374 Royal Society : — 



their intercoraparability ascertained, bj' a competent and responsible 

 Authority ; — and provided that no observations be used but those in 

 which careful attention shall have been given to the precautions 

 which it will be necessary to adopt, for the purjjose of obtaining the 

 correct knowledge of the temperature of the external air, amidst the 

 many disturbing influences from heat and moisture so difficult to 

 escape on board ship. In this respect additional precautions must 

 be used if night observalions are to be required, since the ordinary 

 difficulties are necessarily much enhanced by the employment of arti- 

 ficial hght. Amongst the instructions which will be required, perhaps 

 there will be none which will need to be more carefully drawn, than 

 those for obtaining the correct temperature of the external air under 

 the continually varying circumstances that present themselves on 

 board ship. 



In regard to land stations. Professor Dove's tables have shown that 

 data are still pressingly required from the British North American 

 possessions intermediate between the stations of theArctic Expeditions 

 and those of the United States ; and that the deficiency extends 

 across the whole North American Continent in those latitudes from 

 the Atlantic to the Pacific. Professor Dove has also indicated as 

 desiderata, observations at the British Military stations in the 

 Mediterranean (Gibraltar, Malta and Corfu), and around the Coasts 

 of Australia and New Zealand : also that hourly observations, 

 continued for at least one year, are particularly required at some 

 one station in the West Indies, to supply the diurnal corrections for 

 existing observations. 



Whilst the study of the distribution of heat at the surface of the 

 globe has thus been making progress, in respect to the mean annual 

 temperature in difl^erent places, and to its periodical variations in 

 tiiifereut parts of the j'ear at the same place, the attention of phj"-- 

 sical geographers has recently been directed (and Avith great promise 

 of important results to the material interests of men as well as to 

 general science) to the causes of those fluctuations in the tempera- 

 ture, or departures from its mean or normal state at the same place 

 and at the same period of the j^ear, which have received the name 

 of "non-periodic variations." It is known that these frequently 

 affect extensive portions of the globe at the same time ; and are 

 generally, if not always, accompanied bj'^ a fluctuation of an opposite 

 character, prevailing at the same time in some adjoining but distant 

 region ; so that by the comparison of synchronous observations a 

 progression is traceable, from a locality of maximum increased heat 

 in one region, to one of maximum diminished heat in another 

 region. For the elucidation of the non-periodic valuations even 

 monthly means are insufficient ; and the necessity has been felt 

 of computing the mean temperatures for periods of much shorter 

 duration. The Meteorological Institutions of those of the European 

 States which have taken the foremost part in the prosecution of 

 meteorologj% have in consequence adopted five-day means, as the 

 most suitable intermediate gradation between daily and monthly 

 means : and as an evidence of the conviction which is entertained 



