40 FORMATION OF THE HILLS OF BUNDLECUND. 
istic of the granitic formation. Their outhne, contrasted 
with the table-summits of the ranges in the interior, ex- 
empUfies in a striking manner the effect of the rock in 
the figure of the general elevation, from which, at the 
distance of many miles, we can often determine the na- 
ture and position of the strata forming very extensive 
ranges. The bare aspect of these granite piles, in the first 
range, and the irregular surface which they present, lead 
one to conclude, that they are but the remains of hills of a 
much greater extent, which have once existed here. They 
may be said to exhibit the cores of large hills, whose exte- 
rior has suffered in the lapse of time, while the more com- 
pact structure of the granite still enables it to resist the 
common causes of decay. I think it can hardly be doubted 
that all hills, similar to those met with in Bundlecund, have 
been originally formed by a force from below, elevating the 
pi'imitive rocks, and causing a disruption of the secondary 
strata, at the several points at which it had been exerted. 
Where the force was but slightly impressed, and in a limit- 
ed area, a small elevation would be formed. The granite 
would then only break through the superincumbent strata, 
without carrying them along with it, while the broken strata 
would rest on the sides of the mass, after the impelling force 
ceased to act. The figure of the hill then would not be a 
pyramid, which it now resembles, but would approach more 
to that of a cone ; sandstone, trap, &c. lying on and sur- 
rounding the granite, and filling up its inequalities, and the 
direction of the strata of each of these deviating more or less 
from the horizontal line, in proportion to the elevation of 
the central mass. We could thus picture a hill more ex- 
tensive than any of those now existing in the first series, 
whose sides at the surface were originally composed of sand- 
stone ledges, and the summit of a pointed block or mass of 
granite, or, crowning the whole, may have existed a table- 
