462 PROCEEDINGS OF THE AMERICAN ACADEMY. 



for a period of thirty seconds a worker from another colony of Campo- 

 notus americanus, and the number of ants attacking it was noted for 

 each dish. 



2. Observations. — Table VI gives the results of these experiments. 

 The first column gives the number of the colony, the second the num- 

 ber of observations made on each colony. The results for each of the 

 four classes of ants are shown separately ; for each class the first col- 

 umn shows the whole number of individuals of that class under obser- 

 vation, the second column the per cent of activity.* 



Two colonies were observed in these experiments, and in both cases it 

 ■was found that the largest size took part slightly more often than the 

 other sizes. They were, too, more savage in their attacks. For the 

 first time, then, we seem to have an indication of greater activity on 

 the part of the large workers. This might, of course, be due to differ- 

 ences in surrounding conditions, for in the other activities Fielde nests 

 were used (pp. 446, 456, 458), and in fighting, Stender dishes ; but 

 in view of work done in the field on Camponotus pictiis (pp. 468-471), 

 I hardly think that this is the explanation. It seems to me, rather, that 

 it is due to division of labor, though here, again, the distinction is not 

 hard and fast, as all sizes are very active, but the largest workers 

 most so. 



5. Belat ions of Classes to Various Activities. 



Having discussed foraging, tending the young, building, and fight- 

 ing in their various aspects, attention should now be directed to the 

 specific question of the relation between size and the various activities. 

 If we examine Figures 12-14 we see that, while the correlation be- 

 tween the size of the ants and the three first mentioned activities is 

 not in all cases striking, nevertheless, on the whole, the smaller ants 

 are the more active in each of these functions. This is even more evi- 

 dent when the curves given in Figures 12-14 are combined (Figure 

 18) so as to show the total activity for these three occupations for each 

 size. In the case of fighting, on the contrary, it is the larger individ- 

 uals which show the greater activity. 



It is another question whether a given individual shares in all activ- 

 ities, for it is conceivable that certain individuals of a given size-class 

 confine their attention to one mode, and others of the same size-class 

 exhibit theirs in another mode of performance. As a matter of fact, I 



* The per cent active was found in each case by first ascertaining the num- 

 ber of ants active during each observation; the per cent which this number 

 was of the whole number of ants in a given dish was then found, and finally 

 these per cents were averaged for each dish from each colony. 



