Darwirfs Theory of Coral Reefs. 179 



has described a limestone terrace which he regarded as an up- 

 lifted barrier reef ; his text is silent as to the attitude of its 

 strata, but his sections show them horizontal. Gardiner 

 describes* the interbedding of washed volcanic detritus with 

 horizontal limestones in an uplifted reef on Yiti Levu of the 

 Fiji group, but the context does not make clear the relation in 

 which these alternating strata stand to the foundation on or 

 against which they were deposited. Andrews describes + with 

 considerable detail some of the uplifted limestones on certain 

 members of the Fiji group, but he often fails to mention the 

 attitude of the limestone and " soapstone " layers in masses 

 which he describes as stratified, and his cross-sections cannot 

 be safely utilized because it is impossible for the reader to sep- 

 arate observed from inferred structures.^ This brief review 

 would seem to indicate that, for the present, no safe decision 

 can be made for or against the theory of outward growth on 

 the evidence of uplifted reefs, because their structure is too 

 imperfectly known. 



Two other points deserve mention in this connection. First, 

 the theory of outward growth is not an alternative that 

 excludes the theory of upward growth ; the two processes can 

 go on together or alternately. Second, Darwin fully recog- 

 nized this possibility : he accepted outward growth without sub- 

 sidence for a reef three miles wide, in which the ridges earliest 

 formed at the back of the reef stand at the same height as 

 those last formed in the front ; § in another case he keenly 

 decides in favor of a stationary period after subsidence because 

 of " the broad belt of low land at the foot of the mountains " 

 (128), while in a third he concludes in favor of upward growth 

 during subsidence, because of " the small quantity of low 

 alluvial land at the foot of the mountains " (128). He con- 

 cluded that "subsidence supervening after long intervals of 

 rest . . . probably is the ordinary coarse of events " (130) ; 

 also, that a reef " could not increase outwards, without a 

 nearly equal addition to every part of the slope . . . and 

 this would require a large amount of sediment " (74). 



It would thus appear that the outgrowth theory of Murray 

 and the truncation theory of Wharton and Agassiz, when 

 tested by certain consequences that have not been explicitly 

 stated by their inventors, fail to satisfy the requirements of 

 observation. They must, therefore, be either modified or dis- 

 carded. It remains to be seen whether Darwin's theory of 

 subsidence must suffer the same fate when tested in the same 

 manner. 



* Quart. Journ. Geol. Soc, xlvii, 1891, 500-607 ; see p. 600. 

 f Proc. Camb. Phil. Soc, ix, 1898, 417, 503 ; see p. 454. 

 {Bull. Mus. Comp. Zool., xxxviii, 1900, 1-50. 

 § Coral Eeefs, 1842, 74, 75. 



