T U T D I L A. 87 



proved too true, respecting the difficulty of seeing the sun and stars. 

 The wind at times was very strong, almost a gale, accompanied by 

 light rain and mist. I was informed that there is a good deal of 

 rain during the year, but seldom such a continuance of it as we 

 experienced. There does not appear to be any particular rainy 

 season, but they are liable to these high winds during the winter 

 months, or from October to March.* I obtained from the pilot a 

 register of the weather from January 1839, till October of the same 

 year, which will show more clearly the state of the climate. (This 

 will be found in Appendix VII.) 



In our explorations, nearly all the villages of this island were 

 visited by some of the officers of the squadron, and from their report 

 they much resemble each other. Those of Fagaitua and Leone, on 

 the southern coast, are the largest, and are more of the Devil's towns 

 than the others. One of their customs is truly savage. They 

 seldom use pork as a food, consequently it is a great rarity with 

 them ; but at intervals of several months the villagers assemble at a 

 feast, at which thirty or forty hogs are killed, when they gormandize 

 on them for four or five days, or as long as the food lasts. The 

 whole is eaten, entrails and all. Fish and taro are the principal food, 

 and large numbers of the natives may be seen fishing off the coast in 

 fine weather. The kind of fish usually caught are mullet. 



There is a large kind of worm which they esteem a great delicacy, 

 and which is eaten with much relish. It is impossible to see them 

 sucking down the entrails of the Biche de Mar, Holithuria, and 

 E china, without disgust. They also eat many of the shell-fish that 

 are found on the shore. 



The temperature found on the top of Matafoa, at the altitude of 

 two thousand three hundred and fifty -nine feet, was at 4 p. M., 69-4°, 

 whilst that on board the ship was 79-5°. 



We made an endeavour here to search the reefs at night for shells, 

 with flambeaux or torchlight, after the manner of the Chain Islanders, 

 by which means it is said that many species of shells are taken, which 

 are never seen by daylight. We cannot vouch for this being the case, 

 our experiment not having succeeded. The leaves of the cocoa-nut 

 were either too green or too wet to burn. If success really attends 

 this method, it is a singular trait in the economy of mollusca, which 

 are generally supposed to be partial to daylight. It was my deter- 



* During eleven days of our stay, the quantity of rain that fell was 4 T 6 6 inches. 



