The Factors of Evolution 249 



rectly affected the germ cells, or has such influence been trans- 

 mitted to them indirectly, by affecting first the body cells of 

 the parent, which effect has been secondarily transmitted to 

 its germ cells? If the latter is true, it would go far to es- 

 tablish Lamarck's theory of the inheritance of acquired char- 

 acters. In an attempt to solve this problem, Sumner has 

 reared several generations of white mice under different con- 

 ditions of temperature and relative humidity. The mice were 

 kept in unheated and heated rooms respectively, in the for- 

 mer of which the relative humidity was much higher than in 

 the latter, although in neither were the conditions at all con- 

 stant. As a result the mice in the cold room developed a 

 greater amount of hair and shorter tails, feet and ears. These 

 differences however diminished as the mice grew older. Un- 

 fortunately his equipment did not enable him to control his 

 factors properly, so that his comparisons were made between 

 the general conditions of greater and less heat and moisture 

 respectively, and not between definite degrees of these factors. 

 However his results do show a rather definite influence of 

 temperature and humidity, not alone upon the parents, but 

 also upon their offspring of the first generation. These re- 

 sults, so far as they go, show then that environment may di- 

 rectly affect not only the organism itself, but also its off- 

 spring. 



When we come to the question however of how these re- 

 sults were obtained, we are very much at sea. In the case 

 of Tower's experiments described above there can be little 

 doubt that the influence of the external factors upon the germ 

 cells was direct, because inheritable variations were obtained 

 only when the experimental factors were operative at a cer- 

 tain definite time in the development of the germ cells. Here 

 too the latter, in the body of an animal whose temperature is 

 not constant, but varies with that of its environment, are 

 readily subject to environmental influence, at least so far 

 as temperature is concerned. In the case of humidity it 

 is more difficult to see how the germ cells could be directly 

 influenced. In mice however whose body temperature and 

 moisture is constant under normal conditions, regardless of. 

 external factors, it is impossible to postulate a direct ac- 

 tion of such environmental factors upon the germ cells, and 

 the only possible interpretation to be placed on Sumner 's 

 results, assuming their accuracy, is that of an indirect in- 

 fluence of external factors upon . the germ cells, through 

 changes primarily induced in the body cells or soma, and sec- 

 ondarily transmitted from them to the germ cells; i.e., the 

 inheritance of acquired characters. 



Unfortunately Sumner 's results have not been, so far as 



