312 



Messrs. Stewart and Tait on the Heating of a [May 15, 



results were as follows, a result represeutiug ou an average somewhat 

 less than four rotations : — 



Pressure of Heat-indication for 



residual gas, ✓ K ^ 



in inches. Dry hydrogen. Dry air. Dry carbonic acid. 



(A) ^- 27-0' 55-0 55-0 



(B) / 23-0 35-0 32-0 



Treating these results in the same manner as those of the ebonite disk, 

 we obtain :— 



Dry hydrogen. Dry air. Dry carbonic acid. 

 "Whole effect (-4).. 25-0 45-0 43-5 



Gas-effect .... 4-0 2O0 23-0 



Eesidual effect .... 21-0 . 25-0 205 



36. JSTow, if we suppose that there is only one effect due to gas, it 

 follows : — 



(a) That the proportion between the effects due to the various gases 

 experimented on (and all of the same pressure) is nevertheless different 

 for the two disks. 



(13) That the proportion (for the same disk) between the effects due to 

 the various gases experimented on is different according to the pressure. 



If, however, we suppose that there are two effects, one of which is in- 

 dependent of the residual gas, we find : — 



(a) That, as regards the gas-effect, the proportion between that due to 

 the various gases is nearly the same for both disks. Thus in the ebonite 

 disk we have 4, 20, 18, while in the paper disk we have 4, 20, 23 as re- 

 presenting the gas-effect for the various gases. 



(/3) That the residual effect in either disk is nearly the same for the 

 various gases. Thus in the ebonite disk we have 5*5, 5*0, 6*0, while in 

 the paper disk we have 21*0, 25-0, 20*5 as representing the residual effect 

 for the various gases. 



The results are thus much more simple on the hypothesis of two effects, 

 one of these being independent of the residual gas, than on the hypothesis 

 of only one effect. 



37. It was next endeavoured to ascertain whether these two effects 

 were differently influenced by a blind. 



A linen blind was first used before the thin ebonite disk, which it 

 cloaked during rotation, falling down afterwards. During rotation the 

 blind did not appear to touch the disk. A delicate galvanomer was used 

 for this experiment ; and the time of vibration of the needle was 2 seconds. 

 The rotation consisted of 30 turns of the handle in 40 seconds, as before. 

 The f ollowing result was obtained (each number representing the mean 

 of two observations) : — 



Without blind. With blind. 



of dry hydrogen 165 127 



^ hydrogen + 4 an ' 477 248 



