vii ASSIMILATION AND READJUSTMENT 113 



effect in directing or refining impulse. In some such cases 

 we may have to do with experience of results, but in the 

 main they appear to be instances of primary retentiveness. 

 4. Assimilation. 



In human beings, the operation of experience in 

 simple form is seen when impulsive, reflex or random 

 actions are modified by the pleasure or pain immediately 

 resulting. But this operation of experience is, to all out- 

 ward appearance, as familiar among animals as among men. 

 The burnt puppy, as some one has said, dreads the fire as 

 much as the burnt child. Is it making an assumption to 

 say that both dread it for the same reason namely, that 

 both feel ? If it is an assumption, we must make it pro- 

 visionally, for the sake of compendiously describing the 

 simplest cases of the action of experience. After a pre- 

 liminary description, we may refer to our assumption again. 

 Experience does not, it must be remembered, arise in a 

 vacuum. The senses do not furnish a dark room. On the 

 contrary, experience begins to operate on organisms that 

 have already certain tendencies to act, and its first effect is 

 to modify those tendencies. We find this effect in two 

 related but still clearly distinguishable forms : 

 (a) Selective modification. 



Animals of very low organisation possess, as we have 

 seen, certain type reactions, with which they often 

 appear to respond mechanically to a given stimulus. 

 But in certain instances it is found that they have more 

 than one type reaction to which the same stimulus may 

 give rise, and if one fails the other is tried until success 

 is obtained. Thus a stentor stimulated with carmine 

 will bend aside, and, if it does not succeed in ridding 

 itself of the annoyance, will reverse the movement of its 

 cilia. If this too fails it will contract strongly upon its 

 stalk, and finally uproot itself and swim away. 1 This 

 is one of those cases of persistent effort to avoid dis- 

 comfort which have already been discussed. Some- 

 thing more than reflex action is in question here, because 

 different methods are tried until the result is secured, 

 but what follows is the point of importance for our 



1 Jennings, p. 176. 



