This solution lias an alkalinity according to litmus paper 

 of '6 gram Na 2 C0 3 per litre, although "92 gram is present; 

 and there is probably no actual excess of alkali over 

 arsenic, since assuming the compound Na 2 As 2 3 to be 

 formed, we have "92 gram sodium carbonate combining 

 with 1*72 gram arsenious acid, leaving an excess of "28 gram 

 of the latter. 



The figures obtained with litmus paper, though useful for 

 comparative purposes would appear to have no definite 

 significance, and the following method is submitted as being 

 both rapid and reasonably accurate. 



It depends on the fact that a sufficient excess of acid 

 flocculates both the tar and the fine silt particles which are 

 diffused through the dip-fluid when alkaline, this flocculent 

 matter being then removed by filtration, and the filtrate, 

 which will be found practically colourless on dilution, 

 titrated against decinormal sodium hydrate with an indi- 

 cator unaffected by CO L , and As 2 3 . 



To 50 cc. dip-fluid in a small beaker, run in 5 cc. normal 

 sulphuric acid from a burette, stir, filter at once through 

 dry paper, take 10 cc. filtrate with a pipette, transfer to a 

 beaker and dilute with about 100 cc. water ; the solution 

 is now colourless. Add two drops methyl orange solution 

 (cochineal is entirely unsuitable, being rapidly bleached by 

 the arsenic) and titrate to neutrality with decinormal NaOH. 

 The end reaction will be found perfectly sharp. 



Calling n the number of cc. decinormal NaOH used, and 

 /) the number of grams Na,,C0 3 per 100 cc. of dip-fluid, then 



Six samples of dip-fluid were prepared <•<> 

 quantities of arsenic and soda, and gav 

 resnfts by this method :— 



