PKOFESSOE H. B. DIXON AND ME. E. C. EDGAR 



pressure slightly above atmospheric. Then the current was discontinued, the tap P 

 cautiously opened, and the gas allowed to escape, first through mercury contained in 

 the tube M and then through solid potassium hydrate. Any gas other than chlorine 

 was then sucked out by the automatic pump, which, during 'the first part of the 

 electrolysis, was kept constantly working. 



The operation of filling the apparatus with chlorine and exhausting was repeated 

 four times. The gas from the first two fillings was not completely absorbed. In 

 preparing chlorine for our determinations we filled the apparatus five times, and tested 

 the fifth by fusing-off the side tube K and opening it under mercury. The absorption 

 was so complete as to leave no visible gas residue. This test assured us that no air 

 was left in our chlorine. The fact that the chlorine first evolved was allowed to 

 escape was a safeguard against the possible presence of bromine or iodine, for any 

 bromide in the silver chloride would have been decomposed by the chlorine, and the 

 evolved bromine would have been carried over with the chlorine first escaping. 



The chlorine bulb (immersed in a cooling mixture of solid carbonic acid and ether 

 contained in a silvered Dewar tube) was then filled with liquid chlorine. The current 

 was increased and the condensation allowed to proceed until the liquid reached the 

 level of a circular line etched on the bulb, when the current was stopped. About 

 37 grammes of liquid chlorine were collected in each experiment. Finally the 

 chlorine bulb was separated by fusion. 



Irregularities, arising in the electrolytic cell, were shown by an ammeter placed 

 in the electrical circuit. We found it advisable, as SHENSTONE says, to prevent these 

 irregularities by frequently reversing the current for a short interval of time, thus 

 shattering any incipient silver tree. 



4. The Chlorine Bulb. 



Chlorine, prepared by the electrolysis of fused silver chloride in vacuo, and dried 

 by phosphorus pentoxide, was condensed by means of a freezing mixture of solid 

 carbonic acid and ether, or by liquid air, in an apparatus shown in fig. 4. The 

 chlorine vessel, which was made of soft glass, consisted of a stout glass bulb, A, 

 holding about 40 cub. centims. To this was attached one limb of a T-piece, made 

 of capillary tubing ; another limb could be fused to the source of chlorine, whilst the 

 third ended in a cul-de-sac (B). 



B was a contrivance by means of which we got over a difficulty, which threatened 

 at one time to bring our work to a premature end. For a long time we were unable 

 to discover any means by which liquid chlorine could be safely weighed, and, at 

 the same time, be under such complete control as to admit of its subsequent regular 

 entry to the combustion globe. The pressure of liquid chlorine at ordinary 

 temperatures is from 6 to 8 atmospheres, and the difficulties of successfully controlling 

 such a pressure by means of a tap were found very great. 





