414 
CECIL YAMPOLSKY 
proportion of male flowers on what I have called prevailingly female plants 
is probably an insignificant fraction of one percent. 
Swelling of the ovary begins a few days after pollination, and ripe fruit 
appears in a few weeks. 42 seeds were collected during the period in April 
when stamens appeared on the female plant A . The plant then ceased to 
produce male elements and seed production ceased. Two weeks later 
stamens again began to appear and seed production followed. 24 more 
seeds were collected, making the total number of seeds produced by the 
plant 66. There was no further evidence of male elements. The plant 
continued to live but with decreasing vigor, the older branches dying off, 
the younger ones persisting. Female flowers were produced until the very 
end. The plant was nearly two years old when it died. 
Fi Offspring of the Original Mother Plant 
I have reported briefly on the sex of the offspring of this original plant 
A, but will summarize the results here for the sake of completeness. The 
lots of seed gathered from the mother plant during the month of April, 
1914, were sown in pots of previously sterilized soil. In the first lot 36 
germinated and in the second lot 14 germinated. Each seedling was potted 
and labeled and its history was recorded. They were kept in the green- 
house of the propagating house of the New York Botanical Garden through- 
out the experiment. They were repotted from time to time. As they con- 
tinued to grow they were placed further apart, so that there was little or 
no contact of vegetative parts. 
The sex of the seedlings, whether they be prevailingly male or prevail- 
ingly female, can be determined, as noted, within two or three weeks after 
germination. The 50 seedlings raised from the 66 seeds were all prevail- 
ingly female. The plants were carefully examined at three-day intervals. 
All the branches of one plant were examined on the same day. With the 
exception of plants nos. Ill and X (see table i), the time of appearance of 
male flowers and seeds synchronize exactly. In the two exceptions, 3 and 
2 seeds respectively were set in October about six months after the seedlings 
were up. In the case of plant no. Ilia male flower was seen preceding the 
development of seed. In plant no. X, although no male flower was seen 
prior to the setting of the 2 seeds, it is safe to assume that -a male flower 
was developed in their vicinity. As can be seen from the table these 
plants produced seed later, as did the rest of the plants. 
The rather profuse appearance of male flowers and the subsequent 
development of seeds began in March, and lasted until the end of May. 
The stamens in the Fi plants varied in their number and arrangement 
in the flowers as was described for the parent plant A. In one case the 
anthers were found to be shriveled and sterile. Plant D in March produced 
a hermaphroditic flower having two stamens, both of which were abortive. 
The pollen of these two stamens also was shriveled. 
