PATTERN-DISCRIMINATION IN VERTEBRATES 219 



tion. I hope at some later time to attempt an analysis of some 

 of these theories with reference to this question. In so far as 

 such assumptions are made to apply to animals the structure 

 and function of whose eyes are typified by the dog, I am con- 

 vinced that the foregoing experiments present at least presump- 

 tive evidence to the contrary. 



If the dog's eye is not adapted to detail-vision, and if even a 

 4 to 1 brightness-difference is ineffective, as was indicated by 

 the control tests of 14 September to 19 September, we may 

 seriously inquire what useful purpose his vision serves. Orbeli, 

 Kalischer and Smith have seriously attacked the problem of 

 color-vision. Orbeli 14 used the saliva-reflex method of Pavloff, 

 and attempted conscientiously to control his stimuli. He 

 reports negative results. Kalischer 16 used a food-training method 

 with muscular responses as the criterion. He reports definitely 

 positive results. However, he worked without attempting to 

 exclude certain obvious secondary stimuli, and his own report 

 suggests that it was the latter which were actually effective. 

 Smith's 16 use of colored papers for stimuli is open to fatal criti- 

 cism, since they reflect a considerable amount of light of all 

 wave-lengths, and since we know nothing as yet of the relative 

 stimulating value of the different wave-lengths for the dog. It 

 is not improbable that the range of effective wave-lengths is 

 comparatively short for the dog, as it evidently is for rodents. 17 

 Smith seems to have used great care in the training-procedure, 

 and is inclined to draw a positive conclusion from the results. 

 I cannot see that the latter is warranted by the number of trials 

 given and the percentage of correct responses obtained. Re- 

 garding the above and other tests of color-vision, it seems well 

 to remark that all of them have been made only with test-fields 

 of large area. In nature, different colors are usually presented 

 as parts of patterns. This is especially true of the markings of 

 plumage and hair. Proof of the existence of sensitivity to dif- 



14 Orbeli, L. A. Reflexes conditionnels du cote de l'oeil chez le chien. Archives 

 des sciences biologiques (St. Petersburg), 1909. 



15 Kalischer, O. Weitere Mitteilung uber die Ergebnisse der Dressur als physio- 

 logischer Untersuchungsmethode auf den Gebieten des Gehor-, Geruchs- und 

 Farbensinns. Arch. f. Physiol., 1909. 



16 Smith, E. M. Some observations concerning color-vision in dogs. British 

 Journal of Psychology, vol. 5, 1912-13, pp. 119 ff. 



17 Watson, John B. and Watson, Mary I. A study of the responses of rodents 

 to monochromatic light. This journal, vol. 3, 1913, pp. 1 ff. 



