CHAPTER IV.

ISOBARS.

In our introduction to prognostics, we have already explained the leading features of the science of isobars, and of their relation to the changes in the readings of the meteorological instruments which are most usually observed. In our chapter on Clouds, we have also introduced the reader to the idea that over the surface-circulation of the air round cyclones and anticyclones there is an upper circulation of a very different character. But we must now go more deeply into the subject, so as to explain many details which could not then be conveniently given, and we shall not only complete our description of the nature of cyclones, anticyclones, etc., but also describe the two remaining forms of isobars-V-shaped depressions and cols-which were omitted in the previous chapters.

CYCLONES.

We have already sufficiently explained the broad features of a cyclone, and the wind and weather which are associated with it. The reader will now comprehend. what is meant by the centre, the trough, the front or rear,

the intensity, the path, and the velocity of the cyclone; and he will also understand that it is generally associated with bad weather, and rapid shifts of wind according to very definite laws.

What we want to consider now is the kind of circulation which constitutes a cyclone, and some points connected with the propagation and motion of this particular kind of low pressure. Conventionally, we shall not call a low pressure a cyclone, unless the isobars form a well-defined closed-curve. With the exception of V-shaped depressions, any other irregular area will be called by the generic name of a "depression."

GENERAL CIRCULATION.

As the surface-wind in a cyclone is always a little incurved and the upper wind always more or less outcurved, the inference is irresistible that the main body of the air near the centre of a cyclone must be rising; otherwise, as the wind is always blowing in, the cyclone would soon fill up if there was no escape upwards. To this ascensional movement undoubtedly must be attributed the rain and cloud which we find there-rain near the centre, where the ascensional impulse is strongest; cloud round the outside, where the uptake is less strong. From this we can readily understand the effect of what we have called intensity in a cyclone. It is not difficult to conceive a cyclone which possessed so little intensity that it could only develop cloud in the centre. Then, if from any cause the intensity of the ascensional current could be increased, rain would be developed where only

cloud had been formed previously. Thus we get hold of the idea, which we shall work out in some detail in future chapters, of the influences which can modify any existing cyclone.

If, for instance, any cause, such as the heating of the ground by the sun, increased the velocity of the wind, and so poured more vapour-laden air into the centre in a given time, then the uptake would be greater, and the tendency to form rain would be increased. Similarly, if the wind was unchanged, but local causes, such as a range of hills, gave the inpouring currents an increased ascensional impulse, then, too, the precipitation of rain would be still further developed.

AXIS.

Returning now to our conception of the cyclone as a circulatory system, it is manifest that we may consider the whole as constituting an extremely complicated vortex, something analogous to an eddy of water. There is, however, this difference-that a water-eddy sucks down, while an aerial cyclone draws upwards.

The line along which any particle of air may be supposed to move must not only curve irregularly inwards, but also upward, and finally outwards. As the whole is treated as a whirling system, there must be a line, more or less perpendicular to the earth's surface, round which the air rotates in this complicated manner. This imaginary line is called the axis of the cyclone.

There is much uncertainty as to the nature of the circulation round this axis. Some writers have thought of

the axis of a top, and believed that the axis of a cyclone can nutate, always keeping a revolving disc of air perpendicular to itself, so that the cyclone would be pressed down on the ground in the direction towards which the axis inclined, and be lifted off, as it were, on the opposite side. This, they say, would explain the anomalies that are sometimes found both in the position of steepest gradients relative to the centre, and in the variable destructiveness of the wind.

With the same velocity, wind will sometimes unroof houses, at other times do little damage; and they consider that in the first case the direction of the wind is a little upwards, in the latter a little downwards. They believe that this conception of an inclined axis is confirmed by the fact that, in tropical cyclones, the small, clear patch of blue sky in the centre of a cyclone is not always exactly over the point of lowest barometer. They would then consider that the axis of the cyclone is like a telescope pointed upwards, at some angle from the ground, instead of truly vertical.

The insuperable difficulty in the way of all this lies in the fact that a cyclone is often one or two thousand miles across, and certainly not more than ten miles deep; so that the amount of tilt required to give the observed deflection of isobars would be sometimes 20° or more, and that, with a disc of two thousand miles, would be impossible under the conditions of our earth.

If, for instance, we glance back for a moment to our typical cyclone (Fig. 2), we see that the isobar of 29 ins, is not concentric with that of 30 ins. The idea would be that a vertical-axis cyclone would have concentric circular

isobars, but that both the oval form and setting back of the isobars at one side are due to the axis being tilted forward towards the word Front.

Another theory to account for all these facts supposes that a cyclone is made of a series of flat oval horizontal sections, but that these are not superimposed concentrically one on the top of the other, but pressed successively more or less to one side by surrounding influences. In Fig. 2 this shunt would have been towards the rear, instead of forwards as by the preceding hypothesis.

This view is probably partially correct, though it is impossible to suppose that the air does not get more or less inclined upwards at times, for no cyclone is ever absolutely symmetrical.

We often see the conical revolving cloud of a whirlwind or tornado bending about like the trunk of an elephant, with both a true axial inclination and a certain amount of sideways shunt. Here, however, the vertical height is enormously greater than the diameter, which is just the opposite to the proportions of a cyclone. Anyhow, if we suppose that upper winds follow the same laws as surface-currents with relation to isobars, observations on cirrus-clouds tend to the belief that the axis of a cyclone is very often inclined backwards from the direction in which the cyclone is moving, as if the surface portion was going faster than the upper.

This is just the converse of what might have been expected à priori, that surface friction would retard the lower portions, so that the axis of the cyclone would have been inclined forwards. The whole question is, however, still very obscure.

K