How Capillary Action Helps Trees Move Water

Capillary action contributes to water movement in trees, but it is a surprisingly minor player in what turns out to be one of the most impressive feats of physics in the natural world. On its own, capillary rise through the narrow tubes inside a tree trunk could lift water only a meter or two. The real engine pulling water to the tops of trees that stand tens of meters tall is evaporation from the leaves, which generates enormous tension through the entire water column. Understanding how capillary action fits into this broader system reveals a transport network that operates under conditions most engineers would consider impossible.

What Capillary Action Actually Does in a Tree

Capillary action is the tendency of water to climb upward in a narrow tube because water molecules are attracted both to each other and to the walls of the tube. In trees, the water-conducting tubes inside the wood, called xylem, are narrow enough for capillary forces to have some effect. Water clings to the walls of these tiny conduits, and the surface tension at the top of the water column pulls more water upward.

The problem is scale. Capillary rise depends on the diameter of the tube: the narrower the tube, the higher water can climb. In the narrowest xylem conduits, capillary action alone might push water up roughly a meter. Some of the widest conduits found in tropical hardwoods are hundreds of micrometers across, and capillary rise in those tubes is negligible. A coastal redwood can exceed 100 meters in height, and even the narrowest tracheid in a conifer trunk cannot get water remotely close to the canopy on capillary action alone. So capillary forces help keep xylem walls wet and assist in maintaining the continuous water column, but they are not the pump. The pump is something else entirely.

The Real Engine: Transpiration and the Cohesion-Tension Mechanism

The dominant explanation for how water reaches the top of a tall tree is the cohesion-tension theory. When water evaporates from tiny pores on the surface of leaves, called stomata, it creates a pulling force. Because water molecules stick tightly to one another through hydrogen bonds, that pull is transmitted all the way down through the continuous column of water in the xylem, from the leaves to the roots. The water column is essentially under tension, like a rope being pulled from the top.

Measurements with pressure chambers confirm that this tension is real and substantial. In a damp forest, the sap pressure inside xylem is typically around negative 4 to 5 atmospheres. In desert plants, it can plunge to negative 80 atmospheres.¹ That “negative” means the water is being pulled rather than pushed, a state that would cause most human-made pipes to collapse inward or fill with air bubbles. Trees get away with it because the xylem walls are rigid and the water column stays intact through the cohesive strength of water itself.

This theory has held up well for over a century, though some researchers have argued that the full picture involves multiple forces working together rather than tension alone.² Even so, the evaporation-driven tension remains the centerpiece of our understanding, with capillary action, root pressure, and other forces acting as supporting players.

The Plumbing Inside Wood

The architecture of xylem varies dramatically between the two major groups of trees, and those differences shape how efficiently and safely water moves. Conifers, like pines and spruces, conduct water through tracheids: single elongated cells stacked end to end. Flowering trees, known as angiosperms, conduct water through vessels: chains of cells whose shared walls have dissolved away to form long, continuous tubes. Vessels can be an order of magnitude longer than tracheids of similar width, but that does not automatically make them more efficient.³

What does make vessels efficient is that they can grow much wider. A wider conduit offers far less resistance to flow, so angiosperm vessels gain their hydraulic advantage primarily from diameter rather than length. In both tracheids and vessels, the cell walls where one conduit meets the next contribute a large share of the total resistance to flow, roughly 56 to 64 percent.³ These end walls are studded with tiny pores called pits, and the structure of these pits turns out to matter enormously for both flow and safety.

Conifers often have a distinctive pit design with a central thickening called a torus, surrounded by a porous mesh called the margo. This torus-margo arrangement acts like a valve: under normal conditions, water flows freely through the margo, but if an air bubble threatens from a neighboring conduit, the torus can swing over and seal the pit opening. Angiosperms generally lack this specialized valve and instead have uniformly thin pit membranes, which creates a fundamentally different set of trade-offs between moving water efficiently and keeping the system safe from failure.⁴

When the Water Column Breaks

A water column under tens of atmospheres of negative pressure is in a physically precarious state. Physicists call it metastable: it works fine as long as nothing disrupts it, but if the tension exceeds a critical threshold, dissolved gases can come out of solution and form a bubble. That bubble expands rapidly, filling the conduit with air and blocking water flow. This event is called cavitation, and the resulting air-filled conduit is described as embolized.⁵

The most widely supported explanation for how cavitation starts in living trees is air-seeding. Rather than a bubble forming spontaneously in the water column, air enters from an already-embolized neighboring conduit through those pit membranes between cells. If the pressure difference across a pit membrane exceeds what the membrane can hold back, a tiny thread of air is pulled through, and the receiving conduit rapidly fills with gas. In conifers, the torus-margo pit can resist this by sealing the aperture. Species that are highly resistant to cavitation tend to have a torus with a large overlap over the pit opening and a flexible margo that allows a tight seal.⁶

An interesting wrinkle in this story is the presence of lipid-based surfactants in xylem sap. These lipids coat conduit walls and pit membranes, and they dramatically alter the surface tension at gas-liquid interfaces. Measurements show xylem lipids can lower surface tension to around 25 millinewtons per meter at equilibrium, with dynamic values ranging from about 19 to 68 millinewtons per meter depending on conditions.⁷ This matters because surface tension is exactly what keeps air bubbles from expanding through pit pores. The presence of these surfactants means the system is more dynamic and chemically complex than a simple picture of pure water under tension would suggest.

How Trees Fix Broken Water Columns

If cavitation were always permanent, trees would steadily lose hydraulic capacity throughout their lives. Many species can, in fact, repair embolized conduits. The evidence suggests that living cells adjacent to the xylem actively push water back into gas-filled conduits, dissolving the trapped air and restoring flow.⁸ This is metabolically active work, not a passive process, and it appears to involve osmotic gradients generated by cells that surround the conduit.

What makes this process remarkable is that it can happen while the rest of the xylem is still under tension. That seems paradoxical: how do you refill a conduit with water when the surrounding water is being pulled in the opposite direction? One proposed explanation is that embolism repair works much like phloem unloading, the process by which sugars are moved out of the tree’s food-transporting tissue. Cells surrounding the embolized conduit may release solutes that draw water in osmotically, creating a local pocket of positive pressure even while the bulk xylem remains under negative pressure.⁹ Not all species can do this, and the capacity for repair varies widely across the plant kingdom.

Trees also use a longer-term strategy: simply growing new wood. Each year’s growth ring adds fresh xylem conduits to the system, replacing capacity lost to embolism in older wood. In many tree species, only the most recent few growth rings are actively conducting water at any given time.

Stomata as the Control Valve

If transpiration drives the whole system, then the stomata on the leaf surface are the control valves. These microscopic pores open to let carbon dioxide in for photosynthesis, but every time they open, water vapor escapes. The tree faces a constant trade-off: open the stomata wide to grow faster, or close them to conserve water and reduce the risk of cavitation.

When soil dries out or the air becomes very dry, leaves respond by closing their stomata. This was once thought to be a purely passive, mechanical response, but research now shows it involves active metabolic feedback. Leaves sense their own water status and adjust stomatal openings accordingly, with the plant hormone abscisic acid playing a key signaling role.¹⁰ By closing stomata before tension in the xylem reaches dangerous levels, the tree protects its water transport network from catastrophic failure. The cost is reduced photosynthesis and slower growth, but the alternative, widespread embolism, would be worse.

Root Pressure as a Backup Pump

While transpiration pull dominates during the day, a different mechanism takes over when evaporation slows, particularly at night or in very humid conditions. Roots can actively pump ions into the xylem, creating an osmotic gradient that draws water in and builds positive pressure from below. In corn roots, for example, this root pressure can reach about 4.2 bar, roughly four times atmospheric pressure.¹¹

Root pressure is what causes guttation, those droplets of water that appear at the tips of grass blades on cool mornings. In trees, root pressure is thought to help refill embolized conduits overnight, particularly in spring before the leaves have fully emerged. Some species rely on root pressure more than others. Birches and maples, for instance, generate substantial root pressure in early spring, which is why tapping a maple trunk yields sap. For tall trees, though, root pressure alone cannot push water to the top of the canopy. A pressure of 4 bar could lift water only about 40 meters against gravity, and many trees are taller than that.

Why Trees Cannot Grow Indefinitely Tall

The hydraulic system imposes a ceiling on tree height. As a tree grows taller, the water path from root to leaf tip gets longer, and friction through all those pits and conduit walls adds up. The result is that leaves at the top of a very tall tree receive water under greater tension and at a slower rate than leaves lower down. Hydraulic models of coast redwoods, the tallest trees on Earth, show that this longer flow path reduces the amount of water available to upper leaves, forcing their stomata to stay more closed and reducing photosynthesis.¹²

At some point, the carbon a tree gains from adding new height no longer justifies the hydraulic cost. Leaves at the very top grow smaller, grow more slowly, and contribute less to the tree’s energy budget. This is thought to be a major reason why even the tallest redwoods top out around 115 meters and why most tree species are far shorter. The hydraulic limitation is not a sudden wall but a gradual squeeze: the taller the tree, the harder every additional meter of height becomes.

Cold Weather and Freeze-Thaw Embolism

Temperature adds another dimension of risk to the water transport system. When xylem sap freezes, dissolved gases that were held in solution under pressure come out and form bubbles within the ice. When the ice thaws, those gas bubbles can expand into the liquid water and embolize the conduit. This freeze-thaw embolism is one reason why the widest-vesseled tropical hardwoods cannot survive in cold climates: larger conduits trap bigger bubbles, which are harder to dissolve back into solution upon thawing.

Research on spruce wood shows that freeze-thaw embolism in conifers is not simply about gas coming out of solution. The advancing ice front itself creates sharp pressure gradients that can pull air in from neighboring conduits through the pit membranes, a process related to the air-seeding mechanism described earlier. This explains why embolism tends to appear in clusters and why only a few tracheids are affected per freeze-thaw cycle: the ice front that triggers the embolism also stops the process by freezing the affected area.¹³

Visualization work on tulip tree leaves showed a distinctly different pattern of embolism during freezing compared to drought. Freeze-thaw embolism was confined to the midvein and second-order veins, with roughly 78% of those larger veins affected, whereas drought-induced embolism spread into all vein orders.¹⁴ This makes sense because larger veins contain more dissolved gas and produce larger bubbles during freezing. Trees in cold climates have evolved several countermeasures, including narrower conduits, strong root pressure to refill embolized wood in spring, and in some deciduous species, simply abandoning the affected wood and growing a fresh ring of xylem each year.

Sacrificing Leaves to Save Trunks

Trees do not protect all of their hydraulic network equally. Research on compound-leaved trees found a consistent hierarchy: stems are the most resistant to embolism, followed by petioles, with individual leaflets being the most vulnerable.¹⁵ This pattern, called hydraulic vulnerability segmentation, means that as drought intensifies, the cheapest and most replaceable parts of the tree fail first. A leaf can be regrown in weeks; a branch or trunk segment represents years of invested carbon.

By letting leaflets embolize and eventually drop, the tree reduces its total evaporating surface area, which eases the tension on the remaining water column and protects the more expensive stems. You can think of it as a series of hydraulic fuses: the weakest link blows first, preventing a catastrophic failure deeper in the system. This strategy sits alongside stomatal closure as part of the tree’s drought toolkit. Close the stomata first, shed leaves if things get worse, and protect the trunk at all costs.

The Efficiency-Safety Trade-off Across Environments

Every aspect of xylem design involves a trade-off between moving water efficiently and keeping the system safe from embolism. Wider conduits carry more water with less friction, but they are more vulnerable to cavitation. Dense, thick-walled wood resists collapse under extreme tension, but it is expensive to build and conducts less water per unit area. Pit membranes with larger pores let water through more easily, but they also let air in more readily.

This trade-off plays out visibly across different environments. Drought-adapted species from rocky, arid soils show a clear pattern: those with more efficient water transport tend to have lower safety margins against cavitation.¹⁶ Desert shrubs and dry-site conifers tend to build narrow, thick-walled conduits that sacrifice flow rate for resilience. Tropical rainforest trees, growing where water is abundant, invest in wide vessels that move enormous volumes of sap but would fail quickly under drought conditions. This is why transplanting a tropical species to a Mediterranean climate often ends poorly, even if temperatures are suitable. The hydraulic architecture is built for a different risk environment.

Synthetic Trees That Mimic the Same Physics

The physics behind tree water transport has inspired engineers to build synthetic systems that move water passively, without pumps, using the same principles. Researchers have fabricated artificial trees consisting of arrays of millimeter-scale tubes embedded in a nanoporous ceramic disk. The ceramic mimics a leaf’s evaporative surface, generating negative pressure that pulls water upward through the tubes, which act like xylem conduits. After saturating the system by boiling it underwater to remove trapped air, these devices passively lifted water more than 3 meters above the reservoir.¹⁷

Three meters is modest compared to a real tree, and the challenge of keeping a synthetic system free of embolism at larger scales remains a major hurdle. Real trees have had hundreds of millions of years of evolutionary refinement to manage air-seeding, repair broken water columns, and adjust transpiration rates in real time. Synthetic trees cannot yet do any of that. But the fact that the basic principle works in an engineered system confirms something striking about what capillary action, cohesion, and evaporation-driven tension can accomplish together: you can move water against gravity with no moving parts and no external power source, as long as you can maintain a continuous liquid column and generate enough evaporation at the top. Trees figured this out long before anyone built a pressure chamber to measure it.