SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY ? NUMBER 18 Stratigraphy and Preliminary Biostratigraphy of the Flagstaff Rim Area, Natrona County, Wyoming Robert J. Emry SMITHSONIAN INSTITUTION PRESS City of Washington 1973 ABSTRACT Emry, Robert J. Stratigraphy and Preliminary Biostratigraphy of the Flagstaff Rim Area, Natrona County, Wyoming. Smithsonian Contributions to Paleo- biology, number 18, 43 pages, 19 figures + frontispiece, 1973.?About 750 feet of sediments of the early Oligoc?ne (Chadronian) White River Formation are exposed along Flagstaff Rim in south-central Natrona County, Wyoming. About 4,000 specimens of fossil vertebrates have been collected from these outcrops. The White River Formation unconformably overlies rocks ranging in age from Precambrian to medial or late Eocene. The lithology of the White River For- mation is predominantly claystone and conglomerate in the lower part of the section, changing to predominantly tuffaceous siltstone and conglomeratic chan- nel sandstones in the upper part. Four stratigraphie sections are described. A geo- logic map of about 40 square miles illustrates the areal limits of the White River Formation and its relationships to underlying and overlying formations. Several distinct and easily recognizable volcanic ash beds occur at intervals within the White River sequence. These serve as convenient markers for precise strati- graphic zonation of fossils and have also provided minerals for potassium-argon dating. Dates obtained range from 35.7 to 31.6 million years. A boulder conglomerate unit, previously considered to be the basal unit of the White River Formation and/or part of the Wind River Formation is shown to be a distinct, and probably unnamed, unit, and should not be assigned to either of these formations. It unconformably overlies the Wind River Formation and is separated from the White River Formation by an erosional disconformity with several hundred feet of relief. This information allows new interpretations of the structure of the area and adds a previously unrecognized episode of depo- sition and erosion to the history of the area. The most common fossil in the White River sequence is the artiodactyl genus Leptomeryx, which is represented by two morphologically distinct lineages. One lineage is provisionally divided into two and the other into three size groups that are believed to represent different species. The local stratigraphie ranges of the different groups do not overlap. In each lineage, the size increases higher in the section. None of the groups are definitely assigned to named species, pending studies to determine the validity and limits of the named species. Preliminary analysis of other elements of the fauna shows that there is recog- nizable change through time within individual lineages and that the faunal composition as a whole changes through time, within the local sequence. When the entire fauna is analyzed in detail, it should be possible to establish local range zones of the fossil species and, by their use, to gain greater temporal resolution within Chadronian time than has previously been possible. OFFICIAL PUBLICATION DATE is handstamped in a limited number of initial copies and is recorded in the Institution's annual report, Smithsonian Year. SI PRESS NUMBER 4790. SERIES COVER DESIGN: The trilobite Phacops rana Green. Library of Congress Cataloging in Publication Data Emry, Robert J., 1940- Stratigraphy and preliminary biostratigraphy of the Flagstaff Rim area. Natrona County, Wyoming. (Smithsonian contributions to paleobiology, no. 18) 1. Geology, Stratigraphie?Oligoc?ne. 2. Vertebrates. Fossil. 3. Geology?Wyoming Natrona Co. I. Title. II. Series: Smithsonian Institution. Smithsonian contributions to paleobiology, QE70?.S56 no. 18 [QE693] 560'.8s [551.7'8] 72-13884 For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 Price 95 cents domestic postpaid or 70 cents GPO Bookstore Contents Page Introduction 1 Acknowledgments 2 Location and Extent of Area 3 Previous Investigations . . 3 Present Investigation 4 Geography 5 Climate 5 Drainage and Topography 5 Stratigraphy 5 General Features 5 Precambrian 5 Paleozoic Sedimentary Rocks 6 Mesozoic Sedimentary Rocks . . 6 Tertiary System 6 Paleocene Series . 6 Eocene Series . 6 Indian Meadows Formation 7 Wind River Formation .... 7 Unnamed Boulder Conglomerate 7 Oligoc?ne Series 15 White River Formation 15 Definition . 15 Distribution and Thickness 18 General Features 19 Stratigraphie Sections 20 North Fork of Lone Tree Gulch Section 20 Blue Gulch Section 22 Little Lone Tree Gulch Sections 23 Miocene Series 27 Split Rock Formation 27 Structure 27 Preliminary Biostratigraphy 28 Systematic Faunal List 31 Discussion 34 Literature Cited 41 in Doctors W. D. Matthew and R. S. Lull on an American Museum of Natural History expedition in 1899, at the head of Bates Hole, Wyoming, near the present study area. (By permission of the Department of Vertebrate Paleontology, American Museum of Natural History.) Stratigraphy and Preliminary Biostratigraphy of the Flagstaff Rim Area, Natrona County, Wyoming Robert J. Emry Introduction In south-central Natrona County, Wyoming, about 750 feet of strata of the early Oligoc?ne (Chadronian) White River Formation are exposed along a prominent erosion scarp known locally as Flagstaff Rim. In the Frick Collection, American Museum of Natural History, are nearly 3000 speci- mens of fossil vertebrates collected from these out- crops. In addition, at least 1000 specimens were collected from the area for the National Museum of Natural History during the summer of 1971. Nearly all of the specimens were carefully related stratigraphically to a number of distinct volcanic ash beds that occur at intervals within the rock sequence. These volcanic ash beds serve not only as convenient markers for precise stratigraphie docu- mentation of fossils, but have also yielded minerals for absolute age determinations, in terms of years before present, based on potassium-argon ratios. The Flagstaff Rim area is of great potential importance to our understanding of the early Oligoc?ne (Chadronian) of the entire Rocky Moun- tains area. The section here is thicker, and prob- ably more nearly represents all of Chadronian time, than any other known single section. Much of the Robert J. Emry, Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Wash- ington, D.C. 20560 section is richly fossiliferous and all of the speci- mens from the area can easily be tied to a single reference section. The fauna is more diverse than any other known early Oligoc?ne fauna in North America. Chadronian vertebrates are quite widespread throughout the Western United States, but many of the sites from which they are derived are quite localized. In Canada, the Cypress Hills area of Saskatchewan is the most important locality. Mon- tana has Pipestone Springs, McCartys Mountain, Canyon Ferry Reservoir area, and Sage Creek among its more important localities, but also has many smaller or less well-known localities. Wyo- ming has many isolated localities in addition to the larger areas of outcrop of Chadronian sediments in the eastern part of the state and along Beaver Rim in the central part of the state. In the northern end of the Bighorn Range, western part of the Black Hills, eastern side of the Medicine Bow Range, southeast end of the Wind River Range, and immediately south of Yellowstone Park are isolated localities that, because of their relatively high altitudes, have important structural connota- tions. Chadronian mammals have been found in Jackson Hole. The Pumpkin Buttes in the Powder River Basin are capped by Oligoc?ne sediments of yet uncertain age. Farther south, there are isolated localities in the parks of central Colorado, and 1 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY much farther south, the Vieja Group of the Basin and Range Province along the Rio Grande River in Texas contains important Chadronian faunas. But most of these localities have been difficult to place temporally, any closer than Chadronian, for several reasons, among these being scarcity of speci- mens at many localities, local occurrences of species unknown elsewhere, and the difficulty of establish- ing local stratigraphie sequences. The Flagstaff Rim area of central Wyoming then becomes important in relating all of these localities. It has many assets that are lacking in other areas: relatively thick sequence, most of which is fossilif- erous; extremely varied fauna; easily established local stratigraphie sequence with good marker beds for precise stratigraphie documentation of specimens; and, a central geographic location rel- ative to most of the other localities. A study of the distribution of all the fossils within the Flagstaff Rim sequence should provide a section that could be subdivided on the basis of the fauna and which would allow more resolution in correlation of many of the other isolated localities. The primary purpose of the present report is to make available the stratigraphie framework necessary for the faunal studies. This will not only facilitate the publication of my own work on various taxonomic groups but should also aid other workers who are studying other taxa from the area. The geologic map that was made during the course of my field work (Figure 19) covers about 40 square miles. This map illustrates the areal limits of the White River Formation and its relationships to both younger and older rock units, as well as a great deal of other information that would otherwise have required many pages of text. My dissertation also included the results of stud- ies of several mammalian taxa from the Flagstaff Rim area. These are briefly discussed in a follow- ing section of this report but publication of the details of these studies is deferred for reasons also outlined in the same following section. Detailed studies of the artiodactyl genus Leptomeryx and the rodent genus Cylindrodon, and preliminary studies of other genera, are sufficiently advanced to demon- strate that in members of these taxa, as well as in the fauna as a whole, there is recognizable change through time within this single stratigraphie se- quence. When the stratigraphie ranges of all the species in the fauna are determined, it should be possible to recognize temporal units of much less magnitude within Chadronian time than has previ- ously been possible. ACKNOWLEDGMENTS.?Appreciation should first be expressed to the late Mr. Childs Frick, whose generous support made possible the geologic field studies as well as the collecting and preparation of the large number of fossil specimens from the Flag- staff Rim area. I am also grateful to Mr. Morris F. Skinner and Mr. Ted Galusha, whose field parties collected the majority of the fossil specimens from the Flagstaff Rim area that are now in the Frick Collection, for including detailed stratigraphie and geographic information with each of the specimens. I would also like to take this opportunity to thank these two men for a great part of my informal and prac- tical education in vertebrate paleontology?thanks not only for giving me much advice gained through long experience but also for putting me into posi- tions where it was imperative that I use the information. I thank the staff of the Department of Vertebrate Paleontology of the American Museum of Natural History for the use of the fossil collections and other facilities, including a place to work. This report is a part of my dissertation for my Doctor of Philosophy degree (Columbia University, 1970); Dr. Malcolm C. McKenna, as my major professor, provided direction for the research. Assisting me in my field work during the sum- mers of 1967, 1968, 1969, and 1971 were, respec- tively, Mr. Robert M. Hunt, my wife Susan, Mr. Julian Kadish, and Mr. Albert C. Myrick. Figures 1, 14, and 16 were drafted by Mr. Ray- mond Gooris, Department of Vertebrate Paleontol- ogy, American Museum of Natural History. Figures 17 and 18, and the cross sections of Figure 19 were drafted by Mr. Larry B. Isham, Department of Paleobiology, National Museum of Natural His- tory. Mr. Chester Tarka, Department of Verte- brate Paleontology, American Museum of Natural History, has helped and given advice with other aspects of the illustrations. Critical reading of the final manuscript by Dr. J. D. Love, United States Geological Survey, Lara- mie, Wyoming, and by Dr. Richard H. Tedford, Department of Vertebrate Paleontology, American NUMBER 18 Museum of Natural History, is gratefully acknowl- edged. Part of the research for this report was carried out during my tenure as a Fellow of the Faculty of Pure Science of Columbia University, partially sup- ported by a National Science Foundation Graduate Fellowship. Location and Extent of Area The Flagstaff Rim area of this report includes about 40 square miles near the geographic center of Wyoming in south-central Natrona County (Figure 1). It is near the southeast end of the Wind River structural basin on the southwest flank of the basin and at the extreme southeast end of the Rat- tlesnake Range. The city of Casper is about 25 miles to the northeast and the village of Alcova is about 5 miles to the southeast. The area is approximately between latitudes 42?35' and 42?40'3O" north and between longitudes 106?10' and 106?48' west and centers on the mutual corners of the following four United States Geologi- cal Survey 7i/?-minute quadrangle sheets: Benton Basin, Benton Basin N.E., Clarkson Hill, and Al- cova, Wyoming. Previous Investigations The geology of the area of this report has been included in published reports and geologic maps of much larger areas and at much smaller scale with less detail, but no comprehensive report of the i Lara m ?e 1 ' 1 ? Cheyenne I I _> _| ? ? FIGURE 1 Map of Wyoming with stipples indicating the approximate location and extent of the Flagstaff Rim area of this report (see geologic map, Figure 19) . SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Tertiary stratigraphy and biostratigraphy has been published. The Oregon Trail traversed Willow Creek and Ryan Hill across the northwest corner of the mapped area, so it is certain that many thousands of people had passed through this area prior to the first geologic report by Dr. F. V. Hayden in 1869. He was the surgeon and naturalist attached to Captain W- F. Raynold's expedition of 1859 and 1860, and his reconnaissance study led to a report (1969) that briefly described the geography and in- cluded general statements about the tectonic history of the central Wyoming region. His geologic map, included in his report (1869), shows the Granite Mountains and the Rattlesnake Hills. Knight (1895), in a summary of the coal deposits of Wyoming, described the upper Cretaceous and Paleocene coal-bearing rocks of the northeast flank of the Rattlesnake Hills anticline. The same author, in a later report (1900), was apparently the first to recognize the major fault system within the area, parts of which affect the area of the present report. W. D. Matthew of The American Museum of Natural History made a reconnaissance of this area in 1899. After looking unsuccessfully for fossils in Oligoc?ne outcrops in the Bates Creek drainage 15 miles to the southeast, Matthew's expedition then went northwest, crossed the North Platte River and continued westward up Poison Spider Creek, turned southward, crossed the Rattlesnake Hills and continued southeastward back to Alcova on the North Platte River. This route completely encircled the present area of study, a richly fossil- iferous area, but Matthew's expedition was unsuc- cessful in finding fossil mammal remains. The ac- count of this reconnaissance is unpublished but can be found in Matthew's field book of 1899, on file in the Department of Vertebrate Paleontology, American Museum of Natural History. Hares (1916) briefly described the anticlines of central Wyoming and later (1946) published a re- connaissance geologic map of the southeast part of the Wind River Basin. A geologic map of Natrona County, Wyoming (Weitz et al., 1954), and the geologic map of Wyoming (Love et al., 1955) both include the area of the present study, but the small scale and consequent lack of detail make them both unsuitable as bases for this report. A report and geologic map by Rich (1962) have been valuable in the present study. The map in- cluded in Rich's report and another map by Den- son and Harshman (1969) both cover only part of the area of present concern, and both illustrate in- terpretations of the Tertiary stratigraphy and struc- tural relationships that differ from mine in some details. The interpretation of the stratigraphy and struc- tural relationships of the Tertiary rocks of the area of this report are in part dependent upon numerous other reports of surrounding areas. Contributing in this capacity are the reports of Hayden (1871), Endlich (1879), Darton (1908), Granger (1910), Sinclair and Granger (1911), Woodruff and Win- chester (1912), Bauer (1934), Van Houten (1964), Keefer (1965), Keefer and Van Lieu (1966), and Love (1970). Unpublished reports that have had some influ- ence on the present report are University of Wy- oming graduate theses by Berry (1950), Bogrett (1951), Rachou (1951), and Roehler (1957). So far as I can determine, no fossils of Eocene age have been reported from the area of present study. Fossils of early Oligoc?ne (Chadronian) age were reported by Rich (1962). The merycoidodonts of the area have been described by Schultz and Falkenbach (1954, 1956, and 1968). A manid was described by Emry (1970). Shorter papers on new rodents include the description of a new beaver (Emry, 1972a), a new heteromyid (Emry, 1972b), and a new cricetid (Emry and Dawson, 1972). Present Investigation The present report is based on field work done at various times since 1957. The late Charles Fal- kenbach of the Frick Laboratory, American Mu- seum of Natural History, collected mammalian fossils of Chadronian age from the area in 1941 and 1954, but these fossils have insufficient strati- graphic data to be used in this study. In 1957 detailed stratigraphie collection of verte- brate fossils was begun in the area by a Frick Laboratory party under the leadership of Mr. Mor- ris F. Skinner and Mr. Ted Galusha, who collected again in the area in 1958 and 1959. Skinner's field party spent a few days collecting in the area in 1963 and again in 1965. My particular interest in the area dates from 1958. During that field season, NUMBER 18 and the subsequent seasons noted above, I was a field assistant for Morris Skinner. For a cumulative time of about seven months during the summers of 1967, 1968, 1969, and 1971 I studied the stratigraphy, mapped, and collected additional fossils. The part of this report dealing with geology is based on field work done during these four summers. Paleontology is based on fos- sils collected since 1957. The mapping (Figure 19) was done both on aerial stereophotographs and U.S. Geological Sur- vey 7i/?-minute topographic maps. Stratigraphie sections were measured with a hand level, correc- tions for dip being made where necessary. Geography CLIMATE.?The Flagstaff Rim area, like most of central Wyoming, has a semi-arid middle-latitude steppe climate. It has an annual precipitation of from 10 to 20 inches, of which more than half nor- mally falls during the months of April through July as heavy scattered thunderstorms, not infre- quently accompanied by hail. With the exception of these scattered rains, the days of the summer months are usually hot, dry, and clear. Because of the dry climate, all of the streams of the mapped area are intermittent. DRAINAGE AND TOPOGRAPHY.?The mapped area (Figure 19) is within the North Platte River drain- age system, the river itself flowing northeastward only a few miles to the southeast of the area. The area above Flagstaff Rim is drained southwestward by small tributaries of Eagle Creek, an intermittent stream that flows southeastward to empty into the North Platte River at Alcova, Wyoming. The face of the Flagstaff Rim escarpment and the area be- low the rim are drained by Blue Gulch, Lone Tree Gulch, and Little Lone Tree Gulch, the first two flowing generally eastward to the North Platte River, and the last a tributary of the second. The north-central part of the area is drained by the heads of Willow Creek, which flows northeastward to the North Platte River. Maximum relief of the mapped area is about 1300 feet. The altitude above mean sea level ranges from a minimum of 5480 feet at the east- central side of the mapped area to a maximum of 6781 feet at the northeast end of Flagstaff Rim. The main positive features of the area are Flag- staff Rim, Flat Top, and Clarkson Hill. The first is a conspicuous topographic rim, the crest of which rises northeastward from about 6400 feet in the southwest part of the area to 6781 feet above sea level at its highest point. The plateau above the rim is a slightly dissected plain sloping gently southwestward. The escarpment below the rim is steep with the Oligoc?ne rocks of the lower part dissected into well-developed badland topography. Flat Top is a large hill to the east of the rim. The flat top of this hill slopes gently southwestward from a maximum elevation of 6536 feet at the northeast end. Along the southeast and north sides of this hill are small areas of hummocky landslide topography below concave scars, some of which are now completely covered with vegetation, and oth- ers, of more recent origin, showing only bare rock (see Figure 2). Clarkson Hill, at the northeast corner of the mapped area, also has a rather flat top with a maximum altitude of about 6180 feet. The northwest side of this hill slopes gently into the surrounding terrain but the southeast side is steep with badland topography developed in Cretaceous, Paleocene, and Eocene rocks. Stratigraphy GENERAL FEATURES The rocks of primary consideration in this study are of early Oligoc?ne (Chadronian) age. However, within the mapped area, rocks range in age from Permian to Quaternary, and within a short distance outside the mapped area rocks as old as Precam- brian are exposed. Chadronian rocks unconform- ably overlie all of the rock units ranging in age from Precambrian to Eocene and are unconform- ably overlain by post-Chadronian rocks. The pre- Oligocene rocks, particularly the Precambrian, were sources of the locally derived clastic portion of the Chadronian rocks. Aside from this relation- ship and the physical relationships between the pre-Oligocene and Oligoc?ne rocks, the older rocks are not directly relative to this study and will therefore be only briefly treated. PRECAMBRIAN Gneiss, schist, granite, and black dike rocks of Precambrian age crop out a few miles to the south- SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY west of the mapped area. The area of outcrop of these rocks is now relatively small, being mainly hills and knobs of crystalline rocks of the former Granite Mountains protruding through the blan- keting later Tertiary rocks. During Eocene and early Oligoc?ne times, much larger areas of Pre- cambrian rocks were exposed and these areas were apparently at much higher elevations relative to the depositional sites of Eocene and early Oligoc?ne rocks. Large volumes of Precambrian rock were eroded from the Granite Mountains to provide elastics for the greater proportion of the early Eo- cene rocks of the southeast part of the Wind River Basin, which are predominantly coarse arkosic sandstones and conglomerates. By early Oligoc?ne times the Granite Mountains had been worn down to much lower relief but were still contributing a significant amount of detritus to the locally derived clastic portion of the rocks of early Oligoc?ne age. PALEOZOIC SEDIMENTARY ROCKS Paleozoic rocks of Cambrian, Mississippian, Pennsylvanian, and Permian age are exposed either in the mapped area or immediately to the south. These rocks dip generally northeastward on the southwest flank of the Wind River Basin. Because earliest Eocene deposits of the area contain Pre- cambrian rocks, it is safe to conclude that the en- tire sequence of Paleozoic rocks had already been exposed by that time. The coarse early Eocene elas- tics also contain cobbles and, in some places, boulders of Paleozoic rocks, particularly of the more resistant rock types such as the Cambrian Flathead Sandstone and the Pennsylvanian Ten- sleep Sandstone. There are numerous published studies of the Paleozoic rocks of central Wyoming. For more detail the comprehensive review of the Paleozoic formations of the Wind River Basin by Keefer and Van Lieu (1966) is available and in- cludes in its list of references many other works that may be consulted. MESOZOIC SEDIMENTARY ROCKS Mesozoic rocks of the mapped area, like those of the Paleozoic, dip generally northeastward to- ward the axis of the Wind River Basin. Rocks of the Triassic, Jurassic, and Cretaceous systems are present. Within the mapped area, each of the for- mations of Mesozoic rocks is unconformably over- lain by early Oligoc?ne rocks. At the end of the Mesozoic the Wind River structural basin began to form with a slight downfolding. A small angu- lar discordance between strata of the Cretaceous Lance Formation and Paleocene Fort Union For- mation is a manifestation of this tectonic event. A much more spectacular tectonic event occurred at the end of the Paleocene or beginning of Eocene time; Paleocene and Mesozoic rocks were planed off and overlapped by early Eocene rocks, part of which were reworked from the Paleocene and Meso- zoic rocks themselves. Cobbles of red Triassic sandstone and distinctive Cretaceous Cloverly Con- glomerate are found in coarser units of the early Eocene rocks. Keefer (1965) discussed the later Cretaceous rocks of the Wind River Basin, and his reference list includes many other papers that discuss the Mesozoic rocks of the southeast end of the Wind River Basin. Tertiary System PALEOCENE SERIES Paleocene rocks of the Fort Union Formation crop out in the mapped area only in a small area near the base of the east side of Clarkson Hill, near the axis of the Wind River structural basin. The Fort Union Formation is separated from the underlying Cretaceous Lance Formation by a slight angular unconformity of a few degrees and from the overlying Eocene rocks by a somewhat greater angular unconformity. The Fort Union Formation consists of many thin and discontinuous beds of dark brown iron-stained sandstone, which are resistant relative to the thin to thick interbeds of gray to almost white siltstone and fine sandstone. Some of the coarser sandstone beds are arkosic and occasionally conglomeratic, containing fragments of reworked older strata. Fragments of siliceous Mowry Shale and chert peb- bles like those of the characteristic Cloverly Con- glomerate suggest that erosion of the margins of the basin had cut at least through the entire se- quence of Cretaceous rocks by medial Paleocene time. EOCENE SERIES The Eocene rocks of the area warrant more dis- NUMBER 18 cussion than earlier rocks because in some places rocks of Eocene age can be separated only with difficulty from the early Oligoc?ne rocks, which are of primary interest in this report. Indian Meadows Formation The oldest unit of Eocene age within the area is exposed on the east side of Clarkson Hill near the base of the escarpment. This unit unconform- ably overlies the Cretaceous Lance and Paleocene Fort Union formations and is separated from the overlying Wind River Formation by at least an erosional unconformity and perhaps by a slight angular discordance. The outcrop area is of lim- ited areal extent, and there are apparently no other outcrops of this unit in the southeast end of the Wind River Basin. The unit is composed pri- marily of coarse angular grains of quartz and light- colored feldspar with the interstices filled with white to yellowish-gray clay. The upper part of the unit has discontinuous beds of pebbles and cobbles up to a foot in diameter in a coarse arkosic sandstone matrix. Throughout the sequence are layers, up to 5 feet thick, of dark carbonaceous siltstone and clay. The thickness of the unit is from 0 to 120 feet. No fossils are known from this unit, so the age can be determined only by stratigraphie position. Rich (1962:487-488) referred to this unit as the "conglomeratic sandstone unit" of the Wind River Formation and mapped it with this formation. He concluded, however, that it was probably equiva- lent in age to the Indian Meadows Formation since it is unconformably overlain by beds of known early Eocene age and unconformably overlies the Fort Union Formation of Paleocene age. Rich (1962:487-488) interpreted this earliest Eocene unit as representing local deposition along a stream and described a measured section of it. Keefer (1965:37) noted Rich's observations and on his map (pi. 1 of his report) showed the In- dian Meadows Formation extending into the Clark- son Hill area. Love (1970:43) also concluded that the unit in question is the Indian Meadows Forma- tion and reported that it thickens to more than 6000 feet in subsurface sections to the northwest. On his map (Love, 1970, pi. 1), he shows the Indian Meadows Formation in the Clarkson Hill area. I have followed these authors in mapping the unit as the Indian Meadows Formation. It should be recognized, however, that this unit may or may not have been originally continuous with the Indian Meadows Formation at its type area in the northwest part of the Wind River Basin. The assignment is made only because the unit is be- lieved to be a correlative of the Indian Meadows Formation and, as no fossils have been found in the unit at Clarkson Hill, the correlation is based only on physical stratigraphie and structural re- lationships with enclosing rocks at the Clarkson Hill locality. Wind River Formation Rich (1962) divided the Wind River Formation of the southeast part of the Wind River Basin into two units?a lower fine-grained faci?s and an upper coarse-grained faci?s. The lower fine-grained facies does not extend into the extreme southeast end of the basin, and none is shown on Rich's map of the Clarkson Hill area (1962, pi. 7). The lower fine- grained facies, where present, has produced fossils that indicate approximate temporal equivalence to the Lost Cabin Member of the Wind River Forma- tion in the Badwater area of the northeast part of the basin (Rich, 1962:493). The unit that Rich (1962) termed the upper coarse-grained facies over- lies with erosional unconformity the fine-grained facies. The upper coarse-grained facies is therefore younger than any part of the Wind River Forma- tion in the Badwater area where the Lost Cabin Member is the upper unit. Rich (1962:496) concluded that the upper coarse- grained facies is of early Eocene age, however, be- cause of its similarity to a unit in the Gas Hills area farther to the west which is overlain by mid- dle to late Eocene pyroclastic rocks. Keefer (1965: 53) believed that the most compelling evidence for an early Eocene age for the upper coarse-grained facies of Rich was the absence in this unit of vol- canic debris, whereas upper Eocene and lower Oligoc?ne rocks in nearby areas contain abundant volcanic debris. Rich (1962:495) described a generalized section of the upper coarse-grained facies. I have mapped it (Figure 19) as Wind River Formation. Unnamed Boulder Conglomerate The latest unit of probable Eocene age within 8 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY the present map area is directly relevant to this study because on top of Clarkson Hill, this unit was mapped by Rich (1962, pi. 7) as White River Formation and considered by him to be early Oli- goc?ne in age. Denson and Harshman later (1969) mapped both the top of Clarkson Hill and the top of Flat Top, four miles to the southwest, as Wind River Formation. Love (1970, pi. 1) fol- lowed Rich in mapping the top of Clarkson Hill as White River Formation, but mapped what is apparently the same unit on top of Flat Top as Wind River Formation. Obviously, some clarifica- tion is needed. The unit in question unconform- ably overlies the upper coarse-grained faci?s of the Wind River Formation on Clarkson Hill and un- conformably overlies Cretaceous strata on Flat Top. It can be demonstrated that the unit in question is separated from early Oligoc?ne fossil- bearing strata by an erosional unconformity with several hundred feet of relief. The unit should not be assigned to either the Wind River or White River Formation. As already noted, this boulder conglomerate unit covers the top of Clarkson Hill in the northeast part of the map area (Figure 19), the top of Flat Top 4 miles southwest, and laps onto Cretaceous and older strata even farther southwest. What are believed to be remnants of the same unit are pres- ent at various localities from 10 to 15 miles south- east of the present map area, preserved on the tops of rather flat-topped spurs extending out from the base of the Oligoc?ne and Miocene escarpments between the head tributaries of Ledge Creek, Bear Creek, Bolten Creek, and the southwest tributaries of Stinking Creek. These are mapped as Wind River Formation and/or Wagon Bed Formation by Denson and Harshman (1969). In the present map area the unnamed unit is up FIGURE 2.?Boulder conglomerate of medial (?) or late (?) Eocene age unconformably overlying Cretaceous Frontier Formation at the south end of Flat Top. The stratification of the boulder conglomerate is more apparent in this perspective than when on the outcrop. The outcrop is a slump scar, with the slump debris spread out below and to the right of the scar. Telephoto view looking west-northwest from highway U.S. 220, about 2 miles northeast of Alcova, Wyoming. NUMBER 18 to 250 feet thick and is characterized by thick beds of boulder conglomerate in a coarse arkosic sand- stone matrix separated by thick to thin beds of coarse arkosic sandstone and occasional thinner beds of red or grayish green siltstone and clay- stone. On Clarkson Hill boulders 6 feet in diam- eter are common, while on Flat Top, nearer the Granite Mountain source area, boulders of 20 feet in diameter are not uncommon and rarely are 25 feet or more in the largest dimension. The boul- ders are predominantly of light-colored granitic rock, with lesser amounts of gneiss, schist, dark intrusive igneous rocks, and bright green Precam- brian quartzite and Paleozoic and Mesozoic sand- stones and quartzites. The individual boulder beds vary in thickness laterally but are for the most part quite continuous over fairly large areas. The beds of finer ma- terial, on the other hand, are most often lenticular and can be traced for only short distances along the outcrops. The following section is representative of the boulder conglomerate unit. Section of Unnamed Boulder Conglomerate of ? medial or late Eocene age, north side of Flat Top, S i/2 of Sect. 23, and N i/2 of Sect. 36, T 31 N, R 83 W., Natrona Co., Wyoming. (Strata dip ap- proximately 3? southwest; Unit 1 is oldest.) Top of section at present land surface at approximately 6500 feet above sea level, about one-fourth mile west of VABM Kendricks bench mark Unit Feet 14. Conglomerate, massive, poorly sorted, with boulders and cobbles of Precambrian and Paleozoic rock up to 25 feet in diameter, matrix of coarse angular grains of quartz and feldspar; partly covered by vegetation so beds of finer material may be present but not recognized .105 ' : / , m FIGURE 3.?Boulder conglomerate of (?) medial or (?) late Eocene age unconformably overlying Cretaceous Frontier Formation at the south end of Flat Top, in the south side of the NE 14, Sect. 2, T 30 N, R 83 W, Natrona County, Wyoming. Same outcrop as shown in Figure 2. Man at left center gives approximate scale. 10 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 13. 12. 11. 10. Sandstone, white to yellowish white, coarse grained, arkosic 12 Sandy claystone, brownish yellow with bands and lenses of pale green and red, poorly consolidated 5 Sandstone, yellowish white to white with resistant dark brown concretionary lenses, coarse grained, arkosic ... 8 Conglomerate, poorly sorted, with cobbles and boulders of Precambrian and Paleozoic rock up to 6 feet in diameter; covered by vegetation so the character of the matrix and whether or not units of finer material are present cannot be determined 48 Sandstone, pale brown to grayish green, coarse grained, poorly bedded, arkosic 12 Conglomerate, massive, poorly sorted, with cobbles and boulders of Precambrian and Paleozoic rock up to 2 feet in diameter; matrix of coarse-grained, pale yellowish brown, arkosic sandstone; granitic boulders weathered 4 to 6 inches deep 8 Sandstone, pale yellowish brown to white, relatively fine grained, bedded; interstices filled with pale green sandy clay 5 Conglomerate, massive, poorly sorted, with cobbles and boulders of Precambrian and Paleozoic rock up to 10 feet in diameter; matrix of coarse-grained pale greenish gray to yellowish brown arkosic sandstone; granitic boulders not so deeply weathered as in lower units ... 11 5. Conglomerate, massive, poorly sorted, with boulders up to 3 feet in diameter; matrix of coarse-grained yellowish brown arkosic sandstone; granite boulders weathered as in Unit 1 13 4. Sandstone, brown to grayish green, friable, medium tex- ture, relatively well sorted 11 3. Conglomerate, massive, poorly sorted, with cobbles and boulders of Precambrian and Paleozoic rock up to 5 feet in diameter; matrix of coarse-grained brown arkosic sandstone; granitic boulders weathered as in Unit 1 ...18 2. Sandstone, gray to pale yellowish white, medium texture, relatively well sorted and bedded, arkosic 2 1. Conglomerate, massive, poorly sorted, with cobbles and boulders of Precambrian and Paleozoic rock up to 6 feet in diameter; matrix of coarse-grained yellowish brown arkosic sandstone; smaller granitic boulders weathered so that they can be easily disintegrated with geology pick; sandstone and quartzite boulders solid and relatively unweathered 11 Total thickness of unnamed boulder conglomerate 269 Angular unconformity at contact Cody Shale Formation FIGURE 4.?Lowest unit of boulder conglomerate of medial (?) or late (?) Eocene age at south end of Flat Top, in south side of NE 14, Sect. 2, T 30 N, R 83 W, Natrona County, Wyoming. Same outcrop as Figures 1 and 2. Shows the deep weathering of the boulders of crystalline rock; note particularly the concentric weathering of the darker colored boulders of more basic composition. NUMBER 18 11 Sections were also measured on the southeast side, south end, and west side of Flat Top. The section on the southeast side is quite similar to the section just described, but distributed throughout the section are several beds from 2 to 4 feet thick of fine siltstone and claystone that are pale green or purplish red with spots of limonite yellow. At the south end of Flat Top only the lower 140 feet of the section is present but the upper 30 feet of this is pale green to white coarse-grained arkosic sandstone with several bands of light green and two bands of red siltstone and claystone. The upper red band has bright green spots and the lower red band has spots of deep purple and limonite yellow. The section on the west side of Flat Top is gen- erally similar to that of the north end described above. On Clarkson Hill no sections were measured, but the sequence seems to be much like it is on Flat Top except the boulders are generally not so large. The unit here approaches 200 feet in thickness. No fossils were found within the boulder con- glomerate unit, so its exact age cannot be deter- mined. On Clarkson Hill, just north of the North Granite Mountain fault zone, the boulder conglom- erate unconformably overlies the upper coarse- grained faci?s of the Wind River Formation (Rich, 1962:497), so the maximum possible age is late early Eocene. Rich (1962) considered the boulder conglomer- ate on Clarkson Hill to be the basal conglomerate of the White River Formation and early Oligoc?ne FIGURE 5.?Contact of Cretaceous Cody Shale (Kc) and Oligoc?ne White River Formation (Twr) in Little Lone Tree Gulch drainage, in SW 14. SW 14, Sect. 18, T 31 N, R 82 W, Natrona County, Wyoming. The basal conglomeratic unit of the White River Formation varies from practically none to a few feet of cobbles of Precambrian crystalline rock and Paleozoic and Mesozoic sandstones in a coarse arkosic sandstone matrix. Directly overlying the conglomerate are silty bentonitic claystones more typical of the lower part of the White River Formation. 12 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY in age. This can be shown not to be the case. He stated (1962:497) that "the lower 12 to 50 feet of the White River Formation is a massive to poorly bedded conglomerate with granite boulders as much as 20 feet in diameter . . ." and described the composition and rock types of the conglomerate. In his section of the White River Formation (1962: 500) Rich described 12 feet of basal conglomerate. It is curious, then, that, after describing the basal conglomerate as 12 feet in thickness and writing that it varied from 12 to 50 feet, he should map more than 200 feet of boulder conglomerate on top of Clarkson Hill as White River Formation and apparently consider it all to be basal con- glomerate. Less than a half mile south of Clarkson Hill, in Sect. 18, T 31 N, R 82 W, the finer grained sand- stones, siltstones, and bentonitic claystones contain- ing early Oligoc?ne mammals are in places lying directly on the upturned edges of Cretaceous strata, and in other areas there is only a thin layer of pebbles or cobbles at the interface (Figures 5 and 6). The contact can be traced over a considerable area just to the south of Clarkson Hill and the basal layer of cobbles and boulders rarely exceeds 10 feet in thickness. However, the thicker boulder conglomerate on Clarkson Hill is northeast of the North Granite Mountain fault zone and the area to the south, with but little Oligoc?ne basal con- glomerate, is southwest of the fault zone. It is evident, then, that after deposition of the thick boulder conglomerate, the area southwest of the fault zone was moved upward relative to the northeast side and the boulder conglomerate was stripped from most of the southwest block before FIGURE 6.?Lower conglomeratic unit of the White River Formation (Twr) overlying Cretaceous Cody Shale (Kc) in Little Lone Tree Gulch drainage, in SE 14, SE 14, Sect. 13, T 31 N, R, 83 W, Natrona County, Wyoming. The basal conglomeratic unit here is up to 15 feet of cobbles of Precambrian crystalline rock and Paleozoic and Mesozoic sandstones in a coarse arkosic sandstone matrix which is locally cemented and forms ledges. This view is typical of the lower conglom- eratic unit of the White River Formation over much of the Little Lone Tree Gulch drainage. The conglomeratic unit is overlain by silty bentonitic claystones of the "lower banded zone." NUMBER 18 13 deposition of the more typical finer grained early Oligoc?ne rocks. The thin basal conglomerate of the White River Formation southwest of the fault zone is probably primarily reworked from the earlier boulder con- glomerate. It contains a higher percentage of cob- bles and boulders of Paleozoic and Mesozoic sandstones and pale to bright green Precambrian quartzite, but this would be expected. If the boulder conglomerate were eroded, the deeply weathered granite boulders would tend to disinte- grate or be reduced in size, whereas the more re- sistant sandstones and quartzites would remain. That an erosion cycle separates the unnamed boulder conglomerate from the beginning of White River deposition can also be demonstrated by ob- servations on the northwest side of Flat Top. In Sect. 35, T 31 N, R 83 W, the more typical finer grained tuffaceous siltstones and claystones of the White River Formation lap onto the eroded edges of the unnamed boulder conglomerate. In the same area the White River Formation overlies slump de- posits containing jumbled masses of Cody Shale and the unnamed boulder conglomerate with both boulder units and finer red and green bands as described above. More recent analogs of these slump blocks can be seen on the north and south- east sides of Flat Top. Some of these slumps are now completely covered by vegetation but others are so recent that only bare outcrops of boulder conglomerate and underlying Cody Shale are ex- posed in the concave scars above the slump blocks (Figure 2). On the northwest side of Flat Top, the basal FIGURE 7.?White River Formation (Twr) lapping against and over Cretaceous Mesaverde Formation (Kmv) and boulder conglomerate (?Tbc) near Little Lone Tree Gulch, in the SE 14, Sect. 13, T 31 N, R 83 W, Natrona County, Wyoming. The boulder conglomerate uncon- formably overlies the Mesaverde Formation and is probably the medial (?) or late (?) Eocene unnamed boulder conglomerate. Contacts located approximately at dashed lines. There are approximately 150 feet of relief on the base of the White River Formation at this locality. View is to northwest. Mesaverde Formation dips about 25 degrees northeastward; White River For- mation is almost horizontal. 14 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY FIGURE 8.?Contact of White River Formation (Twr) and Cretaceous Cody Shale (Kc) in East Fork of Blue Gulch, west side of Flat Top. Contact approximately at dashed line. Basal unit of White River Formation is here a thin limestone that has within it occasional cobbles and pebbles of Precambrian crystal- line rock and Paleozoic and Meso- zoic sandstones. Slope in right foreground is Cody Shale; the cob- bles on the surface are probably partly derived from the basal unit of the White River Formation and partly from the medial (?) or late (?) Eocene boulder conglomerate which overlies Cody Shale farther upslope to right of photograph. FIGURE 9.?Relationships of the White River Formation (Twr) and underlying units in the South Fork of Lone Tree Gulch, near west side of Flat Top, in the NE 14, SE 14, Sect. 26, T 31 N, R 83 W, Natrona County, Wyoming. Basal unit of White River Formation is here a thin white limestone with occas- ional pebbles and cobbles, and is shown here lapping against Cody Shale (Kc). Within a few yards the limestone is overlying the medial (?) or late (?) Eocene boulder conglomerate (Tbc) which is over- lying Cody Shale on the hill in the left part of the photograph. Con- tacts located approximately at dashed lines. NUMBER 18 15 unit of the White River Formation is a white lime- stone up to 6 feet thick with seams of brown chal- cedony and, near the bottom, a few boulders and cobbles. This relatively resistant limestone can be easily traced over the ancient slump deposits and up the west side of Flat Top to where it laps onto the unnamed boulder conglomerate in undisturbed outcrops. The conclusions to be drawn from these ob- servations are that after deposition of the unnamed boulder conglomerate, an erosion cycle produced several hundred feet of relief. The boulder con- glomerate was stripped from part of the area south- west of the North Granite Mountain fault zone, but the northwest side of Flat Top remained an escarpment capped by the unnamed boulder con- glomerate. Relief was abrupt enough so that slumping occurred, probably in much the same manner as it has in recent times on the north and southeast sides of Flat Top. The limestone im- mediately overlying the ancient slump deposits and lapping onto the eroded edges of the unnamed boulder conglomerate probably represents a rela- tively long interval of nondeposition before the beginning of deposition of the finer grained tuffa- ceous siltstones and claystones of the White River Formation. The unnamed boulder conglomerate cannot yet be definitely correlated with any formation of the Beaver Rim area to the west. The Wagon Bed Formation near the northeast end of Beaver Di- vide, just west of the Rattlesnake Hills, about 40 miles west-northwest of the present map area does, however, have a unit, 25 to 50 feet thick, of giant boulders of Precambrian gneissic rock up to 20 feet in diameter (Love, 1970:54-55). This unit and the unnamed boulder conglomerate could both be manifestations of the same tectonic event that ele- vated the core of the Granite Mountains to pro- vide a source area. Within the present map area the finer units within the unnamed boulder conglomerate are neither tuffaceous nor bentonitic, at least not where examined. Along the north flank of the Shirley Mountains, about 25 miles southeast, however, a coarse boulder conglomerate underlying the White River Formation does have tuffaceous beds within it (J. D. Love, personal communication). This boulder conglomerate is very likely the same as that of the present map area, and that exposed along the heads of Ledge Creek, Bolten Creek, Bear Creek, and Stinking Creek, as noted before. The latter four localities are from 10 to 25 miles south- east of the present map area and from 10 to 15 miles north and northeast of the north flank of the Shirley Mountains. If these units are all the same, the presence of tuffaceous beds lends support to a medial or late Eocene age assignment, because early Eocene rocks of this part of Wyoming, where dated by fossils, are not characteristically tuffaceous. The unnamed boulder conglomerate is appar- ently similar in lithology and stratigraphie rela- tions to the Ice Point Conglomerate, described by Love (1970:59-62), about 50 miles west-southwest of the present map area. But the Ice Point Con- glomerate is on the south flank of the Granite Mountains and apparently had the Wind River Range as its source area, whereas the unnamed boulder conglomerate is on the north flank of the Granite Mountains and had the Granite Moun- tains themselves as a source area. With the source areas being different mountain ranges, it is un- likely that both conglomerate units were even re- sults of the same tectonic event. With these considerations in mind they could hardly be con- sidered the same formation. The age of the unnamed boulder conglomerate then cannot be accurately determined. It is younger than late early Eocene and older than early Oligo- c?ne. Its assignment to a previously named forma- tion or recognition as a new formation must await further study. For purposes of this report, the most relevant conclusion regarding the unnamed boulder conglomerate is that it is not part of the White River Formation. OLIGOC?NE SERIES White River Formation DEFINITION.?The White River Formation of the present study area is generally composed of massive fine-grained tuffaceous siltstones. The lower part of the section has more variegated red and gray- green claystones and lenses of coarse arkosic sand- stone and conglomerate than the upper part. The massive tuffaceous siltstones that make up the bulk of the deposits have interbedded thin lenses of claystone, occasional coarse channel sandstones, 16 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY FIGURE 10.?Panorama of Flagstaff Rim viewed from Flat Top, looking southwest (at left) to northwest (at right), showing outcrops of White River Formation. Letter symbols indicate positions of volcanic ash beds F and G. Contact of White River Formation and overlying Split Rock Formation is at abrupt change in vegetation in upper part of escarpment. Boulders and cobbles in foreground are part of underlying medial (?) or late (?) Eocene boulder conglom- erate. Ferris Mountains can be seen in the distance at left. NUMBER 18 17 FIGURE 11.?Outcrops of White River Formation in Lone Tree Gulch drainage. View is to south- west with North Fork of Lone Tree Gulch in foreground, Flagstaff Rim at right, and Flat Top in left distance. Letter symbols indicate volcanic ash beds discussed in text. 18 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY and distinct and easily recognizable beds of nearly pure volcanic ash or vitric tuff. The entire sequence is of Chadronian (early Oligoc?ne) age. It must be understood that the formation here under study is not lithologically identical to and only partly chronologically coincidental with the White River Group of the Great Plains. Meek and Hayden (1862) originally defined the White River Group in the Great Plains of Wyoming, Nebraska, and South Dakota. This was later subdivided into the early Oligoc?ne Chadron Formation and me- dial and late Oligoc?ne Brule Formation by Dar- ton (1899:736). Later, Darton (1908:463) presented evidence that the Oligoc?ne rocks of central Wy- oming, which had previously been assigned to the Sweetwater Group (Hayden, 1871:29; Endlich, 1879:110-112), were originally continuous with the White River Group of the Plains area and sug- gested that the White River nomenclature be ex- tended into the central Wyoming area. A year after Darton's latter report, Granger (1910:238) found early Oligoc?ne mammalian fos- sils along Beaver Rim, about 80 miles to the west of the present map area, and assigned the rocks producing the fossils to the White River Group. Granger noted that these Oligoc?ne deposits were considered part of the Sweetwater Group by Hay- den and Endlich but assigned them to the White River Group, because he felt that they were a west- ward extension of the "Titanotherium beds" of the Bates Hole area considered by Darton to be part of the White River Group (Granger, 1910: 241). Since that time, the Oligoc?ne rocks of central Wyoming have been variously termed the White River Group, White River Formation, Chadron, Chadron and Lower Brule, Brule, and Oreodon beds (Wood, 1948:39). Van Houten (1954, 1964), in stratigraphie studies of the Beaver Rim area, assigned the early Oligoc?ne rocks there to the White River Formation and because of this work, Rich (1962) assigned the early Oligoc?ne rocks of the present study area to the White River Forma- tion also. The early Oligoc?ne rocks of the area of this report cannot now be traced continuously into those of the Beaver Divide area, 50 to 80 miles to the west. They may, however, be continuous in the subsuriace south of the Rattlesnake Range and were almost certainly originally continuous north of the Rattlesnake Range and probably were con- tinuous over much of the Wind River and Powder River Basins. Oligoc?ne rocks can be traced from South Da- kota and northwestern Nebraska almost continu- ously to the area near Douglas, Wyoming. The generalization can be made that the rocks of Cha- dronian age become progressively coarser and more tuffaceous westward. West and southwest of Doug- las, Wyoming, the Oligoc?ne deposits extend far up some of the mountain valleys of the Laramie Range to the south, to within 25 miles of similar deposits on the southwest side of the Laramie Range. These latter are continuous with those of the Bates Hole area and the area of present study. The claystones and tuffaceous siltstones of the present study area were probably originally con- tinuous with the generally finer sediments of the Great Plains area. Because the Chadron Forma- tion cannot be recognized in the present study area, I have followed Van Houten (1954, 1964), Rich (1962), Love (1970), and others in using the term White River Formation for the Chadronian deposits here. DISTRIBUTION AND THICKNESS.?The White River Formation within the mapped area (Figure 19) is, exposed mainly in the lower part of the Flagstaff Rim escarpment where it is dissected into badland topography in the heads of Blue Gulch, Lone Tree Gulch, and Little Lone Tree Gulch (Figure 10). The lower part of the section is exposed farther to the northeast along Little Lone Tree Gulch as far as the southwest side of Clarkson Hill, where it ends at the North Granite Mountain fault zone. Outside the present map area the formation can be traced in somewhat discontinuous outcrops for several miles to the northwest on the south side of the North Granite Mountain fault zone. A few miles south of the present map area the formation is exposed in Benton Basin. From 10 to 15 miles to the southeast similar rocks are exposed near the heads of Ledge Creek, Bol ten Creek, Bear Creek, and Stinking Creek. The maximum thickness of the White River For- mation in the map area (Figure 19) is about 800 feet in Little Lone Tree Gulch. The original thick- ness cannot be determined, because the upper sur- face is an erosional disconformity. The formation NUMBER 18 19 thins southwestward along Flagstaff Rim. The thin- ning is due partly to thinning of individual beds within it, but, at least in the lower part of the section, some of the beds apparently completely wedge out southwestward. At the beginning of White River deposition the land surface was quite irregular, with at least 300 to 400 feet of relief. The area to the northwest of Flat Top was apparently a broad valley, with the west side of Flat Top forming the east side of the valley. The northwest side of the paleovalley can- not be determined but the valley must have been several miles wide. There is insufficient informa- tion to determine whether the land surface as a whole was predominantly upland with occasional valleys or predominantly lowland with occasional positive features. White River sedimentation be- gan first in the bottoms of the valleys with deposi- tion of claystone and stream channel sandstones with tongues of cobble and boulder conglomerate derived primarily from the ? medial or late Eocene unnamed boulder conglomerate that was being eroded from the tops of the adjacent positive areas. As deposition continued, the sediments became more tuffaceous and lapped farther up onto the sides of the valleys. Before the end of Chadronian time, the older positive features such as Flat Top had been completely covered and deposition of tuf- faceous siltstones was on a broad flat plain. GENERAL FEATURES.?The White River Forma- tion of the Flagstaff Rim area can be roughly di- vided into two parts with different lithologies: the lower part of interbedded silty claystone and con- glomeratic sandstone and the upper part predomi- nantly of tuffaceous siltstones. The lower part was deposited in the lower parts of valleys and is found only where the section is thickest. The upper tuf- faceous siltstones are conformable on the lower part, the upward change in lithology being grada- tional and not necessarily at the same stratigraphie level in different areas. The lower part of the section, the part deposited in the lower part of the preexisting valleys, is best exposed in Little Lone Tree Gulch but can also be studied in Lone Tree Gulch. It is characterized by units of bentonitic or montmorillinitic claystone of variegated pale green and pale to bright red color, separated by tongues of conglomeratic sand- stone. The finer claystone units have within them occasional lenses of coarse arkosic channel sand- stone that is frequently crossbedded. The con- glomeratic sandstone tongues thicken and become coarser laterally so that near the edges of the pre- existing valley in which they were deposited, they become a very coarse conglomerate with cobbles and boulders up to 10 or 12 feet in diameter. These large clasts were undoubtedly reworked from the unnamed boulder conglomerate of ? medial or late Eocene age which was exposed only a short distance to the south at a higher elevation. The cobbles and boulders are of similar rock types except that in the White River Formation there is a higher percent- age of Paleozoic sandstones and bright green Pre- cambrian quartzitet This is the expected result of reworking of the unnamed Eocene boulder con- glomerate since the granitic boulders of the Eocene formation are deeply weathered and can be easily disintegrated, while the more resistant sandstone and quartzite boulders are almost completely un- weathered. The upper part of the White River Formation is predominantly massive tuffaceous siltstones of pale gray-green color. Within these siltstones are occasional thin lenses of pale green to brown clay- stone and lenses, up to several feet thick and 50 feet wide, of coarse channel sandstones. The bases of some of these sandstone lenses have a layer of cobbles, up to 6 inches in diameter, of granitic rock and Paleozoic and Mesozoic sandstones and brown to transparent chert and chalcedony peb- bles, usually with a white coating. Within the tuf- faceous siltstone sequence are many distinct beds, normally from 6 inches to several feet in thickness, of nearly pure vitric tuff or volcanic ash. Showers of volcanic debris were apparently frequent and prolonged with some of the debris falling directly onto the depositional surface as tuff and some falling into streams and ponds or onto bordering uplands and being reworked and incorporated into the tuffaceous siltstones. Some of the beds of volcanic ash (vitric tuff) are of only very local extent, but others are continuous over fairly large areas and can be traced for sev- eral miles along the outcrops. These ash beds, par- ticularly the more widespread ones, are very useful as marker beds for precisely describing the strati- graphic level of each of the fossils collected from the area. Radiometrie dates from some of the 20 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY tuffs (Evernden et al., 1964) are also a useful ad- junct to the fossils in correlating this formation with other rock units in other areas. Dates obtained by Evernden et al., are shown on Figure 16. Rich (1962:498) in his description of the White River Formation observed that individual beds within the White River Formation are lenticular and could be traced only short distances along the strike. This is true of the stream channel sand- stones and conglomerates and also of some of the thin claystone beds, but these together make up only a minor part of the formation. Most of the beds of volcanic ash or vitric tuff, however, can be traced for several miles along the outcrop and could hardly be considered lenticular. The geometry of the ash beds indicates that there was very little relief over most of the depositional surface. The massive nature of the tuffaceous siltstones makes it impossible even to determine a single bed, much less trace it along the strike; but the included ash beds support the inference that deposition of these tuffaceous siltstones was also over broad areas rather than in lenses. Individual concretionary or nodular layers within the siltstones are not con- tinuous over large areas but these probably do not define individual beds but are rather a result of some diagenetic or other postdepositional process. The basal unit of the White River Formation is not everywhere the same. In the lowest areas, in Little Lone Tree Gulch, where deposition began, the basal conglomerate is usually from 10 to 12 feet thick with cobbles and boulders up to 2 feet or more in diameter. In Lone Tree Gulch, along the northwest flank of Flat Top, the basal conglom- erate occasionally has boulders up to 12 feet in diameter and these are in most cases the direct re- sult of slumping of the unnamed boulder conglom- erate. Along the west side of Flat Top the basal conglomerate is thinner, usually less than 4 feet in thickness, with cobbles and boulders up to 2 feet in diameter. As previously noted, in the South Fork of Lone Tree Gulch and the East Fork of Blue Gulch, the basal unit of the White River Forma- tion is a white limestone up to 6 feet in thickness, with seams of brown chalcedony. Within the lime- stone are occasional boulders and cobbles. South- west of this area, in the other parts of the Blue Gulch drainage, the basal conglomerate is even thinner and in some places the tuffaceous siltstones of the upper part of the White River Formation are directly overlying Mesozoic rocks, the only basal conglomerate being reworked pieces of the im- mediately underlying formation. Because the basal conglomerate contains cobbles and boulders of granitic rock only in areas that are down slope from positive areas covered with the ? medial or late Eocene unnamed boulder conglom- erate, it seems likely that these large clasts in the base of the White River Formation were derived from the preexisting boulder conglomerate. If the basal conglomerate of the White River Formation was derived directly from Precambrian outcrops, it should be thicker and coarser toward the Precam- brian source. The reverse is true; in Blue Gulch, the basal conglomerate becomes finer and thinner southwestward. The upper contact of the White River Forma- tion is an erosional unconformity. Along Flagstaff Rim this contact is almost parallel to the White River strata, but in the northwest part of the map area (Figure 19), southeast of Ryan Hill, the over- lying (?) early Miocene rocks fill a broad channel cut into the White River Formation. This was re- ported by Rich (1962:503) and supported by my own observations. Within the White River Formation, the only physical evidences of breaks in deposition are the local channel cuts and fills. Channels up to 20 feet or more in depth were cut into the siltstones and subsequently refilled, usually with coarse, often crossbedded, sandstone at the base and green to brown claystone above. These cycles of cutting and filling were apparently short-lived, because lateral to the channels the tuffaceous siltstones are appar- ently continuous with no evidence that deposition was interrupted. STRATIGRAPHIC SECTIONS.?It is impossible to de- scribe every detail of vertical and horizontal varia- tion of the White River Formation of the present study area. Following are the descriptions of sec- tions in four different areas which provide details of each particular area and illustrate changes from one area to the next. North Fork of Lone Tree Gulch Section: Within the present study area (Figure 19), the North Fork of Lone Tree Gulch is the locality where the sec- tion of White River Formation approaches its maximum thickness and can be measured in con- NUMBER 18 21 -6- ?F- -A- m? FIGURE 12.?White River Formation in North Fork of Lone Tree Gulch. View northwestward up North Fork of Lone Tree Gulch to northeast end of Flagstaff Rim. Letter symbols indicate volcanic ash beds discussed in text. In lower part of this view are variegated bentonitic clay- stones of the upper part of the "upper banded zone" which grade upward into tuffaceous siltstones. Contact of White River Formation and overlying Split Rock Formation is at abrupt break in vegetation in upper part of escarpment. tinuous outcrop in the shortest horizontal distance. The section described below was measured along approximately the same route as a generalized sec- tion measured by Skinner (1957) and used for zona- tion of all the fossils subsequently collected from this area. Skinner's section did not include some of the basal beds so that marker beds used for zonation will not necessarily correspond in footage to those on the section described below. An ab- stracted version of Skinner's generalized zonation section is reproduced later in this report as Figure 16. Section of White River Formation in the North Fork of Lone Tree Gulch, Sy2, Sect. 23, and E]/2, Sect. 22, T 31 N, R 83 W, Natrona County, Wy- oming. (Section starts near the junction of the North and South Forks of Lone Tree Gulch and continues up the North Fork to the northeast end of Flagstaff Rim. Strata dip approximately 3? west-southwest. Unit 1 is oldest.) PEarly Miocene Split Rock Formation Erosional unconformity White River Formation Unit Feet 36. Siltstone, tuffaceous, as in Unit 24 22 35. Tuff, pale gray to white, vitric, with biotite and hard rust-colored spots. Ash J of generalized section used for zonation of fossils 5 34. Tuffaceous siltstone, as in Unit 24 8 33. Tuff, pale gray to white, vitric, with biotite crystals .... 2 32. Tuffaceous siltstone, as in Unit 24 47 31. Tuff, gray, vitric, with biotite and hard rust-colored spots. Ash I of generalized zonation section 1 22 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 30. Siltstone, tuffaceous, as in Unit 24 22 29. Tuff, bright bluish white, very glassy with biotite crystals and hard rust-colored spots. Ash H of general- ized zonation section 2 28. Tuffaceous siltstone, as in Unit 24 55 27. Tuff, dark to black, with biotite crystals 1 26. Siltstone, tuffaceous, as in Unit 24 33 25. Tuff, pale bluish white, very glassy with many biotite crystals 2 24. Siltstone, pale greenish gray to white, very tuffaceous, - with few concretionary bands that weather to reddish brown. Lower 12 to 15 feet of this unit weather to a vertical face 62 23. Siltstone, as in Unit 21 33 22. Tuff, silvery gray to white, distinct, vitric, with biotite crystals. Ash G of generalized zonation section 4 21. Siltstone, pale greenish gray, tuffaceous, few nodular layers that weather to reddish brown surface 57 20. Tuff, pale silvery gray to white, vitric, with biotite crystals. Ash F of generalized zonation section 1 19. Siltstone, as in Unit 17 72 18. Tuff, dark gray to black, vitric, with biotite and quartz grains. Ash D of generalized zonation section 2 17. Siltstone, greenish gray, massive, with thin nodular bands that weather to a reddish brown surface. Has occasional thin lenses of coarse channel sandstone ...43 16, Claystone, darker grayish green, silty in some places. Weathers to rounded slopes with crumbly deeply weathered surface. This unit grades upward to the next unit 15 15. Siltstone, as in Unit 13. The top of this unit is more resistant than the overlying unit and usually forms a bench 12 14. Tuff, similar to Unit 12 1 13. Siltstone, pale greenish gray, tuffaceous, with harder concretionary bands that have a reddish brown weather- ed surface 11 12. Tuff, bright silvery gray to white, vitric, with many biotite grains 2 11. Siltstone, pale greenish gray to white, with reddish brown weathering concretionary bands, tuffaceous 44 10. Tuff, bright silvery gray to white, vitric, with biotite grains up to 1 mm or larger. Ash B of generalized zona- tion section 3 9. Siltstone, pale gray-green, tuffaceous, with harder con- cretionary bands that weather to a rich brown surface. Upper 5 feet of this unit is continuous concretionary zone that usually has vertical face. Upper part of unit has occasional lenses of coarse channel sandstone 11 8. Tuff, vitric, dark gray to black with hard rust-colored spots 1 7. Siltstone, pale greenish gray, tuffaceous, with harder nodular bands that weather to a rich brown surface. Upper part of unit has occasional lenses of coarse channel sandstone 15 6, Tuff, white, vitric, with small biotite grains: upper and lower contacts not sharp but mixed with overlying and underlying siltstones. Ash A of generalized zonation section 1 5. Siltstone, pale greenish gray, tuffaceous, with hard nod- ular bands several inches thick that weather to a rich brown color, but when broken are the same color as the surrounding sediment 42 4. Claystone, pale gray-green and pale to bright red banded and mottled in bottom part; changes upward to brown color and finally to pale gray-green siltstones near top 55 3. Conglomeratic sandstone and claystone. Sandstone pale yellow to brown, poorly sorted, arkosic, coarser upward with occasional cobbles and boulders up to 1 foot in diameter in upper part of unit; lower part of unit with lenses of green, brown, and red claystone 25 2. Claystone and sandstone; claystone variegated brown, green, and red, with more red bands near top of unit, massive; within claystone are lenses of coarse pale yellow to brown arkosic sandstone with occasional pebbles and cobbles. Unit grades upward 45 1. Conglomerate, massive, poorly sorted with cobbles and boulders of granitic rock, quartzite, and sandstone up to 6 feet in diameter in a matrix of coarse arkosic sand- stone 18 Total thickness of White River Formation 775 Angular unconformity Cretaceous Cody Shale Formation Blue Gulch Section: Southwestward from the North Fork of Lone Tree Gulch the section of White River Formation becomes thinner. Most of the thinning occurs in the lower part of the sec- tion. The upper part of the section, equivalent to the upper 350 feet of the North Fork of Lone Tree Gulch Section, remains about the same thickness southwestward into the Blue Gulch drainage but becomes progressively more tuffaceous with fewer clay lenses. Ash B (Unit 10 of the North Fork of Lone Tree Gulch Section) can be traced into the East Fork of Blue Gulch but cannot be traced farther south- westward into the other parts of the Blue Gulch drainage. Ash F (Unit 20 of the North Fork of Lone Tree Gulch Section) is the lowest ash bed that can be traced continuously from Lone Tree Gulch through most of the Blue Gulch drainage. Some of the channel deposits below ash F in the Blue Gulch area contain relatively pure but local deposits of white vitric tuff, but these cannot be definitely correlated with any of the tuff beds of the Lone Tree Gulch Section described above. In the North and Trail Forks of Blue Gulch, the tuffaceous siltstones of the interval from about 20 feet below ash G to 100 feet above ash G have a pale pink color. This color is not like the red claystones of the lower part of the section in Lone NUMBER 18 23 Tree Gulch and Little Lone Tree Gulch. The pink color may be due to elastics derived from the bright red upper Paleozoic and Triassic rocks only a short distance to the south. Section of White River Formation in the Trail Fork of Blue Gulch, in the N i/2, Sect. 34, and SW y4, Sect. 27, T 31 N,R 83 W, Natrona County, Wyoming. Strata dip generally almost due west. Unit 1 is oldest.) (?) Early Miocene Split Rock Formation Erosional unconformity White River Formation Unit Feet 21. Siltstone, as in Unit 7 12 20. Tuff, as in Unit 8 1 19. Siltstone, as in Unit 7 4 18. Tuff, as in Unit 8. This unit ash J (Unit 35 of North Fork of Lone Tree Gulch Section) 6 17. Siltstone, as in Unit 7 107 16. Tuff, as in Unit 8 1 15. Siltstone, as in Unit 7 6 14. Tuff, silvery bluish white, vitric, with crystals of bio- tite 1 13. Siltstone, as in Unit 7 6 12. Tuff, as in Unit 8 1 11. Siltstone, as in Unit 7 51 10. Tuff, as in Unit 8 1 9. Siltstone, as in Unit 7 13 8. Tuff, white, vitric, with biotite crystals 1 7. Siltstone, pale gray-green to white except for lower 5 to 10 feet, which is pink, massive, tuffaceous, has occasional harder nodular bands 35 6. Tuff, dark bluish gray, vitric, with many small biotite crystals 9 5. Siltstone, pink, massive, tuffaceous; has occasional hard nodular layers that weather to a brown surface 33 4. Tuff, bluish white, vitric, with biotite crystals up to 1 mm in diameter. This unit ash G (Unit 22 of North Fork of Lone Tree Gulch Section) 3 3. Siltstone, pale gray green in lower part to pink in upper 20 feet, massive, tuffaceous; has occasional hard nodular layers that weather to a brown surface 48 2. Tuff, white, vitric, with small biotite crystals; upper and lower contacts not sharp. This unit ash F (Unit 20 of North Fork of Lone Tree Gulch Section) 1 1. Claystone, siltstone, and sandstone. Predominantly gray green to brown massive claystones becoming siltier and more tuffaceous toward top of unit. Throughout unit is a complex system of coarse channel sandstones, usually with smaller included lenses of pebbles and cobbles of granitic rock, sandstone, quartzite, chert, and chalce- dony. Within some channels are local lenses of nearly pure white vitric tuff with biotite crystals 90 Total thickness of White River Formation 423 Angular unconformity at contact Cretaceous Cody Shale Formation Little Lone Tree Gulch Sections: The lower claystone and sandstone part of the White River Formation is exposed for about 3 miles along the Little Lone Tree Gulch drainage from near the northeast end of Flagstaff Rim to the North Gran- ite Mountain fault zone at the southwest side of Clarkson Hill. The upper part of the section is also exposed on the lower part of the Flagstaff Rim escarpment, but is immediately adjacent to and essentially identical to the corresponding part of the North Fork of Lone Tree Gulch Section and will not be described here. The upper part of the section has been eroded away to below ash B (Unit 10 of the North Fork of Lone Tree Gulch Section) over most of the Little Lone Tree Gulch drainage. Ash B is present, however, on some of the higher ridges and can be traced almost to the North Granite Mountain fault zone at the southwest side of Clarkson Hill. Ash A and a dark gray tuff with rust-colored spots (Units 6 and 8 of the North Fork of Lone Tree Gulch Section) are also present in some localities and can be recognized within one-half mile of the North Granite Mountain fault zone. A general description of the lower part of the White River Formation in Little Lone Tree Gulch can be given by expanding on the field terms used for the units for zonation of fossils collected from this part of the section. The field terms are (from oldest to youngest): lower yellow sandstones, lower banded zone, middle yellow sandstones, and upper banded zone. The lower yellow sandstones are the basal unit, usually of boulders and cobbles of granitic rock, quartzite, and sandstones in a matrix of clayey arkosic sandstone of yellow to rust brown color. This unit is less than 10 feet thick over much of the Little Lone Tree Gulch drainage but thickens and becomes coarser southeastward. The lower banded zone is predominantly of vari- egated red and greenish gray bentonitic claystones and siltstones with local thin lenses of coarse sand- stones. This unit thickens northwestward and be- comes coarser southeastward so that on the south side of the divide area separating Little Lone Tree Gulch and Lone Tree Gulch, it is composed of greenish gray arkosic sandstone and red sandy clay- stone with interbedded thin layers of cobbles and boulders. 24 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY The middle yellow sandstones are of yellow to rust-brown coarse arkosic sandstone with beds of cobbles and occasional lenses of finer siltstones and bentonitic claystones, especially near the top. Northwestward this unit becomes finer and is diffi- cult to separate from the underlying lower banded zone and overlying upper banded zone. To the southeast the unit becomes coarser and thicker so that along the divide between Little Lone Tree Gulch and Lone Tree Gulch it contains many granite, quartzite, and sandstone boulders in a coarse arkosic sandstone matrix. The upper banded zone is a unit of variegated red and pale greenish gray bentonitic claystones that become progressively more silty and tuffaceous upward into the tuffaceous siltstones typical of the upper part of the White River Formation. The lower yellow sandstones and middle yellow sandstones are interpreted as tongues of coarse ma- terial reworked from the ? medial to late Eocene unnamed boulder conglomerate on top of Flat Top, immediately to the south and at a higher elevation. The lower banded zone is a tongue of finer sediments extending between the two tongues of coarser material, becoming thinner and coarser southeastward. The upper banded zone overlies the middle yellow sandstone and in some places interfingers with the upper part of the underlying coarser unit. Two sections are described below to show the lateral variation within the lower part of the sec- tion. Section I of Lower Part of White River Formation in Little Lone Tree Gulch, in the NW 14, Sect. 24 ; and NE 14, Sect. 23, T 31 N,R 83 W, Natrona County, Wyoming. {Beds are nearly horizontal. Unit 1 is oldest. Top of section at present erosion surface at about 6140 feet above sea level.) Unit Feet 5. Tuff, silvery gray to white, vitric, with biotite crystals. This unit ash B (Unit 10 of North Fork of Lone Tree Gulch section) 3 4. Claystone, siltstone, and sandstone. Lower 35 feet alter- nating bands of pale red and greenish gray massive silty claystone with local thin lenses of coarser sandstone. Above highest red band texture becomes progressively coarser and more tuffaceous and color changes progres- sively from pale red to brown and finally to pale greenish gray or white, which is more typical of the NUMBER 18 25 .*<*,.> w^;'-?'* -*" FIGURE 13 White River Formation in Little Lone Tree Gulch. View generally northwestward with north- east end of Flagstaff Rim at left. Letter symbols indicate units dis- cussed in text: LB = "lower banded zone," MY = "middle yellow sand- stones," UB = "upper banded zone," B indicates approximate position of volcanic ash B. tuffaceous rocks. Upper part of unit, then, is tuffaceous siltstone with occasional lenses of resistant channel sandstones up to 12 feet thick, with cobbles up to 10 inches in diameter near the base. Tuffaceous siltstone of upper part of unit, usually lateral to the channel sandstones, has hard concretionary bands several feet in thickness that weather to a rich brown surface but on a freshly broken surface still exhibit pale gray to white color, typical of the enclosing tuffaceous siltstones 92 Sandstone and conglomerate, pale yellow to rust-brown, arkosic, coarse. Some beds with cobbles and boulders up to 18 inches in diameter. Upper part has harder lenses of brown crossbedded coarse arkosic sandstone. Near top of unit are lenses of siltstone and variegated red and greenish yellow claystone 38 Claystone, variegated bright red and pale green to gray, massive, with occasional lenses of pale yellow to gray coarse arkosic sandstone 62 Conglomerate, very coarse at bottom, with boulders up to 3 feet in diameter of granitic rock, quartzite and sandstone in a matrix of coarse pale yellow to rust- brown arkosic sandstone. Unit becomes finer upward and has many chert pebbles in a finer sandstone matrix at the top 22 Total thickness of section Angular unconformity at contact Cretaceous Cody Shale Formation .217 Section II of Lower Part of White River Forma- tion in Little Lone Tree- Gulch, \/^ mile southwest of Clarkson Hill, in the NW y4, Sect. 18, T 31 N, R 83 W, Natrona County Wyoming. (Beds are nearly horizontal. Unit 1 is oldest. Top of section at present land surface at approximately 5960 feet above sea level?) Unit Feet 11. Tuff, bright silvery gray to white, vitric, with biotite crystals. This unit ash B (Unit 10 of the North Fork of Lone Tree Gulch Section) 2 10. Siltstone, as in Unit 8 20 9. Tuff, dark gray to black, vitric, with hard rust-colored spots. Same tuff bed described as Unit 8 of North Fork of Lone Tree Gulch Section 3 8. Siltstone, pale greenish gray, massive, tuffaceous. Has some resistant concretionary bands that weather to a brown color on the surface 7 7. Tuff, white, vitric, with fine crystals of biotite. This unit ash A (Unit 6 of North Fork of Lone Tree Gulch Section.) less than 1 6. Claystone and siltstone, massive. Unit grades upward in color and texture from brown claystone at bottom to pale gray siltstone at top 35 5. Sandstone, yellowish orange to white, coarse, arkosic, with harder lenses of coarse brown arkosic crossbedded sandstone 33 26 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY NORTH FORK Of LONE TREE GULCH ? LITTLE LONE TREE GULCH, SECTION H FIGURE 14?Fence diagram showing correlation of volcanic ash beds and other important units of the four sections of the White River Formation described in text. Numbers at left in each column correspond to unit numbers of text description; numbers at right, in parentheses, are thicknesses in feet of the respective units. Bases of columns arranged in approximate relative elevations. NUMBER 18 27 4. Claystone, variegated pale red and greenish gray; silty in places and has occasional thin lenses of coarse arkosic sandstone 11 3. Sandstone, pale yellowish brown to white, coarse, arkosic, with beds of pebbles and occasional resistant lenses of dark brown to gray crossbedded arkosic sandstone ... 33 2. Claystone, variegated pale red and pale greenish gray and brown, with some thin bands of lavender. Has local lenses of pale yellowish to gray sandstone 18 1. Conglomerate, very coarse, with cobbles and boulders of granitic rock, quartzites, and sandstones up to 3 feet in diameter in a matrix of sandy clay 3 Total thickness of section 166 Angular unconformity at contact Cretaceous Lance Formation MIOCENE SERIES Although this study is not directly concerned with the post-Chadronian rocks of the area, the fol- lowing comments are added in order to complete the description of the Tertiary deposits. Split Rock Formation Overlying the White River Formation along Flagstaff Rim are up to 250 feet of alternating greenish gray to brown claystones, light gray to pinkish gray sandy tuffaceous siltstones, and white lenticular conglomeratic sandstones. In the north- western part of T 31 N, R 83 W, these rocks fill a broad valley cut into the White River Formation as noted by Rich (1962:503). Love (1961) defined the Split Rock Formation and, on his map (1961, fig. 2) showing areas of outcrop of the formation, included the post- Chadronian rocks of the present study area. Love assigned an early Miocene age to the lower part of the Split Rock Formation on the basis of a verte- brate fossil, Merycoides cursor, which was found by Rich. No fossils have been found in the deposits of the upper part of Flagstaff Rim, but these rocks are apparently assignable to the lower porous sand- stone sequence of the Split Rock Formation as de- fined by Love (1961). Structure Structure involving the White River Formation includes faulting and regional tilting. The major fault system within the present study area is a northwest-southeast trending zone which is, according to Rich (1962) and Love (1970), con- tinuous with the North Granite Mountain Fault zone, named and described by Carey (1954:33) in the Rattlesnake Hills anticline to the northwest. This fault zone is not well exposed within the pres- ent map area but only 1 mile to the northwest of the area it is clearly exposed, with the White River Formation on the south side and the upper coarse- grained faci?s of the Wind River Formation on the north side. Rich (1962:510) reported that the White River strata were displaced about 175 feet, the south side of the fault dropped relative to the north side. Along the south side of Clarkson Hill the White River strata are also dropped downward on the south side of the fault zone. Rich reported the opposite to be true here, but this was no doubt due to his interpretation of the boulder conglom- erate on Clarkson Hill as the basal conglomerate of the White River Formation rather than an un- named unit of probable medial or late Eocene age as already shown in the present report. The small faults shown by Rich (1962:511, fig. 81), with dis- placement downward on the north side, are small faults of adjustment on the south side of the major fault zone. Southwest of the North Granite Moun- tain fault zone, in the Little Lone Tree Gulch drainage, are a number of small normal faults, with displacements of 10 to 30 feet. These are all nearly parallel to the North Granite Mountain fault zone and most have displacement downward on the north side, although a few have the oppo- site displacement. The most southwesterly of these small normal faults is about 3 miles southwest of the North Granite Mountain fault zone, near the northeast end of Flagstaff Rim. The small normal faults are difficult or impos- sible to trace in the areas where only massive clay- stone is exposed, but on some of the higher ridges where ash B of the White River Formation crops out, the faults can easily be seen and the displace- ment measured. The total displacement of all these small faults observed is approximately 150 feet. Rich reported (1962:510) that geophysical data indicate that displacement of Wind River and older strata along the North Granite Mountain fault zone may be as much as 5000 feet, with strata on the north side dropped relative to those on the south side. Because the unnamed boulder conglom- erate of ? medial or late Eocene age is stripped 28 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY FIGURE 15.?Small scale normal faults in upper part of "lower banded zone" of White River Formation, south side of Little Lone Tree Gulch, about a mile and a half southwest of the North Granite Mountain fault zone. View is to southeast. Displacement is downward on south- west side (right) relative to northeast side (left). Total displacement of these three small faults is about 10 feet. from the area south of the fault zone but is at least 200 feet thick on Clarkson Hill, north of the fault zone, movement after deposition of this unit was also downward on the north side relative to the south. This movement preceded deposition of the White River Formation. Post-Oligocene movement along the North Granite Mountain fault zone probably occurred during ? Pliocene time (Love 1952:10; Rich, 1962: 512), and resulted in the White River and younger strata being down-dropped on the south side of the North Granite Mountain fault zone, opposite to the movement that preceded White River de- position. During the ? Pliocene faulting, the central part of the Granite Mountains was dropped relative to the Wind River Basin to the north (Love, 1952). The southwestward tilting of the White River and younger strata is a result of these later tectonic events. Preliminary Biostratigraphy The Chadronian White River Formation of the Flagstaff Rim area, Natrona County, Wyoming, is, by vertebrate standards, quite richly fossiliferous. The Frick Collection, American Museum of Na- tural History, contains roughly 3000 individual specimens from this area. A collection made for the National Museum of Natural History in the summer of 1971, though not yet prepared and identified, contains at least 1000 additional speci- mens. The remains are primarily mammalian, but small reptiles are also well represented. Birds and amphibians are quite rare but present in the fauna. Approximately 80 genera and 100 species of verte- NUMBER 18 29 720 700 680 660 640 620 600 560 560 540 520 500 480 460 440 420 400 380 360 3 40 320 300 280 260 240 220 200 leo 160 140 120 100 80 60 40 20 0 FEET KA 1032 3l.6xl06yrs. Ash E Dork gray vitric tuff,similar to Ash D y*vit xvxvxvxim Fi?M term: "Blacky Camel jVsh"ZE22XXXZ??Q Ash D Dark gray vitric tuff : lc _ orn sh ZgXKjfjiXiflfWD ?. Tuffoceous siltstone,pale greenish Woy-^?^~?.T^a.~: with brown weothering concretionary bands^H.~=^:^^-''? Channel sandstone ^a^^^" ^j^Z'^^^r~*~*~\ Ash C Dork 9raV ri,ric ,u,f- ^