the Embryos and Seedlings of the Cactaceae. 449 
belonging to Phyllocactus. It merges with the latter genus 
through such jointed forms as P. Russellianus and Gdrtneri. 
E. truncation is an epiphytic, hanging, much-jointed form, 
too well known to need description. 
None of my many seeds of Epiphyllion have germinated, 
and hence I have no data as to the embryos and seedlings 
except the two figures cited above. Both figures show 
a rather stout embryo with the hypocotyl angular, or quad- 
rangular, in section, and with short, tapering, pointed coty- 
ledons. This is not at all the type of embryo which one 
would expect theoretically in this genus. Considering the 
marked mesophytic character of the adults reached through 
the increasingly mesophytic climbing Cerei and Phyllocacti, 
one would expect a still further development of surface in 
the most mesophytic Epiphyllum. That this is not the 
case remains to be explained. Possibly we are in error as 
to the affinities of this genus ; there is something about 
the cotyledons, as shown in the figures, which suggests 
Rhipsalis. But it is more likely that the extremely epiphytic 
habit of the species has something to do with it. The 
epiphytic habit always requires some xerophytic characters, 
and of these one of the first is reduction in size of leaf-surfaces. 
In this connexion the embryos of P. Russellianus and 
Gdrtneri would be of great interest. 
The epicotyl in E. truncation , as described by Goebel, is 
flat. It forms at first four polsters as the beginning of four 
ribs, but two of these fail to develop farther, so that the other 
two form a flat structure precisely as has been described and 
figured above for Phyllocactus Phyllanthus. This form of 
epicotyl is precisely that which we would theoretically 
expect. 
6. Genus Piloeereus. 
P. Houlletii , Lem. Figure 22. 
This genus merges with the large columnar Cerei, from 
which it is distinguished by very minor characters. 
The germinated embryos of P. Houlletii (Fig. 22) resemble 
