Bee Balm

What is Bee Balm?

Bee balm is the common name for plants in the genus Monarda, a group of about 25 accepted species in the mint family, Lamiaceae, endemic to North America. Linnaeus published the genus in Species Plantarum in 1753, on page 22, and named it for Nicolás Monardes, the Seville physician who catalogued New World plants between 1565 and 1574 without ever crossing the Atlantic. Three species carry almost all the commerce. Monarda didyma L., scarlet bee balm or Oswego tea, grows from Maine west to Ontario and Michigan and south to northern Georgia. Monarda fistulosa L., wild bergamot, has the widest range of the three, from Quebec to the Northwest Territories and British Columbia south to Georgia, Texas, Arizona and Idaho, with eight recognized varieties. Monarda punctata L., horsemint or spotted bee balm, has nine varieties and is chemically the odd one out. It is also the only Monarda that the FDA lists as generally recognized as safe as an essential oil, at 21 CFR 182.20, under the entry "Horsemint, Monarda punctata L."

Common Names

• Bee balm, beebalm, bergamot, wild bergamot, Oswego tea (Monarda didyma and Monarda fistulosa)

• Scarlet bee balm, crimson beebalm, scarlet monarda, Eau-de-Cologne plant (Monarda didyma)

• Horsemint, spotted bee balm, spotted horsemint (Monarda punctata)

• Lemon bee balm, lemon mint (Monarda citriodora, a fourth species sold mostly as a seed and culinary herb)

• Number Six, the Oneida name for Monarda fistulosa, recorded as the sixth medicine given by the Creator

Primary Active Compounds

• Thymol, a monoterpene phenol, the dominant constituent of most Monarda punctata oil at 47.9 to 61.8 percent and of many Monarda fistulosa populations at up to 77.3 percent

• Carvacrol, thymol's structural isomer, which reaches 71.5 percent in leaf oil from a Montana Monarda fistulosa population and 16.7 percent in a cultivated Italian Monarda didyma

• p-Cymene and gamma-terpinene, the hydrocarbon precursors to both phenols, together often 15 to 30 percent of the oil

• Thymoquinone and thymohydroquinone, oxidized thymol relatives, reported between 2.3 and 41.3 percent in Monarda fistulosa collections from South Alabama

• Geraniol and linalool, which dominate entirely separate Monarda fistulosa chemotypes, with geraniol reaching 86.8 percent in a Manitoba selection

• Rosmarinic acid, at 91.23 mg/g in a Monarda fistulosa dry extract, plus luteolin-7-O-glucoside at 76.30 mg/g

• Didymin, a flavanone glycoside also found in sweet orange, blood orange and mandarin, and the marker compound of the one Monarda extract that has been through a randomized trial

Key Note

There is no single "bee balm." The genus produces at least four distinct essential oil chemotypes, and two plants of the same species, both correctly labeled Monarda fistulosa, can differ more from each other than a thyme plant differs from an oregano plant. A 2021 survey by Lawson, Satyal and Setzer in Plants, volume 10, article 482, found thymol ranging from 8.9 to 62.2 percent across Monarda fistulosa samples taken in a single region of South Alabama. That is not a difference between farms or countries. That is the spread inside one state.

Bee balm is one of the very few North American medicinal plants with a continuous documented record from Indigenous practice through nineteenth century pharmacy into a 2025 clinical trial, and almost none of that record connects to what is on the shelf under the name "wild bergamot."

Primary Functions & Benefits

Monarda has three functional identities that get blurred together in retail copy, and they run on different molecules.

As an aromatic phenol source. Thymol and carvacrol are the reason horsemint was farmed. They disrupt bacterial and fungal membranes on contact, which is why thymol sits in Listerine Original Antiseptic at 0.064 percent alongside eucalyptol at 0.092 percent, methyl salicylate at 0.060 percent and menthol at 0.042 percent. This is a topical and surface function, demonstrated at the concentrations found in a mouth rinse or a distilled oil.

As a polyphenol source. The water-soluble fraction is a different plant entirely. Shanaida and colleagues, publishing in Frontiers in Pharmacology in 2021, article 563436, measured 120.64 mg gallic acid equivalents per gram of total phenolics in a Monarda fistulosa dry extract, with rosmarinic acid at 91.23 mg/g, caffeic acid at 21.62 mg/g, apigenin at 15.12 mg/g and chlorogenic acid at 11.34 mg/g. In rats, 200 mg/kg produced 31.91 percent inhibition of carrageenan-induced paw edema at three hours, and in mice the same dose gave 51.24 percent inhibition of acetic acid writhing.

As a sensory irritant. Ghosh and colleagues, in Molecules volume 25, article 4873, published in 2020, showed that Monarda fistulosa oil activates the TRPA1 ion channel at 20 micrograms per milliliter, with pure carvacrol and thymol activating at roughly 100 micromolar. p-Cymene, alpha-terpinene and gamma-terpinene did nothing. That channel activation is the warmth and the sting, and it is the plausible basis for the traditional use as a counterirritant rub in rheumatic pain.

Forms & Standardization

This is where the category falls apart, because almost nothing sold as bee balm is standardized to anything.

Loose dried herb and tea. The default retail form. Sold as Monarda fistulosa or Monarda didyma leaf and flower, with no marker compound declared, no chemotype stated, and frequently the wrong common name attached. Amazon listings exist that read "Wild Bergamot Petals, Scarlet Beebalm Petals (Monarda didyma)," which fuses the common name of one species with the Latin binomial of another in a single product title.

Steam-distilled essential oil. Yields are low and vary by species and harvest stage. Fraternale and colleagues, in PeerJ 2022, article e14433, recovered 2.55 mL per kilogram of dry weight from Monarda didyma grown at 450 meters near Urbino, Italy, harvested in October 2017. Wieczorek's group, in Molecules volume 31, article 338, published in 2026, reported 0.6 percent oil from the Monarda fistulosa cultivar 'Premiera' and 0.35 percent from the Monarda didyma by Monarda fistulosa hybrid 'Tonya', both harvested at the end of flowering.

Chemotype-declared oil. A small number of aromatherapy suppliers sell "Monarda fistulosa ct geraniol," usually of Canadian origin. This is the only honest way to sell the oil, because the same binomial covers a geraniol chemotype at 86.8 percent geraniol, a thymol chemotype at up to 77.3 percent thymol, and a carvacrol chemotype at 71.5 percent carvacrol. An undeclared bottle tells you nothing about which of those you bought.

Alcoholic tincture. Typically 1:3 or 1:5 in 40 to 60 percent ethanol. Ethanol pulls both the volatile phenols and the rosmarinic acid, but no vendor in this category publishes an assay, so the delivered thymol content is unknown to buyer and seller alike.

Standardized flavonoid extract. Exactly one product line matters here. The Monarda didyma extract used in the 2025 GeroScience trial was standardized on didymin as its predominant flavonoid, present at 20 to 80 times the level of other detected flavonoids, and formulated with 50 percent dextrins as a carrier. It was supplied by Mibelle Group Biochemistry. Note what that means: the only Monarda extract with a randomized human trial behind it is not an essential oil product, contains no meaningful thymol, and cannot be substituted with a bag of dried leaf.

Cosmetic hydroglyceric extract. Cornara and colleagues, in Plants volume 13, article 112, published in 2024, compared an ethanolic and a hydroglyceric Monarda didyma extract, identifying and quantifying 53 polyphenols by LC-DAD-ESI-MS. The hydroglyceric extract carried 105.75 plus or minus 5.91 mg of total phenols per 100 mL against 64.22 plus or minus 3.45 for the ethanolic, and gave anti-angiogenic IC50 values of 10.17 mg/mL against 20.65 mg/mL in the chick chorioallantoic membrane assay. Solvent choice changed the potency by roughly a factor of two.

Food Sources

Bee balm is a culinary herb and a garden plant before it is anything else, and the food route is genuinely the sensible one.

• Fresh Monarda didyma or Monarda fistulosa leaf, steeped as tea. This is the historical preparation, the one John Bartram found settlers using near Fort Oswego between 1743 and 1746, learned from the Haudenosaunee peoples of that region. A cup of infusion extracts a fraction of a percent of the leaf's oil, which is why it tastes of thyme and oregano without burning.

• Flower petals, used raw in salads and as a garnish. The colour in Monarda didyma comes from anthocyanins, present in the Trentino hydroglyceric extract at 3.01 mg/100 mL against 0.32 in the ethanolic version.

• Dried leaf as a seasoning for game and poultry, a documented Indigenous use across Blackfoot, Menominee, Ojibwe and Ho-Chunk practice.

• Commercial food use of the oil is limited to Monarda punctata, which is the only Monarda cleared under 21 CFR 182.20. Monarda didyma and Monarda fistulosa are not on that list.

Realistically sufficient? For the polyphenols, yes. A strong infusion of Monarda fistulosa leaf delivers rosmarinic acid and luteolin glycosides in the same class of amounts as a comparable rosemary or lemon balm tea. For the thymol, no, and that is a feature. Steam distillation concentrates the oil roughly two hundred to five hundred fold relative to the fresh herb. Hood's Florida field data give the scale: fresh entire horsemint herb yielded 0.12 to 0.20 percent oil, and even his best selected line reached only 0.44 percent. Everything the essential oil literature reports is happening at a concentration nobody reaches by drinking tea.

Who Should Take Bee Balm

Nobody needs bee balm. The honest list of people for whom it is a reasonable purchase is short and specific.

• People who want a caffeine-free aromatic tea with a real polyphenol load and like the flavour. This is the strongest case, and it is a food case.

• Gardeners and cooks who want a Lamiaceae herb that is native to North America, supports pollinators, and seasons game meat the way oregano seasons lamb.

• People buying a diluted topical for minor skin complaints, where the thymol and carvacrol content is the point and the exposure is external. Aromatherapy suppliers sell it for this, and the mechanism is at least the same one that put thymol in mouthwash.

• Participants in the narrow use case the 2025 GeroScience trial tested: adults aged 45 to 65 taking 100 mg per day of a didymin-standardized Monarda didyma extract for 12 weeks, in a single trial of 81 people, funded with material supplied by the extract's manufacturer. That is one study, and it is not a reason to buy dried leaf.

Nobody should be buying Monarda expecting anything the citrus bergamot literature reports. That includes LDL reduction, which has never been tested with any Monarda preparation in any human trial.

Who Should AVOID or Use Caution

Contraindications

• Undiluted internal use of any Monarda essential oil. This is a thymol and carvacrol concentrate, and the acute animal toxicity of thymol is real: reported oral LD50 values include 980 mg/kg in rats, 1,050 to 1,200 mg/kg in mice and 88 mg/kg in guinea pigs. The guinea pig figure is more than a tenfold species difference, which is itself a reason not to extrapolate casually.

• Pregnancy, for the essential oil and for concentrated extracts. Monarda punctata was used historically as an emmenagogue, and King's American Dispensatory of 1898 lists suppressed menses among the indications for oil of horsemint. There is no human safety data at any dose in pregnancy.

• Known allergy to Lamiaceae. Cross-reactivity with thyme, oregano, mint and sage is plausible on shared phenol chemistry, though not formally documented for Monarda.

Use Caution

• Anyone on anticoagulants or antiplatelets. Rosmarinic acid and the flavones in the water extract have antiplatelet activity in laboratory systems, and at 91.23 mg/g rosmarinic acid the Monarda fistulosa extract is a concentrated source.

• Infants and young children. Thymol-rich oils should not be applied near the face of a child under two, on the same reasoning that applies to menthol and eucalyptol preparations.

• People with reflux or gastritis. TRPA1 activation is the mechanism behind the warm burn, and a concentrated preparation reproduces it in the esophagus.

• Anyone with a history of photosensitivity who is buying because they read about bergamot. That risk belongs to Citrus bergamia peel oil and its bergapten content. Monarda has no furanocoumarins, so the caution here is against the substitution error, not against the plant.

Critical Safety Point

Chemotype is a safety variable, not a flavour note. A bottle labeled Monarda fistulosa may contain 3 percent thymol or 77 percent thymol, and 3 percent carvacrol or 71 percent carvacrol. A dilution ratio calculated for one bottle is meaningless for the next. Demand a gas chromatography report showing the actual percentages before using any Monarda oil on skin, and treat an unlabeled bottle as if it were the most concentrated version.

Recommended Dosages

There is no established therapeutic dose for bee balm in any pharmacopoeia. What follows is what has actually been used, with the source attached.

Tea, dried leaf and flower

1 to 2 teaspoons of dried herb, roughly 2 to 4 grams, per cup, steeped covered for 10 minutes. Covering matters, because the volatile phenols leave with the steam. This is the traditional preparation and carries the longest safety record.

Essential oil, topical, diluted

0.5 to 1 percent in a carrier for general use, which is 3 to 6 drops per 30 mL. Thymol-rich oils are usually held at or below 1 percent. RIFM's safety assessment for thymol, CAS 89-83-8, found no skin sensitization concern with human maximization testing at 2,760 micrograms per square centimeter, and derived a systemic NOAEL of 13.33 mg/kg/day for repeated dose toxicity.

Essential oil, internal

Not recommended, and no modern dose exists. The historical figure, from King's American Dispensatory in 1898, was 2 to 5 drops of oil of horsemint on sugar, or 10 to 30 drops of the essence in sweetened water. Reproduce that at your own risk with an oil whose thymol content you have not measured.

Tincture

30 to 60 drops, two to three times daily, is the range vendors print. No clinical work supports any specific number, and none of these products publish a marker assay.

Standardized didymin extract

100 mg per day, the dose used in the GeroScience trial by Campisi, Cannella, Pavanello and colleagues, published online 10 March 2025, in 81 adults aged 45 to 65.

Duration

The tea is a food and has no duration limit. The one controlled human trial ran 12 weeks. Topical thymol-rich oils should be used in short courses of a week or two rather than continuously, because sensitization risk to any aromatic phenol rises with cumulative exposure even where a single-exposure test is clean.

Timing & Administration

Tea is best taken between meals if the goal is the aromatic effect, or after meals if the goal is the traditional carminative one. King's lists horsemint as a carminative for flatulence and nausea, and Monarda fistulosa appears in Ojibwe and Menominee practice for the same complaint.

Cover the cup. This is not a decorative instruction. The compounds that give Monarda its character boil at 232 to 237 degrees Celsius as pure substances but are steam-volatile at brewing temperature, which is exactly how they are extracted industrially. An uncovered cup loses a measurable fraction of them to the room.

The didymin-standardized extract in the 2025 trial was taken once daily as a capsule with 50 percent dextrins as carrier. No timing effect was tested.

Topical applications should go on intact skin only. Thymol and carvacrol on broken skin produce burning out of proportion to the concentration, which is the TRPA1 channel doing exactly what Ghosh's group measured at 100 micromolar.

Timeline of Effects

Immediate, seconds to minutes. The aromatic and sensory effects. TRPA1 activation is not a slow process, and the warmth from a topical or the sharpness of a strong infusion arrives on first contact.

Hours. Thymol pharmacokinetics are known in humans, and they are unusual. Kohlert and colleagues, in the Journal of Clinical Pharmacology in 2002, volume 42, pages 731 to 737, gave a preparation containing 1.08 mg of thymol and measured a peak plasma concentration of 93.11 ng/mL at 1.97 hours, with a half-life of 10.2 hours. Systemic availability was around 16 percent.

Two to three hours. The anti-inflammatory window in the rodent work. Shanaida's 31.91 percent edema inhibition at 200 mg/kg was measured at the three-hour timepoint.

Twelve weeks. The only human trial duration on record. The GeroScience study measured leukocyte telomere length and DNA methylation age at baseline and at 84 days.

What counts as a fair trial. For the tea, one to two weeks tells you whether you like it and whether it settles your stomach. There is no slow-building effect to wait for. For the standardized extract, 12 weeks is the tested duration, and there is no basis for extending past that in the absence of a second trial. For a topical, 5 to 7 days. If a diluted Monarda oil has not helped a minor skin complaint in a week, more time will not fix it.

When to conclude it is not working. Immediately, if you bought it expecting a lipid effect. There is no Monarda trial with any lipid endpoint, and the trial that produced that expectation was run on a citrus fruit.

Benefits of Taking Bee Balm

Ranked by the strength of the evidence behind them, not by how they are advertised.

• Antimicrobial activity on surfaces and in vitro. Well established for the oil. Wieczorek's 2026 work in Molecules gave minimum inhibitory concentrations below 0.25 to 0.5 mg/mL for both 'Premiera' and 'Tonya' oils against bacteria and fungi. Li and colleagues, in the International Journal of Clinical and Experimental Pathology in 2014, volume 7, issue 11, pages 7389 to 7398, tested a Monarda punctata oil at 75.2 percent thymol against five respiratory pathogens and found Streptococcus pyogenes, Streptococcus pneumoniae and Escherichia coli most susceptible and MRSA most resistant.

• Antioxidant capacity in vitro. Consistent across preparations. The Monarda fistulosa dry extract gave a DPPH IC50 of 0.285 mg/mL. The Trentino hydroglyceric Monarda didyma extract gave IC50 values 5 to 60 times lower than its ethanolic counterpart across FRAP, DPPH, TEAC and ORAC.

• Anti-inflammatory and analgesic effect in rodents. Dose-dependent, at 50, 100 and 200 mg/kg, in two standard models. Not human data.

• Reduced markers of biological aging in one 12-week trial. In the GeroScience study, leukocyte telomere length rose in the treated group with a p value of 0.0537, which is not statistically significant on its own, while it fell significantly in placebo at p equals 0.0006. DNA methylation age held steady in treatment at p equals 0.4522 and rose in placebo at p below 0.0001. The between-group differences were significant at p equals 0.0041 and p equals 0.0162. Read carefully, the treated group mostly did not change and the placebo group deteriorated. That is a real and interesting result, and it is a single study with manufacturer-supplied material.

• Digestive and carminative use. Long traditional record across Oneida, Menominee, Ojibwe and Ho-Chunk practice and in nineteenth century eclectic pharmacy. No controlled trial.

• Wound and skin application. Documented Blackfoot use as an antiseptic poultice. Mechanistically plausible on thymol content. No controlled trial.

Potential Negatives & Side Effects

• Mouth and throat burning from strong infusions or from oil taken internally. This is the expected pharmacology, not an idiosyncratic reaction.

• Skin irritation from insufficiently diluted oil. The variable here is not the label percentage but the actual phenol content of the batch, which can differ by a factor of twenty between two bottles of the same species.

• Gastric discomfort. Aromatic phenols are gastric irritants at concentration, and this is the most common complaint in the traditional literature about oil of horsemint taken internally.

• Photosensitivity is not a Monarda problem. If you experience it, look at what else is in the product.

• Contact allergy is uncommon. RIFM concluded thymol presents no concern for skin sensitization, and no concern for photoirritation or photoallergenicity, on the basis that its molar absorption coefficients fall below the threshold of concern.

• The realistic downside for most buyers is not a side effect at all. It is paying for an unstandardized product on the strength of a body of research conducted on a citrus tree in Calabria.

Deficiency Symptoms

Bee balm has no deficiency state, because the human body has no requirement for thymol, carvacrol, rosmarinic acid or didymin. Nothing stops working if you never encounter Monarda. Compare that to iodine, where the adult recommended intake is 150 micrograms per day and sustained intake well under that figure produces measurable thyroid consequences through a defined biochemical mechanism. No such threshold exists for any Monarda constituent, and JECFA's evaluation of thymol as flavouring agent number 709 and carvacrol as number 710 treats them purely as exposures to be kept safe, not as intakes to be met. Estimated dietary thymol intake was 59 micrograms per day in Europe and 160 micrograms per day in the United States, both far under the 1,800 microgram structural class I threshold.

What bee balm addresses

Symptoms and preferences, not deficits. A tea that settles the stomach. A diluted oil for a minor skin complaint. A polyphenol-dense beverage for someone who would otherwise drink something with nothing in it. In one 12-week trial, a shift in two aging biomarkers driven largely by the placebo group getting worse.

Bottom line

Nothing in this genus is required. Every use is elective, and the case for it has to be made on evidence rather than on need.

Toxicity Symptoms

At high intake

Concentrated Monarda oil taken internally reproduces classic phenol toxicity: burning of the mouth, throat and stomach, nausea, vomiting, abdominal pain, then central nervous system depression at larger amounts. The animal reference points for thymol are oral LD50 values of 980 mg/kg in rats, 1,050 to 1,200 mg/kg in mice, and 88 mg/kg in guinea pigs. The spread across species is more than a tenfold range, which is a warning against arithmetic extrapolation to humans rather than a basis for it.

Signs to reduce or stop

Persistent burning after swallowing, nausea that does not clear, headache, dizziness, or any skin reaction that spreads beyond the application site. Stop the product, do not dilute it further and continue.

General note

The realistic hazard with Monarda is concentration error, not chronic accumulation. Thymol has a plasma half-life of 10.2 hours and is cleared as conjugates in urine, so it does not build up over weeks. What causes trouble is a single generous dose of an oil whose phenol content the user assumed was 10 percent when it was 70 percent. Shanaida's group gave a Monarda fistulosa dry extract to rats at 500 to 5,000 mg/kg orally with no adverse effects observed, which tells you the water-soluble fraction is not where the risk lives. The oil is.

How Bee Balm Works

The mechanisms are distinct and they belong to different fractions of the plant.

Membrane disruption by the phenols. Thymol and carvacrol are lipophilic monoterpene phenols that partition into bacterial and fungal membranes, increase permeability and collapse the proton gradient. This is a physical mechanism, dose-dependent on the actual concentration reaching the membrane, and it works reliably in a dish and on a surface.

The problem with reading those dish results as body results. Free thymol is essentially not detectable in human plasma. Kohlert's group found that what circulates after an oral dose is thymol sulfate, a conjugate formed before the molecule ever reaches the bloodstream, and thymol glucuronide appears only in urine. The compound tested in every antimicrobial assay is therefore not the compound present in your circulation. That is not a matter of the dose being too small. It is a matter of the sulfate conjugate being a different molecule with different membrane behaviour, and nobody has tested it against the organisms the free phenol kills. The membrane mechanism is real where the free phenol is present, meaning the mouth, the skin, the gut lumen and a jar of preserved cosmetic. It is unestablished anywhere downstream of the liver.

TRPA1 activation. Carvacrol and thymol open the TRPA1 channel at roughly 100 micromolar, and the whole essential oil does it at 20 micrograms per milliliter. This is a sensory nerve mechanism and it is local. It explains the counterirritant tradition, the warmth of a rub, and the sting of a concentrated preparation.

Polyphenol antioxidant and enzyme effects. Rosmarinic acid and the luteolin and apigenin glycosides scavenge radicals and inhibit inflammatory signalling in cell systems. Fraternale's group showed that Monarda didyma oil in LPS-stimulated U937 cells upregulated miR-146a, decreased IRAK-1 and lowered IL-6, pointing at the Toll-like receptor 4 pathway.

Didymin. The mechanism behind the aging biomarker trial is not established. Didymin is a citrus-type flavanone glycoside, and the trial reports outcomes without demonstrating a pathway.

Synergistic Supplements

Genuine synergy data for Monarda is thin. What follows is either documented or explicitly labeled as inference.

• Other Lamiaceae phenol sources. Thymus vulgaris and Origanum vulgare oils show additive to synergistic antimicrobial effects with each other in checkerboard assays, and Monarda oil sits in the same chemical family. This applies to surface and topical use, not to systemic effects.

• Carrier oils for topical use. Not a synergy but a requirement. A fixed oil at 99 to 99.5 percent of the blend is the difference between a usable preparation and a chemical burn.

• Vitamin E as a formulation partner. Aromatic phenol oils oxidize. Tocopherol at 0.1 to 0.5 percent slows the conversion of the terpene fraction to less pleasant and more irritating oxidation products.

• Nothing pairs it with a lipid supplement. Not red yeast rice, not plant sterols, not berberine. If a stack recommends Monarda alongside those, the author has confused it with the citrus.

Interactions & What NOT to Take

• Warfarin, apixaban, rivaroxaban, clopidogrel, aspirin. Theoretical rather than documented for Monarda specifically, but the rosmarinic acid content is high enough at 91.23 mg/g of extract to warrant caution with a concentrated product. Tea is a different matter.

• Drugs metabolized by CYP2A6 and CYP3A4. Thymol and carvacrol inhibit several cytochrome isoforms in vitro. No human interaction study exists for Monarda, and the systemic exposure from tea is low enough that this is a concern for oil and concentrated extract users only.

• Other thymol-containing products. If you are using a thymol mouthwash, a thymol-based bee mite treatment, and an ajwain or thyme supplement, you are stacking the same molecule from several directions without any of the labels saying so.

• Do not combine Monarda with a citrus bergamot supplement in the belief that you are doubling up on one ingredient. They share nothing but a word.

• Do not substitute Monarda oil in a recipe calling for bergamot oil, or the reverse. Bergamot peel oil is used at very low percentages in leave-on skin products specifically because of its bergapten content and phototoxicity. Monarda oil has a different limiting factor, the phenols, and a substitution swaps one constraint for another that the formulation was not designed around.

Quality, Testing & Adulteration

This is where the buyer has the least protection and the most to lose.

The naming problem is the primary quality problem. Four unrelated plants are sold under the name "bergamot": Citrus bergamia in Rutaceae, Monarda didyma and Monarda fistulosa in Lamiaceae, and Mentha aquatica variety citrata, the Eau de Cologne mint, also in Lamiaceae. Citrus limetta picks up the name in some markets as well. A search for "bergamot" returns all of them. A search for "wild bergamot" mostly returns Monarda, but the accompanying copy frequently borrows the citrus fruit's reputation.

Chemotype is undeclared almost everywhere. This is the single most consequential gap. There is no legal requirement to state whether a Monarda fistulosa oil is the thymol, carvacrol or geraniol chemotype, and the three have different aromas, different dermal limits and different uses. At least one supplier of Canadian-origin Monarda fistulosa oil states that its material is high in geraniol and prints, on the product page, that there is no relation to the citrus fruit. Most do not.

What a certificate of analysis should show for a Monarda oil

• Full binomial with author citation and the plant part distilled

• Country and region of origin and harvest year

• A gas chromatography with mass spectrometry profile reporting at minimum thymol, carvacrol, p-cymene, gamma-terpinene, thymoquinone, geraniol and linalool as percentages

• Declared chemotype based on that profile

• Optical rotation and specific gravity, which for historical oil of horsemint ran 0.930 to 0.940

• Absence of synthetic thymol, which is manufactured industrially from m-cresol and propylene or isopropanol and is chemically identical, meaning only isotope ratio methods can distinguish it from plant-derived material

Adulteration to expect. Cheap thyme oil or synthetic thymol added to a low-phenol Monarda batch would be undetectable by ordinary GC, because the target compounds are the same molecules. This is the same structural weakness that makes bergamot peel oil one of the most adulterated essential oils in commerce, where linalool and linalyl acetate ratios have to be checked by isotope ratio mass spectrometry to catch synthetic addition.

Third-party certification does not exist here. There is no USP monograph for Monarda didyma, Monarda fistulosa or Monarda punctata. NSF and ConsumerLab do not run a bee balm program. Nothing in this category has an independent identity standard a buyer can point to. What you can verify yourself is narrow: correct binomial, plant part, harvest year, and a GC report with actual numbers on it. A seller who cannot produce that last item is selling a plant nobody has chemically identified.

Special Considerations

Horsemint is the historically important species and the one nobody sells. In May 1916 the USDA published Bulletin 372, by S. C. Hood of the Bureau of Plant Industry, on commercial production of thymol from Monarda punctata. Hood had noticed horsemint growing as a weed on sandy central Florida land in 1907. Wild herb from Putnam and Volusia Counties yielded 0.12 to 0.20 percent oil at 56 to 62 percent total phenols. Six years of selection between "dark" and "light" plant types raised that: by 1914 a two-thirds acre plot gave 0.44 percent oil at 72 percent phenols, and in 1915 the average was 0.42 percent oil at 74 percent phenols. Hood projected 20 pounds of oil per acre in year one and 30 to 40 pounds thereafter, working out to 12.86 pounds of pure thymol per acre the first year and 19.29 pounds after. At two dollars a pound, that was a gross return of $25.72 per acre in year one. The competing source at the time was ajowan oil from Trachyspermum ammi, grown in northern India and processed in northern Europe, and Hood's ten-year import table for 1906 through 1915 shows annual United States thymol imports swinging between roughly 2,000 and 18,000 pounds. The industry never scaled. Thymol is now made from m-cresol.

Thymoquinone is a live commercial interest in Monarda fistulosa. Prairie Pharms LLC holds United States patent 9,776,939, filed 23 May 2014 and granted 3 October 2017, inventor William G. Rohlfsen, covering cultivation of Monarda fistulosa at 2.5 to 5 pounds of seed per acre to produce oil in which thymoquinone and thymohydroquinone together make up about 25 to 50 percent. Thymoquinone is the compound black cumin seed oil is sold on. That is another cross-shelf identity for this plant that almost no bee balm buyer knows about.

The Oswego tea story is older than the story usually told. Bartram found colonists near Fort Oswego brewing Monarda didyma between 1743 and 1746, and sent seed to England where it flowered in 1746. The Boston Tea Party was 16 December 1773. The tea substitute use predates the protest by about thirty years. What the Tea Party did was popularize a practice already thirty years established, among people who had learned it from the Haudenosaunee of that region.

Bee balm is a serious pollinator plant. Monarda punctata is the sole pollen source for the specialist bee Protandrena abdominalis. If the choice is between buying an unstandardized capsule and planting the species, the second does more good.

Research Status & Evidence Quality

Strong Evidence For

• Chemical characterization. The essential oil composition of all four commercial Monarda species has been repeatedly measured by GC-MS across many populations, and the chemotype variation is thoroughly documented rather than merely asserted.

• In vitro antimicrobial and antifungal activity of the oil, with MIC values below 0.25 to 0.5 mg/mL in current work.

• In vitro antioxidant capacity across DPPH, FRAP, TEAC and ORAC.

• Human pharmacokinetics of thymol as a single molecule, including the absence of free thymol in plasma.

Moderate Evidence For

• Anti-inflammatory and analgesic activity in rodents, dose-dependent across 50, 100 and 200 mg/kg with two standard models and clean acute toxicity up to 5,000 mg/kg.

• Cosmetic and preservative applications of the oil, where the free phenol is present at working concentration and the mechanism is not in question.

• TRPA1 channel activation as the basis of the sensory and counterirritant effects.

Emerging / Preliminary Evidence For

• Effects on biological aging markers. One randomized, double-blind, placebo-controlled trial, 81 participants, 12 weeks, 100 mg per day of a didymin-standardized Monarda didyma extract, with the extract supplied by its manufacturer and the within-group telomere result at p equals 0.0537.

• Anti-angiogenic activity of hydroglyceric extracts in the chick chorioallantoic membrane model.

• Anti-seborrheic use of Monarda fistulosa oil, reported in older Eastern European work that has not been replicated in a controlled setting.

Research Limitations

• One human trial exists for the entire genus, on one extract, at one dose, for one duration, in one population of 81 people aged 45 to 65 recruited from a single Italian university.

• That trial used a flavonoid extract. It says nothing about the essential oil, the tea or the tincture, and it cannot be cited to support any of them.

• No Monarda preparation has been tested against any lipid, glycemic or cardiovascular endpoint in humans.

• No trial has compared chemotypes head to head for any clinical outcome.

• Almost all antimicrobial data comes from broth dilution against the free phenol, which is not the species that circulates after ingestion.

• There is no pharmacopoeial monograph, so even the identity of a commercial sample cannot be checked against an official standard.

Summary & Key Takeaways

Bee balm is a real North American medicinal plant with a documented Indigenous record, a nineteenth century pharmaceutical career, a 1916 USDA agronomy program behind it, and one 12-week randomized trial in 2025. None of that is what sells it. What sells it is a common name it acquired because a colonist thought the leaves smelled like an Italian citrus fruit, and the research base attached to that fruit does not transfer across the gap by a single molecule.

Bottom Line

Buy Monarda as a tea and a garden plant, where the case is honest and the chemistry is mild. Buy the essential oil only with a gas chromatography report in hand, because "Monarda fistulosa" on a label covers oils ranging from 3 percent to 77 percent thymol and 1.6 percent to 71.5 percent carvacrol. Do not buy any Monarda product expecting an outcome reported for Citrus bergamia. The only Monarda extract with a human trial behind it is a didymin-standardized Monarda didyma preparation at 100 mg per day, and that trial's main finding is that the placebo group's markers worsened.

Key Safety Points

• Chemotype is undeclared on nearly every product and changes the appropriate dilution by more than a factor of ten.

• Do not take Monarda essential oil internally. Thymol LD50 values span 88 to 1,200 mg/kg across species, and no human dose has been established since 1898.

• Avoid concentrated preparations in pregnancy. Oil of horsemint has a documented historical emmenagogue use and zero modern safety data.

• Caution with anticoagulants if using a concentrated polyphenol extract at 91.23 mg/g rosmarinic acid.

• Phototoxicity warnings you may have read about "bergamot" apply to the citrus and not to this plant. Do not let that reassurance transfer in the other direction.

Special Note

The lesson here generalizes past this shelf. When two unrelated plants share a common name, the marketing money flows toward whichever one has the better clinical file, and the cheaper plant inherits the reputation without inheriting a single compound. Thymol and carvacrol against brutieridin and melitidin is not a subtle distinction requiring a chemist to adjudicate. It is two different families on two different continents. Whenever a product's headline benefit is easier to find in the search results than in the product's own certificate of analysis, check whether the studies were run on the species in the bottle.