INTELLIGENCE OF EARTHWORMS 



337 



interval of about twenty minutes, gradually diminished to five 

 in the later series, separated the trials. 



For several weeks worms No. 2 and No. 1 were given trials 

 alternately in order that they might obscure one another's 

 mucous trails. From the first the sandpaper proved to be 

 slightly repellent to No. 2. Consequently the chemical stimulus 

 was received in only two of the first series of 10 trials. In ac- 

 cordance with chance, the worm turned as often to the right 

 arm of the T as to the left arm. 

 j Four days after the first series, on October 16, a second series 



In 



Tr. 30 



Hr. 11:18 



Out 



/ 



P. s. 



Tm. 64 



V 



>nr™*~> — 



«*"''"' 



E 



lj«-A_a__ 



Figure 3. Diagram of T, showing path followed by earthworm No. 2 in trial 

 30. 



yielded results strikingly similar to those of the first day, and 

 as strikingly in contrast with those obtained on the third day. 



The data for the first three series of trials are presented in 

 table i. In these results it is remarkable that, after only 20 

 experiences in the apparatus, No. 2 should have succeeded, on 

 October 17, in passing thin the T from entrance tube to 

 exit tube 10 times in succession without a mistake in the direc- 

 tion of turning and usually with directness. 



In trial 21, the first on October 17, the worm followed the 

 left wall of the stem of the T as far as the junction of the arms, 



