522 Dr. C. Ludeking on the Action of the 



may be collected and analysed. The entire sparking-tube is 

 encased in an air-bath capable of being heated above 100° C, 

 and thus preventing any condensation. The delivery-tubes 

 are firmly clamped and held in position with their ends 

 exactly on the same level. This is effected by boiling the 

 contents of the flask when only about three fourths full of 

 water, and collecting the gases given off at once. If the tubes 

 have their ends exactly on the same level, the volumes of gas 

 delivered from each will be equal to one another. By repeated 

 trial it is possible to accomplish this within limits entirely 

 satisfactory. The water in the trough E is maintained at 

 90° C. or upwards by means of a current of steam, in order 

 to reduce the error resulting from the different solubilities of 

 oxygen and hydrogen as much as possible. 



Each experiment lasted from three to four hours, during 

 which time a constant current of steam was given off from 

 the flask and was subjected to the action of the electric dis- 

 charge, leaping over various intervals between the electrodes. 

 As electromotive power, four Grove cells of Browning's make 

 were used. The Ruhmkorff coil was used without condenser, 

 and by its construction could give only a feeble extra current, 

 because the section of the inducing wire was very large and 

 its length very small in comparison with the induced wire. A 

 Geissler tube inserted to remove the extra-current spark did 

 not in any way alter the results obtained, thus substantiating 

 my anticipations ; consequently it was unnecessary to resort 

 to the usual means of throwing out the extra-current spark. 



When passing the spark at the beginning of an experiment, 

 while there is still air in the apparatus, it is surprising to 

 notice how, in proportion as this is expelled and the atmo- 

 sphere becomes more and more one of pure vapour of water, 

 the resistance to the passage of the spark increases. It would 

 seem that, just as pure water is a very good insulator, so also 

 pure water-vapour offers very great resistance : in fact it is 

 possible to obtain sparks three times as long in air as in pure 

 aqueous vapour. 



At the conclusion of each experiment the gases were trans- 

 ferred to a eudiometer, and analysed in very much the same 

 way as that used by Perrot {Ann. de Chim. et de Phys. vol. xli. 

 p. 161, 1861). This savant conducted a great number of 

 experiments on the electrolysis of aqueous vapour, extending 

 his work over years of time. 



In the light of modern theories concerning the cause of 

 electrolytic decomposition, as enunciated especially by Arrhe- 

 nius, it seemed desirable that these experiments should be 

 repeated. 



