Abstract
Molecular methods have become an important tool for studying feeding interactions under natural conditions. Despite their growing importance, many methodological aspects have not yet been evaluated but need to be considered to fully exploit the potential of this approach. Using feeding experiments with high alpine carabid beetles and lycosid spiders, we investigated how PCR annealing temperature affects prey DNA detection success and how post-PCR visualization methods differ in their sensitivity. Moreover, the replicability of prey DNA detection among individual PCR assays was tested using beetles and spiders that had digested their prey for extended times postfeeding. By screening all predators for three differently sized prey DNA fragments (range 116-612 bp), we found that only in the longest PCR product, a marked decrease in prey detection success occurred. Lowering maximum annealing temperatures by 4 °C resulted in significantly increased prey DNA detection rates in both predator taxa. Among the three post-PCR visualization methods, an eightfold difference in sensitivity was observed. Repeated screening of predators increased the total number of samples scoring positive, although the proportion of samples testing positive did not vary significantly between different PCRs. The present findings demonstrate that assay sensitivity, in combination with other methodological factors, plays a crucial role to obtain robust trophic interaction data. Future work employing molecular prey detection should thus consider and minimize the methodologically induced variation that would also allow for better cross-study comparisons.