RESULTS. 21 



Curly X straight will give wavy or straight, never kinky. It is impos- 

 sible to admit the legitimacy of this child. 



Table II. The mother produces two kinds of gametes: gametes 

 with i factor for black and those with 2 factors ; consequently an equal 

 number of light and of medium-colored progeny is to be expected; 

 and this expectation is nearly realized. In one of the offspring the 

 skin color is given as N 7; in another it is N 45. In both cases the 

 resemblance in other respects is rather close to the mulatto mother. 

 Such a range from such a pair of parents is unique, and there is 

 internal evidence that justifies doubt as to the paternity of these two 

 children. 



Table 12. This mating gives the strict mulatto, the generation 

 in which great uniformity of offspring is expected on any hypothesis 

 of inheritance of skin color. All offspring should be medium-colored. 

 Of seventeen, fourteen fall in the expected grade (including two with 

 N 25 per cent, which is the limiting grade between this and the next 

 lower class). The three others fall in the next higher class. Of the 

 two exceptional individuals that are found in the same fraternity the 

 darker (N 53) has "typical negro hair," which is hardly to be expected 

 in a mulatto and justifies a doubt as to the paternity. The other, 

 with N 45 per cent, has mulatto (wavy) hair. This may be an extreme 

 case, possibly associated with the dark skin (N 70) of the mother. The 

 case in pedigree B. 3 lies at N 43, just above the limits of the mulatto 

 class, and her mother is also very dark, N 71 ; so this child, too, may be 

 regarded as a fluctuating extreme. Except for the probable "mistake," 

 i.e., illegitimacy, the expected uniformity is practically realized. 



Table 13. Expectation is the same as in table 9, namely, an 

 equal number of the first and second classes. This expectation is 

 approximately realized. 



Table 14. By hypothesis both parents should produce germ-cells 

 of two kinds: with one factor, and with no factor for black. In fer- 

 tilization the unions giving o factor, 1 factor, and 2 factors in the 

 zygote should occur respectively 1,2, and 1 times. The expected three 

 classes, and they only, are actually realized ; but for some reason there 

 is a larger proportion in the middle class than theory calls for and a 

 corresponding deficiency in the extreme classes; but the arbitrary 

 limits of our classes and developmental changes have constantly to 

 be kept in mind, so that lack of close accord is not surprising. 



Table 15. There are here two possibilities. If the darker parent 

 is a strict mulatto she will form germ-cells of three sorts {i.e., 25 per 

 cent with o factor for N; 50 per cent with 1 factor for N; 25 per cent 

 with 2 factors for N), in which case the four classes of zygotes, o, 1, 2, 

 and 3 factors for N, will have a relative frequency indicated by the 

 numbers 1:3:3:1. But if the darker parent be not a mulatto but the 

 daughter, or more remote descent, of a mulatto then both factors for 



