212 NOTES ON THE ANALYSIS 



this reads 1.40% which v/ill be added to the determination of 

 iron and aluminium oxides. 



The filtrates from the silica were combined and the iron 

 and aluminium contained determined in the following way:^ 



A few drops of bromine water were added and the so- 

 lution boiled to oxidize any ferrous iron present, and a con- 

 siderable volume of ammonium chloride added and the so- 

 lution made slightly alkaline with ammonium hydroxide. 

 The precipitated hydroxides were filtered off and washed 

 carefully, dried in the oven and ignited to constant weight. 

 This with the weight of the iron and alumina carried down 

 by the silica gave a weight of .3225 g. or 32.25% of combined 

 iron and aluminium oxides, assuming the iron to be all in the 

 ferric condition. 



The iron content was determined by preparing another 

 sample as before, removing the silica and making the so- 

 lution up to standard volume. Two accurate fractions were 

 then taken and one titrated direct with standard perman- 

 ganate solution to obtain the proportion of ferrous iron. 

 This process is open to some objection as it is very difficult 

 to ensure that no oxidation of the ferrous iron should take 

 place during the processes of grinding the sample and of 

 fusion, and the proportion calculated from it can only be 

 appromixately correct. The other fraction was evaporated 

 with sulphuric acid and the hydrochloric acid expelled. It 

 was then reduced by means of a Jones reductor and titrated 

 with standard permanganate. Thus the ratio of the ferrous 

 iron to total iron was established and from it the proportions 

 of the two oxides calculated, proving to be ferrous oxide 1.72%, 

 ferric oxide 4.51%— total iron 6.23%. 



The calcium and magnesium were determined by the 

 Richards method. This method is based upon the fact that 

 the amount of magnesium occluded by the calcium oxalate 

 precipitate, depends upon the concentration of magnesium 



