ON THE ATOMIC WEIGHT OF CHLORINE. 



193 



B- 



Fig. G. Gas-analysis apparatus. 



globe, so it was evident that the error in our volumetric determinations of iodine 

 must have been small. The alkaline sodium thiosulphate solution in the wash-bottle 

 was exactly neutralised with very dilute hydrochloric acid, and the residual 

 thiosulphate estimated with the standard iodine solution. 



The determination of the residual gases was effected as follows : Fig. 6 is a sketch 

 of the gas-analysis apparatus employed.* It consisted of a graduated pipette A 

 attached to a bent capillary tube, with stopcocks C and D and a graduated tube B. 

 The weight of the apparatus, filled with mercury from ' 

 the tap C to the end of the capillary tube E, having 

 been determined, the whole was filled with mercury and 

 placed in the trough. The gases, sucked out by the 

 pump from the combustion globe, were collected in B 

 and passed into A. It was assumed that the gases 

 consisted of hydrogen, oxygen, and nitrogen. 



When the residual gases had been collected, the whole 

 was transferred to the balance room and allowed to reach 

 the temperature of the room. During this time the 

 taps C and D (fig. 6) were, of course, left open ; they 

 were then closed. In order to maintain the gases during 

 analysis at not only constant temperature but constant 

 pressure, it was necessary that the height of the mercury 

 in the limb A of the gas-analysis apparatus above the 

 surface of the mercury in the trough should be kept 

 constant. With the aid of the two etched scales on 

 A and B, the divisions of which were 1 millim. apart, this constant pressure could be 

 easily attained by raising or depressing the apparatus in the trough until the mercury 

 in the limb A stood the same height as before above the level of the mercury in 

 the trough. 



When the gases had reached the temperature of the balance room, and the difference 

 in level of the mercury in the limb A and of the mercury in the trough had been 

 noted, the taps C and D were closed, B was emptied and the apparatus was then 

 ready for weighing. The difference in the weights of the apparatus (i.) full of mercury, 

 and (ii.) containing the residual gases of the combustion (corrected for the weight of 

 these residual gases) gave, by an obvious process, their volume. 



After being weighed, the gas apparatus was transferred to the mercury trough, the 

 platinum spiral F was then cautiously heated by an electric current so as to bring 

 about the combination of all the hydrogen with the oxygen. Sometimes the oxidation 

 was attended by an explosion ; this, of course, occurred when the percentage of 

 hydrogen was relatively great. 



As can be seen from Table II., the oxygen was always in excess of that required 

 * This form of gas-analysis apparatus was first used by D. L. CHAPMAN and E. HOPKINSON. 



VOL. CCV. A. 2 C 



