A STUDY OF OXYGEN METABOLISM. 443 



might not be included among the experimental pupas selected 

 the following morning. Accordingly, the maximum age of pupae 

 on which determinations were made was 15 or 1 6 hours, a point 

 which must be kept in mind. 



In preparation for the first reading a lot of pupae was first 

 washed in water with a camels' hair brush, then treated with 

 80 per cent, alcohol for 2 or 3 minutes to destroy any adhering 

 yeast cells, rinsed in water and dried on filter paper. After 

 being weighed on a delicate balance they were placed in a cotton- 

 lined basket and suspended in the oxygen-measuring apparatus. 

 Between determinations each lot of pupse was kept in an individ- 

 ual moist chamber with a piece of moist filter paper. Weighings 

 as well as determinations were continued each day until develop- 

 ment was so far advanced that there was danger of flies emerging. 

 As a final step, record was made of the sexes of the flies which 

 issued from each lot of pupae. 



Rates of oxygen consumption were measured with an improved 

 form of the manometer of Krogh (1915), described by Bodine 

 and Orr (1925). Six manometers were used, a single one only 

 being used for a given lot of pupse during the period of pupal life. 

 During readings the manometers were placed in a water bath 

 which was kept at a constant temperature. From day to day, 

 however, the temperature of the bath varied in accord with the 

 temperature of the room, but this fluctuation is not registered in 

 the determinations as oxygen values are always reduced to o C. 

 Calculation of rates has been made on the basis of oxygen 

 consumption per minute of time and both per gram body weight 

 and per single pupa. 



THE "OXYGEN CURVE." 



The duration of pupal life is influenced largely by temperature. 

 An index of this correspondence is afforded by figures for the 

 number of days on which oxygen determinations were possible 

 even though such figures do not represent the actual duration of 

 pupal life. During the first experimental period 4-day pupae 

 were predominant, 5-day pupae occurring rarely and 3-day pupae 

 only to the extent of 7 per cent. At a temperature of 25 C^ 

 on the other hand, the percentage of 3-day pupae was increased 

 to 60 per cent., while ,S-day pupae did not occur. 



