48 



.ELIZABETH LOCK^YOOD THOMPSON 



the experimenter. The number of trips per day varied con- 

 siderably because it was found that sometimes individual snails 

 refused to start up the labyrinth. At such times they were 

 frequently allowed to remain at the base of the stem for a 

 considerable length of time in the hope that the tests might 

 be continued in regular order. Sometimes, however, work on 

 the individual snail so affected had to be discontinued for the 

 day. When a snail lost its hold on the labyrinth and dropped 

 to the bottom of the tank during its test, it was immediately 

 started up again and this second trip was given the same serial 

 number as the previous one with a small letter after it. The 

 same method of recording the observations was used as that 

 described for the preference tests. A record was also made 

 of the " latent " period, that is, the time between the placing 

 of the animal at the base of the stem and the actual start on the 

 upward path. Although not noted in the tabulations, this record 

 appears to have no bearing on the results of the experiments, 

 since fluctuations were probably due in part to the fact that 

 the snail still retained a small amount of air in its lung, so that 

 its need for a fresh supply was less urgent. In constructing 

 curves based upon these records the percentage of errors or 

 wrong choices for each snail per day was used. An analysis of 

 the records appears to show no relation between the latent 

 periods, oviposition, the time required for the tri]j and the 

 number of errors. 



TABLE X 



Showing the paths followed by three individuals of Physa gyrina Say in attempt- 

 ing to reach the surface of the water for air. U-shaped labyrinth used. The 

 right arm was so short that the snail could not obtain air by following it (punish- 

 ment), while air might be obtained at the top of the left arm (reward). The 

 apparatus was cleaned between trials. 



