436 PERCY SLADEN TRUST EXPEDITION 



The reefs of Cosmoledo, Farquhar and perhaps St Pierre, when under the sea, 

 presumably resembled that of Aldabra, though it is not assumed that they had all reached 

 the same distance from the surface. Astove differs from the above in that its whole reef 

 seems to have been in a flourishing condition, while Assumption, perhaps because the base 

 on which it is built is elongated and narrow, had no primary lagoon. 



The next stage in the history of our atoll was its "elevation" in relation to the surface 

 of the sea and here again we can only guess at the means by which it took place. The 

 natural method would have been for a slight local elevation of the ocean-floor to have 

 placed all the reefs near the surface out of water, but in this connection it must be 

 pointed out that numerous points throughout the whole south-western Indian Ocean 

 show a simultaneous elevation of about an equal amount ; thus Mauritius has coral rock 

 at 65 feet, the Aldabra series formerly at 50 feet. Zanzibar has been slightly elevated 

 (coral rock at 200 feet) as also the Seychelles (30 feet), while the Maldives were probably 

 elevated to something under 100 feet. I am not aware of any similar elevation being 

 recorded on either Ceylon or Madagascar, but it must be remembered that, owing to 

 its small extent, its effects are so slight that they might easily be removed by nature 

 or neglected by observers. However this may be, there is no doubt that an area 

 approximately between lat. 9° N. — 9° S. (without including Mauritius), and long. 40° E. — 

 75° E., has been "elevated" through at most 100 feet, and it seems most improbable that 

 there should occur an elevation of the earth's crust which should be so very regular even 

 in extent over sixch a large area. The only alternative is a change in the level of the sea 

 and it may be well to point out the possibilities of such a theory in order to obtain, 

 if possible, more information on the subject. It has been shown that the level of the sea 

 at Karachi is 300 feet higher than the level of the sea near Ceylon, this being due to 

 a banking of the water in the north owing to the attraction of the large mass of the 

 Himalayas. Again observations show that there is an even greater difference of level 

 between the centre of the Pacific and that near the South American coast, a phenomenon 

 attributed to the attraction of the large land mass of South America and the Andes. It 

 is possible therefore that during the last great land changes (late miocene) an alteration 

 occurred in the level of the sea, sufficient to elevate the land in the Indian Ocean, though 

 it remains for the geologist to say whether this suggestion is reasonable. 



A further, but to me less probable explanation, of a change in ocean level is that 

 water has been accumulating as ice in Antarctic regions for a long period and that this 

 is sufficient to account for a considerable drop. Against such an explanation applying 

 to the Indian Ocean islands I would urge first that the effect so produced would be 

 world-wide, and secondly that such a movement would have been too slow to have formed 

 such an island as Aldabra which would have had to be cut down by the sea more quickly 

 than it was being raised in relation to the sea level. 



After the elevation of our reef and the formation of an island a series of retrograde 

 processes commences and the senescence of the reef, if such an expression is permissible, 

 dates from this period. 



At first Aldabra was not a true atoll, for the primary lagoon was elevated to such an 

 extent that it became dry, a statement which is also true of St Pierre and probably of 

 Cosmoledo and Farquhar. Astove is different, for the lagoon of this atoll is probably 



