to one-fifth of a second, he has found that the flash, the clap, and the splash of rain may be supposed really to occur simultaneously, but that the three impressions reach the earth's surface at different times, because light, sound, and a falling body all travel at various rates. Thus light travels practically instantaneously; sound at the rate of about 1100 feet a second; while rain-drops fall a definite distance in any given time, under the influence of gravitation. This would be proved if we found that the distance of the origin of lightning, as measured by the velocity of the sound of the thunder, was the same as that measured by the velocity of falling rain. For instance, on one occasion the interval between the lightning and the thunder was five seconds, while the rain did not arrive for nineteen seconds. Now, calculating the distance of the origin of the lightning from the velocity of sound, we find the altitude to be 5500 feet; while the distance through which a drop would fall in nineteen seconds would have been 5800 feet. difference is only 300 feet, which is very little considering the nature of the observations, and the unknown retardation of a falling drop from the resistance of the air. In practice the thunder always arrives before the rain; in fact, we may consider that the same disruptive discharge of electricity sends three messages to the earth at different rates, and to different senses-the light to the eye, the sound to the ear, and the rain to the touch.

The

UNCLASSIFIED CLOUDS.

So far for the great subdivisions of cloud-forms, but we must now mention a few minor forms, because they have some importance in judging weather.

CIRRO-NEBULA, OR CIRRUS-HAZE.

Sometimes, as a cyclone approaches, in any part of the world, and we are very nearly on the line of its path, we see a blue sky first get white, then grey, and then work up to drizzling rain, without the formation of any true cloud-form. When this happens, the sky is said popularly to sicken, and this is an almost infallible sign of rain, and probably of wind. Mr. Ley has proposed the name cirro-nebula," or cirrus-haze," for this appearance, and the term seems most appropriate. We may, however, observe here the necessity for our caution about the words cirro, cumulo, etc., conveying a rough idea of the height of clouds.

66

66

This cloud has no fibrous or hairy structure to which the name of cirro could be strictly applied; but if we also lay down that the word cirro is to convey the idea of a high level cloud, then the word cirro-nebula is quite correct. It is invariably formed at a great height, and as it nearly always shows a halo when the sun or moon shines through it, we may assume that it is composed of frozen particles, or ice-dust. After it has formed, we can often layer of cirro-stratus being formed underneath the haze. From all this we may draw the important

inference that, though the front of a cyclone is charac-. terized by excessive warmth on the surface, the upper strata are then very cold.

SCUD, WRACK.

Under any mass of cloud which is verging on the precipitation of rain, we have just mentioned that small detached clouds are frequently seen in rapid motion. In England they are called "scud; " in France, fuyards, or diabletons; while Poey suggests the name of fractocumulus. If, instead of being shapeless, they are raggy, they are then known in England as "wrack," from their drawn-out appearance. In all cases it is obvious, from the above description, that they are rather associates of heavy rain-clouds than true prognostics. What we have to explain is their origin. This seems to be simply, that in very disturbed weather small masses of cloud form like ordinary ragged clouds, from the irregular nature of the rising currents, while the apparently very rapid motion comes from their being nearer the surface than ordinary clouds.

CLOUD-WREATHS.

Sometimes, in front of certain kinds of squalls and thunderstorms, we see a long, narrow roll of black cloud moving rapidly, broadside on, and a very well-developed example will be found in Fig. 56, under the heading of "Pamperos." Dark cloud-wreaths in a very much less pronounced form are very common in England before

certain classes of squalls and showers, and in front of a curious light grey vault of rain-producing cloud, as illustrated in Fig. 54, in our chapter on Squalls.

There are, of course, many other minute differences of the various classes of cloud, to which it is impossible even to allude in an elementary work like the present. Our object will have been attained if we have succeeded in explaining the general principles of cloud-formation, and the method of making prognostications from their varying appearance. A few hours spent in watching the changing and degrading forms of a sky which is covered by detached cumulus, or the very different modifications almost from minute to minute of cirro-stratus, will better assist any one to understand the nature of cloud-forms than reading pages of the best printed matter.

We may conveniently summarize here the various varieties of cloud-forms which we have already described. The general idea of our classification has been that, though for large bodies of observers all practical men are agreed that eight or ten principal varieties are all that can safely be used, still more advanced cloud-observers will not be satisfied with so coarse a subdivision, and that therefore more minute varieties are necessary.

The ten principal varieties are therefore printed in capitals, while the minor varieties are denoted by smaller letters.

But there is another point in our subdivision of varieties. Almost all the smaller varieties are so rare or transient that for practical purposes they may be neglected; but if, on the contrary, the ten main words are restricted to the forms of clouds we have described under

them that is, cumulus, pure rocky cloud; stratus, pure sheet cloud; cirrus, pure wispy cloud; cirro-stratus, thin, high, wispy, or striated sheet cloud of all sorts; stratocirrus, a similar low cloud; cirro-cumulus, fleecy cloud at high level; cumulo-cirrus, the same, lower down; stratocumulus, extended lumpy cloud; nimbus, low rain-cloud; cumulo-nimbus, rocky rain-cloud-then the author can say, from an experience of cloud-observation in all longitudes, and in latitudes ranging from 72° north to 55° south, that ninety per cent. of skies in every part of the world can be sufficiently accurately defined by these ten words.

VARIETIES OF CLOUDS.

With the mean height of the principal varieties at Upsala