114
R. J. Orth, M. C. Harwell and G. J. Inglis
long seeds remain viable in the sediment. Persistence
of a seed in the seed bank will depend, in part, on
whether the species in question has evolved some
mechanism for its seed to remain dormant within the
sediment. Dormancy is generally classified according to the residence time in the sediment, the timing
of initiation of dormancy (e.g. during seed development or seed release), or the mechanisms preventing
germination (e.g. environmental, physical or physiological; Baskin and Baskin, 1998). Several seagrass
genera, notably Posidonia, Thalassia, and Amphibolis, produce seeds that are already germinating
within the fruit and thus have no dormancy period or
seed bank (e.g. see Rollon et al., 2003). The remaining genera produce seeds that can remain dormant
from weeks to months or years. McMillan (1991)
showed that seeds of a variety of species are capable
of surviving longer than 12 months under laboratory conditions: Syringodium filiforme—49 months;
H. uninervis—41 months; Halophila engelmannii—
24 months; Halodule wrightii—up to 46 months.
Whether these estimates of dormancy also apply to
seeds in natural field settings remains a topic for future research. An important question that must be
resolved is whether dormancy is maintained by environmental control, seed coat inhibition, physiological characteristics of the seed, or some combination
of all three influences (Baskin and Baskin, 1998;
Orth et al., 2000).
Salinity, temperature, light, scarification of the
seed coat, and oxygen (oxygen-reduction profiles or
E h ) have been shown to be critical cues influencing
seagrass seed germination from field and laboratory
studies. Of these, salinity has generally been thought
to be the most influential factor (see review in Orth
et al., 2000). Nevertheless, there are significant questions about whether the changes in temperature and
salinity that have been shown to initiate germination
in the laboratory are actually experienced by seeds in
the sediment, where the interstitial water is relatively
poorly flushed. Moreover, most studies have not separated the confounding influences of osmotic shock
and seed coat removal (often a consequence of osmotic shock) in initiating germination. Temperature
stratification, photoperiod, and specific wavelengths
of light are important cues for the germination of terrestrial plants (Baskin and Baskin, 1998) and freshwater submersed aquatic vegetation (e.g. Choudhuri,
1966; Ferasol et al., 1995; Kimber et al., 1995) but
their influence on the germination of seagrass seeds
has not been studied. Recent studies of germination by Z. marina seeds suggest that reducing conditions, which occur in anoxic sediments may be
critical for inducing germination, in particular when
coupled to temperature changes (Kawasaki, 1993;
Moore et al., 1993; Brenchley and Probert, 1998;
Inglis, 2000c). While ecologically meaningful studies are needed to determine proximate and ultimate
causes of seed germination, it appears that sediment
reducing conditions may play a crucial role in this
process and should be considered in any future seed
germination experiment where seeds lie dormant
in sediment for some unspecified period of time.
In some terrestrial plants, seed germination also
appears to be under strong genetic control, so that
populations in temporally variable environments retain more polymorphic germination responses than
those in more stable conditions (Meyer and Allen,
1999). In seagrasses, the importance of seed production and germination also appears to vary among
populations subject to different frequencies of environmental disturbance, with observations suggesting increased flowering in areas subject to increased
stress (Phillips and Backman, 1983; Harrison, 1993).
This may reflect both proximate environmental cues
for germination and genetic predisposition. Rhode
(2002) documented strong phenotypic plasticity in
the reproductive cycle of Z. marina. Williams (2001)
has shown significant positive associations between
the genetic diversity of Z. marina populations and
the frequency and success of sexual reproduction.
More seeds germinated from a genetically diverse,
untransplanted population than from a transplanted
population that had low genetic diversity (Williams,
2001). Similar variation in seed viability and ability to germinate may be expressed in natural populations as a result of variation in the rate of
outcrossing.
IV. Seed Dispersal
All subsequent processes that influence plant population dynamics (e.g. predation and competition)
are dependent on the dispersal of a seed to some
suitable site, although it is possible that factors
influencing seed output before they are dispersed
(e.g. predation on fruit on the parent plant before
seeds are mature) can be equally important (Holbrook et al., 2000). Selective forces hypothesized
to produce plants adapted for seed dispersal include, allowing seeds to escape higher mortality
R. J. Orth, M. C. Harwell and G. J. Inglis
long seeds remain viable in the sediment. Persistence
of a seed in the seed bank will depend, in part, on
whether the species in question has evolved some
mechanism for its seed to remain dormant within the
sediment. Dormancy is generally classified according to the residence time in the sediment, the timing
of initiation of dormancy (e.g. during seed development or seed release), or the mechanisms preventing
germination (e.g. environmental, physical or physiological; Baskin and Baskin, 1998). Several seagrass
genera, notably Posidonia, Thalassia, and Amphibolis, produce seeds that are already germinating
within the fruit and thus have no dormancy period or
seed bank (e.g. see Rollon et al., 2003). The remaining genera produce seeds that can remain dormant
from weeks to months or years. McMillan (1991)
showed that seeds of a variety of species are capable
of surviving longer than 12 months under laboratory conditions: Syringodium filiforme—49 months;
H. uninervis—41 months; Halophila engelmannii—
24 months; Halodule wrightii—up to 46 months.
Whether these estimates of dormancy also apply to
seeds in natural field settings remains a topic for future research. An important question that must be
resolved is whether dormancy is maintained by environmental control, seed coat inhibition, physiological characteristics of the seed, or some combination
of all three influences (Baskin and Baskin, 1998;
Orth et al., 2000).
Salinity, temperature, light, scarification of the
seed coat, and oxygen (oxygen-reduction profiles or
E h ) have been shown to be critical cues influencing
seagrass seed germination from field and laboratory
studies. Of these, salinity has generally been thought
to be the most influential factor (see review in Orth
et al., 2000). Nevertheless, there are significant questions about whether the changes in temperature and
salinity that have been shown to initiate germination
in the laboratory are actually experienced by seeds in
the sediment, where the interstitial water is relatively
poorly flushed. Moreover, most studies have not separated the confounding influences of osmotic shock
and seed coat removal (often a consequence of osmotic shock) in initiating germination. Temperature
stratification, photoperiod, and specific wavelengths
of light are important cues for the germination of terrestrial plants (Baskin and Baskin, 1998) and freshwater submersed aquatic vegetation (e.g. Choudhuri,
1966; Ferasol et al., 1995; Kimber et al., 1995) but
their influence on the germination of seagrass seeds
has not been studied. Recent studies of germination by Z. marina seeds suggest that reducing conditions, which occur in anoxic sediments may be
critical for inducing germination, in particular when
coupled to temperature changes (Kawasaki, 1993;
Moore et al., 1993; Brenchley and Probert, 1998;
Inglis, 2000c). While ecologically meaningful studies are needed to determine proximate and ultimate
causes of seed germination, it appears that sediment
reducing conditions may play a crucial role in this
process and should be considered in any future seed
germination experiment where seeds lie dormant
in sediment for some unspecified period of time.
In some terrestrial plants, seed germination also
appears to be under strong genetic control, so that
populations in temporally variable environments retain more polymorphic germination responses than
those in more stable conditions (Meyer and Allen,
1999). In seagrasses, the importance of seed production and germination also appears to vary among
populations subject to different frequencies of environmental disturbance, with observations suggesting increased flowering in areas subject to increased
stress (Phillips and Backman, 1983; Harrison, 1993).
This may reflect both proximate environmental cues
for germination and genetic predisposition. Rhode
(2002) documented strong phenotypic plasticity in
the reproductive cycle of Z. marina. Williams (2001)
has shown significant positive associations between
the genetic diversity of Z. marina populations and
the frequency and success of sexual reproduction.
More seeds germinated from a genetically diverse,
untransplanted population than from a transplanted
population that had low genetic diversity (Williams,
2001). Similar variation in seed viability and ability to germinate may be expressed in natural populations as a result of variation in the rate of
outcrossing.
IV. Seed Dispersal
All subsequent processes that influence plant population dynamics (e.g. predation and competition)
are dependent on the dispersal of a seed to some
suitable site, although it is possible that factors
influencing seed output before they are dispersed
(e.g. predation on fruit on the parent plant before
seeds are mature) can be equally important (Holbrook et al., 2000). Selective forces hypothesized
to produce plants adapted for seed dispersal include, allowing seeds to escape higher mortality