430 PERCY SLADEN TRUST EXPEDITION 



The lagoon as a whole is not deep (2 — 3 fathoms) ; the bottom is sandy as far as can 

 be seen and living coral colonies are scarce. There are two true passes through the reef, 

 both to the south of the atoll, and there seems to be a tendency to form a deep basin in 

 the south of the lagoon between the passes. 



As Cosmoledo is important theoretically a summary of its chief structural points may 

 be given : 



(1) Cosmoledo is an elevated atoll. 



(2) The land-rim was once almost perfect but has been broken up by erosion : the 

 rock islands left are in process of disappearance from the same cause. 



(3) The reef dividing the islands is in just such a condition as should be required 

 for the formation of land by wave-piling : there is, however, no sign of this. 



(4) Great quantities of sand are everywhere present, and much is being piled up on 

 the islands. 



There is no evidence as to the height of elevation of the atoll ; the summit of 

 the rock is now about 10 feet above high spring tides, and the elevation cannot have 

 been less than 20 feet, and was probably more. If Aldabra and Cosmoledo were elevated 

 simultaneously, the interior of the latter atoll must have been much more cavernous 

 to account for the more rapid disintegration of land. 



Of the flora and fauna little can be said, as the few days of my visit were all occupied 

 in examining the formation of the atoll. The sandy islands, e.g. Wizard and Polyte, 

 are clothed solely with coral-sand plants, the type of vegetation such as was found on 

 Aldabra being quite absent. Menai Island, where sandy, resembled Wizard, but in 

 the centre is a small piece of almost typical Pemphis bush, a form of vegetation which 

 apparently clothed South Island on which I was unable to land. The North-East Islands 

 had been too much disturbed by guano digging to determine their original state, but 

 Goelette was thickly covered with Plumbago a/phylla, a herb which produces leaves at 

 the beginning of the wet season but like the broom (Spartium scoparium) speedily loses 

 them, and for the majority of the year assimilates by means of the chlorophyl in its stems. 

 The fauna seemed rather poor, though a lizard, Zonosaurus madagascarensis, found 

 on the North-East Islands was not observed elsewhere in the Aldabra region *. A rail 

 (Dryolimnas abbotti ?) still exists on South Island, and a Cinnyris perhaps forms a local 

 race, but land birds were scarce on Cosmoledo, which as a whole seemed too broken into 

 small islands to be suitable for a land fauna. 



* The skink Ablepharus boutonii var. peronii and the gecko Hemidactylus gardineri were also obtained. 

 Vide G. A. Boulenger, Report XVII. of this Volume. 



