VASCULAR SYSTEMS OF FLORAL ORGANS. 159 
cords belonging to each capellary leaf are plainly differentiated at 
this early stave. 
The ways by which the tracheæ may change their position from 
the inner angle to a central one in the cord, or to the opposite 
side, are at least three in number. The whole cord may twist 
round, as do the marginal cords of the pistil of Hellebore ; 
or, secondly, it may divide into two which rotate, according to 
Van Tieghem, as in Geranium longipes; or the tracheæ may pass 
obliquely upwards through the phloém, and come out at the 
other side, as do the marginal cords of the carpels of Ivy. This 
last method seems to me to be the more usual case with pla- 
centary, that is marginal, cords, which come up from the axis 
with the trachee oriented inwards, but which finally must 
supply cords to the ovules, which lie on the opposite side of 
them *. 
UNDIFFERENTIATED OR ARRESTED STATE OF CARPELS. 
This feature plays a very important part and tends to obscure 
greatly the proper interpretation where cohesions aud adhesions 
are concerned ; for I believe that with rare, even if there be any, 
exceptions the apparently central medulla with the circle of 
marginal or placentary cords of a pistil is entirely due to the 
fusion of the margins of the carpels, the epidermides being 
arrested, so that the combined hypertrophied mesophylls form 
the central parenchymatous column of axile placentations. This 
is not only testified to by the arrangement of the tracheæ, but 
by the fact that the central mass often divides up into as many 
pieces as there are carpels, in the upper part of the ovary. Thus, 
comparing figs. 1 and 2 with figs. 3 and 4 of Gladiolus (1x. ; 
Pl. XXXII.), it will be seen that the cords in the centre of fig. 1 
look like those of an ordinary endogenous stem; but that they 
are potentially marginal is evident, as the dorsal carpellary cords 
are likewise present. Fig. 3 shows how the central tissue breaks 
up into three pieces belonging to the three carpels, no axis proper 
at all being present. 
Another and conspicuous instance of arrest is seen between an 
inferior ovary and the receptacular tube into which it is merged ; 
so that the internal tissues of the two organs become confluent, 
* For a description of Hellebore, Impatiens, and Ivy, see ‘ Origin of Floral 
Structures,’ p. 64 segg. 
