ARTIFICIAL SELECTION 337 



A comparison of the heads of the wild ancestor and of a prize- 

 winning Tumbler will bring out this feature better than a long 

 description. 



But besides the external modifications of form, which are 

 the result of man's selection, certain internal structural changes 

 have taken place which have had no part in this deliberate 

 scheme of modification, such as a reduction in the number of 

 ribs, for instance, nor does it seem that this reduction can be 

 correlated with any external characters. 



As touching the origin of these breeds, of which examples 

 are given in the accompanying illustration, it would seem that 

 they may have arisen, in the first place, as "sports". That is 

 to say, some more or less sudden departure in this or that 

 direction having caught the breeder's eye, he set to work 

 at once to perpetuate and " fix " such sport, and, by the 

 method of selection already alluded to, gradually built up a 

 new race. 



As an illustration of this it may be mentioned that during the 

 last decade a race of web-footed Pigeons has been established, 

 and this not by the cumulative action of selection, that is to say, 

 by the careful selection of those Pigeons which had the largest 

 expanse of the normal webbing between the toes until by de- 

 grees a well-marked webbing was produced, but by mating 

 from a single example which showed a pronounced inter-digital 

 webbing, and careful in-breeding therefrom. 



The most striking proof as to the contention that our 

 domesticated races of Pigeons have been derived from the Rock- 

 dove is that furnished by reversion or the failure to complete 

 the final ontogenetic stages of development. This was first 

 demonstrated by Darwin who crossed a white Fan-tail with a 

 black Barb, then a black Barb with a red Spot — a white bird 

 with a red spot on the forehead and a red tail. The offspring 

 of this cross, the mongrel Barb-fan-tail and Barb-spot, were then 

 crossed, and their offspring developed all the characteristics of 

 the wild Rock-dove I This reversion, it is to be noted, was 

 complete. That is to say, it was not merely blue coloured 

 birds that resulted — caused by a dilution of black pigment by 

 the mating of black-pigmented birds with birds in which pig- 

 ment was practically absent — but birds that reproduced all the 

 characteristic markings of the Rock-dove, among which are 



22 



