128 



IRENE BARNES TAEUBER 



light, medium, and dark crosses was 6.06; while the fraternal variability, 

 according to the assumptions made as to random mating in the preceding 

 paragraph, would have been 6.12. 



The variability of the offspring appears to be less than it would have 

 been if mating were entirely random, but greater than it would have been 

 if all Negroes had married persons of approximately the same color and 

 genealogical class as themselves. One important consequence of the type 

 of mating that is occurring is that the proportion of unmixed Negroes is 

 much less among the children than among their fathers and mothers, 29 

 per cent among the latter and 14 per cent among the former (table 4). On 

 the other hand, 54 per cent of the parents and 72 per cent of the offspring 

 belong to the NNW and NW genealogical classes, while 17 per cent of the 

 parents and 14 per cent of the offspring belong to the NWW. 



To the extent to which this sample is typical, it would appear that the 

 American Negro population of the future will be more homogeneous as to 

 ancestry, i.e., there will °be a smaller percentage of unmixed Negroes, a 

 larger percentage of persons with half or more Negro ancestry, and probably 

 a smaller percentage who pass as Negroes but have more White than Negro 

 ancestry. The process of social selection apparently operating to give the 

 preference in mating to the lighter Negro woman will mean a less propor- 

 tion of Negro ancestry in the population as a whole only if it means that an 

 undue proportion of dark girls do not marry. 8 In any case, if lighter 

 women marry darker men, and vice versa, the offspring on the average will 

 be darker than their lighter parent and so more likely to remain within the 

 Negro group. However, the segregation process operative in the inherit- 

 ance of pigmentation will prevent the development of a population of one 

 uniform hue. 



8 In Jamaica, there is some indication that the mate selection is the opposite of that 

 in the American Negro population, "black" rather than "white" features being preferred. 

 See Davenport, C. B., and Steggerda, Morris, Race Crossing in Jamaica. Carnegie Insti- 

 tution of Washington, Publication No. 395, 1929, pp. 296. 



