1836. KEELING ISLANDS TIDE-GUAGE. 620 



head, though the horizon was clear) — when a number of gan- 

 nets flew past the ship towards the west. We steered directly 

 after them, and early next morning (after making but little 

 way during a fine niglit) saw the Keehngs right ahead, about 

 sixteen miles distant. 



A long but broken line of cocoa-palm trees, and a heavy 

 surf breaking upon a low white beach, nowhere rising many 

 feet above the foaming water, was all we could discern till 

 within five miles of the larger Keeling, (there are two distinct 

 groups) and then we made out a number of low islets, no- 

 where more than thirty feet above the sea, covered with 

 palm-trees, and encircling a large shallow lagoon. 



We picked our way into Port Refuge (the only harbour), 

 passing cautiously between patches of coral rock, clearly visible 

 to an eye at the mast-head, and anchored in a safe, though not 

 the best berth. An Englishman (Mr. Leisk) came on board, 

 and, guided by him, we moved into a small but secure cove 

 close to Direction Island. 



Many reasons had induced me to select this group of coral 

 islets for such an examination as our time and means would 

 admit of; and, as the tides were to be an object of especial 

 attention in a spot so favourably situated for observing them, a 

 tide-guage was immediately placed. Its construction was then 

 new, and, being found to answer, I will describe it briefly. 

 Two poles were fixed upright, one on shore (above high water 

 mark, and sheltered from wind), the other in the sea beyond 

 the surf at low water. A block was fastened to the top of each 

 pole, and a piece of well-stretched log-line 'rove' through them.* 

 One end of the line was attached to a board that floated on 

 the water ; the other suspended a leaden weight, which tra- 

 versed up and down the pole, on shore, as the float fell or rose 

 with the tide. Simple as this contrivance was, and useful as we 

 should have found it in many places where the surf or swell 

 made it difilcult to measure tides at night, without using a boat, 

 I never thought of it till after we left King George Sound. 



* A very small metal chain would be better, because a line, however 

 stretched, will shrink after being wetted by rain, and give out again as 

 it dries. 



