Download (20MB) - Repository@Napier

abundances of P. ciliata had more diverse meiofaunal and macrofaunal communities
compared with areas of low abundances, while Reise (1983b) found that the presence
of dense assemblages of P. elegans promoted the abundance of small benthic
organisms by approximately 40%. Morgan (1997), using a correlation approach,
indicated that the majority of the most common taxa in the Baie de Somme, France,
e.g., Eteone longa, Hediste diversicolor and Cerastoderma edule, were significantly
positively correlated with P. elegans abundances.
Many studies endeavouring to elucidate the mechanism by which high numbers of
tube-builders affect infaunal community structure have focused on the way in which
recolonisation is effected (e.g., Gallagher et al., 1983; Trueblood, 1991). Resident
adults may influence colonists in many ways determined by their feeding mode and
modifications to the sediments and hydrodynamics (Thrush et al., 1992). However,
experimental studies on such mechanisms have proved equivocal. Since recruitment
is usually assayed some time after settlement and metamorphosis (Bachelet, 1990) the
actual patterns of larval settlement are often obscured (Hadfield, 1986).
Consequently, studies have failed to assess the mechanisms by which tube-builders
affect recolonisation since it is inherently difficult to distinguish between differential
settlement and differential mortality of larvae (Woodin, 1986).
Larvae of some marine benthic invertebrate species have been experimentally shown
to actively select settlement sites using certain cues (Scheltema, 1974; Woodin, 1986;
Butman et al., 1988a; Pawlik and Butman, 1993; Hsieh, 1994). However, it is likely
that in the field, water flow is greater than the swimming speeds of larvae and instead,
larvae are transported as passive particles and deposited via passive entrainment
(Hannan, 1984; Butman, 1987; Butman et al., 1988b). Eckman (1983) predicted that
velocity of near-bed flow through plastic straws (seagrass mimics) would be reduced
to such an extent that suspended particles, such as larvae and meiofauna, would be
passively deposited. His results supported this prediction although his experimental
design was flawed by replication of only one of his treatments, his defaunated control.
Later experiments have incorporated replication of the defaunated controls but the
results have been inconclusive due to a number of factors including a plot size too
small to allow the formation of a fully developed boundary layer (Kern and Taghon,
227