the Embryos and Seedlings of the Cactaceae. 445 
only one in which I have found it regular. One supposes 
at first that it is due to the splitting of the tip of the convex 
cotyledon against the tip of the concave one in the seed, 
but careful examination has not certainly confirmed this. 
Another peculiarity in this species, also unique in my observa- 
tion, is the intercalary basal growth of the cotyledons after 
development of the epicotyl, thus carrying the axillary buds 
away from the latter (Fig. 12 c). This suggests the similar 
basal growth of the tubercles in Mamillaria , Leuchtenbergia , 
&c., so important in producing the configuration of those 
genera. I have made one of these buds develop by removing 
the epicotyl. On the epicotyl at first four ribs develop, 
which often become eight, later dropping back to five. 
In division D occur several forms of much importance. 
First is C. Martianus , which differs so much from others in 
this division as to raise the suspicion either that my material 
is wrongly named, or else that the species does not belong 
to this division. The germinated embryos (Fig. 13) appear 
to stand about intermediate between the columnar giganteus- 
like forms and the climbing species now to be described. 
In C. grandiflorus and nycticaulis we have two climbing 
species very closely related, and very much alike both in 
adults and embryos (Figs. 14 and 15). In both of these, 
and also in C. spimdosus and triangularis (Figs. 16 and 17), 
two other climbing and not-distantly related forms, the coty- 
ledons are unequal, flat and broader than the hypocotyl, 
though they are not broadest at base as are most other 
Cerei, but a little above it. One notices in these cotyledons 
a resemblance to those of the Platopuntiae, but only a little 
reflection is needed to show that this resemblance is not 
genetic, but ecological. These climbing Cerei live under con- 
ditions more mesophytic than xerophytic ; and as all evidence 
shows that they have been derived from forms more like the 
columnar species, this larger size of the cotyledons cannot 
be primitive, but must have been re-acquired in response to 
conditions which permit of that increase of surface towards 
which green parts are continually tending. They are of 
