The Germ-cells of Lophius. 589 



mental, corresponding to the segmentation of the paraxial struc- 

 tures of the embryo. This question is the more pertinent, because 

 of the view, held by Van Wijhe ('89) and some others, that the 

 germ-cells in all vertebrates arise from the gonotome, a part of 

 the segmented mesoblast. The observations on Lophius do not 

 give any support to this view, though the presence of germ-cells 

 in the myotome at an early stage might be so interpreted. The 

 myotome, however, is not their place of origin in Lophius; they 

 are observed before the myotome is formed, in fact before the 

 mesoblast itself is definitely separated, and, as we have seen, 

 there is reason to believe that their actual origin is still earlier. 

 They cannot be said to be mesoblastic in origin and certainly 

 they are not segmental. 



In no later stage, moreover, does their grouping appear to bear 

 any relation to the segmentation of the embryo. At first I 

 thought possibly there might be such a relation, but study of 

 several embryos of different ages showed that the groups of germ- 

 cells in no way correspond to the metamerism of the embryonic 

 trunk. 



d. Criteria of gerrn-cells. During this study, the question has 

 frequently suggested itself, how do I know that through all stages 

 I am dealing with the same group of cells? What is the evidence 

 that the cells here described do not represent merely a cyclical 

 stage of ordinary cells? There are three kinds of evidence which 

 support the view that through the different stages, we have to 

 do with the same group of cells. These are as follows : First, con- 

 stancy of size and appearance. The stability of size and struc- 

 ture of these cells, through a period marked by great change in 

 all other cells of the embryo has already been sufficiently empha- 

 sized. Second, constancy of number. In stages from 8 and up- 

 wards, the number, though varying somewhat in different em- 

 bryos, is constant enough to give strong support to the view that 

 we are dealing with a definite group of cells. During this period, 

 the size, structure and number of these cells is by far the most 

 constant and unchanging feature of the embryo. Third, cor- 

 respondence of position. In embryos of the same age, these cells 

 are always found, with minor variations, in the same part of the 



