FUNCTION OF VIBRISSAE IN BEHAVIOR OF WHITE RAT 65 



of the daily averages taken in groups of 25. It shows that the 

 errors had not materially decreased since the first initial drop. 

 It will be seen that the entire series closes with an average of 

 1.7 errors per trial. This point was reached in the third, fifth, 

 tenth, sixteenth, and eighteenth average taken. Five times it 

 fell slightly below this point but never below one error per 

 trial. The variation per animal was not great. 



Little has been said of time because it is very difiicult to 

 make time a criterion in discrimination tests. We can put an 

 animal in a maze and take the time from the moment he begins 

 to run and this time has significance, but we have no way of 

 knowing when an animal begins to discriminate. 



The other experiments had shown the influence of the vibrissae 

 in such a problem, this by the elimination of all tactile sensi- 



TrrQlS 26 50 75 ;oo ;26 /50 116 200 225 250 215 300 325 550 515 ^00 H15 H50 

 Fig. 13 — Error curve of rats with infra-orbital nerve cut. 



bility in nose and upper lip reduced the animal to a condition 

 where its ordinary conduct could not define the situation. 



(b) Blind rats. — The general conduct of all of the blind rats 

 differed less than had been expected from that of the groups 

 which had preceded. It was first noticed that they were more 

 active in the use of noses and vibrissae than the others and it 

 was thought that forced as they were to depend entirely upon 

 the olfactory and cutaneous senses for guidance in this situation 

 they would learn the problem sooner. They had the initial 

 stimulus of odor to start them off and they had the only other 

 senses of any avail here undisturbed by any visual distractions. 

 The results did not justify our expectations. The average 

 number of trials in learning was somewhat greater but the aver- 

 age time per trial, if it is significant at all, was twice the normal 

 average. How is this to be explained? We would not think 

 that the loss of vision so slight as that of the rat's would be a 



