22 



CORA D. REEVES 



for each fish showed no improvement. Fifty per cent of 

 the choices were to each plate. The fish failed to learn 

 to discriminate between white plates illuminated as 1 to 

 4 (see fig. 7). The early preference for the duller plate 

 in these trials and the slow discrimination of the dace upon 

 other problems (see fig. 13) made it seem probable that 

 with a very long series of tests the fish might learn to dis- 

 criminate. But neither the 150 tests of the fish Bu, nor 

 the aimless behavior of the three fish in their swimming 

 about before the stimulus patches, supports the opinion. 

 Whatever the capacity of horned dace to discriminate 

 intensity differences, they probably make little use of such 



FIG. 10 



Path of the dace along the limiting plane between the area illuminated by light 

 from the red plate and from the white. (See p. 23.) 



differences in discriminating familiar objects. By way of 

 comparison it may be noted that under the same condi- 

 tions of illumination these two large intense plates differed 

 for the human eye when illuminated by slits of 8.0 and 

 8.5 mm., respectively. 



