MEPS 155:189-198 (1997)  -  doi:10.3354/meps155189

Coral reefs are subject to both natural and anthropogenic disturbances that may have detrimental effects on resident fish populations. This study examined the association between the coral-feeding wrasse Labrichthys unilineatus and the coral it feeds on, and investigated how the fish species responds to disturbance. Fish abundance was positively correlated with percentage scleractinian coral cover at 3 different scales of sampling (4, 100 and 250 m² sampling units). Juveniles and, to a lesser extent, mature females were closely associated with patches of Acropora nobilis and A. elseyi, the cover of these corals being much greater inside fish home ranges. The habitat of the larger, wider ranging males reflected the cover of corals within the study area which was dominated by Montipora spp. and A. hyacinthus. Ontogenetic changes in diet reflected both changes in the availability of different coral species and changes in feeding selectivity. Juveniles had a narrower range of coral types available, but were more selective than larger fish, preferentially consuming many species of low abundance. At the other extreme, adult males fed almost exclusively on the common acroporid species in proportions reflecting their availability. Strong ontogenetic patterns in selectivity also extended to the relative consumption of undamaged and damaged tissues of coral colonies, the latter including bleached, diseased and broken edges of live corals. Damaged tissue of coral colonies was an important component of the diet of adult L. unilineatus, particularly males, which consumed more damaged than undamaged tissue, despite its low availability. Results of an electivity index showed that all coral colonies (in the feeding territories) that presented damaged tissue were strongly selected and none were avoided. Experimental physical disturbance to coral patches on a scale of 1 m² quadrats resulted in a dramatic increase in the number of bites taken by L. unilineatus, compared with control plots. The increase was short term, with feeding intensity returning to the pre-disturbance level within 24 h of the time of initial damage. Individual feeding rates were also enhanced by the disturbance event, with fish in the vicinity of disturbed plots exhibiting a 50% increase in feeding rate. The reason for selection of damaged corals is unknown but a loss of nematocysts or release of mucous or chemical attractants may be involved.

Corallivores · Disturbance · Coral reef fish · Labrids · Coral disease