the Embryos and Seedlings of the Cactoceae . 465 
brought about by the increasing approach to a spherical form 
of the hypocotyl, and diminution of the cotyledons. The 
only marked exception to this is found in the Platopuntiae, 
climbing Cerei, Phyllocacti, and Rhipsatis ; but in these cases, 
as I have already pointed out, there is in common a return to 
a more mesophytic habit, which amply explains the larger 
spread of surface and consequent relative increase of cotyledon 
and diminution of hypocotyl. This progressive condensation 
of the embryos, however, runs strictly parallel to the 
condensation in the adults, and in both cases is least in 
Pereskia , more in Opuntia , yet more in Cereus , still more in 
Echinocactus , and most of all — indeed reaching near to its 
utmost possible theoretical limit— in the nearly globular forms 
in Mamillaria . In the adults this condensation is due to 
adaptation to a more and more desert habitat. There is 
however this important difference to be noted between the 
two series, that in the progressive condensation the embryos 
lag behind the adults, and this is plainest in the diminution of 
the cotyledons. Thus in Opuntia the leaves are very evident 
in both adults and embryos, but even here the cotyledons 
have diminished from those of Pereskia less than have the 
leaves of Opuntia diminished from those of Pereskia ; in 
Cereus this is yet more evident, for here the cotyledons are 
still large and broad, while the leaves of the adults are 
reduced to small scales and often are but tiny rudiments. In 
Echinocactus the cotyledons are still somewhat prominent, 
but the leaves of the adults are represented only as micro- 
scopic rudiments soon merging into the stem in later growth, 
while in Mamillaria the leaves are as in Echinocactus , and the 
cotyledons themselves are nearing the vanishing-point. It is 
true there are some seeming exceptions to this rule, as in the 
Cephalocactus division of Echinocactus , but in reality, as this 
is one of the most condensed of the divisions of the genus, it 
illustrates and does not contradict the rule. Indeed, as far as 
my material allows of a judgment, it seems plain that the 
same principle applies within the limits of each genus, i. e. the 
more condensed the adults, the more succulent the hypocotyl 
