Tropisms in Insect Behavior 21 



attach to the right edge of the platform an incline leading 

 to territory that is unfamiliar to the ants. This will 

 cause no cliange in the behavior/ Allow a few moments 

 to be certain that the manipulation has not influenced 

 the actions of the ants and then interchange the two in- 

 clines. On the right you now have an incline that is well 

 scented with the odor of ants ; on the left one that is en- 

 tirely free from such odors. If the behavior of the ants 

 is a chemotropism, the substitution of an unscented for 

 the scented trail should disturb the movements and make 

 it necessary for them to relearn the way home. Such 

 will not be the result. Without a moment's hesitation 

 the ants will continue down and up the incline, although 

 there is no ant odor on it. Evidently that behavior is 

 not a chemotropism. 



Moreover, a careful study of ants shows that in most, 

 if not all, cases the path by which ants leave a nest is not 

 identical with the one by which they return. This is 

 especially true of those ants that have eyes.* 



In that case the trails are often widely separated. It 

 is not intended to ignore the fact that certain species of 

 blind ants deliberately deposit an odor trail which assists 

 in guiding them home.' In that case it is next to impos- 

 sible to determine the exact width of the odor trail ; but, 

 even then, the incoming and outgoing ants travel along 

 parallel rather than identical trails. It seems highly 

 probable that the scented pathway was to one side of the 

 ant that led the others back to the nest If that is so then 

 the odor trail must function as a reference point and not 

 as an inducer of tropisms. 



Enter a dark cellar at night. Turn on the electric 



2 Turner, C. H, Op. cit. Cornetz, V.; Trajet de Fourmis et Retoan 

 du Nid. Inst. Gen. Psychologique, Mem. No. 2, 1910, pp. 1-167. 



3 Santschi, F.; Comment s'Orientent des Fourmis. Reoue Sui»st 

 de Zoologie, Vol. XXI, 1913. pp. 349-426. 



