Flying Ointment: What Was Really in the Witch Salve

I have worked with herbs for many years, and flying ointment is the question I am asked most often. Usually it sounds like this: is there a real recipe?

The honest answer is more uncomfortable than a yes or a no. There are recipes. They are old, they are well documented — and almost all of them were written by men who never received a salve from a witch in their lives. That is exactly what makes the story interesting.

What this article gives you: what the word actually covers, the ancient literary ancestors, the first written recipe and the ingredient its author deliberately left out, the silence in the trial records, the pharmacology, the twentieth-century self-experiments, the American side of all of it — and no recipe.

What “flying ointment” actually means

Two completely different things hide behind the word, and they are confused constantly.

The first is an idea: the salve or paste that people — usually women called witches — were said to rub on themselves in the late medieval and early modern period in order to fly to the sabbath. This idea is extremely well documented. What is documented, though, is the idea, not the practice. Scholars were already arguing in the sixteenth century about whether such salves really existed.

The second is a class of substances: ointment-form preparations of psychoactive plants, above all from the nightshade family. That such preparations can produce intoxication, delirium and delusional dreaming is pharmacologically uncontroversial. That they were actually in use as intoxicants in the Middle Ages is not — the evidence is missing.

The names the thing travels under in the literature are a clue in themselves: Unguentum Sabbati, unguentum pharelis, unguentum somniferum, and in German Buhlsalbe (lovers’ salve) and Schlafsalbe (sleeping salve). Two very different expectations of the same substance — eroticism and sedation. Flight often is not in the name at all.

The ancient ancestors

Not one salve recipe of this kind survives from antiquity. What survives is poetry — and that was enough to carry the motif for centuries.

Homer has Hera anoint herself with ambrosia in the Iliad to reach Zeus on Mount Ida. She moves across the highest peaks without touching the earth, and Zeus wonders how she covered the distance so fast. An ointment that shortens the journey: the motif is there long before there are witches.

It gets more explicit in Apuleius. In the Metamorphoses, Lucius watches the Thessalian witch Pamphile take a box of ointment and rub herself from head to foot. Down sprouts, flight feathers grow, the nose goes horny and hooked, the feet draw up into talons — and Pamphile is standing there as an owl.

Ovid and Seneca know the strigae too, who turn into night birds or ride through the air. It is worth being clear about what this is: entertainment literature, not field research. But these were exactly the texts available to late medieval scholastics when they began to construct the witch systematically. They built a doctrine out of novel plots.

Hartlieb 1456: the first written recipe

The first physician to write down a witch-salve recipe is Johannes Hartlieb (c. 1400–1468), personal physician and adviser to Duke Albrecht III of Bavaria. In 1456 he wrote Das Buch aller verbotenen Künste, the Book of All Forbidden Arts. Chapter 32 contains the passage everything since has leaned on.

Hartlieb calls the salve unguentum pharelis — nobody knows to this day what pharelis means. It is made, he says, from seven herbs, each gathered on the day of the week that belongs to it, plus bird blood and animal fat. Then one smears benches, stools, rakes or oven forks, and off one goes.

Two things about this passage are remarkable.

First, the system. The weekday assignment is not folklore, it is learned order: since the Middle Ages the planetary signs also stand for the days of the week, in alchemy simultaneously for metals, and the herbs are assigned to the planets. Through that chain six of the seven plants can be identified with reasonable confidence. This is a scholar’s desk work, not knowledge from a forest hut.

Second, the gap. The seventh herb — Saturday’s, and therefore Saturn’s — Hartlieb leaves out on purpose, so that nobody should be harmed by it. He censors himself. The obvious scholarly guess is mandrake, the Saturn plant par excellence. If you are wondering what a “complete” recipe is worth when its own author removes the decisive component — welcome to the centre of the problem.

And Hartlieb’s own verdict? Unambiguous. All of it is forbidden. He is not writing a manual. He is writing a warning.

The Malleus, and what became of the salve

In 1486 Heinrich Kramer’s Malleus maleficarum appeared. Its second part holds that witches can rise into the air by means of a salve made from the limbs of children.

Here the motif tips over. A herb paste becomes a crime. The salve is no longer an intoxicant or a dream aid; it is evidence, and its recipe is an accusation. Gerolamo Cardano later describes a lamiarum unguentum of child fat, celery juice, vervain, cinquefoil, soot and ergot. Read these lists not as pharmacy but as the dramaturgy of a charge.

It needs saying plainly: these texts cost people their lives. A recipe that names child fat is not an interesting historical detail. It is the written preparation of a death sentence.

The silence in the trial records

Now the finding that turns the whole narrative around, and that almost never appears in articles on this subject.

No salve recipes are known from the records of the witch trials. None.

The accused knew the plant ingredients at best by hearsay. More often they stated that they had not made the grease themselves but had received it from the devil in person — a statement produced under torture, which confirms precisely what the interrogators wanted to hear.

Every surviving recipe comes from physicians and early natural philosophers. That also explains why their compositions overlap so strikingly with the ordinary medicines of the day: they are reconstructed from the apothecary knowledge of the period, not from the practice of the accused.

In 1545 Andrés de Laguna, a physician at the court of Charles V among others, recorded confiscating a witch salve. Giambattista della Porta gives a formula in Magiae naturalis in 1558 in which alkaloid-bearing plants dominate, alongside symbolic ingredients such as bat’s blood; nobody knows his sources and his account is not verifiable. Johann Weichard von Valvasor in 1689 describes a salve under which the woman merely dreams she has flown, and names deadly nightshade and wolfsbane.

Valvasor’s sentence is the cleverest remark in the entire series. It does not describe a flight. It describes a dream that someone took for a flight.

The plants: why nightshade explains everything and excuses nothing

The plants in question belong mostly to the nightshades: henbane, deadly nightshade, datura, mandrake. Some recipes add monkshood — wolfsbane — which belongs to a different family and is among the most poisonous plants in Europe.

The nightshades contain tropane alkaloids: hyoscyamine, atropine, scopolamine. These act on the central and peripheral nervous system, can trigger delirium, hallucination and a complete collapse of the sense of time, and they are absorbed through skin and mucous membranes. That is precisely where the pharmacological plausibility of the salve idea lies: a fatty base assists absorption through the skin.

And that is precisely where the danger lies, and I will put it as plainly as I can. The distance between a dose that does something and a dose that puts a person in hospital or in the ground is very small with these plants, and individually unpredictable. Alkaloid content varies with species, site, harvest time, plant part and year. There is no household dose. There is no safe application.

That is why this article contains no recipe, no quantity, no extraction method and no application site. Not out of prudishness, and not because I am withholding something from you, but because I have read too many times how it ends. If you are disappointed at this point, that is all right. Disappointment is an acceptable price.

In the United States, if you or someone with you has taken any of these plants, the number is Poison Help, 1-800-222-1222, and for anyone who is not responding normally it is 911 first. The classic presentation of tropane poisoning is taught to American emergency clinicians as a rhyme about being blind as a bat, mad as a hatter, red as a beet, hot as a hare and dry as a bone — which tells you something about how recognisable, and how serious, it is.

The plant in your own yard: Jamestown, 1676

Here is the part that makes this a local subject rather than a European one. Datura — jimsonweed — grows across most of the United States, in vacant lots, field margins and roadsides, and it is the nightshade Americans are most likely to meet.

Its name is a worn-down version of Jamestown weed, and the story behind it is on the record. In 1676, soldiers sent to Jamestown, Virginia during Bacon’s Rebellion gathered the plant and ate it. Robert Beverley set down what happened in The History and Present State of Virginia in 1705: the men were deranged for eleven days, behaving like fools and remembering nothing of it afterwards. They recovered. That is the reason a common American weed carries a Virginia place-name.

Two things follow from this, and they pull in opposite directions in a useful way. The pharmacology in the European salve literature is real, and the plants that carry it are growing within a mile of most American readers. And the best-documented American encounter with that pharmacology is not a flight, not a vision and not a rite — it is eleven days of accidental poisoning that nobody enjoyed and nobody remembered.

Datura also has a serious place in the anthropological literature of the American Southwest and Mexico, which is where a great deal of modern enthusiasm for it actually comes from, by way of Carlos Castaneda — whose accounts have their own well-known problems as evidence.

Salem: the ointment that was not there

Most American readers arrive at this subject through Salem, and Salem is a good reason to be careful with the whole picture.

Flying ointment plays essentially no part in the Salem record. The 1692 trials in Massachusetts turned on spectral evidence — testimony that the shape of the accused had appeared and afflicted the witnesses — on afflicted behaviour in the courtroom, on accusation chains, and on the “witch cake” episode. Salem did not run on salves and broomsticks. The Continental apparatus of the sabbath and the ointment, built up by Kramer and his successors, is largely absent from what happened in Essex County.

So the broomstick in the American imagination is an import. It arrived through European print, through Halloween iconography and through illustration, and it attached itself retroactively to a set of American trials it had nothing to do with.

The self-experiments, and the American version of the same mistake

In the twentieth century several researchers wanted to know. Will-Erich Peuckert carried out a self-experiment with a friend in 1927 following della Porta’s account. Both rubbed armpits and forehead and fell into a sleep lasting some twenty hours. Peuckert published on witch salves in 1960.

The experiments show two things. First: the preparations work. Second, and this is the scientifically decisive point, the reported experiences are now largely regarded as shaped by the researchers’ expectations. Someone who rubs themselves down in order to experience a witches’ sabbath is more likely to experience a witches’ sabbath. That is not an accusation against Peuckert; it is a basic property of intoxicated states. The content comes largely out of the head, not out of the jar.

America ran the same reasoning on Salem, and it is worth knowing how that turned out. In 1976 the behavioural psychologist Linnda Caporael published a paper in Science proposing that ergot poisoning from contaminated rye explained the afflictions at Salem Village. It was elegant, widely reported, and it is still repeated confidently today. In the same journal, the same year, Nicholas Spanos and Jack Gottlieb published a detailed rebuttal: the symptom pattern, the distribution among households and the timing did not fit ergotism, and the historical record gave a better account without it.

The parallel is exact. A pharmacological explanation for a witch panic is enormously attractive, because it converts a moral catastrophe into a chemical accident. It is also, in both cases, not what the evidence supports.

And the “wise women who were wiped out”?

There is a popular counter-narrative: that flying ointment was part of an ancient matriarchal body of healing knowledge which the church deliberately exterminated. It was widely taken up in the 1980s and it is emotionally understandable — it hands the victims of the witch trials back a competence that was taken from them.

The sources do not carry it. If there had been a continuously practised body of salve knowledge, it would have to show up in the records, and it is not there. Historical scholarship now largely reads the persecutions as a product of crisis, legal procedure and denunciation dynamics, not as a campaign against a guild of herbalists.

The victims do not need an invented magical power in order to be victims. They were neighbours, midwives, beggars, wealthy townspeople, men and children. Attributing to them the salve they were once accused of is a strange form of honouring them.

What you can work with instead

The salve was never the route. Smoke was — across every culture, for thousands of years, and without the poisoning risk. That said, “it is a plant, so it is safe” is exactly the error this whole article is about: some plants are dangerous to burn as well, and that list is worth knowing before you light anything — see the plants you should never burn.

Beyond that: non-toxic herbs with a long working tradition, and the weekday structure from Hartlieb’s own account — gathering, drying and working on the day assigned to the plant — which is a genuinely old ritual frame you can use without any of the pharmacology. If you want the practical end of that, start with harvesting and drying.

What we do not promise

This article does not give you a recipe, and it will not give you one if you ask. That is a position, not an omission.

We do not claim that flying ointment enabled a flight, a real journey to a sabbath, or any healing or consciousness-expanding effect. The intoxication reports from the twentieth-century self-experiments are regarded as expectation-shaped.

We do not claim that these plants have a safe traditional dose. They do not. The margin is narrow, variable and not predictable for an individual, and skin absorption makes it worse rather than better.

And if the appeal of this subject for you is mostly about getting out of your own life for a night, that is worth saying out loud to someone. It is a very old wish, it is not shameful, and there is no plant in this article that answers it safely.

Frequently asked questions

Is there a real flying ointment recipe?

There are recipes, but not from the people accused of using them. Every surviving formula comes from physicians and early natural philosophers such as Hartlieb, della Porta and Valvasor. No salve recipe is known from the witch trial records themselves. This article gives no recipe, quantity or method, because the plants involved have no safe dose.

What plants were in witches' salve?

Mainly nightshades - henbane, deadly nightshade, datura and mandrake - and in some recipes monkshood. They contain tropane alkaloids such as hyoscyamine, atropine and scopolamine, which are absorbed through the skin and can cause delirium. All of them are seriously poisonous.

Did witches actually fly on broomsticks?

No. The most clear-eyed early source, Valvasor in 1689, already described a salve under which the person merely dreamed she had flown. The motif of an ointment that carries someone through the air comes from ancient literature - Homer, Apuleius, Ovid - long before there were witch trials.

Was flying ointment part of the Salem trials?

Essentially not. The 1692 Salem trials turned on spectral evidence, afflicted behaviour and accusation chains, not on salves or sabbath flight. The broomstick image reached America later through European print and Halloween iconography and attached itself retroactively.

Why is it called jimsonweed?

It is a worn-down form of Jamestown weed. In 1676 soldiers at Jamestown, Virginia ate the plant and were deranged for about eleven days, an episode Robert Beverley recorded in his 1705 history of Virginia. The plant is datura, one of the same nightshades named in the European salve literature.

Did ergot poisoning cause the Salem witch trials?

Probably not. Linnda Caporael proposed it in Science in 1976 and the idea remains popular, but Nicholas Spanos and Jack Gottlieb published a detailed rebuttal in the same journal that year, arguing the symptoms, distribution and timing do not fit ergotism.

What was the seventh herb Hartlieb left out?

He never says. He assigns one herb to each day of the week and deliberately omits Saturday's so that nobody should be harmed by it. Since Saturday belongs to Saturn, the usual scholarly guess is mandrake.

Sources and further reading

• Johannes Hartlieb, Das Buch aller verbotenen Künste, 1456, chapter 32
• Apuleius, Metamorphoses III, 21; Homer, Iliad; Ovid and Seneca on the strigae
• Giambattista della Porta, Magiae naturalis, 1558; Johann Weichard von Valvasor, Die Ehre dess Hertzogthums Crain, 1689
• Müller-Jahncke on witch herbs, in the De Gruyter encyclopaedia of medical history, 2005
• Will-Erich Peuckert, Hexensalben, 1960; pharmacological and toxicological standard literature on tropane alkaloids
• Robert Beverley, The History and Present State of Virginia, 1705, on the Jamestown weed episode of 1676
• Linnda Caporael, “Ergotism: The Satan Loosed in Salem?”, Science, 1976; Nicholas Spanos and Jack Gottlieb, rebuttal, Science, 1976

Questions about this article? Write to 28@templeofdesire.com. If you have taken any of the plants named here, contact Poison Help at 1-800-222-1222, or 911 in an emergency.