ANIMAL EVOLUTION 25 



Isolated A, B, C blastomeres produce defective larvae 

 without apical organ or lower hemisphere. Isolated 

 D larvae are normal, except that the lower hemisphere 

 and apical organ are disproportionately large. 



The micromeres la, ib, ic, id may be isolated, and 

 they continue to segment, la, ib, ic form curious little 

 larvae with a prototroch and an upper hemisphere ; but 

 no lower hemisphere, no gut, and no apical organ, id pro- 

 duces a similar larva as regards absence of trunk and 

 gut, but it has a well-developed apical organ. It is 

 possible to cut eggs of Dentalium into two, to fertilize 

 the pieces, and to develop larvae from these fertilized 

 fragments. The results are absolutely confirmatory of 

 those I have just described. The lower hemisphere 

 segments as a whole, and produces a dwarfed larva, but 

 with all its parts complete. On the other hand, the upper 

 hemisphere of the egg segments as a whole, but no polar 

 lobe is formed, and the resulting larva has neither apical 

 organ or post trochal lower hemisphere. 



Similar and even more thorough and convincing 

 experiments have been made on the developing eggs of 

 the Limpet. But the details are too complicated for me 

 to do more than refer to them. They prove, in an 

 unequivocal manner, that the various cells formed during 

 the cell-lineage, have each a definite and limited destiny, 

 and this they will fulfil, even if removed from the aggregate 

 of which they form a part. For example, early in the 

 segmentation certain cells are formed which are destined 

 to give rise to the prototroch. One cell arises in each 

 of the four quadrants formed by the four large cells of 

 the four-cell stage of development, and the ceU of each 

 quadrant behaves in the same way. It divides into two, 

 and the two divide again into four which combine with 

 the four similar cells derived from each of the other 

 three quadrants to form a ring of sixteen ciliated cells 

 constituting the prototroch. This is the normal course of 

 development. Let us call any one of the primary proto- 



D 



