An accepted truism among clinicians and researchers attributes the persistence of the quest for a unitary mechanism of anesthetic action to the lasting influence of Hans Meyer and Ernest Overton. This article presents a different view: the experiments that led to the Meyer-Overton rule were the consequence-not the source-of a unitary paradigm that was formulated by Claude Bernard a quarter of a century earlier. Bernard firmly believed that the sensitivity to anesthesia was a fundamental criterion that separated 'true life' from 'mere chemistry.' Bernard's scientific authority in the context of 19 century natural philosophy is responsible for establishing a unified (i.e., unitary mechanism and universality across life forms) paradigm of anesthetic action. Meyer and Overton's work was targeted at systematizing and solidifying existing knowledge within this paradigm, not at discovering novelty, and its publication did not substantially affect contemporary research. Claude Bernard's paradigm, by contrast, still influences investigations of mechanisms of anesthetic action.

PARADIGMS play a frequently unrecognized (or at least underappreciated) role in the progress of science. In fact, it has been claimed that “normal science” does not occur outside of paradigms.1Although paradigms are essential, they do not a priori  ensure ultimate success at solving the scientific puzzle to which they are being applied. Paradigms must prove utility at some stage to gain acceptance, but over time and in the absence of evolution, they may also thwart innovative thinking. Has the pursuit of a unified mechanism of anesthesia been driven by a cumulative aggregation of knowledge? Or has it been simply the result of an implicit, subconscious, unyielding paradigm?

In 1950, Thomas Butler, M.D. of Baltimore, Maryland†(Associate Professor of Pharmacology, Johns Hopkins School of Medicine, 1910–1993) remarked ‘there is reason to doubt whether the search for a single theory … may not have been a misguided effort. The metaphysical desire for unification … may well have delayed arrival at a more satisfactory solution.’2More than 60 yr later, one may wonder whether our understanding of cognitive neuroscience in general and anesthetic mechanisms in particular would be different now if research into the latter had not been shackled to the pursuit of a unifying theory of anesthetic action.

Inasmuch as history of research into anesthetic mechanisms is mentioned at all in contemporary reviews and textbooks, it is mostly limited to a cursory mention of the somewhat antiquated and obscure terms ‘lipoid’ and ‘unitary.’ Separation is frequently not even attempted, and for many readers these concepts remain synonymous and inextricably linked to the names of Hans Meyer and Ernest Overton, the only names remembered by most practitioners. When viewed from this perspective, research in anesthetic mechanisms appears to have inevitably progressed toward the predominance of lipid theories simply by virtue of the strength of experimental evidence and universal consensus. Their decline, by contrast, appears sudden and surprising. This reductionist representation obscures the historic dynamics of research and deprives us of a diverse and intellectually rich history from an epoch when anesthetics were thought to hold the answer to some of life's most pressing questions.

The purpose of this article is to shine light on the lasting legacy of a paradigm that originated in the intellectual spirit of the 19thcentury and was launched by the authority of one individual. The intent is neither to comprehensively review nor to adjudicate on individual theories of anesthetic action but to extract their common features. In the process, I hope to place the well-known work of Meyer and Overton into its wider historical context, distinct from the keyhole perspective of recent literature, challenging their (only recently acquired) hegemony over our minds where the history of mechanistic research is concerned. I will explore the notion that, from Meyer and Overton and until now, research in anesthetic mechanisms has been largely pursued within a single paradigm. This paradigm was set by the arguably greatest representative of 19thcentury life sciences: Claude Bernard. The unified paradigm was his brainchild—the product of 19thcentury esprit du temps , Bernard's inspiring vision and very limited experimental data.

Only 6 months after Morton's successful demonstration of clinical ether anesthesia (and more than half a century before Meyer and Overton's anesthesia-related publications), the first monograph addressing anesthetic mechanisms was published in Erlangen, Bavaria.3Emil Harless, M.D., Ph.D. student (later Professor of Physiology in Munich, Germany, 1820–1862) and Baron Ernst von Bibra, M.D., Ph.D. (Gentleman-Scientist, 1806–1878) described experiments on 1 human (Harless, the junior partner) and 29 animals (from cats, rabbits, rats, and birds to copulating frogs and a salamander) and daringly formulated the first theory of anesthetic mechanisms. Noting the affinity of anesthetics to fatty tissues, these authors proposed that anesthetics act by eluting fat from the brain. Not surprisingly, the proposed mechanism of recovery from anesthesia was rather vague. This monograph was patriotically hailed as a success of German science in competition with its French counterpart. On its scientific merit, however, it received from the outset mostly skeptical reviews from leading contemporary scientists, who pointed out its many obvious weaknesses. Notably, the authors did not pursue anesthetic mechanisms further. Harless embraced a conventional academic career whereas von Bibra—an eclectic gentleman-scientist—went on to become the father of German ethnobotany. Nevertheless, two decades later, the idea of ‘action by elution’ was revived once again by Ludimar Hermann, M.D., Ph.D. student (later Professor of Physiology and Rector of the University in Zürich, Switzerland, 1838–1914).4He was inspired by the discovery of a new organic compound originally described as ‘cérébrote’ in 1834, later rediscovered by Otto Liebreich as ‘protagon’ and thought to constitute a homogenous chemical compound.5After separation and purification, its noncholesterol, i.e. , phospholipid component entered the lexicon of biochemistry as ‘lecithin.’ Lecithin had been found exclusively in tissues from the living world and hence was thought to provide a suitably exclusive target for the diverse class of pharmacologic agents capable of producing anesthesia. Hermann noted that ether, ethyl alcohol, and chloroform were capable of dissolving neurons as well as erythrocytes. As both were rich in ‘lecithin bodies,’ the conclusion that lecithin might serve as the anesthetic target appeared intuitive. However, as a theory of narcosis, this proposal of anesthetics acting as solvents was as stillborn as its predecessor. The main lasting influence of these early forays by Harless, von Bibra, Hermann, and others was to direct the attention of the international research community (mostly French and German at that time) to the remarkable affinity of many anesthetics to fatty tissues. These modest beginnings certainly do not explain why ‘general anesthesia became the first pharmacological phenomenon to have evoked scientific theorizing in any modern sense.’2To address this question we have to explore the scientific spirit of the era into which anesthesia was born.

The 19thcentury witnessed the beginnings of the emancipation of ‘natural philosophy’ (what we would currently call ‘life sciences’) into the autonomous discipline of biology. The scientific revolution of the 16thand 17thcenturies had not transformed the study of the living in the same way as that of the inanimate world. As one consequence, the accumulated factual knowledge in physiology, anatomy, and botany largely remained part of the realm of medicine, these disciplines being viewed as medicine's auxiliaries. Natural philosophy existed but was pursued either as a hobby or in the service of natural theology and as such was subject to interpretation within the framework of Christian doctrine with its traditional view of life's ultimate causation. In the late 18thcentury, however, natural philosophy became embroiled in a grand discourse about the nature of life itself.6The revolutionary implications of some 19thcentury discoveries (those that would later become the bedrock of modern biology: cells as the universal building blocks of plants and animals; evolution by natural selection; cell cycle, fertilization, embryology) provided ample fuel for controversy for scientists and the educated public alike. The principal opponents in this debate can be categorized as physicalists and vitalists. The physicalists were heirs to the mechanistic Cartesian tradition that modeled all living beings (with the notable exception of man) as mere machines. The corollary of this view was that by following a strictly reductionist strategy, i.e. , by disassembling the machine into its constitutive parts, a complete understanding of the animal-machine as a whole could be gained, the whole being just the sum of its (physical and chemical) parts. The goal of science was to reduce all of biology to the laws of chemistry and physics which, in contrast to the nascent field of biology, were considered true sciences.6The vitalists, by contrast, postulated the existence of either a special substance exclusive to organisms, a special state of matter or a special force that provided that extra spark that separated living from inanimate matter. This substance was thought not to be reducible to physics and chemistry which, at that time, consisted mostly of mechanics and inorganic chemistry, respectively.

Vitalists attempted to advance a scientific (rather than a metaphysical or theological) argument that would capture the obvious deficiencies of a purely mechanistic/physicalist explanation of life. Although supporters of the ‘vital force’ (variably referred to as entelechie, élan vital, Lebenskraft) had diverse views on the nature of that force, most agreed that its invisibility neither implied a supernatural explanation nor precluded its study by scientific means. The analogy to Newton's gravity was quite natural.6More materialistically oriented vitalists perceived the ‘colloidal state’ of matter as something unique to the living world, inaccessible to reductive analysis with standard physicochemical tools and a likely candidate for life's essence. Others thought that living organisms were unique carriers of a special substance—e.g. , the newly rediscovered ‘protagon’ or the protoplasm thought to be a single substance ‘colloidal’ in nature—which was supposed to carry the essential life-endowing properties. It should be recalled that the notions of the components and structure of protoplasm were fairly vague—even at the turn of the 20thcentury Ernest Overton still perceived the protoplasm as an emulsion or a ‘swollen mixture’ (aufgequollenes Gemisch ) of lecithin, cholesterol, and water.7The seeds of research into the mechanism of anesthesia fell on fertile intellectual soil.

Claude Bernard (fig. 1) is best remembered for coining the term ‘milieu intérieur’ (now known as homeostasis) but ‘his research contributions pervade every field of modern medicine.’8In contemporary anesthesia literature, only his work on the neuromuscular junction and the discovery of the vasomotor activity of the autonomic nervous system is mentioned at best. His critical influence on research into mechanisms of general anesthesia is forgotten.

Fig. 1. Nineteenth century: searching for the essence of life. Claude Bernard (A ) formulated a unified paradigm of anesthetic action rooted in his vision that susceptibility to anesthesia separated the living from the nonliving worlds. This paradigm had a profound influence on research into anesthetic mechanism for more than a century. Albert Dastre (B , a contemporary cartoon) emphasized the philosophical importance of Bernard's life-defining vision for anesthetic susceptibility. He authored numerous works, among them a textbook of anesthesia (Les Anesthésiques ) published in 1891 in which he described anesthesia as ‘the reagent of life’ (reactif de la vie ). (C ) Course of physiology taught by Dastre at the Sorbonne (a college of the University of Paris), where Claude Bernard became Professor and chaired the Department of Physiology from 1854 to 1868 when he left it for the newly established professorship in general physiology at the Museum of Natural History. Albert Dastre became chair of physiology at the Sorbonne in 1887. The image of Claude Bernard is courtesy of the Wood Library-Museum of Anesthesiology, Park Ridge, Illinois. Albert Dastre's cartoon is copyright Bibliothèque de l'Académie nationale de médecine. The image of Dastre lecturing at the Sorbonne is courtesy of the Bibliothèque Interuniversitaire (BIU) Santé, Paris, France.

Fig. 1. Nineteenth century: searching for the essence of life. Claude Bernard (A ) formulated a unified paradigm of anesthetic action rooted in his vision that susceptibility to anesthesia separated the living from the nonliving worlds. This paradigm had a profound influence on research into anesthetic mechanism for more than a century. Albert Dastre (B , a contemporary cartoon) emphasized the philosophical importance of Bernard's life-defining vision for anesthetic susceptibility. He authored numerous works, among them a textbook of anesthesia (Les Anesthésiques ) published in 1891 in which he described anesthesia as ‘the reagent of life’ (reactif de la vie ). (C ) Course of physiology taught by Dastre at the Sorbonne (a college of the University of Paris), where Claude Bernard became Professor and chaired the Department of Physiology from 1854 to 1868 when he left it for the newly established professorship in general physiology at the Museum of Natural History. Albert Dastre became chair of physiology at the Sorbonne in 1887. The image of Claude Bernard is courtesy of the Wood Library-Museum of Anesthesiology, Park Ridge, Illinois. Albert Dastre's cartoon is copyright Bibliothèque de l'Académie nationale de médecine. The image of Dastre lecturing at the Sorbonne is courtesy of the Bibliothèque Interuniversitaire (BIU) Santé, Paris, France.

Close modal

In the essence-of-life-debate, Claude Bernard was unambiguously in the vitalist camp. He considered the ‘protoplasm’ as the life-harboring substance for all living forms. Between 1854 and his death in 1878, Bernard published seven volumes of ‘Lessons,’ which formed the backbone of the Course of Medicine at the Collège de France where he held the appointment of Professor of Medicine. Two additional volumes, summarizing his Course of General Physiology at the Museum of Natural History where he held the appointment of Professor of General Physiology, were published in 1878 shortly after his death. Bernard's opinion was very influential: with his appointments he determined to a large degree what was taught to physicians and other parties interested in life sciences at these European academic beacons. In two of these nine volumes titled ‘Lessons on the anesthetics and on asphyxia’ (‘Leçons sur les anesthésiques et sur l'asphyxie' )9and ‘Lessons on the phenomena of life shared by animals and plants’ (‘Leçons sur les phénomènes de la vie commun aux animaux et aux végétaux ')10Bernard developed his unitary view of life as defined by the susceptibility to anesthesia, in itself a unitary phenomenon: all anesthesia can be reduced to the same essence; there are different anesthetic agents but only one anesthetized state (Toutes les anésthesies se réduisent au même phénomène intime; Il y a des agents anésthesiques divers, mais il n'y a qu'une anésthesie ).9Reproduced in figure 2is figure 23 from the latter volume. Identical conditions for experiments on respiration, arranged in parallel for a rodent and a plant, illustrate Bernard's commitment to a unified perspective on the essence of life, characterized by the property of ‘irritability.’ It is this ‘irritability’ that is abolished by anesthetics. In a bold stroke of deductive reasoning, Bernard then argues that susceptibility to anesthesia defines ‘life,’ unambiguously separating it from ‘mere chemistry,’ which is resistant to anesthesia. Bernard was also convinced that the protoplasm was the indivisible seat of ‘irritability,’ common to all kingdoms of life, a view that dovetailed into his theory of anesthetic mechanisms (see next paragraphs). He supports this view by narrating anesthesia of a plant (mimosa). Its manifestations are graphically illustrated in figures 19 and 20 of the same volume, reproduced here in figure 3.

Fig. 2. Parallel arrangement of an experiment for mammal and plant, illustrating Bernard's vision that the essence of life, which he thought was embodied in the protoplasm, was shared and comparable across kingdoms. Reproduced from Bernard C: Leçons sur les phénomènes de la vie commun aux animaux et aux végétaux, Librairie J-B Baillière et Fils, 1878. Available in the public domain at http://books.google.com/.

Fig. 2. Parallel arrangement of an experiment for mammal and plant, illustrating Bernard's vision that the essence of life, which he thought was embodied in the protoplasm, was shared and comparable across kingdoms. Reproduced from Bernard C: Leçons sur les phénomènes de la vie commun aux animaux et aux végétaux, Librairie J-B Baillière et Fils, 1878. Available in the public domain at http://books.google.com/.

Close modal

Fig. 3. Claude Bernard's illustration of the susceptibility of plants to anesthesia. (A ) Experimental arrangement and illustration of the anesthetized phenotype of the plant Mimosa pudica  unable to respond to mechanical stimuli. Note the ether-soaked sponge. (B ) Normal response to mechanical stimulus: contraction and lowering of the leaves. Reproduced from Bernard C: Leçons sur les phénomènes de la vie commun aux animaux et aux végétaux, Librairie J-B Baillière et Fils, 1878. Available in the public domain at http://books.google.com/.

Fig. 3. Claude Bernard's illustration of the susceptibility of plants to anesthesia. (A ) Experimental arrangement and illustration of the anesthetized phenotype of the plant Mimosa pudica  unable to respond to mechanical stimuli. Note the ether-soaked sponge. (B ) Normal response to mechanical stimulus: contraction and lowering of the leaves. Reproduced from Bernard C: Leçons sur les phénomènes de la vie commun aux animaux et aux végétaux, Librairie J-B Baillière et Fils, 1878. Available in the public domain at http://books.google.com/.

Close modal

Albert Dastre (Professor of General Physiology, Sorbonne, Paris, 1844–1917, fig. 1B and C), a close associate and mentee of Bernard, provides the following summary in his book ‘Anesthetics’ (‘Les Anesthésiques ') published in 1891: ‘the phenomena that ether abolishes (sensitivity, movement, secretory activity, assimilation) are those phenomena that are true characteristics of vitality; ether respects those which, even though necessary to maintain life, are physical and chemical like digestion and respiration’(‘Les phénomènes que l’éther abolit, la sensibilité, le movement, les sécrétions, l'assimilation, sont les phénomènes véritablement charactéristique de la vitalité; il respecte ceux qui, bien que nécessaire pour l'entretien de l'existence, tels que la digestion et la respiration, sont d'ordre physique ou chimique' ).11Dastre concludes that anesthesia is the ‘reagent of life’ (reactif de la vie ) and that the action of anesthetics establishes a dichotomous classification: all that is resistant falls into the category of mechanics, all that submits qualifies as life. Dastre modestly concludes that it is unnecessary to point out the philosophical value of a criterion that permits a clear separation of what is immanent and essential to the living from what has been simply adopted from the physical world (‘Il n’est pas besoin d'insister sur la valeur philosophique d'un tel critérium, qui permet de séparer ce que la nature vivante a d'immanent et d‘essentiel d’avec ce qu'elle emprunte à la nature physique' ).11Not surprisingly, once formulated, this paradigm influenced the interpretation of experiments: ether and chloroform suppressed fermentation of sugar to alcohol by Saccharomyces cervisiae  and hence fermentation was a true manifestation of life; enzymatic digestion of polysaccharides into fermentable glucose, by contrast, was resistant and hence belonged to the realm of chemistry.

In addition to this transcendental ‘property’ of anesthesia, Bernard also proposed an explicitly unitary theory of anesthetic action (a similar mechanism was concurrently but independently proposed by Carl Binz). According to this theory, anesthetics induced a partial and reversible coagulation of the protoplasm, i.e. , an alteration in its ‘colloidal state.’ All the higher-order manifestations of anesthesia that could vary from organism to organism were secondary to the effect on the protoplasm. The coagulation was reversible—but only up to a certain point: if the coagulation was allowed to progress beyond a threshold it became irreversible and deadly, expediently explaining clinical accidents. Bernard also thought that the same alteration of the protoplasm could be caused by anemia, asphyxia, and heat. The quantitative aspects of these experiments are not mentioned in the ‘Lessons’: Bernard thought his conviction of having found an absolute criterion to separate the living world from the inanimate to be sufficient evidence.

In summary, Claude Bernard's legacy to research in anesthesia was twofold. On one hand he proposed a specific theory of anesthetic mechanism, based on experiments and therefore testable and falsifiable. Not surprisingly, his ‘protoplasmic coagulation’ or ‘colloidal’ hypothesis (referring to the component of the protoplasm that was supposed to coagulate) was soon challenged but, rebranded as ‘gelatinization,’ remained defensible well beyond the second World War.12On the other hand and more importantly, Bernard surreptitiously established a paradigm that would remain essentially unchallenged until the 1990s. By bestowing a fundamental life-defining feature to a heterogeneous group of agents he postulated a unifying quality that became a sine qua non  criterion for testing hypotheses: any proposed mechanism had to be unitary (i.e. , valid for all agents) and universal (i.e. , applicable to all life forms). This unifying paradigm, without attribution to him, will reverberate through many editions of textbooks, dictate the type of questions asked by generations of scientists (Meyer and Overton included) and, most importantly, determine how new data will be interpreted and incorporated into the accepted body of knowledge.

Bernard's paradigm left the exact mechanism of anesthesia open to debate and in the decades after his death its exploration continued fervently. To illustrate the profound influence of Bernard's unitary paradigm (as opposed to his specific coagulation theory) on the development of anesthetic research, I will briefly examine two frequently quoted books published at the end of the first World War. These books, one by James Tayloe Gwathmey, M.D. (first President of the American Association of Anesthetists and a great promoter of ‘colonic anesthesia,’ New York, New York, 1863–1944) and Charles Baskerville, Ph.D., F.C.S. (Professor of Chemistry, College of the city of New York, New York, 1870–1922), the clinical textbook ‘Anesthesia,’13the other by Hans Winterstein, M.D., Ph.D. (Professor of Physiology, University of Rostock, Germany, 1879–1963), and the more researcher-oriented ‘Narcosis in its significance for general physiology’ (‘Die Narkose in ihrer Bedeutung für die Allgemeine Physiologie’ ),14were published in the United States of America and in Germany in 1918 and 1919, respectively.‡From different angles (clinician vs.  scientist), these books provide summaries and the beginning of historical perspectives on the scope of research by two generations of post-Bernard scientists.

In Gwathmey's 900-page book, coauthored with Charles Baskerville, Ph.D., fully 24 pages are dedicated exclusively to mechanistic theories of anesthesia (for comparison, 33 pages are dedicated to theory and practice of ‘colonic’ anesthesia). Not fewer than 18 theories are listed using the names of their principal proponents (Bernard, and Meyer and Overton among them). The section begins somewhat smugly (‘Inasmuch as there are still ‘mysteries of anesthesia’…) and goes on to describe the theories in appropriate detail. Only the theory of Moore and Roaf is singled out because Gwathmey and Baskerville think it is the one most likely to be correct. In this context, it is interesting that Moore and Roaf dismissed lipids and favored proteins as anesthetic targets on the grounds that proteins, as opposed to lipids, were found in both animal and plant tissue, i.e. , were more universal.15 

Hans Winterstein's book marked, as the title indicates, a golden age of anesthetic mechanisms research. Of the 319 pages, 180 discuss mechanistic theories of narcosis. The 720 references, mostly to original publications, testify to the enormous interest of the scientific community in anesthesia and its relationship to physiology. It is worth noting that Winterstein arbitrarily compares results obtained in completely different preparations and experimental conditions under a wide variety of agents (Gwathmey lists approximately 600 anesthetic drugs or combinations). In the spirit of Bernard's unifying paradigm, he then uses these comparisons to support or discredit specific theories. Winterstein's approach is more systematic and rigorous than Gwathmey's. Instead of listing theories with partially overlapping ideas separately by author, Winterstein organizes the theories into three chapters according to fundamental mechanisms. The third and longest chapter is titled ‘Physicochemical Theories of Narcosis’ (Physicalisch-chemische Theorien der Narkose). It contains five separate general theories: the lipoid theory, with many contributors but anchored on Meyer and Overton; the adhesive pressure theory (Haftdrucktheorie) popularized mostly by J. Traube; the molecular binding theories—Gwathmey's favored Moore and Roaf are to be found in this group; the coagulation or colloidal theory by Bernard, Binz, and others and, last, the permeability theories that include Ralph Lillie's contemporary sounding ‘reduction of the permeability of the semipermeable plasma membranes of irritable tissues.’ Each of these five general theories is further subdivided into multiple more specific theories. It is noteworthy that, although due space is given to the lipoid theories and the contributions of Meyer and Overton are discussed in detail, Winterstein ends the section with a stringent critique and dismisses lipoid theories as fundamentally flawed due to lack of universality (!) and inherent contradictions.

The first and second chapters are titled ‘Special theories of brain narcosis’ (Spezielle Theorien der Hirnnarkose) and ‘Suffocation theory of narcosis’ (Erstickungstheorie der Narkose), respectively. The former group infers a, surprisingly modern-sounding, special sensitivity of the brain to anesthetics due to specific effects on neurons. The ‘dendritic theory’ attributed higher cognitive function to changing connections between neurons enacted by amoeboid mobility of the dendrites. Anesthetics supposedly immobilized the ‘dendrites’ thereby interrupting the normal connectivity between neurons, ‘paralyzing’ the central nervous system (resonates with current developmental anesthetic toxicity hypotheses). The other theory in this chapter exploits analogies between natural and drug-induced sleep. Although this analogy is obvious, the complete lack of understanding of sleep physiology at that time precluded a productive development of this hypothesis. In any case, both ‘special theories’ represent a partial departure from Bernard's paradigm—the unitary aspect is maintained but universality is limited to organisms with brainlike neuronal networks and/or sleeplike behavior.

The essence of the ‘suffocation theory’ (Erstickungstheorie) was anesthesia-induced inhibition of cellular oxygen ‘assimilation’ (details of mitochondrial function were largely unknown), i.e. , cellular respiration. Numerous research groups, well into the 20thcentury, followed the lead of Max Verworn in approaching anesthetic mechanisms assuming a disturbance of energy metabolism. Among them was Otto Warburg, the laureate of the 1931 Nobel Prize in Physiology and Medicine for his work on (not surprisingly) respiratory chain enzymes. This group of theories was firmly unitary and universal as enzyme preparations, unicellular organisms, germinating wheat, yeast, sea urchins, and mammalian brain were all used to test it. Mitochondrial energy effects also remain a part of the current anesthetic neurotoxicity debate.

As the novelty of anesthesia wore off and it became part of routine clinical practice, the number of original new theories dwindled. Publications revolved around existing theories that were either supported with new data in the original or somewhat altered form or criticized. Most theories of the pre-World War I era were reexamined within the same paradigm of a unitary mechanism and many investigators attempted to support their favored theory by demonstrating universality across phyla.

Plant cells continued to be used16but the difficulty of extrapolating from them to the animal kingdom portents their decline as models for anesthetic mechanisms. Kurt H. Meyer, Hans Meyer's son, published additional experiments in the 1930s reviving and refining Meyer and Overton's data. His innovation was the use of oleic acid as a model of the lipoid target (the use of olive oil, e.g.  by his father, was a popular target for criticism) and determined that the same concentration of any narcotic agent must be present in certain lipoids for narcosis to ensue. Bancroft and Richter revived Claude Bernard's colloidal-coagulation theory.17Juda H. Quastel further refined the ‘suffocation theories’ describing anesthesia as ‘histotoxic anoxia.’18It is therefore not surprising that ‘The Theories of the Action of General Anesthetics’ chapter in the second edition of Gwathmey's Anesthesia, published in 1929, remained largely unchanged and closed with the same support for the theory of Moore and Roaf as in the first edition. Even the novel sounding ‘electric theory,’ proposed by R. Beutner in 193119was in essence an advancement of Lillie's early permeability theories published in 1913.20,21As emphasized by Beutner, this theory was paradigm-compliant as it rationally explained both the increased irritability and the depression (at low and high concentrations, respectively) by concordant effects on cell negativity and impulse propagation along nerves. During these years, the scientific literature saw an increasing number of publications reporting ‘transient increases in irritability’ under low anesthetic concentrations. These observations originated from an exotic variety of models (and correspondingly bizarre phenotypes of ‘irritability’): bacteria (growth), yeast (metabolism), amoeba (movements), ciliated epithelium from various species, infusoria (violence), and limulus (heart beat) as well as other species.22Under Bernard's unified paradigm, increased irritability had to fit into the same mechanistic matrix as the anesthetized state. This necessity generated increasingly arcane theoretical explanations. With disarming simplicity, macroscopic changes in behavior were directly correlated with microscopic changes in any of the observed parameters. In Beutner's case, the manifold presentations of ‘irritability’in vivo  were seen as resulting directly from proportionate changes in permeability in vitro : increased and decreased permeability led to excitation and suppression, respectively. One more of Bernard's ideas celebrated a renaissance—the similarity between heat and anesthetics: research in bacterial luminescence, presaging important work to come a half- century later, proposed ‘reversible denaturation’ of proteins (coagulation of protoplasm) induced by heat and anesthetics as a physical mechanism of narcosis.23 

A review in the British Medical Journal  in 1946, marking the centennial of the Ether Dome event and laconically titled ‘Theories of Anesthetic Action,’ confirms the impression of a largely ‘steady state’ in a mature field of anesthetic theories between the two world wars, i.e. , a period of ‘normal science’ (in T. Kuhn's sense) marked less by novelty than by increased accuracy of measurement, quantity of observations, and scope of experiments in support of existing ideas.1For the purposes of this paper, however, the discussion in Burn and Epstein's review,24written with the benefit of a long-term perspective, is worth a closer look. Burn and Epstein explicitly recognized that the work by Meyer and Overton shouldn't be called a theory at all but merely a rule that fits easier into the framework of some theories than others. Interestingly, they did not consider congruence with the Meyer-Overton rule as an absolute requirement for a theory of narcosis to be valid. By contrast, they did explicitly fold the relatively isolated observations of a concentration-dependent increase and decrease in permeability and excitability into the unified principle by stating that ‘any theory of narcosis which is to be satisfactory must explain them.’

In the first decades of the 20thcentury, a plethora of original theories were competing with each other (almost) all framed in the spirit of the unitary-universalist paradigm set by Bernard. Those that were even marginally straying from one component of the paradigm were appropriately labeled as noncanonical, e.g. , the ‘special  theories of brain narcosis' in Winterstein's book. For contemporaries, experiments by Meyer and Overton fit very much into the pattern of what T. Kuhn calls ‘science as usual’1and were not burdened by the singular importance that would be attached to them much later in the 20thcentury. The reasons are easily understood: neither the experiments of Meyer nor those of Overton were aimed at discovering a new phenomenon or at overthrowing an existing theory. They sought to provide a firmer, more quantitative and more precise base for one of the established observations, namely the affinity of anesthetics to lipids. Their achievement was that by experimenting with unprecedented meticulousness and systematic quantification they increased the accuracy and credibility with which this fact became known. To do so they needed a framework to plan the experiments and to interpret the results. To illustrate how a paradigm affects the interpretation of data, I juxtapose a graphic rendition of the original findings25from which Hans Meyer derived the famous rule26and contemporary data illustrating exceptions to the Meyer-Overton rule27(fig. 5). It is obvious that substances that currently are considered to be exceptions are actually following the rule more closely than those from which the rule was deduced in the first place. The reason Meyer and Overton were able to see a correlation was that they worked within a conceptual framework (unified) and with a literature-supported target (lipids). They were primed to see regularity in their data. A scientist approaching the same data within a different paradigm (e.g. , all agents work by different mechanisms) might conclude that the data points are too scattered to support any regularity, much like we consider the data in fig. 5B as illustrating exceptions to the Meyer-Overton rule.

Fig. 4. The 20th century: compliance with Bernard's paradigm and early skepticism. Hans Horst Meyer (A ) and Charles Ernest Overton (B ), working within the unified paradigm, formulated the eponymous rule at the turn of the century. Although ubiquitously quoted today, the Meyer-Overton “rule” had only a limited effect on theories of anesthetic action until late in the 20th century. To the best of my knowledge, Thomas C. Butler (C ) was the first to publicly question the unified paradigm2but, judging by the trajectory of research in the decades following this publication, his critique had little effect. The images of Hans Horst Meyer and Charles Ernest Overton are courtesy of the Wood Library-Museum of Anesthesiology, Park Ridge, Illinois. HHM is pictured many years after his well-known publication of 1899. The picture of Thomas C. Butler was kindly provided by Todd Dawson from the Office of Medical Alumni, University of North Carolina, Chapel Hill, North Carolina.

Fig. 4. The 20th century: compliance with Bernard's paradigm and early skepticism. Hans Horst Meyer (A ) and Charles Ernest Overton (B ), working within the unified paradigm, formulated the eponymous rule at the turn of the century. Although ubiquitously quoted today, the Meyer-Overton “rule” had only a limited effect on theories of anesthetic action until late in the 20th century. To the best of my knowledge, Thomas C. Butler (C ) was the first to publicly question the unified paradigm2but, judging by the trajectory of research in the decades following this publication, his critique had little effect. The images of Hans Horst Meyer and Charles Ernest Overton are courtesy of the Wood Library-Museum of Anesthesiology, Park Ridge, Illinois. HHM is pictured many years after his well-known publication of 1899. The picture of Thomas C. Butler was kindly provided by Todd Dawson from the Office of Medical Alumni, University of North Carolina, Chapel Hill, North Carolina.

Close modal

Fig. 5. Theory largely determines the interpretation of experimental data: anesthetic potency versus  lipid solubility for anesthetics and ‘transitional compounds’. Hans Meyer concluded from the data in A  that lipid solubility correlates with anesthetic potency. The red circles  in B  show experimentally determined anesthetic potencies of transitional compounds considered to disobey the Meyer-Overton rule. Their predicted potency calculated from the Meyer-Overton rule is indicated with blue circles . The substances from which the Meyer-Overton rule was originally deduced (A ) obey the rule less closely than the drugs currently considered exceptions to the rule (B ). To facilitate visual comparison, x and y axes in both graphs cover the same range of orders of magnitude. The regression coefficient (adjusted R-square) and SE are 0.73 and 0.17 versus  0.91 and 0.16 for the data in A  and the measured values in B , respectively. Data in A derived by F. Baum,25one of Meyer's doctoral students, published back-to-back with Meyer's much referenced paper in the series ‘On the Theory of Alcohol Narcosis’ (Zur Theorie der Alkoholnarkose ) in 1899. The data in B  were published in 1994 by D. Koblin et al.  27Data in A  and B  based on different partition coefficients and indices of potency. Data in B  assumes summation of anesthetic fractions with respect to the behavioral endpoint, e.g. , immobility.

Fig. 5. Theory largely determines the interpretation of experimental data: anesthetic potency versus  lipid solubility for anesthetics and ‘transitional compounds’. Hans Meyer concluded from the data in A  that lipid solubility correlates with anesthetic potency. The red circles  in B  show experimentally determined anesthetic potencies of transitional compounds considered to disobey the Meyer-Overton rule. Their predicted potency calculated from the Meyer-Overton rule is indicated with blue circles . The substances from which the Meyer-Overton rule was originally deduced (A ) obey the rule less closely than the drugs currently considered exceptions to the rule (B ). To facilitate visual comparison, x and y axes in both graphs cover the same range of orders of magnitude. The regression coefficient (adjusted R-square) and SE are 0.73 and 0.17 versus  0.91 and 0.16 for the data in A  and the measured values in B , respectively. Data in A derived by F. Baum,25one of Meyer's doctoral students, published back-to-back with Meyer's much referenced paper in the series ‘On the Theory of Alcohol Narcosis’ (Zur Theorie der Alkoholnarkose ) in 1899. The data in B  were published in 1994 by D. Koblin et al.  27Data in A  and B  based on different partition coefficients and indices of potency. Data in B  assumes summation of anesthetic fractions with respect to the behavioral endpoint, e.g. , immobility.

Close modal

The true (lack of) effect of Meyer and Overton's work on their contemporaries, however, is better appreciated if one realizes that their publications, even though widely read, did not transform this field of science (in the long run, Overton's contribution to the role of extracellular Na+and membrane permeability might be his more important legacy). Except for those by Hans Meyer's son, attempts to reproduce correlations between potency and partition coefficients were mostly inconclusive. In fact, in a review article in 1930, Henderson called for an exact repetition of Meyer's experiments because of the lack of reproducibility.15Tellingly, no competing theories were abandoned following their publications—at least not based on their data—and they themselves largely abandoned this field to others. Indeed, for more than half a century, theirs was only one hypothesis among many others. The situation changed only in the late 1960s when ‘the Meyer-Overton rule’ started its gradual ascent to hegemony. This shift coincided with the development of a standard measure of anesthetic depth, the minimum alveolar concentration or MAC.28MAC, while unitary in its conception, need not per se  favor lipid membranes over alternative targets. The elegance of its graphic representation, however, effortlessly correlating an important behavioral endpoint (immobility) with a physicochemical parameter (lipid solubility), paired with unsurpassed clinical utility proved very convincing and may have swayed ‘public opinion’ in favor of lipids and Meyer-Overton. The position of lipids was also buttressed by the fact that nonlipid-based unified theories independently proposed by Stanley Miller29and Linus Pauling30failed to gain traction. When pressure-reversal of anesthesia was rediscovered23,31this puzzling phenomenon also appeared easiest to explain with unitary lipid-based hypotheses, further reinforcing an already fashionable trend (pressure reversal was later shown to be a mechanistic ‘red herring’).32For these ancillary reasons, not because of rigorous experimental proof, lipid membranes continued to gain momentum well into the 1970s. In fact, a direct experimental attempt to decide between Stanley Miller's hydrate hypothesis and lipid theories failed to declare a winner.33However, as in other scientific enterprises, once a specific paradigm is somewhat favored it tends to become reinforced as more and more young scientists become increasingly involved in its detailed elaboration instead of the search for alternatives. Hence, from some point in the 1970s on, the a priori  assumption of most mechanistic publications became that lipids were the primary target of anesthetics. Experiments sought, with ever greater sophistication, to provide support for this theory without challenging it and they always remained within Bernard's unified paradigm. Textbooks published in the United States followed this trend and, from the 1980s, largely stopped referring to any early theories other than Meyer and Overton's. Instead, a retrospectively ‘cleaned up’ history presented lipid theories as the zenith of incremental scientific progress reaching back to Meyer and Overton's original insight.

The crisis of lipoid theories began brewing in the 1940s, i.e. , long before they were universally accepted, with independent laboratories publishing evidence that was either inconsistent with lipids34or indicating proteins as viable alternative targets for anesthetics.35,,37However, these publications as well as sound quantitative tests falsifying lipid theory-based predictions remained largely unnoticed.38,39The tide began to turn only after two scientists framed the problem in ‘familiar’ terms. In their landmark publication, Nick Franks and Bill Lieb40(not unlike Meyer and Overton's work more than eight decades earlier) achieved a breakthrough borne out of methodologic advances and experimental rigor (but without challenging Claude Bernard's unified paradigm). As a result, lipid theories were put on the defensive not because of intrinsic inconsistencies (earlier publications to this effect did not resonate) nor because of falsification resulting from the use of an inherently superior experimental model (firefly luciferase is not much more representative of the brain than is olive oil) but because the causal argument on which the entire lipid construct rested (potency correlates with lipid solubility hence lipids are the target) was shown to be also applicable to a protein. Reminiscent of Bernard's own specific theory of anesthesia, the principle demonstrated by Franks and Lieb (conformity with the Meyer-Overton rule is compatible with a nonlipid entity) proved to be more enduring than the specific mechanism (competition with endogenous ligands) that was proposed in the same publication. Within a matter of years, with the lipid theories retreating and scientific attention redirected, an increasing number of inconsistencies in the lipid theories, heretofore mostly neglected, began to surface.41Predictably, once the tide had turned in their favor, an avalanche of publications ‘verifying’ the protein theory followed. However, verifying one hypothesis does not falsify the alternative. Hence, proponents continued to develop ever more sophisticated and refined lipid theories trying to account for their inconsistencies but finding themselves nevertheless marginalized by the proteinaceous maelstrom.

The establishment of proteins as supreme anesthetic targets, however, did not negate Bernard's fundamental paradigm of a unitary mechanism: with proteins substituted for lipids, the search for a unified theory continues to date. Anesthetic modulation of channel transitions, depression of synaptic release and increased leak currents substitutes for protagon, permeability, and protoplasmic coagulation. The great Claude Bernard casts a long shadow.

Currently, three decades separate us from the publications that inaugurated the ‘reign’ of proteins in the realm of anesthetic research, a reign that all currently active scientists have to contend with. This is roughly as much time as separated each of the other milestones that anchor this article from Harless and von Bibra to Franks and Lieb.

What paradigms do currently frame our thinking? Are we conscious of them? Do they assist or impede progress? Paradigms are two-edged swords: on one hand they help funnel the communal research effort into a specific direction rendering it more efficient at solving identified problems. On the other hand, by ‘searching where the light is bright’ truly unconventional approaches may go unrecognized. In this author's opinion, the most beneficial paradigm in our context is the recognition of the complex nature of the brain. In a complex system, collections of diverse, interconnected, interdependent, and adaptable elements generate bottom-up emergent phenomena on each hierarchical level of organization (i.e. , novel properties that cannot be predicted even from detailed knowledge of the lower-level elements). This insight alone should reduce if not prevent the type of extrapolation errors (jumping from microscale to macroscale) that have plagued the history of research in anesthetic mechanisms over long periods of time. Freed from the dictate of universality, solutions for key features of the anesthetized central nervous system (e.g. , amnesia, lack of awareness, immobility) can be sought individually while remaining rooted in sound neurobiology. On the other hand, complexity does not preclude the possibility that, on some levels of organization (e.g. , synchronization of neuronal networks) shared mechanisms for the different endpoints may emerge. The challenge of complexity, however, also encourages simplification (both justified and erroneous) as well as mystification for the sake of Procrustean hypothesis-conformity in the desire to declare the puzzle as either unsolvable or already solved.42, 44 

On the molecular level, the shift from lipids to proteins has, simply by virtue of the diversity of possible drug-target interactions, fostered pluralism and with respect to certain agents provided answers that lie beyond reasonable doubt. The protein doctrine, however, also harbors the risks of any hegemony as, simply by virtue of its dominance, it quenches unorthodox approaches.45 

For many scientists, anesthetics offer unique tools to approach the brain. Progress in understanding their mechanisms is inextricably linked to and benefits from advances in the neuroscience of higher cognitive functions, an endeavor that unites some of mankind's most brilliant minds. In fact, paraphrasing a famous quote, the essential reason why we see further than our forebearers is that we have more and taller shoulders to stand on.

The author thanks Robert A. Pearce, M.D., Ph.D., Professor and Chair, Department of Anesthesiology, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin, for thorough reading of and thoughtful comments on the manuscript. Thanks also go to the staff of the Ebling Library of the University of Wisconsin, Madison, Wisconsin, and the Woods Library Museum, Park Ridge, Illinois, for their enthusiastic help with researching bibliographic sources. The author also acknowledges the assistance of Estelle Lambert from the Bibliothèque Interuniversitaire Santé, Paris, France.

1.
Kuhn TS: The Structure of Scientific Revolution, 3rd edition. Chicago, IL, The University of Chicago Press, 1996
2.
Butler TC: Theories of general anesthesia. J Pharmacol Exp Ther 1950; 98:121–60
3.
von Bibra E, Harless E: Die Wirking des Schwefelaethers in chemischer und physiologischer Beziehung. Erlangen, Bavaria, Heyder, 1847
4.
Hermann L: Ueber die Wirkungsweise einer Gruppe von Giften. Archiv fuer Anatomie, Physiologie und Wissenschaftliche Medicin 1866; 27–40
5.
Rosenheim O, Tebb MC: The non-existence of ‘protagon’ as a definite chemical compound. J Physiol 1907; 36:1–16
6.
Mayr E: This is Biology. Cambridge, Massachusetts, London, England, The Belknap Press of Harvard University Press, 1997
7.
Overton E: Studien uber die Narkose. Jena, Germany, Verlag von Gustav Fischer, 1901
8.
Larson MD: Miller's Anesthesia, 7th edition. Churchill Livingstone Elsevier, 2010
9.
Bernard C: Leçons sur les anesthésiques et sur l'asphyxie, Librairie J-B Baillière et Fils, 1875
10.
Bernard C: Leçons sur les phénomènes de la vie commun aux animaux et aux végétaux, Librairie J-B Baillière et Fils, 1878
11.
Dastre A: Les Anesthésiques. Paris, France, G. Masson, 1890
12.
Seifriz W: The effects of various anesthetic agents on protoplasm. ANESTHESIOLOGY 1950; 11:24–32
13.
Gwathmey JT, Baskerville C: Anesthesia, 1st edition. New York London D. Appleton Co., 1918
14.
Winterstein H: Die Narkose in Ihrer Bedeutung für die Allgemeine Physiologie. Berlin, Germany, Julius Springer, 1919
15.
Henderson VE: The present status of the theories of narcosis. Physiol Rev 1930; 10:171–221
16.
Lepeschkin WW: Some aspects of the causes of narcosis. Physiol Zool 1932; 5:479–90
17.
Bancroft WD, Richter GH: Claude Bernard's theory of narcosis. Proc Natl Acad Sci U S A 1930; 16:573–7
18.
Quastel JH: Respiration in the central nervous system. Physiol Rev 1939; 19:135–83
19.
Beutner R: A rational theory of narcosis in comparison with the hitherto discussed empirical theories, 22nd Annual Meeting of the American Society of Pharmacology and Experimental Therapeutics. Montreal, CA, J Pharmacol Exp Ther 1931, pp 258
20.
Lillie RS: The physico-chemical conditions of anesthetic action. Science 1913; 37:959–72
21.
Lillie RS: The physico-chemical conditions of anesthetic action. Correlation between the anti-stimulating and the anti-cytolytic action of anesthetics. Science 1913; 37:764–7
22.
Beecher HK: Effects of anesthesia on the nervous system, The Physiology of Anesthesia. 1, Oxford University Press, 1938, pp 20–43
23.
Johnson FH, Brown D, Marsland D: A basic mechanism in the biological effects of temperature, pressure and narcotics. Science 1942; 95:200–3
24.
Burn JH, Epstein HG: Theories of anaesthetic action. BMJ 1946; 2:533–5
25.
Baum F: Ein physikalisch-chemischer Beitrag zur Theorie der Narcotica Archiv f.experiment.Pathol.u.Pharmakol. 1899; 42:119–47
26.
Meyer H: Welche Eigenschaft der Anasthetica bedingt ihre narkotische Wirkung? Archiv f.experiment.Pathol.u.Pharmakol 1899; 42:109–18
27.
Koblin DD, Chortkoff BS, Laster MJ, Eger EI 2nd, Halsey MJ, Ionescu P: Polyhalogenated and perfluorinated compounds that disobey the Meyer-Overton hypothesis. Anesth Analg 1994; 79:1043–8
28.
Eger EI 2nd, Saidman LJ, Brandstater B: Minimum alveolar anesthetic concentration: A standard of anesthetic potency. ANESTHESIOLOGY 1965; 26:756–63
29.
Miller SL: A theory of gaseous anesthetics. Proc Natl Acad Sci U S A 1961; 47:1515–24
30.
Pauling L: A molecular theory of general anesthesia. Science 1961; 134:15–21
31.
Lever MJ, Miller KW, Paton WD, Smith EB: Pressure reversal of anaesthesia. Nature 1971; 231:368–71
32.
Daniels S, Smith EB: Effects of general anaesthetics on ligand-gated ion channels. Br J Anaesth 1993; 71:59–64
33.
Eger EI 2nd, Lundgren C, Miller SL, Stevens WC: Anesthetic potencies of sulfur hexafluoride, carbon tetrafluoride, chloroform and Ethrane in dogs: Correlation with the hydrate and lipid theories of anesthetic action. ANESTHESIOLOGY 1969; 30:129–35
34.
Butler TC: The anesthetic activity of optical antipodes II. The arabinochloraloses. J Pharmacol Exp Ther 1940; 69:229–35
35.
Featherstone RM, Muehlbaecher CA, Debon FL, Forsaith JA: Interactions of inert anesthetic gases with proteins. ANESTHESIOLOGY 1961; 22:977–81
36.
Ueda I: Effects of diethyl ether and halothane on firefly luciferin bioluminescence. ANESTHESIOLOGY 1965; 26:603–6
37.
Halsey MJ, Smith EB: Effects of anaesthetics on luminous bacteria. Nature 1970; 227:1363–5
38.
Seeman P: The membrane expansion theory of anesthesia: Direct evidence using ethanol and a high-precision density meter. Experientia 1974; 30:759–60
39.
Franks NP, Lieb WR: The structure of lipid bilayers and the effects of general anaesthetics. An x-ray and neutron diffraction study. J Mol Biol 1979; 133:469–500
40.
Franks NP, Lieb WR: Do general anaesthetics act by competitive binding to specific receptors? Nature 1984; 310:599–601
41.
Franks NP, Lieb WR: What is the molecular nature of general anaesthetic target sites? Trends Pharmacol Sci 1987; 8:169–74
42.
Hameroff S: Anesthesia, consciousness and hydrophobic pockets–a unitary quantum hypothesis of anesthetic action. Toxicol Lett 1998; 100–101:31–9
43.
Brown EN, Purdon PL, Van Dort CJ: General anesthesia and altered states of arousal: A systems neuroscience analysis. Annu Rev Neurosci 2011; 34:601–28
44.
Ries CR, Puil E: Mechanism of anesthesia revealed by shunting actions of isoflurane on thalamocortical neurons. J Neurophys 1999; 81:1795–801
45.
Yang L, Sonner JM: Anesthetic-like modulation of receptor function by surfactants: A test of the interfacial theory of anesthesia. Anesth Analg 2008; 107:868–74