

Chap. VII. REVERSION AND ANALOGOUS VARIATION. 245 



Gr. banhiva ; and we have seen that with this species living 1 in 

 a state of nature, the ear-lappets vary in colour, beino- red in 

 the Malayan countries, and generally, but not invariably, white 

 in India. 



In concluding this part of my subject I may repeat that 

 there exists one widely-ranging, varying, and common species 

 of Gallus, namely Gr. banhiva, which can be tamed, produces 

 fertile offspring when crossed with common fowls, and closely 

 resembles in its whole structure, plumage, and voice the Game 

 breed ; hence it may be safely ranked as the parent of this, the 

 most typical domesticated breed. We have seen that there is 

 much difficulty in believing that other, now unknown, species 

 have been the parents of the other domestic breeds. We 

 know that all the breeds are most closely allied, as shown by 

 their similarity in most points of structure and in habits, and 

 by the analogous manner in which they vary. We have also 

 seen that several of the most distinct breeds occasionally or 

 habitually closely resemble in plumage Gr. banJciva, and that 

 the crossed offspring of other breeds, which are not thus 

 coloured, show a stronger or weaker tendency to revert to this 

 same plumage. Some of the breeds, which appear the most 

 distinct and the least likely to have proceeded from Gf. banhiva, 

 such as Polish fowls, with their protuberant and little ossified 

 skulls, and Cochins, with their imperfect tail and small wings, 

 bear in these characters the plain marks of their artificial 

 origin. We know well that of late years methodical selection 

 has greatly improved and fixed many characters ; and we have 

 every reason to believe that unconscious selection, carried on for 

 many generations, will have steadily augmented each new pecu- 

 liarity and thus have given rise to new breeds. As soon as two 

 or three breeds had once been formed, crossing would come 

 mto play in changing their character and in increasing their 

 number. Brahma Pootras, according to an account lately pub- 

 lished in America, offer a good instance of a breed, lately 

 formed by a cross, which can be truly propagated. The well- 

 known Sebright Bantams offer another and similar instance. 

 Hence it may be concluded that not only the Game-breed 

 but that all our breeds are probably the descendants of the 



