[17] CHEMICAL COMPOSITION AND VALUE OF FISH FOR FooD. 247 
mination. “ B” (the air dried material) was employed for determination 
of ash, as well as for those of phosphorus, sulphur, and chlorine. 
Phosphorus. (Phosphoric Acid.)\—About 1 gram of substance was care- 
fully burned in a platinum capsule, with some 10 grams of a mixture 
of equal parts of sodium nitrate and carbonate, previously proven free 
from phosphates. The white mass was dissolved in water, acidulated 
with nitric acid, evaporated, and treated with nitric acid again, the ope- 
ration repeated when necessary to remove chlorine, and the phosphoric 
acid estimated with ammonium molybdate solution. A number of tests 
were made by Mr. Merrill to determine the circumstances in which phos- 
phoric acid, in presence of large amounts of sodium nitrate, as here, is 
completely precipitated. The results will, I hope, be published here- 
after. In brief, at the temperature of the trials, which in each case was 
not far from 29°, two precautions are necessary: use of, a large excess 
of molybdate solution, and allowance of ample time for the precipitation. 
Neglect of these involves risk of loss of phosphoric acid, as is shown 
by the following brief recapitulation of results. The effect of higher 
temperature was not tested. 
z i 
a by § 
on wo > 
n a Ye) 
3 n n 
Time of precipitation and amount of molybdic solu- | a 5 I 
tion used. | = 5 £ 
\ 
N oO oO 
| an oO | (sr) 
a b a b a@ b 
; 4 te . | ) | 
10° Naz HPOsz Solution with no NaNOs yielded | 
OO POTTING. cee serene enon esse net seh. cesses cis <0212)}' 20211 | | | 
10° Nay HPO3 Solution witb 5 germs. NaNOyyielded | 
2) 5 POTN Besser eee mor cinck esses nace: seep eee | .0211 | .0212 
10° Naz HPOs3 Solution with 10 grms. NaNOs yielded | | | 
Spe ring fos eal aa te ts “head | .0207| .0211 | .0211| .0211| .0213| .0212 
10° Naz PHOs Solution with 20 grms. NaNO; yielded | . | 
PLO EV OTMSS Sac os cae fos ak Bae cotter na ciee sine 0190 | .0191 | .0211 | .0211 | .0212] .0211 
Practically we use 25°¢ of molybdic solution, and allow 36 to 48 hours 
for the precipitation. 
Sulphur.—About 1 gram of szbstance is oxidized as for determination 
of phosphorus, the mass dissolved in hydrochloric acid, and the sulphuric 
acid determined with barium chloride. Quite an extended series of de- 
terminations were made to learn the effect of varying quantities of sodi- 
um chloride upon the precipitation of barium sniphate in solutions, hot 
and cold, concentrated and dilute. The details, which the limits of this 
article compel me to defer for subsequent publication, showed conclu- 
sively that, although when precipitated cold from concentrated solutions, 
the barium sulphate is apt to be too heavy, 7. e., to bring down sodium 
chloride; yet, when precipitated hot, or even precipitated cold from 
dilute solutions, it is pure. This is in accordance with the previous 
observations on which general practice is based. Our determinations 
served simply to show the limits of concentration and amount of sodium 
