STUDIES ON GERM CELLS 473 



in vertebrates back into cleavage stages — something that has 

 not been accompHshed as yet. 



It seems too early to speculate as to the influence of the mito- 

 chondria upon the primordial germ cells. That they are bearers 

 of hereditary qualities and that those brought into the egg by the 

 spermatozoon fuse with those of the egg as described by Meves 

 ('08, '11) is doubted by many observers. Montgomery ('12) has 

 shown that in Peripatus the mitochondria are entirely cast out 

 of the spermatozoon during its metamorphosis, and that, at 

 least in this species, the male cell does not contribute any of these 

 bodies to the egg during fertilization. The same writer (Mont- 

 gomery, '11) proposes an hypothesis to account for the segre- 

 gation of germ cells as follows: "Any cleavage cell which failed 

 to receive mitochondria, or failed to receive particular ones or a 

 particular amount of them, would be incapacitated from engender- 

 ing such somatic specializations (fibrillar structures), it would 

 thereby become a germ cell" (p. 791). This substitution hypo- 

 thesis is also offered by Montgomery, that the mitochondria of 

 the prospective germ cells remain unaltered or latent, while those 

 of the other cells undergo developmental changes. As we have 

 already noted, evidence of such a condition had already been 

 supplied and a similar hypothesis proposed by Rubaschkin ('10) 

 and Tschaschin ('10) ; but so far as I know there are no data which 

 enable us to sustain the first hypothesis. The data given in a 

 previous part of this paper show that the pole-disc in Chrysome- 

 lid eggs may arise from mitochondria, but this does not seem very 

 probable. 



An examination of the various Keimbahn-determinants listed 

 in table 1 (p. 461) has led the writer to conclude that none of 

 them is of a mitochondrial nature, but the results obtained by 

 the special methods employed by students of mitochondria give 

 us good reason to hope that other substances may be made 

 visible which will help to clear up the problem of primary cellular 

 differentiation. 



Metabolic products. Among the most difficult cases to explain 

 are those of Sagitta and certain copepods, since here the Keim- 

 bahn-determinants apparently arise de novo in the cytoplasm. 



