THE ORIGIN OF CORAL REEFS. 77 
by jumps than by a gradual outward growth. This inference is of 
considerable importance, since it connects all classes of reefs together 
in the matter of their seaward growth, the degree of inclination of 
the submarine slope being the chief determining factor. 
Following Le Comte, I have previously shown (Proc. Roy. Soc., 
Edin., 1885-86, p. 884) that where there is a very gradual sub- 
marine slope the deposition of sand and the presence of much 
sediment in the water will prevent the growth of corals in the 
shallow water outside the seaward edge of the reef, and that in 
consequence a line of living corals will spring up in the clearer 
and deeper waters a considerable distance beyond. The ap- 
pearance of this line of coral at the surface will result in the 
production of a barrier-reef with a lagoon-channél inside. In 
a similar manner the submerged line of growing corals immediately 
outside the weather-edge of the reef of Keeling Atoll would form 
a barrier-reef, if it was removed some miles from the shore instead 
of being only about 100 yards distant. As it is now situated, it 
lies too close to the edge of the present reef to prevent the oblitera- 
tion of the channel inside it after it has reached the surface. Its 
lagoon-channel would be very quickly filled with sand and reef- 
débris, and as a result we should merely have a permanent addition 
to the present reef-flat, which, when the process was complete, would 
be 100 yards wider. The process is the same as in the case of a 
barrier-reef, the difference in the result being due to the submerged 
line of corals being too close to the edge of the reef for the preser- 
vation of the interior channel; and this circumstance is due to the 
fact of the submarine slope being greater than in the case of a coast 
fronted by a barrier-reef. These remarks are merely intended to be 
suggestive. They may, perhaps, direct the attention of other ob- 
servers to the examination of the outer slopes of atolls and to their 
mode of seaward growth. This can only be done during unusually 
calm weather. 
I have discovered many other new features of minor interest in 
connexion with Keeling Atoll, to which I will refer in my full de- 
scription of these islands. The island of North Keeling, lying 
fifteen miles to the north, is a small atoll connected with Keeling 
Atoll by a bank. I hope to describe it at some future time. 
In conclusion, I may state that most of my observations in these 
islands were directed towards estimating the age of Keeling Atoll. 
These data have yet to be worked up, and I am fairly confident of 
getting a satisfactory estimate. The lagoon is rapidly filling up 
