06 Professor Horsford on the 



ment yields at once to the exhaling carbonate of ammonia the 

 framework of stone. 



With this view, there is no difficulty in finding a supply of 

 carbonate of lime for the vast masses of coral. The sulphate 

 of lime, decomposed to furnish the carbonate, is perpetually 

 renewed through rivers from the continents and islands. 



The following inferences are legitimately deducible from 

 this view : — 



1st, Corals would soon die in bodies of salt water wholly 

 cut off from the ocean. 



2d, They might flourish to some extent in waters accessible 

 to the sea only at high tide, 



In Dana's Report on Coral Reefs and Islands,* he states 

 that "-where there is an open channel, or the tides gain 

 access over a barrier reef, corals continue to grow, &c. At 

 Henuake the sea is shut out except at high water, and there 

 were consequently but few species of corals, &c. At Ahii 

 there was a small entrance to the lagoon ; and though com- 

 paratively shallow, corals were growing over a large portion."! 



These facts seem to me to give some support to the view 

 expressed above. 



It was of interest to ascertain, in the case of corals, 

 whether the formation of new coral without was attended 

 with absorption or partial solution in the interior, and a cor- 

 responding reduction of its specific gravity. Specimens of 

 coral, from the centre, periphery, and midway between, of a 



* Am. Jour. Science, [2] xii., 34 to 41, and Geol. Report Expl. Exp., p. 63. 



f In my article, as published in the Proceedings of the Association, I have 

 further quoted from Professor Dana's papers in support of other inferences de- 

 duced from the foregoing view. I have since learned from the author that I 

 had misconceived the sense in which the quotations were to be understood, and 

 have become satisfied, especially after examination of the map of the Feejee 

 Islands accompanying Professor Dana's last article, that the inference, that 

 fresh-water streams, by their supply of sulphate of lime, exerted any consider- 

 able influence upon coral formations, is not sustained. The sulphate of lime 

 of sea-water, however, being one-sixteenth of the chloride of sodium, is abundant 

 for the supply of the carbonate of lime, without the aid to be dorived from such 

 :i source. 



