88 EUPHA FOLEY TUGMAN 



placed at 38 cm. from the stimulus area, thereby giving it a 

 measured intensity of .098 c.p. The variable light was placed 

 at 238 cm. from the stimulus area, thereby giving it an intensity 

 of .0025 cp.. After two consecutive days of correct choices the 

 variable light was moved closer to the standard, which was 

 always in the same position. Thus the intensity of the variable 

 was increased each time, and the difference between the standard 

 and the variable decreased in the same ratio. Each shift in the 

 variable light necessitated the bird learning the problem again. 

 It usually required a shorter time than the original problem, 

 yet each shift was really a new problem to the bird. The differ- 

 ence between the standard and variable was decreased step by 

 step until the bird was unable to discriminate between the two 

 areas. Then the difference between the standard and the vari- 

 able was increased step by step until the bird was able to dis- 

 criminate between the two areas. The threshold of discrimi- 

 nation was taken as the difference between the least discrim- 

 inable difference in the descending series and the least discrim- 

 inate difference in the ascending series. 



Full records were kept during the progress of ' each series, 

 showing the relative positions of the standard light at each 

 trial, the time consumed in making each choice, the success or 

 failure of the trial, and full data regarding the actions of each 

 bird during each trial. The characteristics of the individual 

 birds, methods of learning, etc., and the tables and results are 

 taken from these records. 



(5) Calibration of the lights.^— The light sources and the stim- 

 ulus areas were calibrated near the beginning of and frequently 

 during the investigation with a Lummer-Brodhun photometer 

 against a Tungsten light, standardized at 1 c.p. in Hefner units. 



The photometer readings were first taken with the light sources 

 38 cm. 12 back of the ground glass stimulus area. After the in- 

 tensity of the stimulus area was calculated with the light source 

 at this position, the intensities of the stimulus areas were cal- 

 culated for the other positions of the light source. 



The intensities shown in Table II are calculated from the 



12 The first experiments were made before this reading was taken. It was un- 

 fortunate that the light was not set nearer the glass and thus have given a reading 

 of I.e. p. instead of the decimal .098. 



