Chap, ii.] FLYING GURNET HOOKED. 45 



A sheer precipice led down from my feet to the .surf and the 

 sea driven into white crested waves by the trade wind, which 

 was blowing with more than ordinary violence, so that it was 

 difficult to stand on the edge of the cliff. 



I found a chasm in the cliff where it was possible to descend. 

 At about 200 feet from the bottom of the cliff, where the 

 stratified volcanic rock was intersected in all directions by 

 dikes, was a very small spring, from which issued perhaps a 

 quarter of a pint of water in an hour. It was the only natural 

 spring I saw in the islands, although a few others exist. There 

 was green slimy matter round the spring composed of diatoms 

 and other low algse, and a small mass of vegetable mould, in 

 which grew two plants which I had not met with elsewhere in 

 the island, a yellow-flowered crucifer {Sinapidendron Vogelli) 

 and Samolus Valerandi. 



This miniature oasis was only about four feet in circumfer- 

 ence, and absorbed the whole of the water yielded by the 

 scanty spring. A number of wood-lice sheltered in it. I 

 suppose the seeds of these two plants must have been carried 

 to the spring by birds coming to drink. 



On returning to the town down the leeward slopes, I passed 

 the principal wells of the town ; they are dug in a now dry 

 stream bed, and are about 15 feet in diameter, and 25 to 30 

 feet in depth. There was plenty of water in them, but it was 

 slightly brackish, and probably partly derived from the sea. 



The trammel net was set nightly in the harbour by Mr. Cox, 

 the boatswain, and yielded some fine fish ; amongst these were 

 some large flying gurnets, which evidently, from their being 

 caught in the trammel, frequent the bottom a good deal like 

 our wingless gurnets. One was caught with a line at the 

 bottom. I hooked one, however, near the surface, when fish- 

 ing with a rod and trout tackle for small mackerel and silver 

 fish. This was quite a novel experience in fishing. The flying 

 fish darted about like a trout and then took a good long fly in 

 the air, and in an instant was down in the water again and out 

 again into the air, and being beyond my skill in playing with 

 such light tackle, soon shook itself loose and got free. 



A species of Balistes, called the trigger-fish, because it has 

 a stout trigger-like spine on the back and the belly, which can 

 be erected as a defence, was caught in the net. The living 

 fish when held in the hand makes a curious metallic clicking 

 noise by grating its teeth ; similarly Diodoii antentiatus makes 

 a curious noise by the movement of its jaws, as noticed by 

 Darwin.* I have heard the sound in the case of a Diodon 

 hystrix caught at St. Thomas ; it is a sort of grunting sound. 

 * Darwin, "Journal of Researches," p. 14. 



