88 CORRELATION OF PHYSICAL FORCES. 



rous vibrations. Metals electrified or magnetised will emit a 

 sound ; so they will if struck, or if a musical note with which 

 they can vibrate in unison be sounded near to them. 



Even chemical decomposition, in cases of feeble affinity, 

 may be produced by purely mechanical effects. A number of 

 instances of this have been collected by M. Becquerel ; and 

 substances whose constituents are held together by feeble affi- 

 nities such as iodide of nitrogen and similar compounds, are 

 decomposed by the vibration occasioned by sound. 



If, instead of being regarded as a fluid or imponderable 

 matter sui generis t electricity be regarded as the motion of an 

 ether, equal difficulties are encountered. Assuming ether to 

 pervade the pores of all bodies, is the ether a conductor or 

 non-conductor ? If the latter that is, if the ether be inca- 

 pable of transmitting the electrical wave the ethereal hypo- 

 thesis of electricity necessarily fails ; but if the motion of the 

 ether constitute what we call conduction of electricity, then 

 the more porous bodies, or those most permeable by the ether, 

 should be the best conductors. But this is not the case. If, 

 again, the metal and the air surrounding it are both pervaded 

 by ether, why should the electrical wave affect the ether in the 

 metal, and not stir that in the gas ? To support an ethereal 

 hypothesis of electricity, many additional and hardly recon- 

 cilable hypotheses must be imported. 



The fracture and comminution of a non-conducting body, 

 the fusion or dispersion of a metallic wire by the electrical 

 discharge, are effects equally difficult to conceive upon the 

 hypothesis of an ethereal vibration, as upon that of a fluid, but 

 are necessary results of the sudden subversion of molecular 

 polarisation, or of a sudden or irregular vibratory movement of 

 the matter itself. We see similar effects to those of electricity 

 produced by sonorous vibrations, which might be called con- 

 duction and nonconduction of sound. One body transmits 

 sound easily, another stops or deadens it, as it is termed i.e. 

 disperses the vibrations, instead of continuing them in the 

 same direction as the primary impulse ; and solid bodies 

 may, as has been above observed, be shivered by sudden im- 

 pulses of sound in those cases where all the parts of the body 

 cannot uniformly carry on the undulatory motion. 



