122 SECTION A. GENERAL ZOOLOGY. 



to the slope generally found, so far as is known, outside reefs, i.e. 

 a gradual slope to 15 to 50 fathoms and then a steep at an angle 

 of something over 45'. The sections run by the Penguin off 

 Funafuti exhibit considerable variation in the depth at which the 

 steep commences ; and it seems to be probable that the different 

 depths off different sides of the same, or off different reefs, may be 

 explained by the depth of the ocean currents, local currents, etc. 

 From the reef as observed at Rotuma there appeared to be a strong 

 undercurrent seawards, both with the rising and falling tide. This 

 sweeps out the sand and small masses of rock, forming the steep, 

 which simply represents the angle of rest of such detritus under the 

 water. In support of the latter I may notice that the muddy inner 

 reefs of Viti Levu and Vanua Levu have a slope of 40'' to 60° in 

 many places, the slopes too off Falcon Island, Tongan Group, which 

 is merely a heap of cinders and ash, appear from the soundings on 

 the chart to be equally steep. 



The next stage would be the gradual broadening of such a small 

 reef by the fusion to it of buttresses and masses, formed outside, by 

 the direct upward growth of nullipores and corals. The reef, as are 

 all the living reefs which I have seen, would be of a more or less 

 open structure underneath with a rather flat, solid crown. The 

 corals and nullipores on the central part would be killed as the reef 

 broadened by sand, thrown on them, and by exposure to the sun, 

 being at some distance from the breaking edge, and the middle no 

 longer being protected by the living organisms would be slightly 

 hollowed out by solution as are the reef-flats of fringing, atoll and 

 barrier reefs. On the solution continuing the pool might be com- 

 parable to the " barachois " of Diego Garcia, described by Bourne', 

 and the so-called "mangrove swamp" and other pools in the islets of 

 Funafuti atoll. Later by the further broadening and solution an atoll 

 would be formed, the depth of which would be comparable to the 

 number and particularly to the depth of the passages through the reef. 



In support of my contention of such a considerable amount of 

 solution taking place 1 would point to the Lau, or Eastern Group of 

 the Fijian Islands. The limestone here is extremely hard, dense and 

 crystalline, and its cliffs are always very extensively undermined at 

 the base. In the Lau Group there are over 20 islands formed of 

 this rock, the highest of which is 1030 feet. Among them may be 

 found forms with no reef, a fringing reef, or a distant barrier reef, 

 perhaps only surrounding one or two small rocks. Many of the 

 larger seem to be raised atolls, having marked depressions in their 

 centres. Fulanga, the two Ongea and Namuka appear as if the 

 sea has broken through the rims of the former atolls into their 

 lagoons. In Fulanga the solution has gone further, for the island 

 merely consists of an incomplete ring of high land, surrounding a 

 lagoon, 4 — 10 fathoms deep, studded with rocky islets up to lOO feet 



' " The Atoll of Diego Garcia and the Coral Formations of the Indian Ocean," Proc. 

 R. S., vol. XLHI, p. 440 {1888). 



