Multitrophic interactions along a plant size gradient in Brassicaceae

Abstract

Plant size is hypothesised to be a major driver of biotic interactions, as larger plants are more conspicuous and offer a wider range of resources and niches for associated animals. However, the role of interspecific differences in plant size for associated animals (antagonists or mutualists) and the resulting plant reproductive fitness is little explored. In this thesis effects of plant size on species richness of herbivores and their natural enemies as well as on species richness of pollinators were tested. Endophytic and ectophytic herbivores and their natural enemies were considered separately as were herbivores and their natural enemies associated with different plant components. Further, the effect of plant size was studied for feeding damage to different plant components and the associated impact on plant reproductive fitness parameters, namely seed number, thousand seed weight and total seed weight per plant individual. Finally, the focus was placed on the effect of plant size on antagonistic and mutualistic flower associated insects and their impact on plant reproductive fitness along the plant size gradient. A common garden experiment with an interspecific plant size gradient (from 10 to 130 cm length) among 21 annual Brassicaceae species was established. In this way, we realised a broad gradient in plant size across different plant species with standardisation of the habitat and the surrounding landscape features, overcoming a common problem flaw in the analysis of within-species variation and naturally grown plants. Plant size, number, biomass and the size of the different aboveground plant components (flowers, fruits, leaves and stems) were quantified along with flower cover and colour. Relative feeding damage to the different plant components and the resulting reproductive fitness of each plant species were assessed. Finally arthropods on and in flowers, fruits, leaves and stems were sampled, including herbivores, their natural enemies and pollinators. Plant size was positively related to the species richness of herbivores, of their natural enemies and to the species richness of pollinators. This was likewise true for endophagous and ectophagous herbivores and their natural enemies as well as for fruit and leaf associated herbivores and their natural enemies. Furthermore, data showed increasing feeding damage to flowers and fruits with increasing plant size, while feeding damage to leaves and stems was driven by their biomass rather than by plant size. Feeding damage to flowers had the strongest effect on reproductive fitness, decreasing seed number, thousand seed weight and total seed weight. Focusing on flower associated insects, plant size had a positive effect on abundance and species richness of pollinators (but only when flowers were not superabundant) and also on pollen beetle abundance, despite the associated higher rates of parasitism of pollen beetles. Pollen beetles reduced seed number and thousand seed weight. Pollinators positively affected seed number only. Overall, increasing plant size led to decreasing thousand seed weight but did not significantly alter seed number and total seed weight, indicating a balance between increasing pollen beetle damage and positive effects of increasing pollinator visits. In conclusion, increased detectability and attractiveness to herbivores leads to important fitness costs for large plants, including flower damage by pollen beetles, which had the strongest negative impact on plant reproductive fitness in terms of seed number, thousand seed weight and total seed weight. These fitness costs for large plants may be counteracted by their detectability and attractiveness to pollinators, which positively influenced seed number. Purely in terms of seed numbers, being large is advantageous in places dominated by pollinators, while being small is advantageous in places dominated by herbivorous flower visitors. Contrarily, plants suffer from being large with regard to their thousand seed weight, which was driven by herbivores only. In general, plant size is a hitherto underestimated driver of interactions, and its effects on plant fitness through interacting insects are highly complex.