In the long history of the study of the nervous system, there have been a number of major developments that involved radical and permanent changes in fundamental beliefs and assumptions about the nervous system and in tactics and strategies for studying it. These may be termed Revolutions in Neuroscience. This essay considers eight of these, ranging from the 6th century BCE to the end of the 20th century.
THE PRE-SOCRATIC PHILOSOPHERS AND THE IDEA OF SCIENCE
The pre-Socratic philosophers were responsible for the very idea of formal science: the idea that the physical and biological universe was governed by consistent and universal laws that were amenable to understanding by human reason. This was a revolutionary change from the previous prevailing view of the universe as a plaything of gods and ghosts acting in an arbitrary and capricious fashion.
The pre-Socratics lived in the 6th–4th century BCE in various Greek city-states. They called themselves “physiologia,” which is best translated as “natural philosophers.” They conceived their inquiries as demanding rational criticism and public debate and involving observation and measurement (systematic experimentation especially in biology would be almost unknown for several centuries).
The earliest pre-Socratics were Thales, Anaximander, and Anaximenes. All three came from the city of Miletus, a Greek city-state in Ionia on the West shore of modern Turkey, and thus are often known as Milesians or Ionian philosophers. There are several likely reason why Miletus was the “cradle of science.” It was a wealthy port city, which brought together products and ideas from Greek, Phoenician, and Egyptian sea traders and caravan merchants from as far as India and China. Power was moving from the landed aristocracy to the merchant classes interested in new techniques and concepts. In this new city-state, there was open debate about the nature of society and the world.
These early natural philosophers sought fundamental principles underlying the universe. For example, Thales thought that water was fundamental, whereas Anaximenes argued for air as the basic element and discussed the various forms it might take. Among the other major pre-Socratics were Heraclitus, Pythagoras, Empedocles, Zeno, and Democritus. Many of them were interested in sensory processes as sources of knowledge.
The first writer to advocate the brain as the site of sensation and cognition was the pre-Socratic Alcmaeon of Croton, a Greek colony in what is now southern Italy. He is said to have written:
The seat of sensations is in the brain. This contains the governing faculty. All the senses are connected in some way with the brain; consequently, they are incapable of action if the brain is disturbed or shifts its position, for this stops up the passages through which senses act. This power of the brain to synthesize sensations makes it also the seat of thought: the storing up of perceptions gives memory and belief, and when these are stabilized you get knowledge (Theophrastus, 1917).
Alcmaeon is reported to have been the first to use dissection as a tool for intellectual inquiry. He dissected the eye and described the optic nerves and chiasm and suggested that they brought light to the brain. He observed phosphenes after a blow to the head and inferred that the eye contained light and that this light was necessary for vision, a view only disproved in the middle of the 18th century.
Democritus is among the other pre-Socratic philosophers who extended Alcmaeon's views on the functions of the brain. He taught that everything in the universe is composed of atoms. The psyche (soul, mind) is made up of the lightest, most spherical, and fastest-moving atoms. Psychic atoms are found mostly in the brain. Slightly cruder atoms are found in the heart, making it the center of emotion, and still cruder ones are in the liver, making it the seat of lust. This trichotomy developed into Plato's hierarchy of the parts of the soul, and through his treatise, “Timaeus” was very influential into the 18th century.
The various pre-Socratic natural philosophers had a variety of views on the nature of the universe, the body, and the brain. What they held in common was the view that these could be understood by the use of observation and reason. This assumption was a revolutionary one, marks the beginning of formal science in the West, and is still the basis of modern science.
This account is based on Gross (1998); the translations of Freeman (1954); the accounts of Longrigg (1993), Lloyd (1970), Sigerist (1961), Sarton (1959), and Farrington (1944); and the appreciations of Popper (1969) and Schrodinger (1954).
GALEN AND THE EXPERIMENTAL INVESTIGATION OF THE NERVOUS SYSTEM
Galen of Pergamon (129–213 CE) was, by far, the most important physician, anatomist, and physiologist of classical antiquity. His ideas were so pervasive that Western medicine well into the 18th century largely saw the structure and function of the body through his eyes. Today, his voluminous writings provide a vivid account of the controversies and achievements of the 600 years of classical biology and medicine.
Galen was the first to carry out systematic experiments on the nervous system, thereby initiating a major revolution in the study of the nervous system. Before Galen, there were only a few isolated examples of systematic experiments such as in acoustics, those of Strato and Hero on vacuums and Ptolemy on optics. In biology and medicine, observation and dissection were the only methods of investigation, unless one wants to consider trying out possible poisons on prisoners as a type of experimentation.
Galen was born in Pergamon, which is now the Turkish city of Bergama, and studied medicine there. His first three medical treaties were published while he was a medical student. (Although most of his writing was lost, the extant ones comprise 22 large volumes, more are discovered every few years and many are still untranslated into English). He then continued his medical studies in several other medical centers including Alexandria, which was a leading center of medical research. He then returned to Pergamon as the physician to the gladiatorial school there where he acquired considerable clinical experience. When the “games” were closed, Galen moved to Rome where he rapidly rose in medical and social circles, eventually serving as physician to four Roman emperors starting with Marcus Aurelius.
Galen made a number of major discoveries, particularly on the anatomy and physiology of the nervous system. He described in detail the course of nine, if not 10, of the cranial nerves (although he grouped them as seven pairs) as well as the sympathetic nerve trunks. He distinguished sensory and motor nerves for the first time and thought that this distinction derived from their source in the brain, a clear statement of Muller's doctrine of specific nerve energies. Galen's descriptions of the gross anatomy of the brain were very accurate, particularly with respect to the ventricles and the cerebral circulation, which are both important in his physiological system. Galen usually presented his dissections as if they were of the human, but in fact, except for the osteological ones, they were invariably of animals, which are usually the ox in the case of brain anatomy, the Barbary “ape” (the macaque M. sylvana) for cranial nerve anatomy, and pigs for vivisection. (He did not like to vivisect monkeys because, he said, of the expressions of their faces when he did so.) It was only very recently when Galen's descriptions were evaluated in terms of the actual species dissected that their great accuracy was recognized (Rocca, 2003).
However, Galen's truly revolutionary work was in carrying out the first systematic experiments on the functions of the nervous system. Galen realized that the spinal cord was an extension of the brain. In his brilliant and systematic experiments on sectioning the cord, he compared the effects of hemi-transection and total transection at different levels and noted that injuries interfered with sensory and motor function below the level of the section, that hemisection affected only one side, and that sagittal section did not produce paralysis. He accurately described the different roles of the spinal nerves in respiration. He even came very close to the Law of Spinal Roots: “The physicians do not even know that there is a special root at the origin of the nerves which are distributed to the entire hand and from which sensation arises; [nor do they know] that there is another root for the nerves moving the muscles” (Galen, 1962).
Galen used piglets in his experiments on brain lesions. He found that anterior brain damage had less deleterious effects than posterior. He viewed sensation as a central process because he knew from his clinical observations and animal experiments that sensation could be impaired by brain injury even when the sense organs were intact. Because animals could survive lesions that penetrated to the ventricles, Galen thought that the soul was not located there but rather in the cerebral substance. He taught that all mental diseases were brain diseases.
Galen's most famous experiment was the public demonstration of the effects of cutting the recurrent laryngeal nerve on squealing in a pig. Although the encephalocentric view that the brain controlled sensation, movement, and cognition remained strong in the Greco–Roman medical community, the opposing cardiocentric view that the heart was the center of sensation and cognition was also active in Rome at this time, being advocated by the Stoic school and its leader Chyrsippus (280–207 BCE). To refute their view that the heart, not the brain, controlled cognition, Galen arranged this public demonstration and invited leading Roman intellectuals on both sides of the question.
Previously, Galen had carried out a series of anatomical and physiological studies on the recurrent laryngeal nerve in a variety of birds and mammals. He showed that cutting this nerve would eliminate vocalization. Because vocalization was seen as reflecting the cognitive activity language, Galen's public demonstration that cutting a nerve originating in the brain would eliminate squealing in a pig was the first and most famous demonstration that the brain controls cognition. It inspired Leonardo to produce a beautiful drawing of the human recurrent laryngeal nerve. Vesalius gave it a prominent place in his great On the Fabric of the Human Body (without mentioning Galen) and repeated the demonstration in his own public lectures in Padua. The Renaissance edition of Galen's works included an engraving of him carrying out the experiment on a huge pig in front of a very distinguished audience.
GALL AND PHRENOLOGY: LOCALIZATION OF FUNCTION IN THE CEREBRAL CORTEX
The revolutionary idea that different regions of the cerebral cortex have different functions begins with Franz Joseph Gall (1748–1828) and phrenology. Before Gall, the cerebral cortex was almost always dismissed as a mere covering or “rind,” which indeed is the translation of the Latin word “cortex.” It was usually drawn resembling intestines following the description of the 2nd century Alexandrian anatomist Erasistratus.
Marcello Malpighi (1628–1694), the discoverer of capillaries, was the first to examine the cortex under the microscope. He saw it as a makeup of little glands or “globules,” and Leuwenhoek (1632–1723) and others followed suit. This was a common view in the 17th and 18th century perhaps because it fits with the much earlier view of Aristotle that the brain was a cooling organ and with the Hippocratic theory that it was the source of phlegm.
The other common view was that the cortex was largely made up of blood vessels; as Frederik Rusch (1628–1731) put it, “The cortical substance of the cerebrum is not glandular, as many anatomists have described it…but highly vascular.” Albrech von Haller (1708–1777), who dominated physiology in the 18th century, also held a vascular view of cortex. He found mechanical and chemical stimulation to be without effect throughout the cortex and declared it completely insensitive.
Given this view of the insignificance and uniformity of the cerebral cortex, the phrenological ideas of Gall and his collaborator J. C. Spurzheim (1776–1832) were indeed revolutionary.
The central aim of phrenology was to correlate brain structure and function. It had five basic assumptions:
The brain was an elaborately wired machine for producing behavior, thought, and emotions.
The cerebral cortex was a set of organs, each corresponding to an affective or intellectual function.
Differences in traits among people and within individuals depend on differential development of different cortical areas.
Development of a cortical area is reflected in its size.
Size of a cortical area is correlated with the overlying skull (“bumps”).
These otherwise reasonable hypotheses had one fatal flaw: the nature of the evidence. Gall and Spurzheim relied almost entirely on obtaining supportive or confirmatory evidence. They collected large numbers of skulls of people whose traits and abilities were known, examined the heads of distinguished savants and inhabitants of mental hospitals and prisons, and studied portraits of the high and low on various intellectual and affective dimensions. Throughout, they were seeking confirmation of their initial hypothesis usually deriving from a few cases. For example, the idea for a language organ in the frontal lobes came from Gall's experience of a classmate who had a prodigious verbal memory and protruding eyes (being pushed out by a well-developed frontal lobe, Gall thought). The idea for an organ of destructiveness came from the skulls of a parricide and of a murderer that were sent to him, from noticing its prominence in a fellow medical student who “was so fond of torturing animals that he became a surgeon,” and from examining the head of a meat-loving dog he owned. All their methods were used to seek confirmations; contradictions were explained away. Gall and Spurzheim's cortical localizations were of “higher” intellectual and personality traits. They accepted the prevailing view that the highest sensory functions were in the thalamus and the highest motor functions, in the corpus striatum.
Phrenology met with considerable opposition from political and religious authorities, particularly on the continent, largely because it was viewed as implying materialism and determinism and denying the unity of the mind (and soul) and the existence of free will. On the other hand, phrenology widely spread particularly in the United States and Great Britain both as a medical doctrine and as a “pop” psychology. It generated widespread interest both among the general populace and among such writers and savants as Honore de Balzac, Charles Baudelaire, George Eliot, August Comte, Horace Mann, Alfred Russell Wallace, and George Henry Lewis. In fact, it rapidly became a popular fad and drawing room amusement, particularly in Great Britain and the United States. Phrenological societies and journals continued to flourish in both countries well into the 20th century.
In spite of its absurdities and excesses, phrenology became a major spur for the development of modern neuroscience in a variety of ways. Gall's mistaken assumption of a correlation of skull and brain morphology was soon recognized, at least in the scientific community. Phrenology generated an interest in the brain and behavior. It directed attention to the cerebral cortex. It stimulated study of both human brain damage and experimental lesions in animals. It inspired tracing pathways from sense organs and to the muscles to identify “organs” of the cerebral cortex. It spurred the anatomical subdivision of the cerebral cortex (cytoarchitectonics, myeloarchitectonics) to find organs of the brain.
When Broca reported a language area in the frontal lobe in 1861, he claimed it as a double confirmation of Gall: both in the specific location of a language area in the frontal lobe and the more general idea of punctate localization of psychological function in the cerebral cortex.
The cytoarchitectonic, PET, fMRI, and other imaging maps of the cerebral cortex that are now ubiquitous in neuroanatomy, neurophysiology, and neuropsychology textbooks bear more than a coincidental resemblance to phrenological charts. They are the direct descendants of the iconoclastic, ambitious, and heavily flawed program of phrenology to relate brain structure and behavior.
FRITSCH AND HITZIG: THE DISCOVERY OF MOTOR CORTEX
Modern neurophysiology began with Fritsch and Hitzig's (1870/1960) discovery that stimulation of the motor cortex produces movement. Their discovery was a major revolution in neuroscience because it was the first experimental evidence that the cortex was involved in movement, the first demonstration that the cortex was electrically excitable, the first strong experimental evidence for functional localization in the cortex, and the first experimental evidence for somatotopic representation in the brain.
At the time of their 1870 experiment, Fritsch and Hitzig were young physicians associated with the Berlin Physiological Institute. Fritsch, before and after this collaboration, was interested in studying hair and eye color in non-European societies with a view toward establishing the superiority of the White race (Grundfest, 1963). Hitzig continued research on motor cortex after his article with Fritsch and, in addition, became a successful psychiatrist. He came from a distinguished Jewish family; he was later described by one of his biographers as “a stern and forbidding character of incorrigible conceit and vanity complicated by Prussianism” (Finger, 2000; Kuntz, 1953).
In their now classical experiment, Fritsch and Hitzig strapped their dogs down on Frau Hitzig's dressing table, as there were no animal facilities at the institute (Kuntz, 1953). In their early experiments, they used no anesthesia or analgesic, although ether surgical anesthesia had been introduced in 1846 and morphine analgesia in 1803 (Magner, 1992). Later, they did use “morphine narcosis.” They began by removing the cranium and cutting the dura, the dog showing “vivid pain.” They stimulated the cortex with platinum wires with “galvanic stimulation”: brief pulses of monophasic direct current from a battery at the minimum current that evoked a sensation on their tongue.
The usual response to this stimulation was a muscle twitch or spasm (“Zückung”). Their central findings were that (a) the stimulation evoked contralateral movements; (b) only stimulation of the anterior cortex elicited movements; (c) the stimulation of specific parts of the cortex consistently produced the activation of specific muscles; and (d) the excitable sites formed a repeatable, if rather sparse, map of movements of the body laid out on the cortical surface. They went on to show that lesion of a particular site impaired the movements produced by stimulation of that site. The loss of function was not complete, suggesting to them that there were other motor centers that had not been impaired by the lesion.
Fritsch and Hitzig had no hesitation in announcing the general significance of their discovery:
by the results of our own investigations, the premises for many conclusions about the basic properties of the brain are changed not a little…some psychological functions, and perhaps all of them…need circumscribed centers of the cerebral cortex. (Fritsch & Hitzig, 1870/1960)
What led Fritsch and Hitzig (1870/1960) to electrically stimulate the cortex of a dog? This was a period of great activity and interest over the new discoveries about electricity in both salons and laboratories: such intriguing gadgets as electrostatic machines, the Leyden jar, and the gold leaf electroscope (Gross, 2009). At this time, it was realized that man-made electricity and lightning were the same phenomenon as that found in the electric fish (an animal whose shocking properties had been known since classical times). There were a number of attempts to use electricity for therapeutic purposes (including that by the French revolutionary Marat and the American savant and revolutionary Benjamin Franklin) and even studies of electrical stimulation of various brains from frog to dead human (Finger, 2000; Brazier, 1959, 1984).
Although there were reports of effects of stimulation of the spinal cord and brain stem, all attempts at eliciting effects of stimulation of the cerebral cortex had been universally ineffective. As Fritsch and Hitzig put it in the introduction to their article:
Even in other fields than in physiology, there can hardly be a question about an opinion which seems to be so unanimous, which seems to so completely settled as that of the excitability of the cerebral hemisphere. (Fritsch & Hitzig, 1870/1960, p. 75)
One impetus to their experiment was the paradox that some CNS structures were excitable and yet the cortex did not seem to be. Another was their own previous observations. Hitzig had tried electrical stimulation of the human head for therapeutic purposes and had noticed that it caused eye movements (Hitzig, 1870). He then tried rabbits and also elicited movements. Fritsch, while working as a battlefield surgeon, had apparently noticed that the contralateral limbs twitched while dressing an open head wound (Walker, 1998).
Soon after the Fritsch and Hitzig (1870/1960) article appeared, the young Scotch physician David Ferrier set out to follow up the Germans' work on motor cortex (Viets, 1938). Ferrier had been heavily influenced by John Hughlings Jackson and realized that they had confirmed Jackson's ideas. In a variety of species including primates, Ferrier replicated their basic findings that stimulation of the cortex can produce specific movements and that there is a topographic “motor map” in the cerebral cortex (Taylor & Gross, 2003; Ferrier, 1873, 1874–1875, 1875).
Although not noticed at the time, there were actually considerable differences in the methods, results, and interpretation of Fritsch and Hitzig's and Ferrier's experiments on motor cortex (Taylor & Gross, 2003). The former used brief, direct current pulses and obtained localized muscle twitches, whereas Ferrier used much longer duration biphasic stimulation that tended to produce complex integrated movements rather than muscle twitches.
Both Fritsch and Hitzig's and Ferrier's articles on motor cortex were initially greeted with considerable and equal skepticism. Their results went against the generally accepted views that the striatum was the highest motor center and that the cortex was inexcitable. The critics usually interpreted the results of Fritsch and Hitzig and of Ferrier as artifactual due to “spread of current” to the striatum, then considered the highest motor center. To overcome these criticisms, Horsley, Sherrington, and others began meticulous “punctate” mapping of cortex using the minimum current to elicit the smallest discernable movement (e.g., Brown & Sherrington, 1915; Horsley & Schafer, 1888; Beevor, 1887). This resulted in both an acceptance of the reality of a motor cortex and a stress on motor cortex controlling individual muscles, that is, in a Fritsch and Hitzig view rather than a Ferrier one.
The extent to which motor cortex controls muscles, as opposed to complex movements, continues to remain as an issue in the study of motor cortex, with most contemporary investigators leaning toward a Ferrier-like view of complex movements rather than a Fritsch-and-Hitzig-type view of the punctate control of individual or small groups of muscles by a circumscribed region of cortex (e.g., Graziano, Taylor, & Moore, 2002; Graziano, Taylor, Moore, & Cooke, 2002; Kakei, Hoffman, & Strick, 1999; Georgopoulos, Schwartz, & Kettner, 1986).
NEURON DOCTRINE: GOLGI AND RAMÓN Y CAJAL
The neuron doctrine, the idea that the nervous system is made up of discrete nerve cells that are the anatomical, physiological, genetic, and metabolic bases of its functions, may be viewed as the single most important revolution in the entire history of neuroscience. The neuron doctrine was built on the work of several generations of scientists. However, the work of two men were crucial in its final acceptance, Camilio Golgi (1843–1926) and Santiago Ramón y Cajal (1852–1934). Ironically, although Golgi provided the technique that allowed Cajal to accrue convincing evidence for the neuron doctrine, he bitterly opposed this doctrine until the end of his life.
Matthias Schleiden (1804–1881) suggested in 1838 that all plant tissues are made up of cells, and Theodor Schwann (1810–1882) extended this to animal tissue the next year. However, the nervous system resisted interpretation in terms of cell theory for about another 50 years. This was because, with the stains and microscopes available, independent self-contained cells were not discernible in the nervous tissue. Rather, the nervous system often looked like an anastomosing network or “reticulum.” Some neuroanatomists did hold the neuron doctrine, the extension of cell theory to the nervous system; whereas others supported the “reticular doctrine,” the idea that the nervous system was a network of interconnecting fibers.
The resolution of this question came, eventually, from the discovery by Golgi in 1873 of a new silver stain that stained a small proportion cells but did so in their entirety. Using this stain Golgi (a) confirmed Otto Deiters's (1834–1863) earlier observation of a single axon (“axis cylinder”) coming from each nerve cell, (b) found that dendrites (“protoplasmic prolongations”) ended freely, (c) discovered axon collaterals and thought that they merged with the axon collaterals of other nerve cells to form a diffuse reticulum, and (d) classified nerve cells by their processes.
Golgi believed that the functions of dendrites, and not the conducting of messages, was nutritive. He had a holistic view of brain function and thought that the reticulum, made up of anastomosing axon collaterals, was the basic mechanism of brain function. This, he thought, accounted for such phenomena as recovery from brain damage. His holism led him to disbelieve the localization results of Fritsch and Hitzig and Ferrier.
Fourteen years later, Ramón y Cajal first came across the Golgi silver stain and was flabbergasted:
Against a clear background stood black threadlets, some slender and smooth, some thick and thorny…. All was sharp as a sketch with Chinese ink on transparent Japanese paper. And to think that that was the same tissue which stained with carmine…left the eye in a tangled thicket where sight may stare and grope forever fruitlessly, baffled in its effort to unravel confusion and lost forever in a twilit doubt. Here on the contrary, all was clear and plain as a diagram. A look was enough. Dumbfounded, I could not take my eye from the microscope (translated by Sherrington, 1935).
Cajal immediately began making the often-capricious Golgi method more reliable, particularly by working with younger animals with less myelin because myelin is resistant to silver staining. He soon confirmed Golgi's findings on single axons from each cell, on freely ending dendrites, and on the existence of axon collaterals. However, unlike Golgi, Cajal concluded that axon collaterals did not anastomose but ended freely: neurons were separate independent units. Although microscopes were not able to actually see the gap between neurons, Cajal inferred (“intuited” might be a better term) its existence on several grounds. One was by using immature or even fetal animals where he observed axons growing out of cell bodies before approaching other neurons or muscles. Another was that, when cutting a nerve fiber, it would degenerate but only up to the border with the next cell.
Cajal's adaptation of Golgi's silver staining methods spread rapidly among the neuroanatomists of Europe, and they used it to confirm Cajal's support for the neuron theory: both axon collaterals and dendrite ended freely; there was no evidence for an anastomosing network.
Beyond confirming the idea of the neuron as an independent unit, Cajal went on to make further developments of the neuron theory. The first was the “Law of Dynamic Polarization,” the idea that information transmission was from the dendrites to the cell body and out along the axon. This had been put forward earlier by the psychologist William James (1880) and the physiologist Charles Sherrington (1906) but had little impact on the anatomical community (both cited in Shepherd, 1991). Cajal then used this “law” to work out several neural circuits that began with sensory receptors in the retina or in the olfactory bulb. A further development was the demonstration of the reality of dendritic spines, which had earlier been thought of by Golgi as staining artifacts.
In 1906, the Nobel Prize was shared by Golgi and Cajal in “recognition of their work on the anatomy of the nervous system,” which had made them “the principal representative and standard bearers of the modern science of neurology.” Golgi's Nobel address was a vigorous defense of the reticular theory, beginning with the claim that the neuron theory “is generally recognized as going out of favor” and going on to attack the idea of the neuron as an anatomical, physiological, and developmental unit as well as Cajal's law of dynamic polarization. In general, he ignored the previous 30 years of work by Cajal and much of the neuroanatomical community. Finally, he defended his ideas of a nerve network to explain the holistic aspects of brain function. Many of his holistic arguments anticipate contemporary nerve network theories. Cajal's Nobel address answered Golgi and reviewed in detail his evidence supporting the neuron theory and the law of dynamic polarization.
Over 100 years later, the neuron doctrine still stands as the bedrock of neuroscience. Its fundamental tenet of discontinuity between neurons was finally confirmed by the electron microscope in the 1950s, only to be soon modulated by the discovery of a very small number of gap junctions in which the cell membranes of adjacent neurons are immediately contiguous even under the electron microscope and synaptic transmission is electrical. The Law of Dynamic Polarization is still a property of neural circuits, although the existence of axon-less neurons, dendro-dendritic, and axon-axonal synapses has complicated the picture. As Gordon Shepherd points out, although the neuron is still the functional unit of the nervous system, it is useful to think of subcellular units such as dendritic trees, microcircuits, and synapses as well as supra cellular ones such as local circuits and interregional systems (Shepherd, 1991).
Charles Sherrington (1857–1952) named the gap between neurons implied by the neuron doctrine as the “synapse.” By this time, it was realized that conduction down the axon was electrical, so it was often assumed that conduction across the synapse was electrical too. There had also been some suggestions of a chemical mediator particularly at the neuromuscular junction. For example, in the 1840s, Claude Bernard (1813–1878) showed that curare (from a poison arrowhead obtained from Brazil) caused muscle paralysis and did so without affecting either nerve conduction nor muscle response but appeared to act on the junction between nerve and muscle. In 1878, Emil du Bois-Reymond (1818–1896) suggested that muscle could be activated by “secretion” of a “powerful stimulatory substance” from the nerve endings.
The first clear suggestion of chemical transmission came from Thomas Elliott (1877–1961), a student of John Langley (1852–1925). Langley had been the first to distinguish the sympathetic and parasympathetic divisions of the autonomic nervous system. Elliott showed that adrenaline (from the adrenal glands) reproduced the effects of sympathetic nerve stimulation and suggested that adrenaline might then be the chemical that stimulation liberated when the nerve impulse arrives at the periphery (Elliott, 1905). In 1914, Henry Dale discovered that acetylcholine mimics the action of the parasympathetic system and that adrenaline mimics the action of the sympathetic system, but he did not yet realize that both chemicals are produced by nerves.
The chemical transmission revolution developed over the first 50 years of the 20th century, but the single most important experiment was that of Otto Loewi in 1921. It was a very simple experiment, and Loewi claimed it came to him in a dream. Here is his description of it:
The hearts of two frogs were isolated [in a bath of Ringers], the first with its nerves attached, the second without…. The vagus nerve of the first heart was stimulated for a few minutes. Then the Ringer solution that had been in the first heart during the stimulation of the vagus was transferred to the second heart. It slowed and its beats diminished just as if the vagus had been stimulated. Similarly when the accelerator nerve was stimulated and the Ringer from this period transferred, the second heart speeded up and its beats increased (Loewi, 1960).
In 1936, Loewi and Dale shared the Nobel Prize for their discovery of neurotransmitters. Even after the prize was given to Loewi and Dale, there was still a great deal of skepticism about the significance and generality of their results. Although it was conceded that adrenalin might be a transmitter at some visceral organs, most neurophysiologists rejected acetylcholine as a possible transmitter at skeletal muscles. One reason was that they thought that chemical transmission was much too slow for their measurements of action at the neuromuscular junction. Another problem was that available estimates of the width of the synaptic cleft made chemical transmission across it seem very unlikely, if not impossible. The possibility of chemical transmission at central synapses was generally thought even more impossible for similar reasons. It was certainly not possible to detect transmission at central synapses at this time. Positive evidence for chemical transmission usually involved collecting perfusates or applying drugs and was often dismissed as “pharmacological” (i.e., artifactual) and not “physiological.”
Valenstein (2005) has pointed out a social–ideological background of this reluctance to accept chemical transmission. In this “battle of the soups and sparks,” most of the sparks advocating electrical transmission were neurophysiologists using high-tech electronic and quantitative methods and looked down on the soups advocating chemical transmission, who were usually pharmacologists using what the sparks consider crude methods. Furthermore, even then pharmacologists were associated with the odium theologicum of drug companies in the minds of neurophysiologists (actually, Dale did work for Wellcome for most of his career).
J. C. Eccles was the leading “spark.” The intensity of the soup versus spark debate often surprised outsiders to the English academic scene. Bernard Katz, later Sir and a Nobel laureate, but then in 1935, a newly arrived Jewish refugee from Germany noted:
To my great astonishment [at the meeting of the Physiological Society] I witnessed what seemed to be almost a stand-up fight between J. C. Eccles and H. H. Dale, with the chairman E. D. Adrian acting as a most uncomfortable and reluctant referee. Eccles had presented a paper in which he disputed the role of acetylcholine as a transmitter in the sympathetic ganglion.… When [he] had given his talk, he was counterattacked in succession by Brown, Feldberg, and Dale… [However] it did not take me long to discover that this form of banter led to no resentment between the contenders, it was in fact a prelude to much fruitful discussion over the years and indeed to growing mutual admiration between Dale and Eccles (Katz, 1996).
Alexander Forbes, Professor of Physiology at Harvard and one of the founders of modern neurophysiology commented on the debate in a 1939 review:
So goes the controversy. Dale, in discussing it, remarked that it was unreasonable to suppose that nature would provide for the liberation in the ganglion of acetylcholine, the most powerful known stimulant of ganglion cells, for the sole purpose of fooling physiologists. To this, Monnier replied that it was likewise unreasonable to suppose action potentials would be delivered at the synapses with voltages apparently adequate for exciting the ganglion cells merely to fool physiologists (Forbes, 1939).
The potential charge observed is directly opposite to that predicted by the Golgi-cell hypothesis, which is thereby falsified…. The experimental observations on synaptic excitatory and inhibitory action require for their explanation two specific transmitter substances. (Brock, Cooms, & Eccles, 1952)
The “war of the soups and sparks” was over: The chief spark had capitulated. There are now hundreds of chemicals that are known to act as transmitters at synapses. Almost as soon as the dust had settled and chemical transmission was enthroned, electrical synapses were found by Furshpan and Potter (1957). Although very small in number, they are widespread in both invertebrates and vertebrates.
NEUROGENESIS IN ADULT MAMMALS
Virtually from the beginning of the neuron doctrine, a basic corollary was that no neurons are added to the brain after infancy in mammals. In Ramón y Cajal's (1913/1928) words:
In the adult centers the nerve paths are something fixed, ended and immutable. Everything may die, nothing may be regenerated.
Because the elaborate architecture of the adult mammalian brain did not change even under the most powerful microscopes, the idea that neurons were being continually added to the brain was quite inconceivable. Further support for this view was the total lack of evidence that damaged neurons were ever replaced in the CNSs of adults. This dogma of “no new neurons in the adult mammalian brain” remained dominant until the adult neurogenesis revolution in the 1990s. There are few, if any, views of the brain in modern times that have persisted so long with so little challenge.
In the first half of the last century, there were a few dissents from the dogma, but their accounts were invariably ignored. Presumably, this was because of the weight of authority opposed to the idea and the inadequacy of the available methods both for detecting cell division and for distinguishing glia from small neurons (Gross, 2000, 2009).
A major advance in the study of neurogenesis came in the late 1950s with the introduction of tritiated thymidine autoradiography. Thymidine is incorporated into the DNA of dividing cells, and therefore, the cells that have just divided can be labeled and their time and place of birth can be determined by autoradiography. Almost immediately, Joseph Altman at MIT began publishing a series of articles in which he reported thymidine autoradiographic evidence for new neurons in the olfactory bulb, the dentate gyrus of the hippocampus, and the neocortex in the adult rat, the guinea pig, and the cat. He suggested that the new neurons played a crucial role in learning and memory. Although published in prestigious journals such as Science and Nature, his findings were ignored or dismissed for over two decades (Altman, 2011; Gross, 2009).
One reason why Altman's work was so ignored was probably because the available techniques may not have been totally unambiguous for an objective demonstration that the adult generated cells were neurons rather than glia. Rather, the distinction rested on having a so-called “expert eye” to tell neurons from glia, and experts knew that adult neurogenesis was impossible. Another reason may have been because Altman was a self-taught junior faculty member working on his own in a Psychology Department and purporting to overturn a central and, by now, universally held tenet of neuroscience. Altman was denied tenure at MIT, presumably on the authority of Walle Nauta, one of the leading anatomists of the day, who was one of three tenured members of his department at this time. Altman then moved to Purdue University where he eventually turned to more conventional developmental questions, perhaps because of the lack of recognition of his work on adult neurogenesis (Altman, 2011; Gross, 2009).
Fifteen years after Altman's first report of adult neurogenesis, direct support for his claim came from a series of electron microscopic studies by Michael Kaplan and his co-authors. They showed that thymidine labeled cells in the olfactory bulb, dentate gyrus, and neocortex of adult rats had the ultrastructural characteristics of neurons, such as dendrites and synapses, supporting the earlier claims of Altman. In spite of his evidence for adult neurogenesis, Kaplan's work had little impact at the time (Kaplan, 2001). Again, as in Altman's case, publication in prestigious and rigorously reviewed journals by an unknown figure was not sufficient to make any significant dent in the central dogma of “No New Neurons.”
A primary reason for the submergence of Altman and Kaplan's evidence for adult neurogenesis was probably a study by Pasko Rakic, professor at Yale Medical School and, arguably, the leading student of primate brain development. He carried out a [3H]-thymidine study of adult rhesus monkeys in which he examined “all major structures and subdivisions of the brain including association cortex, hippocampus and olfactory bulb” and found “not a single heavily labeled cell with the morphological characteristics of a neuron in any brain of any adult animal.” He concluded that “all neurons of the rhesus monkey brain are generated during prenatal and early postnatal life.” (Rakic, 1985)
In a subsequent article with Eckenhoff, the results were again negative, and they claimed that the supposed lack of adult neurogenesis in primates made good sense because “a stable population of neurons may be a biological necessity in an organism whose survival relies on learned behavior acquired over a long period of time” (Eckenhoff & Rakic, 1988). In other words, they suggested that the absence of adult neurogenesis in primates was crucial for their advanced cognition.
At about this time, Fernando Nottebohm and his coworkers showed that neurons were added to the adult avian forebrain. They did so by (a) demonstrating the new cells with tritiated thymidine labeling, (b) producing ultrastructural evidence that the new cells were in fact neurons, and finally, (c) showing that the putative neurons responded to sound with action potentials (Nottebohm, 1996). This work was readily accepted probably in part because Nottebohm was already an established figure, because it was a very elegant work, and because, after all, it was just birds. In spite of this overwhelming evidence of neurogenesis in parts of the adult bird brain that were known to be homologous to primate cerebral cortex and primate hippocampus, Nottebohm's studies tended to be viewed as irrelevant to the primate or even the mammalian brain. Rather, the evidence for avian neurogenesis was viewed as an exotic specialization related to the necessity for flying creatures to have light brains and to their seasonal cycles of singing. It was viewed as quite irrelevant to the human condition. Rakic's view of no new neurons was not particularly unique and was standard in the developmental neuroanatomy texts of the time.
In the 1990s, there were two developments that led to the revolutionary end of the “no new neurons” dogma (Gross, 2000). The first was the introduction of the synthetic thymidine analogue bromodeoxyuridine (BrdU). Like thymidine, BrdU is taken up by cells during the S-phase of the cell cycle and is a marker of proliferating cells or their progeny. One advantage of BrdU is that it can be used to estimate the actual number of new cells. The second advance was the introduction of immunochemical markers that can detect and distinguish neurons (such as neuron-specific enolase, NeuN) and glia (such as glial fibrillary acidic protein, GFAP).
Combining BrdU and immunochemical markers made it possible to objectively detect and count new neurons. Using these techniques, Elizabeth Gould and her colleagues were able to confirm Altman's findings of new neurons in the hippocampus of rats and extend them to a variety of other mammals including nonhuman primates (Gould, 2007; Gould & Gross, 2002). Eriksson and his colleagues (Eriksson et al., 1998) then demonstrated neurogenesis in the hippocampus of adult human, and Rakic recanted his position of no adult neurogenesis for macaque hippocampus (Kornack & Rakic, 1999). Altman's findings of olfactory bulb neurogenesis were also confirmed at about the same time (Luskin, 1993).
However, Altman's findings of cortical neurogenesis are still in dispute. Gould's and several other laboratories have replicated his results, whereas Rakic's and several other laboratories have not (Gross, 2009; Gould, 2007). More sensitive and more reliable methods are required to resolve this issue as well as the existence of adult neurogenesis in other structures, which also has been both reported and denied (Breunig, Arellano, Mackliss, & Rakic, 2007; Gould, 2007).
In summary, Altman's revolutionary challenge of the doctrine of no new neurons in the mammalian brain has now been universally accepted at least for the hippocampus and olfactory bulb.
OTHER REVOLUTIONS IN THE HISTORY OF NEUROSCIENCE
There were a number of other major revolutions in the history of neuroscience that have not been covered in this article, such as the shift from animal spirits to animal electricity as the basis of nerve conduction in the 17th and 18th centuries, the discovery of sensory coding by Adrian in the early 20th century, and perhaps, Hubel and Wiesel's use of single neuron recording to reveal the hierarchical processing of visual information. What revolutions are underway now (neuroimaging? neurocomputation? neurogenetics?) remain to be seen.
Reprint requests should be sent to Charles Gross, Department of Psychology and Neuroscience Institute, Princeton University, 18 East Shore Dr., Princeton, NJ 08540, or via e-mail: firstname.lastname@example.org.