As there are many natural local influences which either increase the amount of carbonic anhydride in the air (respiration, combustion, decomposition, volcanic eruptions, &c.), or diminish it (absorption by plants and water), the reason of the great constancy in the amount of this gas in the air must be looked for, in the first place, in the fact that the wind mixes the air of various localities together, and, in the second place, in the fact that the waters of the ocean, holding carbonic acid in solution,37 form an immense reservoir for regulating the amount of this gas in the atmosphere. Immediately the partial pressure of the carbonic anhydride in the air decreases, the water evolves it, and when the partial pressure increases, it absorbs it, and thus nature supplies the conditions for a natural state of moving equilibrium in this as in so many other instances.38
Besides nitrogen, oxygen, moisture, and carbonic acid, all the other substances occurring in air are found in infinitesimally small quantities by weight, and therefore the weight of a cubic measure of air depends, to a sensible degree, on the above-named components alone. We have already mentioned that at 0° and 760 mm. pressure the weight of a cubic litre of air is 1·293 gram. This weight varies with the acceleration of gravity, g, so that if g be expressed in metres the weight of a litre of air, e = g × 0·131844 gram. For St. Petersburg g is about 9·8188, and therefore e is about 1·2946,39 the air being understood to be dry and free from carbonic anhydride. Taking the amount of the latter as 0·03 per 100 volumes, we obtain a greater weight; for example, for St. Petersburg e = 1·2948 instead of 1·2946 gram. The weight of one litre of moist air in which the tension40 of the aqueous vapour (partial pressure) = f mm., at a pressure (total) of air of H millimetres, at a temperature t, will be (i.e., if at 0° and 760 mm. the weight of dry air = e) equal to
e / 1 + 0·00367t ×
H - 0·38f / 760 . For instance, if H = 730 mm., t = 20°, and f = 10 mm. (the moisture is then slightly below 60 p.c.), the weight of a litre of air at St. Petersburg = 1·1527 gram.41
The presence of ammonia, a compound of nitrogen and hydrogen, in the air, is indicated by the fact that all acids exposed to the air absorb ammonia from it after a time. De Saussure observed that aluminium sulphate is converted by air into a double sulphate of ammonium and aluminium, or the so-called ammonia alum. Quantitative determinations have shown that the amount of ammonia42 contained in air varies at different periods. However, it may be accepted that 100 cubic metres of air do not contain less than 1 or more than 5 milligrams of ammonia. It is remarkable that mountain air contains more ammonia than the air of valleys. The air in those places where animal substances undergoing change are accumulated, and especially that of stables, generally contains a much greater quantity of this gas. This is the reason of the peculiar pungent smell noticed in such places. Moreover ammonia, as we shall learn in the following chapter, combines with acids, and should therefore be found in air in the form of such combinations, since air contains carbonic and nitric acids.
The presence of nitric acid in air is proved without doubt by the fact that rain-water contains an appreciable amount of it.
Further (as already mentioned in Chapter IV.), air contains ozone and hydrogen peroxide and nitrous acid (and its ammonia salt), i.e. substances having a direct oxidising action (for instance, upon iodized starch-paper), but they are present in very small quantities.43
Besides substances in a gaseous or vaporous state,44 there is always found a more or less considerable quantity of substances which are not known in a state of vapour. These substances are present in the air as dust. If a linen