DISCRIMINATIVE ABILITY OF THE TURTLE 11 



predominating. The appearance of these conspicuous lines first 

 suggested to the writer the possibility that Chrysemys marginata 

 might readily distinguish between differences in line markings 

 since a linear design forms so prominent a feature in the decora- 

 tion of this species. 



All of the individuals used were collected around Ann Arbor, 

 Michigan, where much of the work was done. Those shipped 

 to Texas did not well undergo the hardships of the journey, 

 nor have they since reacted as satisfactorily as before the trip. 

 For this reason it has been necessary to discontinue the experi- 

 ments somewhat earlier than had been planned, since their 

 continuance with apparently abnormal animals would vitiate 

 the results of the entire work. 



Fifteen turtles were used during the course of the experiments 

 but the records of only seven will be given. For various reasons 

 the results obtained from work on the others do not justify 

 publication. Some sickened and died before the completion of 

 a series of tests; others became sulky, lost appetite, were slug- 

 gish or developed permanent habits of right or left turning. The 

 following data refer to those directly mentioned in this paper: 



RESULTS OF EXPERIMENTS 



The experiments may be divided into three groups, each 

 group being characterized by the nature of the test to which 

 the animals were subjected. At the beginning of the work the 

 writer desired to obtain some estimate of the ability of Chrysemys 

 to form associations. For this purpose the Black and White 

 Discrimination tests were tried, since similar experiments had 

 been conducted by other workers with several other animals. 

 The results of these tests do not bear so particularly upon the 

 problem in hand but are here given for the sake of those who 

 care to make comparisons. The other two groups, those show- 



