514 
L. DOXCASTEl?. 
epistatic over M, giving a female, but that either, when 
homozygous, is always epistatic over the other when the latter 
is heterozygous. In this case, in the majority of insects and 
in mammals the male would be represented MMFf, the 
female M M F F, while in Lepidoptera and birds the male would 
be MMFF, the female MmF F. This hypothesis is in accord 
with the observed fact that in many insects and some mam- 
mals the male has one heterochromosome, the female two, 
while in the peculiar strain of Abraxas described, it is the 
female which has one chromosome less than the male. 
A second problem, which is really more closely connected 
with this one than may appear at first sight, arises from the 
considerable number of cases in which sex is actually or 
apparently altei*ed by conditions acting on the unfertilised 
egg or on the developing individual. In addition to such 
instances as that of the crab infected with Sacculina, in 
which Geoffrey Smith has shown that males may so far 
assume female characters as to produce ova in the testis, 
there are many cases in which such conditions as staleness of 
the eggs cause alterations in the sex-ratio (usually by 
increasing the proportion of males) which cannot be explained 
by selective mortality, nor, in fact, by any known cause 
except a change in the sex-determining power of the eggs. 
Such cases would seem to offer an almost insuperable 
obstacle to the belief that sex is determined by the presence 
or absence of a particular chromosome, and yet the facts with 
regard to chromosomes are too clear and definite to be dis- 
regarded on this account. It is possible, however, that 
suggestions towards a l econciliation can be found by taking 
into consideration some of the facts which are known about 
the relations of sex to general metabolism. G. Smith f53) 
has shown that the change from male to female secondary 
sexual characters in the crab infected with Sacculina is 
accompanied by deep-seated metabolic changes which are 
induced by the parasite. Steche (54) has brought forward 
somewhat similar evidence in the case of Lepidopterous larvae. 
He finds that there are important metabolic differences 
