Every September, in nineteen states from Alabama to Vermont and on the Menominee Indian Tribe Reservation of Wisconsin, ginseng season opens. A root pulled from an Appalachian hillside cannot legally leave the country without a state or tribal certificate, a federal export permit, and an inspector from the US Department of Agriculture counting scars on the root neck to confirm the plant was at least five years old.
That is an extraordinary amount of machinery for a forest herb. It exists because demand for one species, sustained over roughly three centuries, changed what was left in the woods.
American ginseng is not alone, but the list of medicinal plants under that kind of oversight is short. It is built species by species, through a treaty whose members meet every three years, and it moves at the speed of negotiation rather than the speed of a market.
A companion article in this library covers sourcing standards and certification. The question here is what collection has done to particular species.
How CITES sorts species into three lists
The Convention on International Trade in Endangered Species, usually called CITES, is the legal backbone of all of this. It does not ban harvest. It controls international trade, through three appendices that mean genuinely different things.
Appendix I, in the Convention’s own description, includes species threatened with extinction, and trade in them is permitted only in exceptional circumstances. Commercial trade is effectively closed, and both an import and an export permit are required.
Appendix II is where most medicinal plants sit, and its wording is the most commonly misread. CITES describes Appendix II as covering “species not necessarily threatened with extinction,” but in which trade must be controlled to avoid use incompatible with their survival. An export permit may be issued only if the specimen was legally obtained and the export will not harm the species’ survival. That second condition is called a non-detriment finding, and it is the heart of the system.
Appendix III is different in kind. A country that already regulates a species at home asks other Parties to help police the trade, and it can add the species on its own. Appendices I and II change only by a vote at the Conference of the Parties.
So a plant in Appendix II is not necessarily endangered; it is a plant whose trade somebody decided to watch. And a plant missing from all three is not necessarily secure; it may simply be that no Party has proposed it.
Dates matter more here than in most regulatory subjects. The plant appendices used for this article are valid from 7 February 2025, which puts them before the twentieth Conference of the Parties, held in Samarkand, Uzbekistan, from 24 November to 5 December 2025. Check the validity date on any appendix list.
American ginseng, Appendix II since 1975
The US Fish and Wildlife Service states that American ginseng, Panax quinquefolius, has been in Appendix II since 1975. Anyone shipping roots internationally needs an FWS export permit, issued only after the agency makes legal-acquisition and non-detriment determinations.
The federal regulation at 50 CFR 23.68 draws the edges. Whole plants and roots are covered, wild or artificially propagated. Manufactured parts, products and derivatives are excluded, which is why a ginseng extract on a shelf is not itself a CITES-regulated item although the root behind it was.
One provision deserves more attention than it gets. Cultivated ginseng that does not meet the regulation’s definition of artificially propagated is treated as wild for export. A grower who sows ginseng seed in a forest and waits a decade has a cultivated crop in ordinary language and a wild root in regulatory language.
The federal permit depends on state and tribal work. FWS cannot make an export finding without harvest management programs where the digging happens, and approving one requires licensing of buyers and sellers, dealer record-keeping, root inspection and, for wild ginseng, population assessments and county-level distribution data. Approved programs report every year by 31 May.
FWS reports that harvest season starts in September in all nineteen participating states, that exported wild and wild-simulated roots must come from plants five years of age or older, and that USDA inspectors count stem scars to check age. Some states require ten years. Harvest is restricted or prohibited on National Park Service land, national wildlife refuges and most national forests.
Notice what drives all of it. The oversight exists because an export market does. A plant with the same biology and no export demand gets none of this attention.
Why harvest totals are not population data
A 2010 paper in a USDA Forest Service research proceedings compiled reported ginseng harvest across eighteen states from 2000 through 2007: 480,868 dry pounds in total, from a high of 76,644 pounds in 2003 to a low of 42,085 in 2005. Five states accounted for nearly 70 percent, Kentucky alone around a quarter, and the annual harvest was worth roughly $25 million in producing counties.
Those figures fluctuate without a sustained trend, and the authors claim none. Reported harvest records what diggers sold to licensed dealers, so it tracks price, weather, enforcement and how many people went looking. A drop could mean fewer plants or simply fewer diggers.
The harder evidence comes from population biology. A 2004 study in the American Journal of Botany examined 16 allozyme loci across 21 wild populations and found greater genetic diversity within protected populations than within those open to harvest, and far greater genetic differentiation among the unprotected ones, which had also shifted toward smaller, non-reproductive plants.
A 2014 paper in Ecological Applications modeled harvest and warming together. Drawing demographic matrices from 12 populations representing 75 population-years, the authors projected a 1 degree Celsius rise in maximum growing-season temperature over 70 years. At the median population size of 140 plants, extinction risk was 8 percent from harvest alone and 6 percent from warming alone, but 65 percent when both applied. The effects were not additive, and the authors concluded that harvest rules need revising for a changing climate.
At the 2025 Conference of the Parties, the United States did not propose moving American ginseng to Appendix I. It proposed narrowing the existing listing to exempt certain finished products made from thin-sliced roots of artificially propagated plants. Canada, Indonesia and the European Union opposed it on laundering grounds, and the proposal was rejected. American ginseng remains in Appendix II.
Which other medicinal plants are listed, and where
In the plant lists valid from 7 February 2025, two things matter for each entry: which appendix it sits in, and the annotation, which decides what parts of the plant are covered.
Goldenseal (Hydrastis canadensis) is in Appendix II, annotated to cover underground parts, whole, in pieces and powdered. The United States proposed the listing in 1997, and that proposal is worth reading for its numbers: almost 150 tons of root collected in 1905, and 1.3 million pounds sold by herbal products association members across 1990 to 1992.
There is an uncomfortable symmetry there. The National Center for Complementary and Integrative Health, updating its goldenseal page in February 2025, states that there is not enough evidence to determine whether goldenseal helps any health condition, and that no rigorous human studies have evaluated it. A plant was pushed into trade control by demand for benefits the federal research center says have not been shown.
Agarwood trees, the genera Aquilaria and Gyrinops, are in Appendix II. Hoodia is in Appendix II, annotated to exempt material carrying a specified controlled-harvest label.
Orchids are the broadest entry: the entire family Orchidaceae is in Appendix II, with a subset in Appendix I, including Dendrobium cruentum. Cacti are nearly as sweeping, the whole family in Appendix II apart from species placed in Appendix I and three genera left off.
Rhodiola entered Appendix II in 2022, proposed jointly by China, the European Union, Ukraine, the United Kingdom and the United States, annotated to cover all parts and derivatives except seeds, pollen and retail-packaged finished products. The proposal states that uncontrolled harvest has reportedly driven declines in wild Rhodiola rosea and Rhodiola crenulata, and that most traded material is wild-sourced. It also contradicts itself on trade volume: the same estimate of concentrated extract in US import shipments appears as 2,464 kilograms in one section and about 62,464 kilograms in another.
Saussurea costus, sold as costus or kuth, is in Appendix I rather than II, which effectively closes commercial trade.
Sandalwood is the trap. True sandalwood, the genus Santalum, does not appear anywhere in the plant appendices valid from 7 February 2025. Two sandalwood-adjacent species do: Osyris lanceolata, East African sandalwood, and Pterocarpus santalinus, red sanders. When a seller calls sandalwood a CITES-listed plant, ask which plant is meant, because the answer changes the facts.
Listings keep moving, and following them is harder than it should be. Guggul, Commiphora wightii, shows how. The Earth Negotiations Bulletin reported that the proposal to list it failed in committee on 26 November 2025, 69 in favor to 42 against. The American Herbal Products Association, reporting on 10 December 2025, says debate reopened in the final plenary and the listing was adopted, effective 90 days from 5 December 2025.
The number everyone cites, and why this article will not give it
You will meet a statistic somewhere between 60 and 90 percent for the share of medicinal plants in trade that are wild-collected. It turns up in press materials, trade coverage and other explainers.
Trace it back to its source and it turns out to be backwards.
The source is a case study by Uwe Schippmann, Alan Cunningham and Danna Leaman, published by the UN Food and Agriculture Organization, on how cultivation and gathering of medicinal plants affect biodiversity. Describing a survey run for the Rainforest Alliance, the authors write: “On average, companies reported that 60–90 percent of material was cultivated, with the remainder harvested wild.”
That is a cultivation statistic. The authors go on to say that when companies were asked about numbers of species rather than volume of material, the figures generally invert. The two measures answer different questions, and the famous number is the wrong one of the pair.
The supportable version is this: most of the tonnage moving through the medicinal plant trade is cultivated, while most of the individual species in that trade are not. Even the species side is thinner than it looks. The same text cites a 1997 study finding that of 1,543 medicinal and aromatic species traded on the German market, only 3 to 6 percent came exclusively from cultivation. That is a species count for one national market in the 1990s. The FAO page itself carries no publication year, and its newest references are from 2002.
Treat the widely repeated figure of 15,000 threatened medicinal plant species with the same caution. Related estimates in that text are openly described as extrapolations.
There is no reliable global percentage for the wild-collected share of this trade, and inventing confidence would be worse than admitting the gap. Species-level statements are different: the Rhodiola claim above comes from a document five governments scrutinized, and it covers one genus.
United Plant Savers keeps a list, and it carries no legal force
United Plant Savers is the organization behind the Species At-Risk list that circulates widely among herbalists, and knowing what it is prevents a common mistake. It is a 501(c)(3) nonprofit founded in 1994 by the herbalist Rosemary Gladstar. Its list, published 16 December 2024 and last modified 26 August 2026, names 49 species across three tiers: 8 Critical, 34 At-Risk and 7 In Review. The three tiers replaced an older two-tier scheme in 2022.
At-Risk includes American ginseng, goldenseal, black cohosh, white sage, ramps and bloodroot. Critical is reserved for species the organization asks people not to wild-harvest or buy at all, and includes the Venus flytrap, peyote, lady’s slipper orchids and Hawaiian sandalwood.
Here is the part that matters. This is a private conservation nonprofit’s assessment: not regulation, not a government program, not third-party certification. Nothing happens legally if a company sells an At-Risk plant. The organization frames the list as a recommendation, urges cultivated forms where possible, and says it is not asking for a moratorium.
That is not a criticism. A private list can respond in months to what a treaty takes years to reach, and it covers North American species no CITES Party has proposed. It is simply a different kind of object, and should not be cited as though it carried a treaty’s force.
What the IUCN Red List adds, and why its dates matter
The International Union for Conservation of Nature’s Red List is the third system in play, and the one most often misquoted. Its categories run from Least Concern through Near Threatened to the three that count as threatened: Vulnerable, Endangered and Critically Endangered. Data Deficient means a species was examined and there was not enough information to classify it, which is not reassurance.
The Red List, established in 1964, reports more than 175,900 species assessed, of which more than 49,500 are classed as threatened with extinction, against a target of at least 260,000. The version current as this was written is 2026-1.
Two cautions. An assessment is a snapshot with a date on it, and plant assessments can be decades old. A 2025 review in Frontiers in Pharmacology compiled conservation statuses for seven medicinal case-study species, and the Red List status its table reports for the frankincense tree is dated 1998. Look for the assessment year every time.
The systems also need not agree, because they measure different things. A widely distributed plant can be globally secure and still CITES-listed, because its trade concentrates on a few wild populations.
Forest farming, and an honest answer about cultivated material
The cultivation alternative for woodland medicinals has a name. The USDA Forest Service’s National Agroforestry Center defines forest farming as the cultivation of high-value crops under the protection of a managed tree canopy, and is explicit that it is not the same as gathering wild plants, because deliberate management is part of the definition. Ginseng and goldenseal are the two medicinal species the agency names. It carries that regulatory wrinkle from earlier: forest-farmed ginseng that fails the artificially propagated definition is handled as wild for export.
Does cultivated material contain the same compounds? The evidence is thinner than confident claims on either side suggest.
The most direct recent test is a 2026 study in Life on Cibotium barometz, a CITES-listed medicinal tree fern still supplied largely from wild-dug rhizomes. Researchers profiled wild-collected, field-domesticated and tissue-cultured material. Of 2,828 annotated chemical features, 98.6 percent appeared in all three, and fewer than ten were unique to one group, so the differences were a matter of quantity rather than presence or absence. Across nine compound classes drawn from prior literature, tissue-cultured material retained 93 percent of the wild response and field-domesticated material 63 percent. The authors’ own caution is the line to keep: chemical similarity alone does not establish therapeutic equivalence.
That is one species, three replicates per group, one laboratory. Studies on other plants point in various directions, and the question is confounded by genetics, growing conditions, plant age, harvest timing and post-harvest handling. No study located for this article supports the common claim that wild material works better in people. Nobody has run that trial.
Knowing where a plant can live is part of the answer too. A 2025 paper in Ecology and Evolution modeled goldenseal habitat in Pennsylvania from 51 occurrence points and 28 field plots, and found bedrock type and average winter temperature were the best predictors of suitable ground.
What to do with this
If you buy wild-harvested botanicals, four questions do most of the work. Each one follows from what the systems above check, rather than from a published checklist.
Is the species named in Latin? Common names are shared across unrelated plants, which is exactly how the sandalwood confusion happens.
Is the plant on a CITES appendix, and does the annotation cover the part you are buying? Roots are often covered when extracts and powders are not. A legal extract is not evidence that the root behind it was legally dug.
If a seller cites a conservation list, which list is it? A CITES appendix, an IUCN assessment and a nonprofit’s at-risk list carry three different kinds of authority.
Is there a cultivated or forest-farmed version? For ginseng and goldenseal there is, and buying it is the most direct way a shopper reduces pressure on wild stands.
One health point belongs in any article about these plants. The National Center for Complementary and Integrative Health advises talking with a health care provider before using goldenseal or other herbal products, especially if you take any medicine, and reports that goldenseal cut blood levels of the diabetes drug metformin by about 25 percent in an NCCIH-funded study of healthy adults. NCCIH also states that people who are pregnant or breastfeeding should not use goldenseal and that it should not be given to infants. Raise any botanical with your pharmacist before starting it, and sooner if a prescription stops working the way it did. A long history of use is history, not evidence of effect, and nothing here is medical advice.
If you dig rather than buy: harvest rules are state law, permits are often required on public land, and plant collection is not allowed in National Parks. Checking before the shovel goes in is the cheapest step in the system.
Keep exploring: botanical sourcing, Endangered Species Act, traditional knowledge and culinary herbs health.
Sources
- CITES Secretariat. “How CITES works.” cites.org
- CITES (flora appendices, mirrored by the North American Rock Garden Society). “CITES Appendices I, II and III – plants, valid from 7 February 2025.” nargs.org
- US Fish and Wildlife Service. “American Ginseng.” fws.gov
- US Government Publishing Office / eCFR. “50 CFR 23.68 – How can I trade internationally in roots of American ginseng?.” ecfr.gov
- CITES. “CoP10 Prop. 73 – Inclusion of Hydrastis canadensis in Appendix II (proponent: United States, 1997).” cites.org
- CITES. “CoP19 Prop. 45 – Inclusion of Rhodiola spp. in Appendix II.” cites.org
- International Institute for Sustainable Development. “Earth Negotiations Bulletin, CITES CoP20 summary for 26 November 2025 (Vol. 21 No. 117).” enb.iisd.org
- American Herbal Products Association. “New botanical species added to CITES Appendix II during 20th Conference of the Parties meeting (10 December 2025).” ahpa.org
- USDA Forest Service, National Agroforestry Center. “Forest Farming.” fs.usda.gov
- USDA Forest Service Southern Research Station. “Prisley SP, Chamberlain J, McGuffin M. "Relationships between harvest of American ginseng and hardwood timber production." 2010 Joint Meeting of the FIA Symposium and the Southern Mensurationists, pp. 95-102.” srs.fs.usda.gov
- United Plant Savers. “Species At-Risk List (published 16 December 2024, modified 26 August 2026).” unitedplantsavers.org
- International Union for Conservation of Nature. “The IUCN Red List of Threatened Species: Background and History.” iucnredlist.org
- IUCN SSC Cetacean Specialist Group. “IUCN Red List categories and what counts as threatened.” iucn-csg.org
- UN Food and Agriculture Organization. “Schippmann U, Cunningham AB, Leaman DJ. "Case Study No. 7: Impact of cultivation and gathering of medicinal plants on biodiversity: global trends and issues." Biodiversity and the Ecosystem Approach in Agriculture, Forestry and Fisheries.” fao.org
- National Center for Complementary and Integrative Health. “Goldenseal (last updated February 2025).” nccih.nih.gov
- PubMed. “Souther S, McGraw JB. "Synergistic effects of climate change and harvest on extinction risk of American ginseng." Ecol Appl, 2014. DOI: 10.1890/13-0653.1.” doi.org
- PubMed. “Cruse-Sanders JM, Hamrick JL. "Genetic diversity in harvested and protected populations of wild American ginseng, Panax quinquefolius L. (Araliaceae)." Am J Bot, 2004. DOI: 10.3732/ajb.91.4.540.” doi.org
- PubMed. “Su Y, Qin G, Zhou C, Guan D, Song J. "Untargeted LC-MS/MS Metabolite Profiling of Wild, Field-Domesticated and Tissue-Cultured Rhizomes of the Endangered Medicinal Fern Cibotium barometz." Life (Basel), 2026. DOI: 10.3390/life16091435.” doi.org
- PubMed. “Houston E, Burkhart EP, Zuiderveen G, Chen X. "Identification of Goldenseal (Hydrastis canadensis L.) Habitat and Indicators in Pennsylvania, USA." Ecol Evol, 2025. DOI: 10.1002/ece3.71050.” doi.org
- PubMed. “Mykhailenko O, Jalil B, McGaw LJ, Echeverria J, Takubessi M, Heinrich M. "Climate change and the sustainable use of medicinal plants: a call for 'new' research strategies." Front Pharmacol, 2025. DOI: 10.3389/fphar.2024.1496792.” doi.org