I9I0.] DAVENPORT— NEW VIEWS ABOUT REVERSION. 293 



not appear but when two parents together produce the required 

 factors the combination may be an apparently new, compound char- 

 acter; which we find elsewhere only in remote ancestry. 



The facts of reversion are most notorious among domesticated 

 animals and plants. The reason is that man has preserved just 

 those strains of germ plasm that are peculiar in the absence of some 

 typical characteristic or the presence of some new characteristic. 

 These new conditions either cause the ancestral condition to fail of 

 development or mask it over. In hybridizing we restore the factor 

 that is missing from one strain by introducing it from another strain ; 

 or we remove the added factor that veils the ancestral condition. 

 Thus the ancestral condition is restored — a reversion occurs: 



The foregoing general statement may now be illustrated by some 

 examples.- The goldfinch, which has plain chestnut sides, when 

 crossed with a plain yellow canary produced hybrids that have 

 stripes on back and flanks. Darwin mentions this case^ and adds : 

 " this streaking must be derived from the original wild canary." 

 The results of breeding indicate that the yellow canary has one 

 factor for these stripes — as it were, the potential pattern — but no 

 pigment to bring it out. Adding the factor of pigment from the 

 goldfinch the pattern appears in the ofifspring. 



Reversion in poultry was studied by Darwin. He crossed a 

 White Silkie hen with a Spanish cock which is perfectly black 

 except for the iridescent glossy black in hackle, saddle, wing bar 

 and some of the tail feathers. " As the cocks grew old one . . . 

 became a gorgeous bird. When stalking about it closely resembled 

 the wild Callus hankiva, but with the red feathers rather darker." 

 The hens were black. I have made the same cross with the same 

 result. Moreover, when the hybrids were mated together some of 

 the females as well as males assumed a perfectly typical Jungle 

 fowl coloration. The " reversion " was now complete. 



The interpretation of this case on the factor hypothesis is some- 



' The colored lantern slide illustrations are not here reproduced; they 

 appear in part in the author's works entitled, " Inheritance in Poultry," 

 " Inheritance in Canaries," and " Inheritance of Characteristics in Domestic 

 Fowl," all published by the Carnegie Institution of Washington. 



'Var. Dom. A. and P., Chap. XIII. 



