From these remarks I shall draw the following conclusion, that after passing the Cape de Verde Islands, if you do not make above 4 deg. or 5 deg. easting, and cross the Line in, or to the westward of, the meridian of St Jago, you may expect to find your ship 3 deg. or 4 deg. to the westward of her reckoning by the time you get into the latitude of 10 deg.  S. If, on the other hand, you keep well to the E. and cross the Line 15 deg. or 20 deg. to the E. of St Jago, you will be then as much to the E. of your reckoning; and the more you keep to the eastward, the greater will be your error, as has been experienced by some India ships, whose people have found themselves close upon the coast of Angola, when they thought its distance was above 200 leagues.

During the whole of our passage from England, no opportunity was omitted of observing, with all the attention and accuracy that circumstances would permit, the variation of the compass, which I have inserted in a table, with the latitude and longitude of the ship at the time of observation.  As the longitude may be depended upon, to a quarter or half a degree at most, this table will be of use to those navigators who correct their reckoning by the variation.  It will also enable Mr Dun to correct his new Variation Chart, a thing very much wanted.

It seems strange to me, that the advocates for the variation should not agree amongst themselves.  We find one[91] of them telling us, as I have already observed, “that with 8 deg.  W. variation, or any thing above that, you may venture to sail by the Cape de Verde Islands by night or day, being well assured, with that variation, that you are to the eastward of them.”  Another, in his chart,[92] lays down this variation ninety leagues to the westward of them.  Such a disagreement as this, is a strong proof of the uncertainty of both.  However, I have no doubt the former found here, as well as in other places, the variation he mentions.  But he should have considered, that at sea, nay even on land, the results of the most accurate observations will not always be the same.  Different compasses will give different variations; and even the same compass will differ from itself two degrees, without our being able to discover, much less to remove, the cause.

[Footnote 91:  Nichelson.]

[Footnote 92:  Mr Dun.]

Whoever imagines he can find the variation within a degree, will very often see himself much deceived.  For, besides the imperfection which may be in the construction of the instrument, or in the power of the needle, it is certain that the motion of the ship, or attraction of the iron-work, or some other cause not yet discovered, will frequently occasion far greater errors than this.  That the variation may be found, with a share of accuracy more than sufficient to determine the ship’s course, is allowed; but that it can be found so exactly as to fix the longitude within a degree, or sixty miles, I absolutely deny.[93]