Terpenes are the most oversold subject in cannabis. They are also genuinely interesting. The two facts are related: the interesting parts were stretched into claims the evidence does not support.
This post covers what terpenes are, why the plant makes them, how much survives drying, and what each major terpene has actually been shown to do. Receptor-level tests of the “entourage effect” are covered in the entourage-effect topic (bai69). The terpene hub is topic 321.
What terpenes are
Terpenes are volatile aromatic compounds built from five-carbon isoprene units. Monoterpenes (C₁₀), such as myrcene, limonene and pinene, are small and evaporate easily. Sesquiterpenes (C₁₅), such as β-caryophyllene and humulene, are heavier and last longer. Cannabis makes them in the same glandular trichomes that make cannabinoids. Monoterpenes are made in the plastids (MEP pathway) and sesquiterpenes in the cytoplasm (mevalonate pathway), using a large family of terpene synthase genes. More than a hundred terpenes have been reported in cannabis.
Terpenes supply most of the plant’s smell and much of its taste, but not all of it. Cannabinoids are essentially odourless. Since 2021, much of the “skunk” and “gas” character has been traced to volatile sulfur compounds, mainly 3-methyl-2-butene-1-thiol, which is not a terpene (Oswald et al. 2021). When one cultivar smells of citrus and another of diesel, the difference comes partly from terpenes and partly from sulfur compounds.
Why the plant makes them
Terpenes are not there for us. They are the plant’s chemical defence and signalling system. Some repel herbivores or harm insects. Some attract the predators of those insects, and others attract pollinators. That is why terpene output varies with genetics, environment, ripeness and stress.
The popular idea that heat stress “boosts terps” is not supported by consistent evidence. Studies of stress and terpene output give mixed results. What is not in doubt is that high temperatures speed up the evaporation of monoterpenes, both from the living plant and after harvest. For growers in Thailand, where heat is a constant problem, cooling the grow space and the drying room protects aroma. Heating plants on purpose to improve aroma is not a reliable technique.
Volatility: what drying actually removes
The best-known measurement is Ross & ElSohly (1996). They steam-distilled the oil from fresh indoor-grown buds and measured it as volume per fresh weight:
- Fresh buds gave about 0.29% oil.
- After one week of air-drying this fell to 0.20%, a loss of roughly 31%.
- After three months it fell to 0.13%, a loss of roughly 55%.
- Monoterpenes dropped from about 92% of the oil to about 62%, but no major component disappeared completely.
So drying causes a large but partial loss, mostly of the fresh, bright monoterpenes. It does not remove nearly everything, as is sometimes claimed. Steam-distillation yields also understate the total. Lab certificates (COAs) for dried flower today commonly report roughly 1–4% total terpenes by weight, and some batches come in below 1%. Figures from different methods (distilled oil vs. GC on solvent extracts), or from fresh vs. dried material, cannot be compared directly.
This explains why fresh-frozen (“live”) material keeps more monoterpenes, why stored flower loses its top notes over months, and why heat at any processing step costs aroma. Losses at each processing step are covered in topic 1363 and bai67.
Thai conditions. At 30+°C and 80–90% humidity, drying is a trade-off. It has to be fast enough to prevent mould but cool enough to keep monoterpenes. Open-air drying in Thai heat risks failing on both counts. The usual solution is a closed, dark room with air-conditioning and dehumidification. Details are in bai225.
The main terpenes: what is known
Most of the properties below come from isolated compounds tested in cell or animal studies, usually at doses far higher than flower contains.
| Terpene | Aroma | Reported activity | Evidence status |
|---|---|---|---|
| Myrcene | earthy, musky | Anti-inflammatory and pain-reducing effects; sedation at high injected doses | Animal only; sedation not shown in humans; the “mango” claim is unsupported |
| Limonene | citrus | Antimicrobial in vitro; reduced THC-induced anxiety when vaporised with THC | One small human RCT, in combination with THC (Spindle et al. 2024) |
| β-Caryophyllene | pepper, spice | Selective CB2 agonist; anti-inflammatory in mice via CB2 (Gertsch et al. 2008) | Receptor binding established; clinical effects not shown in humans |
| α-Pinene | pine | Antimicrobial in vitro; proposed to reduce THC-related memory impairment | Preclinical hypothesis; see note below |
| Linalool | floral, lavender | Oral lavender oil (Silexan) reduced anxiety in trials | Human evidence is for oral lavender oil, not inhaled cannabis |
| Terpinolene | fresh, piney | Reduced activity in mice after inhalation (Ito & Ito 2013) | Animal only |
| Humulene | earthy, hoppy | Anti-inflammatory in rodents (Fernandes et al. 2007) | Animal only; see topic 1349 |
| Ocimene | sweet, herbal | Limited data | See topic 1350 |
- Limonene. Spindle et al. (2024) ran a double-blind trial in 20 healthy adults, who inhaled vaporised THC with and without d-limonene. Limonene alone did not differ from placebo. Taken with THC, it reduced ratings of anxiety and paranoia. This is one small trial using purified doses. It does not show that “citrus” flower is calming.
- Linalool. Silexan at 80–160 mg/day did better than placebo in generalised anxiety disorder (Kasper et al. 2014). That was a fixed oral dose of lavender oil, so the link to linalool in cannabis is indirect.
- Pinene. Russo (2011) describes α-pinene as a bronchodilator in humans at low exposure, citing Falk et al. (1990). However, that small study of 8 volunteers reported no acute change in lung function. Treat the bronchodilator claim as unconfirmed.
- Terpinolene. In mice it had a sedating effect, the opposite of its “energising sativa terpene” label. This shows how poorly animal data match effect labels.
What a terpene profile cannot tell you
A GC-MS or GC-FID profile accurately shows which terpenes were in that sample when it was tested. It cannot tell you:
- what effect a given person will feel, because human dose-response data are missing for most terpenes;
- what the profile was at harvest, because drying, curing and storage change it;
- whether it matches the named “strain”, because growing conditions shift terpene expression, not just genetics;
- the cannabinoid content. Terpene and cannabinoid levels are separate traits. Some datasets show them rising together and others do not, so neither reliably predicts the other.
This is also why terpene-to-effect charts (“myrcene for sleep, limonene for uplift, pinene for focus”) are unreliable. They rarely give concentrations. They treat each terpene as if it were taken on its own. They often rely on fresh-plant profiles that drying has since changed. And direct receptor tests were negative (see bai69).
Thailand: effect claims based on terpenes (“calming”, “for sleep”, “anti-inflammatory”) on cannabis products, labels or ads can break Thai FDA health-claim and advertising rules. See topic 1293.
Isolated and botanical terpenes
A separate industry extracts terpenes from non-cannabis sources, such as citrus (limonene), hops and pine (pinene), and blends them into “strain profiles”. These blends are used to put aroma back into distillates and in vape cartridges. Importing and selling e-cigarettes and vape devices is banned in Thailand, so that use has no legal market here.
- Botanical vs. cannabis-derived. Limonene is the same molecule whether it comes from lemon peel or cannabis flower. “Cannabis-derived” describes the source, not a proven difference in effect.
- Enantiomers. (+)- and (−)-limonene are mirror-image forms with different smells and possibly different activity. Telling them apart requires chiral GC. Standard GC-MS cannot do it, and labels rarely say which form is present.
- “Strain” blends (for example “Blue Dream terpenes”) are formulations based on lab profiles. No one has validated how closely they match the original cultivar.
Quality and storage
In Thailand, cannabis flower is a controlled herb under the Notification of the Ministry of Public Health Re: Controlled Herbs (Cannabis) B.E. 2568. It may be dispensed only against a PT 33 prescription, and it may not be sold or advertised online. The points below are about quality control.
- Aroma is a useful quality check. A strong, clean aroma usually reflects careful drying and storage. A weak aroma can mean poor drying or age, or simply a low-terpene cultivar.
- Drying temperature and duration are the biggest factors you can control in what survives.
- Store material cool, dark, sealed and at controlled humidity. Warmth, light and air in a half-empty container drive off monoterpenes first.
- Potency and terpene profile are measured separately. Check both on the COA rather than guessing one from the other.
Terpenes are worth preserving, testing and describing, and their role in aroma is established. Claims about effects need hedged wording: “rodent studies suggest”, “one small human trial found”, “not confirmed in humans”.
Sources: Ross & ElSohly, J Nat Prod 1996;59:49–51; Oswald et al., ACS Omega 2021, doi:10.1021/acsomega.1c04196; Gertsch et al., PNAS 2008, doi:10.1073/pnas.0803601105; Spindle et al., Drug Alcohol Depend 2024;257:111267, PMID 38498958; Kasper et al., Int J Neuropsychopharmacol 2014, doi:10.1017/S1461145714000017; Russo, Br J Pharmacol 2011, doi:10.1111/j.1476-5381.2011.01238.x; Falk et al., Scand J Work Environ Health 1990, doi:10.5271/sjweh.1771; Ito & Ito, J Nat Med 2013, doi:10.1007/s11418-012-0732-1; Fernandes et al., Eur J Pharmacol 2007, PMID 17559833.