Claude Bernard’s Legacy: Milieu Intérieur, Experimental Medicine and Curare in Anaesthesia

Claude Bernard died in Paris on 10 February 1878, but three of the things he left behind are still at work today. The first is an idea: that a living body keeps its own inner world — the blood and the fluids that bathe the cells — steady, and that this steadiness is what lets an animal live freely in a changing outside world. A generation later the American physiologist Walter Cannon gave that idea the name it carries now, homeostasis. The second is a method, set out in his 1865 book Introduction à l’étude de la médecine expérimentale: medicine advances by experiment, by comparison and by taking a failed hypothesis seriously, not by authority or grand systems. The third is a substance. The arrow poison curare, which Bernard used in the 1840s and 1850s to show where a nerve hands its signal to a muscle, walked into the operating room in Montreal in January 1942 and changed surgery.

This page follows those three threads forward from Bernard’s lifetime. It looks at how the milieu intérieur became homeostasis and, much later, a founding idea of systems biology; what the Introduction actually argued about controls, averages and mistakes; Bernard’s own theory of how anaesthetics work; the first clinical use of curare in anaesthesia and the 1947 study of whether it touches consciousness; Daniel Bovet’s synthetic curare-like drugs and the neuromuscular blockers that followed; and the research on carbon monoxide that grew out of Bernard’s scarlet-blood experiments. The experiments themselves are told on the page on his poison experiments, and the plant and chemistry of curare on Curare: The Arrow Poison, Its Plants and Its Science.

Table of Contents

  1. The Milieu Intérieur
  2. From Milieu Intérieur to Homeostasis
  3. An Introduction to the Study of Experimental Medicine (1865)
  4. Controls, Averages and the Failed Hypothesis
  5. Bernard’s Theory of Anaesthesia
  6. Montreal, 23 January 1942: Curare Enters the Operating Room
  7. Paralysed but Awake: The 1947 Question
  8. Daniel Bovet and the Synthetic Curares
  9. Modern Neuromuscular Blockers
  10. Carbon Monoxide After Bernard
  11. Key Research Papers
  12. Connections
  13. Featured Videos

1. The Milieu Intérieur

Bernard’s best-known idea grew out of his laboratory work rather than out of philosophy. His studies of the liver — the discovery that the liver makes sugar (reported with the chemist Barreswil in 1848) and his isolation of glycogen, the body’s “animal starch”, in 1857 — showed him an organ that stores and releases sugar to keep the blood supplied. His work on the vasomotor nerves showed nerves that widen and narrow blood vessels and so shift blood and heat around the body. Again and again he found the body acting to hold its internal conditions within limits.

He drew this together in a phrase. Cells, he argued, do not live in the outside world at all. They live in an internal environment — the milieu intérieur, the blood and the fluid around the cells — and the body works to keep that environment constant. In his lectures on the phenomena of life common to animals and plants (given in 1872–73 and published in 1878–79) he put it in the form that has been quoted ever since: la fixité du milieu intérieur est la condition de la vie libre — in the usual English rendering, the constancy of the internal environment is the condition of free and independent life.

Why the idea mattered

The point of the aphorism is the word “free”. An animal whose blood stays warm, salted and sugared within narrow bounds can move through cold and heat, feast and fast, without its cells feeling the change. The historian Frederic Holmes traced how the concept took shape across Bernard’s career and how it fed into later regulatory physiology, and Charles Gross wrote on the same theme for the neuroscience readership in 1998.

It is worth being exact about what Bernard did and did not say. He did not use the word “homeostasis”, and he did not describe the hormones and feedback circuits that later physiologists found carrying out the regulation. What he gave was the principle and the evidence that started the search.

Back to Table of Contents

2. From Milieu Intérieur to Homeostasis

The word came from Harvard. In 1929 Walter B. Cannon published a long review in Physiological Reviews titled “Organization for physiological homeostasis”. Cannon coined the term from the Greek for “same” and “standing” to describe the coordinated processes that keep most of the steady states of the body in place. Historians of physiology, among them Jensen and Puissant in their 2026 account of Bernard’s route to glycogen and Gomes and Engelhardt in their 2014 bicentenary paper, describe homeostasis as the later name for Bernard’s constancy of the internal environment.

Cannon’s contribution was not only a name. He made the idea measurable and searched for the mechanisms — the nerves and glands that correct a drift in the blood. Bernard’s own glycogen work is one of the earliest pieces of that picture: the liver laying down sugar as glycogen and releasing it again is part of what keeps blood glucose within limits. The site’s animation of blood sugar and insulin shows the modern version of that control loop, with the hormones Bernard never knew.

“The first systems biologist”

In 2008 the Oxford physiologist Denis Noble returned to Bernard in a paper titled “Claude Bernard, the first systems biologist, and the future of physiology”. Noble’s argument is that the constancy of the milieu intérieur implies control processes — something must sense a change and act to reverse it — and that recognising this was the first analysis of an organism as a system rather than as a pile of parts. Noble also noted that Bernard foresaw a role for mathematics in biology. In Noble’s reading, the modern effort to model whole cells and organs from their interacting parts is a return to Bernard’s way of thinking after a century dominated by reduction to single molecules.

Back to Table of Contents

3. An Introduction to the Study of Experimental Medicine (1865)

In 1865 Bernard published the book that made him known far outside physiology: Introduction à l’étude de la médecine expérimentale. It is less a textbook than an argument about how medicine can become a science. An English translation by H. C. Greene, with an introduction by the Harvard biochemist Lawrence J. Henderson, appeared in 1927 and is still in print.

Experiment over systems

The book’s central claim is that medicine had been held back by “systems” — grand theories accepted on authority and defended against the facts. Bernard set against them a cycle of observation, an idea that explains it, an experiment designed to test the idea, and a fresh observation of the result. Ideas are necessary, he argued, but they are tools; the experiment has the last word. The historian Sebastian Normandin, writing in the Journal of the History of Medicine and Allied Sciences in 2007, describes the book as marking the end of conventional vitalism — the belief that living things obey a separate “vital force” outside physics and chemistry — and calls Bernard’s own position a “physical vitalism”: living things are distinctive in how they are organised, but every process in them follows physical and chemical law. Normandin also traces the book’s influence on the philosopher Henri Bergson.

The poisons as worked examples

Much of the book’s power comes from Bernard telling the stories of his own research, including the two poison studies. He describes how, given a sample of curare with nothing known of its action, he started with a “physiological autopsy” of a poisoned frog, checking which tissues still worked; and how his first idea about carbon monoxide failed and led him to the real answer. These passages are told in full on the page about his curare and carbon monoxide experiments.

Back to Table of Contents

4. Controls, Averages and the Failed Hypothesis

Three ideas from the Introduction keep turning up in modern discussions of how medical evidence works.

Comparing treated with untreated

Bernard insisted that a treatment can only be judged by comparison. In the epidemiologist Alfredo Morabia’s 2018 reading in the Journal of the Royal Society of Medicine, Bernard argued for comparing patients who received a remedy with patients who did not — without such a comparison, a recovery that would have happened anyway is credited to the remedy. This is the logic of the control group that sits at the heart of the modern clinical trial, written down long before randomised trials became routine.

Explaining variation rather than hiding it

Bernard is often quoted as hostile to statistics, and he did attack the medicine of averages. Morabia argues that the picture is more careful than that: Bernard did not reject statistics, but he refused to make them the foundation of medicine. His objection was to averaging over results that differed and treating the average as the truth. In Morabia’s summary of Bernard’s view, researchers have to explain why their results varied and not hide the variation in averages; a variation is a clue to a cause that has not yet been found. Modern medicine’s interest in why patients respond differently to the same drug — the study of subgroups and of individual variation — runs along the same line.

Learning from a failed hypothesis

The third idea is Bernard’s respect for being wrong. In his own telling of the carbon monoxide work, his first explanation of the scarlet blood of poisoned animals did not survive the test, and he wrote plainly: “My preconceived idea was therefore false.” The failure is what pointed him to the true answer, that carbon monoxide takes the place of oxygen in the blood. For Bernard, an experiment that overturns the experimenter’s idea is not wasted; it is often the most useful experiment of all, and an investigator has to be ready to let the facts defeat a favourite theory.

Back to Table of Contents

5. Bernard’s Theory of Anaesthesia

Ether and chloroform came into surgery in the late 1840s, during Bernard’s early career, and nobody knew how they worked. Bernard took up the question in his later work and looked for a single explanation: one mechanism by which anaesthetics suspend the activity of living matter, temporarily and reversibly.

The anaesthesiologist and historian Misha Perouansky examined this in a 2012 paper in Anesthesiology titled “The quest for a unified model of anesthetic action: a century in Claude Bernard’s shadow”. Perouansky argues that Bernard formulated a unitary paradigm of anaesthetic action about a quarter of a century before the Meyer–Overton theory, which linked an anaesthetic’s potency to how readily it dissolves in fat. In Perouansky’s account, Bernard saw sensitivity to anaesthesia as the mark that separated true life from mere chemistry. As the paper’s title puts it, the search for one unified model of how anaesthetics work spent a century in Bernard’s shadow.

A link with curare

There is a quiet connection between this theory and Bernard’s curare work. His curare experiments had shown a poison that silenced motor nerves while, in his own words, “without affecting the sensory nerves” — a drug that could take away movement without taking away feeling. Anaesthetics, by contrast, took away feeling and awareness. The two kinds of drug were fundamentally different, a distinction that became critical when curare entered surgery.

Back to Table of Contents

6. Montreal, 23 January 1942: Curare Enters the Operating Room

For almost a century after Bernard, curare stayed in the physiology laboratory. One obstacle was the poison itself: natural curare was a crude plant paste of unpredictable strength. That changed in the late 1930s. The explorer Richard Gill returned from the Amazon in 1938 with more than 11 kilograms of curare, a plant source (Chondrodendron tomentosum) was identified, and a standardised extract was produced by a drug company — the story told in 2024 by Elsherbini and Backman from previously unknown correspondence between Gill and the Montreal anaesthetist Harold Griffith. The plant side of that story is on the curare plant and science page.

On 23 January 1942, in Montreal, Griffith and his resident Enid Johnson gave the standardised curare to a young man undergoing removal of his appendix, during general anaesthesia, to relax the muscles for the operation. Griffith and Johnson reported their first series that year in Anesthesiology under the title “The use of curare in general anesthesia”. Raghavendra’s 2002 history of neuromuscular blocking drugs in the Journal of the Royal Society of Medicine records that date as the first clinical use of curare in anaesthesia.

Why it mattered

Before curare, deep surgical relaxation could be reached only with deep levels of anaesthetic, with the risks that came with them. A drug that relaxed the muscles directly let anaesthetists separate three jobs that had been bundled together: unconsciousness, freedom from pain, and stillness of the muscles. The pharmacologist W. C. Bowman, in his 2006 review in the British Journal of Pharmacology, notes that neuromuscular blocking drugs became established as surgical muscle relaxants after 1943 — and that the site of their action had been worked out by Claude Bernard in the mid-19th century.

Back to Table of Contents

7. Paralysed but Awake: The 1947 Question

Once curare was in clinical use, a hard question needed an answer. Did the drug act on the brain? If curare dulled consciousness, it was partly an anaesthetic. If it did not, a patient given curare without enough anaesthetic would be fully paralysed, unable to move or signal, yet still awake.

In 1947 the journal Anesthesiology published a study by Scott M. Smith, Harold O. Brown, James E. P. Toman and Louis S. Goodman titled “The lack of cerebral effects of d-tubocurarine”. As the title states, the authors concluded that d-tubocurarine, the purified active compound of curare, did not have effects on the brain: it paralysed muscle without abolishing consciousness. The study is often retold as a dramatic self-experiment; those details are left out here because the paper’s full text was not checked for this page.

Bernard’s finding, confirmed in people

The 1947 result fits what Bernard had concluded in animals nearly a century earlier, when he wrote that curare acts on the motor nerves “without affecting the sensory nerves”. A curarised animal is paralysed but not anaesthetised. This is why, in modern practice, neuromuscular blocking drugs are given together with anaesthetic drugs rather than in place of them: the blocker produces stillness, and the anaesthetic produces unconsciousness. The modern concern with awareness during paralysis traces back to this same distinction.

Back to Table of Contents

8. Daniel Bovet and the Synthetic Curares

Natural curare had obvious drawbacks: it depended on plant material gathered on the far side of the world, and its strength varied from batch to batch. The answer came from chemistry. In the 1860s Fraser and Crum Brown had already noticed that quaternary ammonium compounds, a family of nitrogen-containing chemicals, produced curare-like paralysis; and in 1935 Harold King isolated tubocurarine from a museum sample of curare. Chemists now had a pure compound to study and a chemical family in which to design new ones.

The Swiss-born pharmacologist Daniel Bovet, working first in Paris and then in Rome, set out to do exactly that. According to Raghavendra’s history, Bovet synthesised gallamine in 1947, the first synthetic neuromuscular blocker to be used clinically, and in 1949 worked on suxamethonium (succinylcholine), a very short-acting blocker that works in a different way. Bovet received the 1957 Nobel Prize in Physiology or Medicine; the citation covered his work on synthetic compounds that block the actions of the body’s own signalling substances on blood vessels and on skeletal muscle. His whole career — including the antihistamines that make up the other half of that prize — is told on the Daniel Bovet page.

The chemical messenger behind the block

Bovet could design these drugs because the chemistry of the nerve–muscle junction had finally been worked out. In 1936 Henry Dale, Wilhelm Feldberg and Marthe Vogt reported that acetylcholine is released at voluntary motor nerve endings — the chemical signal that Bernard’s curare had been blocking all along without anyone knowing it existed. That discovery, and the Nobel-winning work around it, is told on the Loewi and Dale page.

Back to Table of Contents

9. Modern Neuromuscular Blockers

Bowman’s 2006 review sets out how the field grew after Bovet. The neuromuscular blocking drugs fall into two kinds that act at the same place Bernard identified, the junction between motor nerve and muscle.

Nondepolarising blockers

The first kind act the way curare does. They are reversible antagonists at the acetylcholine receptor on the muscle: they occupy the receptor without switching it on, so the nerve’s chemical signal cannot get through and the muscle stays relaxed. Tubocurarine and gallamine were the early members. Later chemists produced families with different speeds and ways of being cleared from the body — Bowman describes the atracurium group, the vecuronium group and rocuronium among them.

Depolarising blockers

The second kind, of which suxamethonium is the main example, works the other way round. It switches the receptor on and keeps it on, so the muscle membrane stays electrically depolarised and cannot respond to fresh signals. The result is again paralysis, but reached by overstimulation rather than by blockade.

Reversing the block

A drug that paralyses also needs a way of ending its effect at the close of an operation. Bowman’s review describes the arrival of a cyclodextrin reversal agent able to bind rocuronium, a newer approach than the older method of boosting acetylcholine at the junction. The site’s animation of the neuromuscular junction shows the signal these drugs interrupt, and the page on myasthenia gravis describes an autoimmune disease that attacks the same receptor.

Back to Table of Contents

10. Carbon Monoxide After Bernard

Bernard’s second great poison study left an open question. He had shown that carbon monoxide takes the place of oxygen in the blood, volume for volume, and that oxygen could not easily drive it out again. In the same book he described carbon monoxide as deadly because it unites more firmly than oxygen with haemoglobin. What he could not do was put numbers on that competition.

Haldane and the laws of combination

That task fell to the Scottish physiologist John Scott Haldane. In 1895 the Journal of Physiology published his long paper “The action of carbonic oxide on man” (carbonic oxide being the older name for carbon monoxide). In 1912 C. G. Douglas, J. S. Haldane and his son J. B. S. Haldane published “The laws of combination of haemoglobin with carbon monoxide and oxygen”, a quantitative study of how the two gases compete for haemoglobin — the competition Bernard had first demonstrated in a tube of blood over mercury. When the biochemist Otto Warburg described his own research on the respiratory enzyme in 1928, he credited Bernard with the discovery of carbon monoxide haemoglobin in the middle of the previous century.

A clinical problem today

Carbon monoxide poisoning remains a common and serious form of poisoning. Lindell Weaver’s 2009 clinical review in the New England Journal of Medicine gives a modern overview of how the poisoning is recognised and treated. The present-day science, including how carbon monoxide shifts the oxygen-binding curve of haemoglobin, is covered on the site’s pages on carbon monoxide poisoning, carbon monoxide as an indoor toxin and the haemoglobin–oxygen curve animation.

Back to Table of Contents

Key Research Papers

  1. Holmes FL. Claude Bernard, the milieu intérieur, and regulatory physiology. Hist Philos Life Sci. 1986;8(1):3-25. PubMed PMID: 3534926
  2. Gross CG. Claude Bernard and the Constancy of the Internal Environment. Neuroscientist. 1998;4(5):380-385. DOI: 10.1177/107385849800400520
  3. Cannon WB. Organization for physiological homeostasis. Physiol Rev. 1929;9(3):399-431. DOI: 10.1152/physrev.1929.9.3.399
  4. Noble D. Claude Bernard, the first systems biologist, and the future of physiology. Exp Physiol. 2008;93(1):16-26. PubMed PMID: 17951329
  5. Jensen J, Puissant C. Claude Bernard’s route to the isolation of glycogen: the journey that changed scientific views on the physiological role of the liver and animal metabolism. Eur J Appl Physiol. 2026;126(2):629-647. PubMed PMID: 41417056
  6. Gomes Mda M, Engelhardt E. Claude Bernard: bicentenary of birth and his main contributions to neurology. Arq Neuropsiquiatr. 2014;72(4):322-5. PubMed PMID: 24760099
  7. Breathnach CS. Claude Bernard and his revelations in physiology. Ir J Med Sci. 2014;183(1):139-46. PubMed PMID: 24297053
  8. Normandin S. Claude Bernard and an introduction to the study of experimental medicine: “physical vitalism,” dialectic, and epistemology. J Hist Med Allied Sci. 2007;62(4):495-528. PubMed PMID: 17576723
  9. Morabia A. Claude Bernard, statistics and comparative trials. J R Soc Med. 2018;111(9):335-336. PubMed PMID: 30226095
  10. Perouansky M. The quest for a unified model of anesthetic action: a century in Claude Bernard’s shadow. Anesthesiology. 2012;117(3):465-74. PubMed PMID: 22801051
  11. Elsherbini N, Backman SB. The connection between Dr. Harold Griffith and Richard Gill: new insights into the history of curare use in anesthesia from previously unknown correspondence. Can J Anaesth. 2024;71(12):1664-1671. PubMed PMID: 38653929
  12. Griffith HR, Johnson GE. The use of curare in general anesthesia. Anesthesiology. 1942;3(4):418-420. DOI: 10.1097/00000542-194207000-00006
  13. Smith SM, Brown HO, Toman JEP, Goodman LS. The lack of cerebral effects of d-tubocurarine. Anesthesiology. 1947;8(1):1-14. PubMed PMID: 20281549
  14. Raghavendra T. Neuromuscular blocking drugs: discovery and development. J R Soc Med. 2002;95(7):363-7. PubMed PMID: 12091515
  15. Dale HH, Feldberg W, Vogt M. Release of acetylcholine at voluntary motor nerve endings. J Physiol. 1936;86(4):353-80. PubMed PMID: 16994763
  16. Bowman WC. Neuromuscular block. Br J Pharmacol. 2006;147 Suppl 1(Suppl 1):S277-86. PubMed PMID: 16402115
  17. Haldane J. The Action of Carbonic Oxide on Man. J Physiol. 1895;18(5-6):430-62. PubMed PMID: 16992272
  18. Douglas CG, Haldane JS, Haldane JB. The laws of combination of haemoglobin with carbon monoxide and oxygen. J Physiol. 1912;44(4):275-304. PubMed PMID: 16993128
  19. Weaver LK. Clinical practice. Carbon monoxide poisoning. N Engl J Med. 2009;360(12):1217-25. PubMed PMID: 19297574

PubMed Topic Searches

  1. Claude Bernard milieu intérieur
  2. Claude Bernard and experimental medicine
  3. History of curare in anaesthesia
  4. History of neuromuscular blocking agents
  5. Carbon monoxide, haemoglobin and Haldane

Further Reading

Back to Table of Contents

Connections