ABSOLUTE THERMAL CONDUCTIVITY OF NICKEL. 369 



in the interior of a large copper vessel which protected it from draughts in the room. 

 The thermometer used in the calorimeter was one of Ducretet's precision thermometers 

 divided into tenths of degrees centigrade, and it was read by means of a telescope fixed 

 a little distance off, hundredths of a degree being estimated by eye. The steam heater 

 was used in the ordinary way, but in order to get determinations at different tempera- 

 tures the same heater was used with methylated spirits instead of water. A tube was 

 arranged to conduct all the spirits which condensed in the apparatus back to the boiler 

 by a pipe leading in at the bottom of it. Only a small portion of the spirits was dis- 

 tilled off, as the flame of the burner heating the boiler was so arranged that very little 

 vapour was formed over and above that required to produce heat enough by its con- 

 densation to maintain the heater at a uniform temperature. The boiling point rose by 

 less than one degree in the course of a day on account of loss by distillation of the 

 more volatile constituents of the spirits. The top of the chamber in the heater, in 

 which the nickel was suspended while being heated, was closed by a large cork in which 

 were two holes, one letting in the thermometer which indicated the temperature of the 

 lickel, and another through which passed a fine wire supporting the nickel. The 

 >ottom of the chamber was closed by a slide padded with cotton wool. This slide was 

 Irawn aside when the nickel was dropped into the calorimeter, an arrangement being 

 nade for doing all this with great rapidity. The wire which had supported the nickel 

 n the heater remained attached to it, and the end of it, which was usually found pro- 

 ecting out of the mouth of the calorimeter, was at once seized and the contents of the 

 alorimeter stirred by moving the nickel up and down and to and fro in the water. 

 Vhile this was being done the thermometer was being watched through the telescope, 

 wo persons were thus necessary. No correction has been made for the thermal equiv- 

 lent of the work done in stirring, as it has been assumed to be negligible. It was im- 

 ossible to observe if stirring the water produced an appreciable quantity of heat as 

 nder all circumstances its effect was quite obscured by the disturbances produced by 

 ther causes. 



The correction for cooling was applied in the following way : — The observer at the 

 lescope had in his hands a stop watch, with two hands so arranged that, by pressing 

 le stop, they would start together ; pressing another stop made one of them remain 

 here it was at the instant of pressing ; a second press made it overtake and go en as 

 :fore with the other hand. In this way, the time at which the thermometer indicated 

 uy reading could be noted to a fraction of a second, the reading being written down 

 subsequently ; and a fresh reading could be then taken in the same way. A curve was 

 ten plotted from the readings thus obtained, with temperatures as ordinates, and times 

 '<■ abscissae. The part of the curve corresponding to times after the maximum temperature 

 Id been reached was produced backwards to the axis of zero time, and in this way the 

 tnperature which the calorimeter must have cooled from, had its rise of temperature 

 ten instantaneous, was found. This is very nearly the temperature which would have 

 en reached if the calorimeter had not lost any heat at all. This correction is probably 



VOL. XXXIX. PART II. (NO. 12). 3 K" 



