296 G. V. HAMILTON 



which cannot be regarded as expressions of a tendency to seek 

 sexual satisfaction, but which have the essential objective 

 characteristics of sexual activity ? 



(2) Do such sexual reaction- types as homosexual intercourse, 

 efforts to copulate with non-primate animals and masturbation 

 normally occur among any of the primates, and if so, what is 

 their biological significance ? 



It is always a difficult matter to collect scientific data which 

 shall be specifically relevant to behavior problems when such 

 problems do not lend themselves to strictly experimental methods 

 of investigation. Even under carefully prearranged experi- 

 mental conditions one cannot always be sure that adventitious 

 stimuli may not have played a part in bringing about a given 

 response. This uncertainty is much greater w T hen the animal 

 subject is either at large with his fellows or confined w T ith them in 

 a cage of sufficient size to allow a reasonable approximation to. 

 natural conditions. Under such circumstances the best that one 

 can do is to supply a set of conditions which are apt to lead to a 

 fairly definite and uncomplicated development of the desired 

 situation. 



The monkey's marked variability of response presents a further 

 difficulty, for in seeking to identify a definite situation-response 

 sequence the observer is called upon to distinguish activities 

 that are essential components of a given reaction- type from purely 

 fortuitous activities. For example, in a given case a monkey's 

 manipulation of his genitalia may be nothing more than reflex 

 scratching of a momentarily irritated area, and not at all a part 

 of his response to the situation of which the observer wishes to 

 determine the reactive value. In many cases I have been unable 

 to ascertain the essential components of a given reaction-type 

 until prolonged contact with my subjects has enabled me to pre- 

 dict with reasonable certainty that whenever the appropriate 

 situation developed a sequence of activities composed of such and 

 such members would be manifested. 



A difficulty of another kind is encountered when one seeks to 

 present results that have been obtained by non-experimental 

 methods. Such results can seldom be indicated by tables of 

 figures, habit-formation curves, etc., because reaction to 

 " natural " situations are usually complexes of activities which 

 call for detailed description. When, as in the present instance, 



