Chromatic discrimination in fixed saturation levels from tufted capuchin monkeys with different color vision genotypes.

Recent research has proposed new approaches to investigate color vision in Old World Monkeys by measuring suprathreshold chromatic discrimination. In this study, we aimed to extend this approach to New World Monkeys with different color vision genotypes by examining their performance in chromatic discrimination tasks along different fixed chromatic saturation axes. Four tufted capuchin monkeys were included in the study, and their color vision genotypes were one classical protanope, one classical deuteranope, one non-classical protanope, and a normal trichromat. During the experiments, the monkeys were required to perform a chromatic discrimination task using pseudoisochromatic stimuli with varying target saturations of 0.06, 0.04, 0.03, and 0.02 u'v' units. The number of errors made by the monkeys along different chromatic axes was recorded, and their performance was quantified using the binomial probability of their hits during the tests. Our results showed that dichromatic monkeys made more errors near the color confusion lines associated with their specific color vision genotypes, while the trichromatic monkey did not demonstrate any systematic errors. At high chromatic saturation, the trichromatic monkey had significant hits in the chromatic axes around the 180° chromatic axis, whereas the dichromatic monkeys had errors in colors around the color confusion lines. At lower saturation, the performance of the dichromatic monkeys became more challenging to differentiate among the three types, but it was still distinct from that of the trichromatic monkey. In conclusion, our findings suggest that high saturation conditions can be used to identify the color vision dichromatic phenotype of capuchin monkeys, while low chromatic saturation conditions enable the distinction between trichromats and dichromats. These results extend the understanding of color vision in New World Monkeys and highlight the usefulness of suprathreshold chromatic discrimination measures in exploring color vision in non-human primates.

Chromatic discrimination in fixed saturation levels from trichromats and subjects with congenital color vision deficiency.

Color vision tests use estimative of threshold color discrimination or number of correct responses to evaluate performance in chromatic discrimination tasks. Both approaches have advantages and disadvantages. In the present investigation, we compared the number of errors during color discrimination task in normal trichromats and participants with color vision deficiency (CVD) using pseudoisochromatic stimuli at fixed saturation levels. We recruited 28 normal trichromats and eight participants with CVD. Cambridge Color Test was used to categorize their color vision phenotype, and those with a phenotype suggestive of color vision deficiency had their L- and M-opsin genes genotyped. Pseudoisochromatic stimuli were shown with target chromaticity in 20 vectors radiating from the background chromaticity and saturation of 0.06, 0.04, 0.03, 0.02, 0.01, and 0.005 u'v' units. Each stimulus condition appeared in four trials. The number of errors for each stimulus condition was considered an indicator of the participant's performance. At high chromatic saturation, there were fewer errors from both phenotypes. The errors of the normal trichromats had no systematic variation for high saturated stimuli, but below 0.02 u'v' units, there was a discrete prevalence of tritan errors. For participants with CVD, the errors happened mainly in red-green chromatic vectors. A three-way ANOVA showed that all factors (color vision phenotype, stimulus saturation, and chromatic vector) had statistically significant effects on the number of errors and that stimulus saturation was the most important main effect. ROC analysis indicated that the performance of the fixed saturation levels to identify CVD was better between 0.02 and 0.06 u'v' units reaching 100%, while saturation of 0.01 and 0.005 u'v' units decreased the accuracy of the screening of the test. We concluded that the color discrimination task using high saturated stimuli separated normal trichromats and participants with red-green color vision deficiencies with high performance, which can be considered a promising method for new color vision tests based in frequency of errors.

Simultaneous Expression of UV and Violet SWS1 Opsins Expands the Visual Palette in a Group of Freshwater Snakes.

Snakes are known to express a rod visual opsin and two cone opsins, only (SWS1, LWS), a reduced palette resulting from their supposedly fossorial origins. Dipsadid snakes in the genus Helicops are highly visual predators that successfully invaded freshwater habitats from ancestral terrestrial-only habitats. Here, we report the first case of multiple SWS1 visual pigments in a vertebrate, simultaneously expressed in different photoreceptors and conferring both UV and violet sensitivity to Helicops snakes. Molecular analysis and in vitro expression confirmed the presence of two functional SWS1 opsins, likely the result of recent gene duplication. Evolutionary analyses indicate that each sws1 variant has undergone different evolutionary paths with strong purifying selection acting on the UV-sensitive copy and dN/dS ∼1 on the violet-sensitive copy. Site-directed mutagenesis points to the functional role of a single amino acid substitution, Phe86Val, in the large spectral shift between UV and violet opsins. In addition, higher densities of photoreceptors and SWS1 cones in the ventral retina suggest improved acuity in the upper visual field possibly correlated with visually guided behaviors. The expanded visual opsin repertoire and specialized retinal architecture are likely to improve photon uptake in underwater and terrestrial environments, and provide the neural substrate for a gain in chromatic discrimination, potentially conferring unique color vision in the UV-violet range. Our findings highlight the innovative solutions undertaken by a highly specialized lineage to tackle the challenges imposed by the invasion of novel photic environments and the extraordinary diversity of evolutionary trajectories taken by visual opsin-based perception in vertebrates.

Uniform trichromacy in Alouatta caraya and Alouatta seniculus: behavioural and genetic colour vision evaluation.

Primate colour vision depends on a matrix of photoreceptors, a neuronal post receptoral structure and a combination of genes that culminate in different sensitivity through the visual spectrum. Along with a common cone opsin gene for short wavelengths (sws1), Neotropical primates (Platyrrhini) have only one cone opsin gene for medium-long wavelengths (mws/lws) per X chromosome while Paleotropical primates (Catarrhini), including humans, have two active genes. Therefore, while female platyrrhines may be trichromats, males are always dichromats. The genus Alouatta is inferred to be an exception to this rule, as electrophysiological, behavioural and molecular analyses indicated a potential for male trichromacy in this genus. However, it is very important to ascertain by a combination of genetic and behavioural analyses whether this potential translates in terms of colour discrimination capability. We evaluated two howler monkeys (Alouatta spp.), one male A. caraya and one female A. seniculus, using a combination of genetic analysis of the opsin gene sequences and a behavioral colour discrimination test not previously used in this genus. Both individuals completed the behavioural test with performances typical of trichromatic colour vision and the genetic analysis of the sws1, mws, and lws opsin genes revealed three different opsin sequences in both subjects. These results are consistent with uniform trichromacy in both male and female, with presumed spectral sensitivity peaks similar to Catarrhini, at ~ 430 nm, 532 nm, and 563 nm for S-, M- and L-cones, respectively.

Morphological Plasticity of the Retina of Viperidae Snakes Is Associated With Ontogenetic Changes in Ecology and Behavior.

Snakes of the Viperidae family have retinas adapted to low light conditions, with high packaging of rod-photoreceptors containing the rhodopsin photopigment (RH1), and three types of cone-photoreceptors, large single and double cones with long-wavelength sensitive opsins (LWS), and small single cones with short-wavelength sensitive opsins (SWS1). In this study, we compared the density and distribution of photoreceptors and ganglion cell layer (GCL) cells in whole-mounted retinas of two viperid snakes, the lancehead Bothrops jararaca and the rattlesnake Crotalus durissus, and we estimated the upper limits of spatial resolving power based on anatomical data. The ground-dwelling C. durissus inhabits savannah-like habitats and actively searches for places to hide before using the sit-and-wait hunting strategy to ambush rodents. B. jararaca inhabits forested areas and has ontogenetic changes in ecology and behavior. Adults are terrestrial and use similar hunting strategies to those used by rattlesnakes to prey on rodents. Juveniles are semi-arboreal and use the sit-and-wait strategy and caudal luring to attract ectothermic prey. Our analyses showed that neuronal densities were similar for the two species, but their patterns of distribution were different between and within species. In adults and juveniles of C. durissus, cones were distributed in poorly defined visual streaks and rods were concentrated in the dorsal retina, indicating higher sensitivity in the lower visual field. In adults of B. jararaca, both cones and rods were distributed in poorly defined visual streaks, while in juveniles, rods were concentrated in the dorsal retina and cones in the ventral retina, enhancing sensitivity in the lower visual field and visual acuity in the upper field. The GCL cells had peak densities in the temporal retina of C. durissus and adults of B. jararaca, indicating higher acuity in the frontal field. In juveniles of B. jararaca, the peak density of GCL cells in the ventral retina indicates better acuity in the upper field. The estimated visual acuity varied from 2.3 to 2.8 cycles per degree. Our results showed interspecific differences and suggest ontogenetic plasticity of the retinal architecture associated with changes in the niche occupied by viperid snakes, and highlight the importance of the retinal topography for visual ecology and behavior of snakes.

Specificity of the chromatic noise influence on the luminance contrast discrimination to the color vision phenotype.

Many studies have examined how color and luminance information are processed in the visual system. It has been observed that chromatic noise masked luminance discrimination in trichromats and that luminance thresholds increased as a function of noise saturation. Here, we aimed to compare chromatic noise inhibition on the luminance thresholds of trichromats and subjects with severe deutan or protan losses. Twenty-two age-matched subjects were evaluated, 12 trichromats and 10 with congenital color vision impairment: 5 protanopes/protanomalous, and 5 deuteranopes/deuteranomalous. We used a mosaic of circles containing chromatic noise consisting of 8 chromaticities around protan, deutan, and tritan confusion lines. A subset of the circles differed in the remaining circles by the luminance arising from a C-shaped central target. All the participants were tested in 4 chromatic noise saturation conditions (0.04, 0.02, 0.01, 0.005 u'v' units) and 1 condition without chromatic noise. We observed that trichromats had an increasing luminance threshold as a function of chromatic noise saturation under all chromatic noise conditions. The subjects with color vision deficiencies displayed no changes in the luminance threshold across the different chromatic noise saturations when the noise was composed of chromaticities close to their color confusion lines (protan and deutan chromatic noise). However, for tritan chromatic noise, they were found to have similar results to the trichromats. The use of chromatic noise masking on luminance threshold estimates could help to simultaneously examine the processing of luminance and color information. A comparison between luminance contrast discrimination obtained from no chromatic and high-saturated chromatic noise conditions could be initially undertaken in this double-duty test.

Genetic variability of the sws1 cone opsin gene among New World monkeys.

Vision is a major sense for Primates and the ability to perceive colors has great importance for the species ecology and behavior. Visual processing begins with the activation of the visual opsins in the retina, and the spectral absorption peaks are highly variable among species. In most Primates, LWS/MWS opsins are responsible for sensitivity to long/middle wavelengths within the visible light spectrum, and SWS1 opsins provide sensitivity to short wavelengths, in the violet region of the spectrum. In this study, we aimed to investigate the genetic variation on the sws1 opsin gene of New World monkeys (NWM) and search for amino acid substitutions that might be associated with the different color vision phenotypes described for a few species. We sequenced the exon 1 of the sws1 opsin gene of seven species from the families Callitrichidae, Cebidae, and Atelidae, and searched for variation at the spectral tuning sites 46, 49, 52, 86, 90, 93, 114, 116, and 118. Among the known spectral tuning sites, only residue 114 was variable. To investigate whether other residues have a functional role in the SWS1 absorption peak, we performed computational modeling of wild-type SWS1 and mutants A50I and A50V, found naturally among the species investigated. Although in silico analysis did not show any visible effect caused by these substitutions, it is possible that interactions of residue 50 with other sites might have some effect in the spectral shifts in the order of ~14 nm, found among the NWM. We also performed phylogenetic reconstruction of the sws1 gene, which partially recovered the species phylogeny. Further studies will be important to uncover the mutations responsible for the phenotypic variability of the SWS1 of NWM, and how spectral tuning may be associated with specific ecological features such as preferred food items and habitat use.

Adaptations and evolutionary trajectories of the snake rod and cone photoreceptors.

Most vertebrates have duplex retinas, with two classes of photoreceptors, rods and cones. In the group of Snakes, however, distinct patterns of retinal morphology are associated with transitions between diurnal-nocturnal habits and reflect important adaptations of their visual system. Pure-cone, pure-rod and duplex retinas were described in different species, and this variability led Gordon Walls (1934) to formulate the transmutation theory, which suggests that rods and cones are not fixed entities, but can assume transitional states. Three opsin genes are expressed in retinas of most snake species, lws, rh1, and sws1, and recent studies have shown that the rhodopsin gene, rh1, is expressed in pure-cone retinas of diurnal snakes. This expression raised many questions about the nature of transmutation and functional aspects of the rhodopsin in a cone-like photoreceptor. Extreme differences in the retinal architecture of diurnal and nocturnal snakes also highlight the complexity of adaptations of their visual structures, which might have contributed to the adaptive radiation of this group and will be discussed in this review.

LWS visual pigment in owls: Spectral tuning inferred by genetics.

Owls constitute a diverse group of raptors, active at different times of the day with distinct light conditions that might be associated with multiple visual adaptations. We investigated whether shifts in the spectral sensitivity of the L cone visual pigment, as inferred by analysis of gene structure, could be one such adaptive mechanism. Using Sanger sequencing approach, we characterized the long wavelength-sensitive (LWS) opsin gene expressed in the retina of five owl species, specifically chosen to represent distinct patterns of activity. Nocturnality was epitomized by the American barn owl (Tyto furcata), the striped owl (Asio clamator), and the tropical screech owl (Megascops choliba); diurnality, by the ferruginous pygmy owl (Glaucudium brasilianum); and cathemerality, by the burrowing owl (Athene cunicularia). We also analyzed the presence of the L cone in the retinas of four species of owl (T. furcata, A. cunicularia, G. brasilianum and M. choliba) using immnunohistochemistry. Five critical sites for the spectral tuning of the LWS opsin (164, 181, 261, 269, and 292) were analyzed and compared to the sequence of other birds. The sequence of A. cunicularia showed a substitution on residue 269, with the presence of an alanine instead threonine, which generates an estimated maximum absorption (λmax) around 537 nm. No other variation was found in the spectral tuning sites of the LWS opsin, among the other species, and the λmax was estimated at around 555 nm. The presence of L cones in the retinas of the four species of owls was revealed using immunohistochemistry and we observed a reduced number of L cones in T. furcata compared to A. cunicularia, G. brasilianum and M. choliba.

Characterization of the melanopsin gene (Opn4x) of diurnal and nocturnal snakes.

BACKGROUND: A number of non-visual responses to light in vertebrates, such as circadian rhythm control and pupillary light reflex, are mediated by melanopsins, G-protein coupled membrane receptors, conjugated to a retinal chromophore. In non-mammalian vertebrates, melanopsin expression is variable within the retina and extra-ocular tissues. Two paralog melanopsin genes were classified in vertebrates, Opn4x and Opn4m. Snakes are highly diversified vertebrates with a wide range of daily activity patterns, which raises questions about differences in structure, function and expression pattern of their melanopsin genes. In this study, we analyzed the melanopsin genes expressed in the retinas of 18 snake species from three families (Viperidae, Elapidae, and Colubridae), and also investigated extra-retinal tissue expression. RESULTS: Phylogenetic analysis revealed that the amplified gene belongs to the Opn4x group, and no expression of the Opn4m was found. The same paralog is expressed in the iris, but no extra-ocular expression was detected. Molecular evolutionary analysis indicated that melanopsins are evolving primarily under strong purifying selection, although lower evolutionary constraint was detected in snake lineages (ω = 0.2), compared to non-snake Opn4x and Opn4m (ω = 0.1). Statistical analysis of selective constraint suggests that snake phylogenetic relationships have driven stronger effects on melanopsin evolution, than the species activity pattern. In situ hybridization revealed the presence of melanopsin within cells in the outer and inner nuclear layers, in the ganglion cell layer, and intense labeling in the optic nerve. CONCLUSIONS: The loss of the Opn4m gene and extra-ocular photosensitive tissues in snakes may be associated with a prolonged nocturnal/mesopic bottleneck in the early history of snake evolution. The presence of melanopsin-containing cells in all retinal nuclear layers indicates a globally photosensitive retina, and the expression in classic photoreceptor cells suggest a regionalized co-expression of melanopsin and visual opsins.

Photoreceptors morphology and genetics of the visual pigments of Bothrops jararaca and Crotalus durissus terrificus (Serpentes, Viperidae).

Snakes inhabit a great variety of habitats, whose spectral quality of light may vary a lot and influence specific adaptations of their visual system. In this study, we investigated the genetics of the visual opsins and the morphology of retinal photoreceptors, of two nocturnal snakes from the Viperidae family, Bothrops jararaca and Crotalus durissus terrificus, which inhabit preferentially the Atlantic Rain Forest and the Brazilian Savannah, respectively. Total RNA was extracted from homogenized retinas and converted to cDNA. The opsin genes expressed in snake retinas, LWS, RH1, and SWS1, were amplified by polymerase chain reactions (PCRs) and sequenced. The absorption peak (λmax) of the opsins were estimated based on amino acids located at specific spectral tuning sites. Photoreceptor cell populations were analyzed using immunohistochemistry with anti-opsin antibodies. Results showed the same morphological cell populations and same opsins absorption peaks, in both viperid species: double and single cones with LWS photopigment and λmax at ∼555 nm; single cones with SWS1 photopigment and λmax at ∼360 nm; and rods with the rhodopsin RH1 photopigment and λmax at ∼500 nm. The results indicate adaptations to nocturnal habit in both species despite the differences in habitat, and the possibility of a dichromatic color vision at photopic conditions.

Electrodiagnosis of dichromacy.

Retinal and cortical signals initiated by a single cone type can be recorded using the spectral compensation (or silent substitution) paradigm. Moreover, responses to instantaneous excitation increments combined with gradual excitation decreases are dominated by the response to the excitation increment. Similarly, the response to a sudden excitation decrement dominates the overall response when combined with a gradual excitation increase. Here ERGs and VEPs were recorded from 34 volunteers [25.9 ±â€¯10.4 years old (mean ±â€¯1 SD); 25 males, 9 females] to sawtooth flicker (4 Hz) stimuli that elicited L- or M-cone responses using triple silent substitution. The mean luminance (284 cd/m2) and the mean chromaticity (x = 0.5686, y = 0.3716; CIE 1931 color space) remained constant and thus the state of adaptation was the same in all conditions. Color discrimination thresholds along protan, deutan, and tritan axes were obtained from all participants. Dichromatic subjects were genetically characterized by molecular analysis of their opsin genes. ERG responses to L-cone stimuli were absent in protanopes whereas ERG responses to M-cone stimuli were strongly reduced in deuteranopes. Dichromats showed generally reduced VEP amplitudes. Responses to cone-specific stimuli obtained with standard electrophysiological methods may give the same classification as that obtained with the Cambridge Colour Test and in some cases with the genetic analysis of the L- and M-opsin genes. Therefore, cone-specific ERGs and VEPs may be reliable methods to detect cone dysfunction. The present data confirm and emphasize the potential use of cone-specific stimulation, combined with standard visual electrodiagnostic protocols.

Effect of the Decrease in Luminance Noise Range on Color Discrimination of Dichromats and Trichromats.

Color vision assessment can be done using pseudoisochromatic stimuli, which has a luminance noise to eliminate brightness differences between the target and background of the stimulus. It is not clear the influence of the luminance noise on color discrimination. We investigated the effect of change in the luminance noise limits on color discrimination. Eighteen trichromats and ten congenital dichromats (eight protans, two deutans) had their color vision evaluated by the Cambridge Colour Test, and were genetically tested for diagnostic confirmation. The stimuli were composed of a mosaic of circles in a 5° circular field. A subset of the circles differed in chromaticity from the remaining field, forming a letter C. Color discrimination was estimated in stimulus conditions differing in luminance noise range: (i) 6-20 cd/m2; (ii) 8-18 cd/m2; (iii) 10-16 cd/m2; and (iv) 12-14 cd/m2. Six equidistant luminance values were used within the luminance noise limits with the mean stimulus luminance maintained constant under all conditions. A four-alternative, forced-choice method was applied to feed a staircase procedure to estimate color discrimination thresholds along eight chromatic axes. An ellipse model was adjusted to the eight color discrimination thresholds. The parameters of performance were threshold vector lengths and the ellipse area. Results were compared using the Kruskal-Wallis test with a significance level of 5%. The linear function model was applied to analyze the dependence of the discrimination parameters on the noise luminance limits. The first derivative of linear function was used as an indicator of the rate of change in color discrimination as a function of luminance noise changes. The rate of change of the ellipse area as a function of the luminance range in dichromats was higher than in trichromats (p < 0.05). Significant difference was also found for individual thresholds in half of the axes we tested. Luminance noise had a greater effect on color discrimination ability of dichromats than the trichromats, especially when the chromaticities were close to their protan and deutan color confusion lines.

Electroretinographical determination of human color vision type.

It has been previously demonstrated that electroretinography (ERG) elicited by heterochromatically modulated stimuli can be used for objective determination of color vision type. Color vision of trichromatic, deuteranopic, and protanopic participants was psychophysically assessed by the Cambridge Color Test and confirmed genetically. ERG responses to red and green lights modulating in counterphase at 12 and 36 Hz were recorded, while the fraction of red modulation was varied. At 36 Hz (and second harmonics at 12 Hz), the responses were minimal at red fractions that differed significantly in protanopes. At 12 Hz (fundamental component), the responses of the trichromats differed significantly compared to those of the dichromats. An improved protocol shows that the three subject groups can be separated with no overlap.

L-/M-cone opponency in visual evoked potentials of human cortex.

L and M cones send their signals to the cortex using two chromatic (parvocellular and blue-yellow koniocellular) and one luminance (magnocellular) pathways. These pathways contain ON and OFF subpathways that respond to excitation increments and decrements respectively. Here, we report on visually evoked potentials (VEP) recordings that reflect L- and M-cone driven increment (LI and MI) and decrement (LD and MD) activity. VEP recordings were performed on 12 trichromats and four dichromats (two protanopes and two deuteranopes). We found that the responses to LI strongly resembled those to MD, and that LD and MI responses were very similar. Moreover, the lack of a photoreceptor type (L or M) in the dichromats led to a dominance of the ON pathway of the remaining photoreceptor type. These results provide electrophysiological evidence that antagonistic L/M signal processing, already present in the retina and the lateral geniculate nucleus (LGN), is also observed at the visual cortex. These data are in agreement with results from human psychophysics where MI stimuli lead to a perceived brightness decrease whereas LI stimuli resulted in perceived brightness increases. VEP recording is a noninvasive tool that can be easily and painlessly applied. We propose that the technique may provide information in the diagnosis of color vision deficiencies.

A novel nonsense mutation in the tyrosinase gene is related to the albinism in a capuchin monkey (Sapajus apella).

BACKGROUND: Oculocutaneous Albinism (OCA) is an autosomal recessive inherited condition that affects the pigmentation of eyes, hair and skin. The OCA phenotype may be caused by mutations in the tyrosinase gene (TYR), which expresses the tyrosinase enzyme and has an important role in the synthesis of melanin pigment. The aim of this study was to identify the genetic mutation responsible for the albinism in a captive capuchin monkey, and to describe the TYR gene of normal phenotype individuals. In addition, we identified the subject's species. RESULTS: A homozygous nonsense mutation was identified in exon 1 of the TYR gene, with the substitution of a cytosine for a thymine nucleotide (C64T) at codon 22, leading to a premature stop codon (R22X) in the albino robust capuchin monkey. The albino and five non-albino robust capuchin monkeys were identified as Sapajus apella, based on phylogenetic analyses, pelage pattern and geographic provenance. One individual was identified as S. macrocephalus. CONCLUSION: We conclude that the point mutation C64T in the TYR gene is responsible for the OCA1 albino phenotype in the capuchin monkey, classified as Sapajus apella.

Comparative study of photoreceptor and retinal ganglion cell topography and spatial resolving power in Dipsadidae snakes.

The diurnal Dipsadidae snakes Philodryas olfersii and P. patagoniensis are closely related in their phylogeny but inhabit different ecological niches. P. olfersii is arboreal, whereas P. patagoniensis is preferentially terrestrial. The goal of the present study was to compare the density and topography of neurons, photoreceptors, and cells in the ganglion cell layer in the retinas of these two species using immunohistochemistry and Nissl staining procedures and estimate the spatial resolving power of their eyes based on the ganglion cell peak density. Four morphologically distinct types of cones were observed by scanning electron microscopy, 3 of which were labeled with anti-opsin antibodies: large single cones and double cones labeled by the antibody JH492 and small single cones labeled by the antibody JH455. The average densities of photoreceptors and neurons in the ganglion cell layer were similar in both species (∼10,000 and 7,000 cells·mm(-2), respectively). The estimated spatial resolving power was also similar, ranging from 2.4 to 2.7 cycles·degree(-1). However, the distribution of neurons had different specializations. In the arboreal P. olfersii, the isodensity maps had a horizontal visual streak, with a peak density in the central region and a lower density in the dorsal retina. This organization might be relevant for locomotion and hunting behavior in the arboreal layer. In the terrestrial P. patagoniensis, a concentric pattern of decreasing cell density emanated from an area centralis located in the naso-ventral retina. Lower densities were observed in the dorsal region. The ventrally high density improves the resolution in the superior visual field and may be an important adaptation for terrestrial snakes to perceive the approach of predators from above.