Morgan, Are the Germ-Cells nf Mcndclian Hybrids ''Pure"? 291 



be given, as explained more fully below. A cell division would 

 seem to be the simplest form of such a mechanism, but it is not 

 necessary to assume that the formation of the alternating domi- 

 nance takes place in this way, and the question may be profitably 

 left open for the present. 



When the two kinds of sperm-cells nieet the two kind of 

 egg-cells three kinds of individuals will result in the proportion of 



1:2:1; tluis : 



G(W) (G)W 



G(W) . (G)W 



1, G(W) + 2, G(\V) (G)W + 1, (GjW. 



In this formula G(W) is the extracted dominant, with white 

 in the latent condition; ' (G)W is the extracted recessive with gray 

 latent; and G(W) (G)W is the dominant recessive; in which the 

 white is recessive or free, and in a different relation to the gray 

 than in the other cases. It is, in fact, in the same relation as in 

 the first hybrid, GW, formed by crossing directly a gray and a 

 white monse. Therefore this midie group gives the same result 

 as does the hybrid of the first generation. 



The different relation of the gray and white in the middle 

 group is owing to the way in which it has been formed, viz., by 

 he union of the two germ-cells G(W) and (G)W in each of which 

 a different character, W or G, is in the "free" condition, henee in 

 the next generation these two characters alternately dominate and 

 become latent. 



It may be claimed for my idea that it explains as well as 

 does the ordinary view the Mendelian proportion; second, that it 

 accounts for the presence in the extracted recessive of latent cha- 

 racters; third, that it puts the extracted dominant and the ex- 

 tracted recessive on the same footing; and fourth, that it shows 

 why the middle term of the Mendelian equation gives a different 

 result from the end terms. 



In the next place I should like to point out a theoretical 

 possibility regarding the extracted dominants and the extracted 

 recessives. What should we expect to happen if an extracted 

 dominant, G(W), is crossed again with a pure white (if such really 

 exist)? The hybrid will be gray as has been shown, but it will 

 be a gray, according to my hypothesis, formed by the union of 

 G(W) and W, i. e. it will be G(W)W. Will the additional white 

 produce an effect on the gray color, and if not in one generation 

 can the influence be increased by repeating the process with other 

 extracted dominants in subsequent generations? Perhaps the 

 problem could be better tested for two colors not so strongly 

 opposed as gray and white. Gray and black, or gray and choco- 

 late might give better results. In the literature there are, in fact, 



19* 



