Thermotic Effect of Compression and Expansion of Air, 117 



stopcock 2 partially (No. 1 was always wide open), thus 

 allowing the air to issue continually, and strike on the bulb 

 of the thermometer. But if the stopcock was opened too 

 wide, and its bore was fully one inch, or the same as the 

 tube, then the air escaped so fast that the whole tube was to 

 be looked upon as the escaping hole, and the air had partly 

 escaped from the reservoir compression, and was therefore 

 in the act of cooling by expansion. By keeping, however, the 

 stopcock about one-third open, it was found that not only was 

 the highest temperature obtained, but it was so constant^ and 

 varied so little as the opening was either enlarged or de- 

 creased, within certain limits on either side of that proportion, 

 that the mean thermometer reading of 92°, as obtained, was 

 thought to be true to a less quantity than the first element. 



3e/. In the matter of the pressure of the air as shewn in 

 the mercury gauge, that was determined when stopcock No. 

 2 was quite closed ; for any opening of it, by causing a main 

 stream through the pipe, detracted from the pressure exerted 

 on the mercury. The gauge consisted of a bent tube open 

 at the external end, and containing mercury in the lower 

 bend, the air entering from the pipe into the glass tube 

 pressed the mercury down on that side, and the height of the 

 fluid in one above the other, was read oflP by a scale of 

 inches attached, and was generally about seven inches. 



And 4:th. With reference to the temperature of the air on 

 escaping from compression, if the outer stopcock was too 

 widely opened, the mercury pressure gauge fell, and the ther- 

 mometer stood high, on account of the air not having had, time 

 or space to expand. But when the passage was rather 

 throttled, the gauge rose, and the temperature fell, the smaller 

 quantity of escaping air having decidedly room to expand and 

 rarify. Care had to be taken, however, in this case, not to 

 close the openmg too much, but always to keep up such a 

 stream of expanding air as should be able to completely 

 envelope and impress itself upon the bulb of the thermome- 

 ter, which then stood at the average height of 65°, and the 

 same result was obtained when thermometer No. 2 was 

 taken out of its socket under other degrees of opening, and 

 held at some distance away from the mouth of the pipe, so 



