246 Prof. Magnus on the Condensation of Vapours 



cooling at the surface of bodies must be, but they do not enable 

 us to measure it ; for the alteration of temperature must be vari- 

 able, depending upon the thickness or extent of the body at the 

 surface of which the absorption takes place, and upon the velo- 

 city with which heat traverses it. This velocity seems to be very 

 great in all substances, since the alterations of temperature which 

 took place at the surface of the plate, and which could be only very 

 small, were visible on the scale after the lapse of a very short time. 



In the instance of rock-salt and other diathermanous sub- 

 stances, the heating must begin immediately. But with metal 

 plates, in which heat travels only by conduction, the shortness 

 of the time in which the action takes place is very striking, and 

 still more so when wood and other bad conductors are employed. 

 It seemed, therefore, desirable to institute some researches which 

 might furnish data concerning this velocity, and thereby at the 

 same time give certainty that no circumstance remained unno- 

 ticed which might have contributed towards the results mentioned 

 in the above experiments. 



In these researches a cubical box, H K (PL III. fig. 4), consisting 

 of wood and with sides 30 centims. in extent, was used. In one 

 of the sides of the box there was a circular opening, s s, of 10 

 centims. diameter. The plate p p, in which the passage of heat 

 w r as to be observed, was fastened in front of this opening. The 

 plate was pressed against the side of the box by means of screws 

 or other suitable means. In the box there was a thermo-pile, of 

 which one side was closed by means of its brass cap t. The other 

 side, which was uncovered, was pressed against the plate fastened 

 to the box, and, moreover, was placed in the middle of the circular 

 opening ; the pressure was managed by means of a spiral spring, 

 just as in the other instance given above, where the pressure was 

 exercised in a vertical direction. 



The wires which connected the pile with the galvanometer 

 came out through small openings in the box, which was other- 

 wise quite closed. Outside of the box, at a distance of 60 millims. 

 from it, there was placed the flame of a lamp, so arranged in 

 relation to the plate to be examined that it was situated in the 

 prolongation of the axis of the thermo-pile. 



The cubical box and the flame were enclosed in a great paste- 

 board box, which, to avoid catching fire, was provided with a 

 hole perpendicularly over the flame. By means of this case, all 

 disturbance which might arise from accidental motion of the air 

 of the room was obviated. After the temperature of the pile had 

 become constant, and consequently the indication of the galva- 

 nometer constant also, the candle was lit through a hole in the 

 paper box, which could be closed by a curtain, and the deflection 

 of the galvanometer observed with the telescope. 



