AN mvmmmn or: m: morimmcs 09 me seats on GLACIAL momma or my AND MANmAm me an SANILAC scum mmm rhea: Cor the Dunn of M. s. MICHIGAN 5mm mums Bonnie L. Alien 193‘! THESIS This is to certify that the thesis entitled An Investigation of the Characteristics of Some Soils on Glacial Moraines of Cary and Mankato Ave in Sanilac County, Michigan presented by Bonnie L. Allen has been accepted towards fulfillment of the requirements for Master of Science , Soil Science __ degree in — Major professor Date September 14; 1951 0-169 AN INVESTIGATION OF THE CHARACTERISTICS OF SOME SOILS ON GLACIAL MORAINES OF CARY AND MANKATO AGE IN SANILAC COUNTY, MICHIGAN BY BONNIE L. ALLEN M A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Soil Science 1951 THESlS ACKNOWLEDGMENTS The author wishes to express his sincere thanks to Dr. E. P. Whiteside of the Soil Science Department for suggesting the problem and for his constant guidance in conducting the research. To Mr. I. F. Schneider of the Soil Science Department and to the personnel conducting the soil survey of Sanilac County during 1950, the author is indebted for the help received while studying some of the soils of Sanilac County. To Dr. A. E. Erickson of the Soil Science Department for helpful suggestions concerning the laboratory work and to Dr. S. G. Bergquist of the Ge010gy Department for information regarding the ge010gy of eastern Michigan, the author is indeed grateful. 268390 TABLE OF CONTENTS INTRODUCTION REVIEW OF LITERATURE . . . . . . Soil Forming Factors Vegetation Climate . Parent Material Relief The Time Factor Leaching of Pleistocene Calcareous Materials FIELD STUDIES . . . . . . Depth of Leaching of Calcareous Morainic Soils . Soil Profiles LABORATORY STUDIES Procedures Till Sample Preparation Determination of Carbonate Content Me chanical Analyse s Page 10 12 21 21 28 38 38 38 38 39 Infiltration Rates Non—Capillary Porosity Capillary Porosity Total Porosity Volume Weight Experimental Results DISCUSSION OF FIELD STUDIES DISCUSSION OF LABORATORY STUDIES SUMMARY . BIBLIOGRAPHY APPENDIX iv Page 41 42 42 43 43 43 5 l 56 7O 73 76 INTRODUCTION Sanilac County is located on the eastern side of the Lower Peninsula of Michigan in what is pOpularly known as the "thumb. " The complexity of the glacial deposits in the area is well known among glaciologists, and these variations, together with variations in other soil formation factors, especially relief and age, have given rise to a complex pattern of soils. The county is situated in a transitional zone between the Gray— Brown Podzolic and Podzol Great Soil Groups. As the soil survey of Sanilac County has prOgressed over a period of years numerous ped010gic problems have arisen. The surveyors in the field, as well as other soil scientists, noticed that the soils deveIOped on the glacial moraines of Cary and Mankato age in the area had been leached of carbonates to dif— ferent depths, but that the leaching depth on the moraines of each age was fairly constant. Sanilac County presents an excel— lent Opportunity to study soils deveIOped on glacial materials of these different ages. An investigation of this problem was initiated in an at- tempt to quantitatively evaluate the variations in the depth of leaching and to determine what soil forming factor or factors have been dominant in determining the depth of leaching and the relationship of the factors to soil prOperties. In initiating the study it was believed that any informa— tion obtained on the problem would be, at least, some contribu— tion to the existing information on soil formation in the region, and that the information would be of benefit in any future clas— sification of the soils. Further, it was believed the information would be of some use to geologists in their study of the glacial materials in the region. Since the depth of the solum would certainly affect man— agement practices on almost any soil, the depth to which the carbonates have been leached would, to a certain extent, affect the use of the soils under investigation. Already, there have been implications that the abundance of calcareous materials only a few inches from the surface in certain places has affected adversely the release of certain vital minerals to cr0p plants. This problem could become even more serious if erosion were allowed to proceed rapidly on the steeper lepes in the cultivated areas. The depth of the carbonates could, in addition, affect certain physical prOperties of the soils which, in turn, would influence drainage as well as irrigation. REVIEW OF LITERATURE Soil Forming Factors According to Jenny (12) the fundamental equation of soil forming factors may be expressed as: s = f(cl, o, r, p, t . . ..) The s in the equation stands for any soil prOperty such as pH, porosity, alumina content, etc.; f stands for ”function of," and the symbols enclosed in parentheses stand for the independent variables or soil forming factors as follows: c1 = Climate o = Organisms r = T0p0graphy p = Parent material t = Time The dots in the equation indicate that possibly there are unknown factors that enter into soil formation. As an example, assume that the equation be evaluated for one factor, namely climate; then we obtain the following expres- sion: 5 = f(C11mate)t, o, r, p . If the climatic function is to be evaluated accurately all the other factors must remain constant. This may be done under ' controlled laboratory conditions but under natural conditions the value can only be approximated. It is generally believed by soil scientists that the fac— tors, parent material, time, and relief, have been dominant in determining the leaching depth of carbonates in the region under investigation because the climate and vegetation is approximately the same in each of the two areas studied within the region. Consequently, the literature concerning the time, parent mate— rial, and relief factors as they have affected soil deve10pment in the areas has been searched thoroughly. Vegetation The original vegetation of Sanilac County was of the mixed coniferous—deciduous type with the hardwoods being more prominent on the heavier morainic soils and evergreens being dominant on the sandier and poorly-drained areas (10). A large percentage of the land is now in cultivation and the lands that are timbered are occupied mostly by second—growth trees. Smith (23) reported that vegetation has exerted a pro— nounced effect on the depth of leaching of loess in Illinois. The leaching depth of soils deve10ped under forest vegetation averaged 82 inches; whereas, those deve10ped under grass showed an av— erage of 42 inches. The soils in both areas studied in this investigation originally supported a very similar type of vegetation; there— fore, it seems unlikely that vegetation has been an important factor in determining the differences in depth of leaching. Climate The present climate of the area is characterized as continental grading into a semimarine type near Lake Huron. There are wide ranges in temperature varying from -—250 in winter to slightly over 100° in summer; however, these extremes are the highest ever recorded for the county and are of rare occurrence. The annual precipitation average at Sandusky, the county seat of Sanilac County, is 28.87 inches. The time be— tween the first and last killing frosts averages 130 days (29). The soil is generally frozen for several weeks during each winte r . Parent Material Geologically the larger part of the county consists of rather level till plain and lake bed plain with an average ele— vation of 200 feet above the level of Lake Huron to the east. In the east, north, and southwest portions of the county there are morainic formations ranging from 60 to 100 feet above the general plain level (10). Since it was the soils of these mo— raines that were investigated the manner of origin of these deposits will be reviewed briefly. The Pleistocene history of the region has been studied in great detail by Leverett and Taylor (18). The region was overridden four times by continental ice sheets moving from the general area of what is now Labrador. These advances were during cold cycles, and the retreats of the ice sheet were a result of warmer interglacial periods. These interglacial periods are believed to have been of much longer duration than the glacial periods (8). During these warm periods soil forma— tion and erosion were active on the newly deposited glacial material (26). The present surface features of the county and soil parent materials are, for the most part, a result of the fourth and final glaciation (the Wisconsin). However, much of the Wisconsin drift is underlain by pre-Wisconsin glacial material which may be observed in places along the lake shore and along the Black River. The Black River roughly parallels the Port Huron moraine in the eastern part of the county. Underneath the indurated glacial drift the Coldwater shale appears in the eastern part of the area, including the Port Huron moraine; whereas, the Marshall sandstone is encountered above the Cold— water formation in the western part of the county, and lies immediately below the glacial accumulations (10). The thick— ness of the drift in the moraines varies from 20 to over 250 feet. Relief The Deanville and Yale moraines (the latter merging into the interlobate area in the vicinity of Marlette) in the south- western part of the county, are dated from the Cary Substage of the Wisconsin age. These moraines are conspicuous ridges of glacial materials deposited when the ice front was stationary for a short geological time on its northeastward retreat and are broadly oriented with the axis in a general northwest— southeast direction. The tOpography in most of the morainic areas is characterized by Leverett and Taylor (18) as swell and sag but containing a considerable number of basins and swampy recesses in some parts. The elevation of the area is approximately 800 to 850 feet with a relief varying locally from 10 to 50 feet. In (certain areas of) the southern part of the county there are located some so—called transverse ridges. They are oriented almost perpendicular to the moraines described above. One of these ridges is crossed by the Yale moraine in the vicinity of the village of Speaker. According to Leverett and Taylor (18) the origin of these ridges is somewhat obscure but seems to be related in some way to drainage lines in or under the ice, but these ridges are not very similar to esker forms. These transverse ridges bear a strong resemblance to morainic deposits and have a t0pography and relief much like the near—by moraines. The strongly deve10ped Port Huron moraine, roughly paralleling the Lake Huron shore, is in the eastern part of the county. It marks the limits of a distinct readvance of the ice as distinguished from the halts which were responsible for the deveIOpment of the afore—mentioned moraines. While the ice 10 front was standing at this point much of the immediate area to the west was occupied by glacial Lake Whittlesey which discharged its water to the west through the Tyre—Ubly Channel into Lake Saginaw (8). The Port Huron moraine has elevations of some 30 to 40 feet above the old lake plain on the west and rises to a maximum of 180 feet above the present level of Lake Huron. The t0pography may be characterized as swell and sag with comparatively few basins and with a local relief of 50 to 60 feet. Actually the Port Huron moraine is not just one large moraine but is composed of three or four parallel subordinate ridges which have long sags between them, thus, though the bulky ridge is compound, its separate elements are related in the simplest possible way. The Time Factor Robinson (22) considers the time factor to be on a dif— ferent footing from other factors in that in itself, it cannot have any effect, but intervenes in governing the amount 01‘ extent of Operation of other factors. By examining the equation of Jenny (12) it can be seen that the length of time that it would take a so—called mature ll soil to deve10p would depend on the other factors. Jenny con— siders that a material becomes a soil from the moment that the active soil—forming factors begin to act. It has been reported that certain visible changes can be noted in till deposited as end moraines by present—day glaciers in a few decades (7); nevertheless, pedologists generally agree that many of the glacial soils of the United States are immature. Certainly, the soils under consideration in this study cannot be considered as fully mature. As to the length of time that ac— tive soil—forming factors have been in action on these materials only approximate answers are available. Antevs (2) has esti— mated the time elapsed since early Mankato as 27,000 years and the time since the culmination of the Cary as 36,000 years. Most geologists have believed until recently post—Mankato time to have been something like 28,000 years and the Cary—Mankato interval, called the Two Creeks interglacial period by many, to have been approximately 10,000 years. Within the last few years there has been much interest among ge010gists, as well as other scientists, in determining the age of well preserved carbonaceous materials by means of 14 radioactivity. Since radiocarbon (C ) has a long half—life of 12 5,568 :i: 30 years it is supposedly possible to determine the age of material'by measuring its present radioactivity. In a series of measurements Libby and Arnold (19) estimated the age of several spruce samples from Two Creeks, Wisconsin, from 1,877 :h 740 to 12,168 :I: 1,500 years. It is thought that the trees were buried by the Mankato ice sheet. This new approach to the problem of dating relatively recent organic deposits of— fers much promise. It is possible that the commonly used estimates for post—Mankato time are too high. Accordingly, the time since other events in the Wisconsin are possibly over— estimated. Bartlett (3) has discussed some of the limitations in using this method. Leaching of Pleistocene Calcareous Materials Kay (13), using 25,000 years for post—Mankato time, esti— mates the time elapsed since the Iowan (earliest substage of the Wisconsin) at 55,000 years. As a basis for this estimate he used the depth of leaching of materials of the two ages. Since the Mankato deposits were leached of calcium carbonate to a depth of approximately 30 inches, and since the Iowan materials were leached to a depth of somewhat less than 6 feet, he reasoned 13 that the Iowan was 2.2 times as old as the Mankato drift. Kay found various textures of till, gravel, and loess of the same age to be leached to about the same depth. He recognized the fact that as the carbonates were leached to greater depths the rate would probably be slower but did not take this into account in making his estimates. Neither did he take into account the original carbonate content of the material in making his esti— mates. In making a study of weathered zones on pre—Wisconsin drift in Illinois, Leighton and McClintock (17) made a division of the weathered zone as follows: Horizon l: The surficial soil. Horizon 2: Chemically decomposed till, composed chiefly of alteration products and resistant constituents of the original till and strikingly unlike the original till. Horizon 3: Leached and oxidized till, otherwise but little altered. Horizon 4: Oxidized till, but unleached and otherwise un— altered. Below the oxidized till the material was found to be essentially unweathered till. They recognized that several factors deter— mine the depth of the various layers, and that the rate of their formation would certainly decrease with depth. Considering the 14 time factor to be approximately the same for the area studied, they recognized that differences would not only depend on the rate of weathering, but on the composition of the original drift, especially its calcium carbonate content, and on the position of the ground water table. Leighton and McClintock emphasized particularly the effects of topography, noting that the till in a well—drained position was weathered appreciably deeper than that deve10ped in a poorly—drained position. They also empha— sized the effects of the permeability of the parent material, finding that weathering proceeded more slowly on the heavier textured tills. Thornbury (25), working in Indiana on tills of Cary and Tazewell age, found that the border of the Cary (younger) drift could be mapped by determining the differences in depth of carbonate leaching. He found the difference averaged about 12 inches. He rec0gnized the fact that other factors, including the texture and composition of the drift, the height of the water table, and variations in tOpOgraphy, could likewise cause varia— tions in depth of leaching. Thornbury pointed out that a large number of samples should be taken to determine the average leaching depth with a fair amount of accuracy. Although the 15 depth of leaching on pre—Wisconsin drift was found to be only about twice as great, where covered with later drift, as that on the Tazewell drift, it was emphasized that the former repre— sented a very much longer interglacial period because of the decrease in rate as the carbonates were leached ever deeper. He also noted that the till was much more calcareous in areas overlying limestone formations, and that this, as well as better drainage in districts crossed by major streams, would affect the depth of leaching. It was reported by Winters and Wascher (33) that the clay content (4 5u) and especially the colloid content ((lu) exerted a pronounced effect on the depth of leaching of Wiscon— sin drift where the difference in age was supposedly slight, and where there was little variation in the carbonate content (19 to 22 per cent). When the clay content reached 68 to 75 per cent, with a colloid content of around 30 per cent, the percolation rate was extremely slow with a consequent reduction in the thickness of the overlying solum. The content of sand did not seem to be important until the amount exceeded 10 per cent, unless the amount of clay fell below 40 per cent, in which case the effects of the sand content were marked. Winters and Wascher 16 found that the soil prOperties were influenced to such an extent, that, for the purpose of soil classification, they rec0gnized and defined four till groups which are given below. Till Group Permeability Rating Textural Characteristics . Clay: 60-65% C1 arence Impermeable, plastic Colloids: 34—31% Plastic Elliot Slowly permeable, Clays 50—54% plastic C01101ds: 34—31% . Slow to moderate Clay: 36-38% Elhot . . . permeab111ty C01101ds: 16-18% Clay: 36-38% Saybrook Permeable Colloids: 18-16% Sands: 15% Although Winters and Wascher recognized these four textural groups, they realized there were gradations between the groups, and that there were continuous variations in depth of leaching, also. As a basis for estimating the permeability rating of the tills they used the known rating of certain soils from southern Illinois with a comparable textural composition. In the tills studied the samples with low carbonate content seemed to be high in colloidal content, the low carbonate 17 content being attributed to the less calcareous nature of the original material. Stauffer (24) was of the Opinion that soils deve10ped from Wisconsin (Tazewell) till in east—central Illinois were largely dependent on the pr0perties of the underlying till. In this in— vestigation it was pointed out that in the Clarence soil no C horizon had deve10ped; whereas, in the Saybrook soil a C horizon had deve10ped because of the more permeable nature of the till from which it was derived. In this study it was also noticed that the unleached till immediately below the leached horizon in the more permeable soil showed more pronounced indications of weathering, i.e. , color change and changed appearance of pebbles. Additional work concerning the leaching of calcareous glacial drifts is that of Andrew and thafles (l) and that of Kruger (15). The former, in comparing an undisturbed soil on glacial till and a very young soil deveIOping on a railroad cut, found that the material with an original carbonate content of approximately 15 per cent had decreased to 3.7 per cent in the surface 2 inches in 75 years. There was a gradual increase in carbonate content downward. In the undisturbed 18 soil the upper 44 inches had been leached thoroughly ((1 per cent) of carbonates; the horizon immediately below this leached horizon contained 15.1 per cent. Kruger suggested that the "gray" drifts of Minnesota had not been leached so thoroughly as the "red" drifts in the district because of the clayey im— pervious nature of the gray drift; however the analyses showed the "gray" drifts to contain a much higher percentage of car- bonates originally than the "red" drifts. Other reports concerning the leaching of glacial materials include that of Norton (21) in southern Illinois. The old glacial drift in the southern part of the area under investigation had been leached of its carbonate to a depth of 14 feet; whereas, the drift in the northern part showed a leaching depth of 11 feet. Norton suggested possible reasons for the difference in— cluding the original lime content of the drift or the more pro- nounced effect of the weathering in the southern part. The time factor was not mentioned. Mick (20) found that carbonates occurred at greater depths in the poorly—drained (Brookston and Napanee) soils studied in Sanilac County than in the better—drained and sup— posedly more thoroughly—weathered (St. Clair and Conover) 19 soils. This is contrary to what is generally expected. Pos- sible explanations offered by Mick included the dolomitization of the carbonates below the B horizon of the poorly—drained soils and the variation in the susceptibility to geologic erosion. Very often glacial morainic material has been reported as being quite heterogeneous in composition; but Krumbein (16), in a study of morainic material south of Lake Michigan, found the material deposited by any one ice sheet to adhere to a well—defined frequency distribution. The immediate underlying bedrock did not seem to be directly related to the litholOgic composition of the pebbles nor to the mechanical composition of the till; however, a change in the bedrock from limestone to shale was reflected in the lithology of the pebbles 40 miles away in the direction of the ice movement. The study showed that mechanical composition could be used as a means of dif— ferentiating tills under certain favorable conditions if "suites" of samples were used to counteract the sampling error. Although rec0gnizing that high lime content was usually a deterrent to soil formation, Smith (23) found the Muscatine catena to be deve10ped on loess deposits in Illinois with car— bonate content ranging from 16 to 33 per cent. Although the 20 vegetative factor (see page 6) has undoubtedly exerted a pro— nounced effect on the leaching depth, an examination of the table showing the original CaCO equivalent reveals that the 3 samples under grass vegetation averaged 5.5 per cent more than those under forest vegetation. It seems possible that the original carbonate content, as well as the vegetation, has played a role in determining the leaching depth. The authors whose works have been cited seem to be of the general Opinion that a combination of factors usually de- termines the depth of leaching of calcareous materials. How— ever, in a humid temperate forested region the factors, time, permeability of the parent material and resulting soil, and the chemical composition, principally carbonate content, of the parent material, are considered to be the dominant factors. The investigators have presented ample evidence to substantiate their Opinions regarding the effects of time and permeability; and, although the carbonate content was referred to often as a contributing factor, little evidence was presented showing any direct influence of the factor. FIELD STUDIES Depth of Leaching Of Calcareous Morainic Soils As has been pointed out earlier, several investigators have determined the depth of leaching on morainic materials in various places, and the importance of using numerous repli— cations in such a study has been emphasized in the literature. In this study 12 sampling sites were chosen on moraines (Port Huron system) of Mankato Age; and a like number was chosen on Cary Age moraines (Yale, Deanville, and Transverse Ridges). The areas chosen for sampling sites had, with few exceptions, a lepe Of approximately two per cent; showed no visible erosional effects; and the soil of the areas was a loam or silt loam. If the underlying drift showed decided stratifi— cation the site was not used. Further, the sites were chosen so as to cover a rather extensive area of the moraines. The depth Of carbonate leaching on each site was determined by measuring the depth at which the soil effervesced with cold dilute hydrochloric acid. This information, together with other data concerning the sampling sites, is shown in Tables I and II. 22 .mm .3 M .2 NH H. .om .Oom mvlfiv on v IN 6va ammo mdofid T9538 85¢. m .mcmxnoguuumcs moor» .H mu M .2 NH .H .mm 1.3 0&8 finance 21.3. m: 2 a .95 .03.... £3... muons 2.: v .23 backwmuw tnodwm «no? 95: 0:5 v.0 854 .MmHM.ZN~B.m Smog}. 33w :5. Slow. MN NM +N .oom .oEm Anson 95? on: m undo Scum 0:8 «.6 dnd. .23 oflmfld I I .M ma M .Z Ammppou P30: >uo> ow NV >N mm M NH .H. .mH .oom .uonuoo 32 N I .M 2 M Ev mv VN w. +N .2 NM H. .m .oom ..Honnoo HZ a Am Onocwv Am canvas .0 m .350 gamma Athena *OPAH R. :0” moo o ”H x M mafia was flow Omofim .u in H 18mm IgomOJ Igmm a M4M3 625mg no one: swim and monk: Sow .vofiflouuoo noun a?» no: 033 megs. OmuAH. c3550 odfindm 5 mmoumoum 5 >o>udm Mom 05. GM puma wfion omofi and manning: 09$ Saw a. .mfmiZad smut. 2 mm a .uom 63... £8: was... on: 2 «mm? 50.3 OZE v.0 dads .uCO Imopm Human pawn .H o.— M .Z Emnm .wfifiofi 3:2. mm mm M . . . : consm “mom mo ago? 2 H. mm com Monaco 3m Chapman no pugmood Amonudfiv Amoaonwv .02 £me £an in: a. mmeEoM mafia w“: Mom omofim GOSMOOA 3Q IEom Iamm Inowoq Aeosnanaoov H mama‘s 25 I .M NH M .Z Nm om cm a. N HH H. .3 .00m £0930 3m m .M NH M .2 OH H. .oH omlhm Hm NN N .00@ .H0un00 0umafinonnm¢ v .H NH M I .2 OH H. .2 .00m .vmouHHdM mm pm pm as +N 030.3932 0.3m mdon .03."... m $03 50.3 mums.» ooH 854 I .nm NH M 9v EV wN NH N .2 OH H. .: .00m ..H0nuoo HZ N . .M NH M .2 3.8 E a m 2 a .2 .oom .mm :3ng H waoHd 030322 Ho «and £qu Am0£on$ $053ch .0 m .380 fid0Q 5&0Q *0HHH. R. deadwood 0W x M mafia ma: Mom 0QOHm . I H Ifimm Igomod Eon MO< VMGM0Q .33 leom om NH +N . . . . «0 O0 .3980 access... 3238 z a s S m 3m .00.; .H 2 M .2 OH H. .3 33385 2a 5 3.1:. mm 2 IN .3m .03." $8 mac? 2.: w “an?“ 3:: in.» £58 89: 02% ms .93.. Scoot/0 down IOH0 OZ .3d0m0nm .3550 mHoomSH. .H HH M mQEMBm Odd 03me $qu 3 NN H .2 2 H. ON .00m 6:: 3mM0 N. ~33? Handymomg H0350 85: 03c: N6 :54 OMEOHOE MGHHHOM .H NH M .Z 2 H. .2 Stem 2 m: a .3m .02... £28 muons 8: a $00 Eon.“ 035 m.o £54 Am0non5 Am0£on$ .02 £an £95 1.25. a. mxunE0M mafia NE 30m 0.3% £038.04 03 IEOm Iawm 13004 Aeosasaoov : MAS: 27 >3 vommma mm 0.35 GuZm wad mun? 20m .353.»qu >u>udm Sow an... 4.330.200 noun a?» «on Erma mum: umuAH. .5550 uddadm 5 mmvumoum aw >~>ufim 20m 23 aw Uvmd wfiufi «won» wad mavens: 2:... 30m .w .330£ 333 lacs .315»: £0.95 I I .M a; M 4 .mvon« mm S. m# cm .3 N .Z a H. .MN .omm .nuauou Hm NH 26am. :3 33m £533 .. .m 3 m .z m. .H .3 .8m .8: mnmvdpoa ode #0 mm mm m: N am»? macaw HMBEE £54 2 6333 . . . common.» ufiom I do . M v“ “57% m. H. : .coflmuflwudflm mm om mm mm m um “Wm” «M E GWEN“: 0H Emflm ofinmmmom 5 on o m S N o 4 Amugona Amoaona .02 Q Q 9% o mxumEmM 63 0Q A» MD *0 H. x“ dogwood ma mafia we: Sow omoam Iadm Ifimm Inumud gonnuaoov a mafia 28 By using a barrel—type auger till samples of 5 to 7 pounds were obtained at least 18 inches below the leaching depth. The samples were taken 18 inches below the point of leaching to avoid any accumulation or gradual diminution, as has been noted in some areas, of the carbonate content. The samples were placed in cloth bags and stored for future labor— atory determinations. Soil Profiles Along with the till studies it was decided that two mem— bers of a t0posequence (ll) deve10ped on parent material of each age should be studied in the field and in the laboratory. Perhaps, at this point, it might be well to review the catena (toposequence) concept briefly. By definition, a catena is a group of soils within one zonal region, deve10ped from similar parent material, but differing in characteristics of the solum owing to differences in relief or drainage (30). The catena is usually deve10ped around (and named for) the zonal soil on any given parent material. The range in parent material prOperties within a catena is approximately the same as that allowed within a soil series (31). 29 If the parent material occurs on a gentle 510pe or level areas where poor drainage results various intrazonal soils will become members of the t0posequence. In a humid region the intrazonal soils may include Planosol, Half—Bog or Humic Gley (Wiesenboden), and BOg soils. Jenny expresses the idea of a t0posequence (ll) mathe- matically: Soils : 13(5) = f(r) c1, 0, p, t, . . where E(s) = ensemble of soil profile characteristics. Jenny has theorized that the t0posequence may include clinosequences (effects of 510pe only) and hydrosequences (effects of a water table) and perhaps others. These sequences he designates as (i) and (w), respectively. Jenny states that certain soil proper— ties. including soil depth and erosion, are primarily conditioned by i (clinofunction), whereas, some other soil prOperties such as degree of oxidation and mottling are more nearly a function of w (hydrofunction). The t0posequences (to which the soils studied belong) are quite similar to the Miami catena reviewed by Brown and Thorp (6). The well—drained member is found in a position usually I! . lvi. .ili} . .. u r 3O occupied by Miami and the poorly—drained member is in a typ— ical Brookston location. After much searching over a large area of the Port Huron moraine the author finally located a site that was deemed suitable for study. A site was also chosen on the Yale mo— raine for investigation, but because of the difficulty in getting equipment for taking laboratory samples to that site a second site suggested by Mr. D. R. Gardner, was used instead. Representative profiles were studied in pits approximately four feet deep in a poorly—drained area and in a well—drained area at each of the chosen sites. At the same time core sam— ples (5 replicates) and bag samples were taken from each dis- tinct horizon for future laboratory study. In taking the core samples a shelf about two feet wide on one side and one end of the pit was used to take samples in the AZ and B horizons. The core samples of the C horizon were taken in the bottom of the pit; around the edge so as to avoid packing effects resulting from the digging of the pit. Al core samples were taken in the surface in the immediate vicinity of the pit. The equipment described by Uhland and O'Neal (28) was used in taking the samples. 31 Field descriptions of the four soils studied in detail fol- low with the soils from the older drift being described first. Well—drained Soil. Type No. 141 (Cary Age). Locality: Midway along west side of Sec. 30, T 11 N, R 12 E, 1 mile west and l. 5 miles north of Marlette. T0pography: Rolling. Cradle knolls present. Vegetation: Second growth species mostly white birch and aspen interSpersed with Open grassy areas. Drainage: Well drained. Land Use: Pastured at present. Never cultivated. SIOpe: 11 per cent. Erosion: Very slight. 2 Soil Profile Horizon Depth Description A1 0-2" Silt Loam. Very dark brown (10 YR Z/Z) Weak medium granular. pH 6.5 Soil type numbers are those being used in the Sanilac County Soil Survey now in progress; these types have not yet been correlated. 2 This profile description is the field description taken by Mr. D. R. Gardner. The nomenclature is the type used by Mr. Gardner in describing Podzol—Gray Brown Podzolic transitional soils, a group, to which he believed this soil to belong. The symbols, GBP and P refer to Podzol and Gray—Brown Podzolic respectively (9). The color names and notations are those used in the Mun- sell Soil Color Charts, distributed by the Munsell Color Company, Inc. , Baltimore 2, Maryland. All colors are for moist samples. 32 Horizon Depth Description AZ 2—3" Fine sandy loam. Pale brown p (discon— (10 YR 6/3). Weak medium tinuous) granular. pH 5.8. sz 3-6" Silt loam. Yellowish brown 2GBP 2GBP (10 YR 5/6). Weak coarse granular. pH 5.5. 6—12" Silt loam. Light yellowish brown (10 YR 6/4). Weak fine blocky. pH 5.5. 12—24” Silty clay loam. Dark yellowish brown (10 YR 4/4). Medium blocky. pH 7.5. 24" and Sandy clay loam. Brown (10 deeper YR 5/3). Medium thick platy. pH 8.0. Calcareous. Poorly-drained Soil. Type No. 18 (Cary Age). Location: TOpography: Vegetation: Drainage: Land Use: SIOpe: Erosion: 100 yards east of midpoint west side, Sec. 30, T 11 N, R 12 E, 1 mile west and 1.5 miles north of Marlette. Rolling. Site occupies a swale. Clusters of second—growth elm and grass. Poorly drained. Drainage ditch 3.5 feet deep nearby. Ground water table 46 inches (summer). Pastured at present. Never cultivated. l per cent (11 per cent on adjoining knoll where profile No. 14 was sampled). None . 33 Soil Profile Horizon Depth _ ll A0 0 l _ H Al 1 5 A _ n 3 5 9 l G 9—12" 1 _ H G2 12 26 CI 26—29" Description Leaves and grass in all stages of decomposition. Silt loam. Very dark gray (10 YR 4/1). Weak medium gran- ular. pH 6.0 to 6.2. Pieces of undecomposed wood and char- coal throughout horizon. Loam. Gray to dark gray (5 Y 5/1 - 4/1). Very fine fragmental. pH 6.6 — 6.8. Pieces of charcoal present but not as prominent as in Al. (Transitional). Loam to silt loam. Gray to dark gray. Few reddish and yellowish mottlings present. Fine fragmental to fine blocky. pH 6.2 - 6.4. Few pieces charcoal present. Silt loam. Very dark gray to very dark brown (10 YR 3/1 — 2/2). Numerous reddish and yellowish mottlings present. Fine blocky. pH 6.0 — 6.2. Loam. Dark yellowish brown to dark brown. Some mottlings colored very dark grayish brown (2.5 Y 3/2). Fine fragmental. pH 7.8. Calcareous. l The symbol G, as used in the profile descriptions in this paper denotes gley horizons (30). ‘ .KPI ..‘ rfilpf.\-‘ “(2.“... ‘6' ll ..I..‘ 34 Horizon Depth Description CZ 29-38" Sandy loam with gravel present with loam and silt loam pockets. Color the same as C . Struc- ture difficult to determine - pos— sibly fine fragmental. pH 7.8 — 7.9. Calcareous. 38" and Clay loam. Color approximately deeper the same as C . Structure dif- ficult to determine - possibly fine fragmental. pH 8.0 — 8.1. Calcareous. There are reddish colored colloidal accumulations in root channels in the G and C horizons. Earthworm holes are very prominent above the C horizons. Crayfish holes are pres— ent throughout the profile but are more conspicuous in the G and C horizons. Pebbles are few in A and G horizons but are more numerous in the stratified material of the C horizons. Well-Drained Soil. Type No. 4 (Mankato Age). Location: TOpOgraphy: Vegetation: Drainage: 300 yards west of point 0.2 mile from south side along east line of Sec. 11, T 11 N, R 15 E, 2 miles south and 1 mile east of Carsonville. West side Port Huron moraine. Undulating to the east. A long SIOpe to the lake plain to the west. Hardwoods consisting mostly of elm, ash, bass— wood, maple, and birch. A few scattered hemlock. Well drained. 35 Land Use: Pastured woodlot. Slepe: 10 per cent. Erosion: Very slight Soil Profile Horizon Depth Description AO 0-1" Leaves and twigs in varying stages of decomposition. A1 l—3" Silt loam. Very dark brown (10 YR 2/2). Medium granular. pH 6.7 — 6.8. Horizon contains many root channels and living tree roots. B 3—6" Loam (silt loam). Yellowish p brown (10 YR 5/4). Medium platy. pH 6.2 - 6.3. AZGBP 6-9" Silt loam. Light brownish gray (2.5 Y 6/2). Platy - very fine fragmental. pH 6.1 - 6.3. B 9—12" (Transitional). Silty clay loam. lGBP . Dark brown (10 YR 4/3). Fine fragmental. pH 6.0 — 6. l. B 12-19" Clay loam. Dark brown (7.5 YR 2GBP . 4/4). Medium fragmental. pH 6.0 - 6. 1. C1 19—22" (Transitional). Silt loam. Yel— lowish brown to dark yellowish brown (10 YR 5/4 - 4/4). Fine fragmental. pH 7.2 — 7.3. Ca1— careous. 36 Horizon Depth Description C2 22" and Silt loam. Yellowish brown (10 deeper YR 5/4). Compact massive till. pH 7.9 - 8.0. Calcareous. There was a thin podzolized horizon (A2) between the A1 and Bp in places attaining a thickness of one inch. All horizons contain pebbles with a greater amount present in the C. Poorly—drained Soil. Type No. 18 (Mankato Age). Location: 200 yards west of point 0.2 mile from south side along east line of Sec. 11, T 11 N, R 15 E. 100 yards southeast of site No. 4 described above. 2 miles south and 1 mile east of Carsonville. TOpography: Undulating to gently rolling. Site occupies a depression. Vegetation: Hardwoods including elm, ash, basswood, maple, and birch. Many white birch in the more poorly- drained areas. Drainage: Poorly drained. Variable water table; at time pit was dug water table approximately 20 inches deep. Land Use: Pastured woodlot. SIOpe: l per cent (10 per cent on the adjoining knoll where profile No. 4 was sampled). Erosion: None. Possibly slight accumulation. ll 37 Soil Profile Horizon Depth _ ll A0 0 2 _ H Al 2 6 _ ll A3 6 9 G 9—20" C 20” and deeper Description A layer of leaves and twigs in varying stages of decomposition. Silty clay loam. Very dark gray to very dark brown (10 YR 3/1 -- 2/2). Medimn granular. pH 7. 1 - 7.2. Very high percentage of organic matter. Many roots and root channels. Silty clay loam. Dark grayish brown (7.5 YR 4/2). Fine frag- mental. pH 7.0. A lower per- centage of organic matter than A . Mixtures of material from A carried downward in root channels. Silty clay. Dark brown (10 YR 4/3). Much yellow and gray mottling present. Fine fragmen— tal (very angular). Slightly larger aggregates than A3. pH 6.5. Silty clay loam. Yellowish brown to grayish brown (10 YR 5/4 — 2.5 Y 5/2). Evidence of poor drainage (mottlings colored yellow— ish). Medium fragmental. Cleav— age present. pH 7.5. Calcareous. Shell fragments present. There are pockets of lighter textured material (mostly sandy loam) throughout the profile. Very few pebbles are present in A and B horizons, but they are more prominent in the C horizon. LABORATORY STUDIES Procedures Till Sample Preparation The till samples were air dried and gently crushed while still in the bag. Five hundred grams of the material was then crushed with a rolling pin and screened through a 2 mm screen to remove the pebbles. A small portion of the screened ma— terial was taken for determination of hygrosc0pic moisture. The screenings from each sample were put in beakers and soaked in water for several days. After soaking, the smaller particles were easily washed through a screen. The pebbles were then dried and weighed. Determination of Carbonate Content The calcium carbonate equivalent was calculated from the determination of total carbon in the samples. The method used was the dry combustion method of Winters and Smith (32), omitting the use of MnOZ. The total carbon method was used since the amount of organic carbon 18 inches deep in the till 39 was believed to be negligible. Duplicate determinations (trip- licate in some cases) were made on all samples. Mechanical Analyse s Hydrometer methods outlined by Bouyoucos (5) were used in making the textural determinations on the till samples. 50- dium hexametaphosphate, some prOperties of which have been discussed by Tyner (27), was used as a dispersing agent. Total sands (fractions not separated), silt, and clay (( 0.002 mm) were determined. Percentages are reported in per cent of the screened (2 mm) oven dry weight of the sample. Results were calculated as illustrated below. Per cent Sand : Corrected Hydrometer Reading at 40 seconds x 100 100 ’ Oven Dry Soil Weight Per cent Clay = Corrected Hydrometer Reading at end Of 2 hours x 100 Oven Dry Soil Weight ' Per cent Silt .—. 100 — (per cent Sand + per cent Clay). The method in present use by the Soils Division of the Bureau of Plant Industry, as reported by Kilmer and Alexander 40 (14), was used for the mechanical analyses of the profile sam— ples. The samples were not pretreated with HCl since the treatment is thought unnecessary by many and even objectionable in some soils. Total sands (2.0 — 0.05 mm), as well as frac- tions, were determined by sieving and weighing. Clay (( 0.002 mm) was obtained by the following method: (A — B)KD = per cent clay where A 2 weight in grams of pipetted fraction; B = weight in grams correction for dispersing agent; K = [1135 (vol. of cylinder)] {— [24.9 (vol. of pipette)]; D = 100 + [organic free, oven—dry weight of total sample]. Silt (0.05 - 0.002 mm) was found by subtracting the sum of the percentages of sand and clay from 100. In addition, for a check, silt was obtained by taking an aliquot immediately after vigor— ously stirring the suspension soon after setting the suSpension in the constant temperature bath. With few exceptions reason— ably good agreement in determinations using the two methods resulted. 41 Infiltration Rates The core samples were prepared by attaching filter paper covered with muslin over one end with a rubber band. Then an empty cylinder was attached to the t0p of the sample by means of a wide rubber band. The samples were then set in a large container and saturated for 24 hours; the water level in the container being adjusted to somewhat over the level of the cores. At the end of 24 hours the excess water was re— moved from the tOp of the samples and the samples set on a screen. One hundred cc of water were added to the core sam— ple surface and the time recorded for the water to disappear from the surface. One hundred cc more were added and the time again recorded. This procedure was followed until sev— eral readings were obtained (sometimes taking many hours) and until the water infiltrated at a fairly constant rate. The total number of cc that infiltrated over the period of time was re— corded and this figure converted into an inches per hour figure; this conversion was possible since the cross section of the empty cylinder was known. 42 Non—Capillary Po rosity Baver (4) defines non—capillary porosity as the sum of the volumes of the large pores, which will not hold water by capillarity. To determine the non—capillary porosity the soils were saturated overnight, weighed, set on a tension table at pF 1.6 (40 cm of water) for 24 hours and again weighed. The results were calculated as percentage of volume as illustrated below: Non—Capillary Porosity = Weight of water lost at pF 1.6 x 10 . Volume of cylinder 0 Capillary Poro sity Baver (4) defines capillary porosity as the sums of the volumes of the small pores that hold water by capillarity. In determining the percentage of capillary pores the procedure illustrated below was used: Capillary Porosity :-. Weight of Water Held at pF 1. 6 x 100. Volume of Core 43 Total Porosi_ty The total porosity was determined by adding the percent— ages of capillary and non-capillary porosity. Probably "total porosity" is a misnomer in this case, because, undoubtedly, some small pores were still filed with air, even after satura— tion for 24 hours. However, it was believed that a comparison of the porosities effective in moisture relationships could be ob— tained by this method. Volume Weight Volume weight was obtained by dividing the oven dry weight of the soil core by the volume of the cylinder (347 cc). Expe rimental Re sults Mean values of the duplicate (or triplicate) carbonate determinations, along with the depth of leaching at the sampling site and the hydrometer mechanical analysis results, are shown in Tables III and IV. Tables III and IV, in addition, give the percentages of pebbles 2 mm in size determined as percent— age of the total (500 g) air dried sample. u r. 44 TABLE III ANALYSES OF TILL SAMPLES OF MANKATO AGE Depth CaCO3 Mechanical Composition1 of , Peb— No. L h Equiv— bl 2 e,“ ‘ alent Sand Silt Clay es mg 1 24 33. 1 33.8 38.3 28.2 6.52 2 27 18. 6 38.2 32.5 32.3 3.58 3 21 32.5 48.6 26.2 28.2 19.32 43 18 25.6 69.9 22.1 8.0 20.44 5 20 32.8 31.5 38.3 30.2 5.46 6 20 32.6 33.5 36.3 30.2 10.08 7 20 37.2 31.5 38.3 30.2 9.70 8 21 30.9 31.3 38.4 30.3 4.38 9 19 38.1 33.5 35.2 31.2 4.92 10 11 28.8 29.4 40.3 30.3 10.28 11 22 27.9 41.7 38.2 20.1 6.30 12 17 33.7 31.5 40.3 28.2 4.16 20.2 31.47 Mean :1: 4 :l: 34. 396 36.57 29. 039 7.70 1.20 1.53 Percentages of oven dry weight of 4 2 mm material. Percentages of air dry total sample. 3 Not used in figuring the mean values because of the marked difference from the other samples in mechanical com- position. Standard error. 45 TABLE IV ANALYSES OF TILL SAMPLES OF CARY AGE Depth CaCO3 Mechanical Composition1 of . Peb- No. L h Equ— bl 2 e.ac - alent Sand Silt Clay es mg l 24 26.6 39.4 28.3 32.3 7.62 2 28 27.8 33.3 34.3 32.3 7.60 3 37 26.9 27.3 32.3 40.4 8.90 4 31 21.5 39.6 30.2 30.2 14.44 5 36 23.1 43.5 28.2 28.3 6.36 6 31 27.0 45.7 26.2 28.2 12.38 7 19 22.8 49.7 26. 1 24.2 16.00 8 28 22.6 43.5 30.3 26.2 6.94 9 30 22.7 43.7 36.2 20.1 6.94 10 35 19.0 38.6 34.2 30.2 9.08 11 32 16.9 29.2 40.4 30.3 4.70 12 26 17.8 27.3 38.3 34.3 4.54 29.8 . 22.90 Mean 5: .:I: 38.14 32.10 29.76 8.792 1.503 1.04 Percentages of oven dry weight of ( 2 mm material. Percentages of air dry total sample. Standard error. 46 Mean values of the mechanical analyses of the profiles studied are given in Tables V and VI. Tables VII and VIII summarize the mean values of the infiltration rate, porosity values, and volume weights of the replicates from each of the four profiles. 47 TABLE V MECHANICAL ANALYSES1 OF MANKATO AGE PROFILES Sand Fractions Total Ho ' o S'lt Cl r1z n Sands 1 ay G::::e Coarse Medium Fine 1,:er d d 1ne Sand San San Sand Sand Poorly—drained Soil A1 9.7 49.7 40.5 0. 1 0.6 1.6 3.4 4.0 A3 7.8 49.2 42.9 0.1 0.5 1.3 2.8 2.4 G 8.1 53.6 38.4 0.2 0.7 1.5 3. l 3.0 C 8.3 56.3 35.2 0. l 0.7 1.4 2.9 3.1 Well—drained Soil A1 28.6 56.3 15. 1 0.9 3.9 5.2 10.6 8.5 BP2 29.3 57 4 12.3 0.9 3.2 5.3 11.0 8.8 BZGBP 24.3 42.2 33.4 1. 1 3.0 4.2 8.7 6.7 C 22.8 57.4 19.8 2. 1 2.9 3.5 7.3 6.4 1 In all the mechanical analyses the fractions were sep— arated according to the following classification: Fraction Name Diameter (mug Very coarse sand 2.0 — 1.0 Coarse sand 1.0 — 0.5 Medium sand 0.5 -- 0.25 Fine sand 0.25 — 0.1 Very fine sand 0.1 - 0. 05 Silt 0.05 - 0.002 Clay ( 0. 002 2 The Bp horizon of this profile is compared to the AZGBP horizon of the profile on the Cary till because samples were not taken of the thin BP horizon on the Cary till. 48 TABLE VI MECHANICAL ANALYSES OF CARY AGE PROFILES Sand F ractions Total Horizon Sands Silt Clay CVery Coarse Medium Fine Very 2:)? Sand Sand Sand 2:: Poorly—drained Soil A1 30.1 40.3 29.5 0.8 2.7 4.4 11.3 10.4 A3 34.6 39.5 28.9 0.7 3.3 5.7 13.5 10.8 G2 26.1 39.8 34.1 0.5 2.08 3.5 9.7 9.8 CI 43.1 35.5 21.4 1.1 3.8 6.2 17.1 13.5 Well—drained Soil A1 48.5 43.2 8.3 1.4 5.0 7.9 19.8 14.2 AZGBP 44.9 40.9 13.8 1.8 4.3 6.6 18.3 13.3 BZGBP 33.8 31.9 34.3 1.6 3.6 5.3 13. 1 9.8 C 39.5 37.9 21.9 3.5 5.9 5.6 13.4 10.7 49 TABLE VII STUDIES OF CORE SAMPLES FROM SOILS ON THE MANKATO TILLl Infiltration Non- Horizon Rate Capillary (in./h our) Porosity Capillary Total Volume Porosity Porosity Weight Poo rly—drained Soil Al 0.498 4.44 49.65 53.25 1.10 A3 2.051 8.36 40.78 49.64 1.32 G1 2.853 7.61 38.45 46.07 1.44 C 0.354 5.07 41.90 46.97 1.46 Well—drained Soil Al 0.923 4.76 43.30 48.05 1.27 Bp 0.553 3.52 33.73 36.96 1.59 . . . 42.4 1. 0 BZGBP 0 235 7 11 38 32 6 6 C 0.032 3.12 31.89 35.01 1.88 Values shown are means of five replicates. 50 TABLE VIII STUDIES OF CORE SAMPLES FROM SOILS ON THE CARY TILLl Infiltration Non- Horizon Rate Capillary (in./hour) Porosity Capillary Total Volume Porosity Porosity Weight Poorly—drained Soil Al 0.963 10.15 45.65 55.84 1.13 A3 1.390 12.62 38.51 51.12 1.34 G1 0.411 5.07 41.38 46.46 1.55 CI 0.157 4.70 38.14 42.84 1.61 Cz 0.276 4.46 32.57 37.03 1.81 C3 0.009 3.46 23.05 26.51 2.02 Well—drained Soil A1 0.576 6.34 35.45 41.79 1.39 AZGBP 0.356 4.96 30.99 35.45 1.54 BZGBP 0.075 4.96 31.79 36.77 1.59 C 0.024 2.54 26.56 29.28 1.87 Values shown are means of five replicates except C , C , and C of the poorly—drained soil, which are averages Of duplicate determinations. . d . Includes horizons A1, AZP’ an B2P DISCUSSION OF FIELD STUDIES An inspection of Tables I and II reveals that most Of the till sites sampled on Cary age drift are leached to a greater depth than those Of Mankato age. Because of previous notations to this effect by soil surveyors in the region these results, to a certain extent, were expected. The mean value of the leaching depth of the Mankato drift is 20.2 :1: 1.20 inches (see Table 111); whereas the mean leaching depth Of the Cary drift is 29.8 :1: 1.20 inches (see Table IV). These results are statistically highly significant. There seems to be no apparent relation between the leaching depth and lepe or between the leaching depth and soil type on drift of either age. However, with 510pe percentages so nearly the same, any relation between slope and leaching depth could hardly be expected. No Obvious differences between different areas sampled on the drift of the same age could be seen. In the field studies no easily detectable differences, i.e., texture, degree of stoniness, etc., were noted between the dif— ferent age tills. Because of the variation encountered in depths 52 Of leaching the necessity of taking several samples, and pref— erably many, is Obvious. The profile field examinations of the soils, especially of the better—drained members, were complicated because they are supposedly transitional soils between the Gray—Brown Pod— zolic and Podzol great soil groups (9). Double profile deve10p— ment has begun on both the better—drained soils but is more advanced on the Cary drift. The factors, lighter texture and greater age, and possibly more have contributed to the more pronounced deveIOpment on the Cary drift. The depth of the solum is greater on each member of the Cary soil catena than on the soil in the corresponding po— sition on the Mankato material. In the well—drained members the soil on the Cary drift is leached to a depth of 24 inches; but, the soil on the Mankato drift is leached only 19 inches. The difference in leaching depth of the poorly—drained soils on the differentage material is 6 inches. Since the solum is thicker on the Older drift, some or all Of the soil horizons must be thicker than on the soil de— veloped on the younger drift. In comparing the two better- drained 50115 It can be seen that both the AZGBP and the BZGBP 53 horizons of the soil on Cary material are somewhat thicker than the corresponding horizons on the Mankato till. The other horizons are quite similar. The horizons showing the greatest variation in comparing the poorly drained—soils on the different age drifts are the gley horizons (designated by the symbol, G). The soil deve10ped on the Older material has gley horizons 17 inches thick but on the younger drifts the gley horizons are only 11 inches thick. Undoubtedly, the position of the variable water table, along with other factors, has had a profound influence on the development of the gley horizons. The acidity Of most of the horizons of the soils devel— Oped on the Cary till in both drainage situations is a little greater (lower pH) than in the corresponding horizons in the soils on the younger till. One exception to the acidity Observa— tions worthy of note is the difference in the BZGBP horizons of the well—drained member; the soil on Cary material has a pH of 7.5 in this horizon, but the similar horizon of the soil on Mankato material has a pH of only 6.0 - 6. 1. In general, the greater acidity of the soils O’n Cary drift would tend to re— flect a more advanced stage of weathering in a climate such as now exists in the area. 54 No great color differences are noticeable on soils de- ve10ped from material of the two ages. Reddish mottlings and the accumulation of reddish colloidal coatings in root channels seem to be more conspicuous in the poorly—drained soil on the older drift than in the comparable soil on the younger drift. This condition is probably caused by a greater variation in the water table each year. Differences in the mechanical composition of the profiles deve10ped on the two drifts are brought out in the laboratory discussion. The low—lying, poorly—drained soils have been leached to a greater depth than the better—drained members of the same catena. These differences, while not of sufficient magnitude to be significant here, are in qualitative agreement with the findings of Mick (20). Several possible reasons for these differences have been suggested, including those of Mick, i.e., dolomitization of the carbonates and geologic erosion. If appreciable ge010gic erosion has occurred the texture of the surface horizons of the poorly—drained soils should be much like the better—drained soils of the same catena; this was not found to be true in the laboratory investigations. Another pos- sible explanation that has been offered is the relative amounts 55 of water that each site receives over a period of time. All the water that falls on the lepes does not penetrate the soil. The depressions receive not only the water which falls there but, also, some from the surrounding slopes. In most de- pressions, some of the water is lost by runoff in streams; but this is probably not great enough to nullify the effect of the additional amount received. Jenny (12) characterizes similar soil sites as l'locally arid" and "locally humid" associates; because, they have a more arid and humid soil climate, re— spectively, than the soils would have on flat t0pography. In some regions the soils of the depressional areas are more im— permeable to water than are the adjoining upland soils but this was not found to be the case in the soils investigated, as is shown later in the laboratory discussions. DISCUSSION OF LABORATORY STUDIES The carbonate analyses in Tables III and IV reveal the fact that the samples from the Mankato age material are con— sistently higher in carbonate content than are the samples from the Cary till. The mean difference of carbonate percentage is statistically of high significance. The carbonate content of the samples from the Cary drift was found to be quite constant; the highest CaCO3 equivalent contained by any sample was 27.77 per cent and by the lowest sample 16.89 per cent. An interesting fact concerning the Cary till samples is that samples 10, 11, and 12, taken from the area where the Yale moraine crosses one of the transverse ridges, were somewhat lower than the other samples. The one sample taken from the Deanville moraine contained a carbonate per— centage that is very near the mean value for the Cary till. Samples taken in the immediate vicinity of Marlette seem to have contained somewhat higher percentages than the samples from other areas. There was considerable variation in the CaCO equiv— 3 alent content of the different samples from the Mankato till with "I h 57 the highest from any sample being 38.11 per cent and from the lowest being 18.60 per cent. There were no groups of samples from any one vicinity that either contained a constant amount of carbonates or that varied from the mean to a large degree. There is no apparent relation of the carbonate content of any one sample of the till from either age to the depth of leaching where that sample was taken. For an example the Mankato sample (No. 10) with lowest carbonate content (18.60 per cent), also, had the greatest leaching depth (27 inches); but, on the other hand, the site with the least amount of leaching (11 inches) had a carbonate content of 28.84 per cent—-—some distance below the mean. The mechanical analyses of the till samples given in Tables III and IV did not reveal any great differences in the composition of the two ages of drifts, though the sand content is somewhat greater for the Cary than for the Mankato till. The difference in sand content is reflected in the amount of silt, since the silt percentage is obtained by difference, and since the clay content of both tills is very nearly the same. No consistent relation can be detected between the mechanical com— position and the leaching depth; in fact, some of the sandier soils have the thinnest solums. 58 The percentage of pebbles in the Cary samples was found to be slightly greater than in the Mankato samples. By casually observing the lithology of the pebbles it could be seen that the prOportion of sandstone in the Cary samples was some— what higher (excepting samples 2 and 5) than in the Mankato (excepting samples 3 and 4) samples. Furthermore, the sand— stone pebbles were especially conspicuous in samples 9, 10,, 11, and 12 of the Cary drift. The Cary samples perhaps con- tained a few more igneous and metamorphic pebbles than the Mankato samples. A limonitic material was present in prac— tically every Cary sample but was only noticed in two (Nos. 4 and 7) Mankato samples. A thinly laminated blue shale was present in many Mankato samples but was not found in any of the Cary samples. Buff-colored indurated pieces of till were quite conspicuous in many Cary samples, but where the mate— rial was present in the Mankato samples, it was more of a bluish color. Limestone was by far the dominant material in the Mankato pebble samples. A possible explanation of the higher content of sandstone, which has probably caused the soils to be coarser textured, is the Marshall sandstone underlying the drift of the Cary moraines and a considerable portion of the till 59 plain to the northeast; whereas the Coldwater shale is imme— diately beneath the glacial debris of the Port Huron moraine and the area to the east between the moraine and Lake Huron (10). Perhaps, at this point, it would be well to consider the relative permeabilities of the C horizons from the profiles on the different age drift which are given in Tables VII and VIII. Since the carbonates have not been leached from the C horizon and since no visible eluviation or illuviation has occurred, the material should be rather similar to the till samples. In both the better—drained and poorly—drained soils on the Cary drift the C horizon was more impermeable than the related soils on the Mankato drift. The C horizon from the better—drained soil on the Cary till had an infiltration rate of 0.024;1 whereas the soil derived from the Mankato till had an infiltration rate of 0.032. Measuring the infiltration rate of the C horizon from the poorly—drained soil on Cary till was complicated by con- siderable stratification of the material in that horizon. The infiltration rate of the Cl, C , and C3 (in which duplicate 2 core samples only were taken) was 0.157, 0.276, and 0.009, All infiltration rates presented here are expressed in terms of inches per hour. 60 respectively. The C horizon of the poorly—drained soil on the Mankato till showed an infiltration rate of 0.354. The author believes a comparison of the infiltration rate of the well—drained soils to be more reliable than a com- parison of the poorly—drained soils because the core samples of the poorly—drained Mankato derived soil were taken when the soil was much too wet. The mechanical analyses (see Tables V and VI) show the C horizon of the well-drained soil on the Cary till to be of a coarser texture than the C horizon of the related soil on Mankato till. Therefore a more permeable horizon might be expected, but such is not the case, as has already been shown. Texture is an important clue to the per- meability of a horizon but when considered alone it is not fully reliable. Sandy horizons with relatively small percentage of silt, clay, or colloidal material often have much slower perco— lation rates than the scarcity of such materials might suggest. Apparently the fine material fills the spaces between the sand grains and retards water movement (28). Other characteristics, especially structure, are impor— tant in determining infiltration capacities of a soil. The per- centage of overlap of the aggregates is probably the most 61 important structural characteristic. The field descriptions of the profiles under investigation give the structure of the C horizon of the Cary age derived soil as thick platy and that of the similar horizons of the soil on the Mankato till as fine fragmental. Platy structures often show a high degree of overlap, but since the overlap of the structural aggregates was not determined in either soil it is not possible to evaluate the importance of structure in the permeability of these horizons. The clay percentage of the two horizons in question are not far different (with the soil on the Cary till being slightly higher). It is probably the more thorough dispersal of the clay rather than the total clay content among the sands causing the lower degree of permeability of the soil horizon on the Cary till. In attempting to apply the data obtained on leaching depths and carbonate percentages to an estimation of the rela- tive ages of the two drifts the following procedure was used: [Mean-leaching depth x Mean CaCO equivalent (Cary)] 3 -:- [Mean—leaching depth x Mean CaCO3 equivalent (Mankato)] = [Relative difference in age]. 62 The mean leaching depth multiplied by the mean carbon— ate percentage should give an approximation of the amount of carbonates lost through leaching on the drift of each age. Sub- stituting numerical values in the above formula, we have: [30 x 23 (Cary)] -:- [31 x 20 (Mankato)] = 1.1. Assuming an age of 11,000 years (19) for the Mankato drift we obtain a figure of 12,100 (1.1 x 11,000) years for the age of the Cary drift. Since this is considerably below most estimates a correction is believed necessary. If a correction is made for the relative permeabilities of the C horizons of the well—drained soils another step in the procedure must be added as illustrated below: [CaCO equivalent x leaching depth]/Infiltration rate (Cary) 3 -:- [CaCO equivalent x leaching depth]/Infiltration rate (Mankato) 3 = Relative difference in age. Substituting data obtained in the investigation in the above equation, we have: [30 x 23]/o.024 (Cary) + [31 x zo]/o.osz (Mankato) = 1.5. Using the same age of 11,000 years for the Mankato drift, we arrive at 16,500 years for the age of the Cary drift. It is be— lieved that the correction figure for permeability is a maximum 63 and that if many sites were sampled and a mean value of the permeabilities obtained the difference would be smaller. Another reason for believing the factor to be rather high is the fact that only when the solum was saturated and the rainfall rate exceeded the permeability of the C horizon would the permeability of the C horizon (the least permeable horizon) be a limiting factor in determining the amount of leaching. Some other assumptions have had to be made in arriving at the respective ages of the two tills. One assumption made is that the rate of leaching does not decrease with the depth of leaching. Supposedly, the rate of leaching decreases with depth because some of the carbonic and other acids are neutralized in their alteration of the silicate minerals in the upper horizons before reaching the calcareous material, so that as time passes, there is a lower and lower concentration of acid available for leaching at lower depths; The supposed precipitation of lime carbonate in the calcareous zone, derived by leaching from above, would add that much more lime carbonate to be leached again later (17). But Smith (23) has shown that there is no zone of carbonate accumulation, at least, for several feet below the leaching depth, in Illinois loess. Thus, carbonate 64 accumulation does not appear to be an important factor in hu- mid climates in determining the leaching rate, although it is an important factor in semi—arid and arid regions. Kay (13) assumed a constant rate of leaching to a depth of more than 30 feet on gravels in estimating ages since various glacial stages and intervals. Kay rec0gnized that the rate of leaching prob— ably decreased with depth but made a statement to the effect that it had never been proved. As to whether the rate of leaching has decreased with depth in the relatively young Cary material poses a question. It seems rather unlikely, however, that the leaching rate has decreased appreciably in a material that has been leached such a relatively short distance downward. The mechanical analyses of the profiles given in Tables V and VI show that the soils on the Cary drift are consider- ably sandier, with consequent reductions in the percentages of clay and silt, than the soils on Mankato material. Probably this is a reflection of the more sandy till from which the soils were formed and which is revealed in the composition of the C horizons and in the mechanical analyses of the till samples. In comparing the two better—drained soils the soil on the Cary till seems to have been subjected to more thorough 65 weathering as revealed in the content of clay in some of the horizons. The clay has been removed to a greater degree from the A1 horizon and accumulated in the B horizon relative to the C horizon 'in the soil deve10ped from the Cary drift than from the Mankato drift. In both the poorly—drained soils there has been an in— crease of clay in all horizons of the solum over the percentage contained in the C horizons; however, these differences could partly be inherited from. the parent material. This is substan— tiated by the fact that C horizon of the soil on Cary material shows stratification; moreover, there is reason to believe, be— cause of the high percentage of silt and clay in all horizons, that the soil on Mankato material has deve10ped in part from lacustrine sediment. It is a safe assumption that the increase in sands in the A and A horizons over the C horizons in the better—drained members of the catenas on material of both ages is not an ac— tual increase. The increase is relative because of the removal of the clay through weathering processes. The size distribution in the sand fractions does not pre— sent any striking differences between soils developed on different 66 age material or between members of a catena. The fine sand fraction is the largest fraction in nearly every horizon. Two exceptions were the A and C horizons of the poorly—drained 1 soil on Mankato till. The C horizon was the most impermeable horizon in each of the four soils studied. The other horizons on the better— drained soil from the Cary material are also less permeable than the comparable horizons from the similar soil on Mankato till. In each case the B horizon was the next most im- ZGBP permeable horizon, and the A horizons were the most permeable. 1 The more thorough diSpersal of colloids and clay caused by a more advanced degree of weathering is believed responsible for the lower permeabilities of the core samples from the soils on Cary drift. The relative permeabilities of the C horizons from the soil on the till of the two ages have been discussed previ— ously. All the horizons, except the Al, from the poorly—drained soil on the Cary drift exhibited greater permeability than the horizons of comparable depth from the poorly—drained member on the Mankato till. As has been pointed out previously the core samples of the A , G, and C horizons from the Mankato 3 67 area were taken when the soil was practically saturated and the A1 samples taken at a different time of the year (spring) from the others (fall), because of this the data are probably not too reliable. While discussing the poorly—drained soils the differences in the infiltration rate of the C1, C2’ and C3 horizon from the Mankato till should be pointed out. As has been mentioned earlier some stratification was evident and textural differences were discernible in the field. Texture seems to be the principal differentiating factor since structural differences were not great between the Cl and C2. Structure could not be determined in the C3 because of the water—legged condition at the time of the field examination; in the wet state it was rather tenacious, although containing considerable quan— tities of sand. The relative infiltration rates are reflected in the por— osity values to a considerable extent. The well—drained soil deve10ped on the Mankato till is somewhat more porous in all horizons than the similar soil deve10ped on Cary till. The Al, B and the C exhibit greater difference than the A (1 GBP ZGBP an BP horizons. In the poorly—drained soils the Al and A3 hori- zons on the Mankato drift have a greater total porosity than the 68 Al and A3 horizons from the soil on the Cary till. The porosity values for the G horizons are essentially the same but the values for the C horizon from the Mankato till is greater than that from any part of the C horizons on Cary drift. The infiltra- tion rate supposedly depends to a large extent on the amount of non—capillary porosity. An examination of Tables VII and VIII reveals that there is some relationship between the two values but that there is no direct correlation. The depth of leaching is one criterion used in distinguish— ing soil types. The fact that there are significant differences in leaching depth on different age tills in Sanilac County should be of use in correlating soil types in the county. It is prob- able that many of the soil types on the Mankato drift should not be correlated with those on the Cary drift in the county because of the shallow solum they exhibit. Other characteristics of the soils, including double profile develOpment, should also be a help in distinguishing and correlating the types. The thinner solum and more calcareous nature of the underlying till indicate that certain mineral deficiencies, espe— cially manganese, are more likely to occur on the soils on the Mankato drift than on those deve10ped on the Cary drift. Plants 69 growing on eroded areas would be especially susceptible to these mineral deficiencies. The farmers in the eastern part of the county growing crOps on soils deve10ped from Mankato till would need to apply lime only under rare circumstances. Liming many of the soils could be exceedingly harmful. It is conceivable that under in— tensive cr0pping some soils on the Cary drift would be benefited by lime applications. As time prOgresses other soil character- istics attributable to the differences in depth of leaching will probably become evident. SUMMARY In this study involving till analyses and soil profile studies on glacial drift of Mankato and Cary age several dis— tinct differences between the two tills, as they relate to soil formation, were noted. The mean depth of leaching on the drift of Cary age is significantly greater than on the drift of Mankato age. However, the carbonate content of the Cary till is significantly less than the carbonate content of the Mankato age till. This has prob— ably affected the leaching depth of the two tills appreciably. The till of Cary age is somewhat more sandy than the Mankato age till. Although containing a higher percentage of sand, the C horizons of both soils on the Cary drift are less permeable than the C horizons of the comparable soils on the Mankato drift. Considering the information obtained concerning the depth of leaching, carbonate content, and the permeability of the C horizons, the author believes the Cary till in the area studied to be between 1,100 and 5,500 years older than the Mankato till. 71 The well—drained soil deve10ped on till of Cary age shows a more advanced stage of weathering as evidenced by the smaller clay content of the A horizon and the greater clay content of the B horizon relative to the C horizon. Additional evidence that the soils on Cary drift have been weathered more severely is indicated by the deeper solum of both soil profiles investigated on Cary drift. The more pronounced double profile deve10pment on Cary till is a further demonstration of a more advanced weathering stage. All the horizons in the soils on Cary till are less per— meable than the horizons of a comparable depth in the soils developed in a similar t0p0graphic position on Mankato till. The poorly—drained members of the t0posequences have higher percentages of clay in the surface horizons than the well—drained members of the same t0posequence. This is prob- ably due either to the more rapid weathering of the poorly— drained soils or to the greater clay decomposition in the well—- drained profiles. The amount of clay contained in the solum relative to the C horizon in the poorly—drained soils is greater in the soil deve10ped on Cary till than in the soil on Mankato till. 72 This has probably been caused by differences in mechanical composition of the parent materials and/or more severe weath— ering of the soil on Cary drift. Some of the similar soils on the two tills probably will not be correlated as the same soil series because of the dif— ferent depths of leaching they exhibit. Soil management prac— tices on some of the soils deve10ped on the different age ma— terial should be determined in part by the different depths of leaching of the soils. 10. BIBLIOGRAPHY Andrew, L. E., and H. E. Rhoades, 1947. Soil forma— tion studies on fresh Nebraskan till. Soil Sci. Soc. Amer. Proc. 12:409-412. Antevs, E. , 1938. Climatic variations during the last glaciation in North America. Amer. Met. Soc. Bull. 19:172—176. Bartlett, H. H. , 1951. Radiocarbon datability of peat, marl, caliche, and archaeological materials. Science 114:55-57. Baver, L. D. , 1948. Soil physics. John Wiley and Sons, Inc. , New York. Bouyoucos, G. J. , 1951. A recalibration of the hydrometer method for making mechanical analyses of soils (in press). Brown, I. C. , and J. Thorp, 1942. MorphOIOgy and com— position of some soils of the Miami family and the Miami catena. USDA Tech. Bull. 834. Chandler, R. F. , 1942. The time required for podzol profile formation as evidenced by the Mendenhall glacial deposits near Juneau, Alaska. Soil Sci. Soc. Amer. Proc. 7:454-460. Flint, R. F. , 1947. Glacial ge010gy and the Pleistocene epoch. John Wiley and Sons, Inc. , New York. Gardner, D. R. , 1951 Some observations of the "pod— zolized" gray—brown podzolic soils of Michigan. Master's Thesis, Michigan State College. Gordon, C. H., 1900. Ge010gical report on Sanilac County. Geological Survey of Michigan. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 74 Jenny, H., 1946. Arrangement of soil series and soil types according to functions of soil forming fac- tors. Soil Sci. 61:375-391 , 1941. Factors of soil formation. McGraw—Hill Book Co., New York. Kay, G. F , 1931 Classification and duration of the Pleistocene period. Bull. Geol. Soc. Amer. 42: 425—466. Kilmer, V. J. , and L. T. Alexander, 1949. Methods of making mechanical analyses of soils. Soil Sci 68:15-24. Kruger, F. C. , 1937. A sedimentary and petrographic study of certain glacial drifts of Minnesota. Amer. Jour. Sci. 34:345-367. Krumbein, W. C. , 1933. Textural and lithological vari- ations in glacial till. Jour. Geol. 41:382—408. Leighton, M. M., and P. McClintock, 1930. Weathered zones of the drift sheets of Illinois. Jour. Geol. 38:28-54. Leverett, F., and F. B. Taylor, 1927. Pleistocene gla— ciation. USGS MonOgraph 53. Libby, W. F., and J. R. Arnold, 1950. Radiocarbon dates. Sci. 113:111-121. Mick, A. H. , 1949. The pedology of several profiles de— ve10ped from the calcareous drift of eastern Mich- igan. Mich. Agr. Exp. Sta. Tech. Bull. 212. Norton, E. A. , 1927. Profiles of soils in southern Illinois. Proc. First Int. Cong. Soil Sci. 4:283—290. Robinson, G. W., 1949. Soils. John Wiley and Sons, Inc. , New York. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 75 Smith, G. D., 1942. Illinois loess. Ill. Agr. Exp. Sta. Bull. 490. Stauffer, R. S., 1937. Local variability in Wisconsin till and its influence on soil character. Amer. Jour. Sci. 34:235-243. Thornbury, W. D., 1940. Weathered zones and glacial chronology in southern Indiana. Jour. Geol 48: 449—475. Thorp, J., W. M. Johnson, and E. C. Reed, 1951. Some post—Pliocene buried soils of central United States. Jour. Soil Sci. 2:1—19. Tyner, E. H., 1939. The use of sodium metaphosphate for dispersion of soils for mechanical analysis. Soil Sci. Soc. Amer. Proc. 4:106—113. Uhland, R. E., and A. M. O'Neal, 1951. Soil permea— bility determinations for use in soil and water conservation. SCS—TP-lOl. U. S. Department of Agriculture Yearbook, 1941. Climate and man. Washington, D. C. U. S. Department of Agriculture Yearbook, 1938. Soils and men. Washington, D. C. Winters, E., 1949. Interpretative soil classifications: genetic groupings. Soil Sci. 67:131—141. , and R. S. Smith, 1929. Determination of total carbon in soils. Jour. Ind. and Eng. Chem. Anal. Ed., 1:202—203. , and H. Wascher, 1935. Local variability in the physical composition of Wisconsin drift. Jour. Amer. Soc. Agron. 27:617L622. APPENDIX Figure l . Sampling till with barrel-type auger on site No. on Cary drift. 77 10 f 73 Taking core samples of C horizon of well—drained Figure 2. soil on Mankato drift, October, 1950. t x mm Figure 3. Tepographic position and vegetation of well—drained profile site on Mankato drift, October, 1950. 79 ., \\\\—~ WA Figure 4. Poorly—drained profile on Cary drift, July, 1951. Figure 5. Poorly—drained profile on Mankato drift, April, 1951. (Note position of water table.) 31293 03037 9527