STUDIES IN THE PHYSIOLOGY OF THE NERVOUS SYSTEM. XXI.: RHYTHMIC MOVEMENTS AS SUCCESSIVE (OR TERMINAL) REFLEX PHENOMENA OF COMPOUND STIMULATION [PROGRESSION CONDITIONED BY ANTAGONISTIC REFLEX STIMULI]

1. In the experiments described in this paper the occurrence of rhythmic phenomena in successive phases of compound reactions is shewn. Rhythmic phenomena thus occur in the third phase of such a compound reaction— that is, in a phase in which the background reflex is continued after the withdrawal of the interrupting reflex. They also occur in the fourth phase of such a compound reaction—that is, in the phase which immediately succeeds the final withdrawal of the background reflex. The rhythmic movements are reciprocal in the two muscles—that is, when contraction occurs in one muscle reciprocal relaxation is taking place in the other; often pauses occur between the beats; the rhythm is slow; and there can be no doubt that this form of rhythmic phenomenon is similar to the form which occurs during double stimulation as an immediate compound reaction—nor can there be any doubt that the two are forms of the rhythmic phenomenon which underlies the act of mammalian progression—“galloping,” “running,” “walking,” or “stepping.” 2. The rhythmic phenomenon in successive phases of compound reactions may be exhibited in spinal preparations. 3. This has been observed to occur in the third phase of compound reactions in which the background was one of contralateral extension. 4. It has also been observed when the background was one of ipsilateral flexion. The rhythmic phenomena may change in appearance and disappear when the relative strengths of the two stimuli are varied. 5. Rhythmic reactions may occur as successive phenomena of compound stimulation in decerebrate preparations. 6. Where the two stimuli are synchronously withdrawn a rhythmic terminal phenomenon has been observed to occur where there is no rhythmic rebound phenomenon in either of the “pure” reflexes. This, for instance, has been observed where there was flexor after‐discharge in the “pure” flexion‐reflex and extensor after‐discharge in the “pure” extension‐reflex. In such cases it is possible that there occurs a compounding of the two antagonistic terminal phenomena. In other instances a rhythmic terminal phenomenon in the “pure” flexion‐reflex has been observed to become better marked in the terminal phenomena of the compound reaction. In yet other cases a slight rhythmic phenomenon in the “pure” flexion‐reflex has been observed to become completely abolished in the terminal phenomena when that reflex is compounded with an extension‐reflex which, when “pure,” has a strong extension after‐discharge. 7. Where the extension‐reflex forms the background, and the interrupting flexion‐reflex is withdrawn before the extension‐reflex, rhythmic phenomena may occur in the third phase of the compound reaction—that in which the background extension‐reflex continues alone after the period of double stimulation. 8. This may be the case where there is a flexion after‐discharge in the “pure” flexion‐reflex. In such cases the rhythmic phenomenon may be conditioned by the balance between this and the maintained immediate extension of the extension‐reflex. In other instances the rhythmic phenomenon has appeared where there was no maintained flexion after‐discharge in the “pure” flexion‐reflex; but in these cases there was flexor augmentation of contraction during double stimulation in the compound reaction, and this may have conditioned a flexor terminal after‐discharge. Decrease in strength in the interrupting ipsilateral stimulus under these conditions serves to abolish the rhythmic phenomenon. Where rhythmic movements occur in the period of double stimulation as well as in the third phase of the compound reaction a comparison may be made between them. It then appears that where the factor of extension in the rhythmic phenomenon is relatively increased (that is, in the third phase) there occurs a slowing of the beats and they tend to become less incomplete, as seen in the flexor. Where there is a rhythmic rebound phenomenon in the “pure” flexion‐reflex there may also be one in the third phase of the compound reaction in which extension is the background. The flexor beats are then more complete and are slower than in the “pure” flexion‐reflex—there is a greater element of extension in the rhythmic phenomenon of the compound reaction. 9. Rhythmic phenomena may occur in the fourth phase of the compound reaction in which extension is the background. This may be the case when there is no rhythmic terminal phenomenon in the “pure” extension‐reflex which is used as background, and when the termina phenomena of that reflex consist only in a maintained extension after‐discharge. In experiments in which the terminal phenomena of the “pure” extension‐reflex are rhythmic and are exhibited in both the antagonists these movements may be modified in the phenomena of the fourth phase of a compound reaction in which a flexion‐reflex is superposed upon the extension‐reflex. This modification takes the form of an increase in the flexion factor in the rhythmic phenomena. Thus that phenomenon may still remain rhythmic, but the flexor beats may be of increased extent and of greater rapidity—the pauses in complete relaxation being here shortened. Sometimes the rhythmic phenomenon may be depressed by an increase in the foundation of maintained flexion. In such cases the relaxation phases of the flexor beats are reduced in extent, and no pauses occur at the level of the lowest points in the beats. These results may be explained on the assumption that there persists an exaltation of flexion as an after‐effect of the interrupting flexion‐reflex, and that this compounds either against a background of maintained extension (the extensor after‐discharge) to give a rhythmic phenomenon, or against a rhythmic phenomenon already present, to overset the nice balance of the centres towards the side of flexion. 10. Where the flexion‐reflex forms the background and the interrupting extension‐reflex is withdrawn before the flexion‐reflex, rhythmic phenomena may occur in the third phase of the compound reaction—that in which the background flexion‐reflex continues alone after the period of double stimulation. 11. This may be the case where the terminal phenomena of the “pure” extension‐reflex are arrhythmic and consist only in a maintained extensor after‐discharge. When this is compounded against the immediate flexion of the third phase of the compound reaction there may appear a rhythmic phenomenon. If then the relative value of flexion in the balance is increased by increasing the strength of the flexion background or by decreasing the strength of the extension‐reflex, the rhythmic phenomenon is found to be changed. There is a greater degree of the foundation of maintained flexion; the relaxation phases of the flexor beats are diminished in extent; and the beats are increased in rate. If the extension factor is increased the beats show more completeness; pauses appear at the level of complete relaxation in the flexor beats, or become of greater duration if already present; the extents of the flexor beats diminish; the beats become slower in rate of rhythm; and the foundation of maintained flexion diminishes whereas that of maintained extension—as exhibited in the intervals which correspond to the pauses between the flexor beats—increases. 12. Where a rhythmic terminal phenomenon is present in the “pure” extension‐reflex, that may be changed in the third phases of compound reactions in which it interrupts a flexion‐reflex. If the rhythmic phenomenon is not entirely suppressed in the third phase of the compound reaction the flexion factor in it is increased. The beats of the flexor muscle are faster, of greater height, but of smaller relaxation phases. The rhythmic phenomenon may, however, be completely suppressed; and in series of successive compound reactions the suppression may be observed to be partial or complete as the relative value of the factors in the third phase is changed. 13. The variation of the relative values of the factors of extension and flexion in the third phases of the compound reactions may thus be produced by changing the strength of the background flexion‐producing stimulus or that of the interrupting extension‐producing stimulus. In these cases it is found that increase in the relative value of flexion thus produced leads to an increase of the factor of flexor contraction in the rhythmic phenomenon, and when that value of flexion is progressively increased it culminates in the substitution of a state of maintained flexor contraction for the rhythmic phenomenon in the third phase. 14. The variation of the relative values of flexion and of extension in the phenomena of the third phase may be brought about by varying the temporal arrangement of the stimuli without altering their strengths. The later the extension‐reflex is applied in the period of the flexion background the greater is its relative value during the period of double stimulation, and the greater also is probably the value of extension in the third phase. In these circumstances the rhythmic phenomenon in the third phase may be absent, or suppressed by over‐balance to the side of flexion, when double stimulation falls early in the period of the background; and it may appear, or become ever more complete, as that phase of double stimulation is made to fall later and later in that period of the background flexion‐reflex. As it becomes more complete the flexor relaxations become greater, as do also the reciprocal extensor contractions. 15. Where the flexion‐reflex forms the background and the interrupting extension‐reflex is withdrawn before the flexion‐reflex, rhythmic phenomena may occur in the fourth phase of the compound reaction—that after the withdrawal of the background reaction. 16. This may be the case where there is no rhythmic terminal phenomenon in the “pure” flexion‐reflex. In such cases the rhythmic phenomenon of the fourth phase of the compound reaction may be conditioned by the balance between a state of flexion after‐discharge in the flexion‐reflex and an extensor after‐discharge due to the extension‐reflex and carried over the third phase of the compound reaction. Where there is a rhythmic terminal phenomenon in the “pure” flexion‐reflex, that may be modified in the fourth phase of the compound reaction. The beats may become more complete, and slower in their rate of rhythm. By increasing the strength of the background flexion‐reflex it may be surmised that the factor of flexion (flexor after‐discharge) in the fourth phases of the compound reactions is increased in value. It occasionally happens in these circumstances that there is a slowing and not a quickening of the rate of rhythm of the flexor beats where the terminal phenomena of the fourth phase are rhythmic: at the same time the relaxation phases of the beats may be reduced in extent. 17. The results described in this paper strengthen the view already put forward that rhythmic reactions occur when two central activities (here reflex and of peripheral origin) of opposite nature are pitted against each other with a certain nicety of balance. The central activities here used are those of flexion and extension in the hind limb; and it is shewn that when these are varied in relative value in the successive or terminal phases of compound reactions in the manners already described in previous papers of this series (that is, by altering the relative strengths of the background and interrupting stimuli without altering their temporal relations; or by altering their temporal relations without altering their relative strengths) rhythmic reactions may be made to appear in these phases, and may be modified by increasing the relative value of flexion or of extension. 18. Some rhythmic reactions appear to have in them a great preponderance of flexion. Others appear to have a great preponderance of extension. Between these are intermediate forms in which the balance of flexion and extension is more nicely adjusted. It is possible to make a series of rhythmic reactions which will demonstrate at one end the effect of increasing the factor of flexion in the balance, and at the other end will demonstrate the effect of increasing the factor of extension. 19. It should theoretically be possible to obtain such a series in each decerebrate preparation of the cat. In some experiments such series may be obtained. But there is a practical difficulty in that it often happens that the range of rhythmic reactions is obtained with a very small variation in the relative strength of the antagonistic stimuli. 20. Different individual cats vary much in the range of differences in strength of stimuli at which the rhythmic phenomena in compound stimulation (whether immediate or successive) occur. Thus in some cases a small increase or decrease in the strength of one of the antagonistic stimuli is sufficient to abolish the rhythmic reaction. In other cats that rhythmic reaction persists over a large increase or decrease in the strength of one of the antagonistic stimuli. 21. When the typical series of rhythmic reactions is examined it may be described (starting with reactions in which there is a preponderance of flexion and ending with those in which there is a preponderance of extension) as follows:— ( a ) Maintained flexor contraction and extensor inhibitory relaxation (that is, a reaction similar to the “pure” flexion‐reflex of arrhythmic form). ( b ) The maintained flexor contraction is broken by relaxation phases of short extent and duration. That is, the movement of relaxation does not carry the level of flexor contraction far below the level of maintained contraction, and when it reaches its lowest point restitution of contraction at once supervenes. In the extensor muscle there is little or no movement demonstrated in the tracings. ( c ) The relaxation phases in the maintained flexor contraction become of greater extent and longer duration. Reciprocal movements occur in the extensor. These take the form of small and sharp contractions which commence slightly after each flexor relaxation and terminate before that relaxation reaches its lowest point. There then succeeds an equally sharp extensor relaxation, and this is followed by a pause at the level of complete relaxation. The rhythmic phenomenon is usually slower than in b, and continues to become slower; but sometimes it appears to become faster. ( d ) The flexor relaxation phases become “complete”—that is to say, that the lowest point in the flexor relaxation phase corresponds in level with the level of rest of the muscle. No pause as yet occurs at the level of complete flexor relaxation. In the extensor muscle the extensor contractions become of greater extent. The whole rhythmic phenomenon tends usually to become slower in rate of rhythm. ( e ) Pauses occur in the flexor movements at the level of complete relaxation. In the extensor muscle there appear pauses at the level of maximum contraction. In the flexor curve the rhythmic phenomenon is now exhibited as a series of discrete beats. ( f ) The highest points in the flexor beats diminish in height; the pauses between the beats become of greater duration; the height of maintained extensor contraction increases and the extents of the extensor relaxations diminish. ( g ) Flexor movements disappear. In the extensor curve there are shallow relaxation phases. ( h ) Flexor movements not present (or maintained flexor inhibitory relaxation is present); maintained extensor contraction (that is, a reaction similar to the “pure” extension‐reflex of arrhythmic form). 22. These different types of rhythmic reactions are seen in the immediate and terminal phenomena of some “pure” reflexes; in “pure” reflexes when modified by compounding in temporal succession, and in the immediate and terminal phases of compound reflexes. 23. They are also seen in the different phases of the progression phenomenon described by the author as induced by rapid division of the spinal cord in decerebrate cats. There can be no doubt that these are all forms of rhythmic mammalian progression; and the suggestion may be hazarded that the progression phenomenon above referred to serves as a map in the different districts of which all the different reflex phenomena are figured.