ox THE TERRESTRIAL MAGNETIC FORCE. 25.3 



one of which consists in observing the time of vibration of a 

 needle. But this operation, from its very nature, requires a 

 considerable time, as the number of vibrations from which we 

 deduce the duration of a single vibration ought not to be too 

 small. Now, supposing the magnetic intensity to be con- 

 stant during the period employed in the observation, the re- 

 sulting time of vibration will correspond truly to the intensity ; 

 but if the latter has varied in the interval, the time of vibration 

 will only correspond to its mean value. Whatever changes may 

 have taken place during the interval are entirely concealed from 

 us, the instrument giving only average values. If, in order 

 to approximate more closely, we were to choose shorter in- 

 tervals, or to base the results upon a smaller number of vibra- 

 tions, we should sacrifice accuracy and certainty, and be in dan- 

 ger of considering errors of observation as anomalies in the in- 

 tensity. 



But the more interesting the magnetic disturbances in short 

 intervals appeared, — as shown by the experiments of last year, 

 in regard to the declination only, — the more important it was 

 to possess a means by which the effects of similar disturbances 

 in the intensity might be foUow^ed and measured with the same 

 ease, cei'tainty, and accuracy. 



We have already seen that the unfitness of the method hitherto 

 employed for this purpose consists in the circumstance, that it 

 is based on observations of the times of vibration, which, from 

 their very nature, must al^'ays require a long interval. Now- the 

 time of vibration serves in this case only to determine indirectly 

 the moment of rotation w hich the earth's magnetism imparts to 

 the needle when it is not situated in the magnetic meridian. If, 

 then, we can determine accurately this moment of rotation in a 

 direct manner without observations of vibi'ation, and if we can 

 measure its variation with accuracy, quickness, and certainty, 

 our main object is attained. The method to be described for 

 this purpose rests on the follow ing basis. 



The necessary conditions of equilibrium of a body of any form 

 suspended to two threads, — its parts being supposed, in the first 

 instance, suljjcct to gravity alone, and firmly connected with 

 each other, — may be thus briefly described : the vertical passing 

 through the centre of gravity of the body, and the straight lines 

 coinciding with the threads, arc in one plane, and are either 

 parallel with each other, or intersect in a fixed point. In all 



