264 On the Geology of the Key of Sombrero, W. I. 



ticularly described under D. Its fine comminution denotes 

 either its formation by the lagoon-wavelets or the distance of 

 the present area from its source. Its chemical composition, 

 unprecedented in connexion with a coral formation, signifies its 

 derivation from some foreign rock, not calcareous, such as occurs 

 in neighboring islands. Its general superposition above the 

 bedding-line and little intermixture with its material in the 

 veins denote the greater hardness of its source. 



It is plain besides, from the identity of their material with 

 that of the upper layer, from their angular shape, and from the 

 entire absence of any coarse sand, shells, or beach-refuse inter- 

 mixed, that the pebbles scattered over A were not thrown over 

 the barrier reef. All the evidence, therefore, tends to show 

 that a peak of more ancient material, perhaps volcanic, existed 

 within or in some connexion with the atoll ; that quantities of 

 its fine debris were formed ; that, at a certain period of the sub- 

 sidence, the ground-swells forced their way in through passages 

 on the leeward side of the barrier (perhaps perforated by the 

 widening of the fissures), tore up portions of the crust upon the 

 bottom and of the surface of the bed beneath (wherever it was 

 already cracked up into fragments and loosely cemented with 

 the lagoon-mud), swept along these fragments and the masses of 

 green clay around the peak, and distributed them in irregular 

 patches over the bottom towards the E. and S.E. sides of the 

 lagoon. 



At least at times the lagoon may have been brackish, but as 

 the subsidence continued the pure and ever-changing ocean- 

 waters again found entrance. Instead of the impure dark- 

 colored lagoon mud, snow-white limestone was again formed by 

 the growth of coral reefs, especially near the shores. Again 

 the barrier-passages were gradually closed and deposits of shells 

 and sand successively formed. If the compactness of the last 

 deposits be the proper criterion, the depth reached by this sub- 

 sidence was not so great as by the former. 



