1893.] 



The Electrolysis of Steam. 



101 



which it will be seen that the electrodes at which the hydrogen 

 appears are next each other, as are also those at which the oxygen 

 appears, instead of being arranged alternately as they would have 

 been if two water voltameters had been placed in series. 



A very large number of experiments were made to test the truth of 

 this law, platinum as well as gold electrodes were tried, and I replaced 

 the metal tubes which serve as the electrodes in my apparatus by 

 wires surrounded by glass tubes; so as to make the form of the 

 apparatus approximate as closely as possible to that used by Perrot ; 

 the results, however, were all perfectly definite and uniform ; the 

 hydrogen always appeared in the tube next the positive electrode, the 

 oxygen in the one next the negative electrode. 



A good method of showing the way in which the hydrogen follows 

 the positive electrode is to keep the coil on one way until about 2 c.c. 

 of hydrogen have been collected in the tube next the positive and 

 about 1 c.c. of oxygen in the tube next the negative electrode, then 

 reverse the coil ; the volume of residual gas will be found gradually to 

 diminish as the mixed gases are exploded, and this diminution goes 

 on until the hydrogen and oxygen previously collected have quite dis- 

 appeared, and the water reaches right to the top of the collecting 

 tubes. After this, if the sparking is continued, oxygen begins to 

 appear where hydrogen had previously been, and vice versa. 



The reversal of the coil supplies a very useful means of telling 

 whether the apparatus is properly adjusted. If the rates of flow 

 through the two delivery tubes are very different, hydrogen may 

 appear in one of the tubes, and oxygen in the other, from some cause 

 which is not electrical ; this can be detected at once by reversing the 

 coil. 



The following table (p. 102) contains the results of some measure- 

 ments of the relation between the excesses of hydrogen and oxygen in 

 the collecting tubes attached to the steam tubes and the quantity of 

 hydrogen liberated in a water voltameter placed in series with the 

 discharge tube. The ordinary vibrating break supplied with induc- 

 tion coils was used, except when the nature of the break is 

 indicated. 



The results tabulated above show that the excesses of hydrogen 

 and oxygen are approximately equal to the quantities of hydrogen 

 and oxygen liberated in the voltameter. 



Medium Sparks. 



When the spark length is greater than 4 mm., the first of the pre- 

 ceding results ceases to hold. The second of these, that the hydrogen 

 comes off at the positive electrode, remains true until the sparks are 

 some 11 mm. long ; but, instead of the hydrogen from the steam being 



