256 THE INFLUENCE OF INANIMATE SUKROUNDINGS. 



stead of it, we are compelled to have recourse to various other 

 forces to aid us to a solutiou, the problem of showing that a 

 subsidence must nevertheless have taken place belongs more 

 than ever to the prevailing theory. But in the case here under 

 discussion such an attempt is met by an insuperable diflficulty ; 

 this was briefly alluded to above, and must now be discussed 

 somewhat more fully. 



The submarine mountain district which serves as the foun- 

 dation of the whole group is extremely narrow ; at the widest 

 part the western reef is at most from twelve to fifteen miles 

 from the eastern one. Nevertheless, these reefs are extraor- 

 dinarily dissimilar in structure ; even in the small and indepen- 

 dent atoll of Kriangle this diffei-ence is conspicuous. The 

 eastern reef is everywhere much nearer to the shore than the 

 western. A navigable channel between it and the included 

 islands occurs only at about the middle of Babelthuap, where 

 Altngot Passage (see Map I.) has been formed, according to 

 Friedrichsen's map, by an outer reef lying in front of the 

 island reef proper. The eastern reef, so far as I have seen it, 

 nowhere exhibits a line of dead coral-blocks above the highest 

 tide mark. And yet this is the weather side, on which only, it 

 is said, such blocks ever occur. The western reef, on the con- 

 trary, is invariably characterised by them. Moreover, while 

 the western clifi" is throughout a true barrier reef, almost down 

 to the southern point of Pelelew, the eastern reef can hardly be 

 regarded as a barrier reef even at the north of Babelthuap ; and 

 southwards for about the lower third of the group, it assumes 

 the character of a true fringing reef. 



This difference, and the incontrovertible fact that fringing 

 reefs predominate on the eastern coast, cannot by any means 

 be reconciled with Darwin's theory unless we suppose, as 

 Darwin has done and as Dana evidently fain would do, that 

 such a reef formation might take place even during subsidence. 

 But, according to Darwin's own admission, this involves the 

 assumption that the seaward declivity of the outer reef is ex- 

 tremely steep. This hypothetical steepness, however, occurs in 

 the Pelew Islands only on the west side, while, in absolute con- 

 tradiction, the eastern declivity is remarkably gradual. From 



