S2 ROYAL SOCIETY OF CANADA 



H. purshiana. In the Salicaceae the cells are narrow and many times 

 longer than broad. 



In transverse section the medullary rays of the Salicaces" are 

 chiefly one cell wide, while in the Bhamnaceae they are from 1-3 cells 

 wide. In the tangential section, where some of the most characteristic 

 features are to be found, a careful examination of seven species of 

 poplars and nine species of willows shows that the rays are almost 

 all of one kind and uniseriate. Partial exceptions occur among the 

 poplars in P. fremonti in which the rays become 3-seriate in part; 

 and among the willows in S. longifolia and S. bigelovii where the rays 

 are more commonly 2-seriate through the central portion. There are 

 two kinds of cells, however, distinguishable by their difference in size 

 and situation, being primarily terminal or again sometimes inter- 

 spersed. This twofold form of the cell is a striking feature of the 

 Ehamnacese in which the larger cells are chiefly terminal, but in R. 

 purshiana they occasionally lie in the central portion of the ray which 

 then becomes contracted to one cell in width. 



In the radial section the rays of Rhamnus show two kinds of cells 

 so distributed that the low, relatively long and more numerous cells 

 lie in the centre, while the high, very short cells occupy the margins. 

 In R. earoliniana the walls of all the cells are thin and devoid of pits, 

 and the same is also true to a large extent of R. purshiana, which 

 nevertheless not infrequently shows the entire ray to be composed 

 of thick-walled cells^ the walls of which bear numerous fine pits, 

 and through these features mere is developed a very striking similarity 

 to the fossil. In the Salicaceœ two kinds of cells are also present, 

 and in their relative dimensions and positions, they somewhat closely 

 resemble those of the Rhamnacege, though in some respects they more 

 nearly resemble those of the fossil. So much variation is possible in 

 details of this character that one must not place too much reliance 

 upon them, although they might serve to influence a final decision. 

 Where the cells of the medullary ray lie opposite a vessel, they are 

 always perforated with rounded pores disposed in two or more series, 

 and this character, which is common to the SalicaceaB, Rhamnaceœ and 

 the fossil, offers no opportunity for differentiation. 



A careful comparison of the foregoing facts with the diagnosis 

 of the fossil, will show that the latter resembles the structure of 

 Rhamnus in the following particulars: — 



1st. The aggregation of the vessels in transverse section. 



2nd. The presence of (both scalariform? and) pitted vessels. 



3rd. The presence of wood parenchyma. 



4th. The multiseriate character of the medullary rays in transverse section. 



5th. The uniseriate and multiseriate rays of the tangential section. 



