336 



FRANCIS B. SUMNER 



difference of only 0.05° in favor of the warm-room lot. If we 

 include these less reliable readings, the difference becomes 0.24°, 

 in favor of the warm-room lot. 



TABLE 4 

 Nearly adult males (3§ months) 



Passing to the younger males, we find a close agreement be- 

 tween the warm-room and cold-room figures, whichever pair of 

 averages is chosen. The difference of 0.17° between the thirty- 

 second figures is surely not significant, in view of the small number 

 and the variability of the readings. These last range from 35.3° 

 to 38.2° for the warm-room lot, and from 36.2° to 37.5° for the 

 cold-room lot. Indeed, this series, by itself, would have no sig- 

 nificance for our present purposes. 



TABLE 5 

 Females 



