330 Mr H. Meikle m the Specific Heat 



rizontal tube could be connected with a pump or condenser, for 

 the purpose of rarifying or condensing the included air at plea- 

 sure. 



Things being thus prepared, a -slight change was effected in 

 the density of the included air ; and, after waiting a little till the 

 former temperature was regained, the stop-cock was opened, and 

 great care taken just to have it shut again by the very nick of 

 time that the liquid within the gauge-tube had acquired the level 

 of the outside, it being supposed that this was a proof that, at 

 that instant, the included air had exactly regained the atmo- 

 spheric pressure. A small interval being again allowed to re- 

 store the former temperature, the column of liquid in the gauge 

 now shewed the change of pressure due to the last variation of 

 temperature. 



In this mode of operating, there is some ground for suspect- 

 ing two sources of error, but which fortunately would be op- 

 posed to each other. In the first place, the air would take a 

 sensible time to pass through a moderately sized stop-cock ; and, 

 during that interval, a considerable portion of the change of 

 heat due to the change of density, would be lost on the sides of 

 the vessel ; especially considering how quickly heat might be 

 communicated between air in its then agitated state, and a vi- 

 treous surface. On the other hand, the liquid in the gauge-tube 

 might have acquired a force from its motion capable of carrying 

 it to the common level of the cistern, before the spring of the 

 air within had come into equilibrio with the atmosphere. If so, 

 it is evident that, in the above arrangement, the stop-cock has 

 been shut before that equilibrium was attained ; and which shut- 

 ting would, therefore, have been too soon, were it not that it hap- 

 pens nearly to be balanced by the other source of fallacy. To 

 illustrate the second case, let one end of a glass-tube be stopped 

 with the finger, and then let the other be immersed vertically in 

 a jar of water. On removing the finger, the water, which had 

 been depressed by the included air, will start considerably above 

 the common level ; so that, were the finger only partially re- 

 moved, and suddenly re-applied to shut the tube again, at such 

 a nick of time that the water within did not spring higher than 

 the common level, it is clear -that the force of the included air 

 jnust still have exceeded the atmospheric pressure ; and that it 



