THESE reports contain matter of very considerable general interest, showing, as they do, what has been and is being done to perfect the system of weather forecasts in this our ever-changing climate, as well as to compile a register of the winds and weather that may be expected at sea in different parts of the world, including the tracing and laying down the courses of cyclonic disturbances in the Atlantic and other oceans, together with the probable distance travelled whilst they last, and their rate of travelling.

This last subject, so far as regards the North Atlantic Ocean, comes first in the Report for 1889, and merits a perusal in order to learn the outline of the method adopted to obtain this very desirable end.

The one source from which all information can be obtained as to what wind and weather exists, or may be expected to be found, in any spot at sea, must be the logs and other records of ships whilst on that spot. To extract this from a large number of logs, and to carefully compare and connect the various items, necessarily entails a large amount of very correct and exact labour, which is demonstrated by the fact that the results for the thirteen months ended September 3rd, 1883, were not published until 1890, and even now the complicated series of phenomena contained in these charts is undergoing further investigation. The history of all the clearly-marked barometrical depressions or cyclonic systems that occurred between August 1882 and 1883 inclusive, has been investigated and their course and distance travelled, etc., laid down on charts with what is considered justifiable accuracy.

These systems or whirlwinds, of greater or less extent and force, appear to travel at an average rate of 400 to 450 miles a day, and 273 of them have been traced. Of these, 143 had blown themselves out and disappeared in less than 5 days. Very few lasted 10 days, and of these again a very small proportion lasted 13 days. Of 63 of these depressions that appeared between the 40th and 35th parallels of latitude, only 15 crossed the Atlantic to the Meridian of Greenwich ; some of these again, passed it in high latitudes. But of those that originated south of latitude 35°, half possessed sufficient energy to cross the ocean. The cyclones which existed more than 13 days occurred in September, October, and November. The attempt to trace the anti-cyclonic system has not however yet been equally successful.

Charts have been prepared and issued, showing the barometrical pressure in the Atlantic, Pacific and Indian Oceans during the months of February, May, August and November. Further charts on the subject are now ready; and it is hoped that the information they give will prove of great value to seamen. Current charts for the same parts are also being prepared, with which considerable progress was made during the year.

"We now come to the weather forecasts.

The preparation and issue of these, including the storm warnings, is founded on the contents of reports received by telegraph from a large number of Stations in the British Isles, supplemented by a considerable number from the Continent of Europe. Those from America are not now used for this purpose, the conclusion having been come to that the information given by them was as yet of no practical utility in forwarding the object in view. The reports are made out for three different hours, viz., 8 A.M., 2 P.M., and 6 P.M. ; 58 are sent in in the morning, 17 at 2 o'clock, and 29 at 8 P.M. Besides the actual reading of the barometer and thermometer at the hours named, they give the alterations which have occurred in both during the previous 24 hours, the direction and force of the wind, and the state of the weather, together with any changes of importance that may have taken place during the previous day. These are all entered on a chart, and forecasts of the weather for the following 24 hours are deduced from them, and issued ia the well-known form for the eleven districts into which the country is divided for this purpose. The charts obtained from the 8 A.M. reports are the fullest and most important on account of the larger number made at that hour.

The forecasts from these are issued about 11 o'clock, and foretell the weather from noon until the following noon. The 6 P.M. reports are issued about 9 o'clock, and are for the day, commencing at 8 AM. the next morning.

The 2 P.M. reports, of which the forecasts are completed by about half-past three, are used for storm warnings only, excepting during the hay season. Various methods are adopted for making public the forecasts. They are exhibited at the Mansion-house, Lloyd's Booms, the libraries of the Houses of Lords and Commons, and at seven or eight other different places in London. They are also supplied to the newspapers, the Government Offices, and to seaports for public exhibition, to correspondents with the Office, to foreign Meteorological establishments, and about two hundred copies are supplied to various subscribers. Any one applying at the Meteorological Office between 11 A.M. and 8 P.M. on week days, and 7 P.M. and 8 P.M. on Sundays, can obtain in writing the latest information, and the latest forecast for any particular district, on payment of one shilling for each enquiry. Application may also be made by either letter or telegram, in the former case the fee and postage for the reply must accompany the application, and in the latter, the fee must be paid at the telegraph office when despatching the message with a prepaid reply.

By an arrangement between the Meteorological Office and the Royal Agricultural Society, the Royal Dublin Society and the Highland and Agricultural Society, forecasts are sent gratis and daily to about thirty observers, selected by these Societies, at various places, during the hay season, on the two conditions that the information shall be made as widely known as possible and that A record of the weather actually experienced be sent to the Office. The issue of these commenced on the 11th of June, and was carried on into August for the later districts.

A table compiled from the reports of the different observers of the weather experienced, shows the remarkable accuracy of these particular forecasts, and the service they must have been to all they reached having hay at stake. The results are classified in the tables, under four heads: 1. Complete success; 2. Partial success; 3. Partial failure; 4. Total failure. In the year 1888 the percentage of each was No. 1, 49; No. 2, 35; No. 3, 11; No. 4, 5; so that 84 per cent, of the forecasts were correct, or partially so, whilst they only failed partially in eleven, and totally in five per cent. In the year 1889 these predictions were even more successful, the figures being 57, 32, 9 and 2, respectively, or 89 per cent, of success and partial success, against 9 of partial and only 2 of complete failure. The general correctness, and therefore usefulness, of these special forecasts is testified to by several of the observers from different parts of the country who had undertaken the task of receiving them, also by the fact of nine large landowners in different parts of the country having applied to be furnished with them at their own expense.

The Council expressed their readiness to co-operate in any way in their power for increasing the benefits of this movement, the utility of which had been sufficiently established; but the Agricultural Department of the Privy Council replied that they had no funds at their disposal, and that the Treasury had declined to make a grant for this purpose. Nothing further therefore, has been done to extend the publication of this very useful information.

The general forecasts issued are carefully watched, and their accuracy checked by the -weather actually experienced in the several periods. The result is tabulated in the same way as above described for the hay harvest, and for 1888-89 shows the percentage of success and partial success respecting wind to have been 81 • 4, and for weather 83; for the next year i.e., 1889-90, the results are 80-25, and 80-75 respectively. These percentages have varied so little for some years, that it looks as if they represent the amount of accuracy obtainable under present knowledge and conditions. One great disadvantage the Council works under is the impossibility of knowing what is occurring in the Atlantic until the disturbances reach our own western coasts.

Whenever the preparation of the forecasts shows the existence of an atmospheric disturbance which will probably cause a gale of wind to blow on any parts of the coast, those parts are at once warned by telegraph, and the well-known storm signal, a black canvas cone with the point upwards or downwards, according to the expected direction of the wind, is hoisted at the appointed place at the stations warned. These signals are displayed during daylight for forty-eight hours after the telegrams are despatched, unless previously countermanded. There are 72 of these stations in England, 16 in Wales, 44 in Scotland, 15 in Ireland, 3 in the Isle of Man, and 3 in the Channel Islands.

In the year 1888-89 the warnings were justified by complete or partial correctness to the extent of 83'9 per cent, were not verified in 14-3 per cent., and were given too late in 1 • 8 cases per cent.; in 1889-90 the respective figures are 81 • 2,16 • 9, and 1 • 9.

During the years 1888-89 four gales of wind visited different parts of the coast, for which no warnings were issued, and during 1889-90 five occurred under the same conditions. The four in 1888 appear to have evaded notice in most cases, by advancing very rapidly to our coasts between the hours of observation. In one ease at 6 P.M., the time of the evening reports, there was no indication whatever of an impending storm on the coast; by 10 P.M. a shallow depression appeared on the S.W. coast of England, which grew deeper with extraordinary rapidity. At 8 A.M. the next morning, the centre was over Pembroke, and at six o'clock the same evening it had reached Lincolnshire.

Of the five in 1889-90, two which had apparently passed or were passing our islands for the coast of Norway, suddenly turned to the southward, and brought the northern coasts of Scotland and England within their range.

The report shows that the work of the Meteorological Office does not end with the subjects referred to in this short notice but embraces other labours for studying the phenomena of our climate for the advance of science, which we have not space to refer to in this month's Journal.