From our very earliest studies on this subject (18,19,20), we
have routinely used as an estimate of the incidence of improper em-
bryogenesis in cattle the proportion of the cow population which
returned to service in more than a standard interval of 46 days mean
length after insemination (22).

The final fertility of a cow population has always been that
residual of the inseminated population which did not show estrus for
at least a mean of 167 days after that first insemination (22), and
in some of our studies the data are for exactly 200 days (21). The
statistical significance of differences in variance levels associ-
ated with the potential sources always has been rigorously tested
by least squares analysis in each of the experiments.

Results

In the figures (lower left) are presented actual data from a
total of 423,054 first inseminations and a total of 159,038 returns
to insemination in which the absolute date of the return to estrus
was known in every case (21). The data are for semen stored at 4°C
for 5 days. The figure presents in a visual form the general re-
sults on sperm storage found in all of our investigations.

Note that there is an increase in fertility from the first to
the second period of use, followed by a period of optimum fertility
the duration of which is inversely related to the storage tempera-
ture, which is then followed by a constant decrease, in this case a
straight-line regression., These differences are found irrespective
of the time after insemination chosen for the fertility assay. Con-
versely, using the difference between the estimated fertility level
at 24 days after insemination and the final estimate, at 200 days,
as an absolute estimate of the number of cows bred to semen of each
age class which lost a conceptus, it is clear that prenatal losses
vary inversely as the fertility level. The differences shown in
fertility from day to day for the several time periods for assay of
fertility after insemination are highly significant statistically.
So, too, are the differences shown among the days of semen storage
in the prenatal losses for fertility estimates based on the 24-day
minus the 200-day and also for the 46-day minus the 167-day non-
return data (22). The absolute decrease in apparent pregnancies and
in relation to the initial estimate of apparent pregnancies at 24
days for each day of semen storage is shown in Table 1. Table 2
gives conversion factors showing the comparative magnitude of the
two estimates of embryonic losses.

From these investigations it was proven not only that fertil-
ity level and embryonic mortality changed as the stored semen aged
but that we could use fertility estimates routinely obtained in com-
mercial A.l. for continued studies of the effect of sperm aging on
fertility and embryonic losses.

1190