The Integrative Action of the Nervous System

remained the same in the two observations (Figs. 30A and 30B). The observation of Fig. 30B was made from the same preparation as Fig. 30A and about four minutes later, The moment of delivery of the individual stimuli is marked by the abscissae on the myo- gram : in Fig. 30A six were delivered before the reflex contraction set in ; similarly in Fig. 30B, six were delivered before the reflex relaxation set in. The mtensity of the stimulat- ing shocks was feeble, hence the relatively long latent period. Time recorded in hun- dredth seconds above, in seconds below.

Figure 31. — A and B. The "flexion-reflex" observed as reflex contraction (excitation) of the flexor muscle of the knee (Fig. 31A) and as reflex relaxation (inhibition) of the ex- tensor muscle of the knee (Fig. 31B). Conditions the same as in Fig. 30, except that the secondary coil of the inductorium is nearer to primary, and therefore stimulation more intense. The latency is therefore shorter than in the pair of observations yielding Fig. 3c. In Fig. 31 A the first three stimuli fall within the latent period; in Fig. 31B the

first two stimuli only. The reflex contraction excited is more vigorous and prolonged than with the weaker stimuli of Fig. 30. (The myograph lever at the top of its ascent has touched the carrier of the electromagnetic signal, and its further record is retarded until it begins to descend). Electromagnetic records of interruptions of constant current in primary circuit, and of time, as in Figs, 30A and 30B. ing the reflex contraction that until then had been in progress j (Fig. 32). The speed with which the reflex inhibition occurs and is accomplished is much the same as that which reflex con- traction itself exhibits. It hardly seems slower. But it is often noticeable that the relaxation thus induced in the contraction does not reduce the contraction to zero (Fig. 32). The relaxa- tion ensues down to another grade of contraction, at which grade the inhibition often continues to hold it; at least, the muscle continues to remain at that length. In such cases thcj contraction is reduced suddenly from a high level of intensity' to a lower level, but a remainder of contraction persists. It! may be that this lower grade represents another functional act! in which the muscle is simply adjuvant towards steadying thei levers for other muscles which replace itself in its previous r61ej

Figure 32. — Myograph record of reflex contraction of semimembranosus induced by stimu \ lation (unipolar faradization) of the skin of the homonymous foot. The duration of thi i stimulus is marked by the upper signal. The lower signal marks the time of applicatioi 4 of a stimulation (unipolar faradization) of the skin of the contralateral foot : this stimula \ tion caused immediate relaxation of the contracting hamstring muscle, but the relaxatioi % did not proceed beyond a certain grade. Time is marked above in fifths of seconds.

of principal actor. The condition under which I have most frequently met it has been when the exposed and freed tendon of the semitendinosus (dog) has been attached to the myograph. It is interesting that when (Fig. 33) the inhibiting stimulus is strong the relaxation of the extensor muscles is actually to a point beyond their initial length obtaining at the time the " crossed extension-reflex " began. The pre-existent " decere- brate" tonus is inhibited as well as the intercurrent reflex. The relaxation is indeed down to, as I expressed it in one of my earlier notes,^^ the post mortem length of the muscle. The relaxation, if the crossed-reflex stimulus continues, is rapidly recovered from, and the interrupted reflex reasserts itself (Figs. 33 and 34).

Concordantly with these results examination with the myo- graph of the contractions of the pretibial and post-tibial muscles of the frog ^^^ during alternate flexor and extensor strokes of the hind limb shows in many cases, though not in all, that the con- tractions of the two antagonistic muscles are not synchronous, but are conversely timed. The contraction of the pretibial muscle breaks down just as that of the post-tibial ensues, and the post-tibial relaxes just as the pretibial contracts.

An early noted and in various ways typical example of this r61e of reflex " inhibition " was that discovered by E. Hering*^ and J. Breuer** (1868) in the " self-regulating " respiratory vagus action. Distension of the lung by exciting afferent fibres in the pulmonary vagus inhibits inspiration and excites expiration. I reverted to this as a fundamental instance in my first note ^^ on the subject. If we regard the heart and ring-musculature of the arteries as two antagonistic muscles, v. Cyon's *^ still earlier dis- covery (1866) that the afl*erent nerve of the heart, and aorta (A. Tschermak and Koster)^^^ — from this point of view, a tendon of the heart muscle — evokes reflex inhibition of the arterial ring musculature, is another instance. The sugges- tiveness of these facts for the co-ordination of skeletal muscles was not recognised generally. But Meltzer, the discoverer with Kronecker^^ of the r6le of inhibition in normal deglutition.

wrote in 1883,^^ "Of a purposeful arrangement we could ex- pect that a nerve, the stimulation of which causes flexion, ought to contain also inhibitory fibres for the extensors. Now such an arrangement is indeed present — at least in the re- spiratory mechanism. Of the superior laryngeal nerve, of the second branch of the trigeminus, and of the splanchnics, we know that stimulation of their central end causes inhibition of the inspiratory and contraction of the expiratory muscles."

Now in the case of the skeletal muscles of the mammalian limb, no efferent nerve-fibres appear to be supplied to them which under stimulation produce inhibition of their contraction. Such have been sought for by various observers, but without success. I have myself looked for them and obtained no un- equivocal evidence of their existence. Moreover, Verworn^^" has shown that during the inhibitory relaxation produced by the reflex induced from the nerve of the antagonistic muscles the excitability of the relaxed muscle and its motor-nerve to electrical stimuli remains undiminished.

Moreover, in the condition of decerebrate rigidity, when the elbow is being kept in extension by the heightened tonic action of the extensor muscles, their contraction can be inhibited not only by stimulation of the crossed hind foot, but by direct electrical stimulation of a point in the lateral column of the transected spinal cord in the hind thoracic region, as has been shown by A. Frohlich and myself.^^ The inhibition reflexly produced has therefore its seat in the spinal part of the reflex- arcs. It is therefore a central inhibition.

This central inhibition appears more than equivalent to merely arresting the play of an excited afferent channel upon Figure 33 (opposite). — Myograph record of reflex contraction of extensor of knee interrupted by areflex inhibition (relaxation). The reflex contraction was induced by stimulation (uni- polar faradization) of the skin of the opposite foot: this stimulation was applied during the time marked by the lower signal ; its moments of commencement and ending are marked by abscissae on the myogram. Towards the height of the reflex contraction a brief stimulation (unipolar faradization) was applied to the skin of the foot homonymous with the knee extensor yielding the myogram : the duration of this inhibiting stimulus is marked by the upper signal. The knee extensor at outset was in some tonic contraction due to " decerebrate rigidity." The reflex inhibition relaxes this in addition to inhibit- ing the current reflex from the crossed foot. Time is marked below in fifths of seconds.

Figure 34, — Similar to Fig. ^^j except that the inhibiting stimulus was weak faradization applied to the proximal end of the severed " hamstring nerve." the motor centre. Were that all, the phenomenon should re- semble the effect of suddenly stopping the stimulation of the afferent nerve causing the reflex. What happens is often not like that; the arrest is more rapid. The "after-discharge," whatever its seat, can be at once arrested by the inhibition (Fig. 35). The "after-discharge" of a centre, with its con- comitant persistence of contraction of muscles, might well be disadvantageous to the organism. That it is rapidly arrested

by the inhibitory side of a succeeding reflex, is an adaptation which facilitates the successive interchange of reflexes. The inhibition can arrest some forms of clonic spasm arising during experimentation (Fig. 36). The motor neurones of the flexor muscles of the hind limb can be excited to the clonic discharge characteristic of the scratch-reflex at a time when the flexion-reflex is inhibited from employing them. When the scratch-reflex is in progress

Figure 35.— Myrograph records of reflex contractions of the extensor of the knee in 'decere- brate " cat. The exciting stimulus was, in the observation reproduced on the left of the figure, a brief compression — lasting less than a second — of a digit of the contralateral foot. After this stimulus had been given and discontinued, and while the after-discharge of the reflex was still in progress, the proximal end of a branch of the severed hamstring nerve was stimulated by faradization for about a quarter of a second. The time of this inhibiting stimulus is marked by the sigjnal. The reflex after-discharge is seen to have been at once inhibited and in this case not to have returned.

The observation reproduced on the right was from the same experiment, but later ; in it the stimulation exciting the reflex contraction was faradization of the proximal end of a twig of the internal saphenous of the contralateral leg. This stimulation lasted about two fifths of a second or less. Its cessation was quickly succeeded by faradization of the proximal end of a branch of the severed hamstring nerve as in the previous observation. The signal marks the time of this inhibiting stimulation. The after-discharge of the con- traction reflex is cut short as before. Time is marked below in fifths of seconds.

Figure 36. — Myogram of convulsive twitching of semitendinosus in a " spinal" dog. The spasms are reduced and temporarily suspended by stimulation (faradization) of the proxi- mal end of a branch of the internal saphenous nerve of the contralateral leg. The time of application of the inhibiting stimulus is shown on the signal line below. Time is marked above in seconds. it is more difficult to excite a " flexion-reflex," and vice versa. One reflex seems to be precluded from acting on a motor neurone at a time when another and difi"erent reflex is employ- ing it.^^ The preclusion of the motor neurone from one reflex while it is still left open to it to respond to other reflexes ap- pears to be one of the services of inhibition to the organism. The motor neurone itself seems not the actual seat of the in-

hibition, for if so, it would be inhibited for all reflexes; un- less the motor neurone is functionally divisible, and one part of it, e. g., one set of dendrites, can be inhibited at a time when another is not. The seat of the inhibition appears, therefore, with some likelihood, to lie neither in the afferent neurone proper nor in the efferent neurone proper, but in an internun- cial mechanism — synapse or neurone — between them. I say " neurone proper," meaning to exclude from that term the synapse, although in a synapse the neurone terminals are included.

The striking correspondence observed {y, j.) between the reflex inhibition and the reflex contraction, when examined in one and the same type-reflex, allows the inference that the nerve-fibres from the receptive field of the reflex each divide in the spinal cord into end-branches {e.g.^ collaterals), one set of which, when the nerve-fibre is active, produces excitation, while another set, when the nerve-fibre is active, produces inhi- bition.2^> ^^ The single afferent nerve-fibre would therefore in regard to one set of its terminal branches be specifically excitor^ and in regard to another set of its central endings be specifically inhibitory. It would, in this respect, be duplex centrally (Fig. 37). There is analogy between the structural arrangement for reflex reciprocal innervation and that of Astacus claw, if it be sup- posed that the individual nerve-fibres of the crayfish-claw prepa- ration dichotomise, one division of the nerve-fibre passing to the closing muscle, the other to the opening muscle; so that one division of the fibre exerts the excitor action, the other the well-known inhibitory, studied by Richet,^ Biedermann,^^ Piotrowski,^*^ and others.

In denoting one set of central terminations of an afferent arc " specifically inhibitory^' it is here meant that by no mere change in intensity or mode of stimulation can they be brought to yield any other effect than inhibition. But the fact that stimulation of a single set of afferent arcs, namely a single small afferent nerve, excites frequently a reflex movement of alternating direc- tion in which, for instance at the knee, extension succeeds pri- mary flexion, shows that a change of internal conditions may

presumably convert an intraspinal connection that under the primary conditions is inhibitory into one that under later super- vening conditions becomes excitatory. The fact that under certain forms of cerebral action true antagonistic muscles can be thrown synchronously into contraction, points to the same limitation of the term " specific " in this connection. Further, there is the intraspinal action of strychnine. There is the long recognized fact that under strychnine practically all the skeletal muscles of the body may be reflexly thrown into contraction simultaneously, and this is obviously inclusive of, and was proved for, antagonistic muscles.^^ Evi- dently strychnine in some way must alter or obscure reciprocal innervation. I have furnished (1892) tracings showing that the pretibial and post-tibial muscles of the frog, although in normal reflex movements so frequently exhibiting concurrent contrac- tion and relaxation in the two groups reciprocally, under strychnine reveal in the double myogram perfectly synchronous contraction in both groups.^^

Such a result may be explicable in several ways. In order to discover what the nature of the change wrought by strychnine really is, there have to be fulfilled in the test experiments cer- tain conditions which not every preparation of antagonistic muscles can supply. Muscles acting over two joints are to be avoided in such a test. Thus the gastrocnemius of the frog ex- tends the ankle but flexes the knee ; it antagonizes the action of the pretibial muscles which flex the ankle, but since flexion of the knee so commonly accompanies flexion of the ankle, it is synergic with the pretibial muscles in the great flexion-reflex that draws up the limb. If it acts synchronously with pretibial muscles under strychnine, we are still left in a dilemma as to whether the co-ordinate reciprocal action at the ankle is essen- tially destroyed, or whether a reflex attempt to flex the knee has not been simply added to it under a lowering of the intra- spinal resistances. And this dilemma is the greater in that the aflerent nerves and surfaces used for exciting reflexes contain admixed afferent channels, some exciting contraction in one group of the opposed muscles, and some exciting contraction in

the other. Thus in the afferent nerves from the foot, both in the dog^ and the frog,^^ there are commingled with fibres which excite the flexor muscles those which excite the extensor muscles — witness the extensor-thrust and flexion-reflex, both elicitable from the dog* s foot, and Baglioni's extension-reflex of the leg and the flexion-reflex, both elicitable from the frog's leg. That the extensor muscles of the Hmb should under strychnine be thrown into contraction synchronously with the flexors in these cases might be due either to the two reflexes being elicited together when spinal resistance has been lowered, or to a con- version of the inhibition part of one reflex into an excitation. And in this latter case it is still left undecided whether the extension under strychnine is due to prepotent extensor-reflex with its accompanying flexor inhibition changed into excitation, or whether it is the flexion-reflex which is changed conversely.

On similar grounds the " spontaneous " convulsions due to strychnine afford no deeper insight into the problem. These "spontaneous" convulsions are really reflex (Stannius, CI. Bernard, H. E. Hering, and others) in the sense that they originate in the afferent arcs ; and in the convulsive movements antagonistic muscles contract simultaneously. But the difficulty here again is, that the reflex source may, and probably does, operate in many afferent arcs concurrently. Some of these arcs excite extensor muscles normally, while others excite flexors. The simultaneous contraction of both flexors and extensors might thus be naturally explicable by lowered spinal resistance, both sets of reflexes being equally induced together, or the explanation might be of an alternative kind, such as suggested above. On the former view the reciprocal innervation of antag- onistic muscles would merely be obscured by a simultaneous double reflex; on the latter a more profound alteration would have taken place. The occurrence and form of the convulsion fail to decide among these possibilities.

Conditions for determining the nature of the action that really occurs seem offered, however, in certain instances. Thus, in the hind limb of the cat we have two afferent nerves which never, under any normal conditions ^^ in my experience, yield as their primary reflex in the vasti-crureus muscle any action but relaxa- tion; in other words, they, without exception, produce reflex inhibition of that muscle. To suppose that these nerves contain afferent fibres which evoke reflexly at the knee any action other

Figure 37. — Diagram indicating connections and actions of two afferent spinal root-cells, 9. and a', in regard to their reflex influence on the extensor and flexor muscles of the two knees, a, root-cell afferent from skin below knee ; a', root-cell afferent from flexor muscle of knee, /. *., in hamstring nerve ; « and «', efferent neurones to the extensor muscles of the knee, left and right ; 5 and 6', efferent neurones to the flexor muscles ; E and E', extensor muscles; F and F', flexor muscles. The "schalt-zellen" (v. Monakow) probable between the afferent and efferent root-cells are for simplicity omitted. The sign + indicates that at the synapse which it marks the afferent fibre a (and a') excites the motor neurone to discharging activity, whereas the sign — indicates that at the synapse which it marks the afferent fibre a (and a') inhibits the discharging activity of the motor neurones. The effect of strychnine and of tetanus toxin is to convert the minus sign into plus sign.

than flexion would be mere hypothesis. These two nerves are the internal saphenous in its course below the knee, and the hamstring nerve, coming from the flexor muscles of the knee. Further, the vasti-criireus is a single-joint muscle, and unlike the rest of the quadriceps extensor of the thigh is not a flexor of the hip ; therefore contraction in it cannot mean merely its participation in the synergy of the flexion-reflex itself, which includes flexion at the hip. A reflex preparation suited for ex- amining the action of strychnine on reciprocal innervation can, therefore, be obtained in the hind limb by severing, in the de- cerebrate animal for instance, the following nerves : the external popliteal, the internal popliteal, the obturator and pudic in the pelvis, the superior gluteal, the external and cutaneous divisions of the anterior crural and the hamstring nerve. The last named is ligated and cut close to its entrance in the muscles, so that its central end can be stimulated. A branch of the internal saphenous nerve below the knee is also prepared for stimulation of its central end. When this is done it is found that no change^ in intensity or other conditions of excitation of the afferent s nerve ever provokes anything but inhibition of the extensor of the knee, but a small dose of strychnine at once transmutes the inhibitory eff*ect into an excitation eff*ect^^ Reflex con- " traction is obtained in place of reflex relaxation. If small doses^ are carefully graded, it is possible to see a state in which the ' reflex relaxation is diminished but is not replaced by excita- tion.

This phenomenon shows well how little competent is the view of lowered spinal resistance to really explain the action of strychnine ; for at this stage the stimulated arc that normally acts on the extensor muscle by inhibition is less able to aff"ect it than before, so that on the spinal resistance view the resistance at this stage is actually heightened. A similar conversion of inhibitory effect into excitatory is produced more gradually but not less potently by tetanus toxin.304

This conversion sets in before and under smaller doses of strychnine or toxin than are required to produce the convulsive seizures characteristic of strychnine poisoning, or general tetanus. The transformation of effect by strychnine holds good not only for the nerves above mentioned but for skin-stimuli, and also for those skin points remote from the hind limb itself, which pro- voke reflex inhibition of the test muscle. For instance, in the case of the knee-extensor as test muscle, the skin of the fore paws.

The conversion of inhibitory effect into excitation effect by strychnine is more easily obtained in the case of some nerves than of others. In the instances of the nerves above mentioned the conversion is least facile, i. e.y requires larger doses or longer time for development, in the case of the hamstring nerve, than in the others. The inhibitory effect belonging to that nerve is readily lessened by the strychnine, but its actual replacement by excitation effect, e.g.^ contraction of knee-extensor, not only requires larger doses of strychnine, but is even then phasic rather than continuous. When this nerve is tested by stimulation at regular short intervals during one of these phasic periods, it can be seen that, starting from the phase in which it still evokes in- hibition little, or perhaps not at all, less obviously than in the normal state, its inhibitory effect then becomes progressively less, until it is replaced by excitation effect (contraction), at first mild, later violent. This periodic phase will repeat itself many times.

The conversion of inhibition effect as thus tested on the knee-extensor might be attributable to the afferent nerves stimu- lated containing two kinds of afferent fibres admixed, one kind causing reflex contraction of the muscle, the other kind reflex inhibition. Strychnine might, by augmenting the action of the former or by depressing the action of the latter, change the effect of stimulation of the mixed nerve. But the latter fibres would be expected to be associated in their action with — or, as urged above, to be even the self-same fibers which evoke — con- traction of the flexor muscles. Now there is at the stage of strychnization, at which the change of inhibitory into excitatory effect occurs, no trace of any paralysis or even depression of the flexor contractions. The protagonist and the antagonist muscles are thrown together into synchronous contraction as an effect of strychnine. This and other considerations ap- pear to me to weigh against explaining the conversion of