174 c - M - CHILD - 



be a complex. All that has been said above regarding com- 

 plexes of substances applies here. The result produced is the 

 result of the sum total of conditions existing in the complex. 

 When we assume that the result is due to the presence of certain 

 specific substances in the complex which manifest themselves in a 

 peculiar " formative " fashion we are not only making an inference 

 not warranted by the facts, but we are involving ourselves in 

 various difficulties which become manifest only on careful analysis. 

 Some of these I have endeavored to point out above. All the 

 substances in the complex are or may be formative, and under 

 certain conditions the region gives rise to a characteristic differ- 

 entiation. But if all the substances are formative the distinction 

 of formative substances is entirely unnecessary and we must regard 

 the region merely as a formative complex. 



But it may perhaps be maintained that the formative substance 

 hypotheses do not differ essentially from this since the formative 

 substances are really complexes, not definite chemical substances. 

 If this is the case then it is certainly preferable to avoid the use 

 of the vague term substance in this connection, especially since 

 " formative " factors are known in many cases to be related to 

 substances only very indirectly. Moreover, it is difficult to under- 

 stand the grounds for distinguishing a particular complex either 

 of substances or of conditions as formative since this implies that 

 others are not formative. The only conclusion justified with 

 regard to " self-differentiation " in isolated blastomeres or egg- 

 regions is that the conditions for the observed differentiation reside 

 in the part. This, however, is really quite different from the con- 

 clusion that the piece contains certain formative substances for the 

 particular structures to which it gives rise. The region may dif- 

 fer chemically in greater or less degree from other regions of the 

 egg and this difference may be more or less closely related to the 

 structural result and to that extent formative. But to dignify 

 the chemical peculiarity of a region by the term formative sub- 

 stance involves unwarranted assumptions. The strongest evi- 

 dence in support of formative substance hypotheses is found in 

 these cases of ontogenic self-differentiation of parts and this, as 

 has been seen, is far from conclusive. 



In order to avoid misconception it should perhaps be stated 



