452 Mr. G. C. Bourne. The Atoll of Diego Garcia [Mar. 22, 



as the Paumotu Islands, seem to lie altogether to one side of the pre- 

 vailing currents. The islands north of Madagascar (Cosmo Ledo, 

 Farquhar groups, &c.) do not lie in the track of the Mozambique 

 current, but to one side of it, and the Chagos group does not lie in 

 any constant current, but is at one season of the year washed by the 

 currents caused by the S.E. trades, at another by the irregular cur- 

 rents caused by the N.W. monsoons. During the latter season there 

 are often long periods of absolute calm during which the currents 

 are merely tidal. The S.E. trades, however, are the dominant winds 

 both in force and frequency, and if the coral growth were dependent 

 chiefly on the supply of food brought by surface currents, those corals 

 growing on the windward side would naturally have the advantage in 

 the food supply. Situated in the direct tract of the current, they 

 would receive an abundant supply of living organisms, and then the 

 impoverished current, sweeping past the sides of the reef, would 

 become poorer and poorer in organic life as it flowed towards the 

 leeward side, till finally on the further shore, the backwash would 

 hardly bring any sufficient supply of food for coral growth. A reef, 

 therefore, would tend to extend in the direction of the current, and 

 the longer diameters of the atolls might be expected in the Chagos 

 groups to lie S.E. and N.W. This is not the case. Diego Garcia and 

 Speaker's Bank lie north and south ; Peros Banhos is nearly square ; 

 the Solomons lie N.E. and N.W. ; the Great Chagos Bank lies east 

 and west. Pitt's Bank does lie S.E. and N.W., but the rim on the 

 northern and north-western side is nearer to the surface by some 

 5 fathoms than it is on the southern and south-eastern side, indicating 

 a more vigorous coral growth on the side turned away from the pre- 

 vailing current. In the case of a submerged bank it is difficult to 

 see why the corals situated to leeward should be better off as regards 

 food supply than those living in the interior of the lagoon, for the 

 superficial parts of the current would flow freely over the windward 

 rim and bring abundant food into the lagoon. But in the Great 

 Chagos Bank the northern rim is, on the average, higher than the 

 south-eastern, and all the islets are placed on the northern and western 

 parts of the rim. A study of the corals growing within the lagoon 

 of Diego Garcia is in this case of considerable interest. If their 

 existence depended upon pelagic life brought to them by currents, it 

 would be expected that the most numerous coral patches would be 

 found at the northern end of the lagoon in the track of the tidal 

 currents. There is a considerable area of active coral growth to the 

 south and west of Middle Islet, but beyond this there is no relation 

 whatever between the luxuriance of the coral patches and the mouth 

 of the lagoon. Corals grow most vigorously along the shore between 

 Minni Minny and East Point, and most vigorously of all at the 

 southern end of the lagoon, where they are most remote from the 



