PATTERNS IN LEAF HERBIVORY ON SEAGRASSES

We assess leaf herbivory on several populations of four temperate and five tropical seagrass species and examine, along with a compilation o f published reports, the extent of herbivory variability and how it af fects its general magnitude to help solve the apparent controversy abo ut the importance of herbivory on seagrasses brought out by recent rep orts. We further test whether herbivory variability is related to diff erences in leaf specific growth rate (SGR) as a descriptor of leaf nut ritional quality for herbivores. The extent of herbivory varied broadl y, both within and among species, ranging from negligible values up to 50% of leaf production removed within some species. This variability, along with that shown by other published reports, claim that the clas sical statement that herbivory represents minor losses of seagrass pro duction, being considered negligible in most cases, can be misleading and lead to the neglect of important seagrass-herbivore interactions. Differences among species in the percentage of leaf production removed were associated with differences in SGR, pointing to herbivore select ive feeding upon faster-growing species resulting from their higher nu tritional quality. This selection seems to be independent of leaf nutr ient concentrations suggesting that, in agreement with past reports, n utrient levels are a poor descriptor of seagrass nutritional quality a s most nutrients can be bound to indigestible fibre. No relationship b etween herbivory intensity and SGR was found among populations of a si ngle species. On the other hand differences among species in the areal flux of production transferred to herbivores seemed related to differ ences in the level of production attained. These results point out tha t whereas SGR should be a descriptor of the variability among species in the extent of herbivore pressure (i.e. percentage of production rem oved), the level of production should be indicative of differences amo ng species in their capacity to support herbivore production (i.e. flu x of production channelled). (C) 1998 Elsevier Science B.V.