584 TKANSACTIONS OF SECTIO.N A. 



such and such voluniL-s of data and mean values measure the contribution to the 

 stock of knowledge leaves me rather cold and unimpressed. 



But to return to the endeavour after the delicate adjustment between 

 speculation and routine, which will reduce the waste of such an institution to a 

 minimum ; experience very soon teaches certain rules. 



I have said elsewhere that the peculiarity of meteorological work is that an 

 investigator is always dependent upon other people's observations ; his own are only 

 applicable in so far as they are compared with those of others. Up to the present 

 time, I have never known anyone take up an investigation that involved a re- 

 ference to accumulated data, without his being hampered and harassed by uncer- 

 tainties that might have been resolved if they had been taken in time. 1 shall 

 give you an example presently, but, in the meantime, experience of that kind is so 

 universal that it has now become with us a primary rule tliat any data collected 

 shall forthwith be critically examined and so far dealt with as to make sure that 

 they are available for scientific purposes — that is, for the purposes of comparison. 

 A second rule is that as public evidence of the completion of this most important 

 task there shall be at least a line of summary in a published report, or a point on 

 a published map, as a primary representation of the results. Such publication is 

 not to be regarded as the ultimate application of the observations, but it is evidence 

 that the observations are there, and are ready for use. 



You will find, if you inquire, that at the Office we have been gradually lining 

 up these troops of meteorological data into due order, with all their buttons on, 

 until, from the commencement of this year, anyone who wishes to do so can hold 

 a general review of the whole meteorological army, in printed order — -first order 

 stations, second order stations, rainfall stations, sunshine and wind stations, seii 

 temperatures and other marine observations — on his own study table, within six 

 months of the date of the observations, upon paying to his Majesty's Stationery 

 Office the modest sum of five shillings and sixpence. For all the publications 

 except one the interval between observation and publication is only six weeks, 

 and as that one has overtaken four years of arrears within the last four years, 

 I trust that by the end of this year six weeks will be the full measure of the 

 interval between observation and publication in all departments. This satisfac- 

 tory state of affairs you owe to the indefatigable care and skill of Captain 

 Hepworth, Mr. Lempfert, and Mr. 11. H. Curtis, aud the members of the staff of 

 the Office who work under their superintendence. I need say little abDut cor- 

 responding work in connection with the Daily Weather Report, in which Mr. 

 Brodie is my chief assistant, although it has received and is receiving a great deal 

 of attention. The promptitude with which the daily work is dealt with hardly 

 needs remark from me, though I know the difficulties of it as well as anyone. If 

 I spend only one long sentence in mentioning that on July 1, 1908, the morning 

 hour of observation at twenty-seven out of the full number of twenty-nine stations 

 in the British Isles was changed from 8 a.m. to 7 a.m., aud the corresponding 

 post-offices, as well as the Meteorological Office, opened at 7.15 a.m., so that we 

 may have a strictly synchronous international system for Western and Central 

 Europe, and thus realise the aspiration of many years, you will not misunderstand 

 me to mean that I estimate the task as an easy one. 



The third general rule is that the effectiveness of the data of all kinds, thus 

 collected and oi-dered, should be tested by the prosecution of some inquiry which 

 makes use of them in summary or in detail. It is here that the stimulating force 

 of speculative inquiry comes in ; and it is in the selection and prosecution of 

 these inquiries, which test not oulj' the adequacy and effectiveness of the data 

 collected but also tlie efficiency of the Office as contributing to the advance of 

 knowledge, that the most serious responsibility falls upon the administrators of 

 Parliamentary funds. 



Scientific Shylocks are not the least exacting of the tribe, and there have been 

 times when I have thought I caught the rumination : — 



Shy. Three thousand ducats ? 'tis a good round sum ! 

 Bas, For the which, as I told you, Antonio shall be bound, 



