between Metallic Masses having different Temperatures. 187 



the properties of some metals at a definite higher temperature, 

 I heated several bars in a cast-iron vessel full of sand, along 

 with a thermometer, having a very long scale: this vessel was 

 placed in another containing oil, and when the temperature 

 had risen to 350°, the bars were placed upon cold lead. On 

 one occasion I employed copper, brass, iron and antimony; 

 on another iron, tin, platinum and bismuth. I did not find, 

 however, that the additional temperature thus gained facili- 

 tated my inquiries, and it was, in the first place attended with 

 considerable practical difficulties. The experiments, however, 

 confirmed a fact which I had previously suspected, and which 

 forms an exception to what may be considered the general 

 law, namel}', that the intensity of vibration is proportional to 

 the difference of temperature of the metals ; I found that at 

 350° iron was far more sluggish in its vibrations than at 212°. 

 I cannot say that I remarked this in the case of copper, brass, 

 or platinum. The fact, however, hardly admits of doubt. At 

 an early period I had been much perplexed with some ano- 

 malies in the vibration of iron. When first taken out of a 

 hot open fire, and just cool enough not to melt lead, its action 

 with that metal appeared very unsatisfactory. This effect was 

 so sensible, that I have frequently repeated with success a 

 singularly paradoxical experiment. A bar of iron heated, 

 suppose to 212°, being placed on a lead block, and the vibra- 

 tions commenced, if a spirit-lamp was applied to the lower 

 portion of the bar, the vibrations are completely stopped, and 

 may actually be restored by immersing the lead, to which the 

 lamp had been applied, in cold water : these singular effects 

 I have been able to produce several times in succession during 

 one experiment. 



48. The same effects, though less striking, have been pro- 

 duced with zinc instead of iron, which vibrates with consider- 

 able difficulty when the temperature is raised above 212°. I 

 have been disposed to consider that every metal has its own 

 most favourable temperature, though on what principle it is 

 not so easy to explain. 



49. It is probable that the softening of the heated metal di- 

 minishes the resiliency of the two bodies when impact takes 

 place. I do not think that it is attributable to the softening 

 of the lead, for I have found that iron is more disposed to vi- 

 brate on platinum when at a moderate temperature, than when 

 red hot. The effect may, however, be connected with the 

 theory of the vibration. 



50. Having now discussed the phamomena of sound, and 

 of the vibrations to which we have shown these sounds to be 

 referable, we shall next consider 



2B 2 



