A Second Paper on Hygrometry. By J. A. De Luc, Esq. F. R. S.

I. A Second Paper on Hygrometry. By J. A. De Luc, Esq. F. R. S. Read December 9, 1790. In a Paper which I had the honour to present to the Royal Society in the year 1773, I sketched the following propositions, as fundamental for the construction of an hygrometer. 1st, That fire, considered as the cause of heat, was the only agent by which absolute dryness could be immediately produced (§ 5.). 2d, That water, in its liquid state, was the only sure immediate means of producing extreme moisture in hygroscopic bodies (§ 8. and seq.). 3d, That there was no reason, a priori, to expect, from any hygroscopic substance, that the measurable effects effects produced in it by moisture were proportional to the intensities of that cause; and, consequently, that a true hygrometrical scale was to be a particular object of inquiry (§ 2.). 4th, lastly, That perhaps the comparative changes, of the dimensions of a substance, and of the weight of the same or other substance, by the same variations of moisture, might lead to some discovery in that respect (§ 72.). The same propositions will be the subject of this Paper. Of absolute dryness. 1. An hygroscopic body, which is not brought into contact with any other body drier than itself, cannot lose any part of its moisture but by evaporation; and if this is entirely produced by fire, there may be such a degree of heat as will cause the total evaporation of that moisture. This is the principle on which the above first proposition was founded; but at the same time I mentioned, that I had not brought it into practice, because of the impossibility of submitting the substance of the hygrometer to such a degree of heat. However, I soon removed that difficulty by considering, that the degree of heat necessary to produce extreme dryness, might be applied to some substance that could bear it; and that dryness be transmitted to the hygrometer, by inclosing it with that substance in a proper vessel. The substance I chose was potash; and I prepared, for this and some other hygrometrical purposes, an apparatus which was made by Messrs. Nairne and Blunt in 1776. But a new objection stopped me again in that pursuit, and led me for some time to a very great and now almost useless labour. The degree of dryness produced by potash so used, could be only proportionable to the degree of heat that it had received; and not not conceiving yet any known limit to the intensity of heat, I could not expect any limit to dryness, nor even a fixed degree of it. 2. I remained at that point, with however a comparable hygrometer laboriously constructed, till I came to conceive, that heat must be at its maximum in a body, when it is incandescent; which opinion I have explained in my work, Idées sur la Météorologie. From that first idea I soon after concluded; that every hygroscopic substance, which could retain that property after having been brought into incandescence, would answer my first purpose. The following is the theory resulting from the whole of the above considerations. 1st, The hygroscopic substance which has the most capacity for moisture, and receives it the most readily, being placed in any quantity in a given space, cannot bring that space to a degree of dryness greater than its own; and if that degree is undetermined, it cannot afford any fixed point for the hygrometer. 2d, The hygroscopic substance which has the smallest capacity for moisture, and is the slowest in receiving it, if it is really reduced to extreme dryness, will have the power of producing it in a given space, provided its small capacity be compensated by a greater quantity, and its slowness by more time. 3d, Every hygroscopic substance, which may be brought to white heat without losing its property, is fit to produce extreme dryness in a close space. 4th, It is indifferent for that purpose, that the substance used be of the class which has a chemical affinity with water, it being sufficient that, after having been reduced to extreme dryness, it be still capable of receiving it from the ambient medium, as may every porous substance. 5th, But for the practical purpose of fixing the point of extreme dryness on hygrometers, such a substance must be chosen as, with a great capacity capacity for moisture, receives it but slowly: as by that first property it may be taken in less quantity; and by the last, it will be less subject to acquire a sensible quantity of moisture in the time necessary for the operations. 3. Pot-ash and some other alkaline substances afforded the first of those properties, but not the last; and I had not fixed on any substance, when, being at Birmingham in the autumn of 1782, Mr. James Watt informed me, that his friend Dr. Black had found in quicklime a great capacity for moisture, and much slowness in retaking it: this he knew, by having kept a long time the same lime in a close vessel, for drying salts and capillary tubes for thermometers. These were the very properties I wanted for my purpose, which thereby I executed as soon as I came home. I made those first operations in small glass vessels, using old lime, which I brought again to white heat every time I used it. These first trials agreed with my theory in its first point; that of producing constantly the same degree of dryness: as for the second, namely, whether that degree was extreme, it depended on other experiments. 4. Being sure then of a fixed degree of dryness, the number of experiments I undertook made me wish for a means of avoiding the frequent repetitions of bringing again my lime to white heat; and having found one which has succeeded, I am going to describe the apparatus. The vessel, fig. 1. is of tin, 3 feet high and 1 in diameter. A glass plate \(a, a, a, a\) is fixed at the top, forming a vertical section of the cylinder at 1 inch distance forwards from the axis. A woven brass-wire cage \(b, b, b, b\) is fixed in the vessel through its diameter, in order to keep a space for the instruments; for the same purpose it is open at the top, and also opposite the glass, where the dials of the instruments are to be seen. For my experiments, ments, which required instruments of various sizes, I made that cage 18 inches high and 2 deep; but it may be much smaller for common hygrometers. The whole vessel, except that space, has been filled, through the openings c, c, with quicklime taken from the kiln, and suffered only to lose the red heat; after which the openings were covered with heaps of the same lime, which absorbed the moisture of the air entering the vessel while it was cooling, and then the openings were shut with tin plates and putty. The top of the vessel has four square openings d, d, d, d, correspondent to the wire cage, for the introduction of the instruments, which are hung to hooks. I use a hooked wire for putting in or taking out the instruments, to avoid bringing my fingers near the openings. These are kept shut with tin plates and putty: I never open but one at a time, which I leave open as little as possible; and to prevent the introduction of the external air in those short operations, I make them as nearly as possible at the same temperature, which being 60° of Fahrenheit may be obtained in every season. With these precautions, and also by moist air being lighter than dry air, there is scarcely any moisture introduced in the vessel except by the instruments themselves. 5. This application of my method afforded me a very strong confirmation of the practical fixity of the point of dryness produced in that manner by lime: for the apparatus was different from the former ones; 1st, by the quantity of the lime; 2d, by the lime having been put very hot into the vessel, while, when I used glass vessels, I had suffered it sometimes to cool down to 60°; 3d, by that lime being of the first calcination, instead of old lime brought again to white heat; and all these differences produced no sensible effect on the point of dryness. Since that time Messrs. Nairne and Blunt, Mr. Hurter, and Mr. Haas, Haas, have made apparatuses of the same kind, and I have made myself some others of different sizes and shapes; and they all produce sensibly the same degree of dryness. 6. The described apparatus was ready in the month of October, 1787, and I put in it one of my first hygrometers; which, in a few days, came to its fixed point of dryness, and there it has remained ever since, though I have opened the vessel above four hundred times. That degree of constancy, much beyond my expectation, has enabled me to make a variety of experiments, which else had been next to impossible: it proceeds partly from the great capacity of quicklime for moisture, which I shall determine hereafter; and partly from its slowness in receiving it; which circumstances, added to the small size of the openings, to their being at the top of the vessel, and to the care of putting in and taking out the instruments nearly at the same temperature, prevents the lime from acquiring any sensible degree of moisture during these operations. 7. I did not trust at first the apparent continuance of the same degree of dryness in that vessel. At the end of nine months of frequent use, I began to fear, lest the whalebone, of which the standard hygrometer is made, had been impaired; and I took it out, to try its point of extreme moisture (of which I shall speak hereafter;) but it came exactly to that point, and when put again into the lime-vessel it returned where it stood before. I have repeated many times that trial, with the same result; the last time was at the end of three years, when, instead of a loss of expansibility in the whalebone, I found a small increase, but probably accidental; it went a little farther than its point of extreme moisture, and came back to its constant point of dryness. 8. The Hygrometry. 8. The principles of hygrometry being now my only object, it would not be proper to enter into particulars on its practical part; but I shall here mention for once, that the steadiest hygroscopic substances are subject to anomalies: for instance, after an hygroscope has remained fixed in water for many hours, if it is taken out, suffered to dry a little, and then put again into water, it may sometimes happen to overpass that point. In the same manner, after an hygroscope has been long fixed in the lime-vessel, it may happen also, that in taking it out only for a quarter of an hour, and putting it in again, it will move a little farther than it was before. Again, if in taking it out of the lime-vessel, where it had long remained fixed, it is put into water, and then back into the lime-vessel, it may happen, that it will fix itself a little short of its former point, and never move thence, except by repeated great variations of heat; but if, when it shews that disposition, it is taken out for a short time, and put in again, it will then attain its usual point. This was the case in the last trial of my standard. Lastly, the same anomalies may take place at every other point of the scale of every hygroscope, only more or less according to the substances; some of which, for that reason, cannot be used for practical hygrometry. 9. Those anomalies of the steadiest hygroscopic substances, will probably prevent our ever having in the hygrometer an instrument nearly so exact as the thermometer; and this I was to premise, that when I mention the results of particular hygroscopic experiments, it may be understood, that they have only the degree of exactness that belong to their clasps. Luckily those anomalies are yet of no consequence for the great objects of hygrology and meteorology; the present state of hygrometry being sufficient to excite on those objects, questions of great importance for natural natural philosophy. And in the mean time those anomalies are very interesting in themselves; as, from their laws, they seem to point out some modification of cohesion, as being the immediate cause of elasticity in solids. If I can find time to put in order a number of observations and experiments I have made in that respect, I intend to make it the subject of a Paper, in which I shall examine, from a general result of those phenomena, the comparative use of weights and springs for keeping stretched the hygroscopic substance in hygrometers. 10. After fixity in the degree of dryness produced by lime in the manner I have explained, the next point to be examined in respect of my theory, was, if the nature of the substance brought to a white heat, had any influence on the degree of dryness thereby produced; and in order to try at once the effect of a very great disparity, I chose such a sand-stone as is not affected by acids, and strikes fire from steel, before and after having been incandescent. The first experiment I made was with a view of finding the comparative capacities for moisture between that stone and lime. For that purpose I took such pieces of them as might be readily reduced to half an ounce while incandescent; which being done, I put them into brass cups, fitted to a scale, and I inclosed them under a glass vessel inverted over water. I weighed those substances from time to time; each of them continued to acquire weight during five weeks; at which time the sand-stone had gained $\frac{1}{25}$ part of its original weight, and the lime $\frac{1}{25}$: this last was at that time all cracked and fallen in small fragments, easily reduced to powder; the sand-stone struck fire as before. I next prepared a cylindrical tin vessel, 10 inches in diameter and 14 inches high, with a glass top, which I filled with fragments of that stone, treated as the lime; and when it was cooled, I put into it an hygrometer, whose fixed point of dryness has been taken in the lime apparatus: and in five weeks it was fixed to the same point. This is a demonstration, that the nature of the substance does not interfere with the degree of dryness produced, and that incandescence is the only cause of its fixity. 11. Lastly, in respect of hygrometry, a degree of dryness thus determined might have been sufficient; but for hygrology, and even for natural philosophy in general, it was desirable to discover if that fixed degree of dryness was also absolute: and the following are the considerations which directed me in that enquiry: if evaporation is produced by heat only, and if incandescence is the maximum of heat; an hygroscopic body, which is brought to incandescence, cannot contain any evaporable water; and if that body has such a mass, as to be capable of absorbing all the water evaporated in a certain space, without acquiring any measurable moisture, that space may be called absolutely dry. Now, if an hygroscopic substance which is inclosed in that space, contains any sensible quantity of evaporable water, when heat increases, that substance must lose a part of its moisture in the medium, and take it back by the diminution of heat. Consequently there was a means of discovering, if hygroscopic substances, reduced to the above degree of dryness, still retain a sensible quantity of evaporable water; it was that of observing their weight, by changes of heat: and from these previous considerations I made the following experiments. 12. I hung successively to a very sensible beam, shewing the changes of weight by an index, different sorts of vegetable and animal substances, the beam being inclosed in a glazed tin-vessel, containing a sufficient quantity of quicklime; and during the operations, I produced from time to time great changes in the temperature of the vessel. As long as these substances retained Mr. de Luc on retained a sensible quantity of evaporable water, the increase of heat made them lose some weight, which they regained partly when the heat returned to the same point. But that effect diminished by degrees; and, at last, a change of 30° of Fahrenheit did not produce any sensible change of weight in those substances, though they were such as had a great capacity for moisture. An hygrometer placed near the beam, was then at the point taken in the lime-vessel. That single experiment confirms all the previous considerations from which I had expected absolute dryness from incandescence. Of extreme moisture. 13. The second proposition I had sketched in my first paper, is this "that water, in its liquid state, is the only sure immediate means of producing extreme moisture in hygroscopic bodies." 14. Moisture, the nature of which we are first to determine, may be considered in three different cases.—1st, In substances which have an affinity with water; by which their molecules and those of water may unite, and form a new compound.—2dly, In substances which have no affinity with water, but to which water has a tendency to adhere; by which cause it enters their capillary pores.—3dly, In the medium, or space free from visible bodies. I have not undertaken to discover what, in the first of these cases, might properly be called moisture, and its degrees; as I foresaw great difficulties in that undertaking, which besides was unnecessary to my principal pursuit: therefore I come immediately to the second case. 15. When I wrote my work Idées sur la Météorologie, having not yet made some experiments I had in view to verify the opinion I entertained, that vegetable and animal substances, as well well as the porous mineral ones, received water merely by the faculty of capillary pores, I used sometimes the expression hygroscopic affinity, in treating of the hygoscopic equilibrium; but before that work was come out of the press, having conversed on that object with Dr. Blagden, and found him inclined to that opinion, I had time and opportunity to express it, in § 276, as follows: "There are reasons to doubt, whether some of the substances, which share amongst them the water disseminated in a space, do not suck it, by a faculty similar to that of capillary tubes, without any chymical affinity with water." That opinion will be now confirmed by the following experiments. 16. 1st. Exp. Sugar has an affinity with water, and no sensible one with alcohol: however, a lump of sugar will imbibe this last liquid as readily as the first. Consequently, water is not imbibed by sugar in consequence of their affinity; since alcohol is also imbibed: they both ascend in sugar, by the faculty of its capillary pores, as they do in sandstone or in sponge. But when water has thus entered sugar, it dissolves it; which then is a chymical effect; whereas alcohol evaporates, and leaves the sugar sensibly as it was before. 17. 2d. Exp. If water penetrated hygroscopic substances of the vegetable and animal kinds, by an affinity with them, it would not be natural to expect, that other liquids, which do not shew an affinity with the same substances as water, should penetrate the above-mentioned substances. Being led by that consideration, I made two hygoscopes of different elastic animal substances; and after having marked the point where they stood in water, I immersed them successively in alcohol and in ether; by which liquids they were expanded nearly as much as by water, and they contracted as much in coming out of them. In those experiments a singular phenomenon happened in both hygroscopes. The first effect of the immersion of those animal substances in alcohol (and I suppose it would have been the same with vegetable ones) was contraction, soon followed by expansion; and when they came out into the air, the first effect was expansion, soon followed by contraction. The cause of that reciprocal phenomenon is undoubtedly the affinity of alcohol with water. In the immersion, some of the moisture came out of the substance, to unite with the surrounding alcohol; by which loss of water, a contraction took place in the substance, till it came to imbibe the alcohol itself. In the immersion, some of the moisture of the air, uniting immediately with the alcohol retained by the substance, expanded it to the same degree as if it had been in water; after which, the alcohol evaporating, the substance contracted. Those phenomena did not happen with ether; this not uniting readily with water; but it expanded those substances as much as alcohol, and nearly as much as water. From those phenomena we may conclude, that the penetration of animal substances by water, and consequently by moisture, is produced, as that of sugar, sandstone, and every other porous substance, by the faculty of capillary pores, without any affinity between them and water. 18. 3d. Exp. In that theory, of a mere imbibition of water by hygroscopic substances of the elastic kind, a circumstance, which seems to point out affinity, was to be explained; it is that of the hygroscopic equilibrium. In view of that object, I made the following experiments, not new in themselves, but directed to my purpose. I took some glass tubes, of different small bores, which I first bent in the shape of syphons; after which I cut them in the middle of the bent part. This was to enable me, to bring into exact communication the lower end of two tubes, though though held in a vertical position, as an inverted syphon (Fig. 2.) The following are the experiments.—1st. Exp. When a column of the liquid had ascended in one of these tubes; if I applied to it an empty tube of the same bore, the liquid column divided itself equally between them.—2d. Exp. When the empty tube was of a smaller bore, that column rose more in it, than it sunk in the other; and the contrary happened when the empty tube was of a larger bore.—3d. Exp. When some more liquid was supplied to the united tubes, it rose in both, in proportion to the respective heights of the former columns.—4th. Exp. When a superabundant quantity of liquid was supplied, it rose to a maximum in each tube, and the heights of the columns increased in some proportion with their former heights. 19. These known facts have a clear analogy with the hygroscopic equilibrium in elastic substances.—1st. Case. When the quantity of liquid common to capillary tubes is not sufficient for them to receive their respective maximum, they share it between them, and the equilibrium takes place, when there is, in each of them, the same ratio between its specific capillary power and the weight of the raised column. In the same manner; when the quantity of water disseminated in a space, is not sufficient, for several hygroscopic substances to receive the maximum of water which they can contain in their pores, they share it amongst them; and the equilibrium is produced, when there is in each of them the same ratio, between its specific capillary power, and the resistance of their pores to be more dilated.—2d. Case. When there is a superabundant quantity of liquid common to some capillary tubes, each of them receives its maximum; which is determined by an equilibrium, between its total capillary power, and the weight of the raised column. In the same manner, when there is a superabundant quantity of water common to several hygroscopic substances, each of them receives its maximum; which is determined by the equilibrium, between their total capillary power, and the resistance of their pores to be more dilated. That final equilibrium, which, from its very nature, cannot be overpassed in any elastic substance properly used, determines the specific capacity of those substances for moisture. 20. Moisture then, considered in porous bodies not soluble by water, may be defined, "A quantity of water, which is invisibly contained in their pores; without any other connection with their substance, than that which it has with the glass of the capillary tubes into which it has ascended." 21. We may see now whence proceeds the hygroscopic equilibrium between elastic substances inclosed in a space, either filled with air or deprived of it. In this explanation it is unnecessary to determine, how water is invisibly disseminated in spaces free from visible bodies, therefore I shall not enter here into this subject; that dissemination is a fact admitted in every hypothesis, consequently the medium is only to be considered as the stock and standard of moisture. By the cause, whatever, of evaporation, hygroscopic substances lose or gain water in the medium, according to its degree of moisture, till they are in equilibrium with it; which implies the equilibrium amongst themselves according to the laws resulting from their own nature. 22. Now moisture, in a general sense, will appear to be, "a quantity of invisible water, either evaporable, or evaporated." And from that definition, the maximum of moisture will exist, when, "every circumstance remaining the same, no more water can be admitted in a space, without becoming visible; on solid bodies, by their surface being wet; and in the medium, "by a spontaneous precipitation of water." Lastly, as immersing solid hygroscopic bodies in water, or exposing them in a medium where there is an actual precipitation of water (as in a fog) is an effectual means of furnishing their pores with the whole quantity of water they can imbibe; it is evidently a sure means of producing extreme moisture in them: and this point cannot be overpassed, neither in water, nor in fog, since it depends upon the resistance of the pores to further dilatation by the mere introduction of water; but it must be attained in an hygrometer, to fix its point of extreme moisture. 23. When formerly I had fixed upon that method for procuring to my first hygrometer a true point of extreme moisture, it occurred to me, that the temperature of water might influence sensibly the expansion of its ivory tube; and in order to discover if it was so, I made some experiments, related in §§ 104 &c. of my former paper, the result of which was, that the temperature of water had a sensible effect on the expansion of ivory. But soon after I distrusted some modifications of that complicated hygrometer; and especially this particular result. 24. I then changed that first method; which consisted in measuring the changes of capacity of hollow cylinders, into that of measuring the changes in length of hygroscopic substances; and for some preliminary experiments on many of them, I made particular frames, in which, by a combination of glass and brass, the effects of heat on those materials compensated each other; by which means the indices of those instruments were only affected by the modifications of the hygroscopic substances tried in them. I have mentioned these frames in a paper on gyrometry, printed in the Phil. Trans. for 1778. With these frames I first tried ivory, in water of different temperatures; and and I found a very little difference in its expansion, comparatively with what had appeared from my first experiment. Then, continuing the same trial on various substances, I found the effect of different temperatures of water very small in general; and even in some substances, as deal taken lengthwise, and hemp, I could not ascertain any. 25. These experiments led me to think, that the small variations produced by heat in hygroscopic substances dipt in water, were not hygroscopic modifications, but the mere effects of heat, by the cessation of all hygroscopic modifications; these having then attained their maximum: which is a discrimination of effects, that I had vainly attempted to produce by other means. When afterwards I had found the method of producing extreme dryness, I made a lime apparatus, for the purpose of repeating in it the same experiments with my compound frames; and I found that theory confirmed, by the effects of heat in that apparatus being nearly the same, on the same substances, as when they were in water. 26. From the whole of the foregoing experiments there cannot remain any doubt, that water, in its liquid state, is a sure means of fixing the point of extreme moisture on hygrometers. Particularly, in respect of elastic substances, as ivory, quill, whalebone, all sorts of wood, and a number of others which I have tried, the last experiments in water of different temperatures, afford an immediate proof, that their faculty of sucking water has a fixed limit, proceeding from a final resistance of their pores, to be more dilated by the introduction of water. Consequently, their utmost expansion is a true sign, that moisture is extreme in them; which point cannot be exceeded. But my proposition extended farther: I had said, that water was the only certain means of obtaining immediately the point of extreme extreme moisture on hygrometers; and this is a most important question, both of hygrometry and hygrology, which remains to be examined. On the maximum of evaporation, and its correspondence with the maximum of moisture in a medium. 27. Since moisture consists in invisible water, an excess of water is the only immediate means of ascertaining that the maximum exists; as, if a reservoir is above our reach, the only means of knowing if it be full of water, is when it overflows. From that principle, a fog gives the point of extreme moisture on hygrometers, like water itself; because it covers very soon the hygroscopic substance with a coating of water: sometimes even it expands it a little more than an immediate application of water; but this belongs to an object that I have waved before, as relating to some modifications of elastic solids (§ 8.). No other means then but an excess of water over the surface of the hygroscopic substance of the hygrometer, can ascertain that it is arrived at its point of extreme moisture; and the first immediate demonstration I shall give of it, will be afforded by dew; a very uncertain, though apparently certain sign of extreme moisture in the air. We say, that there is dew, when some solids exposed in the open air in a clear evening are wet; but if that was the effect of a precipitation of water happening in the air, all the solids thus exposed would be wet; which is far from being the case; consequently, that phenomenon must proceed from some particular causes, by which, though no water is yet disposed to abandon the medium, it gathers on some particular solids. It is very long since the phenomena of dew have perplexed natural philosophers; and they were the first which I studied in the beginning of my researches in meteorology; but all that I concluded from my experiments and observations was, that we could not understand those phenomena without first having a sure hygrometer. This is the reason why, soon after I had made my first hygrometer, I exposed it in the open air in the country, suspended very little above the grass, from the morning of a fine day to the time of dew in the evening; the grass grew wet, and the hygrometer remained at a great distance from the point which had been fixed in water. I have related that experiment in § 91, of my first Paper. 28. When I had made hygroscopes of various sorts of slips; for instance, of different woods and of whalebone, cut across the fibres; of ivory and horn, reduced first into thin tubes, and then cut in screw; and of quills, by cutting also in screw their barrels; I repeated, with those instruments, my observations on dew; and to give a short, but determinate idea of the phenomena I observed, I shall reduce them to some general cases, as indicated by one only of those hygroscopes, that of quill, which, like all the others, is divided into 100 parts, from extreme dryness to extreme moisture. These hygroscopes were suspended in the open air, three feet above a grass-plot in the country. 1st Case. When a clear and calm evening succeeds to a clear and warm day, the grass frequently grows wet, though the above hygroscope stands many hours, and sometimes the whole night, between 50 and 55. 2d Case. If the dew increases, so that taller herbaceous plants and shrubs grow wet in succession, the hygroscope moves more and more towards moisture; and when it is come to about 80, plates of glass and oil-paint also grow wet; but at that period, neither metallic plates, exposed like the glass ones, nor some shrubs and trees, are wet; and this also may last whole nights. 3d Case. If the dew proceeds to its maximum, the hygroscope moves from 80 to 100 (and sometimes a little farther, § 27.). Then we have also a certain proof that extreme moisture exists in the air; for every solid body exposed to it is wet. But it is only at that moment that we can depend on extreme moisture existing; for, if in the other described stages of the phenomenon, the appearance of water on the surface of some solids had proceeded from a spontaneous precipitation in the air, all the other solids ought to have been wet; but they only become wet in a certain succession, and in the mean time the slip of quill, and all the other above-mentioned hygrosopes, move more and more towards their point 100, in sign of moisture increasing in the air. Consequently (as I had concluded from my first observations), instead of having in dew an hygroscopic standard for the hygrometer, we have in its phenomena many circumstances which will only be explained with the assistance of that instrument. 29. Some previous observations had also warned me against the general idea, that moisture was to be extreme in the air, when there was a sufficient quantity of water in the space, even though that air might be supposed to be filled with evaporated water to its maximum; and the doubts I entertained in that respect were the cause of the difficulties I expressed in the beginning of my first Paper, which I only got over when I thought of water itself, to obtain the point of extreme moisture on my hygrometer. This was also the reason why, as soon as my first hygrometer was finished, I placed it in a cellar, the walls and ground of which were wet, and where it continued two months, without ever attaining its point of extreme moisture. I have related that experiment in § 54. et seq. of my first Paper. 30. When also I had the hygrosopes mentioned above in the observations on dew, I undertook a very long course of various sorts of experiments on that important point of hydrology, Mr. De Luc on grology, of which however I shall only give here the general and constant results, as furnished by those hygrosopes. "The maximum of evaporation in a mass of inclosed air is far from being identical with the maximum of moisture; this being dependent also, even to a very great degree, on the temperature of the space, supposed to be the same, or nearly so, as that of the water which evaporates in it. Moisture may arrive to its extreme in an inclosed air, if that common temperature is near freezing point; but it becomes less and less, even to a very dry state, as that temperature rises, though the product of evaporation, thereby increasing, continues to be at its different maxima, correspondent to the different temperatures." This is a very important proposition in hygrology; which, from my experiments, would not be subject to any objection, if there were no other hygrosopes than those I have mentioned above, of which I have thirteen different species; but there is another class of such instruments, from which some doubt might first arise; and I come now to that point. On two distinct classes of hygrosopes. 31. As I shall now frequently speak of slips and threads, which constitute those two classes of hygrosopes, I must first explain what I mean by those words. The slips compose the class of hygrosopes used in the above experiments; they consist of very thin and narrow laminae cut across the fibres of vegetable or animal substances, either in their natural or artificial breadths (as boards), or by reducing natural or artificial thin tubes of them into helices. By threads I mean the same kinds of substances taken lengthwise, either from their being naturally in thin threads, or by reducing them to that state, in tearing tearing from them thin fasciculi of fibres; which operation is easy in some, as hemp, whalebone, and gut, but very difficult in others, as quill and some sorts of wood. 32. The first hygroscopes of the class of threads, which I observed comparatively to the class of slips, were of hemp, gut, whalebone, and some woods, and they exhibited a phenomenon which at first I could not understand: when they were exposed with the slips in damp air; as, for instance, in open air during the second period of dew above determined, or in a glass vessel inverted over water; the threads had only very small motions backwards and forwards round their point determined in water, while the slips had considerable motions within that point, without coming near it, if the temperature was sensibly above freezing point. Thence arises the objection against the general proposition above stated: for these two classes of hygroscopes contradicting one another on the changes of moisture, nothing could be asserted in that respect, till there were sufficient reasons to exclude one of those classes of informers, and to trust the other class. 33. Proceeding to multiply the species of those two classes of hygroscopes, I found always the same fundamental march in slips, all of them constantly moving in the same direction; but in multiplying the species of threads, I found such variety between them, that in their own class they created distrust; some of them, as of deal, aloes pitta, liber of lime tree, quill, and thin stems of gramen, in coming out of water, increased in length: they went farther that way, to a certain point, as the air was dryer: they retrograded then with accelerated steps, when dryness increased, thereby returning to the point where they had stood in water; and they continued to move in the same retrograde way, with great acceleration, by a still increasing Moreover, they did not follow one another in those motions contradictory to the evident march of moisture: each of them changed direction at different periods, thereby often contradicting also each other, while the slips constantly agreed together in the direction of their motion, and also with all the other symptoms of moisture. From these comparative phenomena I first concluded; that the motions of the same kind, which I had observed in the first-mentioned threads, were also anomalies, proceeding, only to a smaller degree, from a cause of the same nature as that of these last. After which, similar symptoms, which I had formerly observed in the water thermometer near the freezing point, made me first conclude, from a general analogy; that the perceivable modifications of the threads, were the compound effects of two contrary operations of moisture which followed different laws. 34. Another phenomenon led me soon after to a more determined theory in respect of those two opposite effects of moisture on threads. I have said above, that hemp and gut have only a very little retrogradation; their greatest difference from the slips consisting in their being stationary, while the slips have still great motions. But when these same threads are twisted, they acquire a very sensible elongation beyond their point of extreme moisture succeeded by retrogradation. From several trials I have made in twisting these threads more and more, I do not consider as impossible, if some difficulties, which I only could obviate in part, were completely prevented, that they might be brought to such a state, as to have their point of extreme dryness coincide with that of extreme moisture; by which means, in the progress of moisture from one extreme to the other, they would move first in one direction with decreasing steps, then in the opposite direction by increasing steps; the whole, however, Hygrometry. with great irregularities. Here then we see two opposite effects of moisture; one which lengthens the fibres; the other which, by swelling the twisted strings, shortens them; and we see those effects follow different laws, from which is produced a retrogradation that we may change ad libitum. 35. Now, the texture of animal and vegetable fibrous substances must be a sort of reticle, which exists in those which are naturally in thin threads, and in the most minute fasciculi that we can separate from a mass; and we see it in the last case, for in subdividing those fasciculi, there are always fibres breaking in the points where they were anastomosed with others; consequently, the primary fibres of those substances form between them meshes similar to those of a net; and those meshes, which are widened by the introduction of water, must produce in the threads the same effect as the twist in the above strings. 36. If then moisture, in acting on vegetable and animal threads, natural and artificial, produces on their length two opposite effects; one of which, small at first but increasing gradually, compensates at some period the other which is first visible, and surpasses it afterwards, sooner or later, according to the nature of the threads; it is evident, that they cannot be proper for the hygrometer; since, from the indication of some of them it might sometimes be concluded, that moisture changes in one sense, while it really changes in the contrary sense; and from some others, that moisture is extreme, long before it is really so. As for the slips; since moisture has only one effect on their length, that of widening more or less the meshes of the cross fibres, I concluded; that all their hygroscopic indications, in every part of their scale, were true in respect of increase and decrease of moisture; and that consequently, that class of hygroscopes might be depended upon on that important point. point. As for the exact ratio between the indications of those last hygroscopes, and the changes of moisture, that was to be the object of a particular inquiry, to which I now come. Of the scale of the hygrometer between the two fixed points. 37. The long attention I had formerly given to the comparative expansions by heat of various kinds of liquids and solids, made me expect the variety I afterwards found in the modifications of hygroscopic substances by moisture; therefore, as I expressed it in § 2. of my first Paper, my view was only at that time, to find some means of producing a steady comparable hygrometer; but afterwards I pointed out, in § 72, a means which had occurred to me, for attempting to find the ratio between the expansions of some determined hygroscopic substance, and the correspondent increases of moisture; which was, to compare the first with the correspondent changes of weight, of the same or of any other substance; an idea which I did not then much scrutinize, not yet thinking of its execution. 38. From what I have said above, I did not want any other motive of choice between the slips and the threads than their comparative marches; but though the slips agree always in the direction of their motions, there are differences in the progression of their comparative steps; and that difference led me to examine more attentively the above means of finding which of those marches agreed best with that of moisture. The result of that examination was distrust, at least in respect of an immediate decisive means. With the view of rendering the changes of weight more easily measurable, I had first thought of some substance possessed of a strong affinity with water; but on considering Hygrometry. fidering the hygroscopic phenomena of those substances, it appeared to me, that their changes of weight could not be proportional to the degrees of moisture in the medium; and that even the sense of the word moisture applied to them was very difficult to determine (§ 14.). As for the substance of the hygrometer itself, I did not find any reason to think, that its changes of weight could be more proportional to moisture than its degrees of expansion; since on these last depended in part the quantity of water that could be admitted into its pores at each degree of moisture in the air. 39. Being thus disappointed in my first scheme, I thought of a more direct method, which was to act on moisture itself, by first producing, in a glass vessel containing an hygroscope, as much dryness as I could conceive possible at that time, and then introducing into it successive equal quantities of water, for which I had found a sure means without opening the vessel. But then again some previous experiments destroyed my confidence in that method; having found,—1st, That the evaporated water had a tendency to deposit itself against the glass by the smallest difference between the inside and outside temperatures, even to the degree of becoming visible on some part of the vessel, long before I had any reason to expect extreme moisture.—2d, That by the common temperature of my room, the hygroscope in the vessel remained always at a considerable distance from its point of extreme moisture, though the bottom of the vessel was covered with water; and that it varied with the temperature; which could not have happened if moisture had been extreme.—3d, That when I endeavoured to increase moisture in the vessel by cooling it, I produced very often the contrary effect, at the same time that a quantity of the disseminated water gathered over the glass. Vol. LXXXI. 40. As in those trials the contradiction between the marches of the slips and the threads was very evident, I was the more disappointed to find, that the uncertainty of the real degrees of moisture increased at that very period: for instance, setting out from some dry point, and moisture increasing, the march of a thread of whalebone was evidently in a decreasing progression, comparatively to that of a slip of the same substance; and when at last there was a superfluous quantity of water in the vessel, and the temperature was made to change, the thread was almost stationary, while the slip had considerable motions, often contrary to the small motions of the other; while it was at that very period, that the real degrees of moisture in the vessel could not have been ascertained but by an already settled hygrometer. I had no doubt that those anomalies were to be attributed to the thread of whalebone, not only because of the excess of the same modification in other threads; but also considering the analogy between the comparative marches of the two kinds of hygroscopes, and those of the thermometers, of water and quicksilver: but as this was a very important object for natural philosophy, I would not decide it from those first appearances; and that consideration led me to a very great number of various sorts of experiments, which I made in view of multiplying at least the indirect facts with which my theory might be compared. However, as at last I found a more direct means of verification, I shall only mention that last class of experiments. Expe- Experiments on the comparative changes of weight and dimensions of some hygroscopic substances. 41. I have said above, that I could not find any solid reason to consider the changes in weight of a substance, as being more proportional, than its changes in dimensions, to the correspondent changes of moisture in the medium; which doubt had prevented me from undertaking that course of experiments. But it occurred to me at last, that if my theory on the comparative marches of the slips and the threads were true, it might be rendered certain by comparing those marches with the increase of weight of the same substance: for instance, taking a slip and a thread of deal, and having some deal hung to a scale, I was to find, that while the slip continues to lengthen, and the thread to shorten, the substances continue to receive water; and, in general, that the march of slips, in every part of their scale on which the experiment may be regular, was more proportional, than that of threads, to the correspondent changes in weight of every hygroscopic substance of the elastic kind. 42. This having struck me as a sure means of deciding the question, I set immediately to work; and since that time I have made a great number of experiments of that kind. They were not at first very accurate; but successively I have mended both the instruments and the apparatus; and, after having settled every part that I looked upon as essential, and made in consequence a new apparatus and new instruments, I have begun a regular course of experiments, of which I shall give here the first results. 43. The apparatus consists in two tin vessels; the first of which, and the most used, is $16\frac{1}{2}$ inches high, $15\frac{1}{2}$ wide, and 5 deep. The front of this vessel is a plate of glass, and the back a tin-plate slider, which, being taken off, leaves that side of the vessel quite open. The second vessel has the same dimensions as the first; but its back is soldered, and its front is of woven brass wire. This vessel may be applied to the back of the former, in such manner as to make of both one single vessel, which, when the slider of the fore-part is taken off, is only divided by a vertical partition of woven brass wire. The use of that second vessel is to produce extreme dryness in the other; for which purpose it is filled with large pieces of quicklime taken from the kiln. When that vessel is not used, it is kept in a tin box which it fills entirely; and when it is in, as well as while it is out for use, that box is kept shut with putty, by which means the same lime may serve many times in the following manner. 44. When I want to produce extreme dryness in the first vessel, I apply to it the second, fastened with hooks; I then pull out the slider of the first, and stop with putty the chinks between them. When that first operation is completed, I put again the slider to the fore-vessel, and take off the other. In this last operation, some moisture might be introduced through the chinks of the slider, before they are again stopped with putty; especially as the destruction of moisture in the vessel has made room for more air to come in; but I prevent it by making first the apparatus sensibly warmer than it was when I put on the lime-vessel; by which means, in the little time employed for the operation, the motion of the air is from the inside to the outside, which prevents all access of moisture in the vessel. 45. To that first operation succeeds that of a gradual introduction of moisture into the apparatus. It would have been useless to apply to that process the means I had imagined, for spreading successive equal quantities of water in a close vessel; for the evaporated water depositing itself in part, more or less, on every surface, and the surfaces, especially of glass, being very very much multiplied in that apparatus when filled with instruments, no real correspondence could be expected between the quantity of evaporated water and the motion of the instruments; therefore I again gave up the view of determining that ratio, and proceeded in the following manner. At the bottom of the apparatus, on one of its sides, there is an opening, \( \frac{1}{2} \) an inch high, and \( 2\frac{1}{2} \) inches broad, which usually is kept shut with a tin plate and putty. The taking off, only for a moment, that tin plate, is the first operation by which I introduce moisture into the apparatus; which being then reduced to the temperature at which I make all my observations, namely, 60° of Fahrenheit, permits the external air to enter the vessel, and come to an equilibrium with it at the same temperature. By keeping off the tin plate longer and longer, I admit new quantities of moisture into the vessel; and when that means is become ineffectual, I introduce through the same opening, a brass frame, which extends under all the instruments, on which is stretched a cloth that I wet by degrees more and more, as long as it may produce some effect at the above temperature. In order to have that temperature when I want it, I make those experiments in a season when I may have fire in a stove at a proper distance from the apparatus. The time when that equal temperature is necessary in many respects, is that of the observations, which I make twelve hours after each new introduction of moisture. 46. The instruments I place in that apparatus are of two kinds; the first of which are beams, made on the principle used by Mr. John Coventry in his paper hygrometer, which principle I have found of great use; for, with beams of that sort, as delicate as mine are, if the total change of weight in an experiment is not above 1 grain, \( \frac{1}{1000} \) part of it may be distinctly observed on a weight from 3 or 4 to 20 grains; but in my experiments, in which the total variation was from 5 to 6 grains, the observable part was only $\frac{1}{10}$ part of a grain. These two beams are placed on the same line through the middle of the depth of the vessel, and their indices move in that plane; their motions are in an opposite sense by the same changes of weight, because I wanted the two substances suspended to the beams, to hang near one another in the middle of the vessel. The other instruments are frames, in which an index is moved by the variations in length of a very thin slip or thread. These frames are placed before and behind the beams. 47. In the first experiment that I made with that apparatus, the substances suspended to the beams were deal and quill, reduced to very thin shavings, which were stretched edgewise in thin brass-wire frames. The weight of each kind of these shavings was 12 grains, at a certain degree of the thermometer and of my hygrometer. The other hygroscopes were slips and threads of the same substances as the shavings, and also of whalebone. These six last instruments have their point of extreme dryness taken in my lime-vessel, and their point of extreme moisture in water; the interval between these points is divided into 100 parts; and on the scales of the threads the degrees are prolonged beyond this last point. The hygroscopic scale of the shavings could not be fixed before the operation; therefore the scales of the beams served only to indicate the comparative motions of the index; but afterwards, taking for 0, the point where the index stood by extreme dryness, and for 100, the point of extreme moisture, which I shall explain, the interval between these two points became a modulum by which I have also reduced into 100 parts of the whole the changes observed in the weight of the shavings. I shall not give here the absolute quantities, either of the changes of weight, or of those in length of the other hygroscopes, having not had time to make the necessary calculations, of which however I have the data. 48. The necessary time for a complete diffusion of the newly introduced moisture in the vessel renders it impossible to proceed, in that introduction, by regular steps. The method I use is, to observe the motion of my usual hygrometer, which is a slip of whalebone, and to remove the cause of increase of moisture before it has moved 5 degrees. In that manner the steps of the increasing moisture have been in general less than 5 degrees of that instrument; but, by interpolation, I have reduced them to what they would have been if the same instrument had been moved successively 5 degrees. 49. I have said before, that when the maximum of evaporation is produced in a close vessel by a temperature sensibly above the freezing point, there is no regularity to be expected in any farther attempt to increase moisture; the disseminated water being then abundant, the smallest difference of temperature between different parts of the apparatus makes it deposit itself on some surface, and pass from one to the other (§ 39.); which circumstance is also mentioned by Professor Pictet of Geneva, in his late work, Essais de Physique. For that reason, when the evaporated water in the vessel was near its maximum, my last operation was, to put in again the wet cloth, while I kept the temperature at 60°, and to take it out when the indices of the beams were fixed. My usual hygrometer was then at 87°; and as it had still 13 degrees to move towards its point of extreme moisture, and all the others in proportion of their known marches, I have added to the observed increases of weight in the shavings, a quantity proportional to their former marches comparatively comparatively to that of their respective slips; and thus are completed their hygroscopic scales. Repeated trials have shewn me, that the weight of the shavings increases as long as their slips increase in length; but as there is no regularity in their comparative marches at that period of moisture in a vessel, nor any possibility of making these experiments in open air, because of the beams; the addition mentioned, which forms the three last terms of the columns of the shavings in the following table, is to be considered only as having determined the modulum of the observed terms; since it has not changed the ratio between them, nor consequently the correspondent marches of weight and length so far; which were the only object of the experiment. 50. Before I come to the general result of that experiment, I shall place here a comparative view of the two kinds of phenomena, which, by their analogy, led me first to the theory I have exposed: I mean the comparative marches of the slip and thread of whalebone on one side, and those of the thermoscopes of quicksilver and water on the other side. In this table the correspondent terms of the hygrosopes, from 0 of both to 85 of the slip, have been observed in the above experiment; the four following are the results of observations in time of increasing dew. The correspondent terms of the thermoscopes are deduced from the table of their comparative expansions which I have given in § 418. m. of my work, Rech. sur les Mod. de l’Atmosphère; from which table this only differs, 1st, by a change of the modulum, in the ratio of 80 to 100; 2d, by an inversion, which brings the terms of this table to express comparative condensations of the two liquids. ### Hygrometry #### Hygroscopes. | Whalebone Slip | Whalebone Thread | |----------------|------------------| | Extreme | 0 | | 5 | 12.1 | | 10 | 30.1 | | 15 | 41.1 | | 20 | 51.1 | | 25 | 59.1 | | 30 | 65.6 | | 35 | 71.1 | | 40 | 76.5 | | 45 | 81.8 | | 50 | 85.8 | | 55 | 88.8 | | 60 | 91.3 | | 65 | 93.3 | | 70 | 95.6 | | 75 | 97.6 | | 80 | 98.6 | | 85 | 99.6 | | 90 | 100.1 | | 95 | 100.5 | | Extreme | 100. moisture | #### Thermoscopes. | Quicksilver Boiling | Water Boiling | |---------------------|---------------| | Extreme | 0 | | 5 | 9.3 | | 10 | 18.3 | | 15 | 26.3 | | 20 | 35.0 | | 25 | 42.7 | | 30 | 49.2 | | 35 | 56.7 | | 40 | 63.1 | | 45 | 69.0 | | 50 | 74.5 | | 55 | 79.1 | | 60 | 83.8 | | 65 | 87.9 | | 70 | 91.8 | | 75 | 95.0 | | 80 | 97.5 | | 85 | 98.9 | | 90 | 99.9 | | 95 | 100.5 | | Extreme | 100. point | 51. The first part of that table shews, the great steps of the thread of whalebone, comparatively with those of the slip of the same substance, at the beginning of their correspondent increase in length; but the thread relents by degrees, and a nearly stationary state succeeds, in which, while this hygroscope moves first 1.9 degree from 98.6 to 100.5, then retrogrades 0.5 degree. degree to come to its point of extreme moisture; the slip, continuing to move in the former direction, goes over 20 degrees. The phenomena are the same in the part of the table relating to the thermoscopes; that of water proceeds also at first by great steps, comparatively to that of quicksilver; after which it relents, and while it moves only .3 degrees, and retrogrades 0.5 degree to come to the freezing point, the quicksilver one, continuing to move in the same direction, goes over 20 degrees. It was from that phenomenon of the water thermometer that I concluded formerly, that there was in its march a stationary state, during which the heat decreased nearly as indicated by the quicksilver one; and that conclusion was afterwards confirmed by direct experiments. From that ascertained fact, I was led to conclude, with respect to moisture, that there was also a stationary state in the march of hygroscopic threads, even in those which had the smallest retrogradation, as that of whalebone; and this theory will be confirmed by the results of the above-described experiment. 52. The following table contains those results, namely, the correspondent marches of all the mentioned hygrosopes; the shavings increasing in weight, and the slips and threads in length. The last comparative terms do not result from that particular experiment; for the shavings (as I have said above) they are concluded from the former comparative steps; for the other instruments, they have been obtained by observations in the open damp air. ### Hygrometry | Extr. dryn. | Whalebone | Quill | Deal | |-------------|-----------|-------|------| | | Slip. Thread | Shavings | Slip. Thread | Shavings | Slip. Thr. | | 0 | 0,0 | 0,0 | 0,0 | 0,0 | 0,0 | | 5 | 12,1 | 7,0 | 4,8 | 40,0 | 6,2 | 5,4 | 42,0 | | 10 | 30,1 | 13,0 | 9,7 | 72,0 | 9,4 | 11,2 | 69,4 | | 15 | 41,1 | 20,0 | 14,4 | 85,0 | 15,6 | 16,5 | 94,8 | | 20 | 51,1 | 26,0 | 19,2 | 95,0 | 22,6 | 21,9 | 107,0 | | 25 | 59,1 | 31,0 | 23,9 | 101,0 | 27,0 | 27,2 | 113,6 | | 30 | 65,6 | 36,0 | 28,5 | 105,0 | 33,2 | 32,7 | 118,6 | | 35 | 71,1 | 42,0 | 33,3 | 107,0 | 36,0 | 38,3 | 122,6 | | 40 | 76,5 | 43,8 | 38,3 | 102,0 | 41,2 | 43,7 | 120,6 | | 45 | 81,8 | 48,3 | 42,9 | 104,0 | 46,7 | 49,2 | 123,6 | | 50 | 85,8 | 52,3 | 47,4 | 107,0 | 49,7 | 54,6 | 126,6 | | 55 | 88,8 | 56,5 | 52,4 | 103,0 | 56,1 | 59,9 | 119,7 | | 60 | 91,3 | 60,5 | 56,9 | 105,0 | 59,9 | 64,9 | 122,7 | | 65 | 93,3 | 64,4 | 61,9 | 106,0 | 63,7 | 69,7 | 119,7 | | 70 | 95,6 | 69,4 | 67,2 | 108,0 | 67,1 | 74,5 | 117,6 | | 75 | 97,6 | 74,0 | 72,2 | 107,0 | 73,4 | 79,0 | 115,6 | | 80 | 98,6 | 78,0 | 77,8 | 106,0 | 78,1 | 83,5 | 112,6 | | 85 | 99,6 | 84,0 | 82,8 | 105,0 | 83,8 | 87,5 | 110,0 | | 90 | 100,1 | *88,0 | 88,2 | 103,6 | *88,8 | 92,0 | 107,0 | | 95 | 100,5 | *94,0 | 94,0 | 102,0 | *93,8 | 96,0 | 103,6 | Extr. moist. 100 *100. From the 18 first terms of that table, which are the immediate results of the experiment, we are now to examine my opinion, that the lengthening of the slips of whalebone, quill, and deal, beyond these terms is a sure sign that, till they have attained their point 100, moisture continues to increase in the medium where they are placed. 53. There could not be any doubt on that proposition, if it were not for some threads similar to that of whalebone; which threads, having sensibly attained their utmost length at the period when the experiment was stopped, seem to indicate, that moisture is then at its maximum. But if the lengthening of hygroscopic threads in general, is the compound effect of two opposite causes which follow different laws; it may be that, in some threads, those causes happen to compensate each other at that very period; by which means they are stationary, though moisture continues to increase. This was my opinion, and the above experiments were undertaken to verify it, first, by comparing the march of the slips with the increase in weight of their own substance; secondly, by comparing the marches of different kinds of threads with each other, and also with the increase in weight of their substance: and from that now we are to examine the above proposition. 54. In respect of the slips my theory is, that as moisture cannot act on their length but by widening the meshes of their transversal fibres, they cannot go on lengthening but by imbibing more and more moisture, from its increase in the air; and this we see to be the case, by comparing the marches of the three kinds of slips with the correspondent increases in weight of the quill and deal shavings, during the whole progress of the experiment. There are differences in those marches as I expected (§ 37.); but they are not such as to give the smallest reason to suspect, that afterwards, during the period of the three last terms of the table, in which we have no correspondent observations of increases of weight in the shavings of deal and quill, the same law does not take place as in the 17 antecedent terms. If the experiment was only made with one kind of slip, it might be objected, that though that slip lengthens regularly during the whole increase of moisture from its minimum to its supposed maximum, it is not impossible but that immediately after, by some peculiarity of its nature, it will lengthen, without any farther increase of moisture in the medium. medium. But that surmise cannot be admitted when the slips of such dissimilar substances as whalebone, quill, and deal, sensibly agree in their motions at that period, and when a number of other slips of the vegetable and animal kinds follow also the same general march. 55. In respect of the threads, which are the only cause of the above doubt, my theory, which removes it, is also confirmed by that experiment. That cause of doubt is exemplified in the table by the thread of whalebone, which has almost no motion, while its slip moves from 85 to 100. At that period of moisture, no regularity can be expected from experiments made in clay vessels; by which circumstance, not having the correspondent observations of weights, it cannot be demonstrated by immediate experiments, that the thread of whalebone is then stationary from its nature; but the threads of quill and deal, which are in that state during the regular course of the experiment, will guide us in that enquiry. The thread of quill is stationary during that great part of the observed increase of moisture, by which the thread of whalebone moves from 71 to 97.6: the thread of deal is also stationary, while the same thread moves from 71 to 91.3. They both afterwards retrograde; the quill from 107 to 100, and the deal from 122.6 also to 100; and it is during the latter part of that retrogradation, correspondent to a continued direct motion of the slips, that the thread of whalebone, and some more of its class, after a very decreasing march comparatively with all the slips, are at last stationary, and then a little retrograde. In that stationary state of the threads, while moisture proceeds in the same direction, they move backwards and forwards, more or less, according to the duration of that state, and to the quantity of the retrogradation. This may be seen by the table, in respect respect of the thread of deal and quill; and I have observed it constantly, in a smaller proportion, in all the threads which have their stationary state at the last period of moisture, with this particular circumstance in all those motions backwards and forwards, that they are never the same in two different experiments. That symptom already points out a complication of causes; but we shall soon see a more distinct proof of their existence. 56. My theory on the march of hygroscopic threads is founded on this general principle, that a retrograde effect, how small soever it may be, if it is not produced by a correspondent change in the cause itself, is preceded by a stationary state, during which, and the retrogradation, the intensity of the cause continues to increase: and this also is exemplified in the experiment. The stationary state of the thread of deal begins, when its shavings have only imbibed from the air a quantity of water = 36; it is still at the same point when the quantity of imbibed water has increased to 59.9; and in the part of its retrogradation, which is still contained in the regular course of the experiment, that quantity of imbibed water increases to 88.8. The same, with only some differences in the degree, is seen in the thread of quill; therefore, as we see also some, though a very small, retrogradation in the thread of whalebone, as well as in other threads of the same class, we have reason to conclude, that their apparent immobility before they come to that point, while the slips continue to move, is also a stationary state, during which they continue to receive water, by the increase of moisture in the air. 57. The experiments I have now analysed are only one set amongst others which, though made with less accuracy, have given the same general results. These, relating to various kinds kinds of substances, I intend to repeat, and to have the honour of communicating their results to the Royal Society; and I shall conclude this Paper, with an immediate demonstration, that the hygroscopic motions of the slips are simple, while those of the threads are the combined effects of two opposite causes: which will be a further confirmation of the whole of the above theory. On the recoil of hygroscopic threads. 58. When formerly I concluded from the phenomena of the water thermoscope, that its condensations were the combined effects of two opposite causes, which followed different laws, it was not for having distinguished those two effects; but only because of a small retrogradation near the freezing point, preceded by a stationary state, comparatively with the march of quicksilver; but in the case of hygroscopic threads and slips, in which we have the same phenomenon, the two opposite effects are distinguishable in the threads, by one being operated more rapidly than the other. If, for instance, I transport from a drier to a damper place (or inversely) the two kinds of quill hygrosopes, the slip proceeds in an even course to a certain point, where it remains fixed; but the thread moves in an interrupted manner also to a certain point, whence it recoils. If that experiment is made within the limits of the stationary state of the thread, it may recoil as much as it has gone the other way, and be fixed at the same point in both places. The case of the slip of quill is common to every slip, and that of its thread to all others which have a quick motion. Here then we have separately the two effects of moisture on the threads; that on the fibres themselves is the soonest produced, and at first predominates: the slowest, by which afterwards the first produced is more or less compensated, is that operated on the width of the meshes; and it is because the last of those effects is the only one that can affect the length of the slips, that, in every change of moisture, they move evenly, without any recoil. 59. To that demonstration of the existence of two opposite effects of moisture in the threads, I shall now only add an example of a similar phenomenon, in which the causes also are visible. The compound frames, mentioned in § 24, are formed of two glass rods 4 feet long, fixed together at the bottom and the top. A thin slip of brass of determined length, fixed to one of the glass rods towards the top, comes down from thence, passes over a pulley at the bottom, and turns up for half an inch. To this end of the brass slip is fixed the lower end of the hygroscopic substance, the upper end of which is connected with the index of the instrument. It is by that means that, whatever be the changes of heat, provided they are slow, the lower end of the hygroscopic substance remains sensibly at the same distance from the axis. But if I take that instrument out of water at a low temperature, and plunge it immediately into warmer water, the index instantly moves as if the hygroscopic substance had lengthened; which is the effect of the brass slip dilating sooner than the glass rods; then the index recoils, and this is the effect of a slower dilatation of the glass. Conclusion. 60. I have concentrated in these pages an account of twenty years assiduous labour in hygrometry, mostly occasioned by the anomalies of the hygroscopic threads; and the principal results have been, some determinations of the four principles that directed directed me from the beginning, which now are as follows. 1st, Fire, as cause of heat, is a sure, and the only sure, means of obtaining extreme dryness: this is produced by white heat in every hygroscopic substance that can bear it; and it may be thus transmitted to the hygrometer. 2d, Water, in its liquid state, is a sure, and the only sure, means of determining the point of extreme moisture on that instrument. 3d, It is not to be expected, à priori, of any hygroscopic substance, that its changes be proportional to those of moisture; but it may be affirmed, that no fibrous or vascular substance, taken lengthwise, is proper for the hygrometer. 4th, A means of throwing light on the march of a chosen hygrometer, may be, to compare it with the correspondent changes in weight of many hygroscopic substances. 61. From those determinations in hygrometry some great points are already attained in hygrology, meteorology, and chemistry, of which I shall only indicate the most important. 1st, In the phenomenon of dew, the grass often begins to be wet when the air, a little above it, is still in a middle state of moisture; and extreme moisture is only certain in that air, when every solid exposed to it is wet (§ 28.). 2d, The maximum of evaporation in a close space, is far from identical with the maximum of moisture; this depending considerably, though with the constant existence of the other, on the temperature common to the space and to the water that evaporates (§ 30.). 3d, The case of extreme moisture existing in the open transparent air, in the day, even in time of rain, is extremely rare: I have observed it only once, the temperature being 39°. 4th, The air is dryer and dryer as we ascend in the atmosphere; so that in the upper attainable regions, it is constantly very dry, except in the clouds. This is a fact certified by M. de Saussure's observations and mine. 5th, If the whole atmosphere passed from extreme dryness to extreme moisture, the quantity of water thus evaporated would not raise the barometer as much as half an inch. 6th, Lastly, in chemical operations on airs, the greatest quantity of evaporated water that may be supposed in them at the common temperature of the atmosphere, even if they were at extreme moisture, is not so much as $\frac{1}{100}$ part of their mass. These two last very important propositions, have been demonstrated by M. de Saussure.