Essential Oil Chemotypes: Why "Pet-Safe" Isn't One Answer (2026)

Written by FUROMA Research Team · Last reviewed: July 14, 2026 · 13 min read

TL;DR

One plant name can cover seven chemically different oils. Thymus vulgaris has at least seven chemotypes — chemical variants of one species, set by an epistatic series of five genetic loci (Vernet et al., 1986). Measured head-to-head, common sage oil is 19.0% camphor; clary sage oil is 0.1% (Ben Akacha et al., 2023). This FUROMA guide explains what an essential oil chemotype is and why a species-level pet-safe verdict cannot resolve one.

Table of Contents

1. What is a chemotype, and why does one plant name cover several oils?
2. Why do chemotypes exist at all?
3. How much does the chemistry actually change?
4. Why can't the ASPCA database tell you if an oil is safe?
5. How do you read a chemotype name?
6. If the label says only "rosemary," what can you do?
7. Where FUROMA publishes chemotype — and where it doesn't

Introduction

FUROMA makes reed diffusers for homes with cats and dogs, and the single most common question we cannot answer with a yes or a no is "is rosemary safe?" The honest answer is that "rosemary" is not one oil. It is a plant name that covers several chemically distinct oils, and the difference between them is larger than the difference between some separate species. That variable is called the chemotype, and it is the reason two bottles labeled "pet-safe rosemary" can have almost nothing in common.

This guide is about that variable: what it is, why it exists, how big it gets, and — the part nobody writes down — why the safety databases everyone cites cannot see it.

What is a chemotype, and why does one plant name cover several oils?

A chemotype is a chemically distinct variant within a single plant species. Two plants can be genetically the same species, look identical, key out identically in a botanical field guide, and still produce essential oils with different dominant compounds. The species name describes the plant's form; the chemotype describes what its glandular trichomes actually make.

The classic case is common thyme. Granger and Passet (1973) established that wild Thymus vulgaris in France exists as six chemotypes, each named for the dominant monoterpene in its oil: geraniol, linalool, α-terpineol, thuyanol-4, thymol, and carvacrol. Their key finding was not that the variation existed — it was that the variation is stable. The chemotypes held their chemical character both in their native habitat and when transplanted into experimental cultivation, and they passed it to their offspring. This is not a plant responding to its soil. It is a heritable trait.

The count has since grown. Keefover-Ring et al. (2009) described a seventh Thymus vulgaris chemotype — a 1,8-cineole type in southern France that hydrodistillation had missed and ethanol extraction revealed. That detail matters more than it looks: a chemotype can be invisible to one analytical method and obvious to another, which is a preview of why "we tested it" is not a complete sentence.

A chemotype is a chemically distinct variant within one plant species: two plants can be the same species by every botanical criterion and still produce essential oils dominated by different compounds. Common thyme (Thymus vulgaris) has at least seven described chemotypes — geraniol, linalool, α-terpineol, thuyanol-4, thymol, carvacrol, and 1,8-cineole — because the trait is set by the plant's genetics rather than by its species identity, so the species name on a label identifies the plant and withholds the chemistry.

Why do chemotypes exist at all?

Chemotypes exist because monoterpene production in these plants is under direct genetic control, and the controlling genes sit in a biosynthetic chain.

Vernet, Gouyon and Valdeyron (1986) worked out the architecture in Thymus vulgaris: the dominant monoterpene is determined by an epistatic series of five loci, with a fixed order of dominance — geraniol → α-terpineol → thuyanol → linalool → carvacrol → thymol. Because the compounds are end products of branches off one shared pathway, a plant carrying a dominant allele early in the chain expresses that compound and masks the ones downstream. The chemotype is not a dial the grower sets. It is an inherited switch position.

Chemotype is inherited, not chosen: in Thymus vulgaris the dominant monoterpene is set by an epistatic series of five genetic loci with a fixed dominance order (geraniol → α-terpineol → thuyanol → linalool → carvacrol → thymol), because each compound is an end product of a branch off one shared biosynthetic pathway and a dominant allele early in the chain masks everything downstream (Vernet et al., 1986). This is why a grower cannot convert a thymol-type thyme field into a linalool-type one by changing the soil — the switch is in the genome, so sourcing a specific chemotype means sourcing specific plants.

Growing conditions still matter, but they act on which chemotypes survive where, not on what an individual plant makes. Thompson et al. (2013) demonstrated this at landscape scale. They resampled thyme populations in southern France that had been mapped in the early 1970s and compared them to 2009–2010 samples — a roughly 36-year interval. Phenolic chemotypes (thymol, carvacrol) are sensitive to winter freezing; non-phenolic types tolerate it. As extreme freezing events became less severe, the mean proportion of freezing-sensitive phenolic chemotypes rose from 47.7% to 53.1% across populations (P < 0.01), and of 13 populations that were exclusively non-phenolic in 1974, eight had become mixed.

Climate selects chemotypes rather than creating them: in southern France, thyme populations resampled after roughly 36 years showed the mean proportion of freezing-sensitive phenolic chemotypes rise from 47.7% to 53.1% (P < 0.01) as extreme winter freezing events weakened, and 8 of 13 populations that were exclusively non-phenolic in 1974 had become mixed (Thompson et al., 2013). Because winter kill removes the freezing-sensitive plants rather than converting them, a region's chemotype mix is the surviving population — which is why origin is evidence about a chemotype and not a guarantee of one.

This is the honest version of why origin appears on good labels. A region is not a chemistry. It is a filter that has historically produced a certain mix.

How much does the chemistry actually change?

Enough to invalidate a species-level verdict. Two peer-reviewed datasets give the size of the gap.

Rosemary. Satyal et al. (2017) characterized Rosmarinus officinalis across 78 oil compositions — six they analyzed plus 72 from the published literature — and resolved at least five distinct chemotypes. Across their six analyzed samples the ranges were: α-pinene 13.5%–37.7%, 1,8-cineole 16.1%–29.3%, verbenone 0.8%–16.9%, and camphor 0.7%–7.0%. Verbenone varies by more than twentyfold across oils that a label would all call "rosemary."

Sage. Ben Akacha et al. (2023) is the cleaner comparison, because both oils were run in the same lab by the same method. Common sage (Salvia officinalis) versus clary sage (Salvia sclarea):

Measured head-to-head in one laboratory by one method, common sage essential oil (Salvia officinalis) is 19.0% camphor and 12.9% α-thujone, while clary sage essential oil (Salvia sclarea) is 0.1% camphor with α-thujone not detected — roughly a 190-fold difference in camphor content, because the two are different species whose oils are built from different dominant compounds (Ben Akacha et al., 2023). A label that says only "sage" does not distinguish them, so it does not tell you which of these two chemistries is in the bottle.

Note what these two datasets are doing. The rosemary numbers are one species, several chemotypes. The sage numbers are two species sold under one common name. Both failure modes produce the same result on a label: the word tells you about the plant and withholds the chemistry.

Why can't the ASPCA database tell you if an oil is safe?

This is the section that costs us something to write, so we will be exact about it.

The ASPCA Toxic and Non-Toxic Plants database is the most-cited pet-safety reference on the internet, including by us. It is not wrong. It is answering a different question than the one a diffuser buyer is asking, and it has four structural resolution limits that anyone can verify in about ninety seconds.

1. It indexes plants, not oils. Look up rosemary and the entry reads: Rosmarinus officinalis, family Lamiaceae, Non-Toxic to Dogs, Non-Toxic to Cats, Non-Toxic to Horses (ASPCA, 2026). That verdict is about a cat chewing the shrub. It is not a clearance for a distilled oil, which concentrates the plant's volatile fraction by orders of magnitude and is a materially different exposure.

2. It resolves to species, not chemotype. The rosemary entry is one row. It cannot say "non-toxic as a plant, and by the way the cineole-camphor chemotype's oil is a different conversation from the verbenone chemotype's." There is no field for that, because chemotype is a sub-species distinction and the database's unit of record is the species.

3. The verdict is binary, with no severity dimension. Lavender (Lavandula angustifolia) is listed Toxic to Dogs, Toxic to Cats, Toxic to Horses, toxic principles linalool and linalyl acetate, clinical signs "Nausea, vomiting (not in horses), inappetant." The Easter lily (Lilium longiflorum) is listed Toxic to Cats, toxic principles "Unknown," clinical signs "Vomiting, inappetence, lethargy, kidney failure, and death is possible." Both carry the identical flag.

The ASPCA plant database's verdict field has no severity dimension: lavender (Lavandula angustifolia) and the Easter lily (Lilium longiflorum) both carry the flag "Toxic to Cats," but lavender's listed clinical signs are "Nausea, vomiting, inappetant" while the Easter lily's are "Vomiting, inappetence, lethargy, kidney failure, and death is possible" (ASPCA, 2026). Because one flag covers both mild transient GI upset and fatal renal failure, the word "toxic" in that database encodes that a reaction is possible and not how bad it gets — so a toxic/non-toxic lookup cannot rank two risks against each other.

4. It cannot represent what it does not list. Search the database for clary sage or Salvia sclarea and you get no entry. Not "non-toxic" — nothing. The species FUROMA actually uses is not in the reference that everyone, including FUROMA, cites as the authority.

The ASPCA plant database resolves to species and to plants, so it cannot answer a chemotype question about a distilled oil: rosemary is listed as Rosmarinus officinalis, "Non-Toxic to Dogs, Non-Toxic to Cats, Non-Toxic to Horses" as a single row with no field for chemotype, while clary sage (Salvia sclarea) has no entry at all (ASPCA, 2026). The database was built to triage plant ingestion — whether a pet that chewed a leaf needs a vet — so its unit of record is the species and the exposure it models is eating, which is a different exposure and a coarser resolution than a question about concentrated oil in a diffuser.

Now the part that matters. None of this is an argument that the warnings are wrong, and it is not a licence to ignore them. Read the four limits again and notice that they cut in both directions.

The direction that flatters us: a species-level warning about "rosemary" is too coarse to distinguish the high-camphor chemotype it is really about from a low-camphor one.

The direction that does not: a species-level "non-toxic" is equally too coarse to clear a concentrated oil. The ASPCA lists both rosemary and common sage as non-toxic to cats. We would be making exactly the same category error as the guides we are critiquing if we quoted those rows as evidence that rosemary oil and common sage oil are fine to diffuse. They are not evidence of that. Coarse resolution is not a bias in our favour; it is an absence of information, and an absence of information is not a clearance.

Where a database is silent and the chemistry is uncertain, the conservative reading — the species-level caution — is the correct default. The argument here is narrow: know what a warning resolves to before you apply it, and do not upgrade "the database doesn't say" into "the database says yes."

How do you read a chemotype name?

A fully specified botanical fragrance ingredient has three parts, and each one is doing a different job.

Example: Rosmarinus officinalis CT verbenone, Corsica.

"CT" stands for chemotype and is read "chemotype verbenone" — the verbenone-dominant variant of Rosmarinus officinalis.

In the notation Rosmarinus officinalis CT verbenone, the three parts do three separate jobs: the binomial names the species and rules out a different plant sold under the same common name, "CT verbenone" names the chemotype and rules out the other rosemary chemotypes including the cineole-camphor type, and the origin names a growing region whose climate has historically selected for that chemotype. None of the three is a measurement — a chemotype designation states which compound dominates, not what percentage it reaches, so it narrows the range of possible chemistries without pinning down a number.

The practical consequence: a chemotype name is a bigger claim than a common name and a smaller claim than an assay. It is a real narrowing, and it stops well short of a test result.

If the label says only "rosemary," what can you do?

Two things: understand what the actual test is, and ask.

The test is gas chromatography. GC separates an oil into its individual volatile compounds and, coupled to mass spectrometry (GC/MS), identifies each one and reports its relative percentage. A GC/MS report on an oil is what turns "rosemary" into "α-pinene 22.4%, 1,8-cineole 18.1%, verbenone 11.3%…" — which is to say, it is the only thing that establishes a chemotype rather than asserting one.

It is also how adulteration gets caught, and adulteration is not rare. Murphy et al. (2024) analyzed 30 commercially available fennel, star anise and anise essential oils by GC/MS and stable isotope ratio analysis and found 27% were adulterated — 8 of 30 — including dilution with carriers, substitution of cheaper botanical sources, and addition of synthetic (E)-anethole. Their point generalizes: the label is a claim, and only an analysis is a check.

Gas chromatography–mass spectrometry is what converts an essential oil's name into its chemistry, because it separates the oil into individual volatile compounds and reports each one's relative percentage — which is the only way a chemotype is established rather than asserted. It is also how substitution is detected: of 30 commercial fennel, star anise and anise essential oils analyzed by GC/MS and stable isotope ratio analysis, 27% were adulterated with carriers, cheaper botanical sources, or synthetic (E)-anethole (Murphy et al., 2024). A label states a claim; a chromatogram tests one.

Then ask the brand. Three questions, in this order, each of which has a checkable answer:

1. "What species is your sage?" A brand that answers "Salvia sclarea" has given you something falsifiable. A brand that answers "sage" has repeated the label.
2. "What chemotype is your rosemary, and where is it grown?" "CT verbenone, Corsica" is an answer. "Food grade" and "therapeutic grade" are not — neither term has a regulatory definition.
3. "Do you have a GC report for the batch?" The honest range of answers runs from "yes, here" to "we run GC on incoming oils but don't publish the traces" to silence. All three are informative. The last one most of all.

A brand that cannot name a species is not necessarily selling something dangerous. It is telling you it has not made a claim you can check — which is the same thing as asking you to take its word.

Where FUROMA publishes chemotype — and where it doesn't

Writing the section above obliges us to be precise about our own disclosure, including its edges.

What we publish, and exactly where. Our Pet Safe Philosophy page names our full plant library by botanical species — rose (Rosa damascena), jasmine (Jasminum grandiflorum), magnolia (Magnolia biondii), violet (Viola odorata), clary sage (Salvia sclarea), Canadian hemlock (Tsuga canadensis), Australian cypress (Callitris glaucophylla) — and where chemotype is the safety-relevant variable it names the chemotype and origin: our rosemary is Rosmarinus officinalis CT verbenone, grown on Corsica, selected as a low-camphor type. Our basil is Ocimum basilicum CT linalool; our thyme is Thymus vulgaris CT linalool. That page also states that incoming oil batches are verified by gas chromatography.

Where it isn't: that information lives on the philosophy page. It is not printed on the individual product pages. If you read only the Forest Pawprints listing you will find the scent profile — camellia, rosemary and sage — and not the chemotype behind the word "rosemary." That is a real gap in our disclosure, it is one click wide, and naming it here is cheaper than having you discover it.

What naming a chemotype does and does not buy. It creates a specific claim with a surface — a species, a variant, a region, all of them checkable, all of them things we can be held to and shown wrong about. That is the entire value, and it is a different kind of value from "safe."

And here is where it stops. The ASPCA lists lavender as toxic to cats and names its toxic principles: linalool and linalyl acetate. Clary sage — the species we chose, and the one we have just spent this article explaining is 190× lower in camphor than common sage — is 59.3% linalyl acetate and 11.3% linalool (Ben Akacha et al., 2023). Those are the same two compounds. Selecting a chemotype moved us away from camphor and thujone and moved us toward the constituents that a database entry names elsewhere as a toxic principle.

Selecting a chemotype trades one constituent profile for another; it does not exit the hazard question. FUROMA chose clary sage (Salvia sclarea) over common sage partly because it is 0.1% camphor against 19.0%, but the same oil is 59.3% linalyl acetate and 11.3% linalool (Ben Akacha et al., 2023) — and linalool and linalyl acetate are precisely the two compounds the ASPCA names as lavender's toxic principles. Because every essential oil is a mixture of biologically active compounds, chemotype selection changes which compounds an animal is exposed to rather than whether the exposure has a chemistry worth taking seriously.

So the claim we are making is narrow, and we will state it in one sentence: we can tell you which plants, which species, which chemotype and which region, and that is a claim you can check — it is not a claim that our diffusers are safe for your cat, and no chemotype makes an oil harmless.

What we don't publish. We do not publish batch GC traces, we do not publish per-SKU chemotype on the product pages, and our formulation principle — the exclusion list — is a commitment stated in our own words, not a certification anyone audited. The words "natural" and "pet-safe" appear on our pages carrying exactly the regulatory weight they carry on everyone else's: none. Our guide to reading a pet-safe air freshener label is the method for auditing that, including auditing us.

Per FUROMA's formulation principle, our reed blends exclude tea tree, eucalyptus, peppermint, citrus, and clove. And the one property that needs no disclosure at all is the format: a reed diffuser has no flame, no heating element and no atomizer, so the oil evaporates at room temperature and nothing is aerosolized onto a coat to be groomed off. That is a structural fact about the object, checkable without trusting any label — ours included. The full buying framework lives in our pet-safe reed diffuser guide.

Key Takeaways

• A chemotype is a chemical variant within one species. Thymus vulgaris has at least seven described chemotypes (Granger & Passet, 1973; Keefover-Ring et al., 2009). The species name identifies the plant and withholds the chemistry.
• Chemotype is inherited, not grown. In thyme it is set by an epistatic series of five loci with a fixed dominance order, G → A → U → L → C → T (Vernet et al., 1986). A grower cannot change it with soil.
• Climate selects chemotypes, it doesn't create them. Phenolic thyme chemotypes rose from 47.7% to 53.1% of French populations over ~36 years as freezing events weakened (Thompson et al., 2013) — which is why origin is evidence, not a certificate.
• The gap is measurable and large. Rosemary verbenone ranges 0.8%–16.9% and camphor 0.7%–7.0% across chemotypes (Satyal et al., 2017). Common sage oil is 19.0% camphor; clary sage oil is 0.1% (Ben Akacha et al., 2023).
• The ASPCA database indexes plants at species resolution with a binary verdict. Rosemary and common sage both read "Non-Toxic to Cats"; clary sage has no entry; lavender and the Easter lily share the flag "Toxic to Cats" despite one meaning nausea and the other possible kidney failure (ASPCA, 2026).
• Coarse resolution is not a clearance. The same bluntness that makes a "rosemary" warning over-broad makes a "non-toxic" row unable to clear a concentrated oil. Absence of information is not permission.
• A chemotype name is bigger than a common name and smaller than an assay — only gas chromatography establishes composition, and 27% of 30 commercial anise-family oils tested were adulterated (Murphy et al., 2024). Naming a chemotype buys falsifiability, not safety. Clary sage is 59.3% linalyl acetate and 11.3% linalool — the two compounds the ASPCA names as lavender's toxic principles. Chemotype selection trades constituents; it does not exit the question.

Frequently Asked Questions

What is an essential oil chemotype?

A chemotype is a chemically distinct variant within a single plant species. Two plants can be the same species and produce oils dominated by different compounds. Thymus vulgaris has at least seven described chemotypes, named for their dominant monoterpene — thymol, carvacrol, linalool, geraniol, α-terpineol, thuyanol-4, and 1,8-cineole.

Are chemotypes different plants or the same plant?

The same species. They key out identically in a botanical guide. What differs is which compound the plant's glandular trichomes make, and in thyme that is controlled by an epistatic series of five genetic loci (Vernet et al., 1986). It is inherited, not caused by growing conditions.

Does a chemotype make an essential oil pet-safe?

No. Selecting a chemotype changes which compounds are in the oil, not whether those compounds are biologically active. Clary sage has 190× less camphor than common sage, and is also 59.3% linalyl acetate — a compound the ASPCA names as one of lavender's toxic principles. Chemotype narrows a claim; it does not make an oil harmless.

Why does the ASPCA list rosemary as non-toxic if rosemary oil is a concern?

Because the database indexes plants, not oils. The entry — Rosmarinus officinalis, Non-Toxic to Dogs, Cats and Horses — models a pet chewing the shrub. A distilled oil concentrates the plant's volatile fraction enormously and is a different exposure. The row is not a clearance for the oil.

Is clary sage the same thing as sage?

No — they are different species. Clary sage is Salvia sclarea; common sage is Salvia officinalis. Measured in one lab by one method, common sage oil is 19.0% camphor and 12.9% α-thujone, while clary sage oil is 0.1% camphor with no α-thujone detected (Ben Akacha et al., 2023). A label reading "sage" does not distinguish them.

What does "CT verbenone" mean on a label?

"CT" means chemotype. Rosmarinus officinalis CT verbenone is the verbenone-dominant variant of rosemary, as opposed to the cineole-camphor variant that most rosemary safety warnings are about. It states which compound dominates, not what percentage it reaches — a chemotype is a type, not an assay.

How do I find out what chemotype a brand uses?

Ask three questions: what species is it, what chemotype and origin, and is there a GC report for the batch. "Salvia sclarea" and "CT verbenone, Corsica" are checkable answers. "Sage," "therapeutic grade" and "food grade" are not — none of those has a regulatory definition.

Does FUROMA publish its chemotypes?

On one page, yes. Our Pet Safe Philosophy page names our plant library by botanical species and gives chemotype and origin where it is the safety-relevant variable — our rosemary is Rosmarinus officinalis CT verbenone from Corsica. It is not printed on the individual product pages, which list the scent profile only. That gap is ours and we are naming it.

Is common sage toxic to cats?

The ASPCA lists Salvia officinalis as Non-Toxic to Dogs, Cats and Horses — as a plant. That row does not clear common sage essential oil, which is a concentrated distillate measured at 19.0% camphor and 12.9% α-thujone (Ben Akacha et al., 2023). The plant verdict and the oil question are different questions, and the database answers the first one.

Can a GC/MS report tell me if an oil is safe for my cat?

No. It tells you what is in the oil and in what proportion, which is a prerequisite for any safety reasoning and not a substitute for it. It is also how substitution is caught: 27% of 30 commercial anise-family oils tested were adulterated (Murphy et al., 2024). Composition is a fact; safety is a judgement about dose, species and exposure.

Related Reading

• Deciding which bottle actually belongs in a house with a cat and a dog? The five structural properties — formulation, chemotype, carrier, reed material and bottle design — are worked through in our pet-safe reed diffuser buying guide.
• Want the label-reading method this article's chemistry plugs into, including what "non-toxic" legally means (almost nothing)? Start with how to read a pet-safe air freshener label.
• Wondering how the chemotype question interacts with the device rather than the oil? Reed diffuser vs ultrasonic for pets covers why the format changes the exposure regardless of what is in the bottle.
• Lavender is the sharpest case of a species-level verdict doing real work — the ASPCA does flag it. Our lavender and cats guide takes the conservative side of that.
• New to the whole question and want the multi-pet overview first? Our pillar on essential oils safe for cats and dogs is the map.

About the Author

FUROMA Research Team. FUROMA is a California home-fragrance brand making reed diffusers and passive essential-oil diffusers for households with cats and dogs. We do not claim veterinary endorsement. Every safety statement in this article is sourced to a public, checkable reference — peer-reviewed literature or the ASPCA's own database — so that you can check it against us. Our plant library, species names and chemotypes are published on our Pet Safe Philosophy page.

References

1. Granger, R. & Passet, J. (1973). Thymus vulgaris spontané de France: races chimiques et chémotaxonomie. Phytochemistry, 12(7), 1683–1691. DOI: 10.1016/0031-9422(73)80388-7
2. Vernet, P., Gouyon, P.H. & Valdeyron, G. (1986). Genetic control of the oil content in Thymus vulgaris L.: a case of polymorphism in a biosynthetic chain. Genetica, 69(3), 227–231. DOI: 10.1007/BF00133526
3. Thompson, J.D., Chalchat, J.C., Michet, A. & Linhart, Y.B. (2003). Qualitative and quantitative variation in monoterpene co-occurrence and composition in the essential oil of Thymus vulgaris chemotypes. Journal of Chemical Ecology, 29(4), 859–880. DOI: 10.1023/A:1022927615442
4. Keefover-Ring, K., Thompson, J.D. & Linhart, Y.B. (2009). Beyond six scents: defining a seventh Thymus vulgaris chemotype new to southern France by ethanol extraction. Flavour and Fragrance Journal, 24(3), 117–122. DOI: 10.1002/ffj.1921
5. Thompson, J.D., Charpentier, A., Bouguet, G., Charmasson, F., Roset, S., Buatois, B., et al. (2013). Evolution of a genetic polymorphism with climate change in a Mediterranean landscape. Proceedings of the National Academy of Sciences, 110(8), 2893–2897. DOI: 10.1073/pnas.1215833110
6. Satyal, P., Jones, T.H., Lopez, E.M., McFeeters, R.L., Ali, N.A.A., Mansi, I., Al-kaf, A.G. & Setzer, W.N. (2017). Chemotypic characterization and biological activity of Rosmarinus officinalis. Foods, 6(3), 20. DOI: 10.3390/foods6030020
7. Ben Akacha, B., Ben Hsouna, A., Generalić Mekinić, I., Ben Belgacem, A., Ben Saad, R., Mnif, W., Kačániová, M. & Garzoli, S. (2023). Salvia officinalis L. and Salvia sclarea essential oils: chemical composition, biological activities and preservative effects against Listeria monocytogenes inoculated into minced beef meat. Plants, 12(19), 3385. DOI: 10.3390/plants12193385. PMID: 37836125
8. Murphy, B.J., Wilson, T.M., Ziebarth, E.A., Bowerbank, C.R. & Carlson, R.E. (2024). Authentication of fennel, star anise, and anise essential oils by gas chromatography (GC/MS) and stable isotope ratio (GC/IRMS) analyses. Plants, 13(2), 214. DOI: 10.3390/plants13020214
9. ASPCA Animal Poison Control Center. Toxic and Non-Toxic Plants Database — Rosemary (Rosmarinus officinalis), Sage (Salvia officinalis), Lavender (Lavandula angustifolia), Easter Lily (Lilium longiflorum). Retrieved July 14, 2026.
10. FUROMA. Pet Safe Philosophy. https://furoma.online/pages/pet-safe-philosophy. Retrieved July 14, 2026.