Claim register
Machine-readable records connect retained claims to sources and state the limits of each source.
Ten substances. Fifty-seven centuries of confirmed use. One legal contradiction. The most ancient human practice — sacred plant medicine — is a federal crime. This atlas maps the archaeology, the neuroscience, the case law, and the phenomenology of why that matters.
"To fathom hell or soar angelic / just take a pinch of psychedelic." — Humphry Osmond, in correspondence with Aldous Huxley (1957) — the letter in which Osmond coined the word "psychedelic"
This is a research and editorial reference. It is not medical advice, not legal advice, and not a current-law service — statutes, agency positions and litigation move faster than any archive. Verify the official authority and consult a qualified professional before acting on anything here, and check clinical figures against the cited paper and ClinicalTrials.gov.
Dosing, interaction clearance, individualised legal-risk scoring, live-law tracking and living-tradition protocols are deliberately not published here. Evidence & cultural governance sets out why, and how to request a correction. Built by RN Collins, M.S.
Entheogen Atlas is a research and editorial reference site built by RN Collins — neuroscientist, technologist, AI content creator, and law student at Northeastern University (FlexJD, class of 2029). Built as an editorial research interface for readers exploring entheogen science, legal history, and culture. Reliance-prone dose, interaction, legal-risk, RFRA-strength, current-law, and living-tradition protocol outputs have been removed from this edition. The atlas also supports Ask a Neuroscientist in Fat Nugs Magazine and writing for Cannabis Law Report. The citation library contains a mixture of peer-reviewed research, judicial opinions, government material, and foundational or contextual texts; citation presence does not itself establish claim-level verification.
| SUBSTANCE | US STATUS | RECEPTOR | DURATION | ONSET | TRADITION |
|---|---|---|---|---|---|
| Cannabis | Schedule I (fed); legal 24 states | CB1/CB2, 5-HT₁A | 2–6hr | Min–1hr | 10,000+ BCE |
| Ayahuasca | Schedule I (DMT); religious exemptions | 5-HT₂A (DMT+MAOI) | 4–8hr | 30–60min | 1,000+ BCE |
| LSD | Schedule I | 5-HT₂A, D₂ | 8–14hr | 30–90min | 1943 CE |
| Peyote | Schedule I; NAC exempt | 5-HT₂A (mescaline) | 8–14hr | 1–2hr | 5,700+ BCE |
| Psilocybin | Schedule I; FDA Breakthrough | 5-HT₂A (psilocin) | 4–8hr | 30–60min | 9,000+ BCE |
| Ibogaine | Schedule I; legal NZ, Gabon | NMDA, κ-opioid, SERT | 18–36hr | 1–2hr | Bwiti tradition |
| MDMA | Schedule I; FDA Breakthrough | SERT/DAT/NET | 3–6hr | 30–60min | 1912 synthesis |
| Ketamine | Schedule III; Spravato approved | NMDA antagonist | 1–3hr | Minutes | 1962 synthesis |
| DMT | Schedule I | 5-HT₂A, σ-1 | 5–30min | Seconds | Amazonian |
| Salvia | Unscheduled most states | κ-opioid (KOR) | 5–30min | Seconds | Mazatec |
The previous substance-specific dose ranges and contraindication summaries have been removed from the public interface. Static ranges can be mistaken for individualized instructions and incomplete interaction lists can omit material risks.
The scientists, chemists, psychiatrists, and advocates whose work built the field — and whose careers were shaped by prohibition.
fMRI findings, receptor binding affinity, cross-tolerance, and microdosing evidence — mechanistically, not metaphorically.
The framework that governs entheogenic substances — legal, medical, religious — was not built to hold them. This is the argument this atlas makes.
Peyote has been used ritually for 5,700 years: two buttons from Shumla Cave on the Rio Grande are radiocarbon dated to 3780–3660 cal BC and yielded mescaline on GC-MS, which makes them the oldest plant drug to have given up its active compound to chemical analysis. Cannabis was burned in a Judahite shrine at Tel Arad in the late eighth century BCE and smoked in braziers at Jirzankal in the Pamirs by about 500 BCE, both established by residue chemistry rather than inference. A ritual bundle from the Bolivian altiplano dated to around 1000 CE carries harmine and DMT together — ayahuasca’s two pharmacological halves in one kit. Older claims exist, and several are weaker than they are usually reported to be; the evidence ladder separates what is demonstrated from what is inferred. LSD, the youngest of the substances here, was synthesized in 1938 — and within fifteen years had been administered therapeutically to tens of thousands of patients.
Against this record, the Controlled Substances Act of 1970 placed all five substances in Schedule I — alongside heroin — as having "no accepted medical use" and "lack of accepted safety." Schedule I was designed as a temporary placeholder pending study. The study never came. The substances stayed. That gap between what the archaeological and clinical record shows and what the legal framework asserts is the legitimacy problem.
Psilocybin received FDA Breakthrough Therapy designation in 2018 and 2019 — meaning FDA acknowledges it may be substantially better than existing treatments for depression. The designation coexists with Schedule I. The law has not caught up with the agency's own evidence assessments.
The War on Drugs was not a neutral policy response to substance harms. John Ehrlichman, Nixon\'s domestic policy chief, told journalist Dan Baum in 1994: "We knew we couldn't make it illegal to be either against the war or Black, but by getting the public to associate the hippies with marijuana and Blacks with heroin, and then criminalizing both heavily, we could disrupt those communities." This was published in Harper's Magazine in 2016.
LSD was federally prohibited in 1966 — during the peak of therapeutic research. The research was producing results. The prohibition ended the research. The scheduling of cannabis in 1970 came despite the Shafer Commission — Nixon\'s own appointed body — recommending decriminalization. Nixon rejected the report without response.
This is not conspiracy. It is documented political history. The criminal architecture of entheogen law was built on race, politics, and counterculture suppression — not pharmacology, harm evidence, or religious freedom analysis.
The Native American Church had used peyote ceremonially for generations before the United States existed as a legal entity. In 1990, Employment Division v. Smith ruled that the Free Exercise Clause of the First Amendment does not require religious exemptions from neutral, generally applicable laws — meaning the State of Oregon could deny unemployment benefits to NAC members fired for peyote use.
Congress responded with RFRA in 1993. The Supreme Court then limited RFRA to federal law in 1997 (City of Boerne v. Flores). In 2006, Gonzales v. O Centro held unanimously that RFRA protects UDV ayahuasca use — but the decision applies only to that organization, with other churches required to litigate individually in each circuit.
The result: a Shipibo curandero practicing in the Amazon has cultural sovereignty. A UDV member in New Mexico has a Supreme Court ruling. A Santo Daime member in Oregon has a circuit court ruling. An independent practitioner in Texas has a criminal charge. The same substance, the same intent, the same phenomenology — different legal outcomes depending on zip code and organizational membership.
Carhart-Harris et al. (2021) published in the New England Journal of Medicine that psilocybin was non-inferior to escitalopram (the most prescribed SSRI globally) across all secondary outcome measures, and superior on several. The study was published while psilocybin remained Schedule I, while researchers required DEA Schedule I researcher licenses, and while the cost of that research was increased an estimated 5–10× by the regulatory burden Schedule I imposes.
Griffiths et al. (2006) found that 79% of participants reported sustained increases in well-being at 2-month follow-up — changes corroborated by friends and family. Not one participant reported decreased well-being. The Griffiths data has been replicated. The Carhart-Harris data has been replicated. The Pahnke data has been replicated at 14 months (Griffiths 2011) and at 25 years (Doblin 1991).
The clinical evidence for psychedelic therapeutic value is among the most consistent in the history of psychiatric research. The legal framework governing these substances was built in explicit defiance of the evidence that existed at the time it was built, and has been maintained in defiance of fifty years of accumulating evidence since.
Stanislav Grof, the Czech psychiatrist who administered LSD to over 4,000 patients in clinical settings between 1956 and 1967, documented what he called the perinatal and transpersonal dimensions of psychedelic experience: layers of the unconscious inaccessible to ordinary therapy, including what he termed COEX systems (systems of condensed experience) and experiences of cosmic unity indistinguishable from the mystical states described in every major contemplative tradition. Grof's clinical archive — the largest in history — was ended not by evidence of harm but by Schedule I in 1970.
Pahnke (1967) documented nine characteristics of mystical experience produced by psychedelics: unity, transcendence of time and space, deeply felt positive mood, sense of sacredness, noetic quality, paradoxicality, alleged ineffability, transiency, and persisting positive changes. These are indistinguishable, by self-report and by external rater assessment, from spontaneously occurring mystical experiences reported across all major religious traditions.
The law treats the chemical occasion of such experiences as a criminal act. It simultaneously grants constitutional protection — through RFRA — to the religious belief that such experiences are sacred. The contradiction is not theoretical: you may believe psilocybin produces communion with God; you may not consume it to have that communion, unless you belong to the right organization, in the right state, with the right DEA exemption.
The framework cannot hold this. The Legitimacy Gap is not a regulatory problem awaiting a technical fix. It is a category error at the foundation of how the United States has classified substances that produce what Plato called "our greatest blessings" — "provided madness is given us by way of divine gift."
The essay above is an argument. What follows is not: it is the chain of authorities the argument runs along, each one linked to the primary source — a Library of Congress scan of the United States Reports, a court’s own opinion PDF, the Statutes at Large, or the text of the regulation. Read in order it shows something the summary version misses. The exemption line was drawn administratively in 1971, contested in the courts for two decades, nearly erased by a peyote case in 1990, rebuilt by statute in 1993, cut in half in 1997, and has since been extended exactly twice, both times to ayahuasca churches, both times on their particular records.
Two things are worth holding onto while reading it. First, most claims that fail do not fail on the compelling-interest question that gets the attention; they fail earlier, on whether the belief system is a religion at all (Meyers) or whether prohibition substantially burdens it (Oklevueha). Second, the constitutional ground is not settled. Fulton (2021) decided for a religious claimant while explicitly declining to overrule Smith, and several Justices wrote separately to say Smith should go. If it does, this entire line is re-argued on First Amendment terms rather than statutory ones.
Almost every popular claim about how long humans have used psychoactive plants rests on a different kind of evidence from the one beside it. This page separates them, and dates the atlas from the strongest rung rather than the oldest story.
The figure comes from two dried peyote buttons collected from Shumla Cave No. 5 on the Rio Grande and held at the Witte Museum in San Antonio. El-Seedi and colleagues radiocarbon dated them to 3780–3660 cal BC and then ran thin-layer chromatography and GC-MS on the same material: the buttons yielded roughly 2% alkaloids, and mescaline was identified in both. Nothing else was.
That combination is what makes the date usable. A single object supplies the plant, the human context, the calendar date and the psychoactive compound, and each is established by an independent method on the same specimen. El-Seedi et al. describe it as the oldest plant drug yet to yield its major bioactive compound on chemical analysis, and the claim has not been displaced in the twenty years since.
Older claims exist. Some are serious. None of them close all four gaps at once, and the difference between them is not a matter of degree.
Samorini (2019) divides archaeological evidence for psychoactive plant use into direct evidence — material, chemical and genetic — and indirect evidence, which covers iconography, literature, paraphernalia and skeletal indicators. Guerra-Doce (2015) sets out what an individual claim must satisfy before it counts. The ladder below is adapted from both, and the entries beneath it are sorted onto its rungs rather than into a single undifferentiated list.
Several entries below are ranges rather than dates, and one is a claim rather than a finding. Plotted on a linear axis the last three thousand years would collapse into a smear and the contested Neanderthal claim would push everything else off the right-hand edge, so the scale here is logarithmic in years before present. Width is the dated range; colour is the class of evidence.
Fourteen entries. Every one carries the excavation report or the dating paper rather than a secondary summary, and every source below has been resolved before publication. Where the literature disagrees, both sides are cited.
The claim that Neanderthals buried their dead with medicinal flowers 60,000 years ago is the single most repeated statement in popular writing about ancient drug use, and it is the one this atlas has removed from its headline.
Its basis is narrow. Soil samples taken around the Shanidar IV burial in Iraqi Kurdistan contained clumped pollen, and Leroi-Gourhan (1975) identified taxa including Ephedra. Solecki (1975) read the clumping as deliberate deposition. Sommer (1999) pointed out that the Persian jird, a burrowing rodent that hoards flower heads and is present at the site, produces exactly that signature. Fiacconi and Hunt (2015) sampled modern pollen at Shanidar and found that insect and animal activity concentrates whole anthers and pollen clumps in the way the 1975 report treated as anthropogenic. Renewed excavation (Pomeroy et al. 2020) supports intentional burial at the site while leaving the floral reading unsupported.
Two things are worth keeping separate. The scholarship genuinely disagrees about whether Shanidar IV involved deliberate floral deposition; that argument is live. It does not disagree about entheogens, because no entheogen appears in the assemblage at all. Ephedra is a stimulant, not a psychedelic, and its presence in a pollen sample is not evidence that anyone consumed it. A contested burial rite became, by repetition, a 60,000-year date for sacred plant medicine. The atlas lists it as a claim and dates itself from Shumla Cave.
The last decade has moved several claims off the iconographic rungs and onto the chemical one, and it has done so in a consistent direction: the practices are real, and the dates are younger and better constrained than the popular literature assumed.
Cannabis is the clearest case. The ritual reading of the Yanghai tomb cache (Jiang et al. 2006) rested on the character of the burial. Thirteen years later Ren et al. recovered cannabinoids from burnt residue in ten wooden braziers at Jirzankal in the eastern Pamirs, dated to about 500 BCE, at concentrations indicating plants unusually high in psychoactive compound — smoking, in a funerary rite, demonstrated rather than inferred. In the same year Arie, Rosen and Namdar found Δ9-THC, CBD and CBN on the smaller altar of the Judahite shrine at Tel Arad, mixed with animal dung as a slow fuel, with frankincense on the larger altar beside it. Two laboratories, working independently with comparable extraction methods, reported the same result.
The same shift applies in the Andes. Miller et al. (2019) analysed a fox-snout ritual bundle from a Bolivian rock shelter dated to around 1000 CE and identified bufotenine, DMT, harmine, cocaine and benzoylecgonine on a single artefact — including harmine and DMT together, the two pharmacological halves of ayahuasca, in one kit. The plants involved grow in ecological zones far apart, which makes the bundle evidence of botanical knowledge and exchange as much as of consumption.
How structurally divergent substances converge on a single receptor — and why mescaline (phenethylamine) producing the same mystical experience as DMT (tryptamine) is one of pharmacology's most remarkable facts.
The 5-HT₂A receptor (serotonin 2A) is the molecular keystone of the psychedelic experience. Located primarily in the prefrontal cortex, its activation disrupts the Default Mode Network — the brain's self-referential processing hub. DMN suppression correlates directly with ego dissolution and mystical experience intensity (Carhart-Harris 2014).
Richard Evans Schultes, the Harvard botanist who conducted fieldwork in the Amazon from the 1940s through the 1970s, documented psychoactive plant traditions across dozens of Amazonian cultures — work that essentially founded ethnobotany as a discipline. Many of the traditions he documented no longer exist. His student Wade Davis extended this work to ibogaine and the Haitian zombie phenomenon. The ethnobotanical record is the empirical foundation beneath the legal and clinical debates.
What makes this remarkable: DMT, psilocybin, LSD, and mescaline reach the same receptor by completely different molecular routes. Tryptamines (DMT, psilocin, LSD) approach the binding pocket from one angle. Phenethylamines (mescaline) approach from an entirely different angle. Same destination. Convergent pharmacology from divergent structure — which is why the phenomenological experience across all four is recognizably similar despite structural differences.
Each entheogen has a structural relative already present in the human body. This is not coincidence — it explains why these molecules have receptors to bind. The brain did not evolve 5-HT₂A receptors for exogenous psychedelics; psychedelics activate receptors built for endogenous compounds.
That classical psychedelics bind 5-HT2A is not in dispute. What binding explains is a harder question, and imaging alone cannot answer it: a receptor can be occupied during an experience without being the reason for it. The test that settles it is pharmacological blockade.
Preller and colleagues gave LSD with and without pretreatment by ketanserin, a 5-HT2A antagonist. The general subjective effects of LSD were fully blocked. So was the drug’s characteristic effect of making previously meaningless stimuli feel personally relevant — the attribution of significance that participants describe afterwards as the substance of the experience. Block one receptor and the meaning goes with it. This is the strongest single piece of evidence that 5-HT2A agonism is not merely associated with the psychedelic state but constitutive of it.
The mechanistic account above that receptor has moved. Carhart-Harris and Friston’s REBUS model (2019) reframes the loose 2014 “entropic brain” idea in predictive-processing terms: psychedelics relax the weight the brain gives its high-level priors, letting bottom-up signal revise expectations that are normally insulated from it. Siegel and colleagues (2024) then produced the sharpest imaging evidence yet by abandoning the usual design. Instead of scanning many people once, they scanned a few people around eighteen times each, before, during and for three weeks after 25 mg psilocybin, with 40 mg methylphenidate as an active comparator. Psilocybin disrupted functional connectivity more than three times as much as methylphenidate, and did so by desynchronizing the brain across spatial scales — dissolving the distinctions between networks rather than simply turning one down. The effect was strongest in the default mode network. “DMN suppression” is a serviceable shorthand; desynchronization is closer to what the data show.
One question divides the laboratories, and it is not a technical one: is the subjective experience doing the therapeutic work, or is it a side effect of the molecule that does?
The case for necessary is clinical and old. Pahnke’s nine characteristics, Griffiths’ mystical-experience scores, and the repeated finding that outcome tracks the intensity of the experience rather than the size of the dose all point the same way: something happens to the person, and the compound is the occasion for it. Bogenschutz’s alcohol-dependence work found reductions in drinking correlated with mystical-experience intensity, not with plasma levels.
The case for incidental is molecular and recent. Ly and colleagues (2018) showed that serotonergic psychedelics promote neuritogenesis and spinogenesis, with real increases in synapse number and function, through TrkB, mTOR and 5-HT2A signalling — a structural change in cortex that has nothing to say about what the participant saw. If that is the mechanism, the visions are epiphenomenal. Cameron and colleagues (2021) pushed the argument to its conclusion by engineering tabernanthalog, a water-soluble, non-hallucinogenic, non-toxic analogue of ibogaine that still promoted structural plasticity in rodents. A psychedelic-derived drug that does the plasticity without the trip is either the future of the field or a category mistake about what these compounds are for, depending which laboratory is asked.
This atlas does not adjudicate it, because it is not adjudicated. It is worth naming what rests on the answer. If the experience is necessary, psychedelic therapy is irreducibly a practice involving a prepared person, a trained guide and many hours — and the religious-use record on this site is continuous with the clinical one. If the experience is incidental, the therapeutic future is a pill with no ceremony attached, and the continuity between a Mazatec velada and a phase 3 protocol becomes a historical coincidence rather than a fact about the compounds. The law, which currently criminalises the ceremony and fast-tracks the pill, has already made its assumption.
The Default Mode Network is the brain's baseline state — active during rest, self-referential thought, mind-wandering, and rumination. It is the neural substrate of what we call "the self." Psychedelics suppress DMN activity acutely and dramatically. The suppression correlates with ego dissolution and mystical experience intensity. The Carhart-Harris "entropic brain" model (2014) proposes that psychedelics increase neural entropy — freeing the brain from its habitual self-organizing patterns — which both explains the mystical experience and the therapeutic mechanism for depression (a disorder of rigid, ruminative self-processing).
Pahnke's nine characteristics of the mystical peak experience — expanded with neural correlates, cross-substance occurrence rates, and first-person testimony.
Score your own psychedelic experiences using the validated 30-item Mystical Experience Questionnaire. Pahnke (1963), revised Barrett et al. (2015).
Active and recently completed clinical trials across all substances. NCT numbers, phases, institutions, and key findings.
Two claims are commonly made about psychedelic medicine, and only one of them survives contact with the published trials. The first is that the effects are large. They are. The second is that the evidence base is settled. It is not, and the reasons are structural rather than a matter of more studies being needed.
Psilocybin for depression. Goodwin et al. (2022) randomised 233 people with treatment-resistant depression to 25 mg, 10 mg or 1 mg of a synthetic psilocybin formulation with psychological support. Least-squares mean change on the MADRS at week 3 was −12.0, −7.9 and −5.4 respectively; the 25 mg versus 1 mg difference was −6.6. Raison et al. (2023) ran a phase 2 trial across eleven US sites comparing 25 mg psilocybin against niacin, chosen as an active placebo. Both are phase 2. Neither is an approval.
MDMA for PTSD. The two phase 3 trials are Mitchell et al. (2021), 90 participants with severe PTSD, and Mitchell et al. (2023), the confirmatory trial in 104 participants with moderate to severe PTSD. In the confirmatory trial, least-squares mean change on the CAPS-5 was −23.7 for MDMA-assisted therapy against −14.8 for placebo with identical therapy (P < 0.001, d = 0.7). These are the trials. They should not be confused with the widely cited pooled analysis of six phase 2 studies, which the journal retracted in November 2024 and which this atlas lists in its citation library expressly so the retraction can be found rather than missed.
The blinding problem. A 25 mg dose of psilocybin is not something a participant can be uncertain about, and neither is MDMA. Functional unblinding is close to total in this field, expectancy effects are large in depression and PTSD specifically, and the therapy hours are part of the intervention in both arms. Raison’s niacin comparator and the 1 mg arm in Goodwin are attempts to narrow the gap, not to close it. This is why an FDA advisory committee voted against MDMA-assisted therapy in 2024 despite two positive phase 3 trials: the question was never whether the effect was large, but whether the design could attribute it.
Read the trials rather than the summaries, and check any figure here against the cited paper and against ClinicalTrials.gov before relying on it.
Every retained substantive claim must identify its source type, scope, verification date, and correction route. Unverified dose, interaction, current-law, legal-risk, RFRA-strength, and living-tradition protocol tools have been removed from production.
Machine-readable records connect retained claims to sources and state the limits of each source.
Community-origin material is not treated as public-domain permission. Living-tradition protocols are excluded unless the relevant authority has approved publication.
This edition does not provide individualized medical advice, dose guidance, interaction clearance, or current legal advice. Those functions require qualified review and continuously maintained jurisdictional data.
Archaeological claims are ranked by the kind of evidence behind them rather than by age, and the ranking is published rather than implied: chemical, botanical, textual, ethnographic, iconographic, contested, and not supported. A claim on a lower rung is not omitted; it is labelled.
Case law and statute cite the United States Reports scan, the court’s own opinion PDF, the Statutes at Large, or the text of the regulation — not a secondary summary. Archaeology cites the excavation report or the dating paper. Every DOI on this site is checked against Crossref and every URL fetched before publication.
Where the literature genuinely divides — the Shanidar pollen, the Eleusinian ergot hypothesis, whether the subjective experience is necessary to the therapeutic effect — both positions are stated and both are cited. Consensus is not manufactured by omission.