But Mr. Apinus has fhewn that this fuppofition is inadmiffible, becaufe agreeable to the hypothefis the magnetic nucleus must be very profound, and it is well known that the folar heat does not penetrate to very great depths; there are caves at no great diftance from the furface of the earth, in which a thermometer remains always at the fame height. The diurnal heat does not penetrate even thefe, there is therefore no probability of it's effects extending to ftill greater depths.

OF THE DIP OF THE NEEDLE.

If a needle, which is accurately ballanced and fufpended fo as to turn freely in a vertical plane, be rendered magnetical, the north pole will be deprejfed, and the fouth pole elevated above the borizon this property is called the dip of the horizon. As it is very difficult to ballance a needle accurately, the poles are generally reverfed by a magnet, fo that it's two ends may dip alternately, and the mean of the two is taken.

This property was difcovered by Robert Norman, about the year 1576. I fhall give the account of the difcovery in his own words.

"Having (fays he) made many and divers compaffes, and ufing always to finish and end them before I touched the needle, I found continually that after I had touched the yrons with the ftone, that prefently the north point thereof would bend or decline downwards under the horizon in fome quantity; infomuch, that to the flie of the compafs, which before was made equal, I was ftill contrained to put fome fmall piece of wax in the fouth part thereof, to coun-. terpoife this declining, and to make it equal again.

"Which effect having many times paffed my VOL. IV.

Hh

hands

hands without any great regard thereunto, as ignorant of any fuch property in the ftone, and not before having heard nor read of any fich matter; it chanced at length that there came to my hands an inftrument to be made, with a needle of fix inches long, which needle after I had polished, cut off at juft length, and made to ftand level upon the pin, fo that nothing refted but only the touching of it with the ftone: when I had touched the fame, prefently the north part thereof declined down in fuch fort, that being conftrained to cut away fome of that part to make it equal again, in the end I cut it too fhort, and fo fpoiled the needle wherein I had taken fo much pains.

"Hereby being stroken into some cholar, I applied myself to feek further into this effect, and making certain learned and expert men (my friends) acquainted in this matter, they advised me to frame fome inftrument, to make fome exact trial, how much the needle touched with the stone would decline, or what greateft angle it would make with the plane of the horizon." Thus far Mr. Norman.

The dip is faid to be fubject to a variation. At this time in London it is about 72 degrees; from fome late obfervations it appears to diminish about fifteen minutes in four years. The nature of this phenomenon is pleafingly illuftrated by carrying a small dipping needle from one end of a magnetic bar to the other; when it ftands over the fouth pole, the north end of the needle will be directed perpendicularly to it; as the needle is moved, the dip will grow lefs, and when it comes to the magnetic center it will be parallel to the bar; afterwards the fouth end will dip, and the needle will ftand perpendicular to the bar, when it is directly over the north pole.

Of

OF THE INFLUENCE OF THE AURORA BOREALIS ON THE MAGNETIC NEEDLE.

Meffrs. Wilcke and Van Swinden have clearly proved, that there is a connection between the aurora borealis and the magnetic needle; they have fhewn it to be fo evident, fo general, and fo conftant; that no one who examined the affections of the one and the other with attention, could, have any doubts on the fubject. It remained however for Mr. Dalton* to give a complete and fatisfactory account of this connection, and it is with great pleasure I take this opportunity of recommending his work to your attentive pe

rufal.

From various obfervations he has fhewn,t 1. When the aurora appears to rife only about 5° 10' or 15° above the horizon, the needle is very little disturbed, and often infenfible. 2. When it rifes up to the zenith, and paffes it, there never fails to be a confiderable difturbance. 3. This difturbance confists in an irregular ofcillation of the horizontal needle, fometimes to the eastward, then to the weftward of the mean daily pofition, in fuch fort that the greateft excurfions on each fide are nearly equal, and amount at Manchester to about half a degree on each fide. 4. When the aurora ceafes, or foon after, the needle returns to it's former ftation.

From these facts alone, fays Mr. Dalton, independent of other obfervations, we cannot avoid inferring, that there is fomething magnetic conftantly in the higher regions of the atmosphere, Hh 2

that

Meteorological Obfervations and Effays, by John Dalton,

1793.
+ Dalton's Obfervations, p. 183.

that has a fhare at least in guiding the needle; and that the fluctuations of the needle, during the aurora, are occafioned by fome mutations that then take place in this magnetic matter, in the incumbent atmosphere.

OF THE SIMILARITY BETWEEN ELECTRICITY AND MAGNETISM.

The powers of magnetifm, like thofe of electricity, are excited and feparated by friction. This effect is wonderful in both, but more fo in magnetifm, where two powers, naturally attracting each other, remain feparated in the fame fteel bar for many years, and yet they may be reduced to their natural ftate by the friction of two other magnets, acting in a contrary order to that by which the poles were originally feparated.

Magnetism and electricity act powerfully at corners, edges, and points.

Magnetism may be communicated to a small fteel needle, by paffing the difcharge of a large electrical battery through it.

The difcharge of an electrical battery through a fmall magnetic needle, will fometimes deftroy the magnetifm, and fometimes invert the poles of the magnet. Similar effects have been produced by lightning.

OF THE THEORY OF MAGNETISM.

Here, as in other parts of natural philofophy, we must content ourselves with mere conjecture. Of the various hypothefes that have been formed to account mechanically for the phenomena of magnetifm, that of Mr. Prevost is undoubtedly the

beft;

* Prevost De l'Origine des Forces Magnetiques à Genêve, 1788.

beft; but as it depends on a knowledge of Mr. Lè Sage's mechanical fyftem of the univerfe, it will be impoffible for me to lay it before you in a fatisfactory manner; you must therefore be contented with a very imperfect sketch thereof.

HYPOTHESIS.

There exifts in and about our globe, a very fubtil fluid poffeffing the following properties.

1. It is expanfive, and confequently difcrete. 2. The molecules of this fluid are formed by the union of two kinds of elements, A, B, united by affinity.

3. The elements of the different kinds have a greater tendency to each other, than thofe of the fame kind.

4. That excepting the preceding property (n. 3), thefe attractions follow the fame laws as univerfal gravitation.

5. This fluid has an affinity with the particles of iron, and which probably acts only at contact, or when very nearly in contact. This fluid is decompofed by iron, and feldom by any thing elfe.

The foregoing properties of the magnetic fluid. may be all mechanically explained on the principles of Mr. Le Sage.

To explain the magnetic phenomena of the earth, it is fufficient to fuppofe that one aliment of the magnetic fluid is furnished by nature, in greater abundance in one hemifphere than the other; or that a small portion thereof be decompofed by fome of the caufes perpetually acting in nature, by which means the terreftrial globe is maintained in a charged ftate, having a greater abundance

Hh3