( 1.351 

 (0.53^ 



3 PROCEEDINGS OF THE AMERICAN ACADEMY 



H j J 1.3594 gr. lost on ignition 0.1862 gr. water = 13.70% 

 ' ( 0.5340 gr. gave 0.4155 gr. W0 3 = 77.80% 



jy ( 1.1660 gr. lost on ignition 0.1624 gr. water = 13.93% 

 ' X 0.5981 gr. gave 0.4659 gr. W0 3 = 77.89% 



)94 gr. lost on ignition 0.1862 gr. water = 13.70% 

 .5340 gr. gave 0.4155 gr. W0 3 = 77.80% 



The formula 10 W0 3 . 4Na 2 -j- 23 aq requires 



Calc'd. 1. 2. 3 4. 5. Torcher. Marignac. 



IOWO3 2320 77.80 77.99 77.85 77.80 77.89 77.80 77.82 77.88 



4Na 2 248 8.32 8.13 8.21 8.50 8.18 8.50 8.16 8.39 



23H 2 6 414 13.88 13.88 13.94 13.70 13.93 13.70 13.88 13.53 



2982 100.00 99.86 99.80 



The means of all these analyses may be compared with the three 

 formulas above given. 



W0 3 Na 2 H 2 



For the ratio 10 . 4 . 22 78.27 8.37 15.36 



" " 10.4.24 77.63 8.16 14.21 



" « 10.4. 23 77.80 8.32 13.88 



Means of new analyses, 77.86 8.30 13.83 



The analyses therefore leave no reasonable doubt as to the true 

 constitution of thesalt. The solution of the 10 : 4 sodic tungstate has a 

 distinct acid reaction, but it is very difficult to determine the limits of 

 the basicity in this series, because no salts could be obtained having a 

 number of molecules of fixed base higher than four. On the other 

 hand, the white insoluble mass obtained by igniting the crystalline 

 hydrate must have the formula 



10 W0 8 . 4 Na 2 0, 



and I consider it a true pyro-salt. When boiled for some time with 

 water, the pyro-salt dissolves and the original salt crystallizes from 

 the solution. The case appears to be exactly analogous to that of 

 sodic metatungstate, the insoluble 



4 \V0 3 . Na 2 

 of Scheibler and Marignac, giving the normal sodic metatungstate, 

 4W0 3 . NaX>-4- 10 aq, 



when heated with water in a sealed tube. The reactions of the 10:4 salt 

 with metallic solutions are extremely similar to those of the 12:5 salt, 



12WO a . 5Na 2 + 28aq, 



