ufficiently appear. The needle points exactly north
only in a few places, and perhaps not constantly in them; but in most it
declines a little to the east, or to the west, whence arises eastern and
western declination: when this was first observed, it was attributed to
certain excavations or hollows in the earth, to veins of lead, stone, and
other such-like causes. But when it was found by repeated experiments
that this variation varied, it appeared plainly that none of those causes
could take place; since if they had, the variation in the same place must
always have been the same, whereas the fact is otherwise.

Here at London, for instance, in the year 1580, the variation was
observed to be 11 degrees 17 minutes to the east; in the year 1666, the
variation was here 34 minutes to the west; and in the year 1734, the
variation was somewhat more than 1 degree west. In order to find the
variation of the needle with the least error possible, the seamen take
this method: they observe the point the sun is in by the compass, any
time after its rising, and then take the altitude of the sun; and in the
afternoon they observe when the sun comes to the same altitude, and
observe the point the sun is then in by the compass; for the middle,
between these two, is the true north or south point of the compass; and
the difference between that and the north or south upon the card, which
is pointed out by the needle, is the variation of the compass, and shows
how much the north and south, given by the compass, deviates from the
true north and south points of the horizon. It appears clearly, from
what has been said, that in order to arrive at the certain knowledge of
the variation, and of the variation of that variation of the compass, it
is absolutely requisite to have from time to time distinct accounts of
the variation as it is observed in different places: whence the
importance of Captain Tasman's remarks, in this respect, sufficiently
appears. It is true that the learned and ingenious Dr. Halley has given
a very probable account of this matter; but as the probability of that
account arises only from its agreement with observations, it follows
those are as necessary and as important as ever, in order to strengthen
and confirm it.

CHAPTER III: REMARKS ON THE VARIATION OF THE NEEDLE.

On the 6th of November, I was in 49 degrees 4 minutes south latitude, and
in the longitude of 114 degrees 56 minutes; the variation was at thi