10.— THE CHEMICAL COMPOSITION AND NUTRITIVE VALUES 
OF FOOD-FISHES AND AQUATIC INVERTEBRATES. 
By W. O. Atwater, th. d., 
Professor of Chemistry in Wesleyan University , Middletown , Connecticut . 
INTRODUCTION. 
The Reports of the 0. S. Commissioner of Pish and Fisheries for 
1880 (Washington, 1883) and 1883 (Washington, 1885) contain prelimi- 
nary accounts of the progress of an investigation of the chemical com- 
position and nutritive values of American food-fishes and aquatic inver- 
tebrates, the details of which are given herewith. The investigation in 
its present status includes : (1) Chemical analyses of the flesh of Amer- 
ican food-fishes and invertebrates ; (2) experiments upon the digesti- 
bility of the flesh of fish ; (3) studies of the chemical constitution of 
the albuminoids of the flesh of fish. 
Analyses have been made of the flesh of 123 specimens of American 
fishes belonging to 52 species ; of 3 specimens and 2 species of Euro- 
pean fishes, and of 64 specimens representing 11 species of American 
mollusks, crustaceans, etc. Two of the European specimens were of a 
species included in the American fishes. The total number of specimens 
of fish analyzed is, therefore, 126, belonging to 53 species. These, with 
the invertebrates, make 190 specimens, belonging to 64 species. Since, 
in many cases, analyses were made of more than one specimen, the 
actual number analyzed is larger than these figures imply. 
Nearly 1,000 different species of fish are used for food in the United 
States. When we consider that almost no other analyses of American 
food-fishes have been made (I am aware of but two); that different 
specimens of the same species vary considerably in composition, So that 
several analyses of each are necessary to show the ranges of variation 
and the average composition under different circumstances of locality, 
season, size, age, sex, etc.; and that, from both chemical and economical 
standpoints, studies of the constitution of the ingredients of the flesh 
and of other tissues are needed to make our information at all com- 
plete, it is evident that the work here reported can be regarded as only 
the beginning of a much-needed research. 
Along with these studies, a series of analyses of meats, dairy prod- 
ucts, and other food materials, animal and vegetable, have been under- 
679 
