Behaviour controls post-settlement dispersal by the juvenile bivalves Austrovenus stutchburyi and Macomona liliana

Post-settlement dispersal is a key process affecting the population dynamics of many soft sediment benthic invertebrates. Despite its importance, few studies have quantified those factors that influence juvenile dispersal. In a laboratory flume, we examined the effects of three flow velocities and two substrate types on the dispersal of two common intertidal bivalves: the deposit-feeding wedge shell Macomona liliana and the suspension-feeding cockle Austrovenus stutchburyi. Juveniles of three size classes ( < 2, 2-4 and 4-8 mm shell length) were added to cores of defaunated natural sediment or glass beads. We recorded the number of bivalves remaining in cores versus those recovered downstream either on the flume floor, in bedload traps or in a plankton net at the end of the working section of the flume after 48 It at three freestream velocities (4.8, 11.0 and 16.6 cm s(-1)). At flow speeds of 4.8 cm s(-1), < 5% of individuals were recovered outside the cores for both species. At higher flows, the dispersal mode (crawling, bedload or in suspension) and frequency of dispersal differed markedly between species. Austrovenus dispersed primarily by crawling in the low flow treatments. The frequency of dispersal increased substantially (2-6 x) between 11.0 and 16.6 cm s(-1), and most Austrovenus were found in bedload traps at the highest flow. At the highest flow, twice as many Austrovenus individuals left the glass bead treatment as from the natural sediment. The number of dispersing Macomona also increased with increasing flow from the natural sediment, but numbers dispersing from glass beads were similar for the two higher flows (11.0 and 16.6 cm s(-1)). Macomona dispersal mode was associated with size; smaller size classes were collected in the bedload traps and the plankton net in approximately equal proportions, while only a small proportion of the largest size class were collected in the plankton net. In contrast to flume dispersal experiments with live bivalves, most dead Austrovenus did not move at the highest flow speed, while most dead Macomona were transported at the highest flow speed. The live-dead comparisons, in conjunction with our primary experimental results, imply that there are active behavioural components to both water column and bedload transport. Our research emphasises that both species actively influence post-settlement transport in response to both substrate type and flow regime, and that bedload transport, often categorised as a passive transport process, is also greatly influenced by active behavioural processes. (C) 2004 Elsevier B.V. All rights reserved.