L. DONCASTER 15 



and 28 in the inner, which would indicate that both egg nuclei would 

 be male-determining. 



To obtain the complete answer to the questions raised in the 

 preceding paragraphs will require the examination of many more eggs, 

 and it is not unlikely that fresh material will be needed. Meanwhile, 

 one point of great importance appears from the observations made up 

 to the present time — they show that in Abi-axas, and therefore probably 

 in other Lepidoptera, the presence of two heterochromosomes is charac- 

 teristic of the male, and one, of the female. This follows of necessity 

 from the fact that some eggs have 27, others 28, and that females of 

 the strain used have 55 and males 56. In strains with 56 in the 

 oogonia and 28 in all the eggs, it must be concluded that the additional 

 chromosome is functionless as regards sex, as in the case of other insects 

 where there is a large and a small heterochromosome in the male. The 

 Lepidoptera thus show a condition which is the exact converse of that 

 found in all cases hitherto described. The connexion of this fact with 

 sex-limited inheritance will be referred to again in the next section. 



e. The Exceptional Faiiiilij 12.25. 



In my previous communication on the chromosomes of the unisexual 

 families (Journ. Genet. III. p. 9), I mentioned that one female of family 

 '12.25 showed an oogonial figure with 56 chromosomes, while others of 

 the same brood had clearly 55. At that time I had no other clearly 

 countable figures in females which were daughters of unisexual broods, 

 but which themselves belonged to bisexual broods, and I therefore 

 concluded that some females of such broods had 55, others 56. Sub- 

 sequent work, however, as described above, has shown that with this 

 one exception, all females of the strain with 55 chromosomes, whether 

 they belong to unisexual or to bisexual families, have 55. The single 

 female of brood '12.25 is thus alone in apparently having 56 (cf Figs. 17, 

 18). Very careful examination, with both 2 mm. and 3 mm. Zeiss 

 apochromatic lenses, shows that in the single good figure two of the 

 chromosomes counted separately are in contact, and might conceivably 

 be dividing halves of one (Fig. 18). This, however, is made improbable 

 by the fact that one of the halves is itself somewhat divided. Many of 

 the chromosomes show some extent of doubleness, in preparation for 

 division, but none of them are so conspicuously double as the chro- 

 mosome in question, if it is really one in division, and the distinct 

 doubleness of one of the halves is also inexplicable if the pair are really 

 halves of one chromosome. The evidence therefore seems to me to be 



