5o6 Journal of Comparative Neurology and Psychology. 



reaction part of one breathing curve with the reaction part of 

 another, and thus reach utterly unjustifiable conclusions.^ Wc 

 may even question the accuracy of comparing reactions after 

 they have been severally expressed as percentages of their re- 

 spective rates immediately before stimulation, for this proced- 

 ure assumes that a given percentage of rate variation has the 

 same meaning regardless of considerable differences in normal 

 rates. Although the gain is the same, we cannot be sure that 

 a rise in rate from 25 respirations per minute to 27.5 per min- 

 ute, for example, has the same significance as an increase from 35 

 to 38.50. The possible error arisingfrom this source in the pres- 

 ent investigation is lessened by the fact that, in case of any one 

 animal, the average normal rates did not differ greatly, as will 

 be shown in the tables. 



/. Reactions to light of different quality. Each of the 

 ten animals used was stimulated once daily for 25 successive 

 days with red, yellow, green and blue light. To prevent the 

 order of stimulation from influencing the reactions, this was 

 changed each day. The colors were given the first day in the 

 succession red, yellow, green and blue ; the second, blue, green, 

 yellow and red ; the third, yellow, red, blue and green ; the 

 fourth, green, blue, red, yellow; the fifth, the same as the first, 

 and so on to the end of the series. The tracings were measur- 

 ed directly to half mm., and the rates before and after stimula- 

 tion were arranged under the four colors. 



The general statement of results is made in Table II. The 

 averages were computed from a total of 15,000 measurements. 

 The line at the bottom of the table shows a constant decrease 

 in the breathing rate before the stimulus (29.6, 29.5, 29.5, 29.2 

 and 29.1). This condition was probably due to the gradually 

 diminishing influence of the kymograph, whose start at the be- 

 ginning of a record always produced a temporary disturbance. 

 Each animal showed a quickening in rate after stimulation. The 

 average maximum reaction was attained in the second respira- 

 tion (32.2), though individual birds varied in this respect. 



^Fere, loc. cit. 



