CENTRE-EDGE DIFFERENCES IN BEHAVIOUR, TERRITORY SIZE AND FITNESS IN CLUSTERS OF TERRITORIAL DAMSELFISH: PATTERNS, CAUSES, AND CONSEQUENCES

Variation in behaviour and fitness with spatial position in a group (centre vs edge), as well as the causes and consequences, was studied in the permanently territorial threespot damselfish, Stegastes planifrons. These fish have individual territories, clusters of which occur on coral patch reefs. Fitness correlates varied with group position. Although, survival and age did not differ between fish resident in the centre or edge, central fish were larger and grew faster than edge fish. Central males also received more clutches, eggs/clutch, and cumulative number of eggs to defend than did edge males. The higher fitness in centre positions was then correlated with differences in behaviour by group position. Territories defended were smaller in the centre of groups. Aggressive interactions with conspecifics were more frequent for central fish, but interactions with heterospecifics, and overall interactions, occurred at much lower rates for fish in central positions. Centre fish lost less of the algal food in their territories to intruders. Feeding and courtship rates, and forays from the territory, did not differ by position. These behavioural observations suggest that the positional differences in fitness measures and territory size are due to lower energy costs of territory defence for central fish, permitting the greater investment in growth and reproduction. Centre-edge differences in behaviour and territory size could be due to a selfish-herdlike group position effect (i.e. position relative to neighbours determines rates of behaviour), the microhabitat variation that was found between these positions, or a combination of both factors. To test among these alternatives, fish peripheral to a central treatment individual were removed experimentally, thus altering that individual's relative position from the centre to the edge of a group, without changing the microhabitat features of its territory. Territory size and behaviour measures of position-altered fish changed in ways consistent with the interpretation that group position has a much greater effect on behaviour and territoriality than does microhabitat. Group position may thus influence the fitness measures that differ by spatial position. These results suggest that habitat geometry and/or fragmentation may affect individual territory size, and hence maximum population density, as well as per capita reproductive output, by altering the relative number of edge individuals in groups. Does S. planifrons recognize and compete for residence in the better central positions? If there is competition, then there should be (1) slower reoccupation, (2) lower rates of intraspecific aggression, and (3) fish resident in the centre should not compete for open edge space in newly opened edge territories, relative to newly cleared centre territories. Also, (4) colonization to completely cleared areas should preferentially be to the centre positions. The first three predictions were supported by a paired removal experiment, while the fourth prediction was partially supported in another experiment. Competition thus appears to be an important factor determining the local distribution of threespot damselfish on patch reefs.