56 20 DR, LH. CB. GUPP YR: 
varying from 100 to 2,000 fathoms. In more than one 
upraised atoll, denudation had also exposed to view the 
original volcanic peak, thus corroborating in a remarkable 
manner the theory of Mr. Murray, that submerged volcanic 
peaks are levelled up to the zone of reef-corals by the acct- 
mulation upon them of organic deposits, and that these 
deposits are finally crowned by the atoll. 
In another paper on the living Coral Reefs of this region, 
which was published in the Proceedings of the same society 
for 1886, I removed the chief objection against Mr. Murray’s 
views, one which gave powerful support to Mr. Darwin’s 
theory, namely, the awkward fact, as it was then believed to 
be, that lagoons and lagoon-channels are sometimes deeper 
than the zone in which reef-corals thrive. This belief was 
shown to be founded on a misconception of the conditions that 
limit the downward extension of this zone. Observers in 
different regions have variously estimated its depth between 
five and thirty and even forty fathoms. This great variation 
is due to differences in the local conditions, not only in differ- 
ent localities, but, as I found in the Solomon Islands, in the 
same locality. The main determining condition of the depths 
of reef-corals in all regions is to be found (as Professor A. 
Agassiz has also shown) in the injurious effect of sand and 
sediment rather than in the general influence of depth, the 
distribution of these materials being dependent on such local 
conditions as the angle of the submarine slope, the presence 
and situation of submarine declivities, the amount of sediment 
held in suspension, the force of the breakers, and other 
influences. Local conditions will usually restrict the reef- 
coral zone to depths less than twenty fathoms; but where 
there are a moderate submarine slope, clear water, and 
breakers of no great size, reef-corals may be found flourishing 
in depths of fifty and even sixty fathoms. 
In the same paper I also referred to the conditions 
described by me as the determining causes of a barrier-reef. 
Where there is a rapid submarine slope,—for instance, more 
than 10 degrees,—the sand and gravel produced by the action 
of the breakers on the outer edge of the shore-reef will extend 
to depths far beyond those in which reef-corals thrive; but 
let the slope be small, say 1 or 2 degrees, the lower margin 
of the belt of sand and débris will then le within the zone of 
reef-building corals, and in consequence a line of barrier-reef 
will ultimately be formed with a deep channel inside. This 
explanation I afterwards found to be precisely the same as 
that advanced by Professor Le Conte thirty years before in 
the instance of the Florida reefs, a view which I have pre- 
