CALCULATION OF RESULTS. 75 



as can be inferred from the slight difference in their specific gravities. 

 The effect of such difference would be to drive air from the carbon- 

 dioxide absorbers into the closed circuit by virtue of the increased 

 volume of the absorbent. That the increase in volume would be suffi- 

 ciently large to affect our results is, however, very questionable, and 

 owing to the lack of data we have made no correction for it. 



In addition to the possible difference in volumes of the reagent at 

 the beginning and end of the period, there is a loss of water from the 

 carbon-dioxide absorbers corresponding to the amount taken up by the 

 air and absorbed by the last water-absorber. Inasmuch as the quantity 

 rarely exceeds 50 grams per two-hour period, it has been assumed for 

 purposes of calculation that it is the equivalent of distilling 50 grams 

 of water occupying a volume of 50 cc. from the three carbon-dioxide 

 absorbers into the water-absorber, where the 50 cc. become reduced in 

 volume by the contraction taking place when mixed with sulphuric acid 

 to 50 X 0.7 = 35 cc. Thus, even in maximum cases, there is a difference 

 in volume due to absorption of water in the last water-absorber of only 

 15 cc. In the work so far this amount has been entirely neglected. 



It is thus seen that no correction is applied at present for any possible 

 changes in the air volume of the carbon-dioxide absorbers. 



INTERCHANGE OF AIR THROUGH FOOD APERTURE. 



The double door on the food aperture (see fig. 8) makes it possible 

 to put into the chamber or remove from the chamber food and excreta, 

 and the vessels in which they are contained, without causing any great 

 change in the air admitted to or removed from the chamber during this 

 process. When the inner door is closed and the outer door is open, it is 

 assumed that the air in the food aperture has the composition of that of 

 the laboratory. On the other hand, when the outer door is closed and 

 the inner door open, it is assumed that the air in the food aperture 

 has the composition of the air inside of the chamber. It is possible, 

 however, that while air ordinarily diffuses quite rapidly, the composi- 

 tion of air in the food aperture does not change as rapidly or as com- 

 pletely as the above assumptions would imply. 



We have assumed that laboratory air contains 20 per cent of oxygen 

 and 80 per cent of nitrogen, not allowing for a small quantity of water 

 vapor and carbon dioxide. The composition of the air in the chamber 

 is always very different from that of the air in the laboratory, the 

 difference being most pronounced as to the carbon dioxide present, as 

 this may be anywhere from 12 to 60 times the normal amount. The 

 percentage of oxygen is in general lower than normal, at times being 

 as low as 17 per cent, though ordinarily it is not far from 19.5 to 20 



