168 



upon ; neither can it be denied, that in resorting to such a proof we 

 are only reasoning in a circle. It was known indeed, long ago, that 

 both the declination and inclination in the same place are far from 

 being invariable ; that both of them, in the course of time, undergo 

 very considerable progressive variations, independently of those pe- 

 riodical ones by which the nicety of observation is affected in different 

 seasons and parts of the day. It is, therefore, no matter of doubt 

 that the intensity of terrestrial magnetism must likewise be subject 

 to them 5 indeed, the periodical diurnal variations are clearly per- 

 ceptible in delicate observations. Hence, even if, after a considerable 

 lapse of time, the same time of vibration is again observable in a 

 given place, we are not, on that account, warranted in ascribing this 

 circumstance to anything but a casual compensation of the variations 

 which the intensity of the magnetism of the earth in that place, and 

 the magnetic state of the needle itself, may have experienced during 

 that interval. But even allowing the certainty of the comparative 

 method to be only diminished to a certain degree, not entirely an- 

 nulled, provided too long a space of time do not intervene, that mode, 

 at all events, becomes entirely useless in cases where it is required to 

 ascertain what changes the intensity of terrestrial magnetic force 

 undergoes in a given place during a very long interval. This ques- 

 tion, of considerable interest in a scientific point of view, must, there- 

 fore, remain unanswered until the merely comparative method shall 

 be superseded by one which reduces the intensity of terrestrial mag- 

 netism to unities perfectly determined and manifest, and entirely 

 independent of the individual nature of the needles employed in the 

 experiments. 



It is not difficult to lay down the theoretical principles on which 

 such an independent method is to be founded. The time of oscilla- 

 tion of a given needle depends on three quantities ; namely, the in- 

 tensity of the terrestrial magnetism, the static momentum of the free 

 magnetism in the needle, and the momentum of the inertia of this 

 needle. The last of them may readily be ascertained by suitable 

 methods ; and thus, from the observed duration of the oscillation, is 

 deduced, not the quantity of the intensity of the terrestrial magnetism, 

 but the product of this quantity into the static momentum of the free 

 magnetism in the needle. But it is impossible to separate these two 

 factors from one another, unless observations of quite a different kind 

 be superadded, that involve a different combination of them ; and 

 this end is attained by the use of a second needle, which, in order to 

 ascertain the ratio of these forces, is subjected both to the influence 

 of the magnetism of the earth and to that of the first needle. These 

 two effects do, indeed, partly depend on the magnetic state of the 

 second needle ; but, by suitably conducting the experiments, the 

 observer may eliminate that state, inasmuch as the ratio of both 

 forces becomes the more independent of it, the greater the distance 

 of the two needles from one another is assumed. Here, however, it 

 is obviously necessary, at the same time, to consider the position re- 

 lative to the magnetic meridian, of the magnetic axes of both needles, 

 and of that of the straight line connecting their centres, as also the 

 magnetic state of the first needle ; all which cannot be Subjected to 



