EFFECTS OF PREDATORS ON GROWTH, MORTALITY AND ABUNDANCE OF A JUVENILEREEF-FISH - EVIDENCE FROM MANIPULATIONS OF PREDATOR AND PREY ABUNDANCE

Studies of open populations, in which offspring are dispersed away fro m their natal site, often emphasise density-independent processes, nam ely stochastic variation in the input of newly arrived young (recruitm ent). Studies of closed populations, in which offspring remain close t o their natal site, emphasise the role of density-dependence, particul arly through competition and predation. The usefulness of this dichoto my was tested on spiny chromis Acanthochromis polyacanthus, which is o ne of the few marine fish that lack a pelagic larval phase and is thou ght to be the archetypal species for closed population status among re ef-fishes. Experimental manipulations showed that juvenile A. polyacan thus suffered lower rates of mortality in locations where large predat ory fish were excluded. The magnitude of differences in mortality betw een treatments (cages and open plots) matched natural differences in m ortality between locations representing low and high numbers of predat ors, suggesting that large predatory fish were a primary source of juv enile mortality. Growth did not increase when prey and predator abunda nces were reduced. This finding rejected 2 models: that growth was red uced in large schools and these slower growing individuals suffered gr eater rates of predation; and that intraspecific competition occurred. Growth of prey increased when released from predation pressure, a pro cess not previously considered in studies of growth in reef-fish. The repression of growth in response to predation pressure indicates that predators may have a more far-reaching effect on demography of reef-fi sh than previously thought. Predation was dens:ty-dependent in that ju veniles in larger schools suffered greater rates of mortality in the p resence of predators but not in their absence. Predation appeared to l imit the upper size oi schools, but this did not eliminate a linear re lationship between input and subsequent abundance. Hence, density-depe ndent predation was weak and large fluctuations in recruitment (births ) persisted and contributed importantly to fluctuation in prey density . These results support the model that when density-dependent mortalit y is weak, both input and subsequent mortality will be important limit ing factors. Consequently, this study questions the assumption that va riation in predation and competition has greater consequences for clos ed populations.