208 The Botanical Gazette. [May 
were the rule in these two genera. My sections of embryo- 
sacs, which were just approaching maturity, resemble Martin's 
figures and lead me to suspect that his conclusions have been 
drawn from material in an early stage of development. 
Strasburger has noted that there are sometimes more than 
three antipodal cells, and Mottier* figures a case in which 
each of the three cells has divided. 
In rare cases I found just three antipodal cells, each with a 
single nucleus, three cells with doubled nuclei were not quite 
so rare, while the condition represented in the figures was not 
at alluncommon. In fig. 2 we have six antipodal cells, ar- 
ranged in a single longitudinal row, with the divisions ap- 
proximately in the same plane. Fig. 11 shows seven antip- 
odal cells which present more complexity in their arrange- 
ment. Fig. 13 has nine antipodal cells with still another 
variation in the plane of division. Fig. 3 goes a step farther 
and displays thirteen cells. The first three of these cells, of 
course, arise from free cell formation, but when the number 
exceeds three, the extra cells are produced by cell division 
with the formation of partitions. If the partition is not 
formed at the first division of the nucleus, as in fig. 11, 4, I 
am inclined to think that it will not be formed later, at least 
I have not seen anything which would lead me to believe that 
partitions are formed after the nuclei have begun to multiply, 
as in fig. 10. 
I am well aware that the doubling of nuclei is not unusual 
in these cells, but neither my reading nor my preparations of 
other embryo-sacs foreshadowed the condition represented 
in fig. 13 where we have thirty nuclei in a single section. 
Fig. 10 shows twenty nuclei in one section of a single cell. 
The occurrence of mitotic figures proves that these nuclei 
multiply by indirect division, but whether they multiply by 
fragmentation also, I am not prepared to say, although some 
of the cases figured would suggest such a possibility. 
The homology of the antipodal cells has long been a sub- 
ject for controversy. Without reviewing theories, it seems 
to me that Strasburger was correct in making them homolo- 
gous with the prothallium of the gymnosperms. This homol- 
ogy seems sound when we compare their origin with that of 
the gymnosperm prothallium, but the gymnosperm prothal- 
. On the embryo-sac and embryo of Senecio aureus L. Botanical Gazette 
18: 245. Jl. 1893. 
