208 INFLUENCE OF ALUMINIUM SALTS ON THE 



but not by sodium hydroxide. On heating some of the washed 

 precipitate on charcoal, allowing it to cool, and heating again 

 after moistening with cobalt nitrate solution, the mass became 

 sky blue on cooling a second time. A qualitative analysis proved 

 this precipitate to contain aluminum only. 



As the only source of aluminium seemed to be the glass balls 

 in the condensing tower, many of the bottom ones being greatly 

 corroded, an analysis of them was made. The sodium carbonate 

 used for fusing was free from aluminium. These balls were 

 found to contain a quantity of aluminium. 



The weight of this white precipitate which was present in 

 the alkaline " sulphur solution " in the form of hydroxide, 

 usually varied from 10 to 40 milligrams. 



Peckham* has observed that when sulphates are precipitated 

 by barium chloride in the presence of aluminium, the p-recipitate 

 is contaminated with aluminium, and hence its weight is greater 

 than that which corresponds to the actual amount of sulphur 

 present. This case is similar to that which occurs in the precipi- 

 tation of sulphates in the presence of ferric salts. The following 

 experiments were carried out to determine whether the amounts 

 of aluminium usually present in the " sulphur solutions" 

 caused any marked error in the sulphur estimation. 



Two sets of experiments were carried out. On the one hand, 

 varying amounts of an aluminum salt were added to constant 

 amounts of sulphate solution; on the other hand, constant 

 amounts of the aluminium salt were added to varying quanti- 

 ties of the sulphate solution. The sulphate in each case was 

 determined by precipitation with barium chloride. 



The sulphate used was the ammonium salt. The strength 

 of this solution was about ten grams per litre. The object of 

 using the ammonium salt was to reproduce as far as possible the 

 conditions existing in the above mentioned sulphur estimation. 

 - The aluminium salt used was the chloride. This solution, 

 which was prepared by dissolving pure aluminium in hydro- 



*Peckham J., Arner. Chem. Soc., 1899, 21, 772. 



