21 



through the fact that it is primarily the part stimulated that responds. 

 This part contracts or extends, thus partly determining the direction of 

 movement. 



But the localization of the external stimulus is not the only factor in 

 determining the direction of locomotion. Especially in the negative 

 reactions certain other factors are evident, which are of much impor- 

 tance for understanding the behavior. After stimulation at one side or 

 end, the new pseudopodium is as a rule not sent out in a direction 

 exactly opposite that from which the stimulation comes. It usually ap- 

 pears, as we have seen, on some part of the original anterior end of the 

 body, and at first alters the course only slightly. This is evidently con- 

 nected with the fact that only the anterior end is attached to the sub- 

 stratum, and without such attachment locomotion cannot occur. If the 

 pseudopodium were sent out from the unattached posterior part of the 

 body, it would have to overcome the resistance of the contraction existing 

 there, and would have to find the substratum and become attached to it. 

 The new pseudopodium thus starts out from the region of least resist- 

 ance, and in such a way that the new movement forms a continuation 

 of the original one, though in a different direction. If the new direction 

 still leaves the anterior part of the body exposed to the action of the 

 stimulus, then a new pseudopodium is sent out in the same way, still 

 further altering the course. This may continue till the original direc- 

 tion of locomotion is squarely reversed. 



This is the method of changing the course that is usually seen in 

 the reactions to mechanical (Fig. 9), chemical (Fig. 15), thermal, and 

 electric (Fig. 17) stimuli. From Davenport's figures (Fig. 16) it ap- 

 pears to be likewise the method in the reactions to light. 



From these facts it is clear that the direction of movement in a nega- 

 tive reaction is not determined entirely by the position of the stimulat- 

 ing agent or the part of the body on which it acts. The moving Amceba 

 is temporarily differentiated, having two ends of opposite character, 

 while the two sides differ from the ends. These internal factors play 

 a large part in determining the direction of movement ; the present action 

 of Amceba, even when responding to stimuli, depends, as a result of these 

 temporary differentiations, partly on its past action. The new pseu- 

 dopodium will be sent out under most circumstances from some part 

 of the anterior end, only under special conditions from a side, and still 

 more rarely from the posterior end. We have here the first traces of 

 relations which play large parts in the behavior of animals higher than 

 Amceba. Structural differentiations have become permanent in most 

 animals, and as such play a most important role in determining the 

 direction of movement. Further, in practically all animals the past 



