666 Mr. J. A. Crowther on the Secondary 



the apparatus was pumped out and refilled several times in 

 order to completely eliminate all traces of the previous gas. 



Oxygen and nitrous oxide were obtained from cylinders of 

 the gas, and sulphur dioxide from a bottle of liquid S0 2 . 



Arsine was prepared in the apparatus used for hydrogen 

 from an alloy of zinc and arsenic containing 40 per cent, of 

 the latter. As the evolved arsine always contains some 

 hydrogen, the gas in the apparatus was analysed immediately 

 after the experiment by sweeping it through absorption-tubes 

 containing silver nitrate solution, and weighing the silver 

 deposited. In this way the actual amount of arsine present 

 in the apparatus could be found. 



The liquids whose vapours were measured were obtained 

 from Kahlbaum (with the exception of the sample of nickel 

 carbonyl, for which I am indebted to the kindness of 

 Mr. H. 0. Jones). They were supplied by that firm as pure. 



To introduce the vapours into the gas-chamber a small 

 bulb, fitted with an exit tube with a well-fitting glass tap, 

 was used. This is shown at G (fig. 2). A few cubic centi- 

 metres of the liquid were placed in the bulb, and it was 

 then connected to the entrance tube of the gas-chamber. The 

 latter was pumped down to a fairly low pressure which was 

 read off on the gauge. The radiation due to the small amount 

 of residual air in the chamber could easily be calculated and 

 allowance made for it in calculating the results. The taps 

 connecting the bulb and the gas-chamber were then opened 

 and the liquid allowed to evaporate, the process being hastened 

 if necessary by gently warming the bulb. When sufficient 

 vapour had passed over into the gas-chamber, the taps were 

 closed, and the pressure again read. The difference of the 

 two readings gave the pressure of the vapour in the apparatus. 

 Some time was allowed to lapse before making readings, in 

 order to allow of the equal diffusion of the vapour throughout 

 the chamber. 



Ethylene bromide and stannic chloride, whose vapour 

 pressures at ordinary temperatures were too small to allow of 

 the employment of this method, were introduced into the 

 apparatus by exhausting the gas-chamber, and passing in air, 

 saturated with the vapour at a known temperature by passing 

 through a series of bulbs (such as are used for absorbing 

 carbon dioxide) filled with the liquid, and maintained at a 

 temperature of about 30° by a water-bath, and then through 

 a long coil of glass tubing, moistened with the liquid, and 

 kept at a known temperature (about 16°) by another water- 

 bath. This ensured the complete saturation of the air at the 

 temperature of the cool bath. 



