What it is
Chemistry asks what compounds a natural source contains, how much of each, how that changes from one specimen to the next and from one week of handling to the next, and how sure anyone can be of numbers from instruments and procedures with errors of their own.
A few terms recur. Tryptamines are a family of alkaloids, nitrogen-containing compounds made by living organisms, built on the same skeleton as the amino acid tryptophan, and serotonin belongs to it. An analyte is the specific substance a method is designed to measure, and the matrix is everything else in the sample, such as sugars, water and enzymes, which can interfere.
Most of the page concerns psilocybin-producing mushrooms, where the most published chemistry exists, with a short section on vine, cactus and root-bark sources.
What is known
The short version. The chemistry of psilocybin mushrooms is well established as chemistry. Psilocybin is usually the main tryptamine, several related compounds sit beside it, and the amounts vary a great deal between species, strains, individual mushrooms and batches. Some measurements are fragile. Whether the companion compounds change what a person experiences is open, and the evidence so far is mixed and mostly preclinical.
The tryptamines in psilocybin mushrooms
Psilocybin is a phosphate ester of psilocin, a prodrug that the body converts into psilocin, the compound that acts mainly on serotonin receptors [2]. Psilocybin weighs 284.25 atomic mass units and psilocin 204.27, a ratio of about 1.39 that matters when totals are calculated [16].
A 2026 laboratory study measured six tryptamines: psilocybin, psilocin, baeocystin, norbaeocystin, norpsilocin and aeruginascin. Psilocybin dominated in every one of 14 strains, and norpsilocin, norbaeocystin and aeruginascin were below the limit of detection, the faintest signal a method can separate from noise, in most strains: norpsilocin turned up in one strain, norbaeocystin in one and aeruginascin in four [1]. A 2024 analysis of one species also reported 4-hydroxytryptamine, but its average in whole fruiting bodies fell below the limit of quantification, so that figure is uncertain [4].
Two chemical changes explain much of what appears in the measurements. Dephosphorylation, the loss of the phosphate group, turns psilocybin into psilocin, and moisture, heat, acidity and tissue enzymes can drive it, after which oxidation removes psilocin as dark, poorly soluble products [2]. Psilocybin itself does not turn blue without a phosphatase, and a laccase oxidizes psilocin into blue-colored chains [42], but in one study frozen mushrooms turned deep blue without psilocin levels tracking the color, and one species turned intensely blue despite low psilocin [3]. Bluing is a visible sign of chemistry in progress, no measure of potency and no test of identity.
How the fungus builds psilocybin
A 2025 study of the enzymes summarizes the route in Psilocybe as four steps from tryptophan, with baeocystin as the intermediate. Psilocin is not a step on this route. A fibercap mushroom, Inocybe corydalina, uses different enzymes in a different order, which the authors take to mean the ability evolved twice [41]. This is test-tube biochemistry on recombinant enzymes and says nothing about what the compounds do.
How much: variation at five levels
Between species. In six Psilocybe species grown indoors under controlled conditions, species averages for psilocybin ran from 0.84 percent of dry weight in P. stuntzii to 1.46 percent in P. zapotecorum, about 1.7-fold. The single-sample range of 0.45 to 1.89 percent is made of individual samples, not species averages, each average rests on 5 to 25 samples, and the visual fingerprint comparisons rest on one sample track per species [3]. A larger analysis of 226 wild and herbarium fruiting bodies found a highest psilocybin value of 15.54 mg per gram of dry mass in one variety of P. serbica, while several collections, among them P. fuscofulva and Inocybe calamistrata, held no tryptamines at all. Handling varied, so these samples cannot be pooled with the controlled-cultivation numbers [5].
Between strains of one species. In a 2026 study, 14 strains of P. cubensis were grown on one substrate and processed by one procedure, each in a single chamber, with each strain's averages from one powder made of several mushrooms, so strain cannot be separated from chamber. Total tryptamines ranged from 2.62 to 20.65 mg per gram of dried material, more than 7.8-fold, and psilocybin alone about 9.7-fold (a calculation from the two published means). The tryptamines did not rise and fall together, and the ranking is soft: in one strain a pooled powder gave 13.53 mg per gram of psilocybin against 6.43 for five individual mushrooms, about 2.1-fold (a comparison from the paper's two tables). Strain identity rested on appearance and supplier names, because no reference genetic database exists [1].
Between mushrooms of one strain. Five fruiting bodies of each of six strains from one chamber gave a coefficient of variation of 12.81 to 23.39 percent for psilocybin [1]. Seven P. zapotecorum specimens in another analysis gave 17.9 ± 1.7 mg per gram, where the 1.7 is a standard error and the standard deviation is near 4.5 (a calculation from n of seven) [4], and the species-reference study gave about 14.6 mg per gram across 25 samples by a different method [3]. The same order of magnitude is reassuring, though not close agreement.
Within a mushroom and across time. Caps held about twice the tryptamine of stems in one P. cubensis analysis, short of significance [6], while in P. zapotecorum the difference was significant, in another species and method [4]. In the species-reference study, psilocybin and psilocin fell as mushrooms of four species matured, which the authors call preliminary, and three flushes of P. stuntzii matched by visual fingerprint on one cultivar, not a quantified match [3].
Between batches of a product. A laboratory in Jamaica reported across 26 batches of P. cubensis grown under uniform conditions a mean psilocybin and psilocin content of 1.14 percent with a standard deviation of 0.17, and real batch-to-batch differences. Encapsulated biomass kept its content for 11 months, shown in a figure, not as a number. Only psilocybin and psilocin were measured, and the study was commercially funded, with authors linked to testing laboratories [7].
Summaries give different headline ranges. The 2026 Biomolecules review is narrative, followed no formal search protocol, and has two numbers that did not match the primary papers it cites, so this page relies on it only for its framing and its own statements [2].
How composition is measured
HPTLC, a plate method, quantified psilocybin and psilocin in the reference-material project but could only identify norbaeocystin and baeocystin. The abstract says aeruginascin was identified too, but the results text says it was not visible, so this page does not rely on the abstract [3]. The 14-strain study quantified all six tryptamines by HPLC [1].
A 2026 single-laboratory LC-MS/MS validation covered 26 psychedelic and psychoactive compounds across fungi, cactus, leaf, tea and a maltodextrin powder simulating a finished product. The fungal recovery data came from a surrogate matrix, a reishi mushroom, and an actual Psilocybe sample entered only the intermediate-precision work. Recovery stayed within 88 to 124 percent for the fungal surrogate, cactus and leaf but ranged from 81 to 174 percent in maltodextrin, where norbaeocystin and harmaline fell outside the acceptance range and norbaeocystin also failed repeatability. Norbaeocystin missed the range in the reishi matrix at the low level too, so the problem is not confined to the finished-product simulation. No other laboratory repeated the validation [8]. The HPTLC project came from a mushroom products company and the validated method from authors at a commercial testing laboratory, and both have an interest in the standards that emerge [3] [8].
qNMR, quantitative nuclear magnetic resonance, reads psilocin from hydrogen signals and psilocybin from phosphorus signals, which avoids the large sugar signals, though for psilocin the trehalose signal still crowded the spectrum and was corrected by calculation. The authors call the method calibration-free, but it uses an external reference standard (PULCON). Against LC-MS on four samples, psilocybin agreed in three and read 3.7 times higher by NMR in one P. cyanescens sample, and psilocin read about 5 to 20 percent lower by NMR [9].
Procedure matters as much as instrument. The qNMR method needed four rounds of ultrasound-assisted extraction to recover 95 percent of psilocybin and psilocin, so single-pass analyses can undercount, and the 14-strain study used two rounds, so its absolute values may be underestimates, which is an inference [1] [9]. A systematic review of 25 extraction studies found that method and species both change yields, but it is qualitative, has no meta-analysis, and has mis-formatted values in its yield table, so no table numbers are reused here [10]. Psilocybin standards dissolved in water vanished from a mushroom sample while psilocin rose, because tissue phosphatases converted it, and water-free standards avoided this [9].
Totals and equivalents. Studies sum tryptamines differently [1] [9], and because psilocin is lighter, one scale means multiplying psilocin by about 1.39 [16], so comparisons between studies have to check which total was used.
Community testing (level 3). A 2024 community-laboratory article says many commercial mushroom-branded products contained no psilocybin and instead held 4-AcO-DMT, a synthetic compound, but it gives no count or share and its test data did not load [36]. A 2022 community "Psilocybin Cup" report was not read, and nothing here rests on it [35]. These reports are not peer reviewed and their methods were not visible.
Stability and handling as a variable
If a compound changes between collection and measurement, the measurement describes the handling as much as the source. The findings below concern sample integrity in research and are not guidance for anyone handling material.
An analysis of P. cubensis cut fresh fruiting bodies lengthwise before every storage condition, and the authors attribute part of the losses to that cutting. The largest loss was in fresh pieces held at minus 80 degrees Celsius, which held 94 percent less psilocybin than pieces dried in the dark. Once dried and ground, the powder lost about half of all its tryptamines in the first month under every storage condition tested, then leveled off, and the conditions looked alike after 15 months [6]. In the reference-material project, a day at minus 20 degrees before freeze-drying cut psilocybin in P. zapotecorum from 1.29 to 0.08 percent, with no cutting reported [3]. The two handling chains differ, but both found large psilocybin losses after freezing, which supports the direction of the effect and nothing more.
Psilocin rose in both reference-material samples, only from 0.04 to 0.08 percent in P. zapotecorum, which by a calculation from the published figures covers a small fraction of the psilocybin that vanished, and above the instrument's calibration range in P. cyanescens, so the balance cannot be closed there [3]. Oxidation of psilocin to colored products, documented in injured fruiting-body tissue, is a candidate for the missing mass [42], but that is a hypothesis, not a result of the freezer study. The Biomolecules review states that stability of the minor alkaloids has not been systematically studied [2]. Two partial reports exist: the 15-month biomass study measured baeocystin, norbaeocystin and aeruginascin alongside psilocybin and psilocin [6], and a 2024 preprint found, in water, that the dephosphorylated forms of the three compounds were far less stable than the phosphorylated ones [13].
In whole blood spiked in the laboratory, psilocin lost 46 to 66 percent after a day at room temperature, while psilocin glucuronide stayed stable for at least 14 days, and refrigeration did not protect psilocin, which differs from an earlier report the authors cite. A one-year freezer figure for the glucuronide rests on a single sample [11].
What happens in the body
The measurements in this section come from phase 1 studies of purified, synthesized psilocybin in healthy adults. They describe the compound and say nothing about a whole mushroom.
After swallowing, psilocybin is not found in blood: in 12 healthy adults it appeared in neither plasma nor urine, and psilocin, the compound that could be measured, peaked at about two hours with a half-life of about three hours [38], and about 1.4 to 2.7 hours in 51 participants from two trials [39]. Less than 2 percent left in urine as free psilocin, about a fifth as psilocin glucuronide and about a third as 4-HIAA, inactive at serotonin receptors in cell tests [38] [39] [40]. The 12-person study had no control group, and its glucuronide pool was inferred [38].
In a 2024 study, the liver enzyme CYP2D6 broke down nearly all of the psilocin in the test tube and CYP3A4 about 40 percent, while seven other enzymes did not. In a small group of volunteers, though, CYP2D6 genotype did not change psilocin levels in blood, a result the authors call preliminary, and genotype group sizes were not in the text read for this page. Response was not measured, so the result says nothing yet about whether genotype changes how a person responds. The study also shows the limits of test-tube work: UGT1A10 and liver microsomes did not glucuronidate psilocin, although that is a prominent route in living people, and a proposed metabolite, norpsilocin, appeared in the test tube and in mice but not in the human samples, with its identity unconfirmed [40].
The companion compounds: what has been measured
In 2020, chemists synthesized norbaeocystin, baeocystin, norpsilocin and aeruginascin and tested them. Baeocystin alone showed no psychedelic-like activity in the mouse head-twitch response, a behavioral proxy for activation of the 5-HT2A receptor, yet its likely breakdown product norpsilocin was a potent activator of that receptor in cell assays [12]. A 2024 preprint, the only version read for this page, found nearly identical rates of dephosphorylation and monoamine oxidase metabolism for baeocystin, norbaeocystin and aeruginascin. In PAMPA, a passive test across an artificial membrane that is not a true blood-brain barrier, the dephosphorylated forms of baeocystin and norbaeocystin crossed as well as psilocin, though with three samples per compound the differences were not significant. Norbaeocystin lowered immobility in the rat forced-swim test at a borderline p value of 0.05, in males only, with some test statistics not matching their p values, and part of the funding came from a company whose authors hold patents on the biosynthesis of these compounds [13]. A 2025 rat study, read at abstract level, found that baeocystin and norpsilocin barely entered the brain, with no significant behavioral effect of baeocystin at the two amounts tested [14], and a 2024 norpsilocin-analog study fits the idea that brain access limits norpsilocin, though access was not measured directly [15].
The public ChEMBL database reflects this. On 4 October 2026, norpsilocin had nine activity records, including a human 5-HT2A activation value near 8.4 nanomolar with a maximal response of 93 percent, while a different assay format showed only 2.9 percent, so a low EC50 in one readout is not the same as high efficacy. Baeocystin had one record, and none turned up for aeruginascin or norbaeocystin [16]. The companion compounds are pharmacologically real in cells and mostly quiet in the animals tested, and whether they matter in people at the amounts in a mushroom is a question that none of these studies answers.
The entourage question
The "entourage effect" is the idea that minor compounds in a whole source change what the main compound does, either by acting on receptors themselves or by altering how psilocin is absorbed or broken down. The Biomolecules review states what would be needed: the minor compounds would have to show independent psychoactive effects or alter the effects of psilocybin [2].
Animal studies. A 2015 Polish study compared methanol extracts of two wild mushrooms with pure psilocin in mice. The extracts held roughly a tenth of the psilocin of the pure sample, yet produced a similar twitch response and were nearly as toxic, and the authors read this as synergy. The full text shows why the case is weak: the authors report the lethal-dose differences as insignificant and give p < 0.001 beside them, an internal contradiction, the extracts held psilocybin at five to six times the psilocin content, and psilocybin converts to psilocin in the body, so the comparison cannot isolate the minor compounds. No control extract from a mushroom without psilocybin was tested, and the lethal-dose groups were ten mice [2] [17].
A 2009 Japanese study of dried Psilocybe argentipes mycelium, not fruiting bodies, reported reduced marble-burying in mice, with purified psilocybin acting only at the highest amount tested. The paper's printed psilocybin amounts for the mycelium are about a thousand times larger than its own reported content of 0.024 percent implies, a units error, and on the corrected arithmetic the mycelium acted at a lower psilocybin content than the pure compound. The authors say the effect is probably unrelated to the hallucinogenic action [18]. In a 2024 mouse study, synthetic psilocybin and a P. cubensis extract at the same psilocybin amount gave similar twitch responses and blood psilocin, and each was compared with vehicle, not with each other (synaptic proteins rose in 6 of 16 region-and-protein combinations with the extract and 5 of 16 with synthetic psilocybin). No metabolic difference remained after correction for multiple testing, the mice were all male, and the authors state that the data are not conclusive evidence of an advantage for the extract [19]. None of the three animal findings had a control extract from a mushroom without psilocybin.
Monoamine oxidase. β-carbolines inhibit monoamine oxidase, an enzyme that breaks down serotonin and related compounds. A 2020 analysis found them in several Psilocybe species and showed by isotope labeling, that the fungus makes them itself, but in very small amounts: in one P. cubensis isolate harmane measured 21 micrograms per gram in cultured mycelium and 0.2 in fruiting bodies, and in sclerotia of P. mexicana harmane was 1.4 and harmine 1.6, roughly four to five orders of magnitude below psilocybin in fruiting bodies, though the comparison crosses studies and species. The authors did not measure enzyme inhibition themselves and call for pharmacological research before concluding that these compounds shape the effects [20], and the review says detected concentrations appear too low for clinically meaningful effects [2]. A qNMR study found none in one P. cubensis extract, though its extraction was not designed for them [9].
Computation. A 2026 study applied network pharmacology, docking and molecular dynamics to mushroom compounds, and its title and abstract speak of entourage mechanisms and enhanced efficacy. Nothing was measured, each simulation was a single run, four phosphorylated compounds including psilocybin were screened out in advance, and one of the β-carbolines it included was not reported in these mushrooms by the 2020 isotope study [20] [21]. It is a hypothesis for the laboratory to test.
Human evidence. The review states that some preliminary preclinical evidence points to possible entourage effects, that human validation is lacking, and that the evidence rests on a small number of unreplicated studies. The only direct human comparison it found had four patients in an open-label setting, who found synthetic psilocybin and whole mushrooms broadly similar, and the review cautions that expectation and ceremony may have shaped those preferences. Its ranking of dose precision, from purified psilocybin to whole biomass, concerns measurement reliability, not whether whole mushrooms work better or worse [2].
Registered trials. The three registered trials of whole-mushroom material found for this page (obsessive-compulsive disorder, refractory depression, PTSD) all enrolled clinical populations [24] [25] [26]. One lists "accepts healthy volunteers" as yes while its own criteria require depression, which looks like a registration slip [25], and none describes a purified-psilocybin comparison arm. One double-blind, placebo-controlled trial gave 34 adults who had planned to start microdosing on their own, listed in the registry as healthy volunteers, a low amount of whole P. cubensis against placebo. Acute-effect ratings were higher with the mushroom material only among participants who correctly identified the condition, 75 percent of the weeks were identified correctly, so the blinding broke, and the trial found no support for improved well-being, creativity or cognition. The material held about 0.16 percent psilocybin and psilocin combined, and what each participant received is unknown. The registry entry was first submitted on 5 May 2021, after the recruitment period of December 2019 to August 2020, although the paper states it was registered before it started [27] [28].
Safety signals and the limits of composition data
A 2026 review reports that the alkaloid content of non-standardized mushroom products can vary by more than an order of magnitude, and pools supervised studies to a mean blood pressure rise of about 19 mmHg systolic and 8.7 diastolic, peaking 60 to 90 minutes after administration [22]. Those figures come from supervised studies of a standard pharmaceutical product, so they describe the compound and do not show an effect of varying composition, and the review was read at abstract level only. A separate 2026 review of 219 publications reports hypertensive urgency or emergency when monoamine oxidase inhibitor drugs were combined with psilocybin mushrooms [23]. A case report describes an 18-year-old who reached an emergency department with anxiety, tremor and hallucinations after taking a commercial "mushroom extract" product and recovered fully. Neither the patient nor the product was tested for psilocybin or any other psychoactive compound, so the product's composition could not be verified and the compound responsible is unknown [37].
Other natural sources, briefly
Ayahuasca's vine β-carbolines inhibit the gut enzymes that would otherwise inactivate its DMT, and a 1984 analysis of eight samples found that this inhibition in vitro tracked β-carboline concentration and that β-carbolines combined additively, not synergistically [29]. A 2019 Brazilian study validated a method for DMT and three β-carbolines [30], a 2024 study rebuilt the mescaline pathway of peyote in a model tobacco plant and in yeast [31], and test-tube work on iboga showed that two enzymes carry out the last two steps of the ibogaine pathway, which appears to branch, with a further enzyme not excluded [32]. The body quickly demethylates ibogaine to noribogaine [33]. Quantitative variation across these plant sources is not covered here.
Where the evidence is weak, and what came back empty
Baeocystin alone did nothing in the mouse twitch assay [12] and little in rats [14]. The one paper whose title claims entourage mechanisms measured nothing [21]. The mouse work most cited for synergy has an internally contradictory significance statement and no control extract [17], the human comparison has four participants [2], and CYP2D6 genotype did not change psilocin levels in the small human group, though response was not measured [40].
What is not known
How much reported variation comes from the material, and how much from drying, storage, extraction and the laboratory? Studies show that both matter [1] [3] [9], but none has partitioned them across the same strains with shared reference material.
Whether the minor compounds do anything in a person at the amounts present in a mushroom is unanswered. No trial has isolated them in humans [2], and the best-studied, baeocystin and norpsilocin, barely reach the rat brain [14].
How the minor tryptamines hold up over time has not been studied systematically, which is the wording the 2026 review uses [2]. Two partial reports exist [6] [13], so the question is partly answered and not closed.
Whether distinct chemotypes exist, meaning stable patterns in the relative proportions of compounds, is open. The 14-strain study found no coordinated rise and fall among compounds [1], and a difference between two varieties of P. serbica rests on one collection per variety, not separated genetically [5]. Whether such patterns survive changes in growing conditions has not been tested.
Whether a finished whole-source product can be measured as reliably as a purified compound is unsettled, and part of the imprecision attributed to whole mushrooms may be analytical, which is an inference and not a published finding.
What Metanoia asks
Metanoia's long-term goal is a comprehensive research center: an institution that keeps monitoring and reporting the field's evidence, convenes researchers, registries and regulators around shared reporting standards, and also conducts its own research, beginning with a validated measure of flourishing and a naturalistic observational cohort studied under those standards, and growing toward laboratory work in cultivation, chemistry and genetics under full legal authorization. The center is planned, not built, and its location has not been decided.
Metanoia does not hold or supply any material. Any laboratory work with controlled compounds would take place only under the legal authorization the law requires, which for Schedule I compounds in the United States includes a DEA researcher registration.
The questions this discipline is set up to ask:
- Standardization. What must a shared description of a material contain so that two studies can be compared, and which totals, units and methods should it fix?
- Measurement. How much apparent variability in whole-source material is analytical, where do methods break, as with norbaeocystin, and what reference materials would close the gap?
- Chemotypes. Do natural sources show reproducible chemical patterns across strains, growing conditions and time, described well enough to report beside any outcome?
- Companion compounds. Beyond receptor pharmacology in cells and rodents, is there measurable activity in people at the amounts found in a mushroom?
History in brief
Psilocybin and psilocin were isolated and characterized in the 1950s by Albert Hofmann and colleagues, as recounted in the modern papers cited here [1] [20], and the original reports were not read. Traditional use of these sources is documented mostly in ethnographic work, which was not read for this page, apart from one ethnomycological study of a Thai species read at abstract level [34]. This page makes no claim that tradition shows safety.
Sources
Reliability levels follow Metanoia's scale. Level 1 is a primary source or any peer-reviewed publication, reviews included, and it also covers database records and registry entries. Level 2 is a reliable secondary source that is not peer reviewed. Level 3 is grey literature, community material or a preprint. Level 4 is a traditional or ethnographic source. Levels 3 and 4 describe the type of source and say nothing about quality. Every peer-reviewed publication below is graded 1, and reviews and weak designs carry a design note. Where the version read was a preprint, the entry says so.
- Sharchaton A et al. J Fungi (Basel) 2026, 12(7):486. DOI 10.3390/jof12070486 PMID 42506248. Level 1. Laboratory study, 14 P. cubensis strains, one chamber per strain. Read in full 4 October 2026, supplementary tables not read. Detail moved from the text: psilocybin alone ranged from 2.03 to 19.60 mg per gram, DNA barcoding confirmed one species, and the study sums all six tryptamines by mass.
- Enriquez-Geppert S et al. Biomolecules 2026, 16(7):983. DOI 10.3390/biom16070983 PMID 42509777. Level 1 (narrative review, no formal search protocol). Read in full 4 October 2026, reference list not read. Two of its figures do not match the primary papers it cites. Detail moved from the text: the review describes 0.2 to 5.3 mg per gram as a 25-fold variation and elsewhere gives a species range of 0.01 to 3.04 percent of dry weight, and in summarizing the Polish mouse paper it says synthetic psilocin induced the greater twitch response.
- Windsor C et al. J AOAC Int 2026, 109(1):37-50. DOI 10.1093/jaoacint/qsaf007 PMID 39960890. Level 1. Cultivation and HPTLC profiling, six Psilocybe species. Read in full 4 October 2026, figures and supplement not read. The authors work for a company that states a financial interest in developing Psilocybe cultivars. A published correction exists (PMID 41422512) concerning the mobile-phase ratio of the method. Its body was not re-read here. Detail moved from the text: a day at minus 20 degrees before freeze-drying also cut psilocybin in P. cyanescens from 1.16 to 0.27 percent, and the project's visual fingerprint comparisons rest on one sample track per species.
- Miller DR et al. J Psychedelic Stud 2024. DOI 10.1556/2054.2023.00332 PMID 41836496. Level 1. P. zapotecorum cultivation, chemistry and genome, one clone. Read in full 4 October 2026, figures and supplement not read. Funded by the NIH, the Entheome Foundation and Tryp Labs. The text and table disagree on several norbaeocystin values, and the plus-minus values are standard errors. Detail moved from the text: the seven P. zapotecorum specimens ranged from 10.6 to 25.7 mg per gram, and caps held 25.5 against 15.6 mg per gram of psilocybin in stems.
- Gotvaldová K et al. Int J Mol Sci 2022, 23(22):14068. DOI 10.3390/ijms232214068 PMID 36430546. Level 1. 226 fruiting bodies, 82 collections, convenience sample with variable handling. Read in full 4 October 2026, supplement not read.
- Gotvaldová K et al. Drug Test Anal 2021, 13(2):439-446. DOI 10.1002/dta.2950 PMID 33119971. Level 1. Stability and distribution of tryptamines in P. cubensis, short communication. Read in full 4 October 2026 (publisher PDF hosted by a state health authority), figure data and supplement not read. The abstract and the body differ on one mycelium value, and the body is used. Detail moved from the text: caps held about twice the tryptamine of stems in P. cubensis (1.03 against 0.52 percent), and losses differed with storage condition and with fresh or dried material.
- Foster K et al. J AOAC Int 2026, 109(1):51-60. DOI 10.1093/jaoacint/qsaf072 PMID 40802522. Level 1. Standardized P. cubensis biomass, 26 batches, 11-month stability. Read in full 4 October 2026 (publisher page), supplement and figure data not read. Commercially funded. The paper's dose calculations are not reproduced here. Detail moved from the text: the authors concluded that each batch should be tested.
- French T et al. J AOAC Int 2026, 109(1):61-74. DOI 10.1093/jaoacint/qsaf081 PMID 41014027. Level 1. Single-laboratory validation, 26 analytes. Read in full 4 October 2026, chromatogram figures not read. The authors are at a commercial testing laboratory. Detail moved from the text: recovery is the share of a known added amount that the method gets back, the authors recommend isotope-labeled internal standards (reference molecules tagged with heavier atoms), the reishi mushroom used as surrogate does not contain these compounds, and the authors cite low signal intensity and say the norbaeocystin results need more work. The January 2026 issue of the Journal of AOAC International carried four papers on psychedelic plants and fungi, three of them cited here [3] [7] [8], and AOAC International publishes consensus analytical methods, so this is early groundwork for a shared language of measurement.
- Rodríguez L et al. ACS Omega 2025, 10(46):55725-55732. DOI 10.1021/acsomega.5c07092 PMID 41322617. Level 1. qNMR method, six P. cubensis samples. Read in full 4 October 2026, supplement not read. Detail moved from the text: the qNMR study sums psilocin and psilocybin by mass and found 0.154 to 0.456 percent across six samples, relative standard deviations were 11.3 percent for psilocin and 16.1 percent for psilocybin, and user-supplied samples that may have met moisture were mostly psilocin while a freshly dried laboratory sample was almost all psilocybin, an explanation the authors give as likely and did not test.
- Galdino TP et al. Pharmaceuticals (Basel) 2025, 18(3):380. DOI 10.3390/ph18030380 PMID 40143157. Level 1 (systematic review, 25 studies, qualitative, no meta-analysis). Read in full 4 October 2026, reference list not read. Its table 4 contains mis-formatted values.
- Bergh MS et al. J Anal Toxicol 2026, 50(4). DOI 10.1093/jat/bkag015 PMID 41723823. Level 1. Whole-blood method and stability, then 23 forensic cases. Read in full 4 October 2026, supplement not read. Detail moved from the text: psilocin glucuronide stayed stable for at least 14 days at room temperature or in the refrigerator, and the forensic lesson for research is that measuring a stable breakdown product can beat trying to preserve an unstable one.
- Sherwood AM et al. J Nat Prod 2020. DOI 10.1021/acs.jnatprod.9b01061 PMID 32077284. Level 1. Synthesis and testing of four minor tryptamines in vitro and in mice. Abstract only. Full text not accessible.
- Rakoczy RJ et al. Br J Pharmacol 2024. DOI 10.1111/bph.16466 PMID 38825326. Level 1 for the published paper, level 3 for the version read. Enzyme, receptor and rat studies of three minor tryptamines. Read in full as the bioRxiv preprint (DOI 10.1101/2023.10.19.563138, version 1, 23 October 2023). The published version was not accessible and may differ. Partly funded by a company, and some authors hold patents on tryptamine biosynthesis. Detail moved from the text: in rats only psilocybin produced head twitches.
- Danda H et al. Prog Neuropsychopharmacol Biol Psychiatry 2025. DOI 10.1016/j.pnpbp.2025.111439 PMID 40619050. Level 1. Baeocystin in Wistar rats. Abstract only. Full text not accessible.
- Sherwood AM et al. ACS Chem Neurosci 2024. DOI 10.1021/acschemneuro.3c00610 PMID 38189238. Level 1. Eight synthetic norpsilocin analogs, mouse twitch response. Read in full 4 October 2026, some synthesis details and the supplement not read. Detail moved from the text: eight synthetic norpsilocin analogs were tested, and adding one carbon to the nitrogen group restored psilocin-like twitch behavior.
- ChEMBL database, release 37 (1 May 2026), public REST API, queried 4 October 2026. https://www.ebi.ac.uk/chembl. Level 1 (database record). Molecule, activity and property records read for psilocybin, psilocin, norpsilocin and baeocystin. Name and structure searches for aeruginascin and norbaeocystin returned no molecule. Detail moved from the text: baeocystin's single record is from the mouse twitch assay, and no molecule record for aeruginascin or norbaeocystin turned up by name or structure.
- Zhuk O et al. Toxins (Basel) 2015, 7(4):1018-1029. DOI 10.3390/toxins7041018 PMID 25826052. Level 1. Mouse study of two mushroom extracts versus psilocin. Read in full 4 October 2026, figures not read. Detail moved from the text: both extracts produced head twitches, most with pure psilocin, the pure compound was slightly more toxic, the route was injection, and the material was wild collections.
- Matsushima Y et al. Biosci Biotechnol Biochem 2009, 73(8):1866-1868. DOI 10.1271/bbb.90095 PMID 19661714. Level 1. Marble-burying in mice, mycelium of P. argentipes. Read in full 4 October 2026 (publisher PDF from J-STAGE), figure data not read. The printed amounts contain a units error of about a thousandfold. Detail moved from the text: the dose-response curve was bell-shaped.
- Shahar O et al. Mol Psychiatry 2024, 29(7):2059-2073. DOI 10.1038/s41380-024-02477-w PMID 38378926. Level 1. Synthetic psilocybin versus mushroom extract, mouse brain proteins and metabolites. Read in full 4 October 2026, supplement not read.
- Blei F et al. Chemistry 2020, 26(3):729-734. DOI 10.1002/chem.201904363 PMID 31729089. Level 1. β-carbolines in Psilocybe species. Read in full 4 October 2026, supplement not read. Detail moved from the text: the paper found harmane and harmine in several Psilocybe species, isotope labeling traces atoms tagged with a heavier form of the element, and the authors take the inhibition figures from earlier literature.
- Murray Z et al. Sci Rep 2026, 16(1). DOI 10.1038/s41598-026-39483-7 PMID 41691031. Level 1 (computational prediction only). Read in full 4 October 2026, supplement not read. Detail moved from the text: the study predicted stable binding of several compounds to the 5-HT2A and monoamine oxidase A proteins, and the software predicted that the four phosphorylated compounds would not cross the blood-brain barrier.
- Szarpak L et al. Eur J Clin Pharmacol 2026, 82(8). DOI 10.1007/s00228-026-04151-2 PMID 42538465. Level 1 (narrative review). Abstract only. Full text not accessible. Detail moved from the text: the review calls a heart-valve concern tied to the 5-HT2B receptor plausible under chronic exposure but not shown clinically.
- Rached G et al. CNS Drugs 2026, 40(3):359-417. DOI 10.1007/s40263-025-01256-7 PMID 41546846. Level 1 (narrative review of 219 publications). Abstract only. Full text not accessible. Detail moved from the text: the review also draws on 28 non-peer-reviewed forum reports.
- ClinicalTrials.gov NCT07347405. https://clinicaltrials.gov/study/NCT07347405. Level 1 (registry record). Identification, status, design, eligibility, arm and summary fields read 4 October 2026 through the registry's v2 interface. Detail moved from the text: NCT07347405 is a phase 1 study in obsessive-compulsive disorder at the University of Arizona, planning 30 participants, recruiting since May 2026, with "healthy volunteers: no".
- ClinicalTrials.gov NCT06898606. https://clinicaltrials.gov/study/NCT06898606. Level 1 (registry record). Same fields, read 4 October 2026. Detail moved from the text: NCT06898606 is a Brazilian phase 1/2 pilot in refractory depression, planning 24 participants, with and without fluoxetine.
- ClinicalTrials.gov NCT07275970. https://clinicaltrials.gov/study/NCT07275970. Level 1 (registry record). Same fields, read 4 October 2026. Detail moved from the text: NCT07275970 is a phase 1 PTSD study with up to 24 participants, active but not recruiting.
- Cavanna F et al. Transl Psychiatry 2022, 12:307. DOI 10.1038/s41398-022-02039-0 PMID 35918311. Level 1. Double-blind placebo-controlled trial, 34 adults, whole P. cubensis. Read in full 4 October 2026, supplement not read.
- ClinicalTrials.gov NCT05160220. https://clinicaltrials.gov/study/NCT05160220. Level 1 (registry record). Registry entry for the trial in item 27, read 4 October 2026. First submitted 5 May 2021. Detail moved from the text: the trial did not compare whole with purified material, and the psilocybin content was lower than other sources would predict.
- McKenna DJ, Towers GHN, Abbott F. J Ethnopharmacol 1984, 10(2):195. DOI 10.1016/0378-8741(84)90003-5 PMID 6587171. Level 1. Alkaloids in ayahuasca samples, MAO inhibition in vitro. Abstract only. Full text not accessible. Detail moved from the text: the 1984 analysis tested chemically why the traditional two-plant brew pairs a DMT source with an enzyme-inhibitor source, and DMT by mouth alone is inactive because gut enzymes break it down.
- Souza RCZ et al. J Chromatogr B 2019. DOI 10.1016/j.jchromb.2019.06.014 PMID 31220748. Level 1. Method validation, 38 Brazilian ayahuasca samples. Abstract only. Full text not accessible.
- Berman P et al. Mol Plant 2024. DOI 10.1016/j.molp.2024.05.012 PMID 38835170. Level 1. Mescaline biosynthesis in peyote. Abstract only. Full text not accessible. Detail moved from the text: the study sequenced the peyote genome and identified four groups of enzymes covering the six chemical steps of the pathway.
- Farrow SC et al. J Biol Chem 2018. DOI 10.1074/jbc.RA118.004060 PMID 30030374. Level 1. Final steps of ibogaine biosynthesis in Tabernanthe iboga, enzyme assays. Read in full 4 October 2026, supplement not read.
- Mash DC. Pharmacol Res 2023. DOI 10.1016/j.phrs.2022.106620 PMID 36907284. Level 1 (invited review). Abstract only. Full text not accessible.
- Gartz J, Allen JW, Merlin MD. J Ethnopharmacol 1994, 43(2):73-80. DOI 10.1016/0378-8741(94)90006-X PMID 7967658. Level 1 (ethnomycological and biochemical study). Abstract only. Full text not accessible.
- Oakland Hyphae and Hyphae Labs. Spring Psilocybin Cup 2022 final results. https://www.oaklandhyphae510.com/post/oakland-hyphae-hyphae-labs-present-the-spring-psilocybin-cup-2022-final-results. Level 3 (community testing). Post dated 16 May 2022. The landing page was read in full on 4 October 2026. It links to a report that was not downloaded, and no statement on this page rests on that report. Detail moved from the text: the report is by the same laboratory and a community group.
- DoubleBlind and Hyphae Labs. Product testing article, 4 April 2024. https://doubleblindmag.com/double-blind-hyphae-labs/. Level 3 (magazine). Page text read in full on 4 October 2026. The test data appear as images that did not load, so counts and methods were not seen. Detail moved from the text: the article reports work by a community testing laboratory, Hyphae Labs, with a magazine on commercial mushroom-branded products.
- Udoro et al. Cureus 2026. DOI 10.7759/cureus.110791 PMID 42445624. Level 1 (single case report). Read in full 4 October 2026, figure and reference list not read.
- Brown RT et al. Clin Pharmacokinet 2017, 56(12):1543-1554. DOI 10.1007/s40262-017-0540-6 PMID 28353056. Level 1. Open-label pharmacokinetic study in 12 healthy adults, synthesized psilocybin, no control group. Read in full 4 October 2026 (publisher PDF hosted by a university research center), reference list not read.
- Holze F et al. Clin Pharmacol Ther 2023, 113(4):822-831. DOI 10.1002/cpt.2821 PMID 36507738. Level 1. Pooled secondary analysis of two phase 1 trials in healthy participants, synthesized psilocybin. Read in full 4 October 2026 (open-access publisher page), supplement not read. Partly funded by a psychedelic drug developer, and one author consults for it. Detail moved from the text: the half-life was between about 1.4 and 2.7 hours depending on the dose group and the calculation method, and body weight, age and kidney filtration did not predict blood levels.
- Thomann J et al. Front Pharmacol 2024, 15:1391689. DOI 10.3389/fphar.2024.1391689 PMID 38741590. Level 1. Test-tube, mouse and human-sample work on the metabolism of psilocin. Read in full 4 October 2026, supplement not read. Authors report ties to the Heffter Research Institute, the MIND Foundation, Roche and Definium Therapeutics (formerly Mind Medicine Inc.).
- Schäfer T et al. Angew Chem Int Ed 2025, 64(46):e202512017. DOI 10.1002/anie.202512017 PMID 40977073. Level 1. Test-tube biochemistry of recombinant enzymes and predicted structures, Inocybe and Psilocybe. Read in full 4 October 2026, supplement not read. Its account of the Psilocybe pathway is a summary of earlier work whose full text was not accessible for this page. Detail moved from the text: a decarboxylase gives tryptamine, an oxygenase adds a hydroxyl group at position 4 of the ring, a kinase adds the phosphate, a methyltransferase adds two methyl groups, the same kinase can put a phosphate back on free psilocin, and in the fibercap mushroom baeocystin is a second end product beside psilocybin.
- Lenz C et al. Angew Chem Int Ed 2020, 59(4):1450-1454. DOI 10.1002/anie.201910175 PMID 31725937. Level 1. Enzymes from P. cubensis fruiting bodies behind injury-triggered blueing. Read in full 4 October 2026, supplement not read. Detail moved from the text: the enzymes came from P. cubensis fruiting bodies, the laccase oxidizes psilocin into blue-colored chains of 3 to 13 linked psilocin units, and psilocybin itself does not turn blue without the phosphatase.
The book on all twelve dimensions Back to the twelve dimensions