the Embryos and Seedlings of the Cactaceae . 455 
this division I have had no seeds ; but E. robustus i of which 
figures of embryos are given by Pfeiffer, belongs here 1 . 
Though in these figures the breadth of the cotyledons cannot 
be seen, the whole aspect, particularly of his figure c } is very 
Cereus- like, thus confirming the primitive position of this 
sub-genus. 
Most important, and typical of this genus, is Schumann’s 
division of Ancistrocactus. In it I have studied embryos of 
E. Wislizeni (Fig. 32), viridescens (Fig. 33), texensis (Fig. 34), 
longihamatus (Fig. 35), cornigerus (Fig. 36), while a figure of 
E. recurvus is given by Pfeiffer, and of setispinus by Maxwell. 
It will be noted that all of these figures agree in showing 
a much swollen hypocotyl with thick, pointed, flat-faced 
cotyledons which are much less broad than the hypocotyl, 
and do not merge gradually into it all around, but on the 
sides and faces are separated by a sharp angle which is at 
times almost a slight constriction. The cotyledons are 
smallest in E. cornigerus. The form of cotyledons shown in 
this group seems to be as characteristic for the genus E chino - 
cactus as the broad and flatter ones are for Cereus , and the 
constancy of form through this large sub-group implies 
marked constancy in this character, and hence reliability as 
a test for affinities. In this group I have followed the develop- 
ment of the epicotyl in E. Wislizeni , texensis , and a few 
others. The axillary clusters are at first distinct, but stand 
in lines, and very soon run together to form ribs. In 
E. texensis new ribs appear among the older ones, starting 
abruptly with a small piece of rib below a spine-cluster. 
This occurs also in other species, and shows that in these 
forms rib-formation has become so fixed that the rib is 
itself a morphological element. They, together with most 
1 Of this species he figures three very different forms, and draws conclusions 
therefrom that the form of embryos and seedlings in the same species is very 
variable. I have only to say that I have never seen in any one of the many 
species of Cactaceae I have studied any such variation as Pfeiffer figures and 
describes, aside of course from specimens injured in some way or drawn for want 
of light. Seeds are apt to mix in the pots, as I have pointed out, and some of his 
different forms may be different kinds. 
