Coevolution is the process where at least two species put some selective pressure on each other, thereby reciprocally influencing each others evolution. To explain co-adaptations invoked by coevolution of interacting species, evolutionary biologists predominantly use host-parasite interactions as model systems. These enable the study of adaptations and counter-adaptations that might evolve in the arms-race between a parasite pursuing maximum gain and a host trying to avoid exploitation. One such system is the socially parasitic butterfly Maculinea alcon and its host the ant Myrmica rubra. Throughout the first instars M. alcon lives on a specific food plant, however, in the last instar before pupation it develops into an obligate social parasite, posing a considerably cost to its host ant colony. I here focus on the different exploitation strategies of M. alcon throughout its lifecycle and the host ant's response.

There has been an ongoing debate on whether or not oviposition behavior in female butterflies of Maculinea is dependent on the direct proximity of host ant workers. I show that oviposition behavior is not host ant dependent; instead development and exposure of the plant are the most important factors for oviposition choice of the female butterflies.

Whilst development from egg to the 3^rd instar exclusively occurs on the food plant, the caterpillar stops feeding when it reaches the 4^th instar. Hereafter it will only survive if it is actively adopted by a host ant worker into the nest where it will feed on ant regurgitations and ant brood. It is thus crucial for the caterpillar's survival to attract the host ant, get picked up and brought back to the ant's colony. My study shows that 3^rd and 4^th instar caterpillars are distinct from each other not only morphologically but also in their surface chemistry. 3^rd instars are passively gaining parts of their surface chemistry from the food plant, while 4^th instar caterpillars are actively mimicking the cuticular hydrocarbons of the host ant's brood, ensuring their adoption and integration into the ants nest.

As the caterpillar constitutes a fitness cost to infected host ant colonies, the host ants are expected to have developed defense mechanisms in response to the presence of the social parasite. I was able to demonstrate that the efficiency of ant colonies to defend themselves against intruders depends on a multitude of often correlated factors including genetic-, chemical- and physical-distance, queen number, and whether they are from an area where the social parasite is present.

Ants rely on chemical cues to discriminate nestmates from non-nestmates. The recognition cues are not static though and ants have evolved to adapt to the ever changing recognition cues. I tested the hypothesis that M. alcon caterpillars can actively alter their host's chemical profile by producing a great surplus of cuticular hydrocarbons themselves and spreading them throughout the colony, but found no support. However, I show that the chemical profile of ant colonies is very flexible as it can change significantly over the course of only one month even when rearing conditions are kept completely stable.

Local co-adaptations are the basic module for coevolutionary change and can have a great impact on the population genetic structure of the coevolving species. Investigation of 24 sites of M. alcon at the northern most distribution range including Denmark, Sweden and northern Germany, revealed high levels of isolation by distance and low levels of gene flow between sites in the five regions sampled. Furthermore, a temporal comparison over ten years showed a significant decrease in genetic diversity which strongly argues for conservation measures to take place at those sites.

My thesis thus highlights the need for management initiatives for this rare butterfly and the new insights provided here on the specialized adaptations to its host ant can help to improve such future conservation efforts.