Take a core
Heartwood lets you date wood. This is the half before that: how a dendrochronologist pulls half a century out of a living tree with a hand drill and a spoon, without felling it, and what that pencil of wood goes through before it can be read.
A field guide · built around an interactive increment borer · 2026-10-17
i — the borer
Fifty years in a pencil
Move across the drawing to bore in, slide the spoon, snap the core and draw it out. Or step through it with the buttons. Every line on the core is the same ring you can see on the sawn top of the round.
The rings in this figure are modelled, not measured: 48 of them, laid out to look like oak, so the drawing can show the bore at its true scale. The bore stops 2 mm short of the pith, so the core comes out with 45 rings and the three innermost stay in the tree. That is the normal result, not a failure; the trouble section below says what to do about it.
ii — the kit
The bag
All of it fits in a daypack. Almost none of it is exotic, apart from the borer itself.
- the toolIncrement borer
A hollow auger with a T-handle. Pick a bit longer than the tree’s radius. The common bores are 4.3 mm and 5.15 mm; two-start threads cut slowly and suit hardwoods, three-start threads go faster in softwood.
- the toolExtractor spoon
The thin trough that rides with the borer. It slides in under the core and carries it out. Bend it once and it will scratch every core after.
- the bit’s friendBeeswax
Rubbed on the threads before each bore so the auger turns smoothly and backs out cleanly. Clean and dry the bit at the end of the day, and keep it sharp.
- the carryPaper straws
Drinking straws or folded paper, never sealed plastic: a wet core sweats and moulds. One core per straw, labelled on the straw.
- the recordPencil, tags, notebook
Tree ID, species, diameter at breast height, which side you bored, height, date, location. A core with no label is a stick.
- back at the benchGrooved mounts & glue
Wooden strips with a channel the core sits in. Wood glue, plus string or tape to hold the core in place while the glue sets.
- back at the benchSandpaper
A run of grits from coarse to very fine: roughly 80, 120, 220, 320, 400, and 600 if the rings are tight. The core is ready when single cells show under the lens.
- the eyeLens or microscope
A hand lens in the field. At the bench, a stereo microscope over a measuring stage, or a high-resolution scan you measure on screen.
iii — seven steps
Bark to pith, then pith to calendar
The first five happen at the tree and take a few minutes. The last two happen at a bench and take longer.
-
Ask, then choose the tree
Get permission before you bore anything: a landowner, a park, a forest service permit. Then choose trees the way the question needs. For a climate record, that means trees at the edge of what they can survive, where one thing (water, or summer warmth) rations their growth.
A comfortable tree writes wide, even, uninformative rings. The signal Heartwood crossdates against lives in stressed trees, which is why its measured master comes from bristlecones at treeline.
Coring a landmark, a yard tree with someone’s name on it, or a tree with obvious rot. Healthy trees usually seal a core hole off, but it is still a wound you put there, so it has to be worth one.
-
Find the line
Most work cores at breast height, 1.3 m up. If you want the tree’s age, go lower, because every metre of trunk below the core is years you will not see. On a slope, bore from the side, along the contour, not from the uphill or downhill face. Then picture the pith and aim the bit straight at it.
Trees on a slope lay down lopsided reaction wood on the uphill or downhill face, and those rings are pinched or swollen for reasons that have nothing to do with the weather.
Aiming a few degrees off. The bit walks past the pith and you get a core that ends in long, curved rings that never reach the centre.
-
Bore in
Wax the threads. Set the tip against the bark, push hard and turn clockwise until the threads bite. The first two or three turns are the hardest part of the day. After that the thread pulls the bit in by itself, and all you do is turn. The borer in Fig 1 advances 4.44 mm a turn: 22.5 turns for 100 mm.
The bit is hollow. As it advances, a cylinder of wood rises inside it, untouched by the cutting edge. That cylinder is the core.
Rocking the handle. A wobbling bit cuts a wavy, spiral core that snaps into pieces. Keep your wrists square and the turns even.
-
Slide the spoon in
When the bit is at or just past the depth you want, push the extractor spoon into the bore, under the core, all the way to the tip.
The spoon is what brings the core out. On its own the core stays attached to the tree at the far end.
-
Half a turn back, then draw it out
Turn the handle back, anticlockwise, about half a turn. You will feel the core let go at the tip. Draw the spoon out slowly with the core lying in it. Then back the borer out of the tree straight away, all the way.
The reverse turn shears the core off where it still joins the tree. Backing out at once matters because a borer left in a living tree tightens as the wood swells around it.
Pulling the spoon before the snap. The core stays behind, or breaks in the middle and you lose its order.
-
Straw, label, second core
Slide the core into a paper straw, bark end marked, and label it on the spot. Then take a second core from the same tree at 90° or 180° to the first.
One radius can lose a ring that the other side of the tree kept: a year so hard the tree never grew a ring all the way round. Two radii per tree, and many trees per site, are how crossdating catches missing and false rings.
-
Mount, sand, dot, measure
Let the core dry, then glue it into a grooved mount and tie it down while the glue sets. Turn it so the wood fibres stand upright: you want to sand into the end grain, the same face you would see on a sawn stump. Sand through the grits until single cells show under the lens. Then, from the bark ring inward, put a pencil dot on every decade.
1979 · pith 2000 bark · 2026The core from Fig 1, decade-dotted. One dot for each decade, two for each half-century, three for each century, so 2000 gets three. The dots let you, and anyone after you, find the same ring again without recounting.
Last comes measurement: each ring’s width to a hundredth of a millimetre, under a microscope on a measuring stage or on a high-resolution scan. That column of numbers is what goes into crossdating.
Mounted sideways, the core shows split grain, and the ring boundaries blur. Mounted with the fibres upright and sanded fine, every boundary is a clean line.
Counting the core and calling that the date. Counting drifts, because rings go missing and false rings appear. Only matching the pattern against other trees fixes the year, which is what Heartwood’s crossdating act lets you do.
iv — trouble
When the wood fights back
Every one of these will happen to you. None of them ends the day.
- You missed the pith.
Look at the inner end. If the last rings curve sharply, the pith is close; if they run nearly straight, you went wide. Field crews estimate the missing rings from that curvature, often with a transparent overlay of concentric circles, and write the estimate down as an estimate. The core in Fig 1 is a mild case: three rings left in the tree.
- The core came out in pieces.
Keep the pieces in order and mark which end faced the bark before they go in the straw. Most breaks glue back end to end in the mount. Where a break crushed a ring, write that down, so nobody later reads its width as a hard year.
- The bit will not start.
Usually a dull tip, thick or loose bark, or not enough push on the first turns. Sharpen it, wax it, lean your weight on it. A sharp borer bites on the first turn.
- The borer is stuck.
Do not force the handle. A bent or snapped bit is the expensive way to learn this. Ease it back a little at a time, and rewax next time. Never leave a borer in a tree overnight to deal with tomorrow.
- The centre is rotten or hollow.
The bit suddenly turns easily and the core comes out as crumbs. Stop there: the record is gone for this radius. Note it, and core another tree.
v — the tree
Is it bad for the tree?
Usually not much. Sometimes yes. It depends on the tree.
A small, real wound
A 5 mm hole through the sapwood is an injury. A healthy tree responds the way it does to any wound: it walls the damaged wood off and grows over the opening. Old field writing describes bore holes closing within about a year. Thin-barked, stressed or decay-prone trees cope less well.
Clean bit, few holes
Take only the cores the question needs, and clean the bit between trees where disease is a concern. Many crews now leave the hole open, because studies of the question have not found that plugging it helps. The old advice to fill every hole has largely fallen away.
vi — the tool
The borer’s own rings
Field crews often call it a Pressler borer or a Swedish borer. The two names are the tool’s whole history.
- 1800s
Max Robert Pressler (1815–1886), professor of mathematics at the forestry academy at Tharandt in Saxony, designs a hand borer for measuring how fast standing trees grow, the Zuwachsbohrer. Forestry, not dating, was the first job: how much timber is this tree adding?
- 1889
An increment borer is shown at the Paris World Exhibition.
- 1901
A. E. Douglass starts measuring ring widths in Arizona pines, and dendrochronology begins. Over the next decades the tool built for timber volume becomes the standard way to sample a living tree for the calendar.
- by 1920
In Mora, Sweden, the entrepreneur Anders Mattson buys the invention, and the Haglöf family opens a workshop in the same town. A photograph from the Haglöf workshop is dated 1920. That is where the Swedish name comes from.
- 1991
Haglöf takes over Mattson’s borer manufacturing. The makers still describe their borers as largely made by hand, work that takes, in their words, a great deal of patience and more than ordinary workshop experience.
- now
The same three moves, bore, snap and draw, still produce most of the ring-width data taken from living trees, including the networks behind the drought atlases. It is still a hand tool, turned by a person standing at a tree.
You have a mounted, dotted, measured core. It still has no date. The next move is to slide its pattern along a master until the narrow years line up, and to know when they don’t.
crossdate a core in Heartwood → the century of the method →