i4 



THE QUARTERLY REVIEW OF BIOLOGY 



could solve it. Thereupon Rolf proved 

 himself worthy of such opinion by tapping 

 out the proper answer with his paw. 

 This led to a systematic education of the 

 dog involving not merely the simpler 

 arithmetical problems but also square 

 roots and so on. The dog also learned to 

 spell, using taps of the paw to represent 

 letters. It is said that at the suggestion 

 of his mistress he manufactured his own 

 alphabetical system, and it is claimed by 

 his proud owners that, apparently in the 

 interests of efficiency, this system involved 

 the fewest number of taps for those letters 

 which are ordinarily used most frequently. 

 It is difficult to evaluate the so-called tests 

 made of this dog because of their uncritical 

 nature. Every opportunity was given for 

 the operation of the Clever Hans error. 

 Only one incident reported cannot readily 

 be explained on that basis. In this case 

 the cards, on which the problem given 

 the dog was presented, were shuffled by 

 the observers and shown the dog at ran- 

 dom and without any of the observers 

 knowing just which card had been 

 presented, and thus what reply to expect. 

 Unfortunately these conditions were not 

 maintained often enough to yield results 

 of statistical validity. The attention of 

 Claparede, one of the scientists to test 

 the horses of Elberfeld, was called to the 

 dog, and he planned to subject him to 

 critical tests. Probably the most brilliant 

 thing Rolf ever did was to be taken sud- 

 denly ill soon after Professor Claparede 's 

 arrival. 



Other modalities 



Of the other sensory modalities little 

 can be said, since there are at present no 

 data upon which we might base an esti- 

 mate of sensitivity and threshold values. 

 That kinesthetic stimulation is highly 

 important has already been suggested 

 in connection with our consideration of 



the role of vision. It apparently domi- 

 nates visual stimulation in determining 

 the behavior of the dog under ordinary 

 conditions. Beyond the fact that dog and 

 man are sensitive to the same general 

 types of tactual, thermal, gustatory, and 

 electric stimuli little is known. To sum- 

 marize : it seems probable that the average 

 dog is far more sensitive to odors than is 

 man; that he is not strikingly unlike man 

 with respect to sensitivity to sounds; 

 that his vision for still objects is decidedly 

 inferior to that of man, while his acuity 

 with respect to moving objects is great, 

 although there is not sufficient data to 

 warrant a comparison with man in this 

 respect. 



Learning ability 



The learning ability of dogs has been 

 tested by means of the usual laboratory 

 methods. Dogs have been used by Thorn- 

 dike (53), Johnson (zo), and others on the 

 problem box, by Szymanski (49) in the 

 maze, by Hamilton (14) in a quadruple 

 choice device, by Hunter (18) and by Wal- 

 ton (54) on the delayed reaction set-up and 

 by Shepherd (41) in situations involving 

 the pulling of a string to draw food into 

 their reach. Apparently they have yet to 

 be tested by the Yerkes multiple choice 

 method. One gets a clearer impression 

 of the dog's learning ability not by 

 perusal of the quantitative results but by 

 a comparison of these results with those 

 obtained on other animals. 



Without going into detail it may be 

 said that as tested on such devices the 

 dog appears to be superior to the cat, 

 slightly inferior, perhaps, to the raccoon, 

 and probably inferior to monkeys and 

 apes. This last conclusion is, however, 

 based only upon tests which are much 

 more suited to the motor equipment of the 

 monkeys and apes. Even though a dog 

 and a monkey might be equal in intelli- 



