Experimental Studies on Germinal Localization. 265 



the inequality is caused by, or at least correlated with, a preceding 

 segregation of different materials in the cell before division. 

 Hence it is an interesting fact that all the typically unequal divi- 

 sions of the normal development show a tendency to become less 

 unequal upon isolation of the cell. This has been observed in 

 the first division of the isolated M> of the 3^-micromeres, of the 

 i' cells, of the 5^-micromeres, the yV-macromeres, and in the 

 CD ^-blastomere of Dentaliiim (where It is expressed by a re- 

 duction in the size of the polar lobe), all of which are qualitative 

 divisions. This may be explicable as a result of relatively simple 

 physical conditions, but it is at the same time not improbably an 

 expression of a tendency for the segregation to recede, as it were, 

 towards a less definitely localized condition. The possibility is 

 thus suggested that the segregative process in the cells when in 

 their normal position in the whole embryo may, even in relatively 

 late stages, be in some measure influenced by their relation to their 

 fellows or to the whole. I believe that important light may be 

 thrown on this question by an accurate comparison of the later 

 development of isolated blastomeres that vary in this respect. 

 The most important question remaining is whether after com- 

 plete segregation and isolation of specific cytoplasmic stuffs has 

 been once effected by qualitative division the missing materials 

 may be restored by regulative metabolic processes. Such remark- 

 able facts as those determined in regard to the regeneration of 

 the lens in Triton, or Spemann's hardly less striking results on the 

 formation of double-headed monsters in the same animal, leave 

 no doubt that specific cell-characters may, within the limits of the 

 germ-layers, be very widely altered through a response to a local 

 defect, or to a change as simple as a mere mechanical alteration 

 of form in the growing mass ; and the facts of regeneration even 

 seem to show that one of the differentiated primary germ-layers 

 may produce structures which in the typical development arise 

 from a different layer. If the hypothesis of formative cytoplas- 

 mic stuffs be valid there seems to be no escape from the conclu- 

 sion that in such cases the necessary formative stuffs may be form- 

 ed anew. But if the potentiality of the cytoplasmic system be 

 primarily given in the nuclear organization, and if this be the 



