Following each stage of glaciation there was one in which the climate was not very different from that of the present time. Such stages are called interglacial. Should glacial conditions return again, the epoch in which we live would, of course, be interglacial. If we assume the current century to be in the middle of this stage, ours would be a very short interglacial stage as compared with some of the others, probably the shortest of all.
The time of deglaciation which followed the Illinoian glacial stage is called the Sangamon interglacial stage. The name is taken from Sangamon County , Ill. (containing Springfield), where the evidences of this stage are clearly shown, and have been well studied.
Topography Left by the Illinoian Glacier.-When the ice of the Illinoian stage finally disappeared from southwestern Ohio, a topography appeared which was not very different from that of the present. The drainage was along present lines, though the narrows, marking the places where great streams had found new courses, were even narrower than at present. The great valleys were the same then as now, (except, that the axes of some were from 100 to 200 feet higher than now; that is, the deep valley filling, now represented only by terrace remnants, was then continuous, and the streams ran over its surface -at the level of the present terrace tops. The small valleys which open into the large ones were, of course, correspondingly less deep in their lowest -courses. The axes of most small valleys were less sharply entrenched than now, and most of these streams were shorter at their heads, and had fewer tributaries than at present. Some other details were different because the covering of surface silt to be described below had not yet been spread.
The above description applies primarily to Hamilton County. In Butler County, or speaking more strictly, in the area covered by the later drift sheet, the difference between the present topography and that at the close of the Illinoian stage would be of the same general character, perhaps less than in Hamilton County.
Erosion and Soil Making.-On the surface of the newly made Illinoian drift sheet, the great streams began at once to entrench themselves by cleaning out the drift from their valleys. The smaller streams began at the same time to deepen their narrow valleys and elongate their heads, and send out more tributaries. The chemical and organic agencies described in Chapter IV began to change the character of the till and make a soil. Before the old soil made in this interglacial stage was covered up, it became considerably deeper than the one which was formed on the newer drift since the last glacial stage. This is one of the evidences from which it is inferred that the interglacial stage was much longer than the postglacial.
Relative Amounts of Erosion on the Older and Newer Drifts.-Another kind of evidence bearing on the length of time which has elapsed since each glacial stage, is the amount of erosion on the drift of different ages, where it has remained uncovered by later drift. On the uplands and in the bluffs, the size of a valley indicates little or nothing as to the amount of erosion since the ice departed. This is because the streams are running so largely in valleys cut before the glacial epoch. In the great valleys, however, the erosion which followed the Illinoian stage started on the nearly flat surface of the deposits described above. The valleys in these deposits, therefore, represent the postglacial work of the streams which now occupy them. Illustrations of this erosion are so widespread as to make mention of them almost unnecessary. The immediate valley of Duck Creek, in the Norwood trough, is wholly postglacial, and almost everywhere in its vicinity its larger tributaries have cut out valleys twenty to thirty feet deep, and having flat bottoms from fifty to several hundred feet wide. The valley of Ludlow Run, followed by the Southern Ohio traction line west of Spring Grove Cemetery, and heading at College Hill only two miles away, is seventy-five to one hundred feet deep and forty rods wide, cut wholly since the Illinoian glacier departed. Many streams not more than two or three miles long in the deposits of this age have valleys cut to their local base levels, and have already developed flat bottoms several hundred feet wide.
If these valleys be compared as to size with those of similar streams on the newer drift sheet to the north, choosing places where the latter is also flat, it will be seen that the valleys in the older drift are many times as large; ten to twenty times would not be an unreasonable estimate. It thus becomes apparent that the interglacial stages, even though they do not attract attention by any deposits made at the time, are very important in the history of the present surface.
There is one important deposit in this region which appears to have been made at a time distinctly later than the Illinoian glacial stage, and distinctly earlier than the last or Wisconsin glacial stage. In other words, the time of its making seems to be separated from the Illinoian by a distinct stage, the Sangamon, and from the Wisconsin by an equally distinct stage. This deposit is loess or windblown dust, accumulated locally in this area to a thickness of twenty or thirty feet. Farther west, in the upper Mississippi Valley, it is even thicker.
Character.-This dust deposit appears now as a mealy earth, generally of light yellow color. When dug out of a bank in a fairly dry condition it breaks into fragments. Such a fragment, when crushed in the hand may offer considerable resistance, but gives way all at once, and crushes into a meal too fine to be properly called sand, but too coarse to be called clay. It does not form crumbs, balls, or lumps as clay does. Careful search will often disclose within it small coiled shells belonging to species still living. Not infrequently there are also calcareous concretions from a fraction of an inch up to several inches in the longest dimensions. These may have odd forms, suggesting the German word Loess Kinder (loess children). Within this area such concretions are rare.
When examined under the microscope this loess is seen to be made up of small angular fragments of quartz and other minerals from the igneous rocks which make up the glacial drift from the north. Such an examination has not been made for this area, but the material here is evidently identical with that farther west. Professor Salisbury found the loess at Kansas City to consist of particles ranging in size up to one-tenth of a millimeter (one two-hundred-fiftieth of an inch). But this size is extreme; only 4 per cent of the particles measured more than one ten-thousandth of an inch in their greatest dimensions; but this is still much larger than the average of clay particles.
Near the base of the loess, where it rests on the bowlder clay, or even on bed rock, may often be found a few pebbles of the kind which occur in the till. Its exact base may thus be hard to determine. The upper two or three feet may also be difficult to recognize as loess, because the decomposition of the constituent mineral grains yields so much true clay that the substance becomes stiff and may form clods.
A striking characteristic of loess is its tendency to "pack" into a mass which has a certain amount of rigidity. This is an essential property of molders' sand, for which loess is much used. It depends in part on the presence of angular grains of all sizes. Along with this goes a tendency to split in vertical planes or joints. These joints subdivide the mass into vertical prisms, not noticeable except on the weathered face of a loess cliff or bank. Partly on account of this packing and vertical jointing the loess has the peculiar property of standing in vertical faces. In excavations such vertical faces often endure for many years despite the mealiness of the material when crushed. This is only true where the surface drainage goes the other way and does not pass over the face of the exposure.
Distribution in the United States.-The formation here described is most abundantly represented in the states bordering the Missouri south of the Dakotas, and the Mississippi south of Minnesota. South of Illinois it appears chiefly in a belt bordering the Mississippi on the east. In these southern states it is generally known as the "brown loam," especially at some distance from the river, where its color is generally darker and its constitution more clayey. It is also spoken of as "bluff loam," especially near the river, and sometimes merely as "bluff."
Speaking of the entire area in which the true loess occurs, it is most prominent near the larger streams, especially those flowing south. In a very general way, also, its thickness decreases toward the south. It is generally much thicker on the east bluff than on the west, and on both sides it diminishes in thickness and becomes more clayey as the distance from the stream increases. It is also a. very significant fact that it is generally (though not universally) absent from the surface of drift sheets which are younger than the Illinoian.
Distribution in This Area.-In the area under consideration the true loess is found only in small spots, a very few square miles, in the immediate vicinity of the Ohio. The bluffs on the Kentucky side from Bellevue to Ludlow are well covered, except on steep slopes from which it may have been eroded. Probably the maximum thickness, at least thirty feet, is found in the Covington City Park, on the high ground southeast of Ludlow and south of West Covington. Probably the largest deposit on the Ohio side is at the foot of the bluff on the road leading northeast from Delhi.
Origin of the Loess.-The origin and mode of accumulation of the loess may be inferred from its constitution and distribution. The fact that its grains consist in large part of minerals which make up the glacial drift, and the further fact that some of these particles are still fresh, indicate that it was reduced to its present fineness by mechanical means rather than by decomposition. Its relation to the glacial drift suggests that this comminution might have been in large part mere grinding beneath the ice, and, therefore, that the loess is derived from glacial drift by some process of assortment. Its association with the great streams indicates that they were the primary agents of its distribution, and probably also of its assortment from the body of the till. Its presence on hills and bluffs, regardless of height, leaves no doubt that it was transported by wind as well as water, while the greater thickness on east bluffs points to the prevalence of westerly winds, and suggests that the flood plains of these streams were the source from which the wind gathered the material.
Following the inferences thus drawn, a picture of conditions during the loess-making time may be reconstructed as follows: (See note 15)
The streams which flowed south from the source of supply were heavily charged with silt. It might be assumed from this fact that the streams were aggrading their channels, filling their valleys; and this was probably the case. In doing this they not only made wide flood plains but were subject to great floods. In times of flood the silt spread out on the flood plains, and in times of drought it was picked up and carried by the winds over the uplands. The sand was drifted to and fro on the flood plain, sometimes building dunes, as in a later time at Ludlow, Ky. These would be quickly carried away when the streams began to degrade their valleys. Perhaps none remain from the loess stage. The silt would be carried higher and farther, settling in greatest abundance near to the source of supply, that is, on the bluffs. The deposit should be thickest on the bluffs, not only because they are near the source of supply but because the coarser grains settle first, and that which is regarded as typical loess is of relatively coarse grain when compared with other atmospheric dust. A similar process of wind deposition is now going on at the same place, but the amount of dust now carried is too small to accumulate as a distinctive deposit. It is not improbable that more is being removed by wind from the bluffs than is being added. Small mollusks, which lived on the land and in pools, were often covered by the drifting dust, and their shells were thus fossilized in the loess.
The location of the true loess in this area is in general consistent with the above view of its origin. This is especially true of the bluffs behind Newport and Bellevue. They lie on the leeward side of a broad alluvial flat (the Cincinnati basin). It is true that a prevailing north-west wind would be better suited to this hypothesis than the prevailing southwest wind now observed. The same is true at Delhi, where the valley to the northwest offers a free sweep for the wind. The locality of thickest deposit west of Covington does not lie in the lee of the broadest flats when the winds are westerly, but on the other hand it has flats on three sides, from all of which it might have received accretions.
Relation to the Loess and Drift. - Over most of Illinois and southern Indiana, except near the Mississippi, Illinois, Wabash, and other large streams, the Illinoian driftsheet and the adjacent driftless region is covered with a sheet of loam or silt, which has some features in common with the loess. It is finer in grain and of more clayey appearance, and is generally less than ten feet thick, frequently only two or three feet, and entirely absent from many steep slopes. It resembles the loess in covering hill and valley alike, regardless of altitude. Like the true loess also, it is apparently independent of the material which it covers. Not only does it seem improbable that it should be formed by the weathering of the material beneath, but there is much evidence that the drift sheet was old and deeply weathered before this surface covering was added. Moreover, it extends without abrupt change of character beyond the limits of the drift into the driftless region on the south.(See note 16)
This sheet, often called the "white clay," extends east over the here described, probably as far as Parkersburg, W. Va. It is limited on the north by the edge of the newer drift.
So far as can be detected, this surficial sheet of silt is continuous with the true loess. The coarser grain and greater thickness of the latter seems to give way gradually to the more clayey character and diminished thickness of the former. In this locality the thickness of the surface silt is rarely less than two feet, except on steep slopes from which it may have been eroded. Locally the thickness is four or five feet.
The Phase Called "White Clay." - Generally this surface silt weathers into a rather mealy soil of light color. It is especially light in color on flat uplands, which are poorly drained. From such areas, which are abundant in parts of Illinois, southern Indiana, and even southern Ohio east of this area, the formation has taken its popular name "white clay." Such spots are not abundant in the area here treated, but their character is well exemplified in the flat lands about Bevis, three miles west of New Burlington, and likewise on the flat upland strip followed by the Cincinnati, Lebanon & Northern Railway, from Deer Park and Rossmoyne north to the edge of the newer drift near the Hamilton-Butler County line. The flat fields including the brickyards south of Brecon are good examples. Similar illustrations are found on some of the flat uplands of Boone County, Ky.
Beneath the top soil it is more yellow and, where poorly drained, is often bluish at the base. At places, scattered throughout the entire thickness, but more abundant near the base, are small pebbles of chert, quartz, quartzite, and igneous rock, All representing the more resistant materials of the drift.
Analyzed both chemically and physically this formation is shown to consist of particles very similar to those which form the loess, though smaller. There are also many small concretions of limonite, the common oxide of iron which constitutes rust.
Other Phases of the Surface Silt.-With better drainage this surficial formation assumes darker colors, the surface soil formed from it being light yellow rather than pale gray, and the subsoil more ruddy. At depths of two or three feet the material is often brown, traversed by blue cracks, or mottled with blue and carrying concretions of limonite. Either at this depth or lower are generally found the chert and other pebbles mentioned above. Either at the horizon of these pebbles or below it is found, at many places, a stiff brown gummy clay, of the kind which may be formed by the decomposition of our local bed rocks or of the till.
Generally speaking, the body of this formation below its weathered, surface is silty rather than clayey, and yellow rather than white or even gray. For the purpose here in hand, therefore, it is commonly referred to as the surface silt, a merely descriptive and not a technical term.
Origin of the Surface Silt.-If this surface sheet of silt is a direct continuation of the loess, and grades into it, then it was presumably deposited by wind, as was the loess. Its finer grain and texture are accounted for by the fact that wind, like water, when depositing its load, drops the larger pieces near to the place where the load was picked up. This would leave only the finest dust to be deposited at a great distance from the major streams, whose flood plain furnished the material of the loess. The same principle would account for the relative thinness of the deposit.
The chief difficulty in assuming such an origin for this formation lies in the presence of the pebbles. It has been suggested by Leverett (See note 17) that some of these were brought up from the underlying till by crayfish or other animals during the gradual accumulation of the deposit, just as the same animals working in the same material occasionally bring up small pebbles now. The fact that these stones consist exclusively of chert, quartz, quartzite, and other rocks which are very enduring, is assumed by him to indicate that all limestones had already dissolved out of the soil before this formation was deposited.
Instead of assuming that the material of this entire formation was imported from the same localities that furnished the loess, and, therefore, from a long distance, it is possible to conceive of the greater part of an aeolian deposit as being laid down not far from the place where the material was picked up. According to this supposition the loess is coarser than the widespread silt, not so much because it represents the material first dropped, as because the locality from which it was derived afforded a supply of coarser material. The broad flood plains, of course, afforded sand and particles of all sizes smaller than sand grains, but the sand generally fell again on the flood plain and would be again quickly carried away by the water.
If the surface silt be a wind deposit, it, like the loess, indicates an arid climate at the time of its making, or at least a time of extreme intermittent droughts. In such times dust is not brought to a place and deposited once for all, but is blown here and there many times, or many hundreds of times, before it is effectually buried and kept in place, or the aridity ceases. If a very limited surface like that of a county start in such an epoch with no wind deposit whatever, and none is brought in, it will nevertheless gradually acquire such a sheet, derived from its own soil. One spot will furnish dust which will settle at another. On another day the deposit will be shifted and new dust added to it from the exposed ground. The sites of erosion and deposition shift from day to day, or from storm to storm, but bare spots subject to fresh erosion become smaller and fewer as the amount of drifting material becomes greater.
It should not be assumed that the material thus subject to wind action was all laid down locally, or derived from nearby. There have probably always been "prevailing winds" from some direction, which would result in an aggregate movement of dust in some one direction. Moreover, the uniformity in thickness of the surface silt does not permit this conception of to and fro shifting to be adhered to exclusively. It is probable, however, that a large part of the area furnished dust as well as received it (See note 18)
Over somewhat more than the southern half of this area, the surface silt was the last formation deposited on the uplands. When it ceased to form, the present cycle of erosion began. In the northern part of the area the later glaciation interrupted this cycle, or brought it to a close. The time, between the deposition of the surface silt and the newer glaciation, constitutes a distinct stage. The same changes which are now at work at the surface were proceeding in that stage, namely, rock decay, soil-forming, and erosion. At a few places we have records of how far these processes went before the Wisconsin glacial stage, and therein we have some evidence of the length of this stage. Thus, on Twomile Creek in Hamilton (see Table V, also PI. VI-B), six feet of this old silt are seen in the bank between the older till below and the younger till above. The silt is much weathered, and for the most part converted into a soil. If this weathering was all done before it was covered up by the twenty-five feet of younger till, the time during which it was exposed must have been very long, much longer in fact than that which has elapsed since the last ice withdrew. A similar section bearing the same evidence is found in a high creek bank two miles northeast of Glendale.
Early and Late Wisconsin.-The last stage of glacial conditions in this area, and the last in the United States, is called the Wisconsin. If the area here described extended ten miles farther north it would be necessary to subdivide this stage into early and late Wisconsin, for the stage represents two distinct advances of the ice, the latter falling short of the former by about twenty-five miles in this region. Only the early Wisconsin glacier invaded this area. These advances were not separated by a very long time, but each has its own deposits, which at many places are distinguishable.
Limits of Advance. - The limits to which the glacier of this stage advanced are best seen on the one inch to the mile topographic map of the Cincinnati area. . Briefly summarized, it crossed the New Haven trough to the bluffs on the south., but at the western edge of the area it did not reach so far south, the local trend of the ice front being from northwest to southeast. Above New Baltimore the present channel of the Miami was buried, but below that point it remained free from ice. Between New Baltimore and Banklick Creek (distinguish from Banklick Creek, Ky.) the ice was checked at the foot of the bluff or part way up on the slope. East of that creek the glacier scaled the bluff and spread south over the upland, but it was still unable to override the highest points. The effects of the 920-foot knob in section eight (Ross Township, Butler County, see topographic map) is particularly noticeable in causing a re-entrant curve in the ice front. From here the line trends southeast until within a mile or more of Mill Creek Valley. Here the effect of that large valley in developing an ice lobe is distinctly seen. This lobe advanced southward to Hartwell, its western edge for four miles being almost parallel to the valley. Turning north again at Hartwell, the eastern edge of this lobe follows the bluff almost to Sharonville, a distance of more than four miles, in a direct line. North of that the ice sheet covered the upland on the east, its edge trending northeast to near Foster. It does not appear to have crossed the Little Miami within the limits of this area.
In a general way the front of the ice seems to have trended both northwest and northeast from this area. The great southward advance within this area is therefore very marked. This is probably due mainly to the influence of the broad Miami Valley south of Dayton.
Thickness of the Drift.-The deposit of till made by the Wisconsin ice is on the whole thicker than that made by the Illinoian. Probably the combined thickness of the older and younger tills would average fifteen to twenty feet. Most of this is the newer deposit. The older was always thin, and in addition was doubtless, to some extent, disrupted by the advancing ice, and incorporated into the new drift. The aggregate thickness of the two on relatively plain ground may occasionally reach thirty feet, but greater thicknesses are for the most part confined to old valleys.
Not only is the general thickness of the later till greater than that of the earlier, but there is in some places a distinct thickening as the edge is approached. Many wells in the vicinity of Mason are reported to show a thickness of approximately forty feet of till, and in isolated cases not far away, much greater depths are reported. Near Pleasant Run, on the C. D. & T. Traction Line, similar depths are reported.
Topography. - The thicker drift here described does not constitute a good terminal moraine, but there are, near the border, spots of considerable size which have a morainic topography, that is, they are characterized by hummocks, and in a few instances by undrained hollows. One such spot is found on the western border of the area from one to two miles north of Shaker Village. North of New Baltimore is a similar topography. A considerable area near the Hamilton-Butler County line, west of the C. D. & T. Traction Line, is distinctly morainic; likewise a few small patches northeast of Sharonville.
As already described, the topography of the Wisconsin drift sheet is due in the main to pre-glacial erosion, the valleys now being at the same places as they were before the advent of the ice. The newer drift sheet has this in common with the older. On the older drift there has, however, been much erosion since the ice disappeared; on the newer drift, there has been very little. Post-glacial streamlets have searched out the lowest line on which to run, and there may be something of a channel along such a line, but along most of the subordinate drainage lines there are, properly speaking, no stream valleys. Only streams of some size and permanence have erosion valleys cut beneath the levels which they found, and bounded by slopes of their own making, either ravine slopes or bluffs.
Character of the Till.-The drift of this stage is not sharply distinguished in character from that of the earlier. It has in general a lighter color and is softer and easier to dig, partly, no doubt, because it contains more sand, both mixed with the clay and localized in pockets. It is probably stonier also, and the percentage of igneous and metamorphic rock to limestone is greater than in the older drift; it may be as much as four or five per cent. The joints described in the Illinoian drift are less prominent in the Wisconsin, though by no means wanting.
The till of this age is calcareous like that of the Illinoian, and it has been less deeply leached in the formation of soil. The yellow zone due to surface oxidation is also less deep, but the contrast between young and old drift in this respect is not so great as might be expected, probably because the more porous character of the younger till has permitted the agencies of chemical weathering to work more rapidly.
The Wisconsin till is not covered by the surface silt described as covering the Illinoian, but its surface portion, from one to three feet deep, is in many places almost free from stones. In places this want of stones, at and near the surface, is very striking. Along the road eastward from Hamilton, past the infirmary, a trench two to two and one-half feet deep was dug for a pipe line. In two miles east from the infirmary scarcely a stone was thrown out. Generally the number of stones increases gradually with depth, but at places there is a sharp line separating the stoneless clay above from the stony clay below. Where this is the case it seems to represent a mode of deposit not described above. Glaciers may carry on their surfaces considerable layers of sediment. Such deposits on continental glaciers must necessarily be local and near the edge. When they consist of silt it may have been brought partly or wholly by wind from the land surface in front. In the final melting of the ice, such "super-glacial drift" is quietly let down on the till below. It is doubtful if this process in detail can be invoked to account for all of the stoneless clay found in this area. Such clay is, however, sufficiently abundant here and elsewhere to call attention to the fact that the deposit of the topmost foot or few feet of glacial material is often to be explained by a different process from that which explains the lower deposit.
Contact of the Two Drift Sheets. - The distinction between Wisconsin and Illinoian till, when exposed in the same section, is not always easy. In some exposures, believed to show till of both ages, the oxidized joints described above cross the boundary and extend many feet up into the younger formation. In at least one bank, two miles west of Glendale, water issues at the base of the softer till, evidently because it fails to percolate into the denser formation below. Many ravines cutting deep into the younger till expose below, a bowlder clay which has all the features of the typical Illinoian; yet the silt which commonly covers the latter is wanting and there is no suggestion whatever of a surface of contact. While some of these exposures are doubtless of Wisconsin till from top to bottom, it is likely that the base of some of these sections is of Illinoian age. The want of any evidence of contact may be due to the disrupting of the surface of the older formation as the younger glacier advanced, and the incorporation of some of the older material in the newer formation. The same process would obviate any sharp contrast in physical character. It is not to be expected -that the two formations can always be distinguished even where found together. Much less is it possible to judge of every exposure at sight, whether it be one or the other.
The best section showing the deposits of the several glacial stages is found in the north bank of Twomile Creek in the city of Hamilton, about forty rods west of the C. D. & T. Traction Line. It is shown in Table V. In this section, number six is typical Illinoian till. One hundred feet away the oxidized fissures are strikingly exemplified. Number five is the oxidized phase, the topmost one or two feet being leached of its lime. Number four is plainly the surficial silt with its highly characteristic pebbles near the base. The whole thickness is here well oxidized, in part, no doubt, by weathering and soil-making before the deposition of the overlying till; in part, also, by ground waters which later followed this porous bed between the two tills. The dark color becomes more prominent near its base and extends over a little into the underlying till. Numbers three and two (Wisconsin till) do not differ greatly from the corresponding phases in the Illinoian till as seen at many places. As compared with the Illinoian till in this section, they are slightly less hard and paler in color. The lower till also lacks the silt and gravel lenses.
Behavior of the Ice.-From the description here given of the later drift sheet the behavior of the glacier may be inferred. The greater thickness of the drift, especially near the edge, indicates that, as compared with the older ice sheet, the Wisconsin glacier kept up its power better as it approached its limit. Probably it moved faster, and thus carried a greater thickness of ice close to the limit.
Evidences of Abundant Water.-Water flowing out from the edge of the Wisconsin glacier, and due largely to its melting, seems to have been more abundant than that which flowed from the former ice sheet. This tends to confirm the inference already drawn from the deposits of till, namely, that the thickness of the later ice sheet was better maintained as its limit was approached. The evidences of this abundant water are found partly in the main drift sheet itself, which contains pockets of stratified, that is, waterlaid sand and gravel. It is seen more clearly in the kames built at the edge of the ice, and still more clearly in the great valley trains, fragments of which constitute the sand and gravel terraces in the great valleys
Outwash Deposits of Different Ages. -The deposition of valley trains by outwash from the glacier has been described above. Those in this area were deposited in part when the ice front stood at its extreme limit, shown on the accompanying topographic map; but their accumulation began as soon as the advancing ice invaded the upper basin of the Miami, and continued during its retreat to the same place. Necessarily, those parts which lie north of the limit of the Wisconsin ice sheet were deposited during advance and retreat only.
As stated above, the Wisconsin glacial stage was double, the ice coming down a second time to a line about half way between Dayton and Hamilton. A portion, perhaps a large portion, of this outwashed sand and gravel was due to this later advance, which did not otherwise affect the area here described. It is not possible at present to distinguish the late Wisconsin outwash from the earlier.
Probably when this outwashed material began to be laid down, there remained, at least locally, in the bottoms of the great south-leading valleys, some of the till and some of the outwashed sand, gravel, and clay from the Illinoian stage. (See Fig.50) Of the several hundred feet of sand, gravel, and clay passed through at Hamilton before reaching the rock, it is not possible to say how much remains over from the older glacial stage and how much was deposited in the newer. This is equivalent to saying that it is not now known how deep the inter-glacial streams in these valleys had cut their channels before the approach of the newer ice sheet caused them again to aggrade.
The "Forest Beds."- There is one feature in the banks of the Ohio which has sometimes been interpreted as giving the answer to this question. This is an occasional exposure of muck or dark mucky clay, seen only in spots and at favorable times after scouring has been effected by floods. Here and there it may contain wood, leaves, etc. It suggests driftwood buried beneath the flood plain deposits of a shifting stream. As such, it might have been laid down at any time, however recent, as similar deposits are being made and covered up now. The exposures of this deposit have, however, all been about ten feet, or a little more, above low water, which suggests that they were all made at the same time on the surface of an ancient and now buried flood plain. More convincing evidence was noted by Dr. Edward Orton at Lawrenceburg, Ind. There, at the corresponding level, was an exposure of several beds, called by him the "Forest beds."(See note 19). Here he found stumps in upright position as though still standing as they grew in the ancient soil. There were thin beds of ocherous clay both above and below the deposit containing the stumps. The overlying beds were reported by Dr. Orton to be not recent alluvium, but constituent beds of the great terraces. This deposit he interpreted as marking the surface of the flood plain at the close of the inter-glacial stage, when the outwash began to deposit in advance of the oncoming Wisconsin ice sheet. In the Miami Valley, below Hamilton, some drilled wells are cited by Leverett (See note 20) as having passed through a bed of dark mucky substance about sixty feet below the surface. He suggests that this material may be muck accumulated on the Miami flood plain at the same time that the deposit was made at Lawrenceburg, and may thus mark the level of the inter-glacial valley floor near Hamilton. The differences in elevation between the beds at the two places correspond roughly with the fall of the river.
Character of the Outwashed Material.-While it is not possible to say of a well, at what depth the older deposit begins, and not always possible to distinguish the two deposits when seen in exposures, there are, nevertheless, certain characteristic differences. The newer gravel contains a larger proportion of stones from the far north. Not infrequently ten to fifteen per cent of the pebbles are of igneous rocks and quartzite. This is a much larger percentage than that found in the bowlder clay. Such a contrast is to be expected, since the water-laid material represents the material dropped by the ice at a point farther north. The stones are usually well rounded, indicating much wear during their journey. It is not possible to make any general statement as to their size, because the material differs greatly in this respect in the different large valleys. It varies from the coarse material along the Little Miami at Loveland, where many stones exceed one foot in size, to almost pure sand, as found in parts of Mill Creek Valley. The newer outwash contains less clay than the older. Generally speaking, the sand of this age is sharper than that of the older formation.
Cementation into conglomerate is less frequent in the newer gravels, chiefly because there has been less time for the process. There are, however, conspicuous instances of such cementation, as in the banks of the Little Miami one-half mile north of Milford, and in Langdon's travel pit at Linwood. (See Plate I-B.)
Miami Valley Train. - Remains of this valley train constitute a very broad terrace north of Hamilton. Most of the "Hickory Flats" is of this origin. The altitude ranges from 620 feet to about 660 feet. The slope from the north bluff toward the axis of the valley is due to the fact that much of the material was brought in by tributaries from the north, and was deposited in excessive measure where the small valleys open into the larger. The material brought down by the Miami is sand and fine gravel, stones larger than two or three inches being exceptional.
Farther down the valley little is preserved of this formation until near the junction of the Miami and Whitewater. North of Miamitown at Valley View, a terrace of this age rises a little above 560 feet. This is in the narrower part of the Miami trough. It consists of coarse and fine gravel and sand, at places confusedly crossbedded. A large tract here is owned and reserved for gravel by the Chesapeake & Ohio Railroad. The Willey Construction Company also operates a large plant for the supply of commercial gravels a little farther north.
At Valley Junction, near the mouth of the Whitewater, the large gravel pits worked by the Big Four Railroad are in a terrace of this age rising 540 or more feet above the sea.
From the altitudes here given it appears that the down stream slope of the valley train was decidedly less than the present fall of the stream. The Miami falls from 625 feet at the northern edge of the area to about 440 at Valley Junction, a distance of thirty-five miles, measured along the axis of the valley, but ignoring the turns of the stream. In the same distance the altitude of the terraces falls from about 650 feet at Middletown to 540 feet at Valley Junction. The fall of the stream is 185 feet; that of the terrace level is 110 feet. This suggests two principles; first, large stream need less fall than small ones in order to carry their loads. The Miami no doubt carried more water then than now. Second, it was a depositing stream. The Miami is now essentially a graded stream, but is eroding rather than depositing. If its gradient were again reduced to what it was then, it would probably be unable to carry even its present load.
Little Miami Valley Train.-The corresponding terraces along the Little Miami begin at Loveland where the terrace rises above 600 feet. North of that the newly adopted valley is so narrow that the stream has succeeded in washing out whatever deposits may have been made. The stream here was necessarily swift, and the materials dropped were chiefly heavy stones. The gravel exposed at Loveland is made up largely of heavy fragments of limestone, some of them a foot in diameter. Descending the stream, the valley widens in irregular fashion, the terrace level falls, and the material becomes less coarse. It is present on one side of the river or the other for most of the distance to Terrace Park. That village takes its name from a beatiful level terrace nearly a mile square, rising abruptly sixty feet above the flood plains, and abutting against the steep bluff on the west.
South of Terrace Park, remnants of the old valley train become more and more sparse to the mouth of the river, but fine remnants are seen both south and east of Newtown. The flat-topped ridge or hill stretching from Red Bank to Linwood is of the same origin. Here the material has a large proportion of sand, but through most of it is distributed gravel of medium coarseness. Cementation is locally complete.
The same relations as noted on the Miami between the gradient of the valley train and that of the present stream, are observed again here. The altitude of the terrace remnants falls from above 600 feet at Loveland to about 530 feet at Red Bank, a total fall of 70 feet. In the same distance the stream level falls from about 565 feet at Loveland to 455 opposite Red Bank, a fall of 110 feet. In other words, the terrace surface is 35 feet above the river at Loveland and it is 75 feet at Red Bank, indicating that the fall of the stream when the outwash was deposited was much less than its present fall.
Mill Creek Valley Train.-As Mill Creek Valley was the seat of the chief lobe of the Wisconsin ice sheet and the locality of its most southerly extension, it should also have furnished the chief channel of outwash. This does not seem to have been the case. As seen below, the Cincinnati basin appears to have been filled more largely from the east than from the north. Remnants of a valley train of Wisconsin age are not abundant south of Lockland. North of that there are few exposures, either natural or artificial, and it is not known how much of the "sand and gravel" mentioned in well sections is of this age and how much is older.
All of the gravel in this valley is relatively fine, and the terrace remnants are much more largely sand than gravel. One of these on the west side of Bond Hill has been excavated for sand to a depth of forty feet. A part of the sand is picturesquely crossbedded, indicating active local currents, but there is very little gravel. Its surface is flat and a little above 540 feet, this being the same level as the surface of the similar filling in the Cincinnati basin. It is a fair inference that the entire valley north of this place was filled to this altitude or higher. Similar deposits in Reading are twenty feet higher. North of that the floor of the open valley, parts of which are on outwash of Wisconsin age, rises gradually to more than 600 feet near Lindenwald, where it falls off into the more recent trench of the Miami.
How far below Bond Hill the filling of Mill Creek Valley continued is not known. There are no remnants lower down in the valley which rise to the full height of 540 feet. Between Cumminsville and the mouth of Mill Creek there is evidence that stagnant water existed during the latter part of the time that these valley trains were being constructed, or very soon after. This evidence consists in fine laminated silt which is exposed near the stock yards to a thickness of more than twenty feet. Its surface is at least a few feet lower than that of the terraces at Bond Hill and in the Cincinnati basin, indicating that this stagnant basin was not completely filled with silt.
A suggestion as to the origin of this basin is that the mouth of Mill Creek was dammed by the valley train along the Ohio. If the 540 foot deposit which now partly fills the Cincinnati basin was extended across the mouth of Mill Creek, the possible fall in the total length of Mill Creek Valley was about sixty feet, that is, from 600 where it joins the Miami Valley to 540 where it joins the Ohio. The distance is about twenty-five miles, measured along the axis of the valley, without allowing for any meanders. This is less than two and one-half feet per mile, a fall sufficient for a large stream, but not sufficient to enable the smaller volume of water which traveled through the Mill Creek Valley to carry its load through to the Ohio.
At the north the Mill Creek Valley may be said to end at the 600 foot contour line, which runs nearly south from the bluff east of Hamilton and Lindenwald. The Miami Valley west of this line is from twenty to thirty feet lower than the Mill Creek Valley on the east. Locally the fall is abrupt, and the higher level on the east stands out as a distinct terrace. From this line eastward to Flockton the valley contains no stream. It is followed by the Miami and Erie Canal, whose constant level shows that the valley level falls slightly toward the east. The floor of this valley is in part flat and in part very faintly rolling. The latter parts are ice-laid drift, while the flat parts are sand, gravel, and clay, that is, water-laid.
The altitude of the sand and gravel floor of this valley is exactly the local level of the Miami Valley train. It appears that at some time after the glacier had receded northward beyond the junction of these two valleys, and the valley to the south had been much aggraded with outwash, the Miami divided at Lindenwald, one stream following its present course, the other following Mill Creek Valley. To what extent this occurred during the recession of the glacier which invaded this area, and to what extent during the life of the late Wisconsin glacier which stopped farther north, is not clear. At all events the Miami during its aggrading stage was bifurcated at this point. The river at that time was not cutting a channel. It flowed, no doubt, in many small and shifting channels between shifting sand bars, as is the habit of overloaded streams. Its behavior could not have been very different from those parts of the present Platte River which are aptly described as "braided." When the glacier disappeared and the overloading stopped, the stream began to cut a channel in its former deposits. The fork which followed the present course of the Miami then had the advantage of a shorter and more direct course, and hence larger fall than the fork which followed Mill Creek Valley. Hence the former was cut down more rapidly and took more and more of the water until the old course on the east was left dry.
The conditions thus depicted offer another suggestion for. the deposit of silt in the valley of Mill Creek near its mouth. It will be remembered that this part of the valley failed to be filled with outwashed sand and gravel as was the valley north of St. Bernard. When the Miami no longer discharged a part of its water through the Mill Creek Valley, the unfilled lower course would be without current, and periodically filled by back water from floods in the Ohio. From such quiet flood waters, deposits of silt would be made very similar to those now found in this valley.
Ohio Valley Train.- Many terraces in the immediate valley or trench of the Ohio show that that valley was filled to an almost uniform height somewhat above 540 feet. From east to west across this area there is little or no fall in the height of these terraces. The map shows their locations at the mouth of Fourmile Creek, Ky., at California, in the Cincinnati basin, at Sedamsville, opposite Delhi, at Sayler Park (formerly Home City), and opposite North Bend. So far as observed, their materials are a medium grade of gravel with much sand.
By far the most important remnant of this valley train is the terrace on which the business portion of Cincinnati stands. Deep wells indicate that the maximum thickness of mantle rock is 185 feet, that is, from the surface at 545 feet to bed rock at 360 feet. Sand and gravel form the bulk of the material, with probably more sand than gravel. It appears that the upper thirty to sixty feet are generally free from clay, but that considerable clay is found at intermediate depths.
A study of the pebbles of this terrace was made by Orton (See note 21) who classified them with reference to their derivation. His results are shown in Table VI.
The striking thing in this classification is the ten per cent of stone which could not have been derived from the basins of tributaries from the north, but must have come down the Ohio. This estimate is for the gravel exposed in the digging of cellars on the top of the terrace. The proportion of sandstones from the east may be still larger in the lower beds. Taking this fact in connection with the incomplete filling of Mill Creek Valley, it appears that the Cincinnati basin was filled chiefly by material which came down the Ohio and not down the Mill Creek Valley. However, the large proportion of stones from western Ohio indicates that the river derived its load largely from the Little Miami, Scioto, and other streams which enter the Ohio not far above Cincinnati. To what extent the valley train in the Cincinnati basin was completed at full height across the Mill Creek Valley is uncertain.
Silt-Filled Valleys.-The valley of the Licking, which heads to the south and flows north could, of course, not be filled with glacial outwash. It is observed, however, that the larger part of the valley of this stream (within this area) is floored by terraces whose level is essentially the same as that of the valley train. Covington stands on such a terrace and the entire Milldale (or Latonia) basin to the south has its floor at the same level. Many exposures show that these terraces on the Licking are composed of silt and not of sand and gravel like those on the other side of the Ohio. The explanation of this is that the detrital filling in the Ohio constituted a dam across the mouth of the Licking, thus ponding its waters. In the valley thus filled with standing water the mud brought down by the Licking itself was allowed to settle. Before the Ohio again entrenched itself, and thus removed the obstruction from the mouth of the Licking, the valley of the latter had been filled up almost or quite to the level of the ponded waters. To an undetermined extent, such filling was doubtless aided by back water from the Ohio during floods.
A small terrace in the valley of Fourmile Creek, where it leaves the Kentucky bluffs southeast of Fort Thomas, is of the same character. A similar remnant at an altitude of 560 feet or more is found on the north side of the valley of the East Fork of the Little Miami, some two miles east of Milford. It is only to be expected that the valley of the East Fork would be obstructed by the valley train along the Little Miami. Probably the former was in large part filled with silt from its own basin which has been since removed. A similar instance is found in a small valley one mile south of Terrace Park.
Two of the best examples of such silt terraces are found in the eastern tributaries of the Miami. The valley of the west branch of Taylor Creek, which joins the main stream opposite Miamitown is largely occupied by such a silt deposit, which has been exposed at various places by the grading of the Chesapeake & Ohio Railroad (C. C. & L.) and by stream erosion. Here the fine silt in the small valley becomes first sandy, then gravely, as it is seen in successive exposures nearer and nearer to the main valley. Doubtless the silt deepened in the side valley at equal pace with the deepening of the sand and gravel outwash in the main valley, the currents in the latter affecting the water in the former only a short distance from the mouth. Jordan Creek which enters the Miami two and one-half miles north of Cleves has some terraces which are remnants of similar filling.
An excellent example of what is probably the same feature is seen in the valley of Ludlow Run, a small northern tributary of Mill Creek, followed by the Cincinnati, Dayton & Toledo Traction Line west of Spring Grove Cemetery, Cincinnati. No remnant of the Mill Creek Valley train at full height is found so far down the valley, but less than three miles to the northeast is the excellent terrace near Bond Hill, already described. The surface of this is above 540 feet. It may well be that this filling extended down the valley far enough to obstruct the mouth of Ludlow Run. The consequence of such obstruction would be the aggradation of the side valley with silt from its own basin. The silt terrace is best preserved and exposed about one-half mile north of the terminal station of the traction line. Here it consists of finely laminated ' silts interbedded with more sandy layers. The uppermost silt layers abound in calcareous concretions, and in some of the layers are small coiled shells. The surface of this terrace, already somewhat lowered by erosion, is 540 to 545 feet above sea level. It is not entirely certain that it is not of the same formation as the silts lower down in Mill Creek Valley already discussed. Its sandy layers and its greater altitude seem, however, to indicate wash from up the valley rather than subsidence from backwater of the Ohio. (Plate. VII-A.)
Kames.-Among deposits made by glacial waters at the edge of the ice, the most important are the kames which cover about two square miles centering at Camp Hageman where the Cincinnati, Lebanon & Northern Railroad crosses the old pre-glacial valley of Todd's Fork. These are abrupt gravel hills, some of them rising 100 feet above the flat valley floor, and affording the most striking morainic topography in the region. Their material is coarse and fine gravel and sand. Much of it was confusedly crossbedded when laid down and large masses have since slumped, perhaps when the ice melted away, so that the bedding is now very confused (Plate. V-B). Many of the individual stones exceed one foot in diameter and a considerable proportion of the materials is of crystalline rock from the far north. Nowhere is the lithologic variety of the younger drift better illustrated.
The exact circumstances which brought about so large a deposit of water-laid material at this place are not apparent. The kames were of course built when the edge of the ice was at this place. In general, however, it is safe to say that where a not very thick ice sheet crosses obliquely so large a valley as that of the former Todd's Fork, there is apt to be a depression of its surface corresponding to that of the ground beneath. Both the surface drainage and the subdrainage of the ice are likely to be concentrated along such a line. Moreover, the chances of an irregular, and locally indented edge of the ice are greatest where disrupted in crossing a valley. All these conditions favor kames.
Another good group of kames, but inferior to those at Camp Hageman, stretches from near Schencks Station, south of Hamilton, to beyond Furmandale. They lie mainly in a narrow north-south belt to the west of the Mill Creek Valley Traction Line, but are crossed by that line at Furmandale. The highest of these rise about fifty feet above the surrounding flat.
The circumstances giving rise to kame deposits at this place are apparent. The ice sheet moving south across the broad valley came against the upland which presents an angle to the north. There was a tendency of the ice to split on this headland and divide into two lobes, one going southeast, the other southwest. Probably it actually did this to some extent. The important thing here is a tendency to longitudinal crevassing, or at least to a separation between two lobes during the recession of the ice front. In such an indented front kame gravels would tend to accumulate. Another factor favoring kames at this place is the tendency to stagnation. This may indeed be the most important factor. Cracks in stagnant ice enlarge by melting and furnish appropriate recesses or pockets into which glacial waters carry their load, and where they leave it partly supported by ice walls. In the gravel pits in these kames the beds may be seen bent, broken, and faulted by the withdrawal of the supporting ice.
On these kames near Furmandale are thin patches of true loess, containing the ordinary loess fauna. It is noteworthy because of its occurrence on the younger drift. As a vast area of valley bottom lies to the west (windward) it will be observed that this was an appropriate place for the accumulation of loess according to the principle stated above.
The great changes in the drainage of this region were probably all made during the first glaciation. Certain minor changes which are of interest seem to be connected primarily, if not exclusively, with the later advance of the ice.
Narrows.-Fourmile Creek is a western tributary of the Miami near the northern edge of the area. It follows essentially its old preglacial course, generally between bluffs one-half mile to one mile apart. Its old valley is, however, filled with alluvium (partly glacial outwash) to a depth of 150 to 200 feet, and the river runs over this material. This is known partly by the depth of the filling in the Miami Valley which necessarily raised the mouth of its tributary, and partly from wells in the valley itself. At Oxford (just beyond the boundary of this area at the northwest) a well near Fourmile Creek was drilled through 187 feet of alluvium before reaching bed rock. Running over this material, the stream, while confined to its old valley, is not in its old channel. Locally it has meandered so near the bluffs that rock is exposed in its present channel. Several fords, and at least one bridge, are located at such points. The only marked dislocation from its former trench is two or three miles west of Darrtowm. Here the stream makes a sharp bend to the south and runs some distance in a gorge, which, at the narrowest place, has just the width of the stream. The gorge is more than 100 feet deep and is cut 30 feet into the bed rock. It is evident that the stream has here been crowded far up on the south side of its former valley. The hills within the bend on the north side are, so far. as seen, entirely of drift, and, no doubt, obscure the old valley. Indian Creek is similarly crowded to the south at Bunker Hill and flows in a similar rock gorge, while elsewhere it follows its old trench now deeply filled, but bounded by its old bluffs one-half to three-fourths of a mile apart.
Simpson Creek is a small stream three or four miles long which flows east by north and enters the Little Miami at Foster. It has its upper course outside the area of the later drift and in a wide open valley. The last mile of its course is through a narrow and rocky valley which plunges steeply into the Little Miami. The interpretation here, as in the cases mentioned above, is that the lower course was crowded southward by the ice. This case is of interest because of the excellent contrast between the deeper and more fertile Wisconsin drift on the north and the thinner, silt-covered, more eroded, and less fertile older drift on the south. As in all the other instances cited of minor drainage changes, erosion in the new portion of this valley has exposed a good section of the underlying rock.
Lakes.- Where the edge of the ice rested against high uplands, some of the smaller streams were ponded, and the resulting ponds or lakes were partly filled with lacustrine sediments. On Banklick Creek in Colerain Township, Hamilton County, an interesting record of such conditions has been preserved. Here the Wisconsin ice sheet, coming from the north, covered only the edge of the upland south of the Miami, thus obstructing the north-flowing creeks. The ponded waters of the Banklick here rose to a height of at least 690 feet, that being the approximate level of the deposits made. This lake was filled in its central portion with sand and silt, and near its edge with sand and gravel. Near the bridge at the west side of section 7, the creek has cut into its bank, making an exposure forty-five feet high. Four beds of alternating sand and silt are shown, the topmost bed being silt. The sand is much crossbedded. Apparently the dip of the crossbedding is down the valley toward the north. This would indicate considerable current in that direction, probably during intervals when the ice dam was not .effective. A number of small remnants of deposits at the same level are found farther up the valley.
The West Fork of Mill Creek flows north from Mount Airy to near New Burlington, where it turns and takes an easterly course to near Glendale. Its upper course is outside the limits of the newer drift sheet. In its eastward course it enters the area of the newer drift about two and one-half miles southwest of Glendale, in the northwest comer of section 16, Springfield Township. For nearly a mile east of that point the creek runs in a narrow valley 75 to 150 feet deep, which is locally a sharp gorge cut many feet into the solid rock. This is due to the local crowding of the stream to the south out of its old valley. At the end of this gorge it again finds its old valley, or that of another pre-glacial stream, and follows this to its junction with Mill Creek proper. The ice, during its presence, and after that for some time the drift, ponded the upper stream, forming a lake, whose level was a little below 700 feet. This lake was almost filled with silt, remains of which are now well shown in the flat ground in the northwest comer of section 16. Most of the valley above this place is so narrow that the stream has since carried out the lacustrine silt, but small remnants are still found. Along a small northern tributary once known as Whiskey Run, about one mile east of New Burlington, the silt appears in the banks. About one-half mile above its mouth the skull and two teeth of a hairy mammoth (Elephas primigenius) were found. This animal flourished in late glacial time. The fossil was found embedded in gravel just beneath the lake silt and a little below the level of the stream.
Sharon Creek is a small stream four or five miles long which flows southwest, and emerges from the east bluff of Mill Creek Valley at Sharonville. Its upper course lies in a broad valley just at the edge of the early Wisconsin ice sheet. About one mile above Sharonville it suddenly enters a sharp gorge more than 100 feet deep, and cut fully fifty feet into rock. The Pleistocene map shows that at this place the glacier pushed to the southeast in a small lobe which crossed the course of the creek and crowded it out of its old valley. The former valley is now filled with drift and cannot be seen. The valley above the obstruction became a lake, and was filled with lacustrine clay to an altitude between 760 and 770 feet. Some of this clay is finely laminated and very plastic. Most of this filling was accomplished before the ice withdrew. This is known from a section just above the rock gorge where a sheet of till rests on the clay. Most of it, however, remains uncovered. The cutting of the rock gorge was necessarily slow as compared with erosion in clay. Hence the stream above the gorge has not only trenched the clay to a depth of thirty feet, but widened this trench to about 400 feet.
Development of Drainage.-The Pleistocene or Glacial epoch ended with the retreat of the Wisconsin ice sheet, and all subsequent time is known as the Recent epoch. The geological work of this epoch has been largely erosion, but there have also been deposits, especially of alluvium. These include that which is now forming on our flood plains and that which covers the alluvial terraces lower and younger than the Wisconsin outwash. The changes brought about during this epoch are therefore similar to those described after the Illinoian stage. The streams which run over the drift sheet are, in their present cycle, largely consequent, that is, they follow courses determined by slopes which they found. In the older chapters of their history, many or most of these valleys were no doubt subsequent, that is, due essentially to the work of the stream instead of being found and followed by it. The larger side streams have altered the original troughs which they followed, by carving within them stream-made valleys with bluffs, in some cases separated by flats, but the smaller drainage lines with temporary streams are not true stream valleys. They are purely constructional features. In some cases gullies have developed either in the constructional troughs or on their slopes. These have been growing headward, thus lengthening some of the original valleys or increasing the number of tributaries.
Revived Streams.-Certain valleys immediately tributary to the great valleys, show the features of recent revival very strikingly. One of these is the valley of West Fork which enters Mill Creek at Cumminsville. Up stream from the schoolhouse (about one mile West of Cumminsville) the immediate valley of this stream is a gorge from ten to thirty feet deep, cut largely in the Eden shales. The cutting is so fresh that the walls are at many places precipitous, and the channel shows a succession of small falls or rapids where the stream passes over the limestone beds which occur at intervals in the shale. The level in which this gorge is cut is itself the bottom of an old valley more than 200 feet deep and one-half mile wide at the top. The relatively level bottom is from several hundred to 1,000 feet wide. These are the prominent features of a revived stream. (See Fig. 41 and Plate IX.)
It is plain that at a time not very remote, the stream was flowing at the level of the top of the present gorge, that is, at the level of the floor of the broad valley. This floor near Cumminsville is well below 500 feet, and is 400 to 600 feet wide. It therefore represents a well established graded course at a time later than the Wisconsin stage, for at the close of that stage the mouth of this valley must have been filled to the level of the partly filled Mill Creek Valley. Up stream from Cumminsville the old valley floor rises with a normal stream profile. One-half mile above Cumminsville the present stream flows at the old level, not being entrenched. One-half mile farther up, the stream flows in a vertical sided gorge thirty feet deep and its tributaries do the same. Up stream from that point, the depth of the new gorge gradually decreases.
It is plain that the stream does not now need so steep a gradient as it did in its former condition. This in turn indicates that the stream has greater carrying power and cutting power than before. It is therefore cutting down its upper course without changing the level downstream, thus flattening its profile or reducing its gradient. This much is certain, though the exact reason for its increased efficiency may not be clear. Probably it receives no more water than it did in the former condition. On the other hand it is plainly handling much larger stones than formerly. This may be seen by comparing the great slabs of limestone in the present channel with the rubble which occupied the channel at the higher level (Plate. III). The natural assumption is that since the removal of the once luxuriant forests the run-off is much more prompt, causing greater freshets, and thereby increased power; for it is well known that, so far as erosive power is concerned, a given amount of runoff is vastly more effective if concentrated into freshets than if discharged at a uniform rate. This supposition as to the reason for the "rejuvenation" of West Fork agrees with observations on certain other creeks in this area, some of which (though much smaller than West Fork) have been known to cut canyons ten feet deep in less than fifty years since the removal of the forests.
The post-glacial work of the large streams has consisted mainly in removing most of the material of the valley trains. In doing this they have lowered their own levels from 30 to 200 feet. If all the larger valleys were again filled up to the level of the present terraces of Wisconsin age, the amount of material thus used would be that which has been carried away since the glacier departed. Along the Ohio this would mean a fill of 190 feet in the axis of the valley. In the Miami Valley at Hamilton it would require a fill of forty or fifty feet. The work of deposition since the last glacial stage, has been done mainly by streams on their flood plains. In this way they have coated their bottom lands with sand and gravel to a depth somewhat greater than that of their channels, and have spread over this a sheet of loam ranging in thickness from zero to ten or fifteen feet.
Wind has not performed much work of geologic importance since the last glacial stage, but it has raised some sand dunes and even deposited a little loess. The former are exemplified at Ludlow, Ky., where, unfortunately, they are being fast cut away in excavation for gravel. They partly covered the gravel terrace of the Wisconsin stage between the railroad yards and the Lagoon, rising ten to fifteen feet above the gravel surface. It is significant that the activity of the wind in recent geologic time should thus be attested in the same vicinity where there is such abundant evidence of its work the epoch when the loess was deposited. It would not be surprising to know that the loess on the uplands to the east had again deepened at the time the dunes were made.
What seems to be dune sand covers a small area in the Mill Creek Valley north of Reading and east of the International Agricultural Works. The area thus covered is the south end of the elliptical spot shown on the map as underlain by till of Wisconsin age. The sand here covers the mildly rolling surfaces three to five feet deep. Its grains are well rounded and consist of quartz and other minerals derived from the igneous rocks of the drift. Probably it was carried up by winds from the alluvial flats on the west.
That wind was an active agent in this valley in post-Wisconsin time is made certain by the presence of true loess resting on the newer till at the foot of the east bluff at the northern edge of Reading. It is identical in character with that at Covington and Newport, containing also small pulmonate shells. The deposit is from three to seven feet thick and may well have settled on the bluffs at the same time that the heavier sand was drifting on the flats to the west (Fig.52). The thin covering of loess on some of the kames south of Hamilton is exactly analogous to this occurrence.
From the economic standpoint, by far the most important geologic process since the glacier disappeared, has been the making of the soil. This has been accomplished on the newer drift by the weathering process described in Chapter IV and by the intermixture of decaying vegetation. The soils of the area are described in Chapter VIII.
