806 REPORT—1904. 
basic slag. With the weak citric acid solution this is not the case, so that the 
results of the two methods of analysis are contradictory. 
It is very remarkable that the weak citric acid solution dissolves more phos- 
phate from basic slag than the acid ammonium citrate solution, although the 
latter contains more free acid. This is doubtless because citric acid contains 
one =COH.COOH group and two —CH,.COOH groups; and the former group, 
being the more strongly acid, is the first to be neutralised by the ammonia. The 
acid ammonium citrate solution therefore acts as a solvent in virtue only of the 
—CH,.COOH groups it contains, while the weak citric acid solution contains 
the stronger =COH.COOH group in addition. The former solution would there- 
fore be likely to have a more discriminating action as a solvent, and to give analytical 
results of greater value as far as the availability of the phosphate is concerned. 
The authors are not satisfied, however, with the use of the acid ammonium 
citrate as a solvent in analysis. In the first place it is extremely troublesome to 
work with, and in the second place it in no way resembles the solvents at work in 
the soil, the soil water being usually alkaline, not acid. The same two objections 
can be urged against the use of a solution of carbonic acid which the authors have 
tried but abandoned. The principal solvents in the soil are undoubtedly bi- 
carbonates, and it therefore occurred to them to make experiments with these. 
Of the bicarbonates of calcium, magnesium, sodium, potassium, and ammonium, 
the bicarbonate of ammonium solution is by far the best solvent for phosphate of 
lime. The following reaction probably occurs :— 
Ca,2PO, + 8NH,IICO, =Call,2PO, + 4(NH,),CO, + 2CaH,2CO,, 
whereas in the case of sodium bicarbonate probably no CaH,2PO, is formed, but 
a mixture of CalIPO, (insoluble) and Na,HPO,; and, at any rate, no complete 
solution can be obtained, however small the quantity of Ca,2PO, taken. 
The presence of ammonium carbonate in the solution of ammonium bicarbonate 
does not influence its solvent action. Ilence, the authors used in their experi- 
ments a saturated solution of the ‘carbonate of ammonia’ of commerce. That 
such a solution is far more convenient to work with than either citric acid or acid 
ammonium citrate is obvious, since it is only necessary after extracting a sample 
of basic slag with the solvent to filter and evaporate to dryness to get a residue of 
nearly pure phosphate and carbonate of calcium, 
The analytical process adopted is as follows:—Digest one gram of the 
sample of basic slag with 200 c.c. of the saturated solution of ‘carbonate of 
ammonia,’ shaking from time to time, for twenty-four hours. The flask should be 
closed with an indiarubber cork, wired on to prevent loss of carbonic acid gas and 
precipitation of calcium carbonate. Filter, and wash the residue with cold water, 
evaporate the filtrate and washings to dryness, and determine the phosphate in the 
residue in the usual way. 
A number of experiments made with various samples of basic slag showed that 
the quantity of phosphate dissolved was not appreciably increased either by 
increasing the time of digestion or increasing the volume of solvent. 
The results of analysis of four samples of basic slag are compared with those 
obtained by other solvents in the following table :— 
| 
No. of Total Tricalcic phosphate | Tricalcic phosphate | Tricalcic phosphate | 
sample | tricalcic dissolved by acid dissolved by 1 per | dissolved by satu- | 
of | phosphate citrate of ammonium!) cent. citric acid solu- rated carb. of ammonia, 
/ basic per cent. | solution per cent. of tion per cent. of solution per cent. of | 
| slag of slag total phosphate total phosphate total phosphate | 
| 
| 1 | 25-94 916 920 10°9 | 
2 | 30-00 89°3 | 91:2 40-4 | 
| 3 36°15 80:3 93-4 10-4 | 
4 43:10 84°6 | 863 40:2 | 
* 150 grams citric acid and 27-9 grams ammonia in 1,000 e.c, 
Further experiments are in progress. 
