Increases in axon diameter increase velocity, largely because the longitudinal resistance of a larger axon is lower, so local circuits can spread further. Myelination reduces membrane capacitance and hence reduces the amount of current required to effect a particular membrane potential change which again means the currents can spread further. These effects are additive, so the fastest velocities are achieved in large myelinated axons.
One might ask why not have all nerve fibres large and myelinated? Apparently evolution did not result in this.
Cost benefit considerations can sometimes help. For example, the benefit could be to maximise the speed of transmission.
First, it may not be entirely true that speed is always beneficial. Consider muscle: high speed attack and escape behaviour seems a clear evolutionary advantage, yet clearly there is a need for slow controlled movements in life maintaining movements like feeding, grooming, and even stealthy attacks and escapes. So we have evolved both slow and fast muscles, muscle fibres and muscle proteins. There would be little point in maximising the speed of an axon innervating a slow muscle fibre - not unless that maximisation could be without cost.
So what is the cost of speed? Myelination itself is one - many cells to subserve the action of a single axon, but possibly some energetic benefits from the ionic pumping efficiencies. Size is another. Myelinated axons are all large, and the myelination makes them larger. The space available to contain axons is not unlimited - nerves have to be able to fit inside the body parts they subserve. So the cost of making all axons large and myelinated to achieve uniform maximised speed is a smaller number of axons. Fewer axons means poorer control of effectors and poorer sensation. Poor sensation can be just as big an evolutionary threat as low ground speed.
So while there is an evolutionary benefit to fast axons, there is also a benefit to compromises where some axons are slow and space efficient, allowing for more parallel paths, and richer information capabilities at a compromised speed.
The conduction velocity vs. axon diameter graph that follows is instructive.
This prediction comes from theoretical analysis by W.A.H. Rushton (1951, J. Physiol. 115: 101-122). The myelinated fibre diameter used is that including the myelin.
It is most useful to examine the lower end of this curve.
Myelination gives faster speeds only above about 1μm diameter. Below that diameter unmyelinated axons are slightly faster. (This is partly because the fractional space occupied by myelin starts to be very great in small axons.) It appears that evolutionary pressures have taken this design limitation into account, for myelinated axons less than 1μm are rare, and unmyelinated axons larger than 1μm are similarly rare.
Department of Physiology,
University of Sydney
Last updated 5 May 2010