NEW FORM OF BOMB CALORIMETER. 151% 
determinations of heats of combustion can be made quite rapidly 
and with a high degree of accuracy. Each determination of a 
heat of combustion involves: (1) either three or four weighings; 
one of the calorimeter cylinder and water to insure the right 
amount of the latter, one of the substance, one of the iron wire, 
and sometimes one of a small portion of the iron wire left 
unoxidized; (2) twelve observations of temperature of which 
eight are actually used; (3) one titration of nitric acid with 
standard alkali. The calculations for one determination can 
be-made in the intervals of waiting between the observations 
of temperature of the next one. When the materials of which 
the heats of combustion are to be determined are ground and 
prepared for weighing in advance it is not difficult to make 
nine determinations and calculations in a working day of seven 
and one-half hours. The accuracy of the results is very satis- 
factory. Not only do duplicates agree with the greatest close- 
ness, but we have made a considerable number of determinations 
with pure substances such as sugar, and have obtained results 
practically identical with those obtained by Berthelot and by 
Stohmann for the same substances. This identity of results 
obtained in the three laboratories is certainly very favorable 
for their correctness and for the reliability of the apparatus 
and method. 
Table 45 gives the results of a considerable number of deter- 
minations of the heats of combustion of different foods and feed- 
ing stuffs made by the method and with the apparatus above 
described in the course of investigations here. Many of the 
combustions here reported were made by Dr. R. L. Slagle, whose- 
skill and accuracy call for especial recognition. The special 
purpose in collating them is to compare the fuel values as 
obtained from calculation from the analysis and as determined 
by the bomb calorimeter. In ‘calculating the potential energy 
each gram of protein is taken at 5.5 Calories, each gram of fat at 
9.3 and each gram of carbohydrates at 4.1 Calories. The differ- 
ence between the 5.5 Calories here used for one gram of protein 
and the factor 4.1* which.we have ordinarily assumed is explained 
by the allowance made for the nitrogenous materials which are 
incompletely consumed in the body. 

* See Report of this Station, 1890, pp. 177, 178. 
