Responses of small- and large-field bipolar cells to GABA and glycine

Morphologically distinct subtypes of retinal bipolar cells transmit information along parallel pathways to convey different aspects of the visual scene, but the synaptic mechanisms that regulate signal transmission are largely unknown. The all-rod retina of skate provides a comparatively simple system in which to correlate bipolar cell morphology with responses to the inhibitory neurotransmitters GABA and glycine. Two subtypes of bipolar cells can be identified when isolated in culture: large-field bipolar cells with extensive dendritic arbors, and small-field bipolar cells with one or two dendritic branches. Under voltage-clamp, glycine elicited significant current responses from small-field cells, but not from large-field bipolar cells. Although all bipolar cells displayed GABA-activated chloride currents mediated by both GABA(A) and GABA(C) receptors, the small-field bipolar cells showed a significantly greater contribution from GABA(A) receptors. The results of the present study reveal for the first time that the relative expression of the two classes of GABA receptor on each bipolar cell type correlates with cell morphology and the presence of the glycine receptor.

Excitatory amino acids and serotonin uptake blockers reveal two physiologically distinct serotonin systems in the retina of the skate, Raja erinacea

The retina of the skate (Raja erinacea) contains at least 2 types of cell (amacrines and bipolars) that can be visualized with an antiserum against serotonin. We have employed serotonin immunocytochemistry in combination with pharmacological manipulation of retinal tissue to analyze physiological properties of serotonergic amacrine cells and serotonin-accumulating bipolar cells. Excitatory amino acids (NMDA, aspartate) had no detectable effects on serotonin-immunoreactivity in bipolar cells but decreased staining in amacrine cells. High K+ Ringer increased staining in bipolar cell somata, however, it depleted the inner plexiform layer of the retina of serotonin. Zimelidine, a serotonin uptake inhibitor, completely blocked serotonin accumulation by bipolar cells but had no effect on amacrine cells, whereas incubation of the retinas in fluoxetine (Prozac), a different inhibitor of serotonin uptake, did not block serotonin accumulation into bipolar cells which was actually enhanced in some cases. We conclude that amacrine and bipolar cells of the skate retina employ two different serotonin uptake carrier systems, thus generating two distinct pharmacological components that are capable of interacting with each other as they compete for extracellular serotonin. Similar mechanisms may exist in the vertebrate CNS and further examination of the interaction of these systems could provide important insights into the action and possible side effects of serotonin-related drugs.

GABA receptors of bipolar cells from the skate retina: actions of zinc on GABA-mediated membrane currents

GABA receptors of bipolar cells from the skate retina: actions of zinc on GABA-mediated membrane currents. J. Neurophysiol. 78: 2402-2412, 1997. gamma-Aminobutyric acid (GABA)-induced currents were recorded from isolated bipolar cells of the skate retina using perforated patch-clamp methodology. Pharmacological analysis of the responses, using selective agonists and antagonists of the major classes of GABA receptor, revealed the presence of both GABAA and GABAC receptors at both the dendrites and axon terminals of the bipolar cells. The two receptor types showed very different reactions to zinc, a divalent metallic cation that was detected in the synaptic terminal region of skate photoreceptors. Currents mediated by the activation of GABAC receptors were down-regulated by zinc, a feature that is typical of the action of zinc on GABAC receptors. On the other hand, the effects of zinc on GABAA receptor-mediated activity was highly dependent on zinc concentration. Unlike the GABAA receptors on other neurons, responses mediated by activation of the GABAA receptor of skate bipolar cells were significantly enhanced by zinc concentrations in the range of 0. 1-100 mu M; at higher concentrations of zinc (>100 mu M), response amplitudes were suppressed below control levels. The enhancement of GABAA receptor activity on skate bipolar cells showed little voltage dependence, suggesting that zinc is acting on the extracellular domain of the GABAA receptor. In the presence of 10 mu M zinc, the dose-response curve for 4,5,6, 7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP; a GABAA agonist that suppresses GABAC-activated currents) was shifted to the left of the curve obtained in the absence of zinc, but without a significant change in the response maximum. This finding indicates that the enhancing effect of zinc is due primarily to its ability to increase the sensitivity of the GABAA receptor. The novel enhancement of neuronal GABAA receptor activity by zinc, observed previously in the GABAA-mediated responses of skate Müller (glial) cells, may reflect the presence of a unique subtype of GABAA receptor on the bipolar and Müller cells of the skate retina.

Two classes of bipolar cell in the retina of the skate Raja erinacea

We have used immunoreactions against serotonin and protein kinase C to visualize two distinct classes of bipolar cell in the all-rod retina of the skate, Raja erinacea. To enhance the immunoreaction in serotonin-accumulating bipolar cells, prior to fixation, some retinas were incubated in Ringer's solution containing serotonin and pargyline. We found the somata of serotonin-accumulating bipolar cells to be located slightly distal to the midline of the inner nuclear layer. With increasing eccentricity from the visual streak, the size of the perikarya increases, concomitant with a decline in density of their distribution. Dendrites emanate from stout primary stalks and branch out before reaching the outer plexiform layer. Axons are bistratified within the inner plexiform layer with ramifications at the border of strata 1 and 2 and in stratum 4. The overall morphology of serotonin-accumulating bipolar cells is similar to that of serotonin-accumulating OFF bipolar cells of other non-mammalian vertebrates. Protein kinase C immunoreactive cells display the typical appearance of rod bipolar cells. Somata of protein kinase C immunoreactive bipolar cells are spindle-shaped and located distal to the serotonin-accumulating bipolar cells. Dendrites of these bipolars do not ramify before reaching the outer plexiform layer. Thin axons of protein kinase C immunoreactive bipolar cells end in large, club-shaped terminals in stratum 5 of the inner plexiform layer, bearing a striking similarity to axon terminals of mammalian ON rod bipolar cells. Our findings suggest that the all-rod retina of the skate contains at least two distinct vertical pathways including an OFF bipolar cell pathway in addition to a classical rod ON bipolar pathway.

Zinc enhances ionic currents induced in skate Müller (glial) cells by the inhibitory neurotransmitter GABA

We describe here a novel effect of zinc on GABA receptors of glial cells in the skate retina. The GABA-induced currents of skate Müller cells, the radial glia of the retina, are mediated by activation of GABA(A) receptors (GABAARS). In other parts of the nervous system, GABz*-)ediated currents are inhibited by zinc. However, in isolated, voltage-clamped Müller cells, coapplication of zinc (10 microM) and GABA (1 microM) resulted in enhancement of the GABA(A)R current. Surprisingly, zinc alone induced a current similar in many respects to that elicited by GABA, i.e. the reversal potential was the same as for the GABA-induced current, the current was blocked by bicuculline and picrotoxin, and the current-voltage relation obtained in the presence of 10 microM zinc was virtually identical to that obtained with 1 microM GABA. Both bicuculline and picrotoxin suppressed a current that was present with cells bathed only in Ringer, suggesting that some of the GABA channels were spontaneously open in the absence of externally applied GABA. This possibility was supported by cell-attached patch recordings. Under conditions in which potassium and calcium currents were suppressed, spontaneous channel activity was observed. Moreover, the frequency of these channel openings was greater when zinc was included in the pipette solution, and reduced when bicuculline was added. These findings suggest that zinc acts directly to enhance the GABA(A) receptor activity of the Müller cells, and raise the possibility that the subunit composition of the GABA(A)Rs of skate Müller cells differs from that of GABA(A)Rs identified previously in other neuronal and glial preparations.

Pharmacology of the skate electroretinogram indicates independent ON and OFF bipolar cell pathways

Organization of afferent information into parallel ON and OFF pathways is a critical feature of the vertebrate visual system. All afferent visual information in the vertebrate retina reaches the inner plexiform layer (IPL) via bipolar cells. It is at the bipolar cell level that separation of ON and OFF information first appears for afferent information from cones. This may also hold true for the rod pathway of cold-blooded vertebrates, but not for mammals. The all-rod retina of the skate presents an opportunity to examine such pathways in a retina having but a single class of photoreceptor. Immunocytochemical evidence suggests that both ON and OFF bipolar cells are present in the skate retina. We examined the pharmacology of the skate electroretinogram (ERG) to test the hypothesis that independent ON and OFF bipolar cell pathways are functional as rod afferent pathways from outer to inner plexiform layer in the skate. 100 microM 2-amino-4-phosphonobutyric acid (APB) reversibly blocked the skate ERG b-wave. A small d-wave-like OFF component of the ERG revealed by DC recording of response to a prolonged (10 s) flash of light was reduced or blocked by 5 mM kynurenic acid (KYN). We found that addition of 200 microM picrotoxin to the Ringer's solution revealed prominent ON and OFF components of the skate ERG while reducing the c-wave. These ON and OFF components were reversibly blocked by 100 microM APB and 5 mM KYN, respectively. Reversible block of the OFF component by KYN was also accomplished in the presence of 500 microM N-methyl-DL-aspartate. From these findings, we conclude that ON and OFF bipolar cells are likely to be functional as parallel afferent interplexiform pathways in the all-rod retina of the skate.

Ultrastructural and electrophysiological changes associated with K(+)-evoked release of neurotransmitter at the synaptic terminals of skate photoreceptors

Bathing the skate retina in a Ringer solution containing a high concentration (100 mM) of potassium ions depolarized the visual cells, depleted the receptor terminals of synaptic vesicles, and suppressed completely the b-wave of the ERG and the intracellularly recorded response of horizontal cells (the S-potential). The depletion of synaptic vesicles was accompanied by a large increase in the extent of the plasma membrane resulting in distortion of the normal terminal profile, i.e. distension of the basal surface and elaborate infolding of protoplasmic extensions. Morphometric analysis showed that despite the changes in vesicle content and terminal structure, the combined linear extent of the vesicular and plasma membranes was unchanged from control (superfusion with normal Ringer solution); the increase in plasma membrane was equivalent to the observed loss of vesicular membrane. When returned to a normal Ringer solution, the terminals rapidly began to reform, and in about 10 min they were morphologically indistinguishable from receptor terminals seen in control preparations. After 30 min in the normal Ringer solution, the amount of membrane associated with the vesicles and the plasma membrane had reverted to control values, and once again the total membrane estimated morphometrically remained essentially the same. Thus, there is an efficient mechanism at the photoreceptor terminal for the recycling of vesicle membrane following exocytosis. The K(+)-induced depletion of synaptic vesicles was paralleled by a precipitous loss of responsivity in both the b-wave of the ERG and the S-potential of the horizontal cells. However, after 30-min exposure to the high K+ and a return to normal Ringer solution, the recovery of electrophysiological activity followed a much slower time course from that associated with the structural changes; 60 min or longer were required for the potentials to exhibit maximum response amplitudes. It appears that the rate-limiting step in restoring normal synaptic function following massive depletion of vesicular stores is transmitter resynthesis and vesicle loading rather than vesicle recycling.

Sensitivity transformation for vertebrate vision

The visual response to a flash given in the dark is known to saturate according to the Michaelis-Menten relationship. Nevertheless, the incremental response from increasing levels of mean luminance tends to follow a Weber-Fechner relationship well into the saturation range determined from the Michaelis-Menten results. This sensitivity transformation from Michaelis-Menten to Weber-Fechner is an important characteristic of light adaptation in the vertebrate retina. Recent studies concerning the role of calcium in photoreceptor adaptation have shown that the relaxation from peak to plateau in the response of isolated photoreceptors was absent under conditions in which adaptation was blocked. Comparing the pronounced relaxation from peak to plateau in turtle horizontal cells with the absence of such relaxation in the catfish response, we noted also that turtle incremental sensitivity shows a Weber-Fechner relationship while catfish incremental sensitivity more closely follows the local slope of the Michaelis-Menten relation. Based on these observations, we have obtained an expression to relate the relaxation from peak to plateau with the sensitivity transformation. We assume that adaptation shifts the half-maximum point of the Michaelis-Menten curve so that the light response relaxes to a plateau value equal to a specified fraction phi of the peak response. We show that this manipulation alone results in a transformation from Michaelis-Menten kinetics to Weber-Fechner sensitivity.

Membrane current responses of skate photoreceptors

Light-evoked membrane currents were recorded with suction electrodes from the outer segments of individual photoreceptors enzymatically dissociated from the skate retina. The intensity-response relation of dark-adapted cells closely followed a Michaelis function for which a half-saturating response was elicited by a flash intensity that produced about 36 photoisomerizations. Dim-light responses, as well as the early rising phase of the responses to a wide range of flash intensities, could be described by a reaction scheme that involved a series of four first-order delay stages. The number of delay stages required to model the rising phase of the photocurrents did not change in light adaptation. However, background illumination that reduced sensitivity by 1.5 log units, or a bleaching exposure that resulted in a nearly equivalent desensitization, shortened significantly the time scale of the responses. In both instances there were two- to threefold increases in the rate constants of the transitional delays, and almost complete suppression of the tail current that characterized the response of the dark-adapted cell. These findings suggest that although light adaptation alters the gain and kinetics of the transduction mechanism, the nature of the intervening processes is the same in dark- and light-adapted photoreceptors. Moreover, the results show clearly that there is no need to postulate the existence of a second class of cone-like rods to account for the remarkable ability of skate photoreceptors to respond to incremental stimuli presented on "saturating" background fields or after exposure to an intense bleaching light.

Dynamics of skate horizontal cells

The all-rod retina of the skate (Raja erinacea or R. oscellata) is known to have the remarkable capability of responding to incremental flashes superimposed on background intensities that initially block all light-evoked responses and are well above the level at which rods saturate in mixed rod/cone retinas. To examine further the unusual properties of the skate visual system, we have analyzed responses of their horizontal cells to intensity-modulated step, sinusoidal, and white-noise stimuli. We found that during exposures to mean intensities bright enough to block responses to incremental stimuli, decremental stimuli were also initially blocked. Thereafter, the horizontal cells underwent a slow recovery phase during which there was marked nonlinearity in their response properties. The cell first (within 2-3 min) responded to decrements in intensity and later (after greater than 10 min) became responsive to incremental stimuli. After adaptation to a steady state, however, the responses to intensity modulation were nearly linear over a broad range of modulation depths even at the brightest mean levels of illumination. Indeed, examination of the steady-state responses over a 5-log-unit range of mean intensities revealed that the amplitude of the white noise-evoked responses depended solely on contrast, and was independent of the retinal irradiance as the latter was increased from 0.02 to 20 muW/cm2; i.e., contrast sensitivity remained unchanged over this 1,000-fold increase in mean irradiance. A decrement from the mean as brief as 2 s, however, disturbed the steady state. Another unexpected finding in this all-rod retina concerns surround-enhancement, a phenomenon observed previously for cone-mediated responses of horizontal cells in the retinas of turtle and catfish. While exposure to annular illumination induced response compression and a pronounced sensitivity loss in response to incremental light flashes delivered to the dark central region, the cell's sensitivity showed a significant increase when tested with a white noise or sinusoidally modulated central spot. Unlike horizontal cells in other retinas studied thus far, however, response dynamics remained unchanged. Responses evoked either by a small spot (0.25-mm diam) or by a large field light covering the entire retina were almost identical in time course. This is in contrast with past findings from cone-driven horizontal cells whose response waveform (dynamics) was dependent upon the size of the retinal area stimulated.

Dynamics of turtle cones

The response dynamics of turtle photoreceptors (cones) were studied by the cross-correlation method using a white-noise-modulated light stimulus. Incremental responses were characterized by the kernels. White-noise-evoked responses with a peak-to-peak excursion of greater than 5 mV were linear, with mean square errors of approximately 8%, a degree of linearity comparable to the horizontal cell responses. Both a spot (0.17 mm diam) and a large field of light produced almost identical kernels. The amplitudes of receptor kernels obtained at various mean irradiances fitted approximately the Weber-Fechner relationship and the mean levels controlled both the amplitude and the response dynamics; kernels were slow and monophasic at low mean irradiance and were fast and biphasic at high mean irradiance. This is a parametric change and is a piecewise linearization. Horizontal cell kernels evoked by the small spot of light were monophasic and slower than the receptor kernels produced by the same stimulus. Larger spots of light or a steady annular illumination transformed the slow horizontal cell kernel into a fast kernel similar to those of the receptors. The slowing down of the kernel waveform was modeled by a simple low-pass circuit and the presumed feedback from horizontal cells onto cones did not appear to play a major role.

Dynamics of turtle horizontal cell response

The small- and large-field (cone) horizontal cells produce similar dynamic responses to a stimulus whose mean luminance is modulated by a white-noise signal. Nonlinear components increase with an increase in the mean luminance and may produce a mean square error (MSE) of up to 15%. Increases in the mean luminance of the field stimulus bring about three major changes: the incremental sensitivity defined by the amplitude of the kernels decreases in a Weber-Fechner fashion; the waveforms of the kernels are transformed from monophasic (integrating) to biphasic (differentiating); the peak response time of the kernels becomes shorter and the cells respond to much higher-frequency inputs. The dynamics of the horizontal cell response also depend on the area of the retina stimulated. Smaller spots of light produce monophasic kernels of a longer peak response time. The presence of a steady background produces three major changes in the spot kernels: the kernel's amplitude becomes larger (incremental sensitivity increases); the peak response times become shorter; the waveform of the kernels changes in a fashion similar to that observed with an increase in the mean luminance of the field stimulus. A similar enhancement in the incremental sensitivity by a steady background has also been observed in catfish, which shows that this phenomenon is a common feature of the horizontal cells in the lower vertebrate retina.

Turtle and catfish horizontal cells show different dynamic response

Horizontal cell responses of catfish and turtle have been found to differ in a characteristic way. These characteristics established by white-noise analysis show that the impulse response (first order Wiener kernel) of the catfish horizontal cell has a substantially shorter latency as well as peak response time than that of the turtle. The turtle horizontal cell, on the other hand, has a dynamic gain which is twice that of the catfish. Since these differences were established under conditions of ambient illumination, they may be expected to be functionally important in the visual experience of the animals under normal environmental conditions.

Synaptic feedback onto photoreceptors in the ocellar retina

Intracellular studies from photoreceptors and second order neurons in the dragonfly ocellar retina suggest that the hyperpolarizing OFF oscillation in the photoreceptor reflects synaptic feedback from second order neurons onto receptor terminals. The receptor OFF response was normally observed when recording more proximally, closer to the nuclear and synaptic region, but it was not seen when recording more distally, closer to the rhabdomeric end of the cell. Both the hyperpolarizing OFF response in the receptor and the depolarizing OFF response in the second order cell are apparently generated in the ocellar plexiform layer because they were not eliminated when the second order processes were isolated from the brain. In both intact and cut nerve preparations, the receptor OFF response was blocked by curare and enhanced by picrotoxin, the same drugs that were reported to selectively modify the response of the second order cell. In addition, a normal-appearing OFF response was recorded intracellularly from the dark-adapted photoreceptor in response to the application of brief hyperpolarizing current pulses to the ocellar nerve. These results support a model of sign-conserving feedback from second order neurons onto receptor terminals and are consistent with the hypothesis that the receptor transmitter may be acetylcholine and the feedback transmitter could be GABA.

Changes in ERG b-wave and Müller cell structure induced by alpha-aminoadipic acid

The role of the Müller cell in the generation of the ERG b-wave was studied in the skate eye by examining the effects of a gliotoxic agent (alpha-aminoadipic acid; alpha-AAA) on retinal structure and function. Superfusing the eyecup for 1 h in 50 mM alpha-AAA resulted in the loss of the b-wave and extensive damage to glial cells, i.e. disruption of the cells' plasma membranes, and a marked loss of cytoplasmic substance. Of the other retinal elements, only the horizontal cells showed some signs of injury in alpha-AAA. On returning the retina to a normal Ringer solution, the widespread loss of cytoplasmic electron density persisted, but the Müller cell membranes appeared to have undergone repair, and the b-wave recovered fully its normal amplitude and waveform.

Lateral ocellar nerve projections in the dragonfly brain

The central projections of the lateral ocellar neurons of the dragonfly were examined using whole nerve cobalt iontophoresis, supplemented by sectioning of the nerve and brain for inspection in the light and electron microscopes. At E.M. level the presence of cobalt in filled axon profiles and cell bodies was confirmed by analysis of X-ray energy spectra in the microscope. The pathways, cell body sites and terminal arborizations of four large (7--25 micrometer diameter) lateral ocellar neurons are described. Two of these fibers arborize in the ipsilateral posterior neuropil of the protocerebrum and two cross the brain and arborize in the contralateral posterior neuropil. Within each half of the posterior neuropil, two spatially separated regions of ocellar input have been identified. These regions receive median ocellar input plus input from either the ipsi- or contralateral ocellus, but not both. The arborizations of the contralateral fibers are more extensive than those of the ipsilateral fibers. One of the contralateral neurons crosses the brain in the region of the protocerebral bridge giving off a collateral in that region before descending to the posterior neuropil. This collateral arborizes almost immediately in a region receiving input from arborizations of a number of small ocellar neurons (those less than 5 micrometer in diameter) from the ipsilateral ocellar nerve, together with small neurons from the median ocellar nerve, forming a region in each half of the brain which receives input from all three ocelli. The small lateral ocellar neurons associated with these arborizations have cell bodies adjacent to the lateral ocellar tracts.

Feedback synaptic interaction in the dragonfly ocellar retina

The intracellular response of the ocellar nerve dendrite, the second order neuron in the retina of the dragonfly ocellus, has been modified by application of various drugs and a model developed to explain certain features of that response. Curare blocked the response completely. Both picrotoxin and bicuculline eliminated the "off" overshoot. Bicuculline also decreased the size of response and the sensitivity. gamma-Aminobutyric acid (GABA), however, increased the size of response. The evidence indicates the possibility that the receptor transmitter is acetylcholine and is inhibitory to the ocellar nerve dendrite whereas the feedback transmitter from the ocellar nerve dendrite may be GABA and is facilitory to receptor transmitter release. The model of synaptic feedback interaction developed to be consistent with these results has certain important features. It suggests that the feedback transmitter is released in the dark to increase input sensitivity from receptors in response to dim light. This implies that the dark potential of the ocellar nerve dendrite may be determined by a dynamic equilibrium established by synaptic interaction between it and the receptor terminals. Such a system is also well suited to signalling phasic information about changes in level of illumination over a wide range of intensities, a characteristic which appears to be a significant feature of the dragonfly median ocellar response.

Action spectra and chromatic mechanisms of cells in the median ocelli of dragonflies

Spectral sensitivities were recorded intracellulary in median ocelli of Anax junius, Aeschnatuberculifera, and Libellulapulcella. All cells had peak sensitivities at 360 and 500 nm while UV-blue+green cells found only in Anax had a third peak sensitivity at 440 nm. Ratios of UV-to-green sensitivities varied from cell to cell in each ocellus, but no UV-only or green-only cells were recorded. Half of the cells tested had a reverse Purkinje shift: They were more sensitive in the green at low illuminations but more sensitive in the UV at high illuminations; their intensity-response curves at 370 and 520 nm crossed but became parallel for large responses. Wave-lengths 420 nm and shorter elicited a family of low intensity-response curves with one slope; wavelengths 440 nm and longer elicities a family of curves with another slope. Orange-adapting lights selectively adapted sensitivity in the green, but UV-adapting lights had little selective effect. Amounts of log-selective adaptation were proportional to log orange-adapting intensity. It is concluded that two spectral mechanisms can be recorded from each cell, possibly by coupling of UV and green cells or possibly because each cell contains two visual pigments. Selective chromatic adaptations may provide the ocellus with a kind of "authomatic color control," while the reverse Purkinje shift could extend the ocellus' sensitivity to prevailing skylight.

Neural organization of the median ocellus of the dragonfly. II. Synaptic structure

Two types of presumed synaptic contacts have been recognized by electron microscopy in the synaptic plexus of the median ocellus of the dragonfly. The first type is characterized by an electron-opaque, button-like organelle in the presynaptic cytoplasm, surrounded by a cluster of synaptic vesicles. Two postsynaptic elements are associated with these junctions, which we have termed button synapses. The second synaptic type is characterized by a dense cluster of synaptic vesicles adjacent to the presumed presynaptic membrane. One postsynaptic element is observed at these junctions. The overwhelming majority of synapses seen in the plexus are button synapses. They are found most commonly in the receptor cell axons where they synaptically contact ocellar nerve dendrites and adjacent receptor cell axons. Button synapses are also seen in the ocellar nerve dendrites where they appear to make synapses back onto receptor axon terminals as well as onto adjacent ocellar nerve dendrites. Reciprocal and serial synaptic arrangements between receptor cell axon terminals, and between receptor cell axon terminals and ocellar nerve dendrites are occasionally seen. It is suggested that the lateral and feedback synapses in the median ocellus of the dragonfly play a role in enhancing transients in the postsynaptic responses.

Neural organization of the median ocellus of the dragonfly. I. Intracellular electrical activity

Intracellular responses from receptors and postsynaptic units have been recorded in the median ocellus of the dragonfly. The receptors respond to light with a graded, depolarizing potential and a single, tetrodotoxin-sensitive impulse at "on." The postsynaptic units (ocellar nerve dendrites) hyperpolarize during illumination and show a transient, depolarizing response at "off." The light-evoked slow potential responses of the postsynaptic units are not altered by the application of tetrodotoxin to the ocellus. It appears, therefore, that the graded receptor potential, which survives the application of tetrodotoxin, is responsible for mediating synaptic transmission in the ocellus. Comparison of pre- and postsynaptic slow potential activity shows (a) longer latencies in postsynaptic units by 5-20 msec, (b) enhanced photosensitivity in postsynaptic units by 1-2 log units, and (c) more transient responses in postsynaptic units. It is suggested that enhanced photosensitivity of postsynaptic activity is a result of summation of many receptors onto the postsynaptic elements, and that transients in the postsynaptic responses are related to the complex synaptic arrangements in the ocellar plexus to be described in the following paper.