unfortunately, no scientific observations have been made on the reputed poisonous or fatal character of these blasts, or of the dangerous quality of heat which they develop.

There is one type of warm weather in Europe for which no explanation can be given at present. We have seen that an anticyclone usually develops cold-radiation weather, but sometimes we find an anticyclone with warm air and a peculiar soft cloudy sky. This anticyclone covers Continental Europe, and is always associated with the eastward passage of distant cyclones on the northern side. No reason can be assigned for this heat; all we can do is to note the fact for future research.

SOURCES OF COLD.

The principal source of all cold is radiation into space. The space which surrounds the earth has a theoretical temperature of at least 226° below 0° Fahr., and it is the influence of this chilly envelope which we feel.

The greater part of the influence is, however, indirect. We do not feel the cold of space as if we were standing near an iceberg, for all our greatest colds are produced by radiation. Bodies on the earth's surface radiate into this cold space till they lose a large amount of their original temperature; and air, which is a bad radiator itself, gets cold by contact with the chilly soil.

For instance, on a calm winter night different bodiessay, a sheet of iron lying on the ground and a patch of grass-begin to radiate into space at different rates, according to their own intrinsic properties. Iron radiates

very quickly, but is also such a good conductor that it brings up an abundant supply of heat from the ground to replace the loss by radiation, so that the plate does not become very cold. The grass is a less good radiator, but at the same time a very bad conductor; so, though it parts with heat slower than the iron, it cannot replace what it has lost by conduction, and therefore, on the balance, becomes much colder. This is the cold which we really feel, and which sends down the thermometer.

A very striking result of all this is, that under these circumstances, the air gets warmer as we ascend up to a certain height, and this proves conclusively that we do not feel directly the chill of space. Of course, the greatest cold will be produced when the greatest number of causes are combined which favour radiation. These are a still air, a clear sky, and an absence of water-vapour in any stratum of the atmosphere. This last condition is very interesting. Professor Tyndall's researches seem to show that water-vapour is a great absorbent of the quality of heat which is radiated from the ground, so that when much vapour is present the ground cannot lose its heat so rapidly as when the air is dry.

All these conditions of great cold are fulfilled in the most perfect manner in Siberia. There we have the centre of a large dry continental area, which in wintertime is persistently covered by an anticyclone; while the latitude is so high that the sun has little power. Here, then, we find calm, dryness, and a feeble sun; and here the greatest known colds are reported, if we except some in the north of Smith's Sound, many degrees further north. A good illustration of this will be found in the

two charts which we give in our chapter on types of the north-east monsoon (Figs. 80 and 81). In them we see that the south of Siberia, which is covered by an anticyclone, has stations in which the mercury marks more than 30° below zero, Fahr.

If we take the less extreme cases which occur in Great Britain, we find that all frosts in that country are "home-brewed;" that is to say, that cold winds never bring extremely low temperatures from the plains of Europe or the mountains of Norway. But when shallow gradients for east and north-east winds cover Great Britain, and a dry chilly air favours nocturnal radiation, then all the hardest frosts are developed. Then we often find the temperatures 10° or 20° lower in the most inland stations of England and Ireland, and the isotherms gradually increase round these cold centres. When we look at a synoptic chart of Europe for 8 a.m., we find, on these occasions, that England and Ireland are separate islands of cold on the general thermal slope from a cold continent to the warm North Atlantic. From the fact that frost depends on radiation, we can readily explain why cold is so local. Radiation is very sensitive; the least breath of wind or any local shelter may interfere with the free play of radiation, and so we find two places only a few miles apart, one of which records 10° or 15° lower than the other.

The next source of cold is found in wind. When this blows from a frozen continent, then, of course, very low temperatures may be recorded; but this is not the same kind of cold as radiation-frost. Here we have another of the innumerable instances of the necessity of distinguish

ing between different kinds of the same nominal phenomenon. The 1st of January may be cold in one year from wind; in another from radiation. These are the products of totally different kinds of weather, and must not be mixed up in scientific meteorology.

THE "BLIZZARD" AND THE

66 BARBER."

A very striking example of wind is found in the "blizzards" of the United States. These are cold snaps which come with a high wind, as opposed to the calm frost of anticyclones. They are the result of the passage of the rear of cyclones or of V-depressions in the winter months, such as we see in Figs. 42 and 43. Then we get high, strong, north-westerly winds, blowing off a frozen continent with a temperature many degrees below zero, and with surroundings which are very destructive to life. The wind drives the cold into the bones even through fur clothing, and raises a blinding dust of powdery snow. Under these circumstances only are the western voyagers ever lost. If wood cannot be found, nature can only resist the cold for a certain number of hours, and the men are frozen to death if no shelter can be reached. A very curious circumstance attends these deaths. In almost every case the victims are found to have begun to strip themselves. When the body is nearly reduced to an icicle, only a very little blood continues to circulate languidly through the brain. Then delirium sets in, with a delusive sensation of heat, under the influence of which the traveller begins to divest himself of his clothes.

Another disagreeable form of cold is found in the

St. Lawrence Gulf. Sometimes with a high wind the air becomes much colder than the open water. The latter, being relatively hot, begins to smoke, and the vapour freezes into peculiarly sharp spicules. The poudré snowcrystals of the north-west are usually small, dry, six-sided petals, and, though penetrating as sand, they are soft. The latter kind of snow is so damp and sharp that, when driven by a gale, it nearly cuts the skin off the face. Hence the popular name of the "barber," which is applied to this phenomenon. The same name of " barber" is applied to another phase of cold along the coasts of Nova Scotia and New England. When a vessel is caught by a gale of wind in a cold arctic current, the spray freezes the moment it touches the deck or rigging. Every block is turned into a lump of ice; men get coated with ice like an icicle; and sometimes such a weight of ice forms on the bow that the stern is lifted out of the water, and the ship becomes unmanageable for want of steering power.

The last source of cold which we need mention is rain. All rain, of course, is not cold. In front of a cyclone rain is warm, and a shower does not send down the thermometer. In the rear, on the contrary, and in thunderstorms and secondaries, precipitation is more or less cold, and turns the mercury downwards. The influence of this varies very much in different countries and at different seasons of the year. In England, during the summer, rainy weather is cold, because it cuts off the sun, independent of any chill of its own. In winter, on the contrary, rainy weather is warm, because an overcast sky prevents loss of heat by radiation. In the tropics cloudy weather is colder, as far as the thermometer is concerned,