896 
Journal of Agricultural Research 
Vol. XXX, No. 10 
by means of centrifugal force exerted 
upon them while they are immersed in 
a liquid of very nearly the same density 
as themselves is possible theoretically, 
and since some preliminary experiments 
along that line had indicated that suc¬ 
cess might actually be obtained in this 
way the experiments which are reported 
in this paper were designed to test that 
point. 
DIMORPHISM IN SIZE OF SPERMATOZOA 
The existence of two distinct sizes 
of spermatozoa in many animals is to 
be expected, particularly in those in 
which the Y-chromosome is absent or 
very much smaller than the X-chromo- 
some. The head of the spermatozoon 
is composed almost entirely of chro¬ 
matin, and therefore the spermatozoa 
which receive an X-chromosome should 
have larger heads than those which re¬ 
ceive the Y-chromosome except in 
cases where the Y-chromosome is as 
large as or larger than the X-chromo- 
some. This has actually been found 
in a number of species of vertebrates 
and insects (AS, 44, 45, 46, 47)- Con¬ 
sidering that the size of the spermato¬ 
zoon head depends upon three dimen¬ 
sions and possibly may vary in any 
one or all of these dimensions more or 
less independently, it is remarkable 
that the results obtained show rather 
uniformly two size groups in nearly 
equal numbers when only the head 
lengths were measured. However, each 
species has a rather characteristic shape 
for its spermatozoon heads and the 
length of the head may be a fairly re¬ 
liable index to its total size. Two 
classes of spermatozoa have been found, 
the differences between their mean head 
lengths approximating very closely in 
most of the species the theoretical dif¬ 
ferences calculated from drawings of 
the chromosomes as seen during ma¬ 
turation divisions of the germ cells. 
EXPERIMENTAL PROCEDURE 
FACTORS INVOLVED IN THE PROBLEM 
In undertaking experimental work 
involving the separation of sperma¬ 
tozoa by centrifugal force, several fac¬ 
tors had to be considered. If there is 
no difference in the density of the two 
sizes of spermatozoa the larger ones 
will offer less resistance than the smaller 
ones, per unit of weight, to locomo¬ 
tion through a surrounding liquid, 
whether the specific gravity of the 
liquid be greater or less than their own, 
because the volumes of the larger ones 
(and therefore their weights, upon 
which depend the forces tending to 
move them through the liquid) are 
proportionately greater per unit of sur¬ 
face or cross section (upon which de¬ 
pends the resistance which they offer 
to locomotion through the liquid). If 
the specific gravity of the liquid were 
very much more or very much less than 
that of the spermatozoa themselves, a 
very little centrifuging would cause 
both classes of spermatozoa to go 
either to the inside or to the outside of 
the centrifuge tube together, but if the 
specific gravity of the liquid were very 
nearly that of the spermatozoa their 
progress during the centrifuging would 
be very much slower, and if the right 
amount of centrifugal force were used 
for the right length of time a higher 
percentage of the large than of the 
small spermatozoa would reach the in¬ 
side or the outside of the centrifuge 
tube, according to whether the specific 
gravity of the liquid were more or less 
than that of the spermatozoa. If, 
however, the two sizes of spermatozoa 
differ in specific gravity, a thing for 
which there is no proof and which 
there are no a priori reasons to expect, 
but which might nevertheless be true 
because the chromatin of the sex chro¬ 
mosome might be of a different spe¬ 
cific gravity from that of the other 
chromosomes, the problem becomes 
somewhat different. If, in this case, 
the specific gravity of the liquid could 
be made intermediate to that of the 
two classes of spermatozoa, it would 
be possible to effect a more complete 
separation than under any other con¬ 
ditions, for prolonged centrifuging 
should result in one class going to the 
inside and the other to the outside of 
the centrifuge tube. If the larger 
ones have the greater specific gravity, 
they would move .through the liquid 
still more rapidly than if there were 
no difference in' specific gravity, since 
in addition to having less resistance per 
unit volume they would also have a 
greater weight per unit volume than 
the smaller ones. If the smaller ones 
have the greater specific gravity, this 
would counterbalance their greater re¬ 
sistance per unit volume, and the only 
means of separating them by centrif¬ 
ugal force would be to adjust the spe¬ 
cific gravity of the liquid to a value 
intermediate to that of the large and 
of the small spermatozoa. 
Thus, there are at least six important 
factors which influence the success of 
the centrifugal method of separation of 
the male-producing from the female- 
producing spermatozoa. The first is 
the difference in size of the two classes, 
a factor which is not under control at 
