Producible by Attrition and Contact of Metals. 21 



to me that this opinion has never been established. A persistent temperature 

 above or below that of the surrounding media, is never the condition of metals 

 brought into contact at unequal temperatures, for the purpose of producing 

 deflections ; hence we know absolutely nothing of what the result would be if 

 they could be presented to each other in such a condition. What we do know 

 is, that when deflections are produced by metals at unequal temperatures, the 

 colder metal is receiving, and the warmer is parting with heat at that moment; 

 tliat is, the mediate agent in causing deflection is the motion of heat, provided 

 .tlie language be permitted of those who consider heat to be imponderable ele- 

 mentary matter which enters into, and passes out of, other kinds of matter. The 

 quantity of heat, even although diflereut in the metals, would not be a sufficient 

 cause of deflection; for the quantity of heat may be different, yet the thermo- 

 mctric indication may be the same: and under such circumstances no deflec- 

 tion would result. 



Thus quiescent heat, no matter what its amount or difference of quantity, is 

 not known to affect metals in such a manner as to produce deflection ; and it is 

 tiie entrance or departure of heat, in other words, its motion, that seems to be 

 effective. It is not probable that a body, during the process of heating or cool- 

 ing, ever remains of the same temperature for one moment of time. Heat, if 

 ever stationary in a body, can only be so when that body and all the surround- 

 ing media are at the same temperature ; but metals, when in this state of equili- 

 brium of heat, do not by contact produce deflection. As soon as heat begins to 

 enter or leave either or both unequally, then they are capable of affecting the 

 needle ; and although in the foregoing pages I have frequently mentioned the 

 efficiency of unequal temperatures, it was for the sake of brevity, and not with 

 a view of conveying any opinion different from what has now been explained. 

 It will presently appear that the distinction here made is important. 



The following experiments coincide with the preceding views. Take two 

 rods of different metals, each about twelve inches long, and a quarter of an inch 

 in diameter; antimony with bismuth, or German silver with brass, and many 

 other combinations will answer. Tie the rods together at one end with thread, 

 so that they shall form the letter V; immerse the tied ends in boiling water to 

 the depth of about one-tliird of the length of the rods, and continue the boiling. 

 It is to be presumed that the immersed portions have at length assumed the 



