest exposed bedrock is Silurian limestone and dolomite in the Paint Creek and Buckskin Creek valleys of western Ross County* Most of the county west of the Scioto River valley is underlain by Devonian shales and Mississippian shales and sandstones. The Mississippian Berea sandstone caps the outliers and uplands in the central and southern portions of western Ross County. East of the Scioto Valley the Mississippian bedrock is a series of shales and thin sandstones which form the uplands* In small areas of northeastern and southeastern 13 Pennsylvanian [j>] Devonian Jackvon I V inton Co I H ocking mi le t fXJ Mississippian pcj Silurian t-i i—i ^

Figure 2* Bedrock geology of Ross County. Ohio (after Bownocker, 1920; Orton, 194?; and Schmidt, 195*3. 15

TABI-E 1 GENERALIZED BEDROCK SECTION OF F.C33 COUNTY,

A Group/ Approximat w System Formation Thickness Character P enn s ylvanian Pottsville Gp 40' coarse, pebbly sandstones Cuyahoga sh 670* alternating ss, Sunbury sh sh, siltstones Mississippian Berea ss 25 - 40* thin-bedded ss and siltstone Bedford sh 90* clay shale Ohio sh 270-300* black shale, Devonian carbonaceous Olentangy sh brown shale Greenfield dol 100* fine-grained Silurian dolomite Pebbles dol 20 - 9C1 porous dolomite

Ross County, the Pennsylvanian Sharon conglomerate caps the higher uplands.

Preglacial Surface and Topography

The preglacial topography of Ross County (Figure 3)t as reconstructed from well logs and exposures, displays two distinctive topographic situations. The variations in bedrock topography closely conform to variations in the distribution of bedrock lithologies (Figure 2). Prior to glacial modification the southeastern two- thirds of the county was a prominent, deeply-incised Highland Co iue? Generalizedbedrock contour mapof Figure Ross?•County, Ohio ia Co Pika (contoursin hundredsoffeet). Ejckjwav Contourinterval 200 feet K * I

17 upland forming a portion of the western margin of the Appalachian Plateau, Deep entrenchment in che Mississippian bedrock by the preglacial drainage created a maximum local relief of approximately 800 feet in southwestern Ross County. The nearly accordant elevation of this plateau (1000 to 1100 feet) was interrupted by small hills which rose 200 to 300 feet above the general upland surface. The northwestern one-third of Ross County had a much more subdued bedrock surface in contrast to the bordering dissected upland. This bedrock surface was developed primarily in Devonian shale and lay 300 to 400 feet below the neighboring upland. Prominent outliers of Mississippian strata of the Appalachian Plateau dominated the southeastern margin of this region. Additionally, small rounded hills and minor tributary stream valleys interrupted the general planar nature of the bedrock topography. Comparison of Figure 3 with the distribution of constructional glacial deposits (Plate I) clearly illustrates the pronounced effect of the bedrock topography on the multiple glaciations that affected Ross County. Chapter II

DRAINAGE HISTORY

Preslacial Drainage - Teays System

The earliest drainage system that can be definitively established in south-central Ohio is the Teays System, named for the Teays Valley in Cabell and Putnam Counties in West Virginia (Tight, 1903). The term Teays refers not only to the primary river of the Teays System but also to the general degradational work of all streams during the erosional period prior to continental glaciation. The Teays River, a mature stream flowing in a broad bedrock valley, had its headwaters in the Piedmont Province of Virginia and North Carolina (Stout et al., 19^3) or in the folded Appalachians (Kanos, 1961). The northerly-flowing Teay3 River entered Ohio near Wheelersburg on the present Ohio River. The river valley is easily traced northward past Minford, Stockdale, and Beaver to Waverly. Prom Waverly to Richraondale in south­ eastern Ross County (Pigure 4-) the Teays River generally followed the valley of the present Scioto River. Prom Richmondale the river flowed northeasterly to near Vigo 18 londoodtfR

thooiay

LickskjJU t

-> Present drainage — Teays System drainage Beep Stage drainage Figure 4. Map of the D-ring area of southeastern Ross County showing drainage trends. 20

and then swung abruptly west-northwest through London­ derry, rejoining the present Scioto Valley near Schooley. Subsequent cutoff of this major meander by the Newark River (Deep Stage) and the Scioto River created a distinctive "D" pattern on topographic maps. This portion of the Teays Valley in southeastern Ross County will be subsequently referred to as the "D-ring" (Figure 4). No streams presently occupy the northern portion of the D-ring (Figure 5)* The eastern and southern areas of the valley are drained by Salt Creek and its tributaries. Three major tributaries joined the Teays River in Ross County. Two parallel, westward-flowing streams joined the Teays River near Londonderry and Vigo in the eastern portion of the D-ring. The northernmost of the two streams, Eagle Mills Creek, eroded much of the major bedrock valley now drained by Salt Creek. The largest Teays System tributary in the county intersected the Teays River near the present location of Mound City Group National Monument. Bainbridge Creek headv/atered in High­ land County and drained northeasterly through the present Paint Creek valley. The angles at which the major tributaries of the present Paint Creek (e.g. Upper and Lower Tv/in Creeks) intersect the main stream were probably established during Teays time. The portion of the Bainbridge Creek valley, extending from two miles south­ west of Slats Mills past Pleasant Valley to Mound City 21

Figure 5- View of the northern portion of the D-ring looking northwest from the Vigo Fire Tower (SEJi Section 22, Liberty Township)* The Lickskillet lacustrine section (L) is located along the railroad in the left center of the picture. Illinoian Higby Outwash (Io) forms the extensive plain throughout the center of the picture. 22

Group National Monument, has been partially filled with

drift* Several factors indicate that the Teays River had reached erosional maturity before the onset of continental glaciation. The northerly gradient of the valley floor averages one foot per mile from Wheelersburg to Circle- ville (Walker et al., 1965)# The Teays Valley was broad and shallow, with a general northward increase in valley width and a decrease in depth. Walker et al. (1965) indicated that valley width varied from 1.25 to 8.00 miles and that valley depth ranged from 52 to 250 feet. Cross- sections in the Scioto Valley (Walker et al*, 1965) indicate the valley floor was not flat but had bedrock highs and deep narrow channels modifying the floor of the bedrock valley* In addition, the uplands bordering the Teays Valley were well-rounded with few original inter- stream tracts extant. The course of the Teays River north of the glacial boundary has been established by well logs, bedrock test holes, and geophysical methods. Figure 6 is a reconstruction of the Teays System drainage in the Ross County region modified after Stout et al. (194-3). Recent studies by Walker et al. (1965), Teller (1970), and Richard (1974-) have locally modified th9 positioning of the Teays System drainage as proposed by Stout et al. (194-3). Wayne (1956) and Horberg (1950) traced the buried 2 ?

Figure 6* Teays Stage drainage in the Ross County region.

---1

v-— ■

Figure 7. Deep Stage drainage in the Ross County region. 2 4

course of the Teays Valley across northern Indiana and central Illinois and indicated the drainage reached the Gulf of Mexico by way of the ancestral Mississippi River.

Deep Stage

Deep Stage drainage was initiated when Kansan or pre-Kansan ice dammed the northwesterly drainage of the Teays System (Stout et al., 1943)* Teller (1970) presented stratigraphic evidence from southwestern Ohio and southeastern Indiana that indicates the Teays System drainage was originally modified by the Nebraskan ice sheet. This evidence includes: 1) Kansan and other pre- Illinoian drift in deeply-incised bedrock valleys up to 150 feet below Teays System valley floors, and 2) presence of Nebraskan till (St. Maurice Till) in southeastern Indiana, south of the Teays River valley. The Nebraskan age of the St. Maurice Till has been questioned by most glacial geologists. On the basis of regional till stratigraphy and relationships between drift and bedrock valleys, Teller (1970) showed the existence of both Aftonian and Yarmouthian periods of Deep Stage drainage in the Cincinnati region. The ice dam created an extensive finger lake in the Teays Valley and its tributaries. In this lake extensive deposits of clay-rich, laminated sediment accumulated during the early stages of damming. These deposits are 25 known as the Hinford silts (Stout and Schaaf, 1931). The ice obstruction caused the streams to seek new outlets south of the glacial boundary. Eventual breaching of cols and divides by ponded waters caused new drainage lines with a southerly gradient to be established with con­ sequent abandonment of many well-formed Teays drainage ways. Figure 7 is a reconstruction of the Deep Stage drainage in the Ross County region after Stout et al. (W). The Newark River, an important tributary of the master Cincinnati River, drained much of central and south-central Ohio during the Deep Stage. The course of the southerly-flowing river was essentially coincident with the present valley of the Scioto River. Three important tributaries Joined the Newark River in Ross County. Adelphi Creek drained portions of Hocking, Pickaway, and northeastern Ross counties before inter­ secting the Newark River near Kinnikinnick. Bournevilie Creek and a major unnamed tributary from the northwest, which joined Bourneville Creek near the present location of Slate Hills, drained portions of Madison, Fayette, Clinton, Highland, Pike, and western Ross counties. Bourneville Creek retained the northeasterly drainage trend established by Bainbridge Creek of the Teays System. Bourneville Creek and its major tributary are the predecessors of the present Paint Creek and North Fork 2 6 respectively. Originally the creek was probably a barbed tributary of the Nev/ark River, intersecting the river near the present Hound City Group National Monument. Eventually in Wisconsin time the present valley between Slate Mills and Chillicothe was established and the barbed intersection abandoned. Vigo Creek drained parts of Hocking, Vinton, Jackson, and southeastern Ross counties before emptying into the Newark River near Richmondale. To account for the depth of entrenchment of the Deep Stage valleys, regional uplift must have occurred soon after establishment of the new drainage lines. During the Deep Stage cycle most channels were deepened below levels established during the Teays System. In Ross County, floors of the Deep Stage valleys lie 100 to 200 feet lower than adjacent Teays System valleys. At Chillicothe the floor of the Newark River lies at an elevation of approximately 500 feet while the Teays River valley floor elevation is about 610 feet (Stout et al., W). The Deep Stage was an interglacial erosion cycle, initiated by Kansan or pre-Kansan glaciation, and terminated by the advance of the Illinoian ice sheet* 27

Glacial Drainage Modifications

Alum Cliffs Diversion - Glacial Lake Bourneville

Glacial modifications of drainage in the Slate Mills area was first noted by Orton (1874) ana later described by Davis (1884) and Fowke (1895)* During advance but primarily during retreat when greater quantities of meltwater were available, Illinoian ice blocked the northeasterly drainage trend established by Deep Stage Bourneville Creek near Ebenezer Church creating Glacial Lake Bourneville (Foster, 1950) (Figure 8). To account for the thickness and areal extent of lacustrine deposits of Glacial Lake Bourneville, Illinoian ice must have also blocked the western margin of the present Faint Creek valley near Bainbridge* There are no channels or other evidence of water escapement between the ice mass and the valley sides at the dam site* This indicates blockage by a thick ice mass near Ebenezer Church* Eventually the impounded water established an out­ let through a col south-southeast of the ice dam coincident with the present location of Alum Cliffs gorge* The col originally served as the head of a southerly- flowing tributary of Ralston Run. Extension of topo­ graphic profiles from the sides of the present 250-foot- deep Alum Cliffs gorge indicates that the original col ---v “ p=------7 S \ \ I28

^ \ \ V - ' i Growl ^ / v Natl IMon. ^ ^ X M ooaint ^v«ll»y ^ ^ % r O

Andorton

I

Church

°e

Present drainage — * Teays System drainage Deep Stage drainage Figure 8. Map of the drainage trends in the Slate Mills- Alum Cliffs area. 29 elevation was about 800 feet. Original elevation of the col must have been less than 890 feet, because a col at that elevation 0.7 mile west-southwest of Bishop Hill Church, Huntington Township, was not used. Lake level was lowered as the lake outlet through the gorge was progressively downcut through the highly jointed and fractured Ohio shale. Following retreat of the Illinoian ice, with con­ sequent termination of Glacial Lake Bourneville, drainage lines established by Deep Stage Bourneville Creek through the present North Fork valley may have been reoccupied and the newly-formed drainage way through Alum Cliffs gorge abandoned. This is suggested by the present under­ fit nature of North Fork and the youthful character of the Alum Cliffs gorge (Foster, 1950)* Deposition of thick drift in the vicinity of Slate Hills by the Wisconsin ice sheet permanently altered the drainage of ancestral Faint Creek from Deep Stage trends to the present Faint Creek drainage through Alum Cliffs gorge. No erosional or depositional features associated with a possible Wisconsin stage of Glacial Lake Bourne­ ville have been identified. However, a shallow lake may have reformed for a short interval during early Late Wisconsin, completing the formation of Alum Cliffs gorge without modification of older glacial deposits in the Paint Creek valley. 5 0

Beech Flats Reversal

During Deep Stage the Beech Flats area of Highland, Pike, and southwestern Ross counties (Figure 9) was drained by a major north-flowing tributary of Bourneville Creek which joined the main stream just west of Bain­ bridge. Beech Flats is now drained by the southwesterly- flowing Baker Fork and its tributaries. Thi3 drainage reversal is clearly defined by barbed tributaries to Baker Fork in Pike County and water-well records which indicate a northeasterly gradient and widening of the buried Deep Stage bedrock valley. During maximum advance of the Illinoian ice sheet, ice invaded Beech Flats from the north damming the north- flowing tributary, creating a proglacial lake (Figure 9). The ponded waters established an outlet over a low col at the southwestern margin of the lake between Fort Hill and Reeds Hill in Highland County (Rosengreen, 1970)* Drainage from the lake generally followed the present course of Baker Fork to Ohio Brush Creek in southern Highland County. The stand of the Illinoian ice sheet at maximum position was probably of short duration since ice-contact deposits along the outer boundary are limited in size and extent. The Illinoian ice then retreated to a position along the northern margin of Beech Flats (Figure 10). North- Proglacial lake Highland 'A County Heltwater drainage t ~ ~~ Illinoian ice margin

Pika County

Adam* County

Figure 9- Hap of the Beech Flats area showing the position of the Illinoian ice margin during maximum advance (after Rosengreen, 1970).

R ob* C ounty Proglacial lake

Heltwater drainage

Illinoian ice margin

Pika County Illinoian glacial limit

Figure 10. Map of the Beech Flats area showing the position of the Illinoian ice margin during a retreatal stand (after Rosengreen, 1970). 32 ward drainage of the Bourneville Creek tributary remained blocked and a more extensive proglacial lake developed. The abundance of Illinoian outwash and ice-contact drift in southwestern Ross County indicates the ice mass remained along the northern edge of Beech Plats for a substantial time. Meltwater deposited large quantities of sediment into the lake eventually aggrading the area to an elevation of approximately 960 feet, creating an extensive flat plain. Water-well records indicate that up to 265 feet of sediment overlies the Deep Stage bedrock valley floor. Drift deposition in the Beech Plats area was sufficient to permanently reverse the local drainage following subsequent retreat of the Illinoian ice sheet.

Northward Shifting of Drainage Divides

As the several continental ice sheets advanced up the north-facing bedrock slope of the Appalachian low plateau, small proglacial lakes were impounded between the ice mass and the hillside. With advance of the ice sheet, lake elevations in some locations became high enough to establish outlets through cols and divides that headed southerly-draining streams. Downcutting at these outlets coupled with drift deposition along the northern front of the bedrock escarpment caused capture of portions of basins formerly drained by north-flowing streams. 55

Freglacial, southerly-draining ancestral Walnut

Creel: healed in a col near Rod: Hill in Section 26 of Colerain Township in northeastern Ross County. As Illinoian and later Wisconsin ice advanced up the Mississippian bedrock slope, a lake was impounded in the area of what is now known locally as Maple Swamp

(centered in Section 22, Colerain Township). Subsequent downcutting at the lake outlet coupled with deposition of thick drift along the escarpment margin resulted in the northward shifting of the Walnut Creek drainage divide* Breaching of the col must have been initially completed in Illinoian time because outwash associated with the initial, maximum Late Wisconsin advance is found in the upper portions of the Walnut Creek valley in east- central Ross County near Tuscon (Section 9* Harrison Township). A similar northward shift of a drainage divide greatly increased the area of the Buckskin Creek drainage basin in western Ross County. This drainage modification is associated with the formation of Glacial Lake Humboldt which is discussed below in conjunction with other lacustrine features. The drainage divide of the southeasterly-flowing tributary to Deep Stage Bourneville Creek, which inter­ sected th3 main creek at Slate Mills (Figure 7)» was dramatically shifted northwestward. The divide shift 3

followed. impoundment of a lake by Illinoian ice and sub­ sequent breaching of a col near Musselman in the west- central portion of the county (extreme southeastern Concord Township). Thick Wisconsin morainic deposits along the north slope of the Appalachian front near Dolphins Ridge and Ryan Hill (west-central Ross County) caused the drainage divide of southeasterly-draining Lower Twin Creek to be shifted more than a mile north of the former divide position between the ridge and hill. Similar minor drainage changes occurred at many localities along the Mississippian escarpment in conjunction with deposition of the Lattaville Moraine. Chapter III

GLACIAL STRATIGRAPHY - METHODS, IDENTIFICATION, AND CORRELATION

Interpretation of the glacial history of an area nust he based primarily on identification and correlation of glacial-stratigraphic units. Several independent empirical methods must he used in conjunction for reliable identification and correlation of the glacial strata. Empirical methods employed during this study included: stratigraphy, pedology, granulometric analysis, calcite- dolomite analysis, clay mineralogy, pebble lithology, heavy-mineral analysis, till-fabric analysis, and radio- metric dating. Results and conclusions determined by these methods are presented in the following sections; the detailed data are listed in the Appendix, Section B.

Stratigraphy

The initial stage in stratigraphic correlation is the establishment of local stratigraphic sections wherever exposures exist. Correlation between identified sections is based on quantitative and semiquantitative methodology. 35 36

Although relative age relationships within individual exposures are usually easily determinable, correlation between established sections is fraught with problems. These difficulties include physical similarities between nonsynchronous glacial strata, lack of datable materials to define absolute age relationships, wide variation in the quality and reliability of water-well logs submitted to the Ohio Division of Water by local well drillers, and scarcity of exposures in key areas. Glacial stratigraphy is characterized by repetitive sequences of varied drift units (e.g. till-outwash-till). In the absence of a distinctive marker bed or an abundance of datable material, correlation based on stratigraphic sequence from section to section or between water-well logs is at best tentative. No multiple-till exposures were identified in Hoss County. The glacial stratigraphy is established on the basis of areal extent of till units, pedologic features, and physical and compositional characteristics of the till units.

Rainsboro Till

Fifteen feet of oxidized, brown (7.5YR5/2) Illinoian till is exposed on the south side of a Baltimore and Ohio railroad cut on the boundary between Sections 16 and 17, 1.8 miles east of Schooley, in Liberty Township. The 57

pebble-poor till is calcareous below an 85-inch thick leached zone. This till is areally, stratigraphically, and texturally correlative with the Rainsboro Till of Highland County (Rosengreen, 197*0• Well-developed alickensides is exposed on the till face near a recent slump at the eastern end of the exposure. The till overlies flaggy Mississippian shale. Directly overlying the finely-striated bedrock surface is a six- to nine-inch thick brownish-yellow (10YR6/6) sand unit which contains many well-rounded quartz pebbles. This exposure is within one mile of the Illinoian glacial boundary in the area where the erosive power of the ice sheet would have been greatly diminished. Thus the sand unit may represent a remnant of a Yarmouthian weathering profile that was not incorporated into the till of the advancing Illinoian ice sheet. Exposures of Rainsboro Till in Ross County are limited to areas outside of the Wisconsin glacial boundary in the southern portion of the county. Generally a thick loess (mean thickness 29.8 inches) overlies the till. Rainsboro Till forms thin, patchy ground moraine or till plain on the bedrock uplands of the Appalachian low plateau. 53 Boston Till

Till associated with the maximum advance of the Late Wisconsin ice sheet is exposed only in tributary valleys to the Paint Creek valley in southwestern Ross County, This till is areally, texturally, and compositionally distinct from all other till units in the county. The type locality of this distinctive till is in the main scarp of a slumped area, one mile south-southwest of Storms in southwestern Twin Township* The exposure is located 0,1 mile southeast of Alexander Hollow Road, 200 feet west of Sulfur Lick Run, Approximately seven feet of friable, pebble-rich, yellowish-brown (10TR5/4) till is exposed in the main scarp of the rotational slump area* The till is leached of carbonates to a depth of 32 inches and is overlain by about 15 inches of loess. Just north of the type section, Boston Till clearly overlies rhythmically-laminated lacustrine deposits of Illinoian Glacial Lake Boumeville. Areal extent in conjunction with compositional and textural distinctiveness suggests that this earliest Late Wisconsin till in Ross County is probably correlative with the Boston Till (Rosengreen, 1970) of Highland County, Lack of datable material and exposures north of the Paint Creek valley hinders a definitive correlation. Boston Till forms the constructional topography of the patchy Knockemstiff Moraine in the tributary valleys 39

south of the Paint Greek valley (Plate I), Isolated patches of Boston Till are also exposed in tributary valleys on the north side of the Paint Valley north of Bourneville.

Caesar Till

Three till units separated by two thin outwash zones are exposed in the west bank of Anderson Hun, one mile northwest of Anderson, southwestern Union Township. This exposure is 0.3 mile upstream from Stop 13A ("Anderson buried forest") of the Fifth Biennial Pleistocene Field Conference (Goldthwait, 1935)* This field conference stop displayed eighteen feet of pebbly blue-gray (10YR5/1) to yellow-brown (10YR5/4) Late Wisconsin till with a "buried forest" near its base. The till overlay a sandy gravel unit but it is now badly slumped and overgrown. Spruce logs from the lowest till at the Anderson Run cut have been dated at 17,990 * 400 years B.P. (W-331) and 16,590 ± 570 years B.P. (CWR-190). This lowest till is unoxidized, calcareous, dark gray (7.5YR4/1) and very compact. This till is areally, texturally, and stratigraphically correlative with the Caesar Till of the southwestern portion of the Scioto Sublobe (Rosengreen, 1970). Caesar Till is the surface till south of the Darby I till limit (Reeseville Moraine equivalent position) to 4 0 the distal margin of the Lattaville Moraine (Plate I). Minor variations in laboratory analyses of the three till units (Figure 11) coupled with the thinness of the intercalated outwash units suggests that all three till units are associated with deposition during minor oscillations of the same Late Wisconsin ice sheet. All variations in particle-size distribution, calcite-dolomite content, clay mineralogy, and heavy mineralogy between the three tills at the Anderson Run cut can be explained by progressive pedologic development.

Darby I Till

Twenty-two feet of homogeneous, pebbly, dark gray (10YR4/1) till is exposed in the north bank of Dry Run, 1.1 miles southeast of Dry Run Chapel in west-central Union Township. The well-jointed till is leached of carbonates to a depth of 24 inches and lacks a loess cover of measureable thickness. Secondary clay accumulations along the joint margins are common. A similar exposure is found in the village of Austin in northwestern Ro3s County (north-central Concord Township). This till is texturally and areally correlative with the Darby I Till (Goldthwait et al., 1965; Rosengreen, 1970) of the south­ western portion of the Scioto Sublobe. Darby I Till is the surface unit north of the Reese- ville Moraine distal boundary position throughout north- Chittick I Clay Heavy-Mineral Anderson Granulometric Analysis Analysis Mineralogy Analysis 4-> OJ Run % % % % % % % * Pt 0 • m O t-c 1 s n w <• Section i T«*al « f t m m d Slit Cl«y f ^ c iu 1 * QpHiii i Mixed zone i 1" A A ’ A s r k n n i k U ’A 1 Jn> 4k ’ A Oxidized friable yellow- brown till

Gray till • 1 • • • • • • • • t * • • 1 1 r t sample 82 1 1 1 t « % 1 ■ip*. I t m n n ) 1 W/////J h t i n t , 7//i n i t 1 1 1 - 2 2 Outwash (HH 7 /• A u i i u 7 / /// * 1 * * 1 1 1 t 1 1 1 t 1 t 1 1 1 / 1 1 1 Light gray 1 \ t \ 1 t 1 t \ 1 * / 1 \ 1 \ 1 \ 1 1 1 t 1 1 / till • • • m • • r 1 f * f f ■ * t 1 1 sample 81 9 1 \ 1 1 • 1 / t 1 i 1 1 1 1 (wood) t 1 t 1 1 1 1 t 1 * t t 1 ■ 1 t ■ / i 1 1 Fine sand

and gravel 3 , ■ I ///* 1 //////// ilium,Ill/Ill,t llll,%1 'iiiih/////7 //// /// 71 1 9 ■ t 1 Dark gray 1 » 9 1 1 1 1 1 1 t 1 till t » 1 1 ilt 1 4 • ; • • sample 80 • • • • 16,590*570 (CWR-190) Figure 11. Laboratory analyses of the tills at the Anderson Run till section.r 4:2

central and northwestern Ross County. The recessional Yellov/bud Moraine is capped with Late Wisconsin Darby I Till. Totten (1969) suggested that the constructional topography associated with the end moraines and till plains of central Ohio was formed during the Altonian. The present subdued nature of this topography resulted from deposition of thin Woodfordian tills on the Altonian moraines and till plains. If such postulated thin tills occur as far south as Ross County they should be detect­ able in thick till sections such as the Dry Run stream cut in west-central Union Township. The Dry Run section was sampled at two-foot intervals and the samples were subsequently analyzed for particle-size distribution (Figure 12), calcite-dolomite content (Figure 13), heavy mineralogy (Figure 14), and clay mineralogy (Figure 15)- Except for an anomalous silt-clay content at a depth of 20 feet, all data variability is expectable as the result of normal pedologic development. If thin Woodfordian tills are present in Ross County they are texturally and compositionally indistinguishable from all other till units of the southern Scioto Sublobe. All sections with a significant stratigraphy are described in the Appendix, Section C. 43

% % % % • a S a n d S i l t C l a y » 2 m m E 1" 1 1 I f 22 » 4^ J 5% S ^ 2\ ‘ r — • 1 2- i i i ! 4 t R U i _ _ iftKttno. _ .

4 - • • • ■47 i i ' m i -4ft ft 4 N. / t / / / / ft- • * • •49 V i \ i \ 10- V. • 4 0 / i t i t 12- • • 41 / / i \ / t \ 1* • 4 2 r t / i i * • * 43 \ t 1 \ \ ft- 1 X • 44 1 1 20 . - - 4 • -59 f

22 « r 4 ft

Figure 12. Results of granulometric analyses of the Dry Run till section (1.1 miles southeast of Dry Run Chapel, west-central Union Township). 4 4 % % % Total Calcitt Dolomita Caicita 1 1 Carbonata Oolomita 0 - s a " ’ *

4 - • * 4 7 \ V s

* ■*' «■ • X •48 %\ 8- •49 1 'T \ 10 ' w 1 i •SO 1 2 - * K • 5 1 \ W- '*. > • 5 2 /• \$ / 9 IB' • • 5 3 1 / Ifr • 5 4 \ 4 \ \ \ • » \ • 5 5 * \ 2 2 • 5 8

Figure 13. Results of Chittick gasometric analyses of the Dry Run till section.

% % Mite. Opaquaa Oarnalt Hornblanda Kooopaaua a ?S L < * I I E os «

2 s L J»v< Ana. -----

4. *47 6- / B- • • 4 3 i t 10- i ■ t i t i i 12' •i 51 \ 14 i 15 53

i : 55 Figure 14. Results of heavy-mineral analyses of the Dry Run till section. *5

% % % % % % % l u l t r i l n i , • £ „ I l l i U V h m Im Hi* CM m K w I M t* 0 « « r ti 10- 14* 14- 17 * a £ ft n a os n ' » r n i (0

- QbiQS 2 * 1 d o p l t u .

4 - • • • • ■47 T f ' i 1 i / 1 * * 6 - ; • V • • • ■46 i i i / i V > / • i \ \ ►40 «• • • • • • • i r i 1 \ i 1 \ 1 4 ; • • ►50 10- \ f i / i i > t • • 4 • • 41 t 2* T i / 1 * % i i / 1 / \ • • • • 4 2 14- F i t i / n 1 \ \ f / ; 4 3 16- • • • • /4 i \ t 1 \ / / / \ 4 4 * • V • • •

,*v \ / •' "• V » i \ \ t \ t i 20- • • • • • 4 5 t \ / 1 i i i ■. i 22- • \ 4 4 • • 4 6

Figure 15- Results of clay mineralogy analyses of the Dry Run till section. 46

P e d o l o g y

Till-Soil Associations

Five Late Wisconsin till-soil associations are recognized in western Ohio (Forsyth, 1955). These associations are distinguished on the basis of variations in: 1) loess cover, 2) amount of clay in the B-horizon, 3) amount of clay in the C-horizon, and 4) depth of the soil profile. These soil groups occur in irregular bnads generally parallel to enc! moraines. Boundaries between most till-soil associations are caused by major variations in parent material usually associated with a significant readvance of the ice sheet. However, the boundary separating the Miami 6A and 60 soils appears to have been established because of the occurrence of an end moraine representing a major glacial readvance. This boundary separating the two Miami soil groups is placed at the distal margin of the Farmersville (Miami Sublobe) - Reeseville (Scioto Sublobe) Moraine correlative position (Quinn, 1972) (Figure 16). Recent soil mapping (Wilding et al., 1965* Petro et al., 19&7) indicates little significant difference between Miami 6A and Miami 60 soil associations in western Ohio. Laboratory analyses during the above studies and this study confirm that there is no significant difference in the parent material composition of 6A soils (Darby I Till) 47

1__ 62 t T T i ' l I f T

[Ch,

62 - St. Clair soils 67s- "shallow" Russell soils 6B - Morley soils 67 - "deep" Russell soils 6A - Miami 6A soils 75 - Cincinnati soils 60 - Miami 60 soils

Figure 16. Map of southwestern Ohio showing relationship between end moraines and distribution of major till-soil associations. Soils are labelled by field mapping symbols. Diagram modified after Forsyth (1965). 48

and 60 soils (Caesar Till). Under current soils term­ inology, areas formerly mapped as Miami 6A and 60 are included under the terra "Miamian" (for well-drained sites), Although the Miami 6A and 60 areas are not easily differentiated throughout western Ohio, the two soil groups are readily distinguishable in Boss County on the basis of variability in loess thickness. Similar distinctive loess variability exists in Highland County (Rosengreen, 1970). For simplicity of explanation the defunct Miami 6A and 60 terminology will be retained for this study. Four Late Wisconsin till-soil associations occur in Ross County (Figure 17)• The characteristics of these associations are listed in Table 2. Characteristics of individual soils which are diagnostic for differentiation are underlined. Illinoian "Cincinnati" soil is included in the table for comparison. The "deep" Russell soil was used to support "early" Wisconsin glaciation of the southern Scioto Sublobe. These soils have a degree of soil development inter­ mediate between knov/n Illinoian and Late Wisconsin profiles. The "deep" Russell soils are areally limited to a narrow belt along the Wisconsin glacial boundary in Ross and Highland Counties. Rosengreen (1970) has shown that the "deep" Russell soil profile is attributable ako Cl itn o I Hocking I Col Vinton Col Jackson 49

Pika Co Miami 60 Miami

Illinoian Cindinnati 7 Cindinnati Wisconsin j t j* -n -n j* Glacial boundaries Glacial Miami / / 6A Miami *> p Pickaw a* Co Russell Russell , V 673/67 \ \ 673/67 \ X N » . 75 in Ross County. Ross in Miami ^ Miami 60 Cincinnat

puimOiH 17- Figure associations till-soil of Distribution TABLE 2 SIGNIFICANT CHARACTERISTICS OF MAJOR TILL-SOIL ASSOCIATIONS IN ROSS COUNTY

Soil Soil Characteristics

Thickness Depth of % Clay % Clay Ratio of silt leaching in B- in C- ;% Clay B Boundary Boundary cap (inches; horizon horizon % Clay C to south to north

Miami 6A generally 6-4-5 33-40 11-27 1.8 Reeseville Powell absent Moraine Moraine Miami 60 18" 11-65 33-40 10-27 1.8 Lattaville Reeseville Moraine Moraine " shallow'* 6?s 18" 35-60 35-40 15-27 1.7 Mt. Olive Reeseville Russell Moraine Moraine "deep1* Russell 67 18" 60-85 33-40 15-27 1.7 Knockem- Lattaville stiff Moraine Moraine

Cincinnati 75 10-80" 66-145 33-40 15-27 1.7 Illinoian Wisconsin boundary boundary Diagnostic characteristics are underlined* Modified from Forsyth (1965) p. 223.

s to a unique parent material and that the soil occurs in drift of early Late Wisconsin (V/oodfordian) age. The Kussell soils are characterized by a greater than 18-inch thick loess cap. "Deep" Russell soil is differentiated from "shallow" Russell soil in a greater depth of carbonate leaching (Table 2). Russell soils hare a colored, more weathered appearance than typical Late Wisconsin "Miamian” profiles. The "deep" Russell soils in Ross County are limited areally to patchy till (Boston Till) and lacustrine deposits in the Paint Creek valley. The "shallow" Russell and Miami 60 soils commonly occur together in a complex distribution; differentiation at a specific locality being based on thickness of the loess cover (Forsyth, 19&5)* Clearly no significant age difference exists between parent materials of "shallow" Russell and Miami 60 soils. In portions of the Miami Sublobe of western Ohio a "Silt Lino" separates areas dominated by "shallow" Russell soils to the south from areas characterized by higher concentrations of Miami 60 soils to the north. A similar "Silt Line", which is nearly coincident with the southern margin of the Latta­ ville Moraine, exists in Ross County suggesting that the major episodes of loessial deposition occurred before construction of the end moraine. Although "shallow" Russell and Miami 60 soils occur on both sides of the 52

"Silt Line", "shallow" Russell soils are sore common south of the boundary where they are found in complex association with "deep" Russell soils. In this complex association variability between the two Russell soil groups is related to local composition differences in parent material and variable loess thickness. Another "Silt Line" separates areas dominated by the loess-free Miami 6A soils to the north from the region characterized by Miami 60 soils. Based on loess distribution in the Miami Sublobe and southwestern Scioto Sublobe, this "Silt Line” is correlated with the Farmers- ville-Reeseville Moraine position.

Depth of Carbonate Leaching

Surface soils are often used by glacial geologists as an important tool in delineating relative ages of glacial deposits. The characteristics of a soil profile in any area depend on the interrelationships of 1) physical and mineralogical composition of the parent material, 2) time of subaerial exposure, 3) climate, 4) topography, and 5) organic activity. If climate, topography, and organic activity are averaged over a given area by careful site selection, then any differences in soil development can be attributable to variations in parent material and/or time. The relative importance of these two factors must be ascertained to allow utilization 53

of pedologic information in interpretation of the glacial history of an area. Four areas of Ross County were differentiated on the basis of variation in mean depth of carbonate leaching (Figure 18). These areas are coincident with the areas delineated by the major till-soil associations (Figure 17) because in each case time and parent materials are the primary controlling factors. Histograms of each till unit (Figure 19) show the distribution of depths of leaching in relation to the observational frequency for Ross County sites. Soil characteristics associated with the till units in Ross County are listed in Table 5. It is important to know whether the differences between the means for any two till units is statistically valid considering the number of investigations and the range of observational data. To determine if the difference between means is significant the statistic 7 - 7

* “ 6p ( V t y + \ w 2 ) 2 can be used where Sp is the pooled mean-square estimate of the variance squared given by 2 CN.-DS,2 + (H2-1)S22 Sp ■ - 5 7 7 - 5 7 7 1

(7 ■ arithmetic mean, N » number of observations, S - 5 4 -

A r e a 1 nil 2570

Are

Area 4 Area 4

Pik« Co.

IS.4 Mean loess thickness Glacial boundaries 56.1 Mean depth of leaching

Wisconsin Area 1 - Darby I Till Area 2 - Caesar Till Illinoian Area 5 - Boston Till Area 4 - Rainsboro Till

Figure 18. Map of Ross County showing areas differentiated on the basis of mean loess thickness and mean depth of carbonate leaching. Darby I Till 55 2 4 0 Caesar Till 36 1 m*an 2 0 r - N - 82 22 K • 115

> > >> o c cO 0) o 3 10 cr O'3 o

■ L T TTw a ¥ 1 * * & 6 fit n £ * Depth of carbonate leaching (inches)

Rainsboro Till Boston Till WR N - 64

o 3 § «t o 3 o* o*3 o 0) Pi U fin

8 3( 8t 5t * ¥ HU k a 5 Depth of carbonate leaching (inches) Figure 19. Histograms showing the distribution of depth of carbonate leaching in relation to frequency for Ross County till units (N - number of observations). 56

TABLE 3 SOIL CHARACTERISTICS ASSOCIATED WITH TILL UNITS IN ROSS COUNTY

Depth of Loess Percent Till Unit Leaching Thickness CaCOx (inches; (inches) equivalent

Darby I mean 24-. 0 generally 31.8 S.D. 7.5 absent 6.4 N 82 82 27 range 9-44 10.8-41.6

Caesar mean 36.1 15.4 27.3 S.D. 9.8 4.8 7.7 N 113 102 25 range 14-63 5-32 9.3-40.0

Boston mean 51.3 16.1 19.1 S.D. 9.3 5.4 7.8 N 56 58 9 range 31-82 5-51 6.7-29.1

Rainsboro mean 97.9 29.8 27.0 S.D. 14.8 10.4 9.3 N 64 46 8 range 67-143 10-79 6.9-37.2

mean - arithmetic mean S.D. - standard deviation N - number of observations range - smallest-largest values 57 standard deviation). The value of t is computed and the level of significance is determined from statistical tables (Dixon and Massey, 1957* P* 121 and 334). Values determined by this method (Table 4) clearly indicate that the mean depth of carbonate leaching is a diagnostic criterion in differentiating Ross County till units.

TABLE 4 CONFIDENCE LEVELS FOR SIGNIFICANCE OF DIFFERENCE BETWEEN MEAN DEPTHS OF LEACHING FOR ROSS COUNTY TILL UNITS Till Units Confidence Level Darby I Till - Caesar Till 99# Darby I Till - Boston Till 97*5 Caesar Till - Boston Till 95 Boston Till - Rainsboro Till 99

Assignment of relative ages to glacial deposits by soil scientists and geologists is often based on the depth of carbonate leaching. Generally depth of leaching and time of subaerial exposure are directly related. Although most investigators (e.g. Merritt and Muller, 1959) agree that carbonate content of the parent material is the most important factor controlling depth of carbonate leaching, there are many other factors to be considered including soil texture, organisms, temperature, permeability, precipitation, ground water environment, clay content, and the type and rate of surface erosion or deposition. If all these factors are considered to be a constant 58

in a given area, carbonate content: of a parent material is inversely related to the depth of leaching whereas depth of leaching and time are directly related. Rosengreen (1970) stressed the importance of proportional concentration of residual materials after carbonate leaching in progressively decreasing the rate of leaching with time (Figure 20). Not only does the depth to carbonates increase as leaching progresses but also an increasingly impermeable zone develops as the non­ carbonate materials are concentrated in the leached zone. Thus two till units of similar age could have distinctly different depths of carbonate leaching because of significantly different carbonate content of the parent materials. Radiocarbon age determinations in the southern Scioto Sublobe indicate that Caesar Till (Area 2) was deposited approximately 800 years before Darby I Till (Area 1). However, the mean depth of carbonate leaching for Area 2 is pO percent greater than in Area 1. Percent calcium carbonate equivalent is similar for both tills. This apparent anomaly is easily explained when one considers the loess component of the soil profile. Since Area 1 lacks a consistently measureable loess cap, the entire leached zone is developed in Late Wisconsin till* Area 2 has a loess cover whose mean thickness is 15.Cl­ inches. If this loossial component of the leached zone iH •rl 7 0 * -P Illinoian Rainsboro Till

•rl JO -

O O 4 0 - 40) * £o i a « Highland County O *rl so- Ross County Late Wisconsin FDarby I and Caesar Tills 3 0 - l (averaged) Caesar Till N -

Thousands of years before present Figure 20. Graph showing changing rate of leaching with time. The slope of a line tangent to the curve represents the rate of leaching in inches per thousand years at that time. The loessial component of the depth of leaching has been subtracted. Highland County data from Rosengreen 60

is subtracted from the mean depth, of carbonate leaching in Area 2, then the mean depth of leaching in the compositionally similar Late Wisconsin till units is 14- percent less in Area 2 than Area 1. Although the loess fraction must have been leached very rapidly compared to the till, the appreciable loess cover in Area 2 served to diminish the depth of carbonate removal in the under­ lying till as compared to the loess-free Area 1* Area 3 (Boston Till) is leached to a mean depth of 51*3 inches which includes a mean loess cover of 16.1 inches. Thus the mean depth of carbonate removal in the till is 33*2 inches, which is 32 and 42 percent greater than in Area 1 and 2 respectively. This greater depth of leaching in Area 3 (Boston Till, Bussell soils) was attributed to an "early” Wisconsin age of the glacial drift (Rogers, 1936; Forsyth, 1961; Forsyth, 1965* and Fetro et al., 196?)• Two radiocarbon age determinations date the correlative Boston Till in Highland County (Rosengreen, 1970) at 21,000 years B.F.; clearly early Late Wisconsin (Woodfordian) not "early" Wisconsin (Altonian). Soils associated with the Illinoian Rainsboro Till (Area 4) are characterized by deeper weathering zones than V/isconsin soils. Compositionally the Illinoian till parent material is very similar to that of the Darby I and Cae3ar Tills (see later sections in this chapter). 61

Soils in Area 4- have a mean depth of leaching of 97*9 inches and a mean loess thickness of 29*3 inches. Sub­ tracting the loess component gives a mean depth of leaching in the Rainsboro Till of 68.1 inches which is 44-. 1 and 4-7.4- inches thicker than the leached zones of the Darby I and Caesar Tills, respectively. The rates of carbonate leaching in Ross and Highland Counties are compared in Table 5. Note that the Caesar

TABLE 5 COMPARISON OP RATES OP MEAN CARBONATE LEACHING IN ROSS AND HIGHLAND COUNTY TILL UNITS. THE LOESS FRACTION IS SUBTRACTED PROM DEPTH OF LEACHING. HIGHLAND COUNTY DATA PROM ROSENGREEN (1970).

Approx. Rate of carbonate leaching Age (inches/thousand years) Till Unit (yrs. B.P.) Ross County Highland County Darby I 17,200a 1.40 b 1.25 Caesar 18,000° 1.13 Boston 21,000° 1.68 1.88 Rainsboro 128,000d 0.53 0.59 a - Radiocarbon date (OWU-256) (Moos, 1970) b - Radiocarbon date3 (tf-331, tf-91, OWU-331) c - Radiocarbon dates (D-46, 3-4-7) d - Pairbridge (1968), p. 923

Till in Ross County was leached at a slower rate than the compositionally similar Darby I Till. This variance is attributable to the presence of a loess cover on the Caesar Till as discussed above. The Boston Till was leached at a much faster rate than any other till unit in either county. This is expectable because of the much 62

lower carbonate content in the parent material of the Boston Till in comparison to the other Late Wisconsin till units. Figure 20 shows the mean depth of leaching plotted against the mean values of the compositionally similar Darby 1, Caesar, and Hainsboro Tills* Note that the rate of leaching decreases with time; the rate being the slope of a line tangent to the curve at any given time. Certainly the difference in weathering profiles between the two Late Wisconsin tills and the Illinoian till is related primarily to age of the drift and not to variations in parent material. Percent calcium carbonate equivalent (see Calcite- Dolomite Analyses for details) for the Boston Till (Area 3) averages 40 and 30 percent less than the Darby 1 (Area 1) and Caesar (Area 2) Tills, respectively. Thus the anomalously high depth of leaching in Area 3 appears to be caused by an anomalously low calcium carbonate content in the original parent material and is not related to a significant difference in time of subaerial exposure. Rosengreen (1970, p. 44, Figure 10) suggests that variability of depth of leaching values within a single parent material results from local variations in the calcium carbonate content of the parent material. 63 Loess

Boundaries between areas of significant change in loess thickness (figure 18) coincide with the distal edges of major end moraines or correlative positions (Reeseville, Lattaville, and Knockemstiff moraines). Fetrographic and x-ray analyses of the loess of south­ western Ohio indicate two episodes of loess deposition; £arly (Altonian) and Late (Woodfordian) ‘.7i scons in (Goldthwait, 1968). Mineralogic and minor textural variations allowed areal and stratigraphic separation*of distinct loess types. A stratigraphic nid-loess break has been recognized at several localities in the southern Scioto Sublobe. This break is defined by textural and mineralogic anomalies and occasionally by a weak paleosol with organic accumulations. Figure 21 is a series of histograms showing the thickness of loess cover in relation to frequency of observation for Ross County till units. Table 3 statistically summarizes the data on the Ross County loe3s cover. Thickest loess (mean thickness 29.3 inches) overlies Illinoian till and outwash and the unglaciated region of southern Ross County. The loess cover generally thins easterly from western Ross County to the Scioto River valley. East of the valley the loes3 cap abruptly Figure Frequency 28 20 IS- - - f C* I 1 Histogramsshowing thethickness 21•of loess Caesar Till i it s i 8 5 Countytill (Nunits » number of observations) coverin relation to frequency for the Ross Pq ® 3 o >> U ® o* 3 Loessthickness (inches) N« 102 Rainsboro Till N* 46 & o h a o' ifl3 ® C ITT

-

HH k « Boston Till 161 Loessthickness HMflH (inches) Ni & 58

64

65 thickens indicating that the Late Wisconsin outwashes in the Scioto Valley were a significant source of loess material. Subdued loess dunes occur in areas of thick loess on Xllinoian outwash terraces, e.g. NWJC Section 51* Springfield Township. Agricultural modification hinders use of these loess dunes in establishment of paleowind directions. Thickest loess accumulations commonly over­ lie a Sangamon paleosol. Clay mineralogy of the buried soil clearly indicates a weathering period longer than post-glacial time (Goldthwait et al., 1965). Measurements at 58 localities in the early Late Wisconsin (Boston Till) area of the Paint Creek valley gave a mean loess thickness of 16.1 inches. This indicates that approximately the lower 14-.7 inches of loess in the Illinoian area was deposited pre-early Late Wisconsin, probably in Early Wisconsin (Altonian). Only by careful delineation of the mid-loess break in thick loess sections can the thickness of the Early Wisconsin loess accumulation be determined. Deposition of the second, upper loess (Upper Melvin loess) probably began at least 20,000 and perhaps as much as 23,000 years ago (Goldthwait, 1968). Loess cover on the Caesar Till averages 15*4- inches thick. Thus only a small (less than one inch) loess increment was deposited during the interval between retreat from the Knockem3tiff Moraine and initiation of 66

retreaz from the Latfcaville Moraine, about 13,000 years B.P. Loess deposition on the Caesar Till area continued until the ice sheet readvanced to the Farmersville- Heeseville Moraine position (c. 17,200 years B.P.). Ice sheet readvance is indicated by the presence of loess, which is stratigraphically correlative with the surface loess on the Caesar Till, beneath the Reeseville Moraine in Fayette County (Moos, 1970)* Most water-well logs north of the Reeseville boundary in Ross County penetrate a "silt zone" or a "buried forest" which is probably this buried loess zone and associated organic deposits* North of the Reeseville Moraine distal boundary surficial loess is generally absent* Except for the Caesar and Boston Tills' loess cap, there is a significant difference (99 percent confidence) between the mean values of loess thickness associated with Ross County till units* Thus loess thickness is a useful criterion for differentiating till parent materials in the county. Means for the Caesar and Boston Tills are significantly different only to a confidence level of 60 percent. Thus use of mean loess thickness to distinguish between these two till unit3 must be done cautiously in conjunction with other diagnostic criteria. The lower loess (Lower Melvin loess), which directly overlies the Sangamon paleosol, occurs stratigraphically beneath till (Caesar Till) of the Cuba Moraine (Goldthwait 67

and Forsyth, 1965)* Thus the loess must have been deposited before 18,000 years B.P, Suggested correlations of the Lower Melvin loess with the Gahanna Till-Lock- bourne Outwash of central Ohio and the Roxana loess of Illinois indicates loess deposition in Early Wisconsin (Altonian, 46,000 - 52,000 years B*P.). The loess above the mid-loess break (Upper Melvin loess) is clearly Late Wisconsin (Peoria equivalent). Since a small component of this loess episode overlies the Knockerastiff and Mt. Olive (Highland County) Moraines, initiation of upper loess deposition may have begun as early as 21,000 years B.P. Cessation of loess deposition after 17,000 years B.P. probably resulted from abrupt decrease in the availability of fine outwash source material due to changing meltwater-stream regimens and vegetational stabilization of some loess source materials. Two thick-loess sites bordering the Scioto River valley were sampled and analyzed in an attempt to delineate the mid-loess break and determine the thickness of the Early and Late Wisconsin loess components.

Massieville Loess Section (Sample locations 40 - 44)

In an excavation associated with the rerouting of Ohio Route 23, 2.0 miles northeast of Massieville in Scioto Township, 73 inches of noncalcareous loess are exposed. The leached loess overlies seventeen feet of 68 coarse Illinoian outwash (High;/ Outwash). Figure 22 shows the results of granulometric analysis and the variable components of the clay mineralogy of the Massie­ ville section (complete clay mineralogy data is listed in the Appendix, Section B). Significant increase in percent silt and decrease in percent clay between samples 42 and 43 indicates the presence of two parent materials in the loess section. Neither increase in percent sand and silt nor decrease in percent clay at depth is expectable during normal soil development. The locally variable clay minerals do not clearly support the existence of two distinct source materials at the Massieville site. Although the heavy minerals analysis (Table 6) is not definitive, it strongly suggests a change in parent material between samples 42 and 43 (depth 40 - 60 inches). Certains variations in heavy minerals, e.g. decreases in percent augite/diopside and hematite/limonite and abrupt increase in percent hornblende, between samples 42 and 43 are not expectable in normal pedologic development. These variations are similar to but less definitive than results from Hocking River valley loess (Goldthwait, 1968). The above data indicate a mid-loess break in the Massieville loess section at a depth of about 50 inches. No recognizable color change or organic accumulation marks 60- 73- * 4 36- 24- 13- Depth (inches) iue£. Laboratory analysesof the Massieville loessFigure section£2. ' i 1 ' ' A • 1 1 1 • 1 1 1 1 1 \ d n a S s i s y l a n A c i r t e m o l u n a r G I 1 V t 1 \ t 1 • £ 1 t / \ / / / / / / / * 9 • * Non-celcareoue y / \ / V / 1 sk ' 1 1 • \ / i y a l C t l i S % t / t i \ i i t \ i m \ \ % \ \ \ • l \ V 1 X • • y y y • / i i lllinoian y i i t i y • • i i i i i i y \ / \ y \ / i y i i \ • f '"rf- ' A • i i i nut* i i • % • < i 1 l 1 t y t sand t < • y t y i * y V / t / 1 1 y • 1 hieliillh ' 4» A • y g o l a r e n i M y a l C y • t i 1 x i t a y I 1 and y 1 t i 1 y 1 • 1 t i t \ X X • • t 1 1 1 1 1 ■

InlWh grave) • si i I i y 1 y y i i y • i y i % y i i • i i t y t i i • i i y • i I t r " " " r k i t % latinltalilM % t \ % i \ t X * 1 i 1 / 1 / I % / t • • 1 ■ * ■ ' / / / § i / t r / • -40 -41 -42 -43 -44 £ E a • w CD fi* CD CD

O H* HEAVY MINERAL DATA OFTHE MASSIEVILLE SEYMOURVILLE AND LOESS SECTIONS P CD 03 00 03 P p P P P < Sample Number O' VJl p VX P vx ro -A O H* H* H H Aw VX VX TO _A vx vx P VX ro P ro 00 \Q ro 00 vO -A o Magnetite • •• « • •• • • -A oi ro oi Ov ro P 01 00 Ilmenite

ro P vx VX _A ro vx p p vD o oi vO 00 01 TO o Hematite • •• • t • •• • 00 VJl o vO VJl vx _A Limonite

"J S3 00 >3 VJl <£ e? oo S3 p vO _a ro vO P 5 O 00 • •• • • ••• • Total Opaques r\> P vO vx S3 01 vx P vO

VJl -s3 oo ov -A (V) vx 01 VJl Augite hj

• •• • • ••• • TABLE _A Dlopside 4 ro vO p vO 00 vO ro o CD b _a <+

ro VX ro si p VX o ro P> 6 • •• • • ••• • Hornblende m ~A _A CD 00 .4 Oi vx >3 p n

ro -A O o vx ro ro _A o • •• • • « •• • Garnet _a p vO vx 00 Oi 03 vD

VJl P VX ro SI VJl VJl ro vx Zircon • • • • • •• • • O vO oo VX _A vD O Monazite

ro ro _ a ro P vx _A _A TO UI o 00 ^3 O VJl _A \D _A Total Non- • •• • •• •• • 00 01 _A S) VX p S3 01 -A Opaques

Sample depth -0 O' P _v S3 01 P TO VJl p to VJl ru O O P 01 (inches)

06 71

this boundary. Thus the upper 30 inches of loess may have been deposited in Late Wisconsin whereas the lower 24 inches may he Early Wisconsin or Illinoian.

Seymourville Loess Section (Sample locations 8? * 86)

In an actively-operated gravel pit (S£3£, NEJ4, Section 20, Springfield Township) in the highest Higby Outwash terrace on the east side of the Scioto River valley opposite Chillicothe, a thick, partially calcareous and fossiliferous loess section is exposed (Figures 23 and 24). The 84-inch-thick loess accumulation is leached of carbonates to a depth of 34 inches. The sparsely fossiliferous zone lies between 66 and 72 inches below the top of the exposure. Most of the individual shells were at least partially leached during soil development, numerous, irregular secondary accumulations of calcium carbonate occur in the lower 18 inches of the loess. Several of the secondary deposits take the form of rhizoconcretions, often with highly decomposed rootlet nuclei still visible. Three species of terrestrial Pleistocene molluscs were identified from the fossiliferous zone: Anguispira alt e ma t a (Say) Stenotrena fraternum (Say) Succinea gro3venori (Lea) According to LaHocque (1970) all three species are adapt- 72

Figure 23. Illinoian Higby Outwash and Parke soil exposed on the east wall of a gravel pit near Seymourvillo (sample location 83)* 3?he Seymourville loes3 section overlies the outwash in the upper left center of the picture. 73

Figure 24. An excavation at the Seymourville loess section. The scale is one yard long with one and six-inch gradations. The lower six inches of loess is calcareous and sparsely fossiliferous. 7*

able to wide ecological ranges but the assemblage is indicative of a scrub woodland to open field environment. Anguispira alternata (Say) ("Tiger Snail”) has been identified in drift ranging from post-glacial to Aftonian age and is thought to indicate a woodland environment (Oughton, 194£)• Stenotrezna fraternum (Say) has not been previously identified in Ohio Pleistocene deposits (LaRocque, 1968). Living species in Ohio inhabit open fields and woodland environments. Baker (1920) indicates that this species has been described in glacial deposits ranging from Tarmouth to "Wabash" (Late Wisconsin) age. Succinea grosvenori (Lea) has no living forms in Ohio. This species was identified at the Sidney cut (LaRocque and Forsyth, 1957) and was thought to be of Early Wisconsin age on the basis of the local stratigraphy and aerial relationships. This species has been identified at Cleveland (Leonard, 1953) in Sangamon loess and lower and upper pro-Tazewell (Early Wisconsin?) loess. These three molluscan species must have lived during the interval between Early and Late Wisconsin loess deposition because of their occurrence within the loess section. Size of some individuals clearly indicates that the fossil assemblage is indigenous and was not accumu­ lated by wind transport. Therefore, on the basis of the fossils the mid-loess break is placed at a depth of 66 75

inches. The 1S inches of loess below the boundary may be Early Wisconsin (Altonian) or possibly Illinoian while the upper 66-inch increment was probably deposited during Late Wisconsin (Woodfordian)• Variations in results of laboratory analyses above the postulated mid-loess break (Figure 25) are generally expectable in normal soil development. However, decrease in percent clay and increase in percent quartz (clay fraction) suggests two parent materials are present in the section. Note that most of the calcium carbonate in the calcareous portion of the section is in the dolomite. This indicates that weathering has proceeded to a point where most of the more easily leached calcite has been removed. Heavy mineral data (Table 6) from the Seymourville loess section is generally inconclusive in establishing the mid-loess break. The abnormalities in mineralogy seen from the Massieville site are absent at this section. Since the two sections are only 5.5 miles apart, the distinct variation in heavy minerals between the sites is puzzling. This variance can be explained only by major variations in source material over a short distance or by anomalously different weathering regimens in topograph­ ically and cliaatologically similar areas. Therefore, both loess sections seem to contain a mid­ loess break separating an Early Wisconsin loess froa a Chittick Granulometric Analysis Clay Mineralogy Analysis • Loess % % % % * % * % % % a i! E M Silt CWy HIM Moiilist W wumUIM • oj Section Caklto VtrmlnoUN Owarts 10*14 JT H-id CO t l * * ■“’"A m r^ib i~ n ■"ira .n> n r a r i A

Non- 10* calcareous reddish • • • •• • • • • • -03 ■ i > t \ * 1 » (7.5YR7/6) t i i i i t i i \ # loess ii i / ■ t i i i 1 < i % t i 1 30* i t i i 9 * i \ Non- i t i i \ \ > i 1 • j 9 i i i i i 1 calcareous 1 40- • I • i • • m • • -84 yellow-red V i / \ % * * $ 1 i I7.5TTR6/6) / 1 JO- / \ t i t loess / \ \ / t > I / \ V i \ / 1 i \ 4A. Calcareous / % \ t \ / I i \ / % * 9 1 \ loess • • • • k 0 9 i1 • 0 -85 t ■ * i 1 % * w (fossils) 1 1 / t 1 4 \ i y 70- 1 i 1 i i / i 1 / • 1 1 V % i / Compact f 1 / f 1 % V\ t w gray loess 0 • • • • 0 • • • -88 00- C7.5YR5/1) lllinoian 90- sand and gravel

Figure 25* Laboratory analyses of the Seymourville loess section. 77 much thicker Late Wisconsin loess (Upper Melvin loess).

Paleosols

The occurrence of a buried soil, developed in out- wash gravel or till, beneath two to ten feet of till has been reported at several localities in western and southern Ohio (Goldthwait, 1952; 1955; 1959)* The paleo­ sols typically exhibit: 1) distinctive red-brown color, 2) clay enrichment, and 3) "ghosts" of former calcareous material. In some cases, leached reddiBh pods of a paleosol were incorporated into the overlying till. Two explanations have been offered to account for these buried soils. Goldthwait (1955; 1959) suggested that the leached zones are true paleosols, developed during interglacial periods and thus imply periods of subaerial exposure during their development. The alternate hypothesis holds that these buried soils are areas of clay concentration (Beta layers) which developed synchronously with the present surface soil through leaching, movement down through the till, and concen­ tration of fine materials in the underlying gravels (Gooding et al., 1959)- Clearly all paleosols noted in Ross County are true buried soils and not zones of clay accumulation. Paleosols developed during interglacials and covered by later glacial deposits are reliable marker horizons 78

in glacial stratigraphic correlation. The existence of a

buried soil in glacial drift can be used t:o document ice sheet retreat and subsequent readvance and therefore, the identity of at least two glacial stratigraphic units. The time interval represented by a buried soil is difficult to determine in Ohio because in most cases the entire paleosol is not preserved. Ice advance after the interval of soil formation usually removed the entire A and most of the B-horizons leaving only the C-horizon in most paleosol localities. Newberry (1874) was the first to note the occurrence of a buried "forest bed" and related paleosol about 30 feet beneath the surface in north-central and north­ western Ross County. Most water-well logs in this area indicate the presence of this zone. This organic deposit is probably correlative with the paleosol described by Moos (1970) from beneath the Reeseville Moraine in Fayette County. If the correlation is correct then the paleosol and "forest bed" must have developed during the short interval (about 800 years) between retreat from the Cuba and Lattaville Moraines and subsequent readvance to the Reeseville Moraine position. A v/ell-developed Sangamon paleosol is exposed in a gravel pit 0.3 mile south of Humboldt in western Ross County (central Faint Township). The paleosol is developed in Illinoian gravel that is overlain by till 79

("early" Wisconsin till of Petro et al., 1967). Depth of leaching, stratigraphy, and particle-size distribution indicate that this is Boston Till (early Late Wisconsin). Determination of the age of the buried soil is critical in establishing the age of Glacial Lake Humboldt deposits which overlie the Boston Till (Reynolds, 1959)* Sangamon paleosols, developed in Illinoian till or outwash beneath thick loess accumulations, are commonly found in' Ross County. Although exposures are limited, (e.g. Stop 15, Goldthwait, 1955) existence of a more highly weathered profile, upon which the present soil development has been superimposed, can be seen by augering in thick loess areas (e.g. Plyley and Poplar Ridges in Concord and Twin Townships, respectively). Two exposures in Ross County show evidence of pre- Sangamon weathering intervals. A brownish-yellow sand unit underlies Rainsboro Till at sample location 53, 1.8 miles east of Schooley in Liberty Township (see page 36 for description and discussion). Unit coloration and the local stratigraphy suggest the sand unit is a remnant of a Yarmouthian weathering profile. At the Massieville section (Figure 26) a 10- to 17- inch thick, black (7-5YR2/1), noncalcareous, organic- rich zone was exposed near the base of the section. The organic zone directly overlies a fractured sandstone surface which rises to the north in the section. The 0 50 fMt Late Wisconsin L *und Early Wisconsin^ ^ non-calcareous loess 'lOYRS/S, maximum thickness 7 5 V |amplfs partially stripped

>d^pCh rskching Si^JV'-.S • o • -o. . 0 . 0 * • £ * ..O' o . • * a o : * . o • o . « o 0*0 -■ o ° o' ° ° coarse calcareous, sand* and gravel” (IlliiToian) T -o Q ' • * : b 17* > . o o o - o O •« . » • O • O 0 . 0 . ■ ° •

0^— 2- Irhythmically-banded lacustrine sediments (Illinoian): 12.5’

om

--'V o | ^ d e ^ 5 - ickness 17'* • .y. r c r u s t j ^ ^ v ^ . A *

a - « X it X fractured,vam lar sandstone X V * ^ floorj Figure 26. Diagram of the Massieville section (0.1 mile west-northwest of the junction of Route 23 and Three Locks Road, south-central Scioto 09 Township) as exposed during the summer 1972* The stratigraphy is O at present poorly exposed due to slumping and human modification. 81 upper surface of the zone is a one-eighth to one-quarter inch thick manganese oxide crust which contains small, decomposed wood fragments. The crust is overlain in turn by fine, cross-bedded sand and gravel which contains many lenses of very-fine sand. The sand and gravel is capped by a slightly calcareous, dark gray (7.5YR4-/1) till. Local stratigraphy and laboratory analyses indicate the till is Xllinoian Rainsboro Till. Thus the organic zone must be related to a period of subaerial or subaqueous exposure in pre-Illinoian, possibly Aftonian, time. Since two episodes of Xllinoian glaciation are indicated by two levels of Higby Outwash in Ross County it is likely that the outwash below the till represents the earlier maxima while the till was deposited during the second Xllinoian interval.

Granulometric Analyses

Farticle-size distribution is a standard tool for correlation of tills in Ohio. While some studies have yielded inconclusive results (e.g. Forsyth, 1956)? in most cases the data has been internally consistent and useful (e.g. Shepps, 1955; Steiger, 1967; Steiger and Holowaychuk, 1971)- Variations in particle-size distribution generally is thought to indicate a difference in source area. Dreimanis and Vagners (1969; 1971) indicated that for each 82

mineral there is a limiting grain-size to which it can be reduced by glacial erosion, Therefore, variable particle- size distribution between mineralogically distinct source areas is expectable. Although grain-size distributions are similar for most samples within a till unit, a few samples vary widely from the mean. These samples usually result from incorporation of unconsolidated material into till of an advancing ice sheet. Thus the anomalous samples are usually restricted to the basal portions of till units. Careful sampling from consistent, non-basal, positions within till sheets is necessary for reliable data which is useful in stratigraphic correlation. Farticle-size distribution was determined for 70 Hoss County till samples to determine the usefulness of this criterion in stratigraphic correlation. Table 7 summarizes the data obtained from this procedure. Grain-size distribution for the Darby I, Caesar, and Bainsboro Tills is remarkably similar (Table 7» Figure 27). Note there is only a three percent variance between the arithmetic means for the sand, silt, clay, and pebble fractions of the three till units. This similarity is expectable because compositionally (see later sections of this chapter) the three till units are nearly indistin­ guishable, The particle-size distributions of the Ross County till units are similar to distributions from other 83

TABLE 7 FARTICLE-SIZE DISTRIBUTION OP TILL UNITS IN ROSS COUNTY n o a Weight Percentages Till « Pt ( 2mm on total sample; sand-■silt-clay Unit *§ | on 2mm fraction) 3 w ft Sand Silt Clay 2mm

Darby I 28 mean 24.8 52.1 23.1 11.7 S.D. 4.3 4.7 6.8 7.8 30.4 61.1 38.6 29.7 rangelow 15.9 48.4 14.0 0.6

Caesar 25 mean 26.9 50.7 22.4 13.1 S.D. 5.5 4.2 4.0 7.6 40.5 62.5 30.1 36.4 ® low 19.5 46.6 16.7 3.2

Boston 8 mean 7.7 48.8 43.4 2.8 S.D. 5.7 4.2 9.1 2.2 ^o-^-high 16.8 55.9 56.1 5.5 ran8«ioS 0.5 43.5 33.4 0.0

Rainsboro 9 mean 24.7 50.9 24.5 10.9 S.D. 8.3 5.9 6.2 5.6 41.0 60.5 35.2 17.1 range^oS 14.1 38.5 17.4 1.7 mean - arithmetic mean S.D. - standard deviation Sand 84

'v.

Silt (0.62- 0 .002mm) Till Unit Age Samples • Darby I Late Wisconsin 28 O Caesar Late Wisconsin 25 A Boston Late Wisconsin 8 A Rainsboro Illinoian 9

------Highland Co. 111. - Late Wise. 51 ------Darby I-Caesar Late Wisconsin 286 ...... Gahanna Early Wisconsin 13 ------Rainsboro Illinoian 51 Figure 27. Three component diagram showing the relation­ ship of mean particle-size distribution of Ross County till units to other Scioto Sublobe tills, (data from Rosengreen, 1970 and Goldthwait and Rosengreen, 1969). 85

southern SciOto Sublobe till localities (Figure 27). Except for the silt fraction, the Boston Till is anomalously different from the Darby I, Caesar, and Rainsboro Tills. Boston Till has about two-thirds less sand, twice as much clay, and about one-fourth as many pebbles as the other three till units. Confidence levels for differentation of the Boston Till from the other Ross County till units (Table 8 ) indicate it can be clearly distinguished from the Darby I, Caesar, and Rainsboro Tills on the basis of particle-size distribution.

TABLE 8 CONFIDENCE LEVELS FOR SIGNIFICANCE OF MEAN DIFFERENCES FOR PARTICLE-SIZE DISTRIBUTION BETWEEN BOSTON TILL AND THE DARBT I, CAESAR, AND RAINSBORO TILL UNITS Percent Confidence Till Unit Pebbles Sand Silt Clay Darby I 99 99 — 99 Caesar 99 99 — 99 Rainsboro 99 99 — 99

The high clay content in the Boston Till is probably the rosult of incorporation of lacustrine sediments of Illinoian Glacial Lake Bourneville by the initial Late Wisconsin ice advance. Since these lacustrine sediments are confined to the Paint Creek valley, the Boston Till north of the valley may have distinctly different textural characteristics than the till exposed in the Knockeiastiff Moraine. Such areal variation in till 86

texture nay hinder identification of Boston Till in till exposures outside of the Paint Creek valley. Correlation between areas mapped as Boston Till in Soss and Highland Counties is supported by the granulo­ metric data. Although the variations between Boston Till and other Highland County till units (Rosengreen, 1970, p. 58) are not as pronounced as between correlative Ross County units, all differences are consistent in character; i.e. Boston Till has less sand and pebbles and more clay than other Highland County tills.

Calc it e-Polomite Analyses

Variations in calcite-dolomite content in tills reflects differences in the composition and quantity of bedrock and soil material incorporated into drift by an advancing ice sheet. Weight percentages for the calcite, dolomite, and

total carbonate (# CaCO^ * % dolomite (1.085) + % calcite) content of 69 Ross County till samples were determined by the Chittick gasometric procedure (Dreimanis, 1962). Statistical results of these analyses are listed in Table

9. The calcite content of the four till units is nearly identical within the operational accuracy limits of the procedure (Table 9, Figure 28). However, the dolomite content and consequently the total carbonate percentage of 87

TABLE 9 CALCITE AND DOLOMITE CONTENT IN THE LESS THAN 2mm SIZE FRACTION FOR TILL UNITS IN ROSS COUNTY

Till Weight percentages (for mean values) CaCO, Unit 5 Calcite Calcite Dolomite equivalent Dolomite samples Number

Darby I 27 mean 7.3 22.5 31.8 0.3 S.D. 1.9 4.4 7.0 0.1 -.--.high 10.6 29.4 41.6 0.7 ® low 2.4 6.0 10.8 0.2

Caesar 25 mean 7.2 18.6 27.3 0.4 S.D. 2.8 4.6 7.1 0.1 -.--.high 13.6 27.9 40.0 0.7 rangeio5 3.1 5.7 9.3 0.1

Boston 9 mean 7.4 10.8 19.1 0.7 S.D. 4.3 3.4 6.9 0.2 -0-„„high rangelow 13.7 14.7 29.1 1.1 1.3 6.0 6.7 0.3

Rainsboro 8 mean 8.0 17.5 27.0 0.5 S.D. 3.4 5.8 8.6 0.2 high 13.0 28.0 37.2 1.0 ranSelor 5.3 3.3 6.9 0.2 mean - arithmetic mean S.D. - standard deviation iue2. Bargraphs ofthe mean calcite andFigure 28.dolomite Weight Percentage 0 2 M- 32 12 - ab Cea Bso Rainsboro Boston Caesar DarbyI JZ. contentof Ross County tillunits. Till Unit QD □ dolomite calcite 8 8

89

Boston Till is distinctly lower than in the Darby I, Caesar, and Rainsboro Tills. Variations in the calcite/ dolomite ratio also clearly separate Boston Till. These differences are consistent with compositional data determined by pebble counting (see Pebble Counts, p. 98). Carbonate content of the Boston Till in Highland County (Rosengreen, 1970, p. 65) has similar variations with respect to other Highland County tills, except in an anomalous calcite/dolomite ratio. Confidence levels for the significance of differentation between the means for the carbonate content of the Boston Till and other Ross County till units (Table 10) indicate that percent TABLE 10 CONFIDENCE LEVELS FOR SIGNIFICANCE OF DIFFERENCE BETWEEN THE MEANS FOR CAR30NATE CONTENT BETWEEN / BOSTON TILL AND OTHER ROSS COUNTY TILL UNITS Percent confidence Boston Till Total Calcite Till Unit Calcite Dolomite Carbonate Dolomite Darby I — 99 99 99 Caesar •— 99 97.5 99 Rainsboro — 95 97*5 95 dolomite and total carbonate are diagnostic criteria for distinguishing Boston Till from the Darby I, Caesar, and Rainsboro till units. Chittick analysis also proved useful in determining the degree of weathering in the calcareous loes3 at the

Seymourville loess section. Nearly all of the calcium 90

carbonate present was in the form of dolomite with little calcite remaining. This indicates that weathering has proceeded to remove most of the more easily soluble calcite and has reached the stage of primarily dolomite leaching.

Calcite-Dolomite Analyses

The clay mineralogy of the till matrix reflects the composition and quantity of material incorporated during ice sheet advance. Although studies of clay mineralogy of Ohio tills has been primarily by soil scientists (Andrew, 1960; Bidwell, 1959* Holowaychuk, 1950; Wilding et al., 1965)t geologists (e.g. Droste, 1956; Teller, 1970; Rosengreen, 1970) have made contributions. Studies in Illinois (Willman et al., 1965; Johnson, 1964; Frye et al*, 1969) have provided detail of the clay mineralogy in glacial deposits elsewhere in the midwest. Some workers (e.g. Bhattacharya, 1962) have used clay mineralogy as a tool in determining degree of weathering in tills. Others suggest clay mineralogy is indicative of the intensity of interglacial erosion and inclusion of previously- weathered regolith. Thirty-one Ross County till samples were analyzed by x-ray diffraction (for procedural details see the Appendix, Section B) for eight clay-size mineral compo­ nents (Table 11). Bata determined by this semiquanti- 91

TABLE 11 CLAY MINERALOGY OF THE LESS THAI T7C MICRON FRACTION OF ROSS COUNTY TILL AMD LOE33 UNITS

co m. - y ® m t W » Till/ r - t 0 % A P 0) 0 € *H *H *H 0 (3 M M _ to O 0 'H >H Loess h p -pcw -P04 M *H *H 0 0 00 Qj O- O *H rH © -p fi.fi. P 3 P -H p K\ p vO TT~4+. ^ • 9 O *rl G N w O r Unit 0 P S *rt P -H P h b ,0 *rt P S OrH P ) 0 O Sr-tflfir-IOCdP P 3r-tQ0,30 3ti 0 fcHS>U«Q'H H

Darby I 15 mean 80 0 10 5 0 5 0 0 S.D. 5 - 5 5 - 5 -- _____high 85 Tr 15 10 Tr 10 Tr Tr ranSelo5 75 0 0 0 0 0 0 0 C.L. 99# 99# — ** 99# ** 99#

Caesar 9 mean 80 0 10 5 0 5 Tr Tr S.D. 10 - 5 5 - 5 — — 85 Tr 30 10 5 15 Tr 5 55 0 0 0 Tr 0 0 0 C.L. 99# 99# — 99# — 99#

Boston 5 mean 65 0 20 0 Tr 10 0 5 S.D. 5 - 5 — - . 5 — — Tr Tr rangei0° 75 0 25 5 10 5 60 0 10 0 Tr 5 0 0

Rainsboro 4 mean 0 10 0 10 0 0 75 5 c S.D. 5 - 5 —- s - - T.OMn..hish 80 Tr 15 5 5 10 Tr Tr ranseiQr 75 0 5 0 0 5 0 0 C.L. 99# — 99# — — 99# — 99# 92

CO © ©01 _ - _ _ j «H 0) T) Till/ P< -p © WB -H (3 *HV| -H CO O © *rl Loess m h S _ © td^5 g^cs O *rl rH © -P H • U 3 +» *H +>K> +»l0 h © O o *rt Q h n r Or Unit— >*. .Q-H-PSOiHh© ©+»S*rlfc*«H-Ph © BrHCjfirHOrt-P +» SHOOXlCdSfl (3 &5HS>OMO*H H

Loess 10 mean 35 0 35 0 Tr 15 Tr 15 S.D. 10 - 10 — - 5 — 5

■A* W high 60 Tr 4-5 Tr 5 20 5 30 ranSelow 20 0 15 0 0 10 0 0 All percentages rounded to the nearest five percent, Tr- trace; less than 2.5#. mean - arithmetic mean S.D, - standard deviation C.L. - confidence level of mean differences compared to the Boston Till value 95

tative method should be reported only to she nearest five percent (L. P. Wilding, personal communication). Since midwestern tills usually contain variable amounts of montmorillonite, this clay mineral is generally used as a member of three-component diagrams in comparing clay mineralogy of tills. However, because no Ross County till sample contained a measurable amount of montmoril­ lonite, percent quartz plus expandables was substituted for montmorillonite as one member of the three-component system (Figure 29)* Unfortunately this substitution hinders comparison with the clay mineralogy of other mid- western tills (Figure 50)• Without this change the Boss County samples would plot along the base of the three- component diagram, rendering separation of the tills based on the clay mineralogy ineffectual. Clay mineralogy of the Darby X and Caesar Tills is nearly identical (Table 11, Figure 29)* These tills are characterized by high illite and low veraiculite content. Rainsboro Till clay mineralogy is intermediate between these two tills and the anomalous Boston Till. The Illinoian till has five percent less illite and five per­ cent more quartz than the Darby I or Caesar Tills. This variation may be the result of sampling of slightly- weathered till instead of fresh, unweathered material. Proportionate decrease in illite and increase in resistant quartz can be attributed to weathering phenomena. If the 9*

Quartz, Expandables

Hite Vermiculite Kaolinite Chlorite

Number of Till Unit Age Samples • Darby I Late Wisconsin 15 O Caesar Late Wisconsin 9

A Boston Late Wisconsin 5 A Rainsboro Illinolan 4- Figure 29. Three-component diagram showing mean clay- mineral composition of Ross County till units. Montmorillonite 95

Illite percent Vemiculite Kaolinite Chlorite

Woodfordian (Late Wisconsin) Altonian (Early Wisconsin) Illinois Buffalo Hart (Illinoian)

Boston Till (Late Wisconsin) Ohio Darby I, Caesar and Rainsboro (Late Wise.) (Illinoian)

Figure 50. Three-component diagram contrasting clay- mineral composition of till units In Highland County with selected tills from Illinois and Indiana (after Rosengreen, 1970, Figure 15, p. 56). 96 differences are related to sampling error, the clay mineralogy of unweathered Rainsboro Till is very similar to that of the Darby I and Caesar Tills. Similarity is expectable because the ice sheets which deposited all three tills advanced over the same middle to late Paleo­ zoic bedrock of Ohio. Boston Till is characterized by lower percent illite and higher content of vermiculite and expandable clay minerals compared to the Darby I, Caesar, and Rainsboro till units. This difference may be related to incorpo­ ration of material weathered during the Sangamon inter­ glacial by the initial Late Wisconsin ice advance which deposited the Boston Till. Confidence levels for distinguishing the Boston Till from the other Ross County till units (Table 11) indicate such differentation can be readily made solely on the basis of clay mineralogy of the till matrix. Taking into account the substitution of one member in the three-component diagrams (Figures 29 and JO), the Ross County tills are similar in clay mineral composition to correlative units in Highland County and Illinois. Fre-Illinoian and Illinoian tills have expectably higher content of v/eathering products with consequent depletion of the percent of easily-weathered constituents. Variations in percent montmorillonite between neighboring Ross and Highland counties may be more the 97

result of differences in data presentation than in actual variation in montmorillonite content. Rosengreen (1970), using similar x-ray diffraction methods as employed during this study, reported clay mineral data to the nearest percent; clearly beyond the range of data reliability in this semiquantitative method. Note that the Darby X, Caesar, and Rainsboro Tills of Highland County (Figure 30) average less than five percent montmorillonite, with only the Boston Till containing as much as 10 percent of this clay mineral. In the Ross County data (Table 11) Darby I, Caesar, and Rainsboro Tills all contain trace amounts (less than 2.3 percent) of montmorillonite. Thus the clay mineralogy variations between the two counties may not be as pronounced as the diagrams indicate. Possibly the differences in bedrock lithology between southern Ross County (shale) and central Highland County (limestone and dolomite) are sufficient to have caused the clay mineralogy to vary over a relatively small area. Analyses of ten loess samples from the tfassieville and Seymourville sites indicate that the loess is characterized by half as much illite, twice as much vermiculite, one-half to two-thirds more quartz, and three times more expandable clay minerals than Ross County till units. Since all of the loess samples are from the upper portion of deeply-weathered profiles, the variations are probably as much the result of weathering as differences 98 in clay content of the unweauhered parent materials.

Pebble Counts

Till Units

Lithology of pebble-3ize material in till has been used to differentiate till units that were deposited by ice sheets that traversed areas of different bedrock lithologies. This method is advantageous because it is simple, rapid, and can be completed in the field. Pebble counts have proven useful in stratigraphic correlation in many studies (Anderson, 1957; Norris et al., 1950; Drake, 1970; Rosengreen, 1970). Results of 28 pebble counts in Ross County tills are listed in Table 12, Darby I, Caesar, and Rainsboro Tills have very similar pebble lithologies (Figure 31). These tills are characterized by 75 to 81 percent carbonate pebbles with approximately twice as many dolomite as limestone pebbles. Boston Till is significantly different from the Darby I, Caesar, and Rainsboro Tills to at least a 95 per­ cent confidence level for all lithology groups (Table 12) except limestone, chert, and metamorphics. This earliest Late Wisconsin till averages less carbonates, more elastics, and more crystalline pebbles than the other Ross County till units. Boston Till also has a much TABLE 12 PEBBLE LITHOLOGIES OF TILL UNITS IN ROSS COUNTY

© © u © © © X! O 09 ® P Sis -p © p P p © 35 Till 0> p o H © XI H o o rH O © g1 rH © Jh p H *rt +> © p o a p p © *H P O © © ® Pi a © p o o © © © P P O a P P & a o © 0.0 + H i—t p O © © © O H Unit a 3 pH a © £4 fH O © H £4 © P p Eh rH P 0 3 O .p at CO 3 XI pH bO © & n o o W CO 03 O H O Darby I 9 mean 51 26 2 79 .54 6 3 4 13 5 3 8 S.D. 6 5 — 6 .19 ——- 5 —— 2 high 64 12 11 21 rangB1

Total Clastics and Crystallines

Limestone ptrc*nl Dolomite and Chert

Number of Till Unit Age Samples • Darby I Late Wisconsin 9 O Caesar Late Wisconsin 9 ▲ Boston Late Wisconsin 6 A Rainsboro Illinoian 4 Figure 31. Three-component diagram showing the mean pebble lithology for Ross County till units. 101 higher calcite/dolomite ratio than the other till units* She low dolomite content in the pebbles of this till is anomalous and can only be related to unknown differences in bedrock source. The distribution of limestone and dolomite in the Boston Till pebbles is consistent with the calcite/dolomite ratio in the less than 2mm size fraction (Table 9, page 87)* Although Boston Till averages only three percent chert pebbles, several samples contained a much higher (up to 11 percent) chert content* Since the chert must be derived from weathering of the Paleozoic limestones of Ohio, it is consistent that the first till deposited after the Sangamon weathering interval would contain locally high concentrations of this residual weathering product* Variations in pebble lithology and other composi­ tional aspects of till, over a relatively small area, indicate that the majority of the material incorporated into the till was of local derivation; i.e. deposition within a few miles of the source area. If distances of material transport were not minimalf then local variations in bedrock types would be averaged out by mixing during long distances of ice sheet advance* Such averaging of local anomalies would make compositional differentation of till units nearly impossible* 102

Outwash Units

Kempton and Goldthwait (1959) found significant differences in the pebble lithologies between Illinoian and Wisconsin outwashes in a single valley such as the Scioto or Hocking River valleys* Thirty-three pebble counts were made in the four Ross County outwash units (Table 13)* Percentages of pebble lithologies for the Worthington, Circleville, Kingston (Scioto Valley), and Higby outwash units are all very similar (Figure 32). These outwashes are character­ ised by 76 to 78 percent carbonate pebbles (with about twice as many dolomite as limestone pebbles), eight to ten percent elastics, and 12 to 16 percent crystalline pebbles* Although the number of pebble counts is minimal, that portion of the Kingston Outwash in the Paint Creek valley can be readily distinguished from the other out­ wash units on the basis of pebble lithologies* This Kingston Outwash has a higher dolomite and carbonate content, a lower calcite/dolomite ratio, similar percent elastics, and a lower content of crystalline pebbles in comparison to the other Ross County outwash units* The pebble lithologies for the Ross County till and outwash units (Tables 12 and 13) are strikingly similar except for the Boston Till and the Paint Creek valley TABLE 13 PEBBLE LITHOLOGIES OP OUTWASH UNITS IN ROSS COUNTY

% * « © e © © XJ O © a p © 4* « A © ft Outwash © p 0 rH © xl *H O o rH O a h r H UJ M © p © a o a P p © *H O © © © H a © p o o Q © PP o P p a X> ft 0 © H •© p o © © © O *rl a a rH a © Eh U © rH EH © P Unit O *H xi m P *rl rH & © Ss5 P P o o XI co-9 CQ O o Worthington 7 mean 46 26 4 76 .65 6 1 2 9 9 6 15 S.D. 5 3 - 3 .19 --— 2 - — 3 high 54 31 6 81 .89 7 3 5 13 13 9 20 ® low 57 20 2 70 .48 2 0 1 5 6 2 11 Circleville 7 mean 53 23 2 78 .47 5 1 4 10 8 4 12 S.D. 3 4 - 3 .15 -— «* 2 - - 3 ^ ^ ^ h i g h 60 30 4 84 .64 2 6 12 10 8 18 ranGe 7 lo5 49 16 0 73 .28 3 0 2 6 5 1 9 Kingston 8 (Scioto River) mean 50 24 3 77 .54 6 0 3 9 9 5 14 S.D. 4 3 — 3 .16 —— - 2 -- 4 59 28 4 83 .74 9 1 5 12 14 7 21 s low 43 19 1 71 .35 4 0 0 5 5 2 10 Kingston (Paint Creek) mean 73 13 1 87 .19 4 3 1 7 5 1 6 S.D. 2 2 — 3 .02 —— — 2 — - 3 76 16 3 91 .24 6 4 2 10 7 3 10 69 11 0 82 .17 2 2 0 6 1 0 1 % -p © © u © © © O 0) P p « PP P 03 © P O H d .© HO o O © p-t © Outwash P « •H -P © p O 0 P p rH *rl P O © © © h n a P p o O © © © © P O a p p ,0 Pi o © P 0,0 + rH rH p P © © © O *H £ H r H a m H e © EH P O P © rH O © R p L 4 Unit P at O ■ H ,© © © R © .P •H E-t H S > i©* 1 4 S W R o o w CO CQ O H t-i— O

Higby mean 54 20 2 76 40 5 1 2 8 10 6 16 S.D. 3 3 — 4 10 _— 2 —_ 4 60 26 84 51 6 2 3 10 15 9 24 raneelow 51 17 0 69 25 3 0 0 6 7 3 11 mean - arithmetic mean S.D. - standard deviation tOU 105 Total Clastics and Crystallines

Limestone pvrctnt Dolomite and Chert

Number of 0utwa3h Age Samples • Worthington Late Wisconsin 7 O Circleville Late Wisconsin 7

A Kingston (Scioto R.) Late Wisconsin 8

A Kingston (Faint Cr.) Late Wisconsin 5 ■ Higby Illinoian 6 Figure 32. Three-component diagram showing the mean pebble lithology for Ross County outwash units. portion of the Kingston Outwash. This Kingston Outwash has a much higher calcite/dolomite ratio and about 50 percent more dolomite pebbles than the areally-related Boston Till. The percentage of abrasion-resistant crystalline pebbles in the Boston Till is two to three times greater than in the Kingston Outwash (Faint Valley) and other Hoss County till units, but is similar to the other outwash units in the county. The Boston Till is also characterized by percentages of easily-abraded, locally-derived clastic pebbles which range from one- third to two times greater than other Ross County till units and two to three times greater than all outwash units. The high percentages of both clastic and crystal­ line pebbles in the Boston Till indicates two phases of ice transport to allow incorporation of materials of widely variable resistance to chemical and physical break­ down.

Heavy-Kineral Analyses

Heavy-mineral analysis often provides the most effective means of differentiating tills from different source areas. The usefulness of heavy-mineral data in glacial stratigraphic correlation has been demonstrated in Ontario (Dreimanis et al., 1957) and in Illinois (V/illman et al., 1963)- The primary source of the heavy minerals in the Ross County tills is the Frecambrian 10?

igneous-metamorphic conples of the Canadian Shield of Ontario. Heavy-mineral analyses were made on the very-fine sand fraction (0.062 - 0.125mm) of 25 Boss County till samples. Results of these analyses are listed in Table 14 (for detailed data see the Appendix, Section B). The heavy minerals were grouped into two primary suites; opaques and non-opaquesv with several subgroupings of non­ opaque minerals differentiated. The miscellaneous grouping includes those minerals with sporadic occtuv rence in the tills such as tourmaline, kyanite, topaz, zircon, olivine, sillimanite, apatite, and andalusite. All four Ross County till units have a similar distribution of heavy minerals (Table 14, Figure 33)* These tills average about 30 percent opaques, 20 percent garnets, and 50 percent other non-opaque heavy minerals. This similarity in data is consistent with other heavy- mineral studies of midwestern tills (Rosengreen, 1970; Teller, 1970). Since all four tills were deposited by the same sublobe of the Brie Lobe and the source area of the heavy minerals is far distant from Ross County, it is expectable that any local variations in heavy minerals in the source area would be normalized during the hundreds of miles of glacial transport prior to deposition. On the basis of weight percent of heavy minerals in the very-fine sand fraction (Table 14) Illinoian Rainsboro TABLE 14 HEAVY-MINERAL DATA OP .THE VERY-FINE SAND FRACTION ( 0 .062-0*125mm) FOR TILL UNITS IN ROSS COUNTY

Opaques Non-opaques Percent Heavy Minerals Till © c © © © 43 O B» © 1 43 © Garnets +> © - p *rl © Unit £4 rH © •rt iH © 43 © o © 43 4) Pt u rl « © a 43 O & B OS os © © © 43 © tI *S © B aJ •P a © •p fl ! cx O 43 to •H 4> ©CO o *rl o O o •a O 3 O £3 EH C4 o EH 1 w tu s to » EH

Darby I 8 *2.73 mean 26.1 31.9 6.7 14.3 21.0 2.7 2.4 32.1 2.0 1.6 0.2 1.6 2.0 1.8 47.1 S.D. 4.5 3.5 1.9 2.3 3.3 3.9 3.4 32.3 36.9 10.4 16.9 23.3 4.1 4.0 38.6 3.5 3.6 0.8 3.9 3.0 3.3 53.8 ran6elow 19-7 27.3 3.9 11.3 14.9 1.4 1.2 26.2 0.9 0.6 0.0 0.9 1.0 0.6 43.3

Caesar 7 *2.74 CVJ CO mean 21.1 25.5 6.4 13.8 20.2 2.4 2.8 35.9 2.6 2.0 0.4 • 2.4 2.2 54.3 S.D. 4.0 2.8 2.7 1.4 2.6 3.4 3.3 ranG. £ f 27.2 30.3 10.6 17.1 25.8 4.7 4.3 41.8 4.6 3.8 1.1 4.0 3.1 2.9 61.0 15.3 21.9 2.7 11.6 16.4 1.0 2.1 30.1 0.8 0.9 0.0 0.8 1.0 1.0 46.8

*

CM Boston 2.54 Cvl CVJ O • mean 26.5 50.3 4.9 15.7 20.6 2*8 • 34.1 2.2 1.5 1.9 1.6 1.6 49.1 S.D. 2.1 2.8 1.1 2.3 2.2 4.3 4.1 ranSe £ f 30.4 35.2 7.3 19.4 24.1 3.7 3.1 39.2 4.2 3.0 1.0 3.9 2.6 2.7 54.7 24.0 26.7 2.6 12.0 18.6 1.8 1.1 25.4 1.0 0.5 0.0 0.8 0.8 0.9 45.5 Opaques Non-opaques

Percent Heavy Minerals 0 © (3 0 •O © 0 P o a © m a P *H Q Garnets P ® - p •H H 0 Till h H •rl « -P 0 O 0 P 0 ft u fe) XI 0 a C PO rH & E3 M < a a) -rl a) a ® P ® •rl ■H a a e aj +3 S3 v 43 P H u 0 P bO ■H p Unit 2 w eb o •H r l o O o ft 4 O 0 a P ft o K s a a $ CO oi

lOQr

+» 0 0) 30- ©

Darby X Caesar Boston Rainsboro Till Unit

H Miscellaneous non-opaques | | Hornblende

| | Garnets

I'm Opaques

Figure 33. Bar graphs of the heavy minerals of Ross County till units. 111

Till can be distinguished, from the Late V/isconsin tills* Rainsboro Till contains 28 to 37 percent more heavy minerals in the same size fraction than the younger tills* This difference may reflect source area variations or concentration of resistant heavy minerals after removal of more mobile constituents by weathering. Variations in ice-flow direction and provenance areas between the Erie and Huron Lobes were indicated by differences in heavy-mineral composition of tills in southern Ontario (Dreiraanis et al*f 1937)* Additionally, comparison of heavy-mineral composition between successive recessional moraines indicated shifting of Erie Lobe ice flow to more easterly trends during deglaciation.

Til1-Fabric Analyses

Many recent studies have shown the utility of till fabrics in determining direction of glacier motion (Lineback, 197*1; Ramsden and Westgate, 1971; Evenson, 1971)* Use of till fabric as a directional indicator is based on the presupposition that preferred alignment of the long axes of pebbles corresponds to the direction of ice motion* There remains uncertainty as to whether the directional component is imparted during transport of the pebbles within the ice or at the time of till emplacement by active ice. In well-developed ("strong") till fabrics (e.g. basal tills of Drake, 1968) the majority of pebbles 112

tend to plunge upstream toward the ice source. Some workers question the validity of till fabrics in representing direction of ice motion during active glacial erosion because the till was deposited during a depositional phase of glaciation (Young, 1962). However, most strong till fabrics parallel other local indicators of ice motion direction (e.g. striae, drumlins, boulder fans). Thus the relationship between fabrics and the direction of ice movement is too consistent to be fortuitous. Thirty till fabrics were measured in the four Ross County till units. Results of these analyses are plotted as "rose" diagrams in Figures 34-, 35, and 36. The well- developed nature of the till fabrics and the number of striated pebbles indicates that all Ross County till is probably basal (lodgment) till. The anomalous lack of ablation drift is unexplainable. Although criteria have been established to differentiate between basal and ablation till (Drake, 1968), no ablation till has been identified with certainty in Ohio. Till fabrics in the Darby I Till (Figure 34*) give a consistent direction of ice movement from the north- northwest. Four fabrics were measured at six-foot intervals at the Dry Run till section (sample locations 4-7, 50, 53, and 56 of Figure 12). Similar till fabric in the upper portion of the section (locations 47, 50, Illinoian glacial boundary Wisconsin glacial boundary -— Reeseville Moraine distal boundary Yellowbud Moraine

Figure 34. Rose diagrams of till fabric of Darby I Till in Ross County. Sample location number at top of diagram; number of pebbles measured at bottom. 114

**-»•- Illinoian glacial boundary Wisconsin glacial boundary Reeseville Moraine distal boundary E H Lattaville Moraine

Figure 35. Rose diagrams of till fabric of Caesar Till in Ross County. Sample location number at top of diagram; number of pebbles measured at bottom. Figure 36. Figure

NI«Mm4 Cm, il (samplesTill 7, 8,58) and in Ross County. Hosediagrams oftill fabric ofBoston Till HeesevilleMoraine distal boundary Wisconsinglacial boundary Illinoianglacial boundary number of pebblesmeasured at bottom. Samplelocation number at oftop diagram; (samples4, 31 , 39, 61, and and62) Rainsboro

115

116

and 53) indicate that only one till is present in this part of the section and that the Darby 1 Till has a homogeneous pebble orientation both areally and vertically within the till unit. Till fabrics at sample locations 17* 38* and 56 (base of the Dry Run section) have a secondary north-northeasterly-trending axis which indi­ cates a slight variation in ice-flow direction and possibly a different till unit. These three fabrics were measured in the basal portions of exposed sections. Their orientation is more closely related to Caesar Till fabrics (Figure 55) and therefore may represent exposures of Caesar Till or an older Darby Till unit. Twelve Caesar Till fabrics (Figure 53) show a general north-to-south direction of ice movement, with an increase in an easterly directional component from east to west in the county. This slight variation may be due to the influence of the escarpment of Mississippian bedrock that generally borders the southern margin of the Caesar Till. Since the escarpment trends northeast-southwest across Ross County, the ice sheet was able to penetrate to a more southerly position in western Ross County and Highland County, thus facilitating some eastward expansion of the Scioto Sublobe in this region. Three similar fabrics (locations 30, 81, and 82) frora the three till units at the Anderson Run section (Figure 11) reinforce the idea that the three till units 117

are all Caesar Till deposited furing minor oscillation of the ice margin. Till fabric in the Boston Till (Knockemstiff Mor­ aine) (Figure 36) illustrates strong topographic control of ice motion. This control hinders delineation of bed­ rock source materials which could account for anomalous carbonate content, distribution of heavy minerals, and pebble-count lithologies in this till unit. Boston Till is primarily distributed in low-relief deposits in north- south trending tributary valleys on the southern margin of the Paint Creek valley. All five till fabrics are aligned with the trend of the valley in which the Boston Till is exposed. The two fabrics that were measured nearest the mouth of tributary valleys (locations 31 and 39) have more of an indication of southwesterly ice motion than the other fabrics. This suggests that the ice which deposited the Boston Till came primarily from the northeast, entering the Paint Creek valley in the vicinity of Alum Cliffs. The three Rainsboro Till fabrics (Figure 36) indi­ cate a general north-south ice motion during Illinoian glaciation. Fabrics seven and eight were measured in two areas of the Rainsboro Till unit at the Kassieville site (Figure 26). Although some variation does exist, the general direction of ice movement indicated by the till fabrics is nearly identical. 118

Radiocarbon Darin-

Late Wisconsin ice invaded Ohio at Cleveland about 24,600 years B.P. Positions of ice sheet maxima were occupied nonsynchronously between the Miami Sublobe (20,500 years B.P,; W-304) and Scioto Sublobe (18,500 years B.F.; Y-448). Many radiocarbon dates in conjunction with the overlapping pattern of end moraines (Figure 37) (eastern Warren County and western Clinton County) indi­ cate the Miami Sublobe reached its terminal position about 1400 years prior to the Scioto Sublobe (Goldthwait, 1959). Dates on buried wood from pre-Cuba Moraine till in the Todd Fork valley of Clinton County indicate occupation of the Scioto Sublobe maximum position about 21,350 years ago (average of OWU-159, 0WU-160, D-46, D-47), roughly 850 years before the Miami Sublobe maxima. Teller (1964) indicates that the radiocarbon dates from the Todd Fork valley (21,137 * 1*^35 years B.P. - OWU-159 and 22,255 * 1,652 years B.P. - 0WU-160) probably date advance to the stand at the Vandervort Moraine in Clinton County. The Boston Till (Mt. Olive Moraine) of Highland County, considered to be "early" Wisconsin in reconnaissance investigations from 1956 to 1969, has been proven to be early Late Wisconsin (Woodfordian) by radio­ carbon dating (Rosengreen, 1970). A date of 21,080 ± 200 I1linoian boundary Pickaway Co Wisconsin boundary Montgomery

7 . 3 4 0 1 C linton Co. •j 1 7 , 2 9 2 * 4 3 6 16.590±5 (CWR-190 1 8 . 0 0 0 A 0 0 J 9 6 5 0 1 41 CWfl- . 0 5 0 * 4 0 0 1,137+14 ( W - 9 1 ) tO W U '15 7 6 9 0 * 2 2 4 ( O W U ' 5 2 ) 20.500*900 20910*240 17.960*400 ( W - 3 0 4 ) 3 5 6 0 0 ( O W U - 3 3 1 1 (V-448) ( W - 7 7 3 )

H ighlandClermont Co Brown Co I H ighlandClermont End Moraines

I IMarcy ¥ X S \ Glendon Mt. Olive H | 3 Yellowbud kWV Reeseville Vandervort B&B Bloomingburg fctEEH Cuba I o I Lattaville III I If Esboro gggl Hartwell Figure 37* Map of the southern Scioto Sublobe showing the location of end moraines and radiocarbon dates (years before present)* 120 years B.P. (D-46) was obtained from wood fragments near the base of the Boston Till while the basal portion of the overlying till unit was dated at 20,910 ± 200 years B.P. (D-47). These age determinations clearly indicate the Boston Till was deposited during the initial Late Wisconsin ice advance. There are no radiocarbon dates from Ross County analogous to this earliest Late Wisconsin glaciation, although the youngest drift in the Paint Creek valley has many of the physical and compositional characteristics of the Boston Till. Note that the end moraine trend in western Clinton County is overlapped perpendicularly by the Hartwell Moraine of the Miami Sublobe in northwestern Warren County (Figure 37). Eastward in Clinton County, the Vandervort Moraine is overlapped by the Cuba Moraine. The Cuba Moraine in turn overlaps the Mt. Olive Moraine in the vicinity of the Highland-Clinton County boundary. This pattern suggests that the drift of the Vandervort and Mt. Olive Moraines may be correlative (Rosengreen, 1970). Grouping of radiocarbon dates from Caesar Till of the Cuba Moraine (Highland and Clinton Counties) and the correlative Lattaville Moraine (Ross County) presents an enigmatic bimodal distribution of age determinations (Table 15). Two explanations can account for this anomalous distribution of radiocarbon ages. The group of three older dates may indicate the time at which the 121

TABLE 15 RADIOCARBON BATES PROM THE CAESAR TILL 0? THE CUBA- LATTAVILLE HORAINE OF THE SOUTHERN SCIOTO SUBLOBE Age (years Lab Ave. Age of County before Dresent Desiemation GrouninK Ross 19,650 ± 419 DIC-200 Highland 20,460 * 700 W-2459 20,182 ± 550 20,820 ± 600 W—2465 Clinton 19,800 ± 400 1-4795 Ross 18,050 ± 400 W-91 17,980 ± 400 ¥-331 18,000 ± 400 OWU-331 18,082 ± 365 17,880 ± 224 0WU-52 Clinton 18,500 ± 400 T-448

trees were overridden in central Ohio and incorporated into the till of the advancing ice sheet. In this explanation the younger group of dates would then indicate the incorporation of wood into till in the southern Scioto Sublobe area, shortly before formation of the Lattaville Moraine. This explanation is supported by the fact that many of the logs, which date in the 18,000 years B.P. range, retained their bark during transport. This is suggestive of a short distance between the points of incorporation and deposition. Alternately the timing of the maximum position may have varied between Highland and part of Clinton County and the rest of the southern Scioto Sublobe (Rosengreen, 1970). The early Late Wisconsin 25,500 ± 600 years B.P.

(I-4797) date from a log in the lower uill along Blinco 122

Branch in Highland County is probably inaccurate. This conclusion is drawn because the dated log v/as found in a till unit that correlates to the Illinoian Rainsboro Till by stratigraphy, clay mineralogy, pebble lithology, texture, and carbonate content. The till is overlain by ten feet of weathered gravel which contains a Sangamon paleosol in its upper portion. Boston Till of the Mt. Olive Moraine (dated at about 21,000 years B.P.) overlies the buried soil. Thus if the date is accurate, the paleo­ sol would have to have formed within a 4,000 year interval during early Late Wisconsin. Rosengreen (1970) considers this time span to be insufficient to develop the exposed weathering profile. Two new radiocarbon dates from Caesar Till at the Anderson Run site (sample locations 80 and 81) were obtained during this study (Table 16). Unfortunately these dates do little to clarify dating of the Lattaville- Cuba Moraine position. The accuracy of at least the younger date is questionable because the older radiocarbon date, 19*050 * 419 years B.P. (DIC-200), is found stratigraphically above the younger date of 16,590 ± 570 years B.P. (CWR-190). The older date (DIC-200) was determined from a group of small wood (Picea) frag­ ments. If this date is valid, it is evidence of till deposition in the outer Lattaville (Cuba) Moraine corresponding to the earlier Cuba Moraine dates from TABLE 16 RADIOCARBON AGE DETERMINATIONS FROM ROSS COUNTY Stratigraphic Material Age (years Lab Unit Location Dated before nresent) Designation interglacial Humboldt, peat peat Paint Twp. 55,000 VJ-773 Bier's Run Picea 18,050 ± 400 W-91 Caesar Till (sample Picea 18,000 ± 400 0WU-331 location 6) Picea 17,880 ± 224 0WU-52 Anderson Run (Stop 13A, Caesar Till Goldthwait, Picea 17,980 ± 400 W-331 1955) NWJC, Section Caesar Till (?) 11, Colerain Picea 17,292 ± 436 OWU-76 Township Anderson Run (sample Picea 16,590 ± 570 CWR-190 Caesar Till locations 80 Picea 19,650 ± 419 DIC-200 and 81) 0WU-26OA Hallsville, Picea 12,855 ± 275 extensive Sections 7 xn 12,685 * 244 0WU-260B marl and 8 of marl 13,695 ± 520 0WU-2G0C Colerain Twp. 13,180 ± 520 0WU-220 ♦Laboratory abbreviations: OWU - Ohio Wesleyan University, W - U.S.G.S. Washington, CWR - Case Western Reserve University, DIC - Dicar Corporation 124

Highland and Clinton Counties (W-2459, W-2465, 1-4795). Thus the idea of non-synchroneity in establishment of the maximum positions of the Cuba-Lattaville Moraine (Rosengreen, 1970) would be refuted. The younger date (CtfR-190) was from a large spruce log near stream level in the lower of the three till units at the Anderson Run section. The horizon containing the log was saturated by ground water seepage. The date indicates a post-Reeseville Moraine age (i.e. less than 17•200 years B.P.) for the till unit. All stratigraphic, textural, and compositional evidence indicates that the till is Caesar Till which is definitely pre-Reeseville Moraine age. Additionally, the dated till unit can be physically traced downstream to the location (Stop 13A, Goldthwait, 1955) where a log in the same till unit was dated at 17,980 ± 400 years B.P. (W-331). Therefore, this new date appears to have been derived from a sample which was contaminated, most likely by addition of recent carbon by ground water. Peat from two buried organic-silt zones in north­ western Fayette County date at 17*540 ± 390 years B.P. (0WU-25S) and 19,755 * 475 years B.P. (OVU-257) (Moos, 1970). Presence of these two buried silt zones indi­ cates two periods of significant, yet short duration, retreat of the ice sheet during active loess deposition; 1) after deposition of the Knockemstiff Moraine (Boston 125

Jill) and prior to formation of the Lattaville Moraine and 2) post-Lattaville Moraine formation but preceding the readvance to the Reeseville Moraine position. The date of 17*292 ± 4-75 years B.P. (OWU-76) from near-surface till in the Lattaville Moraine near Halls- ville (Colerain Township) is thought to be associated with the readvance to the Farmersville-ReeSeville Moraine position (Dreimanis and Goldthwait, 1973). How­ ever, the local stratigraphy, loess distribution, and compositional and textural parameters of the till suggests the unit dated is Caesar Till, not Darby I Till as the radiocarbon age indicates. Since other radiocarbon dates (Figure 37) indicate the Caesar Till and Darby 1 Till were deposited only about 800 years apart, it appears likely that the near-surface spruce fragments that yielded the Reeseville-age date (OWU-76) were contaminated by rootlets or humic wates leached from nearby pastures. The dates on interglacial peat (W-773) and an extensive marl accumulation near Hallsville (0WU-260A, 260B, 260C, and 0WU-220) (Table 16) are associated with glacial and post-glacial lakes. The significance of these dates will be discussed under "Lacustrine Deposits" in the following chapter. Chapter IV

DESCRIPTION OP GLACIAL DEPOSITS AND FEATURES

"The area over which the Drift is spread in Ohio corresponds in a general way with the area of glaciation, but through the action of icebergs, which in the last great submergence seem to have carried their freight, in some instances, beyond the points reached by the glaciers, and especially by the action of local currents of water which flowed down through certain great lines of drainage, the Drift materials have been bourne far beyond the line I have indicated as bounding the erosive action of the ice sheet." -Newberry (IS?1*')

Illinoian Margin and Glacial Boundary

The limit of Illinoian glaciation transects Ross County from northeast to southwest, generally paralleling the bedrock escarpment of the Appalachian Plateau (Plate I, in pocket)* Penetration of a small sublobe of Illinoian ice into the Teays River valley in the northern portion of the D-ring resulted in the only major modifi­ cation of this trend* Throughout Ross County there is no marginal, constructional drift accumulation (end moraine) which delineates the maximum position of Illinoian ice. Com­ plete post-Illinoian removal of a possible Illinoian end moraine is unlikely* Illinoian ice which advanced onto 126 the Appalachian Plateau south of the Paint Creek valley was thin because relatively low-relief (less than 100 feet) bedrock protruberanees on the upland had a pro­ nounced channelling effect on the ice sheet. Minor crenulations in the glacial boundary (Plate I) attest to this local topographic control of Illinoian ice movement. A thin ice mass on an upland surface would be susceptible to relatively rapid removal during early stages of deglaciation. The combination of a thin ice mass and a short duration of occupancy of terminal position would inhibit construction of any significant morainic form. The boundary between the unglaciated area and the area of well-defined drift is normally a transition zone which varies from a mean of several hundred yards to one-half mile in width. Since most of the plateau upland south of the Paint Creek valley is capped by dark shale, binocular examination of soil samples for quartz content has proven to be an effective tool in delineating the glacial boundary (Poster, 1950). In the Illinoian- glaciated area all samples contain greater than ten per­ cent quartz grains while the quartz content decreases to near zero in the unglaciated terrain. However, care must be used during this method because several small knobs on the upland are capped with 3erea sandstone which locally yields high concentrations of quartz grains in the unglaciated area. In addition to quartz grains, the 128 easily identifiable crystalline materials rapidly decrease in percentage across the transition zone from glaciated to unglaciated terrain. Care must be taken in defining the Illinoian boundary in areas where the terminal position is near divides of south-flowing streams. Pebble-size granitic erratics were found near the divide separating the Black Run and Pee Pee Creek drainage basin near Summithill. Similar erratics were noted in the upper portion of the Crooked Creek basin in south-central and southeastern Huntington Township. Poster (1950) reported cobble-size erratics in the upper reaches of the Sunfish Creek system in Scioto and Paxton Townships. These erratics pebbles and cobbles were probably carried beyond the area of actual glaciation by turbulent stream flow during periods of meltwater dis­ charge. The width of the area of the Appalachian Plateau in Ross County, which was glaciated during the Illinoian Stage, is much greater west of the Scioto River valley in comparison with the region east of the valley between Chillicothe and Adelphi (Plate I). This difference is due to more moderate relief at the escarpment margin and on the upland surface in western R03S County. In that portion of Ross County west of the Scioto Valley and south of the Paint Valley, the area of glaciated plateau significantly increases from west to east: indicating the influence of the Deep Stage Bourne­ ville Greek valley (present Paint Creer: valley) on Illinoian glaciation. Illinoian ice which advanced over the portion of the Appalachian Plateau north of the Bourneville Creek valley, then had to traverse the deeply- incised valley before glaciation of the upland south of the valley. This combination of high-relief obstacles inhibited the spread of the Illinoian ice onto the plateau south of the Bourneville Creek valley. The preponderance of ice-contact deposits near the Illinoian margin, suggesting thin stagnant ice, supports this assertion. In the area between Sulfur Lick and Chillicothe in central Goss County, the edge of the Appalachian Plateau is more of a gradual slope than a high-relief escarpment. Also the relief associated with the now-partially-buried Bourneville Creek valley was much less pronounced in this area compared to further southwest in Ross County. Thus the moderation in local relief had less of a barrier influence on the ice sheet which penetrated two to five miles onto the upland surface south of the Bourneville Creek valley. Cobble-size erratics on Jones Hill in extreme south­ western Ross County indicate that Illinoian ice overrode this plateau segment during its advance to near the southern margin of the present Beech Plats area. The Illinoian boundary along the eastern margin of 130

the Walnut Greek valley in east-central Ross County is established on the basis of sparsely distributed granitic erratics on the valley side above the level of Illinoian outwash. In southeastern Colerain Township, the Illinoian limit is defined by deposits of thin, patchy till and several cobble-size erratics. These deposits predominate in cols between bedrock knobs and on the north-facing slopes of these topographic “highs"• Consequently there is a marked reduction in valley and col relief between the drift-covered area and the unglaciated terrain.

Illinoian Ground Moraine

Illinoian ground moraine covers a northeast to south­ west , wedge-shaped zone of approximately 30 square miles through central Ross County (Plate I). The moraine is restricted to the Appalachian Plateau upland except for the drift accumulations in the northern portion of the D-ring. Illinoian drift is absent in the bottoms of the numerous valleys which transect the area of Illinoian glaciation. Ground moraine is patchy to absent on most valley slopes in Ross County. Exposed bedrock, rubble from mass-wasting, occasional erratics, and thin patches of drift form the surface materials on the valley sides. The Illinoian ground moraine is typically thin, patchy, and loess-covered to a mean depth of about 30 inches. The drift averages 3^ feet thick (average of data 1J1 fron 4-7 wacsr-iirell logs) with a range from effectively zero to 71 feet. Thinness of the ground moraine is attributed to a short duration of stand by -Che Illinoian ice near its maximum position. This is consistent with the lack of development of an end moraine near the glacial boundary. Leighton and Brophy (1961) suggested a short duration of the entire Illinoian glaciation on the basis of thin drift accumulation in central and southern Illinois. Exposures of Illinoian till are rare in Ross County. In the limited exposures, the depth of oxidation averages 12 feet (average of measurements at seven sites) with a range of nine to fourteen feet. Illinoian till in the oxidized zone is typically yellowish-brown (10YR5A) while the unoxidized till is dark gray (7.5Y34/1). The till is commonly highly-jointed with secondary clay accumulations often occurring in the lower portion of the oxidized zone along the Joint margins. Subglacial erosion was minimal during deposition of the ground moraine because augerings in thin-drift areas (e.g. Beath Sidge in west-central Twin Township; Farrel Hill in southeastern Paint and western Twin Townships) and the exposure along the Baltimore and Ohio railroad near Schooley (sample location 38) show that residual, highly-weathered zones in till, alluvium, or colluvium overlie the bedrock beneath the ground moraine. The 132 intesity (coloration) and thickness of the weathered zone suggests development during an interglacial (probably Yarraouthian) period. Lack of removal of these paleosol remnants suggests a significant diminishment of the erosive potential of the Illinoian ice sheet near the glacial boundary. The Illinoian ground moraine west of the Scioto River valley is typically planar due to the topographic in­ fluence of the easterly-sloping Appalachian Plateau on the thin drift sheet. Portions of the moraine on the upland take the form of small, poorly-drained till plains. East of the Scioto Valley the high-relief topography along the plateau margin causes the ground moraine to be very patchy with abrupt variations in drift thickness* The drift accumulations are typically thickest on north-facing slopes and in cols between bedrock knobs. Due to the erosional nature of the Appalachian Plateau margin in Ross County, several Illinoian-drift inliers occur on topographically-high plateau remnants within the area of Wisconsin glaciation (Plate I). This is the only area in Ohio where such inliers occur* The largest of these areas is the plateau segment north of the Paint Creek valley and south of the Lattaville Moraine. Water-well logs, augerings, and occasional erratics indi­ cate all of this upland was glaciated by Illinoian ice including the highest summits (about 1300 to 1333 feet 133 elevation) along Brenner Hill and Parrel Hill in south­ eastern Paint Township. Thin Illinoian ground moraine also occurs on several bedrock-cored, topographic "highs11 north of the distal boundary of the Lattaville Moraine. These areas were glaciated during the Illinoian Stage but were nunataks during the Wisconsin glaciation. On most of these drift- inliers the boundary between Illinoian and Wisconsin drift is distinct due to variations in: 1) loess thickness, 2) depth of leaching, and 3) weathering profiles. Sig­ nificant Illinoian-drift inliers occur 1.3 miles east of South Salem (Buckskin Township), one mile southeast of Fruitdale (Paint Township), along a northwest-southeast zone in western Paint Township, along the boundary between Buckskin and Concord Townships, and 1.1 miles south and 2*3 miles southwest of Hallsville in northeastern Boss County. Cobble and pebble-size erratics, marking the former presence of thin ground moraine, were found on the summit flats of Sugarloaf Mountain (Sections 32 and 33t Green Township), Bunker Hill (Section 8, Springfield Township), Mount Logan (Sections 16 and 17* Springfield Township), Rattlesnake Knob (Section 8, Harrison Township), and several unnamed summits in the southeastern one-quarter of Colerain Township and the northeastern portion of Springfield Township. These erratics prove that even the 134 highest areas of the plateau margin within the Illinoian glacial boundary were ice covered during the Illinoian Stage.

Late Wisconsin Moraines

Late Wisconsin (Woodfordian) drift covers the north­ western forty percent of Ross County and small areas of tributary valleys to the Paint Creek valley (Plate X). This drift was deposited in association with four end moraine positions. Only the Reeseville Moraine position lacks a continuous belt of hummocky end moraine topography to delineate the ice margin position. Each end moraine position represents a significant advance, readvance, or recessional stand of the ice sheet and, except for the Yellowbud Moraine, is associated with a separate till unit.

Knockemstiff Moraine

Drift associated with the initial Late Wisconsin ice advance is exposed only in valleys tributary to the Paint Creek valley in southwestern Ross County. In small valleys south of the Paint Valley, extending eight miles northeast from Jimtown Hollow (1.3 miles southeast of Bainbridge) to the Black Run valley, are fifteen scat­ tered, snail areas of hunmocky topography forming a low- relief Late Wisconsin end moraine. The end moraine 155

segments are generally elongated north-south, paralleling the general trends of the enclosing valleys. The total area of all the end moraine patches is approximately 1.5 square miles with individual segments ranging up to 0.25 square mile. This end moraine is herein named the Knockemstiff Moraine after the small hamlet of the same name on the eastern margin of the Black 3un valley in extreme western Huntington Township. Elevations on the end moraine range from 845 feet in Jimtown Hollow to ahout 700 feet at several localities, while mean local relief is approximately 80 feet. Elevations on the Appalachian Plateau south of the end moraine generally decrease eastward from 1200 to 900 feet with relief above the end moraine ranging from 200 to 400 feet. The end moraine commonly borders higher Illinoian ice-contact deposits which range in elevation up to 950 feet. The boundary between the moraine and ice-contact deposits i3 very distinct because of: 1) a topographic break in slope, 2) variation in loess thickness and depth of carbonate leaching, and 3) differences in soil profiles associated with changes in parent materials. In all areas the local relief of the kames is at least double that of the neighboring, lower-elevation Knockemstiff Moraine segments. In Jimtown Hollow and the valley northeast of Sulfur Lick Plat, Knockemstiff Moraine segments occur south of 136

the Illinoian ice-contact deposits in anomalous positions. The combination of low local relief, low end moraine elevation, and valley location of the end moraine areas indicates deposition by a thin-ice, topographically- controlled, one-mile-wide sublobe which extended south­ westerly up the Paint Creek valley. Deposition from a southwesterly-advancing ice mass is also supported by till fabrics and by the predominance of highest elevations and thickest drift accumulations along the western margins of the tributary alcoves and valleys. The Knockemstiff Moraine is composed of a texturally, compositionally, and areally distinct till which is correlated with the Boston Till (Rosengreen, 1970) of Highland County, If this is a valid correlation, the Mt. Olive Moraine of the western Scioto Sublobe correlates to the Knockemstiff Moraine of Ross County, Radiocarbon dates in Highland County (Pigure 31) indicate that these moraines were constructed approximately 21,000 years B,P. Data from 17 water-well logs indicates a mean thick­ ness of 31 feet for the end moraine drift with thicknesses ranging from 13 to 78 feet. These logs indicate that several moraine segments are cored with weathered sand and gravel, presumably Illinoian ice-contact deposits. Thus the hummocky topography of the moraine may be due as much to the underlying sand and gravel as the thin Boston Till cover. 157

Lattaville Moraine

Hummocky, Late Wisconsin drift forms a nearly-contin­ uous, 0.5 to 5*5 mile-wide end moraine which generally parallels the northeast-southwest-trending Appalachian Plateau escarpment from Humboldt (Faint Township of west- central Ross County) to Adelphi (Colerain Township of northeastern Ross County). Major discontinuities in the end moraine result from erosion and Wisconsin outwash deposition through the morainic topography in the North Fork valley near Slate Mills and in the Scioto River valley west of Hopetown. Till stratigraphy and areal position indicate that this moraine is correlative with the Inner Cuba, Outer Cuba, Wilmington, and a portion of the Reeseville mo­ raines of Highland and Clinton Counties. These four end moraines converge by topographic tracing in eastern High­ land and western Ross Counties. Separate moraine elements associated with these end moraines can not be differentiated on the basis of moraine crests, boulder concentrations, or till stratigraphy within the single major Ross County end moraine. Therefore, this large composite end moraine, bordering the plateau escarpment in Ross County, is herein named the Latcaville Moraine after the small village on the moraine in south-central Concord Township. 138

Laboratory analyses indicate that most of the Latta- ville Moraine drift is Caesar Till. This till is compositionally and texturally indistinguishable from the Rainsboro Till, capping the plateau upland south of the end moraine, and the Darby I Till which forms Wisconsin ground moraine north of the moraine. Radiocarbon dates on wood in Caesar Till at Bier's Run (sample location 6) and Anderson Run (sample location 80) indicate that much of the Lattaville Moraine was constructed about 18,000 years B.F. Fedologic criteria indicate that thin Darby I Till forms the surface unit in a narrow, east-west zone along the proximal margin of the Lattaville Moraine, 1.3 miles south and southeast of Roxabell (south-central Concord Township). This is one of the few areas of Ross County where drift associated with the readvance to the Reese- ville Moraine (circe 17,200 years B.P.) forms hummocky end moraine topography. The segment of the Lattaville Moraine west of the Scioto Valley consists of a 0.3 to 3.0 mile-wide, 18 mile- long, northerly-convex, crescentic area along the Appalachian Plateau escarpment from Humboldt to North Fork Village. The hummocky end moraine is continuous except in the area west of Slate Mills where Wisconsin outwash and ice-contact deposits transect the morainic topography. The ice-contact deposits in this area that are covered by thin till probably formed contemporaneously with the 159

Lattaville Moraine. Postglacial erosion on some of the kanes has removed much of the thin till cap, exposing ice-contact sand and gravel at the surface. Although mapped as till-over-ice-contact deposits (Plate I), portions of this region are topographically indistinguish­ able from the bordering end moraine and can be considered kame moraine. Prom Humboldt to Just south of Lattaville, the clearly-defined distal boundary of the Lattaville Moraine closely parallels the high-relief escarpment margin. Moderation in the slope of the plateau edge east of Latta­ ville allowed the Late Wisconsin ice sheet to advance onto the northern margin of the plateau along Plyley Ridge. The distal boundary of the end moraine on the up­ land is a narrow diffuse zone rather than a sharply- defined limit* The distal margin on the ridge is delin­ eated on the basis of: 1) variations in thickness of loess cover and depth of carbonate leaching, 2) augering3 in the thin-drift edge of the moraine which penetrate through Wisconsin till into Sangamon paleosols, and 3) small, sub­ dued ridge segments of Late Wisconsin till (Caesar Till) near the glacial boundary. This Late Wisconsin-Illinoian boundary is most clearly defined by these criteria 0.2 mile south of the Junction of Ohio Route 28 and Poplar Ridge Road (0.8 mile southeast of Lattaville).

The portion of the Lattaville Moraine between Sulfur 140

Lick and Worth Fork Village occupies a topographic basin and has more subdued topography and lower local relief than the remiander of the end moraine* Although less distinctive as end moraine than the rest of the Latta­ ville Moraine, this area is topographically differenti­ able from the bordering Late Wisconsin ground moraine* Mean drift thickness for the Lattaville Moraine west of the Scioto Valley is 69 feet (based on data from 37 water-well logs) with a range from 26 to 210 feet. Thinnest drift occurs in areas of bedrock "highs” such as on Flyley Ridge in southeastern Concord Township and in the northwest-southeast-trending region through Harper where the end moraine attains its maximum width* Thick­ est drift accumulations are in the area between Lattaville and Mussellman where several wells penetrate more than 130 feet of till and sand and gravel. Wood fragments are logged in seven wells near Lattaville at depths ranging from 13 to 46 feet* Except for sand and gravel zones near the ice-contact deposits, most of the moraine drift is till* Elevations on this end moraine segment range up to a maximum of 1120 feet on the thin-drift cover on the upland east of Lattaville. West of Lattaville the higher bordering upland is 100 to 200 feet above the end moraine. The Lattaville Moraine rises 300 to 400 feet above the Darby I Till ground moraine to the north. 141

Illinoian-drift inliers are found within the Latta­ ville Moraine one mile southeast of Fruitdale and 1.2 miles east of South Salem. Late Wisconsin ice-contact deposits interrupt the morainic topography near Lattaville and on the southwestern margin of the North Fork valley opposite Sulfur Lick. East of the Scioto Valley the Lattaville Moraine is a 14 mile-long, up to 3*5 mile-wide, area of well- developed hummocky topography bordering the north-facing escarpment of the plateau (Figure 38). The end moraine is continuous except for a small, isolated area on the east side of the Scioto Valley west of Mount Eyes and two transecting stream valleys; South Fork (Kinnikinnick Creek)-Dry Run valley (Sections 14, 15, 23, and 26 of Green Township) and Bull Creek valley (Sections 2, 3, 10, 11, and 14 of Colerain Township). Large areas of ice-contact deposits are included in the end moraine from 1.5 miles west of Hallsville east to Adelphi and in the region north of Sugarloaf Mountain* Additionally, water-well logs indicate a much greater portion of the drift is sand and gravel in this segment of the end moraine as compared to the moraine west of the Scioto Valley. In this portion of the Lattaville Moraine drift thickness averages 79 feet (data from 48 water-well logs) with a range of 25 to 119 feet, generally increasing in Figure 58. A portion of the Lattaville Moraine bordering the Illinoian-glaciated Appalachian Plateau escarpment and upland (higher wooded areas). View east-northeast from south-central Section 26, Green Township, 4.1 miles south­ west of Hallsville. 143

thickness toward the distal margin of the moraine* Many sand and gravel zones (one to 25 feet thick) are included in the drift. Only wells within one mile of the southern margin of the end moraine penetrate to bedrock (at depths of 25 to 80 feet) indicating that the escarpment margin slopes steeply northward to depth beneath the Lattaville Moraine. 'The end moraine in this region lies 100 to 300 feet below the Illinoian-glaciated summits of the upland south of the Wisconsin boundary and 50 to 300 feet above the lacustrine plain and ground moraine which border the end moraine to the north. Elevations on the end moraine generally increase from the proximal to distal sides of the moraine. Maximum elevation of 1040 feet occurs south­ west of Hallsville near the southern moraine boundary. Evidence of penecontemporaneous deformation in association with construction of the end moraine is commonly found in the numerous, small borrow pits along the northern margin of the Lattaville Moraine. The deformational features include small-scale folds, faults, and various slump and flow features.

Heeseville Moraine

The distal boundary of drift associated with the readvance of the Scioto Sublobe to the Reeseville Moraine is defined on pedologic criteria, because, except for a 144 small area of the Lattaville Moraine south of Roxabell and the till-over-ice-contact deposits east of the Scioto Valley, there is no topographic expression of the drift margin (Plate I). The Reeseville Moraine is a well-defined, topo- graphically-high, accumulation of drift throughout Greene, Clinton, and Highland counties. The distal boundary of the end moraine in these counties is established on the basis of topography and abrupt changes in thickness of loess and depth of carbonate leaching. Although the end moraine is absent throughout most of Ross County, the variations in loess cover and depth of leaching provide tools for defining the extension of the Reeseville Moraine drift-sheet (Darby I Till) boundary. Even though the "soils break" between the Miami 6A and 60 soils is well- defined areally, the boundary separating these two soil groups in Ross County is not a precise line but is a transitional zone. North of the boundary is the area that generally lacks a loess cover and has shallower depths of carbonate leaching compared to the loess-covered soils south of the zone. Since the Darby I Till, which was deposited during thi3 readvance, primarily forms ground moraine in Ross County, the elevation, thickness, and surface expression of this till unit will be discussed later in this chapter under "Ground Moraine". Four areas of ice-contact deposits on the east side of the Scioto Valley, extending north from Hopetown to the Pickaway County boundary, are at least partially covered with loess-free Darby I Till. These areas are mapped as Reeseville Moraine (Plate I). The thin till cover has been removed in some areas by postglacial erosion, expos­ ing ice-contact sand and gravel. Loess-covered Kingston Outwash and ice-contact deposits, which border the eastern margin of these till-over-ice-contact deposits, lack a till cover. This areal relationship of the Darby I Till strongly supports the idea that the Reeseville- age radiocarbon date (17,292 * 436; OWU-76) from the Lattaville Moraine east of Hallsville is from a contam­ inated sample. In order to have deposited till east of Hallsville on the end moraine, the Beeseville-age ice would have had to advance over the older Kingston Outwash and ice-contact deposits and almost certainly would have deposited till somewhere on these deposits. There are no till remnants or indications of till removal on the surface of the outwash or ice-contact features. There­ fore, the eastern limit of Reeseville-age drift is roughly along a north-south line from Hopetown to the Pickaway County line* The location of drift associated with the Reeseville Moraine position in Ross County (Plate I) indicates strong topographic control on the readvancing ice sheet. 146

Southernmost penetration of the ice mass wa3 in those areas of lowest local relief (e.g. Frankfort-Poxabell area and along the western margin of Ross County near Green­ field) . The lack of hummocky drift accumulation along most of the Darby I Till margin in the Ross County region is anomalous because of the significant Reeseville Moraine drift accumulations elsewhere in the southern Scioto Sublobe (Rosengreen, 1970; Teller, 1964; Quinn, 1972). This anomaly may be related to large-scale deflection of the Scioto Sublobe ice-motion by the bedrock escarpment of the Appalachian Plateau. The Glacial Map of Ohio (Goldthwait et al«, 1961) shows that maximum development of end moraines in the Scioto Sublobe occurred in the southwestern portion of the sublobe. Each end moraine becomes less pronounced and more poorly defined toward the southeastern area of the sublobe. The bedrock escarpment trends roughly north-south from northern Ohio south to eastern Fairfield County but then changes to a northeast-southwest orientation from Lancaster southwest through Ross County. This linear orientation of the confining bedrock "high" may have caused a southwesterly deflection of ice movement in the southern portion of the Scioto Sublobe. Primary ice-flow direction would also coincide with the main trend of drift transport accounting for the much larger morainic 147 accumulations in the southwestern portion of the sublobe. 'fill fabrics and striae in Ross, Highland (Rosengreen, 1970), and Clinton (Teller, 1964) counties neither sub­ stantiate nor deny this idea.

Yellowbud Moraine

Bordering Pickaway County in north-central Ross County is a 6.2 square mile area of hummocky end moraine. The end moraine is bounded on the southwest and east by Late Wisconsin outwash in the Deer Creek and Scioto River valleys* The end moraine extends northward into Pickaway County where it grades into ground moraine. This drift accumulation is herein named the Yellowbud Moraine after a small village along the eastern margin of the moraine. Although the moraine topography is similar through­ out the Yellowbud Moraine, the end moraine is composed of two distinct moraine types. The southeastern one-third of the end moraine is kame moraine with the rolling topography developed on ice-contact sand and gravel. Four water-well logs in this area indicate thicknesses of sand and gravel vary from 53 to 106 feet. The north­ western portion of the Yellowbud Moraine is typical till- covered (Darby I Till) end moraine. The boundary between these two end moraine types along Swaney Road is clearly defined on the basis of: 1) a topographic break (relief five to fifteen feet) in some areas and 2) augerings 148 which easily delineate the till/sand and gravel boundary. Average till thickness in the northwestern portion of the end moraine is 28 feet (data from 16 water-well logs) with a range of 10 to 41 feet of "clay" (till). This till overlies thick sand and gravel accumulations which seem to be continuous with the surface deposits in the kame moraine area. Lack of any drainage channels or other indications of surface erosion coupled with the similarity in topography between the end moraine types indicates that no till was ever deposited and later removed from the kame moraine surface. The relationship of the Yellowbud Moraine to the Glendon, Esboro, and Bloomingburg moraines of the southern Scioto Sublobe is problematic (Figure 39)• Correlation of the end moraine is hindered by: 1) lack of morainic topography in northwestern Ross and southwestern Pickaway Counties, 2) no radiocarbon dates on the post-Reeseville moraines in the southern Scioto Sublobe, and 5) the similarity in drift (Darby I Till) composing each of the end moraines. Areal position clearly indicates that the Yellowbud Koraine was constructed by the ice sheet in conjunction with a significant short halt of the ice margin during recession from the Reeseville Moraine. A key to possible correlation of the Yellowbud Moraine is the location of Circleville Outwash valley train remnants in the Scioto River and Deer Creek valleys Madison do* Pickaway Co

ica margin during C ire la villa Outwasl daooaition

(C* location ot I Circlavilla OutwashJ tarracas) I.

Figure 39. Hap of the southern Scioto Sublobe showing the relationship between the Yellowbud Moraine. Circleville Outwash, and Reeseville and later end moraines. 150

(Figure 39). Circleville Outwash is definitely traced to the distal margin of the Marcy Moraine at Circleville (Kempton and Goldthwait, 1959). It defines one ice margin in that area, Circleville terraces are limited to the Ross County portion of the lower Deer Creek valley indicating that the Scioto Sublobe margin was near the Ross-Pickaway County boundary west of the Yellowbud Moraine during the deposition of the outwash, A "Silt Line" and related parent material variations define the distal boundary of the Reeseville Moraine position along the eaBtem edge of the Marcy Moraine east of Circleville. Therefore, the Circleville Outwash has been correlated to the Reeseville Moraine time (circa 17,200 years B.P.) (Kempton and Goldthwait, 1959). This relationship indicates that the Circleville Outwash was deposited synchronously with the Mad River Outwash of the Miami River basin (Quinn, 1972). However, the ice-margin position associated with the Circleville Outwash and Yellowbud Moraine in north-central Ross County is clearly a recessional position after the Reeseville maximum and is therefore definitely post-Reeseville Moraine/Mad River Outwash age. Although the distal margin of the Marcy Moraine is Reeseville Moraine equivalent, the ice-margin position at the Marcy Moraine during deposition of the Circleville Outwash is post-Reeseville and probably correlates with 151

one or more of tie Glendon, Esboro, or 31ooningburg moraines. Thus during the period of general retreat of the ice margin and formation of the Glendon, Esboro, and Bloomingburg moraines in the southwestern portion of the sublobe, the ice margin in the southeastern part of the region remained at or near the Marcy Moraine. Variable rates of retreat within the Scioto Sublobe may be re­ lated to the suggested deflection of ice flow by the Appalachian Plateau escarpment. In the area of most extensive ice movement (southwestern Scioto Sublobe), end moraines formed during halts in relatively rapid reces­ sion, while in the area nearer the bedrock escarpment general stagnation occurred as is evidenced by deposition of outwash, absence of large end moraines, and abundance of ice-contact deposits. Although small areas of outwash associated with the Glendon Moraine have been mapped in the Paint and Walnut Creek valleys in Highland County (Rosengreen, 1970), there is no post-Reeseville Moraine outwash in the North Pork or Paint Creek valleys in Ross County. Since these valleys head near the Glendon, Esboro, and Bloomingburg moraines, it appears that little outwash deposition occurred in conjunction with their construction. Although the evidence is not conclusive, the Yellow­ bud Moraine is correlated to the post-Reeseville portion of the Marcy Moraine and the Glendon, Esboro, and 152

Bloomingburg moraines of the southwestern portion of the Scioto Sublobe. The small size of the Yellowbud Moraine in comparison to the correlative end moraines is attri­ buted to its areal position, which was lateral to the primary ice-flow and drift-supply direction in the sub­ lobe*

Ground Moraine

Late Wisconsin ground moraine covers most of the north-western JO percent of Ross County (Plate I). The ground moraine has generally a low-relief, nearly planar topography except in those areas where thin drift over­ lies topographic-bedrock "highs" and postglacial stream erosion has created up to 150 feet of local relief. Caesar Till ground moraine extends in a one to three mile-wide zone from the proximal margin of the Lattaville Moraine to the edge of the Darby I Till drift sheet. This ground moraine unit is divided into three separate areas by the Reeseville boundary penetration to the Lattaville Moraine south of Roxabell and erosion and out­ wash deposition in the Scioto River valley. The loess- covered Caesar Till ground moraine surrounds several Illinoian-drift inliers west of Pruitdale and Roxabell. Ground moraine elevations generally range from 900 to 1000 feet on the Caesar Till surface except in areas of high bedrock surface where the thin-drift moraine 153

locally is topographically similar to end moraine. Kean drift thickness on this ground rnoraine jg feet (data from 51 water-v/ell logs) with a range from effectively zero (bedrock exposed with occasional erratics on outcrop) to 64 feet. Thinness of this ground moraine is related to its areal position on a high bedrock surface near the escarpment margin. Except in those areas where the northern margin of the Caesar Till drift sheet coincides with a change in bedrock topography (e.g. east of Frank­ fort), this ground moraine is topographically continuous with the loess-free Darby I Till ground moraine to the north. The boundary between these units is established on pedologic criteria (Miami 6A/60 boundary). The northwestern one-fifth of Ross County is covered with Darby X Till ground moraine. This ground moraine takes the f o m of a till plain over large areas (e.g. north of Frankfort). Water-well logs indicate that this moraine is composed of an average of 22 feet of Darby I Till (data from 73 water-well logs) covering several tens of feet of sand, sand and gravel, and "hardpan". Depth to bedrock ranges from 15 to 125 feet beneath the ground moraine surface. Thinnest drift occurs in areas of high bedrock topography such as in eastern Union Town­ ship and southvrestern Concord Township. Surface elevations on the moraine vary from 700 to 850 feet above sea level. 154

Ice-Contact Deposits

Illinoian

The only Illinoian constructional topography in Ross County is a series of ice-contact deposits bordering bed­ rock slopes in valleys tributary to the Paint Creek valley (Figures 40 and 41 ). The topographic anomaly produced by these features was noted by Fowke (1895, p. 17) when he wrote that: "... the rugged hills on the south cease and in their stead appear conical knolls which cause the observer to rub his eyes and wonder if he has been suddenly transported to the region of Omaha, for at no nearer point will he find such a remarkable resemblance to the Missouri river bluffs.,f Typical kamic topography is developed along the southern margin of the Paint Creek valley north of Jones Hill (southwestern Paxton Township), in the Massie Run valley, in Jimtown Hollow, and between Sulfur Lick Flat and Spruce Hill (Plate I). Illinoian kames also are present in the Upper Twin Creek valley, 0.4 mile west of the junction of Tong Hollow and Upper Twin Roads, near the middle of the valley (eastern Paint Tov/nship), and on the north edge of the valley on the boundary between Paint and Twin Townships. The Illinoian kames are typically symmetrical and conical. Lack of erosional modification of these ice- contact deposits is due to: 1) the highly-permeable Figure 40, Kingston Outwash and alluvial terraces (foreground) and Illinoian kames (Ik) bordering the higher bedrock upland along the southern margin of the Paint Creek valley, two miles east of Bainbridge. View south from the southern end of Brenner Hill* Figure 41. Illinoian kames (foreground) and outwash (lo) looking northeast from Houseman Town Road in southwestern Paxton Township down the Paint Creek valley* The tree-covered plateau in the distance is Illinoian-glaciated upland. The silo in the right center of the picture lies on a Kingston Outwash terrace. 157 material forming the Kames, and 2) the isolation of the kames from the main lines of drainage. The unmodified nature also attests to the lack of an extensive Wisconsin glacial lake in the Paint Creek valley. The elevation of the ice-contact deposits varies from 700 to 1070 feet, Relief above the Paint Creek valley floor averages 100 to 200 feet with a maximum local relief of 530 feet north of Jones Hill in the extreme southwestern portion of the county. Exposures of Illinoian ice-contact stratified drift are limited to several small gravel pits southeast and southwest of Bainbridge. A typical exposure occurs in an active gravel pit just west of Ohio Route 41, 0.2 mile north of the Pike County line, east of Jones Hill in southwestern Paxton Township. About 60 feet of coarse, crudely-cross-stratified sand and gravel is displayed on the pit wall. The ice-contact drift has a widely variable particle-size distribution, locally containing erratics up to three feet in diameter. Almost all of the ice-contact material has a surficial yellowi3h-brown or reddish-brown secondary iron oxide staining. Stonecounts in the various groups of Ross County ice-contact features (Table 17) indicate that the Illinoian kames are characterized by lower total carbonate and crystalline content and higher clastic pebble per­ centages compared to the other areas of ice-contact TABLE 1? PEBBLE LITHOLOGIES OF ROSS COUNTY ICE-CONTACT DEPOSITS

Group of *R ice-contact features Total Total Total Samples Igneous Siltstone Metamorph. Cxlines Shale elastics Carbonates Sandstone Dolomite Chert Number of Number Limestone Illinoian 3 mean 29 19 5 53 27 1 15 43 2 2 4 S.D. 1 3 — 4 4 — 1 5 — — 1 high 30 26 5 60 36 1 16 52 3 3 6 ranseioS 28 12 4 4-5 19 0 13 46 1 2 3 Roxabell- A Lattaville mean 50 28 1 79 5 1 4 10 4 7 11 S.D. 2 3 — 3 1 —- 2 —- 2 84 2 4 12 rangelow-a^-^high 52 33 3 6 5 7 15 4-7 19 0 73 4 1 3 8 3 4 7 Kinnikinn ick- 2 Kingston mean 54 25 4- S3 8 0 2 10 5 3 8 S.D. 1 1 — 1 — — — 1 —— 1 56 26 4- 84 8 0 4 11 5 3 8 ranga^oy 52 24- 3 81 7 0 1 9 5 2 7 Hallsville- Adelphi mean 36 34- 3 73 12 1 5 18 5 4 9 3.D. 2 4 3 2 — 1 4 —— 3 high 40 43 6 80 16 2 8 26 8 7 15 range 10 0 4 1 7 31 29 1 68 ? n features. The pebble composition of the Illinoian kames may partially reflect post-Illinoian weathering since all samples were taken from deep in the oxidised zone. The depletion of the more mobile weatherabie constituents such as the carbonates would cause a relative increase in the more resistant materials (e.g. crystalline and sand­ stone pebbles). The Illinoian-kame pebble composition is very different from the pebble lithology of Higby and Kingston Outwash in the Paint Creek valley (Table 13) and the Illinoian Rainsboro Till (Table 12). The Illinoian kames were formed on dead ice stag­ nating against the bedrock margins of the Paint Creek tributary valleys. The lack of till accumulation along the Illinoian boundary contrasts the styles of Illinoian and Wisconsin deglaciation in western Ross County. During deglaciation, that portion of the Illinoian ice sheet which moved over the plateau segment north of the Paint Valley and then glaciated a portion of the upland south of the valley probably became a separate stagnant ice remnant. Since the ice surface sloped naturally to the south and the ice on the uplands was relatively thin compared to the ice mass in the valley, most debris­ laden meltwater naturally gravitated to the southern bed­ rock margin of the present Paint Valley; thus, the pre­ ponderance of Illinoian ice-contact features in this region. The ice-contact deposits to the north in the 160

Upper Tv/in Creek valley may be raoulin kames or crevasse fillings developed synchronously with the ice marginal accumulations to the south or they may represent accumu­ lation of ice-contact drift during a later stage of deglaciation following pronounced shrinkage of the stag­ nant ice mass* Although Late Wisconsin ice did locally stagnate allowing formation of ice-contact deposits, the accumu­ lation of an extensive end moraine along the plateau escarpment margin, and the well-developed series of recessional and readvance moraines in the Scioto Sublobe indicates the Wisconsin ice sheet had a much more active mass balance regimen in the terminal zone during de­ glaciation than did the Illinoian ice sheet* Alter­ natively, the difference in terminal zone deposition may be primarily related to variable duration of these glacial stages.

Late Wisconsin Hoxabell-Lattaville Area (south-central Concord Twp.)

Late Wisconsin kames are present along the Wisconsin glacial boundary south and southeast of Lattaville* These asymmetric accumulations of ice-contact stratified drift along the bedrock escarpment were formed in association with construction of the Lattaville Moraine (circa 18,000 years B*F*)« Maximum elevation on these kames is 1045 161 feet which is about 70 feet lower than the bordering plateau upland. The karaes are topographically indistin­ guishable from the surrounding Lattaville Moraine. The remainder of the ice-contact features in the Roxabell-Lattaville area formed about 17,000 years B.P. in association with the ice sheet readvance to the Reeseville Moraine position. Three separate kame groups were constructed near the Reeseville boundary position, one mile southwest of Roxabell. These symmetrical, conical kames rise 40 to 70 feet above the surrounding drift deposits. Between the southern margin of the easternmost kame area and the bordering Lattaville Moraine is an easterly-sloping drainage channel which indicates the direction of meltwater discharge associated with the formation of these ice-contact features. A gravel pit on the northern margin of the same kame group, near the function of Perry Creek and Davis Roads, exposes 50 feet of chaotically cross-stratified ice-contact drift. Two eskers are located in the vicinity of Roxabell. The larger, northernmost esker, trends northwest-south­ east along the northern margin of Roxabell. This broad (up to 0.2 mile wide) feature is continuous throughout its 1.8 miles length. Local relief between the esker and bordering Wisconsin outwash ranges up to 90 feet. An exposure in a gravel pit on the western edge of Roxabell displays the pseudo-anticlinal structure that is typical 162 of the axial zone of an esker. However, this ice-contact feature nay he partially kame or crevasse filling, as it lacks the continuous sinuous crest that is typical of eskers. This deposit consists of a series of higher conical areas connected by a broad ridge of low relief. The second esker consists of four discontinuous, narrow (less than 0.1 mile wide) segments which trend east-west south of Roxabell over a 2.7 mile-long area. This esker has a distinct crest on each of the segments, with crest elevation generally decreasing eastward. Small borrow pits in each of the segments expose the ice-contact drift. Following readvance, the Reeseville ice sheet stag­ nated in the topographic basin near Roxabell. While kames formed along the southern margin of the stagnant zone, two eskers were constructed by high-energy glacio-fluvial action between ice walls within or beneath the ice mass. The local bedrock surface in the Roxabell area had an easterly slope which controlled the direction of melt- water discharge and thus the formational orientation of the eskers in the easterly-sloping, ice-walled raeltwater channels. The anomalous orientation of the eskers, nearly perpendicular to the direction of Reeseville ice movement, suggests that they might be crevasse fillings, not eskers* However, the lack of ablation till or other supraglacial material in Ross County argues against the 163

formation of crevasse fillings and indicates a subglacial or englacial origin for these features. Pebble lithologies for the Roxabell-Lattaville area (Table 17) shows that these ice-contact features are compositionally similar to the pebble fraction of the Darby I, Caesar, and Rainsboro tills (Table 12) and all outwash units (Table 13) in Ross County except the King­ ston Outwash in the Paint Creek valley. A group of ice-contact features were formed in con­ junction with deposition of a drift "plug" west of Slate Mills at about 18,000 years B.P. These deposits per­ manently altered the northeasterly drainage trend of the Paint Creek valley to the present Alum Cliffs drainageway. Most of these deposits were later covered by a thin till unit (Caesar Till) during formation of the Lattaville Moraine. Kames on both sides of the North Fork valley at Sulfur Lick are related to post-Lattaville Moraine local stagnation in the lee of a narrow bedrock constric­ tion in the valley.

Kinnikinnick-Kingston Area

Many discontinuous Late-Wisconsin ice-contact features occur in a three-by-six mile area extending north from Hopetown to the Pickaway County line (western Green Township) along the eastern margin of the Scioto River valley. This valley was the primary neltwater drainage­ 164

way in the Scioto Sublobe during deglaciation. The areal distribution of the ice-contact deposits indicates that the present features are remnants of a once more extensive area of stagnation features. Most of these ice-contact deposits are asymmetric kames with a local relief ranging from 50 to 170 feet. The largest kame in this area is one-half mile south of Kinnikinnick in the western portion of Section 29, Green Township* This deposit now is modified by extensive extraction of sand and gravel. The original kame-summit elevation of 860 feet has been reduced over 100 feet after two decades of quarrying. Continuous pit excavation exposes up to 100 vertical feet of cross-stratified, slumped ice-contact drift. The Kinnikinnick-Kingston group of ice-contact features are the southernmost deposits in an esker/kame system that extends northward for 32 miles along the eastern side of the Scioto River to near the I-2?0 Outer- belt south of Columbus. A small kame terrace, 1.5 miles southwest of King­ ston (center of Section 8, Green Township), heads the Kingston Outwash deposits delineating the ice-margin position during outwash deposition. Although kames are predominant in this area, several esker segments are included in the stagnation deposits. The most prominent esker extends for 0.8 mile from the western part of 165

Section 5 t o the I'IW> of Section 8 in Greer. Township; 1.5 miles west-southwest of Kingston. Other small eskers are present in the northern one-half of Sections 1 and 6 in Green Township. Darby I Till covers most of the western one-third of these ice-contact deposits probably marking the eastern­ most expansion of the Beeseville-age ice sheet. These deposits accumulated in association with the Lattaville Moraine at which time large areas of the Late Wisconsin ice sheet east of the Scioto Valley decayed near the bed­ rock escarpment* The ice-contact deposits bordering the Kingston Out­ wash clearly predate the outwash. This time relationship is based on: 1) the sharp boundary between the outwash and kames indicating erosion on the margins of the kames prior to outwash deposition, and 2) the tracing of the Kingston Outwash to a kame terrace that is clearly in a recessional position, i.e., the kame terrace formed after the Latta­ ville Moraine to which the ice-contact features bordering the Kingston Outwash are correlative. The pebble composition of the Kinnikinnick-Kingston ice-contact deposits is very similar to that of the Roxabe11-Lattaville group (Table 17). The pebble lithol- ogy of these features is also indistinguishable from the bordering end moraine and ground moraine made of Caesar Till (Table 12) and the Kingston Outwash further down 166 the Scioto River valley (Table 13).

Hallsville-Adelphi Area

A group of discontinuous ice-contact deposits ex­ tends from 1.3 miles west of Hallsville to near Adelphi in northeastern Ross County. The deposits are concentra­ ted along the proximal margin of the Lattaville Moraine and were produced during stagnation following or accompanying end moraine construction about 18,000 years B.P. Most of the ice-contact drift takes the form of large kames which rise up to 140 feet above the bordering lacustrine plain ("The Prairie") and ground moraine. The kames are topographically continuous with the end moraine. There are many small gravel pits in these features that expose slumped, faulted, and folded ice-contact drift, which is typical of areas of disintegration deposits. A large esker (1.8 miles long, up to 0.3 mile wide, and averaging 70 feet high) extends from the west-central portion of Section 9 to the southeastern one-quarter of Section 16 in Colerain Township. This esker is confined to a northwest-southeast trending valley. The south­ eastern margin of the esker is nearly coincident with the drainage divide of the south-flowing Walnut Creek. The esker probably relates to southerly meltwater drainage, glacial lake formation, and consequent northward extension 167

of the Walnut Creek drainage basin (see page 33 ). Pebble counts in the Hallsville-Aieiphi ice-contact deposits (Table 17) indicate these deposits have less dolomite and clastic pebbles than other Late Wisconsin kames and eskers* The pebble composition is unlike any Ross County till (Table 12) or outwash (Table 13) unit* This compositional anomaly may be related to variation in bedrock lithology beneath the drift in the eastern portion of the Scioto Sublobe*

Outwash

Glaciofluvial deposition adjacent to and outside the Illinoian and Wisconsin ice-margin positions created extensive outwash deposits in the Ross County meltwater drainageways. The characteristics of all Ross County outwash deposits are listed in Table 13*

Illinoian Stage Higby Outwash

Illinoian-age outwash in Ross County is named Higby Outwash after the hamlet of Higby which borders Illinoian outwash along the western margin of the Scioto Valley in south-central Ross County. In eastern P.oss County Higby Outwash occurs as: 1) large, flat-top terrace remnants east and southeast of Chillicothe, 2) small terraces along the Lick Run, Dry Run, and Walnut Creek valleys in the TABLE 18 CHARACTERISTICS OF ROSS COUNTY OUTWASH DEPOSITS

to ja O P O P rb O a Unit n o n O O *rl fJH © d d o Soil group deposition a d o O -P a o *h •d*H ©«H o at <0 «J -H S .d .a ft o o o ttfc > d o s s Hi *H d d U o P O P © a d © jd*H 3 O H © © d a PS P o O UK-' S Illinoian near Illinoian boundary Rainsboro- higher in eastern Ross County, 860-670 13 Parke- Higby recessional position in 25-71 145" Negley- lower western Ross County 825-660 10 (9) Pike Wi scons in Knockemstiff Moraine; entire Bartle- Bainbridge ice dam at Bainbridge terrace 800-770 12 96" Pekin- (c. 21,000 years B.P.) silty (6) Markland Lattaville Moraine and later recessional rnoc_cco Kingston positions (c. 18,000 - 3"32 725"6>° 6 46" Ockley 17*800 years B.P.) (14) Tellowbud/Marcy Moraine Roxabell ice-contact r Circleville group (Reeseville Mor) b * 670-615 3.2- 30” Fox (c. 17,000 - 16,800 absent 1.5 (21) years B.P.) Powell Moraine (Delaware Worthington County) (c. 15,000 none 660-600 5.4- 21" shallow Fox years B.P,) 1-?, (16) 169

east-central portion of the county, 3) a pitted outwash plain and terrace remnants throughout the i)-ring, 4) large terraces on the northwest margin of the Scioto Valley front Higby southwest to the Ross-Pike County line, and 5) small terrace remnants near the mouth of Paint Creek, in the Brewer Heights area just west of Chillicothe, and in the vicinity of Renick, Three Locks, and Pride along the western margin of the Scioto Valley* The areal distri­ bution of the terrace deposits indicates that the main Illinoian meltwater discharge passed through the D-ring rather than occupying the previously-formed Newark River valley (Deep Stage) across the D-ring cutoff. Evidently the interglacial (Yannouthian) erosion of the Teays System lacustrine valley-fill (Minford Silt) in the D-ring area and deposition in the Newark River valley created a lower outlet for the Illinoian meltwater through the D-ring, Additionally, the Newark Valley may have been a very narrow bedrock gorge across the D-ring cutoff. This bed­ rock constriction would have inhibited passage of the huge volumes of Illinoian meltwater, thus deflecting the main flow eastward through the D-ring. Pinal post- 111 inoian breaching of the D-ring cutoff and the final establishment of the present course of the Scioto River may have occurred as late as during deposition of the Kingston Outwash (circa 18,000 years B.P.). Two levels of loess-covered Higby Outwash have been recognized in eastern Ross County (Kyde, 1921; Leverett, 19^2; Keapton and Goldthwait, 1959). The higher level heads near Seymourville on the southern flank of Mount Logan, east of Chillicothe. The northernmost portion of this level may be a kame terrace (Kempton and Goldthwait, 1959) although no kettle holes or other evidence of a nearby ice mass have been identified* This higher Higby Outwash level extends as terrace remnants for about five miles southeastward to the northern portion of the D-ring. The two Higby Outwash levels are well-displayed just north of Sandy Bottom Run, west of Higby in southeastern Franklin Township (Figure 42). Elevations on this higher Illinoian level range from 860 to 670 feet within Ross Gounty with a southerly gradient of about 15 feet per mile. The lower Higby Outwash level heads as far north as Section 34 of Colerain Township in northeastern Ross County. Terrace remnants of this level are found along the Walnut Creek and Little Walnut Creek valleys south­ ward to the easterly-sloping pitted outwash plain in the northern portion of the D-ring. The partially-filled kettle holes in this area are broad and shallow. They are highly visible on aerial photographs and following periods of precipitation when the fields are not in cultivation. The lower Higby Outwash level has a southerly gradient of about 10 feet per mile. 171

Figure 42, Two Illinoian-age Higby Outwash levels (Io) and lower Wisconsin inwash terrace remnants (V/i) along the Sandy Bottom Hun valley. View north from Higby School in southeastern Frahhlin Township. 172

Leverett (19'+2) attributed the two levels of Higby Outwash to formation and later destruction of an ice dam on the Ohio River at Cincinnati. The higher terraces were thought to have formed during the period of ice dam ponding while the lower level indicated deposition after the dam was destroyed. However, Kempton and Goldthwait (1959) showed that the two lllinoian-age levels may merge before reaching the Ohio Hiver. This suggests that the two levels may be related to variations in ice and hydraulic characteristics within a single period of Illinoian glaciation or during two successive substages of glaciation. Several workers in Indiana and Ohio (Durrell, 1961; Gooding, 1963; Goldthwait and Rosengreen, 1969; White, 1969) have indicated the existence of two distinct episodes of Illinoian glaciation in other valleys. There­ fore, the two levels of Higby Outwash in Ross County are most likely attributable to deposition in two substages of Illinoian glaciation. Kempton and Goldthwait (1959) reported that Illinoian till overlies the higher level of Higby Outwash in the eastern half of Section 21 and the western half of Section 22 of Springfield Township. Although patches of till are present on this outwash sur­ face, they do not represent a complete till cover. How­ ever, the restricted till cover does indicate an Illinoian ice readvance following deposition of the higher outwash. This evidence supports the idea of two episodes of 173

Illinoian glaciation. The Kigby Outwash has a pebble composition similar to the Worthington, Circleville, and Kingston (Scioto Talley) outwashes and to all of the Ross County till units except the Boston Till (Tables 12 and 13). All outwash units, except the Kingston Outwash in the Paint Creek valley, contain 76 to 78 percent carbonate pebbles with a limestone/dolomite ratio between 0.40 and 0.65. Mean clastic and crystalline pebble content is nine and four­ teen percent, respectively. Data from nine water-well logs which penetrated through the Illinoian outwash to bedrock indicate an average outwash thickness of 79 feet within a range of 42 to 144 feet. The trend of the Walnut Creek valley in south­ eastern Harrison Township suggests that the lower portion of the creek once flowed to the west of Rattlesnake Knob (Section 6 , Liberty Township). The change in stream pattern to the present position east of the knob probably resulted from superposition from Illinoian valley-fill in the northern portion of the D-ring (Goldthwait, 1955). Reconstructed profiles (Plate II) show a pronounced steepening of outwash gradient toward the ice-margin position (south to north). The more pronounced gradient increase in the higher Higby Outwash level indicates a thicker ice mass and closer ice sheet control on outwash deposition than in the lower outwash level. The two 174*

Illinoian profiles converge to the south within Ross County and nay he inseparable in the lower Scioto River valley (Kempton and Goldthwait, 1959). Local relief between the two terrace levels decreases from about 4-0 feet at the latitude of Chillicothe to about 15 feet at the Ross-Fike County line. Illinoian outwash also occurs in western Ross County in the Faint Creek valley and tributary valleys. Higby Outwash in this area occurs as: 1) highly-dissected terrace remnants west and northwest of Bainbridge, 2) high terraces in the Lower Twin, Upper Twin, and Flug Run valleys, and 5) & narrow remnant (identified on the basis of soil profile) on the south wall of the Faint Creek valley opposite Schotts Bridge. The distribution of Higby Outwash in western Ross County indicates primary outwash deposition occurred in association with meltwater dis­ charge south through the Beech Flats area of southwestern Ross, Fike, and Highland counties. At recessional positions during Illinoian deglaciation, meltwater may have drained down the Faint Creek valley into the Scioto Valley by way of the newly-formed Alum Cliffs Gorge or the North Fork valley. Lack of terrace remnants in the vicinity of the Alum Cliffs Gorge and the Slate Mills area indicates that if easterly discharge of Illinoian meltwater did occur, the duration of the episode was minimal. 175

During rise Sangamon Interglacial expensive valley cutting occurred in the Higby Outv/ash deposits. Some of the valleys which formed during this interval in eastern Ross County served as meltwater drainageways in Late Wisconsin tine. The Dry Run, Walnut Creek, Little Walnut Creek, Salt Creek, and Sandy Bottom Run valleys all con­ tain Wisconsin outwash or inwash terraces which lie several tens of feet below the Higby Outv/ash levels* The inwash terraces consist of fine silt and clay which may be lacustrine in places*

Early Wisconsin Substage

Early Wisconsin (Altonian) outwash deposits have not been identified in Ross County* Only in the Hocking River valley have terraces of Early Wisconsin age been clearly identified on the basis of a markedly distinct soil pro­ file and pebble lithology. The absence of such features is probably related to several factors. The Early ’Wisconsin ice mass may have terminated in widespread down- wastage forming extensive ice-contact deposits (Dreimanis and Goldthwait, 1973)* Outwash deposition occurring with this stagnant ice body may have taken the form of a fan rather than a valley fill so that most of the Lockbourne gravel (outv/ash) may have been deposited north of Ross County nearer the glacial limit. Erosion during the Sidney Interstaiial and erosion 176

and deposition in association with the Wisconsin glacia­ tions nay have removed or buried all of the Early 'Wiscon­ sin outwa3h. Buried sand and gravel surfaces such as the one beneath the Yellowbud Moraine and the Darby I Till ground moraine south of the Yellowbud Moraine in north- central Ross County, may represent Early Wisconsin (Lockbourne gravel) deposits.

Late Wisconsin Substage

Following further erosion and valley cutting in the Illinoian valley-fill during the Sidney Xnterstadial, four episodes of Late Wisconsin outwash deposition created extensive accumulations of glacio-fluvial drift.

Bainbridge Outwash (Kempton and Goldthwait, 1959)

Bainbridge Outwash in Ross County consists of a series of intermediate-level (elevation 800 to 770 feet) silt terraces which are confined to the Paint Creek valley west of Bainbridge. Augerings indicate that these silt deposits are at least 10 feet thick. Bainbridge Outwash terraces lie about 150 feet lower than the nearby Higby Outwash terraces and about 20 to 40 feet above the border­ ing Kingston Outwash remnants. The intermediate elevation of these terraces coupled with the transitional nature of the soil profile developed on the silt terraces, led many workers (e.g. Petro et al., 19&7) to conclude that the 177

Bainbridge Outv/ash deposits are "early” .Visconsin (Altonian). However, new radiocarbon chronology from Highland County (Rosengreen, 1970) indicates this outwash formed in conjunction with the Knockemstiff and Mt. Olive Moraines about 21,000 years B.P., i.e. early Late Wiscon­ sin (Woodfordian). On the basis of the limited areal extent of the terraces and the silt-size grains composing the Bain­ bridge Outwash, Kempton and Goldthwait (1959) proposed that these terraces represent "... slackwater deposits accumulated in 'early' Wisconsin time along decaying ice which squeezed into both ends of the Paint Creek valley." However, recent mapping in Highland County (Rosengreen, 1970) and now in Ross County indicates that the earliest Late Wisconsin ice sheet did not advance into the weBtem end of the Paint Creek valley in eastern Highland and western Ross Counties. The Ross County Bainbridge Outwash silt terrace levels project on a profile to sand and gravel terraces associated with the Mt. Olive Moraine in eastern Highland County. Thus the Bainbridge Outwash consists of normal sand and gravel in Highland County but grades to slack­ water, lacustrine silt in western Ross County. Therefore, the thin, Late Wisconsin ice sheet that advanced south- westward up the Paint Valley and deposited the Knockem- stiff Moraine (Boston Till) dammed the Paint Creek valley 178 drainage near Bainbridge. Slackwater deposition in this glacial lake west of Bainbridge (Glacial Lake Bainbridge) may have occurred synchronously with deposition of the molluscan fauna and lake sediments in Glacial Lake Hum­ boldt, three miles to the north in the Buckskin Creek valley# Reconstructed profiles (Plate II, in pocket) indicate that the Bainbridge Outwash had an easterly gradient of about 12 feet per mile, slightly steeper than the later Kingston Outwash. The effect of the Bainbridge ice dam in creating a temporary local base level is clearly seen on the profiles because the Bainbridge Outwash terraces do not project to any similar outwash level downstream in the Scioto Valley.

Kingston Outwash (Kempton and Goldthwait, 1959)

The highest level Late Wisconsin outwash in the Scioto River valley is the Kingston Outv/ash (Figure 43). Outwash elevations range from 725 to 650 feet above sea level. At the latitude of Chillicothe the Kingston Out­ wash is about 115 feet below the higher Higby Outv/ash level and about 45 feet above the next highest Wisconsin outwash terrace (Circleville Outwash). Kingston Outv/ash heads in a kame terrace in the center of Section S, Green Township. There are a few shallow kettles in the terrace segment bordering the ice- 179

Figure 43. Exposure of cross-bedded Kingston Outwash near the junction of Blacksmith Kill Road and Charleston Pike (sample location 87). Note the upper portion of the Cckley soil profile is developed in dark loess. The scale in the center of the picture is one yard long. 180 contact deposit, Kingston Outwash terraces in eastern Hoss County extend about five miles southeast of Chilli- cothe down the Scioto River valley. Terraces are also present in the Walnut Creek and Dry Run valleys east of Chillicothe. An isolated portion of Kingston Outwash terrace is present south of the junction of Paint Creek and North Pork, three miles southwest of Chillicothe, The absence of Kingston Outwash terraces in the Scioto Valley south of the Walnut Creek valley (Rupels) may be related to the steeper gradient of this outwash level as compared to the younger Circleville and Wor­ thington levels (Plate II). Linear projection of the Kingston level profile on Plate II suggests that this level may become buried beneath the Circleville Outwash between Rupels and Higby in the area of the D-ring cut­ off. Alternatively, the Kingston Outwash may have been deposited while the D-ring cutoff was taking place. Thus the higher Kingston Outwash level and the absence of terraces south of Rupels may have been controlled by a bedrock constriction in the D-ring cutoff area. As the Scioto Valley between Rupels and Higby continued to be widened, the bedrock control on outwash deposition was diminished. The portion of the Kingston Outwash east of the Scioto Valley is readily identifiable on the basis of: 131

1) terrace elevation, 2) outwash gradient, 3) loess cover, and 4) the Ockley soil profile (Table 18)* Composition- ally this outwash i3 indistinguishable from all of the bordering Late Wisconsin and Illinoian levels. Kingston Outwash typically consists of medium to coarse sand and gravel but in some areas (e.g. gravel pit at sample loca­ tion 87, Figure 4-3) it takes the form of fine sand and gravel with well-defined cut and fill structures. Kingston Outwash is also present in the Faint Creek and Lower Twin Creek valleys of western Boss County. This Kingston Outwash is compositionally distinct from all other Ross County outwash units (Table 13)- This outwash contains 11 percent more carbonates (with a much lower limestone/dolomite ratio) and about 30 percent less crystalline pebble3 than the other outwash units. King­ ston Outwash in the Paint Valley also lacks the loess cover which typifies the Scioto Valley portion of this outwash level. The projected profiles (Plate II, in pocket) of the outwash deposits show an anomalous situation in which two nonsync'nronous outwash levels in separate valleys lie along the same profile. The high, silt terraces (Glacial Lake Boumeville) near Boumeville project to the valley- fill level west of Slate Mills (Anderson on Plate II). Higher sand and gravel terraces in the Lower Tv/in Creek valley tie to sand and gravel terraces in the North Fork 132

valley. Kempton and Goldthwait (19p9) suggested that following construction of the Lattaville Noraine, melt­ water that discharged down the Lower Tv/in Valley was able to enter the North Fork valley across the valley-fill west of Slate Mills and form the higher level Kingston Outwash terraces in the Lower Twin Creek valley. Following further retreat of the ice sheet from the Lattaville Moraine, meltwater from the wasting ice mass discharged down the Faint Creek and Buckskin Creek valleys and then through the Alum Cliffs Gorge, creating the Kingston Out­ wash level in the Paint Creek valley (circa 18,000 years B.P.). Presumably Kingston Outwash deposition was occurring simultaneously in the North Fork valley. Since only a short time interval (600 to 800 years) elapsed between deposition of the Kingston Outwash and readvance of the ice sheet to the Reeseville Moraine maximum position, only minor entrenchment of the outwash terraces occurred. As the Reeseville ice stagnated near Roxabell, forming the ice-contact features in that area, Circleville Outwash was deposited down the North Fork valley. The gradient of the Circleville Outwash was locally controlled by the only slightly-eroded Kingston Outv/ash valley-fill. Due to the gradient control and the minor duration of this episode of outv/ash deposition in the North Fork valley, these Circleville terraces project to the profile gradient established by the Kingston 183

Outv/ash terrace associated with the Lower Twin Creel; valley. Thus tv/o, nonsynchronous outwash accumulations in adjacent valleys can seemingly represent a single, synchronous episode of outv/ash deposition.

Circleville Outv/ash (Kempton and Goldthwait, 1959)

Circleville Outwash heads at the southern margin of the Marcy Moraine near Circleville in Pickaway County and in the Roxabell group of ice-contact features in west- central Boss County. The intermediate-level Late Wis­ consin Circleville terraces are areally the most extensive Wisconsin outwash deposits in Ross County. These outwash terraces extend up the Beer Creek valley to near the Hoss- Pickaway County boundary, thereby marking the position of ice margin during deposition of the outwash. Circleville Outwash deposition occurred at two distinctly different ice positions; initially from the Reeseville Moraine maximum position near Roxabell (circa 17,000 years B.P.) down the North Pork valley, and later in the Scioto Valley while the ice sheet stood- at the Reeseville recessional position at the Marcy and Yellow­ bud Moraines (circa 16,800 years B.P.). Meltwater drainage during retreat from the Darby X Till limit at the patchy Reeseville Moraine formed a lower level in the Kingston Outwash which extends from Hopetown to north of Kinnikinnick along the eastern margin 184

ox' the 3cioto Valley. This relationship is iniicated by the Darby I Till on the ice-contact features along the eastern valley margin which is lacking on the intervening lower, dissected Kingston Outwash level. A meltwater channel was cut in Late Wisconsin Darby I Till ground moraine in southeastern Deerfield Township during deposition of the Circleville Outv/ash. In this northwest-southeast-trending channel, the Darby I Till has been completely stripped exposing the sand and gravel unit which underlies much of this ground moraine. Although compositionally indistinguishable from the other Wisconsin and Illinoian outwash deposits (Table 13), the Circleville terraces can be identified on the basis of terrace elevation and depth of leaching (soil profile) (Table 18). Circleville Outv/ash consists of at least 60 feet of medium to coarse sand and gravel (few wells penetrate through the outwash to bedrock) which is locally stained or cemented by iron oxide.

Worthington Outwash (Kempton and Goldthwait, 1959)

The lowest Late Wisconsin outwash level in the Scioto River valley heads near the Pov/ell Moraine in southern Delav/are County. Worthington Outwash terraces are identi­ fied in Ross County, 50 miles south of the source, on the basis of terrace elevation and depth of leaching (Table 18). This medium to coarse sand and gravel has a slightly 185

higher lxmestone/dolonite ratio than the other Wisconsin and Illinoian outwash units (Table 13) but the outv/ash is not distinguishable solely on this single criterion* Worthington Outv/ash terraces in Ross County range in elevation from 660 to 600 feet above sea level* Large areas of outwash plain in the Scioto Valley have been mapped as alluvial terraces and alluvium (Plate I) since they are covered with a surficial alluvial unit. The sand and gravel underlying most of these low-lying terrace segments is Worthington Outwash* The projected profiles (Plate XX) show that the Worthington and Circleville levels are nearly parallel in gradient throughout Ross County* Both outwash profiles have an anomalous inflection point near Chillicothe where the gradient decreases from about 3*2 to 1*5 feet per mile* Note that no similar gradient variation occurs in the higher Kingston and Higby Outwash levels* North of Chillicothe the two profiles nearly parallel the present gradient of the Scioto River. South of Chillicothe the outv/ash levels remain mutually parallel but decrease in gradient relative to the Scioto River. The divergent gradients continue down the Scioto River valley (Kempton and Goldthwait, 1959» Figure 3) although the magnitude of divergency is not as pronounced in the lower Scioto Valley as in Ross County. This gradient anomaly appears to be related to 186

narrov/ing of the .Vjioto Valley by tae constrictive nature of rhe bedroch uplands and nigher outwasn terraces. The point of inflection on the outv/ash profiles (Plate IX) is nearly coincident with the Appalachian Plateau escarp­ ment and the northern limit of Higby Outv/ash. Both these features served to confine Late Wisconsin meltwater to a narrower channel compared to the valley north of Chilli- cothe. This confinement caused a rise in the level of meltwater flowing through the valley south of Chillicothe. Outwash deposits formed in association with these higher- level meltwater discharges would have a lower gradient in comparison to areas of unrestricted flow north of the bedrocl:-terrace constriction.

Lacustrine Deposits

Lacustrine deposits in Ross County range in age from pre-Illinoian (Kansan?) Minford Silt in the 3-ring to post-Late Wisconsin sediments in "The Prairie" region of north-central Colerain Tov/nship.

Minford Silt Deposits in the D-ring

Pre-Illinoian (Kansan?) lacustrine sediments are exposed in the D-ring area of southeastern Ross County. These accumulations of Minford Silt (Stout and Schaaf, 1931) formed in an extensive finger lalce which was created by damming of the Teays System drainage by Kansan or pre- 187

Kansan ice. The best exposure of Minford Silt in Ross County is on the north side of a Baltimore and Ohio railroad cut, 0.7 mile northwest of Lickskillet (SEJi Section 16, Liberty Township). Forty-two feet of varved silt and clay are exposed in this section (Figure 44). Augering at the base of the exposure indicates the lacustrine sediments con­ tinue at least eight feet below the railroad level. The varved sediment is slightly calcareous (Table 19) and has been oxidized to a depth of 1? feet. Individual laminations vary in thickness (1mm to 1cm) and grain size (clay to very-fine sand). The laboratory data (Table 19) shows that the term ’’Minford Silt" is actually a misnomer when applied to this clay-rich sediment. Clay-mineral analyses indicate this sediment is composed primarily of illite and vermiculite with little or no montmorillonite, chlorite, or kaolinite present. The Lickskillet varves contain many secondary, white (7.5YR8/1) calcium carbonate accumulations along a well- developed joint system. Coloration of the sediments typically varies from brown (10YR5/5) to yellowish-brown (10YR5/4) in the oxidized zone to dark gray (10YR4/1) in the unoxidized portion of the section. Preservation of Hinford Silt up to an elevation of 750 feet at the Lickskillet section occurred because of the protective influence of higher bedrock topography 188 Sample Elevation Number 750* . • ., o *. ;©• q ,mixed • * • * zone^ JT

:76 (10YR5A: ■ 9 many secondary CaS0^\« > oxidized" accumulations alongll— S/. - -.1 pint 8 ■ v •77 (10YR5/3)' _72L3J .desth. at-oxidation :(5YR4/1): r - - *. -unoxidized- ■78 -laminations 1 mm -/i' '/ sandrV -varves 1 cm— thickness— /sandstone • *■ *■ *r*" stoned 3 ;. and.-.-.

".and-"[ r ■ *■*- t ^ shale'> shaleo- - - ,4i _•

- p * Baltimore and Ohio railroad

Figure 44. Diagram of the Lickskillet lacustrine section (0*7 mile northwest of Lickskillet; SEX, Section 16, Liberty Township).

TABLE 19 RESULTS OF LABORATORY ANALYSES FROM THE LICKSKILLET LACUSTRINE SECTION (MEAN VALUES FROM FOUR SAMPLES) Sand Particle-size Silt Clay 2mm distribution 0.4# 33*6# 66.0# 0.0# Calcite- Total Calcite Dolomite Calcite Dolomite Carbonate DoTomiEe analyses 2.5# 3*6# 6.4# 0.69 Mineral £Llita Mont. Verm. Chi. Kaol. Quartz Inter, analyses 80# 0 15# Tr Tr 5# Tr waxen borders both the eastern and western margins of the section* Evidently the varves were deposited in a small tributary valley or bedrock alcove on the south side of the Teays Valley. Water-well logs in the Illinoian out- wash (Sections 9 and 10, Liberty Township), one mile northeast of the Lickskillet section, indicate the presence of Minford Silt ("blue muck") beneath the out- wash at elevations ranging from 600 to 660 feet. Similar sediments are exposed at an elevation of 600 feet beneath 4-0 feet of Higby Out wash in a stream cut in the NEJC of Section 10, Liberty Township. The differences in elevation of the Minford Silt deposits in the northern portion of the D-ring indicates that locally up to 150 feet of this lacustrine valley-fill was removed prior to deposition of the Higby Outwash. Loess-covered Minford Silt is exposed in a large area of the southern portion of the D-ring, north of Richmon- dale (Sections 27, 33, and 34-, Jefferson Township). Small remnants are also present in the small tributary valleys southeast of Richmondale. Ho Minford Silt deposits are found in the Paint Creek valley. It is probable that deposition did occur in this region but the sediments were later removed by the com­ bined erosive action of post-Kansan glaciations and drainage systems. 190

Indian Hun Valley - Glacial Lake Massieville

The northeasterly drainage trend in the Indian Run valley (southeastern Scioto Township of south-central Ross County) was blocked near the mouth of the valley by both the Illinoian and early Late Wisconsin ice sheets. Drainage impoundment resulted in the formation of Glacial Lake Massieville in the Indian Run valley. Lacustrine sediments from both the Illinoian and Wisconsin stages of this glacial lake are exposed from Massieville northeast to the Scioto Valley. Illinoian lacustrine sediments are exposed in the Massieville section (Figure 23) near the mouth of the Indian Run valley. About 12.5 feet of rhythmically- banded, reddish-yellow (7*5YH6/6) clay and silt lacustrine sediment overlies Rainsboro Till. The lake sediments contain very few dropstone pebbles. The lacustrine unit is unconformably overlain by 17 feet of coarse Higby Out- wash. Varved clay is present in several localities in the Indian Run valley. It is well-exposed in drainage ditches and shallow road cuts along Ohio Route 23* The varved clay is typically dark gray (5XR4/1) and contains many more pebbles than the related Illinoian sediments. Hyde (1921) noted these deposits and considered them to be Illinoian because he did not feel the Late Wisconsin ice had advanced south of Chillicothe. However, striae on shale, formerly exposed one mile north of the Indian Run valley, were considered fresh enough to be of Wiscon­ sin age by Kempton and Goldthwait (1959). These striae coupled with the low elevation (590 to 610 feet above sea level) of the varved deposits compared to the nearby high Illinoian terraces, strongly indicates a Late Wisconsin age for these deposits. Since the initial Late Wisconsin ice advance was the most extensive in the Faint Creek valley, it appears likely that the lacustrine sedimenta­ tion in the Wisconsin stage of Glacial Lake Massieville occurred during this period, i.e. early Late Wisconsin (circa 21,000 years S.P.)* Thus these lake deposits are correlative with the Xnockemstiff Moraine (Boston Till) and Bainbridge Outwash (Glacial Lake Bainbridge) of the Paint Creek valley.

Buckskin Creek Valley - Glacial Lake Humboldt

Lacustrine sediments at the Humboldt deposit (0.4 mile north of 3M 793, east of Ohio Route 41, central Paint Township of west-central Ross County) were described by Reynolds (1959). He considered the lake beds to be "early" 'Wisconsin age on the basis of comparison of the molluscan assemblages from this deposit to assemblages from deposits of known age and interpretation of the local stratigraphy (Table 20). 192

TABLE 20 GENERAL STRATIGRAPHIC SECTION OF THE LACUSTRINE AND RELATED SEDIMENTS IN THE BUCKSKIN CREEK VALLEY (AFTER REYNOLDS, 1959)

Unit Descriotion Thickness 9 Till, silty and clayey, oxidized 15' brown, calcareous 8 Till, silty and clayey, unoxidized 17 blue-gray, calcareous 7 Clay, smooth and somev/hat plastic, 24 blue-gray, noncalcareous, shale fragments 6 Feat, contains many crushed Molluscs 34-2 5 Marl, clayey, very fossiliferous 5-5 4 Marl and clay, laminated in places, 1-6 some fossils in upper portion 5 Sand and gravel, reddish-brown, 0-5 calcareous 2 Till, oxidized brown, calcareous, 5-8 silty 1 Gravel, coarse, stratified 10+

Unit 1 is stratigraphically and corapositionally correlative to a gravel unit which contains a well- developed Sangamon paleosol that is exposed 0.5 mile south of the Humboldt section. Thus this basal gravel unit is probably Illinoian outwash. The till overlying the gravel was mapped as "early” Wisconsin by Petro et al. (1967) because of pedologic variations which differentiate this unit from known Illinoian or Wisconsin drift. Reynolds (1959) indicated that the Unit 2 till wag continuous with the uppermost till (Unit 9) in the composite section. He thus concluded that the interbedded lake deposits were also "early” Wisconsin. Till similar to Unit 2, once mapped as "early" 195

Wisconsin in Highland Country, has been recently shown to be oi' early Late Wisconsin age (Sosengreen, 1970). Since Unic 2 appears to be correlative with tne Boston Till of Highland County and the Boston Till and Bainbridge Outwash of the Paint Creek valley in Boss County, the age of the Humboldt lake sediments must be about 21,000 years B.P. (early Late Wisconsin). The lacustrine sediments are primarily clay and marl and are highly fossiliferoua in restricted zones* These sediments range in thickness from four to ten feet within the valley* The base of the lacustrine section at the Humboldt deposit lies at an elevation of about 810 feet. Time-correlative, silty Bainbridge Outv/ash terraces in the Paint Creek valley range up to 790 feet elevation. The variations in elevation, texture, and fossil content between these two lacustrine accumulations argues against formation in a single glacial lake* Therefore, these lake deposits probably accumulated nearly synchronously in two separate glacial lakes in adjacent valleys. Early Late V/isconsin ice advanced southwestward up the Paint Creek valley, forming Glacial Lake Bainbridge west of an ice dam in the vicinity of Bainbridge. Simul­ taneously, the ice sheet blocked a north-flowing stream, which headed in a col, 2.5 miles southeast of Humboldt, in the Buckskin Creek valley north of Humboldt* In the resultant proglacial lake, Glacial Lake Humboldt, the 194

lacustrine sediments of the Humboldt deposit accumulated. Eventually the col at the southeastern end of the lake was breached and Glacial Lake Humboldt drained into the Paint Creek valley; possibly into Glacial Lake Bain­ bridge. Relatively rapid downcutting at the outlet through the Ohio shale caused permanent reversal of the drainage direction in the Buckskin Creek valley following deglaciation.

Paint Creek Valley - Glacial Lake Boumeville

Illinoian lacustrine sediments associated with Glacial Lake Boumeville are present from Dills (Cali­ fornia Hollow) northeast to Schotts Bridge in the Paint Valley and in the Owl Creek valley north of Schotts Bridge. Although the lacustrine terrace remnants pre­ dominate along the northwestern margin of the Paint Creek valley, highly-eroded lake deposits are present in the Sulfur Lick valley and in the valley betv/een Copperas and Little Copperas mountains along the southeastern valley margin. These latter deposits are exhumed remnants which have been stripped of their former Boston Till cover. The areal distribution of these lake deposits indicates they are minor remnants of a formerly more ex­ tensive lacustrine accumulation. Plate II shows that these lacustrine terraces, which range in elevation from 700 to 720 feet above sea level, 195

project to the level of the valley-fiil was: of Slate Mills where the Illinoian ice darn which created Glacial Lake hourneville was located. The terraces typically consist of silt but also contain laminated-ciay zones and fine sand and gravel lenses. Shallow road cuts along U.S. Route 50 and small borrow pits afford the best ex­ posures of the terrace stratigraphy. At sample location 62 in the Sulfur Licit valley, south of Storms, recent slumping has exposed several feet of Boston Till in the main scarp of the slumped area. Augering at the base of the scarp indicates the presence of a pebble-rich, dark gray clay beneath the till. In non-3lumped portions of the same slope, pebble-rich, laminated clay forms the surface unit. This lacustrine unit is topographically and stratigraphically beneath the Boston Till. This is a clear indication of Illinoian age for the lake sediments deposited in Glacial Lake Boume­ ville. These lake sediments do not contain the expect­ able Sangamon weathering profile. The absence of the paleosol is probably attributable to removal of surficial material by mass movement or normal erosion during the interglacial period or incorporation of the weathered debris by the initial Late Wisconsin ice sheet which deposited the overlying Boston Till. 1 % "Tne Prairie" (north-central Golerain Township)

Park (7-5YH2/1) lacustrine silt and intercalated marls are present in a large area of northeastern Ross and southeastern Pickaway counties, which is known locally as "The Prairie". These sediments have an average thickness of 25 feet (data from four water-well logs) and generally overlie coarse sand and gravel. The lake sediment surface slopes gently to the northeast indicating the glacial lake drained through the Salt Creek valley. Spruce fragments included in a thick marl unit in these lake sediments, 0.8 mile northeast of Hallsville, range in age from 12,685 * to 15,695 ± years B.P. These dates indicate that the lacustrine sediment of "The Prairie" wa3 deposited in a post-Late Wisconsin glacial lake which developed along the proximal edge of the Latta- ville Moraine and associated ice-contact features following ice margin recession. Deposition could have begun as early as 17,000 years B.P. if the mapping of the Darby I Till (Reeseville Moraine) position is correct.

Beech Plats (southwestern Ross County, Paxton Township)

Lacustrine sediments in the Beech Plats area consist primarily of silt. Water-well log data in Ross and Pike counties indicate these silt accumulations average 20 to 40 feet; in thickness and overlie sand, sand and gravel, 19?

or till. Surface exposures of bedded silt, 3ome of which is coarsely-laminated, are limited to ri>e and Highland counties. These lacustrine silts accumulated in a pro­ glacial lake which formed due to blockage of northward drainage to Bourneville Creek (Deep Stage) by Illinoian ice (Figure 10).

Boulder Concentrations

The distribution of Ross County boulders that are larger than one foot in diameter (Figure 4-5) indicates a slightly greater concentration of boulders along the Lattaville Moraine and in an east-west zone of north­ western Ross County which lies in the areal trend of the boulder-rich Bloomingburg Moraine (Figure 39). Goldthwait and Rosengreen (1969) suggested that the greater con­ centrations of boulders on end moraine as compared to the surrounding ground moraine indicates the boulders accumulated as ice marginal, probably supraglacial, drift. Boulders on the Illinoian-glaciated uplands are generally smaller and less numerous compared to the erratics in the Wisconsin-glaciated terrain. Most of the Illinoian erratics are present along the margin of the plateau escarpment. The distribution and size of boulders in areas of Illinoian drift supports the idea of only a thin ice cover on the uplands south of the bedrock escarpment during the Illinoian glaciation. Figure 45< Figure Highland Distributionof boulders andorientation of striaein RossCounty. \ Symbol Orientation ofstriae ReesevilleMoraine distal boundary Illinoianglacial boundary Wisconsinglacial boundary Number ofboulders e 50acres per 4-12 4 - 1 12+

Pika

o c 198

199

Boulder counts at four localities indicates that an average of 78 percent of the boulders are Canadian crystalline lithologies while the remaining 22 percent are composed of local bedrock types (14 percent dolomite, eight percent limestone). Very few of these boulders are striated. The majority of the boulders are found on the surface or in near-surface drift, although several erratics were observed beneath thick drift accumulations in large exposures. Since glaciers have a very poor capacity for sorting, a selective concentration of the boulders must have occurred before the material became incorporated into an advancing ice sheet. The boulders display no tell-tale features which would suggest intense eolian winnowing or any other in situ sorting mechanism. Some­ how the Scioto Sublobe ice must have advanced into Boss County with an anomalous load of supraglacial boulders. Why the composition of the erratics is heavily weighted toward crystalline lithologies remains conjectural. Goldthwait and Rosengreen (1969) suggested that peri- glacial concentration by water or chemical weathering in northern crystalline rock areas could have supplied the advancing ice sheet with the supraglacial load. They also noted that ”... surface concentration of scattered boulders without equivalent till matrix is similar to the debris pattern left by recently surging glaciers and 200 might represent; Just such loba^e activity." The largest erratic observed in Ross County was a roughly spherical granitic boulder, six feet in diameter, which is located in the stream bed of dry Run near sample location 47 in west-central Union Township, The weathered surface of the boulder is rough and unstriated and is locally stained reddish-brown from oxidation of ferro- magnesian minerals.

Directional Indicators

During this study no striations were observed on any rock surface in Ross County. Two striae localities, no longer exposed, have been described in the county (Figure 45)* Leverett (1902, p. 424) reported striae that trended north-south near Buckskin Station (between Lyndon and Greenfield) in western Ross County. In 1954- H* P# Gold­ thwait measured striae that trended S503 (149° to 155°) on a massive shale surface, one-half mile north-northwest of Renick on the western margin of the Scioto Valley. The striae v/ere found on a surface which was partially covered with recent mudflow and slump debris* These striae orientation are nearly coincident with the directions of ice motion derived from till-fabric analyses

(Figures 34, 35* and 36)* The lack of striae on the numerous bedrock surfaces in Ross County is attributed to: 1) post-glacial weathering of exposed surfaces, 2) minimal glacial erosion along the Illinoian and Wisconsin glacial nargins, and

3 ) large areas in which shale, with its typical poor retention of striations, forms the sub-drift bedrock sur­ face. Chapter V

GLACIAL HISTORY

Nebraskan Glacial Stage

No definitive Nebraskan drift has been identified in Ohio. Teller (1970) described a weak, truncated paleosol which was developed in till beneath a Kansan till unit in Decatur County, southeastern Indiana. This Nebraskan till (St. Maurice Till) containing the paleosol is the first possible evidence of Nebraskan drift east of well- documented Nebraskan material in central Illinois (Frye et al., 19&5)* Deposits of questionable Nebraskan age have been described in Ohio and northern Kentucky. These deposits include: 1) Deeply-weathered outwash in extreme eastern Ohio which is distributed south of the glacial boundary. Lessig (1961, p. 36) indicated this outwash accumulated during "... the first glaciation of the Allegheny Plateau11 because this outwash is the highest and most deeply weathered of the four outwash levels in the upper portion of the Ohio River drainage basin.

202 203

2) Crystalline erratics in northern Kentucky (Jillson, 1924a, 1924b, 192$; Leverett, 1929; Campbell et al., 197*0 and southwestern Ohio (Patton and Hicks, 1925; Leverett, 1929; Ireland, 1943; Merrill, 1953). The boulders are generally small and have been interpreted by some as being residual erosional remnants of a Nebraskan till sheet (Ray, 1969; Thwaites, 1946). Since the erratics are found up to several miles south of the generally accepted glaciation limit (Goldthwait et al., 1961), several workers (e.g. Ray, 1969) have shifted the boundary to the south to include within the limit of continental glaciation the area in which the boulders are located. 3) A leached, clay and manganese-rich gravel, exposed in a road cut ten miles southwest of Cincinnati, which has been interpreted as a truncated Aftonian soil profile (Leighton and Ray, 1965). Since the overlying drift is definitely Kansan till, the till unit beneath the paleosol was interpreted to be Nebraskan till. If the initial disruption of Teays System drainage resulted from Nebraskan glaciation (Teller, 1970) then the extensive deposits of Minford Silt, which accumulated in the dammed Teays System valleys, are also Nebraskan age. Lack of reliable dating of drift materials and problematic location of the ice dam which formed the extensive proglacial lake in which the Minford Silt was 204 deposited hinders definitive establishment of either a Kansan (Stout et al., 194-3) or Nebraskan age for the lake sediment. Paleomagnetic data indicates an age of greater than 700,000 years B.P. for the Minford Silt (Marcus C. Boyer, personal communication).

Kansan Glacial Stage

Kansan-age drift has been identified from Iowa to Pennsylvania (Reed et al., 1963; Gooding, 1963; Wright and Ruhe, 1965; Frye et al., 1965; Wayne and Zumberge, 1965; Goldthwait et al., 1965; Durrell, 1961; White et al., 1969; Teller, 1970). The only Kansan drift exposed at the surface east of Illinois is present in south­ western Ohio and adjacent northern Kentucky. Definitive Kansan material has not been identified either in sub­ surface or surficial deposits from southwestern to north­ eastern Ohio, including Ross County. Rosengreen (1970) described a possible pre-Illinoian till unit in central Highland County which overlies a paleosol developed in sand and gravel. Since the till is compositionally and texturally distinct from overlying Illinoian Rainsboro Till, Rosengreen suggested the till above the paleosol may be "early" Illinoian (Alpine Stade?) or Kansan. Gooding (1963) recognized two episodes of Kansan glacia­ tion (Columbia and Alpine Stades) in southeastern Indiana and indicated that stratigraphic evidence suggests 205 that only a short duration interstadial (Garrison Greek Interstadial) separates the two stades. Two truncated Yarmouthian (?) soil profiles were identified in Ross County (sample locations ? and 59). Both weathering zones directly overlie bedrock indicating a probable absence of pre-Illinoian drift in the county. Kansan-age till (Cincinnati Till, Teller, 1970) has been described in southwestern Ohio (Teller, 1970; Durrell, 1961). This deeply-weathered drift is found in valleys as well as on interstream uplands in the Cin­ cinnati region. An associated till, which is more deeply- weathered than typical Kansan drift, is restricted to interstream uplands south and west of Cincinnati. This restricted distribution and depth of soil development led to speculation that this material was Nebraskan drift. However, stratigraphic similarities with exposures in southeastern Indiana indicates this till is actually Kansan age, if the Indiana drift has been properly identified (Wayne, 1956; Teller, 1970).

Illinoian Glacial Stage

Deep Stage drainage was terminated as Illinoian ice advanced into Ross County from the north-northwest. At its maximum extent the Illinoian ice sheet covered the northwestern three-quarters of the county (Plate I). The Illinoian glacial boundary has a northeast-southwest 205

trend across Ross County except in the region southeast of Chillicothe where a sublobe advanced up the Teays River valley into the northern portion of the D-ring. Due to the general absence of Illinoian morainic accumulations, recessional ice-margin positions are delineated on the basis of erosional and depositional features such as ice-contact deposits, heads of outwash units, and ice-dammed lacustrine deposits. While at or near its southernmost position, the ice sheet deposited till on the northern margin of the bedrock "island" in the middle of the D-ring. A thin till was synchronously deposited on the plateau upland along the Illinoian limit in eastern and southwestern Ross County. The absence of morainic accumulations along the crenulated glacial boundary indicates the Illinoian ice sheet was probably thin and only stood near the glacial boundary for a short time interval. Crystalline erratics in the upper portions of the drainage basins of south-flowing streams suggest minor periods of southerly meltwater dis­ charge and outwash deposition while the ice sheet stood at or near the glacial limit. During occupation of near maximum positions, the Illinoian ice sheet caused a permanent reversal in the Deep Stage, northeasterly drainage trend in the Beech Flats area of southwestern Ross County and adjacent Pike County (Figures 9 and 10). While the ice margin stood along the southern margin of Jones Hill (Figure 46), aggradation in the ice-dammed, proglacial lake created an extensive lacustrine plain at an elevation of about 960 feet. Lacustrine sedimentation in the Beech Flats area was sufficient to cause permanent drainage reversal following subsequent recession of the ice margin. Syn­ chronously, ice-dammed Glacial Lake Massieville formed in the recently-deglaciated Indian Bun valley due to ice blockage across the valley mouth by the Illinoian sublobe which had penetrated into the northern portion of the D-ring (Figure 4 6 ) . Besultant lacustrine sedimentation formed the rhythmically-banded sediments which are now exposed in the Massieville section (Figure 26). During various stages of retreat from the Illinoian boundary position, the ice sheet deposited the thin-drift cover on the plateau upland segments. The next establishable Illinoian ice-margin position is defined by ice-contact features in the Faint Creek valley and the northern limit of higher level Higby Out­ wash east of Chillicothe (Figure 47). Ice stagnation occurred along the southern bedrock margin of the Faint Creek valley. The southerly regional slope caused the concentration of meltwater discharge and stratified-drift deposition along the southern edge of the valley. Fossibly the stagnant ice mass was isolated from the main portion of the Illinoian ice sheet by an ice-free, Figure 46. Highland GlacialLake Massieville. Mapof Ross Countyshowing the position of of the Beech lacustrineFlats plain and theIllinoian icemargin during formation Proglaciallake Meltwater discharge Illinoian icemargin position PlChlfflY Co

\ mila*

203

•a-*'*"-* Illinoian ice margin position -» Meltwater discharge

Figure 47. Map of Ross County showing the location of the Illinoian ice margin during formation of ice-contact features in the Faint Creek Talley and the higher Higby Outwash level in eastern Ros3 County. 210

thin-till covered upland north of the Paint Creek valley. The lack of Illinoian ice-contact deposits in eastern Goss County indicates a more active mass balance regimen during ice-margin recession and less topographic control over deglaciation in comparison to the transverse Paint Creek valley. The recessional position in eastern Hoss County which is correlative with deposition of the ice-contact features in the Paint Valley (Figure 47) is delimited by the northern edge of the higher Higby Out­ wash level and a possible kame terrace near Seymourville, east of Chillicothe. This Illinoian ice-margin position during outwash deposition may be the equivalent of the Centerville Stade (Gooding, 1963)* A minor Illinoian ice advance after deposition of the higher level Higby Outwash is indicated by the patchy thin-till cover on the outwash level in portions of Springfield Township. Areas of higher topography in eastern Ross County probably emerged as thin-drift covered nunataks during early stages of deglaciation. Throughout Illinoian recession, Teays System and Deep Stage drainage lines were used as major spillways for meltwater discharge and outwash accumula­ tion. As disintegration continued in the isolated ice mass in the Paint Creek valley, the main Illinoian ice sheet retreated to a new position along the plateau escarpment in western Ross County and near the head of the lower 211

Higby Outwash level northeast of Chillicothe (Figure 48). Outwash deposition in the Buckskin Creek, Upper Twin Creek, Lower Twin Creek, and Plug Run valleys of western Ross County was initiated during retreat to this ice- margin position. Since Illinoian ice in the Schotts Bridge-Slate Mills area probably blocked northeasterly meltwater drainage in the Paint Valley, Glacial Lake Boumeville began to form at this time. Thus, much of the material mapped as Illinoian outwash (Plate I) may actually be deltaic accumulations built into the expand­ ing glacial lake. Eventually the impounded meltwater rose to the elevation of the lowest ool in the southern bedrock margin of the valley. Downcutting at the lake outlet through easily eroded shale created the Alum Cliffs gorge. Following retreat from the ice dam position, the north­ easterly drainage trend through the Slate Mills region may have been reestablished with abandonment of the newly formed Alum Cliffs gorge drainageway. The lower Higby Outwash level (Richmond Stade equivalent?, Gooding, 1963) was deposited in the Walnut Creek, Little Walnut Creek, and Dry Run valleys of eastern Ross County simultaneously with the formation of Glacial Lake Boumeville (Figure 48). The presence of shallow kettles in the pitted plain of the northern portion of the D-ring attests to the proximity of the ice sheet and rapid sedimentation during outwash deposition. Following ! Piki Co

Illinoian ice margin position Meltwater discharge

Figure 48. Map of Ross County showing the location of the Illinoian ice margin during formation of Glacial Lake Boumeville, the Alum Cliffs diversion, and the lower level Higby Outwash in eastern Ross County. 21? glacio-fluvial and lacustrine sedimentation associated with this ice-margin position (Figure 48), the Illinoian ice sheet retreated probably depositing thin-drift ground moraine over the northern two-thirds of Ross County* No surficial Illinoian deposits exist north of the plateau escarpment to delineate subsequent Illinoian ice-margin recessional positions.

Wisconsin Glacial Stage

The Wisconsin Stage of the Laurentide Ice Sheet began following the Sangamon Interglacial (c. 120,000 - 73,000 years B.P.; Suggate, 1974) period of erosion and soil development*

Early Wisconsin Substage (c. 7?,000 - 55*000 years B.P*)

Early Wisconsin (early Altonian) ice did not advance into the southern Scioto Sublobe as was proposed (Petro et al., 1967; Goldthwait et al., 1965) on the basis of pedologic and stratigraphic relationships in Highland, Clinton, and Ross counties. The main Early Wisconsin glacial advance of the Erie Lobe occurred after the well-documented St. Pierre Inter- stade (c. 67,000 - 63,000 years B.P. in the St. Lawrence valley). This ice advanced into central Ohio (Goldthwait and Forsyth, 1965; Goldthwait and Rosengreen, 1969) depositing the Gahanna Drift (Rocky Fork Till) near 214

Columbus, tills below the paleosol at the Sidney cut (LaRocque and Forsyth, 1957i Forsyth, 1965) in west- central Ohio, and the intermediate-level Lancaster Outwash in the Hocking River valley (Kempton and Goldthwait, 1959)* Dreimanis and Goldthwait (1973) indicate probable correlatives of the Gahanna Drift include: 1) Olean Drift south of Lake Ontario (Denny and Lyford, 1963; Muller, 1965; Dreimanis, 1960; Connally, 1964), 2) Sunnybrook Till, Canning Till, and Upper Bradtville Till of southern Ontario, and 3) Whitewater Till (Gooding, 1963) of south­ eastern Indiana, Relict truncated paleosols (Sidney Interstadial soils), developed in gravel (e.g. Lockboume gravel), indicate the Early Wisconsin substage was terminated by rapid downwastage and disintegration of the ice mass with­ out construction of typical morainic forms (Dreimanis and Goldthwait, 1973)* The palimpsest end moraines of north- central Ohio (Totten, 1969) may be an exception.

Middle Wisconsin Substage (c. 55*000 - 23,000 years B.P.)

There are no Middle Wisconsin (middle or late Altonian plus Farmdalian in the Lake Michigan Lobe; Figure 49) glacial deposits in Ohio. Middle Wisconsin features in Ohio include: 1) a paleosol (Sidney Inter­ stadial soil) at the Sidney cut (LaRocque and Forsyth, 1957), 2) the Brush Creek peat (22,000 - 50,000 years B.P. Figure 49* Stratigraphic correlation chart of the glacial deposits of the Miami and Scioto Sublobes of the Laurentide Ice Sheet in Ohio (modified from Dreimanis and Goldthwait, 1973, Figures 3 and 4). i«»rt Unit Tim«/»ir«t.

LAURENTIDE ICE SHEET

a if Us lllf LOU i !! fc M l«nl MM* Scl«t*

TIIW - JO- | T A L I 3 E !

KWbvwM. EEfi IBTn*’ W»*tM«gtaa a

8 WISCONSIN 9 17« • B S r M wi h m A M* fa w p»i8t a m k i w m a TIN VKiiZtUi a M m Oa |J c»ni a U»

IE- Krnww#4 Ol Cmm mr TIN UMavlH*. a M*I»U tawM.fc. "ass.a i * Wl 2 0 - BARLY MIDDLI WISCONSIN SANAA MON IAN MINOIAN - s w ise. Plum

PUidt J □ f r c 4 1 I { i I J; j .

* * * : i i £ 2 J i 1 * i f c • £ J 4 : 1 I 1 ► J 2

1 I { «*

*

1 i i i

6 3 0 - $ I t i * t £ U » | l« » I W « • | M • | |V UBjUOIUSftUUf • • ( • • m i ■HIMNI IN ! -u m u j -«A 216

Forsyth, 1965) from near the Sidney cut in west-central Ohio, 5) paleosol remnants at several localities in cen­ tral and western Ohio (Goldthwait et al., 1965; Gold­ thwait, 195S), and 4) Lower Melvin Loess (Goldthwait, 1968). Dreimanis and Goldthwait (1973) indicate that the portion of the Melvin Loess which is found stratigraph- ically beneath the Caesar Till (Lattaville/Cuba Moraine, c. 18,000 years B.F.) was deposited about 35»000 years B.P. This loessial episode may relate to the Middle Wisconsin glacial interval between the Plum Point and Port Talbot II interstadials (Figure 4-9). However, since no Middle Wisconsin source material is present in Ohio, the Lower Melvin Loess may relate to silt deposition in association with the late Early Wisconsin Gahanna Drift and Lockbourne gravel (c. 55,000 years B.P.).

Late Wisconsin Substage (c. 25,000 - 10,000 years B.P.)

Evidence of the initial ice sheet advance (c. 21,500 years B.P.) following the Middle Wisconsin interstadial (Sidney Interstadial) is found primarily in the Paint Creek and tributary valleys. This earliest Late Wiscon­ sin advance represents the maximum extent of Wisconsin ice in the Scioto Sublobe (Figure 50). A small sublobe advanced down the Scioto River valley forming the striae on shale near Renick and damming the iue5. Mapof Ross Countyshowing locationtheFigure50. of Highland Massieville (c. 21,500 years B.P.). GlacialLakes Humboldt, Bainbridge, and position during depositionof Boston Till theLate Wisconsin ice marginits at maximum (KnockernstiffMoraine) andformation of Proglaciallake Meltwater discharge Wisconsinice margin position Illinoian glacialboundary Mtulrrll

likt

Pickaway Pika

Ca.

217

213

Indian Run valley to fora the Wisconsin phase of Glacial Lake Massieville. The areal distribution of varved silt and clay in the valley indicates that this proglacial lake was much more extensive during the Illinoian stage than in the Late Wisconsin substage. No till deposits of this minor sublobe are known. Luring recession from this position, meltwater discharged down the Scioto Valley completing the cutoff of the D-ring which had been ini­ tiated by the Deep Stage Newark River. At the same time, an ice tongue from the main ice mass in the Slate Mills area advanced up the Paint Creek valley. This minor sublobe deposited the compositionally and texturally distinctive Boston Till in tributary valleys to the Faint Creek valley forming the construc­ tional topography of the Knockemstiff Moraine. At its maximum point of advancement near Bainbridge, the ice mass dammed the easterly drainage to form Glacial Lake Bain­ bridge. Meltwater from this ice mass coupled with dis­ charge from the main ice sheet in Highland County deposited the intermediate level Bainbridge Outwash. This outwash was probably a local source of Upper Melvin Loess. Correlative deposits and features include the Mt. Olive Moraine (Boston Till) of Highland County and the HE-SW striae of the Brassfield limestone surface in northwestern Greene County which are transected by younger NW-SE striae (Dreimanis and Goldthwait, 1975)# 219

Simultaneously, this earliest Late Wisconsin advance (c. 21,500 years B.P.) blocked the northward drainage in the Buckskin Creek valley near Humboldt forming Glacial Lake Humboldt. The unique Molluscs assemblage (Reynolds, 1959) of the Humboldt deposit accumulated in the sediments of this proglacial lake* In the Scioto River tributary valleys whose drainage basins did not extend to the area of actual glaciation, alluvial and colluvial silt and clay accumulated as in­ wash and backwater deposits. Remnants of these deposits are typically found on the lower flanks of Higby Outwash terraces. Such deposits are found in the valleys of Indian Creek, Stoney Creek, Toad Hollow, Coon Hollow, Snake Hollow, Lick Run, Dry Run, Walnut Creek, Salt Creek, Little Salt Creek, Sandy Bottom Run, and Wilson Run. Following a retreat of unknown extent, the Scioto Sublobe readvanced (c. 18,000 years B.F.) to a position generally coincident with the northern margin of the Appalachian low plateau upland (Figure 51). A readvance is indicated by the presence of a "forest bed" (dated at 19,800 A - 20,910 A years B.F* in Highland and Clinton counties) beneath Caesar Till which was deposited in association with this ice-margin position. Additionally, logs up to eight inches in diameter were incorporated by the readvancing ice sheet and are presently exposed in basal Caesar Till in the Anderson Run and Bier's Run iue5- Mapof RossCounty showing Figure51-the location of Highland B.P.). ofthe Lattaville Moraine (c. 18,000years the LateWisconsin ice margin during formation Froglaciallake Meltwater discharge Wisconsinice margin position Illinoianglacial boundary

^ - 9

.-1

q S

0 2 2 Jf

221

valleys. The extensive composite Lattaville Moraine was con­ structed along the plateau escarpment margin during this interval. The ice mass extended onto the upland surface only in the area between Lattaville and Mussellman. At this time in the western portion of the Scioto Sublobe the Cuba, Xenia, and Springfield moraines were construc­ ted, The numerous ice-contact deposits associated with the Lattaville Moraine indicate large areas of ice mass stagnation along the escarpment probably during early stages of deglaciation. The 18,000 years B.P. readvance penetrated up the Faint Creek valley only to the area of Schotts Bridge (Plate I). Local ice sheet stagnation occurred with resultant deposition of the large kame complex west of Slate Mills, These deposits permanently deflected the general northeasterly drainage of Faint Creek through the Alum Cliffs gorge which had formed as an outlet spill­ way of Illinoian Glacial Lake Bourneville. A minor oscillation or readvance of the ice mass deposited the thin Caesar Till cover over much of this group of ice- contact deposits. During retreat from the Lattaville Moraine position the highest level Late Wisconsin outwash (Kingston Out­ wash) was deposited. Meltwater discharge and outwash accumulation occurred in the valleys of Paint Creek, 222

Upper Twin Creek, North .Fork, Scioto River, Walnut Creek, Little Walnut Creek, and Dry Run. The loess-covered Kingston Outwash is probably correlative with the silt- capped Kennard Outwash of the Little Miami Valley (Quinn, 1972) and the Vanatta Outwash of the Licking River valley. The silt cover on these outwash units indicates deposition of the Upper Melvin Loess continued through this period. During ice margin recession from the Latta­ ville Moraine the Caesar Till ground moraine of north­ eastern and western Ross County was completed. Following an ice margin retreat of at least 25 kilo­ meters from the Cuba-Xenia Moraine in Greene, Clinton, and Highland counties (Goldthwait, 197^), the Scioto Sub­ lobe readvanced to the Reeseville Moraine position (Figure 52) about 17,200 years B.F. Several lines of evidence suggest that this readvance may have taken the form of a sublobe surge: 1) lack of drift accumulations along portions of the ice-margin position, 2) anomalous concentration of boulders (correlative to the Farmers- ville Moraine and boulder belt in the Miami Sublobe) that is similar to the tenuous drift deposits of recently surging glaciers (Goldthwait and Rosengreen, 1969), and 3) the short time interval (about 800 years) between construction of the Lattaville Moraine (c. 18,000 years B.P.) and readvance to the Reeseville Moraine position (c. 17,200 years B.P.). 223

9 L-, L-J mHM

CO.

Illinoian glacial boundary Wisconsin ice margin position

«—■ > Meltwater discharge

Figure 32. Map of Ross County showing the location of the Late Wisconsin ice margin following readvance to the Reeseville Moraine position (c. 17,200 years B.P.). 224

Darby I Till was deposited over the northwestern one- quarter of Ross County in association with the Reeseville position* This till overlies Wisconsin ice-contact deposits in a narrow zone along the eastern margin of the Scioto Talley north of Hopetown, delineating the eastern limit of the readvance* Only in the area south of Roxa- bell did this readvance penetrate as far south as the Lattaville Moraine* Local stagnation in the Roxabell area during deglaciation produced a complex of ice-contact features* Southeasterly meltwater drainage down the North Fork valley from the Roxabell area deposited the initial phase of Circleville Outwash* Meltwater discharge during early stages of recession from the Reeseville Moraine position cut the loess-free lower level Kingston Outwash near Kinnikinnick» The lack of silt cover on the Darby I Till and Circleville Outwash indicates deposition of Upper Melvin Loess ceased about 17>000 years B.P. The main episode of Circleville Outwash deposition occurred about 17>000 years B.P. while the Late Wisconsin ice margin stood in a recessional position from the Reeseville Moraine at the Marcy Moraine near Circleville and the Yellowbud Moraine in north-central Ross County (Figure 53)* Correlative position in the western Scioto Sublobe may be the Glendon, Esboro, or Bloomingburg moraines (composite Cable Moraine in Champaign County; Quinn, 1972). The lower Deer Creek and Scioto River **■ ■» Illinoian glacial boundary

j u .o -.j- Wisconsin ice margin position — Meltwater discharge

Figure 55. Map of Ross County showing the location of the Late Wisconsin ice margin during deposition of the Circleville Outwash and formation of the Yellowbud Moraine. 226

valleys served as spillways for meltwater discharge and outwash deposition. Similar loess-free outwashes were synchronously deposited in the Miami Valley (Mad River Outwash), Hocking River valley (Carrol Outwash), and Licking River valley (Utica Outwash). Subsequent ice margin retreat.produced the Darby X Till till plain of Ross and Pickaway counties. A short halt during general recession (c. 16,500 years B.F.) allowed construction of the London Moraine of northwestern Pickaway County and southwestern Franklin County. The Laurentide Ice Sheet then retreated into the Erie basin where an extensive proglacial lake developed during the Erie Interstade (Momer and Dreimanis, 1973)• About 15,000 years B.P. the Erie Lobe readvanced incorporating the lacustrine sediments which had accumulated in the proglacial lake. The resultant unique fine-textured tills (e.g. Hiram Till) were deposited as far south as the Powell and Union City moraines. While the Scioto Sublobe stood at the Powell Moraine (Figure 5^), the lowest level Late Wisconsin outwash (Worthington Outwash) was deposited in the Scioto River'valley. This loess-free outwash is the youngest glacial deposit in Ross County. Postglacial modifications of the Ross County drift deposits include: 1) extensive gullying on higher outwash terraces and end moraines, 2) deposition of low-lying, thin-silt, alluvial terraces overlying outwash terrace /' ■-*! Ml C*. / /

I

Figure 5^* Map of the Scioto Sublobe showing the area relationships of the Wisconsin ice margin position at the Powell Moraine during Worthington Outwash deposition (c. 15,000 years B.P,). gravel, and 3) erosion of outwash terraces and associated formation and abandonment of series of stream channels on the valley-fill surface. APPENDIX

Section A

Sample locations 1 Till, ditch 3E Rt 133, 0,75 mile Stf Jet Ragged Ridge Rd, elev 985'* Concord Twp. 2 Till, stream cut, N bank tributary to Little Ck, 1.4 miles WSW Roxabell, 0.1 mile N of Ragged Ridge Rd, elev 825*, Concord Twp, 3 Till, N bank Paint Creek, 0.2 mile E of Paint Creek Dam, 2.7 miles SW Humboldt, elev 825', Paint Twp. 4 Till, pit, 150' E Dills Rd, 1.5 miles SSE Seip Mound State Memorial, elev 775', Paxton Twp. 5 Till, 150' S Alexander Hollow Rd, 0.5 mile NW Spargursville, elev 820*, Twin Twp. 6 Till, 500* SW Jet Bier’s Run-Cattail Rds, 0.25 mile NW BM 758, stream cut S side Bier's Run, elev 775'* Union Twp. 7 Varved clay, gravel pit, 0.25 mile N Rt 23, 2.0 miles NE Massieville, 250' NW Jet Three Locks Rd- Rt 23, elev 680*, Scioto Twp. 8 Till, sample location 7- 8A Till, sample location 8. 9 Till, road cut, 50' E Lick Run Rd, SE Bunker Hill, 0.4 mile S Jet Lick Run-Rocky Rds, elev 875'* Springfield Twp. 10 Organic zone, sample location 7, elev 670'. 11 Till, gravel pit, 0*2 mile SE Cattail Rd, 0.5 mile SW Jet with Rt 207, elev 880', Union Twp.

229 230

Till, sev.’L'r excavation, 50' E Cactail 0.2 mile V:0:7 Jet v;ich R t 207, elev 755', Union 7::p. 13 Till, road, cut, W side v/estfall Rd, I'M' 3 Jet with Williamsport Pike, elev 695*» Deerfield Twp. 14 Till, road cut, SE Rt 138* 0*1 mile 31 Clarksburg, elev 760', Deerfield Tv/p. 15 Till, road cut, N Asbury Rd, 2.1 miles V/3V/ Clarksburg, 0.7 mile E Jet with Egypt Pike, elev 5351* Deerfield Tv/p.

16 Till, road cut, N Dogtown Rd, 1.0 mile II Plano, 400* NE North Fork Paint Cr, elev 840*, Deerfield Twp.

17 Till, road cut. IT US Rt 351 1*2 miles ESE Payette Co line, elev 860', Concord Twp. 18 Till, road cut, N side Jet Rt 138-Austin Rd, Austin, elev 785*, Concord Twp. 19 Sample location 18, elev 774'. 20 Till, road cut W side Waugh Rd, 1.5 miles N Rt 138* elev 975', Concord Twp.

21 Till, stream cut, N fork Buckskin Cr, 1p0' E Lyndon- Good Hope Rd, 0.25 mile If Rt 28, 0.5 mile NW Lyndon, elev 990', Buckskin Twp. 22 Till, road ditch, N Egypt Pike, 0.75 mile SE Jet Mill Tree Rd, 1.5 miles NW Greenland, elev 765', Deerfield Twp. 23 Till, stream cut, W bank, just NE Jet V/estfall- Simmons Rds, 2.0 miles HE Frankfort, elev 855*, Concord Twp.

24 Till, stream cut, E bank Anderson Run, 200' E Maple Grove Rd, 1*0 mile NE Anderson, elev 710', Union Twp.

25 Till, road ditch, N Rt 138, 0.3 mile E Lattaville, elev 975'» Concord Twp. 26 Till, road ditch, S Rt 28, 0.6 mile V/ Lattaville, 0.2 mile W Jet Porter's Hollow Rd, elev 3151* Concord Tv/p. 231

27 Till, road ditch, 8 v / e stfall 21, 0.2 mil 112 South or.len, elev 045'» Buckskin Tv/p. 28 Till, road, ditch, Clii'i* Run .Rd, 1.3 miles S Jruitdale, 0.1 mile S Jcz Chambliss Rd, elev 900', Paint Twp. 29 Till, road cut, S Rapid Forge Rd, 0.75 mile N Jet Free Ln, elev 1010', Paint Twp. 30 Lacustrine silt, E bank terrace, Ashland Oil tank site, 300* E Rt 41, 0.8 mile S Jet U3 Rt 50-Rt 41, elev 769', Paxton Twp.

31 Till, road cut, Camelin Hill Rd, 0.15 mile S Jet Baum Hill Rd, elev 760*, Twin Twp. 32 Till, road cut, N Baum Hill Rd, 0*5 mile WNW Baum Hill Cem, elev 735*« Twin Twp, 33 Lacustrine silt, 50* N Jet US Rt 50-Blain Hwy, 3.3 miles HE Boumeville, elev 715', Tv/in Twp. 34 Till, stream cut, E bank, 0.3 mile E Jet Swamp Rd (BH 836), elev 850*, Section 20, Colerain Twp.

35 Till, road ditch, S Oak Ln, 0.1 mile E Jet Marietta Pike, 1.0 mile S Hallsville, elev 880*, Section 10, Colerain Twp. 36 Till, road ditch, E Marietta Rd, 50* N Jet Dearth Rd, elev 920*, Section 24, Green Twp.

37 Till, road cut, N Wiley Rd, 0.1 mile SE Jet Marietta Pike, elev 710, Section 34, Green Twp. 38 Till, road cut, E Rt 104, 1.0 mile SW Yellov/bud, elev 675'* Union Tv/p.

39 Till, road cut, W Jet Potts Hill-Schnidt Rds, elev 850', Paxton Tv/p. 40 Loess, sample location 7, depth 6", elev 700'* 41 Loess, sample location 7, depth 24". 42 Loess, sample location 7» depth 40".

43 Loess, sample location ?, depth 60". 44 Loess, sample location 7, depth 72". 232 45 Till,'road cut, 0.05 mile S3.7 BM 7- i Baum Hill Rd, 2.0 miles 33H Boumeville, elev 7-'-', Tv;in Tv/p. 46 Till, stream cut, H bank Dry Hun, 0.25 mile 3;/ Egypt Pike, 1.1 miles G.O Albright Mill Hi, elev 855', Union Twp. 4? Till, stream cut, N bank Dry Run, 0.6 mile 3E Albright Mill Rd, elev 830', depth 4', Union Twp.

43 Till, sample location 47, depth 6'. 49 Till, sample location 47, depth 8'. 50 Till, sample location 47, depth 10' 51 Till, sample location 47, depth 12' 52 Till, sample location 47, depth 14* 53 Till, sample location 47, depth 16* 54 Till, sample location 47, depth 18* 55 Till, sample location 47, depth 20' 56 Till, sample location 47, depth 22* 57 Till, road cut, E Morgan Rd, 0.3 mile EINE Jet Wisecup Hill Rd, elev 1010*, Buckskin Twp. 58 Till, 3 side Baltimore and Ohio railroad cut, 1,5 miles NI7 Lic.kskillet, elev 670', Section 16, Liberty Tv/p, 59 V/eathered-zone sand, sample location 53, elev 655* • 60 Till, pit, 0.15 mile 2 Black Run Rd (Xnockenstiff), 0.5 mile KNE Jet Baum Hill Rd, depth 25% elev 780*, Huntington Twp. 61 Till, sample location 60, depth 45”, 62 Till, pit, 0.1 mile S Alexander Hollow Rd, 0.6 mile H'.7 Spargursville, elev 735’* Twin Tv/p. 63 Till, auger hole, 50* N Bush Mill Rd, 0.1 mile HE Jet Clarksburg-Erankfort Pike, depth 6*, elev 760', Concord Twp. 64 Till, sample location 63, depth 12.3'. 255

65 Till, auger hole, 50' U Beath Rd, 0.1 "lie 15 Jet i-iOrfcon-Zdgeion Rds, depth 4', elev 'il-V?', Buckskin Tv/p. 66 Till, sample location 65, depth 6'.

67 Till, sample location 65, depth 8.5'• 68 Till, sample location 65, depth 141.

69 Till, auger hole, 100' S Rt 28, 0.1 mile SW Jet Norman Hill Rd, depth 2.5', elev 107?', Concord Twp. 70 Till, sample location 69, depth 5'. 71 Till, sample location 69, depth 7'. 72 Tili, sample location 69, depth 10*.

73 Till, sample location 69, depth 16*• 74 Till, pit, 150' W Swamp Rd, 0.5 mile S Rt 180, elev 960', Section 15, Colerain Twp.

75 Alluvial sand, road ditch, N Slate Hill Rd, 0.6 mile SSW Jet Rt 41, elev 950', Paint Twp. 76 Varved clay, N side Baltimore and Ohio railroad cut, 0.5 mile NW Lickskillet, depth 51, elev 750*, Section 16, Liberty Twp* 77 Varved clay, sample location 76, depth 15'• 78 Varved clay, sample location 76, depth 20'. 79 Varved clay, sample location 76, depth 30'. 80 Till, stream cut, W bank Anderson Run, 1.4 miles NW Anderson, lowest till unit, elev 760', Union Twp. 81 Till, sample location 80, middle till. 32 Till, sample location 80, upper till.

85 Loess, Seynourville section, 0.3 mile HR Charleston Pike, depth 15", elev 805*, NEJl Sec-ion 20, Springfield Tv/p. 34 Loess, sample location 83, depth 4CM. 234

S3 Loess, sample location ;5» depth -.,5”. SS Loess, sample location S3, depth 7':". 87 Loess, gravel pit, 301 OS Jet Charleston Pike- Black3mith Hill Rd, elev 680', SC Section 21, Springfield Twp. 83 Till, pit, 501 2 Sugar Run Rd, O.S mile II:/ Jet US Rt 50, elev 715', EC Section 5, Liberty Tv/p. 89 Till, pit, 200' N Spud Run Rd, 0.9 aile HE Spud Run Church, elev 7751, SE;. Section 16, Harrison Twp. 90 Till, pit, 250' E Jones Rd, 0.4 mile SE Jet Ginger Hill-Walnut Creek Rds, elev 755', I'W- Section 6, Liberty Twp. 91 Till, pit, 50' NW Jet McDonald Hill-Hurless Rds, elev 1115 , Twin Tv/p. 92 Till, pit, 100' W Camelin Hill Rd, 0.2 mile N Pleasant Grove, elev 1095', Twin Twp. 235

Pebble count locations 1 Sample location 4. 2 Sample location 5. 3 Sample location 6. 4 Sample location 7, upper sand and gravel. 5 Sample location 7, lower sand and gravel. 6 Sand and gravel, road cut, S of Jet Rt 104— Williams­ port Pike, elev 680', Union Two. 7 Sample location 13. 8 Sample location 14. 9 Sample location 15. 10 Sample location 16. 11 Sample location 18. 12 Sample location 19* 13 Sample location 20. 14 Kame gravel, borrow pit, 0.3 mile S Lattaville, 100* W McDonald Hill Rd, elev 995** Concord Twp. 14A Kame gravel, gravel pit. 1.4 miles SSU Roxabell, 1001 SE Jet Little Creek-Davis Hill Hds, elev 845', Concord Twp. 14B Kame gravel, gravel pit, 0.1 mile W Roxabell, 250' N Johnson Rd, elev 760', Concord Twp. 14C Pebble count location 14B. 15 Sample location 26. 16 Sample location 29. 17 Kame gravel, borrow pit, 50* W Baum Hill Rd, 0.6 mile VJ Jet Black Run Rd, near Baum Hill Cem, elev 900', Tv/in Twp. 236

17 Kane gravel, gravel pit, 200' NW R- -1, 0.5 mile NE Pike Count ,y line, elev 890', Pax tor. 7*.;p.

17 Kame gravel, borrow pit, 75* N 3au:r. Kill id, 0.3 mile HE Jet Jones Levee Rd, elev 715*, Twin Twp. 18 Outv/ash gravel, gravel pit, 200' K Higby Bridge Rd, elev 5901, 8WX Section 32, Jefferson Tv/p.

19 Sample location 34. 20 Kame gravel, gravel pit, 0*3 mile 3 County Line Rd, elev 800', NV/J? Section 3, Colerain Tv/p.

21 Kame gravel, gravel pit, 100* NV/ Rt 150, 1.1 miles NW Hallsville, elev 830', SWJi Section 9, Colerain Twp. 22 Kame gravel, gravel pit, 250* N Delano Rd, 0*2 mile E Jet Sulfur Springs Rd, elev 795*, S-K- Section 29, Green Twp.

23 Outwash gravel, road cut, E Seip Rd, 0.7 mile N Jet Grouse Chapel Rd, elev 700*, EC Section 18, Green Twp. 24 Outwash gravel, borrow pit, 50* H Grouse Chapel Rd near Grouse Chapel, elev 705', SW)£ Section 18, Green Tv/p.

25 Outwash gravel, stream cut, S bank Kinnikinnick Cr, 300' SE Jet Rts 180-159 near Kinnikinnick, elev 685*, SE# Section 19, Green Twp. 26 Outwash gravel, E side Chesapeake and Ohio railroad cut, 0.3 mile W Rt 23, elev 665', £C Section 10, Green Twp.

27 Outwash gravel, stream cut, 0.05 mile V/ :-ietsger near Norfolk and Western railroad, elev ?CO', 3Z# Section 31, Green Tv/p. 28 Outwash gravel, N bank Dry Run, 50* V/ Rt 23, 0.1 milo NW Hopetown, elev 690', Sivi Section 1, Green Tv/p •

29 Sample location 87. 30 Sample location 8?. 237

51 Outwash gravel, borrow pit, 150' Falls 2d, 0.4 mile d ’.V Jet'- dove Fun. HI, elev 730' , Pal.nt Tv/p. 32 Outv/ash gravel, road cut, -I U3 50, 1*1 miles 3« Bainbridge, elev 765', Paxton Twp.

55 Outv/ash gravel, gravel pit, 100' 3 Potts Hill Rd, 0.2 mile 3 Jet US 50, 1.0 mile E3E Bainbridge, elev 725', Paxton Tv/p. 34 Outwash gravel, W bank Paint Cr near DT&I railroad bridge, 0.9 mile E Dills, elev 710*, Paxton Tv/p.

55 Outwash gravel, borrow pit, 75* N Baum Hill Rd, 0.1 mile SW Jet with Jones Levee Rd, elev 695* * Tv/in Twp. 56 Outwash gravel, borrow pit, 150* S Kellenberger Rd, 0.2 mile E Scioto River, elev 655', SC Section 2, Green Twp.

57 Outwash gravel, road ditch, H side Infirmary Ln, across from Ross Co fairgrounds, elev 660*, Union Twp. 38 Outwash gravel, borrow pit, 50' E Rd. 0,5 mile V/ Rt 23, elev 640', WC Section 1, Springfield Twp.

39 Outwash gravel, gravel pit, 250' E Rt 104, 0.4 mile N Mound City Natl Mon, elev 640', Union Twp. 40 Outwash gravel, gravel pit, NE Chillicothe, 0.2 mile NE Mt Logan School, elev 600', Scioto Twp. 41 Outwash gravel, gravel pit, E Chillicothe, 0.3 mile SE McArthur School, elev 590', Scioto Twp. 42 Outwash gravel, gravel pit, 0.1 mile SW Rupels, 300' NE Rt 35, elev 625*, EC Section 14, Liberty Twp. 43 Outwash gravel, 250' S Kellenberger Rd, 0.6 mile E Jet Rt 104, 0.3 mile W Scioto River, elev 640', Union Tv/p. 44 Outv/ash gravel, S bank Scioto River, elev 615', NEJi Section 2, Springfield Twp. 45 Outv/ash gravel, N bank Jet Paint Cr-Scioto R, 2.4 miles 3E Chillicothe, elev 585', Scioto Twp. 46 Outv/ash gravel, gravel pit, 150' 3 Higby Bridge Rd, elev 593 , SV//. Section 32, Jefferson Twp. 238

4? OuOucv/acu Bridge Rd, elev ri'-.O' 43 Pebble count location 47. 49 Outv/ash gravel, borrov; pit, 100* S.v id, 0,3 mile N Ross-Fike Co line, 1.1 miles SE Higby, elev 600', Franklin Tv/p • 50 Outv/ash gravel, gravel pit, sample location 83. 51 Pebble count location 50. 52 Outv/ash gravel, borrow pit, 0.1 mile SW Jet Charleston Pike-Coneord Church Rd, elev 700’, SW% Section 26, Springfield Twp. 53 Outv/ash gravel, borrov/ pit, 200* S Church of the Bretheren, elev 795'* central Section 9» Harrison Twp.

54 Outwash gravel, stream cut, t r i b u t a r y to Blacklick Run, elev 690*, NWJC Section 10, Liberty Twp. 55 Outv/ash gravel, borrow pit, 50' HE Rd, 0.1 mile SE Higby School, 0.1 mile NW canal, elev 610', Franklin Twp. 56 Sample location 39. 57 Sample location 45. 58 Sample locations 47-50. 59 Sample locations 51-56. 60 Sample location 57. 61 Sample location 58. 62 Sample location 59. 63 Sample locations 60-61. 64 Sample location 62. 65 Sample location 74. 66 Sample location 30. 239

67 Ganale location 81. 58 fiaaple location GG.

69 Till, road ditch, 20* N Musgrove 2d, 1.8 miles 3E Schooley, elev 610', UE]v Section 21, Liberty Tv/p. 70 Sample location 89. 240

doc; ion I-

Lesuli,s 0 - Laboratory Analyses

Clay-nineral Analysis Proeelure

Pe'obles and coarser material in tie samples were removed using a if 10 sieve. About 80 grams of the less than 2mm fraction which passed through the sieve was used in the silt-clay percentage determination procedure. Dispersion was accomplished by adding about 5 nil of Na^COj to the 80 grams of silt and clay in a 500 ml bottle. The bottle was half filled with distilled water and then the sample was shaken for 10 to 15 hours on a recipricating shaker at 120 cycles per* minute. The sands were removed by v/et sieving through a #300 sieve and the clay fraction was separated from the silt through use of manual fractionation apparatus. Clay (less than 2 microns) was flocculated with ITaCl and then magnesium saturated, washed, and decanted four times with the aid of a centrifuge. This washing step was repeated sequentially: 1) three times with 1N MgCl^ solution, 2) once with distilled water, 5) once using 50# water, 50# methanol, 4) final washing with 90# methanol solution. Then the magnesium-saturated clay fraction was dispersed in double distilled water ani plated on porous ceramic plates using a micropipet to form oriented clay 241

Ea -e;:>* Three platen were preparei for each sample. One plats was leached with a 10.1' aqueous solution of ethylene glycol and placed in a saturated ethylene glycol atmosphere until x-ray analysis. Immediately preceding analysis, the sample was heated at 40°C for three hours to remove excess ethylene glycol from the surface. The other two samples were air dried for 48 hours. One plate was x-rayed at room temperature, then heated to 450OC for two hours, cooled to about 100°C and x-rayed again. The final plate was heated to 550°C for four hours, then cooled to room temperature and x-rayed. Each sample plate was scanned from 2° to 32° 2Q at the rate of 2° 2©/minute. Analyses were performed on a Norelco diffractometer (Type 12045 full-wave generator) using Cu Ka radiation at settings of 35Kv and 15ma. A diffractogram was obtained from each scan and the resultant peak areas were measured using a planimeter. To compute semi-quantitatively the clay mineral percentages, the following ratios and area factors were used: Illitet The area of the 10,0 - 10.2 peak (ethylene glycol treated sample). MontmorilIonite 1 The area of the 16.7 - 13.0 5 peak (E.G.) divided by 4 tines the 10 £ peak (E.G.). Chlorite: The area of the 14.0 - 14.7 S peal: (4500) divided by 2 tines the 10 ® peak (E.G.). 24-2

r.iuui tne aren 0 :‘ cue sane pea'.: neat); area difference divided by 2 tines tie 10 i? peak area (E.G.). Kaolinite: The area of the 3*5 £ peak (^5C° heat) minus the area of the same peak (550° hear); the area difference divided by 4- times the area of the 10 fi peak (E.G.). Quartz: The area of the 3.3 5 peak (E.G.) minus 3/4- area of the 10 5 peak (E.G.); area difference divided by 4- times the area of the 10 J? peak (E.G.). To compute the clay mineral percentages the sura total of all ratios (Illite is arbitrarily assigned a ratio of 1) is divided into the ratio obtained for each mineral species. 243

TABLE 21 RELATIVE PERCENT OP CLAY MINERALS IN THE LESS THAR TWO MICRON FRACTION OF ROSS COUNTY STRATIGRAPHIC UNITS AS DETERMINED BY X-RAY DIFFRACTION ANALYSIS Relative Percent

' © *d •a Unit/ P * © *H •H •H 0 <000 «0©3 Sample O 0) •H IN •rl rH p p • P * H •H © 0 KN tf IN Number •H »H © p 0V H V 0 2 p *H P I P I © O O •rl a N n o 01 O P a •rl U •rl P 0 • a • ■H p a O rH 0 © K\ ©IN rH a 0 rH O 0 P r* p r- H o o .a © 3 a a H s > o W O' M H Darby I Till 13 80 0 15 0 Tr 5 Tr Tr 16 75 0 15 0 Tr 10 Tr Tr 20 80 0 5 5 5 Tr 5 0 47 75 Tr 15 Tr Tr 10 Tr Tr 48 70 0 15 0 5 10 Tr Tr 49 80 0 15 Tr Tr 5 Tr 0 50 75 0 15 0 Tr 5 5 Tr 51 80 Tr 10 Tr 5 Tr 5 0 52 80 0 10 Tr Tr 5 Tr Tr 53 85 0 Tr 5 5 Tr Tr Tr 54 80 0 0 10 Tr 5 5 0 55 85 0 Tr 5 5 Tr 5 Tr 56 85 0 0 10 Tr Tr 5 Tr Caesar Till 9 70 0 20 Tr 5 5 Tr 0 29 60 0 20 5 Tr 10 0 5 34 80 0 0 10 5 5 0 Tr 35 75 0 15 0 Tr 5 Tr 0 36 55 Tr 25 0 Tr 10 0 5 37 60 Tr 30 0 Tr 5 Tr 5 80 75 0 0 15 Tr Tr 5 Tr 81 85 0 0 15 Tr Tr Tr Tr 82 85 0 10 0 Tr 5 0 0 Boston Till 4 65 0 20 Tr 5 5 0 Tr 31 65 0 20 0 Tr 10 0 5 32 75 0 10 0 Tr 10 Tr 5 39 60 0 25 0 Tr 10 0 5 61 65 0 20 0 Tr 10 0 5 244-

Relative ♦*V SJ-.a ^ a ent

rt 0 0 •rt •rl - •rl O ' ■p -P • - P • •rl 0 PT\ 0 [ N r H a> h r* F i r Sample tt -p -p I • p I O * H N W O n o •rl fn ■P P • u • 0 fO, ocs Number e s d •P r - 0 Si 3 C f 5 * ~ Illite Montmorillonite > o Kaolinite O' H M Rainsboro Till 8A 80 0 10 0 5 5 Tr 0 88 80 0 15 Tr Tr 5 Tr 0 89 75 0 10 Tr 5 10 0 Tr 90 75 Tr 5 5 5 5 0 Tr Loes3 40 20 0 35 0 5 10 0 30 4-1 35 Tr 40 0 0 10 0 15 42 50 Tr 35 0 Tr 10 0 5 45 50 0 45 0 5 10 0 10 44 30 0 45 0 Tr 10 0 15 85 45 0 30 0 0 15 Tr 10 84 60 0 15 0 Tr 15 0 0 85 40 0 25 Tr Tr 15 5 15 86 40 0 40 0 Tr 20 Tr 0 87 30 0 40 0 5 10 Tr 15 Lacustrine sediments 76 80 0 15 Tr Tr 5 Tr Tr 77 75 0 20 0 Tr 5 Tr 0 78 85 0 0 15 Tr Tr Tr 0 79 75 0 15 0 Tr 5 5 Tr All values rounded to the nearest five percent Tr - trace, 0.0 - 2.5# 245 TABLE 22 PARTICLE-31ZE DISTTtlBUTIOK OF BULK SAMPLES Unit/ Percent by weight (>2mra on total sample; e<, sand-silt-clay on less than 2mm fraction) Sample £ {g 36 jiS Humber Sand Silt______Clay 2mm_____ Darby I Till 1 25.7 55.3 19.0 10.2 2 23.1 55.5 21.4 13.1 11 22.7 56.0 21.4 4.5 12 28.2 52.5 19.4 16.4 13 30.3 54.1 15.6 11.9 14 28.5 53.1 18.4 14.5 15 29.7 53.2 17.1 9.8 16 15.9 61.1 23.0 23.3 17 16.6 55.0 28.4 2.6 18 20.7 55.6 23.7 15.0 19 25.8 49.7 24.6 9.6 20 28.2 52.0 19.7 10.8 22 24.7 53.6 21.7 11.0 23 30.0 47.4 22.7 8.8 38 27.7 51.0 21.2 29.7 46 25.3 49.5 25.3 12.2 4? 18.9 53.5 27.6 8.9 48 18.9 5^.1 27.1 13.6 49 23.2 51.6 25.2 12.2 50 22.4 52.9 24.8 13.9 51 24.7 50.8 24.6 11.7 52 27.1 48.4 24.6 13.3 53 27.8 47.9 24.3 10.4 54 24.9 52.0 23.1 12.0 55 28.4 57.7 14.0 12.4 56 30.4 48.4 21.2 9.6 63 17.8 43.7 38.6 0.6 64 27.5 42.7 29.8 7.0 Caesar Till 6 26.8 54.3 18.9 17.8 9 26.1 57.2 16.7 9.9 21 20.6 54.0 25.4 14.0 24 19.7 51.7 28.7 14.1 25 33.0 49.3 17.7 20.0 26 30.5 48.0 21.5 13.1 27 24.0 54.7 21.3 14.4 28 19.5 62.5 18.0 36.4 29 33.4 46.6 20.0 17.0 34 26.2 51.4 22.7 9.4 35 22.7 47.2 30.1 10.4 36 23.2 57.6 19.2 3.2 Unit/ Sample % % S5 * Number Sand Silt Clay 2mm 37 25.8 49.6 24.6 3.4 57 23.0 53.1 24.0 10.6 65 22.3 40.6 37.1 4.1 66 30.0 41.2 28.9 3.9 67 31.2 52.9 15.0 5.9 68 45.6 32.6 21.9 21.0 69 16.0 53.7 30.3 2.8 70 33.6 41.0 25.4 2.8 71 30.0 45.5 24.5 9.0 72 21.7 54.0 24.3 3.3 73 25.6 47.7 26.8 2.3 74 27.4 43.5 29.1 7.1 80 25.7 50.1 24.2 8.0 81 35.1 45.0 19.9 14.1 82 40.5 38.0 21.5 20.4 Boston Till 4 1.6 44.4 54.0 0.5 5 0.5 43.5 56.0 0.0 31 16.8 49.6 33.6 5.5 32 11.8 52.6 35.7 1.8 39 46.1 35.6 18.4 9.4 45 10.7 55.9 33.4 3.7 60 5.2 47.4 47.4 0.5 61 4.3 48.0 47.8 1.4 62 11.0 49.4 39.6 3.3 R&insboro Till 3 19.9 53.6 26.5 2.9 8 14.1 50.8 35.2 11.4 8A 32.8 49.8 17.4 15.7 58 23.6 55.4 21.0 1.7 59 41.0 38.5 20.5 2.3 88 27.6 53.0 19.4 10.9 89 19.8 51.6 28.5 17.1 90 25.8 50.2 24.0 15.4 91 18.1 60.5 22.4 9.9 92 19.3 48.3 32.4 13.8 Loess 40 8.7 50.2 41.1 0.0 41 4.0 52.3 43.8 0.0 42 1.6 48.0 50.4 0.0 43 1.6 54.9 43.5 0.0 44 5.5 57.4 37.1 0.0 33 53.2 37.3 9.5 0.0 247 Unit/ Sample # ■ % £J # Humber______Sand______Silt Clay_____ 2mra 84 68.9 27.3 3.8 0.0 85 15.1 75.3 9.6 0.0 86 12.1 82.1 5.8 0.0 87 24.3 57.0 18.7 0.1 Lacustrine sediments 7 0.5 91.5 8.1 0.0 30 14.6 83.2 2.2 0.0 33 1.8 56.5 42.0 0.2 76 0.4 22.9 76.7 0.0 77 0.4 23.3 76.3 0.0 78 0.1 25.5 74.4 0.0 79 0.5 62.9 36.6 0.0 Miscellaneous 10 10.5 82.9 6.6 0.2 75 40.8 19.3 39.9 9.1 248 TABLE 23 TABULATION OF CIIITTICK-GASOMETHIC AIIALTSHo FOE WEIGHT PERCENTAGES OF CALOTTE AND DOLOMITE IN THE LESS THAN 2am SIZE FRACTION OF HOSS COUNTY UNITS Unit/ Percentages by weight of *2mm fraction Sample % % Total Calcite Numb er______Calcite____ Dolomite Carbonate Dolomite Darby I Till 1 9.9 24.2 36.1 .41 2 8.4 17.9 27.3 .47 11 4.3 6.0 10.8 .72 12 9.4 22.8 34.1 .41 13 8.0 22.9 39.1 .28 14 7.4 26.2 35.8 .28 15 10.6 24.1 36.7 .44 16 7.3 23.6 32.9 .31 17 6.9 19.7 28.2 .35 18 8.9 29.4 40.7 .30 19 8.9 21.5 32.2 .41 20 7.0 12.6 20.6 .56 22 9.2 23.3 34.4 .39 23 8.8 17.9 23.2 .49 38 9.7 29.5 41.6 .33 46 5.8 20.6 25.1 .28 47 4.4 20.7 26.8 .21 43 6.0 21.7 29.5 .28 49 7.4 23.8 33.2 .31 50 6.4 24.6 33.0 .26 51 6.3 24.5 32.9 .26 52 7.0 25.7 34.8 .27 53 7.1 24.6 33.7 .29 54 6.6 24.2 32.8 .27 55 6.0 25.9 34.0 .23 56 6.8 27.0 36.0 .25 63 0.0 0.0 0.0 --- 64 2.4 10.4 15.7 .23 Caesar Till 6 8.0 23.2 33.1 .34 9 6.6 17.3 25.3 .38 21 7.3 22.0 31.1 .33 24 6.8 22.3 51.0 .30 25 9.4 21.4 32.6 .44 26 9.8 27.9 ^0.0 .35 27 8.7 25.3 56.1 .34 23 3.0 21.2 25.0 .14 29 10.6 21.1 .50 34 8.6 18.2 28.3 .47 35 3.1 5.7 9.3 .54 249 Unit/ Sample % % Total Calcite Number______Calcite____ Dolomite Carbonate Dolomite Caesar Till 36 3.3 8.8 12.8 .38 37 5.1 14.6 20.9 .35 57 13.6 20.3 35.6 .67 65 Tr 5.2 5.6 66 8.1 17.4 26.9 .47 67 8.7 18.7 29.0 .47 68 10.9 24.4 37.3 .45 69 1.8 1.8 3.7 1.00 70 3.8 17.9 23.2 .21 71 6.9 19.7 28.2 .35 72 5.7 17.4 24.5 .33 73 6.4 13.5 21.0 .47 74 7.3 17.3 26.0 .42 80 8.2 16.5 26.1 .50 81 8.4 16.6 26.4 .51 82 1.9 15.8 19.0 .12 Boston Till 4 11.4 10.9 23.2 1.05 4A 4.3 6.7 11.6 .64 31 4.4\ 8.7 13.8 .51 32 4.6 9.1 14.5 .51 39 9.2 15.1 25.6 .61 45 1.3 6.0 7.8 .22 60 13.1 11.8 25.9 1.11 61 13.7 14.2 29.1 .96 62 4.6 14.7 20.5 .31 Rainsboro Till 3 3.5 8.3 12.5 .42 8 3.3 3.3 6.9 1.00 8A 6.4 12.2 19.6 .52 58 5.2 7.3 13.1 .71 59 0.0 0.0 0.0 —— 88 8.9 26.1 37.2 .34 89 13.0 17.9 32.4 .72 90 11.5 24.8 36.9 .46 91 6.7 28.0 37.0 .23 92 9.3 20.6 31.6 .45 Loess 40 0.0 0.0 0.0 -— 41 0.0 0.0 0.0 --- 42 0.0 0.0 0.0 ----- 43 0.0 0.0 0.0 ----- 250

Unit/ Sample fi ,*5 Do~al Calcite Number______Calcite_____ Dolomite Carbonate Dolomite Loess 44 0,0 0.0 0.0 -- 83 0.0 0.0 0.0 -- 84 0.0 0.0 0.0 -- 85 2.4 24.7 29.2 .10 86 3.1 17.2 21.7 .18 87 0.0 0.0 0.0 -- Lacustrine sediments 7 3.9 12.2 17.1 .32 30 1.7 8.3 10.7 .20 35 7.9 16.6 25.9 .48 76 3.1 5.0 6.3 1.03 77 5 A 5.1 6.8 1.10 78 1.8 2.9 4.9 .62 79 1.5 5.4- 7.3 .28 t4iscellaneous 10 1.5 0.8 2.4 1.88 75 0.0 0.0 0.0 TABLE 24 HEAVY MINERALOGY OF THE VERY-FINE SAND FRACTION ______OF ROSS COUNTY STRATIGRAEHIC UNITS Opaques Percent Non-opaques g e Pi ■rl O © •H Xi © S © a « P O \ G 9 9 ■P bn at -PK Garnets © (3 ft •P w Pi *rl -p © ■p •H © c © © •H rH © -P 9 o © -P & rH rH u rH H © f t U cd G X •P o © ft cd M a) id •rl cd a 9 -p 9 •rl •rl o Gj • M a U -p ft « ■P P a u © ■P fcO •H a •P ■P 'rl cd cd o •rl rH O 3 a o 8 ft O ft •rl O fcc w CO ft CH ft O EH W S w tc CO w EH Darby I Till fOi o

— .

13 30.4 36.8 00

11.0 • 3.9 14.9 2.3 1.7 35.9 1.3 0.6 1.6 0.9 1.0 04 3.21 16 19.7 27.3 7.6 14.7 22.3 4.1 1.2 38.6 1.4 1.0 0.2 0.9 2.2 0.6 0.2 50.4 2.17 38 26.6 30.4 6.5 16.9 23.4 2.8 4.0 32.0 1.1 1.8 0.1 0.9 1.3 2.0 0.2 53.3 2.31 47 29.4 32.0 6.3 11.3 17.6 3.6 1.1 28.4 3.5 3.6 0.5 3.9 1.3 2.9 1.6 50.4 3.00 49 21.9 29.6 10.4 14.9 25.3 2.0 3.4 33.0 1.4 1.0 — 1.1 1.0 1.6 1.6 45.1 2.49 51 32.3 36.9 5.2 14.2 19.8 1.4 4.0 32.9 0.9 1.1 — 1.0 1.1 0.8 0.1 43.3 2.63 53 26.7 32.8 7.6 13.8 21.4 2.6 2.3 26.2 3.7 2.8 0.1 1.8 2.7 3-3 0.3 45.8 2.81 55 32.0 29.2 5.8 17.4 25.2 2.9 2.5 30.1 2.3 1.3 0.8 1.3 3.0 2.9 0.5 47.6 2.74 Caesar Till 21 21.0 24.5 10.6 13.4 24.0 3.6 2.1 29.7 3.6 2.7 0.6 3.4 3.0 2.2 0.6 51.5 2.90 29 17.8 23.6 8.0 11.6 19.6 1.9 2.9 30.1 4.6 3.8 1.1 4.0 3.1 2.8 2.5 46.8 2.13 35 15.3 21.9 2.7 14.4 17.1 2.4 3.0 39.6 2.9 2.8 0.9 3.6 2.7 2.7 0.4 61.0 4.19 37 24.3 26.0 3.4 14.3 17.7 1.2 4.3 41.8 1.9 1.1 — 2*3 1.6 1.6 0.5 56.3 1.78 80 22.7 24.4 9.7 17.1 25.8 1.0 2.7 40.4 0.8 1.6 0.1 1.0 0.8 1.0 0.1 49.8 2.34 81 27.2 30.3 6.1 13.8 19.9 4.7 2.1 33.8 1.2 0.9 — 1.9 2.4 2.2 0.6 49.8 2.97 82 19.5 27.6 4.5 11.9 16.4 2.4 2.5 36.1 3.4 1.3 0.4 3.3 3.0 2.9 0.7 56.0 2.89 ro Opaques Percent Non-opaques

0) S N f . •h to ® UJ v_^ ® G ® -d ® ® P G G Garnets ® (3 £ t p *ri *H O -p ® -P rH ® -,-t ® •H r H P ® o o P P H g H i-H N .O ctf a G p o • HO osl (4 (4 *H C ® p ® -rH •rH -d o (4 • rf p « ® 4* -P G G P* n P bO -rl to P p f-t ® o iH O P» O O B ft O d ■H*—■* O U>rn en o EH K EC X W CO *3 a f—4 EH Boston Till 4 24.0 26.7 5.3 13.3 18.6 3.1 2.2 38.0 2.7 1.9 0.2 1.8 2.6 1.6 0.6 54.7 1.91 45 25.5 31.9 2.6 19.4 22.0 1.9 2.4 33.2 2.0 1.0 - 2.0 1.4 0.9 1.3 46.1 1.96 60 26.3 30.1 4.3 14.8 19.1 3.4 3.1 25.4 4.2 3.0 1.0 3.9 2.7 1.8 2.3 50.8 2.43 61 30.4 35.2 7.3 12.0 19.3 3.7 2.0 34.8 1.0 0.9 - 0.8 1.0 1.1 0.4 45.5 3.01 62 25.5 27.8 5.1 19.0 24.1 1.8 1.1 39.2 1.3 0.5 - 1.0 0.8 1.7 0.7 48.1 2.40

Hainsboro Till 59 31.1 32.6 6.7 15.4 22.1 3.6 2.3 28.7 2.1 2.2 0.6 2.7 2.6 0.9 0.6 46.3 2.91 88 20.9 28.6 10.2 14.4 24.6 3.0 2.9 30.4 2.6 2.4 — 1.0 1.4 1.2 1.9 46.8 4.04 90 32.1 36.1 6.3 15-0 23.3 2.1 2.2 28.8 2.2 1.4 — 0.8 1.1 1.3 0.8 40.6 3.35 92 28.1 29.0 9.9 16.3 26.2 2.6 2.6 26.1 2.9 2.5 1.0 2.2 3.0 1.1 0.9 44.8 3.49

ro ro TABLE 25 PEBBLE LITHOLOOIES (PERCENT) FOR ROSS COUNTY STRATIGRAFHIC UNITS Percentages Unit/ © © O © © •rl Pebble « © © © 4 3 *H © a P © « a c © P* rH rH p o r-t © X3 fp o o H O © U © rH Count ■H p © (3 O -rl p p © * H © O P © B © P P O a © a P P o E O P O © P 0 X 1 o tS p O © © © EH © Number H E © Eh tn H © rH E H © P O •rl J2 « n O 43 •H H & © P R O O m 03 W CO O H O Till 1 4-3 30 1 74 .77 7 2 11 20 5 2 7 8 44 35 0 79 .79 3 1 9 13 5 3 8 9 42 24 2 68 .61 12 2 6 20 8 4 12 10 54 22 6 82 .51 8 0 4 12 4 4 8 11 60 27 0 87 .45 2 0 20 22 6 4 10 12 50 28 0 78 .56 4 4 4 12 4 6 10 13 42 31 2 75 .78 6 8 7 21 2 2 4 58 64 19 3 86 .34 3 2 1 6 6 2 8 59 60 21 1 82 .36 7 4 1 12 4 2 6 >sar Till 3 34 21 0 55 .61 11 9 20 40 4 1 5 15 52 24 2 78 .50 7 5 6 18 3 2 5 16 36 16 4 56 .55 14 9 10 24 18 2 20 19 23 31 6 60 1.60 11 6 7 24 10 5 15 60 67 18 4 89 .32 2 2 0 4 6 1 7 65 63 22 1 86 .36 2 3 0 5 7 2 9 66 69 15 3 87 .26 3 5 0 8 4 1 5 67 59 21 2 82 .38 4 2 0 6 8 4 12 68 67 19 1 86 .29 5 2 2 9 5 0 5 ru Ul v» U) o © GQ ■p -H £3 o © Fh © .cj •rl Unit/ © c -P © © c rH o i H Rt A ■P o 09 c—I i—1 Pebble *d -p at e o *rC -p a O Cl © 5 ta +» ■p o B to o a -P -P V u °*P + O © © o © Count ° S © ^ Fh H +> >» •rl o ,d at to O © Total Si © f-l Shale Siltstone Number « Hi O o h ! Pi CO Clastics w O Boston Till 1 56 16 11 63 .75 15 0 13 28 9 0 9 2 16 5 5 26 .62 14 0 15 29 37 8 45 56 39 31 2 72 .84 8 10 5 23 4 1 5 57 38 41 1 80 1.10 5 7 1 13 5 2 7 63 32 39 0 71 1.20 9 5 3 17 8 4 12 64 42 30 2 74 .76 8 9 2 19 4 3 7 Rainsboro Till 61 55 24 2 81 .47 5 3 2 10 5 4 9 62 61 17 0 78 .27 7 4 2 13 7 2 9 69 6? 21 0 84 .33 6 2 1 9 5 2 7 70 54 26 1 81 .50 8 3 2 13 6 0 6 Worthington Outwash 45 47 29 5 81 .72 5 2 1 8 9 2 11 44 45 24 5 81 .60 5 1 2 8 13 7 20 45 57 50 3 70 .89 7 1 5 13 12 5 17 46 45 51 4 77 .77 2 0 3 5 9 9 18 47 48 26 2 76 .58 6 2 2 10 10 4 14 48 54 20 6 80 .48 7 1 1 9 6 5 11 49 49 23 2 74 .51 6 3 1 10 7 9 16 to O © (0 p •rl £ © « u © ja *iH Unit/ 0) H -p © © a a rH p O rl © Count r~i © O 0} B © EH H © H o © p Eh ^ O •H si © m O s i *H EH rH to © Pi o Hi Uumber o o piR Sandstone CO CO O H O Circleville Outwash 56 57 14 2 73 .28 6 2 6 14 12 1 13 37 60 16 1 77 .28 5 0 2 7 9 7 16 33 55 22 4 81 .47 7 0 3 10 8 1 9 59 48 27 0 75 *56 7 1 4 12 6 7 13 40 51 30 3 84 .64 4 0 5 9 5 2 7 41 53 24 3 80 .50 3 2 6 11 5 4 9 42 4-9 26 1 76 .55 4 0 2 6 10 8 18 Kingston Outwash (Scioto River valley) 25 49 19 3 71 .44 6 0 4 10 12 7 19 24 50 22 4 76 .52 6 1 5 12 10 2 12 2p 4-3 28 4 75 .74 8 0 3 11 8 6 14 26 52 25 4 81 .55 9 0 0 9 8 2 10 27 59 20 1 80 .35 5 0 3 8 5 7 12 28 4-7 27 3 77 .65 6 1 6 10 9 4 13 29 49 22 3 74 .51 4 0 1 5 14 7 21 30 51 30 2 83 .62 5 0 1 6 7 4 11 Kingston Outwash (Paint Greek valley) 31 69 16 1 86 .24 6 3 1 10 3 1 4 32 75 13 3 91 .21 5 2 1 8 1 0 1 33 76 11 2 89 .17 4 2 0 6 4 1 5 3* 70 12 0 82 .17 2 4 2 8 7 3 10 35 74 12 1 87 .18 5 3 1 9 4 0 4 o o u rH i—1 Pebble h -p o cd n) £ oa e -P -P Count ,3 § of O Cfl Total -p EH r^J O * H Total 1-—*0) rt Sandstone Chert Carbonates Dolomite Shale Ls + Ls + chert Clastics Siltstone i. umber Q >4 Igneous pi'-i o i i i Kigby Outwash 4 39 20 2 61 .56 18 2 14 34 4 2 6 5 21 17 6 44 1.09 17 3 18 38 11 8 19 50 49 22 3 84 .51 3 1 1 5 8 3 11 51 57 17 4 78 .36 5 0 2 7 7 8 15 52 54 21 4 69 .46 4 0 3 7 15 9 24 55 51 20 2 73 .43 6 2 2 10 12 5 17 54 60 15 0 75 .25 6 0 0 6 10 9 19 55 53 26 1 80 .50 4 1 3 8 9 3 12 i-contact deposit! 6 56 22 4 82 .46 8 0 2 10 8 0 8 14 49 33 2 84 .71 4 1 3 8 4 4 8 14A 51 28 1 80 .55 5 1 4 10 5 5 10 143 52 21 0 73 .40 6 2 4 12 9 6 15 14C 47 28 3 78 .66 4 2 3 9 6 7 13 17 28 12 5 45 .60 36 0 1b 52 1 2 3 1?A 50 19 4 53 .76 28 0 13 41 3 3 6 173 29 26 5 GO 1.08 19 1 16 36 3 1 4 18 41 35 3 79 .92 8 1 4 13 6 2 8 20 31 31 6 68 1.20 16 2 8 26 6 1 7 21 36 43 1 80 1.20 10 0 3 13 4 3 7 22 40 29 2 71 .77 11 0 3 14 8 7 15 71 52 26 3 81 .56 7 0 4 11 5 3 8 72 56 24 4 84 .50 8 0 1 9 5 2 7 ru v cr> 257 :'Joetion C

sections with significant stratigraphies

Distances are striaght line measurements from town centers. The following abbreviations are U3ed to indicate the various field and laboratory analyses which were performed on the samples: MA - mechanical analysis, CM - clay-mineral analysis, PC - pebble count, CD - calcite- dolomite analysis, HM - heavy-mineral analysis, and TP - till-fabric analysis.

Locality

6 500' SW Jet Bier's Run-Cattail Rds, 0.25 mile NW BH 738, stream cut S side Bier's Run, Union Twp.

feet description 5 alluvium, sandy gravel, nonclacareous 20+ Caesar Till, blue-gray (10YR5/1)* calcareous, logs near base MA, CD, PC, TF 760* stream

7 2.0 miles HE Massieville, 0.25 mile N Rt 25, 250' NW Jet Three Locks RcL-Rt 23, Scioto Twp. feet description 6*1 loess, noncalcareous, 10YR6/6, MA, CD 17.0 sand and gravel, coarse, calcareous, PC 12.5 silt-clay, varved, 10YR5/4* MA, CD 3.0 Rainsboro Till, calcareous, 7.5YR4/1, HA, CD, TP 25.0 sand and gravel, fine-3and lenses, cross-beaded, PC 1.0 organic zone, black (7.5YH2/1), wood fragments, [In-oxide crust, HA, CD S^rQ* pit floor 258

Locality 24 E bank Anderson .inn, 200* E Maple Grove Rd, 1.0 mile NE Anderson, Concord Twp, feet description 18 Caesar Till, yellowish-blue gray (10YR 5/1), MA, CD 1 forest bed, matted logs, uneven upper surface 10+ sand and gravel, coarse, cross-bedded 705* stream

58 S side B&O railroad cut, 1.5 miles NW Lickskillet Section 16, Liberty Twp. feet description 4 loess, noncalcareous, 10YR6/5 15 Rainsboro Till, light gray (7.5YR5/4), calcareous, MA, CD, HU, PC, TF 1 oxidized sand, rounded cuartz pebbles, 10YR6/6 sandstone with finely striated upper surface 645* railroad bed 76 N side B&O railroad cut, 0.5 mile NW Lickskillet Section 16, Liberty Twp. feet description 5 mixed zone, loess/colluviua, non­ calcareous 14 oxidized varved clay, 10YR5/4 - 10YR5/3, white CaCO* stringers, few pebbles, HA, CD, CM 25 unoxidized varved clay, 5YR4/1 - 5YR4/2, MA, CD, CM 5 slumped zone 710* railroad bed 259

Locality

80 E bank Anderson Run, 1.4 miles 1 7 : 1 Anderson, Union Twp. feet description 5 mixed zone, loess/colluviun, dark brown (7.5YR4/2) 11 oxidized Caesar Till, yellowish- brown (10YR5/4), MA, CD, TP, HM, CM 4 unoxidized Caesar Till, calcareous, compact, dark gray (7.3YR4/1), MA, CD, CM, PC, TP, HM 3 colluvium/outwash/alluviua, coarse, many sandstone fragments 11 Caesar Till, light gray (7.5YR7/1), friable, small wood fragments, MA, CD, PC, TP, HM, CM 3 sand and gravel, fine, calcareous 9 Caesar Till, very compact, dark gray (7.5YR4/1), wood, MA, CD, TP, HM, PC 750* Anderson Run

83 Gravel pit, 0.3 mile NE Charleston Pike, NEK Section 20, Springfield Twp. feet description 4-8.5 loess, lower portion calcareous and fossiliferous, many secondary CaCOx accumulations, 10YR5/4, MA, CD, CM, HM 40 Illinoian Higby Outwash, cross-bedded sand and gravel, oxidized, wood 745* pit floor BIBLIOGRAPHY

American Society for Testing Materials, 1964, Grain-size analysis of soils: D-422-63 in Procedures for testing soils, p. 95-106. Anderson, R. C., 1957* Pebble and sand lithology of the major Wisconsin glacial lobes of the central lowland: Geol. Soc. Amer. Bull., v. 68, p. 1415- 1449. Andrew, R. W. Jr., 1960, The relationship of natural drainage and the clay mineralogy of the Miami and Brookston soils in central Ohio: unpubl. M.Sc. thesis, The Ohio State University, 61p. Baker, F. C., 1920. Pleistocene Mollusca from Indiana and Ohio: Jour. Geol., v. 28, p. 439-457. Bhattacharya, N., 1962, Weathering of glacial tills in Indiana: Geol. Soc. Amer. Bull., v. 73, p. 1007- 1C20. Bidwell, 0. W.. 1949, A study of clay minerals of soils in the Miami catena: unpubl. Ph.d. thesis, The Ohio State University, 77p. Campbell, L. J., Bleuer, N. K., Gray, H. H . , Powell, S. L. and Swadley, W. C., 1974, Late Cenozoic features of the middle Ohio River valley: Spring Field Conf. Guidebook, Geol. Soc. Kentucky, publ. by Kentucky Geol. Surv., Lexington, p. 3* Campbell, M. R., 1918, The country around Camp Sherman: text on back of topographic map (Ohio - Camp Sherman quad.), U. S. Geol. Surv. Carman, J. E*, 194-7* Geologic section of the Chillicothe test-core section: Ohio Jour. Sci., v. 47, p. 49- 5*. ------( 1955, Revision of the Chillicothe test- core section: Ohio Jour. Sci., v. 55, p. 65-72.

260 261

Chamberlin, T. C., 1383, Preliminary paper on the terminal moraine of the second glacial epoch: U. S. Geol. Surv. Third Annual Sept., p. 291-402. Connally, G» G.t 1964, Garnet ratios and provenance in the glacial drift of western Hew York: Science, v. 144, p. 1452-1453. Davis, W. M., 1884, Gorges and waterfalls: Amer. Jour. Sci., 3rd ser., v. 28, p. 123-132. Denny, C. S. and lyford, W. H., 1963, Surficial geology and soils of the Elmira-Williamsport region, New York and Pennsylvania: U. S. Geol. Surv. Prof. Paper 379, 60p. Dixon, W. J. and Massey, F. J., 1957* Introduction to statistical analysis: McGraw-Hill Inc., New York, 488p. Drake, L. D., 1968, Till studies in New Hampshire: unpubl. Ph.d. thesis, The Ohio State University, 112p. Dreimanis, A., 1960, Pre-classical Wisconsin in the eastern portion of the Great Lakes region, North America: Internatl. Geol. Cong., 21st, Copenhagen 1960, Rept. Section 4, p. 109-119. ------1962, Quantitative gasometric determination of calcite and dolomite by using Chittick apparatus: Jour. Sed. Petrol., v. 43, p. 520-529* ------# Reavely, G. H., Cook, R. J. B., Knox, K. S. and Moretti, F. J., 1957, Heavy mineral studies in tills of Ontario and adjacent areas: Jour. Sed. Petrol., v. 27, p. 148-161. ------and Vagners, U. J., 1969, Lithologic relation of till to bedrock in Wright, H. E. Jr., ed., Quaternary geology and climate: Natl. Acad. Sci. Publ. 1701, p. 95-93. ------— , 1971, Bimodal distribu­ tion of rock and mineral fragments in basal tills in Goldthwait, R. P., ed., Till/a symposium: Ohio i5£ate Univ. Press, Columbus, p. 237-250. 262

Dreiuanis, A. and Goldtkwait, 3. , 1975, .Wisconsin glaciation in the Huron, Erie, ana Ontario Lobe3 in Black, R. P., Goldthwait, R. , ana Hillman, ITT B., eds., The Wisconsinan Stage: Geol. Soc. Amer. Mem. 136, p. 71-106. Droste, J. B., 1956, Clay minerals in calcareous till in northeastern Ohio: Jour. Geol., v. 67, p. 187-190. Burrell, R. H*, 1956, Illinoian boundary in southwestern Ohio and northern Kentucky: Geol. Soc. Amer. Bull, v. 27, p.1751. ------, 1961, The Pleistocene geology of the Cincinnati area in Guidebook for Field Trips, Cincinnati Meeting, Geol. Soc. Amer., p. 45-57* Evensonj E. B., 1971, The relationship of macro- and micro-fabric and the genesis of glacial landforms in Goldthwaitf R. P., ed., Till/a symposium: TSEio State Unxv. Press, Columbus, p. 335-366. Pairbridge, R. W., 1968, Quaternary Period in Fairbridge, R. tf., ed., The Encyclopedia of Geomorphology: Reinhold Book Corp., Hew York, p. 912-931* Forsyth, J. L., 1956, The glacial geology of Logan and Shelby counties, Ohio: unpubl. Ph.d. thesis, The Ohio State University, 203p. ------, 1961, Wisconsin glacial deposits: Geol. Soc. Amer., Guidebook for Field Trios, Cincinnati, p. 58-61* ------f 1965, Contribution of soils to the mapping a ia interpretation of Wisconsin tills in western Ohio: Ohio Jour. Sci., v. 65, p* 220-227* Foster, J. W., 1950, Glaciations and ice lake of Paint Creel: valley, Ohio: unpubl. M.Sc. thesis, The Ohio State University, 57p* Fowke, G., 1895, Preglacial and recent drainage changes in Ross County, Ohio: Bull, of Sci. Lab of Dennison Univ., v. IX, pt. 1, p. 15-24. Frye, J., Glass. H., Kempton, J., and Willman, H., 1969, Glacial tills of northwestern Illinois: 111. Geol. Surv. Circ. 437, 45p. 263

Frye* J*, Willman, H. B., and Black, R. F., 1965, Outline of glacial geology of Illinois and Wisconsin in Wright, H. E. Jr. and Frey, D., eds., The Quaternary of the United States: Princeton Univ. press, Princeton, p. *1-3-61 • Goldthwait, R. P., 194-7, Geological history of Sandy Bottom Run and vicinity; Higby, Ohio: unpubl. rept. to Norfolk and Western Railroad Co., 21p* — ------, 1952, Guidebook of the 1952 Pleistocene field conference of the Friends of the Pleistocene: minted by Ohio Div. of Geol. Surv., 14p. ------1955, Pleistocene chronology of south­ western Ohio in Guidebook of fifth biennial field conference (iTeistocene), pt. 2 , p. 35-72. ------, 1958, Wisconsin age forests in western Ohio; I. Age and glacial events: Ohio Jour. Sci., v. 58, p. 209-219. ------, 1959, Leached clay-enriched zones in post-Sangamon drift in southwestern Ohio and south­ eastern Indiana: a reply: Geol. Soc. Amer. Bull., v. 70, p. 927-928. ------— , 1962, The 1962 field conference of the Friends of the Pleistocene guidebook, 26p. —------— , 1968, Two loesses in central southwest Ohio in The Quaternary of Illinois; Univ. Illinois Spec/Tubl. 15, p. 41-4-7. ------1 White, G. W., and Forsyth, J. L., 1961, Glacial map of Ohio: U. S. Geol. Surv., Misc. Geol. Invest. Map 1-316. — — and Forsyth, J. L., 1965, Ohio in Great Lakes-Ohio River valley: Guidebook for field conference G: INQUA VII Cong*, 110p. •— ----- , Dreioanis, A., Forsyth, J. L., Karow, P. F., and White, G. W., 1965, Pleistocene deposits of the Erie Lobe in Wright, H. E. Jr. and Frey, D. G*. eds., The Quaternary of the United States: Princeton Univ. press, Princeton, p. 85-97. 264 Goldthwait;, R. P. and Rosengreen, T. E., 1969, Till stratigraphy from Columbus southwest to Highland County, Ohio: field trip guidebook; Worth-central section, Geol. Soc. Amer., p. 2-1 - 2-1?. Gooding, A. M., 1963, Illinoian and Wisconsin glaciations in the Whitewater basin, southeastern Indiana and adjacent areas: Jour. Geol., v. 71, p. 665-682. ------. Thorp, J., and Gamble, E., 1959, Leached clay-enriched zones in post-Sangamon drift in south­ western Ohio and southeastern Indiana: Geol. Soc. Amer. Bull., v. 70, p. 921-925- Hildreth, S. P., 1834, Ten days in Ohio: Amer. Jour. Sci., v. 25, a. 2, p. 217-257- Holowaychuk, N., 1950, Clay mineral studies of soils of the Miami, Hermon, and Worthington catenas: unpubl. Ph.d. thesis, The Ohio State University, 64p. Horberg, L., 1950, Bedrock topography in Illinois: 111. Geol. Surv. Bull., v* 73, p. 41-111.

Hubbard, G. D., 1954-, Terrace interpretation in south­ eastern Ohio: Ohio Jour. Sci., v. 54, p. 365-377- Hyde, J. E., 1921, Geology of the Camp Sherman quadrangle: Ohio Geol. Surv. Bull. 23, 190p.

— ------, 1953, Mississippian formations of central and southern Ohio: Ohio Div. Geol. Surv. Bull. 51, 355p- Ireland, H. A., 1943, Pre-Illinoian glaciation in south­ eastern Ohio: Ohio Jour. Sci., v. 43, p. 180-181. Jillson, W. P., 1924a, Glacial pebbles in eastern Kentucky: Science, v. 60, n. 1544, p. 101-102. ------, 1924b, Glaciation in eastern Kentucky: Pan Amer. Geol., v. 42, p. 125-132. Johns, W. D., Grim, R. E., and Bradley, W. P., 1954, Quantitative estimations of clay minerals by diffraction methods: Jour. Sed. Petrol., v. 24, p. 242-251- Johnson, W. K., 1964, Stratigraphy and petrography of Illinoian and Kansan drift in central Illinois: 111. Geol. Surv. Circ. 373, 33?. 265 Kempton, J. and Goldthwait, 3. P., 1959* Glacial outwash terraces of the Hocking and Scioto River valleys, Ohio: Ohio Jour. Sci., v. 59* p. 135-151. LaRocque, J. A. A., 1968, Pleistocene mollusca of Ohio: Ohio Div. Geol. Surv. Bull. 62, pt. 5, p. 357-553.

------and Forsyth, J. L., 1957* Pleistocene molluscan faunules of the Sidney cut, Shelby County, Ohio: Ohio Jour. Sci., v. 57* p. 81-89. Leighton, M. H. and Brophy? J. A., 1961, Illinoian glaciation in Illinois: Jour. Geol., v. 69, p. 1-31. ------and Ray, L. L*, 1965, Glacial deposits of Nebraskan and Kansan age in northern Kentucky: U. S. Geol. Surv. Prof. Paper 525-B, p. B126-B131. Leonard, A* B., 1953* Molluscan faunules in Wisconsin loess at Cleveland, Ohio: Amer. Jour. Sci., v. 251, p. 369-376. Lessigf H. D,, 1961, Soils of the high terrace remnants m the upper Ohio Valley: Ohio Jour. Sci., v. 61, p. 25-37. Leverett, F., 1902, Glacial formations and drainage features of the Erie and Ohio basins: U» S. Geol. Surv. Mono., v. 41, 802p. — ------, 1929, The Pleistocene of northern Kentucky: Ky. Geol. Surv., ser. VI, v. 31, p. 1-80. ------, 1942, Gravel outwash near Chillicothe, Ohio: Science, v. 95* p. 528-529. Lineback, J. A., 1971* Pebble orientation and ice move­ ment in south-central Illinois in Goldthwait, R. P. ed., Till/a symposium: Ohio Sta^e Univ. press, Columbus, p. 328-334. Manos, C., 1961, Petrography of the Teays-Mahomet Valley deposits: Jour. Sed. Petrol., v. 31* p. 456-465* Melvin, J. H., 1933* The geology of a portion of the Piketon, Ohio quadrangle: unpubl. M.Sc. thesis, The Ohio State University, 87p* Merrill, W. K., 1953* Pleistocene history of part of the Hocking Valley: Ohio Jour. Sci., v. 53* P- 143-153. 266 Herrit4;, E. 3. and Muller, E. H*, 195?, Depths of leaching in relation to carbonate content of till in central Mew York State: Amer* Jour. Sci*, v. 257, p* 4&5-430. Moos, H. H., 1970, The age and significance of a paleo3ol in Payette County, Onio: unpubl* M.3c* thesis, The Ohio State University, 38p. Morner, IT. A. and Dreimanis, A*, 1973. -he Erie Inter- scade in 31ack, R. P., Goldthwait? R. P., and Willman, H. B>, eds*, The Wisconsman Stage: Geol. Soc. Amer. Mem. 136, p. 107-134-. Muller, E. H., 1965, Quaternary geology in Hew York in Wright, H. E. Jr. and Prey, D. H., eds., The Quaternary of the United States: Princeton Univ. press, Princeton, p. 99-112. Hewberry, J. S., 1874, Report of the Geological Survey of Ohio, v. II, p. 41-46. Norris, 3. 2., Cross, tf. P., and Goldthwait, R. P., 1950, The water resources of1 Greene County, Ohio: Ohio Div. of Water Bull. 19, 52p. Orton, 2., 18?4, Report on the third district: geology of Ross County, Ohio: Ohio Geol. Surv., v. 2, pt. 1, p. 643-658. Patton, L. T. and Hicks, C., 1925, Notes on the occurrence of glacial material beyond the border of the drift in Muskingum County, Ohio: Ohio Jour. Sci., v. 25, P. 97-98. Petro, J. H., Shumate, W. H., and Tabb, M. P., 1967, Soil survey of Ross County, Ohio: Ohio Dept, of Natl. Res. and Soil Conserv. Serv., U* 3. Govt. Printing Office, Washington, D. C., 1b8p. Quinn, M.J., 1972, The glacial geology of Champaign County, Ohio: unpubl. H.Sc. thesis, The Ohio State University, 94p. Ramsden, J. and Westgate, 2., 1971T Evidence for reorientation of a till fabric in the Edmonton area, Alberta in Goldthwait, R. P., ed., Till/a symposium, Chio 3ta7e Univ. press, Columbus, p. 335-344. 26? Hay, L. L. ? 1969, Glacial erratics and the problem of glaciation in northeast Kentucky and southeast Ohio - a review and suggestions: U. S. Geol. 3urv. Prof. Paper 650-D, p. D194-D199. Seed, E., Dreezen, V., Bayne, C., and Schultz, C., 1965, The Pleistocene in Nebraska and northern Kansas in Wright, H. E. Jr. and Prey, D. H., eds., The Quaternary of the United States: Princeton Univ. press, Princeton, p. 187-202. Reynolds, M. B., 1959, Pleistocene Molluscan fauna of the Humboldt deposit, Ross County, Ohio: Ohio Jour. Sci., v. 59* p. 152-166. Richard, B. H., 1974, Relationship of Cedar Bog to the ancestral Teays River system^ p. 3-7 in King, C., ed., Cedar Bog symposium: Ohio Biol. Surv* Info. Circ. 4, 71p- Rogers, J. K», 1936, Geology of Highland County: Ohio Geol. Surv. Bull. 38, 148p. Rosengreen, T. E., 1970, The glacial geology of Highland County, Ohio: unpubl. Ph.d. thesis, The Ohio State University, 163p. ------1974, Glacial geology of Highland County, Ohio: Ohio Div. Geol. Surv. Rept. of Invest., n. 92, 36p. Schmidt, J. J., 1934, The water resources of Ross County, Ohio: State of Ohio, Div. of Water, Info. Circ., n. 4, 25p. ------t 1961-62, Underground water resources - Ohio water plan inventory: Scioto River Basin (index M-8, M-11, M-15); Paint Creek Basin (index M-12, M-13, K-14}; Deer Creek Basin (index M-7)i and Salt Creek Basin (index M-10): Ohio Dept. Nat. Res., Div. of Water publications. Shepp3, V. C., 1953, Correlation of the tills of north­ eastern Ohio by size analysis: Jour. Sed. Petrol., v. 23, p* 34-48. Stauffer, C. R., 1909, The middle Devonian of Ohio: Ohio Jour. Sci., 4th ser., Bull. 10, 204p. 268

Steiger* J. 2., 1887, riechanical and carbonate analysis of 3oil parent material derived from glacial tills and lacustrine deposits in western Ohio: unpubl. H.Sc. thesis, The Ohio State University, 127p. ------and Holowaychuk, H., 1971, Particle-siae and carbonate analysis of glacial till and lacustrine deposits in western Ohio in Goldthwait, R. P., ed., Till/a symposium, Ohio S"Uate Univ. press, Columbus, p. 275-289. Stout, V/., 194-1, Dolomites and limestones of western Ohio: Ohio Geol. Surv. Bull. 4-2, 4th ser., p. 67-70.

------and Schaaf, D., 1951, The Minford silts of southern Ohio: Geol. Soc. Amer. Bull., v. 42, p. 665-672. ------and Lamb, G. P., 1958, Physiographic features of southeastern Ohio: Ohio Jour. Sci., v. 58, p. 49-83. ------, Ver Steeg. K., and Lamb, G. P., 1943, Geology of water in Ohio: Ohio Geol. Surv. Bull. 44, 154p. Suggate, R. R., 1974, When did the last interglacial end?: Quat. Res., v. 4, n. 3, p. 246-252. Teller, J. T«, 19&4, The glacial geology of Clinton County Ohio: unpubl. M.Sc. thesis, The Ohio State University, 107p. -— , 1967, Glacial geology of Clinton County, Ohio: Ohio Div. Geol. Surv., Rept. of Invest., n. 67, map with text. , 1970, Early Pleistocene glaciation and drainage in southwestern Ohio, southeastern Indiana, and northern Kentucky: unpubl. Ph.d. thesis, University of Cincinnati, 115p. Thornbury, W. D., 1965, Regional geomorphology of the United States: John Wiley and Sons, Inc., Kew York, 609p. Thwaites, P. T., 1946, Outline of glacial geology: Edwards Brothers, Ann Arbor, 129p. Tight, 17. G., 1895a, Some preglacial drainage features of southern Ohio: Bull, of Sci. Lab of Dennison Univ., v. IX, pt. 2, p. 22-32. 269 Tight, G., 1S9pb, A preglacial tributary to Paint Creel1: and its relation to the Beech Plats area of Pike County, Ohio: Bull, of 3ci, Lao of Dennison Univ., v. IX, pt. 1, p. T >-34. ------, 1903* Drainage modifications in south­ eastern Ohio and adjacent parts of West Virginia and Kentucky: U. S. Geol, Surv, Prof. Paper 13, 111p. Totten, S. M., 1969* Overridden recessional moraines of north-central Ohio: Geol. Soc. Amer. Bull., v. SO, n. 10, p. 1931-1946. Van Tuyl, D. W. and Bernhagen, R. J., 1947, Summary of ground water conditions in Ohio: Ohio Water Res. Board Bull. 5* ?2p. Walker, A. C., Schmidt, J. J., Stein. R. B., Free, H. L., and Bailey, N. G., 1963* Ground water for industry in the Scioto River valley: Buried Valley Invest., n* 1, State of Ohio, Div. of Water, 29p. Wayne, V/. J., 1956, Thickness of drift and bedrock physiography of the Wisconsin glacial boundary: Ind. Geol. Surv. Rept. of Progress, n. 7, 70p. — and Zumberge, J. H., 1965j Pleistocene geology of Indiana and Michigan in Wright, H. E. Jr. and Frey, D. H., eds., The Quaternary of the United States: Princeton Univ. press, Princeton, p. 63-84. White, G. W., 1969* Pleistocene deposits of the north­ western Allegheny Plateau, U. S. A.: Quart. Jour. Geol. Soc. London, v. 124, p. 131-151. ------Totten, S. M., and Gross, D. L., 1969* Pleistocene stratigraphy of northwestern Pennsylvania: Pa. Geol. Surv. Bull. G55* 88p. Whittlesey, C., 1838^ Topographical report (on the region between the Scioto and Hocking Rivers): Ohio Geol. Surv., 1st Annual Rept., p. 99-109. Wilding, L. P., Jones, R. E.. and Schafer, G. M., 1965* Variation of morphological properties within Miami, Celina, and Crosby mapping units in west-central Ohio: Soil Sci. Soc. Amer. Proc., v. 29, p. 711- 717. PLATE I

EXPLANAT ION

Alluvium )iiu« a : Iff

■•df End moraine name mt,-* :*9ansa

Wfl Ground moraine H H3

WtfC '■ Till over ice-contact teatu* ■Mi

< Ice - contact feature* r M W Bamtondge Wok King at on u Wo Outwasn. Woe - C'rciev >'e WOW- Worthington , a e .*1 Wot erosion a te O

Lacustrine deposit

it Till thin and patchy 0 a

lie ice-contact features a c

Outwash, lOh-Higby 270

Jillman, H. E. , Glass, 2. D. , and j'l'yc, 7. 0., 'iCGf, Mineralogy of glacial rills and Tf.e:r v/Gathering profiler in Illinois; 2, gloria] s-Ils: 111. Geol. Surv. Circ. 400, Cop. Wright, G. P., 1390, The glacial boundary in v/estern Pennsylvania, Ohio, Kentucky, Indiana, and Illinois (with introduction by T. C. Chamberlin): U. S. Geol. Surv. Bull. 58, 112p. Y/right, H. S. Jr. and Ruhe, R. V., 1965, Glaciation o Minnesota and Iov/a in Wright, H. E. Jr. and Frs

D. K., eds., The Quaternary of the United State Hi *

\ ~ J Yeliowbud 4 3- Reesevi ite 2 Knockemstitf1 pcfc moraines specific Drift Lattaviiie eae to related THE GL A CIAL GEO

Ge o I o g y by

to nes rEOLOGY OF ROSS COUNTY, OHIO

(fly by Michael J. Quinn 1974 o

Boundary of a mappe

• ■ I *J i i i i r • ~ J- Distal boundary of

h m i l of a ma/or Isis

Seal*

Contour interval

Datum is me a 9

E X P L A N A T io n

______Boundary of a mapped deposit dashed where approximate

^ Distal boundary of the Reesevitie Moraine pos.lu

.. - Limit Of a major ,s,age, g.ac.a-■ advance, dashed where approximate

Scale i, es

Contour intervals 5, 10, and 20 teet

Datum is mean sea level

i A N A T I O N

app'oxi mate

Locat ton

»b«d where approximate

N*lt Cwaty klhnoMn Pi l tic u s. loh - Out ash. w H igby t c e - c o n t a c if e a t u r e s autie deposi Lacustrinet Lacustrine deposit $ f < > i 11 / -) b* y r“ ' V/ t • .5 • * t / , V - V . x■ „ ^ . /*' - J / . f X,/ ' 7 ^ ■ '7*^1 r 1 ■ ' " ' ' V ^ t \ t ' \ . y J ■ C'£*Y' Y * £ ' C B V , f j * j ^ ^ -V"'/ . X r ^ r / J; , ■ O . -/ >5,

V^,- 4

I -~-4 ‘

& ■ < * ? $

b e c o n s i PUATE II

N 900

•SO

•00

7 0 0

CIRCLEVILLE*

WORTHIMGTOK

SCIOTO RIVER

6 0 0

S § 0 oF BOSS COOHTV OUTW.SH bECo n stb o c t 60 pR0F''" t e ^ B A C E S SCIOTO RIVER

# TERR*C* ® 0 silit ****

A HlKl*0* ‘1

• A UNTY OUTWASH TERRACES :s

Legend

• TERRACE ELEVATIONS LT -LOWER TWIN CREAK TERRACES 0 SILT TERRACES NF NORTH FORK VALLEY TERRACES _ ALLUVIAL TERRACE ( ) LOCATIONS ALONG THE XLLINOIAN 0 Cjfl OUfWASH TERRACE HORIZONTAL A MINIMUM /ELEVATION

HORIZONTAL SCALE I INCH - I MILK VERTICAL EXAGGERATION * 200x

700

67S

650 --- -

625 JSL CIRCLEVILLE WORTHINGTON -EL J5L 600 •C ••MW « t © / T

/ 5 / o e T 1 W £ £ •-3 p I ” H ” s i s I H tn - ► h o IN U er re Valley* Creek Deer Anderonville MoundCity ainl Mon.National IN IN O e © o T Mound City National Mon.

"0 FRANKFORT > ~ (Seymourvilll Chillicothe • • Bourneville o Lower Twin X Creek m m 7;

m x Paint Creak * SULFUR LICK > o m c o

u

I Three Locks * .1M to X • n jo SLATK MILLS

Rupefs £ 3 NORTH FORK VALLEY l£» (Londonderry) I

■ ■ M M m 0)

3 Three Locks ► •! I oto > SLATE HILLS o I Rupels IB f NORTH FORK VALLEY (3 (Londonderry)

t* 31 H H

► I North Fork w PAINT CR./NQRTH FORK JUNCTION w (Richmondale) T | L

1 ► 0 s

Higbyl A w M Ross - Pike

Chillicothe i Cf» I / / «* 1* U N / o «* os u s / o