ORDER CARNASSIER. 405 



its habitat, are the only characters which distinguish the 

 Spotted from the Striped Hyssna. 



We begin to fear that, from frequent recurrence to the 

 same position, it may be thought we are rather broaching 

 a particular hypothesis than concentrating by compila- 

 tion to one focus the various labours of others, which, 

 with occasional comments and reflections, and original 

 graphic illustrations, forms in reality the more hum- 

 ble object of our endeavours. Renouncing, therefore, 

 more lofty pretensions, and deprecating the anger of those 

 who may know better and think differently, we again ad- 

 vert to the too great readiness with which some naturalists 

 have established diversity of colour as a ground of diversity 

 of species. There are cases, however, we mUst allow, in 

 which there is no other obvious criterion of distinction, in 

 the present limited state of our zoological knowledge. Two 

 species may have always remained distinct, though the only 

 point of dissimilarity between them may be a very slight 

 variation in the arrangement of their colours. Yet even 

 here we should be cautious, and hesitate to pronounce any 

 more than a provisional judgment on the subject. Many 

 causes may prevent the intercourse of animals, which, from 

 their not intermixing, we refer to different species. When 

 we see two races of animals inhabiting the same country 

 never intermix, and always preserve the same external dif- 

 ferences, we may with confidence pronounce them distinct. 

 But we must always be liable to some error in our judg- 

 ments concerning animals which inhabit different countries, 

 and whose characteristic differences are slight and external. 

 Even if when we bring them together, they refuse to in- 

 termix, it would be no sufficient proof of specific difference, 

 for we know how much the instincts of wild animals are 

 weakened or perverted in the unnatural state of captivity. 

 This is true even of those animals of whose specific iden- 

 tity no doubt can be entertained. Transpose the habitats 



