38 INDIVIDUAL DIFFERENCES AND FAMILY EESEMBLANCES 



at the open gates. The errors are thus given in the second column 

 for each mouse, while the number 1 in the third column merely 

 means that the animal has successfully passed through the open 



gates. The record indicated in the table by: 1 1 and found to 

 occur on the fourth day for both mice, is repeated several times for 

 each animal. This rhythm may be interpreted by saying that the 

 animal successfully passed the first open gate, without making any 

 errors in the first compartment, but ran around to the closed gate in 

 the second compartment and was thus credited with an error before 

 it finished the day's record. By a glance at the table it will be seen 

 that the single error that was made in the second compartment was 

 not eliminated until the tenth day for mouse No. 131, and one day 

 later No. 132 made a perfect record. This same error occurred once 

 more on the fifteenth day for No. 131, and of the twelfth and four- 

 teenth day for No. 132. These examples are typical in that they 

 show that errors in an animal's behavior are not suddenly discon- 

 tinued, but are gradually eliminated. Further, it may be noted that 

 after only two days of the interference test, the eighteenth and the 

 twentieth, No. 131 changed its entire movement habit ; after only six 

 errors it was able to make the correct turn to the left in each com- 

 partment, while before the interference was set the correct turn had 

 been to the right. This record shows a rather unusual amount of 

 flexibility of behavior; only a few animals have been observed to 

 make a similar record. In fact, if such records were found to be 

 common in animal behavior we would have to alter our conception 

 of a kinesthetic, or muscle sense movement; for the quick readjust- 

 ment as exemplified in the behavior of mouse No. 131 would hardly 

 be compatible with our idea of such a sense. The observations of the 

 writer lead him to believe that kinesthesis was the all important ele- 

 ment in adjusting the movements of the animals that were observed 

 in this investigation. It is also to be remembered that in the case of 

 No. 131 any visual, olfactory or gustatory clews, if they existed as 

 such for the animal, were so left under the experimental conditions, 

 that they would favor the production of errors during the interfer- 

 ence period and not tend to their elimination. The behavior of 

 mouse No. 132 in the interference period, with numerous errors, ap- 

 pears to be the more general type of animal reaction. 



Turning again to the subject of the fixed rhythmic type of be- 

 havior it may be seen in Table X. that, in the case of No. 132, the 

 n 1 " 



record which is indicated by 1 1 in the initial learning period, 



