PHENOMENA OF INHERITANCE 289 



which is intermediate between the white parent 

 and the black one, and even in the F 2 and in 

 subsequent generations mulattoes do not usu- 

 ally produce pure white or pure black chil- 

 dren, though the children of mulattoes show 

 considerable variation in color. Hence there 

 seems to be a failure of the Mendelian princi- 

 ple of segregation. 



But white skin is not really white nor is 

 black skin ever perfectly black. Davenport 

 has shown that there is a mixture of black, 

 yellow and red pigment in both white and 

 black skins, though the amount of each of 

 these pigments varies greatly in negroes and 

 whites. The relative amounts of these pig- 

 ments in any given case may be determined by 

 means of a rotating color disk. A white person 

 may have a skin color composed of black (b) 

 8 per cent., yellow (y) 9 per cent., red (r) 50 

 per cent., and absence of pigment or white (w) 

 33 per cent. On the other hand a very black 

 negro may have b 68 per cent., y 2 per cent., 

 r 26 per cent., »4 per cent. The nine children 

 of two mulattoes, the father having 13 per cent. 

 of black and the mother 45 per cent., ranged all 



