Vapour-density of Bromine at High Temperatures. 13 



mine was then allowed to condense above the stopcock a, and the 

 stopcock h was opened to the air ; a was then turned on, when usually 

 U little bromine entered, and finally began to blow out on reaching the 

 hot neck of the globe ; a was then turned off, and the tubes between 

 a J b, and d cleared of bromine by a current of air • a was again turned 

 on for a few seconds, and the tubes cleared again ; this was repeated 

 imtil no more bromine was seen to come out on opening the stopcock. 

 During this process the neck of the globe and the stopcock were 

 kept warm by heating with a Bunsen burner ; considerable difficulty 

 was sometimes experienced in getting the stopcock free from bromine. 



The globe was thus filled with bromine vapour at the atmospheric 

 pressure and the temperature of the furnace; the latter was read 

 on the galvanometer scale at the moment of opening the stopcock a 

 for the last time. The stopcock c was then closed, and the bromine 

 removed from the tube he by repeated exhaustion. The bromine in 

 the globe C was thus cut off from the rest of the apparatus by the 

 exhausted space he. 



Absorption and Estimation of the Bromine. — A strong solution of 

 potassium iodide was placed in the tube and flasks D and E, and the 

 globe A was exhausted to about one-sixth atmosphere as in the filling, 

 the bromine drawn off" being caught by the potassium iodide. Air was 

 then admitted to the globe through h, and the exhaustion repeated. 



After the seventh exhaustion the tube was washed out through the 

 stopcocks e and /, and the solution added to that from the flasks E, 

 and titrated with a standard sodium thiosulphate solution of about 

 N/5 strength. The solution was standardised by means of pure bro- 

 mine weighed in small bulbs and dissolved in potassium iodide solu- 

 tion. When a series of determinations was made at different pres- 

 sures, the globe was filled in the usual way, and then slowly exhausted 

 to the required amount, the bromine di'awn off being absorbed by 

 potassium iodide solution as usual, except that the absorption flasks 

 were replaced by a straight tube with two small bulbs blown in it con- 

 taining some of the solution ; this arrangement was to obviate the back 

 pressure caused by the solution in the flasks. 



The residual amount of bromine was drawn off as before described. 



Determination of Temperature.— thermo-electric couple was stan- 

 dardised by means of boiling selenium and solidifying potassium 

 sulphate. The latter was melted in a small porcelain crucible by 

 means of the oxygen-gas flame ; the porcelain was not appreciably 

 attacked if the operation was conducted quickly. The galvanometer, 

 a dead-beat d'Arsonval, was obtained from Messrs. Nalder Bros., and 

 had a resistance of about 200 ohms ; the focal length of the mirror 

 was 40 inches. In order to bring the zero point and the reading for 

 the highest temperature employed upon the scale, it was necessary to 

 introduce a resistance of 11,000 ohms, which remained in the circuit 



