On 9 August 1850, a crate of forty Chinese and Japanese plants arrived at Kew Gardens from the Bavarian physician and botanist Philipp Franz von Siebold, who ran a commercial nursery near Leiden after years collecting specimens in Japan. Item thirty-four in that crate was labelled Polygonum sieboldii, a knotweed the Japanese called itadori, or “pain puller,” for its use in traditional medicine. The plant had already impressed European growers before that crate reached Kew: it won a gold medal from the Agricultural and Horticultural Society of Utrecht in 1847, three years before Siebold shipped a specimen on to Britain. Kew planted the knotweed, propagated it, and by the following decade nurseries across Britain were selling it as an ornamental border plant valued for its bamboo-like canes and dense summer foliage.
The specimen shipped from Leiden was female and genetically male-sterile, unable to produce viable seed without a compatible pollen source nearby. That single detail turned out to matter enormously. Because the plant Siebold shipped to Kew cannot reproduce sexually here, every stand of Japanese knotweed descended from it has spread by vegetative means alone, broken rhizome and stem fragments regenerating into new plants. Researchers and popular accounts alike have long described the result as a single genetic clone occupying the majority of Britain’s knotweed-infested ground, and the botanical evidence for the main invasive lineage, Fallopia japonica var. japonica, broadly supports that picture.
The story has picked up some rough edges since. A smaller, less aggressive variety of the plant, Fallopia japonica var. compacta, also grows in the UK with its own separate history, and where the main clone has occasionally hybridised with related species such as giant knotweed, it has produced genetically distinct offspring including the plant known as Bohemian knotweed. Studies of knotweed populations across regions have also found measurable epigenetic differences, chemical marks that alter how genetically identical plants behave, between stands that are otherwise clonal copies of each other. So “one clone” is a reasonable shorthand for how the plant spread; individual stems can still differ at the epigenetic level despite sharing near-identical DNA.
From ornamental to national liability
What began as a fashionable garden plant is now the species most associated with devalued property in Britain. Japanese knotweed’s rhizomes can extend several metres from a visible stand and push through cracks in tarmac, brickwork and drainage, which has made its presence a routine trigger for mortgage lenders to demand a management plan before approving a loan. Local and national government bodies list it as controlled waste that cannot simply be dug up and taken to landfill, and removal contracts for a single residential infestation commonly run into the thousands of pounds.
The plant’s scale is now continental rather than local. It has been recorded across the large majority of Britain’s ten-kilometre survey grid squares, and equivalent lineages are established through much of mainland Europe, North America and New Zealand. Decades of herbicide spraying, root excavation and physical barriers have slowed individual infestations without producing a general method that reliably removes the plant from a site. That gap, a lot of practical experience but not much rigorous comparative evidence, is what a team based at Swansea University set out to close with a purpose-built field trial.
The largest trial ever attempted
Led by Daniel Jones, working with Daniel Eastwood and colleagues in the university’s Department of Biosciences and the knotweed-management company Advanced Invasives, the study ran nineteen different physical and chemical treatments across fifty-eight field plots, each measuring fifteen by fifteen metres, or 225 square metres. The plots were spread across three sites in south Wales: Lower Swansea Valley Woods, Swansea Vale Nature Reserve and Taffs Well. Treatment and monitoring ran for three years at each site, from 2012 to 2014 at two of them and 2013 to 2015 at the third, with growth tracked across 348 smaller monitoring patches nested inside the plots.
The treatments compared herbicide applications, including glyphosate delivered by foliar spray and by stem injection, against combinations of digging, cutting and physical barriers, applied at different frequencies and in different seasons. It is, by plot count and treatment range, the largest controlled comparison of knotweed management methods published to date. The results appeared in 2018 in the journal Biological Invasions.
A result that reset expectations
No treatment eradicated the knotweed from any plot within the three-year study window.
That is the study’s central finding, and for an experiment of this scale, a sobering one. Some approaches did far better than others. Herbicide regimes applying glyphosate twice a year, in summer and again in autumn, or annual autumn treatment by stem injection or foliar spray, produced the largest reductions in stem density and the area covered by knotweed growth. Digging and barrier-based methods without a chemical component performed comparatively poorly. But even the best-performing regimes left living rhizome in the ground, capable of resprouting once treatment stopped. The researchers set out to measure something narrower than eradication, which they judged close to unachievable with current tools: how far a site could be pushed toward long-term dormancy.
That distinction has practical weight for anyone managing an infested site today. It means a knotweed management plan should be built around suppression and monitoring over several years rather than a single treatment promising removal, and it explains why credible commercial contracts now typically specify multi-year herbicide programmes with follow-up inspection rather than a one-off dig-and-dispose job. It also strengthens the case for approaches that do not rely on chemicals or excavation at all. CABI, the UK’s Centre for Agriculture and Bioscience International, has spent more than a decade developing a biological control programme using a Japanese sap-sucking insect, the psyllid Aphalara itadori, which feeds specifically on knotweed and was first released into the British countryside in 2010 under strict containment trials. The original population struggled to survive British winters in the wild, which is why CABI has since released a cold-hardier strain gathered from northern Japan and nicknamed the Murakami psyllid, at UK sites from 2021, with its establishment still being monitored. It will not clear a garden overnight either, and unlike a herbicide programme it would not need reapplying every season once established, but it starves the plant from a place herbicide application can miss too — chronic aboveground feeding that, over years, drains the underground reserves a knotweed stand needs to resprout.
Whether that patience pays off is still an open question. What the Swansea trial did settle is the more immediate one: for now, managing Japanese knotweed means budgeting for suppression on a multi-year timeline. A single treatment that makes the problem disappear isn’t on offer.