Medical Officers of the Army of India.

43

dite, with perfect male and female organs, but their construction is such that
self-fertilization cannot easily take place. The anthers are introrse and dehis-
cence apical, but the upper portion of the cone is so closely applied to and
around the stigma, that pollen could scarcely fall spontaneously upon the stigma;
in fact self-fertilization could only be effected at a mechanical disadvantage.
On the other hand, all the circumstances of the case appear to favour the idea
of fertilization by insect agency. But flies and ants are obviously not destined
to be the agents; these are caught and perish, while stronger ones, such as bees
for instance, or butterflies and moths, perhaps having a longer proboscis, may
be able to release themselves (and a long proboscis would be an advantage for
withdrawal through the apex of the cone) and so to carry away the pollen to the
next flower they visit. The fact that the stigma must be reached before the
pollen in the process of withdrawal supports this view. Pollen adhering to
the tip of a proboscis after its extraction from one trap would certainly be
wiped off upon the stigma of the next one in which it may chance to be ar-
rested. Such are the points that I have observed in connexion with this ap-
parently wanton destruction of insect life by the Wrightia , and the interpre-
tation they appear to suggest, but I by no means claim to have finally solved
the mystery, for I admit that I have never actually seen a bee or other larger
insect engaged in the flower, or in the act of releasing itself from the trap.

G2