276 
prolongation of the sexual cycle. It may be 
that also in pregnancy the decidua may 
contribute directly to the prolonged life of the 
corpora lutea and thus indirectly be respon- 
sible for the lack of ovulation during the 
period of gestation. We found further that 
during pregnancy the life of experimental de- 
ciduomata is prolonged very considerably, and 
we may provisionally assume that directly or 
indirectly the presence of an embryo is re- 
sponsible for the preservation of deciduomata 
during pregnancy. It accords with these ob- 
servations that in a case of experimentally 
produced extrauterine pregnancy in which a 
living embryo, but no decidua was present, we 
did not find the life of the corpus luteum pro- 
longed, and consequently an ovulation had 
taken place despite the presence of a living 
embryo.2° In a provisional way we may there- 
fore assume that the life of the corpus luteum, 
which is one of the determiners of the dura- 
tion of the sexual cycle, depends among other 
factors upon the life of the deciduomata or 
decidua and that the latter depends upon the 
existence of a living embryo. We have carried 
out a series of experiments concerning this 
question several years ago, but stress of other 
work has so far prevented a detailed study of 
our results. However, we referred to them in 
a preliminary way in a recent communication 
dealing with those problems. 
We have now to consider briefly the reason 
why it is that while we can consider as firmly 
established the significance of the corpus 
luteum for the sexual cycle, injections of dried 
corpus luteum substance are without a decided 
effect on ovulation, and likewise without effect 
on the growth of the mammary gland in the 
guinea-pig, as we established more recently. 
We can not answer this question definitely at 
the present time, but we can at least consider 
certain possibilities and refer to some interest- 
ing analogies. It might be that the isolated 
lipoid extract of the corpus luteum would have 
been active where we established the lack of 
efficiency in dried gland from which, as was 
the case in Armour’s preparation, the lipoids 
10 Leo Loeb, Biological Bulletin, 1915, XXVIIL., 
p. 59. 
SCIENCE 
[N. 8. Von. XLVIII. No. 1237 
had been previously removed. This ‘would be 
in accordance with the observation made by 
several investigators who found that injection 
of lipoid extracts of corpus luteum or placenta 
causes growth processes in the uterus and 
mammary gland. However, these induced 
growth processes are evidently not identical 
with the cyclic changes normally taking place 
in these organs. Or it might be that the proc- 
ess of drying destroyed the active principle. 
Thus we know that while suspension of living 
cells when injected produce immunity against 
tumor growth in the mouse, cells which have 
been previously treated in a way similar to the 
treatment accorded to the corpus luteum and 
thus killed, have completely lost their effi- 
ciency. We know furthermore, that the anti- 
gens against mouse tumors are species specific; 
on the whole only tissues of the mouse are able 
to immunize against mouse tumors; tissues of 
the cow for instance being completely inert. 
We can not therefore exclude the possibility 
that extracts prepared from homologous cor- 
pora lutea might have been more efficient than 
those from the cow. : 
There remains a last possibility which I sug- 
gested a number of years ago when I found 
that a substance given off by the corpus 
luteum is one of the factors of significance in 
jnitiating the decidual reaction and the de- 
velopment of deciduomata in the uterus. At 
that time I tried to imitate the effect of the 
corpus luteum on the mucosa of the uterus 
through implantation of living young corpora 
lutea obtained from other guinea-pigs in the 
appropriate stage of the sexual cycle. The im- 
plantation of this substance gave either en- 
tirely negative results or at least its effects 
were very weak. I then pointed out that the 
corpus luteum functions by giving off a 
small amount of substance continuously dur- 
ing a relatively long period of time, while in- . 
jection or implantation of corpus luteum sub- 
stance leads only to the temporary introduc- 
tion of a larger quantity which is probably 
rapidly absorbed and eliminated or destroyed, 
and that it is impossible to imitate in this way 
the action which takes place in nature. 
While we can not be certain at the present 
