SCIENCE TRANSMISSION OF ACTIVATION IN 66 point of stimulation and activates the whole. The characteristic functional manifestation then appears-contraction in a muscle cell, motor reaction in a protozoon, cell division and development in a resting egg cell, etc. Conduction is in fact a widely general if not universal cell process. Excitation may thus be transmitted not only between different regions of the same cell or cellular element but also between different cells or elements which are in contact with one another; the transmission between neurones in the central nervous system and from a nerve to its muscle or other terminal organ illustrates this type of conduction. It is thus possible to distinguish between intracellular and intercellular conduction, although there is probably no essential difference between the two types. Physiological transmission of the kind described seems to have in it something mysterious and specifically vital; in fact the problem of the essential physico-chemical nature of nerve conduction-the type phenomenon of this class is still regarded by most physiologists as unsolved, and apparently by many as insoluble. The difficulty of the problem has been accentuated by the apparent lack of any close analogies with known inorganic processes. Comparisons with the electric current, with the transmission of mechanical influences such as elastic strain or vibration, and with the propagation of explosive waves or of germeffects like crystallization in supersaturated solution, have all proved inadequate and often highly misleading. Yet it would seem that any phenomenon which is so universal in organisms and upon which many of their most characteristic activities directly depend-especially in animals-must have some general physico-chemical basis present also in inorganic nature. The problem is to find some simple and readily reproducible inorganic process, involving transmission of chemical influence, which is similar in its most general features to the conduction process in living cells, initiated under similar conditions, and dependent upon the same fundamental factors. What is to be looked for is not complete or detailed identity of the physiological process with its inorganic model, but rather a class resemblance of a definite and unmistakable kind; the inorganic process should exhibit peculiarities which stamp it clearly as a phenomenon of the same essential kind as the physiological process. If the comparison is a true one, the transmission of chemical influence to a distance in cells or nerve fibers and the transmission of similar influence in the inorganic model should take place at similar rates, be influenced similarly by external conditions, be initiated by the same means, have the same external manifestations, and be dependent upon the same underlying physico-chemical factors. Just as the passage of the pulse wave in an artery and that of a distension wave in a simple elastic tube are both determined by general physical factors common to both objects, so the transmission of chemical or metabolic influence along a living conducting element like a nerve should-in the case of a valid comparison-depend upon certain fundamental features of physico-chemical constitution present also in the inorganic model. Is there in fact any known general class of non-vital physico-chemical phenomena to which we can thus assign the phenomenon of protoplasmic conduction? In the stimulation of an irritable living structure by an external agent, the primary or releasing event is undoubtedly a surface process of some kind; the characteristic activation or "response" of the whole irritable element follows automatically upon this surface change. In most irritable cells any local mechanical or chemical alteration of the protoplasmic surface layer (or "plasma membrane "), or a slight change in its electrical polarization due to an electrical current, may cause excitation. There is little doubt, however, that the essential determining factors in any form of stimulation are electrical; and that mechanical and chemical stimuli excite the cell indirectly by means of the local electrical effects which they produce. The stimulating agent alters locally the structure or composition of the surface film; the state of electrical surface polarization is there changed; and the bioelectric circuit arising between altered and adjoining unaltered regions completes the activation. This view at once explains why the electric current is the most universal stimulating agent. It is well known that stimulation of any cell, by whatever means induced, is always accompanied by an electrical variation of the cell surface, or current of action; and we find the same to be true of the propagation of the excitation wave. This last process, which is evidently essential to the stimulation of the cell as a whole, is apparently dependent upon the bioelectric circuit formed at the boundary between the active and inactive regions of the cell surface; that part of the local current which traverses the still inactive regions stimulates these electrically; the regions thus secondarily excited act similarly upon the resting regions next adjoining; the process repeats itself automatically at each new active-inactive boundary as it is formed, and in this manner the state of excitation spreads continually from active to resting regions. A wave of activation thus travels over the surface of the element.1 If this theory of conduction is well founded, the chemical alteration of a surface film of material under the direct influence of local electrical circuits would seem to be indicated as the essential basis for the transmission. Changes of this kind are in fact a frequent phenomenon at the surfaces of metals in contact with solutions; and in a recent paper2 I have called attention to the many striking analogies between the effects of such local electrolytic action in metals and the effects of local stimulation in living cells. For example, in the rusting of iron in aqueous solutions the formation of local electrical circuits between different regions of the metallic surface is now generally recognized to be the chief factor in the process. The surface layer of metal is typically not homogeneous, but exhibits local anodal and cathodal areas; at the former regions the ions of the metal enter solution and are precipitated as oxide or carbonate, while nascent hydrogen and alkali are presumably formed at the cathodal regions. 1 Cf. Amer. Jour. Physiol., 1915, Vol. 37, p. 348; 1916, Vol. 41, p. 126. 2 Loc. cit., 1916. Each of the areas of local chemical action thus represents an electrode-area in a local electrical circuit; and electrolysis at these areas is what determines the chemical changes there taking place. Now electrolysis is a process in which the transmission of chemical influence to a distance without transfer of material is an essential and constant characteristic; the very flow of the current depends in fact upon this condition. Any electrochemical change at one electrode of a battery or other electrical circuit due to chemical action necessarily involves a corresponding change of a chemically opposite kind at the other electrode. Oxidation, the general effect at the anode, thus involves simultaneous reduction at the cathode; an oxidizing substance placed in contact with one electrode will thus instantly oxidize a reducing substance at the other electrode. Spatial separation of the two regions is a matter of indifference except in so far as it increases the electrical resistance of the circuit, thus retarding the rate of the electrochemical process. The transmission of the chemical influence between the electrodes is automatic and instantaneous. This chemical distance action 993 suggests a possible basis for the protoplasmic type of transmission, since distance action is a feature of all electrochemical circuits, including those present in local action at metallic surfaces. If therefore it could be shown that the cell surface can act like a metallic surface the essential difficulties of the problem of protoplasmic transmission might be regarded as overcome. An inconsistency, however, appears in the fact that the transmission of electrochemical influence in a circuit is instantaneous (i. e., 3 X 1010 cm. per sec.), while the most rapid protoplasmic transmission-in the motor nerves of mammals-is only 120 meters per second; again, the intensity of chemical distance action decreases with the distance between the electrodes, because of the increase in electrical resistance, while in the nerve impulse there is normally no decrease in intensity (or "decrement") as the local change 3 Cf. Ostwald, Zeitschr. physik. Chemie, 1891, Vol. 9, p. 540. passes along the fiber. Such difficulties are only apparent, however; in nerve conduction it is quite certain that an entirely new state of activity is aroused at each successive region of the fiber as the impulse passes; and all of the evidence indicates that the speed of transmission is determined mainly by the sensitivity and local rate of response of the nerve,* and not at all by the rate of transmission of the electric current in the bioelectric circuit. It is probable that in the local bioelectric circuit set up by the initial stimulus the direct chemical influence of the current extends for only a short distance, at most a few centimeters from the original site of stimulation; but one of its effects is to originate a new and similar circuit in the adjoining regions of the fiber; this process repeats itself as already indicated, and in this manner the impulse spreads. The observed speed of the activation-wave has thus nothing to do with the speed of the purely electrochemical distance effect. What we seem to observe is a local electrical circuit which travels along the nerve together with the activation wave; but in reality there is a succession of new circuits, each of which automatically arises at the boundary between resting and active regions as the front of the activation wave advances. The relatively slow rate of movement of the impulse and the absence of a decrement may thus be understood. The rapid passage of a wave of chemical decomposition (probably oxidative in nature and involving some structural change) over the surface of the reacting element, followed immediately by a reverse change which restores the original or resting condition, is what appears to take place in a nerve or other living structure during conduction. Associated with the chemical process is a local electrical circuit by whose electrolytic action the chemical change is apparently determined. Have we examples of similar processes in inorganic systems? It appears in fact that this general type of process is not unusual in metals in contact with solutions. Especially clear and 4 Cf. Amer. Jour. Physiol., 1914, Vol. 34, p. 414; Vol. 37, p. 348. striking examples are seen in the transmission of the state of activity over the surface of metals, especially iron, which have been brought into the temporarily non-reactive or "passive" condition by immersion in strong nitric acid (or other suitable oxidizing agent) and are then placed in dilute acid and made to react. It has long been known that iron which has been thus "passivated" becomes resistant or refractory to reaction and (for example) no longer dissolves spontaneously when placed in dilute nitric acid (s. g. 1.20). But if while immersed in the dilute acid it is touched momentarily with a baser metal, or with a piece of ordinary non-passive iron, it is at once activated" and reacts vigorously with the acid until dissolved. The experiment is a striking one and easily performed. In my own demonstrations a piece of pure iron wire (No. 20 piano wire, bent at one end into a hook for handling) is passivated by immersion in strong nitric acid (s. g. 1.42) for a few seconds, and is then placed (by means of a glass hook) in a flat dish containing dilute acid (s. g. 1.20). The wire if left undisturbed remains bright and unaltered for an indefinite time. If then it is touched at one end with a piece of ordinary iron, or with zinc or another baser metal, the bright metallic surface is at once darkened (through formation of oxide) and active effervescence begins; this change is transmitted rapidly, though not instantaneously, over the entire length of the wire; the velocity of transmission varies with the conditions, and is of the order of 100 or more centimeters per second in this experiment. The wave of activation may also be initiated mechanically, e. g., by bending the wire or tapping it sharply with a glass rod; or chemically, e. g., by contact with a reducing susbtance such as sugar; or electrically, e. g., by making the wire (while immersed in the acid) the cathode in any battery circuit (of two or more volts potential), preferably with another piece of passive iron wire as anode; B For a recent extended study of the passive state in metals with full references to the literature, cf. Bennett and Burnham, Jour. Physical Chem., 1917, Vol. 21, p. 107. the cathodal wire is instantly activated, while the anodal wire remains unchanged. Activation with the electric current is thus typically a polar phenomenon, just as is the excitation of a living irritable element like a nerve. Activation by contact with active iron or a baser metal is in reality an instance of electrical activation, the activating metal forming the anode of the local circuit arising at the region of contact. At the local cathode, i. e., the adjoining passive iron, the metal is at once activated, and the effect spreads in the manner already indicated by means of the circuit which automatically arises at the boundary between active and passive areas. Any metal which thus activates by contact must be of such a nature that the passive iron becomes the cathode of the local circuit formed. A metal which is nobler than passive iron, like platinum, not only does not cause activation, but it renders the iron locally more resistant to activation; thus the passage of the activation wave may be blocked by the contact of a platinum wire. This latter effect depends upon the formation of a local circuit of the reverse orientation, the iron becoming anodal, a condition which furthers passivation and hinders activation. Active iron is a base metal in relation to passive iron, being more negative that the latter by ca. 0.75 volt in 1.20 HNO3; hence when any region of a passive wire is rendered active it immediately activates the adjoining areas. In passivation the surface layer of the iron. is modified in a peculiar manner, apparently by the formation of a thin resistant layer of higher oxide. Any condition that interrupts locally this surface film of altered iron forms necessarily a local circuit by whose action the whole metal is activated in the manner just described. Apparently at any cathodal area the surface film of oxide is reduced to metallic iron; contact of a reducing substance has a similar effect; while a mechanical agent breaks the continuity of the film and exposes the unaltered iron beneath, thus forming the local circuit. The reason why mechanical, chemical and electrical influences all produce the same effect is thus evident. The parallel to the living irritable tissue is plain; local alteration of the protoplasmic surface film produces effects of a closely comparable nature, which spread in an analogous manner by means of the local electrical circuits formed. We are thus enabled to understand why any rapid local alteration of the cell surface may activate the whole cell-in other words why the cell is so characteristically "irritable." The iron wire in its passive state may be compared to the irritable living element in a state of rest. The state of inactivity continues in both cases only so long as the surface layer is intact and homogeneous. The reason why the whole cell (or the whole iron wire) responds completely to a local stimulus is simply because transmission over the entire surface follows automatically and inevitably upon local activation. The 66 all-ornone "behavior thus becomes intelligible. Under normal conditions an irritable nerve or muscle returns spontaneously to an inactive or "resting" state after stimulation, and for renewal of activity a second stimulus is required. The resting condition thus represents a condition of equilibrium, which is temporarily disturbed by the stimulating agent. The same is true of the passive condition of iron in strong solutions of nitric acid. In weaker solutions, of s. g. 1.20 and less, the reaction once initiated continues unchecked until all of the iron is dissolved; but in stronger solutions the reaction is temporary and the metal returns spontaneously to the passive condition. A wave of temporary activity thus sweeps over the surface of a passive iron wire which is activated (e. g., by touching with zinc) in nitric acid of s. g. 1.25 or higher; the state of local activity lasts in such a solution for a brief period only, which is the shorter the higher the concentration of the acid. An interesting gradation of effect may thus be shown by activating a series of passive wires in different dilutions of strong (s. g. 1.42) acid, e. g., 90, 80, 70, 60 and 55 volumes per cent. (i. e., 90 c.c. 1.42 HNO plus 10 c.c. water, etc.). When a wire immersed in pure 1.42 acid is touched at one end |
