CLE A FACE AND DIFFERENTIATION 



The gastrula may develop from a morula (a solid mass of 

 cells) or a blastula (a mass of cells that contains a cavity); 

 the gastrulation may be that of invagination, delamination 

 or epiboly; finally the gastrula is a two-layered radially 

 symmetrical structure enclosing a cavity whose polar axis, 

 now the embryonic, for the most part traverses empty space, 

 and at whose poles only are cells located which are part of 

 the two-layered covering. Components of an axial gra- 

 dient, if there be such, can become effective only in the 

 cell-layer.^ 



Eggs of most bilaterally symmetrical animals begin their 

 development as radially symmetrical structures and there- 

 fore show a polar axis. But at the moment after fertiliza- 

 tion when bilaterality appears in such an egg, we can no 

 longer speak of an axis. In a bilaterally symmetrical 

 organism — egg or adult — there exists no line common to 

 planes as in a radially symmetrical one. Here, accurately 

 speaking, we can use only the term, plane of symmetry. 



Certain eggs out of which develop bilaterally symmetrical 

 animals — as, for example, those of Loligo, Hydrophilus, 

 Amia — reveal bilateral organization before fertilization. 

 In these, obviously, we can not speak of a polar axis. For 

 even if we know that at some time in their history the eggs 

 were radially symmetrical, the moment that bilaterality 

 appears in them there can be no axis with respect to which 

 the parts are symmetrically arranged. 



But even if we allow the incorrect use of the term, axis, 

 in bilateral embryos, we find no constant relation between 

 this axis and that of the egg. A survey of embryogenesis 



^ /. JV. Wilson and I were never able to repeat Child's observations 

 on the effect of KCN on eggs of Arbacia which he interpreted as 

 demonstrating an axial gradient. My own experiments {lQ28c) 

 on this egg, which gave results similar to Child's, strongly indicate 

 that these are due to changes in the ectoplasm. Child's ozvn work, 

 as that on protozoan forms, can be explained in terms of effects on 

 the superficial cytoplasm. 



301 



