VARYING COLOURS IN FISHES. 83 



colours being like that of a tortoise ; the nature of 

 tlie water, exposure to light, and, probably, their 

 kind of food, producing these effects." (Salm. 41.) — 

 Here, then, are instances of a permanent change 

 in colour, and constituting the distinction of a va- 

 riety. Examples of this sort are, we believe, by no 

 means rare ; and the operation of such accidental 

 circumstances and artificial influences is strikingly 

 illustrated by the well-known Gold-fish, Cyprimcs 

 auratus. Its frequent companion, the common Sil- 

 ver-fish, is of the same species, with the mere dif- 

 ference of metallic tinting ; and M. Sauvigny, in his 

 learned work on this fish, has represented no fewei 

 than eighty-nine varieties in form and colour, ma- 

 nifesting all shades of silvery-white and purple 

 orange, red, and gold. 



The causes of these changing hues, whether merelj 

 versatile or permanent, are so latent and obscure, 

 that scarcely a conjecture has been hazarded regard- 

 ing them ; Suhtilitas naturce suhtilitatem census et. 

 intdlectus tnultis partibus super at. The cause in 

 certain instances has been recently hinted at, and 

 with a short reference to these we must dismiss the 

 subject. Mr. M'Lelland, in an able and elaborate 

 Paper on the Indian Cyprinidee (Ann. of Nat. Hist., 

 viii. 35), wherein he proposes a new arrangement 

 into which colour enters as one element, connects the 

 livery which is assumed with the circumstance of 

 the food of the fish being animal or vegetable. He 

 informs us th at the whole Sub-family of the Poeono- 

 7nince is remarkable for their uniform plain colours, 



