s 4 ROYAL SOCIETY OF CANADA 



of his figures and descriptions with those of Felix, I have little he;?ita- 

 t]on in regarding the two as belonging to the same species. 



Rhamnacinium tkiseriatim, n. sp. 



Figs. 17-20. 

 Eocene of the Porcupine Creek and Great Valley Groups. 

 Bib. :— B. N. A. Bound. Comm., 1875, App. A, 331. 



The wood originall}' referred to by Sir William Dawson as that 

 of a poplar, really comprises two species which are to be referred with- 

 out question, to the genus Ehamnacinium. The first of these has 

 already been described, and the second, which differs from it in essen- 

 tial features, may now be diagnosed as follows: — 



Transverse. — Growth rings rather narrow. Summer wood not conspicuous, 

 narrow and composed of 2 — 4 radially flattened and somewhat thicker 

 walled cells. Wood cells in radial rows, rather thick walled- 

 Medullary rays numerous, from one to three cells wide. Vessels 

 numerous in radial rows; not large; radially elongated elliptical or 

 oval, when double in radial or tangential series; gradually diminish- 

 ing in size and number toward the summer wood where they are 

 small or entirely wanting. 



Radial. — Rays numerous, the cells of two kinds with their lateral walls multi- 

 porose when opposite a vessel; the short arfd higher marginal cells 

 thick walled and pitted, variable, rarely twice longer than high, 

 usually of the same length, often interspersed; the narrower central 

 cells about four times longer than high, thick walled. Scarlariform 

 vessels numerous and conspicuous. Pitted vessels not determinable. 

 Wood parenchyma not determinable. 



Tangential. — Ray cells of two kinds; the uniseriate rays chiefly small and few; 

 the multiseriate rays narrow and high, the cells of two kinds; the 

 small and thick-walled cells chiefly 3, or sometimes 4-seriaite. the 

 large, thinner walled cells oval or oblong, terminal and uniseriate, 

 often interposed between regions of smaller cells. Vessels as in the 

 radial section. 



Among existing species of Ehamnus which it has been possible to» 

 • bring into comparison, the resemblance is in many respects closest to» 

 E. purshiana, from which, however, it differs in the general aspect of 

 the transverse section, and in the detailed structure of the medullary 

 ray. The radial aspect of the medullary ray is very similar to what 

 may be seen in Salix longifolia with respect to the relative distribution 

 of long and short cells. The great excess of the multiseriate over 

 uniseriate rays, is also another feature which establishes similarity 

 with Ehamnus purshiana, and this resemblance is further emphasized 

 by the generally triseriate character of the larger rays. It only remains 



