THE PHYSIOLOGICAL ANALYSIS OF BEHAVIOR 153 



is selected from among several positions occupied at random. 

 Not only are individual responses the material for selection, but 

 those individuals also whose initial responses are especially favor- 

 able to the rapid attainment of success; that is, the fit survive. 



Now there can be no more doubt of the development of types 

 of behavior with the aid of selection than the development of 

 structural variations with similar aid — and no less. Scanty and 

 uncertain as is the experimental evidence for natural selection, 

 I do not wish to deny that it may have had large influence in 

 organic evolution of whatever kind. 



This is far from agreeing, however, that the precise responses 

 of organisms to what for the sake of brevity may be called 

 directive stimuli, have been effected, in all or in most cases by 

 selection, whether of individual reactions or of individuals them- 

 selves. On the contrary, evidence is accumulating in favor of 

 the view that organisms respond typically without trials; that 

 what have been called, by a figure of speech, trials, are actually 

 definite responses to stimuli that are neglected by the observer; 

 that behavior is the resultant of many stimuli of which the 

 directive stimulus is but one; that, in fact, the tropic response 

 is rigidly determined as to direction by factors which complete 

 analysis may be expected in all cases to bring to light. 



V 



The serious difficulty with a thoroughgoing application of 

 selection theory to behavior is that it so emphasizes certain 

 types of reaction as to divert attention from others, especially 

 from the analysis of those physiological states on which it is 

 generally recognized behavior depends. Nothing illustrates this 

 fact more clearly than the criterion of stimulation that was 

 introduced into studies of behavior with the method of trial. 



Let us return again to Paramecium. How shall we determine 

 when this organism is under stimulation ? It is obvious that 

 our criterion will depend on our reason for ascertaining the 

 fact. If we are interested primarily with selection among such 

 motor reflexes as Paramecium executes upon contact with ob- 

 structions, a sufficient criterion for pur purpose might well be 

 the reflex itself, however pronounced, that would bring the 

 organism into a recognizably new position with reference to 



