precedence is given to phlogiston, by reafon of the extreme levity of the purer inflammable air of metallic folutions, and of the quan tity of fiery matter which it emits in the inftant of its union with empyreal air, the gravitating matter of this last air must not be placed beneath any other that is now known; because no air that is free from phlogifton, is lighter fpecifically than empyreal air, or emits more fire in the inftant of aggregation.

"I confider the fpecific gravity as a fafe guide in our investigations of thefe affinities and of their order, in regard only to the elastic fluids which feem to confift of no more than one kind of gravitating matter engaged in the repellent atmospheres and of fixable air, denfe inflammable air, acid airs, the phlogistic alkaline air, and others, I would obferve, that the atmospheres include molecules in tead of folitary ultimate parts; for without this chemical union of heterogeneal parts, and the formation of molecules, an elaftic fluid of the kind that I now fpeak of, could not differ, as it does, from either kind of matter of which it is compofed.

"From this confideration of the attractive forces which tend to form molecules, and of the atmospheres, which, in compound elastic fluids, encompass the molecules, but not the ultimate parts feverally, we derive an eafy explanation of the phenomenon fo often noticed in the preceding pages; I mean the convertion of a fubftance, not into one but into two or three different elaftic fluids, by mere ignition.

"When elastic fluids are formed in folutions and other procefles, in which the fubjects are not ig

nited it is to be observed that the gravitating parts of the emitted claftic fluid, were distant from each other, by reafon of the interfpofed matter, at the inftant of their extrication, and that at this distance there is a great diminution of the powers which restrain them in their tendency to engage all the matter of fire that the menftruum or folvent could extricate during their union. Thus it happens, as Dr. Black originally fuggefted, and as Bergman has obferved in his excellent Differtation of Elective Attractions, that cauftic alkali, in the union with an acid, excites great heat, that is to fay, in their union they exclude a part of the matter of fire which they feverally held in a fixed ftate; bur mild alkali, in uniting with an acid, gives little or no heat, and for this reafon; that the gravitating parts of the fixable air, engage all the liberated matter of fire in forming atmospheres around its molecules. This expofition is applicable to every other claftic fluid that is extricated in folutions or combinations attended with little or no incalefcence.

"Since the particles which attract the matter of fire, exc.ude a part of their refpective charges in the inftant of their clofe approximation or contact, there is no difficulty in accounting for the cold produced in folutions, expanfions, and evaporations; in every one of which, the particles which refume the matter of fire, and in fixing it produce the cold, are previously removed to fome diflance from each other, either by the interpofition of the parts of a menftruum, as when ice is diffolved by nitrous acid, or by the prevalent powers of their proper fixed fire, of which we have an inftance in ether placed

in the exhausted receiver of an airpump.

"As the mere acetous acid confifts of the gravitating parts of empyreal air, of phlogifton, and the acid principle of vegetables, and by the acceffion of the matter of fire only makes two different elastic fluids, namely fixable air, and denfe inflammable air, it follows from thefe notions, that the gravitating parts of these two airs would make acetous acid again with water, if the matter of fire could be withdrawn from them; and I venture to prognofticate that this, or fomething equivalent, will be done by the first ingenious experimenter that attempts it, either by the means of a body which may absorb the gravitating matter of both, and exclude their atmospheres, or by breaking the atmospheres, in the manner which I am to defcribe in the next fection.

"There is undoubtedly a natural limitation, as we formerly obferved, of the forces with which the grofs parts of divers elattic fluids compel, and are compelled by their refpective atmospheres; but still thefe forces are fo great, that we cannot form adequate ideas of them, without reflecting on the reliftance which they give, even in fmall quantities of the elaftic fluids, to

any preffure which tends to condenfe them, and caufe an approximation of their parts; or without adverting to the operations which fhew the ability of these forces to maintain the elafticity of divers aeriform fluids, in oppofition to the greatest mechanical powers hitherto employed to fubdue them. The air-gun, and other inftruments exhibit thefe forces in atmospheric air, and chemical operations and explofive compofitions too often fhew them in the other elaftic fluids.

"Although the force of chemical attraction reaches not far from the particles, with any fentible effect, we are not thence to conclude that the attractive virtue ceases at any distance.

But from all the known phenomena we may infer, that the attractive forces, tending to the approximation and cohefion of grofs particles, decrease in a duplicate, or fome higher ratio of the distances, reciprocally that the natural refiftance to the interpolition of the fiery matter, is leffened at the smallest, and totally overpowered at fmall diftances which no eye can measure; and finally, that where the fenfible effect of attraction ceafes, there repulfion fucceeds."

"TH

HAT dephlogifticated and inflammable air, at least with the element of heat, conftitute water, feems to be clearly inferred

from the experiments in the preceding fection. Some difficulties, however, have occurred refpecting this theory, from my fubfequent experiments,

experiments, which I fhall propofe with the fame fidelity as I have done thofe which favour the hypothefis. But as I mean to throw into this fection all the obfervations that I have made upon the fubject, I fhall first recite fome experiments which perfectly agree with the former, and evidently lead to the fame conclufion. Some of them are thofe of which only the general refult was given before.

"I have obferved, that, when that iron which has been melted in the open air (or that which has been altered by the paffing of fteam over it red-hot) was heated by a burning lens in inflammable air, the air difappeared, and a confiderable quantity of water was pro. duced. I had the fame refult with feales of copper. Thefe fcales became of a genuine copper colour in this procefs; fo that I had no doubt of their being copper revived. The water was fo copious, that when only 3 ounce measures of air were abforbed, the water formed in drops on the infide of the veffel, and fome of them ran down it.

"I alfo procured water when I decompoted dephlogisticated and inflammable air from iron by the electric fpark in a clofe veffel, which is an experiment fimilar to thofe that were made by Mr. Lavoifier, at Paris. I put 3,75 ounce meafures of a mixture of air, of which one-third was dephlogifticated, and two-thirds inflammable from iron, in the clofe veffel; and after the explosion I found in it one grain of moisture. The dephlogisticated air in this mixture would have weighed a little more than 0,75 of a grain. But repeating this experiment with half as much dephlogificated as inflammable air, I could not perceive any

water after the experiment. Neither was there any fixed air formed in it. Again, ufing inflammable air from wood, in the fame proportion from 3,8 ounce meafures of the mixture, I got 0,8 of a measure of fixed air, which was actually obforbed by lime-water. lime-water. Some water alfo was evidently produced.

Uling more precautions to exclude all water from either of thefe two kinds of air before the experiment (both the dephlogifticated air, which was from nitre, and the inflammable air which was from charcoal, being from the first received in mercury, and always confined by it) I still found a little water after the explofion.

I varied this experiment by producing the inflammable air in the dephlogisticated air as follows. Into a veffel containing dephlogifticated air confined by mercury; I introduced a piece of perfect charcoal, as hot from the fire as I could bear to handle it, and threw upon it the focus of the lens, fö that a quantity of the air was imbibed; but I could not perceive that any moisture was formed. Afterwards, when I refumed the procefs, the air which the charcoal had imbibed was expelled again, and very little more was absorbed. From feven ounce measures there remained four, of which more than two was pure fixed air. No water could be expected after this procefs. For even had there been a small quantity of moisture in the veffel, it would have been abforbed by the charcoal, and have enabled it to yield inflammable air. The phlogiston the charcoal contained uniting with the dephlogisticated air, free from moisture, formed, I prefume, the fixed air that was found after this process.

"But the greatest difficulty that H 3 occurred,

occurred with respect to the preceding theory of the conftitution of water, arofe from my never having been able to procure any water when I revived mercury from red precipitate in inflammable air, or at least not more than may be fuppofed to have been conta ned in the inflammable air as an extra neous fubítance.

"In order to make the experiments with the fales of iron, and that with the red precipitate as much alike as poffible, and that I might compare them to the greatest advantage, I made them immediately one after the other, with every circumftance as nearly as I could the fame. The inflammable air was the fame in both the experiments, and both the fcales of iron, and the red precipitate, were made as dry as poffible. They were heated in veffels of the fame fize and form, and equally confined by dry mercury. And yet when I heated the former, water was formed as copioufly as I have defcribed it before, viz, actually running down the infide of the veffel in drops, though only four ounce measures of inflammable air were abforbed. But though I heated the red precipitate till eight ounce meafures of the inflammable air was abforbed, and only threefourths of an ounce meafure of air remained, there was hardly any fenfible quantity of water produced, certainly not one-tenth of what appeared in the experiment with the fcales of iron.

"There was this difference, however, in the two refults. In what remained from the experiment with red precipitate, I at this time perceived a flight appearance of fixed air, whereas there was none in what remained from the fcales of iron. The refiduum alfo from the

red precipitate had in it a very fmall portion of dephlogisticated air. For being mixed with an equal measure of nitrous air, the ftandard of it was 1,8. I must also observe that the inflammable air difappeated much more rapidly in the procefs with red precipitate than in that with the feales of iron.

"Fixed air, however, was no neceflary refult in this experiment. For I particularly obferved, that when, upon another occasion, I reduced eight ounce measures of inflammable air to 0,22 of a meafure, I found no more than what I judged to be much less than a grain of water, and without any appear ance of fixed air whatever. Again, 1 reduced fix ounce measures of inflammable air in this procefs to 0,12 of a measure, without producing any more water than before, or any fixed air at all. I alfo had the fame refult in ufing the black powder of lead and mercury, in itead of red precipitate: reducing 6,5 ounce meafures to 0,4 of a meafure, without finding any very fentible quantity of water, or any fixed air.

In this experiment there can be no doubt but that the dephlo gifticated air diflodged from the red precipitate, united with the inflammable air in the veffel; as no water equal to the weight of the two kinds of air was produced, they must have formed fome more folid fubftance, which, in the fmail quantities I was obliged to ufe, could not be found.

"It may be clearly inferred from this experiment, that red precipitate is a fubftance by no means fimilar to the scales of iron, as the latter appears to contain within itself all the elements of water; and, therefore, as the inflammable air enters into it, a quan

tity of water, equal in weight to the lofs fuftained by the feales in their revival in the form of iron, is found in the recipient. Whereas in the experiment with the red precipitate, there is certainly no more water found than may be fuppofed to have been contained in the inflammable air which difappeared, or to have been lodged without being perceived among the particies of the precipitate itself. Confequently the feales of iron must be confidered as the calx of iron united to water, and red precipitate as mercury, united to dephlogifticated air, or rather, perhaps, as Mr. Kirwan fuppofes, to fixed air, the phlogifton belonging to which revives the mercury, while its other component part, the dephlogifticated air, is fet at liberty, forming an union with the clement of heat.

"The difficulty with refpect to what becomes of the two kinds of air, was not leffened by the attempts which I made to collect all that I could from repeated decompofitions of inflammable and dephlogifticated air in a clofe veffel.

As I had produced water in this procefs when I made no more than a tingle explosion at a time, I thought that by continuing to make explosions in the fame veffel, the water would not fail to accumulate, till I might collect what quantity I pleated; and I intenced to have collected a confiderable part of an ounce, And as I fhould know exactly what quantity of air I decompofed, I had no doubt of being able to ascertain the proportion that the water and air bore to each other.

"With this view I made a mixture of a large quantity of air, one third dephlogiticated, and two. thirds inflammable from iron and

oil of vitriol. But though I had a fenfible quantity of warer at the first explofion (in each of which was ufed between four and five ounce meafures of the mixture of air) I was furprised to perceive no very fenfible increase of the quantity of water on repeating the explotions. Having, therefore, expended 48 ounce meafures of the mixture, I difcontinued the procefs; and collecting the water with all the care that I could, I found no more than three grains, when there ought to have been eleven.

"In this procefs the infide of the veffel was always very black after each explosion, and when I poured in the mercury after the explosion, though there was nothing viable in the air within the veffel, there iffued from the mouth of it a dense vapour. This was even the cafe, though I waited fo much as two minutes after any explofion before I proceeded to put in more mercury in order to make another; which if the vapour had been steam, would have been time more than fufficient to permit it to condenfe into water. I even perceived this vapour when I had a quantity of water in the veffel, and the explofion was confequently made over it, as well as in contact with the fides of the veffel which were wetted with it; fo that as this vapour had paffed through the whole body of water when the veffel was inverted, it is probable that it must have confifted of fomething elfe than mere water. But I was never able to collect any quantity of it, though it must have been fomething produced by the union of the two kinds of air.

"In order to collect a quantity of the matter that formed this vapour, I contrived the following apparatus. In a cork with which