EVAPORATION.

hour 2.72 grains, and to have had its temperature reduced to 43°. By repeating the process, continually diminishing the pressure one half, the successive losses from evaporation in intervals of half an hour each, were as follow:—

When the exhaustion was pushed to the utmost, which was .07 inch, the evaporation in the half hour was 87.22 grains. During this last experiment the water was frozen in about 8 minutes, while the thermometer under the ice was at 37°. There must, however, be some inadvertency respecting the loss from evaporation in the second experiment; because in the table it is 2.87, and in the account before it only 2.72, which last, we presume, is the more correct of the two.

Before inferring from these experiments the rate of evaporation under different pressures, Mr Daniell applied to the results a correction for the variation of temperature. Taking the evaporation as proportional to the elasticity of the vapour, he estimated the latter from the mean of the temperatures at the beginning and end of each experiment, and calculated the amount for a fixed temperature. This might have been supposed to give a near approximation, were it not evident from the last experiment, that, owing to the bulb of the thermometer not being close at the evaporating surface, it indicated often too high a temperature; but the following table presents the results computed on that principle for the temperature of 45°; and here again we suspect the second number 2.97, being computed from 2.87, is too great.

Notwithstanding the slight irregularity of the series, Mr Daniell thinks we can run no risk in concluding that the amount of evaporation is, ceteris paribus, in exact inverse proportion to the elasticity of the incumbent air. But perhaps it would be more correct to say, that, ceteris paribus, the rate of evaporation is inversely as the density of the air; for that is Dr Dalton's view of the matter, and is equally conformable to these experiments. It may however be observed, that the rate of evaporation in the first case, which is in air of the ordinary density, is much smaller than that given by Dr Dalton; but the process going on in so small a volume of still and confined air, seems a sufficient reason for this.

It was hinted above, that evaporation tends to lower the temperature of an evaporating surface, by abstracting heat from it for the formation of vapour. For, from the well known fact, that in the formation of aqueous vapour as much heat is absorbed as would raise the temperature of 1000 times its weight of water one degree, it follows, that were it not for the heat derived from surrounding bodies, the vaporization of the 1000th part of a mass of water would lower the temperature of the whole one degree; so that, by the time the 100th part had evaporated, the tem-

perature of the whole would have fallen ten degrees. But surrounding bodies, particularly the air, by participating in the cooling effects, supply heat in such a manner to the water, that a limit is soon set to the fall of temperature. Although this cooling influence has been long known in a general way, it is remarkable how few experiments have been made, or, so far as we know, published, to show the extent of cold produced in different circumstances; and still more remarkable, that those few results of different experimenters should differ so widely, as we shall presently see they do. Dr Anderson, in the article Hygrometry, in the Edinburgh Encyclopædia, gives the following as the results of his experiments on the cold produced by the evaporation of water in air of different densities. For this purpose he placed Leslie's hygrometer, along with a cup of sulphuric acid, under a receiver of the air pump.

The first column is the pressure in inches, the second the Fahrenheit temperature of the dry bulb, the third that of the moist, the fourth their difference or the depression, and the fifth the same depression in degrees of Leslie's hygrometer. The third and fourth columns, though not in the original, are obviously obtained from the others, and inserted in degrees of Fahrenheit, for the sake of comparison with those which follow.

In an article on hygrometers and evaporation, in the Edinburgh Philosophical Journal for October 1826, Mr Meikle gives the two following series of similar experiments, which were not made with Leslie's hygrometer, but with what was reckoned preferable for the purpose, two common thermometer tubes fitted upon one broad and doubly graduated scale; the one bulb being dry and the other covered with wet linen. The columns respectively denote the same things as the first four in Dr Anderson's.

All sealed thermometers stand too low when the pressure of the atmosphere is removed from their bulbs. The two just mentioned were found to be each 1.5 too low in an exhausted receiver. Hence most of the numbers in the second and third columns still require to be more or less corrected for this, according to the degree of exhaustion; but since both thermometers were equally affected, the depressions have nothing to do with it. However, the included air seems to have been really a little cooled from the influence of the cold wet bulb.

We have always been puzzled to account for the extreme smallness of Dr Anderson's depressions in the fourth column. Perhaps he had used very weak sulphuric acid, or too small a surface of it; or he might not be aware that the full effect in a single case is seldom attained in less than half an hour. The pressure and temperature in his first case are nearly the same as in Mr Meikle's; and yet the depression 4.86 is not half of 11.6. From many trials we know that the weather is by no means very droughty, if a wet thermometer do not fall more than 4.86, when only exposed in the open air at 48.5, without any sulphur-

Evaporation. ric acid at all. To increase the cold in a wet thermometer, Dr Lardner recommends exposing it to the sun. But whoever takes the trouble of properly trying this, will find that the sun has quite the contrary effect; and therefore the doctor is just in the same delusion, only on a greater scale, when he directs his readers to wrap wine bottles in wet cloth, and expose them to the sun as a source of greater cold! The sun, like the fire, aids evaporation, by supplying heat, which will more or less counteract the cooling effect, and may even raise the temperature; but common sense revolts at the very idea of the heating rays of the sun being a source of cold.

In 1827, Professor Daniell published, in the second edition of his Essays, p. 499, a series of experiments on the cooling effects of evaporation, and which, like Mr Meikle's, were made by means of two common thermometers, the one dry and the other moist; but he seems only to have read off to the nearest half degree; and his depressions, though not so deficient as those of Dr Anderson, still lean considerably to the same side. The following are his results, arranged as the preceding:

From repeated experiments we have found that at the temperature of 50°, and under a pressure of about thirty inches, a wet thermometer, when inclosed in a receiver along with a sufficiently large surface of strong sulphuric acid, should fall 12°·1 or 12°·2, in place of 9°, as Mr Daniell has it. The rest of his depressions are still more deficient; but when we attend to the disposition of his apparatus, our surprise rather is that the depressions should be so great as they are. What we particularly object to is his not only inclosing a vessel with water under the receiver, but interposing it directly between the wet bulb and the sulphuric acid. This he did for the purpose of occasionally dipping that bulb in the water; a precaution which was not only unnecessary, but which, by rendering the air moist and obstructing the drying of the bulb, must obviously have rendered the whole of his depressions much smaller than they should have been in perfectly dry air. When Mr Meikle made his experiments, and they were published before those of Mr Daniell, he satisfied himself that if the thermometer was properly covered and moistened at first, no renewal of moisture was necessary. This he ascertained by repeating, in a retrograde order, the different cases when he had just completed a series of these experiments; for on finishing with the case in which the air was most rarefied, he readmitted as much air as just brought back the gauge to the next case, which being repeated, he readmitted more air for the next, and so on, till he had got back over the whole series. In every case, the wet thermometer was found to be as much depressed in the backward process as in the forward, a clear proof that there was no lack of moisture. Neither Dr Anderson nor Professor Daniell say whether they made any allowance for the effects of pressure on the bulbs of their thermometers.

Cold not yet observed in vaporizing turpentine. Oil of turpentine is well known to evaporate very quickly, but we have as yet failed to detect any cold that it produces, though we do not therefore suppose it forms any exception to the general law. Some liquids, as alcohol, ether, sulphuret of carbon, &c. both evaporate more rapidly than water does, and produce a greater cold; but the expenditure of heat is greater in vaporizing a given weight of water than of any other liquid yet examined; a fact from

which it had often been concluded that alcohol could be evaporated with so much less fuel than water, that its vapour would be an economical substitute for steam as a first mover; nay, this had become a standard doctrine in almost every scientific compilation. It however involved a fatal oversight, as was first pointed out by Mr Meikle in the Philosophical Magazine for July 1826, p. 41, and afterwards by Mr Ainger in Brande's Journal of Science for March 1830; for it is not with equal weights, but with equal volumes of these vapours, that the comparison falls to be made; and when this is done, the greater rarity of steam puts it on at least an equal footing with alcoholic vapour, so far as regards the economy of heat. In other respects, steam is decidedly preferable; for, independently of the enormous expense of alcohol, the density of its vapour, being about 2·5 times that of steam, would prevent it from moving with the same facility through the pipes, valves, &c.

Perhaps it may not be altogether foreign to our subject to mention, that Mr Hutton froze alcohol by condensing air on it in a strong vessel, which being very much cooled by freezing mixtures, was next suddenly opened, when the dilatation of the air, and the consequently rapid evaporation and re-expansion of the alcohol, absorbed so much heat as induced a cold sufficient to freeze it.

The bad effects of damp clothes, beds, &c. are usually ascribed to the cold attending the evaporation of the damp; but since considerable injury often follows where scarcely any cold is felt, we suspect the chief cause lies rather in the moisture suppressing the natural evaporation from the skin. The heavy smell which accompanies most dogs, and the fetid grease upon their hair, show that they perspire in some other shape besides what some suppose the only one from their tongues.

Various instruments have been employed to measure the rate at which moisture escapes into the atmosphere. We have already alluded to the method of exposing water in a vessel, and noting the loss. The vessel is generally circular, with a roof a little above it to keep off the rain, &c.; and, for greater accuracy, the contents may either be weighed from time to time, or measured in a graduated tube. When the vessel is brim full, the wind is apt to blow some of the water over; and when not full, the water is in some measure sheltered from the drying effects of the wind. The only use of such an instrument or gauge is to give us a very vague idea of the drying power of the air. It affords no measure whatever of the evaporation from the ground; for when the soil is exceedingly parched, and consequently has next to no moisture to give off, the gauge generally shows the greatest evaporation of all. Yet in stating the particulars of the climate of a place, it is usual to mention the amount of evaporation derived most probably from no better source. But it cannot even be a measure of the evaporation from a lake, because in that case the indications of the gauge depend very much on the direction of the wind; for it is obvious that if a previously dry wind traverse an extensive lake before reaching the gauge, it may by that time have become so humid as to show little or no drying power at all upon the gauge. When, on the other hand, the wind passes first over the gauge, its indications are to be suspected of erring in excess, because the farther a wind continues to traverse the surface of a lake, the more will its drying power be impaired, and therefore the average rate of evaporation will be less than at the side next the gauge. For this reason the evaporation from a smaller surface of water is, ceteris paribus, greater in proportion to the area than from a larger. Whether any general rule could readily apply to this, we are unable to say. Perhaps something like an approximation to the relation between the area and evaporation in such circum-