# Nerve Impulse

Laplace’s Mini World · Plate 55 · Life & Medicine · Life

> A nerve cell is about 70 millivolts negative inside. Pushed past threshold, the voltage flips and runs down the axon.

- Interactive model: https://world.lapla.si/en/impulse/
- Embed: https://world.lapla.si/embed/en/impulse/
- 中文: https://world.lapla.si/zh/impulse/
- Updated: 2026-10-29

## 01 Lower Inside Than Out

At rest the inside of a nerve cell is negative to the outside, typically about −70 mV; across neurons, −40 to −90 mV. The voltage sits across the cell membrane, about 5 nm thick: 70 mV over 5 nm is a field of about 14 million volts per metre.

The ions on the two sides differ. In a mammalian neuron, potassium is about 140 millimoles per litre inside and 5 outside; sodium the other way round, 5⁠–⁠15 inside and about 145 outside.

In the resting membrane the open channels are mostly ones that let potassium through. Potassium leaves down its gradient, taking positive charge with it, and the inside goes negative; at about −89 mV the pull of the voltage balances the push of the gradient: the potassium equilibrium potential. Sodium and chloride leak a little too, so the resting potential is near it, not at it.

*Rest about −70 mV · membrane about 5 nm · K⁺ 140 in, 5 out · Na⁺ 5⁠–⁠15 in, 145 out (mM)*

## 02 A Pump and a Leak

The gradients are kept by the sodium–potassium pump. For each ATP it uses, it moves 3 sodium ions out and 2 potassium ions in. At 37 °C one pump can turn about 160 times a second; it normally runs at 10⁠–⁠15% of that.

The pump’s own voltage is small: in the squid axon it changes the resting potential by about a millivolt or less. The voltage comes mostly from potassium leaking out down the gradient the pump builds. Stop the pump and the resting potential decays only slowly; the potassium gradient typically lasts many minutes.

The work is not cheap. A textbook estimate puts the pump at 20⁠–⁠40% of the brain’s energy.

*3 Na⁺ out · 2 K⁺ in · per ATP · the pump’s direct voltage is small*

## 03 Past the Threshold

A stimulus lifts the membrane voltage a little. If it is not lifted far enough, the voltage falls back and nothing travels.

Past a certain level, voltage-⁠gated sodium channels start opening in numbers. Sodium rushes in, the voltage rises further and opens still more channels: the process feeds itself. The impulse is all-⁠or-⁠none: a stronger stimulus does not make a bigger impulse; strength is carried by how many impulses come each second.

The threshold is not one fixed number; it varies with the cell and the stimulus. The figure is this page’s patch of membrane, computed with the equations of Hodgkin and Huxley (1952) at 18.5 °C: shocks of 0.5 and 0.9 times threshold fall back, 1.1 and 3 times give impulses of nearly the same height. On the model, the start of the axon gets 0.9 times, then 1.1 times.

*Below threshold: falls back · above: all-⁠or-⁠none · strength as rate*

## 04 Sodium First, Then Potassium

The impulse rises and falls through two kinds of voltage-⁠gated channel. As the voltage rises, sodium channels open quickly, sodium rushes in and the membrane swings positive, close to the sodium equilibrium potential. Soon after opening, the sodium channels inactivate and let no more through.

Potassium channels open more slowly. As they open, potassium flows out and pulls the voltage back negative, briefly below rest; then they close.

An impulse usually lasts about 1 ms; in the largest mammalian fibres at body temperature, 0.4⁠–⁠0.45 ms. The figure is one impulse from the 1952 equations: sodium channels open first and close first, potassium channels open later. On the enlarged membrane, the sodium channel’s inner lid is a drawing.

*Sodium in: up · sodium inactivates, potassium out: down · about 1 ms*

## 05 Only Forwards

In a stretch of membrane that has just fired, the sodium channels are still inactivated and need a while to recover. In that time no stimulus, however strong, starts a second impulse: the absolute refractory period. For a while after, it takes a stronger stimulus than usual: the relative refractory period.

So a travelling impulse does not turn back: the membrane ahead has not fired and is easily pushed past threshold; the membrane behind is still refractory.

In this page’s model (18.5 °C), for 1.5 ms after an impulse even ten times the threshold fails; at 2 ms it takes 5.3 times; by 5 ms about once is enough. Real neurons differ with the cell and the temperature.

*Absolute refractory · relative refractory · no turning back*

## 06 A Bare Axon

The impulse is made again, stretch by stretch, along the axon. The stretch that is firing is positive inside; charge moves through the fluid inside the axon to the stretch just ahead and pushes it past threshold, and that stretch fires. Each stretch makes it afresh, so it does not fade however far it goes.

That is slow. Unmyelinated axons conduct at 0.5⁠–⁠10 m/s, faster when thicker, roughly with the square root of the diameter. C fibres, which carry dull pain, are about 1 µm across or less and conduct at under 2 m/s.

On the model, the myelin is gone. This page computes the same axon (10 µm across, 18.5 °C) bare: the impulse travels at 2.7 m/s and creeps along the model, slowed 4,⁠000 times.

*0.5⁠–⁠10 m/s · thicker is faster · model: 2.7 m/s*

## 07 Node to Node

Most thick axons are wrapped in myelin. In peripheral nerves a Schwann cell winds its own membrane round a length of axon many times, up to about a hundred layers, over about 1 mm; in the brain and spinal cord oligodendrocytes do it, each wrapping dozens of lengths. Between two sheaths a short stretch of membrane is left bare: a node of Ranvier, about 1 µm long in the rat optic nerve.

Sodium channels crowd onto the nodes: in rabbit sciatic nerve, about 12,⁠000 per square micrometre at the node, no more than about 25 under the myelin. The impulse is made again only at the nodes; between them the current passes along inside the axon, losing little. So it seems to leap from node to node. What leaps is where it is made again: the current flows all the time, inside the axon and in the fluid outside, not across a gap.

In this page’s model the same axon, with 100 wraps of myelin and a node every millimetre, conducts at about 39 m/s, about 14 times faster than bare, about 25 µs per node. An impulse is far longer than one internode: of the model’s 31 nodes, up to 16 are above 0 mV at once.

*Node about 1 µm · internode about 1 mm · about 12,⁠000 sodium channels per µm² · model: about 39 m/s*

## 08 Fast Fibres and Slow Ones

Nerve fibres fall into classes by size and speed. In the cat’s hind limb, the largest, group I (Aα) fibres from muscle receptors conduct at about 90 m/s; the Aβ fibres of touch in the skin at about 66; the Aδ fibres of pain and temperature at about 20; unmyelinated C fibres at under 2. These are measurements in cats.

In myelinated fibres the speed goes roughly with the total diameter: about 6 m/s per micrometre in the largest fibres, 4.4⁠–⁠4.6 in smaller ones. The fastest myelinated fibres reach over 100 m/s.

Over 1 m, 94 m/s takes about 11 ms; 1 m/s takes a whole second. An injury often brings a sharp first pain and then a duller, longer one: the first travels in Aδ fibres, the second in C fibres.

*Aα about 90 m/s · Aβ about 66 · Aδ about 20 · C under 2 (cat)*

## 09 The Squid’s Giant Axon

These rules were first worked out on the giant axon of the squid. It can be up to about 1 mm across, wide enough to slide a fine glass capillary along its inside. In the summer of 1939, at Plymouth in England, Hodgkin and Huxley recorded the impulse from inside the axon: the voltage did not just fall to zero, it went past zero to positive. That same summer Curtis and Cole did so at Woods Hole in the United States.

With the voltage clamp (made by Cole and Marmont in 1947⁠–⁠1949, improved by them and Katz), they held the membrane voltage steady, measured the current that flowed and split it into sodium and potassium parts. In 1952 they described the two currents with a few equations and worked out, on a hand-⁠operated Brunsviga calculator, how the impulse travels; Hodgkin recalled that this step took about three weeks.

The computed speed was 18.8 m/s; the same axon measured 21.2 m/s (18.5 °C). Recomputed on this page with the same equations: 18.7 m/s. In 1963 Eccles, Hodgkin and Huxley shared the Nobel Prize in Physiology or Medicine.

*1939: the inside goes positive · 1952: 18.8 m/s computed, 21.2 measured · recomputed here: 18.7*

## 10 When Myelin Is Damaged

Multiple sclerosis is a disease in which the immune system attacks the myelin in the brain and spinal cord. Where it is damaged, signals slow down or are blocked; the symptoms depend on where the damage is and differ from person to person. The most common course alternates relapses and remissions.

The MS International Federation’s Atlas of MS counts about 3.1 million people with MS worldwide (2024 data); the average age at diagnosis is 32, and about seven in ten are women. It is not contagious. There is no cure yet, but treatments can reduce the number and severity of relapses and delay long-⁠term progression.

On the model, one sheath has thinned from 100 wraps to 10: this page computes that the impulse reaches node 7 about 130 µs later. With the whole sheath gone, and no sodium channels under it, the impulse stops there. This is a model axon, not any patient’s.

*The immune system attacks central myelin · slowed or blocked · about 3.1 million people*

## 11 At the Axon’s End

At the axon terminal the impulse opens voltage-⁠gated calcium channels; calcium flows in and makes vesicles release their neurotransmitter into the synaptic cleft, where it acts on the next cell. A motor neuron’s terminals end on muscle fibres; neurons in the brain end on other neurons.

Between the impulse arriving and the transmitter acting there is a short delay: about 0.5 ms at the frog nerve–muscle junction at 20 °C, about 150 µs at rat cerebellar synapses near body temperature.

On the speech neuroprosthesis page, electrodes record exactly these impulses from neurons in the motor cortex.

*Calcium channels open · transmitter released · synaptic delay about 0.5 ms (frog, 20 °C)*

## 12 Not a Wire

The impulse is often likened to a current in a wire, but the two are very different. An electrical signal in copper cable travels at about 200,⁠000 km a second, about two-⁠thirds of the speed of light, 5 ns per metre; the impulse in the squid’s giant axon takes about 47 ms per metre.

An axon is a poor wire: salt water inside, a leaky membrane outside. Passively, a voltage does not get far: in this page’s 10 µm axon the space constant is about 1 mm, and each millimetre leaves about 37%. The impulse goes a metre because it is made again in each stretch of membrane, or at each node.

Nor do ions run from one end of the axon to the other. They only cross the membrane, a little: one impulse costs a 1 µm axon about one three-⁠thousandth of its potassium, the squid’s giant axon about one millionth. The current along the axon is charge handed on, step by step, through salt water.

*Copper about 5 ns/m · the impulse milliseconds per metre · made again stretch by stretch · ions only cross*

## 13 A Node of Ranvier, Taken Apart

Lift away the myelin on either side of a node: on the node, rings of crowded sodium channels. Next to it, at the paranodes, junctions seal the edges of the myelin to the axon. A little further out, hidden under the myelin, are potassium channels (Kv1).

Each sheath is wound by one Schwann cell; the swelling on it is the cell’s nucleus. The cut sheath shows the layers of membrane.

*Myelin (two sheaths) · potassium channels · sodium channels*

## 14 Laboratory: Stimulus and Myelin

This is the page’s model: an axon 10 µm across, its channels from the equations of Hodgkin and Huxley (1952) at 18.5 °C. The stimulus is a 0.1 ms pulse of current at the start of the axon, chosen as a multiple of the threshold.

With myelin, the axon has 100 wraps and a node every millimetre, the nodes carrying ten times the 1952 channel densities (the page’s choice); without, the channels are spread evenly along the whole membrane. The speed is measured well away from the stimulus, between nodes 12 and 24.

Each press runs it afresh, slowed 3,⁠000 times. The readings are the model’s; real nerves at body temperature are much faster.

*Readings: did it reach the end · speed*

**Try this**: Set the stimulus to 0.9 times the threshold, then 1.1 times: does the impulse reach the end of the axon? Then take the myelin away and compare the speed.

## What the model leaves out

The neuron is this page’s motor neuron, enlarged and shortened: a real internode is about a hundred fibre diameters long, here only a few; a node is about 1 µm, here far longer; the dendrites’ branches are generated; the axon has seven sheaths, and the cut one is drawn with five wraps (a real one has up to about a hundred). The colour of the impulse comes from the page’s computation, slowed several thousand times: brightest where the membrane is firing now, fading behind; a faint wash where it is still depolarised. The membrane patch is enlarged about 3.6 million times; the proteins are simplified after their published overall shapes, helices drawn as rods, the sodium channel’s inactivation gate as a hinged lid; the ions are drawn far larger than they are, and their numbers are schematic. The laboratory’s axon is the page’s model: 18.5 °C, the node channel densities the page’s choice.

## Easy to get wrong

The resting potential is not made mainly by the pump directly; it comes from potassium leaking out down the gradient the pump keeps up. The impulse is not electrons flowing in a wire, and sodium ions do not run along the axon: it is channels opening, stretch after stretch of membrane. In saltatory conduction what leaps is where the impulse is made again; the current flows all the time inside the axon and in the fluid outside. A stronger stimulus does not make a bigger impulse. One impulse hardly changes the ion concentrations. Thresholds and refractory periods vary with the cell and the temperature; the page’s numbers are the model’s.

## Common questions

### What is a nerve cell’s resting potential?

At rest the inside of a nerve cell is negative to the outside, typically about −70 mV; across neurons, −40 to −90 mV.

See: [01 Lower Inside Than Out](https://world.lapla.si/en/impulse/#rest)

### Does the sodium–potassium pump make the resting potential?

The pump’s own voltage is small: in the squid axon it changes the resting potential by about a millivolt or less.

See: [02 A Pump and a Leak](https://world.lapla.si/en/impulse/#pump)

### What does all-or-none mean for a nerve impulse?

The impulse is all-⁠or-⁠none: a stronger stimulus does not make a bigger impulse; strength is carried by how many impulses come each second.

See: [03 Past the Threshold](https://world.lapla.si/en/impulse/#threshold)

### Does saltatory conduction mean the impulse jumps through the air?

What leaps is where it is made again: the current flows all the time, inside the axon and in the fluid outside, not across a gap.

See: [07 Node to Node](https://world.lapla.si/en/impulse/#myelin)

### Is a nerve impulse as fast as a current in a wire?

An electrical signal in copper cable travels at about 200,⁠000 km a second, about two-⁠thirds of the speed of light, 5 ns per metre; the impulse in the squid’s giant axon takes about 47 ms per metre.

See: [12 Not a Wire](https://world.lapla.si/en/impulse/#wire)

## Sources

1. [Purves et al. (eds): Neuroscience, 2nd edition (2001), NCBI Bookshelf](https://www.ncbi.nlm.nih.gov/books/NBK10799/)
2. [Alberts et al.: Molecular Biology of the Cell, 4th edition (2002), NCBI Bookshelf](https://www.ncbi.nlm.nih.gov/books/NBK26910/)
3. [Hodgkin & Huxley (1952): A quantitative description of membrane current and its application to conduction and excitation in nerve, J Physiol 117, 500⁠–⁠544](https://doi.org/10.1113/jphysiol.1952.sp004764)
4. [Hodgkin, Huxley & Katz (1952): Measurement of current-⁠voltage relations in the membrane of the giant axon of Loligo, J Physiol 116, 424⁠–⁠448](https://doi.org/10.1113/jphysiol.1952.sp004716)
5. [Hodgkin & Katz (1949): The effect of sodium ions on the electrical activity of the giant axon of the squid, J Physiol 108, 37⁠–⁠77](https://doi.org/10.1113/jphysiol.1949.sp004310)
6. [Hodgkin & Huxley (1939): Action Potentials Recorded from Inside a Nerve Fibre, Nature 144, 710⁠–⁠711](https://doi.org/10.1038/144710a0)
7. [Hodgkin (1963): The ionic basis of nervous conduction (Nobel Lecture)](https://www.nobelprize.org/uploads/2018/06/hodgkin-lecture.pdf)
8. [Huxley (1963): The quantitative analysis of excitation and conduction in nerve (Nobel Lecture)](https://www.nobelprize.org/uploads/2018/06/huxley-lecture.pdf)
9. [Hodgkin (1976): Chance and design in electrophysiology, J Physiol 263, 1⁠–⁠21](https://pmc.ncbi.nlm.nih.gov/articles/PMC1307686/)
10. [Huxley & Stämpfli (1949): Evidence for saltatory conduction in peripheral myelinated nerve fibres, J Physiol 108, 315⁠–⁠339](https://pmc.ncbi.nlm.nih.gov/articles/PMC1392492/)
11. [Keynes & Lewis (1951): The sodium and potassium content of cephalopod nerve fibres, J Physiol 114, 151⁠–⁠182](https://doi.org/10.1113/jphysiol.1951.sp004609)
12. [Boyd & Kalu (1979): Scaling factor relating conduction velocity and diameter for myelinated afferent nerve fibres in the cat hind limb, J Physiol 289, 277⁠–⁠297](https://doi.org/10.1113/jphysiol.1979.sp012737)
13. [Ritchie & Rogart (1977): Density of sodium channels in mammalian myelinated nerve fibers, PNAS 74, 211⁠–⁠215](https://doi.org/10.1073/pnas.74.1.211)
14. [Rasband & Peles (2016): The Nodes of Ranvier: Molecular Assembly and Maintenance, Cold Spring Harb Perspect Biol 8, a020495](https://doi.org/10.1101/cshperspect.a020495)
15. [Arancibia-⁠Cárcamo et al. (2017): Node of Ranvier length as a potential regulator of myelinated axon conduction speed, eLife 6, e23329](https://doi.org/10.7554/eLife.23329)
16. [Morell & Quarles: The Myelin Sheath, in Basic Neurochemistry, 6th ed. (1999), NCBI Bookshelf](https://www.ncbi.nlm.nih.gov/books/NBK27954/)
17. [Skou (1997): The identification of the sodium-⁠potassium pump (Nobel Lecture)](https://www.nobelprize.org/uploads/2018/06/skou-lecture.pdf)
18. [Katz & Miledi (1965): The effect of temperature on the synaptic delay at the neuromuscular junction, J Physiol 181, 656⁠–⁠670](https://pmc.ncbi.nlm.nih.gov/articles/PMC1357674/)
19. [Katz (1970): On the quantal mechanism of neural transmitter release (Nobel Lecture)](https://www.nobelprize.org/uploads/2018/06/katz-lecture.pdf)
20. [Sabatini & Regehr (1996): Timing of neurotransmission at fast synapses in the mammalian brain, Nature 384, 170⁠–⁠172](https://pubmed.ncbi.nlm.nih.gov/8906792/)
21. [Neher (1991): Ion channels for communication between and within cells (Nobel Lecture)](https://www.nobelprize.org/uploads/2018/06/neher-lecture.pdf)
22. [The Nobel Prize in Physiology or Medicine 1963](https://www.nobelprize.org/prizes/medicine/1963/summary/)
23. [NINDS: Multiple Sclerosis](https://www.ninds.nih.gov/health-information/disorders/multiple-sclerosis)
24. [MS International Federation: Atlas of MS, 3rd edition, epidemiology update 2026](https://www.msif.org/wp-content/uploads/2026/09/Atlas-Epidemiology-report-text-core-data-2026-FINAL.pdf)
25. [MS International Federation: What is MS?](https://www.msif.org/about-ms/what-is-ms/)
26. [China National Committee for Terms in Sciences and Technologies: termonline](https://www.termonline.cn/)

## Data

The neuron is this page’s model of a motor neuron, enlarged and shortened: its dendrites are generated, the sheaths are drawn far shorter and the nodes far longer than real. The membrane patch is enlarged about 3.6 million times; its proteins are simplified after their published overall shapes; the ions are a schematic layer. The spike on the model and the laboratory are this page’s computation with the equations of A. L. Hodgkin and A. F. Huxley, J. Physiol. 117, 500 (1952), at 18.5 °C, in docs/plates/impulse/computed-⁠hh.py: an axon 10 µm across, bare, or with 100 wraps of myelin and a node every millimetre carrying ten times the 1952 channel densities (the page’s choice). Ion concentrations and fibre speeds are from Purves et al., Neuroscience, 2nd ed. (2001) and Boyd & Kalu, J. Physiol. 289, 277 (1979, cat hind limb); multiple sclerosis figures from the MS International Federation’s Atlas of MS (2024 data, published 2026) and NINDS. Material presets, parts and the softbox studio of the site’s model kit are ported from AirsupHQ/machine-⁠skill (MIT, © 2026 Air Supply Technologies Inc.).

## Licence CC BY-NC 4.0

The text, pictures and models on this page are licensed CC BY-NC 4.0: free to use and adapt for teaching and other non-commercial purposes, with credit to world.lapla.si. Third-party material listed above keeps its own licence.

https://creativecommons.org/licenses/by-nc/4.0/deed.en

## Cite this plate

Laplace’s Mini World, Plate 55, “Nerve Impulse”, https://world.lapla.si/en/impulse/, CC BY-NC 4.0, updated 2026-10-29.