Chemistry and- Physics. 425 



may be a sharp, premonitory, simultaneous movement, followed 

 after a time by general disturbance, or the movement may at 

 once usher in the disturbance. These initial movements are not 

 always great in magnitude, sometimes, indeed, small, but they 

 have a very definite character, and frequently occur nearly instan- 

 taneously, as is shown by the character of the photographic traces. 



It has been long known that magnetic disturbances occur at the 

 same time over wide areas of the earth's surface, but the accidental 

 comparison in past years of the times of commencement of one or 

 two disturbances at Greenwich with the times at other places has 

 led "the author to suppose that the coincidence in time is much 

 closer than had been before supposed, and the definite, and on 

 occasions isolated, character of the initial movement induced him 

 to undertake the collection and comparison of the times of such 

 movements for a number of days at observatories geographically 

 widely separated. 



The times of such movements cannot be caught by eye observa- 

 tion without continuous watching of the magnets, so that the 

 photographic registers have to be relied upon, which is better, 

 excepting that the scale of time is necessarily contracted ; but, 

 though in individual measures there might be variations, it was 

 conceived that (supposing no systematic error to exist) the mean 

 of a number of comparisons should give a good result. Seventeen 

 days occurring in the years 1882 to 1889 were selected for com- 

 parison, the observatories being those of Toronto, Greenwich, 

 Pawlowsk, Mauritius, Bombay, Batavia, Zi-ka-wei, and Melbourne, 

 and, for a less number of days, Cape Horn (as obtained from the 

 Mission Scientifique du Cap Horn, 1882-83.) It was desired to 

 have times for Pola, but it was found that photographic registers 

 during great part of the period did not exist. The variation in 

 time at each place from the mean of times for all places is given 

 for each day. The mean deviation at the different places varies 

 from + 2 *4 minutes to — 2*9 minutes, the agreement between four 

 of the places, Greenwich, Pawlowsk, Mauritius, and Bombay, 

 being very much closer, the mean values of deviation for Green- 

 wich, Pawlowsk, and Bombay differing, indeed, by only 0*1 

 minute, equivalent to 6 seconds. 



The question arises, Are the differences real, or due (consider- 

 ing the contracted time scale) to accidental error ? If the mag- 

 netic impulse is really simultaneous over the whole earth, it is a 

 striking physical fact, and if not entirely so, the circumstance is 

 no less interesting; but greater attention to accuracy of time scale, 

 or a more extended scale, may be necessary before the point in 

 question can be definitely settled. 



A table is added, showing the character of the magnetic move- 

 ment at the several observatories, from which it appears that at 

 any one place the movements on different days were in most cases 

 similar, though different at different places, indicating on these 

 occasions the occurrence usually of one general type of disturb- 

 ance. 



