2 



every five minutes, day and night. Erom these observations he con- 

 cluded — 



" 1. That the path described by the needle consisted of a regular 

 diurnal motion, subject to disturbances of greater or less magnitude. 



"2. That this motion is due to electric currents passing from the 

 northern to the southern extremities of the telegraph wires, and return- 

 ing in the opposite direction. 



''3. That, exclusive of the irregular disturbances, the currents 

 flowed in a southerly direction from about 8 or 9 a. m. until the evening, 

 and in a northerly direction during the remainder of the twenty- four 

 hours," 



He was thus led to examine whether any relation subsisted between 

 these movements and the daily changes of the horizontal magnetic needle ; 

 and having made, for this purpose, a series of simultaneous observations 

 with a delicate declinometer, he came to the conclusion that although, 

 generally, the currents flow southwards during that part of the day in 

 which the variation of the horizontal needle is westerly (i. e. from 8 or 9 a.m. 

 until the .evening), and northwards, when the variation is easterly (i. e. 

 during the night and early part of the morning), ''yet simultaneous 

 observations showed no similarity in the path described by the mag- 

 netic needle and the galvanometer." 



An examination of Mr. Barlow's galvanometric observations led me, 

 some time since, to an opposite conclusion ; and at the last meeting of 

 the British Association, I stated my conviction, founded on these ob- 

 servations, that the earth-currents, whose continuous flow Mr. Barlow 

 has the merit of establishing, would eventually explain all the changes 

 of terrestrial magnetism, both periodic and irregular. I now proceed 

 to state the grounds of this conviction, and to show, from Mr. Barlow's 

 observations, that the diurnal changes of the earth currents correspond 

 with those of the horizontal component of the earth's magnetic force. 



Let us suppose, then, that the forces which act upon the horizontal 

 needle, and which cause it to deviate from its mean position, are due to 

 electric currents, traversing the upper strata of the earth in a horizontal 

 direction ; and let ^ denote the intensity of the current in the magnetic 

 meridian, positive when flowing northwards, and vice versa ; and i] the 

 intensity of the current perpendicular to the magnetic meridian, posi- 

 tive when flowing eastward, and vice versa. Then the force of the 

 current in any direction, making the angle e with the magnetic meridian 

 (measured to the east of north) is 



(p = ^ cos e + 7] sin e. 



jN'ow ^ is proportional to the force which deflects the freely suspended 

 horizontal needle from its mean position, or to XA^Jr, X being the 

 horizontal component of the earth's magnetic force, and the change 

 of declination expressed in parts of radius. Similarly, is proportional 

 to the force which deflects from its mean position a magnet, which is 

 maintained (by torsion or other means) in a position perpendicular to 

 the magnetic meridian ; and is measured (in terms of X ) by the rela- 



