Skip to content
NatureHQ

ecological relationship

Seed dispersal

Seeds move to get away from their parent, where pathogens and seed-eaters concentrate. Most travel a few metres; the rare journeys that decide where a species can live are governed by weather and by animals that are not trying to help.

The intuitive account of seed dispersal is that plants are seeking new ground, and that is part of it. The larger and less obvious benefit is escape: mortality from host-specific pathogens and seed predators is concentrated directly beneath a parent plant, so simply not landing there is worth a great deal even when there is nowhere new to go. What follows from that is a set of mechanisms that look designed and mostly are not negotiated. A jay burying acorns is not planting a wood; it is caching food and forgetting some. A bird eating fruit is being paid in sugar to carry seeds in its gut, which is a genuine trade. A burr on a dog is theft. And the journeys that decide where a species can actually live — across a valley, onto an island — turn out to depend less on how well a seed flies than on whether it happened to be released into a rising gust.

Developed record · 65% complete · reviewed 2026-08-10

What this page covers

A relationship rather than an organism. Covers seed plants and the wind, water and animals that move their seeds.

Quick facts

Main benefit
Escape from mortality beneath the parent
Long-distance dispersal
Driven by rare updraughts, not seed shape
Ant dispersal
Myrmecochory — thousands of plant species worldwide
Typical distance
Metres to tens of metres for most seeds

Why it is worth leaving home

The biggest benefit of dispersal is not finding new ground — it is not landing under your parent

Well supported

Good evidence backs this, though some details remain open.

Seed dispersal confers three non-exclusive advantages: escape from density-dependent mortality near the parent, colonisation of unoccupied sites, and directed dispersal to suitable microsites. Mortality from host-specific pathogens and seed predators concentrates beneath parent plants, so escape alone can favour dispersal even where no new habitat is available.

Who this applies to
seed plants generally, best documented in tropical forest trees
Studied in
Angiospermae
Why we rate it this way, and what the caveats are
Well supportedModerate confidence

The escape advantage is robustly demonstrated. Confidence is moderate because the three advantages operate together and are difficult to separate empirically in any single system.

How far it can be extended

Density-dependent mortality near parents has been demonstrated in many unrelated species and forest types.

Caveats

  • The relative weight of escape, colonisation and directed dispersal varies by system.
  • Most dispersal is short-distance; the rare long journeys have disproportionate consequences.
  • The disperser usually gains nothing intentional — a forgotten cache is not a favour.

Still unanswered

  • How much of long-term forest composition is set by rare long-distance events?

Last reviewed 2026-08-10

The evidence (2 studies)

Underneath a parent tree, the seeds and seedlings of that species are dense, and everything specialised in eating them — fungi, insects, rodents — concentrates there. Survival rises sharply with distance. That escape benefit alone can favour dispersal in a completely full habitat with nowhere new to colonise.

Two other benefits stack on top. Colonisation reaches ground the species does not yet occupy — gaps, burnt areas, new islands. And directed dispersal delivers seeds somewhere specifically good: an ant carrying a seed into its nest has moved it into bare, fertile, protected soil, which is a better outcome than chance would give.

Words used here
Density-dependent mortality
Death rates rising with crowding, because pests and pathogens find dense patches easily.

How seeds actually move

Dispersal modes and what they cost the plant
ModeHow it worksWhat the plant pays
WindWings, plumes and dust-fine seeds slow the fallSeeds must be small, so little food reserve
WaterBuoyant or waterproof seeds floatRestricted to waterside and coastal species
BallisticThe fruit dries and splits explosivelyShort range — usually a few metres
AttachmentHooks and barbs catch fur and feathersNothing; the animal gains nothing either
IngestionFleshy fruit is eaten, seed passes throughA sugar reward, and seeds tough enough to survive a gut
CachingAnimals bury seeds and fail to retrieve all of themMost of the crop, in exchange for the forgotten remainder
AntsA fatty attachment gets the seed carried to a nestA small oil body per seed

Whether a wind-blown seed travels metres or kilometres is decided by the weather, not by the seed

Well supported

Good evidence backs this, though some details remain open.

Coupling field measurements of seed release, terminal velocity and canopy turbulence to an atmospheric transport model showed that long-distance dispersal is dominated by rare updraught events lifting seeds above the canopy. Variation in the tail of the dispersal distribution is governed by turbulence at the moment of release rather than by seed aerodynamics.

Who this applies to
wind-dispersed forest trees; modelled from two species in one forest
Studied in
Pinus taeda, Liriodendron tulipifera

You may have heard

“Seeds like sycamore keys are designed to fly a long way”

The wing shape governs how slowly a seed falls, and therefore typical dispersal distance — which is usually a few tens of metres. The journeys that let a species cross a valley depend on being caught in a rare updraught, and that is a property of the weather on the day rather than of the seed.

Why we rate it this way, and what the caveats are
Well supportedModerate confidence

Field measurement coupled to an explicit physical model, with the caveat that the events it explains are by definition almost never observed directly.

How far it can be extended

The controlling physics is canopy turbulence rather than anything species-specific, so the conclusion is expected to extend to wind-dispersed trees generally.

Caveats

  • A model coupled to measurements, not tracked seeds.
  • Two species in one forest type.
  • Animal-mediated dispersal can exceed wind and is not addressed.

Still unanswered

  • Can long-distance dispersal events be measured directly rather than modelled?

Last reviewed 2026-08-10

The evidence (2 studies)

The caching relationship is the one most often described as cooperation and is best read as a numbers game. A jay caches thousands of acorns and recovers most of them; the oak’s return is the fraction forgotten, and the acorn crop is sized accordingly. That is also why mast years work — flood the cachers with more than they can ever retrieve.

The animals doing the moving

  • Crows

    Corvid caching, and what they remember about it

  • Ants

    Myrmecochory — seeds with a fatty bribe attached

  • Trees

    Mast years and the seed crops being dispersed

  • Animal memory

    What a caching animal can actually recall

Words used here
Myrmecochory
Seed dispersal by ants, attracted by an elaiosome — a fatty body attached to the seed.
Scatter-hoarding
Storing food in many separate small caches rather than one larder. It is what makes forgetting productive.

Why leaving matters more than travelling far

The ground beneath the parent is the most dangerous place a seed can be, and the reason is its siblings.

Seeds that land under their parent are the most likely to be eaten or infected

Well supported

Good evidence backs this, though some details remain open.

Seed density declines with distance from the parent while per-seed survival rises, because host-specific seed predators, herbivores and pathogens concentrate where conspecific seeds and adults are densest. Where this holds, recruitment peaks at some distance from the parent rather than beneath it, limiting local dominance by any one species.

Who this applies to
tree species with specialised seed enemies, best supported in tropical forestDo not extend this beyond the taxa listed — the popular version over-reaches.
Studied in
Angiospermae

You may have heard

“Seeds are spread so they find room to grow”

Space is part of it and predation is usually the bigger part. The ground under the parent is dangerous precisely because so many identical seeds are there, which is what makes it worth eating for anything specialised on them.

Why we rate it this way, and what the caveats are
Well supportedModerate confidence

Supported by many field studies and a large meta-analytic literature, with genuine variation between systems and continuing difficulty separating enemy effects from seedling competition.

How far it can be extended

Distance- and density-dependent mortality has been demonstrated in many species and not found in others; effect strength varies with the specialisation of the enemies involved.

Caveats

  • Effect strength varies widely and is weaker in many temperate systems.
  • Distinguishing enemy-driven mortality from competition between crowded seedlings is difficult.
  • Independently proposed by Connell; the evidence base is for the joint hypothesis.

Still unanswered

  • What determines whether a species shows strong or negligible distance-dependence?

Last reviewed 2026-08-10

The evidence (2 studies)

The intuitive explanation for dispersal is competition for space and light, and it is part of the answer rather than most of it. The larger effect is that seed predators and pathogens specialised on one plant species concentrate exactly where that species is densest — which is directly under a fruiting adult, in a layer of thousands of identical seeds.

That produces a characteristic pattern. Seed density is highest at the parent and falls with distance; survival per seed is lowest at the parent and rises with distance. Multiply the two and recruitment peaks somewhere out from the tree rather than underneath it. Where the effect is strong, no species can fill the ground around itself, which leaves room for others — one of the standing explanations for why tropical forests hold so many tree species in so little space.

It also reframes what a dispersal mechanism is for. Distance is not the objective; escaping a concentration of enemies is, and a seed carried fifty metres by a bird can achieve that as well as one carried five kilometres by the wind. The showiest mechanisms are not automatically the most effective.

The hypothesis was proposed independently by Daniel Janzen and Joseph Connell around 1970, which is why it carries both names. It has been supported in many systems and not found in others, and what determines the difference is still being worked out.

Words used here
Recruitment
The number of seedlings that actually establish. The measurement that matters, as opposed to how many seeds arrived.
Density-dependent mortality
Death rates that rise as individuals get more crowded — usually because enemies concentrate where food is concentrated.

What a seed goes through on the way

Being carried is not a neutral journey, and for animal-dispersed seeds the trip changes the seed. Gut passage strips the flesh, scours the coat with acid and abrasion, and deposits the seed in a nutrient package. For many species this is not incidental — germination rates rise after passage through the right animal, and some seeds germinate poorly without it.

  • Gut passage: pulp removed, coat scarified, seed deposited with faeces. Time in the gut sets the distance, so a bird that perches and a mammal that ranges produce very different seed shadows.
  • Caching: a jay or a squirrel buries the seed at roughly the depth a seedling needs, and fails to retrieve a proportion of them. The forgetting is the dispersal.
  • External attachment: hooks and barbs on fur, feathers and socks. Distance depends entirely on the animal and the seed falls off wherever it happens to.
  • Ant dispersal: seeds carrying an oil-rich elaiosome are taken into the nest, the elaiosome is eaten and the seed discarded in nest soil — bare, fertile, and below the reach of fire.
  • Wind and water: no processing at all, so the coat and reserves must be adequate on arrival.

The trade-off running through all of this is size against distance. A seed with a large food reserve gives its seedling weeks of independence and is too heavy to travel far; a dust-fine orchid seed travels enormous distances and carries almost nothing, which is why it can only establish where a compatible fungus is already present to feed it. Neither strategy is better. They fail in different places.

A seed shadow — the distribution of where a plant’s seeds end up — is usually steeply peaked near the parent with a long thin tail. The rare seeds in that tail do most of the colonising, and are the hardest to measure.

Words used here
Elaiosome
An oil-rich appendage on a seed that attracts ants. They take the seed home, eat the elaiosome and discard the seed intact.
Seed shadow
The pattern of where a plant’s seeds land, relative to the parent.
Scarification
Damage to a seed coat that lets water in. Done by gut acids, abrasion, fire or frost.

Dispersal is not reproduction until something germinates

A seed that travels a kilometre and lands on rock has achieved nothing. Dispersal solves the question of where; germination decides whether anything comes of it, and the two are linked more tightly than they look.

  • Gut passage abrades a hard seed coat while transporting the seed — one event doing both jobs at once.
  • A cached seed is buried at roughly the depth a seedling needs, by an animal with no such intention.
  • Ant-dispersed seeds arrive in nest soil: bare, fertile and protected from fire.
  • Wind-dispersed seeds must be small to travel, so they carry little reserve and must establish quickly.
  • Dormancy handles the timing that dispersal cannot — a seed in the right place in the wrong season is still lost.

What happens next

What is still unsettled

  • Can long-distance dispersal be measured directly rather than modelled?

    Why it matters: These rare events set the pace at which species can track a changing climate, and almost everything known about them is inference.

  • How much modern plant distribution reflects dispersers that are now extinct?

    Why it matters: Some large fruits appear to have no living animal capable of swallowing them, suggesting they are waiting for partners that no longer exist.

The research behind this page

9 studies, newest first. Each one has a page explaining what it found and what it could not show.

You may also find this interesting

What this page is still missing

NatureHQ publishes its own gaps. This record is at 65% completeness against what we would call a finished subject.

  • no research from the last few years is attached — check for newer work
  • no popular claim about this subject has been checked yet
  • Germination — what happens after the seed lands — is not covered and is heavily searched.
  • Seed banks and dormancy are absent.
  • Invasive species spread is an obvious application and is not addressed.

Last reviewed 2026-08-10 · 9 claims · 45 search questions answered on this page