106 



LONDON PURPLE. 



Owing to the considerable variation in results and the difficulties encountered in 

 the carrying out of the present methods for the determination of total arsenic oxid in 

 London 'purple, methods for this substance have been receiving considerable atten- 

 tion from the association for several years. The three principal objections to the 

 methods thus far proposed are: (1) The difficulty in reading the end point when using 

 up the excess of liberated iodin with sodium thiosulphate after the reduction of the 

 arsenic oxid To arsenious oxid. (2) The difficulty in reading the end point in the 

 final reaction on adding the standard iodin solution. (3) The great tendency to 

 foaming on adding sodium carbonate to neutralize the acid. Both the first and second 

 are caused entirely, and the third largely, by the great amount of organic matter (dye) 

 contained in London purple. Several methods have been proposed, which are more 

 or less successful, for overcoming these difficulties. The two which are most used, 

 however, have not given satisfactory results in the hands of all the analysts using 

 them, in many cases the results running several per cent too low. The modification 

 proposed by the referee and tried this year overcomes these difficulties and renders 

 the determination of total arsenic oxid much easier, particularly for one who has not 

 had considerable experience with the present methods. 



In the following tables are given the results obtained by the different analysts, fol- 

 lowed by their comments: 



TOTAL ARSENIOUS OXID. 



Method I is a provisional method and may be found in Bureau of Chemistry Bulle- 

 tin 107, revised, page 28. Method II may be found on page 29 of the same bulletin. 

 The results are very satisfactory, Method II appearing to give slightly lower figures. 



Total arsenious oxid (As 2 3 ). 



TOTAL ARSENIC OXID. 



Method I is a provisional method and may be found in Circular 10, revised, and 

 also in Bulletin 107, revised, of the Bureau of Chemistry, page 28. Method II may 

 be found in Circular 10, revised. Method III is Davidson's modification of Hay- 

 wood's method for removing a part of the coloring matter and may be found in Bulle- 

 tin 107, revised, where it is designated as Method II, Provisional (p. 29). 



Method IV is that proposed by the referee and is as follows: 



Place 2 grams of the .sample in a 200 cc graduated flask, add 5 cc of concentrated 

 nitric acid and 20 cc of concentrated sulphuric acid. Place on a hot plate or over low 

 flame and heat nearly to boiling; after ten or fifteen minutes add powdered sodium 

 nitrate, in small quantities at a time, until all organic matter is destroyed and the 

 solution is colorless. Cool, add about 50 cc of water (to decompose any nitro-sul- 



