The Hodgkin-Huxley Squid Axon Experiment: How We Cracked the Action Potential

Every thought you have, every muscle twitch, every heartbeat begins with a signal that races along a nerve fiber at up to 120 meters per second (about 270 miles per hour). That signal is the action potential — a precisely orchestrated wave of electrical change. For most of history, the machinery behind it was a mystery. In the late 1940s and early 1950s, two British physiologists named Alan Hodgkin and Andrew Huxley threaded tiny electrodes into the axons of squid, held the membrane voltage artificially still with a feedback circuit, and watched sodium and potassium ions do what no one had seen before.

Why Neuroscientists Needed a Squid

The fundamental problem with studying nerve impulses in 1939 was size. The axons of mammals are, in most cases, too thin to insert an electrode without destroying the fiber entirely. The giant axon of the North Atlantic squid (Loligo forbesi), however, is an outlier so extreme it belongs in a different category.

The squid giant axon evolved to control the animal's escape response — a jet-propulsion maneuver that requires near-simultaneous contraction of the mantle muscles. Speed demanded thickness, so evolution produced an axon up to 1 millimeter in diameter, roughly 1,000 times wider than a typical human axon. A 1-millimeter tube is large enough to insert a wire electrode along its length without causing fatal damage to the membrane.

Axon — the long, slender projection of a neuron that carries electrical signals away from the cell body toward other neurons or muscles.

The Plymouth Marine Laboratory in England gave Hodgkin and Huxley access to freshly caught squid throughout the summer months. Fresh tissue was essential: a squid axon loses its electrical properties within hours of the animal's death, so experiments ran on a tight schedule during the annual summer sessions at Plymouth.

Source: Hodgkin, A.L. (1963), Nobel Lecture, "The ionic basis of nervous conduction" — NobelPrize.org. Hodgkin describes the Plymouth work, the seasonal constraints, and the choice of Loligo as the experimental preparation.

The State of Knowledge Before 1939

By the 1930s, physiologists knew a few things about nerve impulses. They knew the impulse was electrical — a traveling disturbance in membrane potential, the voltage difference between the inside and outside of the axon. They knew it traveled at a fixed velocity along any given fiber. They knew it followed an all-or-nothing rule: either a stimulus is strong enough to trigger a full-size impulse, or it triggers nothing at all.

What they did not know was the ionic mechanism. The leading hypothesis in 1939 was that the action potential resulted from a sudden non-selective increase in membrane permeability — the membrane temporarily became "leaky" to all ions, reducing the voltage difference to zero. This hypothesis was wrong, and Hodgkin and Huxley would prove it.

Membrane potential — the voltage difference across a cell membrane, typically around −70 millivolts (inside negative relative to outside) in a resting neuron.

In 1939, Hodgkin and his Plymouth colleague A.F. Huxley — not yet the Andrew Huxley who would become his long-term collaborator, though the same person — made the first measurement of the membrane potential from inside the axon. By threading a fine capillary electrode down the length of the squid axon, they recorded the resting potential at approximately −45 to −60 millivolts and confirmed that during the action potential the inside actually went positive, rising to around +40 millivolts.

Source: Hodgkin, A.L. & Huxley, A.F. (1939), "Action potentials recorded from inside a nerve fibre," Nature, 144, 710–711 — the first intracellular recording of an action potential, establishing that the interior goes positive during the impulse.

That observation — the membrane potential not merely collapsing to zero but overshooting to a positive value — immediately contradicted the simple "leakiness" hypothesis. A non-selective leak would bring the potential toward zero, not beyond it. Something specific was happening to specific ions.

The Voltage-Clamp Technique: Freezing the Wave in Place

The challenge with studying an action potential in its natural state is that it is a self-propagating event. The moment you trigger one, it moves. Measuring the detailed time course of the ion flows underneath it is like trying to photograph a wave from inside the water while the wave is already past you.

The voltage clamp solved this problem. Developed by Kenneth Cole at Columbia University and the Marine Biological Laboratory in Woods Hole, Massachusetts, the technique uses a feedback amplifier to hold the membrane potential at any chosen value. If the experimenter sets the clamp voltage to −20 millivolts, the amplifier instantly injects or withdraws current to keep the membrane at exactly −20 millivolts, no matter what the ion channels are trying to do. The current the amplifier must inject to maintain that fixed voltage is precisely equal and opposite to the ion current flowing through the membrane — so by measuring the injected current, you measure the ion current directly.

Source: Cole, K.S. (1949), "Dynamic electrical characteristics of the squid axon membrane," Archives des Sciences Physiologiques, 3, 253–258 — original description of the voltage clamp applied to squid axon.

Think of it like holding a garden hose at a fixed shape while measuring the force needed to do so. The force reveals the water pressure inside without letting the hose move. Hodgkin and Huxley adapted Cole's technique for their Plymouth experiments beginning in 1948, when they returned to systematic work after the interruption of World War II.

Voltage clamp — an electrophysiological technique that holds a cell's membrane potential at a fixed level using a feedback amplifier, allowing the direct measurement of ion currents flowing through membrane channels.

Five Papers in One Summer

In 1952, Hodgkin and Huxley published five papers in the Journal of Physiology describing their complete analysis of ion flow in the squid axon. The series remains one of the most concentrated and productive bursts of experimental work in the history of biology.

The core strategy was to separate the total membrane current into its component parts. They knew from prior work that two major ion species were likely candidates — sodium (Na⁺) and potassium (K⁺). By replacing the sodium in the external solution with a non-penetrating substitute, they could eliminate the sodium current and observe potassium current in isolation. Subtracting the pure-potassium current from the total current gave them the sodium component.

What they found was striking. When the membrane was suddenly depolarized (voltage stepped from resting to a positive value):

1. A rapid, large inward current appeared within about 0.5 milliseconds. This was sodium ions rushing into the cell, driven by both the electrical gradient (inside negative) and the concentration gradient (sodium is more concentrated outside).
2. The sodium current spontaneously declined even if the membrane voltage was held constant — a process they called inactivation.
3. A slower, sustained outward current followed. This was potassium ions flowing out of the cell, driven by the concentration gradient (potassium is more concentrated inside).

Source: Hodgkin, A.L. & Huxley, A.F. (1952a), "Currents carried by sodium and potassium ions through the membrane of the giant axon of Loligo," Journal of Physiology, 116, 449–472 — first paper in the 1952 series, documenting the separation of sodium and potassium currents.

The action potential, it turned out, was a two-stroke engine. Sodium flowing in depolarized the membrane (drove it positive). Potassium flowing out repolarized it (drove it back negative). The whole event, from start to finish, took about 1 millisecond.

The Hodgkin-Huxley Model: Four Equations That Described a Nerve

Hodgkin and Huxley did not stop at description. Their fifth 1952 paper presented a mathematical model that could reproduce the action potential from first principles.

The model treated sodium and potassium conductances (the ease with which each ion passed through the membrane) as functions of voltage and time, described by three gating variables: m and h for sodium (activation and inactivation gates, respectively) and n for potassium (activation gate). Each variable followed a first-order differential equation with rate constants measured experimentally at different voltages.

Source: Hodgkin, A.L. & Huxley, A.F. (1952d), "A quantitative description of membrane current and its application to conduction and excitation in nerve," Journal of Physiology, 117, 500–544 — the model paper, containing the full system of equations and predictions.

The resulting system of four coupled differential equations could be solved (by hand, on a mechanical calculator — computers of the era were not accessible) to yield a predicted action potential waveform. The predicted waveform matched the recorded waveform almost exactly, including the overshoot, the repolarization, the afterhyperpolarization (a brief dip below resting potential after the impulse), and even the correct conduction velocity.

This was not a curve fit. The parameters were measured independently from voltage-clamp data, and the predicted action potential was a genuine prediction — one that came true.

Gating variable	Ion	Role	Kinetics
m	Na⁺	Activation (opens fast)	Fast: ~0.5 ms at 0 mV
h	Na⁺	Inactivation (closes channel)	Slower: ~1–2 ms at 0 mV
n	K⁺	Activation (opens slowly)	Slow: ~2–5 ms at 0 mV
g_L	Leak	Constant background conductance	Time-independent

The model's predictive power was its most important feature. Hodgkin and Huxley used it to calculate what would happen if you injected a pulse of current into the axon, applied a ramp of current, or changed the external ion concentrations. In each case, the model gave the right answer.

What the Voltage Clamp Revealed Physically

The ionic currents Hodgkin and Huxley measured implied the existence of specific, voltage-sensitive pores in the membrane — what we now call ion channels. In 1952, the molecular identity of these pores was unknown. The HH model was purely phenomenological: it described how conductances behaved without explaining the physical structures responsible.

The molecular picture came later. In the 1970s and 1980s, patch-clamp recordings by Erwin Neher and Bert Sakmann revealed individual ion channel openings and closings — showing that the smooth conductance curves of Hodgkin and Huxley were actually averages over thousands of stochastic single-channel events. Neher and Sakmann received the Nobel Prize in 1991.

Source: Neher, E. & Sakmann, B. (1976), "Single-channel currents recorded from membrane of denervated frog muscle fibres," Nature, 260, 799–802 — the original patch-clamp paper, directly extending the HH framework to the molecular level.

The sodium channel was sequenced in 1984, and its structure — four repeated domains each containing a voltage-sensing segment — directly explained why the channel opens in response to depolarization. The gating variables m and h in the HH model correspond to real conformational changes in the channel protein's voltage-sensing helices.

Source: Noda, M. et al. (1984), "Primary structure of Electrophorus electricus sodium channel deduced from cDNA sequence," Nature, 312, 121–127 — first molecular sequence of a voltage-gated sodium channel.

The Timeline

Year	Event
1936	Hodgkin begins research at Cambridge; first work on nerve conduction
1939	Hodgkin & Huxley record first intracellular action potential from squid axon; paper in Nature
1939	Cole & Curtis measure conductance increase during action potential
1940–1945	Wartime interruption; Hodgkin works on radar, Huxley on antiaircraft fire control
1947	Hodgkin and Huxley resume Plymouth experiments
1949	Cole develops voltage-clamp technique
1948–1951	Systematic voltage-clamp experiments at Plymouth
1952	Five papers published in Journal of Physiology; HH model presented
1963	Hodgkin, Huxley, and Eccles share Nobel Prize in Physiology or Medicine
1976	Neher & Sakmann invent patch clamp; single channel recordings confirm HH framework
1984	Sodium channel sequenced; molecular structure explains HH gating variables
1998	First crystal structure of a potassium channel (Doyle et al., Science)

Why the Model Is Still Used Today

The Hodgkin-Huxley model is over 70 years old. In neuroscience, that is ancient. Yet it remains the foundation of computational neuroscience and is taught in every graduate-level neurobiology course.

The reason is that the model works at a level of abstraction that has proven enduring. Even though real neurons have dozens of different ion channel types — and some neurons, like Purkinje cells in the cerebellum, have hundreds — the mathematical framework Hodgkin and Huxley invented to describe voltage-dependent gating applies to all of them. Each new channel type discovered is characterized by the same kind of voltage-clamp experiments, described by the same kind of differential equations, and added to HH-type models.

Source: Rinzel, J. (1990), "Electrical excitability of cells, theory and experiment: review of the Hodgkin-Huxley foundation and an update," Bulletin of Mathematical Biology, 52, 3–23 — review of the HH model's enduring relevance.

Large-scale brain simulations today — including the Human Brain Project and similar efforts — use HH-type equations for individual neuron compartments. The Blue Brain Project's model of a cortical column contains thousands of neurons, each with multiple compartments described by modified HH equations.

The model also proved that biology, in principle, is fully reducible to quantitative physical description. That proof mattered enormously for the development of neuroscience as a discipline. If the firing of a neuron can be written as four equations with experimentally measured parameters, then the hypothesis that the brain is, at bottom, a physical system with quantitative laws becomes not just plausible but demonstrated.

The Computational Legacy: From Mechanical Calculators to Neural Simulations

Huxley's calculation of the action potential waveform from the HH equations took weeks of hand computation using a mechanical desk calculator. He worked through the equations using Euler's method — breaking time into tiny increments of 0.01 milliseconds and computing the state of each gating variable at the next time step from its value at the current step. The full integration of a single action potential required hundreds of such steps and thousands of individual arithmetic operations.

The first digital computer used to run HH simulations was the EDSAC (Electronic Delay Storage Automatic Calculator) at Cambridge, used by Huxley in the early 1950s. Even with the computer, the calculations were slow by modern standards — a single simulation took hours. Today, the same simulation runs in milliseconds on a laptop, and a full-scale cortical column containing 31,000 neurons, each with dozens of HH-type conductances, can be simulated on a cluster computer in near-real time.

Source: Huxley, A.F. (2002), "From overshoot to voltage clamp," Trends in Neurosciences, 25(11), 553–558 — Huxley's personal account of the computational challenge and the role of the EDSAC.

The HH framework also became the theoretical backbone of computational neuroscience as a discipline. The Blue Brain Project, launched by Henry Markram at EPFL in Lausanne in 2005, uses HH-type models for each of the thousands of neurons in its cortical column simulations. The Human Brain Project, a 10-year European initiative, adopted the same approach at larger scale. Whether brains can ultimately be understood by simulating every neuron at the HH level of detail remains debated — some argue that higher-level descriptions are necessary and sufficient — but there is no question that the HH formalism is the standard of biological realism at the single-neuron level.

The HH equations also inspired the field of artificial neural networks, though indirectly. The simplified "integrate-and-fire" model widely used in computational modeling strips the HH equations to their essentials: integrate input current until the voltage threshold is crossed, then fire and reset. This abstraction sacrifices biological accuracy for computational speed, allowing simulations of millions of neurons. But the threshold-and-spike concept at its core descends directly from the HH model's demonstration that the action potential is an all-or-nothing threshold event driven by regenerative ion conductance.

Source: Mainen, Z.F. & Sejnowski, T.J. (1995), "Reliability of spike timing in neocortical neurons," Science, 268, 1503–1506 — using HH-type models to show that spike timing in cortical neurons is highly reproducible, linking computational models to experimental observations.

The Squid's Unlikely Contribution to Medicine

The squid giant axon's most direct medical legacy is in the treatment of chronic pain and cardiac arrhythmia. Local anesthetics — lidocaine, procaine, bupivacaine — work by blocking voltage-gated sodium channels in the nerve fibers that carry pain signals. The sodium channel they block is the molecular descendant of the Na⁺ conductance that Hodgkin and Huxley measured in the squid axon. Understanding the channel's voltage-dependence and its activation-inactivation kinetics, worked out in Plymouth using squid, was essential for understanding how these drugs work at the molecular level.

Cardiac antiarrhythmic drugs work by the same principle applied to cardiac muscle. Some arrhythmias arise when sodium or potassium channels in cardiac cells are mutated or pharmacologically disrupted, producing abnormal action potential patterns. HH-type models of cardiac cell electrical activity are used routinely to predict how a given drug will affect heart rhythm before animal or clinical trials — a regulatory application of 1952 equations in twenty-first-century medicine.

Source: Fozzard, H.A. & Hanck, D.A. (1996), "Structure and function of voltage-dependent sodium channels: comparison of brain II and cardiac isoforms," Physiological Reviews, 76, 887–926 — comparative pharmacology of sodium channels across tissues, showing the direct application of HH-framework knowledge to cardiac medicine.

Inherited channelopathies — diseases caused by mutations in ion channel genes — are another direct application. Long QT syndrome, a cardiac condition that can cause sudden death, results from mutations in potassium channel genes that slow the K⁺ current responsible for repolarization — the third gating variable in the HH model. The risk of a mutation can be evaluated by inserting it into an HH-type cardiac model and computing the effect on action potential duration. This computational approach has become standard in preclinical drug evaluation.

What People Get Wrong About the HH Experiment

"The squid axon is exotic and irrelevant to human neuroscience." The squid axon's ionic machinery — sodium channels in, potassium channels out — is the same in every animal with a nervous system, including humans. The HH model, developed on squid, describes the action potential in human neurons with only minor modifications for different channel types. The squid axon was a preparation of convenience, not a special biological case.

"Hodgkin and Huxley discovered ion channels." They inferred the existence of selective, voltage-controlled permeability pathways, but they did not observe single channels. Ion channels as discrete molecular entities were not directly observed until the patch-clamp era of the 1970s and 1980s.

"The HH model described the whole neuron." The 1952 model described a space-clamped patch of axon membrane — a small, uniform piece. Modeling how signals propagate along a branching dendritic tree required extensions of the model developed by Wilfrid Rall in the late 1950s using cable theory.

"The 1952 papers used computers." The Hodgkin-Huxley equations were solved numerically by hand, using a mechanical desk calculator. Huxley spent weeks performing the calculations. The integration of the equations required breaking time into small steps and computing each step sequentially — a procedure now done in milliseconds by a laptop but requiring months of manual effort in 1952.

Source: Huxley, A.F. (2002), "From overshoot to voltage clamp," Trends in Neurosciences, 25(11), 553–558 — Huxley's own account of the computational effort in 1952 and the context of the five papers.

FAQs

Q: What made the squid giant axon ideal for Hodgkin and Huxley's experiments?
A: Its diameter — up to 1 millimeter — is roughly 1,000 times larger than a typical mammalian axon. This allowed electrodes to be inserted along the axon's interior without destroying it, making intracellular recording and voltage clamping feasible. No equivalent preparation existed for mammalian neurons in 1939.

Q: What is the voltage-clamp technique and why was it essential?
A: Voltage clamping holds the membrane potential at a chosen level using a feedback amplifier. Because the membrane voltage stays fixed, the ion currents flowing through it can be measured directly from the current the amplifier injects to maintain the clamp. Without clamping, the action potential propagates and changes voltage simultaneously, making it impossible to separate cause from effect.

Q: What exactly does the Hodgkin-Huxley model consist of?
A: Four coupled differential equations describing how the membrane potential and three gating variables (m, h for sodium; n for potassium) change over time. The parameters in the equations were measured from voltage-clamp experiments at different membrane voltages and temperatures. Together they reproduce the full time course of the action potential.

Q: When did Hodgkin and Huxley win the Nobel Prize?
A: They shared the 1963 Nobel Prize in Physiology or Medicine with John Eccles, who demonstrated that synaptic transmission in the spinal cord also involves ionic conductance changes. The prize cited "discoveries concerning the ionic mechanisms involved in excitation and inhibition in the peripheral and central regions of the nerve cell membrane."

Q: How did World War II affect the research?
A: Hodgkin worked on radar development for the British Air Ministry, and Huxley contributed to antiaircraft fire-control systems. Their Plymouth experiments, begun in 1939, were interrupted until 1947. The wartime experience with electronics and feedback circuits arguably contributed to their technical sophistication in developing the voltage-clamp setup.

Q: Is the Hodgkin-Huxley model still accurate for modern neurons?
A: The basic framework is accurate for all neurons, but real neurons contain many more channel types than the three the HH model includes. Modern computational models of specific neuron types add conductances for calcium channels, various potassium channel subtypes, and hyperpolarization-activated channels, all described using the same HH formalism but with individually measured parameters.

Q: What came after the HH model in understanding ion channels?
A: Erwin Neher and Bert Sakmann's patch-clamp technique (1976) allowed single ion channels to be recorded in isolation, showing that the smooth HH conductances arise from the statistical averaging of thousands of channels that open and close randomly. The 1984 sequencing of the sodium channel protein identified the molecular structures underlying voltage gating.

The Resting Potential and Why Neurons Are Like Charged Batteries

To fully understand the action potential, it helps to understand what the HH model captured about the resting state — the condition the neuron maintains between impulses. A neuron at rest maintains a voltage difference of about −70 millivolts across its membrane, inside negative. This is not a passive equilibrium but an active, maintained state that costs metabolic energy.

The resting potential exists because different ions are distributed unequally across the membrane. Potassium is much more concentrated inside the cell than outside (roughly 140 millimolar inside vs. 5 millimolar outside), while sodium is much more concentrated outside (140 millimolar outside vs. 10–15 millimolar inside). These gradients are maintained by the sodium-potassium ATPase — a molecular pump that uses ATP to move three sodium ions out of the cell for every two potassium ions it brings in, running continuously to counteract the tendency of ions to diffuse down their concentration gradients.

Sodium-potassium ATPase — a membrane protein that uses one ATP molecule to pump three sodium ions out of the cell and two potassium ions in, maintaining the ion concentration gradients that power the action potential. Neurons consume a large fraction of the brain's energy budget running these pumps.

Think of the resting potential as a compressed spring. The concentration gradients represent stored potential energy, and the action potential is the release of that energy as ions flow rapidly down their gradients through channel openings. After the impulse, the ATPase must spend ATP to re-compress the spring — to restore the ion gradients that the action potential briefly dissipated. This is why sustained neural activity is metabolically expensive and why a brain, which is only about 2% of body weight, consumes roughly 20% of the body's energy at rest.

Source: Attwell, D. & Laughlin, S.B. (2001), "An energy budget for signaling in the grey matter of the brain," Journal of Cerebral Blood Flow & Metabolism, 21, 1133–1145 — quantitative analysis of the energy costs of neural signaling, showing that the dominant cost is restoring ion gradients after action potentials via the Na/K-ATPase.

The HH model incorporated a small background "leakage" conductance that, combined with the Nernst equilibrium potentials for sodium and potassium, reproduced the resting potential automatically. This was not a special assumption but a consequence of the measured properties of the membrane at rest — another demonstration that the model was capturing the actual physics, not just fitting a curve.

The Takeaway

Hodgkin and Huxley chose an animal with an accidentally large nerve fiber, built a feedback circuit to freeze the membrane in place, and measured the current needed to hold it there. From those currents they extracted a mathematical description of sodium and potassium flow precise enough to predict, from first principles, the exact shape of a nerve impulse. The fact that their four equations still power computational neuroscience more than 70 years later is not sentimentality — it is evidence that they found something genuinely true about how excitable membranes work, and that truth has not needed replacing.

---

Read next:

• The Loewi Frog Heart Experiment
• The Meselson-Stahl Experiment
• The Milgram Experiment