Turn over a bottle of an herbal supplement and read the small print. Somewhere on the back you may find a phrase like sustainably sourced, ethically wild-harvested, or responsibly collected.
Now ask a plain question. Who checked?
In the United States, the honest answer is usually nobody with legal authority. Federal rules have a great deal to say about what ends up inside the bottle. They say almost nothing about how the plant came out of the ground.
That gap is the subject here. Private standards, trade-association guidance and international recommendations have grown up to fill it, and they promise very different things from one another. Underneath all of them sits a harder problem, and it is the one this article keeps coming back to: you cannot source a plant responsibly if you cannot establish which plant it is.
Four buckets, and most sourcing claims land in the last one
Any claim about how an ingredient was collected belongs in one of four categories. Learning the four takes a minute and it does most of the work.
Regulation. A legal requirement, written by a government, enforced by an agency. Breaking it has consequences.
A government certification program. An agency writes the criteria and authorizes the people who inspect against them. The rules are public and the inspectors answer to the agency.
Private third-party certification. A non-government body publishes a standard, and independent auditors check compliance. Nobody in government writes the rules, but the rules exist in a document you can read, and somebody other than the seller does the checking.
Undefined. A word on a label with no published criteria behind it and no auditor.
“Sustainably sourced” on a supplement is the fourth bucket. No US agency defines the phrase for botanicals, and no rule requires a company to prove it before printing it.
Three neighboring questions belong to other articles in this library and are not re-argued here. Organic certification is a federal regulatory program with its own rules. “Regenerative” is a separate and much younger claim. Pesticide residue limits are a different legal system altogether. Seals and certification marks as objects, including who audits them, also get their own treatment. This article is about the narrower question of harvest and collection.
The two federal hooks, and how little they reach
US law touches botanical sourcing in two places, and both are narrower than people expect.
The first is CITES, the international convention that controls trade in listed species. It applies to a specific list of plants, species by species, and a companion article in this library covers how that list works and which medicinal plants are on it. For most botanicals in the American supplement aisle, CITES simply does not apply.
The second is the Lacey Act, whose plant provisions were added in 2008. Under 16 U.S.C. 3372, it is unlawful to import, sell or transport plants taken in violation of a state or foreign law. But the statute names which kinds of law count: laws protecting plants or addressing plant theft, rules on taking plants from parks, forest reserves and other protected areas or without required authorization, requirements to pay royalties, taxes or stumpage fees, and limits on export or transshipment.
Read that list again and notice what is missing. There is no sustainability test. If a collector strips a hillside bare with the landowner’s permission, in a country with no law against it, the Lacey Act has nothing to say.
The Lacey Act’s other half is the import declaration. Under the US Department of Agriculture’s rule at 7 CFR 357.3, anyone importing a plant must file a declaration listing four things: the scientific name including genus and species, the value, the quantity with its unit of measure, and the country from which the plant was taken. If the species used is unknown and varies, the importer lists every species that may have been used. If the source country is unknown, the importer lists every country it may have come from.
That is a disclosure, not an approval. No agency grades it. And an importer who genuinely does not know can satisfy the rule by listing possibilities.
The exceptions narrow it further. Under 7 CFR 357.4, no declaration is needed when the plant is used only as packaging, or when plant material makes up no more than 5 percent of an individual product unit’s weight and the total plant material in an entry under the same ten-digit tariff code stays under 2.9 kilograms. That exception does not apply to material listed under CITES, listed as endangered or threatened under the Endangered Species Act, or protected by a state law covering threatened species. For a great many finished consumer goods, though, the 5 percent rule means no declaration is filed at all.
Here is the sharp contrast. Federal law does impose a hard requirement about identity. The dietary supplement manufacturing rules at 21 CFR 111.75 require a manufacturer, before using a dietary ingredient, to run at least one appropriate test or examination “to verify the identity of any component that is a dietary ingredient.” Identity is mandatory. Provenance is not.
You cannot source a plant responsibly if you cannot prove what it is
This is the part most sustainability writing skips, and it undercuts everything above it.
A sourcing commitment is a promise about a particular species, taken from a particular place, under particular conditions. If the powder in the drum is not the species on the paperwork, every downstream claim collapses, including the sustainability claim. A company can audit a collection site conscientiously and still ship adulterated material, because adulteration happens after the audit.
The American Botanical Council, the American Herbal Pharmacopoeia and the National Center for Natural Products Research run a joint effort called the Botanical Adulterants Prevention Program, which publishes bulletins describing specific documented substitutions. By January 2019, when the program released its bulletin on ashwagandha root, the American Botanical Council said the program had published 48 peer-reviewed articles and that more than 200 US and international parties had funded or endorsed it. The ashwagandha case is instructive: the documented problem was the addition of undeclared leaf and stem material to products labeled as root, which raises measured withanolide content and makes the material look stronger than it is.
Peer-reviewed work reaches the same conclusion with less diplomacy. A 2019 review in Natural Product Reports by investigators at the NIH-funded Natural Product-Drug Interaction Research Center reported that many studies of botanical products are carried out with poorly characterized material, leaving results that are irreproducible and hard to interpret.
Verification is also harder than it sounds. Scientists at the Food and Drug Administration’s Center for Food Safety and Applied Nutrition collected 112 market samples of five commonly sold botanicals, including ginkgo, valerian and St. John’s wort, and compared chemical marker analysis against DNA barcoding on the same samples. Their point was that the methods answer different questions and each has a failure mode. Microscopy fails once material is processed past recognition. Chemical markers fail when the marker compound is not unique to the species, or when no purified reference standard exists. And the authors raised concerns about whether DNA barcoding is appropriate for finished botanical products at all.
The stakes are not theoretical. A 2015 paper in Scientific Reports reported that roughly 256 traditional patent medicines containing material from the Aristolochiaceae family were still being sold in Chinese markets, and described a combined DNA and chemical system for distinguishing those plants from the look-alike species they get substituted for. Aristolochic acids, which occur in that plant family, cause a documented kidney disease. Misidentification is a safety problem before it is ever a conservation problem.
When to raise this with a clinician
Two specific situations are worth a phone call rather than a web search.
If you take a prescription medicine and start an herbal product, tell your pharmacist and your prescriber the botanical name, not just the marketing name. The National Center for Complementary and Integrative Health reports on its goldenseal page, last updated February 2025, that in an NCCIH-funded study, blood levels of the diabetes drug metformin dropped about 25 percent in healthy adults who took goldenseal extract. NCCIH also notes on that page that some commercial goldenseal products have been found to contain herbs or ingredients not listed on the label.
The second is simpler. If you start a botanical product and develop something new and unexplained, such as a rash that spreads, yellowing of the eyes or skin, dark urine, or unusual and persistent fatigue, stop and get seen. Bring the bottle. An identity problem in the supply chain shows up in a person long before it shows up in a database.
Nothing here is medical advice, and no herbal product should replace treatment your clinician has prescribed.
FairWild, the one widely used standard with a published rulebook
FairWild is the certification scheme most often named when wild-collected botanicals are involved, and it is worth understanding precisely because it is the exception rather than the norm.
Version 3.0 of the FairWild Standard was approved on 19 December 2023, replacing a 2010 version. It contains seven principles and 24 criteria, organized under three themes. Summarized plainly, the ecological theme asks that wild harvesting support conservation of the species being collected and leave the collection area and surrounding habitat no worse off. The social theme covers the human rights of collectors and workers, fairness in the arrangements between operators and collectors, and respect for communities’ customary rights and share of the benefits. The business theme requires that harvest and trade be legal and that the company run itself responsibly. The criteria are unevenly distributed: three under each of the two ecological principles, five on human rights, four on fair arrangements, two on community benefits, two on legal compliance and five on business practice.
The scope is narrower than many readers assume. The standard covers wild plants and plant products, wild fungi and lichens, and wild seaweed and other algae. Animals and animal products are excluded by name, and that exclusion specifically includes honey and beeswax, no matter how the finished product is marketed.
Who runs it matters for the bucket test. The FairWild Foundation lists a registered office at Stampfenbachstrasse 38 in Zürich, Switzerland, and a secretariat hosted by the conservation organization TRAFFIC at the David Attenborough Building in Cambridge, in the United Kingdom. It is a private foundation. No government wrote the standard and no government enforces it. That puts FairWild squarely in the third bucket: private third-party certification, with a published rulebook you can download and read.
Certification runs through accredited control bodies, and FairWild lists nine of them, based in Germany, Switzerland, China, Sri Lanka, Turkey, Ukraine and India. Certification requires an annual audit, and it is species-specific, so a certificate covers named species from named sites rather than a company in general.
Two details are worth carrying to the store. First, FairWild separates certification from registration. Wild collection operations are certified. Processors, traders and the brands that put finished products on shelves are registered. A registered brand is not an audited collection site, and the words are not interchangeable.
Second, even the people inside this system struggle to keep their own pages current. FairWild’s own standard page still describes version 2.0 as current, more than two years after version 3.0 replaced it. A page from one of its accredited certifiers, timestamped August 2023, points readers to Principle 11 of a standard that now has seven principles. If the certifiers and the standard-setter cannot keep their web pages synchronized, a consumer reading a marketing line has very little chance.
UEBT verifies a company, not a product
The Union for Ethical BioTrade is the other name that recurs. Its program description, published on the UN Conference on Trade and Development website, says UEBT is a member-based non-profit association created in 2007 as a spin-off from the United Nations, promoting what it calls sourcing with respect. Members are audited by independent third parties when they join and every three years after, must set long-term sourcing targets, must put a biodiversity management system in place, and must submit annual public progress reports.
That is a real system, and it sits in the third bucket. But the most useful fact about it for a shopper is what it does not do. The same page states plainly that the UEBT logo is not for product-specific marketing and that UEBT has no on-product label. The member mark belongs on a company’s website, catalogue and business cards. It is a statement about the organization’s sourcing practices, assessed as a whole, not a verification that the jar in your hand was checked.
Guidance that reads like rules but is not
Two documents get cited as though they were regulation. Neither is.
The World Health Organization published WHO guidelines on good agricultural and collection practices (GACP) for medicinal plants on 12 December 2003. It runs 80 pages, and WHO’s own catalogue classifies it as a guideline. A document record for it lists three objectives: contributing to quality assurance for medicinal plant materials, helping countries write their own national GACP guidelines, and encouraging sustainable cultivation and collection. That third objective is genuinely about sustainability. But WHO has no authority over an American supplement company, and the document is designed to be turned into national rules by countries that choose to do so. The United States has not done so for botanicals.
The American Herbal Products Association, the trade association for the US herbal industry, publishes a free Good Agricultural and Collection Practices and Good Manufacturing Practices guidance document, published in May 2021 and revised in May 2023, including a dedicated wild collection section and an assessment tool for it. AHPA does not require membership to download it, which is to its credit. And AHPA is candid about status. Of the recommendations in its dietary ingredient processor section, the association says “they are not legal or regulatory requirements” except where noted.
That sentence is the honest summary of this entire layer. Both documents are fourth-bucket material in the sense that matters: nobody audits a company against them and nothing happens if a company ignores them.
Three questions that sort any sourcing claim
When a label or a brand page makes a claim about collection, these three questions settle it faster than reading the marketing copy. They are built from what the schemes above verify, not from anyone’s published checklist.
Can I find the criteria document, with a version number and a date? FairWild’s standard is a downloadable PDF with an approval date on it. If the claim’s criteria are not published anywhere, the claim is undefined and you can stop there.
Who was audited, and what exactly were they audited for? This is where most claims fail quietly. A brand that is a registered licensee of a scheme has not had its collection sites inspected. A company holding a membership mark has had its systems reviewed, not its individual products. Ask whether the audit touched the harvest or only the paperwork.
Does the claim name a species and a place? Sourcing happens to a species in a location. A claim that names neither cannot be checked by anyone, including the company making it. A claim that names both can at least be asked about.
A fourth question is optional but revealing: does the seller publish anything about how it confirms identity? A company that will tell you which test it runs on incoming material is a company that has thought about the problem underneath the problem.
The most useful thing to take away
“Sustainably sourced” is not a regulated term in the United States, and you should read it as marketing until something concrete is attached to it. That is not cynicism. It is the accurate legal position, and knowing it stops you from paying extra for a word.
What is checkable is narrower and more useful. A published standard with a date. An audit of the collection operation rather than the brand. A species name. A documented identity test. Those four things exist or they do not, and a company that has them will usually say so in plain terms, because it cost money to get them.
And if the only thing you change is this: a botanical product that will not tell you its species is a product nobody can make a sourcing promise about, no matter how green the label looks.
Sources
- FairWild Foundation. “FairWild Standard Version 3.0 (approved 19 December 2023).” static1.squarespace.com
- FairWild Foundation. “Launch of the FairWild Standard Version 3.0.” fairwild.org
- FairWild Foundation. “Approved Control Bodies.” fairwild.org
- Union for Ethical BioTrade (profile hosted by UNCTAD). “Union for Ethical BioTrade.” unctad.org
- World Health Organization. “WHO guidelines on good agricultural and collection practices (GACP) for medicinal plants.” who.int
- Medbox. “Document record: WHO guidelines on good agricultural and collection practices (GACP) for medicinal plants.” medbox.org
- American Herbal Products Association. “Good Agricultural and Collection Practices (GACP-GMP guidance, published May 2021, revised May 2023).” ahpa.org
- US Government Publishing Office / eCFR. “7 CFR 357.3 – Declaration requirement (Lacey Act plant import declaration).” ecfr.gov
- US Government Publishing Office / eCFR. “7 CFR 357.4 – Exceptions from the declaration requirement.” ecfr.gov
- Cornell Law School Legal Information Institute. “16 U.S.C. 3372 – Prohibited acts (Lacey Act).” law.cornell.edu
- US Government Publishing Office / eCFR. “21 CFR 111.75 – What must you do to determine whether specifications are met?.” ecfr.gov
- American Botanical Council. “ABC-AHP-NCNPR Botanical Adulterants Prevention Program Publishes Ashwagandha Root and Root Extract Bulletin (15 January 2019).” globenewswire.com
- National Center for Complementary and Integrative Health. “Goldenseal (last updated February 2025).” nccih.nih.gov
- PubMed. “Kellogg JJ, Paine MF, McCune JS, Oberlies NH, Cech NB. "Selection and characterization of botanical natural products for research studies: a NaPDI center recommended approach." Nat Prod Rep, 2019. DOI: 10.1039/c8np00065d.” doi.org
- PubMed. “Pawar RS, Handy SM, Cheng R, Shyong N, Grundel E. "Assessment of the Authenticity of Herbal Dietary Supplements: Comparison of Chemical and DNA Barcoding Methods." Planta Med, 2017. DOI: 10.1055/s-0043-107881.” doi.org
- PubMed. “Wu L, Sun W, Wang B, et al. "An integrated system for identifying the hidden assassins in traditional medicines containing aristolochic acids." Sci Rep, 2015. DOI: 10.1038/srep11318.” doi.org
- Control Union. “FairWild certification programme page (certified collection operations versus registered processors and licensees).” controlunion.com
- FairWild Foundation. “About Us (registered office) and Contact (secretariat hosted by TRAFFIC).” fairwild.org