ArbNation Join free

Flush cuts and stub cuts: what the research record found about decay and closure

A 1993 beech trial found more wood discoloration under flush cuts than under target cuts, and slow closure on stubs. A 2026 Mexican study measured wounds 73 percent larger after flush cuts across ten species.

Checked against the sources at the bottom of this page on October 16, 2026. Rules, fees and pay change: the source has the last word.

The facts on this page come from four research and extension papers. It is not training and does not teach the work. Each paper measured something different.

The 1993 beech trial

The Arboriculture Research Note by D. Lonsdale, issued by the Department of the Environment Arboricultural Advisory and Information Service in October 1993, reports an experiment set up in July 1986 on 40-year-old European beech (Fagus sylvatica). Three cuts were compared, each replicated 10 times. Target cuts were made just outside the branch collar, or outside the branch bark ridge where no collar showed. Flush cuts removed the whole flare of the branch base. Stub cuts left 50 mm of branch beyond the ridge. All branches were at least 50 mm across at the base. After four years the trees were cut open and sampled.

Four-year results on beech, as reported in the note
CutDiscoloration in the parent stemDieback and callus
FlushMost extensive. In some cases it ran up the stem and down it.Slight dieback. Slightly more callus than target, closing as a slit and not a circle.
TargetDownward only.Slight dieback. Better closure because the wound began small.
StubSimilar to target. The wood inside the stub was often discoloured and decayed.Much more dieback, usually to the stub base. Very little callus, always at the base.

Decay fungi were present at similar relative frequencies in the wood under all three cuts.

The note concludes that the trial supports target pruning over extreme flush or stub pruning. Avoiding flush cuts is backed by the discoloration result. Avoiding stubs is backed by delayed closure, not by more decay. The note adds that four years is short. Discoloration around the stubs might have spread further over time, because stubs stay open for many years.

The note also gives limits of its own. Its flush cuts were more extreme than most so-called flush cuts in practice. A trial by Neely (1988) on pin oak, American plane and Norway maple found cuts intermediate between target and flush closed more quickly. German work by Liese and by Dujesiefken on lime and horse chestnut also found more discoloration after flush cuts. Horse chestnut was particularly prone.

The 2010 research review

In ISA's Arborist News for February 2010, James Clark and Nelda Matheny summarise a bibliography of 201 pruning citations from 44 journals in 12 countries. The authors write that before the 1980s it was common to cut branches flush to the trunk and cover the wound with sealant. Alex Shigo's experiments, in which flush cuts gave more discoloration and decay than cuts just outside the branch collar, led to natural target pruning and the CODIT model, they write. Across many species and countries, wound dressings neither improved closure nor reduced discoloration and decay.

The same article marks where research is thin. It says removing stubs outside the collar may improve wound wood development, but its search found no documentation for this. It also finds only a tenuous scientific link between pruning and tree health.

The 2026 wound-size study

A research note in the Revista Mexicana de Ciencias Forestales, volume 17 issue 95, May to June 2026, measured wound size and not decay. The authors cut 20 branches of each of 10 urban species in Northeastern Mexico in winter 2024, branches 3 to 10 cm across. Each branch had one cut by the Hamburg method, after the branch collar and bark ridge, and one flush with the trunk.

Flush cuts left wounds 73 percent larger on average, plus or minus 15. By species the increase ran from 54 percent in Quercus virginiana to 99 percent in Diospyros texana. The differences between species were statistically significant. The authors tie much of the variation to branch angle. They describe larger wounds as reducing the tree's ability to compartmentalize, but the study did not measure decay itself.

Measured results and statements of practice

The Tree Journal Club blog of the University of Florida IFAS Gulf Coast center, November 2025, states that decades of pruning research show flush cuts are bad news. The comparison is with cuts that respect the branch collar. That sentence is a summary of the field and not a result from one trial. The post's own subject is a 2025 New Zealand study of root pruning on mature London plane, by Benson and Morgenroth.

Sources