262 PROCEEDINGS OF THE AMERICAN ACADEMY 



chemists, the last portions of water require a very high temperature 

 for expulsion. The details of the method which I employ are as fol- 

 lows. From 0.5 gr. to 1 gr. of the oxalate is to be gently heated 

 until the greater part of the water and carbonic dioxide have been 

 driven off, and then at a full red heat for fifteen to twenty minutes 

 with a blast lamp to a constant weight. During ignition the cruci- 

 ble is best placed at an angle, and partly uncovered to permit free 

 access of air. The mixed oxides do not retain a weighable amount of 

 carbonic dioxide. To determine C 2 3 , from 0.3 gr. to 0.4 gr. of the 

 oxalate are to be weighed into a 250 c.c. flask ; 20 c.c. of water and 

 30 c.c. of dilute sulphuric acid, 1 : 6 by volume, are then to be added 

 and the flask is to be gently heated upon a sand-bath until the solution 

 is complete, when the hot liquid is to be titrated with carefully stan- 

 dardized permanganate. The following analyses will show the corre- 

 spondence between the results obtained by the above described method 

 and those obtained by the sulphate process. In an oxalate from a 

 perfectly colorless nitrate of lanthanum and ammonium sent me by 

 Dr. Shapleigh, — 



(1) 0.3344 gr. gave 0.1009 gr. C 2 3 = 30.15 per cent. 



The means are 30.10 per cent C 2 3 , and 45.625 per cent R 2 3 . 

 The mean atomic mass calculated from the above is 130.70. Dr. 

 Shapleigh found by the sulphate method 139.75, 139.72, 139.67, mean 

 139.71. 



The necessity of thoroughly mixing the oxalates will appear from 

 the following analyses made with oxalates simply washed and dried : — 



(7) 0.3549 gr. gave 0.1274 gr. C 2 3 = 35.88 per cent. 



(8) 0.3697 gr. " 0.1316 gr. " =35.59 



(9) 0.3807 gr. " 0.1382 gr. " = 36.29 



(10) 0.6550 gr. " 0.2869 gr. R 2 3 = 43.79 



(11) 0.6125 gr. " 0.2699 gr. " =44.07 " 



The same oxalates carefully mixed in a mortar were also analyzed 

 for comparison : — 



