G Chromosomes and Sex in Abraxas 



character is therefore probably a recessive mutation, but shows no sign 

 of being sex-limited in transmission. The Rev. G. Waddington, S.J., 

 writes to me that he also finds evidence that the black form is recessive 

 to the type. In a brood of black larvae, the completeness of the 

 blackening varies somewhat in different individuals, bat even those 

 with least black are conspicuously different liom the normal. 



h. The Oor/omal Chromosomes. 



In my former paper on the strain which produces uni.sexual families, 

 I gave evidence that females belonging to unisexual broods have 55 

 oogonial chromosomes instead of the 56 which I had previously shown 

 to be the usual number for the species. The evidence consisted chiefly 

 in the observations that of brood 'JJ.S, 33 larvae were dissected, all 

 of which were females ; nine of these larvae provided altogether 18 

 oogonial mitotic figures, in which the number 55 was almost or quite 

 certainly correct, and in other larvae a number of figures were found 

 in which 55 was the most probable number. In female larvae of broods 

 '12.1 and '12.32 also, five figures were found each with 55, and several 

 others in which 55 was most probable but not aksolutely certain. Since 

 of these broods '12.8 was the only ope which was known at all certainly 

 to consist only of females (many larvae being still too small to dissect) 

 it was concluded that all females belonging to unisexual broods had 

 55 instead of 56 chromosomes, but that it was doubtful whether females 

 belonging to bisexual families descended directly from unisexual iiad 

 always 55, or sometimes 55 and sometimes 56. 



The work during the spring of 1913 has definitely shown that not 

 only females of unisexual broods, but also females of bisexual broods 

 which are inmiediately descended from unisexual broods, alwaj's have 

 55, with one probable exception to be referred to later. In the following 

 account the observations previously recorded on brood 'li^.8 will not 

 be described again, but the cases of broods '12.1, '12.25, and '12.32, 

 mentioned in the former paper, will be included together with the 

 observations made in 1913. 



In all, in addition to the 33 female larvae of 'Ji'.S recorded in the 

 former paper, the ovaries of about 160 female larvae were dissected out 

 and sectioned. Of these 39 larvae belonging to the unisexual strain, 

 and eight not belonging to this strain, yielded ovaries with figures 

 considered good enough to count. Of the larvae belonging to the 

 unisexual strain, eleven belonged to three broods {'1:^.19, '1J.39, 

 '12.29 B) which consisted entirely of females; fifteen belonged to 



