22 BOTANICAL GAZETTE | JANUARY 
of sections were examined. To examine the position of the 
proto-hadrome in relation to the pericambium, not only were a 
number of sections taken from roots of various individuals, but 
also several roots of the same specimen in their entire length 
from apex to base. The latter method appears to give the most 
satisfactory results, and is sufficient to demonstrate that neither 
the position of the proto-hadrome vessels, nor the number of 
hadromatic rays or of the individual vessels is to be depended 
upon as a constant character. 
A few examples of these structural irregularities are as follows. 
One of the thick roots possessed at the base eight rays of proto- 
hadrome, each consisting of a single vessel, five of which had 
penetrated the pericambium, thus bordering immediately on 
endodermis. Nearer the apex of the same root the number of 
rays decreased from eight to seven and to six, five of which had 
broken through the pericambium, as in the first case. At the 
apex itself there were only five rays, all of which were bordering 
on the endodermis; thus the number of interruptions of the 
pericambium was five in the entire length of this root. In other 
roots only five rays were found at the base, four of which had 
broken through the pericambium ; and here also there was a 
similar decrease in number of rays when approaching the apex, 
where finally all the proto-hadrome vessels bordered on the 
endodermis. The number of interruptions may vary in the same 
root, instead of being five, for instance, throughout the entire 
root; or there may be roots in which all these vessels border on 
the endodermis, the most frequent condition (fg. 7). Van 
Tieghem ascribes six to eight hadrome rays to this species, each 
consisting of two vessels, but such regularity in number was not 
observed in my material. Not only did the number of rays vary, 
usually decreasing from base to apex, but five was the most com- 
mon number, each ray consisting of a single proto-hadrome 
vessel (fig. 1, 1) instead of two. The largest number of rays 
observed was eight, but this number, as well as seven, was rather 
rare, while six or four was not uncommon. While the proto- 
hadrome vessels were observed as being mostly one in each ray, 



