Lack of plasticity in the behavior of queens of the ant Leptothorax rugatulus emery (Formicidae: Hymenoptera)

Abstract

Eusociality has evolved independently several times (Holldobler and Wilson, 1990; Choe and Crespi,1997). Most cases are known from insects, and over 10,000 species of ants are exclusively eusocial, with the exception of some social parasites (Holldobler and Wilson,1990). In ants, eusociality involves in most cases a pronounced morphological and behavioral differentiation between queens, the female reproductive caste, and the nonreproductive, female workers. In an established colony, queens concentrate solely on egg- laying, while workers perform all other necessary tasks, most importantly brood care and foraging. However, the majority of ant queens perform worker tasks in the early stages of colony establishment when they raise their first brood (Holldobler and Wilson, 1990). Thus, they exhibit a full behavioral repertory early in their life cycle, but later they specialize in egg- laying and abandon most other behavioral tasks. In many ancestral ant species workers can take over the queen role and adopt a respective behavioral profile (e.g., Liebig et al., 1998), and in some species queens have been evolutionary lost altogether (Peeters, 1993). Among more derived ant species, similar transitions are not the rule. However, in a considerable number of species some workers in queenles colonies start egg-laying (by arrhenotokous parthenogenesis), accompanied by changes in their overall behavior (Bourke, 1988; Choe,1988). Thus, functional caste changes from sterile worker role to reproductive queen role are widespread in ants. This behavioral plasticity in workers is probably maintained by opportunistic fitness gains when queens are lost (Bourke, 1988; Liebig et al.,1998). In contrast, almost nothing is known about whether established queens are plastic enough in their behavior to take over worker tasks when needed. Queens are shielded by workers from most sources of mortality (e.g., predators, starvation). Consequently a relative queen surplus may arise after adverse conditions, which would be survived by most queens but few workers. This is particularly true for polygynous ant species, where often a large number ofqueens share one colony. While a relative queen surplus leads to queen excecution in extreme cases (Keller et al.,1989), excess queens could increase their fitness (and probably avoid excecution) by performing worker tasks. Some reports about unfertilized queens that remain in the colony without reproducing and engage in worker tasks such as brood care (Liebig, personal communication; Ortius, 1997) indicate that the behavioral repertory of queens is not invariably reduced to egg-laying with age. In this study, we addressed the question whether mature, reproductively active queens expand their behavioral repertoire to worker tasks when necessary.