254 E. P. CHURCHILL, JR. 



of thirteen days after the fortieth trial. The time rose from 

 one minute to 3.5, but fell at the third trial to one minute. 



Some tests were made with the positions of the openings in 

 the partitions reversed so that the fish had to swim down to 

 one side when leaving compartment No. 1 and up to the other 

 side when leaving No. 2. One individual from Group I, which 

 had been running the maze in from one to two minutes required 

 12.5 minutes to run it when thus reversed. The fish tried first 

 the regions where the openings had formerly been situated. An 

 individual from Group II was given a trial in the reversed maze. 

 It tried once at the point where the opening had formerly been 

 but turned then and swam directly to the black-banded opening 

 and passed through. A fish from Group III, when the wooden 

 partitions were reversed, tried quite persistently at the points 

 where the openings has been previously situated before finding 

 them in their new positions. 



The fact that in Groups II and III, where the openings were 

 more apparent to the eye, the improvement was more rapid 

 and the final achievement greater than in Group I leads to the 

 conclusion that the sense of sight was a factor in the running of 

 the maze, especially in the earlier trials. However, the fact 

 that the members of Group I learned the maze only a little 

 less rapidly and thoroughly than did those of the other two 

 groups shows that kinesthetic factors were involved in the pro- 

 cess. They probably came to function more and more exclu- 

 sively as the practice went on. That kinesthesia was an im- 

 portant factor is further proved by the behavior when the posi- 

 tions of the openings were reversed and the fish proceeded to 

 the region where they had formerly found the passage-way. The 

 sense of touch also aided, especially in the case where a visual 

 sign was not used and the fish nosed about the netting. As 

 perfection of the habit was more nearly approached the use of 

 the senses of sight and touch fell more and more into the back- 

 ground and when the physiological limit was reached kinesthesia 

 was largely the factor employed. 



SUMMARY 



Goldfish, although lacking a pallium, are capable of forming 

 a definite habit of a moderate degree of complication and of 

 retaining this habit for some time. 



