Hydrogen with Oxygen at how Pressures. 469 



passed through an ammeter, was read at practically the same 

 time. From these two observations and from the known 

 resistance o£ the voltmeter, it was easy to deduce the joint 

 resistance of the wire and of the platinum pieces which 

 supported the wire. The resistance of those supports was a 

 small quantity which only amounted to a small percentage 

 of the resistance of the wire. Thus* the resistance of the 

 wire, and the temperature to which the current had raised it, 

 was known. It was found that as soon as the current was 

 started the wire almost instantly took up a temperature which 

 remained nearly steady but slightly increased. This 

 temperature is determined by an equation between the heat 

 generated per second in the wire by the current, and that 

 lost per second by radiation and by the convection of the 

 gas. 



Before each experiment the pressure was ascertained by a 

 McLeod gauge. The current was then turned on, and after 

 running for several seconds (about 30 as a rule) was stopped. 

 The resistance of the wire at the moment when the current 

 was broken was known from readings of the voltmeter and 

 ammeter. The pressure was then again determined. If no 

 fall of pressure was detected, the current, slightly increased, 

 was passed again through the wire and then stopped, and 

 the pressure again observed. In this way the electric 

 current was gradually increased and the observations repeated 

 until a fall of pressure was perceived. The first fall of 

 pressure thus detected was, as a rule, quite unmistakable 

 although extremely small. For if some combination had not 

 taken place in the hydrogen and oxygen, a small temporary 

 increase of pressure in the mixed gas, the effect of the hot 

 wire, was noticed just after stopping the current. Moreover, 

 when once the reaction had been started, the temperature to 

 which the wire had to be heated in order to renew the 

 reaction was distinctly lower than the temperature which 

 just initiated the reaction. Thus a repetition of the 

 experiment led to well marked effects, although the 

 chemical combination ceased as soon as the current was 

 cut off. 



The temperature of the wire slowly increased during the 

 passage of the current. The current could, therefore, only 

 be permitted to run for a short time if the temperature at 

 which chemical union began was to be fixed within narrow 

 limits. During these experiments the temperature only 

 increased by about two per cent. Hence, when chemical 

 action was first detected, the mean temperature, to which the 

 wire had just been raised, might, with an error of only one 



