CHEMICAL COMPOSITION OF FOOD-FISHES. 
755 
upon lack of acid in the muscle, upon the way they were killed, or upon the fact that 
the fishes in question had been frozen and that the flesh had thereby suffered some 
alteration, or whether it was due to other causes, I can not say. In the boiling of 
the extract, complete coagulation was not obtained until a small quantity, 5 to 10 
drops, of acetic acid had been added. The flesh of some fishes gives a second filtrate 
of so slight acidity that boiling alone, without addition of acetic acid, does nob cause 
a complete separation of the albuminoids. On this account the filtration jiroceeds 
very slowly, and there is formed on the surface of the filtrate during the subsequent 
evaporation a casein-like film. This occurs when a little acetic acid had been added 
during the boiling for precipitating the albumen. When the amount of the precipi- 
tated albumen is considerable, from 0.5 gramme to 1.0 gramme, a complete drying 
upon the filter is tedious and difficult. But if the partly dried and reasonably solid 
mass which sticks to the filter is separated in thin layers by means of a sharp pen- 
knife and placed upon a watch-glass, the drying is rapid and complete. 
The filtrate from the precipitated albumen was- evaporated in a platinum capsule 
over a water bath and then completely dried at 110° C. and weighed. Thereafter it 
was ignited at a low heat, generally in the same platinum capsule, which, duriug 
the incineration, was covered with a larger capsule, the latter thus serving as a 
loosely fitting cover. The light-colored ash thus produced was weighed and its 
weight subtracted from the total weight of the dried substance, the loss being taken 
as the amount of the extractive matters. If, however, the direct ignition was not 
satisfactory, the amounts of the soluble and insoluble salts were determined separ- 
ately and their sum subtracted from the total weight of the dried substance. 
The portion of the flesh which did not dissolve in water was carefully removed 
from the filter and boiled for about 12 hours in a porcelain dish with a larger amount, 
500 to 600 cubic centimeters of water. The porcelain dish was covered with a large 
funnel and as the water boiled away it was replaced by distilled water. During the 
boiling, thin yellow films formed on the sides of the evaporating dish. These had the 
appearance of gelatin but were not soluble in boiling water. Boiling in a glass flask 
was not successful on account of the heavy bumping, by which the contents were 
at times projected out of the flask. This occurred once or twice in the boiling in the 
porcelain evaporating dish. The solution of gelatin was filtered boiling hot, the 
insoluble portion boiled again with a larger amount of water, filtered, and washed with 
boiling hot water. The gelatin solution, which was usually light yellow in color, was 
concentrated by evaporation in a porcelain dish tc a small volume, then transferred 
to a platinum capsule and dried, first over a water bath and then in the drying oven. 
The residue had the appearance of a fine jelly, but did not become hard on cooling. 
This gelatin is not entirely pure. It contains especially, besides other matters, 
some salts insoluble in cold water. The gelatin from the plaice, for instance, con- 
tained 3 per cent, of ash corresponding to 0.1 per cent, of the total flesh. The parallel 
determinations from the same flesh gave results which agreed well with one another. 
The perch, for instance, yielded in two determinations 3.71 per cent, and 3.77 per cent, 
of gelatin. The amount of gelatin obtained in the second 12 hours’ boiling, though 
appreciable, is very small in comparison with that yielded by the first 12 hours’ boiling. 
In my opinion , the determinations of the gelatinoids are the least reliable of all. They 
are, nevertheless, sufficiently accurate for comparison, since they were all conducted 
by the same method. 
In regard to the remaining nitrogenous material, Almen remarks : 
The insoluble protein, for reasons easily seen, can not be directly determined. 
On this account, it is usual to determine by differences. The total water-free sub- 
stance minus the sum of the salts, fat, extractives, gelatinoids, and soluble albumen, 
gives the insoluble protein. All the errors which may be made, in the determinations 
of the other substances named, thus appear in the estimation of the insoluble protein. 
