7 1 6 Thomas . — Seedling A natomy of 
usual structure. An individual having three cotyledons was examined 
(Text-fig. 27). Two of these cotyledons united near the base and their 
confluent vascular systems became continuous in the hypocotyl with 
one of the groups forming the quite normal elliptical stele, and through 
this with a pole of the diarch root. Thus, contrary to what was recorded of 
a tricotyledonary specimen of Cheiranthus Cheiri , two cotyledons form 
between them only one pole, so that the root is diarch and not triarch. It 
is of interest to note that in the free portion of these cotyledons the vascular 
strand is quite distinctly double, and this not in preparation for the root 
arrangement but contrary to it, as in the lower united portion the strand 
becomes single so as to form with that of the other cotyledon or half- 
cotyledon a double bundle. This would seem to have some bearing on our 
conception of the meaning of ‘ doubleness \ 
Hesperis tristis. The cotyledons insert themselves by broad flat bases 
which unite on one side for a very short distance before fusing with the 
27 
ph 
xj 
Text-figs. 26, 27. Matthiola tri- Text-fig. 28. Draba Aizoon : vascular system of 
cuspidata. 26, Normal seedling ; 27, plumular leaf, ph , phloem ; xy, xylem. x 800. 
Seedling with three cotyledons. Nat. 
size. 
hypocotyl. One minute lateral on either side of the central double bundle 
remains free until quite near the cotyledonary node, when it unites with the 
central. The protoxylem is in a somewhat degenerate condition throughout, 
but persistent and obvious. 
Draba Aizoon. While the main features are entirely according to type 
there are several distinctive subordinate characteristics in this species. 
Thus the hypocotyl shows a very definite central region that may perhaps 
be termed a ‘ stele ’, which approaches protostelic appearance by early 
(i. e. high up) loss of pith and somewhat cylindrical disposition of phloem. 
This appearance is enhanced by the extremely early activity of 
the cambium, which is perhaps most active in the position corresponding to 
the two protoxylem poles of the root. The appearance suggests that there 
is probably no interval of time, i. e. stage of quiescence, between the cell- 
divisions of the primary desmogen and those more definite radial divisions 
termed ‘ secondary \ 
While the cotyledon strands are not markedly double above the 
