Introduction
Historically, herbivory on aquatic plants has been considered negligible. ‘One could probably remove all the larger plants and substitute glass structures of the same form and surface texture without greatly affecting the immediate food relations’, wrote Shelford (1918), cited in Hutchinson (1975) about grazing losses of submerged angiosperms. This misconception might have persisted for so long because grazing by zooplankton on phytoplankton has been the major focus in limnology for decades. Also, herbivore-related biomass losses of higher aquatic plants were estimated to be less than 10% of the total production (Wetzel, 1983). In the past two decades many studies have shown that multiple invertebrate and vertebrate herbivores feed on freshwater angiosperms and that herbivory on vascular plants is quantitatively equally important in terrestrial and freshwater habitats (Lodge, 1991; Newman, 1991; Cyr & Pace, 1993). Thus, we are now ready to critically consider the role of plant secondary metabolites (PSMs) in freshwater plant–herbivore interactions. Whereas the importance and tremendous variety of PSMs is well acknowledged in terrestrial plants and seaweeds, relatively little is known about the presence, levels, types and function of PSMs in freshwater plants (Lodge et al., 1998; Sotka et al., 2009). This is surprising because aquatic angiosperms and most of their insect herbivores are in fact secondarily aquatic, descendant from terrestrial ancestors (Newman, 1991). Thus, similarities in potential feeding deterrents and host-plant selection might be anticipated. Yet there may also be pronounced differences in plant–herbivore interactions in the aquatic environment. For example, water provides different physico-chemical conditions compared with air or soil, which should affect the dispersal of released compounds. Additionally, not all terrestrial plant families and growth forms have relatives underwater, and aquatic herbivores differ in species composition and diet selection from their terrestrial counterparts. These environmental, phylogenetic and ecological predispositions might have shaped the kinds of feeding deterrents that are present in freshwater systems.