EVAPORATION

EVAPORATION is that process by which liquids and solids gradually assume the gaseous form, or by which, in certain circumstances, an invisible vapour or gas is continually detached from the surface of water, and perhaps of every other liquid, as also from the surfaces of many, if not every solid. From this process being generally found to lower the temperature of the evaporating body, an effect which might have been expected from the doctrine of latent heat, it is probable that heat enters largely into the constitution of every vapour, and is therefore supposed to be the principal agent. At a given temperature, a cubic foot or other determinate space, incumbent over a liquid, cannot contain above a certain quantity of the same kind of vapour; and when it contains that maximum, it is said to be saturated. If, therefore, after this is the case, any additional vapour continue to be thrown off, it must either fall back immediately into the liquid, or, if slightly cooled, may apparently remain suspended in the visible form of a cloud, which, however, is not properly vapour, but minute drops, formed by the cooling of it, and which successively fall back into the liquid. For evaporation continues to go on under the pressure of air or other gas, however dense, but ceases in the presence of any gas, however rare, if already saturated with the same vapour, unless such gas be colder than the evaporating body; in which case, however, the vapour being partly condensed into drops, becomes cloudy, as was just mentioned; and sometimes liquids, while in the act of ebullition, which is rather an extreme case of evaporation, throw up small drops, which may be seen alternately rising and falling in the vapour. Such is a summary of a few of the leading facts, several of which we shall have occasion to touch on more fully, along with other particulars, whilst briefly noticing, as we shall now do, some steps in the history of this inquiry, before proceeding farther.

The spontaneous disappearance of water, and the drying of humid bodies, or the tendency which moisture has to escape from them and diffuse itself in the atmosphere, and afterwards to descend as rain, snow, dew, &c. must have been familiar to mankind in every age, condition, or climate; but any thing like an approach to a rational expla-

Evaporation of the facts seems to be of very modern date. Some of the ancients, particularly Aristotle, conjectured that fire was concerned in the process; but such of their notions as have come down to us are so little to the purpose, that they need not now be dwelt upon. Halley supposed that aqueous vapour consisted of small hollow spheres, filled with something so very light, that they readily ascended and floated in the air,—a theory not altogether such as might have been expected from the known sagacity of its author; for, though Descartes seems to have hinted at the same thing long before, it scarcely appears to have had the least shade of truth in it. But so completely was Kratzenstein's imagination filled with this doctrine, that he fancied he could see these vesicles with a microscope, and even discern them to be hollow, notwithstanding their being supposed to form a transparent vapour. If he really saw any thing, it was more probably minute drops of water descending in the air. Nay, almost up to the close of the last century, we find many eminent men, and among these the late able mathematician Mr George Atwood, maintaining and speculating on this vesicular theory, as it was called. Halley, it is true, hinted at another theory, which not a few embraced, and for which some have still a predilection, namely, that air holds moisture united to it by chemical solution or affinity; but the first idea of this seems to be due to Hooke. A different explanation was attempted by Desaguliers, who, after pointing out the defects of former theories, showed that water is capable of being converted into a transparent elastic fluid, which he supposed to be many times lighter than air, and that it could therefore ascend and float in the atmosphere; but he some time afterwards inclined to ascribe the ascent of vapour to an electrical attraction of the air. Desaguliers was no doubt greatly mistaken regarding the specific gravity of steam, which, under the same pressure and temperature, is now known only to fall short of that of air in the ratio of five to eight; but he was perhaps the first who thought of identifying pure steam with the attenuated aqueous vapour diffused in the atmosphere. It would seem that prior to Desaguliers advancing this notion, it was not known that steam is always generated in a transparent state, although minute drops of water may be passing or falling through it; and indeed some are scarcely aware of this at the present day. The fact is, that steam or aqueous vapour preserves its transparency so long as it has neither lost heat nor had its volume enlarged; but so soon as its temperature is lowered, whether from losing heat or from enlargement of volume, it is more or less condensed into minute drops of water, which of course render it opaque. For we can on no account admit the very generally received doctrine of M. Clement, that equal weights of steam, if in contact with water, contain equal quantities of heat, whatever be their volumes or temperatures. On the contrary, we cannot help believing that a given weight of steam requires more heat to maintain it in the transparent elastic form, when under a greater volume, than under a less; and in proof of this, we presume it is only necessary to adduce the following fact. If we open a stop-cock in the cover of a boiler when the pressure is only one atmosphere, the steam which issues preserves its transparency till it gets to some distance from the orifice; but when the pressure within amounts to two or three atmospheres, the steam, if allowed to issue freely, is opaque from the very orifice. This clearly shows that the same quantity of heat (we do not mean temperature) which had been amply sufficient to maintain the steam in the transparent form under a smaller volume within the boiler, is quite inadequate to do so when the steam is allowed to enlarge its volume under a reduced pressure. A more palpable or satisfac-

tory proof than this, which has been long before the public, is scarcely to be wished; yet we find Dr Lardner not long ago reading a paper to the Royal Society on the heat of steam, and which has for its foundation the untenable theory of Clement. Indeed M. Clement's mode of experimenting was so fallacious, that it would have given him the same results whether his theory had been true or not. For when the stream of steam is very small, and of course passes through the stop-cock with difficulty, as it must have done in his experiments, it has, while in the act of dilating from under a force of two or three to that of one atmosphere, the opportunity of absorbing heat from the metal of the scarcely open stop-cock, then much hotter than 212°, the temperature to which the dilating is supposed to reduce the steam. No wonder then that equal weights of steam, at first so different in density, should, when received in equal quantities of cold water, produce equal rises of temperature; as is pointed out more particularly in the Philosophical Magazine for July 1826, p. 38. It is likewise to be remembered, that though in the other experiment the steam must have been absorbing some little heat from the stop-cock, this did not enable it to preserve the transparent form. We have been the more particular in clearing up this point, because we shall shortly have occasion to refer to it.

From the time of Halley and Desaguliers to about 1770, a variety of essays on evaporation made their appearance, which, however, tended but little towards elucidating the subject, because they were either not founded on experiments at all, or on such as were not sufficiently elementary. Some of these sought to establish the vesicular system, while others advocated that of chemical solution and electrical attraction. At length, in 1783, came Researches of Saussure. Saussure, who, though he did not labour so much to establish new theories as to blend together or combine some of those already mentioned, certainly added considerably to the stock of facts. But he at the same time greatly indulged his fancy in supposing aqueous vapour to be capable of existing in four different forms. When it alone occupied a space, he called it pure elastic vapour; when diffused in a transparent state in the air, he named it dissolved elastic vapour, imagining it to be chemically dissolved by the air; but after the air was quite saturated with this sort, he thought some of it assumed the form of vesicular vapour, only differing from that of Halley in being visible. To these he added concrete vapour, which consisted of liquid drops, such as form clouds or fogs; a real form of moisture to be sure, but one which is not properly vapour at all. In these Essais we have an account of a variety of experiments, most of them having the advantage of being executed on a great scale; but, on the other hand, they were often deficient in that elementary simplicity which is so important in a research of this nature, if we would avoid sources of fallacy. By inclosing a known weight of caustic alkali in a very large glass balloon, previously filled with thoroughly damp air, Saussure endeavoured to find what quantity of moisture this air contained, from the increase of weight which it communicated to the alkali; and conversely, by weighing a bit of wet linen, and suspending it in the balloon filled with dry air, he estimated how much moisture had evaporated from the linen to render the air thoroughly damp. These two methods not only gave him nearly the same result, but did so whether he used air hydrogen or carbonic acid in the balloon, and even when he employed a mixture of these gases. During these experiments he also observed, by means of a gauge, what change of pressure took place within the balloon from the change of humidity.

By operating at different temperatures, he obtained re-

Evaporation. suits from which different conclusions might be drawn to suit different theories; for instance, that the quantity of vapour which could exist in the balloon varied in geometrical progression for equal intervals of temperature, and nearly as the square root of the air's density; and that air of the ordinary density could contain five times as much vapour as a vacuum did; all of which conclusions are now known to be incorrect. Prior to this, Lambert had, from his experiments, drawn a very different but still more erroneous conclusion, that the quantity of moisture varied as the square of the air's density.

Theory of Deluc and Dalton. It was natural to think that the great discordance between the results of these celebrated philosophers should have raised doubts whether they might not both have deceived themselves by too complex a mode of operating, and, therefore, that neither of them was likely to be correct. Accordingly we find M. Deluc, in his Idées sur la Météorologie, hinting at an incomparably more simple theory, which he afterwards developed more fully in the Philosophical Transactions for 1792, viz. that the quantity of vapour which can exist in a given space depends upon the temperature alone, and is independent of the presence or density of air, or of any other elastic fluid with which the vapour forms no chemical combination, being the same as if nothing but the vapour occupied that space. In this speculation he also maintained, as Desaguliers had done, that pure steam and the attenuated aqueous vapour diffused in the air are the same. Something similar to this theory was advanced by M. Pictet in his Essai sur le Feu, and still more so by Dr Dalton in his Meteorological Essays; and it is not improbable that each might have hit upon the same theory before knowing what the others had done. By inclosing his hygrometer along with a little water under a receiver, Deluc found that the indications of that instrument were the same whatever was the density of the included air, or however much he exhausted it; and the same thing held true with other gases. Volta, likewise, from a great variety of experiments, was led to acquiesce in the theory of Deluc.

Dr Dalton's researches. The theory was farther investigated, in a much more elementary way, by Dr Dalton, as detailed in the Manchester Memoirs, vol. v.; but preparatory to noticing it, we shall subjoin his manner of expressing himself on the subject in his Essays, p. 201, where, from certain experiments and observations, he infers "that the vapour of water, and probably of most other liquids, exists at all temperatures in the atmosphere, and is capable of bearing any known degree of cold without a total condensation, and that the vapour so existing is one and the same thing as steam or vapour of 212° or upwards. The idea, therefore, that vapour cannot exist in the open atmosphere at a lower temperature than 212°, unless chemically combined therewith, I consider as erroneous; it has taken its rise from a supposition that air pressing upon vapour condenses the vapour equally with vapour pressing upon air; a supposition we have no right to assume, and which, I apprehend, will plainly appear to be contradictory to reason and unwarranted by facts; for when a particle of vapour exists between two particles of air, let their equal and opposite pressures upon it be what they may, they cannot bring it nearer to another particle of vapour, without which no condensation can take place, all other circumstances being the same; and it has never been proved that the vapour in a receiver from which the air has been exhausted is precipitated upon the admission of perfectly dry air. Hence, then, we conclude, till the contrary can be proved, that the condensation of vapour exposed to the common air does not in any manner depend upon the pressure of the air." The method which Dr Dalton took to clear up this question is distinguished by an elegant simplicity. In the first place, he determined by experiment the expansive force of dry air

for each degree of temperature between 32° and 212°; and was the first who did so with any thing like accuracy. He at the same time found that 1000 cubic inches of air at 32° expanded to 1376 on being heated to 212°, which comes surprisingly near 1375, the result published by M. Gay Lussac shortly after. But through some inadvertency Dr Lardner (on Heat, p. 57) makes Dr Dalton's result 1325, and then gravely adds, that "the latter (Gay Lussac's) determination has been proved by subsequent experiments to be the more correct." Secondly, Dr Dalton ascertained the force of pure steam in contact with water for each degree throughout the same range; and, thirdly, he found at what rate dry air expanded, when put in contact with water, and then heated through the various degrees of that range. The result was, that at any particular temperature, the expansive force of dry air in the first case, added to the force of vapour in the second, was exactly equal to their joint expansive force in the third. From this it is obvious that there is either no chemical combination between the air and vapour, or that such combination, if it exist, is quite inert, or goes for nothing so far as this research is concerned. Dr Dalton found, that when other gases were substituted for air, the vapour gave the same results; and similar results were obtained when other vapours were treated as steam. M. Desormes, likewise, by researches operating in a manner very different from that of Dr Dalton, found the quantity of vapour to be independent of the presence of air or other gases. The question was also examined experimentally by M. Gay Lussac, with his well-known accuracy and ability, and his results precisely agreed with those of Dr Dalton. Still more recently Professor Daniell has made a great variety of interesting experiments on mixtures of gases and vapours, in a manner similar to that employed by Deluc, but with the help of an incomparably better hygrometer; and his researches, too, corroborate the theory completely.

Several objections have been made to this theory; but all of them with which we are acquainted are so nugatory and irrelevant that we do not think they require any refutation. There is, however, a point which the foes as well as the friends of the theory have taken for granted, and which, though extremely probable, it would, for any thing we can see, be very difficult to prove in a direct and unexceptionable manner; viz. that the expenditure of heat in forming a given weight of vapour is the same, whether the process take place in air or other gas, or in a vacuum. Since in most combinations heat is given out, it is a wonder that those who contend for the solution of vapours in gases have not founded some objection on this; for we presume it is the only point on which a doubt could be palmed respecting the theory. It is generally taken for granted too, that the expenditure of heat in forming a given weight of vapour is the same at every temperature; but this assumption we cannot help regarding as very erroneous, because to us it appears quite clear that the expense of heat is greater when the vapour is rarer; and this we maintain for the following reason. The density of steam at 212°, for instance, is about 109 times greater than at 32° Fahrenheit; and we have already seen that steam loses in a considerable degree the elastic form by being dilated to two or three times its bulk; how much more then if dilated 109 times? Hence steam rarefied to such a degree would need a large addition of heat to enable it to maintain the gaseous form; in other words, much more heat must be expended or become latent in forming a given weight of vapour at 32° than at 212°. If steam, when kept separate from water, have its temperature raised by compression at the same rate as air has, it might be shown that the expense of heat in forming vapour at 32° would be that at 212° as thirteen to eight nearly, that is, in the

ratio of the logarithms of the number of times the steam is rarer than water. See Philosophical Magazine for April 1832, page 246.

Vapours and gases agree in so many of their properties, that there seems no doubt of their being but one and the same form of matter. Thus, so far as has been examined, every vapour, when kept separate from any sensibly evaporating liquid, expands by heat at the same rate as air and other gases do; and, like gases too, it observes the law of Boyle, by having the force at a given temperature proportional to the density. But since the density of a vapour in contact with its generating liquid increases with the temperature, its force, when so situated, increases much more rapidly than that of air or a mere gas does. However, the precise law which connects the temperature with the density or force seems to be as yet unknown. Dr Dalton having, as already mentioned, ascertained by experiment the force of aqueous vapour between the temperatures of $32^\{\backslash circ\}$ and $212^\{\backslash circ\}$, deduced from the results what he considered to be the relation between the force and temperature, and for some time it was supposed that by means of it the force could be had for temperatures beyond both limits of that range. Accordingly on this principle he constructed a table which extended from $-40^\{\backslash circ\}$ F. up to $325^\{\backslash circ\}$; and indeed he was pretty fortunate in the extension below $32^\{\backslash circ\}$, because the half of it has since been so nearly confirmed by the very accurate experiments of M. Gay Lussac, which were carried down almost to $-20^\{\backslash circ\}$ cent. or $-4^\{\backslash circ\}$ F.; but above $212^\{\backslash circ\}$ the supposed law was soon found to give the force too small, and the more so as the temperature is higher. Yet Dr Lardner, in his recent treatise on heat, gives this part of the table expressly as the result of Dalton's "accurate experiments," and not as it really was, a mere extension by means of a formula; nor does he warn us of its error almost two atmospheres at the upper extremity. It is likewise curious that Dr Lardner, while he notices performances of far less merit, is totally silent respecting the labours of the Committee of the Royal Academy of Sciences at Paris in 1829; though they executed by far the most extensive and accurate experiments that ever were made on steam at high temperatures, as will be found detailed in the Annales de Chimie for January 1830. All the English abstracts from this which we have seen are very incorrect. Dr Dalton has more recently published another table of the force of vapour adapted to some of his theoretical views; but from its differing considerably from the results of others, especially those of Gay Lussac, we do not consider it so correct for low temperatures as his first one, part of which is given farther on in this article.

The method employed by M. Gay Lussac to determine experimentally the force of vapour at low temperatures is particularly commendable, and yet its merits seem to have been very much overlooked, even by those who were engaged in similar researches. It may therefore be useful to give a brief account of it here. Other philosophers had either not carried their experiments below $32^\{\backslash circ\}$ at all, or had mismanaged the matter by allowing water to freeze in the tube in such a manner as to obstruct the free motion of the mercury, and likewise to render its altitude ill defined and difficult to be seen. These inconveniences were dexterously avoided by Gay Lussac, who still employed a barometer tube, but one so much longer than ordinary that a considerable part of it next the sealed end was above the mercury, and a portion of this, being bent to about an angle of $60^\{\backslash circ\}$, was immersed in a cold mixture along with a thermometer. The consequence of this arrangement was, that neither the mercury nor the small portion of water on its top necessarily required to be cooled down at all; because, however moderate their temperature might be, the force of every portion of the included vapour

being in equilibrio with the force of that which touched the coldest part of the tube, was just the same as if the whole vapour had been at the temperature of the freezing mixture. A farther advantage was, that after the apparatus had been allowed to remain for a short time in the state now described, the water entirely disappeared from the top of the mercury, having been transferred by evaporation or distillation to the sealed end of the tube, where it was quickly frozen. By this means the mercury became almost as free from moisture as if none had been in the tube. It is, besides, evident that this method could be carried down, not only to and below the freezing point of mercury, but to any temperature, however low, to which the extremity of the tube could be cooled; and that it is applicable to every sort of vapour which does not act upon mercury. In fine, Gay Lussac having placed a barometer tube filled with dry mercury beside and in the same cistern with the one which contained both mercury and moisture, read off the difference in the heights of their mercurial columns with great precision, by means of a microscope which could be shifted up and down a graduated pillar, and the difference was obviously the force of the vapour. Although mercury evaporates at all temperatures, as appears from a bit of gold-leaf becoming white if long suspended over it in a phial, and from the noxious effects of breathing air which has been for some time in contact with a large surface of mercury, yet the force of its vapour is so very small compared with that of steam, that it could not sensibly affect these experiments. Indeed it seems to be a general law, that the forces of vapours from liquids which have high boiling points, are quite inappreciable at low temperatures.

The most important problem connected with the evaporation of water is to determine the rate at which, under certain circumstances, it escapes from a given surface at a particular temperature. Many cases of this problem are still very imperfectly solved. The experiments of Dr Halley, detailed in No. 189 of the Philosophical Transactions, seem to be among the earliest which were employed for this purpose. He found that water salted to the state of sea water, and exposed to a heat equal to that of a summer's day, did, from a circular surface of eight inches diameter, evaporate at the rate of six ounces in twenty-four hours. Hence he concludes, that a square degree of sixty-nine miles will evaporate 33,000,000 tons in a day; and supposing the Mediterranean to contain 160 square degrees, it would evaporate 5,280,000,000 tons daily. This quantity he considers sufficient to supply all the rains, springs, dews, rivers, &c. It would however be of little use to rehearse the whole of this speculation, because the data, as we shall afterwards see, were too loose, and the experiment was on too small a scale, for such a comparison. Dr Lardner, in touching on this subject in his treatise on heat, says, that in the Straits of Gibraltar, instead of an outward current, there is a rapid and never-ceasing inward flow of water, and that he is therefore compelled to conclude that the evaporation from the surface of this sea carries off the enormous quantity of water supplied from this and other sources. Such doctrine from so respectable a quarter rather surprises us; since it is quite well known that, besides the inward current, which is principally in the middle of the surface, there are generally outward currents, likewise at the surface, next the coasts, particularly the south one, and most probably also in the immense deep beneath. According to Professor Robison, the water sometimes flows outward, even in the middle of the surface. Had there been no outward current, the Mediterranean must long ago have become as salt as the brine of the Dead Sea; or rather perhaps have become a rock of salt, as Dr Thomas Young remarked; for, as we shall afterwards see, no common salt or other muriate rises from the sea by evaporation. In

Evaporation. order that the same degree of saltiness may continue, the outward currents must just bear away as much salt as the inward current and other sources supply, with the very slight exception of what the wind carries off in the form of spray, water spouts, &c. The waters of the Mediterranean being, as is well known, specifically heavier, especially at a great depth, than those of the Atlantic, will naturally incline to buoy them up, and flow out beneath them in the infathomable depth of the Straits.

Dr Dobson's experiments. A series of experiments on the evaporation of water at natural temperatures was continued for four years, beginning with 1772, by Dr Dobson, at Liverpool, as detailed in the Philosophical Transactions, vol. lxvii. He took a cylindrical vessel twelve inches in diameter, and having nearly filled it with water, exposed it beside his rain-gauge of the same aperture; and by adding or withdrawing water, as occasion required, he kept the surface at nearly the same height. By comparing the total quantities thus added or taken away with the proceeds of the rain-gauge, the amount of evaporation was ascertained. The mean monthly evaporation for the four years was, in January, 1.5 inches; February, 1.77; March, 2.64; April, 3.3; May, 4.34; June, 4.41; July, 5.11; August, 5.01; September, 3.18; October, 2.51; November, 1.51; December, 1.49. In all, 36.78 inches. The mean rain for the same time was 37.48. For a great deal more in this way, and on the evaporation of ice and snow, see Howard's excellent treatise on The Climate of London. That

ice and snow evaporate, appears from their losing weight, or even disappearing altogether, if long enough exposed to a brisk and dry wind during frost. But sometimes snow is rendered so much colder by radiation than the incumbent still air, that it condenses the vapour of that air into heart-frost, and by that means augments its own weight or bulk.

But the experiments of Dr Dalton and Mr Hayle, though rather of local application, were no doubt far more accurate rate for evaporation from the soil. Having got a cylindrical vessel three feet deep and ten inches in diameter, two pipes were joined to it, and turned downwards for the surplus rain-water to run off into bottles; the one pipe being near the bottom, the other an inch from the top. The vessel was filled up for a few inches with gravel and sand, and the remainder with good fresh soil. It was then put into a hole in the ground, and the space around filled up with earth, except on one side, for the convenience of afterwards joining bottles to the pipes, and registering their contents. The earth in the vessel was now saturated with moisture, by pouring in water till some of it escaped from the pipes; and, lastly, the bottles were put to. During the first year the soil was bare at top; but for the next two years it was covered with grass, like any green field. A regular register was kept of the rain-water which ran off through the upper pipe, and also of what sunk down through the earth to the lower pipe. A rain-gauge of the same diameter was kept close by, to register the quantity of rain for any corresponding time.

From these experiments it appears, 1st, That the mean annual quantity of water evaporated in the above circumstances is twenty-five inches of rain, to which if we add five for dew, will give thirty inches of water raised annually: 2dly, That the evaporation increases nearly, though not exactly, in the same ratio as the rain; thus 1797 gave most rain, and the greatest evaporation, &c.: 3dly, That there is little difference between the evaporation from bare earth of sufficient depth, and that from ground covered with grass. It is however to be observed, that the numbers in the last or right hand column belong to the years 1799, 1800, and 1801; and that on account of accidents with the frost, these numbers are partly conjectural for January, February, March, April, and December. The mean annual evaporation during the same three years from ground covered with grass, exclusive of dew, was 23.5 inches, which falls a little short of that for the other three, which was twenty-five.

But there is reason to suspect that a less depth of soil in the vessel would have allowed more water to escape into the bottles, and vice versa. It is therefore in some degree a matter of conjecture that this apparatus indicated the true evaporation from the adjacent fields. Lahire having filled a vessel eight feet deep with soil, and sunk it into the earth, left it there for fifteen years. It had an opening at the bottom for the escape of the water, but never a drop came through. The quantity of rain, to be sure, being less at Paris than at Manchester, will so far account for this. In such an article as this it might be expected that, in conformity with custom, we should state the mean evaporation for all England, and a number of other principal countries. This however we are unable to do, because we are persuaded that to obtain even an approximate estimate would require experiments, such as those of Dr Dalton from the soil, to be made at least within every square mile of surface. The usual estimates being only derived from

a vessel with water, as will be noticed more particularly afterwards, afford no measure of evaporation from the soil.

The following relation between the temperature and the rate of evaporation from water, was discovered by Dr Dalton. Having, as above mentioned, determined experimentally the force through a considerable range, he was naturally led to examine whether the quantity of water evaporated in a given time bore any proportion to the force of vapour of the same temperature, and was agreeably surprised to find that they corresponded in every part of the range examined. Thus the forces of vapour at 212°, 180°, 164°, 152°, 144°, and 138°, are equal to 30, 15, 10, 7.5, 6, and 5 inches of mercury respectively; and the grains of water evaporated per minute are in the same ratio. This Dr Dalton considers as quite in conformity with the laws of mechanics; for the atmosphere seems by its inertia to obstruct the diffusion of vapour, which would otherwise be almost instantaneous, as in vacuo; but this obstruction, which is greater as the air is more dense, is overcome in proportion to the force of the vapour. Did the obstruction arise from the weight of the atmosphere, it would prevent any vapour from rising at temperatures lower than 212°; but according to Dr Dalton it is caused by the inertia of the particles of air, and is similar to that which water meets with in descending among pebbles.

The theory of evaporation was thus far settled so much the more easily, because the force of vapour already in the open air being very small compared with what is produced from water at high temperatures, did not sensibly affect the results; but Dr Dalton found, that if the theory was to be verified by experiments for low temperatures, regard must be had to the force of vapour already in the atmosphere. For instance, if water of 59° were the subject, the force of vapour at that temperature being a sixtieth of the force at 212°, one might expect the quantity of evaporation to be a sixtieth also; but if it should happen, as it sometimes does in summer, that an aqueous atmosphere to that amount does already exist, the evaporation, in place of a sixtieth, would be nothing at all. On the other hand, if the previously existing vapour were the 120th, corresponding to 39° F., then the effective evaporating force would be the 120th of that from boiling water. In short, Dr Dalton found that the evaporating force must in every case be equal to that at the temperature of the water diminished by that already existing in the air. To find the force of the aqueous atmosphere, Dr Dalton revived the method which had been employed by Leroy long before. He used in summer to take a tall glass jar, and fill it with cold spring water, fresh from the well. If dew was immediately formed on the previously dry outside, he poured the water out, let it stand a little to rise in temperature, wiped dry the outside of the jar, and then poured the water in again. The like process is to be repeated till dew cease to be formed, and at that instant the temperature of the water is to be carefully noted, for the purpose of obtaining, from a table like that which shortly follows, the corresponding force of vapour. Such temperature is called the dew point, or the point of deposition; because the air, if cooled to it, would be in a state of saturation with moisture, and of course ready to deposit dew, especially on any thing in the least colder than itself. This operation should be performed in the open air, or at an open window, because the air within is generally more humid than without. Spring water in this country being generally within a degree or two of 50°, will mostly answer the purpose during the three hottest months; in other seasons an artificial cold mixture may be used. But Leslie's or Daniell's hygrometer would be still more convenient at all seasons.

To observe the evaporation at atmospheric temperatures, Dr Dalton had two light tin vessels; the one six inches

diameter and half an inch deep, the other eight inches diameter and three fourths of an inch deep, and both made to be suspended from a balance. Water being put in one of these and weighed, it was placed in an open window or other exposed situation for ten or fifteen minutes, and again weighed to ascertain the loss from evaporation. The temperature of the water was at same time observed, the force of the aqueous atmosphere ascertained as above, and the velocity of the current of air noticed; for with the same evaporating force, a strong wind, by quickly changing the air in contact with the water, will double the effect produced in still air. From a great variety of experiments made during both winter and summer, and when the evaporating force was strong and weak, even in the case of ice, the results were found perfectly conformable with the theory. The same quantity is evaporated with the same evaporating force, thus determined as near as can be judged, whatever be the temperature of the air. Thus, if the dew point be 40° while the air is at 60°, the evaporation is the same as if the dew point were at 60° and the air at 72°; for 524 — 263, which is the difference of the forces of vapour at 60° and 40°, is nearly equal to 770 — 524, the difference of the forces at 72° and 60°; but two such differences could be had exactly equal by using fractions of degrees of temperature, and interpolating between the other numbers.

The following table exhibits the ratios and quantity of water evaporated at different temperatures, derived from the preceding theory, and confirmed by experiments. The first column is the temperature; the second the force of vapour; the rest give the number of grains evaporated from a surface six inches in diameter, supposing the air to be previously dry. The third column is calculated on the supposition that at the temperature of 212° the evaporation per minute from the said surface is 120 grains; the fourth, that it is 154 grains; the fifth, 189, according to the strength of the wind; for if the air in contact with the water be continually changed, so that the recently moistened portion may be speedily removed and drier air substituted for it, the process will be proportionably expedited. These columns present the extremes and mean of evaporation likely to be noticed when the process goes on in air of the ordinary density. The effects of different densities will be considered afterwards.

As an example of the use of the table, let the dew-point be 52°, the temperature of the air 65°, with a moderate breeze, to find the evaporation per minute from a vessel six inches in diameter. The number in the fourth column opposite 52° in the first, is 2.06, and that opposite 65° is 3.16; the difference, 1.1 grain, is the evaporation required. Again, if the evaporation per minute, with a brisk wind, be 1.7 grain, while the air is at 62°; required the weight of the aqueous atmosphere, and the dew-point. In the fifth column the number opposite 62° is 3.52, being the whole evaporating force at that temperature in perfectly dry air, from which 1.7 being deducted, leaves 1.82, which, in the fifth column, is opposite to .294 inch of mercury in the second, the weight of the column of vapour, and to 43° the temperature or dew-point. That is by using the nearest numbers

in the table. Greater exactness may be had by interpolating between them. In this case, for instance, the weight should more correctly be .290, and the dew-point 42°.6. Such estimates, however, are liable to uncertainty, where any doubt remains respecting the velocity of the wind.

Dr Dalton found that the same theory held with regard to the evaporation of other liquids, and was even more easily verified by experiment in their case than in that of water, on account of the air previously containing little or none of their vapours. In short, the evaporation in their case is similar to that of water in perfectly dry air.

Since vapours, so long as they maintain the gaseous form, expand and contract by change of temperature precisely at the same rate as air does in like circumstances, and since mixtures of air and vapour do the same, it follows, that if at any temperature above the dew-point the pressure of the unconfined air have to that of the vapour in it a certain ratio, it will have the same ratio to it at any other temperature not lower than the dew-point. Hence, because the entire pressure is constant, the force of vapour at the actual temperature of the air is exactly equal its force at the dew-point or temperature at which the air under the original pressure becomes saturated with moisture, and at which all the moisture in it just barely maintains the gaseous form.

Dr Anderson, in giving a table of the weight and force of aqueous vapour in vacuo (Edin. Encyc. xi. 578), says, it is adapted to a pressure of thirty inches, and that when the barometer differs from thirty, the numbers in the table (none of which denote atmospheres, but grains and inches) must be altered in the same ratio as the pressure. It is difficult to conceive how he could fall into such a palpable mistake. Vapour in vacuo being protected from atmospheric pressure, has obviously nothing to do with its changes or amount; and we have already seen that the maximum force and density of vapour in air, as well as in vacuo, depend solely on the temperature. Yet Dr Anderson (ibid. p. 581) makes the maximum weight of vapour which can exist in a cubic inch of air at a given temperature to be proportional to the barometric pressure, which is remarkable, considering his professed adherence to Dalton's theory. It is further curious that the doctor, not perceiving that the actual force of vapour in the open air must be exactly equal to its force at the dew-point, gives (on same page) a formula for reducing the one force to the other; and he even takes it for granted that Dr Dalton is in the same error, though the very reverse appears from all his writings, and especially those already quoted.

The rate of evaporation is very different in air of different densities, being greater when the density is less, and of course vice versa. Some experiments on this subject are given by Professor Daniell, in the second and much improved edition of his valuable Meteorological Essays, published in 1827, page 493, from which we make the following brief abstract:—Upon the plate of an air pump was placed a Mr. flat dish 7.5 inches diameter, containing sulphuric acid, and covered by a receiver which but just passed over it, so that the base of the included air rested everywhere on the acid. In the middle of the dish was a stand supporting a vessel 2.7 inches diameter, and 1.3 deep, which was filled with water to the depth of an inch, and had a delicate thermometer resting on its bottom. The water, having been previously freed from air, was weighed with a very sensible balance, and then exposed to the action of the sulphuric acid. Its temperature was 45°, and the barometer at 30.4 inches. At the end of half an hour it was again weighed, and found to have lost 1.24 grain by evaporation. Having been replaced, and the air rarefied till the gauge stood at 15.2 inches, it was found to have lost in half an