values were obtained which ranged, for instance, from 1.13 to 1.45 

 per cent of sulphur on a sample which contained 1.56 per cent by the 

 peroxid method, and from 3.91 to 4.50 per cent on a rubber containing 

 4.17 per cent by the peroxid method. Of 75 determinations, each the 

 average of two duplicates, the peroxid value was the highest except in 

 three instances. When the preliminary heating was continued for 

 from two and a half to five hours, rising above 175 C. for at least 

 thirty minutes of that time, it was noticed that the values usually 

 approached the peroxid value by 0.1 per cent. 



These preliminary determinations all indicate that adding reagents 

 in solution, drying, and then ashing causes a loss of sulphur and that 

 the substance must receive a definite preliminary heating before the 

 sulphur can be absorbed by the alkaline reagent. This was noted-after 

 repeated experiments had been made with modifications of Eschka's 

 method and it seemed probable that when the alkaline reagent was 

 added in solution, the mixture dried down, and a coating consisting of 

 an alkaline oxidizing medium was formed, all of the sulphur would be 

 absorbed. This idea was substantiated by subsequent experiments. 



Potassium salts were not tried in any of the experiments as Lunge and 

 Stierling a have found that potassium chlorid dissolves twice as much 

 barium sulphate as sodium chlorid does. The use of sodium carbonate 

 was discontinued after the twelfth experiment as Barlow 6 has shown 

 that as far as the absorption of sulphur is concerned soda can be 

 omitted when magnesia is present. 



The values obtained by these different methods rarely approached 

 those by the peroxid method nearer than 0.1 per cent which suggests 

 that the latter may be too high. It was therefore decided to examine 

 a new lot of samples by the peroxid method, fuse the barium sulphate 

 obtained with soda, reprecipitate, and determine the sulphur by the 

 proposed method. The results obtained are given in Table 1. 



It appears that the first precipitation of the sulphuric acid by this 

 method gives values very nearly 'approaching those obtained by fusing 

 the barium sulphate obtained by the peroxid method and reprecipitating. 

 The differences between the two sets of results thus obtained range 

 from 0.04 to +0.11 per cent (an average of 0.04 per cent) as given 

 in the last column of Table 1 ; these differences are within the experi- 

 mental error. 



G The determination of sulphuric acid by means of barium chlorid in the presence 

 of interfering substances. Report of the Sixth International Congress of Applied 

 Chemistry, 1906, p. 347. 



& Losses of sulphur on charring. J. Amer. Chem. Soc., 1904, 26: 354. 



[Cir. 50] 



