469 



complete and possessed of all the characters of a young skate, thus, 

 in an embryo (no. 245) of 54 mm (fig. 33) they are quite gone. 



The protoplasm of the early germ-cells, large and small, no 

 matter where they be, is very peculiar. It is a colourless, glassy 

 substance, possessing no affinity for any of the ordinary stains. Osmic 

 acid tinges it of a slightly brownish hue. Barely in early stages are 

 the cells rounded, unless they be in a position of rest, among a mass 

 of other cells, or encapsuled in some pecuHar position. More usually 

 they are irregular in form, and show processes, which can only be 

 defined as amoeboid. Though their actual movements have, of course, 

 never been seen, they doubtless possess considerable powers of motion 

 in an amoeboid fashion. 



Indeed, the similarity of many of these cells to amoebae is very 

 striking. Even in the immediate neighbourhood of the germinal nidus 

 one finds them in evident movement, while in many, many instances 

 they are encountered between splanchnopleure and gut in the early 

 embryos (fig. 28), and in the later ones within the connective tissue 

 of the mesentery. 



In fig. 34 one is even depicted in the act of making its way 

 through the splanchnopleure into the body-cavity. And the undoubted 

 germ-cell of fig. 22 was found in three consecutive sections immedi- 

 ately under the epiblast and just beneath the segmental duct in an 

 embryo of 10,5 mm. The nuclei are not always easily found, and 

 this is largely due to the yolk. Sometimes they are large, clear bodies 

 with a weU-defined nucleolus. Nuclei of this kind are most commonly 

 found in the germ-cells of older embryos. In early embryos they are, 

 where well seen, very frequently — if not always — of a bilobed or 

 twin nature — a remarkable character of primary germ-cells, hitherto 

 only recorded in arthropods by Rückert ('95) and Hacker ('96), but 

 now shown to characterise the very early germ-cells of vertebrates. 



We now approach the consideration of the problem, as to how 

 the germ-cells get into the embryo, and, as to.the majority, how they 

 reach the so-called "germinal ridge", or, as here suggested, the 

 germinal nidus. 



In taking up this, it also becomes necessary to consider the large 

 yolk-laden cells, termed by Rückert '"megaspheres" ('87, p. 98). 



In the first place, it may be well to point out, that the present 

 writing concerns itself with these cells in the skate, a form, in which 

 they have not been described by any previous author. 



Rückert's detailed descriptions apparently refer exclusively to 

 Torpedo. Although there is no doubt in my mind as to the general 



