DESCRIPTION OF SPECIES. 105 
C. lanceolatum. These show only the basal portions, and are scarcely sutti- 
cient for accurate determination. Two of these nave the margins coarsely 
serrate; in the third they are represented as entire. Under the cireum- 
stances we are scarcely justified in considering our leaves specifically 
identical with either Heer’s or Ettingshausen’s, but they are very closely 
allied. Another leaf figured by Heer (op. cit., Pl. LXV, fig. 6) he calls 
Celastrophyllum serratum Sap. et Mar., but in this the long wedge-shaped 
base is coarsely serrated nearly to the petiole, a character which we have 
never found in our leaves. Saporta and Marion obtained the leaves upon 
which the description was based from the Upper Cretaceous strata at Gelin- 
den, and it is an interesting fact that leaves so closely allied, if not identical, 
occur in strata approximately of the same age at these so widely separated 
localities. 
CELASTROPHYLLUM MINUS Hollick n. sp. 
Pl. XLII, figs. 51, 52. 
Leaves broadly spatulate in outline, 12" or 13" long by 8™™ broad, 
entire or somewhat undulate-crenate near the apex, narrowed to the base; 
nervation obscure or obsolete. 
These are the smallest leaves which I have referred to this genus. The 
absence of nervation makes it almost impossible to know where to look for 
their affinities, but the spatulate outline and crenate margin give a general 
impression of the genus. 
No memoranda in regard to name or locality were found in connection 
with them.—A. H. 
CELASTROPHYLLUM Brirronianum Hollick n. sp. 
Pl. XLII, figs. 37, 38, 46, 47. 
Leaves lanceolate or slightly lanceolate-spatulate in outline, finely 
denticulate above, entire below, tapering to the petiole; secondary nerva- 
tion fine, but clearly defined, anastomosed in irregular loops, and connected 
by numerous reticulations. 
The several leaves included under this name differ from C. spatulatwmn 
chiefly in having a nearly symmetrical lanceolate outline and finer denta- 
tion. That they are generically related there can be but little question, 
