ESTIMATION OF IRON, ALUMINA AND PHOSPHORIC ACID. 171 



CALCULATED. 



Fe 3 3 *0386 Al 2 O s '0145 P 2 0, '2366 

 In using this method a few precautions, more or less 

 obvious, are advisable in order to obtain good results. The 

 acidity should not be too great, but enough acid should be 

 present to give to the solution a yellow and not a reddish 

 colour in the presence of iron. Monosodic phosphate 

 should be used so that at no time the solution is alkaline 

 to methyl orange during the addition of the phosphoric 

 acid. The standard caustic soda should be added with 

 constant shaking for a similar reason ; as the phosphates 

 are easily acted upon by alkaline solutions, losing a part 

 of their phosphoric acid. 



In testing an original solution to see if any phosphoric 

 acid is present, it is very easy to get an approximation to 

 the amount of that body in solution by using a method 

 which I have found very useful. This consists in preparing 

 standard tubes containing *0001, '0002, '0003, etc. grms. of 

 P 2 5 and treating with a few cc. of ammonic-nitro-molyb- 

 date, keeping to the same bulk and conditions throughout. 

 These tubes can be sealed off and kept for comparison when 

 required. A rough estimation of the amount of phosphoric 

 acid in a solution can be made by suitably diluting and 

 working with quantities similar to those used for the pre- 

 paration of the standard tubes, comparing the densities of 

 the precipitate in the cold. When traces only of phosphoric 

 acid have to be estimated, this method gives the most 

 accurate results. The amount of phosphoric solution 

 required to be added to precipitate the iron and alumina 

 can thus be roughly ascertained. The total phosphoric 

 acid in solution need not exceed twice the quantity neces- 

 sary to form the phosphates of these metals. 



In conclusion I wish to express my thanks to Mr. S. G. 

 Walton, Assistant in the Government Laboratory, Sydney, 

 for his assistance in helping me to carry out some of these 

 analyses. 



