ALCOHOL CHECK EXPERIMENTS. 



105 



mining accurately the amount of oxygen consumed, either by men or 

 by an alcohol lamp. In Table 3 beyond are recorded, first the date 

 of the experiment, then in succession the number of the period, the 

 amount of oxygen residual in the chamber expressed in liters, and 

 the differences in the amounts of oxygen residual in the chamber at the 

 beginning and end of the different periods, expressed first in liters and 

 then in grams. The weight of oxygen admitted from the steel cylin- 

 der is next recorded, followed by the corrected amount of oxygen used 

 in the combustion, the theoretical amount of oxygen required to burn 

 the weighed amount of alcohol, and, finally, the ratio of the amount 

 found to the amount required, expressed in per cent. 



The same considerations which affect the accuracy of computations 

 of this nature for short periods, i. e. , the possible effect of errors in 

 residual analyses, which were discussed when considering the compu- 

 tation of carbon dioxide, also influence, perhaps in a more marked 

 degree, the computation of the amount of oxygen ; but here, as in the 

 previous case, the errors are more or less compensating, and, generally 

 speaking, the longer the period the less the effect of the residual analy- 

 ses. It would seem that in this experiment the determinations for the 

 oxygen were, even in the short periods, extremely satisfactory. 



TABLE 3. Record of Oxygen in Ventilating Air Current. 



Alcohol check experiment, April 6-7, 1905. 



In the discussion of the tests of accuracy of the complete apparatus, 

 i. e., the respiration calorimeter taken as a whole, the heat-measuring 

 ability of the apparatus must also be shown, and in Table 4, on page 176, 

 we have a complete statement of the accuracy of the ' ' respiration calo- 

 rimeter " in measuring not only the chemical factors carbon dioxide, 

 water vapor, and oxygen but also the heat. The table referred to 

 gives a summary of all the results of this particular experiment. 



