The finding
A new hydroponic-plus-synchrotron study of Anthyllis vulneraria subsp. polyphylla — a common European kidney vetch, and a legume — answers the question in its own title, “zinc or thallium?”, with an unexpected split. Grown from a metallicolous population off the former Raibl lead–zinc mine at Cave del Predil (Julian Alps, Italy) and dosed with each metal for 14 days, the plant genuinely hyperaccumulates thallium but merely excludes zinc (Jakovljević et al. 2026, Plant and Soil). It is, the authors argue, the first legume shown to hyperaccumulate thallium under controlled conditions.
The numbers make the split concrete. Under 2000 µM zinc, shoots reached ~4,000 µg g⁻¹ Zn while roots held ~11,000 — a shoot:root ratio well below one. Under thallium, the metallicolous plants did the opposite: at the lower dose, shoots carried ~300 µg g⁻¹ Tl (against ~120 in a non-adapted control population) with a shoot:root ratio of about three. Synchrotron X-ray fluorescence on field plants put thallium where hyperaccumulators put metal — in the root and petiole epidermis (inside the vacuoles) and along the leaf veins, reaching ~2,000 µg g⁻¹ in the basal midrib.
Why the zinc number is a trap
Here is the non-obvious part for anyone screening plants for cleanup. Zinc’s hyperaccumulation threshold is 3,000 µg g⁻¹; these shoots hit ~4,000. Score the plant on shoot concentration alone — the way many field surveys do — and you would file kidney vetch as a zinc hyperaccumulator and a candidate zinc-phytoextraction crop. The root:shoot partition says the opposite: with roughly three times more zinc in the roots than the shoots, the plant is a zinc excluder that happens to tolerate a lot of internal zinc, not a hyperaccumulator. The concentration criterion and the translocation criterion disagree, and only the second describes the physiology. An excluder’s remediation home is phytostabilisation — holding metal in roots and soil under a vegetative cap — not extraction, which needs the metal up in harvestable shoots.
The metals decouple — and that inverts the remediation case
Because the same plant hyperaccumulates thallium yet excludes zinc, the two jobs you might hire it for point in opposite directions. Kidney vetch is otherwise an appealing pioneer for a lead–zinc spoil: a nitrogen-fixing legume that needs no fertiliser on infertile ground, and — in this metallicolous population — one whose photosynthesis held steady up to 2000 µM zinc, roughly ten times the dose that flattened the non-adapted controls. Deploy it to revegetate and phytostabilise the zinc, though, and you are simultaneously pumping the site’s quieter, nastier metal — thallium — up into aboveground biomass. The very traits that make it a good colonist mobilise the wrong element.
The genuine risk: a forage legume as a thallium route
That matters because Anthyllis vulneraria is not an obscure weed. It is a pasture forage legume and the near-exclusive larval food plant of the small blue butterfly (Cupido minimus). Thallium is acutely toxic and, as Tl⁺, chemically mimics potassium, so plants take it up readily (via potassium pathways) and animals mishandle it the same way. A common, palatable legume that sends thallium to its shoots at a mine is a plausible entry point for thallium into grazers and pollinators. Two honest caveats keep this a flagged hazard rather than a demonstrated one: the study measured no bioavailability and no transfer to any animal, and — critically — it did not measure thallium in flowers or seeds, the very tissues the butterfly larvae and seed-eaters depend on. The district’s lead (historically reported up to ~60,000 µg g⁻¹ in earlier surveys) went untested here as well.
As a thallium crop, it is a weak extractor
Could the flip side — thallium phytomining — be the opportunity? Thallium is a scarce, technology-critical metal (infrared optics, superconductors, radiation detectors) and notoriously hard to remediate, so a nitrogen-fixing thallium crop sounds attractive. But this kidney vetch is a modest accumulator: ~300–880 µg g⁻¹ in hydroponics, ~2,000 in a field leaf. The genuine thallium champions run one to nearly two orders of magnitude higher — Biscutella laevigata at this same Cave del Predil site exceeds 32,000 µg g⁻¹ (Fellet et al. 2012), Silene latifolia at Allchar holds the plant record near 80,000, and Iberis intermedia was trialled for thallium recovery in France at ~3,000 (Leblanc et al. 1999). Kidney vetch also appears to hit a ceiling: at the highest thallium dose its shoot:root ratio fell back to ~1 and its stomata closed — a hint that it throttles thallium delivery to the shoot rather than pushing it to extreme levels. Its real value is as a tolerant, self-fertilising pioneer and phytostabiliser on mixed Zn–Pb–Tl spoil, with the thallium behaviour treated as a caution, not a product. One detail worth a follow-up: in the root nodules, thallium collected in the senescent, bacteria-free cortex while zinc sat in the bacteroid-rich core — raising the testable question of whether the nitrogen-fixing symbiosis is spared the thallium, not a demonstration that it is.
Bounds
This is one subspecies, one metallicolous and one non-metallicolous population, six replicates, 14 days, and micromolar hydroponic doses that are not the same as aged soil; several tissue values are read approximately from figures, and the “hyperaccumulator” verdict is explicitly under controlled conditions. Within those limits it reclassifies a plant convincingly, and connects to our thallium and zinc hubs and to the co-occurring zinc hyperaccumulator Noccaea caerulescens — but every “kidney vetch does X” here should be read as “these accessions, under these conditions, did X.”
Provenance: every load-bearing figure above was taken from the paper’s full text and independently cross-checked across multiple vendors; where the paper cites earlier work (site metal maxima, comparator species), those are named as citations, not as this study’s own measurements. See our methodology for how we source and verify.