ASSORTATIVE MATING FOR COLOR IN AMERICAN NEGRO 



125 



in which the two races differ, the mean and standard deviation of the NNW 

 were more like the Negro, the NW intermediate, and the NWW more 

 white. 3 



The present writer utilized Herskovits' data for an analysis of the 

 mechanism of pigmentation inheritance, and found that the black pig- 

 mentation is not inherited as a blend, nor is it produced by the action of one 

 or two factors which act as Mendelian dominants. 4 Both total and fraternal 

 variability increase with increasing intermixture with the white race, and 

 with increasing differences between the percentages of N (black) pigmenta- 

 tion of the parents. 5 If the types of N, NNW, NW, and NWW matings are 

 arranged in order of decreasing amounts of Negro ancestry, from N X N, 

 most Negro, to NNW X NWW and NW X NW, most racially mixed, to 



TABLE 1 



Variability of offspring according to genealogical class of father and mother 



* Sf is the fraternal variability, 5 the family variability, and a the standard deviation 

 or total variability. 



NWW X NWW, least Negro, the fraternal variability increases from 3.19 

 for unmixed Negro crosses to 7.34 and 6.78 for crosses representing the most 

 racial intermixture, and then decreases to 5.91 for crosses representing the 

 least Negro and the most White mixture (table 1). That is, any type of 

 crossing which tended to increase the proportion of Negroes having approx- 



3 Herskovits, M. J., "Does the Negro know his father?", Opportunity, October, 1926. 



4 Barnes, Irene, "The inheritance of pigmentation in the American Negro," Unman 

 Biology, September, 1929, pp. 321-381. See also Davenport, C. B., Heredity of Skin- 

 Color in Negro-White Crosses, Carnegie Institution of Washington, Publication No. 188, 

 1913. 



6 For a discussion of fraternal variability, see Herskovits, M. J., "A further discussion 

 of the variability of family strains in the Negro-White population of New York City," 

 Journal of the American Statistical Association, September, 1925, pp. 1-10. 



