158 TERRESTRIAL MAGNETISM. 



netic storm " only began very late. The greatest abso- 

 lute number of auroras were seen towards the latter part 

 of the month of September, and as March also showed 

 relative frequency as compared with February and 

 April, we may surmise here also, as in other magnetic 

 phenomena, a connection with the equinoxes. In addi- 

 tion to instances of aurora borealis seen in the southern 

 hemisphere in Peru, and of aurora australis see,n so far 

 north as Scotland, I may mention a coloured aurora 

 borealis' seen and observed for two hours by Captain 

 Lafond in the ' Candide' on the 14th of January 1831, 

 south of New Holland, in latitude 45. ( 215 ) 



The sound sometimes attributed to the aurora has 

 been negatived by the French physicists, and by Siljes- 

 trom at Bossekop, ( 216 ) as decidedly as by Thienemann, 

 Parry, Franklin, Richardson, Wrangel, and Anjou. 

 Bravais estimated the elevation or height of the aurora 

 above the earth at 100,000 metres, or more than fifty 

 miles; while a very meritorious observer, Mr. Farqu- 

 harson, had estimated it as only four thousand feet. 

 The grounds on which all these estimations are based 

 are exceedingly uncertain, and are liable to be altogether 

 vitiated as well as by optical illusions, as by improved 

 assumptions of the actual identity of a luminous arch 

 seen simultaneously from two distant places. On the 

 other hand, the influence of the aurora on the declina- 

 nation, inclination, and intensity, both horizontal and 

 total, of the magnetic force (on all the elements, there- 

 fore, of the earth's magnetism), is undoubted, although 

 very unequal at different stages of the phenomenon, 

 and in the effect produced on the different elements. 

 The most complete observations on this subject are 



