CORAL REEFS AND SUBMARINE BANKS 228 
aspect of the problem has been dealt with more fully in some of my 
earlier papers and is further treated in later sections of this article. 
Barrier reefs and their lagoon deposits, if formed under the 
conditions and processes of the glacial-control theory, should, when 
sufficiently elevated and dissected, be found to lie unconformably 
with small thickness, not on a slope of subaérial erosion, but on an 
abraded platform, the outer part of which truncates a series of 
inclined detrital deposits, while the inner part truncates a series 
of volcanic rocks and is limited by an ancient sea cliff, more or less 
weathered. But it should be noted that if such a reef were elevated 
long enough ago to have been well dissected, its formation could 
not have been of postglacial date; it must have been either pre- 
glacial or interglacial; if interglacial, then the continuity of abrasion 
through the glacial period is disproved; if preglacial, the reef should 
not rest on an abraded platform, but should constitute (on the 
supposition that no subsidence took place during its formation) a 
conformable member of the exterior detrital deposits, the lowest 
member of which should lie conformably upon a non-eroded volcanic 
slope. 
The elevated and greatly dissected reefs of Vanua Mbalavu, in 
eastern Fiji, the best examples of their kind that I have seen, do not 
fulfil any of these conditions; their limestones lie on slopes of 
subaérial erosion, the vertical measure of which is, as already noted, 
much greater than the greatest possible change of ocean-level during 
- the glacial period; whatever the date of formation of these reefs, 
the occurrence of subsidence before or during their formation seems 
to me demonstrated.* A single example of this kind cannot, 
however, have great general value; further exploration and 
investigation of other elevated reefs must be made with the con- 
siderations above outlined in mind before this phase of the problem 
can be far advanced. Abundant opportunity for such investigation 
is offered by the Philippines, Pelew, New Hebrides, Solomon, and 
other island groups, where the unconformable contact of elevated 
reefs with their eroded foundations is implied by the incomplete 
records now available. 
1 “The Origin of Certain Fiji Atolls,” Proc. Nat. Acad. Sci., II (1916), 471-75. 
[To be continued] 
