Melissa L. Thomas

Geographic Affinity, Cuticular Hydrocarbons and Colony Recognition in the Australian Meat Ant Iridomyrmex purpureus

Correspondence to: M.L. Thomas The defence of nests from intruders is critical for maintaining the integrity and survival of social insect colonies. Using a sampling method that attempts to minimise genetic influences, we examined whether geographic affinity (the distance between nests) affects non-nestmate recognition. Experimentally staged conflicts between ’chilled’ workers of Iridomyrmex purpureus from different nests revealed that individuals reacted more aggressively towards intruders from nests further away than from neighbouring nests. Further bioassays using filter papers coated with extracted cuticular hydrocarbons revealed a similar pattern. These experiments, which were conducted blind, demonstrate that recognition between conspecific nonnestmates is mediated by these chemicals. However, other cues may also be important since ants responded more aggressively in bioassays using chilled workers than those using only cuticular hydrocarbons. Non-nestmate aggression in I. purpureus was also influenced by the density of surrounding conspecific nests; workers from sites with a higher density of nests tended to be more aggressive than those from less dense sites. These data suggest that past interactions to some extent dictate workers’ reactions to non-nestmate intruders.

Colony size affects division of labour in the ponerine ant Rhytidoponera metallica

In theory, larger colonies of social insects should have greater colony organisation. While inter-specific comparative studies provide support for this idea, there is little direct intra-specific evidence. We investigated differences in task specialisation between large (>450 workers) and small (<80 workers) colonies of the ponerine ant Rhytidoponera metallica. Observations of individually marked young or old workers revealed greater task specialisation in large colonies. Age polyethism was detected in large but not small colonies. In large colonies, old workers spent significantly more time foraging than young workers did, while young workers spent more time caring for brood. In small colonies, young and old workers spent a similar amount of time foraging and caring for brood. This difference in task allocation patterns in large and small colonies was associated with a difference in contact rates between workers. Workers in small colonies have a lower contact rate between nestmates and a greater variability in time between contacts than workers from large colonies.

Ectomised conductors in the golden orb-web spider, Nephila plumipes (Araneoidea) : a male adaptation to sexual conflict?

Abstract. Male genitalia may facilitate sperm protection by acting as a plug that prevents or hinders future matings. The pedipalps (intromittant organs) of males of the orb-web spider, Nephila plumipes, have a conductor with a peculiarly curved ending and a triangular process near the terminal end. The tip of the conductor, including the process, breaks during most matings and remains inside the female genital tract. We explored the possible function of the conductor as a mating plug using the double-mating sterile-male technique. Our data are not consistent with a plug function because males use only one pedipalp in each mating, thus leaving an unobstructed insemination duct available for future matings; conductors of males mating with virgin females are not more likely to break than those of males mating with mated females, and second males show no preference for used or unused spermathecae. In addition, males that inserted their palp in the insemination duct that contained a tip of the conductor from a previous male obtained a share in the paternity of the females clutch of eggs. Interestingly, the conductor is more likely to break if it is inserted in an unused spermatheca. We argue that several lines of evidence suggest that the conductor breaks as a result of intersexual conflict over the duration of copulation.

When supercolonies collide: territorial aggression in an invasive and unicolonial social insect

Some species of ants possess an unusual form of social organization in which aggression among nests is absent. This type of social organization, called unicoloniality, has been studied in only a handful of species and its evolutionary origins remain unclear. To date, no study has examined behavioural and genetic patterns at points of contact between the massive supercolonies that characterize unicoloniality. Since interactions at territory boundaries influence the costs of aggression and the likelihood of gene flow, such data may illuminate how supercolonies are formed and maintained. Here we provide field data on intraspecific territoriality for a widespread and invasive unicolonial social insect, the Argentine ant (Linepithema humile). We observed abrupt and well‐defined behavioural boundaries at 16 contact zones between three different pairs of supercolonies. We visited nine of these zones weekly during a six‐month period and observed consistent and intense intercolony aggression that resulted in variable, but often large, levels of worker mortality. Microsatellite variation along six transects across territory borders showed that FST values were lower within supercolonies (0.08 ± 0.01 (mean ± SE)) than between supercolonies (0.29 ± 0.01) and that this disparity was especially strong right at territory borders, despite direct and prolonged contact between the supercolonies. Matrix correspondence tests confirmed that levels of aggression and genetic differentiation were significantly correlated, but no relationship existed between geographic distance and either intraspecific aggression or genetic differentiation. Patterns of FST variation indicated high levels of gene flow within supercolonies, but little to no gene flow between them. Overall, these findings are inconsistent with a model of relaxed ecological constraints leading to colony fusion and suggest that environmentally derived cues are not the prime determined of nestmate recognition in field populations of Argentine ants.

Contact between supercolonies elevates aggression in Argentine ants

Abstract.Complex recognition systems underlie the social organization of many organisms. In social insects the acceptance of other individuals as nestmates can involve a variety of different cues, but the relative importance of these cues can change in relation to the fitness costs of accepting or rejecting other individuals. In this study we investigate the mechanisms that underlie recognition behaviour in Argentine ants (Linepithema humile). Introduced populations of Argentine ants are characterized by a social structure known as unicoloniality where intraspecific aggression is absent over large distances resulting in the formation of expansive supercolonies. Recent research has identified sites where multiple, mutually aggressive supercolonies co-occur allowing an examination of Argentine ant behaviour at territorial boundaries. We found that workers from different supercolonies always interact aggressively with one another, but that neighbours from different colonies (i.e., workers from nests located in the immediate vicinity of territory borders) consistently exhibited higher levels of aggression compared to those displayed by non-neighbours from different colonies (i.e., workers from nests located far enough away from a territory border so that interactions are unlikely). This difference in the level of aggression displayed between neighbours and between non-neighbours from different supercolonies cannot be explained by differences in relatedness or genetic similarity. Instead our findings suggest that direct contact between mutually antagonistic colonies is sufficient to elevate aggression. A laboratory experiment in which we manipulated the extent to which colonies with no prior history of contact could interact with one another, revealed that aggression increased after colonies were permitted to interact, but dropped after connections between colonies were severed. Moreover, the mere presence of an aggressive supercolony was sufficient to elicit elevated aggression. Overall these patterns are opposite to the “dear enemy” phenomenon and could be the result of the intense territorial aggression exhibited by established supercolonies of this species.

Nest site selection and longevity in the ponerine ant Rhytidoponera metallica (Hymenoptera, Formicidae)

Summary: The size of the nesting cavity and the nature of the nest-building material can constrain colony growth in eusocial insects. Rocks protect colony members against extreme temperatures as well as serve as a supplementary source of heat, thereby maintaining optimal conditions for brood development. The frequency distribution of available rocks and those occupied by colonies of Rhytidoponera metallica in the Avon River State Forest, Victoria, Australia, showed that colonies of R. metallica avoided nesting under small rocks (<200 cm2). A strong correlation between colony size and rock size indicates that nesting under a larger rock promotes colony growth. Laboratory choice experiments revealed that workers are capable of recognizing larger rocks from external physical characters (dimension). Field data indicates that colonies leave nests frequently, with nest abandonment peaking in the summer months.

Food availability and brood number do not influence intraspecific aggression in Argentine ants

Summary.Nestmate recognition cues can derive from both environmental and genetic factors, but can also be modulated in response to context-specific cues. Synchronous changes in nestmate recognition systems occur seasonally in some species of ants, however the mechanisms underlying these seasonal changes are often unknown. We studied two mechanisms, relative brood number and food availability, to determine if they generate temporal variation in intraspecific aggression in an introduced population of the Argentine ant, Linepithema humile. Using data from previous studies we found that seasonal increases in aggression levels correlate with seasonal increases in brood-to-worker ratios in the field. However, when we manipulated brood-to-worker ratios in paired experimental colonies, we found no direct evidence that relative brood numbers influenced aggression levels. To determine if food availability influenced aggression we conducted a second experiment in which we randomly assigned pairs of experimental colonies to starved or fed treatments and then measured aggression levels weekly for five weeks. We observed no difference in the level of aggression between these two treatment groups indicating that food availability also has no affect on aggression levels between hostile conspecific colonies.