384 
Proceedings of Eoycd Society of Edinburgh. [sess. 
c.c. of the molybdate of ammonia mixture* was added, and the whole 
placed in a hath with a thermo-regulator at 50° C. for about twenty- 
four hours. The precipitate was then collected, and the phosphorus 
converted in the usual way into pyrophosphate of magnesium. 
In our first experiments we aimed only at determining the phos- 
phorus, while later we determined carbon and hydrogen in addition. 
The following are the chief results we have as yet obtained : — 
Determinations of Phosphorus only. 
1. In oxide of tribenzyl phosphine — 
(A) 0-4477 gave 0-1582 Mg 2 P 2 O r = 0-04418 P = 9*86 per cent. 
(B) 0*3294 „ 0-1178 „ =0-032899 P = 9-98 „ 
Calculated for (C 7 U 7 ) 3 I J 0 
0-1178 
Obtained. 
A B 
Phosphorus, 9*86 9 -98 
9-68 
2. In dibenzyl-phosphinate of methyl — 
0-2592 gave 0*1088 Mg 2 P 2 0^ = 0*0304 P = 11-72 per cent. 
Obtained. Calculated for C 15 H 17 P0 2 
Phosphorus, 11 -72 11 -92 
3. In tribenzyl phosphine — 
0*3634 gave 0*1352 Mg 2 P 2 0 7 = 0*037758 P= 10*39 per cent. 
Obtained. Calculated for (C 7 H 7 ) 3 P 
Phosphorus, 10*39 10*19 
4. In mono-benzyl phosphinic acid — 
0-2559 gave 0-1608 Mg 2 P 2 0 7 = 0*044908 P = 17*54 per cent. 
Obtained. Calculated for (C 7 H 7 )H 2 P0 3 
Phosphorus, 17*54 18*02 
Determination of Phosphorus , Carbon , and Hydrogen. 
1. In dibenzyl phosphinic acid — 
V 1*4058 C0 2 = 0*3834 C = 67*99 per cent. 
0*5639 gave { 0*3315 H 2 0 = 0*036833 H = 6*53 per cent. 
( 0*2639 Mg 2 P 2 0 7 = 0*073701 P= 13*06 per cent. 
Obtained. Calculated for (C 7 H 7 ) 2 HP0 2 
Carbon, 67*99 68*29 
Hydrogen, 6*53 6*09 
Phosphorus, 13*06 12*60 
These examples show that even when small quantities of the 
phosphorised compounds are submitted to analysis, fairly accurate 
results are obtained — far more accurate, indeed, than might have 
been anticipated, considering the very large quantity of nitrate of 
copper with which the phosphoric acid was diluted previous to 
precipitation by the molybdate. It is probable that with larger 
quantities even better numbers would have resulted. 
* Prepared according to Frsenius’ Manual. 
