A: receptive field map of a simple cell that was subsequently injected with Procion yellow. This cell was driven by stimuli to the left eye; it responded best to a narrow slit of light held in a 2-8 o'clock orientation and moved in either direction orthogonal to the slit (arrows). Stationary spots and slits of light also drove this unit. x, Areas giving “on” responses; Δ, areas giving “off” responses. The cell showed only 5 mV resting potential after penetrations, but it was successfully stained by passing a steady 1 × 10⁻⁸‐A current for 20 min. Large cross indicates the projection onto the visual field of the area centralis (a.c.) of the left eye. B: camera lucida drawing of the same cell, which lay in 2 adjacent coronal sections of the cortex. Drawings were made using bright‐field fluorescence illumination and a 100x oil immersion objective that allowed processes less than 1 μm in diameter to be observed. Most of the branches emerging from the cell body show occasional faint appendages that are likely to be dendritic spines. One descending process (α) is more uniform and is probably the axon.

Excitatory postsynaptic potentials recorded from inside a pyramidal tract cell in response to intracortical microstimulation (STIM) of upper layer III. A: superimposed responses. B: averaged responses after recording on a tape. Stimulating current, 3 μA; potassium citrate electrode; resting membrane potential, 50 mV. Latency of the response was 1.5 ms, indicating that this was a disynaptic connection.

Responses of a fast pyramidal tract cell to current steps. A: antidromic action potential elicited by stimulation of the pyramid at the moment indicated by an arrow. Resting potential, −72 mV; axonal conduction velocity, 31.5 ms. B: intracellular membrane potential (middle trace) and extracellular control record (bottom trace) during passage of depolarizing current step (uppermost trace) through microelectrode. C and D: potential changes during hyperpolarizing currents. E‐J: current step (upper trace) and membrane potential (lower trace) recorded at times indicated in seconds at the upward arrows. Note that peaks of spike potentials are off the records. Oblique arrows (E and J) show spontaneous synaptic potentials. Calibrations: 50 mV and 50 ms for A; 2 nA, 10 mV, and 100 ms for B‐D; 10 nA, 40 mV, and 0.5 s for E‐J. Records consist of approximately 10 superimposed traces for antidromic spike in A and single sweep traces in B‐J. Directions of membrane depolarization and of depolarizing current are indicated by positive signs in calibrations.

Responses of a slow pyramidal tract cell to current steps. A: antidromic action potential from pyramid. Axonal conduction velocity, 10.3 m/s. B‐G: membrane potentials (upper traces) during passage of current steps with different intensities (lower traces); time scale, 100 ms.

A: Photomicrograph of a 3‐barrel microelectrode. B: scheme of tip array of a 3‐barrel microelectrode and its relation to a recorded cell. I, tip of the electrode barrel I for intracellular recording; A, tip of electrode barrel A for extracellular recording in the immediate vicinity of the intracellular recorded cell (tip of A is 50-90 μm remote to the tip of barrel I); B, tip of electrode barrel B used as distant extracellular, reference electrode for barrel A (distance between tip barrel A and tip barrel B was 120-150 μm). C: IPSP field potential of a pyramidal tract cell; antidromic latency, 3.6 ms; depth, 1.3 mm. Single‐shock pyramidal tract stimulation. All traces are averaged responses (N = 120). In intracellular recordings, negativity is downward; in all other traces, negativity is upward. ECoG, electrocorticogram recorded at cortical surface (cf. potential in D, left side at 0 mm). Intracell, intracellular DC record of the cell response to pyramidal tract stimulation; an antidromic spike, which is followed by an IPSP, is elicited by the averaging procedure shown only as dots. Intracell hyp, same as intracell but recorded during hyperpolarizing current injection, the amplitude of the intracellular IPSP is now nearly zero (current trace not shown). ΔV_e (A-B), extracellular potential ΔV_e (A-B) (see text) occurring simultaneously with the intracellular record of full IPSP; AC record, time constant 8 s. ΔV_e (A-B) hyp, extracellular potential ΔV_e (A-B) occurring simultaneously with the intracellular record during injection of hyperpolarizing current. ΔV_e (A-B) difference, ΔV_e (A-B) hyp subtracted by computer from ΔV_e (A-B) during full IPSP; the potential slope occurring simultaneously with the intracellular IPSP (cf. intracell) represents the IPSP field potential. D: left, laminar ECoG potentials from different depths of the cortex evoked by pyramidal tract stimulation with a brief train of shocks. Negativity upward. Dashed line indicates the time when amplitudes were analyzed for the plot on the right of the figure. Right, plot of amplitudes of potentials on left at a time 16 ms after stimulus. A prominent positive amplitude peak is present at 1.2-1.6‐mm depth.

Fast prepotentials occurring in spontaneous discharges of hippocampal neurons. A and B: spikes are preceded by slow depolarizations (slow prepotentials) and are initiated by abrupt depolarizations called fast prepotentials (FPP's). Break between latter and the rest of the spike is indicated by arrow. B shows superimposed tracings to indicate that FPP's superimpose exactly but that slow prepotentials do not. C and D: isolated FPP's occurring spontaneously during passage of weak hyperpolarizing currents (ca. 1 to 5 × 10⁻¹⁰ A). Isolated prepotentials are indicated by large vertical arrows. Diagonal arrow indicates FPP associated with spike in D. No FPP was evident in rebound spike in C. E: scheme to account for data on FPP. Actual location of dendrite trigger zone is conjectural, but evidence does favor concept that dendritic membranes contain patches of membrane capable of active response genesis, with intervening zones of passive, high‐threshold, or graded‐response membrane.

Generation of dendritic spike by LOC (local) stimulus to cerebellar surface and by direct current injection. In A a small all‐or‐none spike was generated by a very weak parallel fiber stimulation. In B‐F the LOC stimulation increased in a graded fashion. In B the large intracellular depolarization was preceded by 3 all‐or‐none components. As the stimulus was increased from C to F the latency of the large depolarization and its duration decreased. In F the different all‐or‐none components (dots) fused into a sharp rising action potential. G‐I: intradendritic responses evoked by outward current pulses of different amplitudes.

Schematic representation of probable set of events following orthodromic and antidromic Purkinje cell activation and inhibition of Purkinje cell dendrites. A: after a slightly out‐of‐line parallel fiber (PF) activation, electroresponsive patches in dendrites (black area) generate full action potentials that are conducted first in a decremental manner and then in an electrotonic fashion to the next site of spike initiation (arrows); in this pseudosaltatory fashion dendritic spikes will be conducted to the large dendritic branches and finally to the somatic level. The diagram further illustrates the thesis that dendritically evoked action potentials do not invade antidromically other neighboring dendritic branches. Thus, after orthodromic activation in one extreme of a Purkinje cell tree, a great many of the remaining dendritic branches remain free from spike invasion. B: an antidromic activation of the same Purkinje cell is shown to invade (arrow) only the lower parts of the main dendritic branches; this is assumed to be the mechanism underlying the somatopetal tendency of dendritic spike propagation orthodromically. C: an idealized representation of the distribution of stellate cell (SC) inhibitory synapses on the dendrites of a Purkinje cell. Although the dendritic hyperpolarization would be electrotonically conducted to the rest of the dendritic tree (density of horizontal lines), the inhibitory action would be restricted, for the most part, to those dendritic branches receiving direct inhibitory input from a given inhibitory interneuron.

Field potentials produced in the chronically deafferented cerebellar cortex by an antidromic volley in Purkinje cell axons. A: potentials at the indicated depths below the cortical surface showing the typical positive‐negative spike and the slow positive wave at depths of 150-400 μm. B: spike potentials in another experiment at very fast sweep speed in order to allow accurate comparison of the rising phases and summits at the different depths. Vertical line at a latency of 0.35 ms shows the progressively delayed onset of the negative spike at more superficial levels. C: measurements from B are plotted to show no significant difference in time to summit (•) from depths of 600-250 μm and to show the progressive increase more superficially. On the other hand, there was a progressively longer latency from 600 μm to the surface when it was measured to a fixed voltage (a negative deflection of 0.15 mV) on the rising phase (○). Constant latency of the antidromic spike summit when set up by another juxtafastigial (J.F.) stimulus in this same experiment is shown by (+).

Field potentials generated by the antidromic propagation of an impulse into a Purkinje cell. Gray shading indicates depolarization. The darker the gray, the more intense the depolarization. The zones occupied by the impulse are shown in black. A single Purkinje cell only is shown in A‐C, but the lines of extracellular current flow are drawn confined to the immediate surround, as they would be in the situation where all Purkinje cells in an area are being simultaneously invaded. In A the antidromic impulse is propagating up the axon and there is a graded electrotonic depolarization of the soma and dendrites. In B the impulse has invaded the soma and dendrites to the maximum extent, there being a terminal dendritic zone not invaded, but merely deeply depolarized by the electrotonic currents as shown. In C the axon soma and the invaded part of the dendrites have almost completely recovered from the impulse, with the consequence that those regions are less depolarized than the uninvaded dendritic zone — hence the reversed current flow.

Spectrum of firing patterns recorded intracellularly in spontaneously active pyramidal cells. A: solitary spikes. B: brief bursts. C: moderate‐duration bursts. Hiatus in base line indicates silent period of 200 ms.

Sample records of intracellularly recorded olivary action potentials following cerebral cortex (CO) stimulation (A), cerebellar (CB) stimulation (antidromic; B and C), and direct intracellular (IC) stimulation by 0.2‐ms depolarizing current pulse (D). A‐C are traces of responses from the same unit. A and C, single traces; B; superimposition of 2 successive traces; and D, superimposition of 7 successive traces. Arrows in A and C point to an inflection on the delayed depolarization.

Dendritic and somatic EPSP in red nucleus cells. A‐C: intracellular EPSP's recorded from red nucleus neurons. The first EPSP (left arrow) is evoked by electrical stimulation of the contralateral interpositus nucleus, the second from a stimulus to the internal capsule (right arrow). Note the different time courses of the 2 synaptic potentials. B and C: membrane potential is hyperpolarized artifically, and an increase in the amplitude of the interpositus EPSP is observed, while the cortical EPSP is not changed. The plot to the right, a current‐voltage relation, shows changes in the amplitude of the somatic (○) and the dendritic (•) EPSP in mV (ordinate), while currents of different intensities were applied across the membrane (abcissa current in 10 × 10⁻⁹ A). Time and voltage calibration as indicated.

Excitatory response associated with increased FPP discharge and lacking detectable EPSP's. Central gray stimulation. A and B: subsequent tests in same cell, stimulation periods marked by bars. C: spontaneous discharge (spikes truncated) showing dissociation of FPP's from full spikes several minutes after records A and B.

Intracellular Purkinje cell records of a climbing fiber synaptic potential, evoked by juxtafastigial stimulation, modified by internally applied currents. An increase in the amplitude of the EPSP occurred when the hyperpolarizing (hyp.) steps were applied through the impaling microelectrode, while with depolarizing (dep.) currents there was a decrease and then reversal of the EPSP. The control synaptic potential is between the arrows. Note that the reflexly activated repetitive climbing fiber response of this cell (marked by the 3 arrows) was altered by the applied current in the same way as the directly evoked EPSP.

IPSP's in identified pyramidal cell neuron. A: spontaneous firing of pyramidal cell interrupted by a first weak shock (left arrow) to fornix, which produces small IPSP. Second shock (right arrow) is stronger and triggers an antidromic spike and larger and longer‐lasting IPSP. B: inversion of hyperpolarizing to depolarizing PSP with KC1 electrode. B1 indicates initial PSP before inversion (lower trace) and final PSP after inversion (upper trace). B2 shows sequence of inversion with superimposed fine‐line tracings. The asymmetric time courses of pre‐ and post‐inversion recordings may reflect occult early excitatory components or spatially distributed inhibitory input.

A‐K: cerebellar‐evoked monosynaptic inhibition of Deiters' neurons. Arrangement of stimulating concentric electrodes on a sagittal plane through the cerebellar vermis is shown. Roman numerals I‐V refer to Larsell's 190 lobular divisions; fp, fissura prima; arrows indicate the interlobular sulci. A‐G, H‐J, and K, intracellular recording from three Deiters' neurons respectively. In A‐G, 0.2‐ms pulse stimuli were applied to the vermal cortex of lobule IV at intensities of 1.9 V (A), 2.1 V (B), 3.2 V (C), 5 V (D), 10 V (E), 20 V (F), and 30 V (G). Note the different voltage scales for A‐C (2 mV) and D‐G (5 mV). Dotted lines in F and G indicate the possible time course of the monosynaptic IPSP's if they were similar to that in D. Vertical arrows indicate the diverging points between the dotted lines and actual potential curves, H, initial part of the potential changes following 30‐V stimulation of lobule III; α and β denote the 2 spiky peaks of the field potential. I, same as in H but reversed during passage of hyperpolarizing currents of 2 × 10⁻⁸ A through the KCl‐filled microelectrode. Dotted line indicates the time course of the potential curve of H. Arrow marks the diverging point of the 2 curves before and during current passage. J, extracellular record taken just after withdrawal; K shows spontaneous spike discharges from a Deiters' neuron and its suppression and later facilitation by the cerebellar stimulation. At the time indicated by arrow 0.2‐ms pulses of 30 V were applied to the ipsilateral lobule III. Time constant of the recording amplifier was 200 ms for A‐J, while there was DC recording in K. The records in this figure were formed by 10-40 superimposed traces. Note that K in this figure was recorded in a single sweep.

A and B: intracellularly recorded IPSP and EPSP from a Purkinje cell in response to local stimulation; C, inhibition of the spontaneous discharges of 2 Purkinje cells caused by a single local stimulus; D and E, extracellular potentials evoked by local stimulation of 30 and 50 V, respectively, and recorded at the indicated depths below the surface of the folium; F, plot of measurements from D and E; G, Purkinje cell with current flow indicated by arrows.

Ultrastructure of the basket synapse. Inset, upper left, situation on light‐microscopic level. The greater part of the Purkinje cell body (P) and dendrite surface is covered by processes of Bergmann glia (G). The terminal branches of basket axons (Ba) have synaptic contacts with the bottom of the Purkinje cell body and the “preaxon.” The real axon (Ax) begins only some 30 μm (or more) deeper and soon becomes myelinated. Finger‐shaped processes of basket axon endings and similar processes of the Bergmann glia are entangled in a relatively loose “axon‐cap neuropile” or “outer basket neuropile” (OBN), which is devoid of synaptic contacts.

Disfacilitation produced in red nucleus by stimulating of the cerebellar cortex. B: control (Con) intracellular records showing EPSP evoked by activation of interpositus nerve cells. Activation followed by a large hyperpolarization produced by the removel of background synaptic bombardment from the same interpositus neurons. In records, A, C, and D, the membrane potential is altered by the application of extrinsic current, and, as seen in record D, both initial EPSP, as well as the long‐lasting hyperpolarization that followed, have the same euilirbrium potential. Time and voltage calibration as illustrated.

Evidence for electrotonic coupling obtained by antidromic stimulations. High‐gain recordings. A₁: top to bottom are successive traces obtained with increasing spinal stimuli. As the stimulus strength is increased from 0.85 to 0.98 T (threshold), a graded depolarization increases to a peak amplitude of about 1.5 mV; a 1.0‐T stimulus evokes an antidromic spike. B₁‐C₁: records from 2 other neurons. In each record, 5 successive sweeps have been superimposed; graded depolarizations up to 2.2 mV and 1.6 mV, respectively, are evoked by stimuli of increasing strength, the largest of which induces a spike. The second depolarizing component in the lower trace of each record presumably represents delayed excitation in nearby cells. A₂, B₂, C₂: extracellular field potentials generated by near‐threshold stimuli outside the cells recorded in A₁, B₁, and C₁, respectively.

Intracellular responses of a neuron and a glial cell during interictal discharge. A: intracellular recordings from a neuron and glial cell penetrated successively 100 μm apart. Both intracellular recordings occurred synchronously with the surface discharge. The “field” record is an extracellular potential recorded just below the neuron and above the glia. B: same glial cell recording as in A but at a lower gain and slower sweep speed. Note the marked difference in time course of the depolarizations in the 2 cells. Positive is upward in all traces.

Schematic representation of one possible process that could result in the observed increase in the number of axon terminals making multiple synaptic contacts. The process illustrated is that of collateral reinnervation of denervated sites by adjacent nondegenerating terminals. This is considered by Raisman and Field as the most likely mechanism operative in their studies. S, spine; D, degenerating terminal of cut axon; N, remaining axon terminal.

Postulated action of impulses in a small bundle of mossy fibers (MF) innervating a focus of granule cells (GrC), as shown in the transverse section adjacent to the larger diagram, where the folium is seen from above. GoC, Golgi cell distributed to that focus. Parallel fibers (PF) are shown in the cross section.

Synaptic connections in the vestibulo‐ocular reflex arc and cerebellar flocculus. Excitatory neurons are indicated as hollow structures, and inhibitory ones are filled in black; gr, granule cells; mf, mossy fibers; cf, climbing fibers.

A: estimates, from human subjects of the locus of the sensation produced by brief shearing motions of single‐ and triple‐probe arrays. A random sequence of single‐ and triple‐probe stimuli was presented at several positions on the middle dorsal forearm. Each bar represents the mean probability that the subject will point to a 10‐mm transverse band of skin on the forearm without visual cues. Negative distances on the abscissa refer to proximal bands on the forearm and positive distances to distal bands. The single‐probe localization histogram was constructed from 193 observations of 7 subjects; responses to the proximal, middle, and distal probes were pooled. The 3‐probe localization histogram shows means of 103 observations of the same 7 subjects. The majority of trials were done with a 210‐μm shear, but some observations were obtained with stimuli of 35-400 μm. Subjects reported feeling a single, distinct phasic stimulus when the triple‐probe array was presented. They localized the sensation produced by the 3‐probe array under the middle probe almost as well as they localized the single‐probe shearing stimulus. B: comparison of the reconstructed population activity evoked in cortical hair units in the cat primary somatic receiving area by separate (dashed lines) and simultaneous (solid lines) actuation of 3 probes 15 mm apart on the foreleg. Each dashed profile represents the cortical population activated by the appropriate single probe in terms of the distance of each unit's receptive‐field center from the stimulus probe. Ordinate plots represent activity in each set of units as a fraction of the activity produced in the neurons whose field center lies directly beneath the active probe. Solid profile represents the active cortical population when all 3 probes are actuated simultaneously. •, Mean activity in each set of units, normalized as a fraction of the activity produced in units whose field centers lie beneath the middle probe of the 3‐probe array. To present both the single‐and triple‐probe data on the same ordinate, the mean ratio of the activity produced by the 3‐probe array with the middle probe on the field center to the activity produced by a single probe on the field center has been calculated. The scaling factor of 1.3 was obtained using the formula [(M:S) + (M:L) (L:S)]/2. It is believed that low‐pass spatial frequency filtering plays an important role in this transformation, although subsequent high‐pass filtering may also shape these contours.

A: responses to stimulation of the deafferented fornix at progressively increasing stimulus intensities. Note the appearance of recurrent inhibition followed by late excitation with repetitive firing. B: effect of ip injection of 15 mg/kg sodium pentobarbital. Average evoked potential (AEP), labeled 0, is taken just before injection. Succeeding AEP's recorded 2, 5, 9, 15, 18, and 20 min after injection. For all AEP's, stimulus was 4 V, 0.01 ms; for each AEP, 100 transients were averaged. C: set of poststimulus time histograms for hippacampal unit (decreasing stimulus intensity). D: neural network for associated wave forms. Pyramidal cells in forward branch send efferent fibers (f₀) over fornix and collaterals (p_e) to interneurons in feedback branch. Fibers from interneurons (i_a) in turn inhibit pyramidal cells. One input to the pyramidal cell population consists of afferent fibers (f_i) in the fornix. Excitatory connections represented by +, inhibitory connections by −.

Extra‐ and intracellular records from cells in the ventrobasal complex of the thalamus in response to stimulation of various contralateral foreleg nerves (left column) and of the foreleg area of the ipsilateral somatosensory cortex (right column). Each horizontal pair of records obtained from the same site (A-B) or cell (E‐J). A‐D, extracellular records. A: initial discharge followed by a large positive wave (P wave) then first and second burst response in response to single shock to ulnar nerve. B: as in A, but stimulation of cortex, activating the thalamic neurons antidromically. C and D: response of a single thalamic relay cell to the same stimuli as in A and B; same cell discharges both in the initial response and in the first burst response. E‐J, intracellular records from thalamic somaesthetic relay cells recorded with potassium citrate‐filled electrodes. E and F: large inhibitory postsynaptic potential in response to stimulation of median nerve (M) and cortex, respectively. Lower records are cuneate responses. G and H: large and rhythmic inhibitory postsynaptic potential induced by ulnar nerve (U) and cortical stimulation. I and J: slow sweeps illustrate the rhythmic nature of the inhibitory postsynaptic potential evoked both by a single superficial radial nerve (SR) and by an antidromic volley. K shows postulated circuitry to account for these data: recurrent inhibition of relay cells mediated by inhibitory interneuron (I CELL) phases rhythmic activity of relay cells.

Genesis of interictal spikes in epileptogenic foci. A, neocortex; B‐F, hippocampus. A: intracellular recording of paroxysmal depolarizing shift (PDS) from pyramidal cell of neocortex in penicillin‐treated interictal spiking motor cortex focus in cat. B: penicillin discharges triggered in fornix stimuli and recorded from the hippocampal surface. All‐or‐none discharges triggered at decreasing latencies by stimuli to the fornix of increasing strengths (indicated at left); 3 penicillin discharges triggered by identical successive fornix stimuli illustrate the jitter in latency of the all‐or‐none triggered discharges. C‐E, intracellular response of pyramidal cell during fornix‐triggered penicillin discharge and underlying evoked response when stimulus was subthreshold for triggering penicillin discharge. C: antidromic spike, small depolarization, and large depolarizing potential that begins with an action potential recorded when the fornix stimulus triggered a discharge. D: identical response during discharge superimposed with the evoked response alone that was recorded when a similar stimulus failed to trigger the penicillin discharge. Both responses are identical until the beginning of the large triggered depolarizing potential. Note that the potential change immediately after the antidromic spike is a small depolarization followed by an isoelectric base line when no triggering occurs, rather than by the normal large hyperpolarizing IPSP. E: extracellular field recorded outside the cell during a triggered discharge. F: sum of the PSP's of 4 cells in the (computer) model array, used as an approximation for the “surface gross recording,” against time, in “millisecond” units, for 4 different initial activating conditions. ○, Response to activating one cell in array; ▪, response to activating 2 cells in array; •, response to activating 4 cells in array; ▴, response to activating 5 cells in array. Note the graded “evoked response” component of each curve at the arrow and the all‐or‐none paroxysmal spike discharges, which occur in response to 2 or more stimuli with a shifting peak latency. These curves are considered analogous to the change in latency with change in stimulus strength of the all‐or‐none penicillin discharge. Amplitude units are arbitrary.

A: experimental arrangement for study of the cerebellopontine reverberating circuit. Stimulation was applied to the branchium pontis (BP), the nucleus interpositus (IP), the red nucleus (RN), and the ventrolateral nucleus (VL) of the thalamus. Intra‐ and extracellular recording was made from IP, RN, and the nucleus reticularis tegmenti pontis (PN). B and C, intracellular recording from RN cell. D and E, intra‐ and extracellular recording from IP cells. F and G, intracellular recording from RN cells. B: EPSP induced in a RN cell by stimulation of IP. C: EPSP produced in the same cell as B by stimulation of the RN. D: upper trace, antidromic activation of an IP neuron from VL; lower trace, orthodromic activation of the same IP neuron by BP stimulation. E, upper trace, intracellular potentials generated by an IP cell as a result of BP stimulation; middle trace, extracellular potential generated just outside the same cell by the same stimulus; lower trace, superimposed tracings of intra‐ and extracellular potentials. The initial depolarization of cell membrane was followed by membrane hyperpolarization. F: intracellular potentials produced in an RN cell by BP stimulation. G: EPSP induced in an RN cell by PN stimulation.

Picrotoxin‐induced depolarization of red nucleus (RN) cells. A‐C: EPSP's and spike potentials induced in an RN cell by stimulation of the ventrolateral nucleus (VL). D: picrotoxin‐induced depolarization and spike potentials of an RN cell evoked by stimulation of the branchium pontis (BP) of increasing stimulus intensities of 2, 2.8, 3, 4, and 5 V, respectively, from 1 to 5. This record was taken in a preparation where the spinal cord and inferior cerebellar peduncle were sectioned in addition to decerebration at A7 in stereotaxic coordinates. E and F, ripples of the membrane potentials superimposed on the picrotoxin‐induced depolarization of evoked EPSP's; RN recordings. E: intracellular potential of a nucleus reticularis tegmenti pontis (PN) cell induced by VL stimulation. Evoked EPSP is followed by additional depolarization. Downward arrows indicate the peak of the depolarization. F: intracellular potential of an RN cell induced by BP stimulation. Initial EPSP is followed by the picrotoxin‐induced depolarization. Superimposed on the rising phase of depolarization, small wavelets of depolarization are noticed as indicated by downward arrows.

Reconstruction of electrode tracks and cell locations within the monkey motor cortex. Electrode penetrations (solid lines, identified by numbers) pass through efferent zones projecting to various thumb muscles. Peripheral motor effects produced by intracortical microstimulation (≤ 5 μA) are indicated by symbols explained in the figure. Cortical spots where 5 μA intracortical microstimulation (ICMS) did not produce motor effects are shown by small solid lines perpendicular to the tracks. Positions of cells encountered are connected by dashed lines to figurines and descriptions that explain receptive fields and adequate stimuli. Spontaneously active cells not driven by peripheral stimulation are indicated as UD. EXT, extension; FLEX, flexion; ADD, adduction; ABD, abduction.

Phasic on‐off activation of a type II neuron elicited from large receptive field. Left: stimuli used for records on right side of the figure. Right: averages of 10 successive reactions. F3, stimuli 1-9 at 20% of the intensity of stimuli A‐D (FO). Bottom record above photocell: extracellular control. Note the same type of reaction to most stimuli with relatively little quantitative differences indicating virtually homogeneous receptive field.

Response of simple cortical cells to a sinusoidal grating, as a function of spatial frequency; amplitude of the averaged responses vs. the passage of a single period of the grating over the cell receptive field. Maximal amplitude of response of each cell has been taken as 1. Average luminance of the grating, 2 cd/m², contrast 20%. The grating velocity was constant and of the order of 1-2°/s; it was chosen to maximize the cell response. •, Unit with an on‐center region of 1.2° width, flanked by 2 off regions. Total width of the receptive field 5°. ○, Unit with a receptive field of the same type as the previous one, but with very weak off flanks. On‐center region of 1° width, ▴, Unit with a bipartite receptive field consisting of an on region of 1.3° flanked by one off region of 0.7° width. Δ, Unit with a bipartite receptive field of total width 0.9°. •. Unit with a bipartite receptive field of total width 0.6°.