ON INVESTIGATIONS IN THE INDIAN OCEAN. 199 
on July 10, 1908, leaving on October 8. I devoted my attention to the 
geography and biology of the islands of Silhouette (2,467 feet) and 
Mahé (2,998 feet). The former is twelve square miles in extent and the 
latter over fifty square miles. Both are formed of coarsely crystalline 
granite rock, cut up by dykes of fine-grained black rock, down which 
the rain has for the most part cut its watercourses. The amount of 
erosion ig enormous, and the general topography of the islands is 
almost entirely to be ascribed to the action of weathering. In all 
there are twenty-seven granite islands of upwards of 150 square miles 
on the centre of a shallow (less than fifty fathoms) bank of over 
20,000 square miles. My impression from a further study of the 
group is that the greater part of the bank must have been at one time 
granite land, and that the present marine and aerial erosive actions 
in progress are sufficient to account for its gradual conversion into a 
series of small islands standing on a relatively enormous bank. 
The fauna and flora of Silhouette and Mahé were carefully studied, 
very large collections being secured. They have a continental facies, 
i.e., the general appearance and variety of form which is usually asso- 
ciated with large land masses. Both are severely restricted as to 
number of genera and species, but not more perhaps than would 
naturally be expected from the cutting down of a large mass of 
land to a few scattered islands, and the consequent change, almost 
to complete uniformity, of climate that would result. Only about four 
square miles of the indigenous jungle now remain. It is of the 
tropical rain-forest type, and a century and a half ago undoubtedly 
covered the islands uniformly from the sea to their highest peaks. 
The conditions of a rain-forest are not altogether favourable for a 
great variety either of plants or of animals. The jungle that remains 
lies on mountain peaks, covered with almost perpetual mist, and 
possesses these conditions in their extremest aspects. Few of the 
plants have been killed out, many persisting by the streams, but 
scarce half of the insect fauna is likely still to survive. Introduced 
insects and spiders are everywhere, and have been peculiarly destruc- 
tive to the indigenous forms, directly by killing them and indirectly 
by consuming their food. 
In a former report I referred to evidence of a change of level of the 
islands of Mahé and Silhouette in respect to the sea. My investigations 
show an elevation of these islands, or a lowering of the sea, of at least 
twenty-five to thirty feet, or four to five fathoms. This could scarcely 
have been sufficient to affect the topography of the Seychelles group as 
a whole, but is an interesting phenomenon when taken in conjunction 
with the general slight elevation found in most of the island groups 
from Madagascar to India. No indication was found of any subsequent 
or previous subsidence, the possibility of which is contradicted by the 
whole topography of the Seychelles. 
Mr. H. Scott devoted himself to the entomology of the Seychelles, 
spending two months in Silhouette, five months in Mahé, and a month 
in other islands. He camped for the most part of the time in the 
mountains, and devoted himself to the insects of the indigenous jungles, 
collecting them and working at their habits and life histories. His 
collections comprise about 50,000 specimens, which are now being 
sorted and mounted for examination by specialists. It is too soon 
yet to say anything about these collections, but it may confidently be 
