(381) 
tion of Schleiden that the placentation is mainly of an axial 
character and that the carpels contribute principally the upper 
portion of the ovary and the style. Here also we are in accord 
with the conclusions of Lestiboudois,” who studied the process 
of development in Caucum7s and Bryonta. He states that the 
carpels appear as projections which elongate and finally close 
the cavity before the formation of the ovules. 
Furthermore, that the carpellary ‘*mamelons” are the 
same in appearance as those which represent the sepals, pet- 
als and stamens (staminodia). This appears most distinctly 
in the case of Fevillea (pl. 58, fg. 6), but in all the other 
cases it is to some extent apparent and in some more than in 
others. After the first appearance of the carpels as minute 
lobes on the peripheral surface of the torus, there is in every 
case an accelerated growth of the periblem layers situated 
just below the insertion of the carpels, that is, toward the 
center of the torus. The growth of the ovary takes place 
then mainly below the insertions of the carpels, the cells 
around the bottom of the cavity retaining for some time a 
distinctly meristematic character. The bottom of the original 
concavity which is indicated by the termination of the stylar 
canal is in some cases (Zwffa) far removed from the lower 
end of the placenta in its later stages, the growth having oc- 
curred mainly in the lower portion of the ovary. There is 
no evidence of the fusion of the backs of carpels, either with 
axial tissue or with portions of the perianth, and such an as- 
sumption here seems unnecessary. 
Text-figures z—7 inclusive show the course of ontogenetic 
development in the gynoecium as already described at length 
in the treatment of the organogeny. The zone of growth 
just below the insertion of the carpels is, in every case, 
clearly apparent. In certain cases, as in Fevz/lea, the pla- 
centa is evidently foliar in origin (text-figure 5, A), and 
possibly in S%cyos (text-figure 6, 7), but in all the others 
examined it arises not from carpellary tissue but by a thicken- 
ing, in certain definite regions, of the ovary wall. 
The influence of the carpels is sometimes manifest in an 
