224 THE BERMUDA ISLANDS. 



of an atoll." It is, indeed, hard to class the logic of this argu- 

 ment. Is it to be presumed that because some geologists have 

 exaggerated notions of the configuration of an atoll and its in- 

 closed lagoon, that the lagoon does not exist? And if it really 

 does exist, how near are we brought to an understanding of 

 its structure by the simple conception of its being a thin film 

 of water perched upon the summit of a flat mountain-top ? 

 The depth of the lagoon still remains the same, and so does 

 the height of the raised border. It might, indeed, as well be 

 urged that there is no necessity to account for the structure of 

 mountain-chains, since the highest of them are mere wrinkles 

 on the earth's surface, corresponding in size to the irregulari- 

 ties on the rind of an orange ! Comparisons are useful, but 

 they are not explanations. 



Johannes Walther. "Die Korallenriffe der Sinaihalbinsel." Abhaiull. d. mathem. 

 physisch. Classe der Konigl. Sachs. Gesellsch. der Wissenschaften, xiv, 1888. 



No new facts tending toward the solution of the coral-reef 

 problem are given in this valuable memoir. The coral-struct- 

 ures described are mainly fringing-reefs, and they occur in a 

 region of existing or recent elevation. The author calls atten- 

 tion to certain atoll-like or ring-formed islets which are found 

 associated with the fringing-reefs, and which in some instances 

 are immediate outgrowths from the latter. This circumstance 

 is immediately seized by Mr. Bourne (? G. C. B. Review of 

 Walther's work in Nature, Dec. 20, 1888) as another instance 

 " added to the many now accumulating of barrier reefs and 

 atolls being formed in an area of elevation." Walther gives 

 no data regarding these islets, except as to form. It seems 

 proper to ask in this connection : Are all circular or crescen- 

 tic growths of coral to be classed as atolls ? Are fringing-reefs 

 which for a short distance leave the coast-line to be classed as 

 barrier-reefs? Are the dunes and sand-hills of the continental 

 areas mountains? 



As touching the question of the formation of these islets in 

 an "area of elevation " it is interesting to note that Walther 

 recognizes a local subsidence of some 6 metres at the southern 



