Chi'omoHome>i <ni(l Sex in Abraxas 



o 



females in the inirnial ratio. The tendency to produce only fenial 

 offspring has usually descended in the direct female line, ])ul has 

 occasionally missed a generation, so that there were one or two cases 

 of females of bisexual families having only female offspring, and others 

 where such females had families in which males were produced, hut 

 only in extremely small numhers. 



The complete pedigree of the all-female strain is given in Table I. 

 The details of the 1912 matings are given in Table II ; those of previous 

 years have been given in the earlier paper. Table II includes the 

 results of all the matings made, with the exception of some which were 

 nearly or quite sterile, or which have no close bearing on the (piestions 

 discussed in the present paper. As in the previous paper, the reference 

 numbers of families consisting entirely of females are printed in italics, 

 in both the table and the text; families in which there were at least 

 twice as many females as males are distinguished by thick type. 



The moths reared in the year 1912-13 (from pairings made in 

 1912) confirmed the previous results, but added some new f;icts. 



The chief new point of importance is the fact that broods which are 

 not entirely of one sex, but in which the females greatly outnumber 

 the males, may arise among the offspring of females which belonged to 

 unisexual broods (e.g. '12.1, '12.21, '12.21 B). It appears, in fact, that 

 there is no sharp distinction between uTiisexual broods and broods in 

 which the sexes are in nearly equal nimibers, but that various grades 

 of excess of females occur, leading to the extreme condition in which 

 all the offspring are female. These results therefore suggest that the 

 existence of unisexual broods is not due to the absence of a male- 

 determining factor from the mother, but that there is a tendency, of 

 varying amount, for this f;xctor to be eliminated from the egg. The 

 facts of sex-limited inheritance have already led to the conclusion that 

 the female in Abraxas is heterozygous for a sex-factor, and in normal 

 families it must be supposed that this factor is present in half the eggs 

 and absent from the other half In the strain which produces female 

 families it may be supposed that the factor, whatever it may be, which 

 determines maleness, has a tendency to be eliminated, in some cases 

 from all the eggs, in others from much mori' than half, instead of its 

 retention or elimination being a matter of chance. It will be seen 

 below that this su{)position is to some extent supported by cytological 

 observations on the maturation of the eggs. 



There are one or two further points in the 1912-18 families which 

 require notice before passing on to the cytological observations. In 



