on, 1, 5,
and 6 in the afternoon, until as the season advanced and the days became
short the earliest and latest of these hours were omitted. Although
several barometers were thus constantly observed, no other use of these
was made but to determine their comparisons with each other, except one
of the barometers of Mr. Hassler, Superintendent of the Coast Survey.
This, from its superior simplicity, being, in fact, no more than the
original Tonicillean experiment, with a well-divided scale and
adjustment of its 0 deg. to the surface of the mercury in the cistern, was
found to be most certain in its results. All the barometers used by the
parties in the field were therefore reduced to this by their mean
differences.

The stations at the two above-mentioned places were near the St.
Lawrence. At Metis the height of the cistern of the standard barometer
was determined by a spirit level. At the river Du Loup the height of the
station was determined by two sets of observations of barometers, taken
with different instruments by different observers, and at an interval of
a week from each other. The results of the two several sets, which were
calculated separately, differ no more than 0.5 of a foot from each
other.

On reaching the highest accessible points of the streams on which the
parties proceeded toward the height of land, stationary camps were
established, as has been already stated. At these series of observations
were made at the same hours as at the river stations. The height of
the former was then calculated from a series of observations taken at
noon and at 1 p.m. for the whole of the time the camp was occupied.
The heights of the points at which observations were made by the
traveling party were then deduced from a comparison with the nearest
contemporaneous observations at the stationary camp. An exception to
this rule was made in the observations to the westward of Temiscouata
Lake, which were referred directly to those made at the river Du Loup,
which was sufficiently near for the purpose.

The height of the stationary camp at Mount Biort having been determined
by observations continued for several days, the level of Lake
Temiscouata was thence determined by using a set of levels taken with a
theodolite by Breithaupt, of Cassel, in 1840. The height of the lake
thus deduced is greater than it would appear to be from the barometric
observations taken in December, 1840. It had been imagined that a
difference in le