126 Effect of Pressure in Lowering the Freezing-Point of Water. 



all doubt. After that I attempted, in a more deliberate ex- 

 periment, to determine as accurately as my means of observa- 

 tion allowed me to do, the actual extent to which the tempe- 

 rature of freezing is affected by determinate applications of 

 pressure. 



In the present communication I shall merely mention the 

 results obtained, without entering at all upon the details of 

 the experiment. 



I found that a pressure of, as nearly as I have been able to 

 estimate it, 8*1 atmospheres produced a depression measured 

 by 71 divisions of the tube on the column of sether in the 

 thermometer; and again, a pressure of 16*8 atmospheres pro- 

 duced a thermometric depression of \6~ divisions. Hence 



71 

 the observed lowering of temperature was —|, or "106° F. in 



the former case, and —2, or -232° F. in the latter. 



Let us compare these results with theory. According to 

 the conclusions arrived at by my brother in the paper referred 

 to above, the lowering of the freezing-point of water b}/ 8*1 

 atmospheres of pressure would be 8"! x*0135, or '109° F. ; 

 and the lowering of the freezing-point by 16*8 atmospheres 

 would be 16*8 x '0135, or -227° F. Hence we have the fol- 

 lowing highly satisfactory comparison, for the two cases, be- 

 tween the experiment and theory : — 



Observed pressures. 



Observed depres- 

 sions of tempera- 

 tures. 



Depressions according to 



theory, on the hypothesis 



that the pressures were truly 



observed. 



Differences. 



8*1 atmospheres ... 

 16-8 atmospheres ... 



•106° F. 

 •232° F. 



•109° F. 



•227° F. 



-•003° F. 

 + •005° F. 



It was, I confess, with some surprise, that, after having com- 

 pleted the observations under an impression that they pre- 

 sented great discrepancies from the theoretical expectations, 

 I found the numbers I had noted down indicated in reality an 

 agreement so remarkably close, that I could not but attribute 

 it in some degree to chance, when I reflected on the very rude 

 manner in which the quantitative parts of the experiment 

 (especially the measurement of the pressure, and the evalua- 

 tion of the division of the sether thermometer) had been con- 

 ducted. 



I hope before long to have a thermometer constructed, 

 which shall be at least three times as sensitive as the aether 

 thermometer 1 have used hitherto ; and I expect with it to be 



