Feverfew: History and Traditional Use
Feverfew is a small white daisy with an unusually long memory. Its very name is a fossil of Latin medicine (febrifugia, "that which puts fever to flight"), its botanical name carries a legend about the building of the Athenian Acropolis, and the plant itself was carried out of the Balkans into monastery gardens and cottage plots across Europe, where it was chewed, brewed and boiled in wine for fevers, headaches and "women's complaints" for the better part of a thousand years. Then it was almost forgotten — until a Welsh doctor's wife started eating three leaves a day in 1973 and, within fifteen years, feverfew had been through the pages of The Lancet and the British Medical Journal. This article follows that documented journey from the ancient names to the migraine trials and the monographs. Where the record is firm we say so; where a story is tradition or legend, we name it as such.
Table of Contents
- Origins and Botany: A Daisy from the Balkans
- What the Names Remember: Febrifuge, Featherfew and Parthenion
- Dioscorides, Pliny and the Ancient Record
- Monastery Gardens and the "Medieval Aspirin"
- Gerard, Culpeper and John Hill: The English Herbals
- Wales, 1973: How Feverfew Came Back
- Into the Journals: The First Clinical Trials (1985–1989)
- Parthenolide: The Chemistry Chapter (1959 Onward)
- Weighing the Evidence: Systematic Reviews, MIG-99 and the Monographs
- Feverfew Today
- Key Research Papers
- Connections
- Featured Videos
Origins and Botany: A Daisy from the Balkans
Feverfew, Tanacetum parthenium, is a bushy, strongly scented perennial of the daisy family (Asteraceae), knee-high at most, with feathery yellow-green leaves and loose clusters of small flowers — a ring of short white rays around a flat yellow disc. It looks so much like chamomile that a tenth-century illustrated Dioscorides made in Constantinople labels a picture of feverfew chamaimelon, chamomile, and the confusion runs through the botanical names too: the plant has been filed under Matricaria, Chrysanthemum, Pyrethrum and Leucanthemum before settling, in most modern floras, in the tansy genus Tanacetum.
Its home is south-eastern Europe. Botanists regard feverfew as native to the Balkan Peninsula, where it grows in rocky mountain scrub, and probably also to Anatolia (modern Turkey) and the Caucasus. Everywhere else — the rest of Europe, the British Isles, North America, Australia — it is an escape from cultivation. In Britain and Ireland it is classed as an archaeophyte: a plant introduced by people before about 1500, most likely carried north by monks as a medicinal herb during the Middle Ages, and now so thoroughly naturalised along hedge-bases, old walls and pavement cracks that it is hard to believe it was ever a foreigner. Feverfew is a garden plant that went wild, and its medical history is the history of a plant people deliberately kept close to the house.
What the Names Remember: Febrifuge, Featherfew and Parthenion
The English word is a small piece of medical Latin worn smooth. Old English already had feferfuge, borrowed from Late Latin febrifugia — febris, fever, and fugare, to put to flight — so the plant was named, before the Norman Conquest, for exactly the job its name still describes. "Featherfew", the form Gerard and his contemporaries often used, is a folk re-shaping that made the strange word make sense: the leaves are feathery. Other English names recorded over the centuries include bachelor's buttons and flirtwort; in German the plant is Mutterkraut, "mother-herb", a name that points to a second ancient reputation, in childbirth and menstrual complaints, that we will meet again.
The species name parthenium comes with a famous story, and the story is worth checking. Plutarch, in his Life of Pericles (chapter 13), records that during the building of the Propylaea — the gateway of the Acropolis, not the Parthenon itself — a workman fell from a great height and was given up by the physicians; Athena appeared to Pericles in a dream and prescribed a treatment, the man recovered, and Pericles set up a bronze statue of Athena Hygieia (the Healer) in thanks. Plutarch names no plant. It is Pliny the Elder, a generation earlier, who attaches a herb to the tale: in Book 22 of his Natural History a slave of Pericles falls from the temple roof and is cured by a plant Minerva reveals in a dream, "and so it began to be called parthenium". The legend is genuinely ancient, but it is Pliny's, not Plutarch's, and even Pliny does not say which modern plant he means. The soberer alternative, favoured by several modern reviews, is that parthenios simply means "maiden" or "virgin", and that the name records the herb's long use for the ailments of women.
Dioscorides, Pliny and the Ancient Record
The earliest medical description that scholars connect with feverfew is in the De Materia Medica of Dioscorides, the Greek army physician who wrote in the first century AD. Dioscorides describes a plant he calls parthenion and recommends it for "hot inflammations" and swellings and for disorders of the womb; later herbalists, from the Renaissance onward, took his parthenion to be their feverfew, and modern reviews of the herb repeat the identification. It is a reasonable identification but not a certain one, because ancient plant names were applied loosely; what can be said firmly is that by the first century a plant matching feverfew was in the Mediterranean pharmacopoeia, used chiefly for inflammation and for women's conditions.
Pliny the Elder, writing at about the same time, gives a fuller picture in Book 21 of the Natural History. His parthenium — "which some call leucanthes and others amaracus; Celsus among us calls it perdicium and muralis" — grows in garden hedges, has a white flower, the smell of an apple and a bitter taste: a fair description of feverfew. He lists it as a sitz-bath for hardness and inflammation of the womb; dried and applied with honey and vinegar to draw off "black bile", and so useful for vertigo and for stones; and as a plaster for sacer ignis, the "holy fire" later called St Anthony's fire (erysipelas), and for scrofulous swellings. He also passes on a charm of the Magi against tertian fever, with a leaf of the plant placed under the patient's tongue. Three of feverfew's later reputations — fever, the womb, and dizziness of the head — are thus already on record in Roman times, as uses, not as evidence that the plant worked.
Monastery Gardens and the "Medieval Aspirin"
Between antiquity and the printed herbals of the sixteenth century the written record on feverfew is thin, but the plant itself tells a story. Its spread out of the Balkans into northern Europe was a matter of gardens rather than seed on the wind, and the monasteries, which kept physic gardens and copied the Latin medical texts, are the likeliest carriers. Feverfew is recorded as cultivated for medicine in central Europe since the Middle Ages, and its presence in Old English under a Latin-derived name shows it had reached England, with its reputation attached, well before 1066.
What it was used for in those centuries can be reconstructed from the uses that surface, fully formed, in the earliest printed herbals. Reviewers who have gone through the old texts group them into three families: fevers (especially the recurring "intermittent" agues later understood as malaria), headache and dizziness; the conditions of women — difficult labour, retained afterbirth, threatened miscarriage, painful periods; and everyday complaints such as stomach ache, toothache and insect bites. Because it was reached for so often and for so much, feverfew has been nicknamed the "medieval aspirin", and one British botanical account calls it "the aspirin of the herbal era". The preparations were simple: the fresh leaves chewed or eaten in bread, the dried herb as a tea, or the herb boiled in white wine. A tradition this broad is what it looks like — a general-purpose bitter herb that was always to hand — and none of these uses had been tested in any way we would recognise.
Gerard, Culpeper and John Hill: The English Herbals
The English herbals put feverfew's traditional uses into print, and two of them add the thread that matters most for its modern story. John Gerard's Herball (1597; enlarged by Thomas Johnson in 1633) treats "Feuerfew" among the daisies and, alongside the expected uses in fevers and for women, reports that it is "very good for them that are giddie in the head" and for those who are "melancholike, sad, pensive and without speech" — dizziness and low spirits, in other words, not yet headache as such.
Nicholas Culpeper, in The English Physitian (1652, enlarged 1653), fits feverfew into his astrological scheme: "Venus commands this herb, and has commended it to succour her sisters (women), to be a general strengthener of their wombs, and to remedy such infirmities as a careless midwife has there caused." He recommends it boiled in white wine to "cleanse the womb" and expel the afterbirth, and then, almost in passing, gives the line that later writers would seize on: "It is very effectual for all pains in the head coming of a cold cause." Culpeper's "planets" and "cold causes" are the framework of his day, not a mechanism; what his entry shows is that by the mid-seventeenth century feverfew was an ordinary English remedy for headache as well as for childbirth.
A century later the claim was made outright. In 1772 John Hill, the prolific physician-botanist of The Family Herbal, published a tract on the virtues of British herbs that included "Cures of Head-Achs by Feverfew Tea, with Two Singular Instances", and wrote the sentence that every modern history of the plant quotes: "In the worst headache this herb exceeds whatever else is known." It is the last important thing anyone said about feverfew and headache for two hundred years. Through the nineteenth century the herb drifted to the margins of the dispensatories — a tonic, an "anti-hysteric", a cottage flower — and it stayed there until the 1970s.
Wales, 1973: How Feverfew Came Back
Feverfew's modern revival is unusual among herbal medicines in that it began not with a researcher or a manufacturer but with patients, and the chain of events is well documented. In 1973 Mrs Anne Jenkins, a doctor's wife in Cardiff who suffered severe, recurrent migraines, began on a friend's advice to eat three fresh feverfew leaves a day. After about ten months her migraines stopped, and they stayed away for as long as she kept taking the leaves. She told people; the story spread through South Wales, where a folk memory of the plant evidently survived; and by the late 1970s the British press was reporting that migraine sufferers in Wales had found relief by eating feverfew leaves. Thousands of people began growing the plant and chewing it, outside any medical supervision. (Feverfew products had in fact held old-style "product licences of right" in the United Kingdom from the early 1970s until 1988, a category that required no proof of efficacy.)
The medical profession took notice through Dr E. Stewart Johnson of the City of London Migraine Clinic, who was reached, as several accounts have it, through Mrs Jenkins's husband. Rather than dismiss the craze, Johnson surveyed it. His 1983 survey gathered some 300 feverfew users; of 253 who had true migraine, about 72% reported that their attacks had become less frequent or less severe since they started eating the leaves, and a similar proportion of a smaller group with tension headache said the same. A survey of enthusiasts proves nothing about efficacy — people who found the herb useless had stopped and were not there to be counted — but it did document the side-effects: just over 11% of leaf-chewers reported mouth ulcers, some badly enough to give up, and others soreness of the tongue and swelling of the lips. The modern literature on feverfew safety begins here.
Into the Journals: The First Clinical Trials (1985–1989)
The first controlled trial followed directly from the survey. Because the volunteers were already eating feverfew daily, Johnson and colleagues at the City of London Migraine Clinic and Chelsea College ran a withdrawal study: seventeen long-term users were given capsules of freeze-dried feverfew leaf or of placebo, without knowing which. The nine switched to placebo suffered a marked increase in the frequency and severity of their migraines, with nausea and vomiting, while the eight kept on feverfew were unchanged. The paper appeared in the British Medical Journal in August 1985. It also named something new: several patients taken off the herb after years of daily use developed a cluster of complaints — rebound headache, anxiety, poor sleep, and stiffness or aching of muscles and joints — which the authors called the "post-feverfew syndrome". That term has stayed in the literature; later estimates put it at roughly one in ten long-term leaf users who stop abruptly.
The stronger test came from Nottingham. In July 1988 Murphy, Heptinstall and Mitchell published in The Lancet a randomised, double-blind, placebo-controlled crossover trial: 72 people with migraine were enrolled, 59 completed, and during the feverfew months they had fewer attacks and less vomiting than during the placebo months, with no serious side-effects. This is the trial that made feverfew a research subject. The same Nottingham group tested the herb's other folk reputation the following year, in a double-blind trial in rheumatoid arthritis published in the Annals of the Rheumatic Diseases (1989), and found no benefit at all — an early and honest negative that the arthritis claim has never recovered from. A Dutch crossover trial in 1996, using an alcoholic extract, also found no preventive effect on migraine, a result most commentators attribute to the extract rather than the herb, for reasons the next section explains.
Parthenolide: The Chemistry Chapter (1959 Onward)
The chemistry of feverfew was worked out before anyone in a laboratory cared about migraine. In the late 1950s the Prague school of natural-product chemistry around František Šorm and Vlastimil Herout was isolating the bitter principles of daisy-family plants, and from Chrysanthemum parthenium, as feverfew was then called, they obtained a crystalline compound they named parthenolide; a first report appeared in Chemistry and Industry in 1959, and in 1961 Souček, Herout and Šorm published its constitution in the Collection of Czechoslovak Chemical Communications. Parthenolide is a sesquiterpene lactone — a fifteen-carbon molecule closed into a ring that carries a reactive "lactone" group — and it is the most abundant of a family of such lactones in the leaf. It is also the main reason feverfew can cause allergic contact dermatitis.
Twenty-five years later the Nottingham pharmacologists connected the molecule to the folklore. In 1985 Stan Heptinstall and colleagues showed in The Lancet that extracts of feverfew inhibit the release of serotonin-containing granules from blood platelets and of inflammatory granules from white cells — a plausible link to migraine, in which platelet serotonin was then thought to play a central role. In 1986 the same group showed that the active extracts all contained compounds with an alpha-methylene butyrolactone unit, the reactive group in parthenolide, and by 1990 parthenolide itself was shown to reproduce the extract's effect on platelets. Out of this came the "parthenolide hypothesis": that parthenolide is the active principle and that a feverfew product is only as good as its parthenolide content.
The hypothesis mattered because commercial feverfew turned out to be wildly inconsistent. A 1992 analysis by Heptinstall, the Canadian government chemist Dennis Awang and colleagues measured parthenolide in authenticated leaf and in products on sale and found that some contained a respectable amount and others almost none; parthenolide also degrades in storage. Canada's Health Protection Branch (the forerunner of Health Canada) responded in the early 1990s by issuing Drug Identification Numbers for feverfew leaf products only if they contained at least 0.2% parthenolide, a criterion associated with Awang and Heptinstall, and the same 0.2% floor was later adopted by the European Pharmacopoeia for the dried herb. And yet the hypothesis was never quite proved. Parthenolide-poor material still shows anti-inflammatory activity in some laboratory assays (a 1997 study made the point directly), and the European Medicines Agency's assessors concluded in 2010 that parthenolide "is not a constituent with known therapeutic activity even though it contributes to the activity". Parthenolide is the best quality marker feverfew has, not proven to be the whole story.
Weighing the Evidence: Systematic Reviews, MIG-99 and the Monographs
By the late 1990s there were enough trials to review. Vogler, Pittler and Ernst's systematic review in Cephalalgia (1998) judged most of the placebo-controlled trials favourable but the whole body of evidence too small and mixed to be conclusive; the same team's Cochrane reviews of 2000 and 2004 reached the same guarded verdict. The trials disagreed partly because the products did, and a German programme set out to fix that with a stable, standardised carbon-dioxide extract known by the research code MIG-99. A dose-finding trial in 2002 was inconclusive overall but encouraging in people with frequent attacks, and a larger multicentre trial published in 2005 found that 6.25 mg three times daily reduced migraine frequency by about 0.6 attacks a month more than placebo. The 2015 Cochrane update by Wider, Pittler and Ernst — six trials, 561 participants — rested its cautiously positive conclusion mainly on that single moderate-sized study, and found no major adverse effects in any trial. In 2012 the American Academy of Neurology and American Headache Society guideline rated the standardised extract "probably effective" (Level B) for episodic migraine prevention, a rating since debated.
The regulatory record follows the same arc. Germany's Commission E, whose monographs of 1978–1994 defined herbal practice for a generation, never processed feverfew at all — there is no Commission E monograph, approved or unapproved. The European Scientific Cooperative on Phytotherapy (ESCOP) gave it a monograph in the second edition of its Monographs (2003) for the prevention of migraine, updated online in 2014. UK feverfew capsules have been registered under the traditional herbal medicines scheme since April 2007, and the European Medicines Agency's herbal committee adopted a Community herbal monograph on 25 November 2010, revised on 8 July 2020, classifying feverfew as a traditional herbal medicine for the prevention of migraine headaches once a doctor has ruled out serious causes — a category that certifies thirty years of safe use, not proof of efficacy.
Feverfew Today
Set the tradition beside the evidence and the fit is narrow but real. Of the three ancient reputations — fever, the womb, and the head — only the third has any modern support, and only for one specific job: taken daily for weeks, feverfew may modestly reduce how often migraines occur in some people, with the best evidence for standardised extracts rather than home-grown leaf. The fever use that named the plant has never been tested and is not used; the rheumatoid arthritis use was tested in 1989 and failed. The womb use survives in reverse, as a warning: an emmenagogue and "expeller of the afterbirth" in every old herbal is, by the same token, a herb to avoid in pregnancy, and the modern monographs say so.
The safety lessons are mostly old ones re-learned. Mouth ulcers come from chewing the fresh leaf — the 1983 survey found them, capsule trials do not — so the traditional way of taking feverfew is the one modern advice discourages. Stopping suddenly after long daily use can bring the rebound "post-feverfew syndrome" first described in 1985. Because it is a daisy, and parthenolide a strong contact allergen, people sensitive to chamomile, ragweed or chrysanthemums can react to it; and its effect on platelets is a reason for caution alongside blood-thinning medicines.
What is most distinctive about feverfew's history is its shape. Most herbal medicines reached the laboratory from the top down; feverfew got there from the bottom up, because a woman in Cardiff kept eating her leaves and a clinic doctor decided that a folk practice used by thousands deserved a survey rather than a sneer. The plant in the cracks of British pavements is still there, still bitter, still smelling faintly of apples as Pliny said, and still a subject of open questions rather than settled answers. For what the modern trials found in detail, see the Feverfew Benefits pages.
Key Research Papers
The list below covers the landmarks of feverfew's scientific history: the 1961 structure of parthenolide, the Nottingham pharmacology, the first clinical trials, the analyses of product quality and the systematic reviews. Historical primary texts (Pliny, Plutarch, Dioscorides, Gerard, Culpeper, Hill) are named in the article as historical sources rather than cited here. Each external link opens in a new tab.
- Souček M, Herout V, Šorm F (1961). On terpenes. CXVIII. Constitution of parthenolide. Collection of Czechoslovak Chemical Communications. — doi:10.1135/cccc19610803
- Heptinstall S, White A, Williamson L, Mitchell JR (1985). Extracts of feverfew inhibit granule secretion in blood platelets and polymorphonuclear leucocytes. Lancet. — PubMed PMID: 2860288
- Johnson ES, Kadam NP, Hylands DM, Hylands PJ (1985). Efficacy of feverfew as prophylactic treatment of migraine. British Medical Journal (Clinical Research Ed.). — PubMed PMID: 3929876
- Murphy JJ, Heptinstall S, Mitchell JR (1988). Randomised double-blind placebo-controlled trial of feverfew in migraine prevention. Lancet. — PubMed PMID: 2899663
- Pattrick M, Heptinstall S, Doherty M (1989). Feverfew in rheumatoid arthritis: a double blind, placebo controlled study. Annals of the Rheumatic Diseases. — PubMed PMID: 2673080
- Heptinstall S, Awang DV, Dawson BA, Kindack D, et al. (1992). Parthenolide content and bioactivity of feverfew (Tanacetum parthenium (L.) Schultz-Bip.). Estimation of commercial and authenticated feverfew products. Journal of Pharmacy and Pharmacology. — PubMed PMID: 1359053
- Vogler BK, Pittler MH, Ernst E (1998). Feverfew as a preventive treatment for migraine: a systematic review. Cephalalgia. — PubMed PMID: 9950629
- Diener HC, Pfaffenrath V, Schnitker J, Friede M, et al. (2005). Efficacy and safety of 6.25 mg t.i.d. feverfew CO2-extract (MIG-99) in migraine prevention — a randomized, double-blind, multicentre, placebo-controlled study. Cephalalgia. — PubMed PMID: 16232154
- Pareek A, Suthar M, Rathore GS, Bansal V (2011). Feverfew (Tanacetum parthenium L.): A systematic review. Pharmacognosy Reviews. — PubMed PMID: 22096324
- Wider B, Pittler MH, Ernst E (2015). Feverfew for preventing migraine. Cochrane Database of Systematic Reviews. — PubMed PMID: 25892430