HEREDITY. 19 



area in hooded-rats. We found, that two dark-hooded animals 

 sometimes gave a minority of light-hooded young, and this was 

 always the case when one of the parents of the animals had 

 been light. Two light-hooded rats however, no matter what 

 their ancestry, never produced any dark-hooded young. 



A circumstance, which has certainly done much to strength- 

 en Castle and his school in their belief in an effect of selection 

 on the quality of the genes, is the over-distinctness of the differ- 

 ence resulting from a presence or absence of such genes, as 

 were first studied by the Mendelians. The very fact that such 

 genes, as produced a difference between black and white colour 

 or between tall and dwarf stature were first studied, made it 

 seem necessary to speak of such genes, as by their cooperation 

 to the development produced a hardly appreciable difference 

 in shade of black or a minute difference in stature, as of modi- 

 fying factors. This unhappily chosen nomenclature, and the 

 tendency to lump genes which happen to influence one and 

 the same quality (no matter how, physiologically spoken) and 

 speak of them as "polumeres," has created the impression which 

 has certainly no foundation in fact, that such modification 

 factors were the same old gene somewhat modified, and that 

 sets of "polymeres" had originated by the splitting up of some 

 one gene. 



If the majority of the Geneticians had not approached Genet- 

 ics from the side of Botany or Zoology, and had not started by 

 observing a few striking inherited differences, but if they had 

 happened to become interested in Genetics as in a study of 

 those numerous factors in the development of the organism 

 which are transmitted through the germ, we would not now 

 find so many authors hampered by conceptions of unit-charact- 

 ers, and illusions about the purity of characters issuing from 

 a cross. 



Are the facts brought to light by the Cytologists compatible 

 with my hypothesis, that the genes are relatively simple auto- 

 katalytical substances? We know, that there are cases in 

 which characters are inherited in an unusual way, that in some 



