CHAP, xxxix] OF THE PAPUAN ISLANDS. 435 



for safety on their resemblance to foliage, their horny 

 shield and wing- coverts, and their spiny legs. 



The large islands to the east of New Guinea are very 

 little known, but the occurrence of crimson lories, which 

 are quite absent from Australia, and of cockatoos allied to 

 those of New Guinea and the Moluccas, shows that they 

 belong to the Papuan group ; and we are thus able to 

 define the Malay Archipelago as extending eastward to the 

 Solomon's Islands. New Caledonia and the New Hebrides, 

 on the other hand, seem more nearly allied to Australia ; 

 and the rest of the islands of the Pacific, though very poor 

 in all forms of life, possess a few peculiarities which 

 compel us to class them as a separate group. Although 

 as a matter of convenience I have always separated the 

 Moluccas as a distinct zoological group from New Guinea, 

 I have at the same time pointed out that its fauna was 

 chiefly derived from that island, just as that of Timor was 

 chiefly derived from Australia. If we were dividing the 

 Australian region for zoological purposes alone, we should 

 form three great groups : one comprising Australia, Timor, 

 and Tasmania; another New Guinea, with the islands 

 from Bourn to the Solomon's group; and the third com- 

 prising the greater part of the Pacific Islands. 



The relation of the New Guinea fauna to that of 

 Australia is very close. It is best marked in the Mam- 

 malia by the abundance of marsupials, and the almost 



F F 2 



