38 ELIZABETH LOCK WOOD THOMPSON 



the surface film when in the least disturbed. After a time they 

 no longer drop from the surface film under any moderate stimu- 

 lation. To secure the expulsion of air from the lungs it is then 

 necessary to use mechanical means. 



(b) The snails recover from the inhibition of the mouth response 

 due to manipulation. This appears in Table III, p. 13. Here 

 inhibition is more marked in the first 60-80 trials, after which 

 the snails recover and the number of reactions per response 

 increases as the trials proceed until it reaches a m.aximum 

 toward the middle of the series. 



(c) A waning of response occurs in the series of trials shown 

 in Table III, p. 13. Following a maximum number of reactions 

 per response toward the middle of the series, the number grad- 

 ually diminishes to the end of the series (figs. 2-7, Plate II). The 

 snails are becoming adapted to the stimulus which is not fol- 

 lowed by its wonted reward. The same thing occurs in the 

 pressure response of trained snails shown in Table IV, p. 25. 

 Here response is entirely lacking toward the end of the series. 



(d) The absence of the response in the final trials on many 

 days' series appears in Table III, p. 13. This may be due to 

 fatigue, but is more likely the result of adaptation. 



Other cases of adaptation in the snail appear in the paper of 

 Dawson (1911) and in the work of Nagel (1894). 



Learning, modifiahility by association. vSnails which gave no 

 mouth response to pressure on the foot, were so modified by the 

 simultaneous application of pressure to the foot and food to 

 the mouth, that they then gave the mouth response to pressure 

 on the foot. The effect of training with the simultaneous stimuli 

 persisted for ninety-six hours after the cessation of training. Ob- 

 jectively this persistence of the training effect simulates memory. 

 Its ps^^chological implication is not discussed here. 



The training process was accompanied by a shortening of 

 the average duration of the individual reactions, in other words 

 in an increase in their rate, as compared with the r^te obtaining 

 when the food stimulus alone was used. This acceleration is 

 no doubt due to the unwonted stimulus applied to the foot and 

 may be taken as evidence (if any is needed) that this stimulus 

 was effective. The training process was accompanied by a 

 reduction in the average number of reactions per response and 



