1896.] Hypochlorites to Chlorates. 83 



The decomposition curves A. i, ii and. iii of similar solutions are very 

 nearly similar. There is a close resemblance between the carves B. i and 

 ii. Special stress need not be given to the curve D because the solutions 

 to "which it corresponds were not heated to the same temperature as 

 the other solutions. Referring however, to all the curves it is seen that 

 the greatest depressions — minimum decompositions — lie very nearly in 

 the same vertical line. The exact strength of the solution which de- 

 composes least has not been determined, but it appears to lie interme- 

 diate between 1*5 and T7 per cent of Sodium hypochlorite. The influence 

 of the other constituents can be neglected, as all the solutions contain- 

 ing widely different quantities of chlorate, chloride and free alkali lead 

 to the same general conclusion. Taking 1 6 per cent to be the approximate 

 number, the ratio between the number of molecules of water to sodium 

 hypochlorite is easily calculated : 



EaOCl (molecular weight 745) = 1-6 % 

 H 2 (molecular weight 18) = 98"4 % 



.'. no. of molecules of water : no. of molecules of sodium hypochlorite 



2**:i!:: 263:1. 

 18 74-5 



In other words an aqueous solution of sodium hypochlorite is most 

 stable when the number of molecules of the salt in solution is approxim- 

 ately 0'4 °/ . As the electric conductivity of such a solution has not 

 been determined I cannot venture to offer any opinion as to what 

 bearing it has on Arrhenius' theory of electrolytic dissociation. It 

 seems probable that this number is different for different hypochlorites 

 and from these numbers the ratio of the molecular weights of these 

 compounds may be determined. 



Leaving the curves of percentage decomposition in relation to 

 concentration, the other curves of percentage decomposition with the 

 duration of heating may be shortly noticed (PI. V). All these lead to 

 the same conclusion, namely the actual increase of decomposition during 

 any interval diminishes as the time from which the initial decom- 

 position began increases. The data however, for the present, are not 

 sufficiently numerous for comparison of the different curves. 



Referring to the columns (g) and (Ji) in the table, 1 it is seen that in 

 all cases oxygen and a chlorate are simultaneously produced. The last 

 column shows in the clearest manner that at the end of the first hour 

 the amount of oxygen that is disengaged varies from one-fourth to one 

 half of the quantity of oxygen fixed in the chlorate. As the duration 

 of heating is prolonged the quantity of free oxygen increases, but in no 

 case is this quantity greater than the oxygen of the chlorate, although 



1 Ante,? 81. 



