694 Prof. 0. W. Richardson on the 



and barium oxide and a trace of poly sulphides. There was 

 no evidence of the presence of sulphites or thiosulphates. 



General Survey of the preceding Experiments. 



The results which have been obtained are extremely 

 complex and difficult to unravel. I think this is largely due 

 to the very complicated chemical actions which take place at 

 these high temperatures. Of the complexity of the chemical 

 actions which occur there can be no question. For example, 

 when the tube with the hot platinum electrode insulated on 

 quartz was taken down, after an experiment in which sodium 

 sulphate had been tested in air at low pressures, the folio wing- 

 observations were made. The central electrode smelt strongly 

 of H 2 S. The lower end was covered with a black deposit and 

 the upper end with a red salt. The crevices between the 

 platinum and the quartz were full of a white salt which was 

 yellow where it touched the platinum. The black and red 

 deposits were mixed with a background of white deliquescent 

 salt. All of these substances must have been formed by the 

 action of the salt vapours on the hot platinum, as there 

 was nothing on the central electrode when it was put into 

 the tube. 



To help towards a more comprehensive view of the pheno- 

 mena rouo-h data bearino- on the absolute value of the currents 

 obtained have been collected in the following table (pp. 696 

 & 697). The substance -tested is given in the first column. The 

 second column gives the least pressure at which observations 

 could be made, and the third the value of the current at this 

 pressure. The fourth gives the pressure of maximum emission 

 where it occurred, and the fifth the value of the maximum 

 emission. The sixth column gives the temperature at which 

 the observations recorded in the preceding columns were made 

 The eighth column states whether the observations were made 

 by the strip or tube method, and the tenth column gives the 

 value of the highest observed rate of emi.-sion reduced 

 roughly to equal areas, and a standard temperature of 

 $00° 0. 



The numbers in column 10 are to be regarded as indicating- 

 only orders of magnitude on account of lack of precision 

 in the data. The variation with temperature of the emission 

 from these substances is not known, but it was assumed 

 that fair results would be obtained by taking it to be 

 the same as that of aluminium phosphate in each case. 

 According to Garrett's observations the emission from this 

 substance doubles for every 30° rise of temperature in the 



