424 ROBERT W. HEGNER 



The discovery of these distinct spindle remains in the spermat- 

 ogonia! divisions enables us to homologize one more period in 

 the cycle of the male germ cells with a corresponding period in 

 the cycle of the female germ cells. 



Thus the ultimate spermatogonium passes through a certain 

 number of divisions — probably five or six — which correspond 

 to the differential divisions so clearly exhibited by the ultimate 

 oogonia of Dytiscus. Just as in the maturation processes, how- 

 ever, where only one female cell but all of the male cells are 

 functional, so these earlier divisions result in the female in the 

 production of a single oocyte and a number of nurse cells which 

 may be considered abortive eggs, whereas in the male every 

 daughter celL is functional. The limited period of division in 

 the cycle of the male germ cells in man (Montgomery, '11; von 

 Winiwarter, '12) is also similar to those in Dytiscus and Lep- 

 tinotarsa. 



(4) Amitosis in the germ cells of Leptinotai'sa. We have al- 

 ready described what appears to be amitotic nuclear division 

 in the vitellophags of Chrysomehd eggs, and shall now examine 

 certain stages in the germ cell cycle where amitosis has been 

 reported. 



Wilson ('00) defines amitosis as "mass-division of the nuclear 

 substance without the formation of chromosomes and amphi- 

 aster" (p. 437), and concludes from a review of the literature 

 up to the year 1900, "that in the vast majority of cases amitosis 

 is a secondary process which does not fall in the generative series 

 of cell-divisions" (p. 119). During the past ten years interest 

 in direct nuclear division has been maintained, principally 

 because of the claims of certain investigators that germ cells may 

 multiply in this way and still give rise to functional eggs or 

 spermatozoa. 



During amitosis the chromatin remains scattered within the 

 nucleus and does not form a spireme nor chromosomes, and 

 therefore its individual elements, the chromatin granules, do 

 not divide. As a result of this ?7zass-division there can be no 

 accurate segregation of chromatin granules in the daughter 

 nuclei as is demanded by the theory that the nucleus, and partic- 



