478 SEMI-CENTENNIAL OF TORREY BOTANICAL CLUB 
pollinated with grains from the smallest anthers were less fruitful 
than in the case of other illegitimate crosses. 
An additional suggestion may be gathered from Halsted’s 
experiments made to determine whether the small pollen grains 
were fertile, since doubt on this point had been expressed by more 
than one previous writer. He found all the grains equally capable 
of germination if sufficient time was given, but that the largest 
grains germinated much more promptly. In Pontederia such 
promptness of germination of the large grains would be of great 
importance for the long- and mid-styled plants, inasmuch as the 
style withers so early that the pollen-tube of a slow-germinating 
grain might be unable to reach the ovule. I have found that large- 
and mid-size pollen-grains of P. montevideensis. both germinate 
very quickly in weak sugar solutions; the only apparent difference 
is that the pollen-tube from the mid-size grains has a diameter 
about three fourths as great as that of the large grains. I had no 
flowers containing small grains for comparison. But this is a 
point which can hardly be settled by study of one small group of 
species. 
In Lythrum and other heterostyled flowers it has been noted 
that the stigma of the long-styled form is larger than those of the 
mid- and short-styled flowers, and it has been considered that 
the longer stigmatic papillae are adapted to receive the large pollen- 
grains. In Pontederia cordata there is very little difference in the 
length of the papillae in the different stigmas (FIGS. 8-10) but 
the stigma of the long-styled flower is frequently, though not 
uniformly, six-parted, and this spreading stigma may be regarded 
as directly correlated with its exserted position, for such a stigma 
would have a distinctly better chance of being dusted with pollen 
by the insect visitor. In the case of the mid- and short-styled 
pistils, however, there is no need for such a spreading stigma, since 
the perianth-tube would almost certainly guide the pollen-smeared 
proboscis of an insect in such a manner as to brush even a narrow 
stigma. 
Observation of the manner in which the flowers are placed on 
seeds per capsule; other flowers fertilized by pollen from the mid-length stamens of 
the same spike produced 121.3 seeds; while still other flowers fertilized by pollen 
from the short stamens of mid-styled flowers produced 113.3 seeds. Kosmos 13: 
298. 1883. 
