Chapter I. On the Properties of Air.

...forms over a height of 10 toises A toise is approximately 6.4 feet or 1.95 meters.. If one wishes to know the ratio of the weight of air to that of water, one need only reduce 70 lbs—the weight of one cubic foot of water—into drachms The "dragme" or drachm in the old French system was 1/128th of a pound., which will be found to be 8,960; and as a cubic foot of air weighs 14 drachms, one may therefore say that its weight is to that of water as 14 is to 8,960, or as 1 is to 640.

793. Messrs. Mariotte and Homberg together conducted several experiments in 1683 on this subject, and they found that the weight of air is to that of water as 1 is to 637; since then, several Scholars have also sought this ratio, but they have not always perfectly agreed, because air expands with heat and condenses with cold. Thus, the same volume of air contains more or less mass at one time than at another; but if one does not take these variations into account, one can conclude that it is 640 or 630 times thinner or more expanded than water.

Regarding the different heights of mercury in the barometer according to the various temperatures of the air, it seems quite surprising to see that when the air is heavily laden with vapor and rain is ready to fall, the mercury descends—whereas it would seem that the column of air pressing directly on the mercury at the open end of the tube ought to be at its heaviest; and that, on the contrary, the mercury rises to its highest point when the air becomes pure and serene.

794. Mr. Leibniz attributes the descent of the mercury in the barometer when it is about to rain to a very natural cause, and one which seems more satisfying to me than all the hypotheses that have come to my knowledge. To understand it, one must recall what was said in article 630: that a foreign body within a liquid accounts for part of the liquid's weight so long as it floats; but the moment it descends, its weight no longer entirely contributes to that of the liquid, which thereby comes to press less upon the bottom of the vessel supporting it.

Similarly, as long as the imperceptible particles of water, in a prodigious quantity, are supported in the air, they increase its weight, which then presses more upon the surface of the bodies on which it rests; and this is what causes the mercury in the barometer to be forced upward. But as soon as the water particles are numerous enough to acquire a weight exceeding that of the air supporting them, they descend, join together, and form drops; as these begin to fall, they cease to constitute such a large part of the weight of the air, which then no longer presses with as much force upon the surface of the bodies on which it rests.