EXPERIMENTAL TESTS OF THE ADVANTAGES AND DISADVANTAGES OF HIGH-DENSITY FOR 2 COEXISTING COCKLES IN A SOUTHERN-OCEAN LAGOON

1. As a strong test of the prevailing generalization that competition plays a relatively trivial role among suspension-feeding invertebrates in the soft sediments on the estuarine seafloor, we manipulated local densities of bivalves, two species of Katelysia, in a Western Austral ia lagoon, characterized by exceptionally high bivalve abundances (ave raging 157 m-2). 2. Competition appeared weakly and sporadically, evid ent in only one of five intervals, when survivorship was 20-30% lower, production of undamaged empty shells 3-5 times higher, and average gr owth of survivors 50% lower for both Katelysia scalarina Lamarck and K atelysia rhytiphora Lamarck in the highest density treatment (320 m-2) than in the others (20-160 m-2). 3. Surprisingly, during both winter- spring intervals survivorship was also 25% lower at the lowest density for both species. The type of mortality was one that resulted in disa ppearance of the empty shell. Addition of sufficient numbers of sham c lams, dead shells filled with sand and implanted in living position in the sediments, acted to prevent the enhancement of mortality at low d ensity. This demonstration of the importance of a structural barrier i mplies that the agent of mortality at low density is a predator. Sampl ing error, emigration and post-mortem transport by scavengers can be r uled out by results of direct tests. 4. In a caging experiment, the pr edatory seastar Coscinasterias calamaria (Gray) replicated the pattern of enhanced predation rate on Katelysia of both species at low densit y. This represents a candidate for the unknown predator that was deter red by the structural barrier created by crowding of shells. A demersa l fish is an alternative candidate. 5. Contrary to previous data for s uspension feeders in estuarine soft sediments, which commonly show neg ative effects of adults on local recruitment of young, recruitment of K. scalarina was greater around higher densities of adult bivalves in one of two tests. Survivorship of recruits did not vary with adult den sity, suggesting that the recruitment pattern may have resulted from g regarious larval settlement around adults rather than from differences in early post-settlement survivorship. 6. Evolution and expression of gregarious settlement may be a reasonable expectation, given the spor adic and small penalties of competition, and the large and repeated be nefits of mutual defences at high density. More broadly, positive inte ractions such as commensalism, mutualism and facilitation might be exp ected to be more prevalent in systems like the soft-sediment benthos w here competition is so ineffective and relatively unimportant.