342 ROBERT M. YERKES 



right, the averages of the data for the first half (5 or 10) of the 

 trials of each daily series. It appears that more than twice as 

 many contacts with sandpaper and salt were made in "first" 

 trials as in the "first half" of the series. There is evidence of 

 daily improvement in this comparison, but no proof of the 

 acquisition of a habit which lasts from day to day. 



The experimental procedure was further modified in the 

 interval between December 2, 191 1, and January 25, 191 2, — 

 trials 511 to 645 — by the use of as few as five trials in a series, 

 and by the frequent changing of the blotting paper under the 

 T and the washing of the walls of the T in order that " tracking " 

 should be rendered impossible. 



The results demonstrated (a) the superiority of five trials 

 per day over a greater number, and (b) the tendency of No. 2 

 to follow its own mucous path. 



Repeatedly it happened that the worm after being permitted 

 to make several direct trips thru the T, without change of 

 floor, made mistakes, hesitated, and wandered as a result of 

 the substitution of fresh blotting paper and the washing of 

 the walls with tap water. But like the other features of the 

 behavior of the worm, the following of the mucous trail laid 

 down in the previous trip is not constant. In many instances 

 the path is wholly disregarded. Only exceptionally, therefore, 

 could a "perfect" series result from the "tracking" tendency. 

 Moreover, since no systematic study of this tendency was made 

 previous to the five-hundredth trial, it is not certain that No. 2 

 did not acquire its ability to follow its mucous trail. Of chief 

 importance in the present connection is the fact that "track- 

 ing" does not suffice to account for the "perfect " series obtained. 



On January 26 an important improvement in the technique 

 of the experiment was made by the substituting of the electrical 

 stimulus for the chemical. The salt solution had proved un- 

 satisfactory because each time the worm came into contact 

 with it a certain amount was converged to the adjacent blotting 

 paper. The electrical stimulus proved to be cleaner, more con- 

 trollable and therefore preferable. 



The electrical stimulus consisted of the induced current from 

 a Porter inductorium, with secondary set at 6-7 on the scale, in 

 circuit with a No. 6 Columbia dry battery. 



From January 26, 1912, till February 13, —trials 646 to 710 — 



