Mechanical Science. 611 



When a magnetic needle is heated, the coercitive force of the steel 

 is diminished, and the strength of the magnet considerably impaired. 

 Besides, the magnetism of the earth acts by induction on a needle, 

 and, when the temper is not very hard, is a quantity which increases 

 or diminishes the magnetism of the needle, according to the position 

 in which it is placed. 



It is also necessary to pay attention to the hour of the day at 

 which the observations are made. At eleven o'clock A. M., the 

 earth's intensity is greatest ; at five o'clock P. M., its intensity is 

 least. Another element, which M. Kupffer has made to enter into 

 the calculation, is the difference of intensity depending on the dif- 

 ference of latitude and longitude of the places of observation. 



The observations were made in the valley of Malka and on Khar- 

 bis, at an elevation of 4500 feet. After making the proper correc- 

 tions due to the causes which we have mentioned, it was found that 

 the duration in one oscillation of an excellent needle, made by Gam- 

 bey, was .063 of a second, which, for every 1000 feet of perpen- 

 dicular ascent (supposing the diminution uniform), is .014 of a 

 second. In comparing these observations with those of MM. Gay- 

 Lussac and Biot, M. Kupffer has arrived at the following remarkable 

 law : l That the increase of intensity which a needle, of the usual 

 degree of hardness, acquires by a diminution of temperature, accord- 

 ing to the height, is almost entirely compensated by the diminution of 

 the intensity in the magnetic forces of the earth, due to that eleva- 

 tion.' Since the magnetic intensity of the earth is thus found to 

 diminish at a small height above the surface, it appears obvious 

 the cause must reside nearer the surface than was formerly supposed, 

 and affords an additional proof that the declination and dip of the 

 needle depend on electric currents circulating about the earth at no 

 very considerable depth. 



ii. It had been long known that, in our hemisphere, the north pole 

 of a magnetic needle marches towards the west from eight o'clock in 

 the morning till two o'clock in the afternoon, and returns towards 

 the east during the rest of the day ; but its oscillations during the 

 night are so irregular, that it has been impossible to ascertain whe- 

 ther or not there be a corresponding period during the night. These 

 irregularities have been carefully attended to by M. Arago, and he 

 has generally found, that during those periods when the oscillations 

 of the needle were very irregular, the aurora borealis has been 

 visible in the north. Hence it has, I think, been too hastily 

 concluded, that the aurora borealis has a decided effect on the 

 needle. It appears more probable that the aurora borealis, and 

 the disturbing force on the needle, are both effects of the same 

 caus e the unknown cause of all terrestrial magnetic phenomena. 

 M. Kupffer has observed at Kasan, that the irregular march of 

 the needle took place on the same day, and almost at the same 

 instant, with those observed at Paris. Could a trifling electrical 

 atmospheric phenomenon in the north of Scotland produce the 

 same effect on a needle in Paris, almost directly south of it, and on 



