that the shooting beams which seem to dart from the arch towards the zenith, properly neither come from the luminous ring, nor are connected with the luminous columns, but that each of them, as well as the ring itself, consists of a great number of short luminous cylinders, lying close together, and almost parallel,* of which each conceals a part of the one lying before it. I have already hinted, that there must be some connexion between the aurora borealis and the magnetism of the earth. Besides the reasons which may be derived from the facts already mentioned, I mean, that the centre of the luminous rings coincide with the four magnetic poles; and also, that the luminous columns shoot parallel to the medium direction of the magnetic powers at any one place; that they, beyond all doubt, follow the laws of repulsion of the magnetic powers, and, consequently, themselves are magnetic,-there are many other reasons as strong, if not stronger, of which I shall briefly mention only the most important. 1. When the aurora borealis is vivid, the horizontal magnet becomes restless, varies in a few minutes from three, four, to five degrees from its ordinary place, and sometimes gets into a quivering motion, which shews, that, at that time, the magnetic powers of the earth are in a state of great agitation. 2. A short time before the aurora borealis appears, the intensity of the magnetism of the earth is apt to rise to an uncommon height; but so soon as the aurora borealis begins, in proportion as its force increases, the intensity of the magnetism of the earth decreases, recovering its former strength by degrees, often not till the end of twenty-four hours. These changes are sometimes so sudden, that I once observed a considerable difference in the space of from two to three minutes. [The Professor here details an experiment made with a fine magnetic needle, suspended by the thread of a silk worm. He counted its vibrations to the amount of 360; and dividing this number into equal portions, and comparing by a chronometer the time occupied by each portion, he found a difference in the number of vibrations performed in the same time, far exceeding any thing which could arise from the inaccuracy of the observation, proving a difference during the time of the experiment in the intensity of the magnetic force.] From They are not completely parallel, as their direction is parallel to the inclination of the needle, which is different in different places. this, he says, it seems to follow, that the polar lights are the effect of an uncommonly high magnetic intensity, which intensity lets itself off, as it were, by the polar lights, and thus sinks under its common strength. 3. During the time of a powerful display of the aurora borealis, Mallet and others have found no uncommon strength of electricity in the air, which seem to disprove the hypothesis of Franklin. 4. The Reverend Mr Steenbuck, who was the editor, and in a great measure, the author of the Description of the Kingdom of Norway, published under the name of Jessen, and who was himself born in the province of Trondheim, relates, concerning the arch of the aurora borealis, that, according to the accounts of old people, this arch was wont to appear lower on the horizon in Nordland, and nearer to the true north; that, since these times, it has risen higher in the sky, and removed from the meridian towards the west. This agrees entirely with the change of place of the North American magnetic pole, by which it lies some nearer to us, and lies in a plane which forms a larger angle with the meridian. 5. The Swedish naturalist Wilcke, has remarked, that, during a vivid borealis, the corona sometimes changes its place, moving backward and forward several degrees. But since the place of the corona is determined by the angle formed by the luminous columns with the surface of the earth, it is evident that the angle must likewise change. And, in such cases, he observed, that the inclination of the needle altered in a similar manner, so that moving up and down, it always pointed to the centre of the corona. A change in the direction of the earth's magnetism, produces also a change in the direction of the luminous columns. The perpendicular height of the arch of the aurora borealis above the surface of the earth, is to be computed from its height above the horizon, observed at two different places, which lie at a considerable distance from one another, nearly in the same meridian. From a number of those arches, the cotemporaneous height of which was observed at Rome, Paris, Copenhagen, Stockholm, and other places, Mr Mairan has found, that the ring of the polar lights above the surface of the earth, almost always exceeds 100 geographical miles. The luminous columns, of which the shooting beams consist, have, in all probability, the same height.* Consider now the diffi When Scoresby talks of the shooting beams appearing to descend to the top-masts, this can be considered as nothing else but an optical deception. In the Heavens, the eye has no measure of distance. The sun and moon seem equally near us, though the first is more than 400 times farther distant than the last. culty of answering the following questions on any other hypothesis than what I have suggested: Why do the polar lights spring not from the poles of the earth, but from four points at a great distance from the poles? Why do the direction and intensity of the earth's magnetism change so considerably, while the streams of this light are issuing forth? Why is the direction of the luminous columns parallel at every place with the direction of the magnetism of the earth, or the inclination of the needle? What can that material substance be, which has such powerful influence at so great a height over the surface of the earth? How can this, at such a height, produce a certain sound? From the manifold experience we have had of this in the north, it must be considered as a fact admitting of no question. Why do the polar lights not appear equally powerful every season, but have periods of from 60 to 100 years, during which they display themselves with great splendour, after which they have pauses of nearly equal length? When we consider, I say, the apparent impossibility of answering these questions, in a satisfactory manner, on any other hypothesis, and how readily an answer to the most of them suggests itself, when we adopt a magnetic origin to the polar lights, it seems a matter of course to admit this hypothesis. Nor is it a new one. All the naturalists who have had more than a superficial knowledge of magnetism, and who can be considered as competent judges of this subject, have admitted it. I have only to mention as the first and chief of these, Dr Halley; and next to him, the Swedish naturalists Celsius, Hiorter, Wargentin, and Wilcke. If any person can explain all the external phenomena of the polar lights, by assuming rather an elastic material fluid, on portions of which the magnetism of the earth can operate according to the known laws of attraction and repulsion, there still remains this important question, which our present knowledge of the sciences cannot answer, What is this material substance? Is it electricity in a neutral state, as we have learned to know it in the insulated galvanic chain in the experiment of Oersted? or, Is it an elementary substance, in some other state as yet unknown to us, and on which the magnetic powers only can act ?* (To be continued.) *A copious statement of all the known external phenomena of the polar lights, together with a historical account of their different periods and pauses, will be found in the second part of my "Magnetismus der Erde," if circumstances should ever put it in my power to bring this work to light. ART. XVII.-On the Influence of the Hygrometric State of the THE Connexion between the minimum temperature of the night and the contemporaneous state of the air, in regard to humidity, was first pointed out, I believe, in the Article HYGROMETRY, which I wrote several years ago for the Edinburgh Encyclopædia. In that dissertation I remarked, that, as the temperature of every place for the whole year ranges between two extreme points, corresponding to the alternations of summer and winter; so it exhibits, during the diurnal rotation of the earth upon its axis, a like difference resulting from the vicissitudes of day and night. In the case of the daily change of temperature, there is some interval between the maximum and minimum condition, which may be regarded as the temperature belonging to the season of the year; and though that point is not, at all times, equally distant from the extremes between which it oscillates, it seldom departs far from their mean. If this mean temperature were to rise and sink regularly, as the year advanced and declined, without being subject to daily fluctuation, the quantity of moisture existing in the atmosphere, at any given time, might be determined by the thermometer alone, with tolerable precision; as it would generally be less than the quantity corresponding to the mean temperature, and seldom greater than that which belongs to the minimum temperature, the latter setting limits to the accumulation of watery vapour in the atmosphere, while the former no less effectually secures it against a state of long continued dryness. The truth of these assertions will be readily perceived, by a comparison of the minimum temperature with the point of deposition, or the temperature at which the moisture existing in the atmosphere would begin to deposite itself. I shall select for this purpose the observations which my friend the Rev. Dr Gordon made in the year 1815, during his residence at Kinfauns; not only because they were made without any view to the support of a theory, but still more because his extreme accuracy as an observer, joined to his profound knowledge of this department of physical science, gives them an authority which can be claimed for the materials of few meteorological The absolute humidity of the atmosphere, exhibited in the coluinn entitled Grains of Moisture in 100 cubic inches of air, was computed by a formula which I have given in the article Hygrometry already alluded to, and the correctness of which I have since verified by a careful comparison of the results it affords, with the quantity of moisture in the air, as determined by direct experiment. The column shewing the mean point of deposition for each month, exhibits the temperature at which the air, at the time of observation, if cooled down to that point, would begin to deposite the vapour in admixture with it. The adjoining column contains the mean minimum temperature as given by a self-registering thermometer; and the last column shews the difference between the lowest temperature and the point of deposition. The coincidence is sufficiently great to establish the important fact, that the quantity of moisture dissolved in the atmosphere is so intimately connected with its minimum temperature, that the former may be readily inferred from the latter. It was not difficult, indeed, to arrive at the conclusion, that, as the quantity of moisture which can exist in the atmosphere, under the vaporous form, is entirely regulated by temperature, and totally independent of chemical solution, the hygrometric condition of the air must, at all seasons, nearly |