The jumping muscle of locusts is supplied by three motor axons (Hoyle, 1955 a), one of which gives rise to large twitch contractions when stimulated with a single shock and is termed the fast (F) axon, whilst another gives rise to slow contractions only, when repetitively stimulated, and is termed the slow (S) axon (Hoyle, 1955 b). The third axon was found not to give a contraction when stimulated alone but it did cause a hyperpolarization in some of the muscle fibres Thus, by comparison with peripheral inhibitory action at crustacean neuromusclar junctions (Fatt & Katz, 1953), amphibian heart (Hutter & Trautwein, 1956) and cat spinal cord (Eccles, 1953), its function could be expected to be inhibitory; but Hoyle (1955b) found that stimulation of this axon at the same time as stimulation of either of the two excitatory ones did not lead to mechanical inhibition. Instead, the mechanical responses were enhanced, though the effect was slight so functional significance could not be ascribed to it with confidence. Since this phenomenon could not be fully understood it has remained in the background awaiting further investigation.

In the meantime, some pharmacological aspects of the third axon function have been studied in Schistocerca gregaria by Usherwood & Grundfest (1964, 1965). In conflict with the earlier finding, these authors found the axon partially to inhibit the contraction caused by the S, though not the F, axon. The present work was undertaken to repeat the experiments on the third axon and to extend them to include other species and if possible other muscles. During the course of an investigation on learning of leg-raising, a good preparation for the study of the mode of action of an axon having similar properties to third axon of the jumping muscle was found (Hoyle, 1965 b), and this has been extensively used in the present work.

This muscle is the anterior coxal adductor (a.c.a.), and it receives but a single excitatory axon. The results show that the conflicting data of Hoyle and of Usherwood & Grundfest can be resolved, because different preparations of the same nerve-muscle combination may show either of the two effects—reduction or enhancement of tension —and in two instances the mechanical contraction of the same preparation has at different times been facilitated or inhibited by the same axon. Anticipating this result, the axon which can function so diversely will be termed the ‘inhibitory-conditioning axon’ and will be given the abbreviation I-C. Its function has been studied both in isolated preparations and in intact and headless insects, where the objective has been to discover, by the long-term examination of the patterns of its discharges, the ways in which the axon is utilized in life, and the ways in which its discharges are associated with excitatory axon activity during spontaneous and reflexly evoked discharges.

The work has been done mainly on the metathoracic a.c.a. muscles of the African locust Schistocerca gregaria, and confirmed on the grasshopper Melanoplus differentials, the Florida lubber grasshopper Ramalea microptera and the American locust Schistocerca vaga. In most of the present work the functioning of the axon was examined in the whole insect, with a minimal amount of dissection. This will be referred to as the in situ preparation. In addition, various degrees of isolation from the body of the insect were tried, terminating in a fully isolated preparation. The fully isolated preparation offers several advantages for testing physiological and pharmacological aspects of the functioning. The possibility of electrical and mechanical interference from neigh-bouring muscles, which is always present when using the whole animal, is ruled out. Nevertheless its use alone would require interpretative, rather than direct, appraisals of functioning in the intact animal. In the present work the two approaches have been used to complement each other.

In situ preparation

To examine the functioning under minimal dissection insects were secured, back down, in soft dental wax with the right metathoracic leg twisted forwards (cf. Hoyle, 1965 b). The loose connective between the rim of the coxa and the thorax was then cut away, exposing the underlying coxal and trochanteral muscles. The a.c.a. can be seen emerging from below the posterior rotator of the coxa, immediately posterior to a large tracheal trunk (Fig. 1). The muscle is attached to the coxal rim via a short apodeme. A semicircular cut was made through the rim close to the apodeme to free the muscle. The apodeme was then clamped by the tips of a micro-forceps mechano-electronic force transducer (Hoyle & Smyth, 1963).

Fig. 1.

Drawing of coxal region of right metathoracic leg of Schistocerca gregaria after minimum dissection, clamping of cut apodeme of anterior coxal adductor by force transducer and placement of intracellular recording electrode.

Fig. 1.

Drawing of coxal region of right metathoracic leg of Schistocerca gregaria after minimum dissection, clamping of cut apodeme of anterior coxal adductor by force transducer and placement of intracellular recording electrode.

Dental wax was shaped around the exposed coxa to make a cup which was filled with locust saline (Hoyle, 1953) replenished at frequent intervals. A ground lead was placed in the saline. A glass capillary microelectrode was next micro-manipulated into position and inserted into an a.c.a. muscle fibre. The lead was direct-coupled, and together with that of the force transducer, monitored on a Tektronix 502 cathode-ray oscilloscope and displayed continually on an Offner Dynograph pen oscillograph.

The isolated preparation was described in a previous paper (Hoyle, 1966). During early stages of the investigation attempts were made partially to isolate the preparation, retaining only connexions with the tracheal system intact. Eventually it was realized that the preparation survives sufficiently well in complete isolation.

Tests on fully and semi-isolated preparations

The excitatory axon was usually stimulated by applying leads to the nerve and raising the stimulus strength to just above threshold. In some cases it was found convenient to excite it by preganglionic stimulation applied to a nerve stump. The I-C axon was routinely stimulated by preganglionic stimulation whenever an effective site, usually the left connective with the mesothoric ganglion or sometimes part of the ganglion itself, could be found. Otherwise, the only way I-C action could be studied was by further raising the strength of the stimulus applied to nerve 3 c, thereby exciting the excitor and inhibitor axons together. This method is disadvantageous compared with preganglionic stimulation because it precludes the varying of the interval between excitatory and inhibitory impulses, so that the relative times of arrival at the junctions cannot be varied. The interval may be critical in relation to the inhibitory aspect of functioning.

In all, 72 preparations of either semi-or fully isolated kinds were studied. Of these 27 showed mechanical inhibition of excitatory axon contraction to greater or lesser degrees (Usherwood & Grundfest phenomenon) ; 24 showed a slight enhancement of the excitatory response (Hoyle phenomenon) ; 2 preparations showed a slight inhibitory effect followed by a slight enhancement during continued interaction of the two axons; 19 preparations did not show any significant mechanical influences of I-C upon E contractions, although they did show electrical effects. It cannot be too strongly emphasized that these results related to the use of standard locust saline. Quite different results may be obtained in different media. For example, higher levels of calcium ions increase the size of the excitatory junctional potentials and lead to greater twitch contraction. Standard locust saline of Hoyle (1953) was not intended to be an optimal one as far as mechanical responses are concerned, but to give results similar to average results obtained with the muscles bathed in their own haemolymph. For descriptive purposes, the two different positive effects of I-C will be considered independently. First, inhibitory effects will be dealt with, then facilitating ones.

Mechanical inhibitory effects of I-C

(i) Action on tetanus

The preparations which showed a reduction in tetanus upon I-C axon stimulation behaved in a similar manner to those described by Usherwood & Grundfest (1965). Examples of the interaction are given in Figs. 2 and 3. The extent of mechanical inhibition which was possible at high ratios of inhibitory to excitatory stimulation ranged from barely perceptible to 80 % reduction. When the inhibitory axon was suddenly excited during tetanus (minimum frequency for smooth fusion), three principal effects were seen. One was a relatively quick fall (complete in 1 sec.) in tension to a plateau (Fig. 2b), which was maintained throughout inhibition. This effect reduced tetanus tension by not more than 20%. On cessation of stimulation tension returned to the tetanus level. A second kind was a very slow progressive fall in tension reaching a plateau after as long as 20 sec. (Fig. 2 e); in extreme cases tension fell to about 50% its initial value. The third was a relatively quick fall, followed by a slow recovery of tension, in spite of continued inhibitory stimulation (Fig. 2f), which in some cases was complete; i.e. the inhibitory effect waned progressively until full tetanic tension was restored.

Fig. 2.

(a-f). Legend on opposite page.

Fig. 2.

(a-f). Legend on opposite page.

Fig. 2.

Fig. 2. Responses of various preparations to stimulation of the excitatory and inhibitory-conditioning axons. Upper traces in each give intracellular recordings from average muscle fibres ; second trace gives tension ; third trace, or third and fourth traces, stimulus marks ; fifth trace, time in seconds. Calibration: 10 mV ; 2-5 g. (a), 0·5 g. (b—j).

Fig. 2.

Fig. 2. Responses of various preparations to stimulation of the excitatory and inhibitory-conditioning axons. Upper traces in each give intracellular recordings from average muscle fibres ; second trace gives tension ; third trace, or third and fourth traces, stimulus marks ; fifth trace, time in seconds. Calibration: 10 mV ; 2-5 g. (a), 0·5 g. (b—j).

Fig. 3.

Influence of I-C on rate of rise of tetanus and on twitch height at optimal phasing for inhibition, Schistocerca gregaria, (a) 10/sec.; first response E alone; second E and I. Upper trace, intracellular, second trace, tension; third trace, time in seconds. (b) 20/sec. (smooth tetanus); details same as (a), (c) I-C axon was stimulated at 10/sec., E at i/sec. First six responses at sub-optimal phasing, last three at optimal phasing: note reduction of both e.j.p. and twitch height. I-C j p.s depolarizing in this fibre. Upper trace, intracellular ; second trace, tension; third trace, I-C stimulus; fourth trace, E stimulus; fifth trace, time in seconds. Calibration: 10 mV ; 1·0 g. in (a) and (b) 0·25 g. in (c).

Fig. 3.

Influence of I-C on rate of rise of tetanus and on twitch height at optimal phasing for inhibition, Schistocerca gregaria, (a) 10/sec.; first response E alone; second E and I. Upper trace, intracellular, second trace, tension; third trace, time in seconds. (b) 20/sec. (smooth tetanus); details same as (a), (c) I-C axon was stimulated at 10/sec., E at i/sec. First six responses at sub-optimal phasing, last three at optimal phasing: note reduction of both e.j.p. and twitch height. I-C j p.s depolarizing in this fibre. Upper trace, intracellular ; second trace, tension; third trace, I-C stimulus; fourth trace, E stimulus; fifth trace, time in seconds. Calibration: 10 mV ; 1·0 g. in (a) and (b) 0·25 g. in (c).

Stimulation of the I-C axon when it is acting as an inhibitor, at the same time as the excitor, leads to a slower rise in tetanic tension (Fig. 3 a, b). Stimulation of the axon at the end of a tetanus, during relaxation, gives rise to a quicker decay of tension (Fig. 21).

(ii) Action on the twitch

In a few preparations the twitch height was markedly reduced by a single I-C impulse. In others it was not reduced even during stimulation of I-C at high frequencies. Some preparations showed an inhibitory effect at high, though not at low, frequencies of I-C. This effect could be explained on the basis of a certain amount of facilitation of the inhibitory junctional potentials in these preparations. However, facilitation of I-C junctional potentials was not common and in many muscle fibres the first I-C junctional potential was the largest, with a decline during repetition. This was the case in some preparations which showed partial mechanical inhibition of tetanus. In these cases the polarizing action of I-C becomes effective only when the excitatory depolarizations have summated beyond a critical point.

(iii) Influence of phasing

Whenever the I-C axon could be excited independently the influence of varying the interval between I-C and E upon tension was examined. This alters the relative times of arrival, or phasing, of the two impulses at the neuromuscular junctions. There were many examples in which there was no influence of phasing, the mechanical inhibitory effect, if any, remaining constant (Fig. 4). Nor could any preparations which did not show mechanical inhibition with coincident stimulation come to show it at some optimal phasing.

Fig. 4.

Lack of influence of phasing on both junctional potential and twitch height of whole muscle. This result was not uncommon. S. gregaria preparation. Upper trace, intracellular; second trace, tension; third trace, I-C stimulus; fourth trace, E stimulus; fifth trace, time in seconds.

Fig. 4.

Lack of influence of phasing on both junctional potential and twitch height of whole muscle. This result was not uncommon. S. gregaria preparation. Upper trace, intracellular; second trace, tension; third trace, I-C stimulus; fourth trace, E stimulus; fifth trace, time in seconds.

Some preparations did, however, show variation in extent of attenuation of the twitch height in association with altered phasing. The same preparations did not, however, show any effect of phasing at tetanic frequencies.

(iv) Relaxing action of I-C acting alone

It has already been noted that activity in the I-C axon can cause an increased rate of relaxation of the tetanus. In some preparations, both isolated and relatively intact, in which E was silent, activity in the I-C axon caused relaxation of passive tension (at mean body length) or tension obtained by stretch (Hoyle, 1966). A similar effect was sometimes seen in the present work (Fig. 5 a), but so also was contraction (Fig. 5 d). The majority of muscles showed no mechanical effect of I-C acting alone.

Fig. 5.

a, b. For legend see opposite page.

Fig. 5.

a, b. For legend see opposite page.

Fig. 5.

Fig. 5. Variety of mechanical responses and interactions between E and I-C. Four different S. gregaria preparations, (a) I-C was stimulated first and gave rise to a slight contraction (starts at arrow). E j.p.s markedly attenuated at optimal phasing but mechanical response was not affected significantly. Upper trace, intracellular; second trace, tension; third trace, I-C stimulus; fourth trace, E stimulus; fifth trace, time in seconds, (b) Mechanical response markedly reduced by I-C action. Other details as in (a), (c) Giant-sized I-C j.p.s and marked attenuation of e.j.p.s (70 % in this fibre). Partial relaxation of summated tension is caused in whole muscle. Upper trace, intracellular; second trace, tension; third trace, time in seconds. Calibration; 10 mV. ; 0·5 g. (d) This preparation developed a weak contracture after prolonged stimulation at 0·9 /sec. Brief twitches were superimposed on the contracture. The contracture was markedly reduced by I-C stimulation at 15/sec. and the twitch height was slightly enhanced. Other details as in (c).

Fig. 5.

Fig. 5. Variety of mechanical responses and interactions between E and I-C. Four different S. gregaria preparations, (a) I-C was stimulated first and gave rise to a slight contraction (starts at arrow). E j.p.s markedly attenuated at optimal phasing but mechanical response was not affected significantly. Upper trace, intracellular; second trace, tension; third trace, I-C stimulus; fourth trace, E stimulus; fifth trace, time in seconds, (b) Mechanical response markedly reduced by I-C action. Other details as in (a), (c) Giant-sized I-C j.p.s and marked attenuation of e.j.p.s (70 % in this fibre). Partial relaxation of summated tension is caused in whole muscle. Upper trace, intracellular; second trace, tension; third trace, time in seconds. Calibration; 10 mV. ; 0·5 g. (d) This preparation developed a weak contracture after prolonged stimulation at 0·9 /sec. Brief twitches were superimposed on the contracture. The contracture was markedly reduced by I-C stimulation at 15/sec. and the twitch height was slightly enhanced. Other details as in (c).

Electrical effects of I-C

The inhibitory junctional potentials ranged from markedly polarizing to markedly depolarizing (Fig. 4). Some muscle fibres showed attenuation of excitatory junctional potentials (e.j.p.s) at all phasings (Fig. 5 a). Others showed little or no effect on the height of individual e.j.p.s (Fig. 2b, c) but in many such fibres a general polarization of the membrane was nevertheless apparent. This may be attributed to an undershoot of the repolarizing phase of the e.j.p.s during I-C j.p. action (Fig. 4a, b). Some showed the general polarization accompanied by enhancement of the e.j.p. (Fig. 2d). An enhancement of e.j.p.s is to be expected on biophysical grounds only if the inhibitory junctional potential (i.j.p.) causes a polarizing potential change without greatly in-creasing the membrane conductance. But the observed heights of e.j.p.s in these fibres are even greater than would be expected on the basis of purely electrical hyperpolarization (Fig. 6). In other fibres in the same preparation the e.j.p.s are markedly reduced by i.j.p. action. Very large attenuations in Crustacea are considered to be caused by presynaptic inhibition (cf. Dudel & Kuffler, 1961, for crayfish). Enhancements (Fig. 6) may require explanation by the converse process—presynaptic facilitation.

Fig. 6.

Influence of I-C junctional potentials on height of e.j.p.s seen in some muscle fibres of Schútocerca gregaria. (a) Intracellular records during combined spontaneous and evoked activity. (b) Graph of e.j.p. height plotted against membrane potential from which e.j.p. started. Insert diagram shows effect of membrane potential increase caused by I-C j.p.s on height of e.j.p.

Fig. 6.

Influence of I-C junctional potentials on height of e.j.p.s seen in some muscle fibres of Schútocerca gregaria. (a) Intracellular records during combined spontaneous and evoked activity. (b) Graph of e.j.p. height plotted against membrane potential from which e.j.p. started. Insert diagram shows effect of membrane potential increase caused by I-C j.p.s on height of e.j.p.

Mechanical facilitation

Enhancement of contraction can be obtained by stimulating the I-C axon at the same time as the E axon in about one-third of all preparations. It is seen as a progressive increase in height of the twitch during repetitive stimulation of I-C (Fig. 7). The extent of the enhancement is related to the frequency of firing of I-C, but only within narrow limits and not in a clear manner. The twitch height, in these examples, increases with increasing frequency only up to about 10 I-C/sec. and further increase in frequency has no additional effect; or instead a decrease, accompanied by relaxation of summated tension, may occur, as the inhibitory effect of I-C takes over following extensive depolarization of the fibres.

Fig. 7.

Facilitating action of I-C on twitch response. E axon was stimulated at 1/sec. I-C axon fired spontaneously at progressively increasing frequency; note that twitch force doubles. Upper trace, intracellular; middle trace, tension; lower trace, time in seconds. Calibration: 10 mV. ; 0·5 g.

Fig. 7.

Facilitating action of I-C on twitch response. E axon was stimulated at 1/sec. I-C axon fired spontaneously at progressively increasing frequency; note that twitch force doubles. Upper trace, intracellular; middle trace, tension; lower trace, time in seconds. Calibration: 10 mV. ; 0·5 g.

Electrical events accompanying facilitation of tension development

Tension development by a muscle is considered to be directly linked to the membrane potential and starts when the membrane is depolarized below a threshold (Hodgkin & Horowicz, 1960; Orkand, 1962). Hence contractions could not be caused by the polarizing actions of the I-C axon. Muscle fibres which may be responsible for the mechanical facilitation observed have been located by micro-electrode probing in various parts of the muscle. They appear to be mixed randomly with fibres showing only polarizing responses, but themselves give depolarizing responses (Figs. 8, 9). If these depolarizing I-C j.p.s are responsible for tension enhancement, and if they are sufficiently large, they should give rise to small contractions ; and such contractions have been observed. Contractions caused by I-C are of the slow, tonic kind, never even minutely twitch-like. The occurrence of such slow contractions only suggests the presence of specialized muscle fibres having slow contractile properties.

Fig. 8.

Slow electrical after-responses occurring in certain muscle fibres only (Schiitocerca gregaria) I-C j.p.s were evoked throughout the record at i/sec. In this fibre they were small and depolarizing. A single E j.p. was elicited. After completion of the response, in this case with a small undershoot, the slow response starts, reaching a peak at the point shown by the arrow. That the change is not an artifact is shown by the reversal of the I-C j.p. which occurs at the same time. Full recovery takes 9sec. Upper trace, intracellular record; second trace, tension; third trace, I-C stimuli; fourth trace, E stimulus; fifth trace, time in seconds. Cabbration: 10 mV. ; 0·5 g.

Fig. 8.

Slow electrical after-responses occurring in certain muscle fibres only (Schiitocerca gregaria) I-C j.p.s were evoked throughout the record at i/sec. In this fibre they were small and depolarizing. A single E j.p. was elicited. After completion of the response, in this case with a small undershoot, the slow response starts, reaching a peak at the point shown by the arrow. That the change is not an artifact is shown by the reversal of the I-C j.p. which occurs at the same time. Full recovery takes 9sec. Upper trace, intracellular record; second trace, tension; third trace, I-C stimuli; fourth trace, E stimulus; fifth trace, time in seconds. Cabbration: 10 mV. ; 0·5 g.

Fig. 9.

Summation of slow after-responses in three examples of specialized muscle fibres, (a) Fibre giving large polarizing I-C j.p.s. E axon stimulated at 1/sec; I-C axon firing spontaneously at 4−5/sec. Note progressive rise in twitch height. Upper trace, intracellular; second trace, tension; third trace, E axon stimuli; fourth trace, time in seconds, (b) Fibre giving large depolarizing I-C j.p.s. Both E and I-C axons stimulated at 11 sec. Note slow facilitation of I-C j.p.s. Upper trace, intracellular ; second trace, tension; third trace, I-C stimulus; fourth trace, E stimulus; fifth trace, time in seconds. Calibration: io mV. ; 0·5 g. (c) Another preparation and a fibre giving small depolarizing I-C j.p.s. Both axons stimulated at 1 /sec. Note reversal of I-C j.p.s at 50 mV. membrane potential level. Other details as in (b).

Fig. 9.

Summation of slow after-responses in three examples of specialized muscle fibres, (a) Fibre giving large polarizing I-C j.p.s. E axon stimulated at 1/sec; I-C axon firing spontaneously at 4−5/sec. Note progressive rise in twitch height. Upper trace, intracellular; second trace, tension; third trace, E axon stimuli; fourth trace, time in seconds, (b) Fibre giving large depolarizing I-C j.p.s. Both E and I-C axons stimulated at 11 sec. Note slow facilitation of I-C j.p.s. Upper trace, intracellular ; second trace, tension; third trace, I-C stimulus; fourth trace, E stimulus; fifth trace, time in seconds. Calibration: io mV. ; 0·5 g. (c) Another preparation and a fibre giving small depolarizing I-C j.p.s. Both axons stimulated at 1 /sec. Note reversal of I-C j.p.s at 50 mV. membrane potential level. Other details as in (b).

Specialized muscle fibres

Muscle fibres having unusual electrical properties have been observed in several preparations. In these fibres, starting immediately after the termination of the e.j.p., a slow wave of depolarization occurs, rising to a peak in 1−5 sec. and decaying in about 10 sec. (Fig. 8). These slow waves summate during repetitive stimulation to reach plateaux of as much as 40 mV. (Figs. 9, 10). Recovery from such a peak may take as long as 3 min. This slow depolarization is not an artifact associated with the contraction of the whole muscle, for its time-constant is much longer than that of relaxation of the whole muscle.

Fig. 10.

a, b. For legend see opposite page.

Fig. 10.

a, b. For legend see opposite page.

Fig. 10.

Fig. 10. Summation of slow after-responses at higher stimulus frequencies (Schitocerca gregaria), (a) I-C stimulated at 1/sec., E at 10/sec. (b) Another fibre, same preparation as (a). E stimulated at 3/sec. (c) Another preparation: small polarizing I-C j.p.s. Note that they are not increased during the summated wave but instead decrease, suggesting that I-C equilibrium potential has fallen. Other details as in Figure 9b.

Fig. 10.

Fig. 10. Summation of slow after-responses at higher stimulus frequencies (Schitocerca gregaria), (a) I-C stimulated at 1/sec., E at 10/sec. (b) Another fibre, same preparation as (a). E stimulated at 3/sec. (c) Another preparation: small polarizing I-C j.p.s. Note that they are not increased during the summated wave but instead decrease, suggesting that I-C equilibrium potential has fallen. Other details as in Figure 9b.

I-C junctional potentials of both polarizing and depolarizing kinds occur in these fibres. When evoked at the same time as e.j.p.s they simply summate with the accumulating slow waves. Even quite large polarizing potentials fail to counteract the slow, late depolarization wave (Fig. 9 a). No explanation can be proposed at the present for this strange depolarization.

During repetitive stimulation of the excitatory axon at frequencies as low as 2−3/sec. the slow waves will summate to a level which exceeds the contraction-coupling threshold. Thus a very low frequency of discharge could give rise to a maintained, tonic contraction suitable for postural adjustment. It may be that these muscle fibres represent a specialized set concerned with muscle tone.

Reversal potential for effects of I-C axon

It is of interest to know the equilibrium potential for the ionic conductance changes associated with I-C axon activity. At this potential they should be absent, and above or below it they should be of opposite signs. Tests of the membrane potential at which reversal of the sign of the junctional potential occurs are usually carried out by altering the membrane potential artificially through a second intracellular electrode.

Unfortunately, the more interesting depolarizing effects in this preparation occur in deeper fibres and it has so far proved difficult to study them satisfactorily by the double-electrode technique. However, the late, slow depolarizing wave acts as an effective means of altering the membrane potential. By stimulating the I-C axon during the slow recovery the reversal potential may be determined. At peak depolarization the i.j.p.s are usually polarizing potentials. They progressively decline in height as the membrane potential rises, until they disappear. As the recovery continues and the membrane potential rises further, they become depolarizing and gradually increase in size (Fig. 10). The point at which they vanish is the reversal potential and is presumably equal to the equilibrium potential for the conductance increase caused by the I-C axon. According to Usherwood & Grundfest (1965) this increase is a selective one to chloride ions.

Assuming this to be the case, the chloride equilibrium potential of fibres in the a.c.a. ranges from below 35 mV. (strong depolarizer) to more than 75 mV. (strong polarizer). The very low values are all associated with fibres which give the slow after-depolarization, but values for the other kinds of fibre range from below 50 mV. to 75 mV.

Pre-synaptic effects in fibres showing depolarizing I-C potentials

In these fibres there is usually very little effect of varying the timing of excitation of the two axons. In some, however, there is a slight attenuation. This is probably due to the equilibrium potential for the I-C junctional potential being below the equilibrium potential for the e.j.p.

Preliminary structural evidence for two kinds of muscle fibre

The muscle is being examined electron-microscopically and has already been found to have many interesting features which will be reported in detail elsewhere. The various fibres have similar, long (6μ) sarcomere lengths and have a similar basic ultrastructure, but a few have a distinctive form of sarcoplasmic reticulum. Most fibres have a two/three-layered reticulum with small cistemae, whilst the others have a single layer of very large cistemae. It is considered possible that they are associated with the slow, late depolarization phenomenon and have much slower contraction and relaxation rates. Marked differences in contractile properties of the fibres comprising a single muscle, even one receiving a single excitatory axon, are now well-known for crustacean muscles (Dorai-Raj, 1965; Atwood & Dorai-Raj, 1964; Atwood, Hoyle & Smyth, 1965).

Inhibitory and conditioning effects in the life of the insect

The results described above do not leave one with even the possibility of making a definite proposal regarding the natural function of the I-C axon in the intact insect. Its effects depend so markedly on the state of the muscle at the time of the experiment and on the proportions of the functionally opposed inhibitory and facilitatory effects in different fibres. It is, of course, possible that the state of affairs in the intact animal is quite different. The muscle is bathed in haemolymph which contains, in addition to mineral ions, organic materials including neurohumours which might affect the neuromuscular properties. Therefore there exists the possibility that in the intact organism the function might be clear-cut and either wholly inhibitory or wholly facilita-tory, even though experimental preparations have given confusing and conflicting results.

The mode of action of the axon will depend on the ways in which it is utilized in natural reflex actions. For example if the I-C axon is active at the end of E axon bursts it might be aiding relaxation. If it is active during contraction of antagonistic muscles it is probably a peripheral inhibitor, and so on. If the normal actions in the intact animal can be determined it might be relatively easy to resolve the problem. Recordings of I-C action in relation to E activity made during spontaneous activity of the isolated preparation of ganglion-nerve-muscle (Hoyle, 1966) did not give any clues as to its role.

In crustaceans, where the homologous axon regularly causes mechanical inhibition In preparations, it has not been possible, with the possible exception of the opener of the claw of crayfish (Wilson & Davis, 1965), to show that the axon is actually an inhibitory one.

The problem was examined in the present work in the insects Schistocerca gregaria and Melanoplus differentialis by recording from the metathoric a.c.a. muscles in situ, with haemolymph bathing the muscle, and the minimum of dissection—just sufficient to expose the muscle. These experiments have been carried out both with and without preparation of the muscle for tension recording, and both with and without the head. Preparations have been watched continually for several hours at a time, and several thousand active bursts have been examined. Removing the head has surprisingly little effect on the detailed patterns of activity occurring in the muscle. The spontaneously occurring activity is full of variety and, although certain kinds of activity recur in a roughly similar way it may be stated that no exactly similar pattern of E and I-C impulse bursts ever occurs twice.

Spontaneous activity

During the long periods of fairly regular discharge of the E axon, the I-C axon seldom fires, though sometimes it fires also for long periods, its frequency changing approximately in parallel with changes in the E. From time to time, especially in intact insects, less frequently in beheaded preparations, the continuous background discharge is replaced by periods of silence alternating with bursts of firing. These are probably associated with attempts to turn right-way-up, and to escape or kick away the wax, etc. It is during these periods that one may expect to see the I-C axon put to good use. What invariably happens is that the I-C axon fires during the rare silent periods of the E axon, often in little bursts (Fig. 11). Return to firing of the E axon is almost invariably preceded by a burst of firing of I-C which continues into the E discharge period for a short time, then stops (Fig. 12).

Fig. 11.

Spontaneous burst activity of I-C axon occurring in almost intact preparation during silent period of E axon (Melanoplus differentialis). Note slight falls in resting tension during high-frequency firing of I-C alone. Upper trace, intracellular; second trace, tension; third trace, time in seconds. Calibration: 10 mV. ; 0·1 g.

Fig. 11.

Spontaneous burst activity of I-C axon occurring in almost intact preparation during silent period of E axon (Melanoplus differentialis). Note slight falls in resting tension during high-frequency firing of I-C alone. Upper trace, intracellular; second trace, tension; third trace, time in seconds. Calibration: 10 mV. ; 0·1 g.

Fig. 12.

Spontaneous I-C axon activity combined with sporadic E axon activity. Three preparations; Melanoplus differentials muscle bathed in haemolymph. Note in all: reduction of tension with high-frequency I-C bursts occurring alone ; facilitation of twitches associated with a high frequency of I-C activity. Upper trace, intracellular; second trace, tension; third trace, time in seconds. Calibration: 10 mV. ;0·2 g.

Fig. 12.

Spontaneous I-C axon activity combined with sporadic E axon activity. Three preparations; Melanoplus differentials muscle bathed in haemolymph. Note in all: reduction of tension with high-frequency I-C bursts occurring alone ; facilitation of twitches associated with a high frequency of I-C activity. Upper trace, intracellular; second trace, tension; third trace, time in seconds. Calibration: 10 mV. ;0·2 g.

Another fairly frequent occurrence is discharge of I-C during a particularly vigorous burst of E activity.

In none of these cases has an obvious inhibition of tension been seen. The present data indicate that the phenomenon of mechanical inhibition is largely confined to preparations bathed in insect saline—a markedly different medium from haemolymph.

During late I-C action in the intact insect there is, however, increased relaxation rate. This may be from 130% to as fast as 3−4 times the relaxation rate associated with simple cessation of the E discharge. This effect is probably caused by a more rapid repolarization than occurs passively. Bursts of I-C impulses may occur during silent periods in the E discharge. In many instances the resting tension falls during these, especially when there is extensive summation of the I-C impulses (Figs. 11, 12).

Reflexly evoked activity

Tactile stimulation applied to the animal anywhere on its body calls forth some change in the output to the a.c.a. Usually, this consists of a brief burst of increased frequency in E, followed by a silent period, and perhaps a few additional bursts. Sometimes there is a simple, abrupt cessation. This is most readily elicited by applying air puffs to the anal cerci. Cereal stimulation evokes the familiar leg extension-jump reflex, and the role of the a.c.a. in this is to relax.

The existence of this simple, marked central inhibition (cf. Hoyle, 1966) shows that the E motor neuron can be fully and satisfactorily inhibited centrally, therefore peripheral inhibition is not essential for normal functioning—at least in this reflex. In many instances the central inhibition was found to be accompanied by firing of I-C, as if to accelerate relaxation and, as it were, to assist the central action. However, this action of I-C builds up to a peak immediately before E starts to fire gain, and, as is the case during spontaneous fluctuations, usually overlaps the E burst (Figs. 13, 14). The reflexly evoked activity is, unfortunately, extremely variable in spite of careful control of the stimulation—a common experience with insects.

Fig. 13.

Spontaneous and reflexly evoked combined E and I-C activity recorded from nearly intact Schistocerca gregaria preparations with muscle bathed in haemolymph. Second, fifth and sixth evoked by tactile stimulation. Upper trace, intracellular records; second trace, tension; third trace, time in seconds. Calibration: 10 mV. ; 0·5 g.

Fig. 13.

Spontaneous and reflexly evoked combined E and I-C activity recorded from nearly intact Schistocerca gregaria preparations with muscle bathed in haemolymph. Second, fifth and sixth evoked by tactile stimulation. Upper trace, intracellular records; second trace, tension; third trace, time in seconds. Calibration: 10 mV. ; 0·5 g.

Fig. 14.

Spontaneously occurring complex interactions of I-C and E in nearly intact Melanoplus differentialis preparations. In these results a quick alternation of E and I-C occurred which resulted in rapid contractions altering with rapid fluctuations in tension. Upper trace, intracellular records; second trace, tension; third trace, time in seconds. Calibration: 10 mV. ;0·2 g.

Fig. 14.

Spontaneously occurring complex interactions of I-C and E in nearly intact Melanoplus differentialis preparations. In these results a quick alternation of E and I-C occurred which resulted in rapid contractions altering with rapid fluctuations in tension. Upper trace, intracellular records; second trace, tension; third trace, time in seconds. Calibration: 10 mV. ;0·2 g.

Many questions of physiological and biophysical interest are raised by the foregoing, but our present task is to consider the light shed by these results on the probable natural function of the I-C axon.

No evidence has been obtained from the intact animal which would support the view that the I-C axon is utilized as a simple peripheral inhibitor. Instances of its firing during contraction of antagonistic muscles have been observed, but they do not occur consistently. Firing during vigorous excitor activity in what was considered to be a purposeful response was commonly encountered. The data obtained from semiisolated preparations were equivocal, since not all the a.c.a. muscles tested showed inhibition of contraction by I-C and many showed an opposite effect at low frequencies of firing of E.

One possibility which must be considered is that the I-C axon functions to aid relaxation. The natural relaxation rate is slow, possibly because of the presence in the muscle of ‘slow’ muscle fibres. A peripheral aid to their relaxation might be functionally valuable and I-C undoubtedly increases the relaxation rate. The effect is, however, quite small in many instances and it is difficult to believe that it could be of vital importance. Such marginal effects may, of course, be of significance in the more stressful conditions obtaining in the wild.

The earlier view expressed by the author (Hoyle, 1955b) was that the I-C axon functions as a ‘primer’ of the membrane, preparing it for urgent activity. This view was based on studies on the jumping muscle. In the present work, a muscle with a completely different, mainly postural, function has been examined. This muscle is infrequently called upon to exert quick contractions in normal life. It has been found that here the I-C axon is utilized primarily in the rare periods when the E axon is silent and especially just before it resumes firing. If it were to be functioning in this way in order to ensure a large contraction, then the mechanical response immediately following the I-C burst should be distinctly larger than one caused by a similar E burst which is not preceded by I-C activity. These contractions are indeed larger; a comparison of twitch heights in Fig. 13 will show that those which are preceded by I-C activity are larger than ones occurring alone. Once again, however, the effect is probably not great enough to warrant the development of a special axon.

It may be, of course, that the I-C axon is an evolutionary relic (whose primeval function was peripheral inhibition) from a time before central inhibition had been evolved by arthropods. If so, it may have acquired a new function. However, a full consideration of its functions can probably not be made without taking into account the influence on contraction—and behaviour—of fluctuating haemolymph composition (cf. Hoyle, 1965 a).

In searching for all possible roles it should be recalled that the Russian school of comparative physiology (e.g. Voskresenskaya, 1959; Onianie, 1964) has long been drawing attention to the probable existence of nerves which have a ‘conditioning’ effect on muscular contraction. A function of this kind might be considered for the locust and grasshopper I-C axons.

One finding of note is that the percentage of muscle fibres which give large electrical potential changes when I-C is stimulated is high in the anterior coxal adductor, being at least 60%. In the extensor tibiae it is much smaller, perhaps less than 30%. The bundle of muscle fibres attached to the most proximal margin of the extensor apodeme contains a much higher proportion of fibres responding to I-C than the muscle as a whole. This bundle also has a high proportion of dually excitatory-innervated muscle fibres, including some which receive the S, but not the F, axon (Cerf, Grundfest, Hoyle & McCann, 1959). The muscle fibres which have an S axon supply (not more than about 30% in the jumping muscle—Hoyle, 1955b) do all the work of postural maintenance and ordinary locomotion. They are therefore under almost constant neuromuscular barrage. Indeed, activity which continued non-stop at frequencies from 9/sec. to i3o/sec. over an 8-day period has been observed in a locust nymph with trailing recording leads implanted in its metathoracic extensor tibiae (see Hoyle, 1964, for technique).

The a.c.a. is a continuously active postural muscle discharging at rates from 8/sec. to over 100/sec. in the intact animal. Its muscle fibres are all functioning in a manner comparable to the few dually excitatory-innervated ones of the jumping muscle. Since these muscle fibres do not get an opportunity to rest they may tend to become filled with an undesirably high concentration of some ion species. It is these same muscle fibres which show large I-C axon electrical activity. Perhaps the I-C axon may be functioning to counteract or relieve ionic accumulations (or losses).

The net conclusion is that we do not yet understand the functioning of this axon properly. The data do not warrant its being referred to simply as a peripheral inhibitory axon, though by selecting the data it could have been made to appear as such. Perhaps the term ‘inhibitory-conditioning axon’ is the most satisfactory one to apply to this interesting axon in the present state of our understanding.

  1. An axon which, in addition to the single excitatory axon, innervates the anterior coxal adductor muscles of locusts and grasshoppers causes peripheral effects, some of which are comparable to those of crustacean peripheral inhibitory axons. Tetanus tension is reduced by the axon.

  2. The axon was termed the inhibitory-conditioning axon (I-C). Resting tension was reduced in a few preparations by the action of the axon, and the relaxation rate was markedly increased.

  3. In some preparations, however, the axon caused a slight contraction of the muscle, and paradoxically enhanced twitch tension.

  4. It was found by intracellular recording that in some muscle fibres the electrical response to I-C is polarizing, but in others it is depolarizing.

  5. A few (but variable number) of the muscle fibres in each preparation showed slow, summating, after-depolarizations following termination of the excitatory junctional potentials. This is a new phenomenon, and suggests that these fibres, or their neuromuscular junctions, have special properties.

  6. A study was made of the reflex and spontaneous discharges of the excitatory and I-C axons in the nearly intact animal, and in preparations, in an attempt to determine the function of I-C.

  7. No evidence was found that I-C is actually used as a peripheral inhibitor and its natural function remains enigmatical.

Atwood
,
H. L.
&
Dorai-Raj
,
B. S.
(
1964
).
Tension development and membrane responses in phasic and tonic muscle fibers of a crab
.
J. Cell. Comp. Physiol
.
64
,
55
72
.
Atwood
,
H. L.
,
Hoyle
,
G.
&
Smyth
,
T. J.
(
1965
).
Mechanical and electrical responses of single innervated crab muscle fibres
.
J. Physiol, (in the Press)
.
Cerf
,
J. A.
,
Grundfest
,
H.
,
Hoyle
,
G.
&
Mccann
,
F. V.
(
1959
).
The mechanism of dual responsiveness in muscle fibers of the grasshopper Ramalea microptera
.
J. Gen. Physiol
.
43
,
377
95
.
Dorai-Raj
,
B. S.
(
1964
).
Diversity of crab fibers innervated by a single motor axon
.
J. Cell. Comp. Physiol
.
64
,
41
54
.
Dudel
,
J.
&
Kuffler
,
S. W.
(
1961
).
Presynaptic inhibition at the crayfish neuromuscular junction
.
J. Physiol
.
155
,
543
62
.
Eccles
,
J. C.
(
1953
).
The Neurophysiological Basis of Mind
.
Oxford University Press
.
314
pp.
Fatt
,
P.
&
Katz
,
B.
(
1953
).
The effect of inhibitory nerve impulses on a crustacean muscle fibre
.
J. Physiol
,
131
,
374
89
.
Hodgkin
,
A. L.
&
Horowicz
,
P.
(
1960
).
The influence of potassium and chloride ions on the membrane potential of single muscle fibres
.
J. Physiol
.
148
,
127
60
.
Hoyle
,
G.
(
1953
).
Potassium ions and insect nerve muscle
.
J. Exp. Biol
.
30
,
121
35
.
Hoyle
,
G.
(
1955a
).
The anatomy and innervation of locust skeletal muscle
.
Proc. Roy. Soc. B
,
143
,
281
92
.
Hoyle
,
G.
(
1955b
).
Neuromuscular mechanisms of a locust skeletal muscle
.
Proc. Roy. Soc. B
,
143
,
343
67
.
Hoyle
,
G.
(
1964
).
Exploration of neuronal mechanisms underlying behaviour in insects
.
In Neural Theory and Modeling
, pp.
346
76
. Ed.
R.
Reiss
.
Stanford University Press
.
Hoyle
,
G.
(
1965a
).
Neural control of skeletal muscle
.
In Physiology of Insecta
, vol.
2
, pp.
203
232
, 407-49. Ed.
M.
Rockstein
.
New York
:
Academic Press
.
Hoyle
,
G.
(
1965b
).
Neurophysiological studies on ‘learning’ in headless insects
.
In Physiology of Insect Central Nervous System
, ed.
J. E.
Treheme
.
Academic Press
.
Hoyle
,
G.
(
1966
).
An isolated insect ganglion-nerve—muscle preparation
.
J. Exp. Biol
.
44
,
413
27
.
Hoyle
,
G.
&
Smyth
,
T.
(
1963
).
Neuromuscular physiology of giant muscle fibers of a barnacle, Balanus nubilus Darwin
.
Comp. Biochem. Physiol
.
10
,
292
314
.
Hutter
,
O. F.
&
Trautwein
,
W.
(
1956
).
Vagal and sympathetic effects on the pacemaker fibers in the sinus venosus of the heart
.
J. Gen. Physiol
.
39
,
715
33
.
Onianie
,
T. H.
(
1964
).
Comparative Physiology of Neuromuscular Apparatus. Tbilisi. (In Russian.)
Orkand
,
R. K.
(
1962
).
The relation between membrane potential and contraction in single crayfish muscle fibres
.
J. Physiol
.
161
,
143
59
.
Usherwood
,
P. N. R.
&
Grundfest
,
H.
(
1964
).
Inhibitory postsynaptic potentials in grasshopper muscle
.
Science
,
143
,
817
18
.
Usherwood
,
P. N. R.
&
Grundfest
,
H.
(
1965
).
Peripheral inhibition in skeletal muscles of insects
.
J. Neurophysiol
.
28
,
497
518
.
Voskresenskaya
,
A. K.
(
1959
).
Functional Characteristics of the Nerve-Muscular System of Insects
.
Academy of Science of the USSR (Akademia Nauk SSR
),
Moscow
. (In Russian)
Wilson
,
D. M.
&
Davis
,
J.
(
1965
).
Nerve impulse pattern and reflex control in the crayfish claw motor system
.
J. Exp. Biol
.
43
,
193
210
.