DESCRIPTION OF SPECIES. 45 



down to the petiole as in the leaf from Kansas, an error rectified by the 

 sj)ecimens of Fl. Foss. Arct., vol. G, where especially Figs. 7 and 8 of PI. 

 XVIII have that basilar cnrve well defined. The real difi'erenee between the 

 two species is in the small size and obtnse ixyiex of the leaves of I', atyfua, 

 while those of Populites lancastriensis are apparently lanceolate, pointed 

 (the upper part is destroyed), and especially in the real camptodrome ner- 

 vation, the secondaries forming a series of areoles in following the l)orders 

 in /'. afi/tjid, while in I'opKlitcs lancastriensis the secondaries either reach the 

 l)orders by their extremities or are efiixced toward the borders and not curved 

 in areoles. 



Habitat: Ellsworth County, Kansas. No. 567 of the museum of the 

 University of Kansas; E. P. West, collector. 



PopuLUs ELLiPTiCA Newb. 



Later Ext. Fl., p. 16 ; Illustr. Cret. aud Tert. PL, PI. iii, Figs. 1 aud 2. 



POPULUS MICEOPHYLLA Newb. 



Later Ext. Fl., p. 17 ; Illustr. Cret. and Tert. PL, PL in. Fig. 5. 



PoPULUS ? CORDIFOLIA, Newb. 



Later Ext. FL, p. 18; Illustr. Cert, and Tert. PL, PL ill, Fig. 7. 



Populites Sternbergii, sp. nov. 

 PL VII, Figs. 8, 9. 



Leaf subcoriaceous, broadly ovate, pointed, much enlarged above the 

 base, rounded to the petiole, entire or slightly undulate; primary nerve 

 thick and straight to the apex; secondaries distant, parallel, forking near 

 the border, curved upward in passing to the borders, subcamptodrome ; 

 nervilles simple, distant, at right angles to the secondaries, percurrent. 



These two leaves are apparently referable to the same species, although 

 differing in some parts. In Fig. 8 the secondaries and their branches are 

 more distinctly craspedodrome, and their disposition less regular. The thick 

 median nerve is also in this leaf disproportionate to the very thin, sharply 

 marked secondaries, which are alternate or parallel, inequidistant, at an 

 angle of divergence of 60°, all arched upward in traversing the lamina, 

 simply forking near the borders, the lower of the secondaries on one side 

 being arched downward, contrary to the upward curves of the others. In 

 Fig. 9 the median nerve is not as thick ; the secondaries are equidistant, 

 stronger, and evidently camptodrome, curving quite near the borders, the 

 lowest pair being very thin and marginal. 



