The Gold Standard of Wood Anatomy

Chapter index

Imagine a woody stem as a structure that must remain standing while connecting roots, leaves and growing tissues. A solid, impermeable pole could provide support but would not carry water. A bundle of wide, unsupported openings might conduct fluid but would not solve every mechanical problem. Wood combines cell types, walls and connections rather than relying on one uniform substance.

In the conventional botanical definition, wood is secondary xylem. The vascular cambium produces it toward the inside of a woody stem or root. Secondary phloem develops toward the outside. The distinction matters because xylem and phloem have different transport roles: calling every conducting tissue “wood” loses the mechanism.

The USDA Forest Products Laboratory’s Structure and Function of Wood provides the biological foundation used throughout this guide. Its three connected functions—transport, support and storage—help organise an unfamiliar section before you try to name the species.

Start an observation with three questions. Which spaces could belong to conducting cells? Which walls contribute to support? Which cells and connections participate in storage and exchange? You may not answer every question from one image, but the questions tell you what additional view is needed.

For the wider plant pathway, follow the existing xylem, phloem and transpiration explanation. Wood anatomy supplies a closer view of one part of that system. It does not replace the physiology of roots, leaves, stomata and water potential.

Contents · Next chapter

The vascular cambium is a thin region of dividing cells between secondary xylem and secondary phloem. Newly produced xylem cells enlarge and differentiate. Depending on their type, they develop particular walls, connections and contents. The mature tissue is therefore a record of development, not merely a repeated stack of identical tubes.

As new wood is added, the stem increases in girth. Earlier wood remains toward the interior. A measurement taken close to the pith can differ from one taken near the bark because the tree’s age, developmental position and growth conditions differed when those cells formed. Position belongs in the sample record.

Many conducting and supporting cells are dead at functional maturity. Their walls and spaces remain useful. Other cells, especially living parenchyma in sapwood, continue metabolic work. The statement “wood is dead” is consequently too coarse for a living tree; the statement “sapwood is all living tissue” is also wrong.

Consider a labelled teaching stem with samples A and B. A comes from the outer trunk at breast height; B comes from a small branch. If their cell dimensions differ, species identity is only one possible explanation. The sampling positions already differ. A fair comparison would either match those positions or explicitly investigate their effect.

This is a practical lesson in study design. Before collecting hundreds of measurements, decide which biological comparison they represent. More measurements from mismatched positions can describe the mismatch very precisely without answering the intended question.

Contents · Previous chapter · Next chapter

Wood is organised along the stem and across it. The longitudinal direction follows the main stem axis. The radial direction runs outward from the pith. The tangential direction follows a tangent to a growth cylinder. These directions remain attached to the wood when the sample or photograph rotates.

Three standard section planes reveal different aspects of that organisation. A transverse section crosses the stem axis. A radial longitudinal section runs along the axis and toward the pith. A tangential longitudinal section also runs along the axis but crosses the radial system.

A ray can appear as a line in one view and as a group of cells in another. That change does not mean the specimen has changed identity. It means the plane has intersected the same structure differently.

Draw a cylinder, mark its centre, and label the three directions before interpreting a micrograph. Then state which dimensions the picture can actually show. The existing grain-direction rotation case develops this spatial reasoning with a changed drawing. A page’s horizontal direction is never sufficient evidence of the wood’s anatomical direction.

Contents · Previous chapter · Next chapter

A lumen is the internal space of a cell. The surrounding wall is a structured material containing cellulose, hemicelluloses and lignin. These components are organised in layers rather than mixed into an anatomically featureless solid. Wall architecture contributes to the directional behaviour of wood.

Adjacent cells need connections. Pits are regions where the wall structure is modified, allowing movement between cells through a remaining pit membrane. That membrane is part of the cell-wall system, not a living cell’s phospholipid membrane. Confusing the two leads to an incorrect explanation of transport through mature conducting tissue.

Vessel elements connect end to end through perforation plates. A simple plate has one main opening; a scalariform plate has a ladder-like arrangement. These are different structures from the pits in side walls. The IAWA hardwood feature list supplies standard terminology and illustrated definitions for identification work.

Suppose a student labels every dark boundary “cell wall” and every pale region “empty space.” Ask them to mark the observation first, then the interpretation. A tear created during preparation may also be pale. An out-of-focus wall may look discontinuous. The image alone must support the anatomical label.

The repair is specific: inspect continuity, compare neighbouring cells, change the focal plane where appropriate and consult a correctly oriented reference. Replacing an uncertain label with a more confident word does not add evidence. A useful drawing can leave a feature unresolved while clearly showing where the uncertainty lies.

Contents · Previous chapter · Next chapter

Hardwood and softwood are botanical groupings. In ordinary timber terminology, hardwoods come from flowering plants and softwoods mainly from conifers. The names do not guarantee physical hardness, density or suitability for a particular load. Those are properties to measure or establish separately.

Typical conifer wood is dominated by longitudinal tracheids, which contribute both to conduction and support. Rays provide a radial system. Some conifers have resin canals; others do not normally have them. A large circular opening in a transverse image is therefore not automatically a vessel.

Most familiar hardwoods contain vessels, fibres and parenchyma in varied arrangements. The vessels can be conspicuous, but their appearance is only part of an identification. There are biological exceptions to simple classroom rules, so an expert does not identify every unknown solely by asking whether one image appears to contain pores.

Try a deliberately incomplete case. Image A shows many small cells and three large spaces. Image B shows many large spaces dispersed among smaller cells. Without evidence of the large spaces’ boundaries and longitudinal connections, naming A “pine” and B “oak” is unjustified. The correct next step is to distinguish possible canals from vessels and examine additional features.

The Botany root places these tissues within plant diversity. The Dendrology root adds whole-plant identification. Wood anatomy becomes more useful when its evidence complements those other observations instead of pretending that a small section contains every botanical character.

Contents · Previous chapter · Next chapter

In hardwood identification, presence is often less informative than arrangement. Are vessels mostly solitary or grouped? Are they concentrated in one part of a growth increment or distributed more evenly? Does axial parenchyma occur near vessels, in bands or in another pattern? What do the rays look like in the appropriate view?

Ring-porous wood has a marked concentration of larger vessels in earlywood relative to later wood. Diffuse-porous wood lacks that strong size contrast across the increment. Intermediate patterns occur. These descriptions organise observations; they are not complete species names or universal climate labels.

Fibres provide much of the support in many hardwoods. Parenchyma cells form axial and radial systems associated with storage and exchange. Their proportions and patterns affect both anatomy and the information available for recognition. Looking only at the largest vessel can miss the combination that distinguishes two candidates.

Use a feature sheet rather than a resemblance verdict. For an invented specimen, record “vessels observed,” “larger vessels concentrated near a boundary,” “rays visible but width unresolved,” and “axial parenchyma uncertain.” Each phrase tells another observer what has been seen and what still needs work.

A useful comparison table has separate columns for observed, absent after adequate examination and unknown. The last category is essential. A blurred image cannot establish absence. Keeping unknowns visible prevents an identification key from excluding the correct candidate because the observer supplied more certainty than the image contained.

Contents · Previous chapter · Next chapter

Sapwood participates in conduction and contains living parenchyma. Heartwood has ceased conducting in the same way and may contain deposited extractives. The transition can alter colour, permeability and other properties. However, visible colour boundaries are not equally clear in every species or specimen.

Dark wood is not automatically heartwood. Stains, coatings, weathering and other changes can affect appearance. Similarly, a pale board is not automatically sapwood. Identification should distinguish the observed colour from the biological interpretation of that colour.

Imagine two photographs of the same specimen, one taken under warm indoor lighting and one under cooler daylight. Their apparent colours differ. No biological change has occurred between photographs. A colour-based identification that changes with the lighting has exposed a weakness in the evidence, not discovered a new species.

Record whether the surface is fresh, coated, weathered or otherwise altered. Compare internal anatomy where the question requires it and sampling is permitted. Avoid turning a conservation object into a destructive classroom sample: existing slides, documented offcuts or authorised collection material can provide the teaching evidence.

The broader material question belongs with The Gold Standard of Wood. Anatomical evidence can help explain why two regions behave differently, but it cannot by itself establish every durability claim about a finished object. Species, condition, environment and treatment remain relevant to its actual performance.

Contents · Previous chapter · Next chapter

Growth increments arise when the structure of newly formed wood changes through time. In many trees, seasonal differences produce recognisable earlywood and latewood. Boundaries may be sharp, gradual or difficult to recognise, depending on the plant and its growth conditions.

A visible band is not automatically one calendar year. False boundaries, locally absent rings and uncertain outer or inner limits complicate simple counting. Anatomy helps identify the boundary, but chronological assignment requires additional evidence and a suitable method.

Suppose an invented section contains twelve apparent bands. The sample lacks the pith and its outer surface has been trimmed. “The tree lived twelve years” does not follow. At most, the observer has counted twelve apparent bands in the preserved section; even their annual nature remains to be established.

The existing Dendrochronology guide explains how crossdating connects ring patterns to time. Its teaching specialist distinguishes a last preserved ring from felling and construction.

For a learner, the transfer question is whether the evidence supports a structural boundary, an annual increment, a calendar date or a historical event. Those are increasingly demanding claims. Write the strongest supported claim first, then list the additional observations needed to move further. That preserves the usefulness of anatomy without assigning it a task it has not completed.

Contents · Previous chapter · Next chapter

An anatomical measurement depends on the path from specimen to image. Section orientation, thickness, surface quality, staining, illumination and image processing can all influence what boundaries appear to be present. A precise software output cannot repair a poorly defined feature.

The research paper Quantitative Wood Anatomy—Practical Guidelines describes the chain from sampling to image analysis and the errors that can accumulate across it. Its practical lesson is to check intermediate products rather than treating the final spreadsheet as the first quality-control stage.

For an invented project, retain the original image and a separate annotated version. Mark the measured region, excluded damaged areas and the rule used for cells touching the boundary. Keep the scale information linked to the image. If a colleague cannot reconstruct which tissue was counted, the result is not yet reproducible.

Choose fields systematically or through an explicitly defined sampling design. Selecting only the sharpest, most visually dramatic area may be suitable for illustration but can bias a quantitative estimate. State whether an image is representative evidence, a teaching example or a selected detail.

A good first exercise uses supplied, calibrated micrographs rather than asking beginners to prepare sections independently. Students can compare the same image under two boundary rules and explain why counts change. They learn that method is part of the measurement, while technical preparation remains with trained staff and appropriate equipment.

Contents · Previous chapter · Next chapter

Assume a teaching micrograph contains a scale bar labelled 200 micrometres that spans 400 pixels in the original image. The scale is 200 ÷ 400 = 0.5 micrometres per pixel. A lumen measuring 120 pixels across therefore has a measured diameter of 60 micrometres in that direction.

Now suppose the image is resized to half its original width. The same lumen spans 60 pixels, and the scale bar spans 200 pixels. The correct new scale is 1 micrometre per pixel. The inferred diameter remains 60 micrometres. Resizing the picture has not shrunk the cell.

If someone carries the old calibration into the resized image, they report 60 × 0.5 = 30 micrometres. That is a 50% underestimate of diameter. If they then calculate circular area, they report one quarter of the correct area because area depends on diameter squared.

The original circular approximation gives area = π × 30², approximately 2,827 square micrometres. The mistaken result gives π × 15², approximately 707 square micrometres. The arithmetic is internally consistent in both cases; the error entered before the calculation.

This example shows why units and image provenance matter. A copied magnification label does not necessarily describe the displayed or analysed image after resizing. A correctly embedded scale bar and retained calibration record are much more useful. Also state that the circular model is an approximation: real lumina can be irregular, and the chosen diameter definition affects the result.

Contents · Previous chapter · Next chapter

Vessel density is commonly expressed as a count per transverse area. It is different from the percentage of the image occupied by vessel lumina and different again from the number of conduits along a length of stem. Name the quantity before choosing its denominator.

In an invented transverse region of 2 square millimetres, an observer counts 30 vessel lumina under a stated boundary rule. The estimated density is 30 ÷ 2 = 15 vessels per square millimetre. A second region contains 18 in 1 square millimetre, giving 18 per square millimetre.

Pooling the regions gives 48 ÷ 3 = 16 vessels per square millimetre. Simply averaging 15 and 18 gives 16.5, which weights the smaller region as heavily as the larger one. Neither operation should be chosen accidentally. For a pooled area-based density, total count divided by total measured area is the appropriate calculation here.

Suppose the first region includes a damaged area of 0.2 square millimetres that was excluded from counting. The denominator should describe the area actually examined: 30 ÷ 1.8 is approximately 16.7. Calling the full frame “measured area” would understate the density.

The research review How to quantify conduits in wood? distinguishes several conduit dimensions and derived measures. For a report, write the count, measured area, section plane, exclusion rule and sampling position together. A single density number without those details is difficult to compare fairly.

Contents · Previous chapter · Next chapter

For ideal laminar flow through a straight circular tube, conductance is proportional to radius to the fourth power when fluid viscosity and tube length are held constant. This model explains why conduit dimensions can matter greatly. It does not imply that a tree is a set of identical open drinking straws.

Consider two hypothetical tubes of equal length. Tube B has twice tube A’s radius. Its ideal conductance is 2⁴ = 16 times as large. Four tubes of A’s size provide four times A’s conductance when connected in parallel under the same pressure difference. They do not equal the single larger tube in this simplified comparison.

Now change the assumption: make B twice as long as A. Because ideal conductance is inversely proportional to length, the ratio becomes 16 ÷ 2 = 8. The result changed because a condition changed, not because the earlier calculation was arithmetically wrong.

Actual xylem introduces pit membranes, end connections, branching, changing dimensions and possible loss of conducting pathways. A mean vessel diameter does not fully describe that network. The conduit review linked in Chapter 11 provides the methodological context for this distinction.

Use the model to generate a question: could a measured structural difference affect transport? Then choose physiological measurements and anatomical detail capable of testing it. Avoid jumping directly from “larger lumen” to “this plant must transport more water” or “this species must be less drought resistant.” Whole-plant claims need more than a single geometric feature.

Contents · Previous chapter · Next chapter

An identification should narrow candidates using independent observations, not search for a photograph that feels familiar. Start with the specimen’s condition and section planes. Record features that can be assessed reliably, then compare them with a documented reference collection or key.

InsideWood explains that descriptions can represent groups of anatomically similar species, that wood varies, and that not every woody species is represented. A database match therefore produces a candidate to evaluate. It does not guarantee that the unknown belongs to the first returned name.

Create a small invented candidate set: A, B and C. All have vessels. A and B share the observed parenchyma pattern. Only B is known to have the measured ray arrangement, but that arrangement is unclear in the unknown. The evidence supports A or B; it does not yet support B alone.

The next observation should target the unresolved difference. A suitable tangential section may be more useful than another colour photograph. If the necessary structure is missing or damaged, report the identification at the level the surviving evidence permits.

A strong record includes specimen identifier, observed features, comparison sources, candidate exclusions and remaining uncertainty. “Consistent with this genus group” can be a scientifically better result than an unsupported species name. The level of certainty should follow the evidence, even when a customer, worksheet or software interface asks for a single exact label.

Contents · Previous chapter · Next chapter

Five images of one wood block provide information about that block. They are not five independently sampled trees. This distinction matters when you move from describing a specimen to estimating variation across a population.

Imagine three trees with invented mean vessel diameters of 40, 50 and 60 micrometres, each measured under the same protocol. Their equally weighted mean is 50 micrometres. Now suppose the first tree contributes ten images while the others contribute one each, and every image reproduces its tree’s mean.

Treating all twelve images as independent, equally important observations gives (10 × 40 + 50 + 60) ÷ 12 = 42.5 micrometres. The biological sample has not changed. The apparent average changed because one tree received more imaging effort.

Which average is appropriate depends on the question and sampling design. If the goal is the mean across equally sampled trees, first calculate tree-level summaries or use a model that respects the hierarchy. If the goal is an area-weighted description of a particular sampled tissue collection, define that different target explicitly.

The repair begins in the notebook: retain tree, branch, block, section and image identifiers. Do not discard those relationships when exporting a flat spreadsheet. A useful result can then distinguish within-tree variation from between-tree variation. Large image counts improve some measurements, but they do not automatically supply broad biological replication.

Contents · Previous chapter · Next chapter

Wood interacts with water in cell walls and larger internal spaces. Moisture changes can alter dimensions, and the amount of change differs by direction. The USDA moisture and physical-properties chapter explains these relationships and the limits of simple shrinkage estimates.

Use an invented small-change model for a prepared board: tangential width change is 0.25% of the starting width per percentage-point change in moisture content; radial width change is 0.12%. These are supplied exercise coefficients, not universal wood values.

For a 200 mm tangential width and a moisture-content decrease of four percentage points, predicted shrinkage is 200 × 0.0025 × 4 = 2 mm. The final width is approximately 198 mm. For a 200 mm radial width under the other coefficient, predicted shrinkage is 200 × 0.0012 × 4 = 0.96 mm.

The calculation only applies to the assumed interval, orientation and linear approximation. It does not predict warping, restraint stresses or the response of a glued assembly. Nor does rotating a photograph change a radial width into a tangential one.

This is where anatomy becomes materially useful: it identifies the directions to which a property statement belongs. For an actual product, use appropriate species and product data, moisture measurements and professional design methods. The linked Timber essay connects grain, drying and craftsmanship; appearance alone cannot supply the missing physical conditions.

Contents · Previous chapter · Next chapter

Wood anatomy can remain recognisable in altered materials, but preservation affects which features can be assessed. Charcoal, decayed wood and a fresh reference block are not interchangeable images of an unchanged specimen. Begin by recording the material’s condition and the limits it places on observation.

The existing Anthracology root follows wood charcoal into questions about fuel selection and past environments. Anatomical identification is one stage in that reasoning. Identifying a charcoal group does not directly measure woodland abundance or prove which activity produced the deposit.

Forensic wood work asks a different set of questions. Does the sample support the declared identity? Is the relevant taxonomic level distinguishable anatomically? What other methods and reference material would be needed? Kew’s wood-identification service describes complementary anatomical and chemical approaches, illustrating why one method’s limitations matter.

Suppose a hypothetical product is declared to be species A, while anatomy only resolves the group containing A and B. The result neither uniquely confirms A nor automatically disproves it. The report should state the resolution achieved and recommend an appropriate additional method if species-level discrimination is essential.

Geographical provenance is another claim. A species name is not a harvesting location, and a verified location is not by itself a complete account of lawful sourcing. Keep sample identity, analytical result and documentary claim in separate fields. Combining them in one confident sentence can conceal the exact question that remains unanswered.

Contents · Previous chapter · Next chapter

When a conclusion fails, look for the first step where observation became assumption. Repeating the final measurement may waste effort if the wrong section plane, denominator or biological unit was selected at the start.

Use a two-column notebook exercise: “what I can point to” and “what I infer.” If a statement belongs in neither column because it comes from a supplier, label it as an external claim. This simple separation reveals which evidence would change your conclusion.

For example, “a ray is three cells wide in this measured view” is an observation with a method. “The object was made locally” is an historical claim. The first does not establish the second without an argument and additional evidence. Good repair preserves valid observations while reducing the unsupported claim to its proper scope.

Contents · Previous chapter · Next chapter

Contents · Previous chapter · Next chapter

Is wood anatomy the same as dendrology?

They overlap but ask different questions. Dendrology studies woody plants and their recognition in a broader botanical setting. Wood anatomy concentrates on the structure of woody tissue. A leaf or flower can provide information unavailable in a detached board; a microscopic wood section can provide information unavailable in a distant photograph of a tree.

Can a phone photograph identify wood?

It can document appearance and sometimes useful macroscopic features. Its reliability depends on surface preparation, scale, lighting and the question. A photograph that suggests a candidate does not automatically establish species. Ask which distinguishing features are visible and whether plausible alternatives have been tested.

Does a high vessel count mean high water transport?

Not by itself. Count, dimensions, connections, conducting condition and the rest of the hydraulic pathway matter. The exercise in Chapter 12 deliberately holds several conditions fixed to show one geometric relationship. A living plant requires a richer explanation.

Why report a group rather than an exact species?

Some species share anatomical features, and a small or altered specimen may lack the necessary distinctions. A group-level result is useful when it accurately describes the evidence. False precision makes later historical, conservation or commercial decisions less reliable.

Does identifying the wood establish an object’s age?

No. Identity and chronology are separate questions. Suitable tree-ring work, contextual evidence or other dating methods may help, but none is replaced by naming the wood. Even the age of wood formation differs from the time an object was made or repaired.

What should a beginner learn first?

Learn the three section planes, the difference between an observation and an inference, and the meaning of a scale bar. Then compare documented features across known specimens. These habits transfer more reliably than memorising a collection of surface photographs.

Contents · Previous chapter · Next chapter

A first learning session can use three labelled views of one specimen. Mark the stem directions, identify which structures change appearance between planes and write down what remains uncertain. Success means explaining why the views differ while retaining the identity of the same tissue.

A second session can use the calibration and counting exercises. Keep the original images, units, boundary rules and calculations together. Ask another person to reproduce one result without verbal help. If they cannot, improve the record before adding more measurements.

A third session can compare an unknown with a small, documented candidate set. Require a reason for each exclusion and an explicit stopping point. The answer may be a group rather than a species. Evaluate whether the conclusion matches the available evidence, not whether it contains the most specific name.

For living-plant questions, continue through Botany, Dendrology and the plant transport specialist. For historical evidence, use Dendrochronology and Anthracology. For material choices, return to Wood and its grain-direction case.

Technical sources and continued reading

• USDA Forest Products Laboratory: Structure and Function of Wood, 2021 — tissue development, section planes, cells and structural relationships.
• USDA Forest Products Laboratory: Moisture Relations and Physical Properties of Wood, 2021 — water, density and dimensional behaviour.
• IAWA: microscopic features for hardwood identification, 1989 — standard feature definitions and illustrations.
• InsideWood database details — coverage, variability and interpretation of descriptions.
• Scholz and colleagues: How to quantify conduits in wood?, 2013 — conduit measurements and derived quantities.
• von Arx and colleagues: Quantitative Wood Anatomy—Practical Guidelines, 2016 — preparation, imaging and quantitative quality control.
• Royal Botanic Gardens, Kew: Wood Identification — anatomical and complementary analytical methods.

The lasting capability is to connect a visible feature to the question it can answer. When the specimen, image scale, section plane or intended use changes, return to that connection. The How X Works hub offers the wider route from a structure to its mechanism, evidence and limits.

Contents · Previous chapter · Continue to How X Works