Chapter 5 Seagrass Seeds and Dispersal Strategies
123
(initial seedling recruitment, ISR) of a population
(Eriksson, 1989). Inglis (2000a) suggested that that
seagrasses that produce large fleshy fruits (e.g. Thalassia, Posidonia) might be expected to exhibit an
ISR strategy based on the evidence that seedling recruitment had been rarely recorded in these species.
However, more recent evidence for T. testudinum in
Florida, USA, (Whitfield et al., 2004), T. hemprichii
and Enhalus acoroides in the Philippines (Olesen
et al., 2004) and both P. coriacea and Posidonia australis in Australia (Kendrick et al., 1999, personal
communication; Orth, personal observation) shows
that seedling recruitment for these large species may
occur more frequently than previously thought, suggesting that an RSR strategy may be more common
in these seagrasses.
Pathogens, predation, inter- and intra-specific
competition, nutrient limitation, movement to unsuitable germination sites, smothering by organic debris (e.g. macroalgae; Hauxwell et al., 2001; 2002;
Cambridge et al., 2002) or simply failed germination can all account for seed losses (Harper et al.,
1965; Chambers and MacMahon, 1994). Some of
these factors may be density-dependent, although
one study with Z. marina (Orth et al., 2003) did
not show density dependence. One process in particular, seed predation, has been shown to be highly
significant in terrestrial environments across broad
latitudinal gradients and can account for complete
loss of seed production in some systems (Janzen,
1971; Harper, 1977; Wenny, 2000). The few studies
to date suggest that predation may be an important
process governing the production, survival, and establishment of seagrass seeds, and can occur at both
the pre-dispersal phase as well as the dispersal phase
(Holbrook et al., 2000; Piazzi et al., 2000; Balestri
and Cinelli, 2003) (Table 3).
Terrestrial studies have shown higher seed predation under or near the parent plant (Janzen, 1970;
Connell, 1971; Harms et al., 2000) and while studies with animals have demonstrated higher predation
rates with decreasing latitude, Moles and Westoby
(2003) found no evidence of higher seed predation
towards the tropics. In addition, while larger seed
sizes might be expected to have higher predation
rates, Moles et al. (2003) found either no or weak
relationships between seed mass and post-dispersal
survival rates. There have been no studies on latitudinal gradients in seed predation in seagrasses, nor
predation studies on seeds of different sizes, but a recent study by Orth et al. (2002) found predation on
P. australis seeds in seagrass beds in Western Australia higher in seagrass compared to adjacent bare
sand (Fig. 6) suggesting similar advantages to being
away from the parent as hypothesized for terrestrial
species.
The presence of a seedling represents the end point
of dispersal, and the processes that influence the establishment of a seedling from a seed are often diverse and complex (Wenny, 2000; Wang and Smith,
2002). Even when a seed escapes direct agents of
mortality and can germinate in a suitable ‘safe site’,
its success is not ensured. Duarte and Sand-Jensen
(1990) found seedling appearance for Cymodocea
nodosa in the Mediterranean far exceeded patch formation (470 seedlings ha
−1 year
−1 vs. 45 patches
ha
−1 year
−1 , respectively). Olesen and Sand-Jensen
(1994) also found high rates of seed recruitment in Z.
marina in Denmark (0.16–0.76 m
2 ) but significant
mortality of small patches with less than 24% of
the studied cohorts remaining as individual patches
1.5–2.5 years later. Olesen and Sand-Jensen (1994)
suggested that improved anchoring, mutual physical protection and physiological integration among
shoots were responsible for greater success of larger
patches. Duarte and Sand-Jensen (1996) showed that
nutrient deficiency may be a significant source of
mortality in seedlings. Once its stored reserves are
exhausted, a young seedling must be able to extract sufficient nutrients from the surrounding sediment for maintenance and growth. In many sites,
the availability of nitrogen and phosphorus may be
severely limited by high metabolism within welloxygenated sediments or the presence of adult competitors as suggested by Olesen (1999) who found
the majority of seedlings within an established Z.
marina bed in Denmark were shaded out during
summer.
Biotic agents can negatively influence seedling
establishment. Dumbauld and Wyllie-Echeverria
(2003) showed that the long-term survival of Z.
japonica seedlings in Washington, USA, were influenced by the burrowing activities of thalassinid
shrimps. Although several studies have documented
bottlenecks in seedling survival (Inglis, 2000a; et al.,
2003), the reasons behind these failures require
greater examination.
VI. Seagrass Conservation and
Restoration: Utility of Seeds
Increasing recognition of the numerous ecological services provided by seagrasses (Costanza et al.,
123
(initial seedling recruitment, ISR) of a population
(Eriksson, 1989). Inglis (2000a) suggested that that
seagrasses that produce large fleshy fruits (e.g. Thalassia, Posidonia) might be expected to exhibit an
ISR strategy based on the evidence that seedling recruitment had been rarely recorded in these species.
However, more recent evidence for T. testudinum in
Florida, USA, (Whitfield et al., 2004), T. hemprichii
and Enhalus acoroides in the Philippines (Olesen
et al., 2004) and both P. coriacea and Posidonia australis in Australia (Kendrick et al., 1999, personal
communication; Orth, personal observation) shows
that seedling recruitment for these large species may
occur more frequently than previously thought, suggesting that an RSR strategy may be more common
in these seagrasses.
Pathogens, predation, inter- and intra-specific
competition, nutrient limitation, movement to unsuitable germination sites, smothering by organic debris (e.g. macroalgae; Hauxwell et al., 2001; 2002;
Cambridge et al., 2002) or simply failed germination can all account for seed losses (Harper et al.,
1965; Chambers and MacMahon, 1994). Some of
these factors may be density-dependent, although
one study with Z. marina (Orth et al., 2003) did
not show density dependence. One process in particular, seed predation, has been shown to be highly
significant in terrestrial environments across broad
latitudinal gradients and can account for complete
loss of seed production in some systems (Janzen,
1971; Harper, 1977; Wenny, 2000). The few studies
to date suggest that predation may be an important
process governing the production, survival, and establishment of seagrass seeds, and can occur at both
the pre-dispersal phase as well as the dispersal phase
(Holbrook et al., 2000; Piazzi et al., 2000; Balestri
and Cinelli, 2003) (Table 3).
Terrestrial studies have shown higher seed predation under or near the parent plant (Janzen, 1970;
Connell, 1971; Harms et al., 2000) and while studies with animals have demonstrated higher predation
rates with decreasing latitude, Moles and Westoby
(2003) found no evidence of higher seed predation
towards the tropics. In addition, while larger seed
sizes might be expected to have higher predation
rates, Moles et al. (2003) found either no or weak
relationships between seed mass and post-dispersal
survival rates. There have been no studies on latitudinal gradients in seed predation in seagrasses, nor
predation studies on seeds of different sizes, but a recent study by Orth et al. (2002) found predation on
P. australis seeds in seagrass beds in Western Australia higher in seagrass compared to adjacent bare
sand (Fig. 6) suggesting similar advantages to being
away from the parent as hypothesized for terrestrial
species.
The presence of a seedling represents the end point
of dispersal, and the processes that influence the establishment of a seedling from a seed are often diverse and complex (Wenny, 2000; Wang and Smith,
2002). Even when a seed escapes direct agents of
mortality and can germinate in a suitable ‘safe site’,
its success is not ensured. Duarte and Sand-Jensen
(1990) found seedling appearance for Cymodocea
nodosa in the Mediterranean far exceeded patch formation (470 seedlings ha
−1 year
−1 vs. 45 patches
ha
−1 year
−1 , respectively). Olesen and Sand-Jensen
(1994) also found high rates of seed recruitment in Z.
marina in Denmark (0.16–0.76 m
2 ) but significant
mortality of small patches with less than 24% of
the studied cohorts remaining as individual patches
1.5–2.5 years later. Olesen and Sand-Jensen (1994)
suggested that improved anchoring, mutual physical protection and physiological integration among
shoots were responsible for greater success of larger
patches. Duarte and Sand-Jensen (1996) showed that
nutrient deficiency may be a significant source of
mortality in seedlings. Once its stored reserves are
exhausted, a young seedling must be able to extract sufficient nutrients from the surrounding sediment for maintenance and growth. In many sites,
the availability of nitrogen and phosphorus may be
severely limited by high metabolism within welloxygenated sediments or the presence of adult competitors as suggested by Olesen (1999) who found
the majority of seedlings within an established Z.
marina bed in Denmark were shaded out during
summer.
Biotic agents can negatively influence seedling
establishment. Dumbauld and Wyllie-Echeverria
(2003) showed that the long-term survival of Z.
japonica seedlings in Washington, USA, were influenced by the burrowing activities of thalassinid
shrimps. Although several studies have documented
bottlenecks in seedling survival (Inglis, 2000a; et al.,
2003), the reasons behind these failures require
greater examination.
VI. Seagrass Conservation and
Restoration: Utility of Seeds
Increasing recognition of the numerous ecological services provided by seagrasses (Costanza et al.,