Vol. 58. | PHYSICAL DEVELOPMENT OF DOMINICA, ETC. 353 
dissection of the plateau into what are now the various island- 
masses. 
This portion of the Antillean chain records several epochs of 
denudation :—(1) That of the long Miocene- Pliocene time when the 
Antillean plateau was elevated to 3000 feet or more, and carved 
into several isolated tablelands now forming the various island- 
masses, with broad undulating depressions between them. The 
date has been determined from the evidence obtained in Antigua, 
Guadeloupe, Barbados, and other islands, and sustained in these. 
(2) After a subsidence, or probably two epochs of subsidence, ati 
the close of the Pliocene Period the region was raised to its 
maximum height, shown within the area of the islands to have 
exceeded 7000 feet, though evidently it attained a still greater 
altitude. The erosion-feature of this epoch is that of very deep 
valleys, amphitheatres or cirques dissecting or indenting the 
late tablelands, but the time was not one of relatively long 
duration and belonged to the early Pleistocene Period. Then there 
was a subsidence in the Mid-Pleistocene Period, so that the islands 
were smaller than at present; followed by (3) a re-elevation which 
resulted in the islands being a little higher than now. The gorges. 
and channels formed then have been partly submerged since that 
date. The subsequent minor changes of level need not be followed 
further at present. 
It was during the great: elevation of the early part of the Pleisto- 
cene Period that all the islands were united into a continental mass, 
thus allowing of the migration of the Pleistocene mammals to. 
Guadeloupe and Anguilla. These, however, were exterminated by 
the subsequent almost complete submergence: since when there 
has been no land-connection between the islands and the adjacent 
continent. The elevation of the mountain-districts, in excess of 
the general movements of the Antillean plateau, is a phenomenon 
due to local plutonic forces, where recent volcanic activity has 
obtained; although it is not to such causes that the various 
islands owe their separation one from the other, but to atmo- 
spheric erosion and changes of level. 
EXPLANATION OF PLATE X. 
Geological sketch-map of the Windward Islands, on the scale of about 
35 miles to the inch. Soundings are shown in feet. 
[For the Discussion, see p. 365. | 
