Seagrass plants with associated macrofauna were confined within transparent Perspex tubes in the field at two Western Australian sites, Cliff Head and Seven Mile Beach, and faunal population changes in the microcosms monitored and compared with changes in the field. Three microcosm treatments were set up at Cliff Head: (i) "dark" microcosms, in which primary production was reduced by wrapping the tubes in black plastic, (ii) "faunal reduction" microcosms, in which the food resources available to each animal were enhanced by removing 80% of the fauna, and (iii) "control" microcosms. Faunal abundances in control microcosms remained relatively constant, while the population numbers of nearly half of the common epifaunal species placed in dark microcosms declined relative to numbers in the control treatments. The abundances of almost all species in the faunal reduction microcosms rapidly increased, with proportionately more crustaceans carrying eggs in faunal reduction microcosms than in the other treatments. These results are consistent with the hypothesis that epifaunal populations are food limited. Other studies have shown that the diets of different epifaunal species are broadly overlapping, hence diffuse exploitative competition is probably a major structuring agent amongst the seagrass epifauna. The populations of almost all epifaunal species fluctuated greatly with season in the field. These annual population cycles did not, however, generally correspond closely between the Cliff Head and Seven Mile Beach sites. Therefore, no single factor, including those implicated to be important in other studies (such as temperature, photoperiod, seagrass biomass and epiphytic biomass), had a controlling influence on the population dynamics of the majority of epifaunal species. The annual population cycles of almost all species at Cliff Head nevertheless coincided with highest faunal densities occurring in late summer or early autumn, followed by a rapid decline to very low densities in winter. Animal populations inside microcosms did not change greatly at the time of rapid declines in field populations. A large-scale autumn emigration of mobile crustaceans from the Cliff Head site in response to low levels of microalgal food and/or dissolved oxygen is postulated to have occurred.
