Seeds dispersed far from adult trees on a tail of a long, skewed seed distribution might in fact have the highest chance of survival Nathan and Muller-Landau Spatially explicit demographic analysis could confirm or reject speculation, based on estimates of high seedling mortality close to V.
Such an analysis would indicate whether enough very sparsely distributed seedlings, perhaps at or m from parent plants, offset density-dependent loss of seeds and seedlings from weevils. The Virola example suggests other demographic tests that could clarify the forces acting on the dispersal process. The large birds that do most of the work for this tree in Panama favor individual trees that have small seeds and large ratios of edible aril to indigestible seed.
The trees must balance powerful selective forces: Small seeds are more likely to be dispersed and therefore to escape almost certain death under their crowns, but large seeds, if they are dispersed, produce large seedlings that are better equipped to establish and survive in the shaded understory.
A parent tree cannot maximize both dispersal and establishment by adjusting seed size alone, because a seed cannot be small and large at the same time. The optimization criteria for dispersal of a tree species may differ in different environments Forget and Sabatier For instance, an aril-to-seed ratio that is optimal for Virola in Panama might be inappropriate in Peru, where different birds could favor a different balance of pulp reward to seed ballast, or where other dispersal agents could choose fruits on the basis of other criteria.
The competitive environment for seed dispersal differs in the two sites; many more primates feed on fruits of tress in the forests of the Peruvian Amazon, and more of these trees are adapted for primate dispersal than in Panama Janson Russo found that birds in Peru, as in Panama, favor Virola calophylla fruits with high aril-to-seed ratios.
However, these preferences matter little to the trees, because almost all the fruits of this species are dispersed by a spider monkey, Ateles paniscus , which visits trees with large fruit crops. A challenging but potentially useful demographic analysis might evaluate the actual success of seeds taken by monkeys from large fruiting trees compared with those scattered by birds from all trees in the population.
Comparative studies of dispersal between close relatives indicate that the questions suggested by Virola are more than academic. For instance, comparing bird dispersal of the tree Commiphora Burseraceae in Madagascar and in mainland Africa, Bleher and Bohning-Gaese find that dispersal for the island species Commiphora guillaumini involves far fewer potential frugivores, fewer effective dispersal agents, less seed dispersal, and more spatial aggregration of juveniles around adults and of the adults themselves than for the continental species Commiphora harveyi.
The island tree has larger seeds and a much smaller aril than the mainland species, indicating a strategy of reduced reliance on animals for dispersal. It costs more seeds to make a seedling on Madagascar, and the chances are that the seedling will be under the parent plant. These contrasts suggest that there may be adaptive advantages to broad dispersal and recruitment in mainland Africa and to lowered dispersal and perhaps greater allocation to seed defense in the island flora.
The advantage of escape from the parent plant appears to be general. Focal studies of particular tree species and their dispersal show that bird dispersal of a Spanish cherry, Prunus mahaleb Rosaceae , confers an immediate advantage to local seed dispersal Jordano and Schupp In North America, bird dispersal of black cherry Prunus serotina helps seedlings avoid rapid buildup of a pathogenic pseudo-fungus Pythium spp.
Community-level analyses decisively confirm the advantages of seed escape see below ; it costs several times more to produce a seedling in dense than in sparse aggregations, and many times more to produce a reproductive adult offspring near the parent plant than farther away.
The question is no longer whether the advantage to escape from parents predicted by Janzen and Connell exists. The questions are, when and where is the effect important, and what is the relative importance of spatially defined components of seed or seedling distributions in contributing to population and community structure? Projections of plant population growth, based on the types of dispersal agents that take the seeds, may provide a window into the future in understanding how different animals bias recruitment of plants.
In a dispersal study of the giant cactus Neobuxbaumia tetezo in Mexico, Godinez-Alvarez and colleagues find that a bat Leptonycteris curasoae is probably the most effective dispersal agent for this species.
Compared with other bats, birds, and coyotes Canis latrans , Leptonycteris eats many fruits and deposits a large number of seeds under bushes and trees that could serve as nurse plants.
Given the mismatch between initial recruitment and survival of Ambrosia discussed above Miriti et al. Seed dispersal has other advantages. Seeds can colonize not only distant sites from which a given species is absent but also vacant sites in the local ecological succession.
Once thought to be primarily a means of colonizing distant places, seed dispersal may also be locally important if some species' inability to get to a region or habitat affects community composition Clark et al. The advantages of local colonization are real.
Open ground is occupied by small-seeded, fast-growing species, followed by successive invasions of larger-seeded trees that are slower to arrive but are capable of establishing under a dense secondary forest canopy. In demographic terms, seed-to-seedling and seedling-to-juvenile elasticities change with the advancing front.
For instance, Parker finds that the seed-to-seedling transition explains almost all population growth at the front of advancing populations of the invasive shrub Cytisus scoparius in California, while adult death explains population change behind the front. Directed dispersal occurs when an animal preferentially carries seeds directly to situations that are critical for seedling establishment.
Examples include bird dispersal of parasitic mistletoes to appropriate host trees Davidar , ant dispersal of seeds that establish better in and around ant nests than elsewhere Beattie , and jay or nutcracker dispersal of pine and oak seeds to forest edges or openings Vander Wall Seeds can also be dispersed to forest light gaps by male birds frequenting habitual display sites.
For example, bellbirds Procnias tricarunculata carry the seeds of a montane tree Ocotea endresiana , Lauraceae to clearings where males display to females Wenny and Levey Other birds take Ocotea fruits but do not consistently deliver them to suitable habitats figure 4.
Elaiosomes seed appendages, typically rich in fat, that attract ants or other animals and the sticky seeds of mistletoes are adaptations for directed dispersal, but many plants without these adaptations, including oaks, bird pines, and Ocotea , opportunistically use birds to disseminate seeds to a suitable environment. Colonization and escape may be synergistic. Temperate hickory Carya tomentosa seedlings in abandoned fields have both light and freedom from the rodents, rabbits, and deer that prefer not to forage in the open Myster and McCarthy Synergisms are obvious when large areas are denuded by agriculture and abandoned or are exposed by fire, earthquake, or rain-induced landslides.
At first, the colonization advantage is pronounced, as the land is populated by rapidly growing bird-, bat-, and wind-dispersed pioneers Finegan , Guariguata et al. In extensive regrowing pastures in Brazil, a few small-seeded pioneer species e. Dispersal limitation slows the successional process for larger-seeded species from deep forest into large openings.
In disturbances of a few hectares ha or less, however, dispersal of larger-seeded trees from nearby forests does occur figure 5. Spatial demography may show that seedlings that establish under the canopy of pioneer trees fare better than members of cohorts in the forest itself.
Escape from density-dependent mortality near parents may also help maintain forest diversity. Janzen argued that density-dependent seed and seedling mortality is severe enough to leave openings for other species if seed predators are species-specific. It follows that seeds escaping from parent trees are safer if they land under a tree of another species rather than one of their own.
Much as a pattern of seed distribution is a template for what may follow for a particular population, a community of seeds or seedlings is a community template of possibilities for the future Howe and Miriti The community of established plants that actually develops is the result of an array of density-dependent and density-independent processes that influence later juvenile distributions in different ways.
Spatial demography is one way to tease apart these effects. As dense patches of sibling seeds or seedlings under parents are decimated by insects or pathogens, one community-level result is the release of seedlings of different species, an effect that turns out to be general.
In rain forest, density-dependent mortality may occur in all stages of growth. Harms and colleagues recorded the densities of almost , seeds of 53 species that fell into fruit traps over 4 years in the Panamanian rain forest, and compared the distributions of species in each trap with samples from more than 13, seedlings growing nearby.
Not only were abundances of emerging seedlings inversely correlated with the number of seeds that fell into traps, the species abundance distributions of seedlings were more even, and therefore more diverse, than those of seeds in the traps. Peters further shows that density-dependent mortality occurs in sapling and adult stages in the vast majority of species that are not acutely rare in both Panamanian and Malayan rain forests; rain-forest trees survive better when neighbors are of other species.
The effect of neighbor identity might be expected to be reflected in lower elasticities for seedling-to-sapling and sapling-to-adult transitions in common trees when the trees are close to neighbors of their own species than when they are close to other species. Peters shows that the escape advantage, and the enhancement of diversity through release of less common species when local dominance is suppressed, extends well beyond seed and seedling transitions.
The escape advantage may be magnified in diverse forests, where it actually helps maintain tree diversity. Genetic variation is not random in seed and seedling cohorts. Microsatellites are hypervariable, noncoding regions of chromosomes, with several or many alleles at a locus, that behave like Mendelian genes without selection Dow and Ashley Godoy and Jordano use microsatellites to measure dispersal distance and mosaics of seedling dispositions for the well-studied European cherry P.
Microsatellite techniques are still new to the field of spatial demographics, but the results for P. Microsatellites confirm that most P. More surprising is the finding that deposition sites at some distance from fruiting trees have offspring of only one or a few parents, while those under fruiting trees, though dominated by full or half siblings, include a number of other parentages as well. A demographic question that should be addressed is whether these outsiders have a better or worse chance of growth and survival than closely related members of the cohort.
Hypervariable markers will alter the questions that can be asked of dispersal processes. For instance, does chance immigration of plants from outside a stand contribute genes that have unusual fitness in a new site? The apparent paradox of the profuse adaptations for seed dispersal and the inconsequential prospects of any given seed lies in the confounding of averaged with partitioned demographic effects.
Spatially defined demographies exist, but the benefits of isolation are subtle Miriti et al. Between the extremes of totally closed and open habitats lies most of nature, where a thoughtful application of this distinction can be enlightening. Ecological restoration reestablishes ecological patterns and processes where they have been destroyed by humans, bypassing the slow stages of natural succession.
Intensive agriculture and long-term grazing not only destroy complex and species-rich aboveground communities, they also destroy seed banks that might permit revegetation. A realistic goal of directed succession is to reestablish processes that accelerate the development of community complexity.
Demographic projections, using estimates parameterized from field or garden studies, may offer a much better chance of predicting success of key species than the usual trial-and-error approach. Reforestation of large areas disturbed by humans is often dispersal limited; most forest species are very slow to arrive. Dispersal may be encouraged within a matrix of disturbed and remnant communities.
Janzen argues that effective ecological restoration creates habitat buffers around ecological remnants and connects these habitats with corridors and stepping-stone patches, thereby increasing the movements of pollinators and dispersal agents on which plants depend Tewksbury et al. In the tropics, a dispersal pattern that promotes reforestation may be established by encouraging seed dissemination where many species are strongly dispersal limited.
Managers might place perches in fields to attract birds and their loads of seeds Miriti , Holl or plant cover of short-lived trees to encourage shade-tolerant tree seedlings and suppress competition from grasses Hooper et al. The goal of these methods is to overcome dispersal limitation by promoting seed arrival through birds and plant survival though tree cover , thereby accelerating the growth of buffers and corridors and increasing their effectiveness.
In these cases, demography could be a tool, but not a critical one. A complementary approach is to actively establish a diverse matrix to encourage processes of dispersal by and for diverse assemblages of animals and plants Martinez-Garza and Howe Properties of the matrix between habitat patches determine both recruitment in ecological remnants and migration of remnant plants into the matrix. One way to accelerate succession in tropical situations is to plant late-successional trees capable of handling the rigors of open pasture, but to choose species whose fruits will attract large-bodied mammals and birds Wunderle Small-seeded pioneers that are less dispersal limited will occupy such habitat anyway Ingle By creating food patches attractive to dispersal agents, animal-borne seeds of many species enrich community templates and ultimately change forest composition.
Tucker and Murphy provide an example in their study of reforestation in northern Australia. Trees bearing fleshy fruits planted among pioneers draw a variety of birds and mammals that appear to accelerate the succession of complex forest structure.
A restoration scheme grounded in demographic thinking would revegetate agricultural land with late-successional species that have low variance in seedling-to-juvenile transitions, thereby permitting predictable recruitment of species that provide resources for fruit-eating, seed-dispersing animals. Projections of populations in subsets of communities parameterized by different planting circumstances may permit predictions of successful and unsuccessful species decades before forests mature.
Unreliable seed dispersal has consequences for plants. Less directly, large fruit-eating animals are the first to disappear from small habitat fragments when continuous forests are subdivided into smaller patches Laurance and Bierregaard Subsistence hunting in the American tropics exterminates animals ranging from 1-kilogram kg agoutis Dasyprocta punctata to hefty kg tapirs Tapirus bairdii.
With no thinning of dominant plants by mammals, these aggressive species take over. In forests where defaunation threatens, demographic projections of the same species in hunted and unhunted areas may predict quite different forest communities 20 or 30 years into the future.
Forest fragmentation occurs when continuous forests are divided into smaller patches of varying sizes and degrees of isolation; intervening water or cropland is a barrier for forest plants and animals. Because the number of species is related to habitat area, one expects habitat fragments to lose species randomly as area declines. Where plants depend on dispersal agents, consequences of habitat fragmentation are anything but random. In the East Usambara Mountains of Tanzania, fragments have been isolated from forest by tea plantations for 60 to 80 years Cordeiro and Howe Primates and large fruit-eating birds disappear quickly from isolated forest fragments.
Seedlings and juveniles of 31 animal-dispersed tree species are three times more common in continuous forest and large forest fragments more than 30 ha than in small fragments less than 10 ha , whereas the recruitment of eight wind- and gravity-dispersed trees of the forest interior is unaffected.
Recruitment of 10 endemic, animal-dispersed tree species is 40 times lower in small fragments than in larger patches. A study of one endemic tree dispersed by birds, Leptonychia usambarensis Sterculiaceae , finds that the understory birds that commonly eat its fruits in continuous rain forest are rare or absent in small forest fragments Cordeiro and Howe Moreover, fewer seeds are taken, more seedlings accumulate under parents, and fewer seedlings or juveniles occur away from parents in small fragments than in rain forest.
Overall, seedling and juvenile recruitment near parents in small fragments is about half that in extensive forest. One might expect that an elasticity analysis would show lower elasticities for the seedling-to-juvenile transition in forest fragments, perhaps accounting for local attrition of this species from small habitat patches. In this tropical forest, dispersal matters. As in the defaunation example, projections of the same species in continuous forest and in forest fragments might help a manager gauge the consequences of loss of dispersal agents or other factors that affect tree recruitment.
At what distance, or in what local circumstances, do variances in survival decline, thereby indicating where seed dispersal is predictably effective? Spatial demography shows that average variations in seed and seedling transitions do not reflect the overall population dynamics of a species.
Where long-lived adults saturate species-poor communities, the advantages of dispersal are subtle: Dispersal of the vast majority of seeds is demographically inconsequential in any given generation, although even these subtle impacts may matter over large enough scales of time and space. In successional populations, or among long-lived trees in diverse forests, dispersal can save seeds and seedlings from intense competition or from foci of pathogen or insect infestation under parents or others of the same species.
Here the impact of slightly biased survival of dispersed versus undispersed seeds is pervasive. A challenge will be to discover how spatially contingent consequences of seed dispersal influence the fortunes of species in areas where human activities have reset successional clocks, hunted out dispersal agents, or fragmented once continuous habitats. In such circumstances, spatially contingent demography may be used to show where seed dispersal and seedling recruitment matter most in the past, in the present, and in the projected future.
Sullivan, Roman Yukilevich, Barbara Zorn-Arnold, and two anonymous reviewers for comments on the manuscript, and to Lana Gits for figure 3. Seeds from plants like dandelions, swan plants and cottonwood trees are light and have feathery bristles and can be carried long distances by the wind.
With wind dispersal, the seeds are simply blown about and land in all kinds of places. To help their chances that at least some of the seeds land in a place suitable for growth, these plants have to produce lots of seeds.
Many plants have seeds that use water as a means of dispersal. The seeds float away from the parent plant. Mangrove trees live in estuaries. If a mangrove seed falls during low tide, it can begin to root in the soil. If the seeds fall in the water, they are carried away by the tide to grow somewhere else. They have a hard seed coat that allows them to float down streams and rivers. Birds often fly far away from the parent plant and disperse the seeds in their droppings.
Plants like pittosporum have sticky seeds that can be carried away by birds. Humans can also spread seeds if they get stuck to our clothing or shoes — and if we throw fruit pips and stones out of the car window!
Some plants, like peas, gorse and flax, have seedpods that dry out once the seeds are ripe. This activity brought to you in partnership with Science Buddies. Already a subscriber? Sign in. Thanks for reading Scientific American. Create your free account or Sign in to continue. See Subscription Options. Go Paperless with Digital. Key concepts Biology Plants Evolution Aerodynamics Introduction Have you ever looked outside on a windy day and seen "helicopter" seeds spinning through the air?
Materials Examples of different seeds that are dispersed by the wind Depending on where you live, you may be able to find some of these seeds outside. If you have access to the Internet, you can also do a Web search for maple seeds, dandelion seeds and other types of wind-dispersed seeds to help get ideas.
Small, uniform, lightweight objects that you can use as "seeds" For example, you could use small paper clips or small binder clips; or purchase a bag of real seeds—such as sunflower seeds—at the supermarket.
Craft supplies to build dispersal mechanisms for your seeds These could be as simple as paper and tape or you could also use things such as streamers, cotton balls or even items you find outside, such as blades of grass. Scissors, tape and glue for cutting and attaching your craft supplies to your seeds Be careful when using scissors. A window fan or large box fan Use with caution and appropriate supervision. Stopwatch or timer optional Measuring tape or ruler optional Preparation Clear an open area in the room where you will do the seed-testing activity.
Place the fan on a table or chair, aimed across the room. You can also do the experiment outside on a windy day. Procedure Design and build several—at least four—dispersal mechanisms for your seeds. The activity works best if you can create at least two similar dispersal mechanisms to test against one another see examples below.
You can use your imagination and come up with your own ideas but here are a few to get you started using a paper clip as an example "seed" : Attach a paper clip to a small, square piece of paper, about the size of a Sticky Note, without making any changes to the paper. Attach a paper clip to another small piece of paper, but make a several parallel cuts in one side of the paper to give it "frills," and bend them outward. Attach a paper clip to a cotton ball. Attach a paper clip to a cotton ball that you have pulled on to expand it a bit and make it wispier.
Cut out some paper in the shape of a maple seed and attach a paper clip.
0コメント