50 



THE AMERICAN NATURALIST 



[Vol. LII 



is not difficult to understand. If the resemblance of 

 model and mimic is based on the presence of similar 

 chance- or non-chance combinations of genetic factors, 

 and if geographic variation consists in the specific adapta- 

 tion of the quantity of certain genes to a required veloc- 

 ity of some vital reaction, it is very natural that similar 

 genes in model and mimic should be in exactly the same 

 situation and should undergo parallel changes. 



The third important set of facts to be considered is the 

 problem of domestication. Darwin's view is well known, 

 as well as the solution of a great part of the problem 

 through Mendelism. The latter shows that selection of 

 the recombinations after cross-breeding (besides picking 

 of mutations) is the chief source of success in domestica- 

 tion. (See our demonstration of this fact regarding the 

 improvement of pigs in "Einfiihrung in die Vererbungs- 

 wissensehaft," 2d ed., 1913, pp. 276^80.) That this fact 

 was well known to Darwin is shown, for example, in his 

 report about Lord Orford's greyhounds ("Variation of 

 Animals," etc., Ch. 1). But he believed, in addition, in a 

 positive effect of selection of small variations. Wher- 

 ever he tabulates such characters, most or all of them are 

 quantitative characters of a kind which we can assume 

 to be dependent upon the presence of definite quantities 

 of a gene. Here we may have the solution of the diffi- 

 culties which the problem of domestication affords in 

 spite of mutation and recombination. No doubt the high 

 capacity for fattening was crossed into our hogs with 

 Asiatic forms. But selection of plus-quantities of the 

 responsible gene enabled us to obtain the character as 

 it stands to-day. 



