HABIT FORMATION AND FEELING QUALITIES 297 



ness was unable to supply the needed movement and the whole 

 performance became a blur. The second indirect cause of the 

 break is due to the neuro-motor reflex aroused by the intense 

 effort necessary to maintain, if not to excel, the level of the pre- 

 ceding record. Even holding their own produced a nervous and 

 muscle tension to which the brief time and small practice had 

 not inured them. In a former experiment, and as subject I 

 observed, that the nervous strain was often too severe for the 

 coordinating centers, resulting in confusion " hold-ups" and false 

 movements. 



The beginning of the diminishing section has been described as 

 the end of the middle section. The section proper shows a marked 

 uniformity within the limits of the practice, the gains steadily 

 decreasing in amounts. There are four exceptions, J's already 

 mentioned, and B's, C ; s and D's,- the latter only slightly. A 

 description of the graphs expressing the results of P and U con- 

 ditions follows here : 



It will be recalled that the stimulus was given on the odd 

 periods beginning with the third. In the case of C an exception 

 had to be made and he received his stimulus on the even periods. 



Graph 2. A's (P) curve shows rapid gains in the first section 

 to 6P where the middle section begins. This section is of normal 

 length, five periods, and shows relatively wide variations of 

 increase. The final section begins at 10P and shows marked 

 uniformity in the decrease of increments to the end of the prac- 

 tice. The sections of maximum and diminishing gains are com- 

 paratively uniform and in marked contrast to the middle section, 

 thus emphasizing the storm and stress character of the latter. 

 A's M. V. is 2.78 seconds per card. The rapid approach to the 

 higher records is partly due to systematic methods of work and 

 mental traits. The locations of the boxes were known at 5P and 

 the order of the cards at 9P. This curve furnishes no evidence 

 that the stimulus had any effect on the learning process. 



D's curve (P), plan II, of the same graph conforms closely in 

 its main outlines to K's (N) of graph 1. The variations of 

 increase persisted throughout the practice more as a result of 

 fatigue, and of nervous condition than as an effect of the stimulus, 



