Larval habitat selection, not larval supply, determines settlement patterns and adult distribution in two chthamalid barnacles

1. Two common species of acorn barnacle, Chthamalus montagui and C. stellatus , with similar resource needs, occur together on rocky shores of NW Europe but differ in their adult distribution over two environmental gradients, wave exposure and vertical height on the shore. 2. A sampling programme was undertaken over 2 years to determine the causal mechanisms of this differential distribution. Three alternative, although not mutually exclusive, models were tested: first, adult distribution is determined at settlement as a direct result of differential larval supply; secondly, adult distribution is determined at settlement as a result of larval choice rather than differential larval supply; and thirdly, settlement occurs homogeneously over the two gradients and the adult pattern is set by differential post-settlement mortality. 3. Near-shore larval abundance in the water column over the gradient of wave exposure was determined at eight dates during the main settlement period in 2003. Larval production was assessed by quantifying early-stage larvae, while the supply of competent settlers to the adult habitat was determined by quantifying late-stage cyprid larvae. 4. The pattern of settlement of the two species was determined over both environmental gradients by daily sampling of the proportional abundance of C. stellatus and C. montagui settled cyprids at three tidal heights at exposed and sheltered shores. Differential post-settlement mortality was determined by quantifying day-old, week-old and up to 3 months-old metamorphosed individuals. 5. Larval production reflected the distribution of adults, with early-stage larvae of C. montagui predominating in shelter and C. stellatus in exposure. However, the supply of late-stage cyprid larvae to the near-shore area showed no difference in the proportion of each species over the wave-exposure gradient; differential larval supply was not responsible for dictating adult distribution. 6. Despite no differential supply of larvae, the pattern of settlement of the two species differed over both environmental gradients, with C. stellatus settling in greater proportions low on the shore and in exposure and C. montagui settlement predominating higher on the shore and in shelter. This pattern directly matched the distribution of adults. Postsettlement mortality accentuated, but did not change, the pattern set at settlement. 7. The implications of the results, and the usefulness of the technique employed, to understanding the degree of ‘openness’ in populations of marine benthic invertebrates are discussed.