FRYER— FORMATION OF ALDABRA, ETC. 



429 



on the lagoon side is a small mangrove swamp, which appeared to be decreasing in size 

 owing to insufficient protection from the waves. It is perhaps the last remains of a 

 mangrove swamp, which once extended completely round the lagoon. 



Wizard Island is sandy, and coral rock hardly occurs on the surface, though it 

 can be found by digging. Beach sand-stone is common round the shore. It appears 

 likely that the island will soon be divided into two. Of all islands in this region Wizard 

 was most like such a sand cay as Bird Island, Seychelles*, for it seems to have lost most 

 of the peculiarities of such rocky atolls as Aldabra or Astove. 



The reef was only visited at a point south-east of Moustique Island where it 

 measured half a mile from lagoon to open sea. The seaward edge is conspicuous for 



P M ^r- ^'*2- ''-'^ 





U. U. Trm?y-* 







/ 



Sou* 



COSM©IiEBO GROUP 



"(.-' I.)-- L*i ■'' li »S Luoc *-«lt'E 



Fig. 2. From the Admiralty Chart of Cosmoledo. 



vigorous massive lithothamnia, but no live coral was noted. Inside the edge lithothamnia 

 is still abundant but consists chiefly of encrusting forms found on broken pieces of 

 coral thrown up on the reef. Sloping towards the lagoon is a sort of buttress-zone 

 with channels and pools containing some live coral, the whole resembling the seaward 

 edge of an atoll reef. The lagoon at this point is largely choked with sand, and partially 

 dry at low tide. In places are large beds of Cymodocea. There was no sign of any large 

 blocks (negro-heads) thrown on the reef edge. 



» Trans. Linn. Soc, Ser. 2, Zool. Vol. xiv., pp. 15—17, 1911. 

 SECOND SERIES— ZOOLOGY, VOL. XIV. 55 



