5o8 Journal of Comparative Neurology and Psychology. 



Later the rate constantly diminished again toward the normal, 

 which was almost reached by the tenth respiration. Animals 

 differed widely also in sensitiveness. Some, for example Q 

 and R, gave very weak reactions, and their rate soon returned 

 to the normal, or even went below it ; whereas others, for ex- 

 ample M and N, gave vigorous responses and did not regain 

 the normal rate by the tenth respiration. 



Table III shows the percentages of reaction given by each 

 bird. Although the respiratory rate before stimulation differed 

 considerably from animal to animal, the four normal rates for 

 each subject (before the four kinds of light stimulation) were 

 usually in fairly close agreement. Thus the rates for S were 

 29.7,29.7,29.1 and 29.4 ; for P, 23.1, 23.2, 23. 5, 23. 5 ; for 

 all, 29.06, 29.16. 29.11, 29.07.' This general harmony war- 

 rants the belief that we have a fairly reliable basis upon which 

 to estimate the amount of reaction. Since the average 

 maximum reactions all occurred in the second respiration, it 

 seemed best to use this respiration throughout in estimating 

 the reactions of individual birds. This reduced the amount of 

 reaction in some cases, but it gave uniformity of treatment by 

 placing each response at approximately the same distance from 

 the beginning of the stimulation. The percentages thus obtain- 

 ed show that a considerable difference existed among the sub- 

 jects, althongh there was clearly a greater sensitiveness to blue 

 than to red. For, while four animals were most sensitive 

 to blue and only one least sensitive to this color, only 

 two animals were most sensitive to red, and three 

 were least so. Yellow caused the highest reaction with 

 three birds, but occasioned the lowest in case of five others. 

 One bird gave its highest reaction to green ; another its lowest 

 to this color. The relative stimulating power of the four kinds 

 of light is brought out more clearly in the final averages, which 

 indicate a gradually increasing color sensitiveness as we pass 

 from the red end of the spectrum to the blue, though there is 

 little difference in the stimulating power of yellow and green. 



*These general averages are carried out to the second decimal place ; this did 

 not seem necessary in case of individual birds. 



