1883.] on Weather Knowledge in 1883. 329 



the wliole agrees sufficiently closely with that determined by the 

 office from its own data. 



I find, on enquiry, from all the European offices which issue 

 forecasts, that the percentages of success which they claim officially 

 are almost identical with that shown on the diagrams. They are on 

 the whole about 80, but no one is really contented with the results. 

 The critics of foreign forecasts are just as severe on their own systems 

 as the writers of newspaper letters are on us over here, and as one of 

 my correspondents, the gentleman who manages the Magdeburg office 

 remarks, those who are least content are the forecasters themselves, 

 thoucrh naturally they do not publish their dissatisfaction. 



To give my audience one illustration out of many of the difficulty 

 of our task in England, the same gentleman whom I have just quoted 

 says : " Our greatest trouble is the lateness of arrival of the English 

 telegrams, and, without news from England, no one in Germany can 

 dream of forecasting." Now we, in these islands, as is so often said, 

 can apparently never hope for daily telegraphic news from signal 

 ships in the Atlantic, so that ocean must keep its wonted silence, and 

 our forecasting must be even more hazardous than that of our German 

 Qcighbours. 



Another development of forecasting, and in fact its earliest form, 

 the desire to carry out which gave rise to the whole system of 

 weather telegraphy, and as a result of the latter, to the science of 

 Modern Meteorology, is the issue of Storm Warnings. These were 

 instituted in this country by FitzEoy, in 1866. They were tempo- 

 rarily suspended at Christmas 1866 and resumed a year later, and the 

 diagram gives the average results for the entire United Kingdom, 

 from the year 1870, being the first year for which the figures were 

 regularly submitted to Parliament. The interval of twelve years 

 divides itself naturally into two periods of equal length, the break 

 between which happens to coincide with the institution of an evening, 

 in addition to the afternoon service of reports. 



This was carried on, in the first instance, at the sole cost of the 

 Times newspaper, and in order to furnish materials for di'awing 

 the chart in its morning issue, and it is only within the last three years 

 that the expense of this part of our work has been borne by the 

 Government. 



The figures show at the first glance an improvement of 8 per cent, 

 in the first column, and this is a considerable advance, but my 

 ludience must not take away the idea that if the receipt of four hours' 

 later intelligence raises the percentage of success, we might anticipate 

 the possibility of warning the coasts for all storms, if the reporting 

 hours were extended to midnight, or the service were even con- 

 tinuous. 



The extension of the reporting service would enable us in London 

 CO know what was taking place on the coasts far better than we do, 

 but we could not impart our knowledge to those whom we wish to 

 benefit. Practically, storm warnings must be issued before sundown. 



