The Flower That Flowers Without Leaves
A 3,500-year history of colchicine — from the Ebers Papyrus to neutrophil biology — and what it means that medicine worked for millennia before anyone understood why.
Written by Seth Collins, Pharm.D.
Updated on May 20, 2026
Drug History
10 viewsIn the autumn of 1793, a French physician named Nicolas-Jean-Baptiste-Gaston Guillon watched a man die from gout medicine. The patient had taken too much. Vomiting came first, then cramping, then a systemic collapse that stretched over three days. Physicians across Europe had watched the same sequence for centuries. They kept prescribing the medicine anyway, because it worked, and because they had nothing else.
The medicine came from a plant. Colchicum autumnale. In English it carries several names: autumn crocus, meadow saffron, naked ladies. That last one comes from the way it flowers, which is genuinely odd. Most plants produce their leaves and flowers at the same time. The autumn crocus separates them by months. Leaves push up in spring, die back before summer, and vanish entirely. Then September arrives, and the flowers come up alone, without foliage, pale violet clusters rising from bare ground that looks like nothing lives there.
Botanists call this hysteranthous flowering. From the Greek for late. The autumn crocus blooms so far out of season that for centuries people thought it was two different plants. The spring leaves seemed unrelated to the autumn flowers. They catalogued them separately. They were wrong about this.
The plant has been in the medical record longer than almost anything else we still use. The Ebers Papyrus, an Egyptian medical text from around 1550 BC, describes a preparation from autumn crocus for swollen joints. That manuscript was already copying older documents when it was written. How much older, nobody knows. What the Egyptians understood clearly was that the dose had to be right. Too little accomplished nothing. Too much killed the patient. They established the margins through observation, which is a careful way of saying that some patients did not survive the process of figuring it out.
Gout was, in that era, almost entirely a disease of the wealthy. It requires excess: organ meats, red wine, rich food, anything high in purines. Subsistence farmers along the Nile were not developing gout. The scribes and priests and nobles who produced the Ebers Papyrus were exactly the population afflicted, and exactly the population with access to physicians. There is something circular about this. Wealthy Egyptians documented a cure for a disease that wealth caused, using a plant that could also kill them if a physician miscalculated.
The Greeks inherited the plant and the knowledge and gave it a name. The genus Colchicum comes from Colchis, a kingdom on the eastern shore of the Black Sea, in what is now Georgia. To the Greek imagination Colchis was remote and dangerous and strange. It was where the Golden Fleece was kept. It was where Medea was from.
Greek mythology gave Medea the plant's darker properties. She was a sorceress and herbalist, daughter of the Colchian king, and several ancient texts place autumn crocus in her hands as a poison. The Greeks were not wrong about the toxicity. Every part of the plant contains colchicine, the alkaloid responsible for both its usefulness and its lethality. The seeds and underground corms carry the highest concentrations. Livestock that graze on the flowers in autumn die. Children who confuse the corms with wild garlic die. The gap between a therapeutic dose and a lethal one is narrow enough that trained physicians still miscalculated.
Dioscorides wrote about the plant in the first century AD, in De Materia Medica, a text that would sit at the center of European pharmacology for fifteen hundred years. He recommended it for joint pain and noted the dosing problem with the precision of someone who had watched it fail. Galen came a century later and was more cautious. He believed the plant worked. He was also frightened of it. That fear was medically reasonable and practically useless to his patients, because the alternatives for acute gout were rest, cold water, and time.
Gout pain in its acute form reads consistently across centuries of description: a burning, crushing sensation in the joint, most often the big toe, severe enough that the weight of a bed sheet becomes unbearable. People took their chances with a dangerous plant.
The Arab world absorbed Greek medicine and improved on it. Ibn Sina, the Persian physician whose Canon of Medicine compiled everything pharmacology knew in the eleventh century, included careful instructions for preparing autumn crocus. He specified sourcing and preparation and the importance of the correct dose. Centuries of accumulated observation informed his writing in ways that Dioscorides had not had access to. He also understood that the plant's potency varied with preparation, which was practical knowledge that saved lives even though the underlying chemistry was still completely invisible to him. The Canon of Medicine was translated into Latin in the twelfth century and became a standard teaching text at European medical schools. Autumn crocus came with it, documented and recommended in Ibn Sina's authoritative hand, crossing from Arabic into Latin into the practical medicine of physicians who had never seen Colchis and had no idea that the plant they were using had been treating the same condition for two thousand years before Ibn Sina was born.
European medicine received the plant back through those translations and through contact during the Crusades. By the sixteenth century autumn crocus preparations were standard across the continent. The knowledge moved through time attached to the plant itself rather than to any theory of what the plant was actually doing. Physicians prescribed it because the physicians before them had prescribed it, and because it worked, and because nothing else did.
This continued for three thousand years.
Nobody knew the mechanism. Nobody had isolated the active compound. Nobody had a conceptual framework for what was happening inside the body when someone took the medicine. They had the observation that it worked and the observation that too much of it killed people. That was the entire knowledge base, and it was enough to keep the plant in continuous medical use from ancient Egypt through the French Revolution.
The isolation of colchicine as a distinct compound came in 1820. Pierre-Joseph Pelletier and Joseph Bienaimé Caventou, French chemists who extracted quinine from cinchona bark the same year, pulled colchicine from autumn crocus corms and identified it as the molecule responsible for the plant's effects. This mattered. For the first time, physicians could administer a known quantity rather than a preparation of variable strength. Dosing became more consistent. Fatalities became rarer.
The molecule was named for the plant, which was named for the place, which was named for a kingdom that had been gone for two thousand years. That chain is its own kind of document. Colchicine carries Colchis inside it, a Bronze Age kingdom on the Black Sea coast, preserved inside the nomenclature of a molecule that French chemists named while the kingdom's ruins sat buried under Georgian soil.
What Pelletier and Caventou could not do was explain how colchicine worked. That question sat unanswered in the pharmacological literature for another hundred and fifty years, while physicians continued prescribing a molecule they could isolate and dose with reasonable confidence, with no idea what it did once it entered the body.
The answer, when it finally came in the 1970s, was strange.
Colchicine does not fight inflammation the way most anti-inflammatory drugs do. It does not inhibit the enzymes that produce inflammatory molecules. It does not block receptors. What it does is more basic than that. It binds to a protein called tubulin, which is one of the primary structural materials of cells. Tubulin normally links together into long fibers called microtubules, which form part of the cell's internal skeleton and serve as the machinery the cell uses to move and divide. Colchicine binds to tubulin and stops that linking. The microtubules cannot form properly.
This is the same mechanism, incidentally, that makes colchicine toxic in overdose. It is not selective. At therapeutic doses it disrupts neutrophil migration. At higher doses it disrupts cell division throughout the body. The gastrointestinal tract, which replaces its lining constantly through rapid cell division, is among the first systems to fail. This is why overdose presents as severe gastrointestinal collapse before progressing to multi-organ failure. There is no antidote. The margin between therapeutic and toxic has nothing to do with the type of cells being affected. It has to do only with the dose.
In gout, this matters because of what actually causes the pain. The uric acid crystals in the joint are the underlying problem, but they do not hurt directly. The pain comes from the immune response to the crystals. Neutrophils, a type of white blood cell, detect the crystals as foreign material and migrate toward them through the joint tissue. To migrate, neutrophils depend on their microtubule structure. Colchicine disrupts that structure. The neutrophils stall. The inflammatory cascade cannot fully develop.
Egyptian physicians in 1550 BC were interrupting this process. They did not know about neutrophils. They did not know about tubulin. They did not know about microtubules. They knew the plant worked on swollen joints, and they knew the dose had to be careful, and that was everything. For three thousand years it was enough.
This is not unusual in the history of medicine. Aspirin was in widespread use before anyone understood it inhibited prostaglandin synthesis. Morphine was prescribed for centuries before researchers identified opioid receptors. Lithium treated bipolar disorder for fifty years before there was any coherent explanation of why. The practice of medicine has always moved faster than its own theoretical understanding. Colchicine is just the oldest version of this pattern.
The plant itself did not produce colchicine for human benefit. It produced it to be left alone. The compound deters insects and grazing animals. That it is useful to humans at sub-lethal doses is a coincidence. The autumn crocus solved a human medical problem accidentally, as a side effect of solving its own evolutionary problem, and humans exploited that solution for three millennia before understanding what they were exploiting. This is worth sitting with. The plant was not trying to help anyone. It was trying to survive in a meadow. The fact that its defense mechanism happens to interfere with neutrophil migration in a way that reduces joint inflammation in gout is not a plan. It is luck, from the human side of it, and pressure, from the plant's side.
Contemporary medicine uses colchicine beyond gout. It treats familial Mediterranean fever, a genetic condition causing recurring fever and abdominal pain. It reduces the recurrence of pericarditis, an inflammation of the membrane surrounding the heart. Oncology researchers have examined it because the mechanism that disrupts neutrophil migration also disrupts cell division, and disrupting cell division is fundamental to cancer treatment. None of the physicians who first used it imagined any of these applications. They found them by following the mechanism once the mechanism was finally visible.
Guillon's patient died because plant preparations were inconsistent. The colchicine content of a given batch varied with the plant's age, the harvest season, and how the preparation was processed. A safe dose from one batch could be dangerous from the next. Pelletier and Caventou's work in 1820 ended this problem, which is why the history of colchicine splits cleanly into two periods. Before 1820, it worked but killed people with some regularity. After 1820, it worked with considerably more predictability. The knowledge of what it was doing had to wait another hundred and fifty years on top of that.
The plant grows across Europe and western Asia. Every autumn it flowers without its leaves, in meadows and at forest edges, in the same pale violet clusters it has always produced. Farmers in regions where it grows consider it a hazard. It has been in the pharmacopoeia of every major medical tradition that ever encountered it. The Roman physician Galen was afraid of it. So was Guillon, probably, watching his patient in 1793.
It was old when Rome was founded.
There is a version of this story organized around danger, around the narrow window between medicine and poison, around the patients who did not survive the three-thousand-year process of calibrating the dose. That version is accurate. There is another version organized around the strangeness of what actually happened: a plant evolved a molecule to repel insects, and that molecule accidentally treated one of the most painful conditions in human medicine, through a cellular mechanism so specific that it took twentieth-century biology to describe it, while physicians across Egypt and Greece and Persia and medieval Europe and revolutionary France kept prescribing it correctly for thousands of years without knowing any of this.
Both versions are true. They are the same story.
The flowers come up every autumn, alone, without their leaves, from ground that looks empty. They have been doing this since before anyone wrote anything down. They will keep doing it.
Colchicine costs between twelve and forty dollars for a month's supply at most pharmacies today, depending on where you fill it. The molecule is the same one Pelletier and Caventou isolated in 1820. The plant knowledge behind it is older than Rome. Pillar Drug Club was built on the premise that a licensed pharmacist should be the one explaining this to you, and that what you pay for a drug that has been off-patent for a century should reflect that. If you take colchicine, or any generic that has been in the medical record longer than most countries have existed, it is worth knowing what you are actually paying for it.
References
1. Ebers Papyrus (c. 1550 BC). Papyrus Ebers digital archive. Karl-Sudhoff-Institut, University of Leipzig. https://papyri.uni-leipzig.de/receive/UBLEipzig_cmi_00000017
2. Dioscorides, P. (c. 50–70 AD). De Materia Medica. Wellcome Collection digitized manuscript. https://wellcomecollection.org
3. Ibn Sina (Avicenna). (1025 AD). Canon of Medicine (Al-Qanun fi al-Tibb). Translated and digitized editions available via the National Library of Medicine History of Medicine collection. https://www.nlm.nih.gov/hmd
4. Pelletier, P.J., and Caventou, J.B. (1820). Examen chimique des plusieurs végétaux de la famille des Colchicacées. Annales de Chimie et de Physique, 14, 69.
5. Borisy, G.G., and Taylor, E.W. (1967). The mechanism of action of colchicine: binding of colchicine-³H to cellular protein. Journal of Cell Biology, 34(2), 525-533. https://doi.org/10.1083/jcb.34.2.525
6. Leung, Y.Y., Yao Hui, L.L., and Kraus, V.B. (2015). Colchicine: Update on mechanisms of action and therapeutic uses. Seminars in Arthritis and Rheumatism, 45(3), 341-350. https://doi.org/10.1016/j.semarthrit.2015.06.013
7. Finkelstein, Y., Aks, S.E., Hutson, J.R., et al. (2010). Colchicine poisoning: the dark side of an ancient drug. Clinical Toxicology, 48(5), 407-414. https://doi.org/10.3109/15563650.2010.495348
The Pillars is a series on the foundational molecules of modern medicine. Written by Seth Collins, Pharm.D.
The medicine came from a plant. Colchicum autumnale. In English it carries several names: autumn crocus, meadow saffron, naked ladies. That last one comes from the way it flowers, which is genuinely odd. Most plants produce their leaves and flowers at the same time. The autumn crocus separates them by months. Leaves push up in spring, die back before summer, and vanish entirely. Then September arrives, and the flowers come up alone, without foliage, pale violet clusters rising from bare ground that looks like nothing lives there.
Botanists call this hysteranthous flowering. From the Greek for late. The autumn crocus blooms so far out of season that for centuries people thought it was two different plants. The spring leaves seemed unrelated to the autumn flowers. They catalogued them separately. They were wrong about this.
The plant has been in the medical record longer than almost anything else we still use. The Ebers Papyrus, an Egyptian medical text from around 1550 BC, describes a preparation from autumn crocus for swollen joints. That manuscript was already copying older documents when it was written. How much older, nobody knows. What the Egyptians understood clearly was that the dose had to be right. Too little accomplished nothing. Too much killed the patient. They established the margins through observation, which is a careful way of saying that some patients did not survive the process of figuring it out.
Gout was, in that era, almost entirely a disease of the wealthy. It requires excess: organ meats, red wine, rich food, anything high in purines. Subsistence farmers along the Nile were not developing gout. The scribes and priests and nobles who produced the Ebers Papyrus were exactly the population afflicted, and exactly the population with access to physicians. There is something circular about this. Wealthy Egyptians documented a cure for a disease that wealth caused, using a plant that could also kill them if a physician miscalculated.
The Greeks inherited the plant and the knowledge and gave it a name. The genus Colchicum comes from Colchis, a kingdom on the eastern shore of the Black Sea, in what is now Georgia. To the Greek imagination Colchis was remote and dangerous and strange. It was where the Golden Fleece was kept. It was where Medea was from.
Greek mythology gave Medea the plant's darker properties. She was a sorceress and herbalist, daughter of the Colchian king, and several ancient texts place autumn crocus in her hands as a poison. The Greeks were not wrong about the toxicity. Every part of the plant contains colchicine, the alkaloid responsible for both its usefulness and its lethality. The seeds and underground corms carry the highest concentrations. Livestock that graze on the flowers in autumn die. Children who confuse the corms with wild garlic die. The gap between a therapeutic dose and a lethal one is narrow enough that trained physicians still miscalculated.
Dioscorides wrote about the plant in the first century AD, in De Materia Medica, a text that would sit at the center of European pharmacology for fifteen hundred years. He recommended it for joint pain and noted the dosing problem with the precision of someone who had watched it fail. Galen came a century later and was more cautious. He believed the plant worked. He was also frightened of it. That fear was medically reasonable and practically useless to his patients, because the alternatives for acute gout were rest, cold water, and time.
Gout pain in its acute form reads consistently across centuries of description: a burning, crushing sensation in the joint, most often the big toe, severe enough that the weight of a bed sheet becomes unbearable. People took their chances with a dangerous plant.
The Arab world absorbed Greek medicine and improved on it. Ibn Sina, the Persian physician whose Canon of Medicine compiled everything pharmacology knew in the eleventh century, included careful instructions for preparing autumn crocus. He specified sourcing and preparation and the importance of the correct dose. Centuries of accumulated observation informed his writing in ways that Dioscorides had not had access to. He also understood that the plant's potency varied with preparation, which was practical knowledge that saved lives even though the underlying chemistry was still completely invisible to him. The Canon of Medicine was translated into Latin in the twelfth century and became a standard teaching text at European medical schools. Autumn crocus came with it, documented and recommended in Ibn Sina's authoritative hand, crossing from Arabic into Latin into the practical medicine of physicians who had never seen Colchis and had no idea that the plant they were using had been treating the same condition for two thousand years before Ibn Sina was born.
European medicine received the plant back through those translations and through contact during the Crusades. By the sixteenth century autumn crocus preparations were standard across the continent. The knowledge moved through time attached to the plant itself rather than to any theory of what the plant was actually doing. Physicians prescribed it because the physicians before them had prescribed it, and because it worked, and because nothing else did.
This continued for three thousand years.
Nobody knew the mechanism. Nobody had isolated the active compound. Nobody had a conceptual framework for what was happening inside the body when someone took the medicine. They had the observation that it worked and the observation that too much of it killed people. That was the entire knowledge base, and it was enough to keep the plant in continuous medical use from ancient Egypt through the French Revolution.
The isolation of colchicine as a distinct compound came in 1820. Pierre-Joseph Pelletier and Joseph Bienaimé Caventou, French chemists who extracted quinine from cinchona bark the same year, pulled colchicine from autumn crocus corms and identified it as the molecule responsible for the plant's effects. This mattered. For the first time, physicians could administer a known quantity rather than a preparation of variable strength. Dosing became more consistent. Fatalities became rarer.
The molecule was named for the plant, which was named for the place, which was named for a kingdom that had been gone for two thousand years. That chain is its own kind of document. Colchicine carries Colchis inside it, a Bronze Age kingdom on the Black Sea coast, preserved inside the nomenclature of a molecule that French chemists named while the kingdom's ruins sat buried under Georgian soil.
What Pelletier and Caventou could not do was explain how colchicine worked. That question sat unanswered in the pharmacological literature for another hundred and fifty years, while physicians continued prescribing a molecule they could isolate and dose with reasonable confidence, with no idea what it did once it entered the body.
The answer, when it finally came in the 1970s, was strange.
Colchicine does not fight inflammation the way most anti-inflammatory drugs do. It does not inhibit the enzymes that produce inflammatory molecules. It does not block receptors. What it does is more basic than that. It binds to a protein called tubulin, which is one of the primary structural materials of cells. Tubulin normally links together into long fibers called microtubules, which form part of the cell's internal skeleton and serve as the machinery the cell uses to move and divide. Colchicine binds to tubulin and stops that linking. The microtubules cannot form properly.
This is the same mechanism, incidentally, that makes colchicine toxic in overdose. It is not selective. At therapeutic doses it disrupts neutrophil migration. At higher doses it disrupts cell division throughout the body. The gastrointestinal tract, which replaces its lining constantly through rapid cell division, is among the first systems to fail. This is why overdose presents as severe gastrointestinal collapse before progressing to multi-organ failure. There is no antidote. The margin between therapeutic and toxic has nothing to do with the type of cells being affected. It has to do only with the dose.
In gout, this matters because of what actually causes the pain. The uric acid crystals in the joint are the underlying problem, but they do not hurt directly. The pain comes from the immune response to the crystals. Neutrophils, a type of white blood cell, detect the crystals as foreign material and migrate toward them through the joint tissue. To migrate, neutrophils depend on their microtubule structure. Colchicine disrupts that structure. The neutrophils stall. The inflammatory cascade cannot fully develop.
Egyptian physicians in 1550 BC were interrupting this process. They did not know about neutrophils. They did not know about tubulin. They did not know about microtubules. They knew the plant worked on swollen joints, and they knew the dose had to be careful, and that was everything. For three thousand years it was enough.
This is not unusual in the history of medicine. Aspirin was in widespread use before anyone understood it inhibited prostaglandin synthesis. Morphine was prescribed for centuries before researchers identified opioid receptors. Lithium treated bipolar disorder for fifty years before there was any coherent explanation of why. The practice of medicine has always moved faster than its own theoretical understanding. Colchicine is just the oldest version of this pattern.
The plant itself did not produce colchicine for human benefit. It produced it to be left alone. The compound deters insects and grazing animals. That it is useful to humans at sub-lethal doses is a coincidence. The autumn crocus solved a human medical problem accidentally, as a side effect of solving its own evolutionary problem, and humans exploited that solution for three millennia before understanding what they were exploiting. This is worth sitting with. The plant was not trying to help anyone. It was trying to survive in a meadow. The fact that its defense mechanism happens to interfere with neutrophil migration in a way that reduces joint inflammation in gout is not a plan. It is luck, from the human side of it, and pressure, from the plant's side.
Contemporary medicine uses colchicine beyond gout. It treats familial Mediterranean fever, a genetic condition causing recurring fever and abdominal pain. It reduces the recurrence of pericarditis, an inflammation of the membrane surrounding the heart. Oncology researchers have examined it because the mechanism that disrupts neutrophil migration also disrupts cell division, and disrupting cell division is fundamental to cancer treatment. None of the physicians who first used it imagined any of these applications. They found them by following the mechanism once the mechanism was finally visible.
Guillon's patient died because plant preparations were inconsistent. The colchicine content of a given batch varied with the plant's age, the harvest season, and how the preparation was processed. A safe dose from one batch could be dangerous from the next. Pelletier and Caventou's work in 1820 ended this problem, which is why the history of colchicine splits cleanly into two periods. Before 1820, it worked but killed people with some regularity. After 1820, it worked with considerably more predictability. The knowledge of what it was doing had to wait another hundred and fifty years on top of that.
The plant grows across Europe and western Asia. Every autumn it flowers without its leaves, in meadows and at forest edges, in the same pale violet clusters it has always produced. Farmers in regions where it grows consider it a hazard. It has been in the pharmacopoeia of every major medical tradition that ever encountered it. The Roman physician Galen was afraid of it. So was Guillon, probably, watching his patient in 1793.
It was old when Rome was founded.
There is a version of this story organized around danger, around the narrow window between medicine and poison, around the patients who did not survive the three-thousand-year process of calibrating the dose. That version is accurate. There is another version organized around the strangeness of what actually happened: a plant evolved a molecule to repel insects, and that molecule accidentally treated one of the most painful conditions in human medicine, through a cellular mechanism so specific that it took twentieth-century biology to describe it, while physicians across Egypt and Greece and Persia and medieval Europe and revolutionary France kept prescribing it correctly for thousands of years without knowing any of this.
Both versions are true. They are the same story.
The flowers come up every autumn, alone, without their leaves, from ground that looks empty. They have been doing this since before anyone wrote anything down. They will keep doing it.
Colchicine costs between twelve and forty dollars for a month's supply at most pharmacies today, depending on where you fill it. The molecule is the same one Pelletier and Caventou isolated in 1820. The plant knowledge behind it is older than Rome. Pillar Drug Club was built on the premise that a licensed pharmacist should be the one explaining this to you, and that what you pay for a drug that has been off-patent for a century should reflect that. If you take colchicine, or any generic that has been in the medical record longer than most countries have existed, it is worth knowing what you are actually paying for it.
References
1. Ebers Papyrus (c. 1550 BC). Papyrus Ebers digital archive. Karl-Sudhoff-Institut, University of Leipzig. https://papyri.uni-leipzig.de/receive/UBLEipzig_cmi_00000017
2. Dioscorides, P. (c. 50–70 AD). De Materia Medica. Wellcome Collection digitized manuscript. https://wellcomecollection.org
3. Ibn Sina (Avicenna). (1025 AD). Canon of Medicine (Al-Qanun fi al-Tibb). Translated and digitized editions available via the National Library of Medicine History of Medicine collection. https://www.nlm.nih.gov/hmd
4. Pelletier, P.J., and Caventou, J.B. (1820). Examen chimique des plusieurs végétaux de la famille des Colchicacées. Annales de Chimie et de Physique, 14, 69.
5. Borisy, G.G., and Taylor, E.W. (1967). The mechanism of action of colchicine: binding of colchicine-³H to cellular protein. Journal of Cell Biology, 34(2), 525-533. https://doi.org/10.1083/jcb.34.2.525
6. Leung, Y.Y., Yao Hui, L.L., and Kraus, V.B. (2015). Colchicine: Update on mechanisms of action and therapeutic uses. Seminars in Arthritis and Rheumatism, 45(3), 341-350. https://doi.org/10.1016/j.semarthrit.2015.06.013
7. Finkelstein, Y., Aks, S.E., Hutson, J.R., et al. (2010). Colchicine poisoning: the dark side of an ancient drug. Clinical Toxicology, 48(5), 407-414. https://doi.org/10.3109/15563650.2010.495348
The Pillars is a series on the foundational molecules of modern medicine. Written by Seth Collins, Pharm.D.
Tags
Issue 001
Drug History
Colchicine
Pharmacology