MAZE STUDIES WITH WHITE RAT 281 



decreasing the number of trials by 28% and the total number 

 of errors by 27%. The following records were obtained for 

 19 blind animals. The average number of trials involved in 

 learning was 25. A perfect record was secured for the various 

 groups on the 30th trial. An average total of 229 errors was 

 made by each rat. The corresponding values for 27 normal 

 animals were 18, 22, and 144. The generalization that vision 

 may aid in the mastery of a stationary maze contradicts the 

 findings of Watson in his study of kinaesthetic sensitivity. I do 

 not question these results but doubt their universality. In these 

 experiments the records of many blind rats and the average, 

 records of many groups of blind animals do not suffer in a com- 

 parison with the records of normal animals. One of the blind 

 rats mastered the maze more quickly than any of the 27 normals. 

 Two of the blind groups gave as good records as those of three 

 groups of normals. On the other hand, six of the nineteen 

 blinds did worse than the poorest of the 27 normals, and two 

 groups of blinds gave a higher average record than the poorest 

 group among the normals. While some individuals and some 

 groups of blind animals do as well or better than the average 

 run of the normal animals, yet there are many blind rats that 

 do considerably worse than the majority of the normals. When 

 the groups compared are rather large, there is likely to be a 

 number of blind rats with extremely poor records and these 

 cases are responsible for the poor group average. The blind 

 rats exhibit the greater range of individual variability in their 

 capacity to learn. 



Vision aids trained rats to learn the rotated maze, decreasing 

 the values by 35-40%. A group of 10 normals learned the 

 rotated maze in 21.5 trials with an average total error score of 

 110. The corresponding values for three blind animals were 

 33.3 and 190. The rats had previously been trained on an 

 alternation problem. The size of the blind group is too small, 

 however, for a confident conclusion. 



Vision is a detriment with untrained rats in mastering a 

 rotated maze, increasing both number of trials and total errors. 

 A group of six blinds learned the maze in 27 trials with an aver- 

 age error score of 117. The corresponding values for 10 normal 

 rats were 30 and 196. 



6. There are certain other peculiarities of blind rats con- 



