KGS Home Geology Home Start of Pleistocene Stratigraphy report

Previous--Report Start || Next--Selected Pleistocene Exposures

Regional Pleistocene Stratigraphy

by C. K. Bayne1, S. N. Davis2, W. B. Howe3, and H. G. O'Connor1

1 State Geological Survey of Kansas.
2 University of Missouri at Columbia.
3 Missouri Geological Survey and Water Resources.


Pleistocene deposits and Pleistocene history are of considerable importance to Kansas and Missouri. The widespread and generally favorable influence of Pleistocene deposits upon the character of agricultural soils of northwestern Missouri and northeastern Kansas combined with the plentiful supply of potable groundwater generally associated with the deposits that occupy the preglacial valleys are of enormous importance to the general economy of the region. Northern Missouri and northeastern Kansas were glaciated during early Pleistocene (Nebraskan and Kansan) time, but escaped glaciation during middle and late Pleistocene time. Pleistocene till and associated deposits covered most of the area north of the Missouri and Kansas rivers and extended south at several localities. Extensive alluvial and eolian deposits of middle and late Pleistocene age were formed in the northern part of the region. Pleistocene deposits in Kansas and Missouri include glacial, fluvio-glacial, lacustrine, and eolian deposits that range in age from early Nebraskan to Recent (Table 1). Extensively developed and widely recognized paleosols are, in many localities, the principal basis for division of the Pleistocene succession. Some carbon-14 dates are also available for late Pleistocene deposits. Compared with surrounding states, Missouri is one in which relatively little modern detailed Pleistocene stratigraphy has been done. In contrast, Pleistocene deposits in Kansas have been studied for more than 30 years.

Table 1--Stratigraphic units of the Pleistocene, northeastern Kansas and northwestern Missouri.

  Kansas Missouri
Stage Substage Rock Unit Soil Formation Substage Rock Unit Soil Formation
Recent*   Fluvial deposits XX   Fluvial deposits XX
Wisconsinan   Bignell Formation   Valderan Bignell Loess  
Bradyan   X Twocreekan   X
  Peoria Formation   Woodfordian Peoria Loess  
  Gilman Canyon Fm. X Farmdalian Roxana? Silt XX
      Altonian (?)  
Sangamonian     XXXXX     XXXXX
Illinoisan   Loveland
  Late Illinoian Loveland Loess  
Medial Illinoian Fluvial deposits XX
Early Illinoian  
Yarmouthian     XXX   Ferrelview Fm. XX
Kansan Upper Kansan Loess, fluvial and
lacustrine deposits
  Late Kansan  
Medial Kansan Cedar Bluffs Till XX Medial Kansan Undifferentiated
Kansan till
Fluvial deposits
Lower Kansan Nickerson Till Early Kansan Atchison sand
Atchison Formation
Aftonian     XXXX     XX
Nebraskan Upper Nebraskan Iowa Point Till X   Nebraskan till
Fluvial deposits
Lower Nebraskan Fluvial and
lacustrine deposits
* Assigned to Holocene Series in Missouri
no scale intended

The preglacial topography of the region is important in the interpretation of Pleistocene history. Because of the considerable information available through water well and test drilling work, it has been possible to delineate the principal preglacial drainageways (see separate report by Dreeszen in this publication) . Most of the buried valley systems were inherited from topography that presumably developed during the late Tertiary. The present drainage system in the region is a composite of exhumed ancient valleys and superposed valleys associated with streams that developed upon Pleistocene till plains. Floors of the lower reaches of the principal buried valleys lie as much as 100 feet below the beds of present-day streams.

Much of the interest in the bedrock of preglacial topography in the region has been associated with the search for increased ground-water supplies. Test drilling has shown that in some localities the buried valleys are completely filled or are blocked by glacial till, while other valleys contain sand and gravel deposits capable of producing large quantities of potable water. In some northernmost counties in both Kansas and Missouri, interpretation of cross sections prepared from well records and test holes indicates that the bedrock surface was generally covered with outwash sand and gravel deposits that were later buried by glacial till and/ or scoured by glacial ice. Knowledge of the distribution of the various types of glacial deposits and their relationship to the bedrock topography is limited. The wide variety of types of deposits that may be encountered is recognized; however, sufficient information about their distribution is not available for satisfactory mapping and description. The delineation of the buried and exhumed topography on the bedrock surface is fairly well defined, but knowledge of the bedrock topography is only a part of the answer to the problems that exist with water supply. Moreover, many details of the stratigraphic succession have yet to be determined. The general succession of deposits and their interrelationships are fairly well defined and can be demonstrated in the area.

The Pleistocene Succession

Essential features of the stratigraphic succession of the Pleistocene Series in northwestern Missouri and northeastern Kansas are discussed in the following paragraphs under subheadings identified by stage ( age) terms. Order of the units in this discussion is from oldest to youngest.

Nebraskan Stage

Oldest tills and alluvial materials encountered in extreme northeastern Kansas and in Missouri north of St. Joseph have been considered by most workers to be Nebraskan. The complexity of early Pleistocene stratigraphy is becoming more evident as studies progress in various parts of the world. The subdivision of the Pleistocene into a simple series of four major glacial advances has been abandoned by a large number of modern workers. Early Pleistocene deposits along the Missouri River reflect many of the complexities reported from other regions. The traditional classification of the Pleistocene of the Midwest has, nevertheless, been retained in the discussions that follow.

Thick Nebraskan deposits are undoubtedly present in many of the buried channels in northeastern Kansas and northwestern Missouri. Reliable criteria for the differentiation of these deposits from overlying Kansan till and outwash, however, are generally lacking.

The principal basis for recognition of Nebraskan till is its association in sequence with one or more overlying tills (Kansan) and an intervening soil profile (Afton). Till of Nebraskan age has been recognized in outcrops in the Moberly and Macon areas in north-central Missouri; in southern Iowa, near Hopkins, Missouri; and in outcrops in Kansas near Iowa Point, Doniphan County, along the Missouri River bluffs. Nebraskan till also appears to be present at many widely scattered points in the north-central and northwestern Missouri areas, where carefully logged test drilling records are available. No evidence of Nebraskan till has been found south of the buried valley of the ancient St. Joseph River between st. Joseph and Gallatin, Missouri. This valley may have been marginal to Nebraskan ice.

Basal till containing few or no northern erratics is moderately common in northwestern Missouri and northeastern Kansas. Davis (1955) described similar material from Platte County, Missouri, where he identified it as a "local" facies of the Kansan till of that area. Glaciers advancing over sedimentary rocks incorporated a preponderance of easily eroded local materials into the basal part of the ice. The lowest till emplaced would, therefore, be dominated by local rock types. Later, overriding ice would not be in contact with as much bedrock and would deposit till containing a greater percentage of northern erratics. Thus, an upward increase of northern erratics within an outcrop of till would not necessarily indicate two episodes of glaciation separated by a significant span of time. Dort (1966) discusses some of the problems in the assignment of ages to tills of various lithologies. Bayne and O'Connor (1967) and Bayne (1968) refer such tills exposed near Doniphan, Kansas, to the Kansan Stage. Such "basal" till facies presumably might be associated with till of either Nebraskan or Kansan age.

A number of exposures of Nebraskan silt, sand, and gravel are found in quarries along the Missouri River in Doniphan County, Kansas. Some of these exposures are described in a later section of this report. Comparable outcrops are not seen on the Missouri side of the river, although silts in St. Joseph and north of St. Joseph along Interstate 29 (Howe, 1968) may be Nebraskan. Further studies in northwestern Missouri may show that periglacial silt of Nebraskan age is more widespread than present knowledge would suggest.

Aftonian Stage

New exposures along Interstate 29 north of St. Joseph include one of the few exposures of a well-defined paleosol known in northwestern Missouri that can be even provisionally referred to the Aftonian. At the Interstate 29 locality, the paleosol in question is not associated with glacial till, but is underlain and overlain by silt tentatively regarded as periglacial deposits of Early Nebraskan and Early Kansan age, respectively.

Afton Soil developed over Nebraskan till was well exposed a few years ago in the overburden of a limestone quarry in southern Iowa just north of Hopkins, Missouri, and has also been identified in exposures at Iowa Point in northeastern Kansas (Frye and Leonard, 1949) and in exposures near Doniphan and Wathena in Doniphan County, Kansas. Gray leached material above glacial till and below calcareous till referred to the Kansan is recorded in logs of test holes scattered over northern and northwestern Missouri. This probably represents the Aftonian paleosol.

Kansan Stage

Most till in northeastern Kansas and northern Missouri is classed as Kansan. At present there is no basis for division of the Kansan tills in Missouri as has been accomplished in Nebraska by Reed and Dreeszen ( 1965) who identify two distinct till deposits. Two Kansan tills are also recognized in northeastern Kansas, as reported by Bayne and O'Connor (1967) and Bayne (1968). According to current interpretation, Kansan glaciers overrode the area affected by Nebraskan glaciers in Kansas and Missouri and extended far to the south where they locally blocked the ancient Kansas River forcing it to develop alternate routes generally marginal to the glacial front.

Kansan glaciation tended to obliterate most of the pre-existing topography in northern Missouri and northeastern Kansas. The early stages of the post-Kansan drainage systems, which presumably were the precursors of those currently in existence, are not clear. Part of the buried sand and gravel fill in some of the early Pleistocene valleys in northeastern Kansas and northwestern Missouri, however, is undoubtedly pre-Kansan material.

Outwash-derived alluvium developed during recessive stages of Kansan glaciation undoubtedly is associated with valley fill in the ancient Kansas River valley and the Missouri River valley below Kansas City which, for the most part, is the locus of the modern Missouri River. Lacustrine and accretion-gley deposits that are provisionally identified as very late Kansan and Yarmouthian in age are widespread in upland regions in Platte, Clay, Clinton, and Caldwell counties in Missouri, and in Atchison, Brown, Doniphan, and Jefferson counties in northeastern Kansas. These deposits may have incorporated considerable quantities of late Kansan loess, as well as volcanic ash, and are a composite of lacustrine and accretion-gley types. They have been named the Ferrelview Formation in Missouri (Howe and Heim, 1968) and have, in Kansas, been referred to as "Nortonville clay."

Yarmouthian Stage

The Yarmouthian Stage is represented in Missouri by a soil profile, the Yarmouth Soil, and by parts of the composite lacustrine and accretion-gley deposits identified as Ferrelview which were deposited at the same time that the Yarmouth Soil was being developed. Well-defined and clearly differentiated Yarmouth Soil is known in only a few localities in northeastern Kansas and northwestern Missouri. A possible reason for this is that the next younger deposits, the Loveland Loess, (although of widespread distribution) is generally thin and in many places incorporated completely in the profile of the next younger soil (Sangamon). Development of the Sangamon Soil was so intense that, in most areas where the intervening Loveland Loess was deposited, soil-forming influences extended all the way through the Loveland and into the underlying Yarmouth. Where the Loveland is absent through nondeposition, the Sangamon is superposed upon the Yarmouth and may represent a continuation of soil-forming processes (Wright and Ruhe, 1965, p. 31; Howe and Heim, 1968, p. 18). In the absence of the Loveland Loess, the Yarmouth Soil is not commonly distinguishable in northeastern Kansas and Missouri, although its presence is inferred at many points. New exposures seen in July 1968 at Kansas City International Airport (under construction) and in Jefferson County, Kansas, indicate clearly that weathering of the till and some accumulation of resistant erratics took place prior to deposition of the Ferrelview Formation.

Illinoisan Stage

Illinoisan glaciers are not known to have reached Kansas or northwestern Missouri, although Illinoisan till is present at St. Louis. Alluvial deposits preserved as terraces have been recognized at a number of points, and as indicated in a preceding section, extensive loess deposits (the Loveland Loess) are recognized. Colluvium of probable Illinoisan age is also common on many hill slopes. Three lines of evidence suggest that the valley of the modern Missouri River between Kansas City and St. Joseph, Missouri, may have been cut principally during Illinoisan time. First, the valley is relatively narrow and youthful looking. Second, till-bedrock contacts along the bluffs facing the river are generally more than 50 feet above the modern floodplain, suggesting that dissection was post-Kansan. Third, tributary valleys along this part of the Missouri River have well-developed late Illinoisan terraces but no indication of earlier terraces. This implies that the stream valleys were graded to the Missouri River by late Illinoisan time but were not carved until after the last advance of Kansan ice.

Sangamonian Stage

The Sangamonian Stage is represented by one of the most intensely developed and most widespread of the Pleistocene paleosols. The Sangamon is recognized over large areas in northern Kansas and Missouri. As already indicated, the Sangamon profile may be intimately associated with older profiles so that the identification of the older one may be in doubt. The paleotopography during Sangamonian time was such that the Sangamon Soil developed on an erosional surface that in places transected an older one associated with the Yarmouth. The general relationship in northeastern Kansas and northwestern Missouri is the development of the Sangamon profile over a wide variety of indurated and nonindurated rock types with the profile itself being overlain by Wisconsinan loess, principally the Peoria Loess.

At several localities in northwestern Missouri and northeastern Kansas, slope-site exposures of till with the Sangamon profile developed on it include typically a prominent "lag" or "stone-line" accumulation of resistant erratics which occurs within the Sangamon paleosol or possibly within a composite SangamonYarmouth paleosol.

Wisconsinan Stage

Wisconsinan deposits in northwestern Missouri and northeastern Kansas include early Wisconsinan loess, identified in Kansas as Gilman Canyon and in Missouri as Roxana?, overlain by younger loesses identified as Peoria and Bignell. Current thinking about the identification and nomenclature of Wisconsinan loesses in the northwestern Missouri and northeastern Kansas region is not in accord with respect to the materials older than those uniformly regarded as Peoria (Peorian). Area geologists have recognized a post-Sangamon, pre-Peoria succession for many years, generally with a tentative assignment to the "Farmdale." This material was given the name Gilman Canyon in Nebraska by Reed and Dreeszen (1965, p. 42), and the name was subsequently adopted in Kansas (Bayne and O'Connor, in Zeller, 1968, p. 62). The difference in viewpoint between Missouri and Kansas geologists relates to both the age of and the appropriate names for this material. In Missouri the term Roxana is being utilized for similar deposits in the St. Louis area and appears to be applicable at Kansas City. In southwestern Iowa, Frye, Willman, and Glass (1968), and Glass, Frye, and Willman (1968) identify loess at this position as Roxana, and note that a weakly developed Farmdale soil caps the loess. This disposition of the matter appears to best fit the relationships in northwestern Missouri and is utilized in a provisional sense, subject to more detailed studies. The material previously identified as Farmdale? is tentatively correlated with the Roxana (of Altonian age), with the overlying paleosol (Farmdale) being the only record of the Farmdalian Substage. In Kansas the same material is identified as Gilman Canyon and thought to be of Farmdalian rather than Altonian age. Radiocarbon dating will ultimately be the deciding factor in resolving the issue, but presently available evidence is inconclusive. In northwestern Missouri only one carbon-14 date (Martin and Williams, 1966, p. 26) is relevant, and it (indicating an age of 25,000 ± 2,200 years BP) relates to material collected at the top of silt referred to the "Farmdale," rather than within it. Wood determined to be 24,500 ± 800 years BP old from a southwestern Iowa exposure (Ruhe, Rubin, and Scholtes, 1957) was described as having been "…extracted from the A horizon of a buried soil in a loess sheet 3 to 4 feet thick that overlies Loveland loess…" (p. 676). The authors identified the associated loess sheet as Farmdale (p. 674, 677). It is likely that the deposit involved is comparable with that noted by Frye, Willman, and Glass (1968) and the Gilman Canyon (Kansas) or Roxana? (Missouri) of this report. The fact that the dated material came from the uppermost part of the deposit at the Iowa locality interjects a factor of doubt about the age of the deposit as a whole. Reported Gilman Canyon dates (Dreeszen, 1970) in Nebraska (23,000-34,900 BP) may also be inconclusive. At present Kansas geologists prefer to assign the early Wisconsinan loess to the Farmdalian Substage rather than to the Altonian as is done on a provisional basis in Missouri. More attention to this problem is needed in both states.

The Wisconsinan loesses are present in maximum aggregate thickness of at least 100 feet along the Missouri River valley bluffs and are widely distributed in northeastern Kansas and northwestern Missouri. Thick exposures of light-colored loess almost invariably show darker color bands which are subparallel to the topography. Some of the color bands may be caused by post-depositional accumulations of iron and manganese staining from subsurface water. Many of the color bands, however, are made of material which is leached of carbonates and contains more clay than the normal loess. These bands are interpreted as buried soils even though the soil profile is not well developed. Better exposures show at least five post-Sangamon "soils." Work is still needed in order to correlate Wisconsinan loesses in this region with better known sections in Nebraska and Illinois. Despite uncertainties, the names Gilman Canyon (Kansas), Roxana? (Missouri), Peoria, and Bignell are used for the loesses. The bulk of the loess deposits are correlated with the Peorian Loess in Illinois. A somewhat poorly defined overlying loess is correlated with the Bignell Loess of Nebraska. The Bignell is profusely fossiliferous at most localities along the bluffs of the Missouri River (see separate report by Pauken in this publication). The Peoria is also fossiliferous at many places where complete leaching has not taken place.

Wisconsinan alluvial deposits appear to compose most of the sand and gravel fill in the valley of modern streams such as the Kansas, Missouri, and Grand rivers. In areas where the modern stream occupies a valley that existed during or even before early Pleistocene time, the alluvial fill is undoubtedly a composite of both early and late Pleistocene materials.

Previous--Report Start || Next--Selected Pleistocene Exposures

Kansas Geological Survey
Comments to
Web version Feb. 19, 2013. Original publication date 1971.