264 Biology in America 



the full number must be present, and (3) as a corollary to 

 (2) if less than the full number are present the result will 

 be more or less intermediate or "blending" between the 

 maximum of the character and its total absence. Thus, let 

 us assume with Davenport that the African negro contains 

 four factors for blackness, while the white man has none. 

 Two factors produce a "mulatto," one a "quadroon," and 

 three a "sambo," while the "octoroon" and the "near- 

 white" resemble the pure white, so far at least as skin color 

 is concerned. The offspring of a cross between a full black 

 and a white will be a mulatto containing two factors for 

 black. If the latter marry a white the offspring will be of 

 three classes, 1 mulatto, 2 quadroons and 1 "near-white." A 

 cross between two mulattoes will result in 1 black, 4 sambos, 

 6 mulattoes, 4 quadroons and 1 "near- white," a result readily 

 derived from the second diagram on page 262 if for the domi- 

 nant factors we substitute the factors for negro color BB. 

 Thus the chance of either original color (black or white) ap- 

 pearing in the second generation is only 1 :16, while there are 

 14 chances of an intermediate or "blending" color appearing. 

 If more than four factors are involved, the chance of either of 

 the original characters reappearing in the second generation of 

 a cross will be correspondingly lessened. Thus if six factors 

 (3 pairs) are involved the chance will be 1:64, with 8 fac- 

 tors (4 pairs) 1:256, with 10 factors (5 pairs) 1:1024 and 

 with 12 factors (6 pairs) only 1:4096. Such an hypothesis 

 readily explains on a Mendelian basis the case of the lop- 

 eared rabbit if we assume the necessity of several factors in 

 the production of a superficially simple, but fundamentally 

 complex result. 



An interesting corollary of Davenport's main thesis, 

 founded on a study of more than a hundred negro-white fam- 

 ilies in Jamaica, Bermuda and Louisiana, is the overthrow, or 

 at least serious weakening of the popular belief that a mar- 

 riage of two "near-whites" may result in children of negro 

 color. His results indicate that the offspring of a cross be- 

 tween persons of negro ancestry, can in no event have 

 more than the sum of the factors for black of the two 

 parents; so that the children of two "near- white" parents 

 can never produce other than white children, while a near- 

 white and a quadroon can at most have only quadroon 

 children. 



The basis of Mendelian inheritance is, as we have seen, the 

 chance combination in calculable proportions of definite char- 

 acters, which are segregable from one another, and do not 

 form permanent "blends." How well do the calculated or 



