Comparative feeding selectivity of herbivorous insects on water lilies: aquatic vs. semi-terrestrial insects and submersed vs. floating leaves

1. The rate of grazing damage experienced by submersed and floating leaves of water lilies (Nuphar variegata and Nymphaea odorata) was monitored in lakes in the Upper Peninsula of Michigan, U.S.A. Herbivores damaged 0.2–1.7% of the leaf surface of water lilies per day. These grazing rates differed between plant species, between submersed and floating leaves, and between lakes. Some leaves had more than 60% of their surface damaged and an overall mean of 16% damage occurred during the 2–3 week monitoring period of this study. 2. Snapshot measurements of grazing damage on randomly collected submersed and floating leaves of Nuphar showed that submersed leaves were more damaged (11.0 6 1.6%, n 5 84) than floating leaves (3.8 6 0.6%, n 5 92). Overall, these 176 Nuphar leaves had 7.2% of their area damaged. 3. Five species of herbivorous insects were commonly found on water lilies (Nymphaeacea). One primarily aquatic insect (sensu Newman, 1991), a caddisfly larva (Trichoptera: Limniphilidae), had a generalized diet of water lilies, other macrophytes, algae, and detritus. Four of the five insects were from primarily terrestrial insect groups (Coleoptera and Diptera; ‘secondary invaders’, sensu Newman, 1991) and consumed only water lilies in food preference experiments. 4. The feeding preferences of the generalist trichopteran were altered when the macrophytes were freeze-dried, ground into a powder, and reconstituted in an alginate gel. This suggests that plant structure may be an important feeding determinant for this insect. In contrast, a specialist weevil preferred its host plant in choice assays, regardless of whether fresh tissue or reconstituted macrophytes were used, suggesting this insect cued on a unique, non-structural property of its host plant. 5. These results suggest that herbivory on freshwater macrophytes is of a similar magnitude to that on terrestrial plants. The findings of this study are consistent with the hypothesis that herbivorous insects of primarily terrestrial groups have a narrower diet breadth than insects of primarily aquatic groups.