THE BANK OF KOSSOL. 243 



KoBSol, rising to within from two to five or six fathoms below 

 the surface of the water. 



The bank itself is open to the south and south-west, but 

 entirely closed everywhere else. Its outer margin is highest on 

 the east side ; on the west there are enormous blocks of coral- 

 line limestone exactly resembling those that lie on the south- 

 west side of Kriangle. At low ebbs the reef is laid quite' dry, 

 but its western portion is conspicuously lower than the western 

 reef of Kriangle, since, while I was able to cross the foimer at 

 five in the afternoon, I had to "wait till eight o'clock, even at 

 the lowest portion of the Kriangle reef, before I could get 

 across. The general outline of the reef is oval ; it encloses a 

 lagoon, very differently formed from that of Kriangle ; for it is 

 almost entirely filled with masses of living corals which vary 

 remarkably in size and form. Towards the south the lagoon 

 gradually deepens, and the blocks of coral increase in size and in 

 number. Towards the other sides the isolated blocks within 

 the lagoon grow together more and more, till at last they form 

 a mass ; at the same time, they are so much raised that they 

 help to form, on the reef itself, a spot which is laid dry at every 

 ebb-tide. This inner surface of the reef is not level and sandy, 

 as in Kriangle, but quite rough, witk knolls of living and dead 

 coral, cut throu^ on all sides by small channels radiating from 

 the centre. We followed up one of these channels, but it only 

 led as far as the outer border of the reef, and we had to wait 

 there a tolerably long time before the tide rose enough to allow 

 us to cross. The outer slope to the west was very abrupt ; at a 

 distance of from 150 to 200 feet the water was almost black, 

 and much, darker than in the channel between Kossol and 

 Babelthuap. The eastern declivity, on the other hand, was far 

 less steep, as is the case in Kriangle ; for the water at some 

 distance from the reef was still quite light blue, and the corals 

 living at the bottom were clearly distinguishable here and there. 



Thus, as will have been seen, the structure of the reef of 

 Kossol, and that of the reef of Kriangle, are equally adver.se to 

 the theory of their origin by subsidence. With regard to this, 

 one point above all must be brought forward. The reefs are 

 beyond doubt connected, as is proved by the soundings between 



