46 



SCIENCE 



[N. S. Vol. XXXIII. No. 837 



fer, in this respect, from seismic disturb- 

 ances which proceed in all directions from 

 the center of disturbance, then a harmonic 

 analysis of the disturbance effects should 

 furnish further evidence. A definite elec- 

 tric or magnetic system must be accom- 

 panied by equally definite effects on our 

 suspended magnetic needles. Knowing the 

 latter, we ought, in turn, to be able to de- 

 termine the general character, at least, of 

 the producing system. 



The beginning of the disturbance of 

 May 8, 1902, may be taken as typical of the 

 general type of perturbation thus far con- 

 sidered, viz., an increase in the horizontal 

 intensity over the whole, or at least the 

 greater part, of the earth, and a decrease 

 in the vertical intensity in the northern 

 magnetic hemisphere, and an increase in 

 the southern. Applying a mathematical 

 analysis, it is found that the system of 

 forces which could produce the observed 

 disturbance was a two-fold one — the first, 

 the stronger, consisted of a set of electric 

 currents in the upper regions, circulating 

 eastward around the earth, if negative 

 currents, and the reverse, if positive ones; 

 the second, a weaker system, contained 

 within the earth and possessing the char- 

 acteristics not of an induced electric sys- 

 tem, but of directly induced magnetism of 

 the same sign as that of the earth's own 

 field. 



Please note that according to this analy- 

 sis the disturbance system is chiefly an 

 overhead set of currents proceeding, if 

 negative ones, in an eastward direction 

 around the earth — but this, in fact, is the 

 direction in which the recorded times of 

 beginning of the disturbance were found, 

 in general, to progress. One method of 

 investigation thus independently supports 

 the other. 



How may we suppose that negative elec- 

 tric currents are brought about in the 



regions above us which could thus affect 

 our magnetic needles? If the progression 

 in the times of beginning of the effect may 

 be interpreted as meaning that, whatever 

 the cause, it is moving with a velocity 

 measured by the differences in the times at 

 distant points on the earth, then the re- 

 sulting velocities are on the order of about 

 one two-thousandth that of electromagnetic 

 waves, or about one four-hundredth that 

 of cathode rays. The question immediately 

 arises — May not the required overhead 

 negative electric currents be brought about 

 by rapidly moving electric charges, whose 

 accompanying magnetic perturbations are 

 but an exhibition of the Rowland effect on 

 a scale far transcending any laboratory 

 experiment within the power of man ? We 

 may be dealing with ionic charges set in 

 motion, as the result of a releasing action 

 from some quarter, by sources of energy 

 already existent in the regions above us, 

 whence currents arise — 



" Of power to wheedle 

 From its loved north the subtle needle," 



as Maxwell said with regard to the convec- 

 tion currents, which "that doughty 

 Knight, Rowland of Troy, did obtain." 



Now before indulging in a bit of scien- 

 tific imagination, let me caution you to dis- 

 tinguish carefully between what is fact and 

 what is hypothesis. The results communi- 

 cated respecting the differences and pro- 

 gression in the times at which sudden 

 perturbations occur, as well as those de- 

 rived from the mathematical analysis of 

 the recorded effects, are independent of 

 theory. You may not agree with me in any 

 hypothesis which I may attempt to estab- 

 lish as to the cause of magnetic disturb- 

 ances and the modus operandi, but please 

 remember that the facts remain, however 

 difficult may be the problems which they 

 present. 



