132 JAMES O. FOLEY. 



around the nuclear membrane (Figs. 49 and 50), leaving the 

 center of the nucleus clear with only a few linin fibers connecting 

 the masses at the periphery. Immediately following this pe- 

 ripheral arrangement of the chromatin in the form of a dense 

 anastomosed ring (Figs. 50 and 51) beneath the nuclear mem- 

 brane and around the clear center to the nucleus, a contraction 

 of the nucleus takes place. Figs. 52 and 53 show final stages in 

 the arrangement of the nuclear material to form the head of the 

 spermatozoon. The heads of the spermatozoa formed by this 

 contraction are somewhat smaller than the original spermatid 

 nucleus. The heads of the mature spermatozoa are practically 

 spherical (Fig. 55) although some may show such shapes as the 

 one pictured in Fig. 54. 



DISCUSSION. 



The literature on the spermatogenesis of the fishes or fishlike 

 vertebrates is very limited. There is apparently no detailed 

 piece of work relating to the Teleosts. 



The prevalent opinion on the origin of the germ-cells in 

 spermatogenesis is that they are formed either from cells which 

 migrate into the testes, assume a fixed position, and produce 

 the ultimate crop of spermatozoa, or that they are derived from 

 reserve germ-cells which have existed as such since the formation 

 of the first or primordial germ-cells in early embryogenesis. 

 Moore ('95), working on the spermatogenesis of several genera 

 and species of Elasmobranchs, Rawitz ('99), working on Scyllium 

 canicula, Cunningham ('86), ('91), in two papers on the spermato- 

 genesis of the Cyclostome (Myxine glutinosa), presuppose the 

 origin of the primary spermatogonia from a germinal epithelium. 

 Upon the spermatogenesis of the fishes proper, there are the 

 papers of Agar ('i i), ('12), on the spermatogenesis of Lepidosiren 

 paradoxa; Turner ('19), on the seasonal cycle in the spermary of 

 the perch; Geiser ('24), on the spermatogenesis of Gambusia 

 //0//>r0oH ; and Winge ('22), on the cytology of Lebistes reticulatus. 

 \Yinge does not discuss the spermatogonia, Agar assumes a 

 development from a definite germinal epithelium, while Turner 

 and Geiser assume an origin from migrating cells. Turner ('19) 

 shows that in the perch the cysts are formed anew each season 



