336 BIRD BEHAVIOUR 



Hybrids have generally to be studied in captivity* 

 where they are so freely produced that in many 

 cases a bird is practically as likely to breed with an 

 ally as with its own species, the reason for which 

 has been suggested at the close of Chapter VIII, 

 as dependent on personal preference ; but in a 

 wild state they are always, with one exception, 

 very rare, even among the Ducks, in which group 

 they occur more frequently than in any other birds, 

 though in the Goose and Swan section of the same 

 family unknown. 



Thus, in India, Hume in all his many years of 

 wild-fowling and collecting, and in spite of his nume- 

 rous correspondents, apparently never came across a 

 specimen, and indeed only one undoubted one to my 

 knowledge has ever turned up there, recorded by 

 Mr. W. L. Sclater a few years before I went out ; 

 this was between Mallard and Gadwall, but in 

 general coloration more recalled a Teal, which must 

 make one careful about assigning ancestral reversion 

 to hybrids which resemble species not concerned, for 

 there is no reason to suppose the Teal to have been 

 ancestral to Mallard and Gadwall, unless the 

 latter has much degenerated in colour. I am there- 

 fore inclined to agree with M. Suchetet, that the 

 comparative frequency of wild-fowl hybrids in 

 Europe is due to the amount of shooting that 

 goes on, and to the practice of keeping waterfowl 

 pinioned, both causes leading to birds being de- 

 prived of the power of flight, and so unnaturally^ 

 isolated in a territory they would not naturally- 



