the wind rotates in a contrary direction to the progressive motion, and the latter, therefore, diminishes the intensity of the former.

One of the consequences of this rotation of the air around a central space, is to produce a rarefaction or diminution of pressure, in the latter region, so that in this central part of the cyclone the barometer falls very low. The course followed by this depression (lowest barometer) is that of the cyclone.

The data collected by vessels crossing the Atlantic, compared carefully with those supplied by the various meteorological stations in Europe, point more and more to the truth of the old opinion that the Gulf Stream is the “parent of tempests" of the Atlantic and its coasts. Its comparatively high temperature the abundant moisture of the air above it, the opposite winds blowing towards its course, all help to make it the theatre of frequent atmospheric perturbations The aërial current which reigns above the Gulf Stream naturally flows towards the east, in the direction o Europe.

Besides these larger commotions that travel s far, it is now admitted that small local cyclones ar occasionally formed in our latitudes, by the sudde combination of a northerly and southerly current Such was the case in September 1876, when a minut storm of this description produced such havoc at Cowes in the Isle of Wight. The author was at Bournemout at the time and experienced nothing. Again, in 1878 a similar occurrence took place when the Eurydi

was lost in the English Channel. It may also be recalled that on the 25th and 26th October 1859, when the Royal Charter was lost, the gale appears to have commenced in the Bay of Biscay and terminated in the Baltic.

The whole question of cyclones is, however, far less simple than this brief sketch might lead my readers to suppose. It has been ably grappled with by Dove, of Berlin, who began his investigations as early as 1827, and his most recent work, the Law of Storms, embodies his views on this important subject.1

1 The Law of Storms, by H. W. Dove, translated into English by Robert Scott, M.A. My copy of this important work was presented to me by the late Admiral Fitz-Roy, F.R.S., together with his interesting work, The Weather Book, a manual of practical meteorology, second edition, 1863.

CHAPTER XVIL

Absolute Weight of the Earth's Atmosphere-Various Optical Phenomena caused by the Air-The greatest height Man has reached on Foot - The Snow-line and Region of Perpetual Ice.

THE absolute weight of the Atmosphere is a very simple problem, though it has long been considered one of the most extraordinary achievements of modern science

The mercury, in a barometer, standing at 30 inches can be poured out of the tube into a balance and weighed. This weight is that of a column of air, o equal basis to the column of mercury. If the basi of the column of mercury is equivalent, say, to on square inch, the weight of the metal in the tube wil be about 15 lbs.

If we can find how many square inches cover th surface of the globe and multiply them by 15 lbs., th result will be the entire or absolute weight of the Earth atmosphere.

In this manner a great astronomer and geographe (Francœur) found that the weight of the entir atmosphere, in tons, is represented by the figure 523,260,000,000,000 tons.

We have already touched upon this subject when alluding, in a previous section, to the absolute amount of carbonic acid in the Earth's atmosphere.

The various optical phenomena presented by our atmosphere are so fully described in works on Physics, and Meteorology, that only a few words need be devoted here to this portion of our subject. As we have already said, the atmosphere is transparent and invisible, except when seen through an immense thickness; it is then blue, the colour of the pure sky. The presence of much moisture lightens this tint, so that it becomes grey. As we rise in the pure air, the vault of heaven becomes of a darker and darker blue, almost black, and the stars become visible, as they do when the sky is viewed from the bottom of a deep pit.

The average refraction of light passing through the entire thickness of the atmosphere causes the sun to be seen for eight minutes after it has set, and eight minutes before it actually rises, so prolonging our day by sixteen minutes at all seasons of the year. In some states of the air, refraction also distorts the figure of the sun near the horizon. It is not, as many have thought, the cause of the apparently increased size of the full moon near the horizon. This is an effect of contrast, caused by viewing the moon in comparison with objects on the Earth's surface; and if viewed through a narrow tube of cardboard, so as to shut out these objects, it appears of the same size as when seen at the zenith.

To refraction of light is also due, as we have already said, the curious phenomenon of the mirage, the prismatic colours of the rainbow, of the solar and lunar halos, and of the sky at sunset and sunrise. Another optical effect that used to be much talked about, is the so-called Spectre of the Brocken-that is, the magnified images of travellers standing at sunrise on the summits of the Hartz mountains, their shadows being thrown upon the mist or clouds by the rising sun at their backs. I have often seen this phenomenon in the neighbourhood of London, at night, when the light from a lamp upon a table at the back of the observer throws his magnified shadow through an open window upon a dense fog.

The electric glows, or phosphorescence of rain-drops, in thundery weather, which have been witnessed on several occasions at Geneva and other places, the light which sometimes issues from the summit of the masts of a vessel (St. Elmo's fire), or from the hats of pedestrians during a thunderstorm, the glow of the Aurora borealis, during which the magnetic needle is disturbed, and a number of other curious emissions of light, often connected with electric perturbations of the atmosphere, have been fully described by m in other works.1

1 Phipson, Familiar Letters on some Mysteries of Nature, London 1876; Phosphorescence, or the Emission of Light by Minerals, Plants and Animals, London, 1862. (A spurious edition of this work wa issued without the author's knowledge in 1870.) Noctilucine (pam phlet), London, 1875. "Phenomènes lumineux, etc.," Comptes-rendus Paris, 1868.