but, as the rule altogether is admitted to be insufficient for its intended purpose, why not discard it at once? The merit of that able navigator rests on a more permanent basis than this rule can furnish; he discovered the error, and the cause of it: but he certainly failed in establishing a formula for computing it; which, in fact, required a much more extended series of observations than he possessed, or than we have probably yet obtained.

A farther observation made by Captain Flinders was that, in every ship, a compass would differ very materially from itself on being removed from one part of the vessel to another; which is indeed the necessary consequence of the nature of the disturbing force; and particularly in different ships, the compasses will vary in a very remarkable degree from each other. Relatively to this part of his inquiry, Captain Sabine states that

The Isabella and Alexander had not completed half their voyage across the Atlantic, before it was found that the binnacle compasses of the one ship differed very materially, in indicating the course steered, from those of the other: namely, one point, or 110. No dependence whatsoever could be placed on the agreement of compasses in different parts of the ship, or of the same compass with itself, if removed but a few inches: even in the neighbourhood of the binnacles the variation, as observed amidships, was from 8° to 10° greater than the result of azimuths taken by a compass placed between two or three feet on the larboard side; and an almost equal difference in a contrary direction took place on removing the compass to the starboard side, rendering it a matter of some trouble and difficulty to make the azimuth compass agree with those in the binnacle by which the ship was steered, and for which it was therefore necessary to determine the variation.'

With respect to the actual observed deviations with the ship's head on different points, we have these remarks:

The Isabella being at anchor in Brassa Sound, Shetland, her head was placed, by means of warps, on each point of the compass successively, and the bearing of a pile of stones on the summit of a distant hill noted by her compass at each point; at the same time that these observations were made on board, her bearing from the hill was also observed by a compass placed on the pile of stones; the agreement in bearing shewed the points of no error, and the differences the errors in each point, without the calculation which azimuths involve.'

These errors are contained in the following table:

TABLE

TABLE of the Errors in the Isabella's Compass. Shetland.

Dip 74° 21'.

Similar experiments were afterward made in the Alexander, and with similar results; with the exception of the direction of the line of no error, which, as we have already stated, was nearly at right angles to the magnetic meridian.

As to the inadequacy of Captain Flinders's rule, Captain S. states that,

In the observations made in the Isabella at Shetland, where the dip is 74° 21, the maximum of error was 5° 34' easterly of the true variation, with the ship's head at E.S. E. and 5° 46' westerly at W.N. W. making an extreme difference of 11° 20'.

By Captain Flinders's rule, the common multiplier for this compass would have been about one-twelfth, or .083, which at a dip of 86° 09', which was the greatest observed during the late voyage, would have given an error of between 7° and 8°, making the extreme difference 15°; whereas repeated observation showed it to be at that time more nearly 50°, if not exceeding that amount.

The inadequacy of the rule will also appear by reference to the observations made by the Alexander in Baffin's Bay. The error at eight points being 6° 46', at a dip of 84° 30'; it ought scarcely to have exceeded 7° at the greatest possible dip, making an extreme difference of less than 15°. No opportunity occurred indeed of making accurate observations at a greater dip than the above; but the difference in the bearing of objects before and after tacking indicated with sufficient certainty, that the error had increased to an amount very far beyond 15°; frequent instances of an extreme difference of from 3 to 4 points being remarked, as the ship ap

proached

proached the farthest western longitude to which she attained in a high latitude; this was in Lancaster's Sound of Baffin, into which inlet the expedition sailed beyond the 81° of west longitude in the parallel of 74° and a few minutes.

It is much to be regretted that the service did not admit an opportunity to be afforded, of making observations on the various magnetic phenomena, with the excellent instruments supplied to the expedition, at this very interesting place; where a nearer approach was made to one of the magnetic poles than had ever been known before.

But in the absence of any actual observation on the dip of the needle, this fact of the error of the compasses having increased from local attraction so greatly beyond the amount which had been before observed, is worthy of notice, as affording an indication that the dip had also increased, and not inconsiderably. The greatest which was observed, was 86° oy'; and after this observatión, the ships continued to sail for six days in the direction in which the dip had hitherto been found to increase.'

The question of the local attraction of the guns on the needle being thus confirmed by a new series of observations, let us offer a few remarks relative to the fatal consequence which may attend it. This, indeed, has been already done by Mr. Bain, in the treatise to which we have referred: but it is a subject which cannot be too strongly impressed on the minds of seamen, and on those who preside over our naval affairs.

The reader will perceive by the extract already given that in Baffin's Bay, where the dip is very considerable, an error of 50°, or more than four points, may be committed in estimating the direction or course of the vessel; and, supposing that, by clouds intervening only for a few hours, the vessel is left wholly to the compass, the great error which may arise in the ship's reckoning is sufficiently obvious:- but, leaving this as an extreme case, let us confine our attention to the errors as they were observed at Shetland, which agree very nearly with those that were recorded by Mr. Bain off the coasts of Scotland, and which probably differ but little for several degrees southward. Let us assume the rate of the vessel to be only six knots per hour, and that for two days no observation has been taken in consequence of clouds or other impediments: let us suppose that she has been standing on the same tack during these 48 hours, and that the error from local attraction is only one point, which is rather less than it is stated by Captain Sabine; then computing the consequent error in the reckoning, we shall soon be convinced of the extreme importance of the inquiry to which these memoirs refer.

Without making this computation according to the strict rules of navigation, we shall arrive at the amount of the con

sequent

sequent error, sufficiently near to the truth, by simply estimating the base of an isosceles triangle, having its vertical angle equal to 11° 15', and its equal sides each 48×6=268 miles: these data give for a base 50 miles; so that, at the end of that time, the vessel will be actually 50 miles distant from the place indicated by her dead reckoning, and perhaps close to land, while she supposes herself to have ample sea-room. If this happens in the night, what will be the consequence?— in all probability, the vessel will be run ashore; and, if there be any sea, the whole or the greater number of her crew may be consigned to a watery grave. This, we are convinced, is no imaginary event, but does in fact often occur; and too much pains cannot be taken to remedy so serious an evil.

Another circumstance, though not so fatal, is still of considerable importance; and we have heard that it was the ground of frequent complaints during the late war, although the cause was not then so well understood as it would be at present. The case to which we allude is that of a ship of war sailing with a convoy. When night comes on, directions are given by signal for all the vessels of the convoy to steer a certain course, which we will imagine to be duly followed by the respective masters: but, in the morning, instead of the different ships finding themselves in company, some will have run several miles to leeward, and others as much to windward. What the consequence might be of an enemy's ship being at hand at such a time is sufficiently obvious: but, admitting that this does not happen, very frequently those which are to leeward cannot get up, and the others are obliged to run down to join them. Much valuable time being lost in effecting these operations, the ships again proceed in company, and are probably again separated in the following night.

That the cause of these errors may be in a great measure traced to the local influence of the iron of the several vessels cannot be doubted, by those who read the extracts which we have given from Captain Sabine's memoir: but it is equally possible that, even independently of that cause, great errors may arise from the imperfect coustruction of the compasses in the different ships. If we have correct information, (and we have the strongest reason for believing it to be so,) this imperfection does not appertain merely to the instruments belonging to transports, &c., but to most of the compasses usually delivered to his Majesty's ships: the irregularities being so great that, in a given number of compasses, (ten for example,) scarcely two will agree in direction, and in some cases the error will amount to between a quarter and half a point. This surely is a subject deserving the serious consider

ation of those who are at the head of our naval department. No expence has been spared, nor any encouragement withholden, in order to improve the construction of our ships of war; and that such improvement has been effected is universally admitted: but surely some attention ought to be paid to their preservation afterward, which is endangered by the employment of compasses of such imperfect construction, sent in (we believe) on contract, at the lowest price; and we are not aware of any officer whose duty it is to see that they are correct. The causes of the deviations to which we have alluded are various: they may arise from a want of perfect centering of the card, and from the north point of it not coinciding with the north point of the needle; or, the needles of these compasses being nearly rectangular, it may happen that the north and the south points may not coincide with the geometrical axis of the bar, but lie diagonally. Other and similar causes exist, against which we cannot too carefully guard; nor can too many precautions be taken to ascertain that all the needles and cards, before the compass is received into store, agree with each other. While these instruments, however, are furnished on contract at the lowest rate, and no person is charged to see that they are accurately constructed, errors must necessarily occur, and their fatal consequences remain undiminished.

As to the remedy of the latter evil, viz. of the disagreement of the different compasses, nothing is more easy: but a cure for the former source of error is a matter of great difficulty. No doubt, the quantity of deviation produced by the guns depends in a great measure on the dip of the needle; and the amount of this is very imperfectly known, except in parts where regular observations have been made. This difficulty, however, instead of discouraging attempts to rectify the evil, ought rather to act as an inducement to push them in every possible way, and to leave no means untried that offer any prospect of success.

Let us now turn to Captain Sabine's second communication; which, as its title imports, is principally occupied with the detail of experiments relative to the inclination of the dipping needle, the variation of the horizontal needle, and the intensity of the magnetic force in different latitudes. On the former of these subjects we are furnished with the following table:

• Observations