G. H. Bishop

The sizes of nerve fibers supplying cerebral cortex

Abstract The nerve fiber supply to various cortical areas of several vertebrate animals has been examined by means of the electron microscope, the fibers measured in cross sections of white matter as it approaches cortex, and plots made of numbers of fibers vs diameters. Most of the fibers are myelinated in mammals, and the relatively few unmyelinated structures are not clearly distinguishable from glial elements. In mammals, each distribution curve shows a maximum at approximately 1 μ, some fibers as small as 0.2 μ, a rapid decrease in numbers up to 4 μ, and a plateau finally declining to zero at diameters of 5 to 14 μ in different cortical areas of different species. In some small areas sampled there may be few or no fibers over 3 μ. In the motor area of the macaque monkey, a second maximum, at 8 to 9 μ presumably represents the Betz cell axons. Otherwise no secondary maxima such as the alpha to delta peaks of peripheral nerves are obvious, but reasons are presented for fividing the total size spectrum range into three ranges, corresponding to the large myelinated afferents passing into the dorsal cord column, medium-sized fibers whose postsynaptics cross the cord to ascend in the ventrolateral column, and the unmyelinated fibers taking a similar course through the cord. There are reasons for considering these three size groupings in spinal nerves to be phylogenetically successive, and to have been extended by relays successively to higher centers. The smaller fiber group is then the most primitive, and evidence here presented indicates that similar-size components have been successively acquired during the phylogenetic development of the mammalian cortex. Similar examination of the white matter of reptilian cortex shows few myelinated fibers, and few of these are as large as 4 μ. They are inferred to be the precursors of the 0.2 to 3 μ range of mammals, to which successively higher mammals have added increasing numbers of fibers of increasing diameter. The fiber size distributions of the pyramidal tract, optic tract and olfactory nerve have been analyzed for contrast with the trigeminal nerve distribution. The results are consistent with the phylogenetic interpretation suggested. Examination of the cortex of an insectivore, the mole, a relatively primitive mammal, shows the typical mammalian pattern of fiber size distribution, but with small areas of white matter containing no fibers over 3 μ, others with none over 5 μ, but an overall of 9 μ. It is proposed that these morphological components, whose axons are recognizable physiologically in terms of thresholds and conduction rates, may contribute phylogenetically successive functional patterns to cortical activity. Tracing the origins of paths activating cortex, and destination of paths originating in cortex, may lead to identification of types of response correlated with fiber sizes.

Potential wave mechanisms in cat cortex

Abstract Evidence is presented for two types of potential-wave responses of apical dendrites of cortex. The shorter-latency of these responses decreases when its afferents are stimulated at 6 per sec. or faster, and the longer-latency process increases in amplitude. The incremental sequences include the classical recruiting waves following thalamic stimulation, and similar responses from cortex to stimulation of other regions of cortex, or from stimulating locally at the lead position. Four groups or size ranges of fibers activating cortex can be differentiated. The largest fibers are those of the afferent radiations to sensory primary projection areas. The next smaller activate somewhat similar responses in association area cortex when projection cortex is stimulated. A third group or division activates the decremental short-latency responses of cortex. The fourth group activates incremental or recruiting waves. From its conduction time, the latter group is inferred to consist of unmyelinated fibers. The recruiting property appears to be a function of the excitability cycle of one type of dendritic synapse, which shows two periods of facilitation. After a first stimulus, a second is effective during the 15 msec. duration of response to the first, after which a period of depression or relative refractoriness follows. At one sixth to one tenth sec. this depression passes over into supernormal excitability. That is, the repetition of an initial stimulus causes an increased response during the supernormal phase. Spontaneously arising recruiting spindles in nembutalized or decerebrate preparations show close resemblances to stimulated recruiting trains, and the interaction of these two sequences of waves indicates that they may occur in the same elements. In fact what are apparently waves of the spontaneous spindle type may appear during recruiting responses to repetitive stimulation if the frequency of stimulation is less than the frequency of spontaneous spindle discharge. This phenomenon seems related to that of “doubling” of response to each stimulus during repetitive stimulation that finally passes over into paroxysmal afterdischarge. Certain relations between incremental and decremental sequences of waves, when stimulated at different loci but recorded at the same leads across cortex, indicate that the two types of response may be those initiated at synapses via two afferent paths of different fiber types, impinging on the same dendrites. Inferences that seem reasonable will account for the alpha rhythm and recruiting and spindle responses as processes in the same or similar elements, activated in somewhat different patterns.

The effects of polarizing currents on cell potentials and their significance in the interpretation of central nervous system activity

Abstract Since no potential could be recorded from the medium surrounding a cell which was uniformly and synchronously active over its whole surface, potentials recorded across cell layers represent the difference in activity at apical and basal regions. Since potentials led from simple cell layers are effectively monophasic, the conduction time along the soma is a minor factor in determining the form of the record. Diphasic records are inferred to be summations of cell discharges some of which, represented predominantly in one phase, have a difference of potential of one polarity, others represented chiefly in the other phase having a potential difference of opposite polarity. If this is valid, potential differences of opposite polarity must be exhibited by cortical pyramids with a uniform anatomical orientation. A difference in the distribution or intensity of response of apical and basal dendrites is suggested as a possible factor. Polarization from an external source across a simple cell layer may either increase the amplitude of the monophasic difference of potential led from across it, or decrease and finally reverse its sign; presumably by altering the relative amplitudes of electrical activity at the two poles of the cells, but without blocking conduction from pre- to post-synaptic axons. The interpretation of this fact involves that polarizing currents act on cells essentially as they do on peripheral nerve axons. The change in polarity of record is in a direction opposite to the sign of the polarizing current. Polarization of optic tract axons may result in prolonged after-effects, negative under anodal polarization and positive under cathodal. This is interpreted as representing a repolarization, after discharge during the spike of the externally applied component of membrane charge, along with its resting metabolically induced component. This effect also occurs at axon end-arborizations, and changes in synaptic conduction may be assigned to such effects of sufficient intensity as well as to polarization of post-synaptic dendrites in the same region. Under polarizing currents the changes in amplitude and reversals of sign of the cortical alpha rhythm, of evoked potentials from shocks to the optic nerve including the evoked alpha waves, evoked potentials from the superior colliculus, strychnine spikes, and paroxysmal spikes of cortex set up by repetitive stimulation or otherwise all may be accounted for by the interpretation offered for the effects of polarization at the geniculate.

The intracortical excitability cycle following stimulation of the optic pathway of the cat

Abstract Like many neurones of the central nervous system, those of the optic cortex of the cat exhibit a brief phase of facilitation, followed by a more prolonged phase of depression, as tested by a shock to the optic radiation following a conditioning shock. Recovery toward normal occurs gradually, and is complete in about 200 msec. The negative phase of the specific response returns only much later than the preceding portions of the response. This cortical depression is not complete, and varies in intensity in different preparations treated similarly. A weak first shock depresses the response to a second to some degree, and as the first shock is made stronger this effect is greater, in some cases to complete occlusion of the second response. The depression is accompanied by a surface-negative swing of the baseline. A more severe depression occurs in the dorsal nucleus of the lateral geniculate body, as previously described by Marshall. Its recovery of responsiveness is slower than is that of the cortex. Thus to optic nerve stimulation the cortical activity is affected by two conditions of depression operating serially. The depressive effect at the geniculate can be recognized in responses from the cortex in terms of the amplitude of the first spike representing responses of radiation axons. Two shocks, the second within the facilitatory period of the first, can break through the depression left by a previous response, when neither can do so alone. Similarly two such shocks may cause a larger single response than can any one stimulus. From these superimposed effects of facilitation and depression exhibited by arbitrary volley responses, certain speculations are offered as to the normal manner of cortical functioning.

ELECTROPHYSIOLOGICAL EVIDENCE OF A COLLATERAL PATHWAY FROM THE PYRAMIDAL TRACT TO THE THALAMUS IN THE CAT.

Abstract Stimulation in the ventralis lateralis nucleus of the thalamus in the adult cat produces an afferent evoked potential in sensorimotor cortex and a projected volley in the pyramidal tract. The latter portion of this tract response probably represents a true cortical relay. The latency of the earliest portion of the tract response is too long to be due to current spread to internal capsule, yet not long enough to be accounted for by a cortical relay. Analysis of this phenomenon leads to the hypothesis of a collateral pyramidal tract projection to ventralis lateralis.

Relations between specifically evoked and spontaneous activity of optic cortex.

Abstract In rabbits, the afterdischarge of an evoked response of optic cortex often has the form and dimensions of an alpha wave of the spontaneous activity. In cats, this is less frequent, and the afterdischarge may consist of one or several bursts of higher frequency activity. Each burst may then show the timing and envelope similar to the afterdischarge of those preparations showing single smooth waves, or other patterns between these extremes may occur. The series of small waves constituting a burst tends to show the same frequency and individual form of responses as does the previous and ensuing spontaneous activity, differing principally by an increase of amplitude. For 100 msec. after a maximal evoked response, during a negative deflection of the base line, no activity appears. This is also the period during which a second stimulus calls forth a lower, or no specific response, Thus after a first specific response, both its after-response and the specific response to a second shock show parallel courses of depression and recovery. The similarity in each preparation between spontaneous activity and aftereffect of response suggests that the latter consists of a return of the former, after its depression following the specific response. The increase of amplitude is interpreted as a result of partial synchronization of spontaneous activity previously firing more or less at random. The simultaneous depression at the instant of the specific response would cause summation of more units into a given interval and would account for increased amplitude. No increase above normal of the specific response to a second volley stimulus can be detected during this period. Speculations are raised as to what extent the mutual depression of one form of response by another indicates that the same neurones are active in each type of response. Some analogies are drawn between the spontaneous rhythmic activity of cortex and rhythmic muscular tremor.

Cortical Response to Stimulation of the Optic Nerve

In the cortex of the rabbit a small area can occasionally be found that is supplied by one artery and one vein. Such a region can thus be isolated by incisions except for the tongue of tissue where the vessels enter, without serious interference with the blood supply. A metal plate slipped under this tissue and connected to ground serves as an indifferent electrode, and the end of a fine wire resting on the cortical surface serves as a test electrode. When the region so isolated is not active no record is picked up from activity in the rest of the cortex. In 2 cases such a preparation has shown activity, and its pathway could be traced via the tongue of intact tissue. The activity consisted of a succession of wavelike action potentials, about 3 per second, lower than those of the adjacent intact cortex and much less complex. The rotating interrupter of the oscillograph apparatus was adjusted to a speed almost synchronous with these waves, but slightly slower, so that successive waves appeared to progress slowly across the screen. If a stimulus is sent in to the tissue at each revolution of the interrupter, it will fall later and later in successive waves. When stimulated during the negative phase or immediately afterward, no response is elicited, the response becoming larger the later the stimulus falls in the cycle. The response consists of a wave but little shorter than the spontaneous wave, and inhibits the following wave, in which case there is a compensatory pause, but it does not otherwise alter the rhythm which is imposed from without the circumscribed region. The response to stimulation is too protracted to be assignable to nerve fibers directly stimulated, and is presumably due to nerve cells.

Correlation between Threshold and Conduction Rate in Myelinated Nerves

In observing the effects of stimulating a nerve in the body, it is possible to lead off the cut end of the nerve into the oscillograph and thus correlate the potential form with the functional result of stimulation. Since the different fibers of a nerve are stimulated at different threshold strengths, if the difference in thresholds of different fibers were known, the oscillograph or other potential recorder could be dispensed with except for an occasional observation of threshold for the most irritable fibers, and the experimental procedure thus simplified. With this end in view, we have examined the ratios between the threshold of the first fibers stimulated in a nerve and that of other fibers, taking as criteria the thresholds of the first fibers in the various potential waves which represent fiber size groups. Gasser and Erlanger 1 have shown that the conduction rates of different fibers tend to vary as the fiber diameters, and by correlating the thresholds and conduction rates of different waves it should be possible to find the relationship between threshold and fiber size. For the potential waves of 4 bullfrog sciatic nerves, the averages of the ratios of the α/β and β/γ conduction rates are 1.64 and 1.60, respectively. The same ratios for similar nerves for which the data are obtained from the table in Gasser and Erlangers 2 paper are 1.62 and 1.60. This is approximately the ratio of fiber diameters shown in Gasser and Erlangers charts of size distribution. The β/δ ratios of conduction rate for 2 nerves available are 1.45 and 1.40. The ratio of average thresholds for β/α for the first 4 nerves above is 1.63, but the γ/β ratio is 2.3. No data are available for δ.

On the Mechanism of Spastic Vascular Disease.

Two opposing views exist as to the mechanism of spastic disease of blood vessels. Some regard it as an expression of dysfunction of the vaso-motor nerve supply to the vessels. Chief support for this conception is found in the symmetrical nature of the lesion and the paroxysms which characterize it. Others, especially Lewis and his coworkers, regard it as a local fault not primarily associated with abnormal innervation. This local defect as studied in Raynauds disease of extremities expresses itself in an abnormal response to cold, in the spatial order of development and disappearance of the vascular constriction, and in the failure of local anesthetization of the nerve supply to prevent or release completely an attack. Our own observations on these aspects of the disease lead us to support the contention of Lewis and others. Evidence is herein presented that the fault is a local one, and represents not a hyper function of a sympathetic innervation, but a change in the blood vessels, namely that they respond to epinephrine in a manner similar to tissues deprived of their sympathetic nerve supply (paradoxical response) while the nerve supply can be demonstrated to be functional. The evidence is derived from the study of 3 patients with vascular abnormalities of the upper extremities, 2 cases of Raynauds disease and one case of acro-asphyxia. The first 2 were subjected to the following tests: (1) 1 cc. or less (graded doses) of epinephrine hydrochloride (1-1000) was given hypodermically and the effect on the diseased and control extremities noted. An attack was invariably induced in the diseased extremities. Then the effect of intravenous glucose or of a meal rich in carbohydrate was noted. (2) Ten to IS units of insulin were administered to produce a physiological secretion of epinephrine. The effect on the diseased and control extremities was noted and again the effect of carbohydrate on the attack determined.

Positive Potentials Recorded from the Superior Colliculus.

A needle electrode thrust into an active nerve pathway may so kill nerve fibers in its vicinity that one obtains an effective dead-end lead, with a positive potential. When 2 electrodes are effectively on opposite sides of a region of synapses, the endings of fibers at synapses act similarly. In the superior colliculus of the cat one regularly finds that a probe electrode thrust downward encounters a shallow negative post-synaptic potential, following stimuli to the optic nerve, then a deeper positive potential, usually of considerably higher voltage. Records obtained as the electrode is withdrawn may show, on the other hand, a reversal of the initial upper negativity, giving a positive spike at any level. The potentials encountered in the colliculus may be explained by conventional physiological considerations without the inference of a relative positivity at an active region of a tissue. From the central region of synapses with optic tract fibers, in the colliculus, pathways radiate out in several directions. In the extreme case, of fibers conducting in all directions from a center, one can consider as an element the cone whose base is a small surface area, subtending a solid angle at the center. Since in the symmetrical case activity in elements adjacent to this will prevent current from flowing across the boundary of such a cone, it may be considered as if insulated in air, and for a small surface area, this element approaches a linearly arranged nerve fiber bundle. The critical feature of the solid figure to be retained is that the shunting material between the fibers will increase, and the resistance decrease, as the square of the distance from the center. The potential distribution along the surface of such an element will be triphasic, and the positive phase toward the center will be of higher amplitude, because less shunted, than the peripheral positivity. From electrodes placed at the center and at the periphery, therefore, an overall central positivity will be recorded, the more so as the peripheral electrode will usually be at some distance from the ends of active fibers. The negative spike recorded peripherally will be increasingly shunted, and therefore decrease in recorded amplitude, as the electrode is moved away from the central synaptic region, and the transition from negative to positive phases should be correspondingly abrupt at the central ends of the conducting elements.