go BOTANICAL GAZETTE [sucust 
the base of the ovule, and of the one-sided epimatium arising from 4 
the scale is uncertain. The origin of the latter may be related, how- 
ever, to the foliar origin of the ovules, and, as Miss ROBERTSON states, 
so that it will not de to lay too much stress on this point, as proving 
that epimatium and arillus are not homologous.” 
Though the external resemblance between the cones of Phy 
cladus and Cephalotaxus is rather striking and the vascular supply is — 
the same, this may merely point back to a common origin of all 
Taxineae. Cephalotaxus, moreover, has two ovules in the axil 
each bract, while Phyllocladus agrees with the Podocarpineae in 
having only one. : 
The presence of mesarch Gon: in the cladodes owes its signifcaul | 
evidence, but have little weight in themselves. 
The most convincing evidence of podocarp affinity lies in the 
entire behavior of the male gametophyte: the formation of prothé 
tissue, the freeing of the nuclei, and the early division of the genera- 
tive cell. The presence or absence of prothallial cells is a de 
group character; where they occur at all they are characteristic of 
entire family, and where absent in one member they are lacking 
all. Permanent prothallial cells are known nowhere in con 
except in Podocarpineae and Araucarineae; and the early divisi 
of the generative cell, which characterizes the former, occurs ! 
where else except perhaps in araucarians. Prothallial tissue 
primitive character, possessed, as we presume, by the ancestors “ 
conifers; but the Taxineae have eliminated it entirely, while it 5' 
remains one of the most characteristic features of the Podocarpin 
The morphology of the different forms of stamen in the con! 
is an open question. The Taxineae, according to COULTER - 
