in the axis of an artificial globe, if iron filings be strewed care- 

 fully over it, the filings would become magnetic, and arrange 

 themselves in curves like the magnetic needles on our globe, as 

 shown in Plate I. The small magnetic ingredients do not con- 

 verge to one mathematical point at each end of the bar, but to 

 a space equal to the transverse section of the axis ; and if this 

 effect is produced by a current of subtile fluid, which may be 

 conceived to emanate from one pole and to enter in at the other, 

 penetrating the substance of the bar, and again to issue from 

 its former outlet, as exhibited by the arrow r s in the above dia- 

 gram, it is reasonable to suppose that the fluid will not be com- 

 pressed more than the transverse size of the bar will require. 

 Taking this simple principle of action as a guide, with its vari- 

 ous consequences under different circumstances, we shall soon 

 perceive that we not only account for the various phenomena of 

 geology, but, in a word, all phenomena connected with terres- 

 trial physics ; and that we are enabled also to reason from the 

 known to the unknown, and actually to predict facts before trial, 

 not merely to satisfy curiosity, but questions of practical utility, 

 especially in mining. Indeed, theories are not worthy of atten- 

 tion unless they can be fairly demonstrated and rendered practi- 

 cally useful. 



The theory of gravitation was founded solely on the observed 

 effects, and not from any knowledge of the cause ; but this mag- 

 netical theory has been founded on both, and has the advantage 

 of merging the whole into one simple system, which does not 

 require any mysterious calculations to comprehend it, but only 

 ordinary abilities with the exercise of common sense. 



To proceed with our subject, let us consider a bar which has 

 been rendered magnetic by the fluid entering into one end, 

 and issuing out at the other, to be placed on a centre, or float- 

 ing on water, we find that it will preserve its meridional position 

 if left undisturbed ; it is therefore manifest, that by some in- 

 ternal action the iron bar is rendered susceptible of allowing the 

 fluid to pass through it only in one direction, viz. from the south 

 to the north end of the bar. If we cut the magnetic bar into 

 several pieces, what would be the result, supposing that by so 

 doing we do not injure its magnetic susceptibility ? Might we 

 not reasonably expect the same phenomena in the separate parts 



