|Original published in D.F. Merriam, ed., 1964, Symposium on cyclic sedimentation: Kansas Geological Survey, Bulletin 169, pp. 183-189|
Brigham Young University, Provo, Utah
A radiographic analysis was made of 123 samples of seemingly homogeneous, medium- to massive-bedded sandstones from the Pennsylvanian Stranger, Tallant, and Vamoosa Formations of Kansas and Oklahoma. This study reveals that the internal structure of these rocks consists of regular rhythmic alternations of "dense" and "light" layers. In many respects the laminations resemble varves, but the individual layers are characteristically undulatory and lenticular and have gradational contacts. The expression of the laminations on a radiograph results from density variations presumably caused by concentrations of accessory minerals and iron oxide cement.
Although it is possible that these laminations are annual or represent some other period of time, they probably result from influx layers which accumulated under a given current velocity and direction without a definite cycle of time involved in the sedimentary process.
Pennsylvanian sandstones of the Midcontinent region are considered by most writers to represent channel and flood-plain deposits of southwestward flowing drainage systems with associated beaches and bars developed along a marginal shallow marine shelf (Tanner, 1956, p. 15; Potter and Pryor, 1961, p. 1218; Doty and Hubert, 1962, p. 35-37; Wanless and others, 1963, p. 458-486). These rocks are characterized by various types of crossbedding and other current structures, but some appear to be massive and homogeneous. Traditionally, the seemingly structureless nature of the homogeneous units has been considered to be the result of a special environment in which sedimentary processes were essentially uniform. Preliminary radiographic analysis of these rocks, however, reveals they are composed of alternating layers of dense and light [see note] material which form a simple rhythmic pattern. In many respects these laminae are similar to varves, but they possess a number of unique and interesting features suggesting an origin without regularity of period. The purpose of this paper is to briefly describe these laminations as a preliminary step in a more comprehensive study of their nature, origin, and significance.
[Note: The terms "dense" and "light" in this paper refer to the degree to which radiation will he absorbed. Dense layers absorb more radiation and appear black or dark gray on a positive print of the X-ray picture. Light layers absorb relatively little radiation and appear white or light gray.]
Acknowledgments--This study was greatly facilitated by a number of individuals who donated their time, equipment, and assistance. Gary F. Stewart, Richard L. Schuman, and Carl B. Kinell of the Kansas Geological Survey helped collect and prepare most of the samples. Special equipment from the J. T. Lardner Co., Topeka, Kansas, was used to cut the extra large specimens and prepare the slabs for radiography. Miss Virginia Detlor, X-ray technician at the Watkins Hospital, The University of Kansas, made most of the radiographs and offered many helpful suggestions concerning X-ray techniques.
Except for variations in degree of cementation these rocks are lithologically similar. They consist predominantly of light-buff, fine-grained, well-sorted, micaceous, quartz sand and are characteristically medium- to massive-bedded (Fig. 1). Thin shale seams separate the horizontal units but little or no internal structure is expressed on the weathered surface. Some suggestion of stratification may be found on a freshly sawed surface, but the detail visible is rarely sufficient to permit identification of the type of structure present. For most purposes the sand within each bed is considered to be structureless and homogeneous. Even under a binocular or petrographic microscope most samples show little or no evidence of stratification (Fig. 2). Parting lineations are common, however, in specimens which can be split parallel to the bedding plane.
Figure 1---Outcrop of sandstone in Tallant Formation on U. S. Highway 60 just east of Osage Hills State Park, Oklahoma. Exposure is typical of seemingly structureless sandstone which contains rhythmic laminations. Expressed stratification ranges from medium- to massive-bedded with thin shale partings separating major units. Little or no internal structure is expressed on weathered surface.
Figure 2--Photograph showing characteristics of sandstone expressed on freshly sawed surface. Area shown in this picture is near notch in top central part of specimen from which radiograph shown in Figure 3 was made.
Iron oxide is the dominant cement and is either concentrated in irregular patches or along layers parallel to the bedding. The degree of cementation, however, is generally very slight and the sandstones are characteristically porous, weak, and friable. Except for minor differences in grain size, the sandstones within the seemingly homogeneous units are identical to the rest of the local Pennsylvanian sandstones in which sedimentary structures are well expressed.
In order to adequately study the nature of laminations, large blocks as much as 2 feet thick and 3 feet long were quarried from outcrops and were prepared by a stone cutter into slabs approximately 1/2 inch thick. In each outcrop a vertical sequence of samples was collected so that a nearly complete composite section of the entire outcrop was obtained. In addition, a number of beds were sampled horizontally at 6-foot intervals so that lateral variations in the laminae could be studied.
A total of 123 radiographs was made from the slabs using one or more 14" x 17" sheets of Kodak type M industrial X-ray film following essentially the same procedures outlined by the writer in an earlier publication (Hamblin, 1962).
Laminations within the seemingly homogeneous sandstones studied consist of regular rhythmic alternations of dense and light layers (Fig. 3, 4). In most samples the rhythmic pattern approaches a varvelike constancy and closely resembles stratification found in shale of the Green River Formation of the Colorado Plateau (Bradley, 1929, Pl. 11), and the Upper Cretaceous shales of the Black Hills (Rubey, 1930, Pl. 4). Pairs of laminae range from 0.1 to 2 mm thick, with between 22 and 29 pairs of laminae to the inch.
As in true varves, the light layers are characteristically thicker than those that are dark (dense). In the Tonganoxie Sandstone the average thickness of light layers is 0.8 mm, whereas the average thickness of dark layers is only 0.1 mm. This ratio of 8 to 1 generally holds true for most specimens, for as thickness of the light layers increases, thickness of the dark layers generally increases in the same proportion.
Contacts between light and dark laminae, and between pairs of laminae are generally indistinct and poorly defined. Grains which define the dark laminae are somewhat dispersed so that a lamina is defined by a series of dots rather than a solid layer. As a result many of the laminae appear to be discontinuous and fade out where concentrations of dense material decrease laterally.
Figure 3--Positive print (actual size) of a radiograph showing characteristics of laminations within seemingly homogeneous sandstones of Vamoosa Formation, Note lenticular nature of light layers and manner in which dark layers merge together and split into two or more units, Specimen from road cut on Oklahoma Highway 99 in NW SW sec, 17, T. 27 N., R, 10 E., Osage County, Oklahoma, Technical factors: specimen 1/2 inch thick exposed for 2 seconds at 35 MA and 40 KV on Kodak type M industrial X-ray film from a distance of 30 inches. A larger version of this image is available.
One of the most striking features of this stratification is that the laminae are rarely perfectly planar. In contrast to true varves, the dark layers are characteristically undulatory. Moreover, where traced laterally, they commonly bifurcate or split into three or more layers or merge to form an abnormally thick dense unit (Fig. 4). The enclosed light layers thus thicken and thin laterally and form elongate lenses arranged in a somewhat staggered pattern. Few if any light layers are more than 10 inches long.
Figure 4--Positive print (actual size) of a radiograph showing characteristics of laminations within medium-bedded units of Tonganoxie Sandstone Member of Stranger Formation. Dark laminae are defined by series of dots rather than solid layer, so many appear to be discontinuous and fade out laterally. Two zones of abnormally dark layers, however, can be traced laterally for more than 50 feet. Contacts between layers are characteristically indistinct and poorly defined. Specimen from Deitmans Crossing on south bank of Wakarusa River, NE NE sec. 14, T. 13 S., R. 20 E., Douglas County, Kansas. Technical factors: same as Figure 3. A larger version of this image is available.
Vertical and lateral control of sampling in the study clearly indicates that the rhythmic laminae are not local structural features but exist throughout most outcrops. At Deitmans Crossing, Douglas County, Kansas (collecting locality number 1), for example, the laminae are found throughout all samples from the medium-bedded units in the upper part of the section. In this exposure the laminae are essentially continuous in a section more than 17 feet thick. Moreover, the zone of abnormally thick dark laminae within a given bed, such as the two shown in Figure 4, can be traced laterally for more than 50 feet. From the evidence accumulated in this study, it is likely that many of the medium to massive horizontal beds in the Midcontinent Pennsylvanian sandstones are composed of rhythmic laminations.
The expression of laminae on a radiograph of seemingly homogeneous sandstone results from density variations which occur in layers only a few grains thick. These variations are not apparent in hand specimens nor under a petrographic microscope so it is difficult to determine precisely their nature and origin. Studies of thin sections suggest, however, that in most samples density variations result from weak concentrations of accessory minerals such as mica, clay, and "the heavies," or concentrations of iron oxide cement.
Regardless of whether the expression of the laminations is due to differential cementation or not, the ultimate control of layering must be primary sedimentation. In all probability laminae were caused by variations in the rate of supply or deposition of different material. As is pointed out by Rubey (1930, p. 40) such variations might have been due to changes in the quantity of material carried by the water or changes in capacity of the currents. These changes may have been cyclic with definite intervals of time or may have occurred without any regularity of period.
Johnston (1922, p. 128) interpreted laminations (15 to 20 laminae per inch) in tidal flood-plain deposits of the Fraser River delta to be the result of ebb and flow of tides. Thus, the laminae are not annual but represent a definite period. Seasonal layers (5 or 6 per foot) on the alluvial flood plain are much thicker. Other laminations similar to those described in this paper in lacustrine sediments and marine shales have been interpreted as seasonal layers or varves (Bradley, 1929; Rubey, 1930).
Laminations within the seemingly homogeneous Pennsylvanian sandstones could possibly be due to tidal fluctuation, but it seems unlikely that seasonal cycles would develop alternating thin, undulatory and lenticular laminae such as described in this paper. Each pair of laminae possibly represents a pulse in the process of sedimentation and could be considered as an influx layer that accumulated at a given current velocity and duration. The quartz would be deposited first and form the "light" layer followed by deposition of quartz with admixed mica, clay, and other accessory minerals to form the "dark" layer. Iron oxide could subsequently be selectively implaced along the "dark" layer. A rhythmic pattern could thus be developed without a definite cycle of time involved in the sedimentary process. The possibility that these laminae are annual or periodic is certainly not rejected for it is obvious that further detailed studies are needed before an adequate theory can be developed.
Bradley, W. H., 1929, The varves and climate of the Green River epoch: U. S. Geol. Survey Prof. Paper 158-E, p. 87-110.
Doty, R. W., and Hubert, J. F., 1962, Petrology and paleogeography of the Warrensburg channel sandstone, western Missouri: Sedimentology, v. 1, p. 7-39.
Hamblin, W. K., 1962, X-ray radiography in the study of structures in homogeneous sandstones: Jour. Sed. Pet., v. 32, p. 201-210.
Hamblin, W. K., 1965, Internal structures of "homogeneous" sandstones: Kansas Geol. Survey Bull. 175, pt. 1, p. 1-37. [available online]
Johnston, W. A., 1922, The character of the stratification of the sediments in the recent delta of the Fraser River, British Columbia, Canada: Jour. Geology, v. 30, p. 115-129.
Potter, P. E., and Pryor, W. A., 1961, Dispersal centers of Paleozoic and later clastics of the upper Mississippi Valley and adjacent areas: Geol. Soc. America Bull., v. 72, p. 1195-1250.
Rubey, W. W., 1930, Lithologic studies of fine-grained Upper Cretaceous sedimentary rocks of the Black Hills region: U. S. Geol. Survey Prof. Paper 165-A, p. 1-54.
Tanner, W. F., 1956, Geology of northeastern Osage County, Oklahoma: Oklahoma Geol. Survey Circular 40, p. 1-76.
Wanless, H. R., and others, 1963, Mapping sedimentary environments of Pennsylvanian cycles: Geol. Soc. America Bull., v. 74, p. 437-486.