Wright : Color Inheritance in Mammals 



231 



granules (Hsajipear first in front part of 

 the iris. \\'ith pigment on the back, but 

 not on tlie front, the effect is bhie. A 

 bluisli or greenish appearance comes 

 in rather abruptly when the anterior 

 pigment is reduced to a certain point. 

 With further reduction of the quantity 

 of pigment the iris becomes a pale gray 



bv considering the proportions of the 

 different classes in such a population. 

 Such a test applied to data in short- 

 horn cattle, in an earlier paper, pointed 

 overwhelmingly to the unit factor hy- 

 pothesis. In this present case, how- 

 ever, the result is not so favorable. 

 The data are summarized below. Bv 



blue as in albinos, and the parallel re- 

 duction in the retina reaches a point at 

 which vision is impaired and a red 

 reflex becomes visible through the pupil. 



Hurst^- and Davenport.' ■' working 

 independently, arrived at the conclu- 

 sion that dark eye colors are dominant 

 over the lighter colors. Hurst drew 

 the line so as to inckide only the purest 

 blues in the recessive class, considering 

 presence and absence of pigment on the 

 part of the iris as the opposed charac- 

 ters. Davenport drew no such sharp 

 line, but concluded that any darker 

 color was dominant over any lighter 

 color. He recognized that the results 

 were not wholly clear-cut and that com- 

 plications must exist. Holmes and 

 I.oomis^^ were more doubtful as to the 

 value of the Mendelian description. 

 They found that the darker colors 

 tended to be dominant over the lighter 

 ones, but that inheritance often seems 

 blending and that offspring might be 

 darker than either parent. 



Hurst based his conclusions on data 

 from a single village in Leicestershire. 

 As the data show that there is no ap- 

 preciable assortative mating for eye 

 color, the population may be assumed 

 to be in equilibrium. A good test of 

 the unit factor hyj:)Othesis can be made 



duplex. Hurst means an eye with pig- 

 ment on both sides of the iris, by sim- 

 plex one with only posterior pigment — 

 ;'. c, a clear blue eye. 



Among the children 37.5% are sim- 

 plex. Among the parents 39.2%. The 

 number of children from matings of 

 simplex by simplex with random mat- 

 ing should be .375 x .375 or 14.1%. The 

 actual number is 14.8%. The number 

 from matings of simplex bv duplex 

 should be 2 x .375 x .625 or 46^9% . The 

 actual number is 47.4%. There is. evi- 

 dently, no appreciable assortative mat- 

 ing. As the number of heterozygous 

 dominants must be twice the product 

 of the square roots of the homozygous 

 classes in a population which is in 

 equilibrium, and is without assortative 

 mating, we can deduce the number of 

 the two kinds of dominants. Where 

 x- is the proportion of recessive 

 ( 37.5% ). the proportion of liomozygous 

 dominants must be (1 — x)-. which 

 equals 15.0%. and the proportion of 

 heterozvgous dominants must be 47.5% . 

 From tliese figures, the number of sim- 

 plex children which should be produced 

 by each type of mating may be calcu- 

 lated. Thus duplex by duplex should 

 produce 57 simplex children, much in 

 excess of the number found (IS) 



'=Hurst, C. C, 1908. Proc. Rox. Soc. B., 80:85-96. 



'"Davenport. C. B.. and G. C. "Davenport, 1907, Sci. X. S., 26 :589-592. 



"Holmes and Loomis, loc. cit. 



