BUCKINGHAM. 



DIVISION OF LABOR AMONG ANTS. 



461 



horizontal nest (77), but in colony 51 it was smaller in the vertical ( V). 

 For my problem, however, this is unimportant. 



In Table III, column 4, it may be noted that out of 43 ants tested 

 in building, five did not engage at all, while only one reacted more than 

 half the time, the largest per cent being 59. As in tending the young, 

 the methods employed allow only positive results. 



Figure 14 (p. 454) shows a distinct correlation between size of the 

 individual and activity. It will be noticed that this activity is almost 

 confined to the smaller sizes (1.5 mm.-4.5 mm.). 



TABLE VI. 



But here again, as in tending the young, both deiilated and winged 

 queens took part. 



4. Fighting. 



1. Methods. — It was impracticable to deal here with the colony as 

 a whole, since it was impossible in the confusion which occurred to 

 distinguish readily the sizes which took part. In order to discover 

 which of four different known sizes (classes) was more active in 

 fighting, a few individuals of each size were placed in four separate 

 Stender dishes, one dish for each size. Each dish measured 5.5 cm. in 

 diameter and contained the same number (given in Table VI) of ants. 

 The limits of the head sizes of the individuals in each of the four classes 

 is indicated in sq. mm. Into each dish in succession was introduced 



