Lavender: One Genus, One Oil, One Famous Molecule

One word, a whole genus

How many plants hide behind the word lavender? More than a bunch of dried stems on a market stall would suggest. Lavender is the common name for a group of woody sub-shrubs in the genus Lavandula, and that genus belongs to the mint family. Botanically, then, the plant keeps company with rosemary, sage, thyme, oregano and basil. You can see the family resemblance without a microscope: squarish stems, leaves set in opposite pairs, and aromatic oil held in small surface glands rather than buried in the wood.

The growth habit is easy to read. A low, branching base that goes grey and woody as the plant ages. Narrow leaves in a dusty grey-green. Flower spikes carried above the foliage on bare stalks. Most of the oil sits in and around those spikes, which is why harvest means cutting the flowering tops and leaving the bush in the ground for the next season.

Then comes the part that catches people out. Lavender is not one species but several, plus hybrids, and they do not analyse the same. Spike lavender, Lavandula latifolia, is the clearest example. Its oil comes back from the laboratory with a higher share of camphor and 1,8-cineole, and a lower ester content, than the material sold simply as lavender oil. Same common name in casual speech. Different numbers on the certificate.

That is the practical reason botanists insist on Latin. A Latin binomial fixes which plant was cut, which means the analysis attached to it can be compared with another analysis. A trade name fixes almost nothing. Anyone who has bought two bottles labelled lavender and found them behaving differently in a formulation has met this problem already. It is not mystery. It is species, cultivar and growing conditions, written down or not written down.

• Family first. Lavandula sits in the mint family, alongside a long list of culinary herbs that also store volatile oil in surface glands.
• A shrub, not a soft herb. The base is woody and perennial, and the flowering stems are the part that carries the oil worth distilling.
• Lavandula latifolia is its own case. Known as spike lavender, it shows more camphor and 1,8-cineole and a lower ester share than standard lavender material.
• Hybrids exist and are grown at scale. Crosses within the genus, cultivated commercially as lavandin, are a separate line of material again and should be read under their own name.
• Cut the spikes. Because the oil glands cluster on the flowering tops, harvest timing sits close to the chemistry of the finished distillate.
• Latin beats marketing. A species name plus a batch analysis tells you what you have; the word lavender on its own does not.

Steam, flower heads and a list of names

Lavender oil is not lavender. It is a fraction of it. Steam runs through the cut flowering tops, carries the volatile compounds with it, and the condensed mixture separates into water and a thin layer of oil. Everything that will not travel with steam stays in the still: the fibre, the pigments, the water-soluble material. What you end up with is a concentrated set of small, light molecules, and the yield from a heap of plant material is modest.

Run that oil through a gas chromatograph and you get a list. Two names usually head it: linalool, a monoterpene alcohol, and linalyl acetate, which is the ester formed from linalool and acetic acid. Terpene chemistry is tidy that way. An alcohol and its acetate are close relatives, and the ratio between them is one of the numbers analysts watch.

Below that pair, the same handful of names keeps appearing in lavender oil analyses, in proportions that vary from sample to sample: camphor, borneol, 1,8-cineole, terpinen-4-ol, lavandulyl acetate, ocimene and beta-caryophyllene. Each belongs to a chemical class, and the classes are worth knowing because they explain why the list is stable while the percentages are not. Camphor is a ketone. Borneol and terpinen-4-ol are alcohols. 1,8-cineole is an oxide, sometimes written as eucalyptol in older literature. Lavandulyl acetate is a second ester in the profile. Ocimene is a straight-chain terpene hydrocarbon with no ring at all. Beta-caryophyllene is the odd one out by size: a sesquiterpene, built from fifteen carbon atoms rather than the ten that make up a monoterpene, which makes it heavier and slower to evaporate.

Why do the proportions move? Because a distillate records the growing season as much as the species. Cultivar, altitude, soil, the weather in the weeks before harvest, the day the field was cut, how long the still ran and which part of the run was collected all leave a mark on the final ratios. Spike lavender demonstrates the point at species level with its higher camphor and 1,8-cineole and lower ester content, but the same drift happens within a single species, field by field and year by year.

This is exactly why anyone working with botanical material asks for the analysis rather than the adjective. A number for the ester share, a number for camphor, a species name at the top of the page. That is a document you can compare. A description on a front label is not. Since 2014 our own working standard has been the same on the hemp side: know what is in the bottle, batch by batch, and publish it.

Reading a lavender oil report, point by point

1. Find the species. Look for the Latin binomial before anything else. Lavandula latifolia on the page means spike lavender, and that carries an expectation of more camphor and 1,8-cineole and less ester than generic lavender material.
2. Check what was distilled. The oil comes from the flowering tops, where the glands sit. A report that names the plant part tells you the material was handled with the chemistry in mind.
3. Note the extraction route. Steam distillation captures the volatile fraction and leaves the rest in the still, so the profile you are reading is a subset of the plant, not the whole plant.
4. Read the two lead compounds. Linalool and linalyl acetate usually sit at the top of a lavender analysis, and the ratio between an alcohol and its acetate is one of the routine markers analysts compare between batches.
5. Look at the ester line. Esters in this profile include linalyl acetate and lavandulyl acetate. A low total ester figure is a real distinguishing feature rather than a fault, and it is characteristic of spike lavender.
6. Look at camphor and 1,8-cineole together. These two rise as a pair in Lavandula latifolia material, which is the single clearest chemical signature separating it from its relatives.
7. Scan the alcohols. Borneol and terpinen-4-ol appear regularly in lavender oil analyses in variable amounts, and both are monoterpene alcohols, chemically not far from linalool itself.
8. Find the hydrocarbon. Ocimene is a straight-chain monoterpene with no ring, and it belongs to the standard cast of a lavender profile even when the percentage is small.
9. Spot the sesquiterpene. Beta-caryophyllene is bigger than the rest at fifteen carbons, which is why it is grouped separately from the ten-carbon monoterpenes on a well-organised report.
10. Accept the variability. The list of names in lavender oil is fairly consistent; the proportions are not, and any analysis is a snapshot of one batch from one harvest.
11. Compare like with like. Two reports are only worth setting side by side when both state the species, the plant part and the method, which is the whole argument for keeping documentation with the material.

Names, categories and dates in one overview

Where lavender meets hemp

Linalool is the reason this page exists in a cannabinoid context at all. It is not a lavender molecule in any exclusive sense. Plants that share no close family tie build the same terpenes, because the biochemical route to them is old and widespread, and linalool is one of the compounds that shows up on both a lavender oil analysis and a hemp terpene profile. Beta-caryophyllene does the same thing, appearing in the lavender constituent list and in hemp analyses alike. The plants are unrelated. The molecules are identical.

That overlap is what the 2011 review by Russo works with. It sets cannabis terpenes next to cannabinoids and asks whether the combined composition explains the character of a whole-plant extract compared with an isolated compound, and the name it puts on that idea is the entourage effect [1]. Worth being clear about the status. Russo presents it as a hypothesis and a research agenda, not as a demonstrated conclusion, and that is still where it sits [1]. We think saying so plainly is more useful than borrowing the term as a slogan.

Meanwhile, lavender has a documented working life that has nothing to do with cannabinoids. It has served as an anchor inside eau de cologne compositions, and it is one of three pillars of the fougere accord, together with coumarin and a mossy-woody base. That accord has a birth date: Houbigant's Fougere Royale, 1882. More than a century later it remains the structure under a large share of masculine fragrance. Not a curiosity. A load-bearing part of the industry.

• The species name is the first useful fact. If a report says Lavandula latifolia, expect the camphor and 1,8-cineole to sit higher and the esters lower, because that is the recorded signature of spike lavender.
• Shared molecules are not shared plants. Linalool and beta-caryophyllene appear in both lavender oil and hemp analyses, which tells you about biochemistry rather than about botanical relatedness.
• Read the entourage effect as an open question. Russo's 2011 review names it and argues it deserves study, and it has not moved from hypothesis to settled conclusion since [1].
• Percentages travel with the batch. Because the proportions in a distillate shift with harvest and distillation, the only number worth trusting is the one measured on the material in front of you.
• Categories help you read a list. Sorting constituents into alcohols, esters, a ketone, an oxide, a hydrocarbon and a sesquiterpene turns a wall of names into something you can actually compare across reports.
• Documentation is the whole point. Working with cannabinoids since 2014 has taught us the same lesson botanical suppliers learned long before: publish the analysis, and let the reader check it.