|Original published in D.F. Merriam, ed., 1964, Symposium on cyclic sedimentation: Kansas Geological Survey, Bulletin 169, pp. 427-439|
Cyclical boundaries coincide with nonsequences in the biostratigraphic record, whereas within the cycles the biostratigraphic succession is an undisturbed sequence.
Duration, or "hiatal value," of the nonsequences between cycles is evaluated on faunal grounds. Although the "hiatal value" of the breaks was found to range as high as 10, there seems to be no relation between the physical characteristics of the initial deposits and the duration value of the underlying hiatus.
This fact confronted the senior author as early as the 1920's in the course of mapping, under the sponsorship of the Wisconsin Geological and Natural History Survey and the Milwaukee Public Museum, of certain quadrangles in central and western Wisconsin. In the virtual absence of units of contrasting lithology (i.e. sandstone, limestone, and shale), the technique of stratigraphic subdivision was based primarily on two factors: establishment (1) of datum horizons through the discrimination of faunal assemblages with lateral distribution but limited vertical extent, and (2) of nonsequences by a combination of faunal and sedimentational criteria. By these means, a series of sequences and nonsequences was determinable which gave consistent results, not only in particular quadrangles, but, as determined through continued studies in the ensuing decades, over the region as a whole.
Whereas the faunal criteria were readily communicable by means of faunal lists and range charts, the sedimentational, based to a large extent on visual inspection of grain size and sorting characteristics, were conspicuously less so. In an effort to document these latter characteristics, the authors, in the early 1950's engaged in sampling the exposed succession in Wisconsin.
The purpose of this paper is to present sedimentational characteristics of a marine environment, expressed as transgressive-regressive cycles; therefore, considerations of stratigraphic breakdown and nomenclature are secondary. Nevertheless, it is necessary to consider the stratigraphic nomenclature and faunal zonation in the interest of supplying a time-stratigraphic frame of reference.
Acknowledgments--The writers wish to express their appreciation to the Illinois State Geological Survey, Urbana, for use of equipment and assistance in drafting, and to Shell Canada Ltd., Calgary, Alberta, for releasing pertinent western Canadian field and faunal data.
Whether this classification satisfies the requirements of the American Commission on Stratigraphic Nomenclature may be, and has been argued, (Nelson, 1953; Berg, 1954; Bell, Berg, and Nelson, 1956). The "formations" and "members" of Twenhofel, Raasch, and Thwaites and of Raasch certainly satisfy the requirements of time-stratigraphic divisions, stages, and substages. Yet all of them are bounded, at top and base, by recognizable physical criteria, and separated by discontinuities, which, at least on a regional scale, constitute disconformities. Whether, on this basis, they are to be regarded also as proper rock-stratigraphic divisions may be a matter of interpretation, and is not particularly relevant to the present purpose.
The following, then, is the stratigraphic subdivision employed.
|Prairie du Chien Group|
|Oneota Formation (Stage)|
|Hickory Ridge Member (base only sampled)|
|St. Croix Series|
|Trempealeau Group (Stage)|
|Sunset Point Formation (Substage)|
|Jordan Formation (Substage)|
|Van Oser Member|
|Lodi Formation (Substage) (not sampled)|
|St. Lawrence Formation (Substage) (not sampled)|
|Arcadia Formation (Substage) (not sampled)|
|Franconia Group (Stage)|
|Bad Axe Formation (Substage)|
|Hudson Formation (Substage)|
|Goodenough Formation (Substage)|
|Ironton Formation (Substage)|
|Dresbach Group (Stage)|
|Galesville Formation (Facies)|
|Eau Claire Formation (Facies)|
|Mt. Simon Formation (not fully sampled)|
The stratigraphic succession is graphically presented in Figure 1, which also carries a brief lithologic description of the units.
Figure 1--Croxian succession in the type region.
In the type region each of the seven genera-zones, with the exception of the Cedaria and Aphelaspis, may readily be divided into a number of species-zones. Thus, four species-zones comprise the Crepicephalus Zone, three the Elvinia Zone, ten the Conaspis Zone, eight the Ptychaspis-Prosaukia Zone and fourteen the Saukia Zone. Many of these species-zones, based on the biochrons of species-assemblages, maintain their identity as far as Oklahoma, Texas, and the Canadian Cordillera.
The species-zone succession within the confines of the type area does not seem to be radically different from that outside, with two exceptions. The first of these relates to the Cedaria Zone. In Alberta, for example, the "Cedaria" woosteri species-zone, of the Upper Mississippi Valley is underlain by at least two additional species-zones, which may be correlated with fauna described by Lochman and Duncan (1944) from central Montana. The absence of the earlier Cedaria fauna in the Mississippi Valley appears logically to have been a result of the later arrival of the Dresbachian marine transgression here.
The other horizon at which fauna known from other regions is absent in the Mississippi Valley corresponds to the Dresbachian-Franconian hiatus. This gap in the type Croixan record is filled in certain other areas by (1) several younger species-faunae of the Aphelaspis Zone, followed by (2) an additional major or genera-zone (Dunderbergia Zone) , and finally (3) by several species-zones of Elvinia older than those in Wisconsin.
The 41 successive species-zones discriminable within some 500 feet of strata in the Croixan type succession in the Mississippi Valley (Fig. 2) are relevant to the problem of sedimentation and sedimentary cycles in that they establish the position of significant nonsequences and also furnish a basis for evaluation of the duration of the hiatus. The fact that these significant hiatuses are also identifiable on the basis of physical, sedimentational evidence is, of course, the theme of the present discussion.
Figure 2--Relation of Croxian faunal zones and sedimentary cycles.
Figure 3--Southwestern Wisconsin showing location of important sections.
Figure 4--Cambrian-Ordovician contact at two Wisconsin localities.
It is to be noted that the initial deposit above the contact, i.e. the basal Oneota sand, yields a histogram distinguished by a wide range in grain size, a marked skewing to the right, and a moderately bimodal character. In addition, the upper sand is notably coarser than the lower. A high percentage of carbonate was removed by solution in acid in the Oneota sample from Stoddard, whereas the Glovers Bluff sample of friable sand was untreated, indicating the negligible effect of carbonate content relative to sand characteristics.
The considerable similarity of the histograms of the two Sunset Point samples from below the contact is also notable.
In Figure 5, it can be seen that the histogram pattern for the Ironton base at the Friendship and Wood Hill localities is strikingly similar to that for the Oneota base at the two localities shown in Figure 4. That is, the sorting is poor and the histogram is bimodal and skewed to the right. At the Sparta locality, the characteristics are somewhat subdued, in that sorting is less poor and the bimodal character is lacking. The Ironton depositional cycle here, where it is 42 feet thick, is complex in contrast to the simple cycles at the other two localities, where the formation is appreciably thinner. However, the contrast in average coarseness between the Ironton and Galesville sands at all three localities is to be noted. Also the Galesville is appreciably less well sorted at the Sparta locality, where the Ironton base lies deeper in the truncated Galesville sedimentational cycle.
Figure 5--Dresbachian-Franconian contact at three Wisconsin localities.
The histogrammic relationships of the uppermost Jordan to the basal Sunset Point sand are illustrated in Figure 6, which compares two localities separated by some 90 miles along a west-northwest and east-southeast axis. As previously discussed, the sands may be differentiated and the contacts located by inspection on the outcrop. Although the Sunset Point Formation seldom exceeds 35 feet in thickness and is generally much thinner, it is widespread over the region but apparently absent in the northwestern part (St. Croix Valley, etc.), and may be locally absent elsewhere, presumably as a consequence of the irregularities of pre-Oneota erosion.
Figure 6--Jordan-Sunset Point contact at two Wisconsin localities.
An interesting corollary to the cyclical independence of the Sunset Point is its heavy mineral character, it being garnet free in contrast to the high garnet ratio in both the underlying Jordan and the overlying Oneota Formations (Ockerman's 1930 paper would seemingly refute this, but a re-examination of his outcrops revealed that no sample of "Madison" (= Sunset Point) sandstone was analyzed).
From an examination of Figure 6, it is readily apparent that the basal sands of the Sunset Point Formation possess the same histogrammic characteristics as those of the Ironton and Oneota, namely, relatively poor sorting, a skewing to the right, and a weakly bimodal character. Moreover, the underlying sand is remarkably well sorted, about 60 percent being accounted for by two adjacent screen sizes. The fact that only 64 percent is so confined in the Mendota sample and 84 percent in the Stoddard sample is a consequence of the pre-Sunset Point removal of the closing or Van Oser phase deposits at Mendota in contrast to their preservation at Stoddard.
The additional horizons at which the dextrally skewed histogram significant of cyclical breaks was recorded are three in number and lie, in descending order, as follows: (1) base of Bad Axe Formation, Franconia Group; (2) base of Hudson Member, Franconia Group; and (3) base of the "Upper Mt. Simon" ("Sooty Zone") in the Dresbach Group. Significantly, each of these lies at a horizon corresponding with the base of a major faunal zone.
Figure 7--Cyclical boundaries in the Franconia Group near Tomah, Monroe County, Wisconsin.
Relative to the sample from the Bad Axe base, it may be noted that it maintains the typical characteristics of a transgressive phase histogram, despite the fact that it is a conglomerate. These "sandstone conglomerates" consist of virtually or totally unconsolidated pebbles in an unconsolidated matrix. Whether conglomeratic or not, the total sample apparently maintains about the same proportions of old (i.e. reworked) and new (i.e. transported or regolithic) sand.
With respect to the overlying Eau Claire Formation, relations have been interpreted as transitional, at least in a gross sense. In most localities for example, some tens of feet below the base of Eau Claire lithology, worm borings appear in the Mt. Simon type sands, followed by beds with a high percentage of comminuted oboloid brachiopods and finally by the fine sands and shales of Eau Claire type with fully marine faunas of trilobites, brachiopods, and pteropods. Accordingly, it has remained somewhat of an arbitrary decision whether to draw the base of the Eau Claire Formation at the first appearance of marine criteria or to draw it at the line of change in gross lithology from the coarser and more thickly bedded sands of Mt. Simon type to the more thinly bedded fine sands and shales of Eau Claire type.
Examination of the mechanical analyses from this portion of the sequence from Mt. Simon Peak and Mt. Washington Bluff, in the suburbs of Eau Claire, Wisconsin, suggests a resolution of this nomenclatural problem.
There, the base of the sand which bears the worm borings indicative of the approach of marine waters yielded histograms similar to those marking such known transgressions as those at the base of the Franconian (Ironton), and of the Canadian (Hickory Ridge). Moreover, the coarse beds with oboloids immediately beneath strata of Eau Claire lithology yielded histograms closely comparable with those from the top of the Ironton. Finally, the basal Eau Claire and basal Goodenough, which overlie the "Mt. Simon" and Ironton respectively, also yielded closely comparable histograms (Fig. 8).
Figure 8--Comparison of early phases of Dresbachian and Franconian sequences.
Thus, it would appear that the "upper" or "marine" Mt. Simon unit and the Ironton Formation are remarkably similar in physical character, in presence of worm borings and oboloid fragments, in stratigraphic relations, and, therefore, in genetic significance. Both comprise the initial deposition of regionally widespread transgressions and bear a similar relationship to the nonmarine, probably freshwater sediments which underlie them. Both, moreover, ought therefore to merit equal nomenclatural status.
Accordingly it is tentatively proposed to designate the "Upper" or marine Mt. Simon as the Shawtown Formation, after that district of the City of Eau Claire in which Mt. Washington (the Eau Claire type locality) is situated.*
(*"Shawtown" is a new name proposed herein for the basal member of the Eau Claire Formation. The name is taken from the district of the city of Eau Claire, Wisconsin, which lies at the foot of Mt. Washington Bluff on west side of the Chippewa River. Because the basal contact of the unit is better exposed in Mt. Simon Bluff, in the north part of the city, that locality is selected as the type, however, with Mt. Washington a supplementary type.
At Mt. Simon the unit is 12 feet thick, and underlies 10 feet of Obolus-bearing Eau Claire sandstone which form the peak of the bluff. The histogrammic character, lithology, and stratigraphic relations are shown in Figure 9. Fauna consists of "worm" borings and comminuted fragments of phosphatic brachiopods. In addition, grains under a lens show a sooty coating as in the sands which underlie the Eau Claire Formation in Illinois, where they are known to subsurface investigators as the "Sooty Zone" (see Templeton, J.S., 1950).)
Figure 9--Shawtown type section, Mt. Simon, Eau Claire, Wisconsin.
Fossil remains, or any traces of organic habitation, are highly exceptional in the coarse, high-energy sediments of the dextrally skewed cyclothem, but these sediments grade rapidly upward into sediments with abundant traces of organic habitation (borings, trails, and organic reworking of the sands) and also identifiable fossils.
The sinistrally skewed histograms occur where rich marine faunas commonly pack the fine sands immediately above the contact, and glauconite is typically present. "Mudcracks" and "current" ripples are common. Shallow-marine waters seem to be clearly indicated.
The terminal histograms, on the other hand, indicate a greater variety of environments, on criteria other than those of mechanical analyses. In the Shawtown and Ironton cycles, the terminal beds are the most richly fossiliferous and bear most of the identifiable organic remains. In the Goodenough and Hudson cycles, marine conditions continue throughout, but there tends to be an overall reduction in faunal abundance upward; but the Sunset Point is sparingly fossiliferous throughout. The Jordan cycle, where fully developed, closes with steeply dipping foreset beds of fairly constant inclination which the writers interpret as beach deposits building seaward under a static profile of equilibrium. Except at one locality (Van Oser Creek, Minnesota; Stauffer, 1940), the beds lack fossils and show little evidence of organic habitation. Finally, the terminal phase of the Galesville, with thin and regular bedding, variously oriented "wave" ripple marks, and total absence of any evidence of organic habitation points toward a fresh water, possibly lacustrine environment.
Despite the considerable variety of indicated environments, the terminal phases of the seven cycles possess in common a histogram type in which over 75 percent of the sediment is confined to two adjacent screen sizes and less than 3 percent passes the 150-mesh screen. The terminal phase of the cycle seems to represent a concentration of the coarser fractions, and an elimination by extraction of the finer fractions. This suggests no change in the character of the sediments received, but a passing on of the fines, due to energy increase, so that only the coarser fractions remain. Wave action under regressive conditions resulting from the progressive filling of the basin to the profile of equilibrium is postulated.
The succession within any particular cycle is not so amenable to generalization, because each may embrace a considerable variety of environments, and not all of these occur within every cycle. A much more impressive quantity of data needs to be gathered before any significant degree of generalization, let alone interpretation, can be made. In general, however, it is apparent that there is an overall upward progression in the direction of better sorting and coarser average texture.
More specifically, the Eau Claire-Galesville cycle embraces the Cedaria, Crepicephalus, and Aphelaspis zones, although the Shawtown cycle probably also falls within the time-span of the Cedaria Zone. On the other hand, the Saukia Zone embraces three demonstrable cycles (Bad Axe, Jordan, and Sunset Point) plus those represented by the unstudied formations in the lower Trempealeau. Thus, while the breaks between the seven faunal zones are remarkably sharp, not only in this region but over the continent as a whole, not all of them, and possibly none of them, can be attributed primarily to sedimentary hiatus, although the two may coincide locally or regionally.
Cyclical boundaries in the Croixan region which do correspond with faunal boundaries are those between (1) the Aphelaspis and Elvinia faunas, (2) Elvinia and Conaspis, (3) Conaspis and Ptychaspis-Prosaukia, (4) Ptychaspis-Prosaukia and Saukia, and (5) Saukia and Symphisurina. Studies within the region or beyond it reveal that a portion of the faunal record may be absent in a position corresponding to the cyclical boundary. This is determinable on the basis of the minor, but equally discrete, species-zones which comprise the respective major or genera-zones. If the total known species-zones for the continent as a whole are considered, and each species-zone is given a value of 1, the magnitude of the faunal break at a hiatus at a particular locality may be expressed integrally in terms of the number of missing zones. This might be termed the hiatal value of the nonsequence.
If we consider the five faunal hiatuses mentioned above, the first or Eau Claire, Galesville-Ironton (i. e. Aphelaspis-Elvinia) hiatus proves to have the highest value. As yet, the intervening beds and faunal succession, as expressed in parts of New Mexico and Texas, have not been closely zoned, but a consideration of Palmer's 1960 paper on the sequence in New Mexico suggests a hiatal value of at least 5.
The faunal succession involving the Elvinia and Conaspis zones is very well known both locally and in other regions. On this basis it is possible to state that in a very small area of the type region (the area surrounding Wisconsin Dells in central Wisconsin), the succession is complete; in other words, the hiatal value is zero. Elsewhere, the hiatus increases in magnitude, and in some parts of the St. Croix Valley may be as high as 10.
Nevertheless, the mechanical-analysis characteristics at the break between the Ironton and Goodenough cycles are the same, regardless of whether the hiatal value be 0 or 10. A further anomaly--where the succession is complete, the highest fauna of the Elvinia Zone (lrvingella major) is physically not a part of the Ironton cycle, as one might logically expect, but of the Goodenough cycle. Thus, even where the succession is complete, faunal and physical boundaries do not coincide, although there is no transition, either faunally or physically. The same relationship seems to exist in other areas, notably Oklahoma and the Canadian Rocky Mountains, where the lrvingella beds, though faunally related to the underlying, are physically inseparable from the overlying strata.
The break between the Goodenough and Hudson cycles, and the corresponding break between the Conaspis and Ptychaspis-Prosaukia groups of faunas has, over most of the type region, a hiatal value of 3; but in the St. Croix Valley, the succession, on the basis of phyllogenetic studies, is complete, i.e. has a zero value. Here, on visual inspection, it seems probable that the boundary between the two sedimentary cycles also disappears, but this has not yet been documented by mechanical-analysis techniques.
The faunal assemblage which characterizes the Bad Axe cycle (the species-zone of Saukiella minor) appears to be extremely widespread, as it has recently been discovered in virtually identical assemblages in the Canadian Rockies. With respect to the underlying Hudson cycle, the hiatus in the type region seems to have a value of from 2 to 3 over most of the region. Where this decreases to zero, in the Baraboo Region, mechanical analyses have yet to be made.
The break between the Jordan and Sunset Point cycles, involving the sporadically occurring species-zones of Saukia, is difficult to evaluate. Where the terminal or Van Oser beds of the Jordan are fossiliferous, the Sunset Point is, unfortunately, absent. It is also not known whether the Van Oser and Sunset Point faunas are directly time successive. On mainly physical grounds, it may be estimated that the hiatal value of this nonsequence is nowhere higher than about 3, and may be zero in some areas, as along the Mississippi River south of La Crosse. The cyclical boundary is everywhere maintained.
Finally, with respect to the Sunset Point-Oneota (Hickory Ridge) boundary, an estimate of the hiatal value is facilitated by the discovery, in the Richardson Mountains of Yukon Territory, of a pre-Symphysurina fauna, characterized by a species-assemblage including Pareuloma brachymetopa Rasetti. Comparison with an underlying Saukia fauna in this section suggests that the Saukia-Symphysurina faunal break at the Cambrian-Ordovician boundary is not great, and might have a hiatal value of no more than 1. Where the Oneota strata, in the absence of the Sunset Point, locally rest directly on some portion of the Jordan Formation, this hiatal value would, of course, be correspondingly increased.
Thus, in summary, the nonsequence of greatest duration in the Croixan type succession would appear to be that between the Eau Claire-Galesville and Ironton cycles, whereas that between the Cambrian and the Ordovician (Sunset Point-Oneota) is considerably less. There seems to be no obvious relationship between the magnitude of the breaks and the physical characteristics of the basal beds of overlying cycles.
Berg, R. R., 1954, Franconia Formation of Minnesota and Wisconsin: Geol. Soc. America Bull., v. 65, no. 9, p. 857-881.
Lochman, C., and Duncan, D., 1944, Early Upper Cambrian faunas of central Montana: Geol. Soc. America Sp. Paper 54, p. 1-181.
Lochman-Balk, C., and Wilson, J. L., 1958, Cambrian biostratigraphy of North America: Jour. Paleontology, v. 32, no. 2, p. 312-350.
Nelson, C. A., 1953, Revision of Croixan Dikelocephalids--a comment: Jour. Paleontology, v. 27, no. 5, p.734-736.
Ockerman, J. W., 1930, A petrographic study of the Madison and Jordan Sandstones of southern Wisconsin: Jour. Geology, v. 38, no. 4, p. 346-353.
Palmer, A. R., 1960, Trilobites of the Upper Cambrian Dunderberg Shale, Eureka District, Nevada: U. S. Geol. Survey Prof. Paper 334-C, p. 663-681.
Raasch, G. O., 1935, Stratigraphy of the Cambrian System of the Upper Mississippi Valley: Kansas Geol. Soc. 9th Ann. Field Conf. Guidebook, p. 302-315.
Raasch, G. O., 1939, Cambrian Merostomata: Geol. Soc. America Sp. Paper 19, p. 1-146.
Raasch, G. O., 1952a, Oneota Formation, Stoddard quadrangle, Wisconsin: Illinois State Acad. Sci. Trans., v. 56, p. 85-95.
Raasch, G. O., 1952b, Revision of Croixan Dikelocephalids: Illinois State Acad. Sci. Trans., v. 44, p. 137-151; reprinted, Illinois State Geol. Survey Circ. 179, p. 137-151.
Stauffer, C. R., 1940, Fauna of the Van Oser beds: Jour. Paleontology, v. 14, no. 1, p. 54-56.
Templeton, J. S., Jr., 1950, The Mt. Simon Sandstone in northern Illinois: Illinois State Acad. Sci. Trans., v. 43, p. 151-159.
Twenhofel, W. H., Raasch, G. 0., and Thwaites, F. T., 1935, Cambrian strata of Wisconsin: Geol. Soc. America Bull., v. 46, p. 1687-1744.