1909.] THEORY OF ATOLL FORMATION. 675 



of the reef are gradually removed by solution of dead coral-rock 

 owing to the action of carbonic acid gas dissolved in sea-water. 



The lagoon is caused by the solution of the calcium carbonate 

 of the coral colonies within its limits, and so the atoll shape is 

 developed. 



The existence of such banks in the open ocean is, of course, 

 undoubted, but the evidence that they are all volcanic seems to 

 be lacking : that they are clothed with Globigerina-ooze is well 

 known from many observations. 



That carbonic acid gas in sea-water can dissolve calcareous 

 skeletons of coral colonies and other marine animals is a well 

 ascertained fact, but that the process forms atoll lagoons is a 

 mere hypothesis. The work of Murray, Irvine, and Ross shows 

 that at greater depths the power of solution of sea-water is greater 

 than at the surface : an explanation is therefore needed for the 

 non-solution of the base of the island bank, while the process 

 proceeds so rapidly at lesser depths as to form deep lagoons on its 

 summit. 



That calcareous matter suspended (but not dissolved) in the 

 sea-water is swej)t in great quantities from lagoon outlets is true, 

 as Sir John Murray observed ; but it is also true that at the 

 inlets it pours into the lagoon in such quantities that in Cocos 

 atoll Dr. H. B. Guj)py (himself an advocate of the solution theory) 

 estimated that 5000 tons of sand and debris were washed in and 

 deposited about the lagoon margins every year. 



The deposition of calcareous matter carried in suspension in the 

 water takes place more rapidly in the lagoon than does its removal. 

 It is true that coconut palms are seen to overhang the waters of 

 the lagoon as though the shore had been dissolved from about their 

 roots : the fallacy of this argument has long been made evident 

 when it is used as a support for the Theory of Subsidence, and is 

 no less evident when urged to support the Theory of Solution. 

 It is an accurate observation that the islets on a reef are 

 commonly situated nearer to the lagoon shore than to the seaward 

 edge, but this is the outcome of their method of making by the 

 waves, and by the outward extension of the reef, and does not 

 necessarily indicate that matter has been removed from within. 



If it be granted that the waters of the lagoon might be specially 

 favourable to the action of the processes of solution : it remains 

 to be explained why the central portions of a reef, twenty fathoms 

 under water, are dissolved more rapidly than are the outer margins. 

 Until this explanation is supplied, basin-shaped reefs or " drowned 

 atolls " do not become any easier to account for. The Solution 

 Theory is urged to account for the formation of lagoons, and to 

 explain how they become "widened and deepened"; but the 

 widening and deepening of lagoons is contrary to experience, for 

 narrowing and shallowing is the common fate of lagoons. 



Granting that solution proceeds rapidly in lagoons, it still must 

 be remembered that calcium carbonate is deposited from solution 

 in large quantities within the lagoon area. Lagoon sandstone 



" 46* 



