Id '/'J-.i ji , >'\ wmmi^mmmmmmmmi\immoii:'m^^^^^^^ LIBRARY OF THE UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN 010.72 no. 7 - 3 LIBRARY ONE lOAH The person charging this material is re- sponsible for its return to the library from which it was withdrawn on or before the Latest Date stamped below. Theft, mutilation, and underlining of books ore reasons for disciplinary action and may result in dismissal from the University. UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN AUG 2 7 m SEP 12 JON 1 8 tel J\]N 2 3 V. SEP 1 8 19^5 SEP 2 JUL 5 AUG 2 1 ID6 01 13S6 1 fji> (J J 988 1989 L161 — O-1096 Digitized by the Internet Archive in 2011 with funding from University of Illinois Urbana-Champaign http://www.archive.org/details/someobservations07alex .# i?^ .-♦^' ,v OCCASIONAL PUBLICATIONS OF THE DEPARTMENT OF GEOGRAPHY SOME OBSERVATIONS ON THE LATE PLEISTOCENE AND HOLOCENE HISTORY OF THE LOWER OHIO VALLEY by CHARLES S. ALEXANDER The Ubranr of the OCT 21 ^^^^ universiiy of (llinois at Urfc3"a-^*v5rT!Dal8n PAPER NUMBER 7 APRIL 1974 i.OO 38 .41 .89 1.00.63 .76 LOO .64 1.00 — 23.2 .71 1002 8.7 Z9A .29 996 9.2 51.3 .31 1291 3.1 ♦ • • • • • • • • ' • • • 28.1 .66 771 2.1 \ ^^VC.v N .ss . XW ♦S'- GARY O. ANDERSON and LUIS E. ORTIZ, editors GEOGRAPHY GRADUATE STUDENT ASSOCIATION UNIVERSITY OF ILLINOIS at URBANA - CHAMPAIGN SOME OBSERVATIONS CN THE LATE PLEISTOCENE AND HOLOCENE HISTORY OF THE LOWER OHIO VALLEY^'- Charles S. Alexander ABSTRACT The late Pleistocene and Holocene stream morphology of the lower Ohio Valley is the result of an interaction of deposition, erosion and deformation (arching) of late Wisconsinan valley train. The terraces and floodplains in the lower valley were field mapped and their elevations were plotted in longitudinal profiles. Trend lines were fitted to the elevation points. Analysis of the trend lines plus information from C-]k datings indicate that stream erosion resulting in the terrace form probably commenced about 1^,000 years ago. Subsequent erosion and deposition plus crustal deformation resulted in two distinct zones of floodplain development along the lower Ohio Valley. Below Dekoven, Kentucky where deformation seems absent a single floodplain predominates. Above Dekoven, in the area of arching, there are two floodplains with an incipient development of a third. The coincidence of the arching and the additional floodplains suggests deformation is a major factor in their origin. Also related to the arching of the floodplains is a changing lateral extent of flooding along the lower Ohio with floods of similar magn i tude. INTRODUCTION The late Pleistocene and Holocene stream morphology of the Ohio Valley downstream from Owensboro, Kentucky, has been described and mapped in several places but, except for Ray's brief synthesis, an overall interpretation of the morphology is lacking (Ray, 1963)- Furthermore, though the valley fill has been studied, the floodplain below Owensboro has received relatively little attention (Ray, 1965; Walker, 1957; Harvey, 1956). The purpose of this paper is to analyze the fluvial morphology of the lower Ohio Valley in an attempt to determine the late Pleistocene and Holocene history of the valley. :An oral version of this paper was read at the "Fifty Years of Berkeley Geography" special session of the Pacific Coast Division meetings at San Diego, California, June 14-15, 1973- Thanks are due to Professor Brigham Arnold for presenting the paper for me. Two important points should be made at the onset of this discussion. First, the fluvial deposits of the lower Ohio River Valley are mainly related to glacial rather than interglacial events. This is the case because the Ohio and its north bank tributaries served as spillways for glacial meltwater. As a result the valleys were filled with valley train material during glacial stages and eroded during interglaci als ; however, this clear-cut dichotomy is subject to modification as will become evident later. Second, the bedrock valley of the lower Ohio is partially filled by a 1^0 to 185 foot thick deposit of outwash and alluvium of V/isconsinan and recent age (Ray, 1965; Walker, 1957; Wayne, 1952). Although there is agreement on the age of the fill, the time of erosion of the valley is open to question. Ray (19^5, p. 24) believes the deep channel was cut prior to the rearrangement of mid-western drainage by the Kansan glacier. Walker (1957) considers that erosion occurred as the result of uplift during late Kansan and early Yarmouthian Stages. But he also thinks that, as the result of glacially lowered sea level, the river removed most of the lllinoian valley train during the early V/isconsinan. In my opinion, the Ohio River may have cut its bedrock valley during the early phases of the lllinoian and Wisconsinan glacial stages when sea level was going down, but before the glaciers reached the headwaters of the Ohio and its tributaries. However, this problem remains unresolved. VALLEY DESCRIPTION In general* the configuration of the lower Ohio Valley is related to the changing nature of the areal geology along its course. As a conse- quence of these changes the lower valley can be divided into three distinct sections (Figure 1). Between Owensboro, and Dekoven , Kentucky, 0) N o — ^ w^— LlI . J_ >- fD D-, o 3 -o \- O — c o t- . — TO J3 3 (D in cr- ■M c — «n 1- 0) J- 4-> o 2 — Q.C j^: T3 o u- C o o TO s_ CO x: >- +j (U c 1- »— 0) O • — 0) c 0! 2 > ■(-» XI (U c 1_ J2 TO (U > (n (U • — c •— a: O ' — O ■t-J > .— o If) Ji (U c o u> TO > 1- (/I LU Q. o • TO 1 — ■M the valley formed In relatively soft Pennsy 1 vanian rocks, and the valley floor is eight to ten miles wide (Willman, et .al . , 1967). This section of the valley is characterized by broad meander loops, well developed upper and lower floodplains, an incipient lowest floodplain, one alluvial terrace (informally called the Brownfield), and adjacent (Owensboro) lake plains. Segments of the alluvial terrace are large and frequent between Owensboro and Mt. Vernon, Indiana. However no clear evidence for the terrace was found between Mt. Vernon and Dekoven. The terms upper, lower and lowest floodplains are used because all are flooded, though frequency of flooding is less for the upper, and all are areas of contemporary deposition. Between Dekoven and Smithland, Kentucky, t*he river passes through a zone of relatively resistant Mi ss i ss ippian rocks (Willman, et .al . , 1967). The valley is narrow and the floodplain is frequently less than two or three miles wide (Figure 1). In this section the upper floodplain virtually dominates the valley floor. Isolated fragments of what appears to be the lower floodplain occur at several localities. Two stream terraces are present but the upper is so badly eroded that its presence has been iden- tified only to a limited extent and the remnants are too small to be shown on Figure 1. The lower or Brownfield Terrace has an intermittent occur- rence and continues into the third section. In this final section below Smithland, Kentucky, the valley widens again as it passes through soft Cretaceous and Tertiary rock (V/illman, et.al . , 1967). Here, the Metropolis lake plain, the Brownfield stream terrace, and the upper floodplain become prominent features in the valley landscape (Alexander and Eyton, 1973; Alexander and Prior, I968). The lower floodplain occurs only at the southern tip of the meander loop immediately east of Brookport, Illinois (The Black Bottom). METHODOLOGY In order to correlate and analyze the Ohio River floodplains and terraces, they were mapped in the field using topographic maps (1:2^,000) and air photographs. Between Brookport and Old Shawneetown, Illinois, field work was intensive, whereas between Old Shawneetown and Owensboro, Kentucky, mapping was based on field work, topographic maps and Ray's work in the vicinity of Owensboro (Ray, 1965). The resulting map was later checked at several locations along the valley with geologic quad- rangle maps prepared by other investigators (Johnson, 1972a, 1972b; Amos, 1967, 1966, 1965; Amos and Wolfe, I966). The floodplains and terraces were plotted in longitudinal profile with elevations taken from 1:2^,000 topographic maps (Figure 2). While most values came from benchmarks, and spot heights, a few were estimated at sites where closed contours cover small areas. All elevation sites were carefully selected so as to minimize errors resulting from human construction or natural erosion. Trend lines representing the upper and lower floodplain and terrace surfaces were fitted to the elevation points by the least squares method (the lowest floodplain was omitted because of its limited extent). The linear, quadratic and cubic solutions were made. The quadratic solution provides slightly better fits for the terraces and lower floodplain. The terraces' surface has a very slight concave curvature whereas the lower floodplain shows a small convex curvature. The cubic solution yields the best fit for the upper floodplain and shows it to have a very slight concave curvature whereas the lower floodplain shows a small convex curvature. The cubic solution yields the best fit for the upper flood- plain and shows it to have a very slight concave profile below Cave-in- Rock, Illinois, and a clearly visible convex profile above. The degree of curvature for the terrace and lower floodplain is so slight at the scale of the profile that it cannot be shown (Figure 2). The advantages of fitting lines to the points by the least squares method are that the lines have an unbiased best fit to the points and the coefficient of correlation (r-value) shows the strength of the rela- tionship between distance and elevation. Because of the high r-values of each line it may be assumed that its points belong to a set of closely related data--that they belong to distinct morpfiologic surfaces. DISCUSSION The terrace-f loodplai n profiles show several interesting aspects of the alluvial morphology in the lower Ohio Valley. (1) The convex curva- ture of the upper floodplain surface above Cave-in-Rock suggests the presence of mild crustal deformation. (2) The Metropolis lake plain appears to have no extension upriver. (3) The profiles demonstrate that the stream terrace remnants above Mt. Vernon and below Cave-in-Rock belong to a single feature--the Brownfield terrace. This result was anticipated by Ray (1965, 1963) in his studies of the Ohio Valley. (k) The major occurrence of the lower floodplain is almost entirely within the zone of arching and is probably related to it. Also related to the arching is the changing pattern of overbank flooding along the Ohio floodplain between Brookport and Owensboro. I I 0) u OJ u u 0) ■M ■D c (0 V) c • a* »D ^^ O. -D O O 9^ *4- tf) ■M •— -o C n L. 4) > • •* CM o: 0) o i- •^ D ^ cn o u. 0) ^ I/) o 1- Q. fU c T3 73 C7) C o 8 Crystal deformation The curvature of the upper floodplain is gentle and continuous-- there is no abrupt change in qradient between the concave curvature below Cave-in- Rock and the convex curvature above (Figure 2). Evidently, once the floodplain had developed, slight crustal warping changed its concave surface upstream from Cave-in-Rock into a gentle convex curve. To my knowlege no direct evidence for Holocene crustal deformation has been reported for the lower Ohio Valley. There is, however, considerable evidence for crustal instability. Numerous faults have been described along the lower Ohio and freguent earthguakes of slight to moderate in- tensity do occur. In light of these facts it seems reasonable to assume the arching of the upper floodplain is the result of local crustal movement (Hei gold, 1968; Baxter, et .al . , 1967; Baxter, et .al . , 1965; Baxter, et.al . , 1963; McGinnis, 1963; Butts, 1925). The movement was slight and probably started after the Rrownfield terrace sediments were deposited. Originally the terrace must have had a concave gradient appropriate for the hydraulic conditions associated with deposition of its sediment. Subsequent deformation would have cancelled the concave curvature resulting in the present nearly straight terrace profile. The almost imperceptible arching of the lower flood- plain may be an indication that crustal movement is still occurring at an extremely slow pace. The Metropolis Lake Plain As the result of trend surface analysis, Alexander and Eyton (1973) believe the Metropolis terrace to be of lacustrine origin. The lack of loess deposits on the terrace older than the Peorian, and the moderate q degree of stream dissection of the terrace surface suggest an Altonian age for the deposition of the lake sediments. They concluded that the lake may have been a slack water body impounded in the lower Ohio Valley by high flows of Altonian meltwater along the Mississippi River. The lack of related deposits up the Ohio Valley support their opinion that the lake plain is mainly a product of an event initiated outside the valley. The Brownfield Terrace Downstream from the junction of the Cumberland River, the Brownfield terrace has been interpreted as the result of lake deposition (Willman and Frye, 1970; Olive, I966; Finch, Olive and Wolfe, 196^; Ray, 1963) or stream origin (Alexander and Eyton, 1973; Alexander and Prior, I968) . The material underlying the first 20 to 30 feet beneath the terrace surface, here and upriver, is silt or silty clay along with deposits of sand. Though the silt and silty clay occasionally occur in fine, horizontally bedded units, this is not diagnostic as to origin because fine, horizontal bedding is characteristic of both floodplain slack water and lake deposits. However, the high degree of fit of al 1 elevation points to a line with a gradient slightly greater than that of the Ohio River strongly supports the theory of stream origin for the terrace. The sediment underlying the Brownfield terrace accumulated rapidly-- so rapidly in fact that it formed dams across the mouths of small tributary valleys. Sedimentation in the ensuing lakes resulted in several small lake plains (Owensboro lake plain) in southern Indiana and central northern Kentucky tributary to the terrace (Thornbery, 1950; Shaw, 1915). It is possible that the valley train sediment in the Brownfield terrace also formed the dam responsible for Lake Saline and the large Lake Saline 10 plain in southeastern Illinois. A brief comment about the origin of the plain is in order because of its great importance to the interpretation of the history of the lower Ohio Valley. The plain results from the time when V/oodfordian outwash sediment clogged the valleys of the Ohio and Wabash Rivers forming a dam that backed up a shallow, semi -permanent lake of three to four thousand years duration. The lake surface was rising above the 360 foot level about 21,000 radiocarbon years ago and finally rose to an elevation of about hOO feet around 20,000 years ago. Eventually the rivers eroded through the fill and the lake slovyly drained. Though this cannot be precisely dated it is thought to have occurred around 17,500 years ago (Frye, 1972). The surface of the Brownfield terrace today is at least 20-30 feet too low to have formed the dam responsible for Lake Saline. However, Leverett (1929) reports that the surface of the Wisconsinan outwash frequently rises down river from the mouths of tributaries that carried outwash to the Ohio. Hence, it is entirely possible that the original surface of the Woodford i an outwash below the junction of the V/abash was higher. The higher terrace in the vicinity of Cave-in-Rock may represent eroded remnants of that surface. The sandy nature of the terrace deposits is compatible with the idea of outwash origin. During the time of lake drainage, eros ion and deposition may have reworked the surface of the high outwash fill down to a level approximately similar to that of the present terrace. Some radiocarbon dates support the interpretation that the Brownfield terrace and Saline lake plain are contemporaneous. Two dates of woody material from terrace deposits in the vicinity of Owensboro indicate an 11 age between 18,^00 and 20,000 years (Pay, 1?65) . Mollusks from terrace depcsitf. in the lov;er Tennessee Valley gave a date of 21,000 years (Oliver, 1966). Dates from the Cache Valley section of the Brcwnfield terrace are much younger. Three samples of vvcod and organic matter from the Cache Valley yielded dates of 1^,000, 13,000 and 6,000 years (Alexander and Prior, 1568). The first two dates are from depths of 3^ and hO feet whereas the latter is from a depth of 32 feet. These younger dates are difficult to explain. Perhaps the Uj,000 and 13, COO year old samples are the result of local scouring and redeposition of the Brownfield deposits during the passage of the Lake Maumee flood through the Cache Valley some 1^,000 years ago. However, no physical evidence has been found to support this speculation. The 6,000 year old sam.ple may have been con- taminated during recovery by the commercial water well driller who made them avai lable--thi s could also apply to the two older dates. Be that as it may, the minimum dates suggest the terrace deposits are younger than the lake plain v/hereas the older dates indicate contemporaneity. I favor the latter possibility v/ith the reminder that the terrace form is substantially younger than the deposits underlying it. The origin of the Brownfield terrace form probably commenced following the discharge of water from glacial Lake Maumee. This is indicated by several radiocarbon dates from alluvium in the Black Bottom, east of Brookport, which suggest the upper floodplain began to take on its present form a little over 10,000 years ago (Alexander and Munnally, 1972; Alexander and Prior, 1971). The flood water flowed from the Wabash into the Ohio east and west of Shawneetown leaving shallow channels in the Saline lake plain (Frye, 1972). Traces of the former channel have sub- sequently been destroyed by stream erosion and deposition. Net incision 12 of the valley fill by the flood water below Shawneetown was probably slight because in that section of the valley the transition between terrace and upper floodplain is usually gradual. Once the Maumee flood ended, glacial events upriver virtually ceased to have much influence upon subsequent developments in the lower Ohio Valley. The river continued slowly eroding the valley fill and commenced to develop its present meander pattern. A variety of evidence indicates that the meander loops migrated to approximately their present position over a period of several thousands of years. However, within the last 1,000 years or so, the banks appear to have been in a state of dynamic equilibrium (Alexander and Nunnally, 1972). The Floodplains The rate of river incision, downstream from Cave-in-Pock , has been slow and apparently continuous. Cross-section profiles of the floodplains show that the crests of meander point ridges and natural levees can be fitted with straight lines that have a pronounced slope toward the river (Figures 2 and 3> Profiles AB , CD, EF) . The erosion defining the Brownfield terrace evidently v/as not an abrupt event. As the river migrated laterally the flood water level necessary for the construction of meander point and natural levee deposits gradually and steadily lowered. The most obvious reason for this is a continuous, slight stream incision. Upstream from Dekoven, stream erosion has been intermittent and produced the Brownfield terrace and the upper, lower and lowest flood- plains (Figures 1 and 3; Profiles GH , !J, KL) . The coincidence of the lower and lowest plains with the zone of arching suggests that local 13 BLACK BOTTOM CENTRAL Brownfield Terrace BLACK BOTTOM EAST »/ower 300 ROSICLARE BOTTOM lower ^ h* upper flood ploin" flood ploin- -Brownfie/d Terroce- 375 350 EVANSVILLE BOTTOM 400 OWENSBORO BOTTOM upper Figure 3 Transverse profiles of the floodplains in the lower Ohio Valley. Profiles AB, CD and EF, downstream from Dekoven show the floodplain with a pro- nounced slope towards the river. The floodplains above Dekoven (GH, IJ and KL) have a slope away from the river. The general trend of each floodplain surface Is shown by a dashed line. See Figure 1 for location of prof I les . crustal warping vjas a major factor in their origin. The initial warping probably commenced while the first floodplain was being established. As the arching progressed the river cut down and began forming a new, lower floodplain. The slow migration of the meander loops, and the fact that the transverse slope of the lower bottom surface is av/ay from the river, indicate that uplift was of short duration and then stopped or became extremely slow. Stable or near stable crustal conditions and channel aggradation permitted the migrating meander loops to deposit a sequence of meander point ridges, each of which had a slightly higher crest than its predecessor. The result is the ideal floodplain cross-section shown by the lower bottom, I.e., the highest land on the floodplain Is adjacent to the river. Segments of the lowest floodplain were identified at two places only One Is a sizable bud-shaped segment south of Mt. Vernon (Figure 2). The second, south of Shawneetown, has a bow-shaped outline and is relatively small. At these localities the segment surfaces are set off from the lower floodplain by reasonably well defined scarplets about six feet high. The lowest bottom may occur elsewhere but could not be Identified because Its bounding scarplet Is obscured by complexes of meander point ridges. Though the lowest floodplain appears to have a limited extent, what can be Identified is of sufficient size to make it doubtful that the feature Is the result of some localized, random event of river activity. The lowest floodplain probably represents a recent, slight renewal of arching or a slight Increase in rate of movement. Because of the arching of the upper bottom, the extent of overbank flooding varies from place to place along the Ohio River. The vertical separation between the upper and lower bottoms varies from about eight 15 to ten feet at the Black Bottom to a maximum of 12 to 15 feet between Mt. Vernon and Evansville. Above Evansville the upper bottom starts to bend gently towards the lower causing their surfaces to converge. The two surfaces may indeed merge a short distance upstream from Owensboro (Figure 2). Because of these differences In vertical separation, a flood of a given crest height will clearly result in different degrees of overbank flooding as the flood wave moves dov/nst ream. ■'' However, flood magnitude is indicated by discharge not by crest height. Due to complex reasons that need not be discussed here, floods of comparable discharge may have different crest heights. Consequently, it Is risky to discuss flood probability in terms of crest heights. Nevertheless, flood height is a convenient way to determine the extent of floodplaln cover by floods of a given frequency. Table 1 below was prepared by using the method described by Speer and Gamble (1965) for calculating flood probability along the Ohio River. Also Involved In the calculations were certain assumptions about normal pool river depth and interpolations in the flood records to convert discharge to flood crest height. The table shows the probability of flooding for each of the floodplalns at several localities along the river. The results shown in the table should not be regarded as accurate predictions. Rather, they Indicate the relative difference in flood magnitude required to cover the several bottoms at different places along the river. Hypothetical flood wave that maintains a uniform crest as it passes downstream. 16 Table 1 Relative Flooding Probability on the Lower Ohio Valley Bottoms Lowest Bottom Lower Bottom Upper Bottom Black Bottom Not Present Mean annual 1 in 15 to 20 years flood Shawneetown Mean annual 1 in 15 years 1 in 20 years flood Evansville Not Present 1 in 20 years greater than 1 in 50 years Owensboro Not Present Mean annual 1 in 15 years flood Clearly, the extent of flooding at any given locality on the lov/er Ohio River is not simply a function of flood magnitude but is also influenced by the degree of deformation of the floodplains. The final event in the valley has been an increase in the rate of floodplain sedimentation. Radiocarbon dates of organic matter from the upper floodplain east of Brookport indicate that a period of increasing rate of sedimentation began some 1,500 years ago. This is especially true of the swales. A reasonable explanation for this increase is that it reflects an increasing extent of cultivation in the Ohio drainage basin during the last one to two thousand years (Alexander and Prior, 1970. CONCLUSION The late Pleistocene and Holocene stream morphology of the lower Ohio Valley results from crustal warping and the deposition and subsequent erosion of late Wisconsinan valley train. The sediment in the valley's one major terrace dates from the Woodfordian sub-stage. Stream erosion, resulting in terrace form, probably did not start until the onset of the Maumee flood some 1^,000 years ago. Subsequent interplay of crustal 17 warping and stream erosion and deposition produced two distinct areas of floodplain form. Below Dekoven , Kentucky, a single floodplain predominates and Its surface slopes gently towards the river. Above Dekoven, as the result of local warping, three distinct floodplalns developed--the upper which is continuous throughout the valley, the lower, with a surface sloping away from the river, that is mainly confined to the zone of deformation and the lowest that occurs near the v/estern end of the deform- ation. Because of the floodplain arching the lateral extent of flooding along the lower Ohio River is a function of both flood magnitude and floodplain morphology. The last event along the river has been an increase in the rate of sedimentation during the last 1,500 years or so, probably as the result of increasing agricultural activity upriver. 18 REFERENCES C. S. Alexander and J. R. Eyton, (1973), "Trend surface analysis of floodplain and alluvial terraces in southern Illinois and western Kentucky," Bulletin, G.S.A ., 84, p. 1069-107^. C. S. Alexander and M. R. Nunnally, (1972), "Channel stability on the lower Ohio River," Annals, Association of American Geographers , Vol. 62, p. 411-417. C. S. Alexander and J. C. Prior, 1968, "The origin and function of the Cache Valley, southern Illinois," Jjt^ R. W. Bergstrom, ed., The Quaternary of Illinois , University of Illinois, College of Agriculture Special Publication 14, p. 19-26. D. H. Amos, 1965, Geology of parts of the Shetlerville and Rosiclare quadrangles, Kentucky, U. S. Geological Survey and Kentucky Geological Survey, Quad, Map GQ-400. D. H. Amos, 1966, Geologic map of the Golconda quadrangle, Kentucky, U. S. Geological Survey and Kentucky Geological Survey, Quad Map GQ-546. D. H. Amos, 1967, Geologic map of the Smithland quadrangle, Kentucky, U. S. Geol. Survey and Kentucky Geological Survey, Quad. Map GQ-651. D. H. Amos, and E. \L V/olfe, 1966, Geologic map of the Little Cypress quadrangle, Kentucky-Illinois, U. S. Geological Survey and Kentucky Geological Survey, Quad. Map GQ-657. J. W. Baxter, and G. A. Desborough, 1965, "Area! geology of the Illinois fluorspar district, Part 2--Karbers Ridge and Rosiclare quadrangles," Illinois State Geological Survey , Circular 385, 40 p. J. W. Baxter, and G. A. Desborough, and C. W. Shaw, I967, "Areal geology of the Illinois fluorspar district, Part 3 — Herod and Shetlerville quadrangles," Illinois State Geological Survey, Circular 4l4, 4lp. 19 J. W. Baxter, P. E. Potter, and F. L. Doyle, "Areal geology of the Illinois fluorspar district, 1963, Part 1--Saline Mines, Cave-in-Rock , Dekoven and Repton quadrangles," Illinois State Geological Survey , Circular 3^-2, hk p. C. Butts, 1925, "Geology and mineral resources of the Equal i ty-Shawneetown area," Illinois State Geological Survey Bulletin , vol. kj , 76 p. W. I. Finch, W. W. Olive, and E. W. V/olfe, 196^, "Ancient lake in western Kentucky and southern Illinois," U. S. Geological Survey Professional Paper , vol. 401-C, p. C130-C133. J. C. Frye, et .al . , 1972, "Geology and paleontology of late Pleistocene Lake Saline, southeastern Illinois," Illinois State Geological Survey Ci rcular 471 , p. 6. E. J. Harvey, 1956, "Geology and ground-water resources of the Henderson area, Kentucky," U. S. Geological Survey Water Supply Paper , vol. 1356, 227 p. P. C. Heigold, 1968, "Notes on the earthquake of November 9, I968, in southern Mlinois," Illinois State Geological Survey Environmental Geology Notes , vol. 2k, 16 p. W. D. Johnson, 1972a , GeologI c map of parts of the Newburger and Yankeetown quadrangles, Kentucky, U. S. Geological Survey and Kentucky Geological Survey, Quad. Map GQ-IO^S. W. D. Johnson, 1 972b, Geologi c map of the Reed quadrangle, Kentucky-Indiana, U. S. Geological Survey and Kentucky Geological Survey, Quad. Map GQ-1038. Frank Leverett, 1929, "Pleistocene of northern Kentucky," Kentucky Geological Survey, Ser. 6, vol. 31, p. 65. 20 L. D. McGinnis, 1963, "Earthquakes and crustal movement as related to water load in the Mississippi Valley region," I 1 1 inoi s State Geological Survey , Circular 3^^, p. ^-5- W. W. Olive, 1966, "Lake Paducah, of late Pleistocene age, in western Kentucky," U. S. Geological Survey Professional Paper, vol. 550-D, p. D87-D88. L. L. Ray, 1965, "Geomorphology and quaternary geology of the Owensboro quadrangle, Indiana and Kentucky," U. S. Geological Survey Professional Paper , vol. ^88, 72 p. L. L. Ray, 1963, "Quaternary events along the unglaciated lower Ohio River Valley," U. S. Geological Survey Professional Paper , vol. 475"B, BI25-BI28. E. W. Shaw, 1915, "Newly discovered beds of extinct lakes in southern and western Illinois," Illinois State Geological Survey Bulletin, vol. 20, 139-157. P. R. Speer, and C. R. Gamble, 1965, "Magnitude and frequency of floods in the United States, Part 3A--0hio River Basin except Cumberland and Tennessee River Basins," U. S. Geological Survey Water Supply Paper , vol. 1675, p. ^-8. W. D. Thornbery, 1950, "Glacial sluiceways and lacustrine plains of southern Indiana," Indiana Geological Survey Bulletin , vol. h, 21 p. E. H. Walker, 1957, "The deep channel and alluvial deposits of the Ohio Valley in Kentucky," U. S. Geological Survey Water Supply Paper , vol. 1^11, 25 p. W. J. Wayne, 1952, "Pleistocene evolution of the Ohio and Wabash Valleys," Journal of Geology , 60, p. 575-585. 21 H. B. Willman and J. C. Frye, 1970, "Pleistocene stratigraphy of Illinois," Illinois State Geological Survey Bulletin , vol. Sh. H. B. Willman and others, "Geologic map of Illinois," I 1 1 inoi s Geologj cal Survey, 1967. Paper A7, Some Observations on the Late Pleistocene and Holocene History of the Lower Ohio Valley , by Charles S. Alexander. Paper ;^6 , Social Problems in a Small Jamaican Town , by Curtis C. Roseman, Henry W. Bullamore, Jill M. Price, Ronald VI. Snow, Gordon L. Bower. Paper #5, Regional Changes in Petroleum Supply, Demand and Flow in The United States: 1$66-1980, by Ronald J. Swage r. Paper ^k. Matrix and Graphic Solutions to the Traveling Salesman Problem, by Ross Ku liner. Paper #3> Regional Components for the Recognition of Historic Places , by Richard W. Travis. Paper #2, Social Areas and Spatial Change in the Black Community of Chicago: 1S50-1960, by Charles M. Christian. Paper ^] , A Theoretical Framework for Discussion of CI imatologi cal Geomorphology , by Dag Nummedal. »>.^ w :i:!-;;>^n:iffi;bin¥!ii'i;ii»ii|iii UNIVERSITY OF ILLINOIS-URBANA 910.72IL6P C001 PAPER URBANA 7-9 1974-75 3 0112 024718931