The Ponto-Caspian mysid Limnomysis benedeni was first recorded in Lake Constance in summer 2006, and a stable population developed at the site of discovery. Although this mysid is common in the Rhine and Danube rivers, little is known about its ecology and impact in systems of invasion. We investigated the autecology of L. benedeni in habitatchoice and food experiments. In the habitat-choice experiments, highly structured habitats, i.e., stones covered with zebra mussels (Dreissena polymorpha), macrophytes, and especially stonewort, were strongly preferred. In food experiments, L. benedeni fed mostly on food sources with a small particle size, e.g., biofilm on leaf litter, biodeposited material of zebra mussels, epilithon, and phytoplankton. We also compared the L. benedeni population data from Lake Constance with that from rivers. In Lake Constance, female L. benedeni were nearly 30% larger and carried more than three times more eggs in spring (9.470.6 mm and 28.475.7 eggs) than in summer (6.770.8 mm and 8.772.9 eggs). The mysids present in spring might be the generation that over-wintered; in summer, this generation was probably replaced by a new generation of smaller individuals. The large brood size and the detritivorous feeding strategy might allow L. benedeni to colonize Lake Constance rapidly.

Efficiency of natural enemies in pest control in the field frequently shows high degree of variability, making it difficult to propose reliable biocontrol strategies. Due to the small size of many beneficals, underlying mechanisms are difficult to evaluate. In the present study the potential of molecular markers to reconstruct individual foraging decisions of a small parasitoid species unsuitable for direct tracking, i.e. Diaeretiella rapae (M'Intosh) (Hymenoptera: Braconidae), a primary parasitoid of the cabbage aphid Brevicoryne brassicae (Linnaeus) (Hemiptera: Aphididae), was evaluated. Nine newly designed microsatellite primer pairs were tested for their reliability in parentage analysis. Therefore, several families of D. rapae with four to six daughters and high or low degree of inbreeding were established in the lab. Individuals were genotyped and parentage and sibship analysis was calculated with the software COLONY. The results indicated that 91 % of the 85 genotyped offspring from all families have been assigned to the correct family when all nine microsatellite loci have been used and 86 % when only six microsatellite loci have been used. Even in families with high degree of inbreeding up to 87 % of the offspring could be assigned correctly. In conclusion, the sibship analysis of D. rapae provides reliable results, especially with all nine selected microsatellite loci. Therefore it will be a valuable tool to reconstruct individual foraging decisions and characterize parasitoid ''personality'' in the field to improve biological control strategies.