MODIFIABILITY OF HOODED CHARACTER 6 



grade —2.75 were crossed with wild males of the same race used in 

 the crosses of the plus-selected race. The extracted hooded Fo young 

 were highly variable, ranging in grade from —2.25 to +3.00, mean 

 — 0.3S, a remarkable change in the plus direction of over 2 grades. 



A second cross with the wild race (table 11) brought about a further 

 movement of the mean in a plus direction but by a somewhat smaller 

 amount, 1.39 grades, the mean of the twice-extracted hooded young 

 being +1.01. A third cross with the wild race (table 12) brought 

 still further contamination of the hooded character, which now ceased 

 to vary below grade +1.00, and had a mean of +2.55 in the case of 

 over 100 thrice-extracted hooded young, this being a change in the plus 

 direction of 1^ grades. It will be observed that the hooded grand- 

 children of cf 2068, table 12, the most plus in character of the hooded 

 grandparents, are very similar in grade to the hooded j^oung resulting 

 from three crosses of the plus-selected hooded character with the 

 same wild race (table 9). In other words, the same wild race, when 

 its residual heredity is made fully effective by repeated crosses, brings 

 both the plus-selected and the minus-selected hooded lines to a pheno- 

 type of common grade. This shows, contrary to my earlier opinion, 

 that what has really happened in the case of the selected races was 

 more largely due to residual heredity than to any change in the gene 

 for the hooded character itself. My critics have been wrong when 

 they insisted that selection could not change racial characters that 

 niendelize and change them permanently, and when on this ground 

 they denied to gradual change through selection an important part 

 in the evolution of characters and thus of races. But my critics have 

 been right when they insisted that evidence is wanting that change 

 in single genes occurs other than spontaneously, uninfluenced by sys- 

 tematic selection. 



