8 



month 24 land observations have been charted. The smallest number of observations for a single day is 19 

 marine for 9th October, 1903, 10 land throughout October, 1901, and 10 to 11 throughout March, 1904. 



In the Eegisters issued by the Meteorological Office, observers were requested to give a few readings of 

 their barometer in ports visited in order that the error of the instrument might be estimated. Facilities 

 were given for recording these readings in the Register by the introduction of a special form. 



Some of the observations received from British ships were recorded by trained observers with properly 

 verified instruments provided by the Office; the errors of these instruments were known ; but the majority 

 of instrumental observations were derived from barometers and thermometers supplied by the owners of 

 the vessels in which the observers were serving. The errors of these instruments were ascertained, when 

 possible, by comparison with standards at various ports, and the corrections registered, verified or 

 otherwise, by the comparison of readings noted in the special form of the Register with the corresponding 

 values published in the Daily Weather Reports of the country in which the observations were taken. 



Entries in the Registers of doubtful accuracy have been discarded, and it must be admitted that a small 

 percentage of the wind observations charted are not in good agreement with the distribution of barometric 

 pressure to which they are related, particularly as regards direction. The discrepancies may, in most 

 cases, be attributed to the failure of the observer to realise, or to make sufficient allowance for, the effects 

 of aberration caused by the vessel's own motion. 



THE DAILY AND MONTHLY CHARTS. 



The result of the work is represented by daily charts for Noon Greenwich Mean Time with charts for 

 monthly pressure and air temperature. 



In some respects the Daily Synchronous Charts are unique. The conditions which they are intended to 

 represent at a given time each day, for a period of thirty months, refer to an area that is far larger than 

 that embraced by any similar set of charts hitherto published. They include localities in the Antarctic, 

 and furnish daily a link between the conditions existing simultaneously in frigid zones and in sub-tropical 

 latitudes, represented on the one hand by three widely separated localities in Antarctica, and on the other 

 by the southern segments of the tropical anticyclones of the Southern Hemisphere. They establish, as a 

 fact, the permanence of these high-pressure areas of the great oceans, and show the gradual seasonal 

 migrations of the South Atlantic, South Pacific, and Indian Ocean anticyclones by the slow oscillations 

 of their southern edges. 



Notwithstanding the large number of observations collated, the localities for which data are available 

 for each daily chart are comparatively few, and frequently isolated, owing to the vastness of the area 

 to which collectively they are related. Despite the sparseness of the observations charted, and the 

 obvious incompleteness in sequence of the conditions which the charts are designed to represent, the 

 information they afford is considerable, for not only are the positions of many high- and low-pressure 

 systems over various parts of the regions under notice indicated, but these " Highs " and " Lows " can, 

 for the most part, be identified day after day, the means being thus afforded for tracing the directions in 

 which their centres moved during more or less prolonged periods. 



The limits of areas over which cyclonic depressions exercised an influence can, moreover, in a few 

 instances be defined and occasionally the places of their origin or extinction roughly determined. 



As regards the charts of monthly pressure and air temperature at sea-level for each of the thirty months 

 under notice, although the number of observations upon which they are based is comparatively small, even 

 for those months in which the more ample data are available, it may nevertheless be claimed that they 

 exhibit broadly the more salient features in the distribution of the respective elements they represent 

 during the periods to which they refer. In addition, the charts of mean monthly pressure and air 

 temperature side by side with those of normal pressure and air temperature supply the means for 

 comparing th distribution of these elements, in the several months, with the average distribution. 



The tables of monthly average wind and gale frequency introduced assist in the interpretation of the 

 mean monthly chart. 



