yE924 .B414 Berry, Edward Vilber The flora of the Cheyenne sandstone of Kansas DEPARTMENT OF THE INTERIOR Albert B. Fall, Secretary UNITED STATES GEOLOGICAL SURVEY George Otis Smith, Director Professional Paper 129 — I THE FLORA OF THE CHEYENNE SANDSTONE OF KANSAS BY EDWARD WILDER BERRY Published April 11. 1922 Shorter contributions to general geology, 1921 (Pages 199-225) WASHINGTON GOVERNMENT PRINTING OFFICE 1922 CONTENTS. Page. Introduction 199 Historical summary 199 Localities 202 Character of the flora 202 Age of the flora 204 Systematic descriptions 207 ILLUSTRATIONS. Page. Plate XLVI. Osage Rock, about 1 mile north of Belvidere, Kans 202 Plates XLVII-LXI. Fossil plants from the Cheyenne sandstone 225 NOTE. — The papers included in the annual volume "Shorter contributions to general geology" are issued separately, with the final pagination, as soon as they are ready. The last paper will include a volume title-page, table of contents, and index for the use of those who may wish to bind the separate parts. A small edition of the bound volume will also be issued, bnt copies can not be supplied to those who have received all the parts. II THE FLORA OF THE CHEYENNE SANDSTONE OF KANSAS. Bv Edward Wilbek Berry. INTRODUCTION. The present study is based on eullections made by Hill in 1894, Ward and Vaughan in 1S90, Ward, Gould, White, and Cain in 1897, and Lee in 1919. These collections were very extensive, but the bulk represented small fragments of the more abundant species, such as the Sequoia and Sapindopsls. The flora itself as at present known consists of a very small number of species. The Cheyenne sandstone comprises about 100 feet of gray to yellow friable ([uartz sand- stone with subordinate lenses of dark shale. The sandstone ranges from fine to coarse and contains a few layers of quartz and chert pebbles. It is in the main only slightly consolidated and is very friable and easily eroded. The bedding is extremely irregular and discontinuous, and cross-bedding is obvious throughout and in places extremely pronounced. Logs of silicified wood and Cycadeoidea minuta from these beds were recorded by Cragin.' The Cheyenne sandstone rests upon " Red Beds" of supposed Permian age and is over- lain by the Kiowa shale — shallow-water and lagoon deposits of alternating layers of marl and bituminous clay shale, with a marine fauna that includes many species character- istic of the Washita group of the Texas Cre- taceous. The invertebrates are said by Twenhofel to' number about 50 species, of which the follow- ing are some of the commoner forms: Cardium kansasense Meek . Cyprimeria kiowana Cragin. Exogyra texana Roemer. Gryphaea corrugata Say. Gryphaea navia Hall. Ostrea quadriplicata Shiimard. Pecten texanus Roemer. Protocardia texana Conrad. Schloenbachia belknapi (Marcou). Schloenbachia peruviana (Von Buch). Trigonia emoryi Conrad. ' Cragin, F. V/., Washburn Coll. Lab. Nat. Hist. Bull., vol. 2, pp. «5-66, 1889. In some places the Tertiary overlies the Kiowa; elsewhere the following units in ascend- ing order have been recognized by Gould: Spring Creek clays, Greenleaf saiulstonc, Kirb}- clays, and lleeder sandstone. The names are those proposed by Gould and Cragin and have not been formally recognized by the United States Geological Survey. These units are chiefly local phases or lentils in the Kiowa, of little significance except as indicative of local and more or less contemporaneous variations in conditions of deposition, with perhaps a basal member of the Dakota sandstone repre- sented in the "Reeder." HISTORIC.\L SUMMARY. The term "Dakota group" was first used in 1861 by Meek and Hayden^ for the lower por- tion of their section of the Cretaceous of Ne- braska, corresponding to No. 1 of the classic Meek and Hayden Upper Missouri section.' Tliis term or simply Dakota or Dakota sand- stone has subsequently been used in innumer- able references to local geologic sections throughout the West. The assumption that the Upper Cretaceous of tliat whole region contained two persistent sandstones — the Da- kota at its bottom and the Fo.x Hills near its top — and the fancied recognition of these sandstones over a wide area have caused much of the confusion and controversy that have arisen over the interpretation of the western Cretaceous. As originally understood the term Dakota was applied to the pre-Benton Cretaceous, no Lower Cretaceous being then recognized in that region. Unquestionably the typical Da- kota sandstone represents the littoral or mar- ginal deposits of the transgressing Benton sea, but that there are similar and somewhat earlier continental or marginal sandstones in 2 Meek, F. n., and Ilaydon, F. V., .Vcad. N'at. Sci. I'hilttdi'lphia Proc., vol. i:i, p. 419, 1861. "Hall, James, and .Meek, F. H., .Vm. .Vcad. Mem., vol. ,■., p. 405, l.S.'ie. Mock, F. U., and Uayden, F. V., .\cad .Nat. Sci. I'tiiladulpbia I'roc., vol. 8, p. 63, 1856. 199 200 SHORTER CONTRIBUTIONS TO GENERAL GEOLOCiY, r.>21. Kansas, Colorado, the western Black Hills, and presumably elsewhere in this region has been pretty well knowTi for a number of years. Their exact age has been a matter of consider- able differences of opinion. The history of paleobotanic discovery of the so-called Dakota flora has been given in Les- quereux's three memoirs and need not be re- counted here except to point out that the col- lections, a study of which resulted in the identi- fication of over 400 species of plants, were made at different times and places by a number of different collectors, who, as in so much of the early exploratory work in the West, paid little attention to stratigraphic position or locality. Any yellowish or reddish sandstone with im- pressions of dicotyledonous leaves was Dakota in age, and for a large number of species "Da- kota group of Kansas," or at most the county from which the specimens were collected, is all we know of the whereabouts of the outcrop. Apparently the first to notice marine fossils at the base of the red Cretaceous (Dakota) sandstones was Le Conte.' Cragin, while at Washburn College, Topeka, Kans., did much work upon the Cretaceous and published many short paleontologic papers. In 1890 he de- scribed a cross-bedded sandstone (the Chey- enne sandstone) which underlay marine beds in southern Kansas and which he considered to be related to the Potomac, Tuscaloosa, Trin- ity, and " Atlantomurus beds," and the next year he published the statement that the Chey- enne sandstone was probably of the same age as the Trinity of Texas, the Potomac of the Atlantic Coastal Plain, and the Wealden or Purbeck of Europe. Invariably in his discus- sions he used the term Comanche as the inter- changeable equivalent of the European Neo- eomian. The first definite announcement of the flora contained in the Cheyenne sandstone was made by Hill,'^ who recorded the following species from collections made by Hill, Gould, and Shattuckin 1894: Rhus uddeni Lesquereux. Sterculia snowii Lesquereux. Sassafras mudgii Lesquereux. * Le Conte, J. L., Notes on the geology of the survey for the extension of the Union Pacific Railway, I'hiladelpliia, 1S68. & Hill, R. T., Discovery of a dicotyledonous flora in the Cheyenne sand- stone: Am. Jour. Sci., 3d ser., vol. 49, p. 47.3, 1895; On outlying areas of the Comanche series in Kansas, Oklahoma, and New Mexico: Idem, vol. SO, pp. 205-234, 1S95. Sassafras cretacciim obtiisum Lesquereux. Sassafras n. sp. Glyptostrobus ^racillimus Lescjuereux. Sequoia .sp. (cones). Cragin 's conclusions were given in a paper published in 1895," in which the section is given as follows : Kiowa shales. Champion shell bed. roi i > ^ rr,,, ,, , , , Stokes sandstone. Elk Creek bedsw u- u i Cheyenne sandstone ILanphier shale. [Corral sandstone. From the "Elk Creek beds" he recorded Sterculia snowii, Sassafras mmlgei, Sassafras cretaceum, Sassafras sp., Rhus uddeni, Sequoia sp., and Ghjptosfrohns gracillimus.'' Only the first two of these are contained in the collections studied by me. Other contributors to the subject prior to 1900 were Mudge, Prosser, Jones, Stanton, and Gould. Their results are not pertinent to my present purpose beyond the fact that they show conclusively the presence of a sandstone, the Cheyenne, containing the remains of a land flora in southern Kansas beneath a marine series, the Kiowa shale, carrying a fauna that is correlated with that of the Washita group at the top of the supposed Lower Cretaceous section of Texas as elaborated by Hill. During his residence in Kansas Twenhofel studied the Cretaceous of the central part of the State, and in a brief paper ^ published in 1917 he confirmed Cragin's earlier results " that a situation identical with that of southern Kan- sas prevails in central Kansas. In a more recent article '" he contends that the Dakota of Kansas and the Washita group of Texas are of the same age, and that both the Cheyenne- Kiowa-" Medicine beds" sequence of southern Kansas and the Mentor-Dakota sequence of central Kansas should be referred to the Comanche series. The " Dakota flora" of the Denver Basin has recently been revised by Knowlton. As a result of field work by Lee and Cannon during 1910 it has been showTi '' that the formation from which Lieut. Beckwith collected the "Dakota" plants from Morrison, Colo., that ' Cragin, F. W., K study of the Belvidere beds: Am, Geologist, vol. 19, pp. 357-385, 1S95. ' Idem, p. 367, quoted from Hill. e Twenhofel, W. H., Kansas Acad. Sci. Trans., vol. 2s, pp. 213-223, 1917. » Cragin, F. W., Am. Geologist, vol. 16, pp. 162-165, lS9o. " Twenhofel, W. H., Am. Jour. Sci., 4th ser., vol. 49, pp. 281-297, 1920. n Lee, W. T,, Am. Jour. Sci., 4th ser., vol. 49, pp. 183-188, 1920 FLORA OK THK CHEYENNE SANDSTONE OF KANSAS. 201 Les(|iierc'ux licscrihod is tho same us that described by Richardson '- in 1915 as the Pur marine sediments would probably be those in dispute. On none of the continents, so far as I can discover, did the sea complete a cycle of inva- sion and withdrawal of what might be called the first magnitude during the Lower Creta- ceous epoch. In the Atlantic (^oastal Phiin no marine Upper Cretaceous deposits earlier than the European Turonian are known ex- cept in the Texas area, where marine forma- tions representing a part but not all of the Lower Cretaceous of Europe advance halt- ingly from the south. The oldest of these formations is the Trinit}', which in my judg- ment is nowhere as old as the Neocomian of Europe. This is followed by the Fredericks- burg group, which llill called Neocomian but which contains a younger fauna. If one dis- regards Bose's correlations of the Mexican Cretaceous on the ground that Mexico is too remote from the north Texas-Kansas area, Whitney's studies of the fauna of the Buda limestone not only clearly show its Cenomanian age but also show that it is' late Cenomanian. Similarly the fauna of the Georgetown lime- stone is Cenomanian. (Whitney has refrained thus far from making any intercontinental correlations.) It is a striking confirmation of this correlation that the Buda limestone near Austin and hence in the region of more continuous marine conditions than farther north should be immediately overlain i)y the Turonian Eagle Ford formation. Tiie |)i()l)!cm of working out the interfingering of forma- tions between north and central Texas is largelj^ a problem of invcrtcljrate paleontology 202 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1921. and would not be mentioned in the present connection if it were not for the necessity of showino; that the correlation of the Texas Comanche series with the Lower Ci-etaceous of Europe is incorrect and can not hope to be ultimately accepted, and because this problem is so inthnatel}- bound up with the age of the Cheyenne sandstone. If there is no marine Lower Cretaceous in Kansas, as I contend, our ideas of the sequence of events from late Lower Cretaceous time into the Upper Cretaceous require to be very greatly modified. With regard to Twenhofel's proposal to refer the Mentor and Dakota of central Kansas to the Comanche, all that I can say is that while he and before him Cragin and others have written about the Dakota flora, this term is altogether meaningless stratigraphically, ex- cept that it denotes in a most general way a change in facies between Lower and Upper Cretaceous floras. The flora of the Cheyenne sandstone, and I presume that in the Mentor formation as well, is no more like that of the Woodbine sand than the Woodbine flora is like that of the several formations of the Montana group, and the reference of the Dakota sand- stone— that is, the post-Mentor Dakota sand- stone ot central Kansas — to the Lower Cre- taceous if correct would of necessity carry with it the Bingen sand of Arkansas, the Tus- caloosa formaticm of Alabama, the Black Creek formation of the Carolinas, and the Magothy formation of New Jersey and Maryland, against whose correlation with the Senonian of Europe by paleozoologists I have been arguing for years, with not very great success. LOCALITIES. All the localities from which fossil plants were collected in the Cheyenne sandstone are in the immediate vicinity of Belvidcre, Kiowa County, Kans. (See PI. XLVI.) I give be- low a transcription of the locality numbers, with the names of the collectors and dates, taken from the United States Geological Sur- vey's records. There appears to be some con- fusion in the two collections numbered 2224, nor do I have locality numbers for the material collected by Ward and Vaughan in 1896. These defects in the record are immaterial, however, for there is no doubt that all the ma- terial studied came from the Cheyenne sand- stone in this immediate region. Many of the numbers are duplications of identical outcrops and are given only as a matter of record. 773. Black hills near Belvideie; collected by Uill, Gould, and Shattuck, 1894. 2217. Osage RockatBeh'idere, from Nos. 1 and 2 of Hill's section; collected by O. L. Cain, 1897. 2218. One and one-half miles northwest of Belvidere, from No. 3 ot Hill's section; collected by Ward and Gould, 1897. 2219. Same as 773. Stokes Hill. 100 yards south of the National Corral: collected by Ward and Gould, 1S97. 2220. Stokes Hill, the most northeasterly of Hill's locali- ties; collected by Gould. 1897. 2221. Thompson Creek near the flume. 2 miles north- west of Bel\-idere; collected by Ward and Gould, 1897. 2222. Champion (Wildcat) Draw, three-fourths mile south of Behddere; collected by Ward, Gould, and White, 1897. 2223. Same locality and collectors as 2222, from the ■ ' T^anphier shales. " 2224. Near Medicine Lodge River. 2 miles west of Bel- videre (original locality of Ward and Vaughan in 1896); collected by Ward and Gould. 2224. Champion (Wildcat) Draw, right (east) branch, in " Lanphier' shales, " half a mile south of Behidere; col- lected by Ward and Gould, 1897. 222.5. One mile southwest of Belvidere, in a draw ("Lan- phier shales" ); collected by Ward and Gould, 1897. 2226. About 2J miles due west of Belvidere (fern bed of 1896) ; collected by Ward aud Gould, 1897. 2227. Hills between Spring Creek and Soldier, 4 miles northeast of Belvidere; collected by Ward and Gould, 1897. 2228. Champion (Wildcat) Draw, right (east) branch, "Lanphier shales," half a mile south of Behidere; col- lected by Ward and Gould, 1897. 2229. Left bank of middle branch of Champion (Wild- cat) Draw, half a mile south of Belvidere; collected by Ward and Gould, 1897. 2230. Draws north of Belvidere ("Lanphier shales"); collected by Ward and Gould, 1897. 2231. Right bank of middle branch of Champion (Wild- cat) Draw, half a mile south of Belvidere; collected by Ward and Gould, 1897. 2232. Osage Rock at Belvidere, "Stokes sandstone" below the so-called Champion shell bed; collected by Ward and Gould, 1897. 2233. First draw west of Champion (Wildcat) Draw, half a mile south of Belvidere; collected by Ward and Gould, 1S97. 740.5. Wildcat Draw, near Belvidere; collected by W. T. Lee, 1919. 7406. Osage Rock, near Belvidere; collected by W. T. Lee. 1919. CHARACTER OF THE FLORA. The flora of the Cheyenne sandstone as dis- closed in the present study numbers but 23 species. It comprises four ferns representing the families Polypodiaceae and Gleicheniaceac, and all four are representatives of widely U. R. OEO LOGICAL SURVEY 1 Uu M.^MONAI. lAPER 129 I'LATE XLVI FLORA OK THE CHEYEITNE SANDSTONE OF KANSAS. 203 rangini^ and more or less well known Crctu- ceous types. The collections include a sup- posed cycadophyte seed, but this is of doubtful relationship. There is also a fragment of a trunk of the genus Cycadeoidea, which is of somewhat uncertain value, as its exact strati- graphic position has been (luestioned. There are four coniferophytes and eleven angio- sperms. One of these is a supposed monoco- tyledon, ten are dicotyledons, and there are two forms of uncertain botanic rclationsliips. The dicotyledons represent the orders ,Sapin- dales, Malvales, Thymeleales, and Umbellales and are remarkable for the total absence of a large number of elements generallj^ found in floras of this age. This absence can not be wholly explained by accidents of preservation and discovery and is due, I believe, to the pecu- liar ecologic grouping resulting from the en- vironment.. The arenaceous portions of the Cheyenne sandstone are very conspicuously cross-bedded. The material is very friable, and the vegetable remains are embedded in all sorts of positions and curled as they are when covered in a diy condition by wind-blown sands. ' All are coria- ceous forms, and the abundant Sequoia cones all have their scales shrunken and widely distended as in thoroughly desiccated modern cones. They appeal' to have blown about and collected in hollows along with the coriaceous leaves that are found in association with them. With a single exception the ferns are found in the claj's and evidently were confined largely to stream banks. The variety of plants in such situations may have been larger than the discovered flora indicates, but it would seem as if in collections so extensive there should be some traces of the other plants preserved if they were growing near at hand. Although the flora is too small and too remote in time from existing floras to afford satisfac- tory ecologic data, it does furnish some sug- gestions. It seems to me to indicate a warm and more or less arid ('limate, with a sparse vegetation. I picture this vegetation as of meager variety and as having been confined largely to the region of watercourses between which were larger areas of sand-hill or beach- dune country over which the dried leaves and fruits were blown, collecting in the hollows and becoming covered by wind-blown sands. The clay lense-s — for example, Cragin's " Lanphier beds" — ^are waterlaid .uid might represent seasonal rainfall and (lood-plain or playa de- posits or normal stream sedimentation, and it is possible that some of the sands had a like origin. There is no evidence of aridity in any of the Cretaceous floras with which the Cheyenne sandstone flora may be compared, whether such comparisons are made with the Patapsco and Fuson floras, on the one hand, or the Woodbine, Dakota, and Tuscaloosa floras, on the other. I believe, therefore, that the Cheyenne flora does not represent general conditions but is puiely an expression of the local environment and perhaps represents a wide sandy coastal plain or fluctuating l)eaches backed hy dunes, and that farther inland a more varied and nor- mal flora probabh' existed throughout the period when the shaUow sea was migrating back and forth across southern Kansas. A sample of the Cheyenne sandstone was submitted to Mr. Marcus I. Goldman, who has kindly furnished the appended observations : Macrnscopir examination. — A solid but friable fine- grained sandstone of a pale lavender-l)ron'n rolor charac- teristic of moderately carlxjnaceous sandstones. Xo lamination. Contains curled and \mnkled leaf impres- sions suggestive of deposition in a dry condition, hence in wind-blown sand. Mechanical analysis. — ^The rock could be easily rubt)ed down int<3 its constituent grains. On Hie\'ing these di\'ided as follows: Fine sand through GO on 100 mesh, 12.9 jjer cent, 0.4.5-0.26 millimeter; very tine sand through 100 on L'(K) mesh, 82.2 per cent, 0.2(M).04 millimeter; extra fine sand through 200 mesh, 4.9 per cent, lees than 0.04 millimeter. Microscopic examination showed that the two finer parts contained thoroughly dLsintegrated grains. The coarsest, however, consisted largely of compound grains which yielded slowly to disintegration, so the following rough figures may l)e taken: I'^ne sand, .5 per cent; very fine sand, 90 ]ier cent; extra fine sand, 5 per cent. In either ca-se the great predominance of the very fine sand is obx-ious. This predominame of a single size at once suggests wind action, but comparison with dune sands (cf. my paper on the Catahoula sandstone,'* where several analyses are assembled) shows that the maximum is in the size next finest to that which forms the maximum in typical dune material. I have looked up the large collection of anah'ses given by Udden " and find that in this character the j^ample resembles the finer sand carried by the wind out of other deposits. Thus it corres|x)nds with only two of his dune sands — No. 219, whiih is the finest material gathered at the crest of a dune, and No. 248, from a blown field. >* Goldman, M. I., eetrograpliicevidprico on the origin oftlie Catahoula sandstone of Texa.s: .\ni. Jour. Sci., 4tli set., vol. 3a, p. 269, 191j. ^'■^ L'dden, J. .v., ^lechanical composilion of clastic sediments: Ceol. Soc. .\merica Bull., vol. Z't, pp. tio5-"44, 1915. 204 SHORTER CONTRIBUTIONS TO GENERAL, GEOLOGY, 1921. On the other hand, the majority of samples of what he calls incipient wind-blown sand, which is sand blown out of other deposits, have this composition, as do also his lee sands, which are the finer material blown beyond the body of a dune. MiirnscoTinc examination. — (I) Thin section: Three char- acters are conspicuous in the thin section of the original rock, namely, («) its porosity, which is evidently priman,' — that is. not produced by grinding the thin section — and is a character of wind-deposited material; '" (h) the angu- larity of most of the grains; {c) the absence of a recogniz- able deposit of secondary quartz on the outside of the grains. In many cases it can be definitely proved, by the presence of inclusions throughout the grain, that there is no outer coat of secondary quartz, and the rough angularit}' of the grains does not suggest secondary growth, which tends to restore crystal form, but. if anything, rather solution. (II) Disintegrated sand: The disintegrated material can be studied to greater ad^■antage because it can be immersed in liquids of different indices of refrac- tion. Thus by immersing in a liquid of index about 1..5.5 the inner structure of the quartz grains and any possible boundary between nucleus and secondary quartz can be more readily recognized. Immersing in a liquid of index about 1.6.5, on the other hand, brings out brilliantly the surface form and texture. Much more could be observed, especially in the latter liquid, than the time at my disposal permitted, but I can make the following generalizations: (a) None of the portions, not even the coarsest, reveal any unusual abundance of rounded grains; rather the opposite. (6) The surface of grains is not frosted, as is characteristic of wind-blown sand, nor pitted, as if subjected to much solution, but rather rough and chipped -loo king, as might be the result of mechanical wear, (c) Under the favorable conditions afforded by immersion in a liquid of index about 1..55, no secondary silicification could be detected. id) In the finest size a few very small well-rounded grains were found, the smallest having a diameter of about O.OSf) millimeter and being exceptionally perfectly rounded. Such a grain could probably be produced only by pro- longed wind action." The main problem of fact, in spite of these observations, still concerns the cause of the present form of the grains. Secondary silicification is often very veiled, and in the absence of comparative studies I am not prepared to commit myself as to the possibility of solution having acted on the grains. These two factors affect not only the form but also the size of the grains. However, the coherence of the observations made favors the belief in their correctness. They lead to the following conclusion: The portion of the Cheyenne sandstone represented by this sample is nothing like an eolian deposit in an arid region, nor even a part of a permanent dune area in a humid climate, but merely an accumulation of material blown by the wind out of a deposit of some other origin. Only an extensive field study could disclose the main accumulation from which this is separated and other possible associated facies and lead to a complete and satisfactory interpretation. From the roughness of the sand grains I would be inclined to assume rather a delta than a beach deposit as the dominant type. The very >« Goldman, M. I., op. cit., pp. 280-281. " Idem, pp. 271-272. small, perfectly rounded grain noted appears entirely out of place as the product of the conditions under which the sandstone seems likely to have been formed and must therefore be assumed to be the product of an earlier cycle in the history of the grain, unless it is assumed that in a deposition of secondary silica the larger grains have been favored so that only the smallest grains retain their original form; but this seems to me incompatible with the uniform size of the larger grains. As Ijrought out in the paper on the Catahoula sandstone there are many more factors that might have been con- sidered, but being limited in time and having no assistance for such work I have selected the few that seemed most rapidly determinable and most important. A single sample whose position in the local section is unknown can be conclusive only for that one sample. The foreset bedding of which I have seen pictures suggests delta deposition, but no one has made field studies of the Cheyenne sandstone with the object of determining the origin of the sediments or the method of sedimentation, and Mr. Goldman's report is therefore chiefly significant in indi- cating what definite results might be obtained in this region of an oscillating Cretaceous strand by a careful study of the problem. The plants and their method of occurrence conclusively indicate a sparse vegetation, at least seasonal dr\Tiess, and accumulation by the wind. This I am sure was a local condi- tion, as previously stated. Whether these facts do or do not indicate beaches, interstream sand hills, or delta deposits can be determined only by future studies. AGE OF THE FLORA. Ignoring for the present the general rela- tions and the implications resulting from the age of the overlying Kiowa shale or the sup- posed equivalent beds in Texas, as discussed in the preceding section, and considering only the evidence of the fossil plants, I may note that ten of the 23 species are positively iden- tified as "Dakota" forms and two additional species are tentatively so identified. Four others are peculiar to the Cheyenne sandstone. The type localities of the remaining seven species were the Cenomanian of Bohemia (one species), the Atane beds of Greenland (two species), and the Patapsco formation of Mary- land (four species) . Of the so-called Dakota forms in the Chey- enne flora the following species were previously known from only the ''Dakota": CladopMehis dakotensis (Lesquereux) , Cycadeospermum line- FLORA OF THE CHEYENNE SANDSTONE OF KANSAS. 205 atum Lesquereux, Sequoia coiulita Loscjuereux, Abietites ernestinae Lcscjuercux, Sterculia mu- cronata Lcsciucreux, and Sassafras muihjii Les-land Geol. Survey, Up|)er Creta- ceous, p. 767, 1916; U. S. GeoL Survey Prof. Paijer 112, p. 53, 1919. This species was described by Heer in 1874 from material found in the Kome beds (Lower Cretaceous) of Greenland. It was subse- quently identified by Heer in material from the much later Atane beds (l^pper Cretaceous) of Greenland; Dawson reported it from a number of localities in the Kootenai formation (Lower Cretaceous) of British Columbia, although these records are questionable: and Fontaine and Ward described it from specimens obtained in the Lower Cretaceous of the Black Hills. It is also reported by both Lesquereux and Wartl from the Dakota sandstone, and by Kurtz from Argentina. It seems very doubtful if these specimens can all be the same plant, and the geologic range alone suggests that the ear- her and the later forms may be distinct. The Lower Cretaceous forms certainly suggest a relationship with those widespread types of sterile fronds variously itlentificd as Thyrsop- teris or Ornjchiopsin, and thcj- nmy be com- pared with Oiiychiopsis (joipperti (Schenk) Berry. The Upper Cretaceous forms suggest Anemia rather than Anplenimn and are much like the widesi)read Eocene species Anemia Jiaydenii (Lesquereux) Cockerell and Anemia subcretacea (.Saporta) Gardner and Ettings- hausen. However, in the absence of repre- sentative material from the different horizons, it seems unwise to attempt any segregation at the present time, and the s'vnionymy is cited in full for the use of some future student who may have access to enough material to enable him to make an accurate revision and segregation of this so-called species. Attention should also be called to its resemblance to the form occur- ring in the Upper Cretaceous of Greenland, the Raritan formation of New Jersey, and the Tus- caloosa formation of Alabama which goes by the name Dicksonia groenlandica Heer, al- though the ground for considering it a Dicfc- sonia is entirel}' inconclusive. Besides occurring at the localities named above the present species is abundant in the Karitan formation of New Jersey and Mary- land, and material that is absolutely identical with the New Jersey liaritan material which I have seen and with that from the Dakota sand- stone is present in the Tuscaloosa formation of 208 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1921. Alabama and the Cheyenne sandstone of south- ern Kansas. I have recently received a fine specimen from northeastern New Mexico from a sandstone that appears to represent the Pur- gatoire formation. The specimens from the Cheyenne sandstone, all of which are fragmentary, were found in brown clay in a draw on Medicine Lodge Creek, 3 miles above Belvidere (collected by Ward and Vaughan, 1896, no number); 1^ miles north- west of Belvidere (2218) ; near Medicine Lodge Creek, 2 miles west of Belvidere (2224, same locality as that first cited); and 2^ miles due west of Belvidere (2226). Order GLEICHENIALES. Family GLEICHENIACEAE. Genus GLEICHENIA Smith. Gleichenia nordenskioldi Heer.^° Plate XLVII, figure 1. Gleichenia nordensholdi Heer, Flora fossilis arctica, vol. 3, Abt. 2, p. 50, pi. 9, figs. 6-12, 1874; vol. 6, Abt. 2, p. 8, pi. I, figs. 1, la, 1882. Lesquereux, U. S. Geol. and Geog. Survey Terr. Ann. Rept. for 1874, p. 334, pi. 2, fig. 5, 1876; Cretaceous and Tertiary floras, p. 26, pi. 1, figs. 1, la, 1883; U. S. Geol. Survey Mon. 17, p. 2.5, 1892. This species was described originally by Heer from material collected in the Kome beds of western Greenland. Species of GleicJienia are very abundant throughout the Cretaceous section of that region, and Heer founded very manj' species on this material, more than seem warranted. The stratigraphic boundary be- tween the Kome and Atane beds has been shown by subsequent workers to be very indefinite, and the Atane beds are present at the Kome locality, a fact which may account for the range accredited to a large number of the species. Lesquereux subsequently identified Glei- chenia nordenskioldi from the Dakota sand- stone at Fort Harker, Kans. His material was not very convincing and perhaps should not have received a specific name. Material "The following do not belong to this species: QUidienia nordenskioldi Fontaine, U. S. Geol. Survey Mon. 15, p. 119, pi. 21, fig. 11, 1S90. aitichenia nordenakioldi Fontaine, in Ward.U. S. Geol. Survey Mon. 48, p. 231, pi. 65, figs. 24-29, 1906. Pezopicris strictineriis Fontaine . Fontaine, in Diller and Stanton, Geol. Soc. America Bull., vol. 6, p. 450, 1895; in Stanton, U. S. Geol. Survey Bull. 133, 1895, p. 15, [18%). Aspidium heterophyllum Fontaine. Fontaine, in Diller and Stanton, op. cit., p. 450; in Stanton, op. cit., p. 15. Qimunda dicksonioides Fontaine . Fontaine, in Diller and Stanton, op. cit., p. 450; in Stanton, op. cit., p. 15. identical with that of Lescjuereux is not un- common in the Cheyenne sandstone, and I have used the same name for it, although it should be borne in mind that neither Les- cjuereux's material nor mine is distinct from what has commonly been called GleicJienia zi-p- pei Heer,-' which has been identified, often wrongly, I believe, at a large number of localities and horizons. The unwarranted determination of uniden- tifiable scraps by Ward and especially by Fontaine has almost completelj' obscured the stratigraphic value of anj' material that they described. The specimen from Dutch Gap, Va., which Fontaine referred to this species not only differs from the type material but might readily represent the terminal portion of half a dozen different Patuxent species of ferns. Similarly the specimens from the Knox- ville formation which Fontaine referred to Gleichenia nordenskioldi are not only not that species but they are not even all the same thing, and the fact that these identical fragments were also referred by Fontaine to Pecopteris, Aspldivm, and Osmunda, as well as to Gleiche- nia, is a fitting commentary on both the char- acter of the material and the critical value of Fontaine's results. What I have called GleicJienia nordensJcioldi is found in the Cheyenne sandstone in clay 2^ miles due west of Belvidere (2226), also de- scribed as a draw on Medicine Lodge Creek, 3 miles above Belvidere (fern bed, no number), collected by Ward and Vaughan in 1896. Gleichenia? bohemica (Corda) Berry. Plate XLVII, figure 2. Pecopteris bohemica Corda, in Reuss, Versteinerungen der bohmisfhen Kreideformation, p. 95, pi. 49, fig. 1, 1846. Heer, Flora fossilis arctica, vol. 3, Abt. 2, p. 96, pi. 26, fig. 17a, 1874; vol. 7, p. 6, pi. 58, fig. 4, 1883. Engelhardt, Naturf. Gesell. Isis in Dresden Abh., 1891, No. 7, p. 86. Kryshtofovich, Coll. Sci. Imp. Univ. Tokyo Jour., vol. 40, art. 8. p. 31, fig. 2, 1918. This species was described by Corda in 1846 from material obtained in the Cenomanian of Bohemia. It was subscquentlj- recorded from the same horizon in Saxony. Heer referred a number of Greenland specimens to it, and lately Kryshtofovich has recorded it from the Upper Cretaceous of Sakhalin Island. If these a Heer, Oswald, Flora fossilis arctica, vol. 1, p. 79, pi. 43, fig. 4, 1868. FLORA Ui' THE CHEYENNE SANDSTONE OF KANSAS. 209 records all represent the same species, it was evidently a wide-ranjiiiig ty[)e in tlie earliest stage of the Upper Cretaceous, wliich spread from the Arctic region southward into North America, Europe, and Asia. There are five specimens in the Cheyenne sandstone that appear to be identical with Heer's Greenland forms, but as they are preserved in a coarse sandstone their detailed characteristics are obliterated. The pinnules arc coriaceous, long, and narrow and somewhat resemble what Heer " called illeicJiema rigida. Although details of frond habit and fructifi- cation are lacking I have ventured to transfer this form from Pecopteris to Gleichenia, as it appears to be congeneric with the numerous Cretaceous forms of that genus. It was found in the Cheyenne sandstone on the left bank of the middle branch of Cham- pion (Wildcat) Draw, half a mile south of Belvidere (2229). Phylum CYCADOPHYTA. Order CYCADEOIDALES. Genus CYCADEOIDEA Buckland. Cycadeoidea munita Cragin. Cycadeoitlea munita Crasriii. Washburu College Lab. Nat. Hist. Bull., vol. 2, p. 6.5, 1889. Ward, U. S. Geol. Sur\'ey Nineteenth .\nii. Kept., pt. 2, p. 541, 1899. Hill -^ states that there is some doubt as to the occurrence of tliis specimen at tliis horizon. Lester F. Ward, who subsequently visited the locality, states that he was satisfied that it could not have come from the Cheyenne sand- stone but may have weathered out from the overlying '"Reeder sandstone." The material, which is only a fragment, has never been studied by a competent person, although Ward states that it is surely a fragment of a cycad trunk. Whatever its true horizon it is of interest as one of the latest authentic occurrences of this type of plant. Genus CYCADEOSPERMUM Saporta. Cycadeospermum lineatum Lesquereux. Cycadeopsermum Unealum Leaquereux, U. S. Geol. Survey Mon. 17, p. 30, pi. 1, fig. 14, 1891 [1892]. This seed, which was found 10 miles north- east of Delphos, Ivans., was described by Lesquereux as foUows: 2= Heer, Oswald, Flora fossills arctica, vol. 1, p. SO, pi. 44, fig. 1, 1S6S. » Hill, R. T., Am. Jour. Sci., 3d ser., %-ol. 50, p. 21-.i, lS9o. Seed oblong-ovate, slightly falcate, rounded at the lower end, short acuminate at the other; testa smooth, transversely lineate, the lines distant, parallel; carena clearly marked longitudinally on both sides, the inner concave, the outer rounded. Length 1 to 1.5 centimeters; width about 6 millimeters, somewhat compressed. Testa thick, shining, and ligneous. Cycadophyte seeds are not so inequilateral, and the present form is probably angiosperm- ous. This genus was proposed for Jurassic forms, of which many have been described. A few have been described from both Lower and Upper Cretaceous material. The Cheyenne form is certainly identical with Lesquereux 's type. Wliether or not it is congeneric with the other species referred to Cycadeospermum, or whether indeed it represents the seed of a cycadophyte and not an angiosperm, can not be determined. My impression is that it belongs to the latter rather than the former. Material identical with Lesquereux's type is found in the Cheyenne sandstone IJ miles northwest of Belvidere (221S) and near Medi- cine Lodge Creek. 2 miles west of Belvidere (2224). Phylum CONIFEROPHYTA. Order CONIFERALES. Family CUPRESSINACEAE. Genus SEQUOIA Endlicher. Sequoia condita Lesquereux. Plate XLVIII, figures 1-11. Sequoia condita Lesquereux, U. S. Geol. and Geol. Survey Terr. Bull., vol. 1, p. 391, 1875 [1876]; Ann. Rept. for 1874, p. 355, pi. 4, figs. 5-7, [1876]; U. S. Geol. Surv'ey Terr. Rept., vol. 8 (Cretaceous and Tertiary floras'), p. 32, pi. 1, figs. 5-7, 1883; in Cook and Smock, Rejwrt on clay deposits in New Jersey, p. 29, 1S78. The inextricable confusion that results from the identification of detached fragments of conif- erous foliage when they can not be checked by fruits or in some other way is well illustrateil by the forms that are variously referred to Ghjptostrohus (jracUlimus Lesquereux, Sequoia (jracilUma Newberry. WiddriiKjtoiiites reichii Heer, etc. In volume 6 of the final reports of the United States Geological Survej" of the Territories Lestjuereux gave figures of a plant which he had named some years earlier Ghjpto- strobus (jraciUintus and which he compared with Frenelites reichii of Ettingshausen. When 210 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1921. Newberry described the flora of the "Amboy days'' (Raritan formation) ' he renamed Les- cjuereux's species Sequoia gracillima because he found associated with similar foliar remains in New Jersey elongate cones with scales resem- bling those of a Sequoia. He commented on the resemblance of these cones to Geinitzia, but the matter rested here until I compared speci- mens of the so-called Seqnoia gracillima cones with those of the European Geinitzia formosa Heer and found the two to be identical. In the European Upper Cretaceous these cones were foimd attached to foliage of a very differ- ent type from Glyptostrobus gracillimus or Sequoia gracillima, although in America the cones were always detached. Foliage like that of Geinitzia formosa was found, however, asso- ciated with them. It seemed obvious that the cones referred to Sequoia gracillima were those of Geinitzia formosa , and accordingly I so assigned them. I had collected hundreds of these cones and had abundant comparative material. When I revised the Raritan flora I had much larger collections than those of Newberry, and I found that the foliage which he had called Sequoia ijracHlima was identical with what he had identified as Widdringtonites reichii (Et- tingshausen) Heer. These remains are abundant in the Atlantic Coastal Plain as far south as Alabama and have never been found with ovulate cones, although the staminate cones are not uncom- mon. Thus the slender conifer in the East is Widdringtonites. Whether Lesquereux's Glyp- tostrohus gracillimus also represents this genus or not I do not know. I suspect that in spite of minor and not very obvious differences Glyptostrohus gracillimus is none other than Sequoia condifa, which Lesquereux described in Hayden's report for 1874 from very incom- plete material collected at Fort Harker and Clay Center, Kans. The most abundant plants in the Cheyenne sandstone are graceful, delicate coniferous branches bearing numerous distinctive cones identical with the one referred to Sequoia condita by Lesquereux, as is the foliage, which has been recorded in the literature of the Cheyenne sandstone, as Glyptostrobus gracilli- mus (Ward) and Sequoia gracillima (Knowl- ton). The collections naturally contain many specimens of detached cones and many speci- mens of foliage lacking cones, but the evidence is as strong as it can possibly be that when cones and foliage are found in union in a dozen specimens those that are found separated in the same bed are none other. I have specimens of cones from six localities and of the foliage from thirteen localities around Belvidere. Moreover, the foliage shows considerable variation in the extent to which the leaves are pointed or obtuse, appressed or spreading, de- pending not only on a natural amount of variation but also on whether it represents shoots of the year or older twigs, and further- more the appearance differs greatly with the nature of the matrix, the extent to which iron salts have been deposited along the channels formed by the twigs, and other con- ditions. The specimens found in the sand- stones appear different from those found in the clays, and some specimens in the clays which were much incrusted suggested at first sight the genus Brachyphylhtm. Before describing the species as fully as the large collection studied permits, I would like to point out that Sequoia condita is not related to Widdringtonites, Juniperus, Glyptostrobus, or Sphenolepis — genera in which the foliage is comparable — and it is perfectly distinct from Sequoia fastigiata. It is known only from the Cheyenne sandstone and from the true Dakota of Kansas. In the absence of attached cones the foliage might be referred to any one of several genera, or its variants might be re- ferred to several different species in as many genera. Taken together, they demonstrate that it is a Sequoia, and I do not feel the slight- est doubt but that all the material from the Cheyenne sandstone represents a single botanic species. It may be described as follows: TAvigs rather rigitl, pinnately branched, slender elongate; covered with small, decurrent, crowded leaves varying from appressed to spreading falcate, thick and coriaceous, acute or obtusely pointed, slightly keeled but with- out vein. The leaves are arranged in a spiral phyllotaxy which becomes higher with the elongation of the twigs. In old twigs 2 to 3 millimeters in diameter they are scattered, spreading, and falcate. Their blunt tip is more apparent than real and is due to their usual or partial preservation in the form of casts. The variations in appearance are well shown in the accompanying figures. The cones vary from prolate to nearly spherical in FLORA OF THE CHEYENNE SANDSTONE OF KANSAS. 211 form and consist of about 22 .scales spirally arranged, as compared with about 30 in the cones of the modern redwood. The axis is stout and fusiform. The scales have a thin rounded peduncle exj)anding distad into a rhomboidal peltate tip only slightl}- wider than high, with a wrinkled marginal face surround- ing a central laterally elongated umbilicus. Length of cone (maximum), 2.4 centimeters; diameter (maximum), 1.8 centimeters. Aver- age size somewhat smaller. Length of scale (maximum), 8 millimeters; width, 6.5 milli- meters; height, 4.5 millimeters. These cones are remarkably like those of the existing red- wood {Sequoia sempervirens) in every respect — size of axis, shape of scales, etc. — except that the scales are less numerous in the fossils and the maximum size of the scales is about six- sevenths that of the average redwood scale. The average size of the fossil cones is from two-thirds to five-sevenths that of the modem cones. The condition of preservation of these cones is a strong argument in favor of the eolian character of the sandstone. All have the scales somewhat shriveled and widely separ- ated and are exactly comparable to thoroughly dried redwood cones. They are exceedingly abundant in the sandy phases of the Cheyenne sandstone, as if they had been blown about by winds and accumulated in hollows. I have not encountered them in the clays, although the claj's contain specimens of the foliage. Whether or not Sequoia condita occui's at any other horizons or localities is problematic and can be determined only by the best of evidence, for, as I have already stated, the foliage is duplicated more or less closely by a variety of unrelated conifers. In particular the Upper Cretaceous conifer known as Widdrhujionites suUilis, a form that I have not mentioned above, has foliage very like the more slender twigs of Stqitoia condita, and in the absence of cones I doubt if the two could be distinguished. However, a single specimen of Widdriiu/tonites suhtilis found in the Tuscaloosa formation of Alabama had small four-valved cones entirely unlike those of Sequoia condita. The Cheyenne sandstone localities are as follows: Cones and foliage, black hills near Belvidere (773) ; 1 i miles northwest of Belvi- dere (2218); Champion (Wildcat) Draw, three- fourths mile south of Belvidere (2222); near Medicine Lodge Creek, 2 miles west of Belvi- dere (2224) ; left bank of middle branch of Champion (Wildcat) Draw, half a mile south of Belvidere (2224) ; right bank of same draw (2231); Osage Rock, near Belvidere (7406). Foliage only, Stokes Hill, 100 yards south of National Corral (2219): ''Lanphier shales'' in Cliampion (Wildcat) Draw, three-fourths mile south of Belvidere (2223): 'Lanphier shales" in a draw 1 mile southwest of Belvidere (2225); hills between Spring CYeek and Soldier, 4 miles northeast of Belvidere (2227) ; Champion (Wild- cat) Draw, right (east) branch, half a mile south of Belvidere, in 'Lanphier shales" (2228) ; first draw west of Champion (Wildcat) Draw, half a mile south of Belvidere (2233). Family ABrETINEACEAE. Genus ABIETITES Hisinger. Abietites longifolius (Fontainej Berry. Plate XLVII, figure 3. Abietites longifolius (Fontaine) Berry, U. S. Nat. Mub. Proc, vol. 40, p. 315, 1911; Maryland Geol. Sur- vey, Lower Cretaceous, p. 407, pi. 67, fig. 7, 1911. Leptostrobus longifolius Fontaine, U. S. Geol. Sur\'ey Mon. 15, p. 228, pi. 101, fig. 2; pi. 102, figs. 1-4; pi. 103, figs. 6-12; pi. 104, fig. 6, 1890; in Ward, U. S. Geol. Survey Nineteenth Ann. Kept., pt. 2, p. 671, pi. 163, fig. 15; pi. 165, fig. 3, 1899; U. S. Geol. Survey Mon. 48, pp. 281, 481, 482, 491, 506, .528, 557, pi. 110, fig. 11, pi. 116, fig. 1, 1906. Leaves narrow, needlelike, 10 to 15 centime- ters in length, aggregated in bundles. Bundles apparently borne on short shoots, with many leaves in «ach bundle. Xo satisfactory vena- tion can be made out. Fontaine described a number of veins in these forms, but as nearly as can be determined these are simply folds due to compression or the angles of the leaf. Tills species has a considerable geologic as well as geographic range, having been recorded from the Kootenai formation of British Co- lumbia, the Fuson formation of tiie Black Hills, and the Potomac group in Maryland and Virginia. In the Potomac group it is of fre- quent occurrence and individually abundant, being found in the oldest as well as the young- est beds, but nuich more commonly in the lat- ter. The remains are always poorly preserved and were evidently much macerated before fossilization. They are closely comparable with Pinites solmsl Seward, of the Wealden, and with Pinus peterseni Heer, from the Kome 212 SHORTER CONTRIBUTIOKS TO GENERAL GEOLOGY, 1921. beds of Greenland. They appear to be iden- tical with specimens from the Atane beds of Greenland which Heer -* described as Pinus vaginalis. I have not, however, included the latter in the foregoing synonymy, as it is an earlier name and would involve changing the well-known and liighly characteristic name longifolius. These remains are very common in the Cheyenne sandstone. Similar forms under different specific names are common and wide- ranging at Lower and Upper Cretaceous hori- zons in North America, Europe, and Asia. The Cheyenne sandstone localities are black hills near Belvidere (77.3); IJ miles northwest of Belvidere (2218) ; Thompson Creek near the flume, 2 miles northwest of Belvidere (2221); Champion (Wildcat) Draw, three-fourths mile south of Belvidere (2222) ; 1 mile southwest of Belvidere (2225) ; left bank of middle branch of Champion Draw, half a mile south of Belvi- dere (2229) ; and right bank of same branch (2231). Abietites ernestinae Lesquereus. Abietites ernestinae Lesquereux, U. S. Geol. Survey Terr. Rept., vol. 6, p. 49, pi. 1, fig. 7, 1874. Pterophyllum haydenii Lesquereux (part), Am. Jour. Sci., 2d ser., vol. 46, p. 91, 1868. Lesquereux characterized this species as follows : Cone oblong, abruptly narrowed to a short pedicel, scales broad, truncate, appressed, and imbricated in spiral. This diagnosis obviously has nothing that would serve to set it apart from what might be written of dozens of fossil cone fragments of diverse relationships. The species was de- scribed from fragments collected near Decatur, Nebr., and similar cone fragments are present in the Cheyenne sandstone. They are not to be distinguished from other' so-called species which I have referred to the genus Abietites of Hisinger.^^ This genus is a convenient and useful reposi- tory for fossils, both stroliilar or foliar, whose real or fancied affinities are with the modern Abietinaceae. These range in age from the Keuper to the Pliocene, though the bulk came from the Cretaceous, and they comprise obscure impressions of foliage and cones, none of which have any real biologic value or present any !' Heer, Oswald, Flora fossilis arctiea, vol. 3, Abt. 2, p. 103, pi. 27, fig. 15b, 1874. « Hisingcr, W., Lcthaea suecica, p. 110, 1837. definite clue to their true relationship. Fon- taine has included in this genus fossils from the Triassic of North Carolina and various indefi- nite remains from the Trinity group of Texas, the Shasta series of California, the Lakota sandstone of the Black Hills, and the Potomac group of Maryland and Virginia. The Potomac fossils he segregated into four species, all of which were liased on obscure cone impressions and none of which possess much specific value. When it is remembered what diverse appear- ances may ])e assumed by a single species of cone, irrespective of individual variation, as a result of different stages, pf^jichaceration before preservation, of differences in the matrix, and of differences in thelUl'i^PUtyi and force of com- pression, it seems very probable that such forms can never be discussed satisfactorily. Similar forms ^om the English Wealden and later Cretaceous are described by Carru- thers, Gardner, Seward, and others and referred to the comprehensive genus Pinites of Endlicher (1847). They are in all probability con- generic if not specifically identical with Ameri- can forms referred to Abietites, and that name is preferable, as Pinites Endlicher is antedated by Pinites Witham, which was proposed for very different objects. Abietites cones are also common in the French and Belgian Cretaceous and have usually been referred to the genus Pin}is, although there is slight warrant for such a procedure. Abietites cones are rare in the Cheyenne sand- stone, being known only from Osage Rock, at Belvidere, in the "Stokes sandstone" below the so-called Champion shell bed at the base of the Kiowa shale (2232). Genus CUPKESSINOXYLON Goeppert. Cupressinoxylon cheyennense Penhallow. Cupressinoxylon cheyennense Penhallow, Roy. Soc. Canada Trans., 2d ser., vol. 6, sec. 4, p. 70, 1900 [1901]; Manual of North American gymnosperms, p. 238, 1907. This species was described as coming from the Cheyenne sandstone east of Stokes Hill, on the Kiowa-Baker County line, and was col- lected by Prosser. There is nothing to be added to the original description of this species, which was unil- lustrated. Nor is it worth while to quote that description, for it is very doubtful if the form could be recognized again, even by the author, FLORA OF THE CHEYEXNE SANDSTONE OF KANSAS. 213 short of comparison with the type sections. There is some doubt as to whether it came from the Cheyenne sandstone. 1 incUide it merely for the sake of completeness. In the case of Araucarioxylon prosseri, which Pen- hallow -" recorded from this re/nus Lesquereux,^" which is somewhat less similar to the known species of Sapindopsis; Sapindns diversifolivs Lesquereux," from Ells- worth County, Kans., which is also less similar to the known species of Sapindopsis; and RJtns powdliana Lesquereux,^' obtained near Fort Harker. Kans., wliich differs from Sapindopsis in the suliordinate lobing and small leaflets developed at the base of the proximal lateral leaflets, in these features resembling RTius, but which is sufficiently like Sapindopsis to be open to more or less suspicion. This species has been found in the Cheyenne sandstone at the black hills near Belvidcre (773); Osage Rock, Belvidere (2217, 2232, 7406) ; Stokes Hill 100 yards south of National Corral (2219); Stokes Hill (2220); Thompson Oeek near the flume, 2 miles northwest of Belvidere (2221); near Medicine Lodge Creek, 2 miles west of Belvidere (2224); left bank of middle branch of Champion (Wildcat) Draw, half a mile south of Belvidere (2229); shale along right branch of Champion (Wildcat) Draw (2228) ; and right bank of middle branch of Champion (Wildcat) Draw (2231). " Lesquereux, Leo, U. S. CeoL Survey Mon. 17, p. 133, pi. 15, flg.4, 1S92. » Idem, p. 93, pi. 16, ng.7. » Idem, p. l.'>2, pi. 55, flgs. 7-9. " Idem, p. l.V>, pi. M, flg. 18. "Idem, p. 15.5, pi. 56, flgs. 4, 5. 216 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1921. Sapindopsis brevifolia Fontaine. Plate LV, figure 1; Plate LIX, figure 1. Sapiruhpsis brevifolia Fontaine, U. S. Geol. Survey Mon. 15, p. 300, pi. 153, fig. 4; pi, 155, figs. 1, 7; pi. 163, fig. 3, 1890; in Ward, U. S. Geol. Survey Mon. 48, pp. 481, 482, 528, 1906. Berry, U. S. Nat. Mus. Proc, vol. 38, p. 644; Mary- land Geol. Survey, Lower Cretaceous, p. 4,73, pi. 87, figs. 2-5, 1911. Leaves odd-pinnate, the terminal leaflet considerably larger than the lateral leaflets, of which but two pairs are known. These are opposite. Leaflets somewhat crowded so that their margins often overlap, with subacute tips, varying in length from 2 to 5 centimeters and in width from 0.8 to 1.6 centimeters, aver- aging about 3 centimeters long by 1.3 centime- ters wide. Inequilateral toward the base and showing considerable variation in decurrence, even among the few specimens known; in some the rachis is conspicuously winged; in others the leaflets are all petioled, the whole having the aspect of some member of the Leguminoseae. Midribs stout; secondaries as- cending, camptodrome, seen with difficulty, as the leaf texture is coriaceous. This is a poorly marked species of infre- quent occuiTence at the same localities where the other species of this genus occur and may simply represent a variant of S. variahilis; in fact, there is no reason for considering it to represent a distinct botanic species, and the name is retained temporarily simply as a geologic convenience, to be eventually dropped entirely. The Cheyenne sandstone localities are 1| miles northwest of Belvidere (2218), Thompson Creek near the flume, 2 miles northwest of Belvidere (2221); near Medicine Lodge Creek, 2 miles west of Belvidere (2224) ; left bank of middle branch of Champion (Wildcat) Draw half a mile south of Belvidere (2229) ; Osage Rock, Belvidere (2232). Sapindopsis belviderensis Berry, n. sp. Plates XLIX-LIV. Leaves of variable size, pinnately compound, ranging in length (in the collected material) from 8 to 19 centimeters and in maximum width from 4.5 to 1 4 centimeters. These leaves are prevailingly odd-pinnate, but a few are even-pinnate. In addition to. the odd terminal leaflet generally' present there are invariably three pairs of lateral leaflets, wliich are gener- ally opposite but sometimes subopposite. These usually decrease regularly in size from the distal to the proximal pair. In some speci- mens the terminal leaflet is equilateral, but all the other leaflets are inequilateral, often markedly so. All except the terminal leaflet are invariably sessile, the latter being sepa- rated from the distal laterals in some of the larger leaves by a considerable interval of ra- chis. Generally, however, the terminal and upper laterals are confluent in the rachial re- gion to form what, if it were broken away from the balance of the leaf, would be considered to represent a palmately trilobate leaf such as is commonly referred to the genus Aralia. The sinuses may be rather broad, narrowly rounded, or pointed. The leaflets vary greatly in size, shape, and marginal characters but agree in being obtuse, generally abruptly and almost truncately mucronate pointed. The leaflets range in form from narrowly spatulate to broadly ovate or obovate. The margins are in- variably toothed, but there is great variation in the amount and degree to which the teeth are developed. Proximally the margins are entire for a greater or less distance. Above this en- tire portion the teeth, which are remote and rather evenly spaced, may be small and ser- rate or very prominent and dentate. Were not all sorts of gradations present one might well doubt that they pertained to the same plants. The accompanying illustrations show these variations much better than they can be described. The lateral proximal margins of the terminal pair of leaflets, except in a single specimen, are decurrent on the rachis, extend- ing dowTiward to the point of insertion of the next lower pair of leaflets and often contin- uous with the distal margins of these. This rachial wing may be broad and triangular, a form which, as the terminal leaflets are the largest, gives the leaf a curious unsymmctrical or artificial appearance. In other specimens the wings are narrow and become reduced to mere marginal hems. In the middle pair of lateral leaflets the proximal margins are only slightly if at all decurrent, and generally they are not decurrent. No decurrence has been observed in the lower lateral leaflets, but they as well as the middle pair have the proximal side of the base fuller than the distal side, the former being generally rounded and the latter FLORA OF TllK CHEYENNE SANDSTONE OF KANSAS. 217 incurved and sometimes disappeiirin;i; some dis- tance above the point of insertion. To judge by the lack of [)eti()hdes and the generally com- plete character of the material it does not aj)- pear that the leaflets were normally shed, and this is also indicated by the concrescence of the terminal leaflets. The leaves are cori- aceous and appear to have been stifl' and strict in habit. The rachis is stout and expanded prox- imad. The midribs are excessively stout and prominent on the lower surface. The seconda- ries are relatively thin, straight, and subpar- allel. They vary from camptodrome to cras- pedodrome. In the entire Viasal i)art of the leaf and sometimes in the apex they are camp- todrome. In many specimens one secondary runs to each marginal tooth, although in other specimens the camptodrome habit is retained and a short branch enters the marginal tooth. All these features are indicated in the accom- panying figures. The tertiary venation is usually obsolete, as the matrix is prevailingly coarse. Occasionally percurrent nervilles are seen. In specimens with broadly winged stripe the venation of the leaf is continued in these wings. This handsome species is represented bj- a large amount of material, which is fortunate, as it would be almost impossible to correlate fragmentary material. It is clearly a repre- sentative of the genus Sapindopfiis and would well merit the specific name of variabilis had that not already been used for the tyj)e of the genus, which came from the Patapsco forma- tion of Maryland and Virginia. In the Patapsco formation the genus Sopin- dopsis may be totally absent from a locality or present in the greatest abundance, and this is equally true of the Cheyenne sandstone of Kansas, indicating possibly a gregarious habit. Various species of existing Sapindaceae show similarities to the present species in form, venation, and variation. All the previously described species of Sapindopsis had entire margins, and no trace of toothed margins has been found in the material from the Atlantic Coastal Plain. The existing genus Matayha Aublet, with which I originally compared Sapindopsis, has leaves with both entire ami dentate margins, and tlie general features of Sapindopsis are sliared by other tropical American genera of vSapindaceae. The genus Matayha comprises about two score existing species and is ciosel}- related to Cupania, also exclusively American in the e.xisting flora — in fact, all the genera of the tribe Cupaiiieae lomatorrliizae as segregated by Kadlkofer are confined to the warmer regions of the Western Hemisphere. Occurrence: Localities 2221, 2224, 2229, 2230, 7406, Medicine Lodge Creek, in draw 3 miles above Belvidere (Cheyemie sandstt)ne No. 3 of Hill) ; collected i)y Waid and \'aughan, October 18, 1896 (unnumbered). Order MALVALES. FamUy STERCULIACEAE. Genus STERCtTLIA Linng. Sterculia towneri (Lesquereux) Berry. Plate LVII, figure 1 ; Plate LX; Plate LXI, figure 1. Aralia lounwri I.esquereux, U. S. Geol. and Geo;;. Survey Terr. Bull., vol. 1, p. 394, 1875 [1876]; Ann. Rept. for 1874, p. 349, pi. 4, fig. 1, 1876; Cretaceous and Tertiary floras, p. 62, pi. 6, fig. 4, 1883; Flora of the Dakota group, p. 132, pi. 23, figs. 3, 4; pi. 31, fig. 1, 1892. Slcradia drakei Cummings, Texas Geol. Survey Third Ann. Rept., p. 210, fig. 8, 1892. Knowlton, in Hill, Am. Jour, ."^ci., 4th ser., vol. 1, p. 213, 1895. Surculia snotrii Lesquereux, Flora of the Dakota group, p. ISli, 1)1. 30, fig. 5; pi. 31, figs. 2, 3; pi. 32, figs. 1-4, 1892. Hollick, U. S. Geol. Survey Men. 50, p. 94. pi. 34, fig. 20, 1907. Amiin lowneri Hollick, \ew York Acad. Sci. Trans., vol. 16, p. 132, pi. 14, figs. 11, 12, 1897. P-erry, New York Bot. Garden Bull., vol. 3, p. 92, 1903. I have long thought that the Aralia toirneri and Sterculia snomi of Les(iuereu.x represented a single species but have never had a chance to test this belief until I received the present collections from the Cheyemie sandstone, in which this is one of the most abundant forms. It shows considerable variation in size but obviously represents a single botanic species, l^nfortunately the name towneri ante- dates snoirii by some 15 years, so that the latter, which is much the better known of the two, becomes a synon3'm. From the large amount of material now available the species may be described as follows : Leaves of variable and often very large size, palmately two to seven lobed. The lobes are prevailingly conical and acuminate, occa- 218 SHORTER CONTRIBUTIONS TO GENERAL. GEOLOGY, 1921. sionally widening somewhat medianly and less acutely pointed, separated by generally open and rounded sinuses extending about halfway to the base. The angles that the lobes form with one another and the form of the sinuses vary with the number of lobes, as does also the character of the base, which ranges from trun- cate to decurrent. The median lobe is gen- erally slightly wider than the others but may be smaller. The normal form is five lobed like the smaller of the two specimens from the Cheyenne sandstone here figured. The texture is so coriaceous that these leaves are well pre- served in the scarcely consolidated wind-blowm sand of the Cheyenne. The margins are entire. Length from S to 20 centimeters; maximum .width from 6 to 24 centimeters. Petiole stout, usually broken away, 12 centimeters long in a medium-sized leaf figured by Lesquereux. Midrib stout, channeled, prominent on the under side of the leaf. An equally stout lateral primary diverges from the midrib, usually at its extreme base but occasionally slightly above. In the five-lobed forms this primary forks almost immediately into two subequal branches, which form the midveins of the respec- tive lobes. In specimens having more than five lobes the additional ones are suljordinate to the basal laterals, their midveins diverge at an acute angle from the midveins of these laterals, and their separating sinuses are less deep. The secondaries are thin and immersed in the leaf substance and are largely obsolete in the Chey- enne sandstone specimens; they are numerous, regularly spaced, subparalleled, and campto- drome in the lobes and in curved anastomosing loops in the body of the lamina. This is an exceedingly well marked species and, like most Sterculias, both ancient and modern, shows the characteristic variability of the genus. It was described originally from material collected in the Dakota sandstone of Kansas and occurs in the Big Tucumcari Mountains of New Mexico in beds referred to the Dakota. It is recorded from the Magothy formation of Massachusetts and New Jersey. In the Cheyenne sandstone of Kansas it occurs at these localities: Black liills near Belvidere (773); Osage Rock, Belvidere (2217); Stokes Hill, 100 yards south of the National Corral (2219); Thompson Creek near the flume, 2 miles northwest of Belvidere (2221) ; near Medicine Lodge Creek, 2 miles west of Belvi- dere (2224) ; left bank of middle branch of Champion (Wildcat) Draw, half a mile south of Belvidere (2229); draws north of Belvidere, in "Lanphier shales" (2230); Osage Rock, in "Stokes sandstone" (2232); Wildcat Draw (7405). Sterculia mucronata Lesquereux. Slerculia mucronata Lesquereux, U. S. Geol. Survey Mon. 17 (Flora of the Dakota group), p. 182, pi. 30, figs. 1-1, 1892. Leaves coriaceous, prevailingly small, pal- matel}' three to five lobed. Lobes entire, conical, separated by open rounded sinuses extending a variable distance, sometimes over halfway to the cuneate or truncate base. Petiole long and stout. Primaries three from the top of the petiole, stout and prominent. In the five- lobed forms subordinate branches from the lateral primaries furnish these with midveins. Secondaries tliin, camptodrone. The tips of the lobes are prominently mucronate, and this feature, which suggested the specific name, is especially obvious in the Cheyenne sandstone specimens, where the mucros are 2 millimeters long and perhaps merit the designation cuspi- date rather than mucronate. The fact that these leaves are prevailingly small suggests that they probably represent small leaves of the associated Sterculia towneri, with which they agree in their main features — the mucronate tips of S. mucronata being the principal differential characteristic. The leaves originally described were obtained from tlie Dakota of Ellsworth County, Kans., and the species is known only from that region and the Cheyenne sandstone of southern Kansas, al- though there is a similar but distinct species, Sterculia minima Berry,'' in the Magothy forma- tion of New Jersey and Maryland. Two speci- mens were found in the Cheyenne sandstone near Medicine Lodge River, 2 miles west of Belvidere (2224). « Berry, E. W.. Maryland Gcol. Survey, Upper Cretaceous, p. S.57, pi. so, figs. 1-3, 1910. FLORA OF THK CHEYENNE SANDSTONE OF KANSAS. 219 Order THYMELEALES. Family LAURACEAE. Genus SASSAFRAS Linnfi. Sassafras mudgii Lesquereux. Plate LXI, figure 3. Sassafras mudgii Lesquereux, Am. Jour. Sci., 2d ser., vol. 4C, p. 99, 18G8; U. S. Geol. Survey Terr. Rept., vol. 6 (Cretaceous floral, p. 7S, pi. 14, fi^'S. 3, 4; pi. 30, fig. 7, 1874. , Ward, IJ. S. Geol. Survey Nineteenth Ann. Rept., pt. 2, p. 705, pi. 170, fig8.'4, 5; pi. 171, fig. 1, 1899. Berry, Bot. Gaz., vol. 34, p. 437, 1902. ?Kurtz, Mus. La Plata Rev., vol. 10, p. 53, 1902. According to Newberry, this is merely a variety of his Sassafras cretaceum, but I fail to see any ground for this association except that it resembles somewhat the narrowor-lobed leaves ascribed to that species. It is somewhat intermediate between these forms and the more typical Sasfsafras acvtUohum but is much more like the modern leaf than either. Los- quereux's figures 3 and 4 of Plate XIV of the "Cretaceous flora" I consider to represent typical forms of this species. In the lengthen- ing of the terminal lobe it approaches the modern Sassafras; and it shows no venation characters which are unlike the modern leaf, for although no marginal veins are discernible, they might have been present in the specimen illustrated in Lesquereux's figure 3, as they are in the identical form from the Cheyenne sand- stone figured on the accompanying plate, and both specimens approach Sassafras in the relations of their secondary members in this region. If it is certain that tlu* fruit has been found in the same strata, as Lestjuereux" asserts, it only serves to substantiate the impression otherwise obtained that they are true Sassafras leaves. The lateral margins of both the base and the lobes are straighter and more ascending than in the existing Sassafras, and the margin shows a tendency to become wavy. Lesquereux's other figured specimen referred to this species differs in the size and direction of the lateral lobes, in the subbasal primaries, and in the acute tip: the venation also is somewhat dissimilar, the a-scending margins bulge outward, and the base is not decurrent on the petiole, as it is most markedly in the specimens shown in his figures 3 and 4. It resembles somewhat the forms wliich Now- « Lesquereu.\, Leo, Flora of the Dakota group, p. 230, 1891 (18921. berry refei-s to Sassafras aciitilobum. Ward's fragmentary leaves from the Black Hills are of doubtful identity. The more perfect speci- men that he originally referred to L'tndera venusta Les(iuiM'eux, wliich it resembles in outline, is a smaller leaf than S. mudgii, with subbasal |)rimaries, considerable breadth of blade, and reduced terminal lobe. Sassafras mudgii was based on material col- lected from the hills along vSaline River in cen- tral Kansas. Up to the present time it has never been found elsewhere, except for the above-mentioned doubtful record by Ward from the supposed Dakota sandstone at Evans quarry, in South Dakota, and a still more doubtful South AmericiU' record by Kurtz that may well be entirely ignored. It may be that the type was from the Mentor formation of central Kansas rather than from the true Dakota sandstone, as the species has never been found in collections from the Upper Cretaceous of the Atluiitic Coastal Plaiji, but no outcrops of the Mentor formation are known as far north as Saline River. The Cheyenne sandstone occurrences are Stokes Hill (2220) and near Medicine Lodge Creek, 2 miles west of Belviderc (2224). Order UMBELLALES. FamUy ARALIACEAE. Oenus ARALIA Linn€. Aralia ravniana Heer. Plate LVIII; Plate LIX, figure 4. Aralia ravniana Heer, Flora fossilis arctica. vol. 6, .\bt. 2, p. 84, pi. 38, figs. 1,2, 1882. Berry, New York Bot. Garden Bull., vol. 3, p. 92. pi. 46, fig. 7; pi. 53, fig. 2; pi. 57, fig. 1, 1903; Torrey Bot. Club Hull., vol, 31, p. 79. 1904; vol. 37, p. 27, 1910; .Maryland Geol. Survey, ITpper Cretaceous, p. 876. i>l. S2, fig. 4; pi. 83, figs. 1-1, 1916. Aralia (jroenlandlai Heer. idem. pi. 46. fig. 17. fSterculia snowii Hollick, New York Acad. Sci. Annals, vol. 11. p. 422, pi. 37, fig. 4. 1898. This species was described by Heer from material collected in the Greenland Upper Cretaceous (Atane beds) and has been found by me in the Magothy formation of both New Jersey anil Maryland. The fragments from Marthas Vineyard, Mass., and Tottenville, N. Y., identilied as this species by Hollick," are not this species, m my judgment. There is a » Hollick, Arthur. U. S. (Icol. Survey Man. 'lO. p. !», pi. 37. rtgs. 1, 2, 1907. 220 SHORTER CONTRIBUTIONS TO GENERAL, GEOLOGY, 1S)21. great display of Aralia-\ike forms in the middle Cretaceous both of this country and of Europe, and these forms are especially abundant in the Dakota sandstone of the West. Compari- sons with existing plants are not so satisfactory, although many tropical Araliaceae show sug- gestive resemblance. The Moraceae in the genus Artocarpus and its allies also show many similar features. This most striking species of Aralia, because of its large size, has always been found in a fragmentary condition. Specimens showing all parts of the leaf have now been collected both from Maryland and from Kansas, and these Conclusively confirm the restoration of this leaf made bj^ me in 1903. They also confirm the supposition based on the venation of the New Jersey material, that instead of a broadly ovate median lobe, as Heer supposed, this middle lobe was sublobate by the greater or less de- velopment of a lateral lobe on each side, as shown in the accompanying illustrations. The species may be more fully defmed in the light of all the material as follows: Leaves large, ranging from 16 to 21 centimeters in length and from 19 to 23 centimeters in maximum width, orbicular in general outline, deeply pinnate- lobate. Apex of the terminal and lateral lobes bluntly pointed. Base broadly cuneate. Mar- gins entire. Texture subcoriaceous. Lobes usually seven, separated by relatively narrow ultimately- rounded sinuses, comprising an ovate medium terminal lobe and two main lateral lobes on each side, the lower pair being more or less divided. In the Maryland mate- rial the auxiliary lobe on the lower side of each main lateral lobe is feebly developed. In the Greenland material it is at least half as large as the main lobe, and the separating sinus extends halfway to the base. Petiole stout, its full length unknown. Midrib very stout and prom- inent, straight. Lateral primaries two on each side, stout and prominent, the lower pair sub- opposite and suprabasilar, the upper pair in some specimens subopposite, more commonly separated by a wide interval. The lower primary may fork a short distance above its base, as it does in the Greenland material at an interval of only about 1 centimeter, or this fork may be at least 4 centimeters above the base, as in the Maryland material, the distance depending on the extent to which the auxiliary lobe is developed. The angle of divergence of the primaries from the midrib is about 40° but varies from specimen to specimen; the basal pair is in general somewhat more divergent than the upper pair. The secondary and tertiary venation is usually obsolete. Some specimens show a few thin remote secondaries diverging from the primaries at angles of about 45° and sweeping upward in ascending camptodrome curves. The Cheyenne sandstone material is not abundant. It comes from the left bank of the middle branch of Champion (Wildcat) Draw, half a mile south of Belvidere (2229) and the right bank of the same branch (2231). Aralia newberryi Berry. Aralia newberryi Berry, Torrey Bot. Club Bull., vol. 34, p. 201, pi. 1.5, fig. 1, 1907; New Jersey Geol. Survey Bull. 3, p. 197, 1911. Aralia palmata Newberry, Flora of the Amboy clays, p. 117, pi. 39, figs. 6, 7; pi. 40, fig. 3, 1896 (not Lamarck). Berry, New York Bot. Garden Bull., vol. 3, p. 93, pi. 44, 1903; Torrey Bot. Club Bull., vol. 31, p. 79, pi. 4, fig. 12, 1904. Aralia rotundiloha Hollick, New York Acad. Sci. Annals, vol. 11, p. 421, pi. 38, fig. 2, 1898. Aralia polymorpha Newberry, Flora of the Amboy clays, p. 118, pi. 39, figs. l-.j, 1896. Aralia sp. Hollick, New York State Mus. Ann. Rept., vol. 55, p. 155, 1903. Leaves very variable in size and outline, palmately three to five lobed. Lobes conical, obtusely rounded. Sinuses open, shallow, rounded. Margins entire, somewhat undulate basally. Petiole long and stout. Midrib stout, more or less curved or flexuous. Primaries three to five, from the base, prominent, run- ning to the tips of the lobes. Secondaries very slender, camptodrome. The middle lobe is usually longest and broadest, and the basal lateral lobes may be reduced to subordinate and but slightly marked parts of the main lat- eral Igbes. The relative development of the apical or basal lobes and the depth of the intervening sinuses greatly alter the appearance of these leaves. Some are symmetrical and others de- cidedly unsymmetrical ; some are preeminently three lobed and sublobate and others are five lobed with additional incipient lobes. The variations are almost exactly comparable with the similar variations in the leaves of the mod- ern Sassafras, Stercnlia, and Araliaceae. There seems to be no basis for maintaining the distinctions between the forms united in FLORA OF THE CHEYENNE SANDSTONE OF KANSAS. 221 the foregoing synonymy. The Cheyenne sand- stone material is more like the irregular .4. poh/morjiha than the more symmetrical A. palwata of Newberry's original material. The species is common in the Raritan forma- tion of New Jersey and survives in the over- lying Magothy formation. In the Cheyenne sandstone it is represented by three specimens obtained near Medicine Lodge Creek, 2 miles west of Belvidere (2224). Genus ARALIOPSOEDES Berry. Araliopsoides cretacea (Newberry) Berry. Plate LXI, figure 2. Araliopsoides cretacea (Newberry) Berry, Maryland Geol. Sur\-ey, Upper Cretaceous, p. 879, pi. 74, fig. 3; pi. 84, figs. 1, 2; pi. 8.^), figs. 1-.5; pi. 88, figs. 1-3, 1916; Torrey Bot. Club Bull., vol. 38, p. 413, 1911. Sassafras (AraliopsU) cretaceum Newberry, New York Lye. Nat. Hist. Annals, vol. 9, p. 14, 1868. [Lesquereux], U. S. Geol. and Geog. Survey Terr., Illustrations of Cretaceous and Tertiary plants, pi. 6, figs. 1-4; U. S. Geol. Survey Terr. Kept., vol. 6 (Cretaceous flora), p. 80, pi. 11, figs. 1, 2; pi. 12, fig. 2, 1874; U. S. Geol. Survey Mon. 17, p. 102, 1892. Newberry, U. S. Geol. Survey lion. 35, p. 98, pi. 6, figs. 1-4; pi. 7, figs. 1-3; pi. 8, figs. 1, 2, 1898. ?Hollick, U. S. Geol. Survey Mon. 50, p. 77, pi. 30, fig. 10. 1906. Penhallow, Roy. Soc. Canada Trans., 3d ser., vol. 1, sec. 4, p. 310, 1907. Berry, Torrey Bot. Club Bull., vol. 37, p. 22, 1910. Leaves petiolate, decurrent at base, very smooth above, strongly nerved below, three lobed ; lobes entire and acute. The nervation is all strongly defined; the central nerve straight or nearly so; the lateral primary nerve springing from it at an angle of 30°; secondary nerves regularly arched till they approach the margin of the lobes, when they are abruptly curved and run together. From these the tertiarj- nerves are given off at a right angle, and from these the quaternary nerves spring at a similar angle, together forming a network of which the areoles are sub- quadrate. — Newberry, 1868. Newberry includes under Sassafras cretaceum the various forms described by Lesquereux as S. mxtdfjii, S. subinte()rifoliiim, S. intecjrifoHum, S. obtusujn, S. cretaceitm dentatum, S. cretaceum obtusum, S. acutilobum, Cissites harkianus, and C. saJisbi/riaefoliits. Although this list shows the undoubted composite nature of iS'. cretaceum, it also shows that the extremes of leaf form above mentioned are so closely connected with the more typical leaf by a series of interme- diate forms that the problem of where one species shall enJ I'LATE XLVilt CONES AM) Fiil.Uc:!': i>r SKQUOIA CONDIIA l.KSQUKREUX. From f'hnyrtitu! siuidslorK! iit. scv(!riil localiliea near Iti-lviiicre,. Kaus. r. S. (:EOLOf;lCAI. SURVEY PROFESSIONAL PAPER 129 PLATE XLIX SAPINDOPSIS BELVIDERENSIS BERRY. Krom Cheyenne sandstone near Relvidere. Kans U. S. CKOIOCICAl. SIRVEY rnoi-EssioNAi. taper 12 SIKVEY PROFESSIOXAL PAPER 129 PLATE LVII FOSSIL PLANTS FROM THE CHEYENNE SANDSTONE. 1, Sterculia towiieri (Lesquereux) Berry; 2, Saptndopsis mai^nirolia Fontaine. Collected near Relvidere, Kans. U. S. GEOLOGICAL SIKVEY PROFESSIONAI. I'AI'ER 129 ri-\.TK I.VIII MtM.lA It.WMANV lllCEK. From Cheyoiiiic yaiuislono near llclvidrrtr, Kans. AImiiiI. roiir-liri}id naluriil sixe. L:. K. C.KOI.OGICAL Sl'RVEY PROFESSIOXAL TAPER 129 PLATE LIX FOSSIL PLANTS FROM THE CHEYENNE SANDSTONE. 1. 2. Sapindopsis brc.vifr.lia Fontaine; 3, Sapindopsis raagnifolia Fontaine; i. Aralia ravniana Heer. All collecl,-.! ..ear Belvidere. Kans. V. S. GEOLOGICAL SURVEY PROFESSIONAL TAPER 129 PLATE LXI FOSSIL PLANTS FROM THE CHEYENNE SANDSTONE. 1. Steroulia lowneri (l^squereux) Berry; 2, Araliopsoides cretacea (Newberry) Berry; 3, SassafrMs iiinilgii Lesqiiereux; 4, Carpolithus helvidoreiisis lierry. All (nllecled near Belvidere, Kans. INDEX. A. rage. A bietUes erncatinae Lesqiiereux, description of 212 longifoliius (Fontaine) Berry, description of 211-212 plate showin^i; , 225 Accuracy in compiitations, excessive 41 Ackjiowledgment for aid 22 Akinetic surface, proposed use of term 8 Alinement charts, nature of 42 prei)aration and use of 44-46, 48, 49-50 Allen, E, 0 ., acknowledgment to 22 Andromeda novaccacsanae Hollick, description of 177-178 notaccacsarcae, plate showing 181 snowii Losquereux, description of 178 Anomalina bilatcralis Cushman, n. sp., description of 97-08,137 bilateralUf, plate showing 115 grosserugosa (Gumbel)H. B.Brady? var., description of. 98 plate showing U5 mississippiensis Cushman, n.sp., descriptions of 98, 137 plate showing , 115 vicksburgensis Cushman, n. sp., description of 137 plate showing 152 Antelope Hill, Ariz., sandstone composing 188 Antelope Wash, Ariz., section on 70 Aralia newberryi Bevvy, description of 220-221 ravniana Heer, description of 21^220 plate showing 225 saportana Lesquereux?, description of. 177 tcellingtoniana Lesquereux, description of 176-177 plates showing 181 Araliopsoides cretacea (Newberry) Berry, description of 221-222 cretacea, plate showing 225 Arizona, northwestern. See Mohave County, Ariz. Arthurs BIulT, Tex. , fossil plants collected at 153 Articulina bt/ramen^is Cushman, n. sp., descriptions of 103, 143 bt/ramen3is, plate showing 121 Arundn groenla ndka Heer?, description of 213 Aspknium dicksojiianum Heer, description of 207-208 Asterigerina subacuta Cushman, n. sp., description of ICK) subacuta, plate showing 118 B. Bancroft, Howland, cited 185 Basalt resting on a conglomerate of basalt and other boulders near Toquerville, Utah, plate showing 67 Bassler, Harvey, Reeside, John B., jr., and, Stratigraphic sections in southwestern Utah and northwestern Arizona 53-77 Beaver Dam Mountains, Utah, section in 76-77 Bedding planes, classification of, difficulties in 5 " Bellerophon limestone," fossils collected in 66 Bend series, correlation of well sections with subdivisions of 15-16 difference between shales in 8-9 distinction between black shale and black limestone in 9 investigation of, dilTiculty with staining matter 2 flint treated separately in 2 lithologic correlation in, inlro*luction to discussion of 6-7 nature of oil-producing beds in 20 object and methods of the investigation of 1-5 position of oil sands in 18-20 " Smith wick lime" in well borings from 16 summarj' and conclusions on 20-22 time required for lithologic work on 20 unit A of well borings from 7 unit B of well borings from 7-9 units C andC of well borings from 9-10 unit D of well borings from 10-1 1 units E, F,and O of well borings from 11-12 units U and 1 of well borings from 12-13 unit J of well borings from 13 Page. Bend series, units K and L of well borings from 13-15 volcanic origin of material from 12 white shale in well borings from 12 Benzoin venustum ( Ix-squereux ) Knowlton, description of 171-172 vtnuatum, plate showing \%\ Berry, Edward Wilber, fossils determined by 153-154 The flora of the Cheyenne sandstone of Kansas 199-225 The flora of the Woodbine sand at .\rthursBlufI, Tex 153-181 Big Horn Mountains, Ariz., hills of chlorilic schist at north end of , plate showing iM BMocuHna orjiata D'Orbigny, description of 143 omata, plate showing ' 150 ap.?, description of 105 Bp.?, plate showing 122 Bingen sand, relation of flora of, to that of the Woodbine sand.. 156-157 Black Butte, Ariz., plate showing 185 Black Rock Canyon, Ariz., section in 70 Black Rock Spring, Ariz., section near 75 Bloomington, Utah, sections on Virgin Rivernear 75-76,77 Bloomington dome, Utah, section at 74 Bolivina amygdalaeformia B.. B. Brady, description of 91 amy^dflfaf/ormw, plate showing 109 fooftri Cushman, n. sp., description of 126 plate showing 146 frondfa Cushman n. sp., description of 126-127 plate showing 146 mlssissippiensia Cushman n. sp., description of 92 plate showing 109 nUida H. B. Brady, description of 91 plate showing 109 Tobiista H. B. Brady, description of 91-92 vicksburgensis Cushman, n. sp., description of 126 plate showing 146 cf. B, punctata U'Orbigny, description of 126 Brachyphyllum macrocarpumformosum Berry, description of. . . 160-161 macrocarpum formosum, piate showing 181 BuHminn ovata D'Orbigny?, description of 92 Qvata, plate showing 110 pupnides D'Orbigny, description of 127 plate showing 146 BuUmindla cnntraria (Reuss) Cushman, description of 128 subteres H. B. Brady var. angusta Cushman, n. var., descrip- tion of 127-128 plate showing 146 BuUrush, Ariz., section near 70 Byram, Miss., fossils found near 83-85 Byram calcareous marl, deposition of 88 expos*ires of, at Byram, Miss 81 at Woodwards, Miss 82 near Vicksburg, .Miss SO-81 on Leaf River, .Miss K2 fauna of, descriptions of 89-105 occurrence of 81, 82-85 relationships of 87-88 feat nrcs of 79-*0 foraminifera found in, at the type station 124-125 C. Caliche, occurrence of, in the lower G ila region , Anz 190 Carpolithus behiderensu Berry, n. sp., description of 224 beliiderensis, plale showing 225 sp. 1, description of 179 sp. 2, description of 180 plate showing ISI sp. 3, description of ISO plate showing IM Cartesian coordinates, disadvantages of 41 Casey, T. L., cited 79 227 228 INDEX. Page. CassidultTia crassa D'Orbigny , description of 12S Charts for stratigraphic computations, preparation and use of; 44- 45, 45-46, 48. 4aicliino adcrna Cushman, n. sp., description of 135 adicna, plate showing - • • '^^ Percentage log, description of 1-2 Phosphate rock, conditions affecting deposition of 4 Plata nus latior ( Lesquereux ) Knowlton, description of 1W-J6S Plummer, F. B., cited *. 1* PodozamUes lanceolatua (Lindley and Hutton) F. Braun, descrip- tion of 159-160 lancrolalus, plate showing 181 Polymorphina adicna Cusbnmn, n. sp., description of 132 advena, plate showing H* a mygdalnides ( Rcu is ) Reuss, description of 95, 132 plate showing 1 12 230 INDEX. Page. Polymnrphina hyramensh Cushman, n. sp.. description of 94, 131 plate showing HI cuspidata H. B. Brady, description of 133 plate showing 1^7 vaT.coi>tulata Cushman, n. var 133 plate showing 1^8 equalis D'Orbigny, descriiition of 132 plate showing I'^S gibba D' Orbigny, description of 93-94, 132 plate showing HI fistulose form, description of ^* plate showing 112 problrmti D'Orbigny ?, description of 94-95, 131-132 plate showing 112 regiTia H . B . Brady, Parker, and Jones, description of 94, 131 plate showing 112, 147 spinosa iW Orhi^n J ) Egger, description of 133 plate showing 148 vicksburgensis Cushman, n. sp., description of 133 plate showing I'lS Populus harkeriana Lesqaereux, description of 163 Pulvinulina advena Cushman, n.sp., description of 99 adiena, plate showing 116 ftyramrnsw Cushman, n.sp., description of 99,138 plate showing 116 glabrata Cushman, n. sp., description of 99, 138 plate showing 116 Q- QuiTiqiielocitlina bicostata D'Orbigny, description of 141 bicostata D'Orbigny var., description of. ., 102 var., plate showing 120 contorta D'Orbigny, description of 142 plate showing 151 cookei Cushman, n. sp., description of 141 plate showing 150 crassa D'Orbigny ?, description of 102 plate showing 121 cuvieriana D' Orbigny, description of 102, 141 plate showing 120 glabrata Cushman, n. sp., description of * 141 plate showing 151 lamarrkiana D'Orbigny, description of 142-143 lustra Cushman, n. sp., description of 141-142 plate showing 150 seminulum (Linnaeus ) D'Orbigny, description of 142 iesseilata Cushman, u. sp., description of 142 plate showing 150,151 venusta Karrer ?, var., description of 102-103 plate showing 120 vicksburgensis Cushman, n.sp., description of 141 plate showing 151 vulgaris D' Orbigny, description of 142 plate showing 149 sp., description of 103 plate showing 120 R. Red Bluff clay, fossils found in 82, 83-85 Red wall limestone, features of, in Utah and Arizona 56-57 Reeside, John B., jr., and Bassler, Harvey, Stratigraphic sections in southwestern Utah and northwestern Arizona 53-77 Reeves, Frank, cited IS Rhamey Hill, Tex., fossil plants collected from 153 Rhamjius tenai Lesquereux, description of 169-170 tenai, plate showing 181 Rhus redditiformis Berry, description of 169 reddUiformis, plate showing 181 Rhythm in movements of ancient seashores 201 Rock Canyon Ariz., section at head of 70-71 section at mouth of 71 Rock Canyon conglomeratic member of the Moenkopi formation, fossils collected from 67 Ross, Clyde P., Geology of the lower Gila region, Ariz 183-197 Eotalia bi/ramen^is Cushman, n. sp., description of 99, 138 byTamensis plate showing 117 Page. Rotalia dentata Parker and Jones, description of 100 plate showing 117 var. parva Cushman, n. var., description of 139 plate showing 152 vicksburgeiisis Cushman, n.sp., description of 139 plate showing 152 Roxana Petroleum Corporation, acknowledgment to 22 Rudd No. 1 well, location of 5-6 log of, plate showing In pocket. section of, correlated with type section 15 sections of, thicker than corresponding sections of Seaman weU 6,7,17 S. Saddle Mountain, Ariz., plate showing 188 St. George, Utah, sections north, east, and south of 73-75,76-77 Salii deleia Lesquereux, description of 163 ksguereuxii Berry, description of 162 Sapindopsis bcl cider eiisis Berry, n. sp., description of 215-217 belvidrnn.'iis, plate showing 225 brevi/oUa Fontaine, description of 216 plate showing 225 •magni/oHn Fontaine, description of 214-215 plate showing 225 variabilis Fontaine, description of 213-214 plate showing 225 Sapindu^ morrisoni Heer, description of 168 Sassafras mudgii Lesquereux, description of 219 mudgii, plate showing 225 Seaman No. 1 well, location of 5-6 log of, comparison of synthetic log with 17-18 plate showing In pocket. oil in 12 sections of, thicker than corresponding sections of Rudd well 6, 7, 17 Sequoia condita, Lesquereux, description of 209-211 condita , plate showing 225 gracillima Newberry, erroneous naming of 209-210 Shinarump conglomerate, features of, in Utah and Arizona 62 north of Virgin City, Utah, plate showing 67 Siphonina advena CLishman, description of 98, 137 advena, plate showing 116 Smith's Mesa, Utah, section in 73 view northward toward, plate showing 59 Spirillina Umbata H. B. Brady var. bipunctata Cushman, n. var., description of 134-135 Umbata var. bipunctata, plate showing 149 subdecorata Cushman, n. sp., description of 95 plate showuig 113 Spiroloculina antHlarum D'Orbigny, description of 140 antillarum, plate showing 150 bip-amensis Cushman, n.sp., description of ." 101 plate showing 119 grateloupi D'Orbigny, description of 101 plate showing 119 imprimata Cushman, n. sp., description of 101-102, 140 plate showing 119 SterciLlia lugitbris Lesquereux?, description of 171 lugubris, plate showing 1 181 mucronata Lesquereux, description of 218 towneri (Lesquereux) Berry, description of 217-218 plate showing 225 Stratigraphic units, use of term 5 Strovibus costatus, plate showing 34 Sulphide, conditions atlecting deposition of 4-5 Supai formation, features of, in Utah and Arizona 57-58 Synthetic log, description of 3 T. Tertiary (?) sandstone, features of, in Utah 65 Tertiary (?) and Quaternary rocks, features of, iu Utah and Arizona 66 Texas, north-central, generah^ed log for lH-20 generaUzed log for, plate showing In pocket. Teitularia agglutinam D'Orbigny, description of 89 agglutiiians, plate showing 108 INDEX. 231 Page. Tcxtularia folium Parker andJoiies, description of 90-91 plate showing 108 missitsippiciisis, description of 90, 125-126 plate showing 108 suhha uerii Cushnaan, description of 89-90, 126 piate showing 108 tumidulum, description of 89, 125 plate showing 109 Thickness of strata, graphic computation of, alinement chart lor.. 44-46 graphic computation of, data needed for 40 geometric construction for 40-41 trigonometric formula for 41 graphic representation of 41-43 mathematical anal>'sis of 43-44 Toroweap Canyon, Ariz., section iii 69-70 Triangle, right, graphic solution of 48-50 right, gra[)hic solution of, preparation and use of chart for 49-50 graphic solution of, aUnement chart for 50 Tricalycites papyraccas HolUck, description of 179 papyraceus, plate showing 181 TTiloculina oblanga { Montagu) D'Orbigny, description of 104-105 oblonga, plate showing 122 peroblonga CiLshman, n. sp., description of 143 plate showing 151 rotunda D'Orbigny, description of 104 sculpturaia Cusiiman, n. sp., description of 143 plate showing 150 trigonula (Lamarck) D'Orbigny, description of 105 Trochodendroides Berry, n. gen., description of 166 rhomboidetis (Lesquereux) Berry, description of. 166-167 plate showing 181 Truncaiulina a mericana Cushman, description of 97 americana, plate showing 114 var., description of 136 byrameTisis Cushman, n. sp., description of 96-97, 136 plate showing 1 14 lobatula (Walker and Jacob) D'Orbigny, description of. . 96, 135-136 plate showing 114 pseudouTigeriana Cushman n. sp., description of 97, 136 plate showing 114 vicksburgeiisis Cushman, n. sp., description of 136-137 plate showing 152 U. Utah, southwestern. See Wasliington County, Utah. Uvigerina byraimnsis, Cushman, n. sp., descriptions of 95, 133-134 byramensii, plate showing 112 pigmea D'Orbigny, description of 134 plate showing 149 V. Vaginulina legumen (Linnaeus) D'Orbigny var. elegam (D'Or- bigny) Fomasini, description of 93,131 Page. VaginuliTia legumen ( Linnaeus) D'Orbigny var. elegans (D'Orbigny) Fornasini, plate showing iu V'aughan, T. W., fossils determined by 82 Verneui'ina spinulosa Reuss var. glabrata Castunan, a. var., de- scription of 92 rtfrfimarjfo Cushman, n.sp., description of 127 plate showing i46 VerUbralina advena Cushman, n. sp., description of 102 advena, plate showing U9 sp., descrii)tion of i02 plate showing UQ sp., descrij)tion of 140 Viburnum robustum L/esquereux, description of 179 robustum, plate showing 181 Vicksburg, Miss., exposures of Byram marl near 80-81 fossils found near 81, 83-85 V^icksburg beds, early paleontologic work on 79 Virgin City, Utah, sections west of 72-73 Virgin Umestone member of the Moenkopi formation, fossils col- lected from 67-68 VirguHna sp., description of 92 sp., plate shoning ii(j W. Washington County, Utah, age of the formations in 66-69 features of 53-54 fossils collected in 66-68 general .section of 54-^ local sections iu 69-77 stratigraphy of 54-66 structure of '. 54 Washington dome, Utah, section in 74 White, David, acknowledgment to 22 Widdringtonites reichii (Ettingshauseu) Heer, erroneous identifi- cation of 209, 210 W^olf Hole, Ari/.., section north of 75 Woodbine, Tex., fossil plants collected at 153 Woodbine sand, fauna of 155 flora of, descriptions of 159-180 distribution of, in other formations 159 early collections of 153 features of 155-156 plates showing 181 relations of, to that of other formations 156-15G naming of 154 nature and distribution of 154-155 Woodwards, Miss., exposure of B>Tam marl at 82 Woolsoy Tank, Ariz., plate showing 189 bank of wash near, plate showing 189 Z. Zizyphus lamarensis Berry, description of 170 laniaremis, plate showing 181 DEPARTMENT OF THE INTERIOR Albert B. Fall, Secretary UNITED States Geological Survey George Otis Smith, Director Professional Paper 129 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY 1921 DAVID WHITE, Chief Geologist WASHINGTON GOVERNMENT PRINTING OFFICE 1922 CONTENTS. [The letters in parentheses preceding tlie titles are those used to designate the papers for advance publication.) Pago. (A) Litholofrio subsurface correlation in the "Bend series" of north-central Texas, liy il. I. fioldman (published Mar. 7, 1921 j 1 (B) Onhaulax, a Tertiary guide fossil, by C. W. Cooke (published Sept. 29, 1921) 23 (C) Graphic and mechanical computation of thickness of strata and distance to a stratum, by J. 15. Mertie, jr. (published Marcli 14. 1922 1 39 (D) StratiKraphic sections in southwestern Utah and northwestern Arizona, by J. H. Reeside, jr., and Harvey Bassler (published March 22, 1922 I .53 (E) The Byram calcareous marl of Mississipj)!. by C. W. Cooke (published March 17. 1922) 79 The Foraminifera of the Byram calcareous marl at Byram. Miss., by J. .\. Cushman (published March 17. 1922 ) 87 (F) The Foraminifera of the Mint Spring marl member of the Marianna limestone, by J. -V. Cushman (pub- lished March 28, 1922) 123 (G) The flora of the Woodbine sand at Arthurs Bluff, Tex., by E. W. Berry (published March 23. 1922) 1.53 (H) Geology of the lower Gila region, Ariz., by C. P. Ross (published March 29. 1922 1 183 (Ii The flora of the Cheyenne sandstone of Kansas, by K. W. Berry (published April 11. 1922) 199 Index 227 III ILLUSTRATIONS. Page. Tlate I. Logs of the Seaman and Rudd wells and a generalized log for north-central Texas In pocket. II-V. Species of Orthaulax ; 33-37 VI. Alinement chart for graphic computation of thickness of strata 44 VII. Alinement chart for graphic computation of depth to a stratum 48 VIII. Alinement chart for graphic solution of right triangles 50 IX. A, Coconino sandstone and Kailiab limestone in Hurricane fault scarp, (i miles south of Hurri- cane, Utah; B, Closer view of middle part of scarp shown in A: C. Upper part of Kaibali limestone and basal member (Rock Canyon) of Moenkopi formation in Virgin Canyon, 1^ miles west of Virgin City, Utah SS X. A, View northward toward Smith's Mesa from a point just south of Virgin City, Utah: B. Panorama along east side of Coalpits Wash near Grafton. Utah 59 XI. A. Chinle formation and overlying massive Jurassic sandstone in Zion Canyon above Spring- dale, Utah: B, Cross-bedding in white upper part of massive Jurassic sandstone 12 miles north of St. George, Utah 02 XII. A, Group of cinder cones of late date and associated lava. r)iamond Valley, 12 miles, north of St. George, Utah: B. Nearer view of larger cone of group shown in A (ifi XIII. A, Shinarump conglomerate resting on upper Moenkopi shale and sandstone. Smith's Mesa. 3 miles north of Virgin City, Utah: B. Late basalt flow resting on a conglomerate of basalt and other bowlders, on road from Toquerville to La Verkin, Utah (57 XIV-XXVIII. Foraminifera of the Byram marl 107-122 XXIX-XXXV. Foraminifera of the Mint Spring calcareous marl 146-152 XXXVI-XL. Fossil plants from the Woodbine sand 181 XLI. A, Hills of chloritic schist at the northern end of Big Horn Mountains, near Palo Verde mine, Maricopa County. .\riz.: B. Gonzales Wells, Dome Rock Mountains, Yuma County, Ariz.. 184 XLII. A. A plug of latite of Tertiary age in the Dome Rock Mountains, about 4 miles southwest of Quartzsite, Yuma County, Ariz.: B. Black Butte, Cactus Plain, near Osborne's Well, Yuma County, Ariz 185 XLIII. A, Osborne Wash, about 2 miles southwest of Osborne's Well, Yuma County, Ariz.; B, Saddle Mountain, Maricopa County. .\riz.. looking south 188 XLIV. A, Bank of wash near Woolsey Tank, Gila Bend Mountains, Maricopa County, Ariz. ; B, Woolsey Tank, ( iila Bend Mountains. Maricopa County. Ariz 189 XLV. Reconnaissance geologic map of the lower Gila region, Ariz In pocket. XLVI . ination of three alinement charts for the solution of right-angled triangles 49 8. Dia^am illustrating the use of an alinement chart for the solution of the tangent condition in right- angled triangles 50 9. Trigonometric computer for the solution of such problems as are readily solved with the 12-inch straight slide rule 51 10. Map showing localities where sections were measured in Washington County, Utah, and Mohave County, Ariz 53 11. Cut in the Woodbine sand near Arthurs Bluff, Tex 180 IV o gen De'f , fjv. J' J 1 1 3 5185 00094 5517 ^j^^