is most generally understood to signify the whole extent of air diffused around this earth, the sun, moon, or any other great body in the universe.

With regard to the component parts of the air we breathe, Dr Priestley hath by undoubted experiments proved them to be the nitrous acid, earth, and phlogiston. To these we may certainly add water, of which the substances he made use of in producing air could never be perfectly free; and as the Doctor inclines to think that the electric fluid may be the phlogiston itself, we will thus have the subtle fluid of electricity as a capital ingredient in the composition of the atmosphere of the earth.

Indeed, whether this fluid is admitted by the Doctor or not as an ingredient in the air he distills, we are absolutely certain that it enters in no small quantity into the composition of the air we breathe. It is also certain, that the higher up we go in the atmospheric regions, the quantity of electrical fluid is apparently greater; neither hath any means been as yet suggested, by which we can determine where this electrical power begins to diminish, much less where it ceases altogether. Hence some have imagined, that though the atmosphere of the earth, as consisting of an heterogeneous mixture of a great number of fluids, extends but for a little way, yet that the electrical part of it may extend to the moon, or farther, and be the cause of the revolutions of that luminary; and that a similar affection of the solar atmosphere may occasion the revolutions of planets and comets.

The first-rate geometricians of modern times have been very solicitous to find out the height of the atmosphere; but the high degree of electricity always existing in the upper parts of it being but very lately, and even yet imperfectly, discovered, it was necessarily overlooked by them, so that their calculations were made without any regard to it, or without suspecting the existence of any such thing. The first attempts that were made with this view were soon after the weight of the atmosphere was found out. It having been discovered, that a column of air whose base was an inch square, and the height of it that of the whole atmosphere, weighed 15 lb., and that the weight of air was to Mercury as 1 to 10,800, it thence followed, that if the weight of the atmosphere was sufficient to raise a column of Mercury to the height of 30 inches, the height of the aerial column itself behoved to be 10,800 times as much, or a little more than five miles high. But as the air hath also a very great elastic power, by which it expands itself when the pressure of the rest is taken off from any part of it, it is impossible the foregoing calculation can be just. Another method therefore behoved to be followed. It being found by repeated experiments in different countries, that the spaces which any portion of air takes up are reciprocally proportional to the weights with which it is compressed, allowances for the gradual decrease of weight behoved to be made in calculating the height of the atmosphere. If we suppose the height of the whole atmosphere divided into innumerable equal parts, the density of each of which is as its quantity, and the weight of the whole incumbent atmosphere being also as its quantity, it is evident, that the weight of the incumbent air is everywhere as the quantity contained in the subjacent part; which makes a difference between the weight of each two contiguous parts of air. By a theorem in geometry, where the differences of magnitudes are geometrically proportional to the magnitudes themselves, those magnitudes are in continual arithmetical proportion; therefore, if, according to the supposition, the altitude of the air by the addition of new parts into which it is divided, do continually increase in arithmetical proportion, its density will be diminished, or, which is the same thing, its gravity decreased, in continual geometrical proportion.

From such a series it is easy, by making two or three barometrical observations, and determining the rarity of the air at two or three different stations, to determine its rarity at any assignable height. Calculations accordingly were made upon this plan; but it having been found that the barometrical observations by no means corresponded with the density which by other experiments the air ought to have had, it was suspected that the upper parts of the atmospheric regions were not subject to the same laws with the lower ones. Another method therefore was had recourse to, namely, by calculating the height from which the light of the sun was refracted so as to become visible to us before the sun himself arose. Thus, it was determined, that at the height of 45 miles the atmosphere had no power of refracting light; and therefore, that, if it extended beyond that distance, it behoved to be the next thing to a perfect vacuum, and not to be regarded.

This theory being extremely plausible, very soon became general; and the height of the atmosphere was commonly spoken of as familiarly as the height of a mountain, and reckoned to be as certainly, if not more certainly, calculated than the heights of mountains are. Some appearances, however, seemed to form insuperable objections. The most remarkable were those globes of fire called meteors, which sometimes appear, and are found to move at vast heights above the earth. A very remarkable one of this kind was seen by Sir Hans Sloane, and the account communicated to Dr Halley, who seems to have been greatly embarrassed by the phenomena attending it. The account given by Sir Hans Sloane is as follows: That "on Tuesday, March 19th 1742, about eight in the evening, passing eastward by the north-east corner of Southampton street in Bloomsbury Square, London, he saw of a sudden a very great light much surpassing that of the moon, which shone then very bright. Upon turning to observe it, he saw a long stream of very bright fire, branched in the middle; but at last it came to be pear-shaped, tapering upwards, and afterwards spherical, though not so big as the full moon. The colour of it was whitish, with a shade of blue, of a most vivid dazzling lustre, which seemed in brightness very nearly to resemble, if not to surpass, that of the body of the sun in a clear day. In about half a minute or less, it seemed to move over about 20° of the heavens, and to go out as much above the horizon, leaving a visible track behind it." All the other observers of this phenomenon nomenon agreed, that the splendour of this meteor was very little inferior to that of the sun; that within doors the candles scarce gave any light; and in the streets, not only the stars disappeared, but the moon, though then nine days old, and very near the meridian, could scarcely be seen, at least the east no visible shade even where the beams of the meteor were intercepted by the houses; so that for a few seconds, in all respects it resembled perfect day.

Sir Hans Sloane, at London, observed this meteor about the Pleiades, descending a little beyond and below the stars in the belt of Orion. At Oxford, from the track it had left in the sky, it was found to have passed about 1½ above the preceding shoulder of Orion, and about 3½ above the middle of his belt, where there appeared a luminous nebula of reddish light, being a dilatation of the track, seeming to have been occasioned by some explosion there; and there, the observer was informed it first broke out. From this it proceeded, as to sense, in the arch of a great circle, and passing in the middle between the tail of Lepus (Bayero) and β in the forefoot of Canis Major, it terminated about 5 in the breast of the same; and at the place of its extinction there remained a large whitish nebula, much broader and of a stronger light than the rest of the track, by which a very strong explosion was thought to be indicated. At Worcester it was observed to have left all Orion and Canis Major to the westward, and divided the distance between Sirius and Procyon, so as to be almost twice as far from Procyon as from Sirius.

From these observations, the distances of the three cities London, Oxford, and Worcester being known, as also the sun's place at that time, and the altitudes of the stars among which it passed, Dr Halley computes the height of the meteor to have been between 69 and 73½ English miles from the surface of the earth, its diameter to have been 2800 yards, upwards of an English mile and a half, and its velocity more than 350 miles in a minute.

The perplexing circumstances here are, that, at such an height, the atmosphere of the earth ought to have no density sufficient to sustain flame of any kind, much less such an intensely dazzling one, and of such magnitude, as this meteor was. Add to this, that, without air, no sound could possibly have attended its explosion, nor indeed could any explosion have taken place; yet all accounts from Devonshire and Cornwall agreed that there was heard there a report as of a very great cannon, or rather a broadside at some distance, followed by a rattling noise, as if some small arms had been providentially discharged. The same was heard at London and in Sussex; nor was it known how far it extended, as we have certain accounts that it was heard beyond the city of Aberdeen in the north of Scotland. What was peculiar to this sound was, that it was accompanied with an uncommon tremor in the air, so as to shake the glass-windows and doors, and, according to some, even the houses themselves, much beyond the usual effect of cannon though near. The learned Dr Halley acknowledges himself unable to reconcile these circumstances with the received theory of the height of the atmosphere; as in the regions in which this meteor moved, the air ought to have been 300,000 times rarer than what we breathe, and the next thing to a perfect vacuum. He offers a kind of dubious conjecture indeed, that the extreme magnitude of the meteor might have compensated for the fineness of the medium. But this we think will hardly account for all the phenomena. According to this supposition, the explosion of a globe of fire 2800 yards in diameter in such a medium ought to be equivalent to the explosion of one 300,000 times less in the denser regions of our atmosphere. But globes are to one another as the cubes of their diameters; therefore, dividing 2800 by the cube-root of 300,000, we have the diameter of a sphere of fire, which ought to produce as extensive effects as this meteor, supposing it to explode at the surface of the earth, or in any part of the atmosphere of nearly the same density. The diameter of such a globe behoved to be almost 42 yards, or 126 feet; and though we must acknowledge the effects of such an explosion to be prodigious in those places that were in its neighbourhood, it is scarce probable that it could extend in such a manner over the whole island of Britain. The truth is, however, that here we have very few data to go upon; the largest flashes of lightening, or meteors that have been observed in the lower regions of the atmosphere, bearing but a very small proportion to a globe of 42 yards diameter.

The greatest difficulty, however, is to account for the brightness of the light. The appearance of meteors of this kind, and indeed of all kinds, are now generally attributed to electricity; but still the difficulty remains. We know that the electrical fluid pervades the vacuum of the air-pump with the utmost facility; but then it appears in long streams resembling the aurora borealis, not in small bright and concentrated sparks, as when drawn from a conductor, or discharged from a vial in the open air. To make this fluid exhibit a very bright vivid flash, the presence of the gross atmosphere seems necessary; and where that is taken off, the electric spark always diffuses itself over a large surface, and therefore becomes proportionally less bright. Experience shows, that though an electrified bottle will discharge itself through a great space of vacuum, yet the spark diverges and loses its force and brightness; whereas this meteor, which at first was long and branched, collected itself afterwards into a lesser compass; which is a very strong presumption of the pressure of a denser fluid; not to mention, that an explosion can be only made by a rare fluid forcing through a dense one. Thus gun-powder fired in the open air explodes, because the flame forces violently against the dense air; but though heated ever so much in vacuo, no such explosion is produced.

Instances have been known of balls of fire similar to this meteor, though vastly inferior in size and brightness, travelling along the surface of the ground, or along the sea, and afterwards bursting with an explosion. A very remarkable one of this kind is mentioned by Priestley in his history of electricity. The substance of the account is, that as the observer (Mr Chalmers) was endeavouring to find the latitude, on board the Montague, Nov. 4th 1749, in Lat. 42° 48', about 10 minutes before 12, he was desired by one of the quarter-masters to look to the windward. Upon which he observed a large ball of blue fire rolling on the surface of the water at about three miles distance from them. They immediately lowered their top-sails; but before before they could raise their main tack, it was advanced within 40 or 50 yards of the main chains. It now appeared as big as a large millstone; and, rising perpendicularly, went off with an explosion as if hundreds of cannon had been fired at once. The noise lasted about half a second; after which they found their main-topmast shattered to pieces, and their main-mast rent quite down to the keel; five men were knocked down, and one greatly burnt, &c.

Now, though we can by no means pretend to give a reason why such large bodies of electric fire should thus be found travelling as it were in quest of adventures; yet as it seems indisputable that they actually do move in this unaccountable manner, both on the surface of the earth and at the height of 30, 40, 50, or 70 miles above it, it seems to some extremely probable, that, excepting the mere want of aqueous vapours, the atmosphere even at that great height is not much different in density from what we breathe. To those who consider the effects of the electric fluid upon light bodies on earth, the decrease of gravity in the superior regions of the atmosphere will be no argument of its want of density. We know, that it is the nature of any electrified substance to attract light bodies; and that, by proper management, they may even be suspended in the air, without either moving up or down, for a considerable time. If this is the case with light terrestrial bodies, it cannot be thought very improbable that the aerial particles themselves should be thus affected in those regions where electricity is so abundant. It is possible, therefore, that where the air is in a highly electrified state, its tendency towards the earth or its gravity may be very much diminished, without its density being at all affected; and if this is the case, it will no doubt occasion great difficulties in affixing the true height of the atmosphere, by rendering every barometrical observation exceedingly precarious.

From these considerations, and many others that will naturally occur to every one who attends to this subject, it must appear that the height of our atmosphere is yet very far from being determined. At first the tinkling of the barometer was thought to be a certain method of determining the degree of the atmospheric density on the tops of mountains, and in some of these it was even said that the air became too subtle for breathing; but this is now found to be a mistake. The French mathematicians, when on the top of one of the Andes, and above the common region of clouds, made no complaint of this kind. On the top of mount Etna, where the smoke of the mountain sinks instead of rising, Mr Brydone found no inconvenience in this respect. Sir William Hamilton indeed says he did; but besides the conjecture mentioned under the article Etna, we apprehend that the respiration may be affected on the top of a volcano, from so many different causes, that nothing can be concluded from thence. With regard to the barometer, M. De Luc has been at incredible pains to ascertain and reduce to rule the irregularities of it, which he found to arise from the different temperature of the air; and it is to be hoped that by his labours the mensuration of altitudes by this instrument will be greatly facilitated, and the conclusions rendered much more certain than before, though whether the difficulties are entirely removed can only be determined by future experience. (See Barometer).

As all bodies which are immersed in the atmosphere of our earth must sustain its pressure, which is various at various times, it is plain, that this variation must occasion very considerable changes in these bodies. We have already mentioned, that the pressure of the atmosphere upon every square inch of the earth's surface is equivalent to 15 pounds. Hence, Dr Cotes hath computed that the pressure of the whole ambient fluid upon the earth's surface is equivalent to that of a globe of lead 60 miles in diameter. From this it also appears, that the pressure upon the human body must be very considerable; for, as every square inch of surface sustains a pressure of 15 pounds, every square foot, containing 144 square inches, must sustain a pressure of 2160; and therefore, if a man's body contains 15 square feet of surface, which is pretty near the truth, he must sustain a weight of $2160 \times 15 = 32,400$ pounds, or fifteen tons, for his ordinary load, which, by the diminution or increase of the gravity of the atmosphere, may become heavier or lighter by near a whole ton. By why we are this enormous pressure we should certainly be crushed to pieces in a moment, were not all parts of our bodies filled either with air, or with some other elastic fluid, the spring of which is sufficient to counterbalance the weight of the atmosphere. Whatever this fluid is, we are sure that it is just able to counterbalance the atmospheric gravity, and no more; for if any considerable pressure is superadded to that of the air, as by going into deep water, &c., it is always severely felt, let it be ever so equable. If the pressure is taken off from any part of the body, as by putting the hand upon the mouth of an open receiver which is afterwards exhausted, the weight of the atmosphere then discovers itself, and we immediately feel strongly sucked down into the glass. See Pneumatics.

To the above-mentioned changes in gravity of the atmosphere, we owe the prognostications of the weather by the barometer; but these changes seem to take place mostly in such places as are at a distance from the equator. In the torrid zone, though frequent and very great changes of weather happen, the barometer stands commonly at the same height, or with little variation. See Barometer.

For the causes of the elasticity of the atmosphere, see the article Elasticity; for the formation and ascent of vapour, see Evaporation; and for the other natural operations, see Congelation, Cold, Rain, Hail, Snow, Meteors, &c.