The First Ascent of Roraima. 45 
face of the shallow pools by the varying force of the 
breezes which are almost continually battling over the 
top of the mountain. They explain also ROBERT Schom- 
BURGK'S statement that the water seems to flow not 
from the top of the cliff but from points some dis- 
tance below ; for the water flows through deep and 
narrow sloping channels, which it has cut for itself 
so as to issue at some distance down the face of the 
cliff, the channels themselves being indiscernible from 
below, sometimes because they are lost to sight in the 
general irregularity of the rock surface, sometimes 
because issuing from behind a false face to some portion 
of the cliff. 
The one observation of previous travellers which I 
find it somewhat difficult to reconcile with fa6ls, is that 
in which it has frequently been asserted, that the top is 
covered with trees. As regards the northern end of the 
mountain, as I have not seen this, I cannot positively 
assert that there are no trees on it. But the remark has 
chiefly been made with reference to the southern end, at 
the point at which we ascended. I can only suppose 
that previous travellers, obtaining only a distant view, 
have mistaken the many and extraordinarily rugged pin- 
nacles and points of rocks for the tops of trees. 
Small fleeting masses of clouds were passing over 
the top during the whole of our visit, though it 
was a fine and, otherwise, a bright day. So many 
and changing were these clouds that I only managed 
to secure sketches by seating myself on a high pinnacle 
of rock from which four or five interesting points were 
visible at once, and turning to sketch each of these 
as each in turn became visible. 1 suppose that there 
