322 Dr. T. Andrews on the Heat disengaged during the 



experiment, were introduced into a copper vessel (Plate III. 

 figs. 3 and 4), whose capacity was about 380 cubic centiineti-es. 

 A vessel made of thin sheet copper will resist the force of the 

 explosion of this quantity of even a mixture of olefiantgas and 

 oxygen. It was closed by a screw, as shown in the figure, 

 the head of which is perforated with a conical aperture (the 

 apex towards the outside) to admit a very tightly-fitting cork. 

 Through this cork a silver wire a a passes, and another b is 

 soldered to the side of the screw ; these wires, as shown in the 

 figure, are connected by a very fine platina wire. When the 

 vessel is closed (fig. 3), the first silver wire is brought into 

 contact with a narrow band of copper c c, which surrounds the 

 upper edge of the vessel, but is at the same time insulated 

 from it. 



The vessel containing the mixed gases, and adjusted in the 

 manner described, was introduced into another of larger ca- 

 pacity, which was then filled with water at the proper tempe- 

 rature. The latter was suspended in a cylinder having a 

 moveable cover at both ends, and the whole was finally intro- 

 duced into an outer vessel, also of a cylindrical shape, and 

 which was capable of being rapidly rotated round its shorter 

 axis. The whole arrangement will be understood by inspect- 

 ing fig. 5, in which the several parts of the apparcitus are re- 

 presented. 



Before observing the initial temperature, the apparatus 

 was rotated for sometime in order to establish a complete uni- 

 formity of temperature through all its parts. The apparatus 

 being fixed in the position shown in fig. 5, the thermometer 

 was next introduced through the apertures shown in the lids, 

 and the temperature observed. On the removal of the ther- 

 mometer, the exterior of the apparatus was brought into con- 

 tact with one pole of a voltaic battery, while the other pole 

 was passed through the water till it came into contact either 

 with the central silver wire, or with the copper band {cc, fig. 3). 

 The position of the wires at this period of the experiment 

 is shown in fig. 5. By this arrangement the circuit was com- 

 pleted through the fine platina wire, which, becoming instantly 

 ignited, caused the mixture to explode. The orifice of the 

 calorimeter was then quickly closed with a good cork, the lid 

 of the outer vessel shut down, and the whole rotated for thirty- 

 five seconds, in which short space of time the heat produced 

 by the combination was found to be uniformly distributed 

 lln'ough the apparatus. This rapid distribution of tiie heat 

 was greatly fiicilitated by the presence of a small quantity of 

 water in the inner vessel. The thermometer, ))reviously 

 brought as nearly as possible to the expecteil temperalure of 



