8. THE REGULATION OF BREATHING
325
2. THE PRODUCITON OF RHYTHMIC A
m
A satisfactory theory cannot yet be produced to account for the generation of the normal breathing rhythm. A single pacemaker neuron or
neuron group cannot be responsible since breathing cannot be stopped by
destruction in any particular region. There must therefore be interaction
between a number of dispersed respiratory neurons. Since two fairly
distinct neuron populations have been described in mammals (Bums and
Salmoiraghi, 1960; Salmoiraghi and Burns, 1960a,b) and in fish (Shelton,
196l), it has been suggested that the interaction takes the form of alternating and reciprocally inhibiting discharges between the two groups
(Fig. 15). A parallel can be drawn with the two triodes in a multivibrator
circuit: when one becomes maximally conducting the other is cut off. Each
group is thought to be self reexciting so that activity increases in individual neurons and spreads throughout the group. In order to explain
how the balance alternates it must be supposed that activity is in some
way self-limiting; in the multivibrator analogy there must be some
phenomenon which corresponds to the extreme positions of the triodes
and to the discharge of the coupling condensers setting off the opposing
cycle. In some way the active group must reach saturation and its activity thereafter decline. It is possible to suggest a number of ways in
which this could happen. Salmoiraghi and von Baumgarten ( 1961)
described a small but progressive accommodation of the postsynaptic
membrane in both inspiratory and expiratory neurons of the cat. As a
burst of activity proceeded, slightly greater depolarization of a neuron
membrane was required to initiate an action potential. If all the cells
of an inspiratory or expiratory group behaved in this way there would
be a substantial fall in excitability throughout an activity cycle and
eventually it would be brought to an end. These observations, which
were made on very few cells because of the difficulty of impaling respiratory neurons with intracellular electrodes, do not exclude other ways
of terminating an inspiratory or expiratory discharge.
These suggestions are extremely speculative and raise a number
of difficulties. In the mammal there is evidence conflicting with that
given above to show that rhythmicity arises, not by an equivalent interaction of two opposing neuron groups, but by periodic inhibition of a
primary, tonic activity occurring in the inspiratory neurons. Centers other
than the expiratory and inspiratory groups of the medulla are thought
to be involved in these processes (Wang et al., 1957; Kalin and Wang,
1967; Wyss, 1954). In fish there seem to be no neuron groups other than
the two medullary ones, but one of these groups (that which fires as the
mouth and opercula are adducted) is numerically very much the
325
2. THE PRODUCITON OF RHYTHMIC A
m
A satisfactory theory cannot yet be produced to account for the generation of the normal breathing rhythm. A single pacemaker neuron or
neuron group cannot be responsible since breathing cannot be stopped by
destruction in any particular region. There must therefore be interaction
between a number of dispersed respiratory neurons. Since two fairly
distinct neuron populations have been described in mammals (Bums and
Salmoiraghi, 1960; Salmoiraghi and Burns, 1960a,b) and in fish (Shelton,
196l), it has been suggested that the interaction takes the form of alternating and reciprocally inhibiting discharges between the two groups
(Fig. 15). A parallel can be drawn with the two triodes in a multivibrator
circuit: when one becomes maximally conducting the other is cut off. Each
group is thought to be self reexciting so that activity increases in individual neurons and spreads throughout the group. In order to explain
how the balance alternates it must be supposed that activity is in some
way self-limiting; in the multivibrator analogy there must be some
phenomenon which corresponds to the extreme positions of the triodes
and to the discharge of the coupling condensers setting off the opposing
cycle. In some way the active group must reach saturation and its activity thereafter decline. It is possible to suggest a number of ways in
which this could happen. Salmoiraghi and von Baumgarten ( 1961)
described a small but progressive accommodation of the postsynaptic
membrane in both inspiratory and expiratory neurons of the cat. As a
burst of activity proceeded, slightly greater depolarization of a neuron
membrane was required to initiate an action potential. If all the cells
of an inspiratory or expiratory group behaved in this way there would
be a substantial fall in excitability throughout an activity cycle and
eventually it would be brought to an end. These observations, which
were made on very few cells because of the difficulty of impaling respiratory neurons with intracellular electrodes, do not exclude other ways
of terminating an inspiratory or expiratory discharge.
These suggestions are extremely speculative and raise a number
of difficulties. In the mammal there is evidence conflicting with that
given above to show that rhythmicity arises, not by an equivalent interaction of two opposing neuron groups, but by periodic inhibition of a
primary, tonic activity occurring in the inspiratory neurons. Centers other
than the expiratory and inspiratory groups of the medulla are thought
to be involved in these processes (Wang et al., 1957; Kalin and Wang,
1967; Wyss, 1954). In fish there seem to be no neuron groups other than
the two medullary ones, but one of these groups (that which fires as the
mouth and opercula are adducted) is numerically very much the