Marcu 29, 1912] 
a hypothesis which has to be earried to such 
lengths, and I think this view is shared by 
many whose experience in breeding work has 
been very extensive. 
Webber has somewhere expressed the view, I 
am sure verbally and I think also in his publi- 
cations, that sport variation is more likely to 
occur in connection with repeated selection 
and in the same general direction as the selec- 
tion. This view, if correct, is highly important. 
A case which may be considered to support it 
has recently occurred in the breeding experi- 
ments in progress at the Bussey Institution. 
In mice, rabbits and eats, a dull black va- 
riety occurs which is known by fanciers as 
blue. The optical effect called blue is in such 
eases the result of a diminished number of 
black pigment granules in the fur. Fanciers 
have long desired to obtain a similar variety 
among guinea-pigs, but thus far without much 
success. Some eight years ago I became in- 
terested in the problem and began experi- 
ments which have continued to the present 
time. Knowing that Andalusian blue fowls 
are heterozygotes of black and white, I tried 
to produce a similar modification in guinea- 
pigs by cross-breeding. I found that crossing 
black with white gave results which varied in 
character with the white strain used. A sooty 
or “ Himalayan ” albino strain which by other 
experiments was shown to transmit intense 
black pigmentation produced no diminution 
of black in the heterozygotes. A very lightly 
pigmented albino which was found to trans- 
mit light yellow (“cream’’) in crosses pro- 
duced heterozygotes with a much duller black, 
but not of the desired blue tone. In other 
words, albinos in crosses with black were 
found to breed exactly like the colored strains 
from which they were extracted. As the 
eross with a cream strain had produced the 
lightest black animals thus far obtained, I 
confined my further experiments to crosses 
with this cream strain or with albinos derived 
from it. In each generation the lightest 
heterozygous blacks were crossed. with the 
lightest creams (or albinos). By this process 
a very considerable reduction in the amount 
SCIENCE 
509 
of pigment in the fur was secured. The hairs 
of the black individuals were now dull black 
at the tip only; elsewhere the hair was sooty 
cream colored, indicating a great quantitative 
reduction of the pigments, both black and 
yellow. These peculiar black individuals we 
may for convenience henceforth call “blue.” 
Blue parents mated with cream ones have in 
the past two years produced 17 blue, 15 cream 
and 13 white young, the Mendelian expecta- 
tion being 3 blue: 3 cream: 2 white, if all 
parents are heterozygous for albinism. In a 
mating of this sort a few weeks ago (Decem- 
ber 2, 1911) a blue mother gave birth by a 
cream male to a female young one which 
closely resembles an albino, its coat being in 
general white and its eyes pink, but on the 
right side of its head and on the hips are 
spots of blue. As in the pink-eyed mouse, the 
color of the fur is decidedly pale. In the iris 
of each eye may also be seen a faint pigmented 
streak. It is noteworthy that the pink-eyed 
mouse also has traces of pigment in its eyes. 
The genetic behavior of pink-eyed mice 
shows that the pink-eyed variation is due to 
modification (or partial loss) of some factor 
necessary for the production of the full pig- 
mentation. This factor, however, is not the 
color factor (C) which albinos lack, nor the 
yellow (Y), brown (Br) or black (B) factors, 
nor yet the agouti (A) factor, since with each 
of these and without it the pink-eyed varia- 
tion may form distinct combinations. For 
the same reason it can be shown to be distinct 
from the condition of spotting with white. 
It is noteworthy that the race of guinea- 
pigs in which the pink-eye variation has ap- 
peared is one in which a reduction of the 
amount of pigmentation was being attempted 
by a systematic selection, and was being actu- 
ally obtained. The pink-eye variation would 
seem, therefore, to be merely a particular long 
step in the general course of modification 
which this race was undergoing directed by 
artificial selection. If so, it has probably 
been brought about by the modification of the 
same color factor or factors that have under- 
gone modification in the blue race. This idea 
