of pure water, may be extended through the whole, and we may say it is rarefied to that dimension; and as the particles in such a case cannot possibly touch one another as they did before, we may imagine there is some repellency in them: but though they do not touch one another, they touch something else, and are sustained upon statical principles. This we suppose (allowing for the difference between the nature of the two fluids) to be the case with the air; which is no otherwise extended than as it is made more dilute by the mixture of a greater pro portion of fire in any given space.

As it does not appear that there is a single experiment to evince any elasticity in air independent of fire, we may therefore deduce from the power of fire all the laws of its elasticity, in the following order:

1. All air is expanded by fire in a certain degree; and this holds universally, because the lowest degree of fire, or that point at which fire ceases to be active, has never yet been discovered, nor can be; for when mercury is frozen with a degree of cold three hundred degrees below the point of freezing, air is still elastic as before. When there is no obstruction in the way to limit the expansion,

DD 2

pansion, it is the effort of fire to expand air in infinitum; for, fire acts as if every point of space which it occupies were a point of radiance. Its centre is every where, and its circumference no where. We have seen above, that air, though greatly expanded by fire, cannot be totally expelled from any space including it, by any degree of heat whatsoever. In a mean state of the air, the heat of boiling water expands it about onethird; as is found by inverting an empty phial (perfectly dry within) in boiling wa ter, and letting it remain there till the air and water are cool, when it will be filled about one-third with water; which shews that so much of the air was expelled by the heat, and consequently that the expansion of the air was in that proportion. By the addition of a small quantity of water to wet the inside of the phial, the fire will totally expel the air, and make such a perfect vacuum, that when all is cool the phial shall be filled entirely with water. The vapour of boiling water is expanded with such force, and unites itself in such a manner to the included air, that both are carried off together, and the space so evacuated is justly thought to be a better vacuum than the Torricellian :

it is, strictly speaking, a vacuum of a vacuum; because nothing remains but a medium more subtile than that common fire of the atmosphere, which remains in the airpump or Torricellian vacuum. The most simple experiment for finding with accuracy the expansions of the air under all different degrees of heat, between the point of freezing and the heat of boiling water, is that described by Mr. Hawksbee in his physicomechanical treatise, p. 218.

2. Air is naturally in a state of compression, from the expansion of other air which lies next to it; to which we may add the weight of the incumbent air, which is more or less according to the density of the air and the altitude of the column.

3. Therefore, the same degree of heat will expand air more in proportion as the compressing force is removed. When a flaccid bladder is placed under the receiver of an air-pump, and the compressing force is withdrawn by exhausting the air from the outside of the bladder, the included air is extended, as it would be if the same bladder were exposed to the heat of a fire. For it is the same thing in effect to subtract the air from without, as to add more heat to the air within.

DD 3

within. Therefore the heat in this case is still the true cause of the additional extension under the receiver, though there is no additional heat. There is no occasion for it the same heat under these circumstances will have a greater effect. If all the compressing force could be withdrawn, and the space were infinite, the air would be infinitely extended. This case of the flaccid bladder explains the reason why the vacuum of an air-pump produces so many of the same effects with actual fire.

4. The resistance of air to any compres ing force is as the compressing force itself, because re-action is equal and contrary to action; in other words, the fire endeavours to re-enter with a force equal to that which expels it. Hence the air is always a counterbalance to itself, and naturally in equilibrio, like other fluids. Air compressed by twice the weight of the atmosphere, is reduced to half the space: by four times that weight, to one quarter of the space; and so on, in a geometrical progression, supposing the degree of heat to be always the same, Air, at any depth under water, is compressed by the weight of the water: and as a column of water 34 feet in depth is equal in

weight to the pressure of the atmosphere, the diver in his bell, when he goes to the depth of 50 fathom, breathes an air compressed by the weight of 9 atmospheres, without being destroyed.

5. The elasticity of any parcel of air which is subject to the pressure of the atmosphere, is equal to the weight of the atmosphere. This holds if the air is ever so much rarefied. Let the cistern of a barometer, with part of the lower end of the tube, be inclosed in a flaccid bladder, so as to be air-tight. When this bladder is held near a fire, and the included air is rarefied to twice or three times the first dimensions, the mercury will still remain at the same height: for the rarer air, by the additional force of the expanding fire, is a counterbalance to the air without, as it was at first. What it loses in density, it gains in elasticity; or in other words, the elasticity is increased in the same proportion as the density is diminished; and therefore the barometer is stationary under all the variations of elasticity and density. If a barometer could possibly be placed where air is thus rarefied by fire an hundred times, it would shew the elastic force of that air to be equal in effect to the Ꭰ Ꭰ 4 ordinary