Edgar Adrian

The claim

Edgar Adrian made the state of a tissue into a number

Edgar Adrian proved that the nervous system speaks one language and that the only thing it can vary is how fast it speaks. In 1926, in the Cambridge physiological laboratory, working with the Swedish physiologist Yngve Zotterman, he recorded the traffic in a single sensory nerve fiber of a frog (Adrian and Zotterman 1926). Every impulse was the same size. Every impulse was the same duration. What changed with the strength of the stimulus was the rate. A light stretch produced firing at about ten times a second. A stronger and faster stretch drove the rate past fifty. In 1932 the Caroline Institute gave Adrian and Sir Charles Sherrington the Nobel Prize in Physiology or Medicine for discoveries regarding the functions of neurons.

Ask what that commits you to. If the impulse is fixed and only the rate varies, then the condition of a tissue cannot be encoded in the kind of signal it sends. It has to live in frequency and in how many fibers join in. That is not a figure of speech about the body being electrical. It is a measurement with units attached. Adrian is the point at which the living state of a tissue stopped being a quality you describe and became a rate you can count, and the vocabulary this library uses for tone, frequency, amplitude, recruitment, adaptation, descends from records he made with a capillary electrometer and a valve amplifier between 1925 and 1934.

The instrument

The obstacle was never the nerve, it was the amplifier

Adrian solved a hardware problem before he solved a biology problem. He said so plainly in Stockholm (Adrian 1932). The potentials to be dealt with, he told the audience, are of the order of a few microvolts lasting for a few thousandths of a second, and the instruments of the previous generation could not see them at all. The capillary electrometer, refined at the close of the nineteenth century by Gotch, Burch, Garten, Samojloff and Keith Lucas, could record the electrical disturbance of a whole stimulated nerve trunk, where thousands of fibers discharge together and their potentials sum into something visible. It could not resolve one fiber working alone in ordinary conditions.

The thermionic valve changed the arithmetic. Adrian coupled the capillary electrometer to a triode valve amplifier and the individual fiber came within reach. He named his debts precisely: Alexander Forbes at Harvard, Herbert Gasser at St. Louis, who was the first to use very high amplification, and Bryan Matthews at Cambridge, who built the moving iron oscillograph that became standard equipment. Behind them stood Julius Bernstein and the membrane hypothesis, which treated the impulse as a wave of surface breakdown spreading by reason of the electric disturbance it creates. Physiology, in Adrian’s telling, advanced whenever physics handed it a better instrument. That is a claim about method, and it holds.

The preparation

One muscle, one sense organ, one fiber

The decisive experiment was an act of dissection, not of theory. Adrian and Zotterman took the sterno-cutaneous muscle of the frog and divided it progressively until it contained only one sense organ. Stretch that muscle and a single muscle spindle fires. The impulses travel up a single afferent fiber. The amplifier makes them audible and the oscillograph makes them visible. The results appeared in the Journal of Physiology in 1926 as a three part series. Part II, on the response of a single end organ, ran in volume 61, issue 2, pages 151 to 171, published on 23 April 1926 (Adrian and Zotterman 1926). Part 3, on the impulses set up by touch and pressure, followed in volume 61, issue 4, pages 465 to 483 (Adrian and Zotterman 1926).

The record showed a train of brief diphasic waves, each one the passage of a single impulse along the fiber. Constant size. Constant duration. Varying interval. For the first time the question of how a nervous system grades a message stopped being an argument between schools and became something you could count off a photographic trace. Adrian was blunt about how contested that ground had been. One school read the electrical changes in contracting muscle as implying a very high frequency of discharge in each nerve fiber. Others believed the frequency was lower. Neither side, in his words, could find convincing evidence. The single fiber record ended the dispute.

All or none

The impulse never gets bigger, so intensity had to live somewhere else

The all-or-none principle was not Adrian’s discovery and he never claimed it was. His Nobel Lecture traces the lineage in public. Francis Gotch had observed that the potential wave in a nerve lasted the same time whether it was set up by a strong or a weak stimulus, and suggested that each fiber always discharges at full intensity, a strong stimulus producing a larger recorded wave only because it brings more fibers into activity (Gotch 1902). Keith Lucas then recorded the contraction of a band of muscle containing only a few fibers and found that with an increasing stimulus the contraction increased in sudden steps, and that the number of steps was never greater than the number of fibers in the preparation (Lucas 1909). Skeletal muscle obeyed the rule.

Adrian’s own contribution to that question came early and is often misdated. He was elected to a Fellowship of Trinity College, Cambridge in 1913 on account of his investigation of the all-or-none principle in nerve, and he worked directly with Keith Lucas before the First World War. What 1926 added was direct evidence from a fiber doing its ordinary job rather than a fiber being shocked by an experimenter. Set that beside his remark in Stockholm that in honouring him with the prize the committee had honoured the master as well as the pupil, and you have a man unusually careful about attribution in a field that rarely is.

The code

Intensity is carried by rate, not by size

If the unit of the message is fixed, grading has to happen in time. That is the whole of frequency coding, and Adrian stated the consequence without decoration. The frequency depends on the extent and on the rapidity of the stretch, he wrote, which is to say on the intensity of excitation in the sense organ, and in this way the impulse message can signal far more than the mere fact that excitation has occurred. Notice the two variables folded into one sentence. How far. How fast. A sense organ is not reporting a single quantity. It reports magnitude and rate of change together, in one stream of identical pulses.

The generality is what makes this a law rather than a frog result. Adrian pointed to Haldan Keffer Hartline’s recordings from the light sensitive receptors in the eye of Limulus, the horseshoe crab, and observed that the discharge was a fairly close copy of that from a frog’s muscle spindle (Graham and Hartline 1935). Light and stretch are unrelated physical events. They produce the same kind of message. Goran Liljestrand put the point cleanly in the presentation speech on 10 December 1932 (Liljestrand 1932) when he said that the signals are the same everywhere, but the receiving stations change and the results with them. Meaning belongs to where a message lands, not to what the message is made of.

Adaptation

A nervous system reports change, not constancy

Hold a stimulus steady and the discharge falls away. This is adaptation, and Adrian’s work on touch and pressure between 1926 and 1928 made it quantitative. The pressure on the skin does not change. The size of each impulse does not change. The rate drops, and the sensation drops with it. He set the pattern out in The Basis of Sensation in 1928 (Adrian 1928): a constant stimulus excites the end organ immediately, that excitation decreases for as long as the stimulation continues, the impulses traveling up the nerve stay uniform in size, their frequency falls away, and the sensation in the brain diminishes in step. Different receptors adapt at radically different speeds, and that spread of rates is itself information.

Ask what that commits you to. If a nervous system fires to change rather than to state, then no tissue can hold a fixed report of itself. Every reading is a reading of a slope. Liljestrand made the same point at the ceremony: the sense organs have a varying power of adapting themselves to their milieu and only respond to changes in it. This is why a load you have carried for an hour drops out of awareness, and why a small unexpected shift in that same load is instantly loud. A nervous system does not store its condition. It keeps recomputing it, and what it computes is a difference.

The motor side

Adrian and Bronk showed force is graded by discharge rate and by motor unit recruitment

Adrian turned the same method on the output side with Detlev Bronk, and the answer came back symmetrical. In Part I, published in the Journal of Physiology in 1928, volume 66, pages 81 to 101 (Adrian and Bronk 1928), they dissected the phrenic nerve of anesthetized rabbits down to single fibers and recorded the drive to the diaphragm during ordinary breathing. In Part II, volume 67, pages 119 to 151, dated 1929 (Adrian and Bronk 1929), they built a tool to reach human muscle. Their concentric needle electrode was an enamelled copper wire of No. 36 gauge, roughly 193 micrometres in diameter, passed up the center of a small hypodermic steel needle, described on page 133. With it they recorded voluntary triceps contraction in man and found units discharging as slowly as six times a second at minimal force.

The half everyone quotes is rate coding. The half everyone drops is recruitment. Adrian and Bronk had already noticed in 1928 that frequency alone would not do the arithmetic. When the air tubes were clamped, the force of contraction rose many times over, whereas a shift in discharge frequency from 20 to 60 a second would account for something closer to a doubling. The rest had to come from more motor units entering the pool. Force is graded by rate and by number together, and neither half is optional. A century of electromyography and motor unit research rests on that pair of papers.

Symmetry

The message going in and the message coming out have the same shape

Adrian found the sensory and motor discharges nearly indistinguishable, and he did not soften the implication. In quiet breathing, at each expansion of the lungs, the sense organs of the vagus send up a train of impulses rising to a frequency of about 20 a second at the height of inspiration, and at that same moment the movement of expansion is being produced by motor trains at much the same frequency. His conclusion was that the motor nerve cells seem to be acting just like a collection of sense organs responding to a rhythmic stretch. In a contraction of gradually increasing force the motor fibers begin at 5 to 10 a second and rise to 40 or 50 at the height of the contraction, the same band as the sensory side.

Then he generalized further. Resemblances of this kind, he said, show that there is an underlying unity of response in the various parts of the neurone in spite of their differentiation into axon, dendrites or terminal arborizations. One cell, one grammar, whichever end of it you are looking at. The population layer behaves the same way. Bronk and Stella’s recordings from the carotid sinus showed that as blood pressure rises the impulses in each nerve fiber increase in frequency and more and more fibers come into action. Adrian played gramophone records of that traffic to the Stockholm audience so they could hear both gradations at once, the change in each unit and the change in the number of units.

The brain

From single fibers to the rhythm of a whole cortex

Adrian’s last great move was upward in scale, and it required him to give the credit away. Hans Berger of Jena had described the human electroencephalogram in 1929 and had been widely disbelieved. In 1934 Adrian and Bryan Matthews published The Berger rhythm: potential changes from the occipital lobes in man in Brain, volume 57, pages 355 to 385. They confirmed the rhythm, traced its source to the occipital region, and named it after the man who found it. Adrian could have attached his own name to the most famous waveform in neurology. He attached Berger’s.

The scientific point outweighs the courtesy. Adrian had spent a decade showing that the unit of nervous action is a discrete, identical pulse. The cortical rhythm showed that when enough of those units interact, the population produces a slow coherent oscillation that no single unit contains. He closed his Nobel Lecture on exactly this frontier. Within the central nervous system, he said, the events in each unit are not so important, we are more concerned with the interactions of large numbers, and our problem is to find the way in which such interactions can take place. That sentence is a research program rather than a conclusion, and ninety years on it is still open.

Adrian and the model

The rate code is where the Unified Model of Tone begins

The Unified Model of Tone defines tone as the organization of all the body’s oscillations at a given moment, and that definition is only possible in a nervous system that speaks in rates. Adrian proved that it does. The model’s opening argument runs through the senses. Pitch is the ear reading air pressure that oscillates 440 times a second. Color is the eye reading light at hundreds of trillions of cycles. Perception itself is the reception of frequency. Adrian’s 1926 record is the same fact one layer in. The traffic behind every perception is itself a frequency.

Look at the loop he mapped. Mechanical tension in a muscle deforms a spindle. The spindle converts that tension into a frequency. The frequency travels to the cord and the brain. Motor cells convert frequency back into tension, graded by rate and by recruitment. Stretch receptors report the result and the cycle repeats. Tension at both ends, frequency in the middle, feedback closing the circuit. Walter Cannon gave that kind of arrangement its general name in the same decade, and Adrian had already measured one of its circuits end to end.

The model then takes the clinical step. Tap a relaxed muscle and it oscillates, measurably, somewhere between eleven and twenty cycles per second, and the number climbs roughly twenty-five percent from relaxed to contracted when read on the muscles beside the spine. Muscle tone, a phrase clinicians have used for more than a century, corresponds to a rate. In the model’s terms, tone was already riding on frequency, with rate as its carrier. It was simply never read as one. Adrian is the reason that sentence can be a measurement instead of a metaphor.

Say plainly where this reading goes beyond the record. Adrian studied signaling in nerve fibers. He did not write about tone as a whole body state, and he made no clinical claims of the kind this library builds on. What is documented is the mechanism: identical impulses, frequency coding of intensity, adaptation to constancy, and grading by rate plus recruitment on both sides of the loop. The Unified Model of Tone takes that mechanism as its carrier and makes a claim of its own past it. The body’s many rhythms compose one readable organization, and health is the regulation of that organization. The extension is the model’s. The measurement is his, and it is exact.

Corrections

Adrian is wrongly credited with the all-or-none law, the EEG, intracellular recording, and working alone on rate coding

The all-or-none law is the first. Adrian investigated it and supplied direct evidence in the nerve fiber, but the principle was established first in heart muscle, where Henry Pickering Bowditch is usually given the credit, and it reached nerve and skeletal muscle through Gotch and Keith Lucas. Adrian’s Nobel Lecture names Gotch and Lucas explicitly and points back to heart muscle as the original case (Adrian 1932). The second is the electroencephalogram, which belongs to Hans Berger in 1929. The third is intracellular recording. Adrian recorded from outside the fiber, in preparations dissected until only one or a few units were active. Recording from inside an axon came later, through Hodgkin, Huxley and the squid giant axon.

The fourth is the habit of citing frequency coding as though Adrian worked alone and as though rate were the whole answer. Yngve Zotterman is a co-author on the 1926 papers and Detlev Bronk on the 1928 and 1929 motor papers, and Adrian and Bronk stated in print that rate alone cannot account for the force a muscle produces. Two smaller cautions are worth keeping. His third book is The Physical Background of Perception, 1947 (Adrian 1947), not The Physical Basis of Perception, a slip that appears even on the Nobel Foundation’s own biographical page (Nobel Foundation 1932), which also dates The Basis of Sensation to 1927 where the book is normally given as 1928. And the well loved story of a toad’s optic nerve signaling Adrian’s movements around a darkened room is a recollection he repeated in later life, reproduced everywhere, not a passage from any research paper. It is charming and it is probably true. It is not a citation.

What the record shows

The nerve code in six dated findings

• 1926. Adrian and Yngve Zotterman recorded a single frog muscle spindle in the Journal of Physiology, volume 61: every impulse identical in size and duration, only the interval varying (Adrian and Zotterman 1926).
• 10 to 50 per second. Light stretch fired the fiber at about 10 impulses a second, and a stronger, faster stretch drove the rate past 50. Intensity lives in frequency, never in amplitude.
• 1913. Adrian’s Trinity College Fellowship was awarded for his investigation of the all-or-none principle in nerve, a decade before the single fiber records, under Keith Lucas (Nobel Foundation 1932).
• 1928. The Basis of Sensation (Adrian 1928) set out adaptation: under a constant stimulus the impulses stay uniform while their frequency falls, and sensation fades in step. A nervous system reports change, not constancy.
• 1928 to 1929. With Detlev Bronk he dissected the phrenic nerve to single fibers, Journal of Physiology volumes 66 and 67 (Adrian and Bronk 1928, Adrian and Bronk 1929), and showed that force is graded by rate and by recruitment. The motor message has the same shape as the sensory one.
• 1932. The Nobel Prize in Physiology or Medicine, shared with Sir Charles Sherrington, for discoveries regarding the functions of neurons (Nobel Prize 1932).