The sun has a dull appearance even when it is high; its light presents a reddish tint. When the sun approaches the horizon, it is of a blood-red colour; it may be looked at without the eyes being dazzled, and its brilliancy is so reduced, that it ceases to be visible, even before it has descended below the horizon. It even happens that the lower edge of the sun is scarcely visible, whilst the upper presents a welldefined red edge.

Sometimes the dry fog has a remarkable intensity; several examples are found in our records. That of 1783 caused a general sensation throughout Europe; it presented the following phenomena: its thickness was such, that in some places objects at the distance of five kilometres could not be distinguished; they sometimes appeared blue or else surrounded with vapour. The sun appeared red, and without brilliancy, and could be gazed at in the middle of the day; at its rising and setting, it disappeared in the haze. At Copenhagen, it was first seen May 29th; it came after a succession of fine days. In other places, it was preceded by a gale. In England, it came after continuous rains; at La Rochelle, it was seen on June 6th and 7th; at Dijon, on the 14th; the atmosphere then became clear at La Rochelle till the 18th. It was noticed almost every where in Germany, France, and Italy, from the 16th to the 18th; on the 19th, it was observed at Franecker and in the Pays-Bas; on the 22d, at Spydberg, in Norway; on the 23d, at St. Gothard and at Buda; the 24th, at Stockholm; the 25th, at Moscow; toward the end of June, in Syria; the 1st of July, in the Altai. Some observers maintain, that they found traces of acid in it, but these observations have no more value than the experiments made on atmospheric electricity. It is accused of having caused the disease of smut in the corn, and diseases in vegetables generally; but we know that these diseases are manifested without being determined by dry fog.

Subsequently, and especially in 1834, a thick dry fog was also observed. I saw it on May 23d on the mountain of Victor, among the Harz. The same day, a violent north wind drove it to Basle, where it remained three or four days; on the 25th, it was seen with a violent N.E. wind at Orleans; and at the same period, chiefly on the 21st, 22d, and 24th, at Munster. Several times, in the course of this summer, I noticed dry fogs of various thicknesses. July 28th and 29th, they were remarked at Halle, Freiberg, and Altenberg, in Saxony, and they lasted several days; they finally disappeared, August 2d, during a rain-storm, and

shewed themselves only for a few isolated days in the month of August.

Dry fog is very common in north and west Germany, as well as in Holland. Finki attributes it to the combustion of peat. Indeed, in order to prepare peat-lands for cultivation, they are dug up and the clods are turned over in autumn, that they may dry in winter; if the month of May is dry, they are set fire to, care being taken that they shall give off plenty of smoke and very little flame. The drier the air and the ground are, the better does the operation succeed; rain, on the contrary, interferes with it; in summer, large surfaces ignite spontaneously. The quantity of the products of combustion may therefore amount to 9,000,000 of kilogrammes.

In these countries the dry fog is coincident with the burning of the peat; when the air is dry, the smoke remains suspended in the atmosphere, and may be carried away by the winds. The wind always blows from the direction of the peat-bogs, when dry fog is manifested; and fogs have frequently been seen coming distinctly from the peat-bogs.

The very thick dry fog of 1834 came partly from the combustion of peat, and partly from the fires for which this year was noted. Whilst it was observed in the end of May among the Harz, and in the neighbourhood of Orleans and Basle, there were fires amongst the peat-bogs. Thus, in particular, the peat-bog of Dachau, in Bavaria, was burned to the depth of 2,5; and the fire was even propagated beneath ditches filled with water; in the neighbourhood of Munster and in Hanover, several peat-bogs were consumed. Subsequently, in July, there were terrible conflagrations of forests and peat-bogs, near Berlin, in Prussia, in Silesia, in Sweden, and in Russia; the drought favoured the propagation of these fires, and the transport of the smoke.

To the same cause, probably, may be attributed the particular aspect of the air in autumn. In the fine days, when the atmosphere is very dry, the air is less transparent; distant objects are not seen distinctly, they appear surrounded by a slight fog. As in many countries, heaps of bad grass and potato-culms are burned at this period, I think that the smoke they produce is diffused throughout the atmosphere.

Can the dry fog of 1783, which was diffused over a great portion of Europe, be attributed to the came cause? At the period when it was manifested many hypotheses were devised in order to explain its origin: Lalande attributed it to the quantity of electricity developed during a very hot

summer, that succeeded a moist winter; Cotte regarded it as formed of metallic emanations united with electricity, in consequence of the great heat and numerous earthquakes; other philosphers have connected this fog with electricity, without our being able to understand how the latter is able thus to disturb the atmosphere. However, Veltmann has shewn that these phenomena are concomitants with the great peat-burnings that took place in Westphalia.

In the same year there was a violent earthquake in Calabria, and a volcanic eruption in Iceland; several philosophers attribute the existence of this fog to these. It is true that volcanic phenomena rarely manifest themselves with so much violence; and we may add, in favour of this opinion, that, in the years when a very intense dry fog filled the atmosphere, the volcanoes were in activity: for example, the years 526, 1721, 1822, and 1834. However, the fog has several times preceded the irruptions. Are we, therefore, to regard volcanic irruptions as an immediate cause of dry fog? Although the column that rises above a volcano has a very great analogy to a column of smoke, yet more accurate researches have shewn that it is for the most part composed of the vapour of water and volcanic ashes, with which are mixed transparent gases in various quantities: no one has observed true smoke. But, when the lava runs over the side of the mountain, it carbonises every thing that it meets, and an immense cloud of smoke rises in the air. If we think of the immense quantity of vegetables that were consumed in Iceland, as well as seventeen villages, we can comprehend that the lava, when running over a soil covered with vegetables, might have been able to produce this smoke, which the north winds immediately spread over a great portion of Europe. Add to this, that the combustions of turf and the conflagrations of forests were as frequent this year as during those that are distinguished by excessive droughts. It is to the latter cause that we must attribute this odour, which is especially perceived in Holland.

The inhabitants of western and southern Europe attribute to the dry fog a great influence over the state of the atmosphere: they connect with its existence the predominance of the north winds which then prevail, and say that it drives away rain and storms, and is a cause of cold. It is true that things are in this condition at the season of peat-burning; but I doubt whether we should attribute them to this cause this opinion, which is so generally prevalent in these countries, arises from considering a general phenomenon under a point of view of narrow locality. Thus it is that,

on the coasts of Scandinavia, the cold that suddenly arises with north winds is thought to be due to the vicinity of the sea; and it is accused of many misdeeds of which it is innocent, because the same thing is experienced in the centre of Germany. These winds blow throughout Germany, accompanied with cold nights and dryness of the air; and young plants suffer greatly. With regard to the dispersion of storms by dry fog, or their transformation into fog, I should be tempted to believe that thick masses of dry fog accumulated in the horizon have been taken for stormclouds, the true nature of which the spectator recognised when they approached him.

SHOOTING STARS AND METEORIC STONES (Sternschnuppen und meteorsteine).-Shooting stars are observed during serene nights. In a part of the sky, a luminous point appears in the form of a star of greater or less brilliancy, it moves through space, and is then suddenly extinguished; but its brilliancy diminishes at the moment when it is about to disappear. Sometimes the star leaves in its passage a luminous train; however, this is not constant: sometimes, also, the star gives forth sparks. The ancients regarded these meteors as stars actually falling; when they are of considerable size, they are called igneous meteors, bolides, balls of fire (feuerkugeln, feuermeteore). A luminous point is first seen, like a shooting star, or a small clear cloud, which is not long before it bursts into flames, or even one or several parallel striæ, which soon form a large flaming globe. This globe moves with a velocity equal to that of the stars, sometimes in bounds, which are a proof of an original impulse, or are an effect of terrestrial attraction; it increases and becomes a burning globe, sending forth flames, smoke, and sparks. This luminous globe generally draws after it a luminous train, which is drawn out to a point, and terminates in smoke. This tail appears to be formed of the substance drawn out from the ball itself; or else it is accompanied by little satellites, which themselves become small luminous globes; finally the ball bursts with a great explosion. These meteors frequently break again, and then the constituent parts that have not been volatilised fall in the form of masses of iron or stone. These meteoric stones or aërolites are of a different composition from that of stones found on the surface of the earth, and occupy a much smaller space than the large ball.

HEIGHT OF IGNEOUS METEORS.— As shooting stars and fire-balls only occur occasionally, it is difficult to

go through the operations necessary for determining their height with certain precision; however, it has sometimes happened that several observers have been able to measure simultaneously the angle of the height of a ball seen from different places, and hence to deduce its absolute height in the atmosphere. Benzenberg and Brandes made the first observations on this point: being at two places distant from each other, they marked on a celestial planisphere every shooting star, in order to have its position and apparent course; knowing the moment of the observation, they were able to deduce the angle of the height, and to calculate the height of the meteor.

The elevation of shooting stars above the earth is very different, in that it oscillates between 16 and 230 kilometres; the greater portion move in a region comprised between 45 and 155 kilometres. The mean height, deduced from all these heights of Brandes, is 116 kilometres.

The greater portion of shooting stars move in a descending direction: however, some are directed horizontally, or even ascending; some have even been observed to describe a semi-circle, first rising and then descending: Chladni quotes several examples. It follows that these bodies are subject to the action of gravity; but that, in addition to this, they receive a very energetic impulse in order to take a direction, which must be always contrary to that of gravity: their velocity is from thirty to sixty-six kilometres per second.

The indications of the height of fire-balls are very discordant; I have given a great number in my Treatise on Meteorology. It follows that their mean height is about the same as that of shooting stars.

FREQUENCY OF SHOOTING STARS.-If shooting stars rapidly succeed each other, they are frequently observed in the same region of the sky. According to Benzenberg, there are often eight in the hour. During the night of the 6th and 7th December, 1798, Brandes counted 480 shooting stars. In these latter times, much attention has been paid to their periodicity; they are observed in a special manner in the nights from the 10th to the 15th of November, and on that of the 10th and 11th of August.

M. de Humboldt was the first who saw a large number on the night of the 11th and 12th November, 1799; they were observed at the same time in Guyana, Labrador, Greenland, and the environs of Weimar. Hemmer had before this been struck with their number at Mannheim, on