409 



September 2nd, 5 A.M., about which time another very abrupt 

 disturbance simultaneously affected all the elements, and continued 

 with great violence until about 4 P.M. of the same day, when it 

 somewhat abated. The elements, nevertheless, remained in a state 

 of considerable disturbance until September 5th, and only attained 

 their normal positions on September 7th. It thus appears that we 

 have two distinct well-marked disturbances, both commencing ab- 

 ruptly and ending gradually, the first of which began on the evening 

 of August 28th, and the second on the early morning of Septem- 

 ber 2nd. 



These two great disturbances correspond, therefore, in point of 

 time, to the two great auroral displays already alluded to. 



The average effect of both of these disturbances was to increase 

 the declination, and to diminish the horizontal and vertical compo- 

 nents of the earth's magnetic force. 



The author then alludes to a disturbance which took place about 

 15 minutes past 11 o'clock on the forenoon of September 1st, simi- 

 lar in its mode of action to the two great disturbances already 

 mentioned, but not equalling them in extent or in duration. This 

 disturbance affected the magnetograph simultaneously with the 

 breaking out, near a spot on the sun's disc, of a bright star, which 

 was fortunately observed both by Mr. Carrington and Mr. Hodgson, 

 independently. 



The study of the curves furnished by the magnetograph during 

 this great storm, in the author's opinion seems to throw light upon 

 the connexion which subsists between magnetic disturbances, earth- 

 currents, and auroral displays. These curves show that at the com- 

 mencement of the disturbance, the horizontal and vertical compo- 

 nents of the earth's force remained depressed below their normal 

 values for at least seven hours. The curves for this portion of time 

 have also a peculiar serrated appearance, as if on the great wave of 

 disturbance whose period was seven hours, there were superimposed 

 smaller waves whose period might be perhaps only a few minutes, or 

 even less. 



Now Mr. C. V. Walker, in a paper communicated to the Royal 

 Society (January 31st, 1861), shows that the earth-currents which 

 accompany auroras are of very short period (say a few minutes), and 

 the auroras themselves also appear to be of a very fitful nature. We 



VOL. xi. 2 G 



