468 Rev. P. J. Kirkby on the Union of 



was then thoroughly cleaned by filling the tube first with a 

 solution of caustic potash, and then with a solution of nitric 

 acid. The tube was then washed out with tap-water and 

 distilled water. 



Both the thin wire and the thicker pieces supporting it 

 were of pure platinum, supplied by Messrs. Johnson and 

 Matthey. The thin wire measured *112 mm. in diameter. 

 The thicker pieces supporting the thin wire consisted of 

 wire measuring *5 mm. in diameter. Underneath the whole 

 length of the wire lay pentoxide of phosphorus, which was 

 to absorb water-vapour as soon as it was formed from the 

 gases and which completed the drying of the gases before 

 experiment. This substance is represented by the shaded 

 part of fig. 1 . 



The glass tube thus prepared was fused together at one 

 end. The other end was fused to the tubing through which 

 the mixed gas had to come. 



The mixed gas was generated in a closed vessel. From 

 this it passed through a stopcock into a drying-chamber. A 

 second stopcock admitted the gas into a large mercurial 

 pump, and a third allowed it to pass on into the apparatus 

 of fig. 1 and the McLeod gauges which were permanently 

 connected with this apparatus. The purpose of this third 

 stopcock was to reduce to the smallest dimensions the space 

 where the falls of pressure were to be observed by means of 

 the McLeod gauges. For when a given amount of hydrogen 

 and oxygen combines into water-vapour, and the water- vapour 

 is absorbed, the resulting fall of pressure is, of course, 

 inversely proportional to the space occupied by the mixed 

 gases. 



The space in question measured about 400 c.c, and, when 

 put into connexion with the pump by means of the third stop- 

 cock, could be rapidly evacuated of its contents. 



It should be added that the stopcocks all carried mercury 

 cups so that any leakage of air through the stopcocks into the 

 apparatus was prevented. Moreover, every joint was made 

 by fusing together glass. No pressure tubing was used. 

 Thus every part of the space containing the gas or traversed 

 by the gas was thoroughly air-tight. This was repeatedly 

 verified. When the apparatus had been supplied with dry 

 hydrogen and oxygen at any appropriate pressure, the 

 platinum wire LM was heated by means of an electric 

 current passing through it. During the passage of this 

 current, the difference between the potentials at the points 

 A and B was read on a voltmeter whose terminals were 

 connected to A and B. Meanwhile the current, which 



