AMOUNT OF PHOSPHORIC ACID IN THE SEA-WATER. 127 
made up and found to contain 0-0100 mg. To this was added 0-0150 mg., 
and analysis showed a loss of 0-0008 mg. The result is not conclusive as 
the blank on the sea-water was so high. 
To test the effect of varying bulk, two lots of natural sea-water were 
examined, one in its natural state, the other after evaporation to small 
bulk. The amounts found were 0-0318 mg. and 0-0316 mg. respectively. 
A third portion, precipitated by strong ammonia only, gave 0-0312 mg. 
Filtrates from the iron precipitate were also examined. The amount 
found by analysis of two lots of a sea-water were 0-0378 mg. and 0-0391 
meg. To one filtrate 0-0237 mg. was added and 0-0266 mg. found, a gain 
of 0-0029 mg. No more phosphate was added to the other and 0-0025 mg. 
was found, using a very dilute standard. In another experiment the 
figures for the original analyses were 0-0336 mg. and 0-0348 mg. ; 0-0237 
mg. was added to each filtrate and gains of 0-0013 mg. and 0-003 meg. 
were obtained. These gains are all small, the average being 0-0018 mg.., 
half the blank on the reagents. Finally the filtrate from the sample 
mentioned above as having been concentrated to small bulk was 
examined after the addition of 0-0237 mg. The gain in this case was 
higher, 0-0058 mg. 
The fact that the errors on the filtrates, though small, were all positive, 
made it seem possible that there was still phosphorus, though not neces- 
sarily phosphoric acid, unprecipitated. To test this three lots of 500 cem. 
were taken ; two were analysed in their natural condition, giving 0-0336 
mg. and 0-0348 me. The third portion was boiled for three-quarters of 
an hour with 10 cem. of decinormal potassium permanganate in Jena 
glass ; it was then acidified with HCl. and boiled for two hours longer. 
On analysis 0-0558 me. was found, a gain of 0-0216mg. Another sea- 
water gave 0-0415 mg. and 0-0566 mg. for the natural and oxidised por- 
tions, a gain of 0-0147 mg. The filtrate from the oxidised portion was 
analysed without the addition of more phosphoric acid and 0-0018 mg. 
found. A similar experiment made earlier by the gravimetric method 
showed an increase from 0-082 mg. to 0-147 mg. on oxidation. In one 
case an untreated water was found to contain 0-09 mg. by the gravi- 
metric method ; the filtrate was oxidised and yielded a further 0-07 mg 
There are two possible explanations of this increase of the phosphates 
by oxidation. One is that there may be in sea-water a small quantity 
of organic matter which hinders hut does not completely prevent the 
separation of the phosphoric acid by iron; the objection to this is that 
the action would probably be irregular and the duplicate analyses would 
differ more widely than they do. The other, to which the writer inclines, 
is that there is a considerable amount of phosphorus present in forms 
