Ch. YI] EFFECTS OF AQUEOTJS ACTION. 89 



amongst them. In each boat stood a man at the bow, hold- 

 ing a bundle of small bamboos, which blazed so brightly that 

 I was hardly convinced that they were not dipped in oil, 

 untn an examination proved them to be dry. This flaming 

 torch, with which about three boats out of four were pro- 

 vided, served to attract the fish, and when the boat was thus 

 smTounded with fish, a signal was made to another boat un- 

 provided with a light, which coming up, drew a net around 

 the iUumiaated boat, and thus secured the shoal. This was 

 done with great shouts and noise, which we often heard, 

 lasting far into the night. I boarded one of these boats, 

 and saw them haul the net ; but the produce seemed to be 

 entirely confined to a small white fish hke whitebait. The 

 fishermen were civil enough, although we rowed right in 

 among their operations; but they did not appear at all 

 anxious to dispose of the fish, which were sweet-tasted, 

 and if cooked at Blackwall would probably rival the real, 

 whitebait. 



The effects of aqueous action upon the sandstone rocks are 

 very conspicuous in some parts of Ke-lung harbour. Near 

 the cave before-mentioned, and immediately upon the verge 

 of high water, is a tall isolated sandstone rock, hating pre- 

 cisely the appearance of an old ruined castle, and appro- 

 priately named Euia Rock, which forms an excellent land- 

 mark by which to anchor a ship. The harder layers of 

 sandstone having defied the effects of weather and the spray 

 which is dashed up during the nofth-east monsoon to which 

 the harbour is exposed, the softer portions have at the 

 same time been more or less excavated, leaving a mimic 

 resemblance of the ruined chambers of a three-storied 

 building. But the most curious and extensive effects of 

 the direct action of the sea are to be found at the entrance 



