CONTINUOUS AND DISCONTINUOUS VARIA- 

 TIONS AND THEIR INHERITANCE IN 

 PEROMYSCUS. II 



DR. F. B. SUMNER 

 Scripps Institution, La Jolla, Calif. 



IV. Heredity of the Racial Differences 

 In view of the long-recognized correlation between cer- 

 tain of these subspecific characters— namely, those relat- 

 ing to pigmentation— -and certain factors of the physical 

 environment, the possibility has suggested itself that the 

 characters in question might be purely " ontogenetic," 

 i. e., produced anew in each generation by the action of 

 external physical factors. The simple experiment of 

 transplanting mice from one habitat to another has dis- 

 posed of this suggestion. 



As I have more than once reported elsewhere (1915a, 

 1917, 19176) entirely negative results have been reached, 

 so far as climatic influences are concerned. Neither the 

 transference of the desert race to Berkeley, nor the trans- 

 ference of both the desert and the redwood races to La 

 Jolla have resulted in any demonstrable change, at least 

 up to the third cage-born ("C 3 ") generation. 



The La Jolla test is the more satisfactory of the two, 

 since the number of animals employed is very much 

 greater. Thus far, however, only the C 2 animals (38 

 nthidus and 96 snHoriensis) have been killed, measured 

 and (in part) skinned. The C 3 generation is still kept 

 alive for breeding purposes, but the characteristic racial 

 differences are obvious. On comparing the skins of the 

 palest and the darkest rubidus, or the palest and darkest 

 sonoripnsis of the C 2 generation, with the extremes among 

 the wild grandparents, it will be seen that the range of 

 color variation has not appreciably changed. 



290 



