446 Prof. FORBES on the Vibrations which take place 



with an intensity proportional (within certain limits) to the diffe- 

 rence of conducting power of the two metals employed for heat (or 

 electricity) : the metal having the least conducting power being 

 necessarily the coldest. I have stated that the difference of con- 

 ducting power must be within certain limits, because the anomaly 

 of antimony and bismuth seems to be caused by this exception ; 

 and it is on the same account probably that the class of bodies 

 which possess the vibrating property is confined to the metals ; 

 other matter being destitute of the requisite conducting power. 

 Here antimony and bismuth almost want this characteristic pro- 

 perty of the other metals examined. My experiments with the 

 thermometer of contact prove their very low rank as conductors 

 of heat, as Mr HARRIS of Plymouth, in reporting to me some 

 experiments which he had kindly undertaken at my request, with 

 regard to their power as electrical conductors, states in regard 

 to bismuth, that nothing in " the form of a metal can be much 

 worse." 



2. Influence of Figure upon the Vibrations. 



37. I have already noticed the form of the apparatus which 

 I have generally employed. The time of the oscillations and 

 their magnitude depend upon the figure of the vibrating mass, 

 which seems to act just as in the case of a pendulum or rather of 

 a rocking-stone, the impulse which it receives at each vibration 

 appearing to be given at whatever instant of time the contact of the 

 vibrating edges with the block is effected. This, however, must 

 be understood within certain limits. There must be a decisive 

 interval of time between the two contacts, for if the surface, in- 

 stead of having two solid angles, as in the bar described above, 

 merges into a continued curve, the vibrations will not take place. 

 If, by any means, however, the period of contact of two portions 

 of the curved surface with the block be prolonged, the impulse 



