504 BOTANICAL GAZETTE [JUNE 
rudiments of its seed cones nearly a year in advance of pollination, 
while here there intervenes scarcely any time at all. In fact, in 
California the greater part of the pollen is likely to be shed before 
there are any ovulate cones to pollinate. Observations in the 
native habitat might show a different state of affairs and one more 
nearly paralleling that of Agathis. Other conifers of course are 
known in which the total time is even shorter than 21 months, but 
I am not aware of any one in which the time is distributed in the 
same manner. 
The very considerable number of free nuclei before cell formation 
is another character that, taken with the very large gametophyte, 
reminds one of more primitive gymnosperms. Gametophytes of 
this size are known only among the Cycadales, Ginkgoales, and some 
taxads (as Torreya). A curious feature of its development is the 
strange absence of mitoses in the free nuclear preparations. 
The manner in which the free nuclear stage passes into a game- 
tophyte with walled cells is apparently different from that reported 
for any other plant. While there is a certain resemblance to the 
centripetal growth with ‘‘alveoli” reported for Sequoia (4) and 
others, yet the exact method is really quite different. The most 
essential feature of this difference lies in the delay of walls. If 
walls formed between the nuclei before the beginning of the centrip- 
etal movement of the inner border of the cytoplasm and were 
extended pari passu with it, and the nuclei divided to keep pace, 
there would be no very essential difference. While far too little is 
known about the exact methods of centripetal growth and wall- 
formation in any considerable number of genera to make any con- 
- clusions based on this feature more than tentative, it is clear that 
Araucaria need not be excluded from relationship either with 
podocarps or with abietineous conifers on this account. 
While there is no proof that the gametophyte invades the 
nucellus in Araucaria, neither is it proved that it does so in other 
genera where similar appearances are commonly observed. The 
homology of the spongy tissue and its functions is none too clear 
anywhere. The peculiar glandular nucellar tip is found elsewhere 
only among conifers of podocarpineous affinities. THompson has 
made the suggestion that the method of pollination found in the — 
