Ganglion Cell Types Research Articles

Optic nerve signals in a neuromorphic chip II: Testing and results.

Seeking to match the brain's computational efficiency, we draw inspiration from its neural circuits. To model the four main output (ganglion) cell types found in the retina, we morphed outer and inner retina circuits into a 96 x 60-photoreceptor, 3.5 x 3.3 mm2, 0.35 microm-CMOS chip. Our retinomorphic chip produces spike trains for 3600 ganglion cells (GCs), and consumes 62.7 mW at 45 spikes/s/GC. This chip, which is the first silicon retina to successfully model inner retina circuitry, approaches the spatial density of the retina. We present experimental measurements showing that the chip's subthreshold current-mode circuits realize luminance adaptation, bandpass spatiotemporal filtering, temporal adaptation and contrast gain control. The four different GC outputs produced by our chip encode light onset or offset in a sustained or transient fashion, producing a quadrature-like representation. The retinomorphic chip's circuit design is described in a companion paper [Zaghloul and Boahen (2004)].

Identification of retinal neurons in a regressive rodent eye (the naked mole-rat)

The retina consists of many parallel circuits designed to maximize the gathering of important information from the environment. Each of these circuits is comprised of a number of different cell types combined in modules that tile the retina. To a subterranean animal, vision is of relatively less importance. Knowledge of how circuits and their elements are altered in response to the subterranean environment is useful both in understanding processes of regressive evolution and in retinal processing itself. We examined common cell types in the retina of the naked mole-rat, Heterocephalus glaber with immunocytochemical markers and retrograde staining of ganglion cells from optic nerve injections. The stains used show that the naked mole-rat eye has retained multiple ganglion cell types, 1-2 types of horizontal cell, rod bipolar and multiple types of cone bipolar cells, and several types of common amacrine cells. However, no labeling was found with antibodies to the dopamine-synthesizing enzyme, tyrosine hydroxylase. Although most of the well-characterized mammalian cell types are present in the regressive mole-rat eye, their structural organization is considerably less regular than in more sighted mammals. We found less precision of depth of stratification in the inner plexiform layer and also less precision in their lateral coverage of the retina. The results suggest that image formation is not very important in these animals, but that circuits beyond those required for circadian entrainment remain in place.

PARALLEL VISUAL PROCESSING: A TUTORIAL OF RETINAL FUNCTION

The mammalian retina extracts a set of 12 different "movies" from the visual scene and send each movie to higher visual centers. These movies exist in physically distinct strata of the inner retina, embodied in the dendritic arborizations of a dozen different ganglion cell types. The dendritic arbors are stacked in discrete layers throughout the inner retina where they are synaptically contacted by excitatory and inhibitory connections from a variety of different cell types. This paper outlines the synaptic and physical structure of these layers, and follows the image of a face through the retina from photoreceptors to final output, graphically displaying the transformations that take place as the image is processed by different neural structures.

The human round window – a perilymph pressure regulator?On a novel mechanoreceptor‐like neuron in the human round window membrane

A medial innervation of the human round window membrane (RWM) was recently discovered in well fixed human tissue using immunohistochemistry. In the present study we used re‐embedding and serial thin section ultrastructural techniques (TEM) to locate and follow these nerve fibres from the spiral ganglion (SG) to the RWM. One neuron projecting through a small perforating bone channel into the RWM could be analysed in detail near the lamina spiralis ossea. Within the middle layer of the RWM it formed a terminal ‘loop’ and was anchored in the stromal tissue. The inward projecting fibre was surrounded by a loose perineural sheath interrupted by gaps where basal lamina faced the extra‐cellular matrix, similar to joint mechanoreceptors. Followed proximally, the neuron branched and formed a unit of several myelinated and un‐myelinated nerve fibres. The cell body was traced to the SG and showed unique ultrastructural features seemingly representing a novel type of ganglion cell unlike the small and large cells nor...

Information transmission rates of cat retinal ganglion cells.

To assess the information encoded in retinal spike trains and how it might be decoded by recipient neurons in the brain, we recorded from individual cat X and Y ganglion cells and visually stimulated them with randomly modulated patterns of various contrast and spatial configuration. For each pattern, we estimated the information rate of the cells using linear or nonlinear algorithms and for some patterns by directly measuring response probability distributions. We show that ganglion cell spike trains contain information from the receptive field center and surround, that the center and surround have similar signaling capacity, that antagonism between the mechanisms reduces information transmission, and that the total information rate is limited. We also show that a linear decoding algorithm can capture all of the information available in retinal spike trains about weak inputs, but it misses a substantial amount about strong inputs. For the strongest stimulus we used, the information rate of the best linear decoder averaged 40-70 bits/s across ganglion cell types, while the directly measured rate was around 20-40 bits/s greater. This implies that under certain stimulus conditions, visual information is encoded in the temporal structure of retinal spike trains and that a nonlinear decoding algorithm is needed to extract the temporally coded information. Using simulated spike trains, we demonstrate that much of the temporal structure may be explained by the threshold for spike generation and is not necessarily indicative of a complex coding scheme.

Glutamate receptors at bipolar synapses in the inner plexiform layer of primate retina: light microscopic analysis.

At least 10 different types of bipolar cells have been distinguished in the primate retina. The axon terminals of these cells stratify in distinct strata in the inner plexiform layer and are involved in parallel pathways to distinct types of ganglion cells. Ionotropic glutamate receptor (GluR) subunits also show a stratified distribution in the inner plexiform layer. Here, we investigated whether different types of bipolar cells are associated with different types of ionotropic glutamate receptors in the inner retina of a New World primate, the common marmoset Callithrix jacchus. Vertical cryostat sections through central retina were double labeled with immunohistochemical markers for bipolar cell types and with antibodies to alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA) receptor subunits GluR1 to 4, kainate receptor subunits GluR6/7, and the NR1C2' subunit of the N-methyl-D-aspartate (NMDA) receptor. The axon terminals of bipolar cell types were reconstructed from confocal sections, and the colocalized immunoreactive puncta were quantified. For all bipolar cell types, immunoreactive puncta for the AMPA receptor subunits GluR2, 2/3, and 4 were colocalized at highest densities, whereas GluR1-immunoreactive puncta were expressed at very low densities. The kainate receptor subunits GluR6/7 were predominantly associated with diffuse bipolar (DB6) and rod bipolar cells. The NMDA receptor subunit NR1C2' was specifically colocalized with flat midget and DB3 axons. These findings suggest that rod and cone bipolar cell types contribute to multiple but distinct glutamate receptor pathways in primate retina.

Retinal ganglion cell projections to the hamster suprachiasmatic nucleus, intergeniculate leaflet, and visual midbrain: bifurcation and melanopsin immunoreactivity.

The circadian clock in the suprachiasmatic nucleus (SCN) receives direct retinal input via the retinohypothalamic tract (RHT), and the retinal ganglion cells contributing to this projection may be specialized with respect to direct regulation of the circadian clock. However, some ganglion cells forming the RHT bifurcate, sending axon collaterals to the intergeniculate leaflet (IGL) through which light has secondary access to the circadian clock. The present studies provide a more extensive examination of ganglion cell bifurcation and evaluate whether ganglion cells projecting to several subcortical visual nuclei contain melanopsin, a putative ganglion cell photopigment. The results showed that retinal ganglion cells projecting to the SCN send collaterals to the IGL, olivary pretectal nucleus, and superior colliculus, among other places. Melanopsin-immunoreactive (IR) ganglion cells are present in the hamster retina, and some of these cells project to the SCN, IGL, olivary pretectal nucleus, or superior colliculus. Triple-label analysis showed that melanopsin-IR cells bifurcate and project bilaterally to each SCN, but not to the other visual nuclei evaluated. The melanopsin-IR cells have photoreceptive characteristics optimal for circadian rhythm regulation. However, the presence of moderately widespread bifurcation among ganglion cells projecting to the SCN, and projection by melanopsin-IR cells to locations distinct from the SCN and without known rhythm function, suggest that this ganglion cell type is generalized, rather than specialized, with respect to the conveyance of photic information to the brain.

Two neuropharmacological types of rabbit ON-alpha ganglion cells express GABAC receptors.

The major inhibitory neurotransmitters GABA and glycine provide the bulk of input to large-field ganglion cells in the retina. Whole-cell patch-clamp recordings were used to characterize the glycine- and GABA-activated currents for morphologically identified ON-alpha ganglion cells in the rabbit retina. Cells identified as ON-alpha cells by light evoked currents were intracellularly stained and examined by light microscopy which revealed dendritic stratification in the vitreal half of the inner plexiform layer and confirmed their physiological identity. All Ca(2+)-mediated synaptic influences were abolished with Co(2+), revealing two types of ON-alpha cell characterized by their different inhibitory current profiles. One group exhibited larger glycine- than GABA-activated currents, while the other group had larger GABA- than glycine-activated currents. Both cell types demonstrated strychnine-sensitive glycine-activated currents and bicuculline-sensitive GABAA-activated currents. Surprisingly, both cell types expressed functional GABAC receptors demonstrated by their sensitivity to TPMPA. In addition, the cells with larger glycine-activated currents also possessed GABAB receptors, whereas those with larger GABA-activated currents did not. Immunocytochemical experiments confirmed the presence of glycine, GABAA, and GABAC receptor subunits on all physiologically identified ON-alpha ganglion cells in this study. In addition, the GABAB receptor immunolabeled puncta were present on the cells with larger glycine-activated currents, but not on the cells with the larger GABA-activated currents. In conclusion, the presence of different functional GABA and glycine receptors determined physiologically correlated well with the specific GABA and glycine receptor immunolabeling for two neuropharmacological types of rabbit ON-alpha ganglion cells.

Differential dendritic shrinkage of alpha and beta retinal ganglion cells in cats with chronic glaucoma.

To study changes in the dendritic morphology of retinal ganglion cells (RGCs) in cats with experimental chronic glaucoma. Chronic elevation of intraocular pressure (IOP) was produced by injecting endogenous ghost red blood cells into the unilateral anterior chamber of the feline eyes for 1 month. The morphologic features of retrograde-labeled RGCs by bilateral injection of horseradish peroxidase (HRP) into layers A and Aa1 of the lateral geniculate nucleus (LGN) were examined and compared between the normal and glaucomatous eyes. Nissl staining was used for measuring the change in cell density in the retina and the LGN. Quantitative analysis of 720 labeled alpha and beta type RGCs showed that the cell density, body size, maximum dendritic field radius, total dendritic length, and number of branch bifurcations of dendrites decreased significantly in glaucomatous eyes compared with normal ones. The cell loss and shrinkage of dendrites in alpha type ganglion cells in the retina was more pronounced than that in beta type cells. The cell density of all kinds of cells in the retina and LGN monotonically declined with time while IOP was elevated, and cell loss was more significant in large cells than in small ones. Progressive cell loss and dendritic damage by chronic elevation of IOP in RGCs and LGN cells are more pronounced in the Y-channel (large cells) than the X-channel (small cells) in feline glaucomatous eyes. The dendritic structure changes and corresponding physiological deficits of RGCs occur before cell death and thus may provide an opportunity for clinical treatment.

Rapid global shifts in natural scenes block spiking in specific ganglion cell types.

The mammalian retina contains more than a dozen different ganglion cell types, each with dendrites ramifying at different strata within the inner plexiform layer (IPL) and each carrying a unique representation of the visual world. We studied the inhibitory and excitatory inputs, as well as the spiking output, of each of the rabbit retinal ganglion cell type during rapid global shifts in 'natural' videos designed to mimic saccadic eye movements. These shifts generated stratum-specific transient inhibitory activity, affecting only those ganglion cells whose dendrites ramify within the central strata of the IPL. The inhibition was GABA-mediated, acted both pre- and postsynaptically and was fast enough to either prevent or delay spiking. These findings indicate that the fast, transient elevation of visual threshold during rapid shifts in scene has a significant retinal component.

Biplexiform ganglion cells in the retina of the perciform fish Pholidapus dybowskii revealed by HRP labeling from the optic nerve and optic tectum

Using retrograde HRP labeling from the optic nerve (ON) or optic tectum (OT), we have visualized biplexiform cells in wholemounted retinas of the stichaeid fish Pholidapus dybowskii and studied their morphology and spatial properties. Biplexiform cells labeled from the ON were similar in their morphology to biplexiform cells found in other fishes. Their distribution across the retina was non-random and independent of that of other large ganglion cell types. Biplexiform cells labeled from the OT, too, formed non-random mosaics, whose spatial properties suggested that most or all biplexiform cells project to the OT in this species. We propose that biplexiform cells in Pholidapus are homologous to biplexiform cells in other fishes (lower vertebrates). Pholidapus biplexiform cells may participate in the tectum-mediated visual reactions.

Intravitreal injections of neurotrophic factors and forskolin enhance survival and axonal regeneration of axotomized β ganglion cells in cat retina

Some retinal ganglion cells in adult cats survive axotomy for two months and regenerate their axons when a peripheral nerve is transplanted to the transected optic nerve. However, regenerated retinal ganglion cells were fewer than 4% of the total retinal ganglion cell population in the intact retina. The present study examined the effects of intravitreal injections of neurotrophic factors (brain-derived neurotrophic factor, ciliary neurotrophic factor, basic fibroblast growth factor, glial cell-derived neurotrophic factor, neurotrophin 4), first on the survival of axotomized cat retinal ganglion cells within 2 weeks, and then on axonal regeneration of the retinal ganglion cells for 2 months after peripheral nerve transplantation. We tested first enhancement of the survival by one of the factors, and then one or two of them supplemented with forskolin, which increases intracellular cAMP. Single injections of 0.5 μg or 1 μg brain-derived neurotrophic factor, 1 μg ciliary neurotrophic factor, or 1 μg glial cell-derived neurotrophic factor significantly increased total numbers of surviving retinal ganglion cells; 1.6–1.8 times those in control retinas. Identification of retinal ganglion cell types with Lucifer Yellow injections revealed that the increase of surviving β cells was most conspicuous: 2.5-fold (brain-derived neurotrophic factor) to 3.6-fold (ciliary neurotrophic factor). A combined injection of 1 μg brain-derived neurotrophic factor, 1 μg ciliary neurotrophic factor, and 0.1 mg forskolin resulted in a 4.7-fold increase of surviving β cells, i.e. 50% survival on day 14. On the axonal regeneration by peripheral nerve transplantation, a combined injection of brain-derived neurotrophic factor, ciliary neurotrophic factor, and forskolin resulted in a 3.4-fold increase of β cells with regenerated axons. The increase of regenerated β cells was mainly due to the enhancing effect of neurotrophic factors on their survival, and possibly to a change of retinal ganglion cell properties by cAMP to facilitate their axonal regeneration.

Choline acetyltransferase is expressed by non-starburst amacrine cells in the ground squirrel retina

We have used immunostaining techniques to reveal a new type of amacrine cell that is immunoreactive for choline acetyltransferase (ChAT), the acetylcholine synthesizing enzyme, in the Ground Squirrel ( Spermophilus beecheyi) retina. Cryostat sections and double immunostained wholemount preparations were examined by confocal microscopy. This new ChAT type III cell is distinct in morphology and neurotransmitter content from the well know ‘starburst’ amacrine cells (types I and II) that are so well represented in the ground squirrel retina [J. Comp. Neurol. 365 (1996) 173–216]. The type III cell colocalizes glycine with the acetylcholine and does not appear to be GABAergic or exhibit calcium-binding proteins like the well-known starburst type. As well, type III cells do not occur as a mirror-symmetric pair with normally placed and displaced varieties. The type III cell is probably a small field amacrine type branching broadly in upper sublamina b of the inner plexiform layer, and is most likely A6 of the Ground Squirrel retina [J. Comp. Neurol. 365 (1996) 173–216]. Type III cells are ideally placed in the architecture of the Ground Squirrel retina to influence ON directionally selective ganglion cell types.

Psychophysical assessment of the L:M weighting of inputs to the ON and OFF S-cone pathways

New data on S-OFF ganglion cells (Dacey et al., 2001) reveal anatomical and physiological differences between ON and OFF S-cone pathways. From transient tritanopia data, McLellen & Eskew (2000) reasoned that ON and OFF S-cone pathways may have different L:M cone weights. Here, in Experiment One, we measured temporal 2AFC thresholds for detecting increment and decrement pulses (abrupt onset, ramped offset; 400 ms) that were defined only by a change in S-cone stimulation. We measured threshold versus radiance (tvr) curves for two adapting chromaticities: high L:M ratio or low L:M ratio. Stimuli originated from a colorimeter with four primaries in centre and in surround that allowed independent modulation of the rods and each of the 3 cone types. In Experiment Two we measured detection of increment and decrement pulses (rectangular; 200 ms) on a neutral reference field, 400 ms after the offset of a long-wavelength adapting field. We measured thresholds as a function of the retinal illuminance of the adapting light, and used two different adapting chromaticities. We used two classes of test stimuli. One offered pure S-cone modulation, and the other included a rod modulation in order to simulate McLellen & Eskew's nominally S-isolating stimuli generated on a CRT. Results from Experiment One were inconsistent with different ON/OFF L:M cone weightings: increment and decrement tvr curves superposed for both adapting fields. Data from Experiment Two replicated McLellen & Eskew's result: the relation between increment and decrement thresholds changed as a function of the L:M ratio in the adapting field. We suggest it is unlikely that the inferred ON and OFF S-cone pathways differ in L:M weighting. McLellen & Eskew's result may in part be ascribed to the rod signal associated with their test stimuli. Dacey, Peterson, Gamlin & Robinson., 2001. “Retrograde ‘photofilling’ reveals the complete dendritic morphology of diverse new ganglion cell types that project to the lateral geniculate nucleus in macaque monkey”. Investigative Ophthalmology & Visual Science, 42 Supplement, 617. McLellen & Eskew, 2000. “ON and OFF S-cone pathways have different long-wave cone inputs”. Vision Research, 40, 2449–2465.

Retinal ganglion cells projecting to the optic tectum and visual thalamus of lizards.

Retinal ganglion cells projecting to the optic tectum and visual thalamus have been investigated in the lizard, Podarcis hispanica. Injections of biotinylated dextran-amine in the optic tectum reveal seven morphological cell varieties including one displaced ganglion cell type. Injections in the visual thalamus yield similar ganglion cell classes plus four giant ganglion cells, including two displaced ganglion cell types. The present study constitutes the first comparison of tectal versus thalamic ganglion cell types in reptiles. The situation found in lizards is similar to that reported in mammals and birds where some cell types projecting to the thalamus are larger than those projecting to the mesencephalic roof. The presence of giant retino-thalamic ganglion cells with specific dendritic arborizations in sublaminae A and B of the inner plexiform layer suggests that parts of the visual thalamus of lizards could be implicated in movement detection, a role that might be played by the ventral lateral geniculate nucleus, which is involved in our tracer injections.

Rabbit retinal ganglion cell responses mediated by alpha-bungarotoxin-sensitive nicotinic acetylcholine receptors.

The responses of many ganglion cells in the rabbit retina are mediated, at least in part, by acetylcholine (ACh) acting on neuronal nicotinic acetylcholine receptors (nAChRs). nAChRs are comprised of alpha and beta subunits; three beta subunits and nine alpha subunits of nAChRs have been identified and these subunits can combine to form a large number of functionally distinct nAChR subtypes. We examined the effects of cholinergic agents on the light-evoked responses of ganglion cells to determine which nAChR subtypes mediate the effects of ACh. Extracellular recordings of retinal ganglion cells were made in intact everted eyecup preparations and nicotinic agonists and antagonists were added to the superfusate. While several ganglion cell classes exhibited methyllycaconitine (MLA) sensitivity, the directionally selective (DS) ganglion cells were most sensitive; exposure to 30 nanomolar MLA, a concentration reportedly too low to affect alphaBgt-insensitive nAChRs, suppressed the stimulus-evoked responses of DS cells without eliminating directional selectivity. Epibatidine, which at low concentrations is an agonist selective for alphaBgt-insensitive nAChRs, stimulated firing of various cell types including DS ganglion cells at low nanomolar concentrations. The effects of the various agents tested persisted under cobalt-induced synaptic blockade. The low nanomolar MLA and epibatidine sensitivity of DS cells suggests that DS ganglion cells express both alphaBgt-sensitive and alphaBgt-insensitive nAChRs. Other ganglion cell types appear to express only alphaBgt-sensitive nAChRs but not alphaBgt-insensitive nAChRs.

Reduced Transition between Open and Inactivated Channel States Underlies 5HT Increased INa+ in Rat Nociceptors

We previously demonstrated that activation of a 5HT4 receptor coupled cAMP-dependent signaling pathway increases tetrodotoxin-resistant Na+ current (INa) in a nociceptor-like subpopulation of rat dorsal root ganglion cells (type 2). In the present study we used electrophysiology experiments and computer modeling studies to explore the mechanism(s) underlying the increase of INa by 5HT. In electrophysiological experiments with type 2 dorsal root ganglion cells, 5HT increased peak INa and the activation and inactivation rate, without significantly affecting the voltage dependency of activation or availability. Studies on the voltage dependency of channel availability, time course of removal of inactivation, and inactivation of evoked Na+ currents suggested that there are at least two inactivation states of the Na+ channel, one (Ifast) that is induced and retrieved faster than the other (Islow). Long (1s), but not short (60 or 100ms), inactivating conditioning pulses (CPs) suppressed the 5HT-induced increase in INa. Computer modeling studies suggest that 5HT increased INa mainly by decreasing the transition rate (kOI1) from an open state to Ifast. Furthermore, 5HT increased INa activation and inactivation rates mainly by increasing the transition rate from closed to open (kC3O) and from Ifast to Islow (kI1I2), respectively. The antagonism of the 5HT-induced increase in INa by 1-s inactivation CPs may be due an enhancement of transitions from Ifast to Islow, via the increase in kI1I2. This may deplete the pool of channels residing in Ifast, reducing the frequency of reopenings from Ifast, which offsets the increase in INa produced by the reduction in kOI1. The above findings fit well with previous studies showing that activation of the cAMP/PKA cascade simultaneously increases voltage sensitive tetrodotoxin-resistant Na+ conductance and inactivation rate in nociceptors. The antagonism of the effects of 5HT by long inactivation CPs suggests that drugs designed to induce and/or stabilize the Islow state might be useful for reducing hyperalgesia produced by inflammatory mediators.

The Diversity of Ganglion Cells in a Mammalian Retina

We report a survey of the population of ganglion cells in the rabbit retina. A random sample of 301 neurons in the ganglion cell layer was targeted for photofilling, a method in which the arbors of the chosen neurons are revealed by diffusion of a photochemically induced fluorescent product from their somas. An additional 129 cells were labeled by microinjection of Lucifer yellow. One hundred and thirty-eight cells were visualized by expression of the gene encoding a green fluorescent protein, introduced by particle-mediated gene transfer. One hundred and sixty-six cells were labeled by particle-mediated introduction of DiI. In the total population of 734 neurons, we could identify 11 types of retinal ganglion cell. An analysis based on retinal coverage shows that this number of ganglion cell types would not exceed the available total number of ganglion cells. Although some uncertainties remain, this sample appears to account for the majority of the ganglion cells present in the rabbit retina. Some known physiological types could easily be mapped onto structural types, but half of them could not; a large set of poorly known codings of the visual input is transmitted to the brain.

The diversity of ganglion cells in a mammalian retina.

We report a survey of the population of ganglion cells in the rabbit retina. A random sample of 301 neurons in the ganglion cell layer was targeted for photofilling, a method in which the arbors of the chosen neurons are revealed by diffusion of a photochemically induced fluorescent product from their somas. An additional 129 cells were labeled by microinjection of Lucifer yellow. One hundred and thirty-eight cells were visualized by expression of the gene encoding a green fluorescent protein, introduced by particle-mediated gene transfer. One hundred and sixty-six cells were labeled by particle-mediated introduction of DiI. In the total population of 734 neurons, we could identify 11 types of retinal ganglion cell. An analysis based on retinal coverage shows that this number of ganglion cell types would not exceed the available total number of ganglion cells. Although some uncertainties remain, this sample appears to account for the majority of the ganglion cells present in the rabbit retina. Some known physiological types could easily be mapped onto structural types, but half of them could not; a large set of poorly known codings of the visual input is transmitted to the brain.

Displaced retinal ganglion cells project to the accessory optic system in the chameleon ( Chamaeleo calyptratus).

Retinal ganglion cells were successfully labelled in the chameleon by retrograde axonal transport of dextran amines that were applied to the nucleus of the basal optic root (nBOR) in an in vitro preparation. Labelled ganglion cells were restricted to the contralateral eye. Many cells were completely stained including their dendritic trees. With few exceptions, all cells had displaced somata that were located at the inner margin of the inner nuclear layer. The labelled ganglion cells had two to six primary dendrites that branched frequently and formed large unistratified dendritic trees within sublamina 1 of the inner plexiform layer. There was extensive overlap of the dendritic trees of neighbouring cells leading to an estimated coverage factor of 2-4. The dendritic field areas varied in size according to the retinal position of the cells and were highest in the central retina around the fovea with a maximum of 0.14 mm(2) and reached a second maximum at the retinal margin with values of 0.08-0.1 mm(2). The smallest dendritic areas (0.04-0.06 mm(2)) were measured midway between the fovea and retinal margin. The size of the soma area was not correlated to the dendritic field size and increased from 100 to 150 microm(2) near the fovea to 150-300 microm(2) at the retinal margin. There was no evidence for a retinotopic organisation of ganglion cell fibres within the nBOR. All cells were of uniform morphology that was identical to the type of nBOR-projecting displaced ganglion cell (DGC) described previously for the bird retina. Similar to birds, the labelled DGCs were the only source of retinal projection to the nBOR. A small fraction of cells had orthotopic somata located in the ganglion cell layer but were otherwise identical to the labelled DGCs. The similarity of chameleon nBOR-projecting ganglion cells to those described in avian retinas mirrors the close phylogenetic relationship of birds and lizards.