During the week subsequent to our departure from Isle des Roches, we anchored successively off the four remaining islets of the group; viz., Etoile, Marie-Louise, Des Neufs, and Boudeuse. They are mere cays, formed of coral and drift sand, and are uninhabited. Owing to the heavy surf which broke all round their shores, we found it unsafe to land.
With our brief visit to the islets just mentioned our survey of the Amirante Group came to an end. I will, therefore, before quitting the subject, make a few general remarks on the group as a whole. The Amirante Group consists altogether of twenty-one low coral islets, resting (with the exception of Isle des Roches, which is on a separate bank) on an extensive coral bank, whose long axis lies in a north-north-east and south-south-west direction, and is eighty-nine miles in length, with an average breadth of nineteen miles. It is included between the limits of 4° 501⁄2′ and 6° 121⁄2′ south latitude, and 53° 45′ and 52° 501⁄2′ east longitude, and is about seven hundred miles distant from the nearest part of the East African coast. Some of the islets and cays of which it is composed, and which are included in the above enumeration, are so grouped into clusters, that for all practical purposes the group may be considered as consisting of nine islets, which have been named African Island, Eagle Island, Darros Island, Poivre Island, Des Roches Island, Etoile Island, Marie-Louise Island, Des Neufs Island, and Boudeuse Island. Of these only three are inhabited; viz., Darros (including the adjoining islet "St. Joseph," which is occupied by part of the same establishment of creoles), Poivre Island, and Isle des Roches; the population consisting of French creoles and negroes imported from Seychelles, who make a livelihood by cultivating cocoa-nuts, and altogether do not exceed forty in number. The islets are all low and flat, are formed entirely of coral and coral-sandstone, and their general surface has an altitude above high-water mark not exceeding fifteen feet, while in the case of African Island, the lowest, it is not more than seven feet. Most of them, however, are conspicuous from a long distance at sea, on account of their possessing clumps and groves of casuarina trees, which tower to heights ranging from eighty to one hundred and eleven feet above the soil, as ascertained by trigonometrical measurement. The casuarinas at Darros Island, which were eighty feet in height, had been planted nineteen years prior to the time of our visit by a Frenchman named Hoyaeux, whom we subsequently met at Providence Island.
All the islets above mentioned possess "fringing reefs," but are distinguished from the coral islets of the South Pacific, and of the other parts of the Indian Ocean, by the entire absence of "barrier reefs." The soundings which we made over the Amirante bank showed a general uniformity in the contour of its surface; whilst at the same time there was abundant evidence that the central portions were more depressed than the margins. Soundings in the latter situation gave a depth ranging from ten to fourteen fathoms, and as each line of soundings reached the central depressed area of the bank, a depth of about thirty fathoms. The islands were for the most part situated near the margin of the bank, and were in every case surrounded by a zone of shallow water. Hence it is obvious that if the entire structure were suddenly to undergo an elevation of about fourteen fathoms, or eighty-four feet, it would present the appearance of an atoll studded with comparatively lofty islets, and enclosing a lagoon of still water sixteen fathoms in depth.
The outer edge of the bank was exceedingly abrupt, for within a ship's length the soundings changed from ten or fourteen fathoms on the margin, to no bottom with one hundred fathoms of line immediately outside the edge. This precipitous character of the reef-edge was found to be the same throughout its entire extent. At various points over this area growing corals were obtained at depths ranging from twelve to twenty-two fathoms, the latter being somewhat greater than the limit in depth at which it is generally agreed that reef-forming corals can live. It therefore follows, that if the entire bank were now to subside bodily at a faster rate than the corals can by their growth raise the surface upwards, these organisms would soon be immersed below their natural limit, and would consequently die. But we have no evidence of a subsidence having occurred, beyond the fact that the bank, as a whole, bears a resemblance to a submerged atoll, while on the other hand there are some positive evidences of elevation to be seen in the overlying islands. At Eagle Island, the general surface—that is to say, all the land within the inner, or coral-drift beach—is level, and consists of dead coral in situ; so that if denuded of its present covering of low vegetable growth, it would present much the same appearance that a broad platform of fringing reef might, if elevated above high-water mark and allowed to remain exposed to the weather for a few years. The same is the case as regards the greater part of one of the African Islands which we visited; while its northern extremity was composed of upraised coral sandstone, standing in situ, and exhibiting excavated grottoes and jagged pinnacles, resulting from old marine degradation. It may therefore be inferred that these two islands have been subjected to a movement of elevation to the extent of at least a few feet at some period subsequent to the formation of their present reef-coral surfaces. Again, at Isle des Roches, which, however, it should be remembered lies on a separate, although adjacent bank, there were along its south-eastern margin stratified beds of hard coral sandstone occupying a position above high-water mark, and presenting to seaward an abrupt eroded face of hard rock which was undergoing degradation, and was being undermined by the action of the waves on a soft subjacent stratum. As regards the other islands of the group, I have seen no evidence of elevation beyond the fact that they are higher than either African or Eagle Island; one of the most southerly being as much as fifteen feet above high-water mark. I may add that the absence of "barrier reefs" throughout the group militates against the probability of subsidence having taken place. There is, therefore, reason to believe that the entire group have undergone elevation rather than subsidence; and if the forces which produced this condition be still in operation, and continue so until a further elevation of fourteen fathoms has been effected, there will result an atoll over eighty miles long by twenty in width, and studded with lofty coral islands, somewhat resembling the high islands of the south-eastern Paumotus, such as Elizabeth Island, which Dana describes as being eighty feet in height.
The Amirante Group furnishes an illustration of the generally accepted position that corals grow more luxuriantly on the weather than on the lee side of banks and reefs. In this region, a wind, varying in direction between east and south-east, prevails throughout ten months of the year, and consequently gives rise to a proportionately constant surface current; and, on looking at the grouping of the islets, we find that of the eight which rest upon the same bank, six are situated on or about its eastern margin, while the remaining two, which are placed on its south-western side, are comparatively insignificant sand-cays. Again, Isle des Roches, which rests on a bank to the eastward of the Amirantes, from which it is separated by a deep-water channel eleven miles wide, is situated on the eastern, or weather margin of its own bank—also a partially-submerged atoll.
CHAPTER XII.
CONCLUSION.
On completing our surveying work at the Amirante Group we steamed back to Port Mahé, Seychelles, in order to replenish our stock of coals. After a stay of a few days we again got under way (17th of April), and shaped a course for Alphonse Island, which occupies an isolated position sixty miles south-west-by-south of the southern extremity of the Amirante bank.
We reached Alphonse Island on the 19th of April, about midday, and saw that in shape, and general appearance, it much resembled one of the Amirantes—for instance, Poivre; but, however, in one important characteristic was different. It possessed a sort of barrier reef little less than a wash, and sufficiently indicated by a long line of heavy breakers. We steamed round the island, holding a course parallel to the line of breakers, and within a few ships' lengths of it, but we got no soundings with fifty fathoms of line. On attaining a position opposite to the southern extremity of the island, we saw a canoe approaching, the occupants of which, an elderly white man and some negroes, soon afterwards boarded us. The information which they gave us confirmed our impression as to there being no anchorage suitable for a large ship anywhere near the island. We learned that it was the property of a Frenchman named Baudon, who resides in Europe, and that the population consisted of twenty-eight, six being whites—viz., John Hickey, the manager, with his wife and children—and the remainder, mulattoes and negroes. The island seemed to us to be covered with cocoa-nut trees, but we were told that only a small number were old enough to bear nuts. The produce consisted of copra, green turtle, hawk's-bill turtle, and pearl-shell. Of the latter, two thousand shells had been exported within the previous two years; and we also learned from Hickey that he then had nine hundred in store awaiting shipment. The shells, which are much smaller than those of Torres Straits, and have a black internal margin like those of Ceylon, are obtained by negro swimming-divers. They are found in the still-water pools, inside the barrier reef, where they lie in four or five fathoms of water; and on account of the danger from sharks they are only sought for in these enclosed pools. Although a good many pearls of small size are met with, the commercial value of the fishery depends on the mother-of-pearl of the shells.