284 WALTER S. HUNTER 



No such behavior was noted in the present experiments with 

 wooden partitions. Head-on colHsions with the walls never 

 occurred. Aside from this comparison of time records nothing 

 can be said with regard to the learning ability shown in the 

 two sets of experiments. Rouse does not state at the end of 

 which trial his birds had learned the problem. He gave, how- 

 ever, seventeen trials under uniform conditions. But the time 

 record was reduced until the last, and it is difficult to say what 

 comparative conclusions may be drawn. 



//. Second series, — metnory records. At the close of the 

 fifteenth trial with labyrinth A, pigeons nos. 3, 4 and 6 were 

 dropped from the course of the experiments, while the others 

 were continued in labyrinth B. Nos. 3, 4 and 6 were handled 

 and fed regularly each morning. At the expiration of twenty- 

 nine days, they were again tested in labyrinth A. 



The results of these tests are given in table H. It will be 

 noticed that pigeon no. 4 made no errors and continued perfect 

 for fifteen trials. No. 3 made five errors the first trial and was 

 perfect thereafter. All of these errors were those of returning 

 over the true pathway. The wrong alley was never entered. 

 No, 6 was nervous upon being placed in E. Passing into the 

 maze, he turned back and entered the second alley. This served 

 to increase his confusion and for the next two trials he became 

 worse. On the fourth trial, i.e., at the beginning of the second 

 day, he was perfect and remained so until the thirteenth trial 

 when, for some unknown reason, one error was made. These 

 results justify the conclusion that the pigeon's memory (using 

 the term in its biological sense) for this grade and type of prob- 

 lem is practically perfect after the lapse of four weeks. 



Rouse gives no quantitative results for experiments upon 

 memory. He tested it, however, and his conclusion is that 

 associations are permanent for some weeks. 



Porter,* in his experiments with birds, made some memory 

 tests upon one vesper sparrow, a cowbird, and an English spar- 

 row. The birds had learned a maze little if any more complex 

 than the present one in from twenty to thirty trials. After a 

 lapse of thirty days ten trials were given and no bird was perfect. 

 From Porter's method of recording results, it is impossible to 

 say how many perfect runs were made in the course of the ten 



•Op. cit., p. 256. 



