No. 467.] STUDY OF THE SALICACEE. 817 
times longer than broad, and either of uniform width throughout 
or somewhat contracted at the ends. The terminal walls are 
always more or less strongly pitted, but they exhibit somewhat 
striking variations in this respect as between Salix (Fig. 17, a) 
and Populus (Fig. 16, 2). The upper and lower walls are rather 
thick and often strongly pitted, but in both of these respects 
there is a marked difference between the two genera in such a 
way that in Populus (Fig. 16, a) there is always a tendency to a 
relatively thin wall with few pits, while in Salix (Fig. 17, æ) the 
tendency is always toward a very thick and strongly pitted wall. 
Such cells are generally resinous, more strongly so in Salix than 
in Populus, and it is in them that resin is found when it may 




[I z — 
s EU, Des 
ix hookeriana. Radial section of a medullary ray showing details in the struc- 
ture of the cell walls: cells æ of the first order, and cells 4 of the second order, the latter 
occupying a marginal position. X 350» 
Fic. 17.— Sali 
have been eliminated from all other parts of the ray. Such 
cells I have designated as “cells of the first order” or (1). 
They seem to be the more primitive parts of the structure, and 
in this sense they are comparable with the parenchyma cells of 
the gymnosperms (Penhallow, :04c). Their radial walls are 
generally devoid of pits except that in a few instances I have 
observed the occurrence of minute and simple pits opposite wood 
cells, but so far as can be ascertained at the present time, this 
feature possesses no special significance. 
