s a common accident, owing to the
neglect of the Indians and voyagers in not putting out their fires, and
in a dry season the woods may be seen blazing to the extent of many
miles. We afterwards crossed the Second, or Swampy Portage, and in the
evening encamped on the Upper Portage, where we were overtaken by an
Indian bringing an answer from Governor Williams to a letter I had
written to him on the 15th, in which he renewed his injunctions to the
gentlemen of the boats accompanying us, to afford us every assistance in
their power. The Aurora Borealis appeared this evening in form of a
bright arch, extending across the zenith in a N.W. and S.E. direction.
The extent of our voyage to-day was two miles.

About noon, on the 25th, we entered Knee Lake, which has a very
irregular form, and near its middle takes a sudden turn, from whence it
derives its name. It is thickly studded with islands, and its shores are
low and well-wooded. The surrounding country, as far as we could see, is
flat, being destitute even of the moderate elevations which occur near
the upper part of Hill River. The weather was remarkably fine, and the
setting sun threw the richest tints over the scene that I remember ever
to have witnessed.

About half a mile from the bend or _knee_ of the lake, there is a small
rocky islet, composed of magnetic iron ore, which affects the magnetic
needle at a considerable distance. Having received previous information
respecting this circumstance, we watched our compasses carefully, and
perceived that they were affected at the distance of three hundred
yards, both on the approach to and departure from the rock: on
decreasing the distance, they became gradually more and more unsteady,
and on landing they were rendered quite useless; and it was evident that
the general magnetic influence was totally overpowered by the local
attraction of the ore. When Kater's compass was held near to the ground
on the N.W. side of the island, the needle dipped so much that the card
could not be made to traverse by any adjustment of the hand; but on
moving the same compass about thirty yards to the west part of the
islet, the needle became horizontal, traversed freely, and pointed to
the magnetic north. The dipping needle being landed on the S.W. point of
the islet, was adjusted as nearly as possible on the magnetic meridian
by the sun's bearings, and found to vibrate freely, when the face of the
instrument was directed to the east or we