The behavioral traits that shape the structure of animal societies vary considerably among species but appear to be less flexible within species or at least within populations. Populations of the ant Leptothorax acervorum differ in how queens interact with other queens. Nestmate queens from extended, homogeneous habitats tolerate each other and contribute quite equally to the offspring of the colony (polygyny: low reproductive skew). In contrast, nestmate queens from patchy habitats establish social hierarchies by biting and antennal boxing, and eventually only the top-ranking queen of the colony lays eggs (functional monogyny: high reproductive skew). Here we investigate whether queen-queen behavior is fixed within populations or whether aggression and high skew can be elicited by manipulation of socio-environmental factors in colonies from low skew populations. An increase of queen/worker ratio and to a lesser extent food limitation elicited queen-queen antagonism in polygynous colonies from Nürnberger Reichswald similar to that underlying social and reproductive hierarchies in high-skew populations from Spain, Japan, and Alaska. In manipulated colonies, queens differed more in ovarian status than in control colonies. This indicates that queens are in principle capable of adapting the magnitude of reproductive skew to environmental changes in behavioral rather than evolutionary time.

BackgroundThe disjunct distribution of several Palearctic species has been widely shaped by the changes in climatic conditions during the Quaternary. The observed genetic differentiation or reproductive isolation between extant populations may be the outcome of their contemporary geographic separation or reproductive incompatibility due to differences in phenotypic traits which have evolved in isolated refugia. In the boreal ant Leptothorax acervorum, colonies from central and peripheral populations differ in social structure: colonies from Central and Northern Europe may contain several equally reproductive queens (facultative polygyny), while in colonies from peripheral populations in Spain only one the most dominant of several queens lays eggs (functional monogyny). By reconstructing the specie’s evolutionary and demographic history in Southwestern Europe we examine whether variation in social organization is associated with restricted gene flow between the two social forms.ResultsWe show that multi-queen colonies from all so far known inner Iberian populations of L. acervorum are functionally monogynous, whereas multi-queen colonies from all Pyrenean populations are polygynous, like those from other previously studied areas in Europe. Our analyses revealed complex spatial-genetic structure, but no association between spatial-genetic structure and social organization in SW-Europe. The population in the western Pyrenees diverged most strongly from other Iberian populations. Moreover, microsatellite data suggest the occurrence of recent bottlenecks in Pyrenean and inner Iberian populations.ConclusionsOur study shows a lack of reproductive isolation between the two social forms in SW-Europe. This in turn suggests that demographic and spatial patterns in genetic variation as well as the distribution of social phenotypes are better explained by co-variation with climatic, ecological, and historical factors. Moreover, we for the first time show the existence of substantial spatial-genetic structure in L. acervorum, suggesting the existence of multiple refugia in SW-Europe, including two extra-Mediterranean refugia in France. While gene flow among inner Iberian refugia may have been larger during the late glacial, extra-Mediterranean refugia in southern France may have contributed to the post-glacial recolonization of W-Europe.Electronic supplementary materialThe online version of this article (doi:10.1186/s12862-016-0711-3) contains supplementary material, which is available to authorized users.