820 THE AMERICAN NATURALIST. | [Vor. XXXIX. 
ment of the more primitive cells (1) by those (2) of a more 
advanced and specialized functional value ; and this tendency is 
carried to such an extent that in low rays the replacement is 
often complete (Fig. 15). It is of interest from a developmental 
point of view to note that this relation is precisely that which has 
>" already been shown to exist among the higher 
gymnosperms, with respect to the relative pre- 
ponderance of parenchyma cells and ray tra- 
cheids, and such a comparison goes far to prove, 
especially in connection with: other relations to 
be described immediately, that the cells of the 
second order are of the higher type of develop- 
ment. 
The features thus described for the radial sec- 
tion of the medullary ray, are of both generic 
and specific value as will be seen from the diag- 
noses given, but it should be pointed out that 
the recognition of two forms of cells in the medullary ray of the 
Salicacez is exactly parallel with what has already been noted 
as characteristic of the Coniferales (Penhallow, :04c), though 
with one important difference. In the gymnosperms the ray 
contains in its more primitive condition, parenchyma cells only, 
although these later become differentiated into two specialized 
forms. As a whole, these are equivalent, from the standpoint 
of development and functional value, to cells of the first order 
in the Salicaceze. But in all the higher Coniferales the ray also 
contains specialized ray tracheids which manifestly provide greater 
freedom in radial circulation, and as already shown elsewhere, 
(Penhallow, : 04c), such development is in direct response to 
the requirements of a higher type of organization. No such tra- 
cheids occur in the rays of the Salicaceze, but they are exactly 
represented in a functional sense by the cells of the second 
order (2), which also serve as evidence of a higher type of 
organization. 
* hes a Sin si present intention to discuss the phylogeny 
rc eiu x din and to establish its relative position 
ital ct > ution, It is desirable to indicate that ‘the 
€ evidence so far collected — geographical, 

sel. X 350. 
