♦ KNOW^LEDGE ♦ 



[November 1, 1888. 



aiul the t;rowth, rising upwards, readied the surface and 

 formed a^fringing reef. The land subsiding, the coral con- 

 tinued its upward growth, so that the central island was 

 reduced in si/e, the intervening channel deepened, and a 

 barrier reef formed, which by continued subsidence became 

 an atoll, inclosing a lagoon, without an island. The follow- 

 insr diagram will make this clear. 



In the case of submerged reefs, Darwin assumed that the 

 subsidence had been so rapid that the coral had been carried 

 into deep water and killed, before it had time to extend its 

 growth in an upward direction. 



Darwin's theory of subsidence was arrived at after an 

 e.xhaustive study of the facts at his disposal ; it was sup- 

 ported by a m;iss of evidence, and put before the world with 

 admirable clearness and force. It was at the same time so 

 simple and satisfactory, explaining at the same time the 

 various forms of reefs and the great depth of water surround- 

 ing them, that it is not surprising that it met with general 

 acceptance. But beautifully simple and complete as the 

 theory is, it presents some serious difficulties. First among 

 these, it involves the assumption that the whole Central 

 Pacific, besides other large areas, has undergone enormous 

 subsidence. Darwin, of course, fully recognised this — indeed, 

 he was led to the belief that a Pacific continent had been sub- 

 merged so completely that only a few of its mountain 

 summits appear above the surface. But most recent deep- 

 sea researches seem to prove the great antiquity and 

 permanence of the large ocean basins, and this evidence 

 must all be put aside if the Darwinian theory is to be 

 accepted in its entirety. 



Besides this general objection to the subsidence theory, 

 many instances have been brought forward of undoubted 

 elevation of the land in coral regions. Piein has shown that 

 the Bermuda reefs have undergone elevation. Alex. Aga,ssiz 

 mentions* other elevated reefs in the West Indies, and 

 with regard to Florida he says : — " All naturalists who have 



* " Mem. American Acad.," vol. xi. 



visited the reefs have felt the difficulty of .applying Darwin's 

 theory to the peculiar conditions existing along the sti-aits." 

 Semper in the Pelew Islands experienced the same difficulty, 

 owing to the clear evidence of upheaval, and Dr. H. I?. 

 Guppy says of the Solomon Islands * : — " These upraised 

 reef masses, whether atoll, barrier reef, or fringing reef, 

 were formed in a region of elevation. jNIr. Darwin's theory, 

 which ascribes atolls and barrier reefs to a movement of subsid- 

 ence, cannot be applied to the islands of the Solomon group." 

 More recently it has been -shown by Mr. G. ('. Bourne t 

 that there is evidence of upheaval in some of the atolls of 

 the Chagos gi'oup in the Indian Ocean. 



Professor J. D. Dana, iu Jiis elaborate review of the 

 whole question,! con.siders some of the.'e instances of eleva- 

 tion, but does not think they contiict with Darwin's theory, 

 which, while insisting upon a general subsidence, would 

 admit of partial and loaxl elevations. 



Ag.ain, it is a difficulty on the theory that fringing reefs, 

 barrier reefs, and atolls are merely diffisrent stages of the 

 same formation, that all three kinds not unfrequently occur 

 in close proximity, as in the Fiji Islands. On the hypo- 

 thesis of general and very extensive submergence, we should 

 expect that all the formations in a given area would be 

 reduced more or less to the same condition, as in the 

 I Indian Ocean, where only .atolls occur. 



Besides this, .all oceanic islands (other than coral islands), 

 I with scai'cely an exception, are of volcanic origin, and 

 volcanic regions are generally areas of elevation rather than 

 of <lepression. 



Moreover, Dr. Guppy found in the Solomon Islands the 

 upraised reefs to be of moderate thickness only — their ver- 

 tical measurement not exceeding the usual limit of the 

 reef-building zone — while, according to Darwin's view, the 

 vertical thickness of a barrier reef or atoll should be 

 enormous. 



But if it is shown that Darwin's theory is at any rate 

 not of universal application, what other view can be 

 adopted 1 One of the most complete alternative theories 

 was brought forward § about eight years ago by Mr. Jolin 

 Murray, one of the naturalists of the Challenger expedition. 

 He thus summarises his views : ( 1 ) " When coral planta- 

 tions build upon submarine banks, they assume an atoll 

 form, owing to the more abundant svipply of food to the 

 outer margins, and the removal of dead coral rock from the 

 interior portions by currents and by the action of the 

 carbonic acid dissolved in sea water. (2) That barrier reefs 

 have been built out from the shore on a foundation of 

 volcanic debris, or a talus of coral blocks, coral sediment, or 

 pelagic shells, and the lagoon channel is formed in the same 

 way as .a lagoon." 



Mr. Murray also called attention to the immense number 

 of other organisms (calcareous algas, foraminifera, jiteropods, 

 itc.) existing in tropical seas, which secrete carbon.ate of 

 lime, and whose remains are found in great abundance on 

 submarine banks, where they accumulate too rapidly for the 

 .sea water to have much effect in dissolving them. Tiiese 

 deposits form banks upon which flourish numerous species of 

 foraminifera, sponges, deep-sea corals, ifec, and thus rise to 

 a level at which reef building corals can live. Alexander 

 Agassiz has also described the gretvt submarine banks in the 

 Gulf of Mexico, largely composed of the remains of minute 

 marine organisms, the form and direction of the banks 

 being determined by the prevailing winds and currents. 



Mr. Murrav shows that submarine volcanoes have been 

 found to be far more numerous than was supposed by 



* " The Solomon Islands." London: 1887. 



f " Proc Roy. See," vol. xliii. 



j " American Journal of Science." Third series, Vol. xxx, 



sj " Proc. Roy. Soc. Edin." Vol. x. 



