THE SPERMATOGENES1S OF AGALENA N.KVIA. 139 



this question was sought in the unequal distribution of the ac- 

 cessory chromosomes. They were found to take no part in the 

 first maturation division and, while their further history was ex- 

 tremely difficult to follow, I gathered, as I then believed, some 

 evidence of their taking no part in the second maturation division, 

 and therefore of their final distribution to but one fourth of the 

 spermatozoa. McClung's theory that the accessory chromo- 

 somes might be sex-determinants was then held to be untenable, 

 so far as its application to the spider was concerned, as it did not 

 seem probable that one sex would be three times as numerous as 

 the other. I then ventured to suggest a new theory viz.: that 

 only the one fourth of the spermatozoa which contain the acces- 

 sory chromosomes the "favored" spermatozoa become func- 

 tional while the remaining three fourths degenerate after almost 

 or altogether reaching maturity. There were a priori reasons for 

 believing such to be the case since, if true, the parallelism be- 

 tween the spermatogenesis and the oogenesis would be even more 

 complete than hitherto supposed, three of every group of four 

 daughter cells descended from a single spermatogonial cell being 

 considered as homologues of the polar bodies which do not be- 

 come functional. I also suggested, in view of the foreseen diffi- 

 culties in the union of the sex-cells, that in the maturation of the 

 egg the accessory chromosomes might be thrown off in the polar 

 bodies and thus, at time of fertilization of the egg the normal 

 chromosomal number would be restored. This second suggestion 

 was overlooked by Boring ('07) in her criticism that if only the 

 "favored" spermatozoa become functional, the egg must neces- 

 sarily contain the accessory chromosomes also and that in the 

 nucleus of the fertilized egg the chromosomal number would 

 exceed the normal number by two. Berry ('06) in her paper on 

 Epeira states that certainly in none of her preparations does she 

 find any trace of degenerating spermatozoa, and other writers 

 have expressed doubt of their existence. Now, while I still find 

 an abundance of degenerating cells in the lumen of the testis and 

 in the sperm ducts, my recent investigations have convinced me 

 of the error of my former results in regard to the distribution of 

 the accessory chromosomes. In the present paper I think it is 

 clearly demonstrated beyond the shadow of a doubt that dimor- 

 phism of spermatozoa is the rule. 



