54 



THE ROYAL SOCIETY OF CANADA 



figures in columns 5 and 8. The gas, when found by difference 

 (column 11), has been calculated from the difference between columns 

 6 and 5, and can be regarded as approximately at normal pressure 

 and room temperature. The figures in column 12, for gas measured 

 and calculated to normal pressure and 15°C. were obtained by cut- 

 ting open the frogs under water, and removing the gas as far as pos- 

 sible. This was transferred to a graduated gas-burette over mercury. 

 In passing through a considerable amount of water any carbon dioxide 

 present would be dissolved ; subsequent treatment with moist sodium 

 hydroxide showed no diminution in volume, indicating carbon dioxide 

 absent. This may in part explain the difference in the figures in 

 columns 11 and 12. In a number of cases the percentage of oxygen 

 was determined by treatment with sodium hydroxide and pyrogallol; 

 the results are shown in column 13. The time during which the frog 

 was buoyant and before death occurred could not be observed in 

 most of the experiments until but two or three frogs survived. As 

 far as could be judged, this time (shown where observed definitely 

 in column 14) increased with the time of survival during the first 

 three or four weeks, and was thereafter irregular (due perhaps to 

 emaciation, and weakness resulting from prolonged immersion); 

 the average time is probably four or five days. 



In the table the results are arranged in order of the gas absorbed 

 (as found by difference). 



TABLE L 



In addition to the results shown in the table itself, particular 

 experiments yielded the following additional data: — 



