Feeding selectivity and a functional trade-off in a benthic fish with a continuous morphological variation: an experimental test

Abstract

Abstract Background Functional trade-offs through ecological specializations are hypothesized to become causes of adaptive phenotypic divergence under divergent natural selection, where intermediate phenotypes may have the lowest fitness. Evidence of phenotypic divergence in a trade-off between populations experiencing different environmental/ecological conditions is abundant. However, traits in divergent selection sometimes present non-discrete (unimodal) variability, including intermediate phenotypes, although the underlying mechanisms are poorly documented. A benthic cyprinid fish, Pseudogobio esocinus, in Lake Biwa, central Japan, exhibits a large non-discrete/continuous variation in mouthpart morphology (from wide to narrow) within a lake population. The variation is linked with individual diets (i.e., the compositions of two different types of prey) even at a single site, and thus the variability is hypothesized to persist under divergent selection for prey usage. As a first step toward understanding the persistence mechanisms, here I examined the presence of morphology-dependent feeding selectivity and a functional trade-off in a laboratory experiment. Results When each experimental fish was simultaneously provided the different types of prey (chironomid larvae and amphipods), the fish mostly utilized chironomid larvae as primary prey. However, compared with the wider-mouthed fish, the narrower-mouthed fish took a larger proportion of amphipods as secondary prey by changing feeding (attacking) behavior. The intermediate-mouthed fish had lower feeding efficiency than the extreme-mouthed fish, indicating potential disadvantage of the intermediate phenotype. Conclusions This experimental result supports the presence of morphology-dependent feeding performance and a functional trade-off with potential impacts on trait variability, which may favor specializations rather than generalizations. In the wild, however, there may be some situations for relaxing the trade-off, such as temporally fluctuating prey availability that could also favor generalizations depending on the conditions, and thus, both extreme and intermediate phenotypes may persist/coexist in a single habitat. Although further examinations, especially focusing on feeding efficiency for each prey type separated from the effects of prey selectivity, are needed, this case represents an opportunity to consider the possible mechanisms of the persistence of phenotypic variation that is maintained without divergence even in a trade-off.