84 A'. W. Skeats—The Coral-reef Problem 



accumulation of talus will he slow and whether or no corals 

 can grow from such a tains and reach the surface will depend 

 on the submarine slope of the island, on the relation between 

 the rates of supply of talus material, of subsidence, and of up- 

 ward growth of coral. Any considerable accumulation of sub- 

 marine coral talus necessarily involves extensive abrasion of 

 the growing reef by wave action and considered by itself will 

 cause a centripetal shift of the position of the outer reef-face 

 at the surface. But here the second and third factors men- 

 tioned above, which involve the relative rates of abrasion 'of 

 the reef face and repair by regrowth of corals, become impor- 

 tant. Von Lendenfeld's and Daly's diagrams can only express 

 the facts in the case of a reef in which the rate of subsidence 

 is much slower than that of upward growth of coral, lateral 

 abrasion of the reef-face and its repair are rapid and a large 

 quantity of submarine coral talus accumulates at depths less 

 than 15 to 40 fathoms giving a foundation for the seaward and 

 upward growth of coral. In the writer's view the combination 

 of these factors in the history of an atoll is likely to be excep- 

 tional rather than normal and the alternative view seems more 

 probable that usually during subsidence the upward growth of 

 coral will occur on dead coral reef and only as the result of 

 occasional long pauses in subsidence or prolonged periods of 

 very slow subsidence will sufficient talus accumulate to provide 

 a foundation for coral growth. If this latter picture of the 

 development of an atoll is correct, on the whole a centripetal 

 shift of the outcrop of the reef will take place. 



Davis* has discussed this question and defined the conditions 

 under which centripetal or centrifugal shift of the reef may 

 possibly take place. Let us turn from these "a priori" argu- 

 ments to consider the facts so far as they may be disclosed by 

 an examination of the submarine profile of Funafuti. The 

 diagram shown in Daly's paper is not helpful in this connec- 

 tion for it is generalized and the vertical scale is exaggerated 

 three times. In the Funafuti report several cross profiles are 

 shown on the true scale and if we limit our attention to the 

 part within 200 fathoms of the surface (slightly greater than 

 the depth of the main bore) it will be noticed that the cross 

 profiles show varying slopes. It may be noted here that the 

 soundings showed that down to 4-00 fathoms from the surface 

 round most of the island the average submarine slope is about 

 40°, a figure which must reach if it does not exceed the upper 

 limit for the angle of rest of submarine talus. One profile, 

 AA in the Funafuti drawings, between the surface and a 

 depth of 200 fathoms shows the existence of 4 very steep 

 walls with angles up to and exceeding 70° and one of these is 

 over three hundred feet in height and slopes at 78°. Between 

 * W. M. Davis, Proe. Nat. Acad, of Sci., ii, pp. 466-471, 1916. 



