the most important differences between these consist in the degree of shelter, from occasional cold winds, which can be secured at each.

Again, the climate varies greatly with the elevation of the station. Not only, on the one hand, does the mean temperature fall with height above sea-level, but, on the other, a position on a slight eminence or on the side of a hill secures the residents immunity from the extreme severity of frost, inasmuch as in calm weather the cold air gradually sinks down to the lowest level of the valley. This is a phenomenon akin to the 'upbank thaw' already mentioned, p. 213. It is a well-known fact in pleasure-grounds that evergreens in low-lying bottoms suffer far more in winter than those growing on the slopes above. Moreover, low-lying stations are also notoriously far more liable to the visitations of fog than the uplands above.

Some localities are specially favoured, even though they do not lie on a south coast, as for instance the coasts of the Moray Firth, Nairnshire, and the Carse of Sutherland. These districts owe their good fortune mainly to the fact of their lying on the lee-side of an extensive mountain district. The prevailing winds, being Westerly, deposit their moisture on the Atlantic slopes of the hills, and the air, thus drained of its moisture by ascent, is further dried and warmed by its descent to the sea-level on crossing the island. The atmosphere of these districts is therefore essentially drier than that of other places in the same latitude, and the dryness, inducing clearness of the sky, allows more frequent access to the sun's rays than is possible on the cloudy west coast. Accordingly crops will ripen in the valley of the Shin, in Sutherlandshire, in 58° N.

which never come to maturity in Argyllshire, two degrees farther south.

The climate of hill stations in tropical countries is also especially enjoyable and salubrious, as compared with that of the plains. This is almost entirely due to the reduction of temperature with height. At the elevation of, say, 6,000 feet, the mean temperature will be reduced, in round numbers, 20°, so that where there is a temperature of 80° on the plains there will be one of 60° above.

Another particular in which high-level stations differ from those below them is the advantage they enjoy in the receipt of a greater amount of direct heat from the sun, owing to the fact already mentioned that the densest and dampest strata of the atmosphere lie the lowest.

It is mainly to this dryness of the atmosphere, and consequent intensity of solar heat, that the favourite Swiss station of Davos, at the level of 5,000 feet, owes its recently-acquired popularity as a winter residence. In some measure, however, it derives the comparative mildness of its climate from its situation in a valley sheltered from all except the warm South-west winds.

The increasing dryness of the atmosphere with elevation may, however, have its serious disadvantages; for when we come to high table lands, like those of Central Asia, we find insufficient moisture in the air to check the radiation to the earth by day or from the earth by night, so that here the range of temperature in the twenty-four hours is often more than the strongest constitution can bear.

It would, however, far exceed the limits at my disposal were I to discuss local climate, and the causes

which affect it, at the length which the importance of the subject demands.

For such a discussion the reader must be referred to such works as the 'Physical Geography' of Herschel or of Mrs. Somerville, while as regards the question of climate in its relation to health and disease, he will find full information in numerous works of deservedly high reputation on medical climatology.

CHAPTER XVIII.

WEATHER.

WEATHER may be described as the combined effect of all the different meteorological conditions which have been treated of in the foregoing pages.

In popular phraseology the term Weather has more especial reference to rain and temperature than to any other phenomena, but it will be explained how all the meteorological conditions are linked together, and how we cannot consider the rain without reference to the wind which brings it, or the wind without reference to the distribution of barometrical pressure to which it is due. The relations of temperature to the other conditions of weather have already come before us frequently.

Prior to the introduction of the modern system of weather observations, taken simultaneously over an extensive area, and telegraphed to some central office where they are inserted on maps, so as to give a picture of the weather existing at the time of observation at each station belonging to the system, observers were compelled to discuss the probabilities of weather for their own locality by the help of the records of observations taken in that locality alone.

This necessity led to the calculation of what may be called weather windroses'; exhibiting the average

values of pressure, temperature, humidity, &c., &c., for the wind from every point of the compass. This work has been carried out in detail for a large number of stations in different parts of the world. The calculation is laborious but simple, and the form of the windroses may be varied in many ways. An obvious modification, from which Professor Dove deduced his practical rules for weather, was the determination of the change in the readings of each instrument in passing from one point of the compass to the next.

The following is a condensed specimen of Dove's first two rules::

The

winds;

barometer falls

thermometer rises with E., S.E., and S.

with a S.W. wind it ceases to {file} and be

rise

rises

gins to fall ; it falls with W., N.W., and N.

winds; with a N.E. wind it ceases to

rise

fall, and be

Dove propounded these and other similar rules as applicable to the entire northern hemisphere, and gave an analogous set, mutatis mutandis, for the southern hemisphere, where, of course, a Southerly wind, blowing from the pole towards the equator, corresponds to a Northerly wind in our latitudes.

A great misconception, if one can apply the word to a man of such eminence as Dove, underlying these propositions, consisted in the assumption that what held good for Western Europe was true for the entire northern hemisphere.