above a span at the firft ftroke; and the external air being let in, impell'd PNEUMATICS. it up again, almoft to the top of the tube: fo little matters it, how heavy or light the cylinder of quick-filver be, provided its gravity overpower the preffure of as much external air, as bears upon the furface of that mercury into which it is to fall. Laftly, we obferv'd, that if more air were impell'd up, by the pump, into the receiver, after the quick-filver had regain'd its ufual ftandard in the tube, it would afcend ftill higher; and immediately, upon letting out that air, fall again to the height it refted at before. But, in order to fill the Torricellian tube with exactnefs, the edges of the open end should be made even, and turned inwards, that fo the orifice, not much exceeding a quarter of an inch in diameter, may be the more eafily, and exactly ftop'd by the finger; between which, and the quick-filver, that there may be no air intercepted, it is requifite that the tube be perfectly full, that the finger, preffing upon the protuberant mercury, may rather throw fome out, than not find enough to keep out the air exactly. It is, alfo, an useful way, not quite to fill the tube, but to leave, near the top, about a quarter of an inch empty for, if you then ftop the open end, and invert the tube, that quarter of an inch of air, will afcend in a great bubble to the top; and, in its paffage, lick up all the little bubbles, and unite them with itself, into one great one. So that, if by re-inverting the tube, you let that bubble return to the open end of it, you will have a much clofer mercurial cylinder than before; and need add but a very little quick-filver more, to fill up the tube exactly. And, laftly, as for fuch lefs, and invifible parcels of air, which cannot be thus gather'd up, you may endeavour, before you invert the tube, to free the quick-filver from them, by fhaking the glafs, and gently knocking on the outfide of it, after every little parcel of quick-filver pour'd in; and afterwards, forcing the bubbles to difclofe themfelves, and break, by applying a hot-iron near the top of the glass; which will raife the bubbles fo powerfully, as to make the mercury appear to boil. I remember, that by carefully filling a fhort tube, tho' not quite free from air, we have made the mercurial cylinder reach to thirty inches, and above an eighth; which is mention'd, because we have found, by experience, that in fhort tubes, a little air is more prejudicial to the experiment, than in long ones. of the mercurial 18. We fill'd a glafs tube, about three feet long, with mercury; and Odd phenomena. having inverted it into a veffel of other quick-filver, that in the tube, barometer.. fell down to its ufual height; leaving fome little particles of air in the space it had deferted: for, by the application of hot bodies, to the upper part of the tube, the quick-filver would be a little deprefs'd. Laftly, having put both the tube, and the vessel whereon it refted, into a convenient wooden frame, we placed them together in a window of my chamber. And during feveral weeks, that the tube continu'd there, I obferv'd, that the quick-filver did, fometimes faintly imitate the liquor of a thermo PNEUMATICS meter; fubfiding a little in warm, and rifing a little in cold weather; which we afcrib'd to the greater, or leffer preffure of that little air, which remain'd at the top of the tube, expanded, or condens'd by the heat, or cold of the ambient air. But, the quick-filver often rofe, and fell in the tube very confiderably, after a manner, quite contrary to that of weather-glaffes, where air is at the top; for fometimes, I obferv'd it, in very cold weather, to fink much lower, than at other times, when the air was comparatively warmer. And fometimes, the quick-filver would, for feveral days together, reft almoft at the fame height; and at others, it would in the compass of the fame day confiderably vary its altitude; tho' there appear'd no change, either in the air abroad, or in the temper of that within my room, nor in any thing elfe, to which fuch a change could reafonably be imputed; efpecially confidering, that the space wherein the mercury continued unfettled for five weeks, amounted to full two inches; defcending in that time about of an inch from the place where it first fettled, and afcending the other inch, and: and when we took the tube out of the frame, after it had ftaid there part of November, and December, a large fire being then in the room, we found the mercurial cylinder to be above the upper furface of the ftagnant mercury 29 inches *. 6 Such That the quick-filver in the barometer blown to another quarter, may, not only fhould ftand lower, when the air is thick condense the atmosphere, but make it and moift, than when it is dry, and clear, heavier. Moreover, heavy dry exhalaseems to overthrow the theory of the air's tions will increase the weight of the air, gravitation. Indeed, to discover the cau- (as falts and metals diffolv'd in proper fes of all the minute variations in the air, is menftrua, increase the specific gravity a very difficult task. The winds have a of them;) and perhaps, at the fame time, great share herein, with the vapours, exha- add to its elafticity. Again, the air, by lations, and expirations of the earth; per- thefe, or the like caufes, being rendred haps alfo, the changes, which happen in heavier, is the more able to fuftain the vathe adjacent regions; the flux and reflux pours; which therefore coming to be incaufed by the moon in the air, no less timately mix'd therewith, and floating ethan in the sea, and many other particu- very where uniformly therein, render it lars, are not unconcern'd. Now, the air fair and clear. But, when from contrary is heavier, than the vapours it fuftains; caufes it becomes lighter, 'tis rendred units particles being more grofs, and arifing able to fuftain the vapours, which always from denfer bodies, than the particles of opprefs it; fo that being, as it were, precivapours. But, winds may change this pitated together, they form clouds, and weight of the air, in any particular re-running into drops, fall, by their increafed gion; either by bringing, and keeping up more air over it, as may cafily happen, when two contrary winds blow; or by fweeping it away, and affording room for the fubjacent air to expand itfelf; as may be the cafe, when two oppofite winds meet, or, when only one blows exceeding ftrong. Thus, 'tis fact, that violent gufts make the mercury in the barometer greatly to fink of a fudden. The cold nitrous particles of the air, or the air itself, being condenfed by cold in the north, and gravity, to the earth. Hence we fee, what caufes render the air heavier, and more able to fuftain the quick-filver in the barometer, namely, fuch as make the air clear and dry: but the caufes, which render the air light and unfit to fuftain the mercury, produce rain. When therefore, the air is lighteft, and the mercury in the barometer loweft, the clouds appear very low, and in very swift motion; and the air having clear'd itself of its clouds by rain, becomes very bright and tranf Such an inequality in the rife, and fall of the mercury will, I fear, ren- PREUMATICS, der. it difficult to determine by the barometer, whether the moon be the caufe of the tides, efpecially, till the reafon of this odd phenomenon be certainly known; which feems principally to depend upon confiderable alterations in the air, in point of rarity and denfity. ment made with water. 19. We took a tube of glafs, about four feet in length, hermetically A like experi feal'd at one end, fill'd it with common water, and inverted the open end, went beneath the furface of a veffel of water. Then this veffel, with the tube in it, being let down into the receiver, the pump was fet on work; when, till the receiver was moderately exhaufted, the tube continu'd quite full of water; it being requifite, that a great part of the air contain'd in the receiver fhould be drawn out, to bring the remaining to an equilibrium, with fo fhort a cylinder of water. But, when once the water began to fall in the tube, each exfuction of air made it defcend a little lower; tho' nothing near fo much, nor fo unequally, as the quick-filver did. The lowest, we were able to draw down the water, was, to about a foot above the furface of that in the veffel. And, when the water was drawn down thus low, we found, that by letting in the outward air, it might be immediately impeli'd up again, to the higher parts of the tube. Upon making this experiment in a final receiver, we obferv'd, that at the firft exfuction of the air, the water ufually fubfided feveral inches; and at the fecond, fometimes near two feet; whereupon letting in the external air, the water was impell'd up, with a very great velocity. 20. That the air hath a confiderable elaftic power, we have abundantly whether water proved: but, whether water participates, in any measure, thereof, feems be elastic? hitherto, to have been scarce confider'd. Into a large glafs bubble, with a long neck, we pour'd common water, till it reach'd about a fpan above the bubble; and a piece of paper being pasted thereon, we put it, unstopp'd, into the receiver; when, the pump any transparent, fo as to afford an excellent | pofite parts; for in the Atlantic ocean, at See was PNEUMATICS was work'd, after the ufual manner, and a confiderable part of the air in the receiver drawn out, before we difcern'd any expanfion of the water; but continuing to pump, the water manifeftly began to afcend in the ftem of the glafs, and feveral bubbles, from the lower parts of the veffel, made their way thro' the liquor to the top of it, and there broke into the receiver. After the water once appear'd to fwell, at each time the air was let out from the receiver into the pump, the water in the neck of the glafs, fuddenly rofe, about the breadth of a barley-corn, and fo by degrees attain'd to a confiderable height, above the mark. And at length, the external air, being fuddenly re-admitted, the water immediately fubfided, and deferted all the additional space, it had gain'd in the glass. 21. We convey'd into the receiver a new glafs-vial, capable of holding about fix or seven ounces of water; into which we had before-hand put only two or three fpoonfuls of that fluid, and stopp'd it close with a fit cork. The receiver being emptied, there appear'd no change in the inclosed water; the air, imprifon'd with it, not having the force to blow out the ftopple. Wherefore, we again put in the vial, lefs firmly clofed than before; but when the air was pumped out of the receiver, that within the vial quickly found little paffages to get out at: for when the vial was put in the time before, the water remain'd all the while perfectly free from bubbles; but now the bottom of the glafs appear'd all cover'd with them, which, upon the return of the excluded air, prefently fhrunk up. Hence it feem'd deducible, that, whilft the vial continu'd well ftopp'd, . the included water fuftain'd, from the air fhut up with it, a preffure equal to that of the atmosphere; fince, till the air could get out of the glass, there appear'd no bubbles in the water, notwithstanding the want of preffure in the ambient body. But, further, we caufed a convenient quantity of water to be hermetically feal'd up in a glafs-egg, whofe long neck was faften'd to one end of a ftring, the other end whereof was ty'd to the cover of our receiver; then the egg being convey'd into the receiver, and that being evacuated, we, by turning the brass-ftopple, fo fhorten'd the ftring, as to break the glass; whereby liberty being given to the air imprifon'd in the egg, to pafs into the receiver, its fudden recefs made fo many bubbles appear immediately, and afcend fo fwiftly in the water, that their motion look'd like that of a violent shower of rain; except that the bubbles did not, like the drops of rain, tend downwards, but upwards; as happens in the diffolution of feedpearl, in fome very acid menftruum, wherein, if a large quantity of the pearls be caft whole, they will, at first, be carry'd in fwarms from the bottom to the top of the liquor. And, without fealing up the glafs, this experiment may be try'd in a small receiver: for the air may here be drawn out fo foon, that the bubbles, lurking in the water, will, immediately, difplay themselves, and afcend in throngs. So that, having made the experiment, in fuch a receiver, with red wine, inftead of water, the wine appear'd all cover'd with a large vanishing white froth. 22. To difcover whether the expansion of the water really proceeded PEUMATICS. from an elastic power in the parts of that fluid; we fill'd a glafs-vial, with a pound and fome ounces of water, and then put into it a glafs-pipe, open at both ends, and feveral inches in length, fo as to reach a little below the neck; then we carefully cemented it thereto, that no air might come into the vial, nor any water get out of it, but thro' the pipe; and the pipe, being warily fill'd about half way with water, and a mark being pafted over-against the upper furface thereof, the whole was, by ftrings, let down into the receiver : when, pumping out the air, the water in the pipe began to rife, while fome little bubbles difcover'd themfelves on its fides; and, foon after, the water ftill fwelling, there appear'd, at the bottom of the pipe, a bubble, about the bignefs of a small pea; which, afcending thro' the tube to the top of the water, ftaid there a while, and then broke. But the pump being nimbly ply'd, the expanfion of the water fo increased, that, quickly getting up to the top of the pipe, fome drops of it began to run down along the outfide of it; which obliged us to forbear pumping a while, and let it fubfide, as it did, within lefs than two inches of the bottom of the pipe. Then the pump being again fet on work, the bubbles began to afcend from the bottom of the pipe; of which we reckon'd about fixty large ones, that afcended one after another. And, at length, letting in the external air, the water, in the pipe, instantly fell down almoft to the bottom of it. When the greater part of the air had been pump'd out of the receiver, the bubbles afcended fo very flowly in the pipe, that their progrefs was fcarce difcernible; their magnitude not permitting them fufficiently to expand themselves in the cavity of the glafs, without preffing against the fides of it. And, what feems ftrange, thefe bubbles were commonly much larger than thofe which rofe before them; fome of them being equal in bulk to four or five peas. And tho', in ordinary bubbles, the air, together with the thin film of water that invests it, commonly fwells above the furface of the water, and conftitutes hemifpherical bodies; the little parcels of air, that came up after the receiver was tolerably emptied, did not make protuberant bubbles; but fuch, whofe upper furface was either level with, or beneath that of the water: fo that, the upper furface being usually fomewhat convex, the lefs protuberant parts had a quantity of water above them. We farther obferv'd, that, in the bubbles which firft appear'd, the afcending air made its way upwards, by dividing the water thro' which it pafs'd; in those that rofe at the latter end of the experiment, the afcending parcels of air, having now little more than the weight of the incumbent water to furmount, were able to expand themfelves, fo as to fill that part of the pipe which they pervaded, and, by preffing every way against the fides of it, to raise what water they found above them, without letting any confiderable quantity glide down along the fides of the glafs: fo that, fometimes, we could fee a bubble thruft on before it a whole cylinder of water, perhaps an inch high, and carry it up to the top Iii 2 of |