50 THE ROYAL SOCIETY OF CANADA 
in the fact that the marginal cells of a ray, which have the greatest 
freedom of growth, are taller and narrower than the central cells. Thus 
Mischke in his discussion of the medullary rays states the following in 
regard to Pinus sylvestris (21, 69): “The marginal cells of the phloem 
ray extend the breadth of one sieve tube, the cells of the other rows 
from one and a half to three and a half sieve tubes in a radial direction. 
A similar relation is exhibited in the rays of the xylem. ; 
These relations can be assigned to the mode of division in the AE 
of the medullary ray. It is allowable to assume a priori that divisions 
will occur more vigorously in the border cells than in the middle rows of 
the medullary ray.” 
Penhallow (25, 79), in speaking of the rays as seen in radial sections 
also states that, “In general terms, the higher the ray the lower the 
component elements, from which it follows that in one-celled rays the 
cells are usually highest, but this feature is only of general interest, 
since it rarely has a bearing upon the chief question at issue,” (the 
question at issue was taxonomic and so does not specially concern us). 
It can thus be seen that both vigor and freedom of growth would be 
followed by a decided tendency on the part of the medullary ray cells 
to elongate in a vertical direction and to undergo rapid division. 
Now if at any period in the cambial growth of the shoot a large 
amount of nutritive material should become available for consumption, 
the following conditions would present themselves. As is evident from 
a consideration of the secondary growth in thickness of a Gymnosperm, 
the radial growth and extension of the medullary ray, more especially 
in the xylem region, is in a sense purely passive, since it is dependent 
not only on the radial extension of the neighboring tracheids, but also 
on the addition of new rows of tracheids. Thus its growth in a radial 
direction is decidedly limited. The vertical elongation of the shoot, 
however, is much greater than its radial growth and thus the rows of 
medullary ray cells, unless capable of keeping pace with the vertical 
extension of the tissues would be torn apart. Hence, medullary rays 
laid down during periods of active growth would be very much more 
numerous and shorter than those laid down during periods of less active 
growth. This was found to be the case in many of the sections examined, 
where a tall ray in the phloem would be seen to be divided into several 
shorter rays as it proceeded through the xylem. Fig. 5 shows a radial 
section through the basal region of a shoot of Pinus banksiana from the 
second lot described above. The drawing extends from the cambium 
to the protoxylem, and shows two xylem rays which are still slightly 
connected by a single ray cell. In the phloem region the two rays were 
seen to originate from one tall ray composed of almost isodiametric 
cells. 
