368 APPENDIX B 



(i. e. an albino), or has an extra digit, or lacks a phalanx, then 

 the inheritance tends to be alternative. We have constantly 

 spoken as if alternative inheritance were a rare phenomenon. In 

 the sense meant the statement was true, and has deceived no one. 

 But in another sense it is not true. In all species, the members 

 of which are divided into males and females, all the larger 

 differences between the mating individuals are subject to alter- 

 native inheritance. The male sexual characters are latent in the 

 females, the female characters are latent in the males. It has 

 been surmised that the inheritance of the sexual characters is 

 Mendelian. It is very much more probable the inheritance of 

 Mendelian characters is sexual. 



Nature, especially in the case of the more complex plants and 

 animals, has devised all sorts of methods to secure cross-fertiliza- 

 tion, the most effectual being complete sexual differentiation. 

 All very complex types, therefore, present two methods of 

 inheritance, the blended and the alternative. Smaller parental 

 differences are subject to the former, larger differences to the 

 latter. In blended inheritance there is of necessity no latency, 

 even though one individual be more or less completely prepotent 

 over the other. In alternative inheritance there is constant 

 latency, but no true prepotency, though there may be something 

 which is often confused with it but is radically different, namely, 

 dominance 1 ; for dominance involves latency of the alternative 

 character, which prepotency does not. This dual method of 

 inheritance provides the machinery whereby a maximum amount 

 of regression is secured. 



Evolution, no matter how stringent the selection from which it 

 has resulted and by which it is maintained, is never perfect. 

 Often as regards the sexual characters, primary and secondary, 

 but more especially as regards the latter, we see, not complete 

 dominance and latency as in perfect alternative inheritance, but 

 some degree of blending. This occurs in its most pronounced 

 form in hermaphrodites. More common is a slighter degree of 

 blending, as when women display rudimentary moustaches, or 

 men unusually large mammae. Distinct from this blending, 

 though perhaps generally combined or occurring with it, is 

 the transference of a male or female character from the set in 

 which it is normally found to the opposite set as when women 

 have long thick beards or men fully-developed female breasts. 

 Nature has evolved alternative inheritance to create and per- 

 petuate sexual differentiation, but she is not a consciously 

 discriminating agency, and, just as the blending of sexual 

 characters sometimes occurs, so, on the other hand, the inherit- 

 ance of non-sexual characters is sometimes alternative. As we 

 have seen, whenever the latter happens the non-sexual differences 

 are, like the sexual differences, usually considerable. Nature 

 1 See Reports to Evolution Committee, I, p. 137. 



