38 CUBIOUS SCREEN, 



appeared to have been placed there singly at differ- 

 ent periods to commemorate the death of the heads 

 of families of the tribe. We saw another of these 

 curious funeral screens, — like the first one it was 

 situated in a httle glade in the forest, but unhke it 

 the front was covered or thatched with cocoa-nut 

 leaves, and it had a sm all door-hke opening in the centre. 



The natives must have left the island either on 

 account of its being now the turthng season, or else 

 from the want of water. A small deep well behind 

 the village, apparently the only one in the place, 

 was almost entirely dried up. From the old man I 

 procured the names of some of the neighbouring 

 islands, and also a few other Kulkalega words which 

 are so similar to those of the Kowrarega language as 

 to corroborate Gi'om's assertion that both have many 

 words in common. By way of illustration I may give 

 a few examples. Thus muto, small bird ; hudulugj 

 dove ; geinow, pigeon ', kalfdr, egg ; hurda, grass ; 

 wardba, cocoa-nut j moda, enclosure round the huts. 



At one place I saw indications of an upheaval of 

 the northern side of the island in a bed of coral con- 

 glomerate six feet thick, with its raised wall-like 

 edge towards the hiU as if tilted up, and the re- 

 mainder sloping down towards the sea. A similar 

 appearance on a small scale exists on most of the 

 coral islands which I have visited, but I had not 

 before seen these sloping beds above the influence of 

 the salt water, or at least beyond reach of the spray, 

 still less supporting luxuriant vegetation, consisting 



