« 이전계속 »
has been emphatically cited. The alleged period of return of this disturbance is 28 days. The essential feature of this squall, as pointed out by the Rev. Clement Ley, who published in Symons' Meteorological Magazine, for April, 1878, a most careful discassion of it, was that it advanced across the country from north to south with a front like a wall, developing great violence of wind in some places where the current was confined by the local configuration of the ground. The very gast which struck the Eurydice was hardly felt at a distance of half-a-mile from her, and had it not been that that luckless craft was in what we may call the line of fire of the blast, the squall would hardly have been dignified with the title of a storm.
The direction of the wind on this occasion was northerly, and its greatest force was off the Isle of Wight, while it was not conDeeted with any great cyclonic disturbance crossing the country; and yet we hear that predictions of its recurrence are completely falfilled by southerly gales connected with a serious barometrical disturbance, and felt in the North of Scotland. The assertion does not deserve a moment's attention !
Again, the announcement of the discovery of a regular recurrence of meteorological events, coincident with the supposed eleven-year period of solar spots, has of late attracted very general notice, but When we come to apply to this theory the rude test of practical utility and common sense, we find that even its most enthusiastic advocates admit that the time for safe prophecy has not yet arrived.
We need only point out that not one of the sun-spot men ventured to predict the prolonged cold of last winter, the all but incessant rain of last summer—at least in England, or the early and severe frost from which we have just emerged. There was the same absence of spots from the sun's surface in the winter of 1877-8 as in that of 1878-9, and yet the resulting weather betrayed no signs of family resemblance.
To take a final instance we may safely ask any unprejudiced seaman what has been the practical value to him of the warnings with which we have been so freely supplied by the munificence of Mr. Bennett, of the New York Herald. In the form in which these telegraphic announcements are usually couched, they apply to
“ the coasts of England and France, affecting those of Norway." This sweep of the net takes in some twenty degrees of latitude at least, and we may safely say that no instance is on record of the same gale being felt in the same way over so enormous an area. Sapposing a captain is in the port of London, or even at Falmouth, we can hardly think that it is of much use to him to receive warning of a gale which only visits the Shetlands and the coast of Norway. At the time that such an occurrence takes place, it is more than probable that he is lying all but becalmed on the south coast, with a barometer ranging considerably over thirty inches ; for the passage of cyclonic disturbances outside our northern coasts is generally associated with the presence of an anti-cyclone over the south of England.
But, even if the Herald warnings were perfectly correct, they rarely profess to give more than 48 hours notice of a coming storm, and this interval is quite insufficient to allow a ship leaving London to get clear of the Channel.
On the whole, the evidence is irrefutable that in the month of January, 1880, meteorologists are utterly unable to tell the farmer whether he will have an early or a late, a dry or a wet, spring, or even to announce to the seaman, with any certainty, his general chances of weather for even a week in advance.
This is a humiliating confession to have to make, and I shall endeavour in the following pages to show what are the inherent difficulties in the problem, and the defects in our mode of attacking it, which render us so helpless when we attempt weather prediction on a large scale.
In the very first instance, we have to admit our all but total ignorance of the proximate causes to which the rise and fall of the barometer, and even the very wind that blows, are due. Our system of weather telegraphy simply derives its efficiency from the fact that telegraphic messages travel more rapidly than the storms which they announce; and so, speaking generally, it is possible to give some intelligence of the approach of a storm. But, to give this usefully, we ought to know the superficial extent of the disturbance, with the rate and direction of its motion, its nature and intensity, and whether it is increasing or diminishing
in severity. There is not one of these points on which the state of our knowledge is at all satisfactory. We lack the data on which to base our induction. We, in these islands, are particularly badly off in this respect, as we have the Atlantic on the weather side of us, for at least nine-tenths of our storms come to us from the west and north.
Our cousins in the United States are far more advantageously circumstanced for gaining information of existing conditions of weather. Their central office is situated to the eastward, and their telegraphic system embraces an area more comprehensive, as regards extent of territory and variety of climatological conditions, than that which is at the service of any meteorological organization in Europe. In the States, therefore, it is possible to learn what is actually occurring over a very extensive area, and thereby to announce the coming weather with a very considerable amount of certainty; but even in America meteorologists have not yet arrived at any important conclusions as to the causes to which storms owe their origin, or to which the rise and fall of the barometer are due. As to this latter point, an influential section of meteorologists attribute all our storms to the condensation of vapour, but they do not explain how it comes to pass that the rain comes with the storm instead of preceding it, and that our heaviest rain deluges are not necessarily accompanied by any cyclonic disturbance at all.
Again, as to the very nature of the storms themselves we have much to learn. The doctrine may be thought heretical to propound, but we do not know accurately what is the real shape of any of these great systems of atmospherical disturbance.
Even the very West India hurricanes and the typhoons of the Indian seas, on which Redfield, Reid, and Piddington based the Law of Storms, are certainly not simple revolving discs of air moving over the earth's surface like a boy's top over the floor. If this were the case, the effect of the motion of translation must be perceptible in a contrast between the violence of the wind on opposite sides of the storm.
Taking a West India hurricane, which advances from south-east to north-west, while crossing the latitades of the North-east trade,
if we assume its rate of progression to be 15 miles an hour, an ordinary speed in that region to which I am alluding, and the average hourly velocity of the wind in the system to be 60 miles, a moment's reflection must show us that the S.E. wind must be felt at a fixed observatory as blowing at the rate of 75 miles an hour, while the N.W. wind cannot attain a higher velocity than 45 miles. Such a contrast as that described between the two sides of a cyclone must surely have attracted notice in the West Indies or the Bay of Bengal, if it really existed.
As for our own storms, which, when they assume the complete cyclonic form, differ only in degree from tropical hurricanes, the contrast would be much more striking, for these systems travel very rapidly, occasionally moving at the rate of 40 miles an hour and even more. When these systems move from west to east the westerly winds should have their velocity increased by 40, while that of the easterly winds should be reduced by an equal amount.
On this hypothesis therefore it would be all but impossible for an easterly wind ever to attain the force of a gale, and yet the eastern coasts of Great Britain can bear testimony to the violence with which that wind, somehow or other, manages to blow.
Moreover, the contrast which always exists in these latitudes between the southerly and south-westerly gales, which blow with a high temperature and falling thermometer, and the north-westerly gales in the rear of the storm which come down upon us, causing the barometer to leap up as much as four-tenths of an inch in two hours (as on March 12, 1876), and the barometer to drop instantly, and bringing on hail-squalls, and their concomitant thunderstorms, is such that it is inconceivable that the actual particles of the air coming from N.W. can have formed a portion of the warm southwest current a few hours before.
It appears absolutely necessary to assume that the air in a storm should be drawn from two independent sources at least, and it seems evident that these storm eddies must be in a continual state of dissolution and reconstruction as the system moves over the surface of the earth.
Again, as to their very character and shape we are far from being quite clear. Distinctly circular eddies they are not. There
is not a single cyclone on record of which the truly circular shape has been proved by a reasonable number (say 50) of strictly simultaneous observations uniformly distributed all round the centre. It is not sufficient to take the wind observations from half-a-dozen logs, and to say that, if these do not accord with the circular theory, the observations themselves must be faulty.
As to our own storms, the overwhelming majority of them have, so to speak, no northern sides at all, as far as Europe is concerned, and the only winds developed in them are those from S.E. veering to N.W., a change which is at times manifested within the space of a few hours over the whole extent of the British Isles from Shetland to Scilly.
There can be no doubt that the character, track, and shape of all storms is in close connection with the general distribution of atmospherical pressure in their neighbourhood, and as far as we are concerned, as the region about Iceland is one of great deficiency of pressure in winter, it is exceedingly seldom at that season that the barometer can rise high enough, say at the Faroes, to produce a gradient of sufficient steepness to give rise to strong easterly winds in the North of Scotland. The prevalent disposition of pressure in our storms is that of a trough sloping northwards, and open on its northern side.
It is for the mathematicians to explain how such disturbances can be formed in the atmosphere; the evidence of the charts published day by day amply proves that they exist.
Again, as to the motion of the air in a cyclone. The older writers always assumed when facing the wind the centre of the hurricane bore exactly eight points away to the right in the northern, and to the left in the southern hemisphere. This rule is very simple and easily remembered, but unfortunately it is not confirmed by recent investigations. Not only did Redfield and Piddington suspect that the motion was vorticose and spiral, but there is not a single independent enquirer into the subject of late years who is not an adherent of that view, while the researches of Meldrum, Toynbee, and, within the last year, of Eliot, have proved its truth. The very last paragraph of Eliot's masterly report on the Madras cyclone of May, 1877, runs as follows: