Prof. Thomson on the Dynamical Theory of Heat. 433 



kept up in the two parts by the middle of the battery being fitted 

 nearly air-tight in the tube by means of wax, or otherwise ; but 

 this arrangement would not be satisfactory, as portions of the 

 bars of the batterj'^, if not the ends themselves directly, would 

 be altered in temperature, even if Mayer's hypothesis were rigor- 

 ously true, on account of the rushing of the air among them. 

 No part of the battery ought to be exposed to the rushing of 

 the air in the neighbourhood of the orifice, and therefore the 

 middle of the battery would have to be external to the tube, the 

 ends being cemented into the tube by some indurating cement 

 sufficiently strong and compact to hold perfectly air-tight on the 

 side where the pressure is different from the atmospheric press- 

 ure. By such means as these, I think a veiy satisfactory series 

 of experiments might easily be performed to test Mayei-'s hypo- 

 thesis for air through a very wide range of temperatures. 



78. Should the differential method of experimenting just de- 

 scribed indicate any difference of temperature whatever on the 

 two sides of the orifice, Mayer's hypothesis v/ould be shown to 

 be not exactly fulfilled, and, according as the air leaving the 

 orifice is found to be warmer or colder than the entering air, we 

 should infer that the heat absorbed, when air expands at a con- 

 stant temjDerature, is less than or greater than the equivalent 

 of the mechanical effect produced by the expansion *. 



79. Calorimetrical methods, like those used by Joule, might 

 then be followed for actually determining the heat emitted or 

 absorbed by the air in the neighbourhood of the orifice, or in the 

 second spiral, in acquiring the temperature of the air in the 

 entering stream; and by careful experimenting, it is probable 

 that excessively accurate results might be thus obtained for a 

 wide range of temperature. 



80. The result of each experiment would be a value of /x, in 

 terms of Joule's mechanical equivalent, to be calculated by the 

 following expression, derived from equations (5) and (G). 



JE 

 1+E/ 



^= H (^)- 



l_j. 11 



p'u' log ^ 



* Experiments on the plan here suggested have been recently made by 

 Mr. Joule and myself, and it has thus been ascertained that the air leaves 

 the rujnds in the neighbourhood of the orifice at a lower tenii)crature than it 

 approached them, even if this temperature be as high as 170° !''■ ; and it 

 follows that the heat absorbed is (jrealer than the equivalent of the mecha- 

 nical elFect of the expansion, even for so high a temperature, and jirobably 

 for nuich higher. See a paper ])ublished in the Sup|)lemcnt to this Volume 

 of the Magazine, in which these experiments are described. — Nov. 11, 1852. 



Phil. May, S. 4. Vol. 4. No. 27. Dec, 1853. 3 E 



