LIGHT-ADAPTATION IN THE PLANARIAN 243 



D-turning was to reduce it from 78% to 38%. Very roughly, 

 then, the effect of compensatory movement may be considered 

 to be proportional to the difference between these two sets of 

 figures; i.e., 14% at first, 36% at the end. 



The increase in compensation is greater for many individuals 

 than the average would lead one to expect. Six individuals 

 (B, C, D, E, G, and H) in the last fraction actually turned 

 through a greater angle toward the light than away from it. 

 This change of direction is entirely as might be expected, so 

 long as the preponderance of preceding continuous movement 

 (as measured by the percentages of distance) has been away 

 from the light, for it is obviously intrinsic in compensatory 

 movement that it should be in the opposite direction from the 

 greater amount of movement. 



An apparent exception in the case of E may be explained 

 as follows. In the first ten minutes, 65% of D-movement may 

 have resulted in sufficient compensation to reduce the angular 

 excess in the same fraction to 31% in the D-direction and to 

 change the angular excess in the succeeding five minutes to 

 30% in the L-direction. By this time, however, the light sen- 

 sitivity may have become so reduced as to reduce the contin- 

 uous movement almost to zero. The actual distance toward 

 the light passed through in the compensatory turns would 

 account for the positive value of g% in the L-direction. The 

 explanation in the other cases is obvious. 



That compensation finally disappears is shown, it is thought, 

 in Table IV. In the column of angular measure it will be ob- 

 served that the planarian at first changes in each period from 

 the D- to the L-direction under directive light from the right, 

 but that later the tendency is to turn first in the L- and then 

 in the D-direction. This behavior accords perfectly with our 

 hypothesis of the final lapse of compensation, if we assume 

 that in each of the later periods the planarian is just so fatigued 

 for D-turns that it responds at the start with a high degree of 

 compensation, which, under increased fatigue, soon drops out 

 altogether. That it repeats this course in successive periods may 

 be due to the fact that the interval of turning under normal 

 light in the opposite direction partially rests it. 



It thus appears that we can account for all the facts observed 

 by a theory of abrupt turns given in compensation to preceding 



