PATTERN-DISCRIMINATION IN VERTEBRATES 187 



When the full effect of practice has been obtained, Monkey 

 2, under optimal conditions, can distinguish differences in width 

 of striae of less than 3%. These values are of the same order 

 of magnitude as those obtained by the method of limits on two 

 human observers possessed of unusual skill in photometry. 

 Chick 2 ceased to discriminate when the difference in width of 

 striae was reduced to a value between 33% and 42%. The 

 relatively poor results of the chick were not due to errors of 

 refraction, as both his eyes were emmetropic. 



The discriminative ability shown by the monkey is on the 

 average roughly ten times as great as that shown by Chick 2. 

 His visual acuity, however, is only four to five times as good 

 as that of the same bird. This disparity suggests that differ- 

 ence of width between two systems of visible striae constitutes 

 a more difficult basis of discrimination for the chicken than 

 the mere presence or absence of the striae. The fact that Chick 

 1 did not learn the problem, although width-difference was effec- 

 tive for him when presented with an ineffective difference in 

 direction, lends support to this belief. 



Nothing in the above work indicates that Chick 2 might not 

 have yielded a lower threshold had training been sufficiently 

 prolonged. A later experiment, however, indicates that the 

 chick's susceptibility to improvement under prolonged training 

 is not sufficiently large to affect the order of difference between 

 his results and the monkey's which appears in this work. 



For all the subjects, both human and animal, the relation 

 between absolute size of detail and effective difference in size 

 approximates an analog of Weber's law. 



For the chick, familiarity was a more effective stimulus- 

 characteristic than relative size of detail, and the bird never 

 overcame the tendency to respond on that basis without con- 

 tinued retraining. The monkey eventually learned to respond 

 on the basis of relative size. His results suggest that he is 

 adaptable to study by a method of much greater precision than 

 the method which was actually employed. 



In conclusion I wish to thank Dr. P. W. Cobb for the work 

 of refracting the eyes of the animals, and for making the photo- 

 metric determinations for me; also, Dr. A. G. Worthing and 

 Mr. B. E. Shackelford for their cheerfulness in undertaking the 

 tedious observations. 



