[PENHALLOW] NOTES ON TERTIARY PLANTS 53 



6th. The presence of two kinds of cells in tangential section, and their relative 

 positions. 



7th. The two kinds of cells shown in radial section, their disposition and rela- 

 tive dimensions. 



By the first five of these characters the genus is also separated 

 from the Salicaceœ, while it also seems probable that the distribution 

 of the two kinds of cells in tangential section, is more nearly like 

 that of the Ehamnacese than of the Salicacese. From these consider- 

 ations it is evident that our fossil must be held to be a Rhamnacinium 

 rather than either a Populus or a Salix. it is true that three species 

 of poplar (P. genetrix, P. richardsoni, and P. arctioa) have been found 

 eomewhat abundantly both at Porcupine Creek and in the Great Val- 

 ley/ while the same localities also show leaves of Salix raeana;^ but 

 on the other hand, the leaves of Ehamnus concinnus and another 

 undesignated species, are well known forms in the deposits at Porcupine 

 Creek and Great Valley,^ so tJiat evidence from this source does not 

 lend material support to one or the other possibility. It now remains 

 to determine in what respects our fossil is related to the specimens 

 of Felix and Knowlton. 



While our specimen and that of Felix are undoubtedly related 

 generically, they differ in such respects as to justify the belief that 

 they may represent different species. In the E. radiatum of Felix, 

 he describes the numerous medullary rays as having only one row of 

 vessels between each pair. In our specimen, on the contrary, each 

 pairs of rays embraces one or frequently two rows of vessels. The 

 most marked dissimilarity is to be found in the proportions of the two 

 kinds of cells in the medullary rays. In E. radiatum the large cells 

 are terminal and do not appear to extend into a single series which 

 forms so prominent a feature of our specimen. It is possible ,that 

 this character does not possess very great importance, but I am hardly 

 inclined to accept this view in consideration of the difference in this 

 respect which obtains between Ehamnus caroliniana and E. purshiana, 

 and I should therefore be disposed to consider the material from Por- 

 cupine Creek as representing a distinct species for which I would sug- 

 gest the name of E. porcupinianum. Among existing species of North 

 American Ehamnacese, the nearest approach seems to be to E. carolin- 

 iana, both with respect to the general character of the transverse section 

 and the structure of the medullary rays. The Ehamnacinium of 

 Knowlton cannot be compared so accurately, as his description does 

 not include some of the essential data, but from a careful comparison 



> Trans. R, See. Can., IV., iv., 27; B. N. A. Bound. Comm., 1875, App. A, 330. 

 - B. N. A. Bound. Comm., 1875, App. A. 330. 

 ' Ibid. 



