'CERATOPS BEDS OF WYOMING AND MONTANA 243 



to say just where the one ends and the other commences. The same 



is true of the beds overlying the Fox Hills The 



sandstones of the entire series are very similar, and since there is 

 entire conformity throughout, it is absolutely impossible to determine 

 just where the marine beds end and the fresh- water beds commence. 

 The Ceratops beds of this region are a natural sequence of the Fox 

 Hills. The materials composing both were evidently derived from a 

 common source. The only safe criteria for distinguishing one from 

 the other are their fossils. 



Above the dinosaur-bearing "Ceratops beds" are similar rocks in 

 which sandstones are a more prominent feature and there are coal 

 beds of workable thickness. The lithologic change is not striking and 

 there is no recognized stratigraphic break but the fossil plants have 

 long been recognized by Doctor Knowlton as belonging to the typical 

 Fort Union flora, i. e., the flora of the "upper Fort Union" as he now 

 interprets it. 



The fresh-water invertebrate fauna and the flora as well as the 

 vertebrate fauna of the Converse County "Ceratops beds" have been 

 collected mainly from the upper half of the formation. Doctor 

 Knowlton's latest revised list* of the plants, which differs considerably 

 from the one previously publi'^hed, ^ need not be repeated here. Atten- 

 tion is called, however, to the fact that of 48 species and varieties 

 enumerated 5 species are figs and 2 are palms, and that of the 16 

 identified species whose names are given 7 are reported to occur at 

 Black Buttes. It seems, therefore, that the flora is still subtropical 

 in character and somewhat closely related to that of Black Buttes. 



A re-examination of the invertebrates has resulted in discrediting 

 or questioning a few of the previous identifications and in recogniz- 

 ing some of the species of Unio described by Whitfield from the " Hell 

 Creek beds." The revised list is as follows: 



fUnio hrachyopisthus White 



^Unio couesi White 

 *^Unio cryptorhyncJius White 



'fUnio endlichi White 

 *\Unio holmesianus White 



^Unio proavitus White 

 ^\Unio stantoni White ^^ 



* Proc. Washington Acad. Sci., Vol. XI, p. 207. 

 »Bull. Geol. Soc, America, Vol. VIII, p. 136. 



'" This is the form sometimes referred to U. dancs M. & H., and later named 

 U . gibbosoides by Whitfield. 



