484 TIUVELS IN TEE EAST INDIAN ARCHIPELAGO. 



ing this man, who had learned some Malay during 

 his stay at Padang mth me as an interpreter. An 

 unexpected event, however, made it necessary to send 

 that boat up the coast, and it would be some days 

 before another would come ; so I concluded to take 

 the mail-boat for Bencoolen, and commence a long 

 journey directly across the island to Palembang, and, 

 reaching Banca, go up to Singapore on the steamer 

 which touches at that island while on her way to 

 Singapore from Batavia. 



While travelling in the interior of Sumatra, we 

 have seen that the mountains^ which extend from one 

 end of the island to the other, range themselves, gen- 

 erally, in two parallel chains, that wall in a long, 

 narrow plateau. The island of Engano is the sum- 

 mit of the southeastern peak in another similar 

 mountain-chain, extending in a northwesterly direc- 

 tion, parallel to tho^e akeady described. After sink- 

 ing beneatli the level of the sea^ this chain reappears 

 in the Pagi, Mantawi, and Batu groups, Pulo Niaa, 

 Pulo Babi, and the Cocos Islands. 



The plateau in the interior, we have also found, 

 is divided into a iituuber of separate valleys^ by 

 transverse ranges, which yoke together the principal 

 chains. In a similar manner transverse ranges ap- 

 pear in Pulo Kapini, one of the Batu Islands, and in 

 the Banyak Islands. These transverse ranges are 

 seen also in the high and well-marked promontories 

 which jut out from the Barizan, or coast-chain of 

 Sumatra, at those places. A thu'd projecting part 

 of the coast is seen at Indrapura. As the valleys in 

 the interior become plateaus, when we compare them 



