known laws, these decompositions and recompositions of atmospheric air, as being a fluid sui generis, and not a mixture of two aeriform fluids, differing in their nature, as has been concluded from specious phenomena produced in our experiments; but these phenomena I have explained in my works, without supposing such a mixture, in itself contrary to a number of atmospheric phenomena. This 1 shall here successively explain, though not with so many particulars as are contained in my works.

Rain will be my first object; and indeed it ought to be so in every general system of chemistry, since no phenomenon, either spontaneous or artificially produced, is more connected with the manifestation of water in the modifications of expansible fluids; and none certainly is attended with greater consequences on our globe. With a view of supporting the new hypothesis of a certain composition of water, from which and its associate hypothesis of two distinct and defined aeriform fluids in the atmosphere, rain, so common a phenomenon, cannot be explained, the ancient and already exploded hypothesis of M. Le Roy, of evaporation being a dissolution of water by air, has been revived. This hypothesis, the only apparent resource of the modern theory of chemistry, was plausible at the time of its first publication, about sixty years ago, when meteorological observations were very little advanced; because it is certain, that evaporation restores, upon the whole, to the atmosphere, the same quantity of water as falls from it in rain, dew, and other aqueous meteors: but from a number of well determined phenomena, discovered by the progress of observation, this compensation is not immediate that water, which ascends in the atmosphere by evaporation, passes through an intermediate state, undergoing a chemical transmutation, which makes it disappear to all our tests, sometimes for many months, it being then transformed into an aeriform fluid; and it must be by some inverse operation, that, all at once, clouds, rain, and the other concomitant phenomena are produced. I shall show hereafter how unfounded, as well as useless, is an hypothesis imagined for evading the consequences of these phenomena, which I have opposed to the new theory of chemistry; but first, I must proceed farther in the account of the phenomena themselves.

The above consequences may be deduced from the most common atmospheric phenomena, even when only viewed from the plain, provided they are observed in all their consequences; but it is on high mountains, the very region of meteors, that, from other circumstances not perceptible in lower situations, the observer is induced to wish for more knowledge in the astonishing operations performed in this laboratory. Such has been the case with M. de Saussure and myself, on account of our frequent visits to the mountains of our native country, for the geological pursuits in which we were engaged at the same time. The surprising phenomena concerning moisture, which we observed in these high regions of the air, led us separately to the pursuit and construction of our respective hygrometers; in order to understand, by experiments and observations with

this instrument, in what really consists moisture in the atmosphere; and to follow certain of its modifications, as its sudden increase and diminution without perceptible cause a knowledge which, if not leading to immediate discoveries on the other atmospheric operations, might at least clear the way to these discoveries by dispelling and preventing errors.

When our experiments and observations were first published, they attracted much the attention of natural philosophers; but by degrees they have been forgotten, from the increasing prevalence of the hypothesis of a composition of water to which they were opposed, in consequence of their connexion with the most common meteorological phenomena; an opposition explained even before this hypothesis was so much relied upon as to effect a change in the whole nomenclature and language of chemistry.

This inattention, for a time, to real and important discoveries, an effect occasioned by prevailing prejudices, is observed under various forms in the history of sciences; but there it is seen also, that an obstacle of this nature could not be perpetual, and it may be expected, that it will not be so in this case; therefore I shall here assemble some uncontroverted results of observation and experience, for the consideration of natural philosophers.

. Article 1. Evaporation, the original source of atmospheric phenomena, is not a dissolution of water by air, as is now so commonly assumed; air has no share in it. The immediate product of evaporation in all its stages, from the formation of steam by boiling water, down to the evaporation of ice in winter, is constantly and uniformly an expansible fluid, composed of water and fire, namely the aqueous vapour. This fluid, in whatever temperature it is produced, acts by pressure, in the same manner as the aeriform fluids, and in particular on the manometer, from the instant of its production, as long as it subsists; and the quantity of its production, attended with a proportional pressure, is the same in vacuo as in air, at its different maxima correspondent to each degree of temperature; a direct proof that air has not the smallest share in evaporation. Lastly, as long as this fluid subsists without any change in its nature, it never ceases to act upon the hygrometer, and its quantity is exactly measured by this instrument, with the addition of the thermometer. I have proved these assertions by the union of M. de Saussure's experiments and mine, in some papers published in the Phil. Trans. of 1793. It is evident, that, if these be real facts, the resource of the new theory of chemistry for explaining rain is overturned (as will be seen hereafter), and with it the theory itself; what then is the reason, that those, who still maintain it, remain silent on these facts? On this however rests (and will continue to rest till the contrary be proved by direct experiments) the whole of meteorology. Art 11. Both M. de Saussure and myself have determined, by direct experiments related in our respective works, as I shall more particularly express hereafter, the quantities of evaporated water contained in one cubic foot of air correspondent to every degree of

our hygrometers, at every temperature; and we have proved, that the maximum of this water, a quantity fixed for every temperature, cannot be exceeded, either by the increase of water in the same space, or by the diminution of heat with the same quantity of this water, without some of the aqueous vapour being decomposed, and water making its appearance by precipatation: and by my experiments it is moreover demonstrated, that no length of time after the production of this fluid, can prevent either its effect on the hygrometer, or its remaining submitted in the same manner to the influence of temperature.

Art. III. The aqueous vapour, i.e., the immediate product of evaporation, is therefore never concealed in the atmosphere; and its quantity, in any part of the latter, can always be determined by the observation of the hygrometer and the thermometer. This fluid, produced by the evaporation that never ceases on the surface of the water aud of the land, being of a specific gravity less than that of air, constantly ascends in the atmosphere, passing through its lower regions, where we do not find that it remains; it ought, therefore, to accumulate in the higher parts. Now, as we ascend on mountains, the hygrometer indicates less and less evaporated water in the transparent air. I shall soon answer the hypothesis already mentioned, as having been imagined for setting aside the conclusion which I have deduced from this phenomenon, namely, a transmutation of the aqueous vapour into atmospheric air; a conclusion, however, which will be found the ultimate result of this series of facts

Art. IV. Another phenomenon, which M. de Saussure and myself have observed, proves that dryness is still greater in the region of the atmosphere above the highest mountains, where it was natural to suppose, and I supposed it at first, that the aqueous vapour was accumulating. On plains and small hills, moisture is increasing in the air after sunset; and before we possessed our hygrometers, we had reason to suppose that it was the same upon high mountains, for there also the grass becomes wet. This being the first common symptom of moisture observed after sunset, and even before, was one of the arguments in favour of the idea that dew proceeds from the ground; but the hygrometer, that neglected instrument, has shown it to be a phenomenon belonging to the physiology of plants, and not to meteorology. On high mountains, while the grass on the ground becomes wet, the hygrometer being suspended at some height above the ground, in some insulated spot where the air is free, shows an increase of dryness, which continues during the night. I have determined the cause of this phenomenon by immmediate observations; it proceeds from the condensation of the columns of air, while the heat diminishes in them; whence results that the part of that air which, during the day, rested on the summits of moontains, descending lower, is followed by the air which was higher before; and this, as long as the condensation continues in the lower parts, descending from higher regions, and thus passing over the

summits in its way downwards, is found, in an increasing degree, drier than that which rested on them in the day.

Art. V. Among the atmospheric phenomena, that of dew, commonly considered as very simple, has been long, and is still now, an object of controversy among natural philosophers who have not attended to the latest experiments and observations. The first and most plausible explanation was, that the dew descended from the air by the condensation of the evaporated water spread in it, when heat diminishes; but some experimental philosopher, finding that this cause was not sufficient to explain all the circumstances of dew, conceived the idea, which I have above mentioned, that it ascends from the ground, because this retains longer the heat of the day than the air above it, which circumstance was considered as increasing evaporation; both parties alleging in support of their opinion certain facts which, though not denied, were not decisive. During the most active time of this controversy, about sixty years ago, I made with my brother various kinds of experiments and observations, which, by turns, favoured one or the other of these bypotheses, but neither of them decisively; and the question would have remained for ever in suspense, had not hygrology and hygrometry been pursued with the degree of attention and labour that M. de Saussure and myself have bestowed upon them; from which the phenomena of dew has appeared under a new and quite different aspect, which excludes both the above causes as fundamental in it, and shows why neither of them could explain its most essential circumstances.

Art. VI. With respect to the experimental part, we have both determined, by direct and unconcerted experiments, the effects produced on our respective hygrometers, placed in a mass of air, wherein the quantity of evaporated water remaining the same, there was no change but in the degree of heat. We have made the same kind of experiments on different quantities of evaporated water in the same space; and combining them, we have formed tables expressing the different effects of heat on moisture, correspondent to different quantities of evaporated water in the same space, and to the changes of heat in each of these quantities; from which tables, after having observed the hygrometer and the thermometer in any part of the atmosphere, the quantity of evaporated water contained in one cubic foot of that air is determined. These entirely distinct experiments have proved the constancy of the laws prevailing in these effects, by the astonishing agreement of our tables, though determined by very different instruments and processes: an agreement which I have shown in the already mentioned papers to the Royal Society.

Art. VII. This determination of the effect produced on moisture, i.e., on the indications of the hygrometer by the changes of heat, in a mass of air wherein the quantity of evaporated water remained the same, was most essential in meteorology; and in particular it was indispensable for the decision of the question, whether the production of dew were principally owing to the cooling of the atmos

phere; which appeared the most natural explanation, but on which however there were sufficient reasons of doubt to produce the obscurity which remained on this phenemenon; because nothing could be either determined or proved, concerning the real effect of the diminution of heat on evaporated water, without such experiments as above defined; and I come now to their immediate application to the phenomenon of dew, in consequence of some observations which were also separately made by M. de Saussure and myself. Towards sunset and in the beginning of the night, moisture increases in the air much more rapidly; and after sunrise and in the first part of the day, dryness increases also much more rapidly; in both cases comparatively with the correspondent changes of heat, than would be the case did the same quantity of evaporated water remain in the air. This is a very succinct account of our experiments and observations concerning this object, the particulars of which may be seen in our respective works; but it is sufficiently distinct to allow me here to conclude, that thus has been pointed out one of the greatest questions and objects of investigation, concerning terrestrial physics, namely: what is the cause of the disappearance in the atmosphere of the greatest part of the aqueous vapour which it before contained, when the sun ascends on the horizon, and of the increase of its quantity when the sun is setting?-while the very reverse should have been expected from all the hitherto known causes, as I shall show hereafter. To this investigation I shall now proceed as far as known phenomenon will lead me.

Art. VIII. I shall first mention a very important course of observations of M. de Saussure concerning the changes in the electric state of the atmosphere. He had erected a high conductor, in a favourable situation, on the brow of a hill in Geneva. The lower part of this conductor was connected with an insulated pair of pith balls, the divergences of which indicated the differences between the electric state of the upper air and that in which the balls stood: he observed during many years the diurnal variations of this difference; and the main result of these observations is the following. In common weather, i.e., when no particular cause disturbs the course of the usual operations going on in the atmosphere during each period of twenty-four hours, the quantity of electric fluid increases in it from sunrise till some time in the afternoon, as is seen by the increase of a positive divergence of the balls. The new electric fluid, the formation of which is thus indicated, accumulates in the air, because it is transmitted but slowly to its lower part near the ground. But afterward, when the hygrometer shows a beginning of increase of moisture in the atmosphere, the divergence of the balls begins to decrease; and when at last dew is forming, the electric equilibrium is soon established between its upper and lower parts, the whole of the electric fluid formed in the day passing then into the ground. Now, it is during the first of these periods, that dryness increases in the atmosphere much more than would happen by the same increase