1907] Jennings. — BeJiavior of the Starfish. 163 



On the following- day (after 48 hours without training), ten 

 additional free tests were made, with the following results: 

 & + c, 4 ; c + d, 3 ; a -f ^j 2 ; cZ -f e, 1. Here the effect of train- 

 ing is shown more strongly, the pair h -\- c being used more 

 often than any other. 



Now the starfish was allowed to remain undisturbed for five 

 days. During this time it did not perform the righting reaction 

 at all, and it had had no training for seven days. Ten free trials 

 were then given. The results showed that the habit had disin- 

 tegrated considerably, but the effects of training were still evi- 

 dent. The righting was rather irregular, two adjacent rays not 

 being used in most cases. The rays were used as follows : e, 5 

 times; c, 5 times; h, 4 times; d, 4 times, a, 2 times. The pair 

 h -\- c was used once in combination. Contrast these results with 

 the thirty trials given before training; h is now employed in 

 six times greater proportion, c in three times greater proportion. 



Thus it is evident that in the starfish A the effects of training 

 are very marked. Selecting for training a method of turning 

 which the individual never used naturally, and which indeed it 

 at first obstinately refused to employ, we find th;^t innnediately 

 after a series of 18 less'ons, the starfish used this method spon- 

 taneously five times in succession. After a longer course of 

 training, the effects are still very marked, after the lapse of 48 

 hours, and decidedly evident after a week. 



Starfish B. — In the second specimen employed the results 

 were similar to those described under A, so that the details will 

 not be given. The general results, and certain points of special 

 interest will be set forth. 



B showed at first a strong tendency to use the ray a -\- e. 

 The result of 25 preliminary experiments gave for adjacent rays 

 the following figures: a -f- ^' 15 ; a 4- &, 3 (p. 143). The pairs 

 h -\- c, c -f- cZ, and rZ -(- e were never used. The ray c was em- 

 ployed less often than any other, being used but three times in 

 25 trials. As soon as the starfish was inverted, this ray c imme- 

 diately began to rise and swing over. 



It was determined to attempt to train the animal to turn by 

 attaching with the rays c and d and swinging the body over by 

 their aid, — a method never used naturally. 



