Protoplasmic Action and Nervous Action

Changes of temperature influence the functional processes and the bioelectric processes similarly. For example, in a special study by Keith Lucas,' in which the temperature-coefficient of the rate of development of the bioelectric variation in the frog's sartorius was compared with that of the propagation-velocity of the excitation-wave, almost identical values were found for the two processes. In a typical experiment, the time required for the rise of the bioelectric variation from zero to its maximum at 8° was .0041 of a second, and at 18°, .0024 of a second; the ratio of these two values, 1.64, was almost identical with that of the propagationvelocities of the excitation-wave (contraction-wave) at the two temperatures. In other words, change of temperature influences the rate of protoplasmic transmission in the same manner as it influences the rate of variation of potential.

It seems clear that the bioelectric variations are inseparably connected with chemical or metabolic processes in the living cells, and that the characteristic rate at which the tissue reacts and conducts excitation is a direct function of the rate of both processes. This rate is determined by the specific chemical and structural constitution of the tissue as well as by external factors, such as temperature and the state of the surrounding medium. The evidence already reviewed indicates that the essential changes underlying the bioelectric phenomena occur at the cell boundary; hence, these phenomena may be regarded as an index of chemical decompositions or other reactions occurring in the protoplasmic surface-films. Apparently these reactions

change the composition of the films and alter their electromotor and other properties (permeability, physical consistency, etc.), and the bioelectric currents are the result. The question of whether the bioelectric currents stand in the relation of cause or of effect to the other physiological activities of the cell is not one to be answered simply, since, as in so many other natural processes, the relations are of a reciprocal kind. Apparently the conditions are of the same general physicochemical nature as in any reversible type of galvanic cell (storage battery) ; a current from an outside source traversing the system may be the means of inducing definite chemical reactions in the latter; or the system may by its own spontaneous chemical action generate an electric current which traverses the surroundings and there produces the usual effects of such currents. Similarly, the electric variation of a cell or nerve fiber may be an accompaniment or effect of other processes, presumably chemical, in the living protoplasm; but once having arisen, a bioelectric current may act in the same manner as any other electric current and influence secondarily other processes in the electrically sensitive living system. Thus there is every evidence that the electric factor, as such, determines the transmission of excitation from one region to another of a conducting nerve fiber or other excitable protoplasmic system; and that the characteristic rate of transmission is determined by the rate at which the local variation of potential rises from zero to its full value.' It is evident that if

' Cf. my article in American Journal of Physiology, XXXIV (19 14), each active region excites electrically the adjoining inactive region, by means of the local bioelectric current accompanying activity, the velocity with which excitation is transmitted from region to region will be higher the more rapidly this current develops. Lucas' observation just cited shows in fact a close parallelism between these two rates, in the same tissue at different temperatures. Variations in transmission-velocity in different tissues, and in the same tissue under different conditions, w^ould on the foregoing hypothesis have a direct causal dependence on the rate of change of potential characteristic of the tissue.

Reference has already been made to the fact that in each species of animal the bioelectric variations of the active tissues have specific peculiarities — of rate of development, normal range, duration, rhythm, etc. — which exhibit a close correspondence with the peculiarities of function and activity characteristic of the species. For example, the normal bioelectric variation of a special organ like the heart is an accurate index of the normal rate and sequence of its different processes; hence, the electrocardiogram may be a delicate means of detecting abnormalities in the action of this organ. Presumably a tracing of the bioelectric variations from the group of muscles involved in the act of speech could, with sufficient knowledge and analytical skill, be translated into the actual words uttered. From what has already been said it will be obvious that this close correspondence between the functional activity of a living system and the character of its bioelectric variations implies a similar correspondence of both with the underlying variations of metabolic activity.

The range of the electric variations obtained from an isolated muscle which is made to contract in a graded manner by single stimuli of different strengths shows a direct correlation with the height of contraction, but the significance of this fact is not obvious. It is possible (i) that the single elements or cells give actioncurrents of varying intensity, or (2) that each element has a constant variation, but that the number of elements excited varies. The probabihties favor the latter alternative; in certain tissues such as heart muscle, the electric variation exhibits a constant range which is independent of the intensity of the stimulus; the whole tissue responds with a complete contraction to any sufficient stimulus — an example of the ''all or none" behavior — and correspondingly the range of the electric variation is constant. There is good evidence that in normal unfatigued voluntary muscle the variations in the strength of contraction depend on the number of cells contracting, and not on variations in the degree of contraction of single cells.^ Similar considerations apply to the bioelectric response, which in the single elements of this tissue and of nerve appears also to exhibit the ''all or none" character. Apparently to any constant manifestation of normal physiological activity a constant bioelectric variation corresponds.

On the other hand, under certain abnormal conditions the bioelectric variation, e.g., in heart muscle, may continue without the normally associated contraction;^ in such cases the internal contractile mechanism of the cell is incapacitated; and apparently this may occur without disturbing the primary processes of stimulation and conduction, which depend on surface-changes associated with the bioelectric variation. Such a dissociation of conduction and excitation from contraction is also seen during the water-rigor of muscles; at a certain stage of water-rigor a muscle will conduct excitation without contracting/ Under normal conditions, however, the electromotor variation and the functional process exhibit close parallelism. The inverse type of case, i.e., where a muscle contracts or nerve conducts without exhibiting a bioelectric variation, does not seem to occur.^ The electromotor variation seems to be inseparable from the process of stimulation. It may be prevented from appearing (by anaesthesia, etc.), but in that case all of the other manifestations of stimulation are also prevented. Such facts again indicate the primary and controlling role of the electromotor variations in cell-activities.

We have seen that the time occupied by a single electromotor variation (i.e., of the unsummated effect resulting from a single stimulus) varies characteristically 'Biedermann, Sitzungsherichte der Akademie, Wien., XCVII (1888), Part III, p. loi; cf. also Overton, Arch. ges. Physiol., XCII (1902), 146; he finds that frog's muscle immersed in 0.2 per cent NaCl loses contractility while still retaining irritability and conductivity. Cf. also Hartl, Engelmann's Archiv /. Physiol. (1904), p. 80. Robertson has observed the same phenomenon in the intestine of the Australian blowfly after bathing in CaCl, solution {Ergehnisse der Physiol., X [1910J, 305).

» Cf . Lucas' discussion of this question in his Croonian Lecture, op cit., p. 502. from tissue to tissue. This time is exceedingly brief in rapidly responding tissues like voluntary muscle and nerve; for example, in frog's muscle the ''electrical response" has a shorter latent period and a much shorter duration than the ''mechanical response"; the muscular twitch begins (at 20°) about o.oi of a second after stimulation and lasts about o.i second, while (according to Snyder) the latency of the electric variation is about 0.003 of a second and its total duration about 0.007 of a second.^ Thus the electric variation may be completed before the muscle has begun to contract; it is the first evident effect of stimulation and apparently is an index or accompaniment of critical changes which determine the succeeding chemical and mechanical processes.

In general the more rapidly a muscle contracts the briefer is its chronaxie and the more rapidly its bioelectric variation develops.'' There is also a direct correlation between the rapidity of contraction and the rapidity of transmission of the excitation-wave; this is true not only for the transmission in the muscle itself, but also for the transmission in the motor nerve supplying the muscle.^ Rapidity in physiological action thus implies rapidity in the associated bioelectric processes. A muscle and its motor nerve constitute a single reactionsystem, and the rate of the bioelectric processes is a close index of the rate of reaction of the entire system. Transmission of the excitation-state from nerve to muscle through the motor end-plate is apparently a phenomenon of the same kind as transmission from region to region

along the same element. Hence if transmission is a case of secondary electric stimulation by the current of the ^ local circuit, it is clear that rapidity of electromotor variation, implying rapidity of conduction, is necessary in a nerve which excites a muscle by means of the bioelectric circuit formed across the junction between nerve fiber and muscle cell. Lucas has furnished evidence that the motor end-plate in vertebrate muscle has a special chronaxie differing from that of either the nerve fiber or the muscle cell,' but it does not appear that the conditions determining the transmission between nerve and muscle are essentially altered by the presence of this intermediary element. According to Lapicque the blocking action of curare results from an alteration (slowing) of the chronaxie of the end-plate.^ Any two contiguous elements, one of which is excited by the bioelectric variation of the other, must have similar time-factors of excitation; dissimilarity in the timefactors or ''heterochronism" (to use Lapicque's term) would be inconsistent with such transmission. This conclusion is a simple corollary of the general laws of electric stimulation described above; a current traversing an irritable element must have more than a certain duration and rate of change, or it fails to stimulate. Similarly, the momentary ciirrent at the myoneural junction, when the excitation-wave traveling along the nerve reaches that region, must have a duration and rate of change corresponding with the chronaxie of the muscle cell.

In the following table^ are collected a considerable number of observations showing the rate of development gastrocnemius. Frog's sartorius . Frog's sartorius . Frog's sartorius . Frog's sartorius . Frog's sartorius . Frog's hyglossus. Frog's hyglossus. ^ From my article, American Journal of Physiology (1914), loc. cit.; the observations cited are from many different authors; for complete references cf. this article. 2 These observations were made with the thread galvanometer. More recent work on vertebrate nerve with other methods indicates that the rise is even more rapid. The rates shown by the recent work of Gasser and Erlanger with the cathode ray oscillograph are almost double those indicated by Garten's observations cited in the table. Cf, American Journal of Physiology, LXII (1922), 517. See also the work of R. Plant with a rheotome method; in the frog's sciatic the rise was estimated at 0.2 to 0.30- (Z.fiir Biol., LXXVIII [1923], 133).

of the local bioelectric variations, as related to the transmission-velocity of the excitation-wave in a variety of vertebrate and invertebrate tissues under different conditions. It will be noted that cold, anaesthesia, and fatigue, which retard the rise of the bioelectric variation, also retard the speed of propagation in about the same proportion. The general correspondence shown seems to leave no doubt that a direct correlation exists between the respective rates of development of the local electric processes and the velocities with which excitation is transmitted from region to region in the different conducting tissues.

This fact, taken by itself, may seem equivocal in its significance, since it is evident that any wave of alteration occupying a definite length of the conducting element and associated with a change of potential would, as it passed one of the electrodes of a recording instrument (e.g., a string galvanometer), cause an excursion, the rate and duration of which would depend on the speed of the wave. The electric variation might thus conceivably be simply a sign or index of the passage of a wave of activation, without having any causal relation to the process of transmission. Other evidence, however, to be considered below, indicates that the electric variation, as such, is the main factor determining the transmission of excitation from the active region to the adjoining resting region (see chap. xv).

It is significant that many normal bioelectric processes, e.g., those accompanying muscular contraction or nervous activity, as they occur under physiological conditions in the intact organism, are typically rhythmical. The underlying chemical reactions must therefore also be rhythmical, and if these reactions are primarily those occurring in the protoplasmic surface-films, it follows that rhythmical variations in metabolic activity are characteristic of this region of the cell. There are various general facts indicating a tendency to rhythm in the activities at free cell-surfaces; for example, the wide distribution of such phenomena as ciliary movement, in which protoplasmic surface-processes show a regular mechanical rhythm which is presumably accompanied by a chemical and electromotor rhythm^. The filamentous processes formed under certain abnormal conditions from the surface of simple cells like blood corpuscles also often exhibit rhythmical movements/ All such movements are probably of an electro-capillary nature, and referable to general conditions similar to those determining the rhythmical phenomena in the poly phasic inorganic systems (mercury in hydrogen peroxide, iron in nitric acid) described above. As already seen, variations in the structure and composition of thin interfacial films are the essential factors in all such phenomena.

The rhythmical bioelectric variations accompanying the normal innervation of muscle have been investigated in much detail since the application of the thread galvanometer to physiological uses by Einthoven. In man, Piper found the rhythmical action-currents obtained from single voluntary muscles (e.g., extensor of forearm) to exhibit a remarkable constancy of rhythm, of about The minute precipitation-filaments first formed when an iron wire is placed in ferricyanide solution frequently exhibit rhythmical movements of a kind suggesting ciliary movement; cf. Biological Bulletin,

fifty per second.^ Records taken from other muscles showed similar but not always identical rates, varying from forty per second in the thigh muscles to sixty per second or more in the jaw muscles; and it appears probable that under normal conditions different muscles have characteristic differences in their electromotor rhythms. There is every evidence that the muscle rhythm corresponds to the electromotor rhythm of innervation, which in turn corresponds to the rhythm of discharge from the nerve cells in the central nervous system. Numerous observations since Helmholtz' time have shown that a voluntary muscle during contraction emits a low musical note, apparently indicating a rhythmical variation in mechanical tension; and it has been shown by stimulating the nerve rhythmically, by tetanizing currents of known frequency, that the note derived from the muscle has a pitch corresponding with the rhythm of innervation, up to a frequency of several hundred per second.^ The same is true of the rhythm of the electric variation obtained from the muscle during rhythmical innervation; in the frog's muscle the rhythmical electromotor variations show the same frequency as that of the stimulus, until an upper limit of about 150 to 200 per second (at room temperature) is reached ; above this limit the electric variations of the muscle are less frequent than those of the nerve, becoming irregular with the higher frequencies.^ In the case of

^H. Piper, Elektrophysiologie mensMicher Muskeln, Berlin (191 2); cf. chap. vii. * Cf. Piper, op. cit., chap, x, p. 143, for a fuller account and references to literature. warm-blooded animals the synchronism between innervation and the electromotor response of the muscle continues up to frequencies approaching i,ooo per second.^ Limits are set to the possible rate of electric rhythm by the refractory period of the muscle cells. Recently Gasser and Newcomer, using an amplifying arrangement, have shown that in the normal innervation of the dog's diaphragm, the electromotor rhythm of the phrenic nerve corresponds exactly with that of the muscle; every electromotor wave in the muscle appears to be produced by a corresponding one in the nerve; the rhythms observed varied between 72 and 104 per second.^ It is thus clear that the normal rhythm of excitation in the muscle cells depends on the rhythm of innervation; in the intact organism the latter rhythm is determined by the special rate of rhythmical discharge characteristic of the motor nerve cells ;^ and this rhythm, under the usual precisely regulated physiological conditions, is remarkably regular.

Experiments of Piper on the influence of temperature on the natural bioelectric rhythm in the tortoise"* have shown that the temperature-coefficient of the rhythm is of the usual order of chemical reactionvelocities. The following frequencies of oscillation per second were observed by him in a string galvanometer connected with the retractor muscle of the neck, which was caused to contract reflexly at different temperatures (see p. 334). The normal tetanic contractions of voluntary muscles are thus summated contractions resulting from rhythmical innervation. We can thus understand why the summated contraction resulting from artificial rhythmical electrical stimulation of the muscle is physiologically indistinguishable from the normal contraction; in reality the natural as well as the artificial tetanus is a result of rhythmical electrical stimulation. Whether the initial electric disturbance in the muscle cell originates

at the motor end-plate or at the point of entrance of a current from an external stimulating electrode is a matter of indifference, so far as the response of the living tissue is concerned. In the intact organism under normal conditions the rate of rhythm is fixed or predetermined by the constitution of the motor cells in the nervous system. These cells, however, are subject to influence by external conditions, such as temperature or the composition of the surrounding medium (e.g., H-ion concentration), so that many physiological rhythms are

subject to variation in frequency. Such variations offer some of the most beautiful examples of physiological adjustment to the varying needs of the organism; e.g., the respiratory rhythm of warm-blooded vertebrates. The normal rhythm of electromotor discharge may be simple, i.e., there may be a regular succession of single impulses, as in the sinus region which controls the vertebrate heart beat; in this case both the unsummated character of the muscular contraction and the form of the galvanometric record show that in the sinus, auricle, or ventricle a single electromotor variation corresponds to each beat. In other cases, however, a rhythmical succession of discharges may occur, each discharge being itself rhythmical; this is the case, for example, in the nerve cells innervating the respiratory muscles of vertebrates. Garrey has recently observed a similar condition in the Limulus heart ;^ corresponding to each beat there is an oscillatory electrical variation or discharge, undoubtedly originating in the ganglion, each discharge exhibiting a constant number, about twelve, of separate electromotor variations. The series or "volley" of secondary waves or pulses forming each discharge has its own rhythm, which is independent of the cardiac rhythm as a whole. Change of temperature changes the rate of oscillation in each. ganglionic "beat," but the number of distinguishable oscillations always remains about twelve. It appears to be a general rule that the electromotor rhythms are influenced by temperature in the same manner as most other physiological rhythms, i.e., show the chemical temperature coefficient;

^W. E. Garrey, unpublished observations at Marine Biological Laboratory, Woods Hole, Massachusetts. their intimate connection with the metabolic reactions of the living protoplasm is thus indicated. In the frog's muscle the rhythmical character of the bioelectric current during tetanic contraction is most readily demonstrated by the ''secondary tetanus" experiment, in which a muscle with its nerve laid along another muscle is thrown into tetanus when the second muscle is tetanized. In man, the slight mechanical oscillation accompanying steady voluntary contraction of the arm muscles can be demonstrated by means of the hot-wire sphygmograph; this mechanical rhythm has the same period as the electromotor rhythm; i.e., about fifty per second.^ This experiment is especially interesting as indicating (as do also the experiments with the string galvanometer) that the various nerve cells inervating a muscle discharge synchronously or in phase with one another. The experiments of Gasser and Newcomer also indicate that this is true of the nerve cells innervating the opposite halves of the mammalian diaphragm.^ Apparently the several nerve cells constituting each motor group are under some control by which all are impelled to react synchronously, or ''keep time." This suggests a co-ordination dependent on some rapidly transmitted influence, presumably electrical; the synchronous activity of spermatozoa when gathered in clumps, or of ciliated cells, seems to be a phenomenon of a closely related kind (see p. 392).

'A. V. Hill, Journal of Physiology, LV (192 1), Proceedings of the Physiological Society, p. 14. We have already given reasons for regarding the bioelectric variations as a result of structural change, controlled by metabolic change, in the protoplasmic surface layers. From this point of view another constant accompaniment of the stimulation-process, the refractory period, receives a consistent theoretical explanation. The presence of the intact surface-film is necessary for stimulation; if the film is broken down as a consequence of stimulation, it must be re-formed and restored to its original state before a second complete stimulation is possible. This deduction is in agreement with our general experience of stimulation-processes. It is a striking fact that in all irritable tissues stimulation is immediately followed by a period of insensitivity and subnormal irritability; this period of temporary depression, which is extremely brief in rapidly responding tissues, is the refractory period, and there is evidence that it corresponds to the period during which the film is undergoing breakdown and reconstruction. During the refractory period the tissue loses at the same time both its susceptibility to electric stimulation and its ability to transmit states of excitation; in other words, there is a temporary loss of both irritability and conductivity.^

The duration of the refractory period varies greatly in. different irritable tissues, and under normal conditions is specific for each tissue. The most significant general correlation is that it is brief in tissues with brief chronaxie, and vice versa. As already pointed out, in any irritable tissue the length of the chronaxie is closely related to the duration of the single bioelectric variation; and the duration of the refractory period shows a similar correlation. This parallelism has frequently attracted attention, and its significance has recently been discussed in considerable detail by Tait,^ who has reached the conclusion that the first part of the period, the interval of complete inexcitability or '^ absolute refractory period," which is of brief duration, corresponds with the period of upstroke of the curve of electromotor variation, while the '^relative" part of the period corresponds with the downs troke or return phase. Tait has shown that the relative refractory period is greatly prolonged by drugs (yohimbine, protoveratrine) which retard the return phase of the bioelectric variation. In general, he finds the return of irritability to run parallel with the Return of the normal resting potential of the muscle.

While it seems clear that the delayed recovery in such cases has a connection with the delay in the recovery of the normal electromotor properties of the tissue, the correlation is apparently not a simple one. More recent evidence shows that under normal conditions the relative refractory period in muscle and nerve may last several times longer than the return phase of the electric variation.^ The curve of the latter is very nearly symmetrical ' Tait, Quarterly Journal of Experimental Physiology, III (igio), 211.

* Cf., for heart muscle, Trendelenburg, Arch. ges. Physiol., CXLIV in unfatigued tissues, while the change in irritabilityfollows a markedly asymmetric course; i.e., there is a rapid and complete loss of irritability on stimulation, followed by a relatively gradual recovery. Evidently in the recovery process additional factors enter which are independent of the bioelectric variation. It is, however, highly significant that in all cases irritability seems to disappear completely during the rising phase of the variation. This partial correlation between the timerelations of the two phenomena indicates that the conditions determining the electromotor variation are closely connected with those determining the temporary loss of irritability.

The division of the whole refractory period into two distinct subperiods, known respectively as ''absolute" and ''relative," is its most interesting feature. The absolute period is the brief interval of complete insensitivity immediately following a single effective stimulus. During this interval a second stimulus, no matter how strong, has no appreciable effect. It is as if the tissue for a brief time lost its irritability completely; hence, two stimuli succeeding each other within this interval produce the same effect as a single stimulus; while if the second is sent in after the completion of this brief interval, its effect is seen in an increased response or summation-eft'ect. The completely inexcitable period in a frog's motor nerve at 20° lasts about o.ooi to 0.0015 of a second;^ in the sartorius muscle it is from two to three times longer; and in both of these cases its duration is closely similar to that of the upstroke of the bioelectric

variation.' It is followed by a second period, the relative refractory period, during which irritability returns progressively to normal; this interval has several times the duration of the absolute refractory period, and in nerves under certain conditions (increased H-ion concentration) it may be followed by a brief period of supernormal excitability.^ In cardiac muscle the return of normal excitability is relatively very slow, even when the increased duration of the bioelectric variation is taken into account; in this tissue the period of complete inexcitability appears to outlast the entire bioelectric variation, and its limits do not seem to be very clearly defined; at 15° Lucas found its duration to be somewhat more than 0.4 of a second.^ Usually it has been supposed that the regular and somewhat slow rhythm characteristic of this tissue is dependent on its long refractory period, which is almost equal in duration to the period of muscular relaxation. A prolonged refractory period is also characteristic of the nerve cells controlling other physiological rhythms of slow period, such as those of certain motor reflexes in higher vertebrates (e.g., scratch reflex in the dog).

In any given tissue the duration of the refractory period is influenced by the chemical conditions in the surroundings as well as by the physiological state of the tissue and the temperature. It is lengthened by fatigue, ^ This is apparently strictly true of frogs' voluntary muscle, but in nerve Adrian finds its duration somewhat longer, equal to that of the whole bioelectric variation. In cardiac muscle it may be still longer. Cf. Adrian, Journal of Physiology, LV (1921), 193.

lack of oxygen, and partial anaesthesia, at least in some cases ;^ but in nerve Lucas found the rate of recovery to be normal in solutions of alcohol sufficient to greatly retard transmission.^ Bazett observed an influence of the salts of the medium, increase of potassium lengthening, and increase of calcium shortening the interval.^ Various poisons such as veratrine, muscarine, digitalis, and barium salts also lengthen the refractory period."*