12 ANIMAL COLOEATION. 



domesticated, and what was the original stock, or probably 

 stocks, from which an infinite variety of dogs have been bred 

 and selected? During this period they have been subjected 

 to every variety of treatment, have been fed with all kinds 

 of food, and have been repeatedly crossed with other breeds. 

 These facts alone give an opportunity for variation such as is 

 not possessed by most wild animals. Secondly, variations in 

 colour do of course occur among animals under natural con- 

 ditions; the pages of our entomological journals constantly 

 contain records of " varieties " ; any one who will take the 

 trouble to consult the beautiful plates illustrating Mr. Buckler's 

 " History of British Lepidopterous Larvae," now being published 

 by the Ray Society, will find that in several of them (e.g. 

 Lophopteryx cimelina—four varieties figured) the colour is by 

 no means constant. 



The Ruff is, of course, the classical instance ; it is said that 

 no two specimens of this bird are alike. One of the plates 

 illustrating Dr. McCook's work * upon " American Orb-weaving 

 Spiders " is devoted to the illustration of eight or nine colour 

 varieties of the female of Epeira trifoUum. In fact, if colouring 

 were really constant for a given species, there would be no 

 chance for natural selection. 



Supposing that a marked variety occurs in a wild species, 

 there is, first of all, a considerable chance against its reach- 

 ing maturity ; secondly, there is a considerable chance 

 against its finding a mate ; thirdly, the hereditary influences 

 on both sides are against the perpetuation of the variety. These 

 appear to be more potent causes of the comparative fixitv of 

 colours in wild animals than the unfitness of the varieties 

 to live. 



In domestic animals the two first difficulties are removed. 

 ♦ Vol. ii., PI. I. 



