Evolution and maintenance of sociality in crab spiders (Thomisidae)

In this thesis I investigated factors that may explain the evolution and maintenance of sociality in crab spiders (Thomisidae). Group-living crab spiders differ from most other social spiders because they lack a capture web, a factor considered to be very important for the evolution of sociality in spiders. Colonies of subsocial and social spiders are usually comprised of family members, and thus another unusual trait of subsocial crab spiders is the regular acceptance of unrelated conspecifics into their groups. Previous work on the subsocial crab spider Diaea ergandros has demonstrated that maternal care as well as the construction and occupation of nests as protective retreats are important factors that may explain group-living in this species. Furthermore, it has been shown that these spiders are able to recognize kin, which offers an excellent opportunity to study group dynamics between relatives and immigrants. Building up on this knowledge, I examined potential costs and benefits of group living with a focus on the effect of unrelated spiderlings. Moreover, I estimated the genetic relatedness within family groups, and studied the broader natural history as well as phylogenetic relationships of subsocial and social crab spiders. With a detailed natural history description of the crab spider Xysticus bimaculatus I have demonstrated that lifestyle and demographics are very similar to the subsocial Diaea ergandros (chapter 2). This new discovery of subsocial behaviour outside the genus Diaea indicated that subsociality may have evolved multiple times independently within Thomisidae. Testing this hypothesis, I investigated the evolutionary history of social behaviour in crab spiders in a molecular-phylogenetic context. The results suggest that subsociality has at least two independent origins confirming that X. bimaculatus is not closely related to any of the other group-living thomisid species (chapter 3). The evolution of sociality in spiders is accompanied by a switch from outbreeding to inbreeding. D. ergandros has been suggested to be at a particularly advanced transitory stage from subsocial to social behaviour, and a previous study has shown that populations are inbred. Considering that low mating rates and inbreeding favour offspring cooperation of highly related individuals, females may benefit from monogamous or even incestuous mating. I studied the mating behaviour of D. ergandros and investigated natural mating rates with microsatellite markers (chapter 4). However, mating trials did not provide evidence for female choice. Accordingly, the genetic analyses did not support the existence of a monogamous mating system but rather supported relatively low mating rates, which may still sufficiently secure offspring cooperation while simultaneously providing some degree of outbreeding. Genetic relatedness has been identified as an important factor promoting cooperation in many subsocial spiders and other organisms studied to date. With two laboratory experiments (chapter 5 & 6) I confirmed that siblings of Diaea ergandros had an advantage over mixed groups that included immigrants. The latter were generally accepted but negatively affected female-offspring and offspring-offspring foraging interactions. Nevertheless, accepting immigrants may have benefits as well. In another experiment, larger groups of D. ergandros outperformed small groups in that they built larger protective retreats and had a lower mortality as well as higher individual growth (chapter 7). Group size varies considerably under natural conditions and small groups may thus benefit from accepting immigrants.