These irregularities are owing to the structure of the needle; for the wires of which it is composed are only hardened at the ends; now if these ends are not equally hard, or if one end be hardened up higher than the other, when they come to be put together, in fixing them to the card, that end which is hardest will destroy much of the virtue of the other; by which means the hardest end will have the most power in directing the card, and consequently make it vary to wards its own direction: and, as the wires are disposed in the form of a lozenge, these cards can have but little force, so that they will often, when drawn aside, stand at the distance of several degrees on either side the point from whence they are drawn: for all magnetical bodies receive an additional strength by being placed in the direction of the earth's magnetism, and act proportionally less vigorously when turned out of it; wherefore, when these kind of needles are drawn aside from their true point, two of the parallel sides of the lozenge will conspire, more directly than before, with the earth's magnetism; and the other two will be less in that direction; by which means the two sides will very much impede its return; and the two latter will have that impediment to overcome, as well as the friction, by their own force

alone.

To remove these inconveniences, some needles are made of one piece of steel of a spring temper, and broad towards the ends, but tapering towards the middle, where a hole is made to receive the cap. At the ends they terminate in an angle, greater or less according to the skill or fancy of the workman. These needles, though infinitely preferable to the other, are, however, far from being perfect; for every needle of this form hath six poles instead of two, one at each end, two where it becomes tapering, and two at the hole in the middle: this is owing to their shape: for the middle part being very slender, it has not substance enough to conduct the magnetic stream quite through from one end to the other: all these poles appear very distinctly, when examined with a glass that is sprinkled over with magnetic sand. This circumstance, however, does not hinder the needle from pointing true; but as it has less force to move the card than when the magnetic stream moves in large curves from one end to the other, it is certainly an imperfection.

These inconveniencies induced the ingenious Dr Knight to contrive a new sea-compas, which came into use on board all the ships of war. The needle in this instrument is quite straight, and square at the ends; and consequently has only two poles, though about the hole in the middle the curves are a little confused. Needles of this construction, after vibrating a long time, will always point exactly in the same direction; and if drawn ever so little on side, will return to it again, without any sensible difference. We may therefore conclude, that a regular parallelopiped is the best form for a needle, as well as the simplest, the holes for the cap being as small as possible.

one

The above observations will be easily understood from viewing the several parts of the instrument as represented on Plate CL. where fig. 6. is the card with the needle KL, and its cap M, fixed upon it, being one-third of the diameter of the real card. Fig. 8. is a perspective view of the backside of the card, where AB represents the turning down of the brass edge, C the under part of the cap, D and E two sliding weights to balance the card, and F, G, two screws that fix the brass edge, &c. to the needle. Fig. 7. is the pedestal that supports the card, containing a screwing needle, fixed in two small grooves to receive it, by means of the collet C, in the manner.of a port-crayon. D, the stem, is filed into an octagon, that it may be the more easily unscrewed. For its further illustration and application to use, see NAVIGATION.

The invention of the compass is usually ascribed to Flavio da Melfi, or Flavio Gioia, a Neapolitan, about the year 1302; and hence it is, that the territory of Principato, which makes part of the kingdom of Naples, where he was born, has a compass for its arms. Others say that Marcus Paulus, a Venetian, making a journey to China, brought back the invention with him in 1260. What confirms this conjecture is, that at first they used the compass in the same manner as the Chinese still do; i. e. they let it float on a little piece of cork, instead of suspending it on a pivot. It is added, that their emperor Chiningus, a celebrated astrologer, had a knowledge of it 1120 years before Christ. The Chinese only divide their compass into 24 points. Fauchette relates some verses of Guoyot de Provence, who lived in France about the year 1200, which seem to make mention of the compass under the name of marinette or mariner's stone; which shew it to have been used in France near 100 years before either the Melfite or Venetian. The French even lay claim to the invention, from the fleur de lys wherewith all na tions still distinguish the north point of the card. With as much reason Dr Wallis ascribes it to the English, from its name compass, by which name most nations call it, and which he observes is used in many parts of England to signify a circle.

Though the mariner's compass has been long in use, the best construction of it was attended with many inconveniences, till the improvements which it received from the invention and experiments of Dr Knight, and the farther emendation of Mr Smeaton.

The compass is sometimes observed to be disturbed by the electricity of its glass cover; and this from so slight an application of the finger as was barely necessary to wipe off a little dust. The same glass, rubbed a little more with the finger, a bit of muslin, or paper, would attract either end of the needle so as to hold it to the glass for several minutes, far out

of

Compass, of the due direction, according to that part of the glass which was most excited. And when the needle, after adhering to the glass, has dropped loose, and made vibrations, those would not be bisected as usual by that point where the needle should rest, but would either be made all on one side, or be very unequally divided, by means of some remains of electrical virtue in that part of the glass which had attracted the needle, until at length, after 15 minutes or more, all the electricity being discharged, the magnetical power took place. The remedy for this inconvenience is to moisten the surface of the glass; a wet finger will do it immediately and effectually. The mariner's compass with a chart is much less dangerously moved than the common compass with a bare needle; and the deeper, or farther distant, the needle bangs below the glass, the less disturbance it is likely to receive.

Plate

CLI. fig. I

Fig.

Fig. 3.

Fig. 4.

Improved sea-compasses have lately been constructed by Mr M'Culloch of London, for which he obtained a patent. The particulars are as follows.

Fig. 1. is a section of the steering compass. aa aaa, The common wooden box, with its lid. bb, The brass compass-box. cc, The glass cover to ditto. dd, The hollow conical bottom. e, The prop upon which the compass is supported instead of gimbals; the spherical top of which is finely polished, and the apex of the hollow cone is fitted in a peculiar manner to receive it. ff, A quantity of lead run round the bottom and cone of the compass-box, to balance and keep it steadily borizontal. gg, The card and the magnetical needle, bent in such a manner that the point of the conical pivot on which it moves and is supported, may be brought very near to the centre of gravity, as well as to the centre of motion. h h, Two guards, which by means of two pins ii, affixed to the compass-box, prevent it from turning round and deceiving the steers

man.

Fig. 2. a perspective view of the steering compass, with the lid off and the front laid_open. hh, The guards. 6, The compass-box. e, The prop, &c. as in fig. 1.

Fig. 3. a view of the azimuth compass. b, The compass-box. h, One of the guards. e, The prop, as in fig. 1. and 2, with this difference, that in an azimuth compass, instead of being screwed to the bottom of the wood-box, it stands in a brass socket, and may be turned round at pleasure. 1. A brass bar upon which the sight-vanes are fixed. 2. A dark glass, which moves up and down on 3, the sight-vane. 4. A magnifying glass, which is also moveable on the other vane. 5. The nonius or vernier. 6. A slide for moving the vernier so as to stop the card in taking the azimuth. 7. A double convex-glass, by which the division on the vernier may be read with accuracy.

Fig. 4. is a section representing another application of the magnetic needle and card, constructed by Mr McCulloch. a a a a, The common wood-box. bb, The brass compass-box. cc, The brass support for the circle and pendulum. d, The pendulum. e, The agate. ff, The magnetic needle and card. gg, The brass circle. hh, The glass cover and brass ring. i, The lead weight. N. B. All the centres of motion are in the same plane.

"In one particular this patent compass is considered as an improvement on the common compasses, in as far

as the needle is both longer and broader; hence its Compass. magnetism must be stronger, and of course the line of its magnetic direction correspondent with the card. In another particular, in order to prevent the motions of the vessel from affecting the needle, which is the most desirable object, the patent compass-box, instead of swinging in gimbals at right angles to each other, is supported in its very centre upon a prop; and whatever motion the other parts of the box may have, this centre being in the vertex of the hollow cone, may be considered as relatively at rest; and therefore gives little or no disturbance to the needle. Again, the pivot or centre upon which the needle turns, is so contrived as to stand always perpendicular over the centre of the compass-box, or apex of the hollow cone, as upon a fixed point; and is therefore still less affected by the motions of the vessel. Thus the centres of motion, gravity, and of magnetism, are brought almost all to the same point; the advantages of which will be readily perceived by any person acquainted with mechanical principles." M'Culloch's Account.

The following is a description of Dr Knight's azimuth compass, with the improvements of Mr Smeaton. Plate CLI. fig. 6. is a perspective view of the compass, when in order for observation; the point of view being the centre of the card, and the distance of the eye two feet. AB is the wooden box. C and D are two milled nuts; by means whereof the axes of the inner box and ring are taken from their edges on which they move, and the friction increased, when necessary. EF is the ring that supports the inner box. GH is the inner box; and I is one of its axes, by which it is suspended on the ring EF. The magnet or needle appears passing through the centre together with a small brace of ivory, that confines the cap to its place. The card is a single varnished paper, reaching as far as the outer circle of figures, which is a circle of thin brass; the edge whereof is turned down at right angles to the plane of the card, to make it grow stiff. O is a catgut line, drawn down the inside of the box, for determining the degree upon a brass edge. PQRS is the index bar, with its two stiles and catgut threads; which being taken off from the top of the box, is placed in two pieces, T and V, notched properly_to receive it. W is a place cut out in the wood, serving as a handle.

The use of the azimuth compass is for finding the sun's magnetical azimuth, or amplitude; and thence the variation of the compass. If the observation be for an amplitude at sun-rising, or for an azimuth before noon, apply the centre of the index on the west point of the card, within the box; so that the four lines on the edge of the card, and those on the inside of the box, may meet. If the observation be for the sun's amplitude setting, or an azimuth in the afternoon, turn the centre of the index right against the east point of the card, and make the lines within the box concur with those on the card: the instrument thus fitted for observation, turn the index be towards the sun, till the shadow of the thread a e fall directly on the slit of the sight, and on the line that is along the middle of the index: then will the inner edge of the index cut the degree and minute of the sun's magnetical azimuth from the north or south. But note, that if, when the compass is thus placed, the azimuth is less than 54°

from

The sun's magnetical amplitude thus found, the va. riation of the needle is thus determined. Being out at sea the 15th of May, 1717 in 45 north latitude, the tables gave me the sun's latitude 19° north, and his east amplitude 27° 25' north; by the azimuth compass, I find the sun's magnetical amplitude at his rising and setting; and find he rises, e. gr. between the 62d and 63d degree, reckoning from the north towards the east point of the compass, i. e. between the 27th and 28th degree, reckoning from the east. The magnetical amplitude, therefore, being here equal to the true one, the needle has no variation; but if the sun at his rising should have appeared between the 52d and 53d degree from the north towards the east, his magnetical amplitude would then have been between 37 and 38 degrees, i. e. about ten degrees greater than the true amplitude: therefore the needle would vary about 10 degrees north-easterly. If the magnetical east amplitude found by the instrument should be less than the true amplitude, their difference would show the variation of the needle easterly. If the true east amplitude be southward, as also the magnetical amplitude, and this last be the greater; the variation of the needle will be north-west; and vice versa.

What has been said of north-east amplitudes holds also of south-west; and what of south-east amplitudes holds of north-west. Lastly, if amplitudes be found of different denominations, e. gr. if the true amplitude be six degrees north, and the magnetical amplitude be six degrees south; the variation, which in this case is north-west, will be equal to the sum of the magnetical and true amplitudes; understand the same for west amplitudes.

The variation may likewise be found from the azimuth but in that case the sun's declination, latitude of the place, and his altitude, must be given, that his true azimuth may be found.

This instrument is also useful for settling the ship's wake, in order to find the leeway; and also to find the bearings of headlands and other objects.

Experience evinces, that the needle of a compass, like every other magnet, whether natural or artificial, continually loses something of its magnetic powers, which frequently produces a difference of more than a point; and we may venture to assert, that the great errors in ships reckonings more commonly originate from the incorrectness of the compass than from any other cause.

Steel cannot be too highly tempered for the needle of a sea-compass, as the more it is hardened, the more permanent is the magnetism it receives; but to preserve the magnetism, and of course the polarity of the needle, it should be cased with thin, well-polished, soft iron. It has been found by repeated experiments, that the cased needle preserves its magnetism in a more perfect degree than a needle not cased; and perhaps

This is not an opinion hastily adopted, but the result of a fair and judicious trial, as the gentleman from whom the above observations were in substance taken, placed a cased and uncased needle in a room for three months, having at first exactly the same direction, and about the same degree of force. At the end of this period it was found that the cased needle had not in the least changed its direction, while the uncased had varied two degrees, and its magnetic power was considerably diminished.

These remarks have the air of novelty, and may perhaps contribute to the improvement of the compass. But the defects of this instrument are not confined to the needle. The heaviest brass compasses are not to be implicitly trusted in a hollow or high sea, as they have the box hanging in two brass rings, thus allowing it to have only two motions, both vertical and at right angles with each other; by which confinement of the box, upon any succussion, particularly sudden ones, the card is always too much agitated, and before it can recover itself, another jerk prevents it from pointing to the pole. It is even not uncommon to see the card unshipped by the violence of the ship's pitching.

All these defects are abundantly supplied by giving the box a vertical motion at every degree and minute of the circle, and combining these motions with a horizontal one of the box as well as of the card. By this disposition of the box, the effects of the jerks on the card are avoided, and it will always with steadiness point to the pole. Mr Bernard Romans found by experience, that the card not only is not in the smallest degree affected by the hollow sea, but that, in all the violent shocks and whirlings which it is possible for the box to receive, the card lies as still as in a room unaffected by the least motion.

A compass was recently invented and made in Holland having all these motions. It is about the size of the brass compass commonly used. The bottom of the brass box, instead of being shaped like a bowl, must be a hollow cone resembling the bottom of a common glass bottle; the vertex of it must be raised so high as to leave only one inch between the card and the glass; the box must be of the usual depth, and a quantity of lead must be poured in the bottom of the box, round the base of the cone, which secures it on the stile whereon it traverses.

This stile is firmly fixed in the centre of a square wooden box, like the common compass, but with a thicker bottom. The stile is made of brass about six inches long, round, and one-third of an inch thick, having its head blunt like that of a sewing thimble, and of a fine polish, and placed perpendicular. The inner vertex of the cone must likewise be well polished. The vertical part of the cone ought to be sufficiently thick to allow a well-polished cavity for holding a short stile, proceeding from the centre of the card on which it traverses. "The compass I saw, (says Mr Romans), was so constructed; but I see no reason why the style might not proceed from the centre of the vertex of the cone, and so be received by the card the common way. The needle must be a mag