14 Chromosomes ami Sex in Abraxas 



mortality among the veiy young larvae, and it is possible that the 

 excess of females is due not to their being families in which few male 

 zygotes were formed, but that there has been differential mortality in 

 favour of the females. For the present, therefore, the question must 

 be left open, whether all-female families are pioduced by the unpaiied 

 chromosome going into the first polar nucleus in all the eggs, or whether 

 a ditferential mortality has some share in their production. Since in 

 some cases, e.g. brood '12.29, more than two-thirds of the eggs laid 

 were reared into females (in this case 111 eggs, 77 females, no males) 

 it seems very improbable that the production of all-female families is 

 due to early death of the males, and the cytological results from '13.4 B 

 and '13.34 described above suggest that there are families in which at 

 least in a high proportion of the eggs the odd chromosome goes into 

 the polar nucleus, but the proof that all-female families are due to the 

 extrusion of this chromosome in every egg caimot be regarded as 

 complete imtil a batch of eggs has been found in which all the inner 

 plates have 27 and the outer plates 28. 



That in some families the unpaired chromosome may be extruded 

 in all or nearly all the eggs is made more probable by the observations 

 of Morgan on Phylloxera^. In this genus, and apparently in other 

 Aphids, the male-producing eggs differ from the female-producing in 

 the fact that in the former one chromosome is always extruded with the 

 polar body, so that males have one chromosome less than females. It 

 seems clear, therefore, that the poles of the maturation spindle are not 

 necessarily alike, since in Phylloxera and other Aphids a particular 

 chromosome always goes to one pole. It is not very improbable, there- 

 fore, that in Abraxas the difference between bisexual and all- female 

 families consists in the amount of unlikenees between the poles of the 

 first maturation spindle. Bisexual fixmilies would arise when it is 

 equally likely that the unpaired chromosome should go to either pole ; 

 all-fcmale families when it always goes to the outer pole, and families 

 with great excess of females when it goes much more frequently to the 

 outer than to the inner. 



Another line of evidence with regard to the part played by the 

 extra chromosome in sex-determination might bo provided if gynan- 

 dromorphs should appear in families which are known to produce 

 binucleate eggs. In the only egg of this kind in which it is possible 

 to count the chromosomes (one of those from '13.42), although the 

 figures are not very good, each appears to have 27 in the outer spindle 

 1 T. H. Morgan, Jonrn. Exp. Zool. Vol. xii. 1912, p. 479. 



