s 
468 

DYNAMICAL GEOLOGY. [Book IIL: 
detailed examination of many coral reefs, has offered another ex- 
planation of the phenomena. He suggests that barrier reefs do not 
necessarily prove subsidence, seeing that they may grow outward 
from the land upon the top of a talus of rock fragments or of their 
own débris broken down by the waves, and may thus appear to con- 
sist of solid coral which had grown upward from the bottom during 
depression, although only the upper layer, 20 fathoms or thereabouts 
in thickness, is composed of solid, unbroken coral growth. He 
points out that in the coral seas the islands appear to have always 
started on volcanic ejections, at least that all the non-calcareous rock 
now visible is of volcanic origin. The portion of a volcanic cone 
(Fig. 179) raised above the sea may be supposed to be cut down 
b wee NS 7 
MMMM 
Fic. 179.—SEcTION oF A VOLCANIC CONE SUPPOSED TO HAVE BEEN THROWN UP ON 
THE SEA-FLOOR AND TO HAVE REACHED THE SEA-LEVEL (B.). 

to the lower limit of breaker action (a a), so as to offer a platform 
on which coral might grow into reefs (¢ k) up to the level of high- 
water (bb). Or, with less denudation, or a loftier cone, a nucleus of 
the original voicano might remain as an island (Fig. 180), from the 
sides of which a barrier reef (7 7) might grow outward, on a talus of 
its own débris, and maintain a steep outer slope. According to this 

Fic. 180,—Srcrion oF VoucANIc IsLhaAnn WITH SURROUNDING CORAL-REEF (B.). 
view the breadth of a reef ought, in some degree, to be a measure 
of its antiquity. 
To the obvious objection that this explanation requires the 
existence of so many volcanic peaks just at the proper depth for 
coral growth, and that the number of true atolls is so great, Mr. 
Murray replies that in several ways the limit for the commencement 
of coral growth may be reached. Volcanic islands may be reduced 
by the waves to mere shoals, like Graham’s Island, in the Mediter- 
ranean. On the other hand, submarine voleanic peaks, if originally too — 
low, may conceivably be brought up to the coral zone by the constant - 
deposit of the detritus of marine life (foraminifera, radiolaria, 
pteropods, &ec.), which this observer has found to be very abundant in 
