Major-General the Hon. J. LINDSAY, M.P., in the Chair.

ROTATORY STORMS.

By Captain Sir FRED. W. E. NICOLSON, Bart. R.N., C.B.

THIS Lecture was originally delivered at a meeting of the Literary and Scientific Society of Shanghae, and contains, in consequence, more elementary matter, and a greater number of local illustrations, than would have been considered necessary, had it been prepared in the first instance for the Lecture Theatre of the ROYAL UNITED SERVICE INSTITUTION. To persons unacquainted with the subject, it may be proper to state that most of the diagrams are copied from various well-known works on meteorology. 1. That branch of meteorology to which the law of storms belongs has been developed only during the last thirty years. Previous to the researches of Redfield, Reid, Dové, and others, all winds were supposed to blow in straight lines. The facts recorded by these observers induce the belief that all violent storms, and probably many, if not all, the variable winds, rotate in obedience to certain laws. Storms are known under various names in different parts of the world. In the West Indies they are termed "hurricanes;" in the Chinese seas, "L tyfoons;" and a new name, "cyclone," indicating their revolving character, is coming into general

use.

I shall use these names indiscriminately in this paper.

2. Colonel Capper of the East India Company's Service was one of the first, if not the very first, to draw attention to the peculiarities of the great storms in the Indian Ocean. He published a work on winds and monsoons in 1801, in which he suggests the possibility of ascertaining the position of a ship in a whirlwind by noting the strength and changes of the wind.

3. This suggestion of Colonel Capper's does not appear to have excited much attention, nor was any progress made in elucidating the phenomena of storms until the subject was taken up by Mr. Redfield of New York. His investigations commenced in 1821; but his records of storms were not published until they appeared in the American Journal of Science for 1831. Mr. Redfield was the first to explain the oscillations of the barometer as connected with the veering of the wind, a phenomenon which had perplexed the ablest philosophers since the invention of that instrument by Torricelli.

4. Sir William Reid, following in the footsteps of Mr. Redfield, has contributed several works to this branch of meteorology. We must all deplore the loss which both services have sustained by his death. He was a most kind-hearted, simple-minded man, and ever ready to impart to others the knowledge he had acquired during a long and laborious life

spent in the public service. Sir William was employed at Barbadoes in restoring the Government buildings destroyed in the hurricane of 1831; when 1477 persons lost their lives in the course of a few hours. This sad catastrophe directed his attention to the investigation of the West Indian hurricanes, and brought Mr. Redfield's recently published paper under his notice. Feeling convinced of the correctness of the views expressed in that publication, Sir William sought to verify the theory thus propounded, and forthwith collected data from every quarter. He has published the results of his investigations in two separate works, which contain a most valuable collection of facts relating to this branch of meteorological science.

5. Professor Dové of Berlin has been engaged, since 1821, in studying the winds and their phenomena, and has published several excellent memoirs on the subject. He claims to rank with Redfield as the discoverer of the law of storms. His researches, like those of Redfield and Reid, were in the first instance confined to storms in the northern hemisphere. From a consideration of the phenomena observed during these storms, these three meteorologists, although working independently of each other, came to the same conclusion, that gales south of the equator would be found to rotate in the opposite direction to those of the northern hemisphere. More extended observation has fully confirmed this theory, which they simultaneously made public.

6. Professor Espy of the United States is another writer on storms and winds. Although his theory differs from that of Redfield and Reid, it is considered by many that the difference is more apparent than real, and that the centripetal winds of Espy's theory will in fact produce the whirlwind of the rotatory theory. An explanation of the means by which the two theories may be reconciled will be given when we come to treat of the causes by which cyclones are produced.

7. Dr. Thom and Mr. Piddington likewise deserve particular notice; the former has devoted his attention to the storms in the Indian Ocean; both north and south of the equator.

8. Mr. Piddington, who, I regret to say, has recently been taken away in the midst of his useful labours, held an appointment at Calcutta; he has thus been enabled to collect a mass of important data respecting storms in the Indian seas; and he has entitled himself to the gratitude of every seaman for the publication of his "Sailor's Hornbook," an excellent handbook for the mariner in the regions of hurricanes and tyfoons.

9. Sir John Herschel's article on meteorology in the last edition of the Encyclopedia Britannica, and the works published by Admiral Fitzroy's department at the Board of Trade, contain much valuable information respecting storms and their phenomena; and the elaborate list of works on meteorology at the end of Sir John's article will be found most useful to those who wish to study this interesting science in greater detail.

I shall now proceed to explain the law of storms, premising, however, that I shall avoid entering upon the more extensive question of the winds in general, except where I may find it absolutely necessary to illustrate the special subject we are about to consider.

10. The law of storms may be briefly stated as follows:-All strong winds to the northward of the equator are whirlwinds, revolving in a direction opposite to that of the hands of a watch. See Plate I.

To the southward of the equator the rotation is in the contrary direction. See Plate II.

11. In addition to this rotatory motion, storms have likewise a progressive movement, which may be thus described :-Taking the West Indian hurricanes as an example for the northern hemisphere, we find that they move in the first instance to the west-north-west. Then as they recede from the equator their course becomes more northerly, and they gradually recurve to the north-east. Their path thus forms a species of parabolic curve, whose vertex lies near the Gulf of Mexico, in about 30° north latitude. See Plate III.

Storms in the southern hemisphere likewise move in a westerly direction in the first instance, then bend to the southward, and finally recurve to the eastward. See Plate IV. The final direction of those best known, which have been chiefly observed in the neighbourhood of the Mauritius, is about east-south-east. Although fewer storms have been accurately observed in the southern than in the northern hemisphere, a sufficient number of storm-tracks have been laid down to prove that their paths are curves of a parabolic form, similar to those north of the equator; and that their vertices are situated in about 20° to 25° of south latitude.

12. The rates of progress in both hemispheres, as ascertained from numerous observations, exhibit most remarkable differences of velocity.

It may not be unworthy of note that the hurricane of 1822, which caused an inundation of the Ganges and Burrampooter, when 50,000 people are said to have perished, is the storm whose rate of progress appears to have been the slowest of those recorded.

13. The diameters of cyclones vary quite as much as their rates of progress. Some are proved to have been only 50 miles in diameter, while others have said to have expanded to diameters of 1,000 miles and even more. These storms of vast circumference have been principally noticed in the North Atlantic. The average diameter of cyclones may be estimated at from 300 to 500 miles.

In the hurricane of November, 1839, which devastated Coringa on the western side of the Bay of Bengal, it was found that the violence of the wind increased as the diameter of the cyclone contracted. The converse is probably true, that the force of the wind decreases as the cyclone's diameter increases. This may possibly explain why the gales on the British coast are less severe than the West Indian hurricanes; for the former are generally storms of very large diameter, while the latter are usually of comparatively small dimensions, expanding, however, after they have recurved to the north-eastward.

14. Cyclones must not be considered as cylindrical in form; they are in fact flat discs, whose height is supposed to be small compared with their diameters, and they have a concave upper surface, the cause of which will presently be shown. The axes upon which they may be supposed

to revolve are often inclined at a considerable angle to the horizon. We may thus conceive the rear of a storm to be tilted up, while its anterior semicircle alone impinges upon the earth; hence we see how it may be possible for a vessel to encounter only a portion of a cyclone, although the centre may have passed over her. Many of these severe storms appear and disappear so suddenly, that they have been supposed to have been whirled down from the upper regions of the atmosphere and then to have soared up again, after touching only a small part of the earth's surface. Many great storms have probably terminated in this manner, for it is frequently found impossible to trace them beyond a certain point; and we can hardly suppose the rotation to have ceased suddenly without previous symptoms of gradual subsidence.

15. It is almost impossible to give any notion of the extreme violence of the wind in a tyfoon or hurricane to persons who have never experienced one of these meteors, although the following Table gives an approximate indication of the force and velocity of the wind. A few incidents selected from numerous records of severe storms may serve as imperfect indications of the fearful force of the wind.

We read of boats being entirely rent to pieces by the wind alone, of masts broken, although no sails were set, and of sails blown away from the masts on which they were furled and secured in anticipation of the tempest.

The most sheltered harbours afford no certain security in these terrific storms. That disasters should occur at Hong Kong need scarcely surprise us, but that vessels should be totally wrecked in Malta Harbour is a convincing proof of the strength to which a wind may rise, even in places supposed to be exempt from actual hurricanes.

When the "Raleigh" was upset in a tyfoon on the coast of China in August 1835, she was under bare poles. The crew scrambled upon the upper gunwale and managed to cut away the rigging; the masts for

tunately were carried away, and the ship righted with no further damage than four feet of water in her hold.

Sir W. Reid records the following anecdote as an instance of the roaring of the wind in a hurricane. During the Barbadoes hurricane of 1831, Colonel Nickle, of the 36th Regiment, was standing under the arch of a lower window of his house; the roof and the upper story fell in, and, although the colonel was covered with dust from the falling ruins, the crash was perfectly inaudible.

The following occurrence is mentioned in Professor Dove's work; the minute details give the story an air of accuracy.

When Basseterre in Guadaloupe was destroyed by a storm on the 26th of July, 1825, General Baudrand states that three 24-pounders were borne away by it (in what manner is not stated, probably by the sea), and a piece of deal board 37 inches long, 9 inches wide, and 10 lines thick was hurled through a palm-tree 16 inches in diameter!

A still more wonderful fact is recorded in Professor Espy's work (Philosophy of Storms, p. 344), where it is stated that tin from the roofs of the houses of Natchez, on the Mississippi, was found at 20 miles' distance, and that a piece of a steamboat window was recognised 30 miles north-east of the town.

16. Such being the terrific fury of the wind in a cyclone, it becomes a question of the utmost importance to the seaman how to escape from the most violent part of these revolving storms. For this purpose the barometer is invaluable.

Since this instrument measures the weight or pressure of the atmosphere, it is evident that, if the height of the column of air above a barometer be diminished, the pressure supporting the mercurial column will be lessened, and the mercury, or, as we loosely phrase it, the barometer will fall. Bearing this fact in mind, let us examine in what manner the rapid rotation of the wind can affect the barometer.

17. When a rotatory motion is given to a fluid in a glass or jar, we find that the centrifugal force drives the water away from the centre, causing a depression at that point. In fact, the surface of the fluid, instead of remaining level, becomes concave. A similar depression takes place in the atmosphere during a cyclone; the air is spread out by the centrifugal force, and the heights of the atmospheric columns diminish gradually from the outer edge or circumference of the cyclone to the centre. We can thus explain, what experiment and observation teach us, that the barometer falls as the centre of a rotating storm approaches; and rises as that point recedes from the place of observation. Consequently it may be assumed as a general rule, that, the lower the barometer, the nearer is the centre of the storm.

18. The bearing of the centre is easily ascertained from the direction of the wind. In Plate I. we find that in the northern hemisphere with the wind at north, the centre bears east; wind at east, the centre bears south, and so on. In the southern hemisphere we must reckon the eight points of the compass between the wind and the bearing of the centre, in the opposite direction. Thus with the wind at north, the centre bears west; wind at east, the centre bears north, and so on. The incurving of the wind near the centre may in some degree modify this rule; but for all practical purposes it will be found sufficiently accurate.