150 ' C. M. CHILD 



absence of other cells anterior to them ; whether a head shall form 

 or not depends on whether the reaction in these cells proceeds 

 far enough so that it becomes the controlling factor in the piece. 

 If it becomes dominant to a sufficient degree a new head forms and 

 the new head region determines the reorganization of other parts 

 from the anterior end backward : if it does not become dominant 

 a head does not arise. In the formation of a new head then the 

 relation between two factors which are in a sense opposed to each 

 other is involved. This relation differs in different regions of 

 the body in consequence of the preexisting axial gradient and the 

 stimulating effect of isolation, moreover, it can be altered in vari- 

 ous ways by the use of reagents and other external factors. Under 

 certain conditions or in certain regions of the body depressing 

 agents like KCN act as apparent morphogenetic stimuli simpl}" 

 because they alter the relations between the two opposed reaction- 

 complexes of the piece in favor of the reaction-complex which 

 leads to head-formation. In certain other regions or under other 

 conditions they alter the relations in the opposite direction and 

 so decrease the frequency of head-formation and consequently 

 the capacity of the piece for regulation. 



If these two opposed factors exist, as they undoubtedly do, 

 then the regulatory capacity of a piece is not definitely and finally 

 determined by its position in the original body, its relation to the 

 axial gradient or its 'organization,' but by the relation of the two 

 opposed factors to each other and it has been shown above that 

 this relation can be altered experimentally. 



It is true, however, that under constant conditions the relation 

 of these two factors to each other is determined within certain 

 limits of variation by the position of the piece within the body, 

 but 'position' in this connection means essentially its relation to 

 the dominant region, i.e., its position in the axial gradient, rather 

 than a certain 'organization.' Since this is the case the regula- 

 tory capacity of pieces taken in sequence along the axis of the 

 animal shows a gradient which is related to the axial gradient in 

 the intact animal but is not necessarily a direct and simple expres- 

 sion of it. The fact that the regulatory capacity of pieces is not 

 fixedly determined, but can be altered experimentally in either 



