The Integrative Action of the Nervous System

Each of the intraspinal neurones in its turn forms two synaptic connections with two neurones, and in these cases also the thresholds at the synapses are of different heights, numerically, two and one respectively. On the view that the action of one neurone upon the next is that of a releasing force liberating a potential system across a barrier whose resistance we do not exactly know, it is impossible to predict how the re- sistance will sum along the whole conductive chain. It is clear that although the total resistance of the reflex-arc A B may be numerically represented by i, the resistance along A D need not, on the numerical values assigned to the synapses in the diagram (Fig, 46), sum to the value 4. Yet it is also clear that the threshold for any whole arc cannot be lower than the highest individual threshold in it. Further, the individual thresholds will tend to sum, for an excitation of neurone A just sufficient to excite neurone a is hardly likely to excite a sufficiently to overcome the threshold of synapse a D. Thus, with even small grades of difference of threshold at different synapses, large differences in the conductive facility of different reflex-arcs can be established.

Similarly, an additive influence of the threshold will make a reflex-chain consisting of several neurones offer caeteris paribus higher resistance than a chain of fewer neurones. The diagram is therefore in accord with the rule that the reflex-chains which conduct to parts segmentally distant require generally intenser stimulation to excite them than do merely local arcs. ^V Short and long reflexes. For many purposes of description ^■t is convenient to divide reflexes into " short " and " long." ^^ ^B The cord may, in its relation to the receptive surface and skeletal musculature, be considered divisible into right and left lateral halves, each subdivisible into regions of neck (cervical, ^including pinna), fore limb (brachial), trunk (thoracic), hind Hmb crural), and tail (caudal). A reflex action in which the stimu- us applied to a receptive area in one of the above regions vokes a reaction in the musculature of another of the regions s conveniently called a lofig spinal reflex. A reflex reaction in hich the muscular reply occurs in the same region as the pplication of the stimulus is conveniently called a short spinal eflex. Short spinal reflexes are, as a rule, more easily and gularly elicitable than are long spinal reflexes. It might urther be convenient to allocate hard-and-fast boundaries to hese regions, but such limits would of necessity be artificial nd arbitrary. The scope of the delimitation is indicated and ts purpose better served by comparison with one retina, say of he bird, the one lateral half of the skin corresponding with one etina ; one optic nerve would correspond with the lateral half of he spinal cord and bulb. Between these comparable surfaces a iflerence exists, in that the receptive field of the skin, unlike e retinal, has instead of one (or two, cf. Kalischer) ^^^ focal region of concentrated responsiveness, several such foci, e. g, the relatively highly responsive skin at the apex of each limb. As the retina has muscles at call, so also the skin. The close- ness of nexus between a retinal point and the visual muscula- ture is graduate in degree, e, g. most close for the muscles of its own bulbus, next for those of the contralateral, then for the neck muscles, etc.

Similarly there are degrees of nexal closeness between a point of skin and the related musculature: its connec- tion is most close with muscles of its own limb, next with those of another limb or other region. The main interest of direc- tion of nervous irradiation per se — apart from light it may incidently cast upon the integrative work of the nervous system — lies in its elucidation of the machinery for working sentient surfaces. That the skin is a region which morphologically

considered is composed of a segmental series seems to have j been allowed greater weight in the estimation of its receptive functions than is fully justified, at least in the higher vertebrata. That its segmental innervation demonstrably limits existing reflex spinal functions in the mammal has not been shown.^^^ Rules observed in the spread of impulses in spinal reflexes. Regarding short spinal reflexes, and the directions taken by the examples of intraspinal irradition which they furnish, it is possi ble to make certain general statements.^^

I. Broadly speaking, the degree of reflex spinal intimacy % between affere^tt and efferent spinal roots varies directly as their "\ segmental proximity. Thus excitation of the central side of a \ severed thoracic root, e, g, seventh, evokes with especial ease '[ contraction of muscles or parts of muscles innervated by the i corresponding motor roots, and next easily muscles innervated \ by the next adjacent motor roots. The spread of short spinal : reflexes in many instances seems to be rather easier tailward \ than headward. This may be related with the oblique correla- \ tion that so largely holds between the distribution of the afferent ^ root in the skin and the distribution of the efferent root in the \ underlying muscles.

II. Taken generally, for each afferent root there exists in \ immediate proximity to its own place of entrance in the cord \ {e. g. in its own segment) a reflex motor path of as low a thresh- \ old and of as high potency as any open to it anywhere, \ Further, in response to excitation even approximately mini- • mal in intensity a single afferent root, or a single filament of ' a single root, evokes a spinal discharge of centrifugal impulses \ through more than one efferent root, i. e, the discharge is plurisegmental. And this holds especially in the limb regions. In the limb region the nerve root is therefore a morphological aggregate of nerve-fibres, rather than a functionally determined assortment of impulse-paths. The view that the efferent spinal root is a functional assemblage of nerve-fibres is certainly erro- neous. The formation of functional collections of nerve paths (peripheral nerve-trunks) out of morphological collections (nerve roots) seems to be the meaning of the limb-plexuses.

III. Motor mechanisms for the skeletal musculature lying in the same region of the cord, and in the selfsame spinal seg- ment, exhibit markedly unequal accessibilty to the local afferent channels as judged by pressor effects. For example, if pressor jffects only, and the primary phase only, of the reflex movement be considered, the flexors of the homonymous knee and the extensors of the contralateral are in many animals much more accessible than the extensors of the homonymous and the flexors of the contralateral. Inasmuch as at many joints the flexors and extensors are both innervated by motor-fibres con- tained in one and the same efferent root, it follows that the reflex movement obtained by excitation of an afferent root in many cases is quite dissimilar from the movement obtained by excitation of the corresponding efferent root, in spite of the rule of segmental proximity.

It is necessary to insert the qualification " pressor " before "effects" ("reciprocal innervation"). It is only in regard to pressor effect that the above statement holds for such contrasted neurones as those of " extensors " and " flexors." I have stated the rule in this way because more in conformity with the oft- quoted rule of spinal reflexes coming to us from Ludwig's labor- atory, which insisted on the rarity or impossibility of obtaining hind-limb extension as a primary homonymous spinal reflex. But how easy and direct is really the reflex nexus between the receptive surface of the limb and its extensor muscles, e. g. at knee, is shown by nothing better than by giving a small dose of strychnine. That alkaloid has, as has been mentioned, the property of converting spinal reflex inhibition into excitation. The same stimulus which normally reflexly excites the knee- flexors to contraction is seen after the strychnine to excite the knee-extensors to contraction. The reflex inhibition of the ex- tensors which was previously the reflex-effect is more difficult to observe, but by turning it into excitation the facility of the reflex nexus with the extensors is found to be as close as with the flexors. Therefore, in the rule before us, if inhibition and excitation are both — as they should be — counted as evidence of the reflex nexus, then the reflex nexus with the homony-

mous knee-extensors and with the crossed knee-flexors, is as close as with the homonymous knee-flexors and the crossed knee-extensors. In the question, therefore, that was put above. How is it that the spread of a reflex reaction, when the reflex is intensified, does not extend to all parts, as it is said to do in strychnine poisoning ? there are two different things involved. It does not spread to some parts because, as argued above, an additive synaptic resistance intervenes across the potentially conductive path. An instance of such a path was given. But the absence of apparent irradiation to certain others is for a different reason. Keeping to the flexion-reflex as elicited by unipolar faradization of the plantar skin of a digit as illustration, the instance of the knee-extensor may be taken. However intensely the stimula- tion may be pushed, although the reflex reaction is thereby more and more intensified, contraction of the knee-extensor does not result, — but for a wholly different reason than that suggested for the absence of spread of the reflex from the leg to the pinna. The muscles of the pinna, in my experience, do not at all easily become involved in the reaction; but the extensor muscle of the knee is really involved in the reaction from the beginning, only it is involved in a way that escapes observation unless special means be taken to reveal it. The reflex-effect upon it takes the form of an inhibition of the efferent path just central to its motor neurone, an inhibitory block, which in pro- portion as the intensity of the exciting stimulus of the reflex is increased simply becomes itself the more intense. There is no evidence that this can be broken down and converted into excitation by merely increasing the intensity of the stimulus that is evoking it.

On the other hand, as shown above, strych- nine and tetanus toxin convert it into excitation, and that is one reason why strychnine seems to increase the spread of re- flexes so greatly; but in this case the increase of spread which that drug appears to cause is really merely apparent. The reflex- effect was there already, but had another form of expression. IV. The groups of motor nerve-cells contemporaneously dis- charged by spinal reflex action innervate synergic and not antergic

muscles. This is the reverse of the view that since Winslow and Duchenne*^ has been common doctrine concerning mus- cular co-ordination. It controverts an argument adduced for the view that the limb movement evoked by excitation of an efferent root represents a highly co-ordinate functional synergism.^* ^ The spinal reflex in its intraspinal irradiation develops a com- bined movement and synthesizes a muscular harmony. V. It follows almost as a corollary from this, and from the rule of spatial proximity (p. 158), that the spinal reflex movement elicitable in and from a7ty one spinal region will exhibit much uniformity despite considerable variety of the locus of incidence of the exciting stimulus. Approximately the same movement, e. g. in the hind-limb flexion of the three great joints, will result, whatever piece of the limb surface be irritated. The locus of incidence of the stimulus will only influence the character of the general movement executed by the limb musculature, in so far that the flexion will tend to predominantly occur at that joint the flexor muscles of which are innervated by motor cells seg- mentally near to the entrance of the afferent fibres from the particular piece of skin the seat of application of the stimulus. Another way of expressing this rule is to say that the receptive field of a ** type-reflex " is usually of plurisegmental cutaneous extent.

Part of the question of spatial distribution of the motor dis- charge of a spinal reflex has long been studied, and a funda- mental contribution to knowledge of it was made by Pfluger.^ His inductions were based chiefly upon observations on the frog and on the records of clinical cases of spinal lesion in man. They were drawn up in the form of four " laws." It was regarding the course of irradiation in long spinal reflexes, namely those spinal reflexes that initiated from 07te of the above mentioned spinal regions spread over into others that Pfluger,2* in 1853, formulated his four " laws." These "Laws" have for many years been widely accepted.^^ They are stated as follows : —

I. The law of homonymous conduction for unilateral re- flexes. If a stimulus applied to a sensory nerve provokes mus- cular movements solely on one side of the body, that move- ment occurs under all circumstances and without exception on the same side of the body as the seat of application of the stimulus. If, as is clear from the context in the original paper, by movement on the same side is meant contraction of muscles on the same side, this statement does not in reality very completely express the facts.^^ It is in part an outcome of the rule of spatial proximity, but certain cases which conform to the latter yet offer striking exception to the former ; for instance, when the skin of the tail is stimulated on one side the organ is very fre- quently moved towards the opposite, and this in a great number of classes, from fish to mammal inclusive.

2. The law of bilateral symmetry of the reflex action. When the change produced in the central organ by excitation of a sensory nerve has already evoked unilateral reflex, it, if it spreads farther, excites in the contralateral half of the cord only those motor mechanisms which are symmetrical with those already excited in the homonymous half of the cord. This statement, although true of a number of instances, fails to con- form with fact in many, even perhaps the majority.

The important cross-reflex from the hind limb of the bird and mammal does not conform to it ; so similarly with the fore limb. The asymmetry of the crossed reflexes of the Hmbs is important because probably connected with the fundamental co-ordination of muscles for progression. Again, the wag- reflex of the tail, and a reflex I have called the "torticollis reflex "^^^ (cervical region), afford important exceptions to the " law." And many other exceptions can be found. In the spinal rabbit, on the other hand, and less often in the dog, the crossed reflex from one hind limb to the other is sometimes not an asymmetrical movement, but a symmetrical one : this seems to stand in obvious relation to the hopping mode of progression of the animal.

3. The law of unequal intensity of bilateral reflexes. When the excitation of a sensory nerve elicits reflex action involving both halves of the body, and the action is unequal on the two ides, the side of stronger contractions is always that homony- mous with the seat of application of the stimulus. This statement is in conformity with a number of instances, he following are examples. When bilateral retraction of the bdomen is excited from the skin of the chest, the contralateral traction is much the less marked : in the bilateral protraction f the " whiskers " (cat, rabbit, dog) on the excitation of the in of the face, the crossed movement is the less ample. An teresting illustration,^^^* ^^ because it involves inhibitory as ell as pressor influence, can be demonstrated in the spinal cat r dog thus: — The animal resting comfortably on its back, if ne hind paw be pressed that leg will be flexed at hip, knee, d ankle, in accordance with the rules laid down on p. 158, and the stimulus be strong, or the reflex excitabihty good, the fellow hind limb will be extended. If instead of one hind paw both hind paws be pressed, both hind limbs are simultane- ously flexed, and there is no trace of extension (Fig. 66, Lect. yi. p. 225). The homonymous reflex is prepotent, therefore, d inhibits the crossed reflex. But there are also a number of exceptions to this " law," mong others, the abduction of the tail from the side stimulated ready referred to.

4. The fourth of Pfliiger's classical " laws " of spinal reflex tion states that with associated spinal reflex centres the irradi- ion spreads more easily in the direction toward than in the irection away from the head. cordant with this statement: in, I think, the majority of in- verse is rare. To elicit by excitation of the hind limb a move- ent of the fore limb, is more difficult than by excitation of the fore limb to elicit movement of the hind limb. To elicit move-

ent of the tail by excitation of the fore limb is easier than to 'move the fore limb by excitation of the tail. The irradiation Figure 47. — a. Position of animal after transection at calamus scriptorius. c. Change of attitude from b evoked by stimulation of left pinna. limb to hind limb than from hind limb to fore limb ; but it is often easier down the cord from fore limb to hind limb than across from fore limb to fore limb. In such reflexes also as the " shake " reflex (a reflex in which the trunk is shaken, as when a dog comes out of water), which implicate the trunk more than the limbs, the radiation is away from the head, for it is well obtained as a rump reflex when the skin of the shoulder is the part rubbed. In the " scratch-reflex," too, the skin stimulus is applied far headward of the region of the muscular contraction evoked.

These so-called ** laws " of reflex irradiation were so generally accepted as to obtain a doctrinal eminence which they hardly merit. It seems here less profitable to attempt adapting them to better fit the observed facts than to briefly describe the salient features of the long spinal reflexes as exhibited in an ordinary experiment on the spinal mammal. The reflex figure. When the animal is supported freely from above, with its spine horizontal and the limbs pendant, a point that early strikes the observer is that there are ten areas whence bulbo-spinal reflexes employing skeletal muscula- ture can be provoked with pre-eminent facility. These areas

b. Change of attitude from a evoked by stimulation of left fore foot. c. Change of attitude from a evoked by stimulation of left hind foot. are the soles, the palms, the pinnae, the mouth, the snout, and the tail and cloacal region. It is significant that nine of these areas are those which possess the greatest range of motility if the axis of the animal be considered fixed. Stimulation at any one of these areas causes a particular attitude — a reflex figure — to be struck. From the pinna is excited movement of each limb, the neck, thfe tail, and the trunk (Fig. 47). The irradia- tion from this reflexigenous area usually presents the following order: (i) Neck and homonymous fore limb, (2) homonymous hind limb, (3) tail and trunk on both sides, (4) contralateral hind limb, (5) contralateral fore limb. From the fore foot (Fig. 48) can be excited besides movements in the fore limb itself, movements in the other limbs and tail. The facility of radia- tion is usually in the following descending series: (i) homony- mous hind limb and the tail, (2) crossed hind limb, (3) crossed fore limb. The relative facility of spread of the reaction to the crossed fore limb seems subject to much variation. In the frog the path between the two fore limbs is, especially in the breed- ing season, very open and facile. In the cat and monkey it seems to be much more open in the bulbo-spinal than in the spinal animal. From the hind foot (Fig. 48) the frequency

and ease of irradiation into other spinal regions usually appears to exist in the following order: (i) extension of crossed hind limb and tail, (2) extension of homonymous fore limb, (3) flexion of crossed fore limb. The facility of " spread " from one lateral half of the cord to the other is very dissimilar at different levels of the cord. It is particularly easy in certain parts of the tail region. Motor mechanisms which are yoked together are for the most part, as with the flexion-extension mechanisms of the hip and knee, of an asymmetrical kind. In the hind-limb region the crossed irradiation is also fairly free, and largely connects asymmetrical muscle-groups; but one of the most facile and persistent of all bilateral reflexes resulting from uni- lateral stimulation is the adduction of both thighs, a bilaterally symmetrical movement.

In the trunk the spread across the median plane is most free for skin reflexes excited from near the midline ; this is seen in the venter of the frog ; the yoking is of bilaterally symmetrical muscles. To excite movement of one fore limb from the other is less easy than to excite one hind limb from the other, at least in many animals. In the neck region irradiation across the median sagittal plane is fairly easy, and the yoking connects in large part asymmetrical muscles.

On the whole, the " long " spinal reflexes are more variable and less validly predictable than the short. They vary in a series of experiments, not only as to order of relative facility of direction of irradiation, but as to the sense of the movement elicited at the joint, whatever it may be, to which irradiation extends. Not unfrequently a region to which the reflex usu- ally irradiates is altogether omitted, and omitted consistently throughout the whole of a lengthy experiment, although the spinal region in question has, so far as known, suffered no damage, nor indeed been directly implicated in any of the procedure. Thus, excitation of the skin of the neck or pinna will sometimes spread back along the cord and produce move- ment in the tail, or in the hind limbs, and in doing so pass by the fore limbs without evoking a twitch in either of them. The motor mechanisms of the fore limb thus skipped over may show

no sign when examined by the local reflexes of being less amen- able than usual. The inconstancy of the irradiation as to the kind of movement roduced, e. g. whether it flex or extend a Umb, is diff'erent in different reflexes. The irradiations from the " drawing-up reflex " of the hind foot {i. e, the flexion-reflex) have great constancy ; the irradiation is shown in Fig. 48 ; to the figure the turning of the head to the homonymous side may be added ; the irradia- tion from the fore-foot reflex is less regular ; sometimes flexion at crossed knee, sometimes extension. The variation is from experiment to experiment, not during a single experiment.

There is some evidence that the influence exerted on a com- mon path by one and the same afferent arc may not always be of the same kind. It is true that the regularity with which the same end-result appears and reappears in observations dealing with certain reflexes is very great, and inclines the observer to regard the reaction of the reflex-arc as perfectly constant. But that is not equally clear of all the reflexes. Instances of incon- stancy seem to occur in some reflex-arcs, and these suggest the possibility that in some cases one and the same aff"erent arc may exert on a final common path even reverse effects at different times ; in other words, under different conditions. The eff"ect of stimuli to the pinna of the " bulbo-spinal " cat seems sometimes to be flexion of the hind limb, sometimes extension of that limb. Stimulation of the aff"erent nerve of a part of the vasto- crureus muscle is often inhibition of the rest of that muscle, but sometimes not. In my experience these results, though variable from experiment to experiment, do not vary during the same experiment. Again, the afferent nerve, stimulation of which excites reflex rise of arterial pressure under curare, is known to yield reflex fall of pressure under chloral. It must be admitted that here other explanations are indeed possi- ble, besides the supposition that the kind of influence excited by the aff'erent arc on the efferent path has changed.

Irradiation of a reflex attaches itself to the problem of the simultaneous combination of reflexes. It does so because it aff"ords clear evidence that by irradiation a reflex assumes use of a number of final common paths which do not in the first instance belong to it, but belong in the first instance rather to reflexes arising in their own immediate segmental locality. From them a '* reflex figure " is formed. Thus, by irradi- ation, the flexion-reflex of the right planta causes reflex- discharge down the motor nerve of the cubital extensor (part of triceps) of the homonymous fore limb. But this reflex motor discharge to the cubital extensor is more easily excited by stimulation of the left fore-paw. Again the flexion-reflex of the right planta, if strong, will irradiate as motor discharge into the flexors of the left elbow, but the reflex motor dis- charge into the flexors of the left elbow is much more easily obtained by stimulation of the left fore paw ; so that the irra- diation welds into a single combined reflex efl"ects belonging primarily, as it were, to several reflexes. But the reflexes whose efl*ects are thus combined are always reflexes of what was termed above " aUied " relation. Thus, if a stimulus ex- citing the flexion-reflex from the right planta be just subliminal for evoking the irradiation to the homonymous cubital extensor and a stimulus be applied to the left fore paw of an intensity by itself just subliminal for provoking crossed elbow-extension, the two stimuli applied simultaneously mutually facilitate and the reflex of the right fore limb results.

Moreover, it seems to me significant that the irradiation extends rather per saltum than gradatim. As the flexion-reflex is continued, flexion at hip (Fig. 45, p. 153) can be seen to add itself almost suddenly to flexion already in progress at knee. Romanes ^2 writes of irradiation in Medusa as follows: "It is not difficult to obtain a series of lithocysts connected in such a manner that the resistance off"ered to the passage of the waves by a certain width of the junction-tissue is such as just to allow the residuum of the contraction wave which emanates from one lithocyst to reach the adjacent lithocyst, thus causing it to originate another wave, which in turn is just able to pass to the next lithocyst in the series, and so on, each lithocyst acting in turn like a reinforcing battery to the passage of the contrac- tion wave. Now this, I think, sufficiently explains the mech-

anism of ganglionic action in those cases where one or more lithocysts are prepotent over the others; that is to say, the prepotent lithocyst first originates a contraction wave which is then successively reinforced by all the other lithocysts during its passage round the swimming-bell." If we read for " pre- potent lithocyst" the "exciting external stimulus" of the arc primarily stimulated and for the other lithocysts the other arcs to which the excitement of the one primarily stimulated extends, it seems to me we have in the above description of Aurelia aurita a description that applies well to the process of reflex irradiation in the central nervous organ of vertebrates.

It may be objected that in the case of Medusa the wave of contraction is reinforced by, on reaching the lithocyst, initiating through that a new reflex which reinforces the one already in progress; whereas in the spread of the flexion-reflex to the reflex-arcs of the fore limb, the reaction does not initiate in these latter anything that can be called a new reflex because the reaction in them is not excited through their local receptors, the normal point of departure for their reflexes. That is a dif- ference certainly, and a significant one. But it does not vitiate the analogy from the point of view under consideration now. In Medusa the irradiation of the reflex is in its propagation reinforced at the certain points mentioned by the reaction in its spread exciting a new neurone, attached to its path, across a definite threshold resistance. In Medusa the threshold lies in Romanes' view at the receptor organ. In the irradiation of the flexion-reflex the reaction also breaks at certain points into new arcs across a threshold resistance, and once over the thresh- old, propagates itself along those arcs, as evidenced by the movements produced. It is in accordance with that mode of propagation that the irradiation of the reflex appears to occur per sal turn rather than gradatim (Fig. 45). Here again it is noteworthy that the places of reinforcement in the spread of the reaction which are peripheral in Medusa are central in the vertebrate; in other words, just as refractory state, inhibi- tion, interference, etc., which are peripheral in Medusa are central in the vertebrate, so with this latter instance of "rein-

forcement." The reason which seemed obvious before, appHes in the present instance also, and is the same as that which ex- plains the centrality of the central nervous system itself (Lect. IX). It is not only when the spot stimulated is a receptor that the reflex shows itself co-ordinate. The stimulation of the central end of any even large afferent nerve-trunk, or even the central end of a whole spinal afferent root, evokes reflexly a movement that is co-ordinate. This result is familiar and commonplace, but it is also remarkable.