6 . 9.8 
Hill and de Fraine. — On the 
arrived at in the upper region of the hypocotyl ; the endodermis, however, 
does not appear until the collar is reached. It is to be observed that the 
protoxylem, during these changes, is always in close contact with the 
metaxylem ; the wood does not divide into two parts, as does the phloem. 
A certain amount of variation occurs both in the cotyledonary bundles 
of one individual and in the transition as seen in different plants of the 
same species. In the first case, although the bundles behave in a similar 
fashion, they may not be in precisely the same phase of transition at the 
same level. In the second case, the relative level in the hypocotyl at 
which transition begins is not constant, for sometimes the traces reach 
the central region of the axis and keep to their collateral condition for 
some little time (Diag. 3, Fig. 2, and Fig. 9, Plate XXXV). The actual 
mode of rearrangement of the vascular tissues is, however, the same as 
has been described above. Further, an occasional xylem element may 
be differentiated well outside the cotyledonary protoxylem after the latter 
has attained, or nearly attained, its exarch position. 
J. Cedrus , Webb & Benth. The structure and transition-phenomena 
of the seedling of this plant are practically identical with those of jf. vir- 
giniana. The main features of difference, which are of relatively small 
importance, are as follows : the vascular bundles of the cotyledons of 
J. Cedrus are relatively larger, the tangential expansion is somewhat 
greater, the transfusion-tissue is more abundant, and the mesarch posi- 
tion of the protoxylem is more marked than in J. virginiana (Fig. 8, 
Plate XXXV). The number of centripetal elements, however, is not great. 
Another feature of difference between these two species is that the 
transition usually begins at a lower level in J. Cedrus ; but, as regards this 
point, the same variations obtain as have been described for J. virginana. 
y. procera in all respects resembles y. virginiana ; attention, how- 
ever, may be drawn to the occurrence in one individual of a well-marked 
bifurcation of the phloem and rotation of the xylem of one of the cotyledon- 
traces immediately on entering the axis ; no other plant of this order, so 
far as has been seen, exhibited these changes at so high a level as in this 
particular individual. The other cotyledonary strand behaved in the 
manner illustrated in Diag. 3, Fig. 3. 
y. bermudiana , Linn. The material of this species available for the 
purposes of this investigation was inadequate, and also unsuitable owing 
to the presence of much secondary thickening. Sufficient, however, was 
made out to indicate that this species does not differ in any essential 
feature from the foregoing. The size of the cotyledon-traces and the 
paucity of centripetal wood corresponds with what obtains in y. vir - 
giniana , while the abundance of transfusion elements is in close agreement 
with y. Cedrus . It is, however, not desired to lay any particular stress on 
this last point, for it is a feature which may vary with the age of the leaf. 
