1^6 



tions of intensity appear to be infinitely feeble 7 

 since they cannot be perceived, if we employ, 

 instead of the magnetometer with a perpendi- 

 cular rod terminated by a ball of iron, the deli- 

 cate apparatus of Coulumb, that is, the oscilla- 

 tions of a small needle contained in a glass case^ 

 and suspended by an untwisted silk thread*. 



been observed ; but a slow change take3 place in the space of 

 several years. 



* At the hospice of Mount Cenis, and at Rome, Mr. Gay- 

 Lussac and myself have observed the oscillations of the same 

 needle by day and by night, in very different atmospherical 

 temperatures. The result of these experiments was, that, if 

 there exists a horary variation in the intensity of the magnetic 

 forces, it does not alter the duration of the oscillations a 

 twelve hundredth. At Milan, the same needle made, on the 

 15th of April, 1805, in the interior of the city, near the 

 cathedral, sixty oscillations in 4'56 , 8 7/ j and on the 7th of 

 October, in a meadow without the walls, in 4' 56*4". At 

 Rome the duration of the oscillations was the same to a few 

 tenths of a second, at the Villa Borghese, at Monte-Pincio, 

 and on the road to Tivoli. Experiments of this sort are sus- 

 ceptible of so great an exactness, that in different experiments 

 made on the top of Mount Cenis, two hundred and fifty oscil- 

 lations lasted .1220-3", 12292", 1229", and 1229\5". At 

 Rome we successively found, with a chronometer of Breguet, 

 1169 2", 1109-2", 1169", and 1169 5". I have thought it 

 right to note in this place these results, in order to prove, that 

 the experiments made on the intensity of the magnetic forces, 

 and recorded in this work, are not subject, in a small extent 

 of ground, to that great number of local and horary influences, 

 which affect the observations on the variation of the magnetic 

 needle. 



