20

Construction of pumps in the last century.

Fig. 4.

21

doctrines) found in this fancied horror of a fancied mind (what else is this that nature abhors?) a ready solution of the phenomena. We shall state the facts, that every reader may see what kinds of reasoning were received among the learned not two centuries ago.

Pumps were then constructed in the following manner: A long pipe GB (fig. 4.) was set in the water of the well A. This was fitted with a sucker or piston C, having a long rod CF, and was furnished with a valve B at the bottom, and a lateral pipe DE at the place of delivery, also furnished with a valve. The fact is, that if the piston be thrust down to the bottom, and then drawn up, the water will follow it; and upon the piston being again pushed down, the water shuts the valve B by its weight, and escapes or is expelled at the valve E; and on drawing up the piston again the valve E is shut, the water again rises after the piston, and is again expelled at its next descent.

The Peripatetics explain all this by saying, that if the water did not follow the piston there would be a void Their ope- between them. But nature abhors a void; or a void ration acis impossible therefore the water follows the piston.counted for by the It is not worth while to criticise the wretched reasoning Peripate in this pretence to explanation. It is all overturned by one observation. Suppose the pipe shut at the bottom, the piston can be drawn up, and thus a void produced. No, say the Peripatetics; and they speak of certain spirits, effluvia, &c. which occupy the place. But if so, why needs the water rise? This therefore is not the cause of its ascent. It is a curious and important phe

• tics.

Galileo first The sagacious Galileo seems to have been the first who accounted seriously ascribed this to the weight of the air. Many before him had supposed air heavy; and thus explained the difficulty of raising the board of bellows, or the piston of a syringe, &c. But he distinctly applies to this allowed weight of the air all the consequences of hydrostatical laws; and he reasons as follows.

23 by the

the atmo

phere,

The heavy air rests on the water in the cistern, and weight of presses it with its weight. It does the same with the water in the pipe, and therefore both are on a level : but if the piston, after being in contact with the surface of the water, be drawn up, there is no longer any pressure on the surface of the water within the pipe; for the air now rests on the piston only, and thus occasions a difficulty in drawing it up. The water in the pipe, therefore, is in the same situation as if more water were poured into the cistern, that is, as much as would exert the same pressure on its surface as the air does. In this case we are certain that the water will be pressed into the pipe, and will raise up the water already in it, and follow it till it is equally high within and without. The same pressure of the air shuts the valve E during the descent of the piston. (See Gal. Discourses).

24

which wa

rise in them.

and preHe did not wait for the very obvious objection, that dicted the if the rise of the water was the effect of the air's pressure, height to it would also be its measure, and would be raised and ter would supported only to a certain height. He directly said so, and adduced this as a decisive experiment. If the horror of a void be the cause, says he, the water must rise to any height however great; but if it be owing to the pressure of the air, it will only rise till the weight of the water in the pipe is in equilibrio with the pressure of the air, according to the common laws of hydrostatics. And be adds, that this is well known; for it is a fact, that

pumps will not draw water much above forty palms, although they may be made to propel it, or to lift it to any height. He then makes an assertion, which, if true, will be decisive. Let a very long pipe, shut at one end, be filled with water, and let it be erected perpendicnlarly with the close end uppermost, and a stopper in the other end, and then its lower orifice immersed into a vessel of water; the water will subside in the pipe upo removing the stopper, till the remaining column is in equilibrio with the pressure of the external air. This experiment he proposes to the curious; saying, however, that he thought it unnecessary, there being already such abundant proofs of the air's pressure.

25

ricelli's ex

It is probable that the cumbersomeness of the neces- His predicsary apparatus protracted the making of this experiment, tion veriAnother equally conclusive, and much easier, was made fied by Toin 1642 after Galileo's death, by his zealous and learned periment. disciple Toricelli. He filled a glass tube, close at one end, with mercury; judging, that if the support of the water was owing to the pressure of the air, and was the measure of this pressure, mercury would in like manner be supported by it, and this at a height which was also the measure of the air's pressure, and therefore 13 times less than water. He had the pleasure of seeing his expectation verified in the completest manner; the mercury descending in the tube AB (fig. 5.), and finally settling Fig. 5. at the height fB of 29 Roman inches: and he found, that when the tube was inclined, the point f was in the same horizontal plane with fin the upright tube, according to the received laws of hydrostatical pressure. The experiment was often repeated, and soon became famous, exciting great controversies among the philosophers about the possibility of a vacuum. About three years afterwards the same experiment was published, at Warsaw in Poland, by Valerianus Magnus, as his own suggestion and discovery: but it appears plain from the letters of Roberval, not only that Toric lli was prior, and that his experiment was the general topic of discussion among the curious; but also highly probable that Valerianus Magnus was informed of it when at Rome, and daily conversant with those who had seen it. He denies, however, even having heard of the name of Toricelli.

This was the era of philosophical ardour; and we think that it was Galileo's invention and immediate application of the telescope which gave it vigour. Discoveries of the most wonderful kind in the heavens, and which required no extent of previous knowledge to understand them, were thus put into the hands of every person who could purchase a spy-glass; while the high degree of credibility which some of the discoveries, such as the phases of Venus and the rotation and satellites of Jupiter, gave to the Copernican system, immediatly set the whole body of the learned in motion. Galileo joined to his ardour a great extent of learning, particularly of mathematical knowledge and sound logic, and was even the first who formally united mathematics with physics; and his treatise on accelerated motion was the first, and a precious fruit of this union. About the years 1642 and 1644, Origin of we find clubs of gentlemen associated in Oxford and Lon- the Royal don for the cultivation of knowledge by experiment; and Society, before 1655 all the doctrines of hydrostatics and pneuma- &c. tics were familiar there, established upon experiment. Mr Boyle procured a coalition and correspondence of these clubs under the name of the Invisible and Philosophical Society. In May 1658, Mr Hooke finished for Mr

Boyle

26

27

Invention of the air

pump.

head of the republic of letters, and cannot brook the concurrence of any foreigners. Roberval was in this instance, however, the champion of Toricelli; but those who know his controversies with the mathematicians of France at this time will easily account for this exception. All now agree in giving Toricelli the honour of the unjustly. first invention; and it universally passes by the name of the TORICELLIAN EXPERIMENT. The tube is called the TORICELLIAN TUBE; and the space left by the mercury is called the TORICELLIAN VACUUM, to distinguish it from the BOYLEAN VACUUM, which is only an extreme rarefaction.

Boyle an air-pump, which had employed him a long time, and occasioned him several journeys to London for things which the workmen of Oxford could not execute. He speaks of this as a great improvement on Mr Boyle's own pump, which he had been using some time before. Boyle therefore must have invented his air-pump, and was not indebted for it to Schottus's account of Otto Guerick's, published in his (Schottus) Mechanica Hydraulo-pneumatica in 1657, as he asserts (Techna Curiosa). The Royal Society of London arose in 1656 from the coalition of these clubs, after 15 years co-operation and correspondence. The Montmorine Society at Paris had subsisted nearly about the same time; for we find Paschal in 1648 speaking of the meetings in the Sorbonne College, from which we know that society originated. Nuremberg, in Germany, was also a distinguished seminary of experimental philosophy. The magistrates, sensible of its valuable influence in many manufactures, the source of the opulence and prosperity of their city, and many of them philosphers, gave philosophy a professed and munificent patronage, furnishing the philosophers with a copious apparatus, a place of assembly, and a fund for the expence of their experiments; so that this was the first academy of sciences out of Italy under the patronage of government. In Italy, indeed, there had long existed institutions of this kind. Rome was the cenThe expe- tre of church-government, and the resort of all expectriments of ants for preferment. The clergy was the majority of the learned in all Christian nations, and particularly of the systematic philosophers. Each, eager to recommend himself to notice, brought forward every thing that was curious; and they were the willing vehicles of philosophical communication. Thus the experiments of Galileo and Toricelli were rapidly diffused by persons of rank, the dignitaries of the church, or by the monks their obsequious servants. Perhaps the recent defection of EngJand, and the want of a residing embassy at Rome, made her sometimes late in receiving or spreading philosophical researches, and was the cause that more was done there proprio Marte.

28

Galileo and Toricelli

rapidly dif

fused.

29

The merit of Toricel

We hope to be excused for this digression. We were naturally led into it by the pretensions of Valerianus Magnus to originality in the experiment of the mercury li's clained supported by the pressure of the air. Such is the by others strength of national attachment, that there were not wanting some who found that Toricelli had borrowed his experiment from Honoratus Fabri, who had proposed and explained it in 1641 but whoever knows the writings of Toricelli, and Galileo's high opinion of him, will never think that he could need such helps. (See this surmise of Mounier in Schott. Tech. Cur. III. at the end).

Galileo must be considered as the author of the experiment when he proposes it to be made. Valerianus Magnus owns himself indebted to him for the principle and the contrivance of the experiment. It is neither wonderful that many ingenious men, of one opinion, and instructed by Galileo, should separately hit on so obvious a thing; nor that Toricelli, his immediate disciple, his enthusiastic admirer, and who was in the babits of corresponding with him till his death in 1642, should be the first to put it in practice. It became the subject of dispute from the national arrogance and self-conceit of some Frenchmen, who have always shown themselves disposed to consider their nation as at the

31

32

The experiment was repeated in various forms, and It was rewith apparatus which enabled philosophers to examine peated in several effects which the vacuum produced on bodies ex- various forms. posed in it. This was done by making the upper part of the tube terminate in a vessel of some capacity, or communicate with such a vessel, in which were included along with the mercury bodies on which the experiments were to be made. When the mercury had run out, the phenomena of these bodies were carefully observed. An objection was made to the conclusion drawn from An objecToricelli's experiment, which appears formidable. Iftion to the the Toricellian tube be suspended on the arm of a ba- conclusion lance, it is found that the counterpoise must be equal to drawn from the weight both of the tube and of the mercury it contains. This could not be, say the objectors, if the mercury were supported by the air. It is evidently supported by the balance; and this gave rise to another notion of the cause different from the peripatetic fuga vacui: a suspensive force, or rather attraction, was assigned to the upper part of the tube.

But the true explanation of the phenomenon is most easy and satisfactory. Suppose the mercury in the cistern and tube to freeze, but without adhering to the tube, so that the tube could be freely drawn up and down. In this case the mercury is supported by the base, without any dependence on the pressure of the air; and the tube is in the same condition as before, and the solid mercury performs the office of a piston to this kind of syringe. Suppose the tube thrust down till the top of it touches the top of the mercury. It is evident that it must be drawn up in opposition to the pressure of the external air, and it is precisely similar to the syringe mentioned in N° 16. The weight sustained therefore by this arm of the balance is the weight of the tube and the downward pressure of the atmosphere on its top.

it obviated

33

The curiosity of philosophers being thus excited by Galileo's this very manageable experiment, it was natural now to original extry the original experiment proposed by Galileo. Ac-periment cordingly Berti in Italy, Paschal in France, and many performed. others in different places, made the experiment with a tube filled with water, wine, oil, &c. and all with the success which might be expected in so simple a matter: and hence the doctrine of the weight and pressure of the air was established beyond contradiction or doubt. All this was done before the year 1648.-A very beautiful experiment was exhibited by Auzout, which completely satisfied all who had any remaining doubts.

34

A small box or phial EFGH (fig. 6.) had two glass An experitubes, AB, CD, three feet long, inserted into it in such ment by a manner as to be firmly fixed in one end, and to reach Auzout nearly to the other end. AB was open at both ends, Fig. 6. and CD was close at D. This apparatus, was completely filled with mercury, by unscrewing the tube AB, fil

35

decisive of the ques

tion.

36

ty of the air there

fore a sta

fing the box, and the tube CD; then screwing in the tube AB, and filling it: then holding a finger on the orifice A, the whole was inverted and set upright in the position represented in figure 8, immersing the orifice A (now a of fig. 8.) in a small vessel of quicksilver. The result was, that the mercury ran out at the orifice a, till its surface m n within the phial descended to the top of the tube ba. The mercury also began to descend in the tube de (formerly DC) and run over into the tube ba, and run out at a, till the mercury in de was very near equal in a level with m n. The mercury descending in ba, till it stood at k, 294 inches above the surface op of the mercury in the cistern, just as in the Toricellian tube.

The rationale of this experiment is very easy. The whole apparatus may first be considered as a Toricellian tube of an uncommon shape, and the mercury would flow out at a. But as soon as a drop of mercury comes out, leaving a space above m n, there is nothing to keep up the mercury in the tube d c. Its mercury therefore descends also; and running over into ba, continues to supply its expence till the tube de is almost empty, or can no longer supply the waste of ba. The inner surface therefore falls as low as it can, till it is level with b. No more mercury can enter ba, yet its column is too heavy to be supported by the pressure of the air on the mercury in the cistern below; it therefore descends in ba, and finally settles at the height ho, equal to that of the mercury in the Toricellian tube.

The prettiest circumstance of the experiment remains. Make a small hole g in the upper cap of the box. The external air immediately rushes in by its weight, and now presses on the mercury in the box. This immediately raises the mercury in the tube de tol, 29 inches above mn. It presses on the mercury at k in the tube ba, balancing the pressure of the air in the cistern. The mer cury in the tube therefore is left to the influence of its own weight, and it descends to the bottom. Nothing can be more apposite or decisive.

The gravi And thus the doctrine of the gravity and pressure of the air is established by the most unexceptionable evidence and we are intitled to assume it as a statical tical prin- principle, and to affirm à priori all its legitimate conseciple from quences. which we obtain

• 37 an exact

of the at

And in the first place, we obtain an exact measure of the pressure of the atmosphere. It is precisely equal to the weight of the column of mercury, of water, of oil, &c. which it can support; and the Toricellian tube, or others fitted up upon the same principle, are justly termed baroscopes and barometers with respect to the air. Now it is observed that water is supported at the height of 32 feet nearly the weight of the column is exactly measure of 2000 avoirdupois pounds on every square foot of base, the pressure or 13% on every square inch. The same conclusion mosphere. very nearly may be drawn from the column of mercury, which is nearly 29 inches high when in equilibrium with the pressure of the air. We may here observe, that the measure taken from the height of a column of water, wine, spirits, and the other fluids of considerable volatility, as chemists term it, is not so exact as that taken from mercury, oil, and the like. For it is observed, that the volatile fluids are converted by the ordinary hreat of our climates into vapour when the confining pressure of the air is removed; and this vapour, by its elasticity, exerts a small pressure on the surface of the

water, &c. in the pipe, and thus counteracts a small part of the external pressure; and therefore the column supported by the remaining pressure must be lighter, that is, shorter. Thus it is found, that rectified spirits will not stand much higher than is competent to a weight 13 pounds on an inch, the elasticity of its vapour balancing about of the pressure of the air. We shall afterwards have occasion to consider this matter more particularly.

of

As the medium height of the mercury, in the barometer is 29 inches, we see that the whole globe sustains a pressure equal to the whole weight of a body of mercury of this height; and that all bodies on its surface sustain a part of this in proportion to their surfaces. An ordinary sized man sustains a pressure of several thousand pounds. How comes it then that we are not sensible of 38 a pressure which one should think enough to crush us A difficulty together? This has been considered as a strong objection solved. to the pressure of the air; for when a man is plunged a few feet under water, he is very sensible of the pressure. The answer is by no means so easy as is commonly imagined. gined. We feel very distinctly the effects of removing this pressure from any part of the body. If any one will apply the open end of a syringe to his hand, and then draw up the piston, he will find his hand sucked into the syringe with great force, and it will give pain; and the soft part of the hand will swell into it, being pressed in by the neighbouring parts, which are subject to the action of the external air. If one lays his hand on the top of a long perpendicular pipe, such as a pump filled to the brim with water, which is at first prevented from running out by the valve below; and if the valve be then opened, so that the water descends, he will then find his band so hard pressed to the top of the pipe that he cannot draw it away. But why do we only feel the inequality of pressure? There is a similar instance wherein we do not feel it, although we cannot doubt of its existence. When a man goes slowly to a great depth under water in a diving-bell, we know unquestionably that he is exposed to a new and very great pressure, yet he does not feel it. But those facts are not sufficiently familiar for general argument. The human body is a bundle of solids, hard or soft, filled or mixed with fluids, and there are few or no parts of it which are empty. All communicate either by vessels or pores; and the whole surface is a sieve through which the insensible perspiration is performed. The whole extended surface of the langs is open to the pressure of the atmosphere; every thing is therefore in equilibrio: and if free or speedy access be given to every part, the body will not be damaged by the pressure, however great, any more than a wet sponge would be deranged by plunging it any depth in water. The pressure is instantaneously diffused by means of the incompressible fluids with which the parts are filled; and if any parts are filled with air or other. compressible fluids, these are compressed till their elasticity again balances the pressure. Besides, all our fluids are acquired slowly, and gradually mixed with that pro portion of air which they can dissolve or contain. The whole animal has grown up in this manner from the first vital atom of the embryo. For such reasons the pressure can occasion no change of shape by squeezing together the flexible parts; nor any obstruction by compressing the vessels or pores. We cannot say what would be felt by a man, were it possible that be could have been pro

duced

39 The weather-glass.

40

The pres

portion to

tion

f

duced and grown up in vacuo, and then subjected to the compression. We even know that any sudden and considerable change of general pressure is very severely felt. Persons in a diving-bell have been almost killed by letting them down or drawing them up too suddenly. In drawing up, the elastic matters within have suddenly swelled, and not finding an immediate escape have burst the vessels. Dr Halley experienced this, the blood gushing out from his ears by the expansion of air contained in the internal cavities of this organ, from which there are but very slender passages.

A very important observation recurs here: the pressure of the atmosphere is variable. This was observed almost as soon as philosophers began to attend to the barometer. Paschal observed it in France, and Descartes observed it in Sweden in 1650. Mr Boyle and others observed it in England in 1656. And before this, observers, who took notice of the concomitancy of these changes of aerial pressure with the state of the atmosphere, remarked, that it was generally greatest in winter and in the night; and certainly most variable during winter and in the northern regions. Familiar now with the weight of the air, and considering it as the vehicle of the clouds and vapours, they noted with care the connection between the weather and the pressure of the air, and found that a great pressure of the air was generally accompanied with fair weather, and a diminution of it with rain and mists. Hence the barometer came to be considered as an index not only of the present state of the air's weight, but also as indicating by its variations changes of weather. It became a WEATHERGLASS, and continued to be anxiously observed with this view. This is an important subject, and in another place is treated in some detail.

pressure

In the next place, we may conclude that the sure of the of the air will be different in different places, according air in pro- to their elevation above the surface of the ocean: for if the eleva- air be an heavy fluid, it must press in some proportion according in its perpendicular height. If it be a homogeneous fluid of equal density and weight in all its parts, the mercury in the cistern of a barometer must be pressed precisely in proportion to the depth to which that cistern is immersed in it; and as this pressure is exactly measured by the height of the mercury in the tube, the height of the mercury in the Toricellian tube must be exactly proportional to the depth of the place of observation under the surface of the atmosphere.

41

first sup

posed by and Paschal, and proved by experiments,

Descartes

The celebrated Descartes first entertained this thought (Epist. 67. of Pr. III.) and soon after him Paschal. His occupation in Paris not permitting him to try the justness of his conjecture, he requested Mr Perrier a gentleman of Clermont in Auvergne, to make the experiment, by observing the height of the mercury at one and the same time at Clermont and on the top of a very high mountain in the neighbourhood. His letters to Mr Perrier in 1647 are still extant. Accordingly Mr Perrier, in September 1648, filled two equal tubes with mercury, and observed the heights of both to be the same, viz. 26 inches, in the garden of the convent of the Friars Minims, situated in the lowest part of Clermont. Leaving one of them there, and one of the fathers to observe it, he took the other to the top of Puy de Domme, which was elevated nearly 500 French fathoms above the garden. He found its height to be 23 inches. On his return to the town, in a

place called Font de l'Arbre, 150 fathoms above the garden, he found it 25 inches; when he returned to the garden it was again 26, and the person set to watch the tube which had been left said that it had not varied the whole day. Thus a difference of elevation of 3000 French feet bad occasioned a depression of 3 inches; from which it may be concluded, that 3 inches of mercury weighs as much as 3000 feet of air, and onetenth of an inch of mercury as much as 96 feet of air. The next day he found, that taking the tube to the top of a steeple 120 feet high made a fall of one-sixth of an inch. This gives 72 feet of air for one tenth of an inch of mercury; but ill agreeing with the former experiment. But it is to be observed, that a very small error of observation of the barometer would correspond to a great difference of elevation, and also that the height of the mountain had not been measured with any precision. This has been since done (Mem Acad. par. 1703), and found to be 529 French toises.

42

Paschal published an account of this great experiment which (Grande Exp. sur la Pesanteur de l'Air), and it was were requickly repeated in many places of the world. In 1653 peated by it was repeated in England by Dr Power (Power's Exper. Phil.); and in Scotland, in 1661, by Mr Sinclair professor of philosophy in the university of Glasgow, who observed the barometer at Lanark, on the top of Mount Tinto in Clydesdale, and on the top of Arthur's Seat at Edinburgh. He found a depression of two inches between Glasgow and the top of Tinto, three quarters of an inch between the bottom and top of Arthur's Seat, and of an inch at the cathedral of Glasgow on a height of 126 feet. See Sinclair's Ars Nova et Magna Gravitatis et Levitatis; Sturmii Collegium Experimentale, and Schotti Technica Curiosa.

43

method of

Hence we may derive a method of measuring the Hence a heights of mountains. Having ascertained with great meer of precision the elevation corresponding to a fall of one-heights tenth of an inch of mercury, which is nearly 90 feet, we have only to observe the length of the mercurial column at the top and bottom of the mountain, and to allow 90 feet for every tenth of an inch. Accordingly this method has been practised with great success: bat it requires an attention to many things not yet considered; such as the change of density of the mercury by heat and cold; the changes of density of air, which are much more remarkable from the same causes; and above all, the changes of the density of air from its compressibility; a change immediately connected with or dependent on the very elevation we wish to measure. Of all these afterwards.

of the air.

These observations give us the most accurate measure Also a of the density of air and its specific gravity. This is measure of but vaguely though directly measured by weighing air the density in a bladder or vessel. The weight of a manageable quantity is so small, that a balance sufficiently ticklish to indicate even very sensible fractions of it is overloaded by the weight of the vessel which contains it, and ceases to be exact: and when we take Bernouilli's ingenious method of suspending it in water, we expose ourselves to great risk of error by the variation of the water's density. Also it must necessarily be humid air which we can examine in this way: but the proportion of an elevation in the atmosphere to the depression of the column of mercury or other fluid, by which we measure its pressure, gives us at once the proportion of this

45

and some

of the height of

the atmo

sphere.

weight or their specific gravity. Thus since it is found that in such a state of pressure the barometer stands at 30 inches, and the thermometer at 32°, 87 feet of rise produces one-tenth of an inch of fall in the barometer, the air and the mercury being both of the freezing temperature, we must conclude that mercury is 10,440 times heavier or denser than air. Then, by comparing mercury and water, we get nearly for the density of air relative to water: but this varies so much by heat and moisture, that it is useless to retain any thing more than a general notion of it; nor is it easy to determine whether this method or that by actual weighing be preferable. It is extremely difficult to observe the height of the mercury in the barometer nearer than of an inch; and this will produce a difference of even five feet, or of the whole. Perhaps this is a greater proportion than the error in weighing.

From the same experiments we also derive some know knowledge ledge of the height of the aerial covering which surrounds our globe. When we raise our barometer 87 feet above the surface of the sea, the mercury falls about one-tenth of an inch in the barometer: therefore if the barometer shows 30 inches at the sea-shore, we may expect that, by raising it 300 times 87 feet, or 5 miles, the mercury in the tube will descend to the level of the cistern, and that this is the height of our atmosphere. But other appearances lead us to suppose a much greater height. Meteors are seen with us much higher than this, and which yet give undoubted indication of being supported by our air. There can be little doubt, too, that the visibility of the expanse above us is owing to the reflection of the sun's light by our air. Were the heavenly spaces perfectly transparent, we should no more see them than the purest water through which we see other objects; and we see them as we see water tinged with milk or other fæculæ. Now it is easy to show, that the light which gives us what is called twilight must be reflected from the height of at least 50 miles; for we have it when the sun is depressed 18 degrees below our horizon.

46

curate.

Why this A little attention to the constitution of our air will knowledge convince us, that the atmosphere must extend to a much is not ac- greater height than 300 times 87 feet. We see from the most familiar facts that it is compressible; we can squeeze it in an ox-bladder. It is also heavy; pressing on the air in this bladder with a very great force, not less than 1500 pounds. We must therefore consider it as in a state of compression, existing in smaller room than it would assume if it were not compressed by the incumbent air. It must therefore be in a condition something resembling that of a quantity of fine carded wool thrown loosely into a deep pit; the lower strata carrying the weight of the upperstrata, and being compressed by them; and so much the more compressed as they are further down, and only the upper stratum in its unconstrained and most expanded state. If we shall suppose this wool thrown in by a hundred weight at a time, it will be divided into strata of equal weights, but of unequal thickness the lowest being the thinnest, and the superior strata gradually increasing in thickness. Now, suppose the pit filled with air, and reaching to the top of the atmosphere, the weights of all the strata above any horizontal plane in it is measured by the height of the mercury in the Toricellian tube placed in that plane; and one-tenth of an inch of mercury is just equal to the weight of the lowest stratum 87 feet thick: for on VOL. XVI. Part II.

t

raising the tube 87 feet from the sea, the surface of the mercury will descend one-tenth of an inch. Raise the tube till the mercury fall another tenth: This stratum must be more than 87 feet thick; how much more we cannot tell, being ignorant of the law of the air's expansion. In order to make it fall a third tenth, we must raise it through a stratum still thicker; and so on continually.

All this is abundantly confirmed by the very first experiment made by the order and directions of Paschal : For by carrying the tube from the garden of the convent to a place I 50 fathoms higher, the mercury fell 17 inches, or 1.2916; which gives about 69 feet 8 inches of aerial stratum for of an inch of mercury; and by carrying it from thence to a place 350 fathoms higher, the mercury fell 13, or 1.9167 inches, which gives 109 feet 7 inches for of an inch of mercury. These experiments were not accurately made; for at that time the philosophers, though zealous, were but scholars in the science of experimenting, and novices in the art. But the results abundantly show this general truth, and they are completely confirmed by thousands of subsequent ob servations. It is evident from the whole tenor of them, that the strata of air decrease in density as we ascend through the atmosphere; but it remained to be discovered what is the force of this decrease, that is, the law of the air's expansion. Till this be done we can say nothing about the constitution of our atmosphere : we cannot tell in what manner it is fittest for raising and supporting the exhalations and vapours which are continually arising from the inhabited regions; not as an excrementitious waste, but to be supported, perhaps manufactured, in that vast laboratory of nature, and to be returned to us in beneficent showers. We cannot use our knowledge for the curious, and frequently useful, purpose of measuring the heights of mountains and taking the levels of extensive regions; in short, without an accurate knowledge of this, we can hardly acquire any acquaintance with those mechanical properties which distinguish air from those liquids which circulate here below.

47

Having therefore considered at some length the lead- Compressiing consequences of the air's fluidity and gravity, let bility of the us consider its compressibility with the same care; and air. then, combining the agency of both, we shall answer all the purposes of philosophy, discover the laws, explain. the phenomena of nature, and improve art. We proceed therefore to consider a little the phenomena which indicate and characterise this other property of the air. All fluids are elastic and compressible as well as air; but in them the compressibility makes no figure, or does not interest us while we are considering their pressures, motions, and impulsions. But in air the compressibility and expansion draw our chief attention, and make it a proper representative of this class of fluids.

48

Nothing is more familiar than the compressibility of A familiar air. It is seen in a bladder filled with it, which we can phenomeforcibly squeeze into less room; it is seen in a syringe, non, which of which we can push the plug farther and farther as we increase the pressure.

49

But these appearances bring into view another, and shews its the most interesting, property of air, viz. its elasticity, elasticity. When we have squeezed the air in the bladder or syringe into less room, we find that the force with which we compressed it is necessary to keep it in this bulk; and 4 Q that