632 



NAIURE 



[February 6, 19 13 



DANA'S PROOF OF BAR WIN'S THEORY 

 OF CORAL REEFS. 



T AMES DVVIGHT DANA, born four years to a day 

 J after Darwin, on February 12, 1813, naturalist 

 of the United Slates Exploring Expedition under 

 Wilkes from 1838 to 1842, and afterwards until his 

 death in 1895 professor of geology at Yale University, 

 was for more than half a century a leading figure 

 among American men of science. On the hundredth 

 anniversary of his birth it is fitting to direct atten- 

 tion to the independent proof that he found many 

 years ago for Darwin's theory of coral reefs, a proof 

 that has long been overlooked, although it supplies 

 the most important confirmation for the theory of sub- 

 sidence that has ever been brought forward. 



Darwin most ingeniously invented his theory of 

 subsidence while he was in South America, before 

 he had seen a true coral reef ; he had afterwards 

 only to test the theory by comparing its consequences 

 with the facts that he observed during the voyage 

 of the Beagle across the Pacific and Indian Oceans, 

 and with the records of other explorers which he 

 studied after his return home. The theory bore the 

 test admirably; it was universally regarded as "true" 

 for a generation, although apart from certain cor- 

 relations of coral reefs with areas of recent uplift 

 and with active and extinct volcanoes, which appear 

 to be less assured now than seventy years ago, the 

 theory of subsidence did not gain that increased prob- 

 ability of correctness which comes to a theory from 

 the capacity to explain facts that were unknown or 

 unnbticed when the theory was invented. 



During the last thirty years several new theories of 

 coral reefs have been introduced, and Darwin's theory 

 has been more or less discredited in the minds of 

 some investigators. Murray re-introduced what may 

 be called the theory of outward growth, which Dar- 

 win had considered and adopted for certain special 

 cases in association with subsidence ; but in its new 

 form subsidence was excluded from this theory, and 

 two provisos were added as to the organic upbuilding 

 of submarine banks until they reach the moderate 

 depth at which they may serve as foundations for 

 atolls, and as to the production of lagoons by the 

 removal of the inner part of reefs bv solution (Proc. 

 Roy. Soc, Edin., ix., 1880, 505-518). Agassiz, in his 

 world-wide explorations of coral reefs, emphasised the 

 possible complications in their history ; he pointed out 

 the frequent occurrence of uplifted " coralliferous lime- 

 stones," which might be worn down and dissolved 

 away while fringing reefs grew around them, thus 

 producing barrier reefs and atolls in association with 

 elevation instead of subsidence. At the same time 

 he reintroduced the idea — which Darwin had rejected 

 on good grounds — that reefs could grow on the outer 

 margin of platforms cut by the waves around volcanic 

 islands, thus producing barrier reefs without subsi- 

 dence, elevation, or solution (Bull. Mus. Comp. Zool., 

 xxxiii., 1899; Mem. Mus. Comp. Zool., xxviii., 1903). 

 Wharton went still farther in suggesting that a 

 volcanic island might be worn down to a depth of 

 twenty or twenty-five fathoms by marine agencies, 

 thereby producing a flat submarine bank on which 

 an atoll could afterwards grow up ; thus accounting 

 for atolls as Agassiz had for barrier reefs, without 

 subsidence, elevation, or solution. 



The possibility of producing barrier reefs and atolls 

 by the wearing down of uplifted "coralliferous lime- 

 stones," as suggested by Agassiz, may be regarded 

 as a modification of any theory that will exf)lain 

 barrier reefs and atolls before the uplifts occurred. 

 Darwin recognised at least one instance of wearing 

 down an uplifted reef (" Coral Reefs," 1842, p. 55), and 

 NO. 2258, VOL. 90] 



would certainly have welcomed the larger application 

 of this process, had he known the results of modern 

 exploration. 



The formation of atolls by up-growth from sub- 

 marine banks of proper depth is eminently possible, if 

 the banks can be provided in suflicient number, but 

 possibility is not proof. When subsidence is demon- 

 strated as having taken part in the production of 

 barrier reefs, as will be shown below, its exclusion 

 from this theory of atolls is unreasonable. 



The development of a foundation of atolls by the 

 marine truncation of a volcanic island, as indicated 

 by Wharton, is eminently possible, provided that 

 floating coral larvae do not establish themselves upon 

 it until truncation is complete ; but the ordinary rela- 

 tion of fringing and barrier reefs to their central 

 islands shows that this proviso is inadmissible. The 

 formation of a fringing reef will be begun as soon 

 as a narrow platform is abraded, and such a reef 

 once established, further truncation of the island by 

 wave work is practically stopped. Moreover, the 

 Alexa and other submarine banks described by Whar- 

 ton can be explained by regarding them as submerged 

 atolls quite as well as by regarding them as trun- 

 cated volcanic islands ; hence this theory is not satis- 

 factory. 



The formation of veneering barrier reefs on the 

 outer margin of sea-cut platforms around still-standing 

 islands, an old idea (see footnote in Darwin's "Coral 

 Reefs," 1842, p. 49) recently given prominence by 

 Agassiz, is open to the same difficulty that is fatal to 

 Wharton's theory of truncation. However, if a 

 barrier reef were ever formed in this manner, the 

 central island should rise from the cut-back shore 

 line in a wall of steep cliffs, as Darwin clearly 

 stated, and the broader the platform, the simpler the 

 outline of the cliff-walled island should become. It may 

 be confidently asserted that the central islands of barrier 

 reefs do not possess tliese significant features; hence 

 there is no more reason for accepting this theory 

 to-day tljan when Darwin rejected it. 



The theory of outward growth and solution, advo- 

 cated by Murray for the production of barrier reefs 

 around volcanic islands, without subsidence, or even 

 in areas of slow elevation, involves several conse- 

 quences which, when compared with the facts, con- 

 tradict its verity ; for during the slow outward 

 growth of the reef around a still-standing island, the 

 streams from the mountainous interior must form 

 deltas in the shallow water at their mouths, and by 

 the time the reef has grown far enough outwards 

 to be called a barrier, the delta plains must become 

 more or less confluent laterally, thus forming a low 

 alluvial plain around the original island, as Darwin 

 clearly saw ("Coral Reefs," 1842, pp. 128-130). When 

 such lowlands occur, they indicate a still-stand of 

 the islandj but their prevailing absence suffices to 

 exclude the general application of the postulated still- 

 stand. 



It would thus appear that the theories of outward 

 growth and solution for the production of barrier 

 reefs, of marine truncation for the production of 

 atolls, and of coral veneers on the margin of sea-cut 

 platforms for the origin of barrier reefs, all fail to 

 satisfy the requirements of observation, when they 

 are tested by certain consequences that have not been 

 explicitly stated by their inventors. It remains to be 

 seen whether Darwin's theory of subsidence suffers 

 the same fate when tested in the same manner. 



The accompanying diagram (Fig. i) exhibits three 

 stages in the subsidence of a dissected volcanic island; 

 the first stage show^s a fringing reef, the second a 

 barrier reef, the third an atoll, as indicated by Dar- 

 win's original figures, which are here reproduced in 



