black even in an hour, although it has been diluted with water. Even vinegar has the same effect. There is no other reason, therefore, why the oil becomes at once black in the air, than that the fire-air present in the air deprives it of its phlogiston, and thereby develops a subtle acid, previously united with this phlogiston, which produces the blackness.

45. Fourth Experiment.—(a.) Into a bottle of 7 ounces, which was filled with fire-air, I put a piece of phosphorus from urine and closed it with a cork. I then heated, by means of a burning candle, the place where the phosphorus lay; the phosphorus took fire with very great brilliancy. As soon as the flame had gone out, the bottle broke into fragments.

(b.) As the bottle in the foregoing experiment was very thin, I repeated it with a somewhat thicker bottle, and after everything had become cold I wanted to take the cork out of the bottle under water. It was not

possible for me to do this, however, so tightly did the external air press the cork into the bottle. Accordingly I forced it inside the bottle; thereupon water entered the bottle and filled it almost completely. Since the first bottle was only very thin, the reason that it was crushed must be ascribed to the external air.

(c.) When I mixed vitiated air with one third of fire-air, and burned a piece of phosphorus in the mixture, only 1⁄3 of it was absorbed.

46. Fifth Experiment.—I also repeated the same experiment which is described in § 19, only with this difference that I took the tube longer, and filled the flask with my fire-air. It was pleasing to observe how the water rose gradually into the flask; and how the flame went out when 7⁄8 of the flask were full of water.

47. Sixth Experiment.—I laid some glowing coals upon the stand (§ 21, letter c), and placed over them a flask which was filled with fire-air. The coals had not even reached the air in the flask before they began to burn very brilliantly.

After everything had become cold, I made an aperture under the flask, whereupon the fourth part became filled with water. But when I removed, by means of milk of lime, the aerial acid which was present in the residual air (§ 22) there remained in the flask only the fourth part. In this air a candle could still burn.

48. Seventh Experiment.—I also examined the behaviour of fire-air with sulphur (§ 23). As soon as the burning sulphur came into contact with the fire-air contained in the flask, the flame became much larger and brighter. When this fire had gone out, the water in the dish had found a way to come through the mass into the flask, which became 3⁄4 filled with it. As I employed for these last 3 experiments a flask which was only of 30 ounces measure, I was obliged to arrange the stand (§ 21) to suit.

49. I have mentioned (§ 16) that I found vitiated air lighter than ordinary air. Must it not follow from this that the fire-air is heavier than our air? As a matter of fact, I actually found, when I accurately weighed as much fire-air as occupied the space of 20 ounces of water, that this was almost 2 grains heavier than the same bulk of common air.

50. These experiments shew, therefore, that this fire-air is just that air by means of which fire burns in common air; only it is there mixed with a kind of air which seems to possess no attraction at all for the inflammable substance, and this it is which places some hindrance in the way of the otherwise rapid and violent inflammation. And in fact, if air consisted of nothing but fire-air, water would surely render small service in extinguishing outbreaks of fire. Aerial acid mixed with this fire-air, has the same effect as vitiated air. I mixed one part of fire-air with 4 parts of aerial acid; in this mixture a candle still burned moderately well. The heat which lurks in the small interstices of the inflammable substance cannot possibly make up so much heat as is felt in fire; and I think I am not mistaken when I conclude from my experiments that the heat is really brought forth and produced in the first place from fire-air and the phlogiston of the inflammable substance....

80. I had long wished to have some of the precipitate of mercury per se, in order to see whether it also would yield fire-air during reduction by means of heat alone. At length I obtained some from my much esteemed friend Doctor Gahn. This so-called precipitate had the appearance of small dark-red crystals resembling cinnabar. Now, as I know that mercury cannot be dissolved in muriatic acid unless it has lost its phlogiston, which takes place during its solution in acid of nitre or in vitriolic

acid; and as this is also the reason why nitre must be present in a mixture of calcined vitriol, common salt, and quicksilver, I therefore poured muriatic acid upon a part of this red precipitate; the solution was soon formed and was somewhat hot; I evaporated it to dryness and increased the heat. Everything sublimed, and a true corrosive sublimate was formed. Hence this precipitate, produced by heat alone, is a calcined mercury. I then placed the other part of this precipitate over the fire in a small glass retort to which I had fastened an empty bladder. As soon as the retort became red hot the bladder became expanded, and at the same time the reduced mercury rose into the neck. In this case no red sublimate arose as customarily takes place with that calx which is prepared by the acid of nitre. The air obtained was a pure fire-air. This is a remarkable circumstance, that the fire-air which had previously removed from the mercury its phlogiston in a slow calcination, gives this same phlogiston up to it again when the calx is simply made red-hot. Still we have several such phenomena, where heat similarly alters the attractive forces between substances.

83. Air is a Dulcified Elastic Acid.

In the foregoing experiments I have demonstrated the two proximate constituents of common air, because it was not necessary to know anything more about it for a clear knowledge of fire. I shall now go further, and see whether a still deeper decompounding of air is possible.

First Experiment.—I placed a rat in a flask capable of holding 4 quarts of water; I gave it some bread softened in milk and closed the flask with a wet bladder. It died 31 hours afterwards. I then held the flask, inverted, under water and made a hole in the bladder, when two ounces of water rose into it. This

small diminution of the air was probably caused by the heat which the rat took with it, which had previously driven the air out.

84. Second Experiment.—I took a large soft bladder and fastened a tube into its opening; then I filled it with the air out of my lungs, and held the tube and bladder with my right hand and closed my nostrils with the left. I respired the air as long as I could, and was able to make 24 inspirations (regarding which it is to be observed that at the last I was obliged to draw the whole bladder full of air into my lungs at once, while at the beginning only the half of it was necessary). I then closed the tube with my finger, and tied up the bladder. This air had properties similar to the preceding in which the rat died. That is to say, it contained one-thirtieth part of aerial acid, which I separated from it by milk of lime; and a burning candle at once went out in it.

85. Third