follow the proportion of the weights with which the air is loaded; therefore the barometer experiments, made at the foot and on the top of the mountains, cannot give us the height of the atmosphere, since these experiments are done only within the lowest part of the air. The atmosphere extends far beyond this; and its rarefactions are all the further away from the previous law, the farther from the Earth it is. This is what prompted de La Hire, after Kepler, to use an older, simpler and safer method to find the height of the atmosphere: this method is based on the observation of the twilight hours.
We will return to the twilight method. The essential fact here is the inference, from the confrontation between pressure measurements made at different altitudes on the mountains and measurements of the relationship between air dilatation and pressure made in the laboratory, of the existence at great heights of air following a law of expansion different from that of the air that directly surrounds us. Thus, the heterogeneous lower air and the elementary upper air do not only differ in their densities, the former being much heavier than the latter because of its load of impurities emanating from the Earth, but also in their nature, the latter extending much higher than it would if it followed the law of expansion of the former. It should be noted that the entry REFRACTION from the Encyclopédie no longer makes any reference to the role of vapors and exhalations, which is emphasized in the Dictionnaire Universel. As we will see, the role of atmospheric impurities in refraction was disproved at the beginning of the 18th century on the basis of observations showing its absence under certain conditions, such as the case of a star seen through a cloud, which led some scientists to postulate the existence of a subtle, lightweight refractive matter. An allusion to refractive matter can be found in the entry REFRACTION of DUF-1727:
The cause of refraction is not yet known; perhaps it will never be known, like many other points in physics. Is it air, is it refractive matter that is in the air, according to Mr. Cassini’s conjecture? This is where we are still on this matter. There are lot of apparent annoyances in one or the other system, and consequently a lot of uncertainties.
The existence of refractive matter was far from being unanimously accepted. It is not mentioned in the entry REFRACTION of the Encyclopédie.
The entry AIR in the Encyclopédie devotes long sections to the different “characteristics” of air. Unlike the vapors contained in a bottle, which, when it is cold, lose their elasticity and attach themselves around the inner walls of the glass, air does not condense. It is the air that provides the means for earthly bodies to burn, while on the contrary, the vapors and exhalations extinguish fire, coals and burning iron. While in stormy weather, the exhalations ignite, producing lightning, air remains intact after a rainstorm. We do not know the nature of the air, because we cannot examine the air alone and purified of the materials mixed in it. For some, air was “a substance sui generis, which does not derive from any other, which cannot be generated, which is incorruptible, immutable, present in all places, in all bodies, etc.”. That is, by definition, elementary air. For others, the elasticity of air, its essential and distinctive character, was conferred on it by the matter of the bodies from which it was derived, “which has become, through the changes made in it, susceptible to permanent elasticity”. This conception was notably that of Robert Boyle, who carried out numerous experiments in the production of air from bodies that did not seem to contain air, the best methods for this purpose being “fermentation, corrosion, dissolution, decomposition, boiling of water and other fluids, and the reciprocal action of bodies, especially saline bodies, on each other”. According to Newton, “particles of a dense, compact, fixed substance, adhering to each other by a powerful attractive force, can only be separated by violent heat, and perhaps never without fermentation; and these bodies, which are eventually rarefied by heat or fermentation, are transformed into truly elastic air”. While the entry AIR in DUF-1690 states that “water resolves itself, evaporates into air”, the Encyclopédie states that not everything that appears to be air is air:
The example of the aeolipile, where water is sufficiently rarefied by fire, comes out with a sharp whistle, in the form of a matter perfectly similar to air; but soon afterwards loses this resemblance, especially in the cold, and becomes water again through condensation, as it was originally. The same behavior can be observed in the spirit of wine, and other subtle and fleeting spirits obtained by distillation; instead of the real air being reduced neither by compression, nor by condensation or any other means, to any substance other than air.
So you can make water take on the appearance of air for a while: but it soon regains its own.
We find the same questioning expressed in the entry for ÉBULLITION (BOILING) in the Encyclopédie:
With regard to the cause of boiling, we have historically related to the word “boiling” that which physicists usually give as the cause of boiling, and which they attribute to the air which is released from the particles of water; but other physicists reject this cause, and believe that boiling comes from the particles of water itself, which are changed by the action of fire into very expanded vapor, and which rise from the bottom of the vessel to the surface. Here are the reasons for their opinion: (1) The boiling is done in the vacuum machine, when water previously purged of air is heated in it. It is therefore not the air that produces it; it is in this case the heat that makes the water scarce: these are the words of Mr. Musschenbroek […] (2) Water does not stop boiling until it is evaporated; but how can one conceive that the air enclosed in water, and which makes up at most one thirtieth the part, can suffice for all this boiling? (3) Although not all liquors contain the same amount of air, all seem to boil equally. (4) The more water is free to evaporate, that is, the more the vase in which it is put is open, the less heat it supports without boiling. (5) The more subtle a liquor is, and therefore easy to reduce to steam, the less heat is needed to boil it. Thus the spirit of wine boils at a lower heat than water, and water at a lower heat than mercury.
Air is divided into “real or permanent” and “apparent or transient”. The vapors, produced by evaporation of water, are apparent air, while dry exhalations are permanent air. Air is to be understood here in the sense of coarse or heterogeneous air, a mixture of elemental air and impurities emanating from the Earth and water. The production of air from solid bodies that appear to be devoid of air is questioned by the author of the entry AIR in the Encyclopédie:
But, after all, there is still reason to doubt whether the matter thus extracted from solid bodies has all the properties of air; whether this air is not transient, or whether the permanent air that is drawn from bodies did not already exist there. Mr. Boyle proves through an experiment conducted in the pneumatic machine with a lit wick, that this subtle smoke, which the fire raises even from dry bodies, does not have as much spring as air, since it cannot prevent the expansion of a little air enclosed in a bladder which it surrounds […] Nevertheless in some later experiments, by dissolving iron in vitriol oil and water, or in etching, he formed a large air bubble which had a real spring and which, as a result of its spring, prevented the neighboring liquor from taking its place; when a warm hand was applied to it, it expanded easily like any other air, and separated in the liquor itself into several bubbles, some of which rose out of the liquor in the open air.
The same physicist assures us that he has drawn a truly elastic substance from several other bodies; such as bread, grapes, beer, apples, peas, beef, etc. and from a few bodies, by burning them in a vacuum, and singularly from paper, from deer horn: but nevertheless this substance, on close examination, was so far from the nature of pure air, that the animals enclosed in it, not only could breathe only with difficulty, but even died there faster than in a vacuum, where there would have been no air at all.
Thus, the nature of the air, its primary character, or on the contrary, secondary to other bodies from which it would be derived, its resemblance to the elastic substances that we draw from humid bodies, by evaporation, or by burning dry bodies, constituted in the middle of the 18th century still unresolved questions, which made the definition of air, and of the atmosphere which was its mass, fluctuating and multiple. Air, depending on whether it is described as elementary, heterogeneous, permanent
