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. 



