Experimental Studies on Germinal Localization. 263 



tified that the power of an Isolated blastomere to produce a com- 

 plete embryo depends upon three conditions : first, upon its vol- 

 ume; second, upon the presence of all the essential elements (and 

 apparently of the cytoplasmic elements) of the system, and third, 

 upon the effectiveness of the regulative process. The production 

 of a complete embryo involves the regrouping of these elements in 

 a disposition essentially like that of an entire embryo, and I sec 

 no escape from Driesch's contention that this is a typical act of 

 regulation that cannot be explained without recourse to a factor 

 that lies behind the primary topographical grouping of cytoplas- 

 mic stuffs/ My observations on Amphioxus, the accuracy of 

 which I see no reason to doubt, seem to show that this regrouping 

 may be effected immediately upon the isolation of the cell, as 

 would also seem to be the case in the inverted single blastomeres 

 of the frog's egg observed by Morgan (though observations on 

 the cleavage in this case are still lacking). In the greater number 

 of cases thus far observed the cleavage-factors, and hence as I 

 think we now may say, probably also the morphogenic factors, 

 do not undergo immediate readjustment; and it is still quite an 

 open question to what extent the cells formed in the ensuing par- 

 tial cleavage undergo changes of prospective potence. But even 

 though all the essential elements of the system be present, in a 

 mass of sufficient volume, a failure of regulation may occur, per- 

 haps owing to merely physical obstacles. As pointed out in my 

 preceding paper, this is not improbably the explanation of the 

 failure of the CD-half in Dentalium to produce a perfect larva. 

 The ctenophore-egg is of exceptional interest in this direction; for 

 there is good reason to conclude (since both the cleavage and the 

 larva are bi-radially symmetrical) that the vertical cleavages — 

 i. e.^ the first and second, and perhaps also the third — are not 

 qualitative, yet, notwithstanding the closure of the embryo pro- 

 duced by the >4 or M-blastomere, the larva remains defective. 

 Driesch's explanation of a failure of the regulative process owing 

 to a "rigidity" of organization or of protoplasmic texture seems 

 in this case perfectly valid; but such explanation must be consid- 



^Cf. Lillie, '01, p. 269; Driesch, '02.1, '02.2; Wilson, '03,. p. 456. 



