RIVER OBSTACLES 85 



one at the bow and one at the stern, in each canoe, in 

 addition to the crews of four Javanese soldiers and con- 

 victs. It was soon evident that without the help of the 

 natives we could not possibly have ascended the river. 

 For a mile or two above Obota the water ran like a mill- 

 race in a very narrow channel full of rocks and sunken 

 trees, and it was only by the most skilful poling and, 

 when a chance occurred, by hauling the canoes along 

 a side channel that we were able to proceed. When 

 we returned a few days later, we skimmed in fifteen 

 minutes down the rapids which we had taken more than 

 three hours to ascend. 



Above the rapids the river widened to about forty 

 yards and the strength of the current was proportion- 

 ately less, but in a few miles we met with another 

 difficulty. At a sharp bend of the river the whole 

 channel was blocked by an enormous barrier of huge 

 trunks and limbs of trees piled high upon each other 

 and wedged below into a solid mass. For larger boats 

 this might have meant a delay of many days spent in 

 cutting a channel, but the dug-out canoe is narrow and, 

 if not flexible, it can be squeezed through the most 

 unlikely openings, so that we passed the barrier without 

 the loss of many hours. 



When we started from Obota we had been doubtful 

 whether it was possible that so small a river could 

 possibly come from the mountains ; but a little way 

 above the barrier of logs our doubts were set at rest, 

 when we found that our river was a mere off-shoot 

 from another more than twice its volume, which flowed 

 down to the sea at a village called Periepia. The main 



