Magnetic Variations at different Places^ $-c. 193 



sponding phenomena afc other places were in these cases similarly 

 definite. It thus appeared that a systematic comparison of the times 

 of these initial movements at different places might afford the means 

 of determining how nearly the movements at such places are really 

 coincident. 



It is practically impossible to catch these first movements by eye 

 observation of the magnets, from the uncertainty as to the time at 

 which they may occur, so that it becomes necessary to rely on the 

 photographic register. Unfortunately, however, the scale of time 

 employed on such registers is necessarily small, one minute of time 

 corresponding usually to about O'Ol inch. This renders difficult any 

 very accurate measurement of time in individual cases. But it 

 appeared to me that by selecting a considerable number of instances 

 of sudden initial movement, and comparing together the times as 

 measured from the photographic registers of magnetic observatories 

 situated in widely different regions of the earth, we might expect in 

 the general average to eliminate any accidental error of measure- 

 ment due to contracted time scale or other cause, and, supposing no 

 systematic error to exist, so obtain some reliable information as to 

 the degree in which, on the whole, such movements, at different 

 places, are or are not simultaneous. 



With this object in view, a selection from the Greenwich registers 

 was made of seventeen cases of sudden distinctive movement, pre- 

 ceding or commencing magnetic disturbance, and occurring during 

 the years 1882 to 1889. The list might have been extended, but the 

 cases selected are typical, and seemed sufficient for the purpose of a 

 preliminary inquiry. Giving the day and approximate hour only of 

 the occurrence of the respective movements at Greenwich, the 

 directors of various magnetic observatories were asked to examine 

 their registers for the corresponding times, and see whether there 

 existed any instance of sudden movement, and, if so, to measure out 

 and supply on each day, and for each magnetic register, the exact 

 local time of commencement of movement to the nearest minute of 

 time, and also the direction and approximate amount of change in 

 each magnetic element. The observatories selected, including Green- 

 wich, nine in number, were, counting from west to east, those of 

 Toronto, Greenwich, Pola, Pawlowsk, Mauritius, Bombay, Batavin, 

 Zi-ka-wei, and Melbourne. This request was complied with in the 

 most obliging manner, and my best thanks are due to the directors 

 of the various establishments for the very kind way in which they 

 endeavoured to supply all necessary information, as well as to the 

 Astronomer Royal for permission to make use of the Greenwich 

 records. In two cases, indeed, Toronto and Zi-ka-wei, copies of the 

 registers were also furnished ; the Toronto report, however, included 

 movements of the declination magnet only. Unfortunately at Pola 



o 2 



