Seed Dispersal

Fast Shingles Cure

Shingles Cure In 3 Days

Get Instant Access

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

Fig. 1. Conceptual model of movement of seeds from the parent plant to its final location emphasizing two distinct but equally important dispersal stages: Phase I which is dispersal of a seed from the parent plant to an initial surface, and Phase II which is dispersal of a seed from its initial surface to a new surface to its final position. Dispersal can be either by biotic (B) and abiotic (A) processes (P) (from Chambers and MacMahon, 1994, reprinted with permission from Annual Review of Ecology and Systematics).

Fig. 1. Conceptual model of movement of seeds from the parent plant to its final location emphasizing two distinct but equally important dispersal stages: Phase I which is dispersal of a seed from the parent plant to an initial surface, and Phase II which is dispersal of a seed from its initial surface to a new surface to its final position. Dispersal can be either by biotic (B) and abiotic (A) processes (P) (from Chambers and MacMahon, 1994, reprinted with permission from Annual Review of Ecology and Systematics).

rates near parents (escape, or Janzen-Connell hypothesis), colonizing distant disturbed, but relatively non-competitive habitats (colonization hypothesis), or finding distinct microhabitats (directed dispersal hypothesis) (Janzen, 1970; Connell, 1971; Howe and Smallwood, 1982; Harms et al., 2000). Numerous studies in terrestrial ecology have focused on various aspects of seed dispersal ranging from dispersal distances from the parent plant, characteristics of the seed that enhances seed dispersal, seed dispersion patterns around the parent plant, and the relative influence of biotic (animal mediated) vs. abiotic (wind or water) processes in mediating dispersal (reviewed in Howe and Smallwood, 1982; Chambers and MacMahon, 1994).

Terrestrial studies on seed dispersal have concentrated on the movement of seeds from the parent plant to a particular surface (often referred to as Phase I, primary, or 'coarse' dispersal). However, subsequent movement of seeds from that surface to another surface (often referred to as Phase II, secondary or 'fine' dispersal) (Fig. 1) may also be significant. Secondary dispersal is less studied than primary dispersal yet can have important consequences for vegetation structure (Chambers and MacMahon, 1994; Nathan and Muller-Landau, 2000; Wang and Smith, 2002). One of the major issues surrounding both primary and secondary dispersal is the mechanism of how a seed is actually dispersed.

While many seeds are often classified as abiotic (wind) or biotic (animal) dispersed it is likely seeds are dispersed from multiple mechanisms (Wilkinson, 1997, 1999). Higgins et al. (2003) recently argued that long distance dispersal of plants can occur frequently from non-standard mechanisms, i.e. seeds that are morphologically designed for wind dispersal can be dispersed by birds over long distances.

Seed dispersal in some seagrasses may follow the sequence outlined by Chambers and MacMahon (1994) and Nathan and Muller-Landau (2000) with Phase I or primary dispersal involving floating reproductive fragments, buoyant fruits with viable seeds, or buoyant seedlings and secondary dispersal occurring when seeds arrive at the sediment surface as 'seed rain' (Figs. 2 and 3). While the dispersal distance of a floating reproductive propagule can be quite large (103 m as noted for Z. marina reproductive fragments (Reusch, 2002; Harwell and Orth, 2002a), or Enhalus and Thalassia fruits (Kaldy and Dunton, 1999; Lacap et al., 2002) (Table 2), seeds on the sediment surface have dispersal distances one to two orders of magnitude less (Table 2). Seeds are negatively buoyant and settle rapidly to the sediment surface (Table 2) when released at the surface. Surface micro-topographic features such as sand ripples, animal tubes, or bioturbation structures such as sediment mounds can influence seed

Fig. 2. Graphic showing potential avenues (following Chambers and MacMahon, 1994) of seagrass seed dispersal to a seed's final resting position. Phase I (Primary dispersal): (a) seed release to water column can also be the fruit or either the whole plant or parts (spathe or rhipidia of Zostera), (b) seed release onto sediment surface, c. seed release below sediment surface, d. air bubble, e. buoyant fruit (Posidonia, Thalassia), f. floating shoot or fragment (Zostera): Phase II (Secondary dispersal): (a) ballast, (b) waterfowl, (c) fish, (d) sub-surface water column transport of reproductive shoot or fragments or fruits, (e) bedload transport; Phase III (Final resting position): (a) seed entrained by bottom microtopographic features (e.g. sand ripples, worm castings or feeding depressions), (b) entrainment of reproductive structure into worm tubes, (c) seed (PhyUospadix) or seedling (AmphiboHs) hooked onto macroalgae, (d) seed in the sediment seed bank.

III c

Fig. 2. Graphic showing potential avenues (following Chambers and MacMahon, 1994) of seagrass seed dispersal to a seed's final resting position. Phase I (Primary dispersal): (a) seed release to water column can also be the fruit or either the whole plant or parts (spathe or rhipidia of Zostera), (b) seed release onto sediment surface, c. seed release below sediment surface, d. air bubble, e. buoyant fruit (Posidonia, Thalassia), f. floating shoot or fragment (Zostera): Phase II (Secondary dispersal): (a) ballast, (b) waterfowl, (c) fish, (d) sub-surface water column transport of reproductive shoot or fragments or fruits, (e) bedload transport; Phase III (Final resting position): (a) seed entrained by bottom microtopographic features (e.g. sand ripples, worm castings or feeding depressions), (b) entrainment of reproductive structure into worm tubes, (c) seed (PhyUospadix) or seedling (AmphiboHs) hooked onto macroalgae, (d) seed in the sediment seed bank.

Fig. 3. Diaspores (dispersing units) for Thalassia testudinum, Posidonia australis, and Zostera marina. The large fleshy fruit of Thalassia (A) and Posidonia (C) and either flowering shoot (i), rhipidia (ii), or spathe (iii) of Zostera (E) provide for the long distance component in these species. Seeds of Thalassia (B) can float but for significantly shorter periods than the fruit. Seeds of Posidonia (D) and Zostera (F) do not float and will settle rapidly to the sediment surface but will not move far from where they settle under most conditions. Occasionally, a Zostera seed is released from the spathe with a bubble and can float several hundred meters from the parent plant (3E, reprinted with permission of Aquatic Botany (see acknowledgement section for additional information); 3A, adapted from photographs provided by J. Kenworthy and J. Kaldy).

Fig. 3. Diaspores (dispersing units) for Thalassia testudinum, Posidonia australis, and Zostera marina. The large fleshy fruit of Thalassia (A) and Posidonia (C) and either flowering shoot (i), rhipidia (ii), or spathe (iii) of Zostera (E) provide for the long distance component in these species. Seeds of Thalassia (B) can float but for significantly shorter periods than the fruit. Seeds of Posidonia (D) and Zostera (F) do not float and will settle rapidly to the sediment surface but will not move far from where they settle under most conditions. Occasionally, a Zostera seed is released from the spathe with a bubble and can float several hundred meters from the parent plant (3E, reprinted with permission of Aquatic Botany (see acknowledgement section for additional information); 3A, adapted from photographs provided by J. Kenworthy and J. Kaldy).

Table 2. Dispersal distances, buoyancy, and fall velocities of seagrass propagules.

Dispersal

Buoyancy

Fall velocity

Distance

Species

Location

unit

time

(cm s_1)

(km)

Source

Thalassia testudinum

Laguna Madre, Texas, USA

Floating fruit

1-10 days

N/A

<1-15

Kaldy and Dunton (1999)

Zostera marina

Chesapeake Bay, USA

Floating reproductive shoot

14 days

N/A

0.7-108.6

Harwell and Orth (2002a)

Zostera marina

Chesapeake Bay, USA

Seeds on sediment surface

N/A

5.96 ± 1.14

0.014

Orth et al. (1994)

Zostera marina

San Juan Island, Wasington, USA

Seeds on sediment surface

N/A

nd

Mean of 0.00127,

Ruckeishaus (1996)

max. distance <0.05

Zostera marina

northern Wadden Sea,

Floating reproductive shoot

nd

nd

33-54

Reusch (2002)

southwestern Baltic Sea

Posidonia coriacea

Perth, Western Australia

Seeds

N/A

17.1 ±2.34

nd

Orth (1999)

Thalassia testudinum

Laguna Madre, Texas, USA

Floating seeds

<1-3 days

nd

0.06-3

Kaldy and Dunton (1999)

Zostera marina

New York, USA

Floating seed with air bubble

> 40 min

nd

0.001->0.2

Churchill et al. (1985)

Enhalus acoroides

Bolinao reef system, Philippines

Floating seeds

14 h

10.0 ±0.1

3.7

Lacap et al. (2002)

Thalassia hemprichii

Bolinao reef system, Philippines

Floating seeds

3.5 h

10.0 ±0.2

18.2

Lacap et al. (2002)

Enhalus acoroides

Bolinao reef system, Philippines

Floating fruits

10.2 days

N/A

63.5 (estimated 400 km

Lacap et al. (2002)

during typhoons)

Thalassia hemprichii

Bolinao reef system, Philippines

Floating fruits

7.2 days

N/A

73.5 (estimated 300 km

Lacap et al. (2002)

during typhoons)

Enhalus acoroides

Bolinao reef system, Philippines

Seeds on sediment surface

N/A

10.0 ±0.1

0.2

Lacap et al. (2002)

Thalassia hemprichii

Bolinao reef system, Philippines

Seeds on sediment surface

N/A

10.0 ±0.2

<0.1

Lacap et al. (2002)

i CD

5"

Fig. 4. Seeds of Halodule uninervis (black spheres) in troughs of small sand waves (scale c. 1:3). Seed densities can be as high as 114 seeds cm-2 in troughs but only 5 seeds cm-2 on hummocks (from Inglis, 2000b, reprinted with permission from Journal of Ecology).

movement (Figs. 4 and 5) (Orth et al., 1994; Luckenbach and Orth, 1999). However, seeds of some species may be moved along with bed load transport of sediments as observed with H. uninervis (Inglis, 2000b). Characteristics of seeds as they settle in the water column, especially seeds with varying sizes, shapes and biomass, as well as transport characteristics of seeds on the sediment surface at different flow velocities, will be important in dispersal distance estimates (Orth et al., 1994).

The movement of seeds from the parent plant has figured prominently in the recent debate about just how far a plant can disperse (Clark et al., 1998; Howe and Miriti, 2000). Dispersal distances of most terrestrial plant species are relatively small (101-102 m). However, these same species have been noted to migrate significantly faster (103 m or greater) than would be predicted by life history and seed dispersal characteristics alone, a phenomena first noted over 100 years ago related to the tree invasion and subsequent migration to the British Isles [i.e. Reid's Paradox of Rapid Plant Migration (Reid, 1899; Pitelka and the Plant Migration Workshop Group, 1997; Clark et al., 1998; Pakeman, 2001)].

In seagrasses, data on dispersal distances of seeds or fruiting structures, such as the detached reproductive shoots in Zostera or the buoyant fruits of

Enhalus, Posidonia or Thalassia, are rare, but the emerging evidence suggests that these distances may be significantly greater than previously considered (Table 2) and may approach the dispersal distances of coconuts and mangroves, generally considered the best examples of long-distance dispersers. Surprisingly, observations made almost 100 years ago by Ostenfeld (1908) and Setchell (1929) suggested that long-distance dispersal could be attained by floating Z. marina reproductive shoots. Studies on the buoyancy potential (Table 1) of floating reproductive structures with viable seeds is a critical need in elucidating maximum dispersal distance that could be attained by various species of seagrasses. As with the terrestrial literature, interest in long-distance dispersal in now gaining attention with seagrasses (Inglis, 2000a; Harwell and Orth, 2002a).

Seeds of many terrestrial species have evolved adaptations to enhance dispersal (e.g. wings or pappi to enhance wind dispersed seeds and hooks or bristles to facilitate animal dispersed seed; Howe and Smallwood, 1982; van der Pijl, 1982). Studies on dispersal characteristics of seagrass seeds are rare but there are indications that several species may have evolved seed morphologies that allow the seed to be retained where it settles. Posidonia seeds are covered with a membranous coat, that in at least

Fig. 5. Zostera marina seedling patterns (arrows pointing to dark bands) following seed broadcasting at two locations in Chesapeake Bay: Mumfort Island in the lower York River (A) and South Bay (B), a coastal lagoon in the Delmarva Peninsula. At Mumfort Island, approximately 250,000 seeds were broadcast in fall, 1988, in a 100 x 100 m plot from both the port and starboard stern of a boat as it made several paths across the plot (close inspection of each path reveals those seedlings arising from seeds broadcast from the port or starboard side of boat. The picture was taken in June, 1989, nine months after seeding. At South Bay, approximately 150,000 seeds were broadcast in fall, 1999, from a boat at two locations in a pattern that resulted in seedlings showing the design of a 'B' and 'W' in a picture taken in July, 2000, nine months after seeding. Seedling patterns result from seeds that settle rapidly and become rapidly incorporated into the sediment matrix.

Fig. 5. Zostera marina seedling patterns (arrows pointing to dark bands) following seed broadcasting at two locations in Chesapeake Bay: Mumfort Island in the lower York River (A) and South Bay (B), a coastal lagoon in the Delmarva Peninsula. At Mumfort Island, approximately 250,000 seeds were broadcast in fall, 1988, in a 100 x 100 m plot from both the port and starboard stern of a boat as it made several paths across the plot (close inspection of each path reveals those seedlings arising from seeds broadcast from the port or starboard side of boat. The picture was taken in June, 1989, nine months after seeding. At South Bay, approximately 150,000 seeds were broadcast in fall, 1999, from a boat at two locations in a pattern that resulted in seedlings showing the design of a 'B' and 'W' in a picture taken in July, 2000, nine months after seeding. Seedling patterns result from seeds that settle rapidly and become rapidly incorporated into the sediment matrix.

one species, Posidonia coriacea, increases the fall velocity (Orth, 1999). Once settled onto the sediment surface the membrane may serve to retain the seed near where it settles by influencing hydrodynamics around the seed, minimizing disturbance by waves and currents, and facilitating burial. Seeds of Phyllospadix have two arms with stiff bristles that facilitate entanglement in a filamentous algae (Turner, 1983; Blanchette et al., 1999). Similarly, seedlings of Amphibolis have a barbed base that can catch onto algal filaments. While seeds of other species may appear at first to have no adaptations to either enhance or limit dispersal, we hypothesize that the distinct morphology of seeds influences dispersal distances. For example, the tests of Halophila seeds are highly sculptured with peg-like (Halophila spinulosa), or honeycombed projections (Halophila ovalis) that may aid seed dispersal (by trapping air bubbles), burial or shedding of the test during germination (Birch, 1981). During germination, a large mass of fine, long hairs develops on the seed surface of H. spinulosa that appear to anchor it in the sediment before the radicle emerges (this process also appears to occur with Thalassia tes-tudinum seedlings). The actual functional purpose of seed coat ornamentation in Halophila, and that displayed on seed coats in other seagrass genera (e.g. Zostera), is still to be determined. H. uninervis produces smooth, hard, oval achenes that lack obvious sculpturing or projections on the pericarp. This morphology appears to be particularly amenable to tumbling dispersal in mobile sediments, where the seeds accumulate in large numbers within small pits and depressions (Fig. 4) (Inglis, 2000b). In terrestrial plants, accessibility to suitable germination sites has been shown to be dependent on seed morphology (smooth vs. rough surfaces and size; Harper, 1977), but no studies have been conducted to date with sea-grasses to test similar hypotheses about the functional importance of seed morphology and size.

Seed dispersal in seagrasses can be influenced by the position of seed release from the parent plant. Within the seagrass genera, seeds can be released either from elevated inflorescences, or at or just below the surface of the sediment (Table 1). For those species that release seeds below the sediment surface (Halophila, Cymodocea, and Halodule) dispersal distances from the parent plant may be on the scale of centimeters. These genera also have highly persistent seed banks, so this strategy may have evolved to maximize seed dispersal in time, rather than space

(Venable and Lawlor, 1980). In this case, appropriate microsites for germination and initial seedling establishment may become available when the parent plants are disturbed or where storms or intense biotic activity (rays or manatees) cause plants, sediments, and seeds to be transported to secondary sites.

Seagrass seed dispersal can be controlled by both abiotic and biotic elements (Fig. 2). Wind and currents are important in transporting floating reproductive fragments or fruits long distances. Most measurements of seagrass propagule dispersal are indirect, involving multiplying some metric of flotation potential to speed of water flow. Harwell and Orth (2002a) showed that currents alone could move a Z. marina reproductive fragment up to 23 km in a single 6 h tidal cycle and strong winds could significantly alter that distance. Small patches of Z. marina have been found at distances of up to 108 km from the nearest source of donor material and Harwell and Orth (2002a) hypothesized that these derived from seeds carried by floating reproductive fragments. Kaldy and Dunton (1999) calculated dispersal distances of up to 3 and 15 km for seeds and fruits, respectively, for T. testudinum, while Lacap et al. (2002) calculated dispersal distances up to 3.7 and 63.5 km for seeds and fruits, respectively, for Enhalus acoroides, and up to 73.5 km for fruit of Thalassia hemprichii. Lacap et al. (2002) also postulated that fruits of Thalassia and Enhalus could be transported 300-400 km, respectively, during typhoons. Kendall et al. (2004) suggest that hurricanes may be responsible for expansion of a Syringudium filiforme meadow in St. Croix, US Virgin Islands by enhancing seed and seagrass fragment dispersal.

Human activities can also influence long distance dispersal if reproductive fragments are moved in ballast water of ocean going vessels, incorporated as wet packing material for shipment of live specimens, or attached to boat trailers. For example, the Japanese eelgrass, Zostera japonica, is thought to have been introduced accidentally to the United States west coast in the early 1900s as seeds transported in live shipments of Japanese oysters (Harrison and Bigley, 1982). Lipkin (1975) suggested that ship transport was a source of Halophila stipulacea into the eastern Mediterranean after Isthmus of Suez was breached. Nienhuis (1983) suggested the rapid spread of Z. marina into Lake Grevelingen in the Netherlands was a result of the closure of the estuary, which may have allowed floating reproductive shoots to be retained within the closed estuary rather than being exported.

Biotic agents such as waterfowl, especially migrating waterfowl, manatees, dugongs, or fish also have the potential to move seeds relatively long distances if they are ingested (Baldwin and Lovvern, 1994; Figuerola and Green, 2002; Figuerola et al., 2002; 2003). Seagrass species with relatively hard seed coats (e.g. Halodule) may survive the passage through the guts of these species better than those with softer seed coats (e.g. Zostera). Figuerola et al. (2002) found that germination of Ruppia maritima seeds was enhanced by passage through the guts of ducks but was species specific and probably related to gut structure (i.e. type of gizzard). An important but overlooked aspect may be the timing of availability of viable seeds and the presence of migrating waterfowl. For example, in Chesapeake Bay, USA, Z. marina seeds are produced in May and June and germinate in late November before wintering waterfowl are present, making it unlikely they could disperse seeds (Orth and Moore, 1986; Moore et al., 1993). Knowledge of seed release and germination periods coupled to an understanding of the feeding strategies of waterfowl that inhabit an area, will be important in elucidating the role of waterfowl in seagrass seed dispersal. In more intensively grazed meadows, the feeding activities of these large herbivores may actually stimulate seed production and germination (Peterken and Conacher, 1997). The decline of many of these large herbivore species, because of over-harvesting or habitat destruction (Jackson et al., 2001), could have implications for the natural development of new seagrass beds distant from the original parent and, ultimately, for the gene flow and genetic diversity of the different populations.

Disturbance of a seagrass bed by biotic (e.g. turtles, manatees or fish) or abiotic (e.g. storms) can result in reproductive fragments being ripped and exported from the bed. Patterson et al. (2001) examined the biomechanical properties of how Z. marina reproductive shoots become fragmented. They suggested that the characteristics of how a reproductive shoots breaks potentially allows for dispersal of some reproductive fragments with viable seeds while allowing some seeds to be retained within the bed for self-maintenance.

Biotic agents can also limit primary dispersal of seagrass seeds. Reproductive shoots carrying viable Z. marina seeds have been found cemented into the tube cap of the common tube-building polychaete, Diopatra cuprea (Harwell and Orth, 2001). Drifting plant fragments appear to be entrained into the tube cap and subsequently incorporated into the construction of the tube. However, this retention may be important in establishment of beds distant from a source if Diopatra inhabiting suitable but unvege-tated bottom captures fragments that have floated from afar. In northern Australia, seagrass seeds accumulate within trails excavated in the meadows by feeding dugongs (Inglis, 2000b).

Was this article helpful?

0 0

Post a comment