the llarejaclion and Condensation of' Air. 

 Table I. 



373 



The result given in the above Table is the difference be- 

 tween the effects of condensation and friction alone, corrected 

 for the slight superiority of the cooling influence of the atmo- 

 sphere in the experiments on friction. We must now, hovk^- 

 ever, proceed to apply a further correction, on account of the 

 circumstance that the friction of the piston was considerably 

 greater during the condensing experiments than during the 

 experiments to ascertain the effect of friction. Jn the latter 

 case the piston worked with a vacuum beneaith it, whilst in the 

 former the leather was pressed to the sides of the pump by a 

 force of condensed air, averaging 2>% lbs. per square inclr. I 

 endeavoured to estimate the difference between the friction in 

 the two cases, by removing the valve of the receiver and work- 

 ing the pump with about 32 lbs. per square inch pressure 

 below it. These experiments, alternated with others in which 

 a vacuum was beneath the piston, showed that the heat given 

 out in the two cases was, as nearly as possible, in the ratio of 

 six to five. When the correction indicated in this manner 

 has been applied to 0°'297 (see Table) and the result subtracted 

 from 0°-643, we obtain 0^-285 as the effect of compressing 

 2956 cubic inches of dry air at a pressure of 30*2 inches of 

 mercury, into the space of 1 36*5 cubic inches. 



This heat was distributed through 45 lbs. 3 ozs. of water, 



