354 Referate. 
other more recent cases (C. 9. in Nos. g and Io below) must be reckoned 
with in formulating hypotheses, and proposes that instead of adopting the 
formula male = ff, female = Ff, as has frequently been done, the male 
should be MM, the female Mm, on the ground that the exceptional trans- 
mission of a sex-limited character from the female to her female offspring 
instead of to her sons is thus more easily explicable. His assumptions 
with regard to chromosomes are perhaps more far-reaching than the known 
facts at present warrant. He suggests also that the somewhat anomalous 
distribution of the characters among the sexes in the crosses by Standfuss 
of Aglia tau and its var. dugens, and of the dwarf guinea-pigs recorded 
by Miss Sollas, may be explained by incomplete sex-linkage of these 
characters. 
Arkell and Davenport (3) have repeated and amplified Wood’s 
experiments in crossing horned and hornless varieties of Sheep; they have 
made all the possible matings and have obtained enough offspring from 
each to give a good idea of the inheritance. They put forward the hypo- 
thesis that the apparent dominance of the horned character in only male 
heterozygotes is due to the presence of a sex-limited inhibitor (I). They 
assume that the sex-limitation is of the Dvosophi/a type (as in Man and 
the Cat), so that the male is heterozygous for I (li), transmitting it only 
to his daughters, white the female is homozygous (II). They further assume 
that I is only effective when the horn-factor H is heterozygous and I is 
homozygous, so that HHII and Hhli are horned, HhII hornless. 
Castle (4) points out that this is inconsistent with the known fact 
that the development of the horns *s conditioned by the presence of a 
functional testis, a fact of which confirmation is given by Marshall (5), 
who finds that castration prevents the further growth of the horns at any 
stage. Castle maintains that this is inconsistent with the explanation by 
means of an inhibitor in the female, and that the inhibitor-hypothesis has 
no basis of fact. Arkell and Davenport reply in ‘Science’ of June 14, 
Igt2, without giving any fresh facts. 
Little (6) records the results of matings between orange and black 
eats, which indicate that the orange male transmits the orange factor only 
to his female offspring; and therefore suggests that the factors for orange 
and black in Cats are sex-limited in the male. 
Doncaster (7) gives confirmatory evidence that the orange factor is 
sex-limited in the male cat, but his data indicate that the sex-limitation 
is not absolute, but that an orange male mated with a black female may 
have occasional black female kittens, and more rarely tortoiseshell male 
kittens. 
Wentworth (8) in a short note gives evidence that the inheritance 
of rudimentary mammae in the pig follows the same ru’es as that of horns 
in sheep, that is, their presence appears to behave as a dominant in the 
male and recessive in the female. It is of course not certain that the 
inheritance is sex-limited in the strict sense of the word. 
Staples-Browne (g), in addition to a quantity of data not directly 
bearing on sex-limited inheritance, shows that certain characters in domestic 
pigeons and in turtle-doves have sex-limited transmission in the female, 
and further that in this case as in that of the Canary exceptions occur 
which can only be accounted for by the assumption that the coupling of 
the characters concerned with the sex-factor is not absolute. Confirmatory 
results are recorded by Cole (ro). In general, the most important results 
