Demonstrating new social interactions in ant colonies through randomization tests: separating seeing from believing

Abstract

Colonies of the ant Leptothorax unifasciatus (Latr.) were examined for the presence of spatial associations between the single queen and each individually marked worker. The results present, for the first time, evidence of queen-worker interactions which would be difficult,if not impossible, to detect solely on the basis of direct observations of behaviour. The spatial description of queen activity in nine colonies showed that (1) the queen tended to be the adult ant closest to the centre of the egg pile (the colony centre), (2) she usually had the smallest spatial fidelity zone (SFZ), i.e., zone of movement within the nest that is of limited area and specific to an individual, and (3) her SFZ did not appear to expand or contract seasonally in contrast to the SFZs of the workers. In six of the colonies randomization tests were applied to the distances between the synchronous spatial positions of the queen and each worker during each of 5 days of observation. These tests established that some workers associated positively with the queen while other workers associated negatively with her. The numbers of these two types of queen-worker associations were not significantly different. There was no significant difference between workers in positive and negative queen-worker associations with regard to their positions in the colony over the whole 5-day period of observation. Positions were measured as the median distance of each ant from the colony centre. In some cases, however, workers in either type of queen-worker association had positions significantly nearer to the colony centre than those in no association with the queen. The median distance, over a day, between the queen and a worker tended to be smaller for workers in positive queen-worker associations than for those in negative queen-worker associations.