thrown on Tarchanoff's observations. Steinach remarked, however, that
one's faith is rather shaken by the fact that in the _Esculenta_, which
in sexual life closely resembles _Rana temporaria_, there are no seminal
receptacles. He therefore repeated Tarchanoff's experiments, and found
that the seminal receptacles were empty before coitus, only becoming
gradually filled during coitus; it could not, therefore, be argued that
the sexual impulse started from the receptacles. He then extirpated the
seminal receptacles, avoiding hemorrhage as far as possible, and found
that, in the majority of cases so operated on, coitus still continued for
from five to seven days, and in the minority for a longer time. He
therefore concluded, with Goltz, that it is from the swollen testicles,
not from the seminal receptacles, that the impulse first starts. Goltz
himself pointed out that the fact that the removal of the testicles did
not stop coitus by no means proves that it did not begin it, for, when the
central nervous mechanism is once set in action, it can continue even when
the exciting stimulus is removed. By extirpating the testicles some months
before the sexual season he found that no coitus occurred. At the same
time, even in these frogs, a certain degree of sexual inclination and a
certain excitability of the embracing center still persisted, disappearing
when the sexual epoch was over.

According to most recent writers, the seminal vesicles of mammals are
receptacles for their own albuminous secretion, the function of which is
unknown. Steinach could find no spermatozoa in these "seminal" sacs, and
therefore he proposed to use Owen's name of _glandulæ vesiculares_. After
extirpation of these vesicular glands in the white rat typical coitus
occurred. But the capacity for _procreation_ was diminished, and
extirpation of both _glandulæ vesiculares_ and _glandulæ prostaticæ_ led
to disappearance of the capacity for procreation. Steinach came to the
conclusion that this is because the secretions of these glands impart