Vol. 69.] GEOLOGICAL HISTOKY OE THE MALAY PENINSULA.. 305 



boulders.' When the decomposed granite is attacked by the 

 mountain-torrents, it is evident that the clay and sand will be 

 washed away, and that the core-boulders will accumulate as huge 

 masses of rock, packed close together and lying on unweathered 

 or partly-weathered granite, in the beds of the streams. The 

 core-boulders themselves act as checks for sand coming down with 

 the stream from above, and the interstices bet-ween the lower 

 core-boulders are partly or wholly filled with sand. Sometimes 

 one finds in the jungle a large deposit of core-boulders, and can 

 hear the stream rushing through them far below and out of sight. 

 Chinamen burrow among them to get the tin-ore concentrated 

 behind them by nature, and not uncommonly are killed in the 

 process by upsetting the equilibrium of some tons of hard granite. 



A demonstration of how such accumulations of core-boulders 

 can be formed by torrents was afforded in December 1911 in 

 Ulu Pahang, when a terrific downpour of rain struck the slopes of 

 the granite-mountains on which the Trunk Road from Selangor 

 descends to a little township called Tras. The effect of this down- 

 pour had to be seen to be believed. Acres of jungle-soil, trees, and 

 core-boulders went crashing down the hillsides under the extra 

 weight of water, the Trunk lload in many places disappearing with 

 them. The small streams became roaring torrents, hurling boulders 

 and trees into the main stream far below, and on to portions of the 

 road that had escaped total destruction. Part of Tras, at the foot 

 of the range, was overwhelmed by the masses of coarse sand poured 

 down the mountain-side. 



There is a difference between these torrential accumulations of 

 core-boulders and the glacial clays of Kinta that is readily dis- 

 tinguished. The former have been formed by the sorting action 

 of water : the case of Tras shows this clearly. The boulders were 

 left heaped up in the hills, boulder touching boulder or resting on 

 hard rock. The sand was swept down to Tras, and dumped there 

 on the flat land. The clay went farther still. Ultimately, we 

 may expect, the interstices between the boulders will be filled up, 

 in part at any rate, with sand brought down by the streams when 

 they are low ; but no one who had seen such deposits would be able 

 to conclude that the boulders had been deposited in the sand, or at 

 the same time as the sand. In the glacial clays, on the other hand, 

 there is striking evidence of the absence of the sorting action 

 of water, which alone makes it impossible to regard them as 

 torrential deposits ; and with the exception of one case, at Pedhills 

 (in Kinta), the boulders are not heaped up together or touching 

 one another. 



XVI. Conclusion. 



The object of this paper being to give some idea of the geological 

 history of the Peninsula, I cannot discuss at any length the con- 

 nexion of rocks in the Peninsula with rocks in the Archipelago. 

 Cochin China, the Shan States, and Burma. That there is a close 



