112 THK KNTOMor.iXilST'S RKCOKD. 



cells, the immediate result will not be seen except in the case of a 

 dominant mutation, but there is evidence that these are much rarer 

 recessive ones. But if a recessive mutation should take place in the 

 sex chromosome oi DvowpJt'da it would show at once in a male in those 

 parts of the bod}^ of which the ctills contained the mutant gene, because 

 the male has only one x (sex) chromosome. Should a recessive muta- 

 tion occur in one x chromosome in the female its effect would not be 

 apparent, because there are two x chromosomes in the female, and the 

 normal allelomorph in the obher would conceal it. From this it will 

 be seen that on the first two theories an equality of the sexes is to be 

 expected in somatic mosaics, whereas on the theory of somatic muta- 

 tion they should all be males in the case of Drosophiia. Actually ten 

 were males out of twelve. One of the females I have mentioned above, 

 the other is a complex case, and to have produced it an additional ab- 

 normality must have taken place. Morgan discusses it very fully on 

 page 69. 



In the case of Lepidoptera a recessive mutation in one sex chromo- 

 some would be apparent only in the soma of the female, because it is 

 the female in the order, which is heterozygous for sex, and has a single 

 X chromosome, whereas the male is homozygous for sex and has two. 

 Thus in a male a recessive mutation in one x chromosome would be 

 concealed as in the female Drosophiia, and cause no effect on the out- 

 ward appearance of the insect. 



In a female, if it occurred at the first cell division of the ovum, in 

 one of the daughter cells it would produce an equally divided somatic 

 mosaic. 



The proportion of the sexes in somatic mosaics in Lepidoptera is 

 very important, a fact which I did not realise when I published the list 

 in the Journal of Genetics. In the present more complete list I have 

 given the sex whenever it was possible. 



It contains in all 88 females, 12 males, and 15 in which the sex is 

 uncertain. If we omit the seven female and the two male M. tiUae, 

 probably due to injury, and the pathological A. uiticae and the male 

 hybrid trinvi, we have left 30 females and 9 males. Omitting the three 

 probable intersexes of coriil<»i and the five females of coridoii and an/ii.s, 

 22 females and 9 males remain. Of these a higher proportion of 

 males than of females are doubtful examples, and the preponderance 

 of females is understated rather than overstated. This is in favour of 

 the explanation that a number are due to a recessive mutation of the 

 sex chromosome of one of the daughter cells, formed by the first 

 division of the fertilised ovum. The true somatic mosaics in males 

 must be due to elimination of an autosomal chromosome or fertilisa- 

 tion of a binucleate ovum, or to some still more uncommon cause, as 

 in the hybrid n-iitni. And since these causes are as likely to produce 

 mosaic females as males, an equal number of females are probably due 

 to one or other of them. 



It must be admitted that all the examples of a mosaic known to 

 be due to Mendelian factors are female, but this is probably because 

 we know so little of the inheritance of colour in Lepidoptera. 

 Asymmetry due to some pathological condition attributable to en- 

 vironment and unconnected with heredity may explam some males, 

 but I feel sure that others, for instance the male Apatiira Ilia, are 

 somatic mosaics. 



