The macrofauna associated with individually collected clumps of pelagic Sargassurn was influenced strongly by the location of sampling. Algae collected in the Sargasso Sea carried more species with lower dominance and higher evenness and variation in abundance than algae collected in the Gulf Stream. Gulf Stream samples were dominated by the shrimp Latreutes fucorum (67.8 %) with little compositional variation whereas the gastropod Litiopa melanostoma was the most abundant but weakly dominant species (25.1 %) in the Sargasso Sea. Generally, species endemic to pelagic Sargassum species were most abundant in the Sargasso Sea while numerous temporary associates were found in the Gulf Stream. Numbers of individuals and species increased with the weight of algal clumps but clump size had little effect on the species composition of algal associates. Variation in species composition and trophic organization of macrofaunal associates with geographic region and with algal age were related to differences in the epiphytization of the algae. Because of the dissimilarity in algal associates of the Gulf Stream and Sargasso Sea, the influx of littoral fauna from the West Indies to the Sargasso Sea is probably very slow and the present forms may have evolved in that water mass.

Epibenthic macroinvertebrate community structure (species composition, numbers of individuals and species, and relative abundance distributions) were examined at 4 study sites in the northeast Gulf of Mexico (Apalachee Bay, Florida, USA). The relative abundance distribution changes with time of sampling (diurnal or nocturnal) because individual species become more or less numerically abundant in nocturnal samples. The number of individuals collected was significantly different between diurnal and nocturnal samples at 3 of the 4 study sites. The degree of day/night variation appears to be associated with habitat complexity (as defined by plant biomass and red algae volume) for number of individuals, but not for number of species, collected. The inadequacies and consequences of reliance on an exclusively diurnal epibenthic sampling program for a good representation of macroinvertebrate community structure are discussed.