CHAPTER III.

CLOUDS AND CLOUD-PROGNOSTICS.

In this chapter we propose to discuss the nature of clouds by first explaining their origin and the varying conditions under which they are formed.

This will lead to a classification of their different shapes and forms, and give us a certain insight into the varying velocity and direction of the upper currents of the atmosphere. But when we come to the more modern developments of cloud-knowledge, we shall have to consider the relation of clouds to the great areas of high and low pressure, which we have already described as cyclones and anticyclones.

Some of this we have already seen in our chapter on prognostics, where we showed that different kinds of cloud are characteristic of different portions of cyclones, etc. But in this chapter we will explain how, from a study of cloud-motion in the upper strata, we are enabled to discover much about the real nature of the circulation of the air in the different shapes of isobars. In the course of our remarks, we shall explain incidentally both the meaning and value of the older cloud-lore, and

also the great development in the science of forecasting. by means of clouds which has been made by recent researches.

NOMENCLATURE OF CLOUDS.

Unfortunately, in approaching the subject of cloudnomenclature, we come to one of the most unsatisfactory branches of meteorology. Though the words of Howard's nomenclature are universally employed, the same word is by no means always employed for the same kind of cloud; and for this reason, though the words we shall employ to designate clouds are those which are used by many, we shall be very careful to describe exactly the kind of cloud we mean by any particular name.

For practical purposes clouds are divided into four different classes, according to their most obvious differences of shape; but these classes are only as a matter of convenience, for in nature they all run into each other by imperceptible gradations. The forms are

1. Cumulus. All cloud which has a rocky or lumpy look is either pure cumulus or must contain the word cumulo in combination with some other name.

2. Stratus. All cloud which lies as a thin flat sheet must either be pure stratus or contain the word strato in combination.

3. Cirrus. All cloud which has a wispy, feathery, or curly look must either be pure cirrus or must contain the word cirro in combination.

4. Nimbus. Any cloud from which rain is falling is nimbus in some form.

It must be fully understood that these names and their derivatives, which we shall give presently, do not by any means exhaust all the varieties of clouds which very experienced observers can detect and classify. All that we propose to give here are the broad distinctions which anybody can understand, for all meteorologists who have to deal with corps of observers are agreed that eight or ten names are as many as can practically be employed.

We will first explain the simple forms of these clouds, and then the more complicated combinations, such as cirro-stratus, cirro-cumulus, cumulo-stratus, etc. But besides giving a broad classification to the leading kinds of cloud, these terms also give a rough relative scale of altitude. Thus in practice stratus and cumulus are usually the lowest, the composites the middle, and cirrus the highest layer of cloud; but no absolute level can be assigned to each stratum at any season, or in any country. For instance, cumulus may be as low down as 2000 feet, and cirrus as about 12,000 feet; and, on the other hand, cumulus may be formed up to at least 25,000 feet, and cirrus probably up to at least 50,000 feet; but true cirrus can never be formed under cumulus, whatever the relative latitudes may be.

These relative heights also partially determine the nomenclature. If a cloud is very high up, we have to add the word cirro, to indicate altitude, to the word which denotes the form only; while cumulo would suggest a lower level. The first word before a compound name gives the idea of relative altitude: thus cirro-cumulus is higher than cumulo-cirrus.

CUMULUS.

Pure cumulus may be described as convex or conical heaps increasing upwards from a flat horizontal base, as in Fig. 11, a. It is undoubtedly formed by the condensation of the summit of an ascensional column of vapourladen air, as shown by the dotted lines. When this is cooled, either by rising into a colder stratum than that from which it started, or by expansion, the water-vapour

FIG. 11.-Cumulus and cirrus. a. Cumulus, surface rapid. b. Cirrus, surface rapid. c. Cumulus, upper rapid. d. Cirrus, upper rapid.

condenses into cloud, like the condensed steam from an engine.

The flat base marks the level where condensation temperature is reached, and the upper rocky summit represents the heads of the air-columns protruding into a cold space. A cumulus is, in fact, the visible capital of an ascensional column of air.

There is one very remarkable feature of all cumulus : it is never seen as such overhead, but only on the horizon, or at a moderate height above it.

The reason is obvious, that as the flat-based mass drifts overhead, the flat under-surface hides the characteristic rocky top, so that we no longer see the typical features of this kind of cloud.

In Northern Europe, and the interior of many continents, cumulus usually only forms during the summer months; for the absolute amount of vapour in the air during the winter months is rarely sufficient to develop lump of cloud.

RELATION TO CIRRUS.

If the ascensional column is stationary, we get a very curious appearance; the top of the cloud seems to be, and is, in a state of commotion, but still the cloud as a whole does not move in any direction. This is very puzzling at first, and is not uncommon before thunderstorms; but we can readily understand its origin by watching the stationary cloud on a hilltop. Then we see the same contradictory appearance-a cloud in rapid motion, but never moving forwards. The reason is that, as each fresh portion of cloud is projected upwards and blown away by the wind, it is immediately evaporated, but a new column of vapour instantly takes its place. But suppose that, whether stationary or moving, the rising column, after depositing a certain amount of its vapour at one level, continues to rise, it will at length reach a second level, at which the condensation-point of