EVOLUTION IN NATURE AND UNDER DOMESTICATION. 237 



rations as that of the Fantail, or the Pouter or the Maltese may 

 be sought in a cross with a Woodpigeon, but if we see how walt- 

 zing rats originate in the second generation of a cross between 

 two outwardly identical species, the Javanese and the Suma- 

 tran field-rat, this hypothesis loses much of its improbabliity. 



We cannot continue to believe that all the tame breeds of 

 pigeons are derived from one species. As we did in the case of 

 the fowl, we think we can here quite logically call the domestic 

 pigeon by the name of Columba livia hybrida, and state that the 

 total potential variation of Columbia livia has been heightened 

 by crossing with Oenas and possibly other species. 



The third case is that of the duck. So many cases of fertile 

 duck hybrids are known, that it is really quite superfluous to 

 believe that all tame ducks must have descended from Anas 

 boschas. As we know that absolutely new characters sometimes 

 originate as the result of crossing, there is no great difficulty 

 in the way of the assumption, that the variability of Anas 

 Boschas, after it was taken into cultivation, was repeatedly 

 heightened by crossing with other species, Dafila acuta for 

 instance. We now know, that in such a case we need not look 

 for characters belonging to these other species in our tame 

 ducks. They may be absent. If after a cross with Dafila acuta 

 animals with a top-knot are produced, or albinos, or blacks, or 

 penguin-like animals, such animals have assuredly later on, 

 been crossed back into the old species, and only the aberrant, 

 varietal character is retained. 



The case of the rabbit is somewhat more difficult, because 

 the only species which can be drawn into account in Europe to 

 have heightened the variability of the wild rabbit, is the Euro- 

 pean hare, Lepus timidus. 



Experiments have been tried in several Zoological gardens to 

 mate the hare with rabbits, by accustoming the animals to- 

 gether when young, but hybrids have never resulted from these 

 trials. In the first place, we must not lose sight of the exist- nee 

 of very many different species of wild rabbits, in Europe as well 

 as in America. The assumption, that from a cross between the 



