SOUFRIERE, AND ON A VISIT TO MONTAGNE PELFE, IN 1902. 
307 
runs upon their surfaces for short distances, and these show them to be rock platforms 
carved by marine erosion out of the tufts and lavas of the island. 
The absence of coral and gravel deposits, the depth and number of the valleys cut 
across the beaches, the imperfect preservation of the lower ones, and the fragmentary 
condition of the higher, all prove that they are not of recent date, but have been for 
a very long time exposed to sub-aerial denudation, and are in process of decay and 
disappearance. The clifts behind the ledges have had their outlines softened and 
rounded, the streams have cut deep channels in the rock platforms, reaching new base 
levels at greater depths, and as the valleys widened they have encroached, to a large 
extent, on the flat-topped ridges between. Any superticial accumulations which 
formed on them have long since disappeared. They may, however, have been quite 
unimportant. Coral is not very abundant around St. Vincent, possibly because the 
streams flowing through the soft and weathered tufls discharge great quantities of 
mud into the sea, discolouring it, after heavy showers, for several hundred yards 
from the shore. The surf, which constantly beats on the weather side of the 
island, has carved a broad submarine plateau opposite the headlands. In the bays 
and across the stream-mouths sand and gravel gather, but on the points between the 
bays there is usually no beach, but only a flat platform, all awash and covered by the 
sea. It is on these headlands that the old raised beaches are most conspicuous, and 
in such situations there is no mistaking their meaning. As seen in profile they have 
mostly a slight but perceptible slope towards the water. In section parallel to the 
coast they are sensibly horizontal. But in the sides of the valleys and up the 
stream-courses it is far less common to find these beaches well preserved. For this 
there are several reasons. The cutting power of the sea is least in these places, and 
accumulation of gravel is more common than erosion and removal of the solid rock. 
Rock ledges may never have been cut, or, if cut, must have been quite small. Any 
gravel beaches formed in such bays would be rapidly carried away by the streams 
which took possession of them when the sea retreated. Moreover, it seems probable 
that the lower courses of most of the less important streams are subsequent to the 
terraces and have been carved out of them, so that in the formation of the stream- 
valleys the beaches must have been removed by erosion. 
Further evidence of the great age of these features is aflbrded by the dejjth to 
which the rocks composing them have weathered under the attack of atmospheric 
agencies. On the leeward side of St. Vincent, and everywhere also in the higher 
elevations, although decomposition is rapid, there are abundant exjjosnres of fresh rock, 
for the steep slopes facilitate removal of loose material, and landslips and the washing 
action of the rain keep the surfaces clear. On the lower grounds often a great thick¬ 
ness of rotted rock may be found in certain situation. The taluses on the valley 
sides, where not exposed to the main streams, may be very deep. The soft ash 
weathers readily into a fine, friable dark soil, and after a heavy rainfall the 
lateral rivulets may cut into this to depths of 30 feet without exposing solid rock 
