Sept. 17, 1 8 74 J 



NATURE 



409 



by their temperature, as well as by //;«> unuil wdsl'ivard ditcitioii, 

 have proved an ubstacle to the transfer of mid-ocean species to 

 the Panama coast. " Tor the same reason the transfer of corals 

 — warm-water species — from the West Indies or the Bermudas, 

 eastward, to /(VjAvk Africa, is impossible. The width of the 

 coral reef region on the African side of the Atlantic is only 15°, 

 while it is 48° toward the American coast, and the tropical 

 current is ois/wnn/. 



A proper understanding of the action of the various causes 

 influencing the growth and distribution of polyps and reefs, 

 which have been mentioned in the preceding paragraphs, may 

 leave much less than has been imagined for that " more recondite 

 cause." 



I did not think to include among the causes a too rapid ufruard 

 change of level — on which Mr. Darwin lays much stress. But 

 I recognised the fact that when a rise, like that which has oc- 

 curred at the island of Oahu [putting an extended range of rt ef 

 thirty feet out of water] takes place, nnd so divides the area of 

 reef into an elevated and non-elevated portion, the latter will lie, 

 on this accoiml, narrower than it would liave been had the land 

 been stationary. But the cause does nut appear tome to have 

 very many examples. 



II. The third sentence of the Preface reads thus : — 



" Professor Dana also insists that volcanic action prevents the 

 growth of coral reefs much more effectually than I had supposed ; 

 but how the heat or poisonous exhalations from a volcano can 

 affect the whole circumference of a large island is not clear." 

 And this is followed by the remark: " Nor does this faC, if 

 fully established, falsify try generaliiation that volcanoes in a state 

 of action are not found within the area of subsidence, whilst they 

 are often present within those of elevation." 



In my discusson of this subject I h-ive attributed the destruc- 

 tion here referred to about islands of active, or recently active, 

 volcanoes, not to aerial eruptions, as might be suspected Irom Mr. 

 Darwin's words, but to subniariiie ; and I happen to have said 

 nothing about "exhalations." I have drawn my conclusions 

 especially from four examples (pp. 303, 305, 306) : the island of 

 Hawaii (Sandwich Islands), about which recent eruptions, and 

 partly submarine, have taken place on the east, south-east, south, 

 and west slopes of the island, or through more than hall of its 

 circumference ; Savaii, the largest of the Samoan or Navigator 

 Islands, and the last of the group to become extinct, as its liva 

 streams show ; the eastern half of Maui, who-e great crater 

 must have been recently in action, whi!e the western half bears 

 the fulleit evidence of long extinction ; and the northern ex- 

 tremity of the Ladrones. I state that reefs o'ten occur on 

 favoured parts of even such volcanic is'ands, as they well might 

 if submarine eiuptions were the cause, and I mention examples ; 

 thus agreeing with Mr. Dirwin's criticism that " the exi-tence of 

 reefs, though scantily developed, and, according to Dana, con- 

 fined to one part of Hawaii, shows that recent vrdcanic action 

 does not prevent their growth." My statement about that 

 Hawaiian reef is worded thus : "the only spot of reef seen by 

 us was a submerged pitch off the southern cape of Hilo Bay." 

 Mr. Darwin ciies an observ.ition with regard to the occurrence 

 also of reefs on the northern coast of Hawaii, which accirds 

 precisely with the principle I have laid down, since the northern 

 part of the island is, as I >tate in my Geological Report of the 

 island, that which was earliest extinct, and is oldest in all its 

 features, and therefore that which would not have been reached 

 by the submarine eruptions. 'I'he western peninsula of Maui, 

 or the old part, has its coral reefs, while the eastern, or pirt 

 recently active, has almost none. Savaii, in like manner, has 

 cor.il reefs on its western and northern shores, while elsewhere 

 without them. 



I failed to find evidence in the ra>e of either of these volcanic 

 regions that they are situated within areas of elevation rather 

 than subsidence. Only l,n miles west of Savaii lies the large 

 i^l.ind of Upolu, having very extensive reels— on some parts of 

 the north side three-f mrtlis of a mile wide ; and it has not 

 seemed safe to conclude that, while Upolu thus bears evidence 

 of no movement or of but little subsidence, Savaii was one of 

 elevation ; or that the north and west sides of Savaii have 

 lilTered in change of level Inim the rest of the isUnd. In the 

 island of Maui, having reefs on its old western half, it can hardly 

 be that the eastern peninsula has changed its level quite inde- 

 pendently of the western. In the ne.ar group of the Ladrones 

 the active volcanoes are at the north end ; the islands of the 

 group are very small at that end, without coral reefs, while large 

 at the other, and with broad reefs. One of them, Assumption 

 Island, near which our Expedition passed, is only a small, steep. 



cinder cone, the vent of a sulimergcd volcanic mountain. Such 

 facts afford, therefore, some reason for my statement that the 

 l^adrones appear to have undergone their greatest subsidence at 

 the northern extremity of the range ; and no observations yet 

 made suggest the contrary view. 



The general proposition, that active volcanoes are absent from 

 areas of subsidence, appears to me to need better proof than it 

 has received. As regards the Pacific Ocean, I have found 

 nothing to sustain it. The subsidence of the coral island area 

 of the ocean was one of so vast extent — the breadth 4,000 miles, 

 according to Mr. Darwin — that the sinking could have been no 

 obstacle to the existence and contemporaneous woiking of 

 volcanoes. 



III. The next point in the Preface is a right correction of a 

 raisunderstinding on my part of one of Mr. Darwin's statements. 

 It says : " Professor Dana apparently supposes (p. 320) that 1 

 look at fringing reefs as a proof of the recent elevation of the 

 land, but I have expressly slated that such reefs, as a general 

 rule, indicate that the land has either long remained at the same 

 level, or has been recently elevated. Nevertheless, from upraised 

 recent remains having been found in a large number of cases on 

 coasts which are fnngcd by coral reefs, it appears to me that, of 

 these two alternatives, recent elevation has been much more 

 frequent than a stationary condition. " 



When my work jiasses to a second edition, I shall make the 

 needed correction. 



But I still hold that, while barrier reefs, as Mr. Darwin urges, 

 are proofs of subsidence, small or fringing reefs are in themselve^i 

 no certain evidence of a stationary level, and are often evidence 

 of subsidence, even a greater subsidence than is implied b)' 

 barrier reefs. I have already stated that one cause limiting 

 distribution of reefs is bold shores, a wall of rock of even a 

 hundred and fifty feet producing a complete exclusion. If Tahiii 

 were to subside two thousand feet, it would be an island of 

 precipitous shores all around, and with deep indentations, like 

 the Marquesas, instead of one with broad shore planes. Such 

 bold shores are evidence of su'^isidence ; and as only very small 

 reefs, if any, could find fooling about such an island, the narrow 

 reef would be another consequence of the subsidence, and no 

 evidence of a stationary condition. Again, the gradual sinking 

 of an atoll, like the Gambler group, or of a Tahiti with iis 

 barrier reefs, at a rate. a little fast for the growing corals, would 

 necessarily contract the reef region, reduce the barrier reefs of a 

 Tahiti to narrow flinging reefs ; and make an atoll, however 

 I large, a small atuU with the reef-ljorder narrow and the lagoon 

 perhaps obliterated. An atoll thus reduced to a sand-bank is an 

 example of the effects of subsidence, and affords no evidence of 

 elevation or of a long stationary condition of the region : and 

 the same may be true of a region of narrow fringing reefs. 1 

 landid on two of the small coral islands of the equatorial 

 Pacific which are in just the condition here described ; and my 

 book contains descriptions of others from a good observer — ^J. D. 

 Hague — who resided on them several monihs "for the purpo e 

 of studying the character and formation of the guano deposits." 

 I found the depression of the old lagoon, in one case partly, in 

 the other wholly, dry; and I found also that the living ree's 

 around were n.rrow. Mr. Darwin inclines to regard i-Lands ol 

 this kind as either evidence of no movement, or, of elevation. 

 On the contrary, since the coral islands of the South Pacific 

 diminish in size toward the region of these small islands, an 1 

 since the region just beyond, to the north and north-east, is free 

 from islands, and since all I he features are such as would come 

 to them from a continuation of the coral-island subsidence to its 

 nearly fatal end, I believe still that I was right in considering 

 the ocean bottom in this part to have undergone a general subsi- 

 dence greater than that to the south, south-west, and west, where 

 the atolls and barrier reefs are large. 



Aga-n, if suljmarine eruptions are destructive, narrow reefi 

 may exist about volcanic islands that are undergoing a sub- 

 sidence. Making a reef is slow work ; and, judging from the 

 eruptions of the present century about Hawaii, reefs would have 

 had a pnor chance in the past to form, except along the coasts 

 that were out of reach of the subm.irine action. 



With so many causes fur the existence of narrow or fringing 

 reefs, or of small patches of corals, it is assuredly unsafe to make 

 them, without other corroborating testimony, evidence of a sta- 

 tionary condition of a region, or of an elevating movement rather 

 than a subsiding. 



IV. The next point in the Preface is stated as follows : — 



* His article is contained in the American jfoimuil 0/ Science, 2nd series. 



