348 PROF. SPENCER ON THE GEOLOGICAL AND [Aug. 1902, 
The Roseau Tufis. 
These deposits have distinct characteristics, and are well exposed 
along the Roseau Valley and on the flanks of Morne Bruce, where 
they are shown to be over 400 feet thick. The tuff is a mixture of 
fine-grained gritty particles and small fragments of the same volcanic 
débris, with some waterworn pebbles varying from a few inches to a 
foot in diameter. The gravels often occur in layers ending abruptly, 
or the pebbles may be scattered throughout the mass. The tuffs are 
of a light cream-colour, due in part to the kaolinization of the felspar. 
These beds only partly refill an older valley. The strata have since 
been raised, so that they dip seaward at an angle of about 10°, the 
slope being somewhat less than that of the surface of the land. While 
the Roseau River has cut through the formation, its surface still 
forms a well-marked sloping terrace, in contrast to the country under- 
lain by the older tuffs previously mentioned. Furthermore, there has 
been no physical break in the succession of the beds, so great as that 
which preceded the submarine accumulation of the Roseau Tuffs : 
consequently one is forced to conclude that the deposit is of much 
more recent geological date than that of the older tuffs. Mr. Guppy 
recognized a conglomerate-formation overlying the older voleanic 
rock, but I do not knew what he included in it. 
V. Tar Gravet-Formations oF Dominica. 
Owing to the extension of the mountain-spurs towards the 
encroaching seashore, the various mechanical formations exposed 
along the coast-line become disconnected and show considerable 
variations, so that it is often difficult to identify their recurrence. 
There are two gravel-formations, which are separated by a great 
unconformity. 
The Morne Daniel and the Grand Savanna Gravel- 
Series.—About 2 miles north of Roseau is an extensive exposure 
of rounded gravel in the lower part of the cliffs overhanging 
the sea, adjacent to the foot of Morne Daniel, where the upper 
portion is composed of a coral-rock along the margin of the cliffs, 
but landward of these the gravel rises to 150 feet or to even 
a greater elevation. In ascending to Bona Vista, about 2 miles 
north of this point, the gravel was seen up to a height of 300 feet. 
Higher up the valley, gravels were also observed to an elevation of 
500 feet, though the identity of the two deposits was not apparent. 
A similar bed overlies the Roseau Tuffs, on the flanks of Morne 
Bruce, up to an altitude of 300 feet; but the Roseau Tuffs are in 
part absent from the locality where the gravels appear beneath 
the coral-bed on the seashore, below Morne Daniel. At other 
points to the northward, what were evidently the same gravels were 
seen in interrupted beds as far as Grand Savanna, the farthest point 
visited. This interruption was occasioned by the very great denu- 
dation, which first removed most of the deposit before the epoch of the 
coral-beds. In places, the gravels were seen resting upon lavas, but 
it is not certain whether these beds belong to the lower or to the upper 
Grayel-Series. In other places, they rest upon reddish-brown tuffs. 
