388 BOTANICAL GAZETTE [MAY 
attention to the fact that in this genus marginal tracheids occur 
with uncertainty, and concludes that they are generally more 
abundant in the mature trunk than in young branches, although 
they are sometimes scarce in the trunk. 
A study of young stems and roots, and of older specimens where 
the phloem region is preserved, shows that when marginal cells are 
present in the xylem they are continuous with erect cells in the 
phloem, but that the latter are frequently present when the xylem 
‘lo | | Petey Ie 
Fic. 1.—Cedrus deodar, stem; radial section through twelfth growth ring of 
xylem, with cambium and a small portion of phloem; in this and the succeeding 
figures the axis of the stem (or root) lies toward the right; X 290. 
has no marginal cells. In such cases a triangular cambial cell 
occurs at the edge of the ray, as is shown in fig. 1, taken from a 
twelve-year old branch of Cedrus deodar. But in cases where such 
a medullary ray can be traced through several annual rings, it may 
generally be made out that scattered marginal cells occur, espe- 
cially at the end of a year’s growth, as may be seen in fig. 2, which 
represents a continuation of the section shown in fig. 1. Both 
margins of the ray in question show not only the scattered occur- 
