114 PROCEEDINGS OF THE AMERICAN ACADEMY 



cases is to be preferred to any other mode of filtering, and which has 

 been in daily use in my laboratory for nearly two years, with scarcely 

 a single instance of failure. Iu many of the earlier analyses the phos- 

 phoric oxide was precipitated as ammonio-magnesic phosphate after 

 the addition of citric acid in quantity about equal ta the weight of salt 

 taken. Dr. Gooch's later experiments showed that this method gave 

 results which were about H c / too high when the precipitate — as was 

 always the case — was not redissolved and thrown down a second 

 time. In some cases I have applied this determination as a correction 

 to the direct result of the analysis. 



As in the analyses of the alkaline tungstates, I have usually deter- 

 mined the alkali by difference, making direct estimations only in doubt- 

 ful cases. Ammonia was always expelled by boiling the salt with 

 sodic hydrate, collecting in chlorhydric acid, and weighing as ammonic 

 chloride. Baric oxide was also sometimes estimated by difference. 

 When precipitated from a solution of baric phospho-tungstate by sul- 

 phuric acid, the sulphate always contains phosphoric oxide, and the 

 same is true when ammonic carbonate and ammonia are employed. 

 Phospho-tungstates which are insoluble in water may be resolved by 

 fusion with an alkaline carbonate, preferably by the mixed carbonates : 

 C0 3 KNa. Water is best determined from the loss by ignition, but 

 in some cases I have found it best to ignite with a weighed quantity of 

 borax. 



Twenty-four Atom Series. Phospho-tungstic Acid. — The acid is most 

 conveniently prepared by decomposing mercurous phospho-tungstate 

 by dilute chlorhydric acid. I have found it best to proceed as follows. 

 To the mixture of 24 molecules of normal sodic tungstate and 2 mole- 

 cules of sodic phosphate in solution, after boiling for some time, dilute 

 nitric acid free from chlorine is to be added until the reaction is quite dis- 

 tinctly acid. The solution is then to be precipitated hot by mercurous 

 nitrate in small excess, and the yellow flocky mercurous salt washed 

 thoroughly by decantation with hot water. Toward the end of the 

 operation a few drops of solution of mercurous nitrate may be added to 

 the water, as the washings are otherwise apt to become milky. After 

 washing, dilute chlorhydric acid is to be added in small quantities at a 

 time until the yellow color disappears, and is replaced by the white of 

 mercurous chloride. It is well to set aside a small quantity of the mer- 

 curous phospho-tungstate. and to add this to the solution of the acid so 

 as to insure the separation of any remaining traces of chlorhydric acid. 

 After complete subsidence the supernatant liquid is to be filtered off 

 clear and then evaporated in vacuo over sulphuric acid. The sirupy 



