244 



Mr. J. Joly. 



This will afford an idea as to the desirability of making the correction 

 in any particular case. It is evident, too, that this additive correc- 

 tion, 0*000037 gram per c.c, might without further calculation be 

 assumed as the correction in many cases, except very great accuracy 

 be sought. 



The construction of the differential calorimeter, it will be seen 

 from PI. 7, differs from that of the single calorimeter, in being single 

 walled, and having a removable box-like covering of wood, fitted on 

 over all, through which the thermometer is inserted, and which is 

 placed in position after the calorimeter is closed. It is left on through- 

 out the experiment. The obvious use of this box is to favour equi- 

 librium of temperature througho at the calorimeter. Steam is admitted 

 centrally, led from the boiler through a thin brass tube. The dia- 

 meter conferred upon this tube in the calorimeter figured in the 

 plate might with no disadvantage be reduced somewhat from that 

 shown. This steam tube is in three lengths : an elbow piece fitting 

 into the boiler ; a straight horizontal piece of any required length ; 

 and finally a vertical double-tee piece. This last is capable of a rock- 

 ing movement about the axis of its lower horizontal member, so that 

 the upper horizontal member may be shifted either to a central 

 position behind the calorimeter, when it is in the position for throwing 

 steam into the calorimeter, or to one side, when it no longer commu- 

 nicates with the interior of the calorimeter. As drawn, it connects 

 boiler and calorimeter. It is further necessary to provide the means 

 of filling the whole steam pipe before filling the calorimeter. For 

 this purpose the upper horizontal member of the double-tee tube is- 

 furnished with two swinging valves, closing it at each end. Before 

 an experiment, and when the tee tube is turned to one side, the outer 

 one of these is drawn inside for a couple of minutes till the whole 

 steam way has been thoroughly cleared of air and heated throughout. 

 During this time the rear orifice of the calorimeter is closed by means 

 of a stopper of cork, bound in soft cloth. When the steam pipe is 

 heated as described this is withdrawn, and the pipe simply shoved 

 across till it comes to a stop provided. In this movement an auto- 

 matic action lifts the inner valve, shoving it completely to one side y 

 so that there is free way into the calorimeter. This, as will be readily 

 understood, is effected by arranging that the edge of the hanging 

 valve strikes against the projecting tubulure of the short steam way 

 leading into the calorimeter. Subsequently, on inclining the tee piece 

 to one side, the valve resumes its old position, closing the steam pipe. 

 By this arrangement there is little or no leakage of steam, and the 

 operation of turning steam into the calorimeter is effected by one- 

 movement of the hand. 



There is but one exit way to the calorimeter. This is placed 

 centrally at the bottom. Before experiment this is closed by means 



