zone which is in the southern hemisphere, and the SW. Monsoon in that part which is in the northern.

In the winter months the NE. Trade-wind is felt to the north of the equator, and the NW. Monsoon to the south of it.

In consequence of this alternation, the NE. Trade is termed NE. Monsoon, and the SE. Trade SE. Monsoon.

In spring and autumn, in the so-called 'change months,' there are calms; on the coast, sea and land winds in the daily period. The change from one Monsoon to the other is usually accompanied by a storm, the breaking-up of the Monsoon.'

The SW. Monsoon extends much farther to the N. (lat. 30°) in the northern hemisphere than the NW. Monsoon towards the S. in the southern; however, the latter extends also far to the S. on the African coast.

The rainy season is observed, as it is in the region of the Trade-winds, when the sun is highest. It is, therefore, during the SW. Monsoon on the northern, and during the NW. Monsoon on the southern, hemisphere.

There is, however, a difference in the behaviour of the barometer. In the region of the Trade-winds it remains nearly constant throughout the year, in that of the Monsoons it varies regularly. During the SW. Monsoon the barometer stands several tenths of an inch lower than in winter, especially in the northern portion of the district; and similarly in the southern hemisphere, it stands lower during the NW. than during the SE. Monsoon. At the equator this annual variation of the barometer disappears nearly entirely, in consequence of the compensation of the two opposite variations.

The rotations of the storms of the China and Indian Seas are the same as those in the same latitudes in the region of the Trade-winds; however, on the coast of

China, their motion is rather from E. to W. than from SE. to NW. One important distinction is to be observed, that, in the district of the Monsoons, the storms are felt with great violence in the southern hemisphere as well as in the northern.

The rotation of the vane during the Typhoons, although the direction of the rotation of the air itself is quite fixed (contrary to the hands of a watch), is yet, in consequence of the uncertainty of the direction in which the centre may possibly move, less fixed than during the West Indian storms. They take place during the SW. Monsoon, and occur up to November, being most frequent in September.

If the Typhoon move from NE. to SW., on the NW. side of its path, the rotation is N., NE., E., or with the sun; on the SE. side WNW., SW., SSE., or against the sun. On the south coast of China the rotation is generally N., NE., E., SE., as the Typhoons which are passing from E. to W. usually pass to the south of the coast.

According to Thom, the storms are never felt in the Indian Ocean excepting at the time when the NW. Monsoon is prevalent between the equator and the tenth or twelfth degree of south latitude, and are most common just after the winter solstice, when the sun is turning back from the Tropic of Capricorn.

The cyclones are most common in the district between the SE. Trade-wind and the NW. Monsoon, which is called the region of the variables.'

The rotatory motion takes place in the direction from E. through S. towards W. and N. The intensity of the cyclone increases regularly towards its centre. At the centre itself there is a dead calm, and the greatest violence of the storm is experienced at the edge of this calm circle. The diameter of this circle is greatest when the storm is just commencing. If the rotatory motion increases in

violence, the diameter of this circle is decreased to about ten or twelve English miles.

The advance of the cyclone, up to lat. 20° S., is at the rate of 200 to 220 miles in the twenty-four hours. From that point it becomes less rapid up to the outer edge of the SE. Trade.

The direction of the advance is from lat. 10° S. near the Indian Archipelago, to lat. 28° or 30° on the east coast of Africa; first towards WSW., then towards SW. by S., and lastly towards SSW.

Throughout the whole of the cyclone torrents of rain fall, which are more violent in front of it than behind it. The clouds are dark, massive, and lead-coloured, as the centre is approaching. Electrical explosions are most frequent on that side of the cyclone which is nearest to the equator.

The sea is disturbed irregularly to the distance of 300 or 400 miles during every such storm.

The barometer falls rapidly as the centre of the cyclone approaches, but the lowest level appears to occur a little before it passes.

3. NORTH TEMPERATE ZONE.

In my Meteorological Investigations,' 1837, I have stated expressly that the principal characteristic of the climate of the temperate zone is the alternation of two currents of air, of which the one flows from the pole to the equator, the other from the equator to the pole; and in my Non-Periodical Variations of Temperature on the Surface of the Earth,' six parts, 1840-59, as well as in the 'Representation of the Phenomena of Temperature by Means of Five-Day Means,' I have proved that these currents move simultaneously in proximity to each other.

6

Hence, in these localities we can no longer speak of a constant direction of the wind, as in the zone of the Trade-winds, nor of one which changes periodically, as in that of the Monsoons, but only of a mean direction. This mean direction is nearly SW.* in the north, and NW. in the south, temperate zone; inasmuch as the equatorial currents are more prevalent than the polar. In Europe this westerly direction is more southerly in winter than in summer; in America the reverse is the case; and the gradual transition from one of these conditions to the other takes place on the Atlantic Ocean. Violent storms are felt here less frequently in summer than in the winter months, and in the Mediterranean Sea at the transition of one season into the other; whence they are called here equinoctial storms.' These storms are either, 1, gales, i. e. ordinary winds whose intensity has been greatly increased, and which cause the vane to rotate with the sun, but through comparatively small arcs; 2, cyclones from the torrid zone, which have changed their path on crossing the outer limit of that zone, and have taken a course from SW. to NE. in the northern, and one from NW. to SE. in the southern, temperate zone; 3, currents, which, by their mutual interference, have checked and then repelled each other; or 4, storms, produced by the sudden intrusion of the cold polar current into the warm equatorial current, a case of which many remarkable instances have been noted. Hence, the barometer during the yearly period neither remains steady nor varies regularly, but is subject to oscillations, which are greater in winter than in summer. The mutual alternation of the currents is indicated by a rotation of the vane with the sun, i. e. S., W., N., E., S. on the northern, and N., W., S., E., N. on the southern, hemisphere. Hence we *See Note, p. 82.

derive the following general rules for the variations of the meteorological instruments :

Since the southern current is warm, moist, and rarefied, the northern, on the contrary, cold, dry and dense, we derive the following rules for their alternation; and we must remember that the cold polar current appears first in the lower strata of the atmosphere, while the warmer equatorial current will always have existed for some time in the upper strata before it is felt below. The changes of weather on the west side of the compass are, therefore, simultaneous with the changes of the barometer, while on the east side the indications of the barometer always precede the fall of rain which takes place. If the wind shifts from S. to N. through W., the barometer rises and the thermometer falls. This transition is characterised in winter by heavy falls of snow, in spring by sleet showers, and in summer by thunderstorms, after which the air becomes much cooler. If the wind veers from N. to NE., we have clear weather, the air is dry, the barometer high, and in winter intense cold follows, with great clearness of the atmosphere. As soon as the barometer begins to fall the wind gets round to the E.; the sky, previously deep blue, covers itself with thin whitish clouds, and the snow which falls comes from the S. wind, which has already set in above. If the barometer falls rapidly, the snow turns to rain, and a thaw sets in, when the wind veers farther through SE. and S. towards SW.*

The transition from a clear sky to an overcast one

*The first notice of this transition is to be found in Drebbel, De Naturá Elementarum (concerning the nature of the elements), 1621. 'If we see a thick cloud rising in summer not far from the south-west, we expect, and also find, that a SW. wind will soon blow, then a W., NW., and lastly a NNE. You see also why the E. or SE. wind brings such heavy and continued rain with it in Holland and the adjacent countries. I could very easily explain, on natural grounds, the reasons of all these phenomena.'