276 A THEORY OF HEREDITY VIII 



of heredity, introducing the subject by the following 

 admirable remarks, which appear to me to assign in 

 the most judicious manner their true value to such 

 hypotheses and to be as strictly applicable to later 

 speculations as to his own. " A positive explanation 

 of heredity is not to be expected in the present state 

 of bioloo-y. We can look for nothing beyond a simpli- 

 fication of the problem, and a reduction of it to the 

 same category with certain other problems wdiich also 

 admit of hypothetical solution only. If an hypothesis 

 which certain other widespread phenomena have 

 already thrust upon us, can be shown to render the 

 phenomena of heredity more intelligible than they at 

 present seem, we shall have reason to entertain it. 

 The applicability of any method of interpretation to 

 two different but allied classes of facts is evidence of 

 its truth. The power which organisms display of 

 reproducing lost parts, we saw to be inexplicable 

 except on the assumption that the units of which any 

 oro-anism is built have an innate tendency to arrange 

 themselves into the shape of that organism. We 

 inferred that these units must be the possessors of 

 special polarities, resulting from their special struc- 

 tures ; and that by the mutual play of their polarities 

 they are compelled to take the form of the species to 

 which they belong. And the instance of the Begonia 

 phyllomaniaca left us no escape from the admission 

 that the ability thus to arrange themselves is latent 

 in the units in every undifferentiated cell. . . . The 

 assumption to which we seem driven by the ensemble 

 of the evidence, is that sperm -cells and germ -cells 



