Type I, II And III: The Chemotype System Scientists Actually Use

The number printed on a jar of legal weed is a percentage of THC. That number is half of something. The other half, the CBD figure, usually sits at 0.1 percent or below and nobody reads it, which is a shame, because the relationship between those two numbers is the basis of the only classification scheme for cannabis that has ever survived contact with a laboratory.

Researchers do not sort cannabis into indica and sativa. They sort it by cannabinoid ratio, into groups they call chemotypes, and they have been doing so since 1973.

The three types, and the two nobody sells

What people mean by Type I and Type II and Type III cannabis is a position on one axis and nothing else. A cannabis chemotype is defined by the ratio between THC and CBD in the flower, and the whole system fits in one line. THC dominant strains have a ratio of THC/CBD > 1, intermediate strains sit at roughly 1, and CBD dominant strains fall below 1, according to a 2021 chemotaxonomic study in Frontiers in Plant Science. That gives the three groups everyone argues about:

• Type I is THC dominant. Almost everything sold in an American dispensary is Type I.
• Type II is intermediate, with THC and CBD close to a one to one ratio.
• Type III is CBD dominant, which is the legal definition of industrial hemp in most of the world.

There are two more. A 2022 paper in the journal Plants describes five recognized chemotypes in total, adding Type IV, plants whose main cannabinoid is CBG instead of THC or CBD, and Type V, fiber-type plants largely devoid of cannabinoids altogether. Type V exists, it is real, and nobody has ever tried to sell it to you.

Notice what the chemotype classification does not contain. There is no indica category, no sativa category, no hybrid category, no place for terpenes and no place for a name. It sorts on chemistry, and only on the two cannabinoids a lab can measure cheaply.

Where the three types came from

The framework traces to a short paper Ernest Small and H.D. Beckstead published in Nature in September 1973, which proposed that drug and non-drug cannabis could be separated on their relative content of THC and CBD. The Roman numerals came later, in the literature that followed, and the shorthand stuck because it was useful.

What made it stick was that the division held up. Fifty years of analytical work kept finding the same three clusters in real plant material, in wildly different growing conditions, on several continents. A classification scheme that keeps reappearing when nobody is trying to produce it is usually describing something real.

One gene doing most of the work

The reason the three types are so clean is that they are largely the output of a single genetic switch.

A 2003 paper in the journal Genetics crossed plants across chemotypes and found that a segregation of the three chemotypes fitting a 1:2:1 proportion appeared in every case. The authors proposed a model involving one locus, B, with two codominant alleles, and that model is still the backbone of how breeders think about cannabinoid ratio.

Read that 1:2:1 closely, because it explains an entire missing product category. A Type II plant carries one copy of each allele. Cross two of them and a quarter of the seedlings come out Type I, half come out Type II, and a quarter come out Type III. Type II cannot breed true. Every generation throws the same lottery, which is why balanced seed lines are difficult to stabilise and why Type II flower is scarce on shop shelves even where there is real demand for it.

The full picture is messier than one gene. Later genome work found the synthase genes sitting inside large, repetitive, awkward stretches of DNA and varying in copy number between cultivars. The 1:2:1 result still holds at the level of the flower, which is what matters to anyone choosing seed.

Why researchers stopped saying strain

Chemotype describes the ratio. Chemovar describes the whole chemical variety, cannabinoids and terpenes together, and chemovar classification is what most current researchers use in place of a name.

Ethan Russo put the objection to strain names bluntly in a 2019 Frontiers paper, writing that while cannabis types are labelled strains in common parlance, this term is acceptable with respect to bacteria and viruses, but not among Plantae. His conclusion was that the only workable solution is to characterise plants by their biochemical and pharmacological properties, and to call them chemical varieties, or chemovars.

The practical difference is small but real. A chemotype is a bucket, and there are only five of them. A chemovar is a profile, and there are as many as there are distinct plants. A strain name is a marketing decision.

The unexamined day is a wasted opportunity. Reflect on what you did, what you learned, and how you can improve.

John Dewey

What ninety thousand samples showed

The obvious test is whether names track chemistry, and somebody ran it. A 2022 study of commercial cannabis across six US states found that labels like indica, sativa and hybrid tell consumers little about what is in their product, based on nearly 90,000 samples, according to the University of Colorado Boulder announcement of the work. The same analysis found the samples falling into a small number of chemical clusters that the commercial labels did not line up with.

A separate Nevada study made the point even harder, reporting that the three chemovars it identified in that state's dispensary supply carried 396 different breeder-reported names between them.

None of which makes a seed catalogue dishonest. It makes the name a poor unit of measurement. Look at what our own pages report and the chemotype is plain in the numbers:

• Bruce Banner Auto lists 25 percent THC and no CBD figure at all, because there is nothing there worth printing.
• Durban Poison, stabilized from South African landrace genetics and about as far from Bruce Banner as the catalogue goes, lists 27 percent THC and the same silence on CBD.

Two plants from opposite ends of the genetic map, one chemotype. That is what Type I means.

Selecting for ratio starts before flowering

Growers tend to assume cannabinoid ratio is something you find out at harvest. It is not, and this is the part of the subject where breeding experience beats theory.

A 2018 study across a diverse cannabis germplasm collection found that chemotype at the vegetative and flowering stages was found to be predictive of chemotype at maturation, which means cannabinoid composition shows very little plasticity. The same abstract notes that accurate early screening of germplasm can accelerate selection.

Barney's Farm has been selecting on that principle for four decades, long before anyone published the confirmation. A plant's ratio is fixed by its genetics and shows up early, so the selection work that decides what a seed line will be happens in the first weeks, not in the drying room. Environment moves the total, sometimes a great deal. It does not move the ratio. A Type I plant grown badly is a weak Type I plant, never a Type III one.

That is also why the CBD-rich seed market is a separate breeding programme rather than a setting. You cannot grow a CBD dominant chemotype out of Type I seed, whatever the light schedule.

Ratios beat names

The chemotype system survives because it measures something the plant actually does. Five groups, defined by the relationship between two cannabinoids, set by a genetic switch that shows its hand within weeks of germination and barely shifts afterwards.

Names do a different job. They tell you who bred a plant, roughly what it tastes like and what the last person to grow it thought of it, and none of that is worthless. None of it is chemistry, and treating it as chemistry is how a shopper ends up disappointed by a jar that matched every expectation except the one that mattered. Read the ratio first and the name second. Barney's Farm has been developing premium cannabis genetics since the 1980s, with over 40 Cannabis Cup wins. Explore our full seed catalog and find strains bred for every climate and skill level.