510
M. V. L. BENNETT
the stimulus. If a sufficiently large cathodal stimulus is given the receptor
discharge is stopped completely during the pulse.
The graph of Fig. 14 shows the number of impulses during the first
100 msec of the stimulus plotted against the strength of the stimulus.
This represents an average frequency over a time when considerable accommodation occurs, but qualitatively the same result is obtained for
shorter intervals. The change in average frequency is linearly related
to the stimulus strength for small stimuli. For larger stimuli, the slope of
the relation between impulse frequency and stimulus strength decreases.
Stimuli of the order of 0.1 mV produce readily detectable changes in
frequency. If sinusoidal stimuli are applied, the maximum sensitivity is
obtained for frequencies of 30-50 Hz (R. E. Poppele and M. V. L.
Bennett, unpublished data). Sensitivity falls off at both lower and higher
frequencies. The falloff at low frequencies is equivalent to accommodation observed with rectangular stimuli. Accommodation reduces the dc
sensitivity of the receptors, but considerable sensitivity remains at frequencies as low as 0.1 or even 0.01 Hz.
Several lines of evidence indicate that transmission at the synapse
between receptor cell and nerve fiber is chemically mediated. First, the
morphological characteristics of the synapse are those typical of chemically mediated transmission (Fig. 12), and there are no regions of close
apposition between the membranes such as are found at sites of electrically mediated transmission (see Figs. 71 and 72 of Chapter 10, this
volume).
A second indication of chemically mediated transmission is that if a
strong brief anodal stimulus is given, the evoked responses long outlasts
the stimulus (Fig. 15A). If such a stimulus is followed by a strong
cathodal stimulus, one much stronger than that required to block the
spontaneous discharge, there is virtually no effect on the maintained
discharge (Fig. 15B). Thus, following a brief stimulus there is a maintained increase in excitation that is inaccessible to what is normally an
inhibitory stimulus. It is difficult to explain this phenomenon assuming
electrically mediated transmission. It is easy to explain it, if transmission
is chemically mediated. The effect could, for example, be a result of transmitter accumulated in the synaptic gap. It is more likely that secretion
persists beyond the depolarization required to evoke it. Possibly a brief
excitatory stimulus “commits” the cell to secrete a certain amount of
transmitter [as might be caused by an accumulation of Ca2+ inside the
cell, a process which is thought to occur at the neuromuscular junction
(Katz and Miledi, 1968)l. The latter kind of explanation is supported
by the fact that the high frequency impulse discharge at the end of a
strong, long-lasting anodal pulse stops almost immediately upon termina-
M. V. L. BENNETT
the stimulus. If a sufficiently large cathodal stimulus is given the receptor
discharge is stopped completely during the pulse.
The graph of Fig. 14 shows the number of impulses during the first
100 msec of the stimulus plotted against the strength of the stimulus.
This represents an average frequency over a time when considerable accommodation occurs, but qualitatively the same result is obtained for
shorter intervals. The change in average frequency is linearly related
to the stimulus strength for small stimuli. For larger stimuli, the slope of
the relation between impulse frequency and stimulus strength decreases.
Stimuli of the order of 0.1 mV produce readily detectable changes in
frequency. If sinusoidal stimuli are applied, the maximum sensitivity is
obtained for frequencies of 30-50 Hz (R. E. Poppele and M. V. L.
Bennett, unpublished data). Sensitivity falls off at both lower and higher
frequencies. The falloff at low frequencies is equivalent to accommodation observed with rectangular stimuli. Accommodation reduces the dc
sensitivity of the receptors, but considerable sensitivity remains at frequencies as low as 0.1 or even 0.01 Hz.
Several lines of evidence indicate that transmission at the synapse
between receptor cell and nerve fiber is chemically mediated. First, the
morphological characteristics of the synapse are those typical of chemically mediated transmission (Fig. 12), and there are no regions of close
apposition between the membranes such as are found at sites of electrically mediated transmission (see Figs. 71 and 72 of Chapter 10, this
volume).
A second indication of chemically mediated transmission is that if a
strong brief anodal stimulus is given, the evoked responses long outlasts
the stimulus (Fig. 15A). If such a stimulus is followed by a strong
cathodal stimulus, one much stronger than that required to block the
spontaneous discharge, there is virtually no effect on the maintained
discharge (Fig. 15B). Thus, following a brief stimulus there is a maintained increase in excitation that is inaccessible to what is normally an
inhibitory stimulus. It is difficult to explain this phenomenon assuming
electrically mediated transmission. It is easy to explain it, if transmission
is chemically mediated. The effect could, for example, be a result of transmitter accumulated in the synaptic gap. It is more likely that secretion
persists beyond the depolarization required to evoke it. Possibly a brief
excitatory stimulus “commits” the cell to secrete a certain amount of
transmitter [as might be caused by an accumulation of Ca2+ inside the
cell, a process which is thought to occur at the neuromuscular junction
(Katz and Miledi, 1968)l. The latter kind of explanation is supported
by the fact that the high frequency impulse discharge at the end of a
strong, long-lasting anodal pulse stops almost immediately upon termina-