NO. 2256. FOSSIL PLANTS FROM OKLAHOMA— BERRY. 635 



from the midrib at rather regular intervals of about 3 mm. at angles 

 of about 45°, and are camptodrome. Areolation not made out. 



These leaves, in both form and venation, are allied to numerous 

 fossil and living species of Bumelia, of which many have been 

 described. In the existing flora about a score of species are known — • 

 all American, and distributed from the southern United States 

 through the West Indies and Central America to Brazil. The fossil 

 species are known from the Upper Cretaceous onward, and the genus 

 is represented in the Miocene of Europe by seven or eight different 

 forms. It occurs also in the Florissant lake bed. 



The present fossil species is very similar to various existing sub- 

 tropical forms, as for example Bumelia angiistifolia Nuttall and 

 Bumelia tenax Willdenow. It is also very much like the leaves of 

 the existing Bumelia lanuginosa Persoon which ranges from Georgia 

 and Florida to Illinois, Missouri, Arkansas, and Texas, and reaches 

 its optimum development in the bottom lands of eastern Texas. 



Holotype.—C^^i. No. 35289, U.S.N.M. 



Family EBENACEAE. 

 Genus DIOSPYROS Linnaeus. 



DIOSPYROS BRACHYSEPALA Al. Braun. 



Plate 95, fig. 3. 



Diospyros brachysepala Al. Braun, Die Tert. Fl. v. Oeningen, Neues 

 Jahrb., 1845, p. 170.— Heer, Fl. Tert. Helv., vol. 3, 1859, p. 11, pi. 102, 

 figs. 1-4.— Lesquereux, Tert. Fl., 1878, p. 232, pi. 40, figs. 7-10, pi, 

 63, fig. 6; Cret. and Tert. Fl., 1883, p. 174, pi. 34, figs. 1, 2.— Knowl- 

 TON, Proc. U. S. Nat. Mus., vol. 51, 1916, p. 285. 



Diospyros princctonia Cockerell, Amer. Mus. Nat. Hist. Bull., vol. 24, 

 1908, p. 105, pi. 10, fig. 36. 



This polymorphous species has been recorded from a very large 

 number of localities and horizons. The type material came from 

 both the earliest and the latest Miocene of Switzerland, but subse- 

 quently this species has been identified from all stages of the Ter- 

 tiary of Europe. In America it has been recorded from beds of 

 the late Upper Cretaceous and at different Teritary horizons. It 

 seems incredible that all of these records should represent a single 

 species and probably several are included, but their segregation on 

 other than stratigraphic grounds is impossible at the present time. 

 This being true I can not do otherwise than to refer the Oklahoma 

 material to this species since it appears to be identical with that 

 from Florissant, Colorado, so determined by Lesquereux and Knowl- 

 ton. At the same time it should be kept in mind that many of the 

 identifications of Diospyros brachysepala, of which an extended 



