Modification of Unit Characters 113 



on Protozoa. JENNINGS, for example, takes a single rhi- 

 zopod (Difflugia) and uses it as the basis of a new strain. 

 Every member of this new strain has come from this 

 original individual by cell division. In this case sex 

 is not involved, and the possibility of new combina- 

 tions of factors is eliminated. It is obvious that such 

 a strain should breed perfectly true, but JENNINGS 

 shows that it does not. Changes of two kinds occur, 

 namely, mutations, which of course are a part of the 

 Mendelian program, but according to JENNINGS "the 

 overwhelming majority of hereditary variations are 

 minute gradations. Variation is as continuous as can 

 be detected." He gradually piles up these minute 

 variations by selection and finally develops a new 

 species. There can be no claim of piling up cumulative 

 factors here, for no sex is involved nor can these gradual 

 changes be mutations. Mutations of course can be 

 minute, but they are not continuous and in the same 

 direction; mutations are jumps, even if the jumps are 

 small, and they occur in every direction. 



The only other possibility is as follows. It will be 

 remembered that WEISSMAN claimed that environment 

 may directly affect the germ plasm in the simplest micro- 

 organisms, just such organisms as JENNINGS deals with; 

 but JENNINGS claims that the characters of his rhizopods 

 are not modified by environment. How he is assured 

 of this is a question, for the factors of environment are 

 very numerous and complex and organisms are probably 

 very sensitive. So far as circumstances permit, the 

 conditions of environment were kept constant, but 

 whether they really were constant or not is another 

 matter. Leaving out of consideration this question of 



