EXPERIMENT STATION BULLETINS. 561 



The potash pipette was disconnected. The thimble and connecting 

 tube were thoroughly washed out. Then, the phosphorus pipette was 

 connected up so that the gas transfer for oxygen absorption was made 

 in a similar way to the one just described. After this absorption was 

 complete, the remaining nitrogen was measured. The difference be- 

 tween this and the last previous gas volume gave the cubic centi- 

 meters of oxygen absorbed. Knowing the volume of the gas-sample at 

 the beginning, of course the percentage of each gas could be calculated. 

 With proper care, the results were so accurate that three successive 

 estimates of the oxygen percentage in any pure sample could be made 

 with a variation less than one-tenth of one per cent. 



The insect was then placed in a respiration container; the stopper 

 was inserted and mercury allowed to flow from the aspirator-bottle 

 until the heavy fluid entirely filled the rubber connecting tube, and 

 floated the insect up in the container. After that the latter could bo 

 set in place in its glass supporting jar. The aspirator was then raised 

 until almost all the air was forced out of the container, the insect being 

 brought as near the top as possible without injury. Some of the air 

 sample to be used in the experiment was then drawn in with the insect 

 and, after a moment, expelled through the other outlet of the two-way 

 cock. This process was quickly repeated two or three times, when the 

 required amount of air plus enough for an extra estimation could be 

 drawn into the container. This extra volume was quickly transferred 

 to the measuring burette for a final check estimation, and the respira- 

 tion experiment was started. At the end of a certain length of time, 

 the gases with the insect were estimated again by the method all- 

 ready described. Thus the percentage, or the actual amount of carbon 

 dioxide given off and of oxygen used during that time might be found. 

 The percentage found was very accurate since 'all measurements of 

 gases for estimation could be made to within 0.05 of a cubic centi- 

 meter. The amount of air placed with the insect, however, could not be 

 measured so accurately because of the wide diameter of the container. 

 On that account, therefore, the respiratory quotient was usually found 

 from the percentage values of carbon dioxide given off and of oxygen 

 used by the insect. 



A single example will illustrate the manner of recording measure- 

 ments and estimating results : 



First, suppose the air sample taken from a respiration chamber %t the be- 

 ginning of an experiment gave 20.697o oxygen, 79.24% nitrogen and 0.05% carbon 

 dioxide. 



Wt. of two specimens of P. cornutus used = 3.287 grms. 



Exp. began 8 p. m., March 8, 1910— Room Temp. 22° C. 



Used about 200 c. c. air in all. 



Exp. ended 8 a. m., March 9, 1910.— Time, 12 hours. 



Too]<, of the respired air, for estimation: 

 82.8 c. c. 

 80.2 c. c. after KOH pipette. 



2.6 c. c. = 3.14% CO, 

 66.75 after Phosphorus pipette = 80.61%N. 



13.45 c. c. = 16.24%02 

 A second sample tried out as soon as possible after this checked closely, 

 showing just a little more carbon dioxide and a little less oxygen (0.2%) — as 

 was to be expected. 



71 



