EXPERIMENTAL-EVIDENCE OF COMPLEX INTERACTIONS BETWEEN BIOTIC AND ABIOTIC FACTORS IN THE DYNAMICS OF AN INTERTIDAL POPULATION OF THE BIVALVECERASTODERMA-EDULE

Densities of the suspension-feeding bivalve Cerastoderma edule (L.) we re manipulated inside field enclosures at two tidal elevations (low wa ter level, LWL, and mid-tide level, MTL) on an intertidal sand flat in Arcachon Bay to test the influence of both adult densities and emersi on time on (1) individual growth rate, (2) settlement rate, and (3) su rvival rate of cockles. These experiments were conducted during two co nsecutive years, in plots with ambient (Ix) and modified (1/3x, 3x, or 10x) densities of adult cockles. Growth rate and condition index of b oth adults and juveniles were significantly higher at the lowest tidal elevation (LWL), which is in accordance with the feeding mode of the species. The highest juvenile growth rate was recorded in the low-dens ity treatments (160-200 adults m(-2)), which suggests a competitive in teraction with adults; for the latter, growth rate was depressed only at the highest density (2000 m(-2)). survival of-adults was affected n either by immersion time nor by densities. Contrasted results were fou nd for the settlement rate and the survival of recruits. During the fi rst year of-experiment, density-treatments had no effect on settlement at MTL, whereas high adult densities negatively influenced settlement at LWL, but only during the period of high settlement; during the sec ond year, a significant effect of-adult densities on settlement was fo und on a single occasion at MTL. However, at the end of-both experimen ts (195 and 252 d, respectively), recruit densities did not significan tly differ between density-treatments. Although the difference in tida l elevation between the two experimental sites was only 1.0 m, settlem ent peaks were clearly distinct: they occurred in April (approximate t o 4000 m(-2)) at LWL but not earlier than July (approximate to 12000 m (-2)) at MTL; however, the final recruitment was higher at LWL than at MTL. Preliminary results of a laboratory flume study show that, in co nditions of smooth turbulent flow (u = 0.51 cm s(-1), Re* = 1.8), juv enile cockles are able to leave the substratum, to migrate by byssus d rifting into the water column over several metres and to avoid resettl ement in areas with high densities of conspecific adults. Although com petition with adults may occur during larval settlement, subsequent mi gration of juveniles between different tidal levels is likely to affec t significantly the growth and recruitment of intertidal cockle popula tions.