1891.] D. Prain— r//e Vegetation of the Coco Group. 289 



camp ; the effects are &o like what must result if land were slowly rising 

 that it is only after careful examination of all the conditions that one's 

 mind becomes disabused of this specious impression. There is no 

 direct evidence that the land is rising and, as will he evident on con- 

 sidering what has been said above, thei'e is no necessity for supposing 

 that it is. But though this is a very common type of bay, it is not 

 the only type. On Great Coco, in some cases, and on Little Coco very 

 generally a different stage may be observed. The shallow pools described 

 as existing between the surf-built embankment at the margin of the 

 fi^nging-reef and the beach, have in them many living corals that raise 

 great rings which rise to almost the surface of the water in the pool 

 at low-tide and, like huge lichens, grow peripherally till they meet and 

 coalesce. The surf, too, breaks off pieces of greater or smaller size 

 which are lodged in the pool behind, and by-and-bye become more or 

 less cemented together. In this way the whole of a pool becomes in time 

 completely filled up with growing coral and cemented blocks, and there 

 are many reefs, especially on Little Coco, that are completely uncovered 

 at low-tide, while small patches of similar reef are here and there seen 

 that ordinarily the high-tides do not cover. The uniformity that the 

 surfaces of some of these exposed reefs display is very striking. They 

 are almost as even as a paved floor and are as bare and destitute of 

 marine vegetation as they are of living coral. The edge of such a reef, 

 in place of being a fairly continuous embankment higher than the floor 

 of the bay behind, is now broken into hundreds of jagged gulleys through 

 which the wave -wash from the almost level platform tears its way back 

 to the deep water beyond the fringing-reef. The main interest of this 

 stage of the reef is less, however, from bhe present point of view, its 

 actual physical condition than its effect on the vegetation of the shore. 



Behind a coral bay like one of those first described, and which charac- 

 terises a less advanced stage of the history of the fringing-reef, has gone 

 on a long and steady growth of land, with some shingle in it doubt- 

 less, especially as one approaches the nearest ridge, but chiefly composed 

 of coral sand with a thin coating of humus derived from the vegetation 

 it has supported. The main force of the surf has for long been spent on 

 the outer embankment, and the force of the waves that at high- water 

 passed over its top has been so much diminished ere these reached 

 the beach that there they did not act destructively. Now all this is 

 altered. At low-tide the force of the surf is still all expended on the 

 edge of the reef, but as soon as the water has risen so high that the edge 

 of the reef is covered, this foi'ce instead of being dissipated in the deeper 

 water of a pool is accentuated as the breakers roll landward across a reef 

 on which the water shallows slightly as the shore is approached ; by the 



