328 THEORIES OF INHERITANCE AND DEVELOPMENT 



The first question raises once more the old puzzle of preformation 

 or epigenesis. The pangen hypothesis of De Vries and Weismann 

 recognizes the fact that development is epigenetic in its external 

 features ; but like Darwin's hypothesis of pangenesis, it is at bottom 

 a theory of preformation, and Weismann expresses the conviction 

 that an epigenetic development is an impossibility. ^ He thus ex- 

 plicitly adopts the view, long since suggested by Huxley, that "the 

 process which in its superficial aspect is epigenesis appears in es- 

 sence to be evolution in the modified sense adopted in Bonnet's later 

 writings ; and development is merely the expansion of a potential 

 organism or 'original preformation' according to fixed laws."^ Hert- 

 wig ('92, 2), while accepting the pangen hypothesis, endeavours to 

 take a middle ground between preformation and epigenesis, by 

 assuming that the pangens (idioblasts) represent only cell-characters, 

 the traits of the multicellular body arising epigenetically by permu- 

 tations and combinations of these characters. This conception cer- 

 tainly tends to simplify our ideas of development in its outward 

 features, but it does not explain why cells of different characters 

 should be combined in a definite manner, and hence does not reach 

 the ultimate problem of inheritance. 



What lies beyond our reach at present, as Driesch has very ably 

 urged, is to explain the orderly rhythm of development — the co- 

 ordinating power that guides development to its predestined end. 

 We are logically compelled to refer this power to the inherent 

 organization of the germ, but we neither know nor can we even 

 conceive what this organization is. The theory of Roux and Weis- 

 mann demands for the orderly distribution of the elements of the 

 germ-plasm a prearranged system of forces of absolutely incon- 

 ceivable complexity, Hertwig's and De Vries's theory, though ap- 

 parently simpler, makes no less a demand ; for how are we to 

 conceive the power which guides the countless hosts of migrating 

 pangens throughout all the long and complex events of development .■* 

 The same difficulty confronts us under any theory we can frame. If 

 with Herbert Spencer we assume the germ-plasm to be an aggrega- 

 tion of like units, molecular or supra-molecular, endowed with prede- 

 termined polarities which lead to their grouping in specific forms, 

 we but throw the problem one stage further back, and, as Weismann 

 himself has pointed out,^ substitute for one difficulty another of 

 exactly the same kind. 



The truth is that an explanation of development is at present 

 beyond our reach. The controversy between preformation and 



1 Gerin-plas//i , p. 14. 



2 Evohition, Science and Culture^ p. 296. 



^ Gerniinal Selection, January, 1S96, p. 284. 



