168 PROFESSOR H. B. DIXON ON THE RATE OF EXPLOSION IN GASES. 
hard glass flask. It was washed with a solution of ferrous sulphate, dried by 
sulphuric acid, and collected over mercury. 
Nitric Oxide .—The nitric oxide was made by mixing eight parts of crystallised 
ferrous sulphate with one part of crystallised nitre, and pouring on to this mixture 
sulphuric acid diluted with three volumes of water. If the acid is w^arm the action 
starts at once, afterwards the flask is gently warmed. 
Nitrogen .—The nitrogen was prepared by Vernon Harcourt’s method, in which a 
mixture of air and ammonia is passed through a heated tube containing copper and 
copper oxide. The nitrogen issuing from the tube was washed with sulphuric acid. 
Chlorine .—The chlorine employed in determining the rates of explosion of hydrogen 
and chlorine was prepared by the electrolysis of aqueous hydrochloric acid, and was 
passed, mixed with hydrogen, into the explosion-tube wdthout standing in a holder. 
The gas was washed first with water, and then with sulphuric acid, and wms dried in 
a glass worm and a series of bulbs filled wdth sulphuric acid. The electrolytic cell 
was made entirely of glass, and the washing bottles and drying vessels wmre all fitted 
together with ground glass joints with cups. 
Fig. ]. 
The electrolytic vessel A was made with three tubulures, into which ground glass 
tubes were fitted. Two of tliese, BB, passed into the vessel, and dipped about an 
inch below the level of the acid. Through these tubes passed the gas-carbon elec¬ 
trodes CC, the small annular space between the rods and tubes being made tight wdth 
paraffin. A little mercury poured on top of the carbon rods served to connect them 
with the battery wires. Into the third tubulure w^as ground a gas delivery tube D, 
containing washing bulbs filled wdth w^ater, and passing through a tubulure in the 
sulphuric acid w'ash-bottle E. From E the gas passed through the wmrm F and 
through the bulbs G to a long flexible glass tube leading into the explosion-tube. 
For mixtures containing an excess of hydrogen, the wmsh-bottle E was replaced by 
a vessel fitted with three tubulures similar in shape to A. Two of the tubulures w^ere 
fitted with tubes of the snme size dipping beneath the acid, the third tubulure serving 
