It is obvious that this principle affords an explanation of the formation of clouds in the atmosphere, and that currents of air, or winds, of different temperatures, when they meet, must produce such mixtures as have been described, and give rise consequently to the condensation of aqueous vapour. When the supply of the humid air, entering into the mixture, is continued, the quantity of cloud formed will continually increase, and the small globules of condensed moisture, uniting into drops, must descend in rain.

But though we are thus in possession of a principle by which rain may be certainly produced, yet whether it be the only one by which rain is produced may require some farther investigation. Dr Hutton accordingly, in order to determine this point, has entered into a very ample detail concerning the rain under different climates, and in different regions of the earth. The result is, that the quantity of rain is, as nearly as can be estimated, every where proportional to the humidity contained in the air, and the causes which promote the mixture of different portions of air, in the upper regions of the atmosphere. Between the tropics, for instance, the dry season is that in which the uniform current of the trade-wind meets with no obstruction in its circuit round the globe; and the rainy season happens when the sun approaches to the zenith, and when the steadiness of the

trade-wind either yields to irregular variations, or to the stated changes of the monsoons.

Thus, too, (to mention another extreme case,) in certain countries distant from the sea, having little inequality of surface, and exposed to great heat, no rain whatever falls, and the sands of the desert are condemned to perpetual sterility. Even there, however, where a mountainous tract occurs, the mixture of different portions of air produces a deposition of humidity; perennial springs are found; and the fertile vales of Fezzan or Palmyra are exempted from the desolation of the surrounding wilderness.

This ingenious theory attracted immediate attention, and was valued for affording a distinct notion of the manner in which cold acts in causing a precipitation of humidity. It met, however, from M. de Luc with a very vigorous and determined opposition; Dr Hutton defended it with some warmth, and the controversy was carried on with more sharpness, on both sides, than a theory in meteorology might have been expected to call forth. For this Dr Hutton had least apology, if greatest indulgence, on the score of temper, is due to the combatant who has the worst argument. The merits of the question cannot be considered here: It is sufficient to remark, that they came ultimately to rest on a single point, Whether the refrigeration of air is carried on by the mixture of the cold

and the hot air, or by the passage of the heat itself, without such mixture, from one portion of air to another. If the former holds, Dr Hutton's theory is established; if the latter be true, M. de Luc's objections may at least merit examination.

Now, it is certain, that if not the only, yet almost the only, communication of heat through fluids, is produced by the mixture of one part of the fluid with another. The statical principle by which heat is thus propagated, was first, I believe, accurately explained by Dr Black, and since his time has been farther illustrated by the experiments of Count Rumford. These last have led their ingenious author to conclude that heat has no tendency to pass through fluids, otherwise than by the mixture of the parts of different temperature. The accuracy of this conclusion, in its full extent, may reasonably be questioned; but this much of it is undoubtedly true, that when the particles of a body are at liberty to move freely among themselves, the direct communication of heat, compared with the statical, is evanescent, and may be regarded as a mere infinitesimal. M. de Luc's objections are

therefore of no weight.

The Theory of Rain was republished by Dr Hutton in his Physical Dissertations several years afterwards, together with his answers to M. de Luc, and several other meteorological tracts, which contain many excellent examples of generalization, in a

branch of natural history where it is more easy to accumulate facts, and more difficult to ascertain principles, than in any other. *

It may be proper to mention here some useful observations in meteorology which Dr Hutton made, but of which he has given no account in any of his publications.

He was, I believe, the first who thought of ascertaining the medium temperature of any climate by the temperature of the springs. With this view he made a great number of observations in different parts of Great Britain, and found, by a singular enough coincidence between two arbitrary measures, quite independent of one another, that the temperature of springs, along the east coast of this island, varies nearly at the rate of a degree of Fahrenheit's thermometer for a degree of latitude. This rate of change, though it cannot be general over the whole earth, is probably not far from the truth for all the northern part of the temperate zone.

For estimating the effect which height above the level of the sea has in diminishing the temperature, he also made a series of observations at a very early period. By these observations he found that the difference between the state of the thermometer in two places of a given difference of level, and not very distant, in a horizontal direction, is a constant quantity, or one which remains at all seasons nearly the same, and is about 1° for 230 feet of perpendicular height. I must, however, observe, that, on verifying these observations, I have found the rate of the decrease of temperature a little slower than this, and very nearly a degree for 250 feet. This seems to hold for a considerable height above the earth's surface, and will be found to come pretty near the truth, to the height of five or six thousand feet. It is not, however, probable, that the diminution of the temperature is exactly proportional to the increase of elevation; and it

After the period of the two publications just mentioned, Dr Hutton made several excursions

would seem, that at heights greater than the preceding, the deviation becomes sensible; the differences of heat varying in a less ratio than the differences of elevation.

In explaining this diminution of temperature as we ascend in the atmosphere, Dr Hutton was much more fortunate than any other of the philosophers who have considered the same subject. It is well known that the condensation of air converts part of the latent into sensible heat, and that the rarefaction of air converts part of the sensible into latent heat. This is evident from the experiment of the air-gun, and from many others. If, therefore, we suppose a given quantity of air to be suddenly transported from the surface to any height above it, the air will expand on account of the diminution of pressure, and a part of its heat becoming latent, it will become colder than before. Thus also, when a quantity of heat ascends, by any means whatever, from one stratum of air to a superior stratum, a part of it becomes latent, so that an equilibrium of heat can never be established among the strata; but those which are less must always remain colder than those that are more compressed. This was Dr Hutton's explanation, and it contains no hypothetical principle whatsoever.

To one who considers meteorology with attention, the want of an accurate hygrometer can never fail to be a subject of regret. The way of supplying this deficiency which Dr Hutton practised was by moistening the ball of a thermometer, and observing the degree of cold produced by the evaporation of the moisture. The degree of cold, cæteris paribus, will be proportional to the dryness of the air, and affords, of course, a measure of that dryness. The same contrivance, but without any communication whatsoever, occurred afterwards to Mr Leslie, and being pursued