The plants looked fine. That is the detail that should make everyone in this industry sit up. Four varieties of medical cannabis, grown under commercial-style conditions, hit harvest at normal height, with normal stems, normal biomass and normal-looking flowers. Nothing about them said “contaminated.” Then the lab ran the numbers on the buds, and two of the four heavy metals tested came back far above the World Health Organization’s ceiling for medicinal plants — one of them by as much as thirty-eight times.
The work comes from Nirit Bernstein’s group at Israel’s Volcani Institute, with collaborators at the Hebrew University of Jerusalem, the Technion and Colorado State University, and was published in the Journal of Cannabis Research on 27 July 2026. You can read the full open-access paper here. It is the first study to track how heavy metals move through a drug-type medical cannabis plant — where they end up, and what they do to the chemistry that makes the plant worth growing in the first place.
Key Takeaways
Researchers fed four medical cannabis cultivars a low-dose cocktail of cadmium, lead, nickel and cobalt through the irrigation water, then measured where those metals travelled and what happened to the cannabinoids. At the higher of the two doses tested — still a modest concentration, far below the levels used in most earlier hemp experiments — cadmium and nickel in the dried flowers exceeded WHO limits for medicinal plants in every single cultivar, while lead stayed comfortably within bounds. The roots hoarded most of the metal load, acting as a filter, but nickel slipped through that defence more easily than the rest and reached the flowers efficiently. Cannabinoid production fell in three of the four varieties, with the drop depending on both the dose and the genetics; one cultivar lost potency at the lower dose. And the variety that soaked up the most metal was, counterintuitively, the one whose cannabinoid profile was completely unaffected — meaning the plant’s chemistry gives you no warning at all about what is in it.
A Dose So Small the Plants Barely Noticed
The experimental design is the reason this paper matters. Bernstein’s team did not poison anything. They took four commercial medical cultivars — PO-10, Dov, LD and QJ — and added a mixture of cadmium, lead, nickel and cobalt to the standard fertiliser solution at just two concentrations: 1 and 5 micromolar of each metal, against a clean control. Five micromolar is a trace. Earlier work on industrial hemp typically used cadmium at milligram-per-litre concentrations, orders of magnitude higher.
At that low exposure, the plants essentially shrugged. Height, the number of internodes on the main stem, stem diameter and the dry weight of leaves, stems, roots and flowers came out statistically indistinguishable from the untreated controls in all four varieties. The only structural casualty was the small sugar leaves around the flowers in two cultivars, which came in lighter at the top dose. There was some visible yellowing of the foliage at 5 µM, and one variety — QJ — showed reduced photosynthesis and leakier cell membranes, a classic fingerprint of oxidative stress. But a grower walking the rows would have seen a normal crop.
That is precisely the problem. Contamination at this level is invisible. Where the metals came from in this experiment was a controlled fertiliser solution, but in the real world the entry points are mundane: irrigation water, trace impurities in fertiliser salts, foliar sprays, metallic processing equipment, and above all the growing substrate. Anyone weighing up whether to reuse a batch of spent cannabis soil should note that metals do not degrade the way organic residues do — they accumulate, cycle after cycle.
Where the Metals Went: Roots Hoard, Nickel Escapes
Cannabis, it turns out, has a decent defensive strategy. The roots grabbed by far the highest concentrations of every metal tested and held onto them — a sequestration tactic that keeps toxic elements away from the sensitive machinery in the shoot. The general gradient at the top dose ran roots, then stems, then leaves, with the flowers and sugar leaves at the bottom of the pile.
Decent, but not airtight. The researchers calculated a translocation factor for each metal — how much of what the root took up actually made it upstairs — and the ranking was unambiguous. Nickel travelled best, lead worst, with cadmium and cobalt in between. Nickel was also the exception that broke the general pattern: it reached higher concentrations in the reproductive organs than in the stems, effectively skipping a stop on the way to the part of the plant people consume. Lead, by contrast, got locked down in root vacuoles and cell walls and largely stayed put.
There is a further wrinkle. For cadmium and cobalt, the translocation factor was actually lower at the higher dose, meaning the root’s filtering response tightened up as the pressure increased. That is good news for the plant. It is not good enough news for the consumer, as the flower data made clear.
The Flowers Failed the WHO Test, and It Was Not Close
The WHO publishes maximum concentrations for toxic metals in herbal medicines: 10 micrograms per gram for lead, 1.5 for nickel and 0.3 for cadmium. The team measured its dried flowers against that yardstick.
Lead passed easily. Across the four cultivars at the top dose, flower lead ran from roughly 0.3 to 0.78 µg/g — well inside the limit, exactly as the low translocation factor predicted.
Cadmium and nickel did not pass. Cadmium in the inflorescences ranged from about 6.7 to 11.5 µg/g, which is between twenty-two and thirty-eight times the WHO ceiling. Nickel ran from roughly 4.4 to 6.7 µg/g, three to four and a half times over. Every cultivar failed on both metals. And remember what produced those numbers: a trace-level cocktail in the feed water, delivered to plants that showed no meaningful loss of growth or yield.
Potency Is a Casualty Too, and It Depends on Genetics
The second half of the study asked what the metals did to the cannabinoids themselves — and the answer is that they interfere, but unevenly.
Cannabis builds its cannabinoids in acidic form first: THCA, CBDA, CBCA and CBGA, which later convert to the familiar neutral THC, CBD, CBC and CBG. At the higher dose, all four acidic cannabinoids dropped in PO-10 and QJ. In LD, only THCA fell. In Dov, nothing moved at all. Totals told a similar story: no change in Dov, broad declines in PO-10 and QJ, and losses limited to THC and CBD in LD. QJ was the outlier in a different direction — all four of its neutral cannabinoids were already down at the lower dose, and QJ was also the only variety with elevated membrane leakage, hinting that oxidative damage and the potency loss are linked.
Two conclusions follow. First, there is a threshold effect sitting somewhere between the two doses tested, which means a cultivation programme could plausibly stay under it. Second, cannabinoid biosynthesis is far more sensitive to the plant’s mineral environment than the industry generally acknowledges — a theme running through years of work from the same lab, including research showing that zinc levels in the nutrient mix shift cannabinoid potency in both directions.
The Dov Paradox: Clean Chemistry, Dirty Flower
Now the finding that should keep quality managers awake. Dov, a CBD-rich cultivar with a cannabinoid profile closer to industrial hemp than to a typical drug-type variety, was the only cultivar whose cannabinoid production was completely untouched by the metals. It was also, consistently, among the highest accumulators of every metal tested — including in the flowers, where it posted the worst cadmium figure of the four at 11.5 µg/g.
Read that together and the implication is stark: the plant’s chemistry is not a warning light. A variety can absorb the most contamination while producing a flawless-looking cannabinoid report. Potency data, terpene data, appearance, yield — none of it flags the problem. Only an elemental analysis does. That is a useful thing to keep in mind when reading a certificate of analysis, and it sits alongside the broader gap between what a dispensary label tells you and what it actually guarantees.
Bernstein’s team also notes the flipside: hemp’s talent for hoovering up metals is exactly why it is studied for cleaning contaminated land. Dov would make a promising phytoremediation candidate, except that it produced the least biomass of the four — you would need to breed it for vigour first.
What This Study Does Not Say
Honesty about limits matters here. This was a controlled pot experiment in perlite with five plants per treatment, four cultivars, one growing cycle and a clean four-metal cocktail delivered through the irrigation line. Real soil is messier: metals bind to organic matter and clay, compete with one another for uptake sites, and become more or less available depending on pH. The study does not tell you how much cadmium is in the jar you bought last week, and it was not designed to. Nor did it measure terpenes, or follow what survives extraction and processing.
What it does establish is a mechanism and a warning. Under conditions that produce a visually perfect crop, a drug-type cannabis plant can deliver flowers carrying tens of times the WHO limit for cadmium, and its potency can quietly erode at the same time — with the extent of both depending on which genetics you happen to be growing. The authors call the results alarming and argue they should feed into international guidelines for cannabis cultivation. On the evidence presented, that is not an overstatement.
Frequently Asked Questions
Not on the basis of this study alone, which grew plants under deliberate metal exposure rather than sampling the market. But the broader concern is legitimate: cadmium and lead accumulate in the body over years, particularly in the kidneys and bones, and inhalation is an efficient delivery route. The practical response is not panic but paperwork — buy from regulated sources that publish a full certificate of analysis, and check that the heavy-metal panel is actually on it rather than assuming it is.
Often not, and this is where the study lands an uncomfortable point. Most cannabis testing regimes in North America were built around the “big four” — lead, cadmium, arsenic and mercury — and requirements vary considerably between jurisdictions. Nickel and cobalt frequently sit outside that mandatory panel, even though pharmaceutical frameworks such as the European Pharmacopoeia and ICH elemental-impurity guidance cast a wider net. In this experiment nickel exceeded the WHO limit in every cultivar. Rules differ by country and by state, so check what your local regulator actually requires before assuming a passing test means a clean product.
Not automatically, and possibly the reverse. The CBD-rich cultivar in this study was the heaviest metal accumulator of the four while showing zero disruption to its cannabinoid profile. Hemp’s well-documented appetite for soil metals is the same trait that makes it attractive for phytoremediation. Cannabinoid content and contamination risk are independent variables.
Rinsing removes surface dust, not metals bound inside plant tissue, and curing does nothing to elemental contaminants. Extraction is more complicated: depending on the solvent and method, metals may partition into or away from the concentrate, and some processing equipment can add contamination rather than remove it. The reliable control point is upstream — clean substrate, tested irrigation water and verified fertiliser inputs.
Disclaimer
This article is intended for general information and science journalism purposes only. It summarises a single peer-reviewed agronomic study conducted under controlled experimental conditions and does not constitute medical, legal or agricultural advice. Nothing here should be taken as a health claim about cannabis, cannabinoids or any cannabis product, nor as guidance on cultivation in jurisdictions where it is restricted. Regulations governing cannabis cultivation, testing and sale differ substantially between countries and regions. Consult a qualified healthcare professional regarding any health concern, and the relevant national authorities regarding legal status and compliance requirements.
