The Relation of Recruitment Rate to Activity Rhythms in the Harvester Ant, Pogonomyrmex barbatus (F. Smith) (Hymenoptera: Formicidae)

The response of a colony of Pogonomyrmex barbatus to bait depends on the activities of the colony at the time bait is offered. Colony activities are temporally patterned within the morning activity period. Both the peak of nest-maintenance work and the onset of the peak of patrolling occur earlier than the peak of foraging. The rate of recruitment to bait is positively correlated with the numbers of ants engaged in nest-maintenance work and patrolling, but, surprisingly, it is not correlated with the number of ants foraging. Thus the stage in the activity period when bait is offered affects recruitment rate. A colony of harvester ants provides a convenient and intriguing system with which to investigate foraging ecology. Studies of ant foraging ecology make fre quent use of two variables. The first is the number of ants foraging at a particular time. In this paper "foraging" means travelling to or from a colony's seed source on a permanent, cleared path. The second variable is rate of recruitment to bait. In contrast to foraging, recruitment is a short-term process. A temporary trail of ants to a new food source builds up, then disappears when the food is gone. "Recruitment rate" is the number of ants per unit time that come to a food source to retrieve the food after it has been discovered. This study examines the relation between recruitment rate and the temporal pattern of foraging and other activities. Recruitment rate is known to be affected by environmental factors (e.g., presence of other colonies) and characteristics of the food source (e.g., amount, type, density, distance from the nest) (H?lldobler, 1976; Davidson, 1977a, b, 1980), and the hunger level of the colony (Wallis, 1962). However, even after taking these factors into account, some authors have noted a great deal of variability in the recruitment rate of particular colonies (e.g., H?lldobler, 1976). Chew (1976) noted that, at times, actively foraging colonies do not recruit to bait at all. The recruitment rate of Pogonomyrmex colonies has been measured to test hypotheses about interference competition in desert ecosystems (Davidson, 1977b, 1980; De Vita, 1979; H?lldobler, 1976) and to census populations of particular species within the community (Davidson, 1977a; Culver, 1974; Chew, 1977). Taylor (1977) found the "general recruitment level" of P. occidentalis to be a significant factor in his optimal foraging model, but he pointed out that recruitment rate had to be "taken as a given which cannot be predicted by our present level of understanding." The present study examines recruitment rate within the context of other colony activities at the time bait is offered. In previous studies of other aspects of ant Accepted for publication 15 December 1982. 278 JOURNAL OF THE KANSAS ENTOMOLOGICAL SOCIETY behavior, the activities of the colony at the time experimental manipulations were performed determined colony response (Gordon, 1983; Meudec and Lenoir, 1982). The following questions are addressed: 1) Is there a temporal pattern in the various activities of the colony's outside work force? Previous studies of daily activity rhythms have measured the numbers of ants entering and leaving the nest (e.g., Bernstein, 1979; Whitford et al., 1976). Daily rhythms in activities other than entering and leaving the nest have been documented but not investigated quan titatively (e.g., Wheeler and Rissing, 1975; Willard and Crowell, 1965; Levieux and Diomande, 1978). 2) Is recruitment rate correlated with any colony activities at the time bait is offered? If so, how is the colony's response to bait related to its activity rhythms?