THE QUEEN ANT 297 



Although the foregoing facts belong to ethology rather than to com- 

 parative psychology, it seemed necessary to review them before empha- 

 sizing their bearing on certain general questions. The behavior of the 

 queen ant may be said to depend, first, on a relatively fixed inherited 

 predisposition, or instinct; second, on inherited plasticity or adaptabil- 

 ity; third, on constantly changing physiological states, and fourth, on 

 stimuli which are partly primary and external and partly secondary, 

 internal or true stimuli. These last are probably identical, as sug- 

 gested by Jennings and others, with the physiological states, which in 

 turn are evidently to be conceived as metabolic processes. That the 

 queen ant profits by her prenuptial sojourn in the parental nest to 

 learn by experience, tradition and imitation, I have no doubt, but 

 queens hatched in isolation show that this acquisition is insignificant 

 in comparison with the inherited instincts. These appear as elaborate 

 catenary reflexes, of which the reactions to light and contact stimuli 

 may be taken as examples. In her callow stages the queen is nega- 

 tively phototropic and positively stereotropic, but as the time for her 

 nuptial flight approaches, these reactions are reversed, so that she seeks 

 the light and avoids contact with the walls of the nest. After fertil- 

 ization she again returns to the prenuptial condition — she shuns the 

 light and tries to bury herself in the soil or under stones. These reac- 

 tions, first described by Loeb, are as irresistible as they are adaptive. 

 It can be shown, moreover, that these changes in tropisms are accom- 

 panied by changes in other instincts. My attention was first directed 

 to the stereotyped character of these reactions in the queen ant by a 

 simple experiment. I found that merely pulling off the wings with 

 the tweezers caused the insect to pass at once from positive photo- 

 tropism and negative stereotropism to the reverse. This shows that 

 the change is not caused by fecundation, since artificially dealated 

 virgin queens went through the complex catenary reflex of founding 

 a colony with the same precision as fertilized individuals. 



These and other observations, which I am unable to give in the 

 space at my disposal, all point to constantly changing metabolic states 

 as the mainspring of the queen ant's behavior. She is, in fact, a 

 veritable chemical laboratory, in which we can see more clearly than 

 in many other animals, a direct relation between behavior and the flux 

 of metabolism. 



I hasten over this matter to another general problem. The dis- 

 covery that the queen ant really possesses, at least in potentia, all the 

 instincts of the worker, besides others peculiar to herself, puts a dif- 

 ferent construction on a matter which has long been puzzling theoret- 

 ical zoologists. It has been taken for granted that worker ants are 

 necessarily sterile and that they possess morphological, physiological 

 and psychological characters not represented in the queens of their 



