CAVENDISH. 433 



For this purpose, he introduced a portion of hydrogen 

 gas into a globe or balloon of glass, sufficiently strong 

 to resist the expansive force of the combustion which 

 had often been observed in mines, and also in experi- 

 ments upon a smaller scale, to produce an explosion. 

 He adapted to the globe two wires of metal, fixing 

 them in air-tight sockets, and bringing their points 

 within a short distance of each other in the inside 

 of the globe ; so that, by an electrical machine, he 

 could send the spark or the shock from the one point 

 to the other, through the gases mixed together in the 

 globe. He found that the whole of the hydrogen gas 

 disappeared by the combustion thus occasioned, and a 

 considerable portion also of the common air. Water 

 was, as usual, found in small quantity, and an acid was 

 also formed. He then weighed accurately the air of 

 both kinds which he exposed to the stream of elec- 

 tricity, and he afterwards weighed the liquid formed by 

 the combustion ; he found that the two weights cor- 

 responded with great accuracy. It was difficult to 

 resist the inference that the union of the two airs had 

 taken place ; and it might further have been inferred 

 that the latent heat which held them in an elastic 

 state had been given out, forming the flame which 

 was produced ; and that water was formed by the 

 union of the two airs, having, of course, less latent 

 heat than was required to keep them in a gaseous 

 state ; but Mr. Cavendish did not approve of this 

 manner of stating the conclusion which Mr. Watt 

 had adopted, because of doubts which he had respect- 

 ing the nature of heat.* The residue of the com- 



* Page 140. 



2 F 



