130 



THE AMERICAN NATURALIST [Vol. XLIV 



tall, crossed with nana, 40 cm. tall, produces offspring of 

 70 to 85 cm., thus failing to attain the dominant stature. 

 Shull finds, in Shepard's purse, the heterozygote between 

 elongated and non-elongated primary lobe to be imper- 

 fectly elongated, and the same investigator has devised a 

 neat chemical experiment illustrating imperfect domi- 

 nance. Morgan states that in hybrids between gray and 

 albino rats, gray fails on the belly. In Lang's snails uni- 

 form shell color dominated over bands, but the hybrid 

 shows pale bands. In poultry, extra toe dominates im- 

 perfectly over normal toe ; complete inhibition of shank 

 feathering dominates imperfectly over non-inhibition, so 

 that some feathers appear ; median comb dominates over 

 absence so imperfectly that it is much reduced in the 

 hybrids ; dominant whites crossed with black give speckled 

 females. Thus, the discoverer of dominance and his fol- 

 lowers have generally recognized and laid emphasis on 

 its imperfection; and were it not for certain careless 

 exponents of Mendelism and a human tendency to sub- 

 stitute a skeleton formula for the living truth this intro- 

 duction might have been spared. 



Of imperfection of dominance there are all degrees, as 

 Correns pointed out in 1905. The extreme case is com- 

 plete failure to dominate; it has been observed many 

 times. Thus Lock found that while black maize domi- 

 nates over white, whites are in excess of expectation in 

 subsequent generations and some of them prove to be 

 heterozygous. Lang found hybrids between red and 

 yellow snails to give exceptionally the recessive yellow. 

 In poultry, median comb crossed with no median will give 

 no median in 5 to 10 per cent, of the offspring. Extra 

 toed crossed with normal poultry give over 20 per cent, 

 of the recessive normal-toed type. The inhibitor of the 

 narial flap when crossed with narrow nostril fails to 

 activate in fully half of the offspring. Finally, cases are 

 not unknown, as in certain rumpless fowl, when the domi- 

 nant inhibitor has failed to act in the heterozygote com- 

 pletely. Several such cases have in recent years come to 



