Chap. VI. MANNER OF FORMATION OF THE CHIEF RACES. 219 



a te of the eye-wattles ; with the reduced or increased size of the feet 



the number of the scutellse would vary ; with the length of 

 the wing the number of the primary wing-feathers would 

 differ; and with the increased length of the body in the 

 pouter the number of the sacral* vertebra would be aug- 

 mented. These important and correlated differences of struc- 

 ture do not invariably characterise any breed; but if they 

 had been attended to and selected with as much care as 

 the more conspicuous external differences, there can hardly 

 be a doubt that they would have been rendered constant. 

 Fanciers could assuredly have made a race of tumblers with 

 nine instead of ten primary wing-feathers, seeing how often 

 the number nine appears without any wish on their part, and 

 indeed in the case of the white-winged varieties in opposition 

 to their wish. In a similar manner, if the vertebrae had been 

 visible and had been attended to by fanciers, assuredly an addi- 

 tional number might easily have been fixed in the pouter. If 

 these latter characters had once been rendered constant we 

 should never have suspected that they had at first been highly 

 variable, or that they had arisen from correlation, in the one 

 case with the shortness of the wings, and in the other case with 

 the length of the body. 



In order to understand how the chief domestic races have 

 become distinctly separated from each other, it is important to 

 bear in mind, that fanciers constantly try to breed from the best 

 birds, and consequently that those which are inferior in the 

 requisite qualities are in each generation neglected; so that 

 after a time the less improved parent-stocks and many sub- 

 sequently formed intermediate grades become extinct. This 

 has occurred in the case of the pouter, turbit, and trumpeter, 

 for these highly improved breeds are now left without any 

 links closely connecting them either with each other or 

 with the aboriginal rock-pigeon. In other countries, indeed, 

 where the same care has not been applied, or where the same 

 . fashion has not prevailed, the earlier forms may long remain 

 unaltered or altered only in a slight degree, and we are thus 

 sometimes enabled to recover the connecting links. This is 

 the case in Persia and India with the tumbler and carrier, 

 which there differ but slightly from the rock-pigeon in the pro- 



