Experience has taught me that the vapours of water can increase the fire in the same manner as common air; and this air, which we may regard as pure, is always mixed with a very great quantity of water; but it must be remarked, as an observation of much importance, that the proportions of the mixtures are not nearly the same in these two elements. It may be said in general that there is much less air in water than water in air. In considering this proportion we must refer to the volume and mass. If we estimate the quantity of air contained in water by the volume it will appear nil, since the volume is not in the least increased. Thus it is not to the volume that we must relate this proportion, it is alone to the mass, that is, to the real quantity of matter in one and the other of these two elements that we must compare that of their mixture, by which we shall perceive that the air is much more _aqueous_ than the water is _aerial_, perhaps in proportion of the mass, that is, eight hundred and fifty times. Be this estimation either too strong or too weak we can derive this induction from it, that water must change more easily into air than air can transform into water. The parts of air, although susceptible of being extremely divided, appear to be more gross than those of water, since the latter passes through many filtres which air cannot penetrate; since the vapours of water are only raised to a certain height in the air; and, in short, since air seems to imbibe water like a sponge, to contain it in a large quantity, and that the container is certainly greater than the contained.