40 A NATURALIST'S WANDERINGS 



what seemed to me signs of recent elevation. At ebb tide 

 there the water was very shallow and quite warm to the hand, 

 and I noticed Ostrseidse, small Tridacnas and other shells -all 

 dead where they grew, doubtless killed by exposure to the sun at 

 low tide and by the fresh water during heavy rains. Of these 

 tropical downpours, Darwin records one as having taken place 

 before his visit, and Mr. Eoss told me that in 1866, there were 

 several months of such continuous rain that the fresh water 

 stood for several inches on the surface of the lagoon, causing the 

 death of large numbers of fish, and no doubt of corals also. 



Completely surrounding this little islet was a thrown-up 

 beach of very white sand, quite different from that I saw 

 anywhere else on the atoll, composed entirely of the minute 

 shells of molluscs. Echini, and of crabs, with a small proportion 

 of coral debris, probably raised by the waves from the seaward 

 slope of the barrier, indicating, perhaps, a less abrupt descent 

 than has been supposed. Since its first occupation (by Eoss 

 Primus) the lagoon has greatly filled up with coi'al patches and 

 sediment, as he could sail his vessel much 'farther up towards 

 South-east Island than now, and several boat channels cut as 

 indicated on the map have become quite obliterated. On the 

 east side of the atoll the islets are much smaller than at any 

 other part, and this may result if such an untoward circum- 

 stance as the irruption of poisoned water, such as I have 

 recorded above, were to occur at frequent intervals. It is 

 possible also that such a stream might issue frequently, if not 

 in great quantity, without being observed. 



I incline to believe, therefore, that the Keeling reef 

 foundation has arisen as Hurray, Semper and Agassiz have 

 suggested ; but that its islets have been the result of the 

 combined action of storms and the slow elevation of the vol- 

 canically upheaved ocean floor, on which the reef is built.* 



The atoll offers to the marine biologist a rich mine that 

 would take not a few years of working to exhaust ; t to the 



* An abstract of an exhaustive resMTOe and discussion by Dr. A. Geikie, 

 F.E.S., of the Coral Eeef theories will he ionnd in Nature, Nov. 29 and Dee. 6, 

 1883, of which the full text has just been published in the Froc. Phys. Soc. 

 Edin., vol. viii. (1884). 



t I have elsewhere (Proc. R. O. S., March 1884) directed attention to the 

 admirable situation of this spot for a Biological and Meteorological Station, 

 where it could be kept up at the most trifling cost. 



