France during that day, which are marked ▲ and û respectively. Such lines are called "isobrontons," or lines of equal development of thunder. Those in the figure may be taken as typical of the march of this class of thunderstorm all over Europe-a long narrow line, advancing broadside on 11.p.m, towards the east or north

5.a.m. 7

A

p.m.

B

1.p.m.

FIG. 53. Track of thunderstorms.

east, pretty nearly independent of the shape of isobars with which it is associated. The first storm, marked A, struck the coasts of the Bay of Biscay at five o'clock in the morning. The shape of the front was bent, the ends being most in advance: this is very often the case in

France. The position of the front of this storm is given for every two hours' interval until 1 p.m., when the disturbance appears to have died out near Calais. The relation which the position of the thunder bore to the secondary cyclone may best be seen by reference to the preceding chart (Fig. 52). There the dotted line shows that the front of the storm A lay to the north-east of the centre of the cyclone; and, as far as the small number of wind-arrows allow us to judge of the nature of the disturbance, there seems to be a small local deflection of the wind in rear of the storm. The wind in rear should have been from south-east with the shape of isobars there represented, whereas the chart shows that in some places the direction was from the west and north-west.

The second series of thunderstorms, marked B, commenced at Biarritz at three o'clock in the afternoon, and moved irregularly in a more westerly direction than the morning storm. The storm, in its course during the day, seems to have increased enormously, for at 9 p.m. we find the front reaching nearly from Brussels to Perpignan in the Pyrenees, a distance of six hundred miles. This will perhaps enable us to realize the disastrous character of hail and thunderstorms in France. Here we have a line of destruction six hundred miles long, and from ten to twenty miles broad, sweeping like a curtain across the country at a rate of about thirty miles an hour, wrecking in a few minutes vineyards which are worth many thousands of pounds, and destroying at the last moment the husbandman's labour for the whole year.

But now we come to one of the most puzzling points connected with thunderstorms. If we look at Fig. 52, where we have marked by a dotted line the position of the storm-front в at 9 p.m., it is very difficult to see any connection between the shape of the isobars and the position of the thunder. So far as we can see, there is no trace of the trough either of a cyclone or of a V-depression, and all the general indications would have been for improving weather after the passage of the secondary cyclone. Whether more numerous observations at stations nearer to one another would have shown the presence of secondaries, we cannot say; but this is by no means an isolated instance in France, and similar cases occur in other countries. For the present, therefore, the explanation of the nature of this class of thunderstorms must await future research; all that we can do here is to

note them as an apparent exception to the general course of weather which we have already explained. We must also specially note them as cases where rain falls with a steady rising barometer; and also understand that, although a forecaster could not have pointed out in the morning exactly when or where thunderstorms would strike, he could have said with certainty that storms would occur during the day in many parts of France. Secondaries can never form in summer without some electrical disturbance.

The details of rain and cloud in line-squalls and thunderstorms are extremely interesting, and for them we are chiefly indebted to the careful researches of Dr. Koppen. The approach of a thunder-squall usually announces itself by the rapid crowding up of heavy clouds. We see a dark, black border, often looking like a long roll or wreath, and beyond this a peculiar light grey uniform sky. The dark, low band passes overhead, and heavy rain commences as the light grey cloud comes on. The first burst of rain is usually the heaviest, and after a longer or shorter period the rain usually clears gradually off. This is the rain with which the sudden rise of the barometer is observed. The wind, which has fallen very light from south-east or south as the clouds begin to bank up, comes in a violent squall from the west, about the time the dark wreath passes overhead, and falls again shortly after the commencement of the heavy rain. A good illustration of a very pronounced cloud-wreath will be found in Fig. 56 in the next chapter under "Pamperos." We give an ideal diagrammatic section of such a squall in Fig. 54. We may suppose

that from general causes, such as the trough of a V-depression, a cold westerly current meets a warmer one from the south-east or south. The latter rises, as shown by the small arrows, and curls over where the black wreath (w) of cloud is found, and then the commingling of the two currents forms a gigantic dark vault (v) of cloud, from which heavy rain (r) pours down. The light

FIG. 54.-General circulation and cloud-vault of line-squall.

grey cloud which an observer sees behind the black wreath is really a peep into the rain falling from this great vault. The big drops of rain bring down mechanically with them a vast amount of cold air, which rushes straight out in front of the squall-it has no time to pick up anything from the earth's rotation-and produces the squall q, marked by a long arrow.

Our section, then, of a squall is that of a vertical whirl, the whole system perhaps not one mile high by one and a half mile across, while the length of the front of the storm may be two hundred miles; and our picture of the whole must be a long, nearly straight horizontal axis, moving broadside on, round which the

wind whirls vertically in a direction opposite to that of the watch-hands. It is always an episode, as it were, in the history of some general form of atmospheric circulation. We can, perhaps, see how a broad-fronted west current can meet a southerly stream of air in the trough of a V or cyclone; but we are unable at present to form any conception of a straight-fronted current advancing across the curved isobars of an anticyclone, as in Fig. 52 for French thunderstorms.

Line-squalls and thunderstorms of a different type are very common in the tropics. The author has observed a very striking instance at the junction of the south-east trade and north-west monsoon in the Indian Ocean. There was no doldrum, but the two currents met along a line whose position was marked by a long dark, black cloud, with heavy rain and squall.

In like manner, the daily thunderstorm which occurs in so many countries at the time when the sea-breeze comes in, charging, as it were, the prevailing wind over the land, is due to a long vertical whirl where the two currents meet, and the whole length can sometimes be watched gradually advancing inland from the coast. The so-called north-westers at Calcutta, during the hot season, belong to this last type.

THUNDERSTORMS WITH SECONDARIES.

We must now just mention a class of thunderstorms which are more complicated than a simple squall, and yet differ in many ways from line-thunderstorms. They are associated with secondary cyclones, and are much