of phenomena. The wind will 'back' or shift against the sun, from South-east through East to North-east and North; and if he is near the centre the rain will fall steadily and heavily, and will not cease suddenly as soon as the barometer begins to rise.

As to an anti-cyclone, it will be perceived, from what has already been said, that no great changes of weather take place during its prevalence, except the occasional formation and clearing away of fogs.

The weather we experience in Western Europe is distinctly related to these areas of depression and anti-cyclones, to the rate at which they respectively travel over the earth's surface, and to the distance which intervenes between their respective centres. As in a system of either kind we may meet with winds from any point of the compass, which will have different qualities as to temperature, humidity, &c., according as they belong to one or the other, we see the great importance of the consideration, first pointed out by W. Koppen,' and subsequently by Capt. Toynbee,2 that the climatic character of a wind depends on its origin, i.e. on its belonging to a depression or to an anti-cyclone. Weather windroses giving figures calculated without any regard to this fact are, therefore, unsatisfactory.

Again, the principle which Dove laid down under the name of the Law of Gyration (Drehungsgesetz), to the effect that the wind generally shifted with the sun; that a South-east wind shifted to West through South, and that any shift in the opposite direction was a sign of unsettled weather, simply means that stations in the

1

Repertorium für Meteorologie, vol. iv., 1875.

? The Meteorology of the North Atlantic during August, 1873. London, 1873, p. 97.

west of Europe lie on the southern side of the paths of depressions, where the wind veers, or shifts with the

sun.

In Germany especially it is comparatively rare for a depression to pass on the southern side of a station, and this is one of the conditions which produce a 'backing,' or a shift against the sun, of the wind. If we examine the changes of wind in Iceland, we find backing is more common than veering, for there the depressions frequently pass to the south of the island. In fact, the evidence which Dove cited in support of his law is conclusive as to its truth in Europe, but not in other parts of the world.

Anti-cyclones are generally more or less stationary; but depressions move over the earth's surface, usually from west to east in these latitudes, their paths, as they advance, though chiefly ruled by the distribution of pressure, being liable to modification by the irregularities of the surface over which they pass; and their effects, as to the amount of cloud and rain to which they give rise, being influenced by the same causes. A South-west wind, for instance, may blow over a flat country with a clear sky, but as soon as the air reaches a hill side and is forced to ascend, the moisture it contains is condensed, clouds are formed, and rain is frequently the result.

I cannot here treat at any length of the subject of forecasting of weather. This, for a region situated like the United Kingdom, is a matter of great uncertainty, from the rapidity with which the changes succeed each other, and the extent to which local conditions affect the phenomena actually experienced.

The possibility of forecasting at all depends on the promptitude with which the earliest signs of a change

of weather which show themselves at our western stations are telegraphed to the Central Office.

Situated

as the British Isles are, with an ocean to the westward, we, in London, can never get many hours' notice of a change. Our neighbours in France and Germany are better off, as they can usually receive intelligence of changes from the westernmost stations in the British Isles. Norway and Spain are nearly as unfavourably circumstanced as ourselves. In the United States the conditions are more favourable than in any part of Europe. There the Central Office is situated on the eastern side of a vast continent, and consequently reports can be collected from an extensive area, and the changes, as they come on, can be watched at head-quarters to an extent which is quite impossible at this side of the Atlantic.

What I have said relates more particularly to the temperate zone and to that part of it in which the British Isles are situated, but it is more or less true for all parts of the world. The region of which I have spoken is more frequently visited by depressions than other districts, and the changeability of its weather is proverbial. In countries where pressure is more steady the weather is proportionately more stable, and weeks may pass over without a change.

In the Trade-wind zones but slight changes of weather are experienced as long as the Trade blows; the air is comparatively dry, but when that current reaches a coast, especially if it be lined with mountains, heavy and constant rains are produced.

In Hindostan we have the alternation of the Monsoons; the North-east Monsoon, a dry wind, bringing with it persistent clear weather; and the South

west Monsoon, a wind which blows from the equator and comes over the sea, bringing deluges of rain to the Western Ghauts and the whole Coromandel coast, as well as to the flanks of the Himalayas, especially in the hill country of Assam.

In the Dutch East Indies we have similar conditions, excepting that, as the hemisphere is changed, the dry Monsoon is South-east, and the wet North-west.

FitzRoy, in his 'Weather Book," devotes four chapters to a description of the general features of weather over the globe, and since many parts of the world were practically known to him as a voyager of great and varied experience, and as a colonial governor, his summary will well repay perusal.

Before leaving the subject of weather, allusion must be made to the views which are held by many meteorologists, as to the existence of a periodicity in the changes of weather, related more or less closely to that which appears to be traceable in the sun's condition, as indicated by changes in the character of his surface. A brief summary of these views will be found in Appendix V.

Longmans & Co. 1863.

CHAPTER XIX.

STORMS.

IN this, the concluding chapter of the book, I shall briefly discuss the principal features of the atmospherical disturbances which are known as Storms.

The forms in which these manifest themselves are very various. In some regions, and at certain seasons, the only breaks in the dull monotony of a tropical calm are sudden squalls of great violence but of short duration. Other regions resemble Kerguelen Island, where there is hardly an interruption to the persistence of the gales which sweep the Southern Ocean. In some parts, again, the wind, when it blows from a certain point, is liable to increase to the force of a storm, while in others the only type of storm which ever occurs is a revolving eddy, in which the changes of direction of the wind are the more rapid and extensive the nearer the station lies to the centre of the eddy.

Lastly we have the districts with which we are most familiar, situated in the higher latitudes of both hemispheres, where the storms are irregular in the frequency of their occurrence and in their character, occasionally blowing for a day or more without change of direction, while at other times the shifts of wind are almost as