Saltatory Conduction

What saltatory conduction is

Picture the axon as a leaky garden hose carrying not water but electrical charge. In a bare (unmyelinated) fiber, the action potential has to be regenerated at every single patch of membrane down the whole length — it crawls. Saltatory conduction is the trick myelinated axons use to skip almost all of that work. The axon is wrapped in myelin, a thick insulating sleeve, broken only by tiny bare gaps called nodes of Ranvier spaced roughly 1–2 mm apart. The spike is fully regenerated only at the nodes; between them, the depolarization races down the insulated internode as fast passive current. The action potential appears to leap from node to node — hence "saltatory," from the Latin saltare, to leap.

The payoff is dramatic. A heavily myelinated A-alpha motor fiber conducts at 80–120 m/s — fast enough that a signal crosses an adult human's 1-metre sciatic nerve in about 10 milliseconds. An unmyelinated C fiber of similar diameter manages only 0.5–2 m/s. Same axon width, the same ions, the same channels — but a 50-fold difference in speed, bought entirely by where the cell chooses to put its insulation and its channels.

How the leap actually works, step by step

Saltatory conduction is best understood through cable theory — the same math used for undersea telegraph cables. An axon's ability to carry a passive voltage signal is governed by two properties of its membrane: resistance (how well it stops current leaking sideways out of the cable) and capacitance (how much charge it takes to change the voltage across the wall). Two derived numbers matter:

• The length constant λ — the distance over which a passive depolarization decays to 37% (1/e) of its value. A larger λ means current spreads farther before fading. λ scales as the square root of (membrane resistance / axoplasmic resistance).
• The time constant τ — how quickly the membrane voltage responds, equal to membrane resistance × capacitance. A smaller τ means the membrane charges faster.

Myelin is up to 100–150 concentric wraps of glial-cell membrane spiralled tightly around the internode. Stacking lipid bilayers in series multiplies the effective membrane resistance and divides the capacitance — so λ grows (current spreads farther, often ~1 mm) while τ shrinks (the membrane charges almost instantly). With both working in the right direction, a depolarization at one node spreads passively to the next node before it has decayed below threshold.

Here is the cycle. (1) An action potential fires at node N — voltage-gated Na+ channels there snap open, Na+ floods in, and the node depolarizes from about −70 mV to +30 mV in well under a millisecond. (2) That positive charge spreads as local-circuit current down the well-insulated internode, almost without loss because little leaks out through the high-resistance myelin and little is wasted charging the low-capacitance wall. (3) The current arrives at node N+1 still well above the ~−55 mV threshold and triggers a fresh full action potential there. (4) Repeat. The spike is regenerated only at the nodes, so it "jumps" the internodes; the just-fired node N is briefly refractory, which keeps the wave moving one way. The whole node-to-node hop takes roughly 20 microseconds, far less than the ~1 ms it would take to regenerate the spike continuously across the same distance.

The molecular cast at the node

A node of Ranvier is not just a gap — it is one of the most precisely organized patches of membrane in the body, and each zone has a distinct molecular signature:

• The node itself (~1 µm). Densely packed with Nav1.6 voltage-gated Na+ channels at 1,000–2,000 per µm², held in place by the scaffolding proteins ankyrin-G and βIV-spectrin. This is where regeneration happens. Fast-spiking K+ channels (Kv3, KCNQ2/3) also sit here to help reset the node.
• The paranode. The myelin loops seal tightly against the axon on either side of the node through septate-like junctions built from Caspr, contactin, and the glial protein neurofascin-155. These junctions act as a fence, blocking lateral diffusion and electrically isolating the node.
• The juxtaparanode. Just under the myelin, this zone concentrates Kv1.1/Kv1.2 delayed-rectifier K+ channels, kept away from the node by the paranodal fence; they dampen re-excitation and stabilize conduction.
• The internode. The long (1–2 mm) myelinated stretch — produced by one oligodendrocyte process in the CNS or one whole Schwann cell in the PNS. Almost no voltage-gated channels live under the myelin; it is electrically silent insulation.

The myelin itself is made by two different cells in the two halves of the nervous system. In the central nervous system, a single oligodendrocyte sends out up to 50 processes, each myelinating one internode on a different axon. In the peripheral nervous system, a single Schwann cell wraps exactly one internode of one axon. Myelin is ~70–80% lipid by dry weight (rich in galactocerebroside and the proteins MBP and, in the PNS, P0/PMP22), which is what makes it such a good insulator.

Saltatory vs continuous conduction

Property	Saltatory (myelinated)	Continuous (unmyelinated)
Conduction velocity	10–120 m/s	0.5–2 m/s
Where the spike regenerates	Only at nodes of Ranvier (~1 µm, every 1–2 mm)	At every patch of membrane
Na+ channel distribution	Clustered at nodes (1000–2000/µm²)	Uniform, sparse (tens/µm²)
Internodal membrane	High resistance, low capacitance (myelin)	Bare; ordinary resistance & capacitance
Length constant λ	Large (~1 mm) — current spreads far	Small (~0.1–0.5 mm)
Energy (ATP) per unit length	Low — only nodes fire and must be re-pumped	High — whole membrane fires
Speed for a given diameter	5–50× faster	Baseline
Cell that builds the sheath	Oligodendrocyte (CNS) / Schwann cell (PNS)	None
Failure mode	Demyelination (MS, Guillain-Barré, CMT)	Na+ channel block, gradient loss
Typical fibers	A-alpha motor, A-beta touch, A-delta fast pain	C fibers (slow pain, autonomic)

The numbers that make it work

Saltatory conduction is a quantitative phenomenon — the speed-up follows directly from a handful of physical figures:

Quantity	Value	Why it matters
Node width	~1 µm	Tiny window where the membrane is exposed and excitable
Internode (node spacing)	~1–2 mm (scales with diameter, ≈100× fiber width)	Distance the spike leaps each hop — ~1000× the node width
Myelin wraps	up to 100–150 lamellae	Each layer adds resistance, drops capacitance
g-ratio (axon/fiber diameter)	~0.6 (optimal for speed)	Theoretical sweet spot balancing axon size against sheath thickness
Na+ channel density at node	1000–2000 / µm²	Enough inward current to re-fire reliably (high safety factor)
Node-to-node delay	~20 µs	Internodal conduction time — far faster than continuous
Resting / peak voltage at node	−70 mV → +30 mV	The full spike that must be regenerated each hop
Velocity rule of thumb (myelinated)	≈ 6 × diameter(µm) m/s	A 20 µm fiber ≈ 120 m/s; velocity scales linearly with diameter
Velocity rule (unmyelinated)	∝ √diameter	Why bare fibers must get enormous (squid 1 mm axon) to go fast

That last contrast is the deep reason myelin won. In an unmyelinated fiber, velocity rises only with the square root of diameter, so to hit 100 m/s a bare axon would need to be roughly a millimetre wide — the trick the squid giant axon uses, and utterly impractical to pack billions of into a brain. In a myelinated fiber, velocity rises linearly with diameter, so a 20 µm fiber hits the same speed at 1/50th the width. Myelin lets vertebrates fit fast wiring into a compact nervous system.

Where it shows up — organisms, reflexes, and disease

• Your knee-jerk reflex. Tap the patellar tendon and the stretch signal races up a myelinated Ia afferent and back down an A-alpha motor axon at ~80–120 m/s, contracting the quadriceps in under 50 ms. The same loop on unmyelinated fibers would be far too slow to be useful as a postural reflex.
• Slow vs fast pain. The sharp "first pain" when you touch a hot stove travels on thinly myelinated A-delta fibers at 5–30 m/s; the dull, lingering "second pain" arrives a moment later on unmyelinated C fibers at ~1 m/s. You can literally feel the conduction-velocity difference as two waves of sensation.
• Multiple sclerosis (MS). Autoimmune destruction of CNS myelin slows or blocks saltatory conduction, producing the disease's hallmark slowed nerve conduction velocities and heat-sensitive symptoms (Uhthoff's phenomenon). Remyelination by surviving oligodendrocyte precursors can partially restore conduction.
• Guillain-Barré syndrome and CMT. Guillain-Barré is acute autoimmune demyelination of peripheral nerves, causing rapid ascending weakness. Charcot-Marie-Tooth disease type 1 is an inherited peripheral demyelinating neuropathy, most often from a duplication of the PMP22 gene that disrupts Schwann-cell myelin.
• Convergent evolution in invertebrates. Myelin-like insulating wraps evolved independently in some shrimp and copepods, giving them fast escape reflexes — direct evidence that saltatory-style conduction is such a good solution that nature reinvented it.
• Local anesthetics target the node. Drugs like lidocaine block the Nav channels concentrated at the nodes; because regeneration depends on those clusters, blocking even a few consecutive nodes is enough to stop conduction entirely.

Common misconceptions

• The signal physically "jumps" across empty space. Nothing leaps a gap. The action potential is regenerated only at the nodes, but the membrane is continuous; what crosses the internode is ordinary passive current flowing inside the axoplasm. "Jump" is a description of where regeneration happens, not of a physical leap.
• Myelin contains the ion channels. Myelin is essentially inert insulation — the voltage-gated Na+ channels are clustered at the bare nodes, not under the sheath. Strip the myelin and you do not expose channels; you expose channel-poor internodal membrane that cannot regenerate the spike.
• Saltatory conduction makes the action potential itself bigger or faster. The spike at each node is the same all-or-nothing event as in any neuron. Myelin doesn't change the action potential; it changes how far the passive current spreads between regenerations, which is what raises the overall propagation velocity.
• Thicker fibers are faster only because they're myelinated. Both matter, but differently. Diameter raises velocity in any fiber; myelin changes the scaling law from √diameter to linear-in-diameter and adds the node-skipping leap on top. The fastest fibers are both thick and heavily myelinated.
• More myelin is always better. There's an optimum. If the sheath is too thick relative to the axon (g-ratio too low) the axon core gets thin and internal resistance rises; if too thin, insulation is poor. The measured sweet spot is a g-ratio (axon-to-fiber diameter) of about 0.6.
• Only vertebrates could ever evolve this. Compact myelin is a vertebrate signature, but functionally equivalent insulating wraps arose independently in some crustaceans — saltatory-style conduction is a convergent solution, not a one-off.