8 Female Fmnilies in Abraxas 



decision between the numbers 55 or 56 was possible. In one of the 

 latter, however, several figures gave 55 and only one 56, and from a 

 slight difference in staining there is reason for believing that one of the 

 bodies counted as a chromosome among the bQ is really an extra- 

 nuclear body at the edge of the chromosome group. Among the 

 offspring of pairing '1^. 8, therefore, we have over 20 figures in which 

 the number 55 is certainly or almost certainly correct, four in which 

 only 54 can be seen, and none in which there is conclusive evidence of 

 as many as 56. The number 55 is also the most probable in several of 

 the ovaries in which the figures are not sufficiently clear to be used as 

 evidence. 



Two other families, '12. 1 and '12. 32, provided sufficient larvae to 

 give considerable probability that they will turn out exclusively female. 

 Both are from females belonging to unisexual families; '12.1 from 

 lact ? ex. '11. 4- X lact. ^ ex. '11. 6, and '12. 32 from lact. $ ex. '11. 9 

 X wild cT. Of '12. 1, four larvae were dissected, all females; two pro- 

 vided ovaries with three countable figures, all with 55. In '12.32, six 

 larvae were dissected, again all females ; four of these show good or fair 

 figures, two figures with 55 clearly, and four others with 55 probably. 

 In the three families, therefore, there are about thirty good equatorial 

 plates showing 55 chromosomes, and only eight in which 55 or 56 might 

 be counted with equal probability. 



In only one other family the mother of which belonged to a 

 unisexual brood have any definite results been obtained up to the 

 present time. Family '12. 25, from gross. % ex. '11. ISxlact. ^ ex. '11. 14, 

 has given ten male and four female larvae. One of the females provides 

 an equatorial plate with 56 chromosomes shown quite clearly; two 

 others probably have 56 and one probably 55 but possibly 56. In the 

 case, therefore, of bisexual families of which the mother belongs to a 

 unisexual family it is clear that some at least of the females have h% 

 chromosomes, but it is probable that others have 55. 



The testes of a number of males from families of which the mother 

 belonged to a unisexual family have been examined, chiefly from '12. 25 

 and '12. 21 B (the mother of the latter was gross. % ex. '11. 26, the father 

 lact. (/• ex. '11. 24 a). The chromosomes in spermatogonial divisions are 

 less easy to count than in the oogonial, but most of the really clear 

 cases have given 56, and both primary and secondary spermatocytes 

 have 28. As far as the work has gone at present, therefore, there is no 

 sufficient reason for regarding the males in bisexual families directly 

 descended from unisexual families as differing in their chromosome 



