434 PERCY SLADEN TRUST EXPEDITION 



(c) The reef is undoubtedly holding its own and may quite probably be gaining 

 by means of the vigorous growth of lithothamnia. This reef, if, as seems probable, it 

 resembles that of Cosmoledo, is really elevated coral rock cut down below low tide mark 

 and now protected from further loss. 



(d) The lagoon is shallow but shows a tendency to form basins which are deeper 

 (6 fathoms, 9 fathoms) than the basin forming on Cosmoledo (3 — 4 fathoms). 



(e) The vegetation consists of coral sand plants. 



In the concluding section I shall show that Farquhar accurately represents the future 

 condition of any of the three atolls visited. 



Providence, which lies 40 miles north of Farquhar, is uninteresting, being a sand 

 cay without evidence of coral rock. A dredging taken to the west of the island in 

 744 fathoms obtained coral rock coated with manganese and nodules of volcanic mud, also 

 similarly coated. These are of great interest as tending to show that the Providence 

 bank is based on a volcanic mound. 



St Pierre, 17 miles west of Providence Bank, apparently resembles Assumption in 

 most respects, being an elevated coral rock island with overhanging cliffs. It differs 

 in containing a dry primary lagoon which however has been elevated to such an extent 

 that it is improbable that St Pierre can ever form an atoll. 



The vegetation rather resembles that of Assumption than that of a sand island, for 

 numerous Pemphis and Euphorbia trees are recorded. 



The present lessees of the island, the Mahe" Syndicate, are exporting guano and, while 

 on the Seychelles, I was shown by Mr Stephens, a member of the firm, an egg obtained in 

 the guano, apparently that of a giant land tortoise. 



Before proceeding it may be well to add a word of warning as to the heights and 

 measurements given. No surveying instruments were taken and estimations of height 

 were made very roughly and indeed could often be only guessed at. Professor Stanley 

 Gardiner records the cliffs of St Pierre as varying up to 30 feet and I feel it quite possible 

 that many of my elevations are underestimated. On the other hand I have every 

 confidence that the various measurements are correct in relation to each other and for 

 purposes of theorisation may be entirely relied on. 



X. THE FOBMATION AND FUTURE OF THE ISLANDS. 



In this, the concluding chapter, I wish to consider from a wider point of view, first, 

 the formation of the Aldabra series of islands, and secondly, their future, basing the 

 various deductions on the processes which can be seen taking place at the present 

 day. A certain amount of repetition is unavoidable, but is perhaps excusable as this 

 section may be regarded as in part a summary of the results of the whole work. 



Before a reef can be built up underneath the sea by the active growth of corals, 

 it is first necessary for there to be some base rising to within at least 40 fathoms of 

 the surface, since reef-building corals, which feed largely by means of symbiotic algae, 

 cannot flourish at greater depths. Of this base in the case of the various members of the 



