28(5 D. Prain — The Vegetation of the Coco Qroup. [No. 4, 



opposed to their crests, the under-jungle is not so dense, larg'ely owing to 

 the mass o£ creepers being carried up to the tall trees above, while on 

 the neck of land that connects the main island with the peninsula at the 

 north-east corner, and on the narrow, more level tongue that forms the 

 south end of the main island and sfci-etches towards Jerry Island, the 

 jungle is rather opener and more penetrable than elsewhere. The 

 height of the outlying north-eastern peninsula which probably at one 

 time has been a separate islet, at least at high-tide, is 80 feet ; the 

 highest ridge of the main island has an elevation of 300 feet ; the other 

 ridges mostly reach from 150 — 200 feet. Jerry Island, the chief outlying 

 islet off Great Coco, consists at the southern end of a low ridge 60 feet 

 high with a vegetation quite like that of the ridges on the main island ; 

 the northern half, however, consists of a level spit stretching towards 

 the main island ; this spit is composed of coral-shingle, and though 

 covered with coco-nut and other trees there is only a sparse undergrowth 

 within its Pandanus belt. Between Jerry and the main island extend 

 wide sandstone reefs on which the waves and currents have thrown up 

 a small eyot of sand, coral-shingle, dead shells, and drift timber on which 

 stranded fruits and seeds are germinating. The other outlying islets call 

 for no remark ; all of them look like detached continuations of particular 

 ridges and most of them have the vegetation characteristic of these. 

 During the first visit daily excursions were made into the island to- 

 wards the northern and eastern parts ; the jungle was found to be so 

 dense on the ridges and the level ground so difficult owing to the 

 ramifications of a considerable creek, which, with its concomitant man- 

 grove swamps, finds an outlet into the northern bay, that it was only 

 on one occasion that the western coast was reached. It was impossible 

 to do anything like jiistice to the interior ; still, the northern and north- 

 eastern peninsulas, the northern half of the east coast, the north coast 

 and about two miles of the west coast at the north end were fairly 

 thoroughly examined. During the second visit, profiting by the experi- 

 ence of the former season, fewer attempts were made to force a passage 

 in a straight line through all obstacles, and the edges of ridges — jugci 

 ipsissima — were in particular carefully avoided. The compass was 

 discarded, no particular objective in the shape of a hill visible from 

 the sea-shore was permitted to occupy the attention exclusively ; the 

 easiest rather than the shortest road was chosen as the route to be 

 followed. In this way the island was crossed in four or five different 

 places, all, however, towards the southern end ; the west coast was 

 explored for about four miles and the east coast examined northwards 

 as far as the point reached when working in the opposite direction in 

 the previous year. The outlying islet, Jerry, was also examined fairly 

 thoroughly and its coasts skirted. 



