Child, Driesch's harmonic eqnipotential systems in form-regulation. (319 



physikalische Speziiltätskombination (Driesch, 1901, p. 187) may 

 exist whose parts are prospectively alike. Such a machine may exist 

 in a isolated part and constitute the basis for future localization. 

 In other words the conditions under which morphogenesis in har- 

 monic eqnipotential systems becomes an „autonomistic" problem 

 cannot be realized in nature. Driesch's „proof" applies, therefore, 

 only to a postulated not a real case. 



It is evident from Driesch's argument quoted above that he 

 makes no distinction between a prospective potential „machine" 

 and a real machine. 



The reason for his failure to make such a distinction appa- 

 rently lies in his belief that external conditions including the rela- 

 tions with other parts play no part in the development of harmonic 

 equipotential systems. I have shown that external conditions cannot 

 be excluded in any case, but even if we admit for the sake of 

 argument that they may be excluded, it is not necessary to accept 

 Driesch's conclusion that each element of such a system according 

 to the mechanistic hypothesis must represent at the same time 

 every part of every machine which it is capable of forming in the 

 future. Conceivably at least localized differences, i. e., parts of a 

 real machine, visible or invisible may exist which are capable in 

 consequence of their relations to each other of bringing about new 

 localizations, i. e., of forming a new machine. But this new ma- 

 chine does not exist as such until it is formed. Up to that time 

 it exists merely as the „properties" and relations of the parts of 

 the old machine and these do not posses the typical space con- 

 figuration which the machine itself possesses when it is formed. 



But as a matter of fact there is no case in regulatory deve- 

 lopment in which these factors alone are involved. As I pointed 

 out above (p. 581) the possible mechanistic factors involved in the 

 formation of a whole from a part are in their lowest terms: first, 

 the constitution of the part and the regional differences which exist 

 in it in consequence of its previous differentiation as a part of an 

 organism; second, the internal changes which result from its iso- 

 lation, which are undoubtedly localized; third, the changes in relation 

 to the extra-organic environment which result form the formation 

 of new surfaces of contact with the medium, new terminal regions etc., 

 and these are also localized to a greater or less extent. Similar 

 possible factors, though differing in specific character are involved 

 in those cases where a whole is formed after dislocation of the 

 parts. And it is impossible to exclude any of these factors in any 

 experiment. 



In short Driesch's assertion that the mechanistic hypothesis 

 must assume that all parts of all potential machines are present 

 as such, i. e., in typical space configuration at the beginning is not 



