Experimental Evolution, Variation, and Heredity 101 



No. 188. (Paper No. 20, Station for Experimental Evolution.) 



DAVENPORT, CHARLES B. Heredity of Skin-Color in Negro-White Crosses. Octavo, 

 106 pages, 4 plates. Published 1913. Price $1.75. 



The heredity of skin color in negro-white crosses is of importance for the theory 

 of heredity and in its application to social conditions. Its theoretical significance 

 lies in the fact that it is the last line of defense of those who deny the general va- 

 lidity of "Mendelism" and the modern science of heredity that has grown out of it. 

 In skin color at least it is asserted we have true blending and no "segregation." 

 This paper is based on the measurement of skin color of between 600 and 700 chil- 

 dren, about 200 pairs of parents, and various other relatives. The result is clear 

 cut: Skin color is inherited in "Mendelian" fashion; and segregation does take 

 place. The evidence for this conclusion rests (1) on the fact that the color of the 

 offspring of two strict mulattoes is much more variable than is the skin color of the 

 hybrids of the first generation (compare table 18 with table 11), and (2) upon the 

 fact that among the offspring of mulattoes are white and black children, who "breed 

 true" to their respective colors. The apparent blend that has long been recognized 

 in skin color is due to the large number of factors that produce negro skin color. 



Studies are made into special phases of the question of inheritance of skin color. 

 The development of pigmentation in the child is considered; the fact that even a 

 negro child is born relatively white, and does not attain full pigmentation until 

 puberty, complicates the main investigation. There is no difference in the skin color 

 of adult negro men and women and the color of the mulatto children is the same 

 whether the father or the mother is the white parent. The tradition that a woman 

 who passes for white may have by a Caucasian consort a black child receives from 

 these studies no confirmation and is probably false ; on the other hand a "light 

 colored" pair may have children who are much darker than either parent. 



Data were studied relative to eye color, hair color, and curliness of the hair. 

 The grades of eye color are numerous ; the darkest browns are duplex in nature, 

 t. e., receive the determiner for brown from both parents ; the hazel and many 

 lighter browns are simplex in origin, or heterozygous. In hair pigmentation many 

 factors are involved. In general the hair of the children is not darker than that 

 of the darker parent. As for curly hair, it dominates over straight; and the curlier 

 types (frizzly, wooly) dominate over the less curly. 



Certain social applications of this study are drawn. There is no correlation, in 

 the children of mulattoes, between the color of the skin and the curliness of the hair. 

 It is proper to infer that many other physical and many mental characteristics of 

 the negroes are not necessarily associated with the color of the skin. It is quite pos- 

 sible that there may be derived, in later generations, from a negro X white hybrid, 

 not only a white person with Caucasian mentality but also a black-skinned person 

 with Caucasian mentality and moral traits. 



There is also no foundation for the view that hybrids are less fertile than the 

 pure races. Some of the most fecund families of Jamaica are those of mulattoes. 



No. 195. (Paper No. 21, Station for Experimental Evolution.) 



CASTLE, W. E., and JOHN C. PHILLIPS. Piebald Rats and Selection: An Experi- 

 mental Test of the Effectiveness of Selection and of the Theory of 

 Gametic Purity in Mendelian Crosses. Octavo, 56 pages, 3 plates. 

 Published 1914. Price $0.75. 



The hooded pattern of tame rats behaves as a recessive Mendelian unit-char- 

 acter in crosses with other color varieties of rat. In the second generation from 

 such a cross, one-fourth of the offspring are hooded. Notwithstanding the be- 

 havior of the hooded pattern as a clearly segregating unit in heredity, the unit is 

 slightly modified by crosses. It may also be modified by "mass-selection" unattended 

 by crossing. Twelve generations of selection in opposite directions (plus and 

 minus) from a stock of hooded rats has produced races widely different in pat- 

 tern. Return selection toward the original racial condition progresses at the same 

 rate as the departure from that condition had occurred. This, with the gradual 

 disappearance of regression upon repeated selection shows the effects of selection 



