A long-term experiment was carried out to determine the relative abilities of five sessile animals to colonize cleared plots in the presence or absence of predators. The experiment was done on two adjacent rocky shores, one sheltered and one exposed, on a small tropical island. The effect of predation in maintaining bare space was greatest on the exposed shore/upper mid-intertidal, less on the sheltered shore and absent on the exposed shore/lower mid-intertidal. The barnacle, Tesseropora rosea, recruited heavily at the exposed shore/lower mid-intertidal and was dominant, but was unable to colonize the exposed shore/upper mid-intertidal or the sheltered shore. The barnacle, Tetraclita squamosa, successfully colonized only the exposed shore/upper mid-intertidal and did not appear to be affected by predation. The oyster, Saccostrea echinata, colonized only the sheltered shore and was very susceptible to predation. Given the variability found on this small spatial scale, it is suggested that consistent differences in ecological processes between regions (tropical versus temperate) are unlikely but, rather, that differences between localities will be found at the level of species interactions and abiotic effects in particular habits.