257 



of three minutes, for one hour. The table contained the 

 direct results of observation with the two instruments ; the 

 differences of these results and the means of the day, (or 

 the disturbances in declination and horizontal force,) esti- 

 mated in parts of the scale; and the same differences re- 

 duced to their proper measures. The extreme disturbance 

 in declination, amounted to 17'.9; and that of the horizon- 

 tal force to .0127. The changes of the horizontal com- 

 ponent of the force arising partly from changes of the total 

 force, and partly from changes of inclination, and the part 

 due to the latter being, in high magnetic latitudes, much 

 the greater, it is manifest that the changes of inclina- 

 tion may be deduced, approximately, from those of the ho- 

 rizontal force, on the assumption that the actual force re- 

 mains unvaried. The changes ofinclination, thus deduced, 

 were given in another column of the table. 



The numerical values of the changes of declination and 

 inclination thus obtained, were laid down in charts, so as to 

 represent graphically the progress of the disturbance of each 

 of the elements of the magnetic, direction. In a third chart 

 the combined effect of the two disturbances was represented, 

 so as to exhibit the successive positions of the pole of the 

 needle, supposed free to move in every direction. From this 

 it appeared, that in the present instance, the effect of the 

 auroral disturbance upon the resultant direction of the earth's 

 magnetic force, has been to impress upon the pole of the 

 needle a kind of epicycloidal movement. It will remain for 

 future observations to determine whether or not this is a 

 general law ; the light that such an inquiry must throw on 

 the nature of the disturbance need not be insisted on. 



The aurora appears to have been frequent about this 

 period. Two days after this observation, on the 21st, at 

 nine o'clock in the evening, the magnetometers were again 

 disturbed. The extreme positions observed, occurred at 

 9 h 10 m , and 9 h 35 ra ; and in this interval the change of decli- 



