22S ON THE GEOGRAPHICAL 



^vrong, then, to suppose, that a being wliich inhabits a 

 country of volcanic origin, can never be found but in soils 

 of a similar kind ; or that soils of a primitive formation 

 ought to nourish onlv particular species. Supposing, 

 even, that this should be the case, as in Java compared to 

 Sumatra and to Borneo, it would be inisafe to attribute 

 these differences to the different age of the former isle, and 

 to the defect of a soil or of a food proper to the existence 

 of tlie animals. It is, therefore, difficult, if not impossible, 

 to class these countries, according to their productions, into 

 cantons, or to seek to establish regions, as has been done 

 for the distribution of plants, on the surface of certain 

 countries. As to the Great Indian Archipelago, we find 

 that the islands of the Straits of Sunda offer, in their pro- 

 ductions, many analogies with the neighbouring parts of 

 the Asiatic continent, and even witli those of Bengal, and 

 the Indian Peninsula. The isles of Sumatra and of Bor- 

 neo support a vast niimber of animals, and some of great 

 size, which are not found in Java ; but there is also, in the 

 animals of small size, a very great analogy between the 

 southern part of Borneo and the island of Java. This last 

 isle produces several animals that arc also found in the 

 southern parts of China, and probably also in the chain of 

 islands that extends eastward to Timor. It is in the vast 

 island of Celebes, that several forms of animals, unknown 

 in the islands of the Straits, begin to shew themselves ; 

 these novel forms become more numerous in the Moluccas, 

 and assimilate to those peculiar to New Guinea, Avhich in 

 several instances recall, in their turn, the heterogeneous 

 productions of New Holland. The Philippine Isles offer 

 an analogy mth those of the Straits of Sunda ; and, what 

 is curious, we there observe several animals of Ceylon 

 and of India. In confining ourselves to the observa- 

 tions suggested by the classes of Mammifera and of Rep- 

 tiles, we see that the principal characters that distinguish 

 the Fauna of these islands are the followino; : — The 



"» 



in Sumatra and Boi'neo, in places entirely analogous, and never frequent 

 situations of another nature ; they do not diffuse themselves over the 

 whole island, although they are not hindered to do so by any physical 

 obstacle : it is thus also with most other animals. 



