Most of what circulates about “minor cannabinoids” is marketing copy wearing a lab coat — here is what the literature supports, and where it runs out.
This article keeps three things apart that consumer writing blends: what was measured in a laboratory, what was tested in animals, and what was shown in humans. Nothing here is a treatment recommendation — in Thailand cannabis is prescription-only and therapeutic advertising is prohibited.
1. How Many Cannabinoids Have Actually Been Identified
The figure that holds up
The review by Radwan, ElSohly and colleagues in Molecules in 2021 reports more than 500 compounds described from Cannabis sativa, of which 125 cannabinoids have been isolated or identified across eleven structural subclasses, plus 120 terpenes.
How the number gets inflated
| Error | What it looks like | What is true |
|---|---|---|
| Constituents read as cannabinoids | “Over 500 cannabinoids” | Over 500 compounds; about 125 are cannabinoids |
| An old count frozen in place | “113 cannabinoids”, quoted as current | A figure from an earlier review generation |
| Identified read as characterised | “Each with unique effects” | Identification is analytical chemistry only |
Other reviews say “more than 100”; thresholds differ. Identified means detected and structurally assigned, nothing more. Of the 125, perhaps six have been studied in humans in any depth.
2. How the Plant Builds Cannabinoids
The acid forms
Cannabis does not make tetrahydrocannabinol or cannabidiol directly; it makes their acid precursors. Cannabigerolic acid is the branch point, converted by plant enzymes into tetrahydrocannabinolic acid, cannabidiolic acid or cannabichromenic acid.
Decarboxylation, and why raw plant material is not intoxicating
Heat drives off a carboxyl group as carbon dioxide, turning acid into neutral cannabinoid. Moreno, Dyer and Tallon studied this from 80–160 °C in Industrial & Engineering Chemistry Research in 2020: first-order kinetics, tetrahydrocannabinolic acid fastest at every temperature, conversion complete in about 90 minutes at 120 °C or 20 at 160 °C.
Tetrahydrocannabinolic acid does not engage the cannabinoid receptor as tetrahydrocannabinol does, so unheated plant material is not intoxicating.
3. Tetrahydrocannabinol
Receptor pharmacology
Delta-9-tetrahydrocannabinol (THC) is a partial agonist at cannabinoid receptor types 1 and 2. Type 1 receptors are dense in the brain and are why THC is intoxicating; type 2 sit largely on immune cells.
Where the clinical evidence is strongest
The 2017 United States National Academies of Sciences, Engineering, and Medicine report graded the whole literature and found conclusive or substantial evidence of effectiveness in three areas:
| Symptom area | Grade |
|---|---|
| Chronic pain in adults | Conclusive or substantial |
| Chemotherapy nausea and vomiting, oral cannabinoids | Conclusive or substantial |
| Spasticity in multiple sclerosis, oral cannabinoids | Substantial, patient-reported only |
The adverse-effect profile, stated plainly
The same report graded harms. There is substantial evidence of an association between cannabis use and development of schizophrenia or other psychoses, highest among the heaviest users; with increased motor vehicle crash risk; and between rising frequency of use and progression to problematic use. There is moderate evidence of acute impairment of learning, memory and attention, and long-term smoking is associated with worse respiratory symptoms and chronic bronchitis. Associations, not proof of cause.
4. Cannabidiol
A contested mechanism
Cannabidiol (CBD) is non-intoxicating because it does not engage cannabinoid receptor type 1 as THC does. Laprairie and colleagues reported in the British Journal of Pharmacology in 2015 that it is a negative allosteric modulator at that receptor. Beyond that, mechanism is unsettled: a 2022 review by Peng and colleagues lists 5-HT1A receptors, TRPV1 channels, GPR55 and PPAR-gamma among proposed targets, many characterised far above blood levels products achieve.
Where the evidence is strong
One place, narrow and specific. A purified pharmaceutical cannabidiol oral solution holds United States Food and Drug Administration approval for seizures associated with Lennox-Gastaut syndrome, Dravet syndrome and tuberous sclerosis complex in patients one year and older, on the basis of randomised trials. That is a prescription medicine, not consumer oil, and its labelling reports somnolence, decreased appetite, diarrhoea and elevated liver transaminases in ten percent or more of patients, with warnings for hepatic injury.
Where the evidence is weak or absent
A review led by Eccleston in the Journal of Pain in 2024 found that of 16 randomised trials of cannabidiol for chronic pain, 15 showed it no better than placebo, with pooled data indicating more serious adverse events, including liver toxicity signals. Retail products held anywhere from no cannabidiol at all to far more than advertised, sometimes with undeclared tetrahydrocannabinol. Everything else attached to the molecule — anxiety, sleep, inflammation, skin, recovery — rests on mechanistic and small-study material.
5. Cannabigerol
Cannabigerol (CBG) is the neutral precursor form; by harvest most has converted downstream, so it is usually a trace compound. Cascio and colleagues reported in the British Journal of Pharmacology in 2010 that it is a potent alpha-2 adrenoceptor agonist and 5-HT1A antagonist, with low cannabinoid-receptor affinity.
The antibacterial work is laboratory and animal, not clinical
Farha, Brown and colleagues published work in ACS Infectious Diseases in 2020 testing cannabinoids against methicillin-resistant Staphylococcus aureus (MRSA).
| Finding for cannabigerol | Detail |
|---|---|
| Potency in culture | Minimum inhibitory concentration of 2 micrograms per millilitre |
| Animal model | In mice with systemic infection, 100 mg/kg gave about a 2.8-log reduction in bacterial counts, comparable to vancomycin |
| Safety margin | Haemolysis at 32 micrograms per millilitre, close above the effective concentration; little activity against Gram-negatives |
That is a petri dish and a mouse: no human clinical trial of cannabigerol as an antibacterial exists, and no safe systemic dose is established in people.
The single human trial
Cuttler and colleagues published a double-blind, placebo-controlled crossover trial in Scientific Reports in 2024: 34 healthy adults took 20 mg of hemp-derived cannabigerol or placebo a week apart, with reductions in self-reported anxiety and stress at some timepoints and no mood effect. Limits the authors state: small healthy sample, single dose, remote testing, no multiple-comparison correction.
6. Cannabinol
Cannabinol (CBN) is not something the plant sets out to make. It is what tetrahydrocannabinol becomes when it oxidises with time, air, light and heat, so a sample high in cannabinol is telling you about its age and storage.
The sedation claim is now this molecule’s entire commercial identity. Corroon reviewed it in Cannabis and Cannabinoid Research in 2021 under the blunt title “Cannabinol and Sleep: Separating Fact from Fiction”: no clinical trials using validated sleep questionnaires were identified, no polysomnography studies were found, the human research that exists is largely from the 1970s and 1980s with small samples, and evidence that it produces cannabis-like effects is mixed — most of it showing a lack of such effect.
7. The Lesser-Studied Cannabinoids
Tetrahydrocannabivarin
Tetrahydrocannabivarin (THCV), the propyl-chain analogue of tetrahydrocannabinol, behaves almost oppositely at low exposures. Haghdoost and colleagues reviewed it in Cannabis and Cannabinoid Research in 2025: it is a cannabinoid receptor type 1 antagonist at most relevant doses, agonist only at higher ones, and a 10 mg oral dose was indistinguishable from placebo in humans.
Cannabichromene
A 2024 review of cannabichromene (CBC) by Sepulveda and colleagues in the Journal of Pharmacology and Experimental Therapeutics calls it a cannabinoid receptor type 2 agonist with affinity in the range of THC, type 1 activity contested, and a potent TRPA1 agonist. Human data is two pharmacokinetic studies; the reviewers note these products sell with little evidence of safety or efficacy.
Cannabidivarin
Cannabidivarin (CBDV) is to cannabidiol what tetrahydrocannabivarin is to tetrahydrocannabinol. As an anticonvulsant candidate it reached a Phase 2 trial in adults with focal seizures, which missed its primary endpoint; later work in autism spectrum disorder produced no approved indication.
8. Delta-8 and the Semi-Synthetic Cannabinoids
How they are actually produced
Delta-8-tetrahydrocannabinol occurs naturally, but only in traces. As Kruger and Kruger noted in the Journal of Cannabis Research in 2022, it has lower affinity for cannabinoid receptor type 1 than delta-9, and a 1973 study found it about two-thirds as potent. It is a weaker relative of THC, and intoxicating.
What is unusual is the supply chain: commercial delta-8 is not extracted, it is made, by acid-catalysed isomerisation of cannabidiol. The United States Food and Drug Administration stated in May 2022 that concentrated delta-8 is typically manufactured from hemp-derived cannabidiol, that it had received adverse event reports including cases requiring hospitalisation, and that poison centres were seeing rising exposures. The same route yields hexahydrocannabinol.
The purity problem
Isomerisation is not clean, and the mixture is the issue. Ray and colleagues analysed ten delta-8 products in Molecules in 2022, and found impurities far beyond what the certificates of analysis declared. Some detected compounds are not identified cannabinoids at all, so their effects in humans are unknown. Testing cited by Kruger and Kruger found delta-9 at 1.3 to 5.3 percent in products sold as delta-8, plus heavy metals. The molecule is understood; the material sold under its name is not.
An unsettled legal position
| Jurisdiction | Position as of writing |
|---|---|
| Thailand, flower | A notification effective 26 June 2025 made cannabis flower a controlled herb; from 1 July 2025 supply needs a practitioner’s prescription, capped at 30 days. Promotion, vending-machine and online sales are banned |
| Thailand, extracts | Extracts above 0.2 percent tetrahydrocannabinol by weight remain narcotics under Category 5 of the Narcotics Code; only low-tetrahydrocannabinol domestic and seed extracts are carved out |
| Thailand, semi-synthetics | No Thai instrument naming delta-8 or hexahydrocannabinol could be located, so the position is unsettled; an intoxicating extract above 0.2 percent in a prescription-only system is no place to assume permission. The Cannabis and Hemp Act had not passed parliament as of mid-2026, so the framework rests on notifications that change fast |
| European Union | The 2025 European Drug Report records hexahydrocannabinol in 27 countries, controlled in at least 22 member states, with 24 semi-synthetic cannabinoids monitored by end-2024 |
| United States | A November 2025 federal law capped hemp at 0.4 milligrams total tetrahydrocannabinol per container and excluded cannabinoids synthesised outside the plant, from 12 November 2026 |
The direction of travel is consistent: the “not technically the controlled molecule” argument is closing.
9. The Entourage Effect
The term describes the idea that cannabis constituents together produce effects differing from the sum of the isolated parts. Reasonable as a hypothesis; as a slogan it travels far beyond its evidence.
Simei and colleagues asked directly whether the effect exists, in a narrative scoping review in Cannabis and Cannabinoid Research. They found limited evidence for it as a stable, predictable phenomenon, inconsistent preclinical findings, and limited evidence for clinical efficacy or safety built on the hypothesis — and urged industry, media and clinicians not to present it as proven.
What is genuinely demonstrated is narrow: a 2021 paper in Scientific Reports reported that several cannabis terpenes are cannabimimetic and can selectively enhance cannabinoid activity — in cell and rodent experiments. Everything after that is extrapolation: whole-plant preparations reliably superior in people, a terpene profile producing a named effect. The term travels because it turns complexity into a selling point and justifies premium positioning without any product-specific trial.
10. Boiling Points, and Why Published Charts Disagree
Most online boiling-point charts trace back to one unsourced table. Turovsky and colleagues, in the Journal of Cannabis Research, used thermogravimetric analysis and differential scanning calorimetry to estimate normal boiling points at atmospheric pressure: 245 ± 6 °C for delta-9-tetrahydrocannabinol and 378 ± 4 °C for tetrahydrocannabivarin. The widely quoted 157 °C figure, they explain, almost certainly refers to a distillation temperature under vacuum. A boiling point given without pressure or citation is folklore.
11. The Major Cannabinoids Compared
| Cannabinoid | Intoxicating | Main receptor interaction | Evidence status in humans | Boiling point |
|---|---|---|---|---|
| Delta-9-tetrahydrocannabinol | Yes | Partial agonist, types 1 and 2 | Strongest of the family; three symptom areas graded conclusive or substantial, plus documented harms | 245 ± 6 °C |
| Cannabidiol | No | Negative allosteric modulator at type 1 | Randomised evidence in three epilepsy syndromes; weak or absent elsewhere | Not documented |
| Cannabigerol | No | Alpha-2 adrenoceptor agonist, 5-HT1A antagonist | Laboratory and mouse antibacterial data; one small trial | Not documented |
| Cannabinol | Not meaningfully | Weak affinity at type 1 | Very thin; no modern controlled sleep trials identified | Not documented |
| Tetrahydrocannabivarin | Not at low exposure | Type 1 antagonist at most relevant doses | Two small human studies | 378 ± 4 °C |
| Cannabichromene | No | Type 2 agonist; potent TRPA1 agonist | Two pharmacokinetic studies only | Not documented |
| Cannabidivarin | No | Cannabidiol analogue | Phase 2 seizure trial missed its endpoint | Not documented |
| Tetrahydrocannabinolic acid | No | Does not engage type 1 as tetrahydrocannabinol does | Chemistry and stability | Decarboxylates on heating |
| Cannabidiolic acid | No | Precursor, distinct from cannabidiol | Preclinical only | Decarboxylates on heating |
| Delta-8-tetrahydrocannabinol | Yes | Type 1 agonist, weaker than delta-9 | Molecule characterised; commercial material not | Not documented |
The blank boiling points are deliberate: figures circulate widely, traceable measurements do not. The pattern is the finding: one compound with a real human evidence base and documented harms, one with a narrow regulator-reviewed indication and a large marketing halo, the rest thin.
If you take one habit from this article, make it a single question: in a dish, in a mouse, or in a person? When you next meet a cannabinoid claim, ask what species it was shown in and at what exposure. Which cannabinoid have you seen described in ways the evidence does not support?
This article is part of the Asiannabis Community educational series on medical cannabis. It is general information, not medical advice — treatment decisions belong with a licensed practitioner. Content is for educational purposes within jurisdictions where medical cannabis is legally permitted.