47 



may, however, not be the exclusive products of the peritoneal 

 epithelium, but the products of a variety of soma cells ; or not 

 being products of soma cells at all, they must have had their 

 origin elsewhere, and have wandered into the various positions 

 they finally come to occupy. 



This latter Beard regards as being the true explanation. 

 If this be so, it must follow that the germ-cells have their 

 origin outside the embryo, and that at some period they 

 migrate into it. 



What then, are the observed facts leading to such a 

 conclusion ? 



Now, the germ-cells of the skate are easy of identification, 

 for they possess very distinctive features. They are glassy, 

 contain a number of yolk-plates, their nuclei are of a bilobed 

 type, and they are considerably larger than the largest somatic 

 cells. These peculiarities are shared by the other elasmo- 

 branch fish embryos in some degree. These distinctive 

 characters enable them to be cotmted, and Beard has found 

 that there is a constant number of primary germ-cells for 

 each species that he has up to the present studied. 



In the skate, the number found in all females approxi- 

 mates to 512. 



If the primary germ-cells are the products of the germinal 

 epithelium and they migrate thence, it follows that the 

 younger the embryo is the greater will be the number of 

 germ-cells normally placed. But what are the facts ? Beard 

 has not only counted the germ-cells in a number of embryos, 

 but he has also tabulated the positions they occupy. Briefly, 

 the results are as follow :— In embryos about 30 m.m. in 

 length nearly 30% of the germ-cells are abnormally placed, in 

 those about k5 m.m. long nearly 40%, those only 20 m.m. long 

 have quite 50% of their germ-cells misplaced, while very young 

 embryos only 2-5 m.m. long are entirely without germ -cells. 

 Beard describes one embryo 2-66 m.m. long very exhaustively. 

 It is, however, but a representative example. It is shown 

 diagrammatically in figs. A and B, 



