2026-08-24 · Prop 65 · 5 min read
Why mercury has no Prop 65 safe harbor level
In short: mercury has been on California's Proposition 65 list since 1990 — methylmercury since 1987 — but OEHHA has never adopted a safe harbor level for it in any form: no MADL for the developmental listings, no NSRL for the cancer listing. That gap does not weaken the warning requirement. It removes the bright line that makes compliance checkable, shifts the burden of proof onto businesses, and, in practice, means mercury gets less enforcement attention than lead — not because it matters less, but because the missing number makes every case harder to bring.
Our safe harbor explainer ended on a wrinkle: hundreds of listed chemicals have no published safe harbor at all. For the product categories we test — protein powders, supplements, foods — the most consequential chemical in that gap is mercury. This post is about what the absence of a mercury Prop 65 safe harbor level actually means, for the businesses that sell these products and for the people who buy them.
Listed three ways, quantified zero ways
Mercury appears on the Proposition 65 list three times. Methyl mercury was listed as a developmental toxicant on July 1, 1987. Mercury and mercury compounds followed as developmental toxicants on July 1, 1990. Methylmercury compounds were listed as carcinogens on May 1, 1996.
For lead, those kinds of listings come with numbers attached: OEHHA adopted a reproductive MADL of 0.5 micrograms per day, and that single figure is what makes our findings checkable — concentration times serving size, compared against 0.5. For mercury, OEHHA's safe harbor table has no entry to compare against. Three listings, spanning nearly four decades on the books, and no adopted MADL or NSRL for any of them.
No safe harbor does not mean no rule
It would be easy to read the gap as a free pass. It is closer to the opposite. The warning requirement in Health and Safety Code section 25249.6 applies to every listed chemical, safe harbor or not: a business that knowingly exposes people to mercury above the statute's risk thresholds must warn.
What the safe harbor provides is a shortcut — a pre-computed number a business can stay under and know it is exempt. Without one, the business has to do the showing itself. For a reproductive toxicant, the statute's exemption requires demonstrating that the exposure, multiplied by one thousand, would produce no observable effect. The burden of that demonstration sits with the business, not the state. A company selling a mercury-containing product in California without a warning is, in effect, holding an exposure assessment it may one day have to defend — or it is holding nothing.
The numbers people borrow instead
In the absence of a Prop 65 number, everyone reaches for benchmarks from other regulators. None of them is a safe harbor, and each measures something different:
- EPA's reference dose for methylmercury is 0.1 micrograms per kilogram of body weight per day. For a 60-kilogram adult, the arithmetic is 0.1 × 60 = 6 micrograms per day — a chronic-exposure risk estimate, not a legal threshold.
- FDA's action level for methylmercury in commercial fish is 1 part per million — a market-removal trigger for one food category.
- USP <2232>, the pharmacopeial standard for elemental contaminants in dietary supplements, sets 15 micrograms per day for total mercury and 2 micrograms per day for methylmercury — a voluntary quality standard unless a product claims USP conformance.
A product can sit below all three of these and still have an open Prop 65 question, or sit above one of them and break no law. That ambiguity is precisely the cost of the missing MADL: with lead, one multiplication settles the compliance question; with mercury, reasonable experts can be paid to disagree.
What the gap means for buyers
The absence of a warning tells you less for mercury than it does for lead. A lead warning decision can be audited by anyone with the lab result and the label. A mercury decision rests on an exposure assessment you will never see.
So the certificate of analysis has to carry more weight. If you buy in the categories where mercury is a live question — fish-derived ingredients, and some imported herbal products with a documented history of metal contamination — look for total mercury on the CoA, note the detection limit, and apply the same skepticism we outlined in how to read a certificate of analysis. To be clear about proportion: a detectable mercury result is a data point for comparison against the benchmarks above, not a danger verdict, and most products in most categories test unremarkably.
What it means for enforcement
Here is the uncomfortable part, and the reason this post exists. Safe harbors are what make Prop 65 enforcement checkable — a number, an arithmetic, a showing anyone can verify. Take the number away and an enforcement case needs a defensible exposure assessment built from scratch: more expert work, more room for dispute, more cost per case. The predictable result is that private enforcement flows toward chemicals with bright lines, lead above all, and away from mercury.
A gap in a lookup table becomes a gap in policing. That is not a criticism of any agency — adopting a MADL is a formal regulatory process — but it is a fact about incentives that buyers deserve to have stated plainly.
The takeaway
CSRC tests for mercury anyway. Where we report results, we will publish the concentration, the serving math, and the comparison against each borrowed benchmark, labeled as exactly what it is — because until OEHHA adopts a safe harbor level for mercury, published arithmetic is what fills the space where the checkable number should be.
This article is educational information about California law and regulatory toxicology, not legal or medical advice. Regulatory descriptions are summaries; OEHHA's regulations control. © 2026 Consumer Safety Research Center Inc.