In the summer of 1772, in a house in Marseille, the Marquis de Sade handed a plate of aniseed sweets to four prostitutes. The pastilles were laced with cantharidin, the powder known for centuries as Spanish fly, and everyone in the room believed it was an aphrodisiac. Within hours two of the women, Marianne Laverne and Marguerite Coste, were vomiting and in serious danger. Coste filed a complaint. By September a court at Aix had sentenced Sade to death, in his absence, for poisoning.
The women lived, and the conviction was later overturned. What did not change is the lesson sitting inside the story. The one substance in the whole history of love potions that reliably does something to the human body is a poison, and it did to those women what it does to anyone: it burned them from the inside. That is the starting point for a subject people would rather keep soft. Humans have brewed, sold, and swallowed love potions for at least three thousand years. Modern medicine can finally sort the short list that works from the long list that does not, and the answer it gives is stranger than either the believers or the debunkers expect.
The oldest trade
Long before anyone could measure a hormone, people were paying for love in liquid form. The classical scholar Christopher Faraone, in his study Ancient Greek Love Magic, gathered the evidence the Greeks left behind: curse tablets and inscribed gems, little bound figures that look like something from a much later century, magical papyri thick with recipes. The Greeks had a whole vocabulary for this. There were violent spells meant to seize a person with uncontrollable eros, and there were the quieter philtra, the affection-potions and drugged foods, more often the tools of women trying to hold a husband’s love than of men.
The church knew the trade well enough to punish it. Around the year 1010, Burchard, bishop of Worms, compiled a handbook for confessors now called the Corrector. Tucked among its questions are the love-charms: the potions brewed to win someone’s affection, the wives who worked their own blood or bodily fluids into a husband’s food to inflame his desire. Each carried its assigned penance. You do not write rules against something nobody is doing.
And the potion sat at the center of the age’s greatest love story. In the legend of Tristan and Isolde, the two drink a love-potion by mistake, meant for Isolde and her husband-to-be, and are bound helplessly to each other for the rest of their lives. The medieval poets could not even agree how long the magic was supposed to last. In Béroul’s version its full force runs three years, in Eilhart’s four, and in Gottfried von Strassburg’s telling it never fades at all. The potion was a device for saying the thing the age found hardest to say plainly: that desire arrives from outside us, uninvited, and does not ask permission.
The Greek word for a love-charm, philtron, is the root of the English name Philter. The same word carried both meanings the whole subject still lives between: a drink that heals and a drink that bewitches.
The one that works, and what it costs
Return to Sade’s sweets, because the pharmacology is worth understanding. Cantharidin comes from blister beetles, the small green insects an old trade ground into powder. It is a genuine chemical weapon at the cellular level. It blocks an enzyme family called the protein phosphatases, and wherever the body concentrates it, the result is blistering and congestion. On the way out, that concentration lands in the urinary tract.
That is the whole trick of Spanish fly. As it is excreted it inflames the genitourinary lining, and in men it can force a prolonged, painful erection, a priapism, with no desire behind it at all. A 2013 report in the Clinical Kidney Journal by Karras and colleagues, reviewing cantharidin poisoning, called it “an old but dangerous precursor of sildenafil,” the drug we now call Viagra. The comparison holds, and it is grim: cantharidin produces the outward sign of arousal by chemical injury, and the dose that irritates sits close to the dose that damages the kidneys and kills. For most of history, the only love potion that did anything worked by hurting the person who took it.
What actually beats placebo
Set the poison aside and the question narrows to a fair one. Take the plants people have trusted for centuries, put them in a proper trial against a dummy pill, and a little does survive.
The strongest case is yohimbine. It comes from the bark of the West African yohimbe tree, and unlike most of this list it has a clear mechanism, blocking a class of adrenergic receptors that otherwise keep blood vessels tight. In 1998 Pittler and Ernst pooled the randomized trials in the Journal of Urology and reached a plain verdict: the benefit of yohimbine for erectile dysfunction “seems to outweigh its risks.” Its purified form, yohimbine hydrochloride, was a prescription treatment for impotence before a small blue pill arrived that year and erased the market. A folk remedy from a tree in Cameroon was, for a while, real medicine.
Saffron has a smaller but genuine record. In a 2012 trial published in Psychopharmacology, men whose sexual function had been flattened by an antidepressant took saffron or a placebo for four weeks. Sixty percent of the saffron group returned to normal erectile function, against seven percent on the placebo. That looks impressive, and it is worth pausing on what did and did not move. Erection improved, and so did satisfaction with intercourse, but desire did not budge, and neither did orgasm. Saffron reached the plumbing and left the wanting untouched.
Ginseng is the plant the sober reviews keep. When Bella and Shamloul surveyed traditional plant aphrodisiacs in 2014, they found what most surveys find: very little human evidence and a great deal of animal research, with ginseng the clearest exception, since it carries at least some supportive trials in men. A 2025 meta-analysis pooling fourteen trials across more than a thousand men found that herbal supplements as a class do beat placebo for erectile function, though the effect is modest and the studies vary widely in quality. The result points in a direction without shouting.
The whole tier has one shape. Where a plant works, it tends to work on the machinery of erection, through blood flow and nerves, and it leaves desire alone. What people actually buy a love potion for is the one thing the working potions do not touch.
The famous names that do not
Now the disappointments, and they are the popular ones.
Maca, the Andean root sold in every supplement aisle, rests on very thin evidence. A 2010 systematic review led by Byung-Cheul Shin, with the noted skeptic Edzard Ernst among the authors, could find only four small randomized trials, 131 people in total, two of them so tiny as to be almost anecdotes. Their conclusion was cautious to the point of a shrug: limited evidence, too little to trust. We have written about maca’s real and fascinating history in its own article; the biography is genuine, the aphrodisiac file is mostly empty.
Damiana, the Mexican shrub brewed into liqueurs and love-teas, tells the same story we told at length in its own piece. Male rats given damiana extract do become more active, and there are plausible chemical reasons why. Humans have never been properly tested. The reputation rests on tradition and a cage full of rodents.
Tribulus terrestris, the current favorite of the gym-supplement shelf, is the sharpest little lesson of the group. A 2020 systematic review of its use for female sexual problems found five small trials, rated the evidence “very low certainty,” and could not even combine them. And in one of the trials that did report a positive result, in women diagnosed specifically with low desire, the plant improved nearly every measured score except the one that mattered: desire itself.
The rest of the famous roster barely reaches even this level. Oysters, chocolate, and the “love molecule” phenylethylamine, figs, asparagus, and the rest live on folklore and on the shape of the food, with no controlled human evidence that eating them changes anyone’s desire. They are worth what a good dinner is worth, which is not nothing and is not chemistry.
The phenylethylamine in chocolate is real, and the brain does release a related compound during attraction. But eaten in food, phenylethylamine is broken down within minutes and never reaches the brain in any meaningful amount. The molecule is right, but the route is hopeless.
The celery that gave itself away
Which brings us to the vegetable that started this line of thought, and to the best small detective story in the whole subject.
The claim is everywhere. Celery, it is said, contains androsterone, a steroid also present in male sweat, so a man who eats it raises his own scent and, without a word, his appeal to women. It appears in men’s magazines, on wellness sites, in a National Geographic piece, all citing “some researchers,” none citing a study you can open.
There is a study you can open. In 2016 a team of Romanian chemists led by Dorina Simedru set out to measure exactly this, publishing in Studia UBB Chemia. They open by naming the folk belief in plain words: that celery is supposed to contain “high quantities” of androsterone. Then they measured it, in three varieties from the local market, with proper laboratory methods. The answer came back at roughly eight to twelve milligrams per kilogram of wet celery, which they themselves translate into about 0.001 percent of the plant, a trace. The one team that weighed celery’s love-chemical had gone looking to confirm the legend, and their own scale refuted it. The myth and its debunking sit in the same paper.
Two more things quietly collapse the claim. It measures a chemical sitting inside a vegetable, and says nothing whatever about whether eating that vegetable changes a person’s smell or anyone’s response to it. No study has ever tested that. And the story leans on a slip between two look-alike names. Androsterone, the weak human metabolite actually found in celery, is not androstenone, the powerful boar pheromone that pig farmers use to detect a sow in heat and that the supplement trade sells in little bottles as a human attractant. The popular articles slide from one word to the other and hope you do not notice.
Even the science underneath is shakier than the confident tone suggests. There is one careful study here: in 2007, Wyart and colleagues showed in the Journal of Neuroscience that women who smelled androstadienone, a genuine component of male sweat, had measurably altered cortisol. A real molecule produced a genuine hormonal change. But a single hormone shift is a long way from attraction, the wider search for human pheromones is littered with results that failed to replicate, and the biologist Tristram Wyatt, reviewing the field in 2015, argued that no human pheromone has ever been properly demonstrated at all. Where the science actually stands, we may not have working pheromones in the way a moth or a pig does. What the wellness aisle claims is that celery will get you a date.
The strongest potion in the room
So the working list is short and unromantic: a tree bark that beat placebo before Viagra retired it, a spice that mends the plumbing but not the wanting, and a root with a handful of trials. Add the one poison that does something by hurting you, and behind all of it a very long list of hopeful foods with nothing to show.
Except that the accounting is not finished, because there is one more thing in every one of these studies, sitting in the group that got the sugar pill, and it is the most powerful force in the entire field.
The numbers are not small. In the pooled trials of sildenafil, the real drug, about eleven percent of men on placebo still reported that most of their attempts at sex succeeded. In women, the effect is larger still. A 2018 meta-analysis of drug trials for female sexual dysfunction found that roughly two thirds of the whole measured benefit of the actual medicines was matched by the placebo. In one careful study, women with arousal and orgasm problems improved significantly on the Female Sexual Function Index after eight weeks of nothing but a dummy pill.
Then the researchers, Andrea Bradford and Cindy Meston, did the thing that turns a curiosity into an explanation. They asked what, in the placebo group, predicted who got better. It was not age, and it was not how bad things had been at the start. The single strongest predictor was plain: the couples who improved were the ones who had more sex during the trial. Signing up, expecting help, and turning attention back toward a stalled part of life had, on its own, started the thing moving again.
That is the real find, and it is not a debunking of anything. Desire is not like blood pressure, a number the body sets on its own and a drug adjusts. It is built partly out of attention and belief, which means it can be reached the same way. A love potion that works because you trust it has not defrauded you into feeling better; it is a treatment aimed at the one organ that has always run the show.
The woman in Marseille who mixed her blood into her husband’s cup, the Greek scratching a name onto a lead tablet, the man today paying for a bottle of pheromones and a bunch of celery are all doing a version of the same thing, and it is not as foolish as it looks. They are trying to bend desire by ritual, and desire, alone among the body’s systems, will sometimes give. The potion in the cup was mostly wine and hope, and the hope was the part that worked.
Sources
Bibliography. The same list is held in the article’s frontmatter for the citation tools that read it programmatically.
- Pittler, M.H. & Ernst, E. (1998). Yohimbine for erectile dysfunction: a systematic review and meta-analysis of randomized clinical trials. Journal of Urology, 159(2), 433-436.
- Modabbernia, A. et al. (2012). Effect of saffron on fluoxetine-induced sexual impairment in men: randomized double-blind placebo-controlled trial. Psychopharmacology (Berlin).
- Bella, A.J. & Shamloul, R. (2014). Traditional plant aphrodisiacs and male sexual dysfunction. Phytotherapy Research, 28(6), 831-835.
- Ho, C.Y., Hsu, C.H. & Chien, T.J. (2025). Herbal dietary supplements for erectile dysfunction: a systematic review and meta-analysis of randomized controlled trials. Journal of Traditional and Complementary Medicine.
- Shin, B.C., Lee, M.S., Yang, E.J., Lim, H.S. & Ernst, E. (2010). Maca (Lepidium meyenii) for improving sexual function: a systematic review. BMC Complementary and Alternative Medicine, 10:44.
- Martimbianco, A.L.C. et al. (2020). Tribulus terrestris for female sexual dysfunction: a systematic review. Revista Brasileira de Ginecologia e Obstetrícia.
- Simedru, D., Naghiu, A., Cadar, O., Dordai, M., Luca, E. & Simon, I. (2016). Determination of androsterone from celery by a new validated LC-MS/MS method. Studia UBB Chemia, 61(3, Tom II), 415-422.
- Wyart, C. et al. (2007). Smelling a single component of male sweat alters levels of cortisol in women. Journal of Neuroscience, 27(6), 1261-1265.
- Wyatt, T.D. (2015). The search for human pheromones: the lost decades and the necessity of returning to first principles. Proceedings of the Royal Society B, 282(1804).
- Karras, D.J. et al. (2013). Poisoning from ‘Spanish fly’ (cantharidin). Clinical Kidney Journal, 6(2), 201-203.
- Moore, R.A., Edwards, J.E. & McQuay, H.J. (2002). Sildenafil (Viagra) for male erectile dysfunction: a meta-analysis of clinical trial reports. BMC Urology, 2:6.
- Bradford, A. & Meston, C.M. (2011). Placebo response in the treatment of women’s sexual dysfunction: a review and commentary. Journal of Sex & Marital Therapy / Journal of Sexual Medicine.
- Bradford, A. & Meston, C.M. (2007). Correlates of placebo response in the treatment of sexual dysfunction in women. Journal of Sexual Medicine.
- Faraone, C.A. (1999). Ancient Greek Love Magic. Harvard University Press.
- Burchard of Worms, Corrector sive Medicus (Decretum, Book 19), c. 1008-1012.
