ESTUDOS SOBRE A BIOLOGIA ELORAL DA MANCHARA INDICA L. 
153 
The ovary contains but one ovule. Néither style nor Stigma present 
ánymarked peculiarity, nor does the latter offer any struicture that 
would faeilitate the adhesion of pollen. 
I believe, with other authors, that pollination is acomplished by in- 
seets. The presence of diptera, specially Musca domestica and Antho- 
myiae is very notieeable on the trees and small Hymenoptera occur also. 
They seem attracted by the perfume and possibly by the secretion of 
néctar. Diptera specially are very plentiful. In the laboratory they be- 
eome a nuisance while flowers are being examined. 
It is however only fair to state that the fertile anther was Some ti- 
mès found at the levei of the stigma, at others slightly below it; occasion 
ally they seemed at first sight to be in direct contact with one another, 
but careful examination shows that itisnotso. 
At dehiscence the anther is pla:ed horizontally, so that sometimes 
one of the pollen sacs seems in a position favourable to -emptying its con- 
tents over the flower ? but not always in a position that might make it 
possible to attain the stigma. The style is somewhat lateral and obliqúe. 
I found slight differences in the maturation of pistil and stamen 
but no clear dichogamy. Direct proof of the nature of pollinisation was not 
obtained. In one of the few flowers in which a pollinated stigma was obs- 
erved, the anther was open. On ancther occasion a dipteron which obli- 
gingly placed itself under the microsope on a flower being examined at 
that moment, showed a few pollen grains one the head, exactly where 
one would expect to find tnem, given the position of the anther. 
I found many variatíons in the flowers, both of th# perianth and 
the sporophylls. According to the variations of the perianth, there are 
pentamerous, tetramerous and hexamerous flowers and intermediate 
forms, with a larger number of petals and disc lobes than sepals. (see 
pictures). At time we found polype:alia and staminoids. Concrescent 
flowers were seen three times, twice between androgynous flowers and 
oneé between male flowers. 
As to sporophyll variation we found the following intermediate 
stages, between male and hermaphrodite flowers. Male flowers with a 
slight cavity between the lobes of the central disk; similar 
flowers with a dark spot in the center of the cavity; flowers with a 
small eminence corresponding to a rudimentary ovary or with more 
developed rudiment, or with an apparently normal ovary, but without 
style or stigma, ending in a cone or even a slight depression. 
Careful investigations of the relative position and distribution of 
male and hermaphrodite flowers in a great many inflorescences did not 
rtisclose definite proportions or constant points of production, but sho¬ 
wed that: 
a) The male flowers were generaíly much more numerous than the 
hermaphrodite. 
