On the Geology of the Key of Sombrero, W. J. 275 



movement, so slow as to allow coral-growth (if not otherwise 

 checked) to compete with it and succeed in keeping the surface 

 of a reef at or near the sea-level. If it were ever faster than 

 coral-growth, the barrier would have been carried down and 

 submerged, the fresh ocean-waters introduced, and the central 

 area perhaps overlaid with another coral-growth of deep-sea 

 species, of which there is no evidence. 



Two conditions seem to have peculiarly affected the cha- 

 racter of the deposits, — the amount of communication between 

 the soa and the lagoon, and the depth and the vertical range of 

 marine species. Both have evidently co-existed, and their 

 effects are similar, so that the latter may have been wrongly 

 attributed, in the foregoing remarks, to one or the other of these 

 causes. Another source of error is our ignorance of the shape 

 of the lagoon and of the position in it of the present remnant of 

 its deposits. The former may be reasonably identified with 

 the outline of the submarine bank, and it may be concluded 

 that the present remnant lay near the centre of the lagoon, 

 from its position near the centre of the bank, from the basin- 

 form of the upper surface of bed A, and from the apparent 

 possession of a coral-rim by each bed. For the beds must in 

 all cases have sloped upwards and thickened outwards towards 

 the shelving shores of the lagoon, along which, whatever the 

 depth in the centre, a coral ring may always have existed when 

 the inlets were open. The present indication may, however, 

 be delusive, if any peak or other occasion for a shoal existed 

 within the lagoon. 



A low knob of foreign rock apparently did occur, and was the 

 nucleus of the whole formation. It must have disappeared in 

 the denudation of the atoll in the course of the first or second 

 guano-period, as will be shown hereafter. It was probably a 

 part of a mountain peak whose base was buried in the lower 

 beds, and whose summit alone affected the beds within our view. 

 The occurrence of the green clay only on the surfaces of two 

 beds (the formation of each of which appears to have ended by 



