248 MAGNETIC VARIATIONS. 



beck has shown us that electrical and magneto- electri- 

 cal phenomena can be produced by the action of heat 

 upon metallic bars, we have, perhaps, approached 

 towards some faint appreciation of the manner in which 

 the solar calorific radiations may, acting on the surface 

 of our planet, produce electrical and magnetic effects. 

 If we suppose that the sun produces a disturbance of 

 the earth's electricity along any given line, in all direc- 

 tions at right angles to that line, we shall have mag- 

 netic polaritjr induced.* That such a disturbance is 

 regularly produced every time the sun rises, has been 

 sufficiently proved by many observers. 



In 1750, Wargentiii noticed that a very remarkable 

 display of Aurora borealis was the cause of a peculiar 

 disturbance of the magnetic needle ; and Dr. Dalton f 

 was the first to show that the luminous rays of the 

 Aurora are always parallel to the dipping-needle, and 

 that the Auroral arches cross the magnetic meridian at 

 right angles. Hansteen and Arago have attended with 

 particular care to these influences of the northern lights, 

 and the results of their observations are : 



That as the crown of the Aurora quits the usual 

 place, the dipping-needle moves several degrees for- 

 ward: 



That the part of the sky where all the beams of the 

 Aurora unite, is that to which a magnetic needle 

 directs itself, when suspended by its centre of gravity: 

 That the concentric circles, which show themselves 

 previously to the luminous beams, rest upon two points 

 of the horizon equally distant from the magnetic meri- 



* It has been observed by Mr. Barlow, in England, and some 

 eminent observers in Austria, that an electric current constantly 

 traverses the wires of the electric telegraph wherever there are two 

 earth connections. 



f Meteorological Observations and Essays : by Dr. Dalton. On 

 the Height of the Aurora Borealis above the surface of the Earth : 

 by John Dalton, F.K.S. Philosophical Transactions, vol. cxiv. 

 p. 291. 



