26 
HILLS NEAR BANDA. 
per part, into two or more smaller elevations, of which the 
central only terminates with a pointed summit. The ap- 
pearance of the hill, from below, is singular and fantastic ; 
huge masses of stone presenting themselves in every posi- 
tion, seemingly quite unconnected the one with the other, 
while a few shrubs, growing out from between them, serve 
as a contrast to the nakedness of the rock. On ascending 
the hill, we found it to be composed of a reddish-coloured 
small-grained granite, having no regular arrangement, but 
lying in blocks of great size, some perpendicular, and others 
horizontal, with a convex or rounded surface in general. 
Many of these are scaling off at the surface ; but the greater 
part remain entire, and possess more compactness of inte- 
grant structure than any rocks of the kind which I have 
met with. This hill at Banda may be reckoned the termi- 
nation of the first of the many series which traverse Bundle- 
cund from W. to E., as no more are observed here follow- 
ing that line. 
Soon after leaving Banda to the south, other hills come 
into view, which, at first sight, appear larger than the one 
at that place. This is chiefly owing to the effect of distance, 
increased by the dewy air of the morning, which is the time 
of travelling ; for, on a near approach, we find these not to 
exceed the congeries at Banda, or the highest does so only 
in a small degree. 
Though evidently entering upon a mountainous country 
here, we are surprised to find no general elevation of the 
surface, the same flatness of the plains continuing as on the 
northern side of the Jumna, and the hills rising abruptly 
from a common level, like so many islands rearing them- 
selves out of the ocean. They are, in fact, mere pictures, 
on dry land, of the rocky Madeira, Forto Santo, or the Ca- 
naries, seen in the voyage from England to India. 
About twelve miles from Banda, at the village of Gc- 
