DESCRIPTION OF THE EXPEDITION. 51 
reach the surface. In their slopes they, as well as the encircling reef, generally arise 
with perpendicular walls for the last 2 to 4 fathoms. They are far more numerous and 
occupy a greater area of the bottom in the shallower waters away from the passage, but 
at the same time the individual shoals are much larger near the passage. 
Judging from the chart there appeared to be on the southern side of the lagoon a 
peculiarly ragged reef-edge. A series of horns extended outwards from its edge into the 
lagoon, bending towards each other so as to enclose areas of the latter. If, on the one 
hand, these were in process of growth, the area of the lagoon would have been reduced 
considerably in size, and if the process continues the lagoon would ultimately disappear 
and leave a flat reef. On the other hand, the more or less enclosed areas might represent 
pits or holes in the reef, velw as they are termed in the Maldives, their inner walls 
breaking down so as to throw them into communication with the lagoon. In this case 
the lagoon would obviously be increasing everywhere at the expense of its encircling 
reef, and it might be supposed that the whole atoll would be growing outwards as a fairy 
ring. While fully aware that our chart was not sufficiently accurate for us to base any 
comparison on a firm foundation, yet we considered the question one of sufficient im- 
portance to claim our special attention. We may now say at once that we found traces 
of every horn shown in the chart represented by shoals, some small and some of them 
large. All the narrower northern extensions shown on the reef between Iles Sudest and 
Lubine were seen to be cut off from the encircling reef proper by channels of at least 
2 to 8 fathoms in depth, the inner edge of the reef running fairly evenly and parallel 
with its seaward edge. To the north of Carré Patte a small pool 50 yards long and 1 to 
2 fathoms deep was enclosed within the reef-edge, but otherwise the whole reef-surface 
was flat with no trace of any velu. 
The above observations on the lagoon all point to the conclusion that it is increasing 
at the expense of its encircling reefs. If this be the case it must be due to the solution 
of the lime by the seawater and to the removal of fine particles of coral and sand held in 
suspension in the water. Seventy years ago Moresby (Joc. cit. p. 64) referred to this 
outwash of mud at Egmont :—* The current and tides sometimes washes the looses and 
off the reef to leeward and discolours the blue water for half-a-mile, which we took to be 
shoal water fit for anchorage until we found we were deceived.” We ourselves noticed 
it in the heavy weather experienced both when we arrived at and left Egmont, as, indeed, 
we have observed it off many other coral-islands. Its effect also was quite clear in the 
passage into Egmont, in that the bottom was largely formed of bare coral-stone, living 
coral being relatively scarce. 
Yet with all this apparent decay and removal of material from the lagoon at Egmont, 
we found in it a type of lagoon-shoal altogether new to us. For while the encircling 
reef had the same general character as that of Salomon, we were surprised to find en- 
crusting Lithothamnia, the most important constituent of its outer edge, to be scattered 
everywhere except just near the islands. While some shoals were covered by corals and 
sand, others showed a surface formed entirely by this group of plants. They were found 
particularly along the north side, being most marked perhaps exactly opposite the 
passage. One shoal that we visited in this position was about 60 yards across, with a 
S* 
