64 



FREDERICK S. BREED 



been somewhat facilitated — through a lessening of the tendency 

 to general excitement during the first tests, if in no other way. 

 In view of these considerations there is no conclusive evidence 

 that the previous training of Group I with black-blue facilitated 

 the formation of the small-large habit, albeit Group I averaged 

 5.4% less errors per day than Group II. 



It should be stated in connection with the above size work 

 that the results indicate difficulty of acquisition. The behavior 

 of the chicks pointed in the same direction. So, any more general 

 conclusion than we have ventured would have to be made sub- 



10 



9 

 8 

 7 

 6 

 5 

 4 

 3 

 2 

 1 



10 11 12 13 14 VS 16 



Figure 10 — Rate of modification in size training, arranged to test the effect of 

 previous color training. Curve I shows the rate of acquisition by Group I of 

 the small-large habit after having acquired the black-blue habit; Curve II 

 represents the rate of acquisition of the same habit by Group II, the same 

 number and the same age as the chicks in Group I, but without previous 

 color training. Preference tests on first day. Distances along axis of abscissae 

 represent series; along axis of ordinates, average number of errors. 



ject to this further fact. It ought also to be pointed out that 

 chicks of the same age are not ahvays of the same size. The 

 members within each group were of unequal size when six weeks 

 old. Yet, fortunately for our experiment, the groups remained 

 comparable as to size, individual for individual, throughout the 

 tests. This matter is vil^al to the experiment, for naturally 

 chicks of different sizes could pass through the small opening 

 with different degrees of ease. 



