basic question

Stephen Black sblack at UBISHOPS.CA
Sun Sep 29 07:56:46 EST 1996

On Sun, 29 Sep 1996, Horace Veery wrote:

> I have a basic question regarding neurons.  I know that neural arbors are 
> generally characterized as afferent (dendritic) and efferent (axonal), but I 
> also know that it is difficult to distinguish one from the other in a 
> practical sense.  Is it actually possible to take any arbitrary segment of 
> neural fiber and determine the directionality of signals flowing along that 
> fiber?  Are there static, physical features that establish a directionality, 
> or is it theoretically possible for impluses to flow in either direction along 
> a neural fiber?  Have experiments been done which measure the directionality 
> of fibers, or which test whether impulses do travel in only one direction?  Is 
> the distinction between axon and dendrite absolute?  Is it a convenient 
> simplification or a well-established neurobiological fact?

I'm surprised to discover that I actually have something useful to say on 
this topic. First, I question the statement that dendrites are afferent 
and axons are efferent. Defining afferent as "into the nervous system" 
and "efferent" as "out of the nervous system" (towards muscles), axons can 
well be afferent, carrying information from sensory receptors to the brain.

I believe the confusion results because of the mistaken idea that 
dendrites should be defined as afferent to the neuronal cell body (soma) 
and axons as efferent to it (i.e. dendrites conduct towards the soma, 
axons away). This lead to conceptual confusion whenever the neuron is 
something other than the textbook motor neuron. For example, in a 
unipolar myelinated sensory neuron, the soma sits on a little stalk apart 
from the myelinated segment, along which the nerve impulse moves, first 
towards the soma, then away from it. (I don't dare attempt a diagram 
here, although it would help). The introductory textbook definition would 
have the part where the action potential is approaching the soma called a 
dendrite, and the part where it's leaving it, the axon. However, 
functionally, it's doing the same thing--conducting the nerve impulse.

Bodian, in an old paper, removed this confusion by defining the dendritic 
zone, concerned with receipt of information and generation of the action 
potential, and the axon, concerned with transmission of the nerve impulse.
The direction in relation to the soma is simply irrelevant. Most intro 
textbooks don't use this definition, and often end up contradicting 
themselves when they consider something other than the multipolar motor 

As for the remainder of this question, axons are inherently bidirectional.
If a nerve impulse is generated by electrical stimulation in the middle of
an axon, the nerve impulse will move simultaneously in opposite
directions--orthodromic (the normal direction) and
antidromic--"backwards". The orthodromic impulse will arrive at the
synapse, cause neurotransmitter release, and if circumstances are otherwise
favourable, generate an action potential on the postsynaptic neuron,
and keep going. However, the backwards antidromic impulse will die out
when it reaches the synapse, because the nervous system has no means of
crossing the synapse in the reverse direction. Neurophysiologists can use 
antidromic stimulation as a means of determining whether there is a 
synapse in a neural pathway.  In sum, conduction in the nervous system is 
unidirectional, but because of the properties of the synapse, not the axon.

As I'm at home, I can't provide references but if you need them, e-mail me
again and I'll dig them up. Bodian's insightful paper was published in
Science in the 60's, I believe, and the rest is available in textbooks of
neurophysiology and physiological psychology. For example, Eric Kandel's
books are higher level, James Kalat provides a simpler treatment in his
textbook Biological Psychology. 

Stephen Black, Ph.D.                      tel: (819) 822-9600 ext 2470
Department of Psychology                  fax: (819) 822-9661
Bishop's University                    e-mail: sblack at
Lennoxville, Quebec               
J1M 1Z7

More information about the Neur-sci mailing list