Search

1. The interaction between K+, Cl- and H2O fluxes was studied in the ventricular membrane of the choroid plexus epithelium from Necturus maculosus by means of ion-selective microelectrodes. 2. Three experimental strategies were adopted: the osmolarity of the ventricular solution was increased abruptly by addition of (i) mannitol or (ii) KCl; (iii) Na+ in the ventricular solution was replaced isosmotically by K+. 3. The mannitol experiments showed that H2O had two pathways across the ventricular membrane. One was purely passive, with a water permeability, L'p, of 0.64 x 10(-4) cm s-1 (osmol l-1)-1. This operated in parallel with an ion-dependent pathway of similar magnitude which was abolished in Cl(-)-free solutions. 4. When KCl was added there was a flow of H2O into the cell. Surprisingly, this took place despite the osmotic gradient which favoured an efflux of H2O. The effect was blocked by frusemide (furosemide), in which case KCl had the same effects as applications of NaCl or mannitol. 5. Replacement of Na+ with K+ caused an influx of H2O. This flux could proceed against osmotic gradients implemented by mannitol. 6. The present data and those of earlier publications show that the interdependence of the fluxes of K+, Cl- and H2O in the exit membrane can be described as cotransport. The fluxes have a fixed stoichiometry of 1:1:500, the flux of one species is able to energize the flux of the two others, and the transport exhibits saturation and is specific for K+ and Cl-. 7. A molecular model based upon a mobile barrier in a membrane spanning protein gives an accurate quantitative description of the data.

1. Necturus gastric mucosa secretes Cl- actively across the gastric glands which are composed almost entirely of acid- and enzyme-secreting oxynticopeptic cells. Single channel studies on Necturus oxynticopeptic cells have shown that the basolateral membrane possesses multiple K(+)-selective channels but no observable Cl- channels while the apical membrane has Cl- channels but no observable K+ channels. To relate these channel properties to the conductance of the whole cell we have investigated the macroscopic membrane currents with conventional whole-cell patch-clamp techniques. 2. When bathed in amphibian Ringer solution, gastric oxynticopeptic cells had a membrane resistance of 47.8 +/- 2.8 M omega and a membrane capacitance of 75.5 +/- 2.7 pF (n = 82). This gave a specific membrane resistance of 3260 +/- 160 omega cm2 (n = 82). Reversal potentials of the oxynticopeptic cells were -13.8 +/- 1.2 mV (n = 45) for an intracellular Cl- concentration ([Cl-]i) of 42 mM and were significantly more negative -24.4 +/- 3.1 mV (n = 31, P < 0.001) for [Cl-]i = 22 mM. 3. In the absence of ATP in the pipette solution, there was an 80% reduction of the whole-cell current with a typical half-time (t1/2) of 5 min. The run-down was not observed when the pipette solution contained 4 mM ATP. 4. A slow and voltage-independent inhibition of 80% of the whole-cell currents occurred after addition of NPPB (35 microM). Ba2+ (10 mM) produced a reversible inhibition of 20% of the total current. Together, 35 microM NPPB and 10 mM Ba2+ eliminated 95% of the whole-cell currents. These data suggest that in the resting oxynticopeptic cells Cl- carried the major fraction of the current while K+ ions carried only a small fraction. 5. Total replacement of Cl- in the pipette and bath solution by gluconate- increased the membrane resistance to 751 +/- 104 M omega (n = 53) and shifted the reversal potential to -38.1 +/- 2.8 mV (n = 53). 6. Increasing the bath K+ concentration from 6 to 91 mM activated a current which had a high selectivity for K+ over choline+, Li+, Na+, Rb+ and Cs+ and was independent of Cl-. The activation of this K+ current (IK*) by high external K+ was not seen with ATP-free pipette solution. 7. Ba2+ or Cs+ had a voltage-dependent blocking effect of this inward K+ current. Ouabain (1 mM) or SCH 28080 (200 microM), specific inhibitors of the Na+,K(+)-ATPase and H+,K(+)-ATPase, had no effect.(ABSTRACT TRUNCATED AT 400 WORDS)

1. The osmotic water permeability Lp and the relations between the flows of H2O, K+ and Cl- were studied in the ventricular membrane of the epithelium from the choroid plexus of Necturus maculosus. 2. The flows were induced by abrupt changes in external osmolarity of the ventricular solution. Solution changes were convective and no effects of unstirred layers could be detected on measured parameters. 3. The initial rate of change in intracellular concentrations of K+ and Cl- was monitored by double-barrelled ion-selective microelectrodes. 4. The initial rate of flux of H2O could be monitored as the changes in the concentration of intracellular choline ions (Ch+i). When 0.5 mmol l-1 of choline chloride was added to the external solutions, Ch+i attained values of 1-5 mmol l-1. The dilution or concentration of Ch+i could be recorded by K+ electrodes since the sensitivity of these to Ch+ is more than 50 times greater than to K+. 5. The Lp of the ventricular membrane of the epithelium was 1.4-2.1 x 10(-4) cm s-1 (osmol l-1)-1 and independent of the direction of the induced water flux. Lp was unchanged in tissues adapted to osmolarities of half the physiological value. 6. The efflux of H2O induced by mannitol was associated with an instantaneous efflux of K+ which was inhibited by furosemide. The fluxes had a ratio of 40 mmol l-1. The influx of H2O induced by the removal of NaCl from the ventricular solution was associated with an instantaneous influx of K+. The H2O influx had a ratio to the flux of K+ of 70 mmol l-1. 7. The efflux of H2O induced by mannitol was associated with an efflux of Cl- which was inhibited by furosemide. The ratio of the two fluxes was in the range 15-44 mmol l-1. 8. The conclusion is that the Ch+ method gives a reliable measure of the movement of H2O across the ventricular membrane. The magnitude of the Lp and its relevance to transepithelial transport are discussed. The osmotically induced H2O movement is accompanied by furosemide-sensitive fluxes of K+ and Cl- of the same magnitude. This suggests that co-transport between H2O and KCl can take place in the membrane.

1. The interaction between Cl-, K+ and H2O fluxes were studied in the ventricular membrane of the choroid plexus epithelium from Necturus maculosus by means of ion-selective microelectrodes. The flux of H2O was measured by means of K+ electrodes as the dilution or concentration of intracellular choline ions, Ch+i. 2. In one series of experiments Cl- was readministered to the ventricular solution of tissues incubated in media with low Cl- concentrations. The resulting influx of Cl- was associated with an instantaneous influx of K+ and H2O. 3. Both the Cl- and the K+ influxes were reduced by the diuretic furosemide but were unaffected by inhibitors of Na+, K(+)-ATPase or changes in membrane potentials induced by Ba2+. Since the influx of K+ proceeds against its electrochemical gradient and is unaffected by changes in membrane potentials, the membrane exhibits secondary active, electroneutral transport of K+. 4. The influx of water, initiated simultaneously with the influx of K+ and Cl-, commenced before these ions had changed the osmolarity of the intracellular solution significantly. The influx of H2O could proceed against an osmotic gradient. The influx stopped when 100 mmol l-1 of mannitol was added to the ventricular solution at the same time as the Cl- ions. The influx of H2O was inhibited by K+ removal, furosemide or high external Ba2+ (10 mmol l-1), but not by strophanthidin, ouabain or low concentrations of Ba2+ (0.5 mmol l-1). The influx could not continue with other permeable anions, NO3-, acetate- or SCN-, replacing Cl-. 5. In another series of experiments Cl- was removed from the ventricular solution of tissues bathed in saline solutions with normal concentrations of Cl-. The resulting efflux of Cl- was associated with an instantaneous efflux of K+ and H2O. This efflux of H2O could proceed against an osmotic gradient of up to 70 mosmol l-1. This effect was inhibited by furosemide, in which case the water fluxes were entirely dependent on the osmotic gradients and the osmotic water permeability Lp of the ventricular membrane. 6. The data suggest that there is a coupling between the flux of KCl and of water in the ventricular membrane, which implies that the reflection coefficient sigma for KCl under the given circumstances is less than one. I suggest that the ability of leaky epithelia to transport against osmotic gradients depends on such a coupling, which derives from the properties of the proteins through which K+, Cl- and H2O leave the cell.

1. The patch clamp technique was used to study the effects of internal and external pH on the Ca(2+)- and voltage-activated maxi K+ channel present in the apical membrane of Necturus gall-bladder epithelial cells. 2. When the pH of the solution bathing the cytosolic side of inside-out patches (pHi) was lowered from 7.9 to 6.9, with internal free Ca2+ concentration ([Ca2+]i) buffered below saturation levels for the channel gating sites, channel open probability (Po) decreased. At saturating Ca2+ concentrations, Po was near 1.0, and unaffected by pHi. The results are consistent with a competitive interaction between Ca2+ and H+ at regulatory binding sites. Kinetic analysis assuming competitive binding yields a Hill coefficient for H+ of 1.3. 3. At sub-maximal [Ca2+]i, changing the pH of the solution bathing the extracellular surface of the patch (pHo) between 8 and 7, had no effect on maxi K+ channel Po, but lowering pHo to 6 or 5 significantly reduced Po. At saturating [Ca2+]i, Po was independent of pHo. 4. There were no effects of either pHi or pHo on single-channel conductance. 5. Inasmuch as reductions in either pHo or pHi decrease maxi K+ channel Po, changes in maxi K+ channel activity account in part for the reduction of apical membrane K+ conductance elicited by acidification of the bathing medium. The dominant effect of pH on maxi K+ channels is on the cytosolic surface of the membrane. 6. The change in Po elicited by small changes in [H+]i (delta Po/delta [H+]i) is -7.6 microM-1, compared to delta Po/delta [Ca2+]i = 2.6 microM-1, both at Vm = -30 mV and at physiological intracellular [H+] and [Ca2+]. This implies that [H+]i and [Ca2+]i have opposite effects on channel Po at physiological levels and underlines the importance of pHi in channel gating.

1. The whole-cell recording mode of the patch-clamp technique was used to investigate the ionic currents of enterocytes isolated from the small intestine epithelium of Necturus maculosus. 2. When enterocytes were bathed in a physiological Ringer solution and dialysed with a K(+)-rich intracellular solution containing 1.5 x 10(-7) M intracellular Ca2+, strongly outwardly rectifying currents were observed. These currents were absent from enterocytes internally dialysed with K(+)-free solutions. 3. The kinetics of activation of the outwardly rectifying current was monoexponential with rate constants decreasing with depolarization from 160 ms at 20 mV to 40 ms at 60 mV. Similar voltage dependence of the relaxations after activation were observed. 4. Strongly buffering intracellular Ca2+ with EGTA inhibited outward currents, while increasing Ca2+ increased both their magnitude and rate of activation. 5. Bath application of the K+ channel inhibitors Ba2+ and TEA greatly attenuated outwardly rectifying currents. This observation plus the fact that tail currents reverse near EK points to K+ as the charge carrier in these currents. 6. Outside-out patches showed maxi K+ channels and lower conductance K+ channels. Averaging fluctuations of the maxi K+ channels gave a kinetic behaviour similar to the whole-cell currents.

1. The ionic permeability of Necturus maculosus small intestine epithelial cells was investigated using intracellular microelectrodes to measure membrane potential in intact tissue or by the single-electrode voltage-clamp technique in isolated cells. 2. The basolateral membrane of enterocytes appears to be K+ selective as demonstrated by the dependence of membrane potential and fractional serosal resistance measured in the intact epithelium on serosal K+ concentration. 3. Isolated cells had membrane potentials similar to those measured in the intact tissue. Voltage-clamp experiments in a physiological Ringer solution showed the presence of both large outward and inward currents. 4. Removal of Cl- from the bathing medium, linear subtraction or the use of a Cl- channel blocker revealed outwardly rectifying currents. The quasi-linear component was also revealed following K+ channel inhibition; it reversed near ECl, suggesting that the charge carrier was Cl-. 5. Outwardly rectifying currents could be kinetically resolved into two components. A fast component (tau for activation < 4 ms) accounted for 60-80% of the total current at positive potentials. A slowly activating component appeared at voltages positive to 50 mV with tau for activation of > 25 ms. 6. The slow outward current showed strong voltage dependence of both activation and relaxation, which were faster at more depolarized potentials. 7. Both fast and slow outward currents seem to be carried by K+ as they were blocked by Ba2+ and tetraethylammonium (TEA). Tail current analysis of the slow component indicated a reversal potential very similar to EK. 8. Fast outward currents were half-activated at about -40 mV whereas slow outward currents were only apparent at more positive potentials. It is proposed that the fast outward K+ current plays a role, together with Cl- currents, in determining the resting membrane potential of Necturus enterocytes.

1. Changes in membrane potential and conductance were measured in on-centre and off-centre ganglion cells during the responses to illumination of different portions of the receptive field. 2. In on-centre ganglion cells the sustained depolarizing response to steady illumination of the receptive field centre was associated with a net increase in conductance. In the presence of centre illumination, stimulation of the surround with an annulus of light caused a hyperpolarization and a net decrease in conductance, and the reversal potential of the light-evoked response was shifted in a negative direction. In the absence of centre illumination the same annular stimulus caused a hyperpolarization and a net increase in conductance. 3. In off-centre ganglion cells the sustained hyperpolarizing response to centre illumination was associated with a net increase in conductance. In the presence of centre illumination, stimulation of the surround with an annulus caused a depolarization and a net decrease in conductance, and the reversal potential of the light-evoked response was shifted in a positive direction. In the absence of centre illumination the same annulus caused a depolarization and a net increase in conductance. 4. The results indicate that illumination of the receptive field surround can affect both the excitatory and inhibitory sustained inputs to a given ganglion cell in a 'push-pull' manner, by decreasing the synaptic input that was increased by centre illumination and increasing the synaptic input of opposite sign. The relative effect of a given surround illumination on these two inputs, and hence the sign and magnitude of the net conductance change, varied with the amount of centre illumination.

1. The cell-attached and excised inside-out configurations of the patch-clamp technique were employed to probe isolated enterocytes of Necturus maculosus for the presence of Cl(-)-selective channels. 2. Chloride-selective channels were rarely observed unless cells were previously stimulated by agonists that raise cyclic AMP. In cell-attached patches forskolin (20 microM) or dibutyryl cyclic AMP 2 mM) evoked single-channel activity that reversed, depending on the cell, between 9 and 27 mV positive to the spontaneous membrane potential. This is close to the Cl- equilibrium potential in those cells; the single-channel current-voltage relationship was linear with a unitary slope conductance between 17 and 25 pS (pipettes filled with 100 mM-NaCl). 3. Large depolarizing voltage steps also activated Cl- channels in excised inside-out membrane patches that were previously quiescent. This mode of activation produced a distinctive single-channel current-voltage relationship with strong outward rectification at depolarizing membrane potentials. Single-channel cord conductance at negative potentials was 15-18 pS and increased to 45 pS at + 100 mV. 4. Altering the Cl- concentration in the bathing solution of excised inside-out patches displaced the observed reversal potential (Erev) to values predicted for Cl- equilibrium potential. Replacement of K+ for Na+ was without effect. 5. The effect of different anions upon Erev was used to determine the channel anion selectivity in excised inside-out patches. The permeability sequence was SCN- greater than I- greater than Br- greater than Cl- greater than F- greater than HCO3- greater than gluconate which corresponds to Eisenman's sequence 1. Neither ionic size nor diffusion rates determine the permeation of ions through the channel. 6. In channels activated by depolarization the open probability (Po) was insensitive to changes in the Ca2+ concentration (less than 10(-8)-10(-3) M) bathing the cytoplasmic face of excised inside-out patches. Depolarization was also without marked effect on Po. 7. Chloride channels in excised inside-out patches were inhibited by stilbene and diphenylamine-2-carboxylate derivatives. 4-Acetamido-4'-isothiocyanatostilbene-2,2'-disulphonic acid (SITS, 5 x 10(-5) M) and 5-nitro-2-(3-phenylpropylamino) benzoic acid (NPPB, 1 x 10(-5) M) caused an irreversible 'flickery' blockade without altering single-channel current. 3'5-Dichlorodiphenylamine-2-carboxylic acid (DDPC, 5 x 10(-5) M) reduced the currents at every voltage without apparent effects on gating properties of the channel.(ABSTRACT TRUNCATED AT 400 WORDS)

1. The steady-state fluxes of Na, K and Cl ions have been measured in Necturus gall-bladder epithelium by a technique that involves labelling the cells with tracer ions from the mucosal bath only, whilst the serosa is kept at low specific activity. After removing tracer, the efflux is followed into the serosal bath, revealing two exponential components. 2. The time constant of the fast component lies between 0.03 and 0.04 s-1 and corresponds to that of the extracellular space. The slow component closely matches the cellular efflux, with constants which lie between 0.14 and 0.46 X 10(-2) s-1. 3. Full unstirred-layer calculations have been performed to determine the specific activities in the mucosal solution, the cell and the corium (subepithelium). These involved measuring the diffusion coefficients of Na and Cl in the isolated corium: they are restricted by factors of 0.17 and 0.11. 4. The partial flux equations for this double-membrane system have been solved to obtain the cellular fluxes for all three ions. The results indicate that: (i) the net transcellular Na flux is 190 pmol cm-2 s-1, equivalent to the transepithelial salt flux during fluid secretion; (ii) the net transcellular K flux is effectively zero because this ion recirculates across the serosal membrane; (iii) the net transcellular Cl flux is 27 pmol cm-2 s-1, or 15% of the net transepithelial salt flux. 5. The permeability of the paracellular pathway to Cl is 1.65 X 10(-5) cm s-1 and the available driving forces will allow a maximum net electrodiffusive Cl transport of 10% through the shunt pathway. 6. 1:1 coupling of Na and Cl net fluxes at the mucosal membrane of this epithelium cannot be present, and processes other than simple electrodiffussion are required to effect net Cl transport by another route. 7. The serosal fluxes of K and Cl do not obey the flux-ratio equation. A component of these fluxes must be present which is neither active (pumped) nor passive (electrodiffusive and independent). If they are symmetrical in the steady state then the ratio of these exchange fluxes lies between 2:1 and 3:1 depending upon the size of the pump flux. They support the view that a mode of coupled K and Cl transport may be operating at the basolateral membrane of these cells.

1. Transepithelial Na transport in Necturus was determined by measuring the rate of isotonic fluid flow. The rate at 20 degrees C was equivalent to 175 pmol cm-2 s-1. 2. Ouabain was effective in Necturus, binding to the Na pump in gall-bladder cells with a mean rate constant of 5.4 X 10(3) M-1 s-1. Measurement of the diffusive time constant of the free space for [3H]ouabain shows that the pump must be fully inhibited within 20 s when ouabain is applied to the serosa at 10(-3) M. 3. The serosal Na efflux from loaded cells was inhibited 36% by ouabain equal to a flux of 73 pmol cm-2 s-1. The remaining flux could not be attributed to either exchange diffusion or electrodiffusion induced by ouabain. 4. The transepithelial potential was 0.3 mV serosa positive. The short-circuit current measured was 6.33 +/- 1.9 microA cm-2, equal to a positive univalent ion flux of 65.6 pmol cm-2 s-1 or 38% of the net Na transfer. The current was inhibited within 1-5 min by 5 X 10(-5) M-amiloride. 5. Fluid secretion was immediately inhibited 34% by ouabain, equivalent to an isotonic transport of Na of 59.7 pmol cm-2 s-1. Thereafter it continued for at least an hour, sometimes declining slowly. Amiloride had little effect (13%). 6. The Na pump rate was measured by titrating the cell content with tracer Na at different times after ouabain treatment. The initial slope was equal to a rate of 61.6 pmol cm-2 s-1 or 35% of the net flux at time zero. 7. The Na pump rate has also been measured by recording the rise in cell Na activity with ion-specific micro-electrodes, and correcting for swelling effects. The Na pump rate was very similar to that estimated from the rise in tracer Na content, equal to 59.3 pmol cm-2 s-1 or 31.4% of the transepithelial rate. Examination of the same experiment in the literature shows a closely similar value, about one-third of that expected from fluid secretion or net flux measurements. 8. A scheme is proposed to explain the results, which requires a flow of NaCl through a parallel pathway of small Na content involving exchange en route with the cytoplasmic Na.

1. The effects of some amino acids and oligopeptides on the electrical properties of the brush-border membrane of the small intestine of the mudpuppy, Necturus maculosus were studied in vitro using micro-electrodes. 2. A number of amino acids (glycine, L-proline and L-leucine) and small peptides (carnosine, glycyl-L-proline, L-leucyl-L-leucine, glycylglycine and glycylglycylglycine) depolarized the brush-border membrane. This was associated with a reduction of input resistance. Tetraglycine did not appreciably reduce the membrane potential. 3. Evidence is presented that the electrical effects caused by application of the peptides are not solely due to the effects of the products of peptide hydrolysis. Furthermore there appears to be more than a single system available to the peptides. 4. Stereospecificity is found for both amino acids and peptides but appears to be more marked for the peptides. 5. Peptide-induced depolarizations are less markedly reduced in the absence of external Na+ than are the depolarizations caused by the amino acids. 6. These results are discussed with respect to the mechanism and significance of amino acid and peptide transport in the intestinal tract.

1. Addition of sodium cyanide to the mucosal or the serosal medium bathing the isolated gall-bladder of Necturus maculosus causes hyperpolarization of both apical and basolateral membrane of the epithelial cells. The effect of cyanide is practically immediate, reversible (if exposure is brief), and long-lasting (greater than 30 min). 2. The hyperpolarization is accompanied by: (a) reduction of the equivalent resistance of the cell membranes, as shown by cable analysis and input resistance measurements, and (b) increase of the potassium selectivity of both cell membranes, as evidenced by the effects of external substitutions of potassium for sodium on cell membrane potentials. We conclude that the cyanide-induced hyperpolarization is caused mainly or exclusively by an increase of the potassium permeability of the cell membranes. 3. Addition of the calcium ionophore A23187 (5 microM) to the mucosal medium in the presence of 1 mM-calcium caused similar effects to those produced by cyanide. After either cyanide or A23187, addition of the other agent did not cause further membrane potential changes. 4. Quinine (100 microM, mucosal medium) reduced the potassium permeability of the apical membrane both under control conditions and during exposure to cyanide. 5. We suggest that the cyanide-induced increase of the potassium permeability of the cell membrane is mediated by an elevation of intracellular calcium ion activity, attributable to release from mitochondrial sources.

1. Intracellular potentials, Cl- activity and membrane resistances were measured in Necturus small intestinal epithelium during Cl- replacement experiments using conventional or Cl- -selective double-barrelled microelectrodes. A Cl- conductance, located in the apical membrane and activated by cyclic nucleotides is demonstrated by ion-substitution experiments. 2. The mean mucosal membrane potential (Em) was -35.5 mV. Removal of Cl- from the mucosal medium by replacement with gluconate, evoked a sudden depolarization of Em and an immediate increase in the fractional resistance of the mucosal membrane (f(Rm)). The size of the change in Em varied between 3 and 65 mV, corresponding to Cl- to K+ permeability ratios between 0.2 and 20. It was inversely related to the initial f(Rm), which ranged from 0.04 to 0.50. 3. Prolonged incubation in low-Cl- solutions led to a reversal of the initial depolarization and to a sustained hyperpolarization accompanied by a marked increase in f(Rm). The new value of Em was close to the K+ equilibrium potential, consistent with a depletion of cellular Cl- and the preponderance of a K+ membrane permeability in the absence of Cl-. This emphasizes the role of Cl- in establishing Em. 4. Removal of mucosal Cl- produced a fast decrease in intracellular Cl-, as measured with Cl- -selective microelectrodes. The efflux was consistent with electrodiffusion across the mucosal membrane. Changes in Em paralleled changes in intracellular Cl- activity, indicating the presence of a large Cl- conductance. 5. Dibutyryl cyclic AMP or forskolin produced a slow depolarization, a decrease in f(Rm) and an increased change in intracellular potential in low mucosal Cl- which on average corresponds to an approximately 15-fold increase in the relative Cl- permeability. These results are consistent with an activation of apical Cl- conductance. 6. The selectivity of Cl- channels of Necturus enterocytes to different anions was obtained from potential measurements. The sequence of permeabilities was SCN- greater than I- greater than or equal to Br- greater than NO3- greater than Cl- much greater than HCO3- greater than gluconate. This is consistent with a model involving a weak interaction of the anions with the selectivity filter. 7. The selectivity of the anion conductance was maintained after activation with cyclic nucleotides, suggesting a single channel for the permeation of the different anions tested, rather than parallel channels. 8. Derivatives of 9-anthracene which are potent inhibitors of Cl- channels in other systems failed to block the apical Cl- conductance of Necturus enterocytes. Chloride conductance was also insensitive to furosemide.(ABSTRACT TRUNCATED AT 400 WORDS)

1. Intracellular responses were recorded from on-centre and off-centre ganglion cells in isolated eyecups of the mudpuppy, Necturus maculosus.2. Current-voltage relations were measured in darkness, during illumination of the receptive field centre, and after chemically mediated synaptic inputs were blocked by 4 mM-cobalt chloride.3. In on-centre cells the membrane potential in darkness was -56+/-6 mV (mean+/-S.D.). Addition of Co(2+) resulted in an average depolarization of 10 mV and an average decrease in conductance of 2.1 nS. These results suggest that in darkness on-centre cells are tonically inhibited by synaptic input which increases conductance and has a reversal potential more negative than the dark membrane potential. In off-centre cells the membrane potential in darkness was -46+/-5 mV. Addition of Co(2+) caused an average hyperpolarization of 6 mV and an average decrease in conductance of 1.5 nS. These results suggest that in darkness off-centre cells receive a tonic excitatory input which increases conductance and has a reversal potential more positive than the dark membrane potential.4. In on-centre cells light causes a sustained depolarization. This response involves an increase in a tonic excitatory input which increases conductance and has a reversal potential more positive than the dark membrane potential.5. In off-centre cells, light causes a sustained hyperpolarization. This response involves an increase in a sustained inhibitory input which increases conductance and has a reversal potential more negative than the dark membrane potential.6. The depolarizing off-response of off-centre cells is associated with an increase in an excitatory input which increases conductance and has a reversal potential more positive than the dark membrane potential. This response may be due to a temporary increase in the excitatory input which is tonically active in darkness or may reflect an additional excitatory input.7. It is suggested that in both on- and off-centre ganglion cells the balance of sustained excitatory and inhibitory synaptic inputs determines the resting potential in darkness. Centre illumination alters the balance of these inputs, by increasing one and decreasing the other, to produce the characteristic sustained light responses.8. The possible presynaptic sources of the sustained excitatory and inhibitory inputs are discussed.