60 THE ROYAL SOCIETY OF CANADA 
pith, and spread out in a fan-shaped fashion in the former region, did 
not start from a vertical strand in the wood, but, on the other hand, 
gave rise to a vertical strand or passage midway between the pith and 
the cambium. The multiseriate character of the ray, as it emerged 
from the pith was clearly seen. 
Fig. 22 is a drawing taken from a cross-section of the shoot de- 
scribed in connection with Fig. 19. A ray present from the beginning 
of the growth of the wood, and of a uniseriate character as seen in cross- 
section, formed the canal seen in the figure, at the boundary of the 
spring and summer wood. Several other canals were present in a ring 
in the same peripheral zone and this tends to prove that an unusual 
amount of food was presented to the plant at this period of its growth. 
As is clearly shown in the figure, the canal at this point is formed 
directly from the cells of the ray, the ray itself continuing through the 
summer wood and into the phloem in an exceptionally broad form, and 
ending in the region of the primary cortex in a large resin cyst. Herice, 
in this case, it is very evident that the same conditions which induced 
the ray to become multiseriate, also caused the formation of a vertical 
strand of cells containing an intercellular passage. 
Fig. 23 is drawn from a radial section obtained at the base of a 
shoot of Pinus banksiana from lot one, and shows a multiseriate ray 
extending from the cambium to the protoxylem. In the latter region 
some of the ray elements are seen to curve upwards and downwards, 
and thus to initiate a vertical strand of parenchyma. This strand was 
of considerable length and in its course was seen to receive cells from 
a large number of rays, both of a multiseriate and uniseriate character. 
In these shoots, where, as has been shown above, the growth of the wood 
is very rapid, the vertical strand seems to connect the various rays in 
a vertical direction, and this would without doubt insure a much more 
uniform and rapid distribution of the foodstuffs received from the 
phloem than would ordinarily be the case. 
Fig. 24 is from a radial section in the region of the apex of a shoot 
of Pinus banksiana from lot two. The two medullary rays represented 
in this drawing were multiseriate and both ended in the phloem in 
the same manner as the ray shown in Fig 7. In the portion drawn, 
which was situated in the immediate vicinity of the protoxylem ele- 
ments, it can be seen that’ the part of the vertical strand represented 
owes its existence wholly to the elements derived from the two rays. 
Fig. 25 is a radial section from a shoot of Pinus banksiana as de- 
scribed under lot two, and shows the relation of the vertical and hori- 
zontal clefts which arise in the aggregation of vertical elements and in 
the medullary ray respectively. Here, as is typical in the shoot in 
question, there is a very rapid growth at the initiation of the woody 
