The Eleven Structural Classes

What "type" means, and what it does not mean

A phytocannabinoid type is defined by its carbon skeleton and its ring arrangement — how many rings there are, which atoms close them, where the oxygens sit. It is a structural classification, and it carries no pharmacological commitment. Two facts make this concrete. The CBD type and the Δ9-THC type contain molecules of identical molecular formula that differ only in whether one ring is closed, and they sit at opposite ends of the intoxication question. Conversely, a single type spans an enormous pharmacological range: the Δ9-THC type contains the propyl homologue that behaves as a CB1 antagonist and the heptyl homologue that binds CB1 more than thirty times as tightly as Δ9-THC itself. Reading "type" as "class of effect" is the commonest error in the secondary literature, and it is the error that makes novel-cannabinoid marketing copy sound authoritative when it is not.

• The type is the skeleton: ring count, ring closure, oxidation state of the rings.
• The side chain is a separate and orthogonal axis (see the side-chain series page).
• The double-bond position and the stereochemistry are a third axis (see the isomers page).
• Any given molecule is one point in that three-axis space, plus its acid or neutral form.

Sources: Hanuš LO 2016 · Radwan MM 2021 · ElSohly MA 2005

The class table

The table below expands the operator source table, keeps its column sense (class, representative members, side-chain lengths present, defining structural feature), and corrects it in three places. The Δ9-THC type row gains the C1, C4 and C6 homologues that have since been isolated. The Δ8-THC type row gains the octyl homologue, because the compound the trade calls THCJD is a Δ8 octyl species, which puts it in this row rather than the Δ9 row. The CBG row is corrected: CBGA is the acid of CBG and not a separate side-chain variant, so it belongs in the members column rather than the chain column.

Sources: Radwan MM 2021 · Hanuš LO 2016 · ElSohly MA 2005 · Van Kush Family Research Institute 2026* · Martin BR 1999

How many phytocannabinoids are there? The counts disagree, and the popular figure is the highest one contested

The figure of more than 180 identified phytocannabinoids is widely repeated, including in the operator source document that this shelf is built from. The taxonomy reviews give lower numbers. Radwan and colleagues, reviewing the field in 2021 from the same Mississippi group that produced the earlier inventory, report 125 compounds isolated or identified as cannabinoids and classify them into the eleven types used above; the same review notes more than 500 total compounds reported from the plant, of which the cannabinoids are one class among terpenes, phenolics and alkaloids. ElSohly and Slade in 2005 counted about 70. Hanuš and colleagues in 2016 performed a deliberately critical inventory and found that a meaningful fraction of published identifications did not survive scrutiny — some were misassignments, some were artefacts of the analysis, some were duplicates under different names. The honest statement is therefore a range with a method attached: on the order of 120 to 150 compounds have been isolated and structurally characterised as phytocannabinoids, the number grows as detection limits fall, and a figure above 180 has not been located in a taxonomy review in this pass. Anyone quoting a single round number should be asked which review it came from.

Contested — caveat. The "more than 180 identified phytocannabinoids" figure appears in the operator source and throughout the trade press but was not found in any of the taxonomy reviews consulted here. Radwan 2021 gives 125; ElSohly and Slade 2005 give about 70; Hanuš 2016 reduces several published identifications. Treat 180-plus as an uncited trade figure.

Sources: Radwan MM 2021 · Hanuš LO 2016 · ElSohly MA 2005 · Van Kush Family Research Institute 2026*

The biogenetic definition, and why cannabis is not the only source

The structural taxonomy above is descriptive. Hanuš and colleagues proposed a biogenetic definition alongside it: a phytocannabinoid is an isoprenylated resorcinyl polyketide, which splits naturally into an alkyl version (the cannabis compounds, where the resorcinol carries a straight alkyl chain) and a beta-aralkyl version (where it carries an aryl-substituted chain). That definition does real work, because it makes the family a chemotaxonomic class rather than a species-specific one. Compounds meeting it occur outside Cannabis — in liverworts of the genus Radula, which produce a cis-configured bibenzyl cannabinoid, in certain higher plants such as Helichrysum and Rhododendron, and in some fungi. For a hemp-sciences programme this matters practically: a reference standard, a chromatographic method and a receptor assay developed for a cannabis cannabinoid transfer directly to these other sources, and the non-cannabis phytocannabinoids sit outside cannabis-specific regulation while sharing the pharmacophore.

Sources: Hanuš LO 2016 · Kinghorn AD 2017*

Acid and neutral forms are not two classes

Every type above has an acidic form in the living plant and a neutral form after decarboxylation. The taxonomy counts them separately as distinct compounds, which is correct — THCA and THC have different masses, different chromatographic behaviour, different receptor pharmacology and different legal treatment — but it inflates the impression of structural diversity. Structurally, the acid and the neutral form differ by one carboxyl group on the aromatic ring. It is worth keeping the two facts in view at once: the acids are the real plant products and the neutral forms are largely post-harvest, and yet almost all of the receptor pharmacology in the literature was done on the neutral forms.

Sources: Sirikantaramas S 2004 · Wang M 2016 · Nadal X 2017