n already described, as forming the ridges in this quarter.
Most of the beds are impregnated wholly, or in patches, with bitumen.
Some of these specimens contain corallines and terebratulae; and at the
lower end of the defile there are horizontal strata of limestone,
covered by a thin layer of flinty slate.

Below the _ramparts_ the river expands to the width of two miles, and
for a reach or two its banks are less elevated. In lat. 66-3/4 degrees,
about thirty miles from the ramparts, there are cliffs which Captain
Franklin in his notes, remarks, "run on an E. by S. course for four
miles, are almost perpendicular, about one hundred and sixty feet high,
and present the same castellated appearances that are exhibited by the
sandstone above the defile of the "ramparts." [Sidenote: 159, 160, 161,
162] The cliffs[38] are, in fact, composed of sandstones similar, in
general appearance, to those which occur higher up the river; but some
of the beds contain the quartz in coarser grains, with little or no
cement. [Sidenote: 163, 164, 165, 166] The beds are horizontal, and
repose on horizontal limestone,[39] from which Captain Franklin broke
many specimens in 1825. [Sidenote: 167, 168, 169, 170] We landed at this
place in 1826 to see the junction of the two rocks, but the limestone
was concealed by the high waters of the river. Captain Franklin's
specimens are full of shells, many of which are identical with those of
the flat limestone strata of the Athabasca River. [Sidenote: 171] One
bed appears to be almost entirely composed of a fine large species of
terebratula, not yet described, but of which Mr. Sowerby has a specimen
from the carboniferous limestone of Neho, in Norway. Some of the beds
contain the shells in fragments; in others, the shells are very entire.

About forty miles below these sandstone walls the banks of the river are
composed of marl-slate, which weathers so readily, that it forms
shelving acclivities. [Sidenote: 172] In one reach the soft strata are
cut by ravines into very regular forms, resembling piles of cannon shot
in an arsenal, whence it was named _Shot-reach_.

The river makes a short turn to the north below Shot Reach, and a more
considerable one to the westward, in passing the present site of Fort
Good Hope. The banks in that neighbourhood are mostly of clay, but beds
of sandstone occasionally show themselves. The Indians travel from Fort
Good Hope nearly due north, reach the summit of a ridge of l