living among us, Mr. John Ball, made the following proposal before the British Association at Swansea :

Now that we have the means of receiving information in an indefinitely short space of time by the electric telegraph, these problems, under favourable circumstances, may be studied à priori. In London we may receive instantaneous intelligence of the condition of the atmosphere, as to the five above-mentioned elements [heat, pressure, moisture, and the velocity and direction of the wind] from nearly all the extremities of Great Britain. With a delay of about four hours we can have similar intelligence from the western part of Ireland, and with a still shorter delay our communications may extend to the centre of France, the banks of the Rhine, and even to the frontiers of Hungary and Poland.

'I do not pretend to say that with such elements for calculation we should at once be enabled to predict changes in the weather with absolute certainty. It would require some time to eliminate the action of accidental and local causes at particular stations; but there is no reason to doubt that in a short time the determinations thus arrived at would possess a high degree of probability. The ordinary rate at which atmospheric disturbances are propagated does not seem to exceed twenty miles per hour; so that with a circle of stations extending about 500 miles in each direction, we should in almost all cases be enabled to calculate on the state of the weather for twentyfour hours in advance.'-Report of the British Association, 1848.

In the Smithsonian Reports for 1850, the idea was suggested of wall maps exhibiting simultaneous weather observations over a large tract of 'country, and in February, 1855, Leverrier exhibited such charts to the Paris Academy; but the time for publication had not yet arrived. Among the earliest attempts to carry this idea of wall maps into practice were the charts which were exhibited at the Crystal Palace soon after 1860, and were continued at the principal stations of the Electric Telegraph Company. These were large maps furnished with arrows at several stations, to show the direction of the wind, which were set daily according to the telegraphic reports. The essential defect of this system was that there was only one map at each station. The record was therefore ephemeral, inasmuch as, to make the chart of any day, you had to obliterate the record of the day before, so that no comparison of successive phases of weather was immediately presented to the eye. In September, 1863, Leverrier issued the first chart with his Bulletin International;' and at an earlier period of the same year Francis Galton published his 'Meteorographica,' in which, by means of coloured symbols, he brought into prominence the idea that different types of weather were associated with winds from different directions, and that these winds again derived their characteristics

teristics from the circumstances to which they owed their origin. Mr. Galton was the first to propose the term 'anticyclone' for a region of relatively excessive pressure, with its concomitant wind system; the converse term 'cyclone' having already been devised by Piddington for a region of relatively deficient pressure, such as a hurricane or typhoon, with its wind system. The air flows round and out with watch hands from an anticyclone, and is essentially dry; while it flows in the opposite direction against watch hands, round and in upon a cyclone, and is essentially damp. All these principles are implicitly contained in Mr. Galton's types of different phases of weather, and we may almost say that, had Admiral FitzRoy recognized in all their fulness the pregnant ideas sketched out in Meteorographica, he would have anticipated almost all the so-called discoveries of the last fifteen years.

The one abiding principle which underlies all weather prediction at present is, that the wind in the Northern Hemisphere, except where modified by local influences of hill and valley, always flows so as to keep the lowest barometrical readings on its left-hand side. This rule is known as Buys Ballot's Law. It was propounded by the Dutch Professor in 1858, and is recognized throughout Mr. Galton's treatise; but it must be admitted that, in an almost forgotten paper in Poggendorf's Annalen for 1853, Adolph Erman had anticipated the discovery by at least five years.

The first attempt to publish a weather chart was apparently a project started about the year 1862, for a daily weather-map, to be issued to subscribers at 2s. per month. The publishing office was to be the present office of the 'Globe.' In this map symbols were to be employed to indicate the motion of the barometer, the direction and force of the wind, and the character of the weather. The symbols were engraved on blocks inserted at definite points on the map. We have not heard that the project ever advanced beyond the stage of a first proof, a copy of which is now before us. A similar idea for the production of a chart was finally carried out for the 'Shipping and Mercantile Gazette' by Captain Chas. Chapman, at the instance of the late Sir W. Mitchell; and from the 1st of January, 1871, the Gazette' has contained a daily chart, which, however, for the first few years of its appearance, only exhibited the conditions of the wind. On the 1st of April, 1875, the Times' published the first of the small charts now familiar to us, which have done so much to popularize the subject of meteorology.

We must now turn to the subject of the practical usefulness of weather telegraphy, with reference, in the first instance, to its

utility

utility for seamen. It was for their benefit that Leverrier projected storm warnings in 1855, and it is mainly in their interest that the Maritime Powers of Europe spend considerable sums on meteorology every year. Our own Meteorological Office sends out warnings to 129 stations on our coasts, as we gather from its last-published report; and it is the only office which has for several years published a comparison between its warnings and the weather subsequently experienced. The following is the tabular analysis of percentage results for the past five years :

The averages of success and failure do not make up the full total of 100, the small residue (4·5), consisting of failures, for which the office can bring forward excuses, such as the inaccuracy or lateness of telegraphic reports from the stations. It is, however, clear that about half the warnings are followed by serious storms; and about a quarter more are followed by strong winds, that is, by a sufficient disturbance of weather to interfere with the comfort of yachtsmen, if not with the possibility of profitable fishing. About one quarter of the warnings con tinue to be distinctly wrong; while there is every year an ugly list of storms not foreseen, owing mainly to deficiency or incorrectness of telegraphic information.

When we analyse these figures dispassionately, we find that we have grounds for fair congratulation in comparing the results attained with the means available for their attainment. We have already stated that meteorologists can only boast of the one new principle, Buys Ballot's Law. This principle does not contain the smallest lurking germ of prophecy. The storm must show itself by the disturbance of barometrical readings, before meteorologists can recognize its existence and issue their warnings.

Now these islands are entirely unprotected on the north

and

and west, and it is on our oceanic seaboard that storms for the most part make their first appearance. When once their presence is recognized, it is an easy matter to announce the fact to our own seaports and to foreign countries; but the idea of successfully warning the captain of a sea-going ship of the storms he is likely to meet before he is well out of sight of land, and before his crew have settled down in their places, is apparently as yet unattainable, at least to the persons at the head of the respective meteorological systems of Europe.

A spirited attempt has been made by Mr. J. G. Bennett, of the New York Herald,' to announce the approach of storms to Europe, and these telegrams have been sent at brief intervals for the space of about two years and a half. From the brilliancy of the attempt, and perhaps also from its utter audacity, these messages, affording as they do no indication of the grounds on which they are based, have at once taken the public fancyalways ready to hold that one successful hit condones a host of failures; but in practical results for the fishing and farming interests of Western Europe their value is absolutely nil. A warning which, like many of the 'Herald' messages, covers the entire coast of Europe for nearly twenty degrees of latitude, from Bodö in Norway to Corunna, must necessarily be superfluous over a large proportion of the ports where it is announced. What possible use can it be to a Wick herring-boat to know that a storm is coming which at its worst may only discommode the bathers at Biarritz, or the 'Jagts' from the Loffoden fishings on their way southwards to Bergen? On a recent occasion within our own knowledge, a single East-coast skipper dared to disregard a 'Herald' warning which kept the rest of the Scilly mackerel fleet in St. Mary's Roads, and he literally netted 1501. for one night's haul; the threatened storm having died out in mid-Atlantic. What is wanted in Europe is a station, or a series of stations, in the mid-Atlantic between the Azores and Iceland. The proposal has been made over and over again, that telegraphic communication should be kept up with ships, or even with large buoys, moored in deep water, but at present no possible solution of the problem has been propounded. It is an old maxim with all our Lighthouse Boards, never to post a light-ship or build a lighthouse where there is not a reasonable probability of a weekly visit in any ordinary weather for the supply of provisions or relief. Those who have read Basil Hall's graphic description, in his 'Fragments of Voyages and Travels,' of his landing on Rockall from H.M.S. 'Endymion,' and his difficulty, first in finding the rock, and next in finding his ship again, will see that the chance of picking up a ship at

anchor

anchor at some distance from the land, and in thick weather, is poor indeed.

We see therefore that there is not much prospect of our obtaining oceanic stations at present, and, until we have gained a much more thorough experience of the phases of Atlantic weather than our meteorologists show that they possess, we must content ourselves with a realization of Leverrier's programme of twenty years ago, which we have already quoted.

When we say that we have to wait until a storm shows itself on our coasts, there is, of course, room for a difference of opinion as to what is meant by the term 'shows itself.' It may be that considerable progress may be made in the delicacy with which our weather doctors feel the pulse of the atmosphere, but at present the most promising line of enquiry in this direction is that pointed out by the Rev. Clement Ley in this country, and by Professor Hildebrandsson in Sweden-the observation of the upper stratum of clouds, especially of those called by Luke Howard Cirrus.' There is no doubt that a careful notation of the motion of these feathery films, which cover the sky hours and even days before the wind begins to change, would be most useful, but the difficulty is to secure trustworthy observers. The phenomena are comparatively seldom visible in this cloudy climate, and when they are, they demand great judgment and a considerable degree of patient attention on the part of the observer. Any one who studies Hildebrandsson's or Ley's charts of upper-cloud observations for any single day, will see that a dozen observers over the whole of Europe is almost the maximum ever obtained.

Such observations as these are essentially personal, and the art of making them is nearly incommunicable. They have something of the same character as the observations by which shepherds and fishermen can and do make for their own districts forecasts which sometimes put to shame the boldest efforts of professed meteorologists.

We have hitherto been speaking principally with reference to storm-warnings for seamen, but the dependence of farmers on weather is to the full as immediate as that of sailors, and it will be interesting to sketch out briefly some of the principal systems which have been organized for conveying information to farmers and other persons interested in agriculture.

The first serious move in this direction was in the interest of the public, more than of the farmers. In May, 1872, Commodore Maury, of St. Louis, suggested a system of international weather reports to enable the corn-merchants and the public in general to form some opinion as to the probable yield of the

harvest