398
S. Gobert, M. L. Cambridge, B. Velimirov, et al.
primarily by epiphytic production. This situation
contrasts with pelagic food webs, which are primarily fueled by phytoplankton, and those occurring below the euphotic zone, which are primarily fueled
by detritus. No consensus exists about the most important primary producers in shallow unvegetated
habitats, although Posidonia and associated epiphyte
detritus are probably both major contributors to food
webs and locally enhance invertebrate and fish production, including at locations away from the beds
where the production occurred.
Direct grazing of Posidonia is relatively unimportant in temperate Australian waters, as evidenced by
considerable empirical data, with the large number
of dietary studies of fishes and invertebrates in temperate beds consistently indicating an extremely low
proportion of species that directly consume Posidonia material (Klumpp et al., 1989). The only
temperate Australian fishes known to ingest large
quantities of seagrass are the garfish Hyporhamphus
melanochir and the leatherjackets Meuschenia freycineti, Monacanthus chinensis, Meuschenia trachylepis, and Acanthaluteres spilomelanurus (Bell
et al., 1978a,b, 1987; Robertson and Klumpp, 1983;
Edgar and Shaw, 1995b), while the crab Nectocarcinus integrifrons is also a Posidonia grazer (Klumpp
and Nichols, 1983; Edgar, 1996).
Seagrass beds can also affect ecosystems through
the filtration capacity of filter-feeding invertebrates
attached to seagrass leaves. Sponges, ascidians, hydroids, and other filter-feeders remove particles from
the water column, and then consolidate the particles and eventually deposit them on the seabed in
the form of feces, pseudofeces, and animal bodies.
In Cockburn Sound and Marmion Lagoon (WA),
the two areas in Australia where estimates have
been made on total filtration capacity of Posidoniaassociated animals, invertebrates were estimated to
filter the water column in some seagrass habitats over
periods of less than 1 day (Lemmens et al., 1996;
Edgar, unpublished data). Whether these estimates
are typical or anomalous it remains to be assessed.
Available evidence supports the nursery bed hypothesis for the New South Wales coast, where luderick, bream, and snapper associate as juveniles with
seagrass habitat in estuaries (Gray, 1991a,b; Gray
et al., 1996). The meagre empirical data available
on temperate coasts by Edgar and Shaw (1995a,b,c),
however, do not support the hypothesis for other
sections of the Australian coast (Jenkins et al.,
1997), with the possible exception of juvenile baldchin groper associating with coastal seagrass beds
along the central Western Australian coast (Howard
et al., 1989). Systematic surveys are urgently required to assess the validity of these hypotheses at
a national scale. Surveys are also required to determine whether the primary response of juveniles
of the important commercial species in New South
Wales is to estuaries or to seagrass habitat within the
estuaries.
In the Mediterranean Sea, P. oceanica, notwithstanding its important biomass, appears to be a minor food source (Dauby, 1989), but supports high
herbivore production due to its large primary production (Cebrian et al., 1996). The low utilization of the
enormous carbon reservoir constituted by Posidonia
leaves can be explained by high C/N ratio, their hard
consistency, coupled with a high encrusting rate by
calcareous epiphytes, making leaves difficult to assimilate by grazers (Havelange et al., 1997; Dumay
et al., 2002). Furthermore, P. oceanica contains phenolic compounds (Agostini et al., 1998; Ferrat, 2001)
able to discourage grazing as well as a large amount
of structural carbohydrates (Buia et al., 2000). In P.
oceanica meadows, no marine mammals graze but
some large marine herbivores (e.g. turtles) spend a
large amount of time (90%) at shallow depth in the
vicinity of the meadow and ingest seagrass (Hays
et al., 2002).
Many grazing studies have been conducted in
P. oceanica meadows, investigating temporal (day
to season) and spatial (small: meters and large:
100 km) patterns (Mazzella et al., 1992; Cebri´ an
et al., 1996; Sanchez-Jerez et al., 1999). Besides
Posidonia, other plant sources are available for animals, e.g. benthic micro- and macroalgae (on sediment or on rhizomes), epiphytes, and phytoplankton.
It seems that epiphyte biomass is partly controlled
by the herbivores (Alcoverro et al., 1997a).
Potential consumers of P. oceanica leaves are the
sparid fish Sarpa salpa (Velimirov, 1984), the echinoid Paracentrotus lividus (Zupo and Fresi, 1984),
the isopod Idotea baltica (Lorenti and Fresi, 1983),
the polychaete Platynereis dumerilli (Mazzella et al.,
1995), and even the green turtle Chelonia mydas in
the eastern basin (Hays et al., 2002). The importance
of these macroconsumers shows a great variation
from one bed to another along the Mediterranean
Coast (i.e. Cebrian et al., 1996). Generally, the authors assumed that these herbivores are a minor
factor in the control of P. oceanica since grazing accounts only for a small percentage of leaf
S. Gobert, M. L. Cambridge, B. Velimirov, et al.
primarily by epiphytic production. This situation
contrasts with pelagic food webs, which are primarily fueled by phytoplankton, and those occurring below the euphotic zone, which are primarily fueled
by detritus. No consensus exists about the most important primary producers in shallow unvegetated
habitats, although Posidonia and associated epiphyte
detritus are probably both major contributors to food
webs and locally enhance invertebrate and fish production, including at locations away from the beds
where the production occurred.
Direct grazing of Posidonia is relatively unimportant in temperate Australian waters, as evidenced by
considerable empirical data, with the large number
of dietary studies of fishes and invertebrates in temperate beds consistently indicating an extremely low
proportion of species that directly consume Posidonia material (Klumpp et al., 1989). The only
temperate Australian fishes known to ingest large
quantities of seagrass are the garfish Hyporhamphus
melanochir and the leatherjackets Meuschenia freycineti, Monacanthus chinensis, Meuschenia trachylepis, and Acanthaluteres spilomelanurus (Bell
et al., 1978a,b, 1987; Robertson and Klumpp, 1983;
Edgar and Shaw, 1995b), while the crab Nectocarcinus integrifrons is also a Posidonia grazer (Klumpp
and Nichols, 1983; Edgar, 1996).
Seagrass beds can also affect ecosystems through
the filtration capacity of filter-feeding invertebrates
attached to seagrass leaves. Sponges, ascidians, hydroids, and other filter-feeders remove particles from
the water column, and then consolidate the particles and eventually deposit them on the seabed in
the form of feces, pseudofeces, and animal bodies.
In Cockburn Sound and Marmion Lagoon (WA),
the two areas in Australia where estimates have
been made on total filtration capacity of Posidoniaassociated animals, invertebrates were estimated to
filter the water column in some seagrass habitats over
periods of less than 1 day (Lemmens et al., 1996;
Edgar, unpublished data). Whether these estimates
are typical or anomalous it remains to be assessed.
Available evidence supports the nursery bed hypothesis for the New South Wales coast, where luderick, bream, and snapper associate as juveniles with
seagrass habitat in estuaries (Gray, 1991a,b; Gray
et al., 1996). The meagre empirical data available
on temperate coasts by Edgar and Shaw (1995a,b,c),
however, do not support the hypothesis for other
sections of the Australian coast (Jenkins et al.,
1997), with the possible exception of juvenile baldchin groper associating with coastal seagrass beds
along the central Western Australian coast (Howard
et al., 1989). Systematic surveys are urgently required to assess the validity of these hypotheses at
a national scale. Surveys are also required to determine whether the primary response of juveniles
of the important commercial species in New South
Wales is to estuaries or to seagrass habitat within the
estuaries.
In the Mediterranean Sea, P. oceanica, notwithstanding its important biomass, appears to be a minor food source (Dauby, 1989), but supports high
herbivore production due to its large primary production (Cebrian et al., 1996). The low utilization of the
enormous carbon reservoir constituted by Posidonia
leaves can be explained by high C/N ratio, their hard
consistency, coupled with a high encrusting rate by
calcareous epiphytes, making leaves difficult to assimilate by grazers (Havelange et al., 1997; Dumay
et al., 2002). Furthermore, P. oceanica contains phenolic compounds (Agostini et al., 1998; Ferrat, 2001)
able to discourage grazing as well as a large amount
of structural carbohydrates (Buia et al., 2000). In P.
oceanica meadows, no marine mammals graze but
some large marine herbivores (e.g. turtles) spend a
large amount of time (90%) at shallow depth in the
vicinity of the meadow and ingest seagrass (Hays
et al., 2002).
Many grazing studies have been conducted in
P. oceanica meadows, investigating temporal (day
to season) and spatial (small: meters and large:
100 km) patterns (Mazzella et al., 1992; Cebri´ an
et al., 1996; Sanchez-Jerez et al., 1999). Besides
Posidonia, other plant sources are available for animals, e.g. benthic micro- and macroalgae (on sediment or on rhizomes), epiphytes, and phytoplankton.
It seems that epiphyte biomass is partly controlled
by the herbivores (Alcoverro et al., 1997a).
Potential consumers of P. oceanica leaves are the
sparid fish Sarpa salpa (Velimirov, 1984), the echinoid Paracentrotus lividus (Zupo and Fresi, 1984),
the isopod Idotea baltica (Lorenti and Fresi, 1983),
the polychaete Platynereis dumerilli (Mazzella et al.,
1995), and even the green turtle Chelonia mydas in
the eastern basin (Hays et al., 2002). The importance
of these macroconsumers shows a great variation
from one bed to another along the Mediterranean
Coast (i.e. Cebrian et al., 1996). Generally, the authors assumed that these herbivores are a minor
factor in the control of P. oceanica since grazing accounts only for a small percentage of leaf