SPERMATOGENESIS OF THE GALL-FLY. 4I 



In the somatic cells of both males and females one occasionally 

 finds mitotic figures containing a much larger number of chro- 

 mosomes, but such cases are in the nature of exceptions and no 

 one would contend that they represent an average condition. 

 If there is such a thing as constancy in the number of chromo- 

 somes in the majority of somatic cells, the constant is in the 

 neighborhood of 12 in both males and females of the bisexual 

 generation. Because this is the number of chromosomes found 

 in the second spermatocyte division, 12 is assumed to be the ap- 

 proximate haploid number. Now in any case where an accurate 

 count is difficult or impossible in the somatic cells, it is always 

 possible to determine with certainty that the number is very 

 much less than the expected diploid number 24. In view of the 

 fact, that in the honey-bee it is said that the somatic mitoses 

 show a very much higher number of chromosomes than occurs 

 in the gonial cells, somatic mitoses should not be used as a safe 

 and reliable method of determining the diploid number. There 

 may however.be some significance in the fact that a large number 

 of somatic cells of both males and females of Dryophanta contain 

 a number of chromosomes that approximates the number found 

 in the dividing spermatocyte rather than a multiple of this 

 number. 



Any definite statement regarding the origin and significance 

 of this condition must await examination of the maturation and 

 cleavage spindles of the egg. However, the facts at hand do 

 suggest that the males and females of the bisexual generation of 

 Dryophanta develop from eggs whose chromosomes have under- 

 gone reduction in maturation. The shghtly large number of 

 chromosomes found in the females somatic tissues may or may 

 not be of significance, but if sex determination has its basis in 

 the chromosomes, a difference in the method of distribution of the 

 chromosomes in maturation may explain why some of these 

 eggs develop parthenogenetically into females and others into 

 males. 



In a recent paper Nachsheim has summed up in a general 

 statement the results of investigations dealing with sex-de- 

 termination in Hymenoptera as follows: "Die Mannchen der 

 Hymenopteren entstehen aus unbefruchtete Eiern, die zwei 



