Hurricanes are generally observed at the period of the greatest heat of the day; but in the interior of continents, especially when they are mountainous, there are also nocturnal storms. According to observations made by the intrepid Caille, this is often seen in the mountains to the south of the western part of Sahara; and, according to those made by Eschwege, in the mountains of Brazil. The latter assures us that no idea can be formed of the violence of a nocturnal storm in the virgin forests of this country.

I am not in possession of sufficient facts for determining the number of storms observed in a year in the different regions of the globe. However, from the observations made by travellers, it appears that they occur for the most part when the regularity of the trade-winds is disturbed, or when the monsoons change.

At sea, in the region of the trade-winds, storms appear to be as uncommon as rain; for I do not remember to have found, in a single traveller, the account of a storm of any violence in this zone. At Madeira, storms appear to be very frequent in winter; and this is also the season when the limit of the N.E. trade-wind passes into the vicinity of this island. During the conflict that occurs between the S.W. wind, which is descending, and the N.E. trade-wind, electric discharges are very common.

STORMS IN HIGH LATITUDES.-North of the Alps there are scarcely any storms except in the hot season. As we advance from the shores of the Atlantic into the interior of the continent, a modification is found, in their number

site winds, which produce the whirlwind. From his researches, M. Dova draws the following indications for the use of nautical men :

1st. In the temperate zone of the northern hemisphere, if the wind blows first from the S.E., and turns to the south, and then to the west, the ship should be steered to the S.E. On the contrary, if it blows first from the N.E., and passes to the north and the N. W., the ship must be steered to the N.W. In the former case, we are in the S.E. region of the hurricane; and in the latter, on the contrary, in the N.W.

2d. In the northern part of the equatorial zone, if the wind first blows from the N.E., and passes through the east to S.E., we must steer to N.E. If it blows first from N.W., and turns by west to S.W., we must steer to the S.W. In the first case, the ship is in the N.E., and in the second, in the S.W. of the hurricane.

3d. In the southern part of the equatorial zone, when the wind blows from S.E, and then turns to south and S. W., we must steer to N.W.; if it blows first from east, and passes through north to N.W., we must steer S.E. In the first case the sailor is N. W. of the tempest, and in the second, S.E.

4th. In the temperate zone of the southern hemisphere, if the wind is first established in the N.E., and then passes through north to N.W., the head must be turned to the N.E. If, on the contrary, it is established in the 8.E., to pass to south, and then to S.W., we must steer to S.W. In the first case, the ship is N.E., in the second, S. W. of the tempest. (POGGENDORFF'S Annals, t. lii. p. 1. 1841).-M.

and distribution, analogous to that of rain. Mountainous countries are an exception to the general law, inasmuch as storms are more frequent on the west side of chains than on the plain. On the west coast of Europe and in Germany, we find about twenty storms in the year; at St. Petersburg and Moscow, seventeen at a mean; at Kasan, nine; at Nertschinsk, two; and at Irkoutsk, about eight. Let 100 represent the number of storms occurring in a year, we have the following distribution in the four seasons :—

On the western coast of Europe only a tenth of the total number of storms occurs in winter; in summer, there is the half. In Switzerland and in Germany, a storm in winter is a very rare phenomenon; two-thirds of the total number occur in summer. In the interior of the old continent there are no storms in winter, three-fourths take place in summer, and the small number of these that are observed in spring and autumn occur only during the hottest months of these two seasons; so that we may correctly say that there are no storms during one half of the year.

STORMS IN SCANDINAVIA.-As there is not any country where the transition from a sea to a continental climate is so sudden as in Scandinavia, so there is no place where there is so great a difference for storms; we shall be convinced of this, by comparing their number in different

towns :

The number of storms is very small, for there are not above ten in the year; but their distribution is very different on the coast and in the interior of the country. At Bergen, the winter rains predominate over those of summer; and, at the same time, storms are more frequent there, as also at Sandmoer, which is in the same district; but, at Spydberg, in the interior of the country, we find the same relations as in Russia: it is the same at Stockholm and at Skara. Stroem, Arentz, and Hertzberg, have perfectly described these winter storms of the province of Bergen: they arrive indifferently at the end of an intense cold, as after a time of continuous thaw, or after rain. They are always brought by west or S.W. winds. Stroem even asserts that a storm may be predicted, if the wind suddenly changes from the S.W. to west or N.W. These storms are violent on the islands along the coast; they are more feeble in the fiords, and almost unknown in the interior of the countries, where storms sometimes burst forth in summer.

Winter storms are formed more commonly on sandy shores. In Iceland, lightning frequently occurs in winter in the neighbourhood of the volcanoes; at the Feroes, the Hebrides, the Shetland, and the Orkneys, it is only during violent gales of wind that thunder is heard to roll. On the west coast of America, and on the east coast of the Adriatic, storms are much more common in winter.

If the number of storms is so small in Scandinavia, compared with Germany and France, it is seen to diminish still more as we advance toward the north, where the quantity of vapours filling the atmosphere is very small. Thus, during a sojourn of six years in Greenland, latitude 70°, Gisecke heard thunder but once; and all travellers agree on this point.

STORMS IN THE NORTH OF THE MEDITER

RANEAN.-The ancients had long ago remarked their frequency in certain seasons. Lucretius thought that violent winds squeezed out the fire contained in the clouds: and he hence deduces the causes of their distribution in the different seasons:

"Autumnoque, magis stellis fulgentibus, alta
Concutitur coli domus undique, totaque tellus,
Et cum tempora se veris florentia pandunt;
Frigore enim desunt ignes; ventique calore
Deficiunt, neque sunt tam denso corpore nubes :
Inter utrumque igitur cum cœli tempora constant,
Tum variæ causæ concurrunt fulminis omnes;
Nam fretus ipse anni permiscet frigus et æstum,
Quorum utrumque opus est fabricanda ad fulmina nobis
Ut discordia sit rerum, magnoque tumultu

Ignibus et ventis furibundus fluctuet aer.

Prima caloris enim pars et postrema rigoris,

Tempus id est vernum; quare pugnare necesse est
Dissimiles inter se res, turbareque mistas,
Et calor extremus primo cum frigore mistus
Volvitur, autumni quod fertur nomine tempus;
Hic quoque confligunt hyemes æstatibus acres:
Propterea sunt hæc bella anni nominitanda.
Nec mirum est in eo si tempore plurima fiunt
Fulmina, tempestasque cietur turbida cœlo."

Book vi. verses 356 et seq.

The testimony of other writers agrees with that of Lucretius; in Greece, storms are frequent in autumn, and in spring, according to M. Peytier's observations. There are probably great differences in their season distribution; unfortunately, we possess but few documents on this point. I associate here those which we have for Rome, Palermo, Padua, and Janina, thanks to M. de Pouqueville.

R

RELATIVE NUMBER OF STORMS, IN DIFFERENT SEASONS, IN

In north Italy, as well as in Greece, there are about forty storms in the year, that is to say, a number the double of that for Germany. At Palermo, the number is not above a third of that in our climates; the air is indeed purer, and the warm air coming from Africa opposes the precipitation of aqueous vapours. There are only sixtyfour days of rain throughout the year at Rome; at Padua, on the contrary, there are 120. At Palermo, storms are very common in autumn, whilst at Rome there is scarcely any difference between autumn and summer; their distribution through the year at Padua completely recalls to mind that of Germany. The observations of the ancients chiefly relate to this town: and, if they have laid stress upon autumn storms, it is because they are more violent, and of longer duration.

FORMATION OF STORMS. · All storms may be divided into two classes; the one class are due to the action of an ascending current, the other are a result of the conflict of two opposite winds; the former occur during the hot season, the latter during winter. Let us begin by examining those of the first class.

In our climates, and in summer, three conditions are necessary for the formation of a storm: a great calm in the