SOUFRIERE, AND ON A VISIT TO. MONTAGNE PPILEE, IN 1902. 363 
structure of the country than the irregular, inconstant, and often tumultuous deposits 
of tuff. 
The lavas, like the ash Ijeds, have certainly accumulated on a land surface. Occa¬ 
sionally we may see where a lava flow has entered an old valley and has partly or 
completely filled it up, and frequently the base of the lavas is so irregular as to show 
that the surface over which they flowed had been sculptured by sub-aerial erosion into 
numerous ridges and furrows. The curved outcrops not unfrequently suggested 
that folding had taken place and thrown the beds into little anticlines and synclines, 
but examination always revealed the eroded character of the underlying material, 
while no proof of folding on such a small scale was ever obtained. 
A layer of bright red earth very commonly lies directly below the lavas, and where 
the rocks are utterly decomposed, this may be one of the best indications of the nature 
of the overlying mass. It is undoubtedly an old terrestrial soil in which the hydrous 
iron oxides have been changed to hematite by the action of the hot lava flow. This 
layer is usually about a foot in thickness and consists of a decomposed ash bed (only 
very rarely do two lavas come together in the sections) in which the fragmental 
structure is becoming obliterated, baked and hardened into a splintery red clay or 
porcellanite, especially where it is fine grained and in immediate contact with the 
lava. We did not find any remains of burnt wood or other fossils in these old soils 
or in any part of the tuffs. 
Intrusive sheets also are not lacking in St. Vincent, though according to our 
experience they are few and small. There are some sections which show junctions 
so decidedly transgressive that they can only he due to the injection of molten 
masses into fissures in the tuffs. Good examples are to be seen at Dark Head, on the 
leeward coast, and Duvernette Island, near Calliaqua. They are so rare, however, 
that it would not be difficult to enumerate every instance which came under our 
notice. In no case are these masses of great size, but some of tiie more important 
lava flows attain to such great thickness, and are so lenticular in character, that they 
greatly resemble laccolites. One of these in the Cumberland Valley, about a mile 
above its mouth, forms a beautiful columnar cliff nearly 300 feet high, overlooking 
the stream. There was not, however, sufficient available evidence to prove that it 
might not be a lava flow occupying an old valley channel, and very thick in conse¬ 
quence. There is another great cliff of crystalline and columnar-jointed rock in the 
Mariaqua Valley, and from the regularity of this mass in dip and thickness it is most 
jDrobably a lava flow. 
In the splendid sections on the leeward side of the island, dykes are as rare as 
intrusive sheets. The exposures are so good that this is sufficient to indicate their 
scarcity, hut in the interior of the island, owing to decomposition of the rocks and 
the thick growth of forest, they would be practically certain to escape detection. 
Three or four dykes were seen cutting the tuffs near Layu. 
The rarity of dykes, intrusive sheets and necks, are characteristic features of 
3 A 2 
