58 Prof. Potter on the Fourth Law of the Relations of the 



temperature. A considerable degree of cold was produced by 

 even a small amount of expansion j but the time required was 

 eight or nine seconds before the thermometer reached its mini- 

 mum. This slowness in a great measure no doubt arises from 

 the quiescence of the remaining air in the receiver, which only 

 then acts slowly by radiation and convection on the helix of the 

 thermometer. This slowness renders the results of little value, 

 beyond showing that a considerable degree of cold is produced 

 by a moderate rarefaction after a short interval of time. 



Many experiments gave nearly the same results ; and averages 

 of five successive good experiments were as follows : — The tem- 

 perature in the receiver in degrees of Breguet's thermometer 

 54 0, 42 : after one stroke of the piston of the pump, in eight 

 seconds the index arrived at the minimum 47°'82 ; after eight 

 seconds more it stood at 49° # 46; in eight more at 51 o, 04; in 

 eight more at 52°-33; in eight more at 53 0, 16; in eight more 

 at 53 0, 72; in eight more at 54 o, 02; after some time more at 

 5 4° *32. The differences in the successive intervals of eight 

 seconds after the completion of the exhaustion were thus 6 o, 60, 

 l°-64, l°-58, l°-29, 0°-83, 0°'56, 0°-30, 0°-30. If we suppose 

 the heat lost in the first interval of eight seconds equal to that 

 in the second, we have 6 o, 60 + l o, 64 = 8 o, 24 for the maximum of 

 cold on Breguet ; s scale with the circle divided into 360 parts. 

 The barometer-gauge rose 1*9 to 2 inches, and therefore the 

 expansion 8 was as follows : 



p-l 30-28 

 °~ p' 28 ** m ' 



Hence 1 degree of cold on the scale was produced by a rarefac- 

 tion ,-7 — Q-r = yr^fi ; which, supposing with M. Prony that 



these degrees are each y^tus of a degree Centigrade when the 



.7 100 7 

 circle is divided into 100 parts, gives ^ x -^rr = o^ of a degree 



Centigrade as the value of each degree of the thermometer used. 

 While Poisson supposed 1 degree Centigrade of cold to be pro- 

 duced by a rarefaction =-y^, we have only found by this method 



one-fifth of that quantity. 



Experiments were also undertaken with dry air. The receiver 

 being exhausted, it was allowed to refill slowly with air passing 

 through a V-tube of glass containing pumice-stone moistened 

 with strong sulphuric acid. The results were not greatly dif- 

 ferent from those obtained with the ordinary air of the room, 

 which was the experimental lecture-room of University College, 

 and was heated by a stove. With the dry air the temperature 



