122 THE LATER EXTINCT FLORAS OF NORTH AMERICA. 



in others much broader. This variability indicates that the leaves having 

 narrow entire lobes found in the Dakota group and named A. quinqiiepartita, 

 A. tripartita, and A. cuneata, by Mr. Lesquereux, are but forms of one 

 species. Aralia WMtneyi Lesq. has seven-parted leaves, these less deeply 

 lobed, and with entire margins; A. angustiloba more deeply cut leaves with 

 narrower and entire lobes (Mem. Mus. Comp. Zool., Vol. VI, No. 2 (1878), 

 p. 22, PI. V, figs. 4, 5). 



Perhaps of all described species of Aralias A. Saportanea Lesq., from 

 the Dakota group of Kansas (U. S. Geol. and Geog. Surv. of Colorado, 

 Hayden (1874), p. 350, PL I), approaches nearest to those under considera- 

 tion, but are distinguished by minor characters, smaller size, less deeply 

 dentate margins, etc. This species is found, however, in our Middle 

 Cretaceous strata, forming part of the most ancient angiosperm flora, and 

 while the species are unquestionably distinct, their great resemblance may 

 be fairly taken as an indication that one is the progenitor of the other. 

 The group of leaves now before us has been, perhaps without sufficient 

 proof, referred to the genus Aralia, and it is highly desirable that this 

 question should be decided by the discovery of fruit or flowers. But 

 whether Aralia or not, they constitute a marked feature in the older 

 angiosperm floras in this country and in Europe, and their geological 

 interest and value is to a certain degree independent of their botanical 

 relations. It has been suggested by Count Saporta that not only the 

 trilobed leaves from the Dakota Cretaceous, which I have described as 

 Sassafras, but also the great leaves of Platanus nobilis, figured in this 

 volume, should be referred to Aralia, as the platanoid leaves described by 

 Unger as P. Hercules, etc., have been; but there is little resemblance 

 between the quinquepartite, narrow-lobed, toothed leaves of A. Saportanea 

 Lesq. and its associates with three lobes, broadly rounded, sometimes 

 almost obsolete and entire, in Sassafras cretaceum, and it only requires a 

 glance at the figure of the huge leaf of Platanus nobilis, given on PI. L of 

 this monograph, to be satisfied that its affinities are with Platanus rather 

 than Aralia. 



Formation and locality: Tertiary (Green River group). Green River, 

 Wyoming. 



