Anatomy and Histology of the Eye of the Nutria Myocastor coypus: Evidence of Adaptation to a Semi-aquatic Life

Abstract

The nutria is a large, semi-aquatic rodent that, being invasive, is having a growing impact on the ecosystem in western Japan. Knowledge regarding physical adaptations to the nutria’s lifestyle and habitual activities would be useful for effectively controlling and preventing their spread. Nutrias spend time on land and in water, feeding on agricultural crops and wild grasses growing near the waterside, as well as aquatic plants and shellfish. In the current study, the nutria’s visual organ was analyzed anatomically and histologically, and aquatic and light environmental adaptations were evaluated. The results revealed that the nutria eyeball was almost spherical, and the cornea was rounded. The lens was convex and slightly thicker than previously reported for other rodents. These features were not characteristic of aquatic adaptations observed in the eyes of fish or marine mammals. The ratio of lens diameter to eyeball diameter was 0.6, similar to that of nocturnal species. The pupil was a vertical slit, suggesting an ability to adjust the amount of light entering the eyeball during twilight. Photoreceptors were sparsely distributed across the whole retina, and no fovea was observed. Retinal thickness was 90–100 μm, thinner than that in other rodent species. Visual acuity was 1.44–1.58 cycles/degree, higher than that in other rodents, likely because of the nutria’s large eyeball and body. These results suggest that the nutria visual system is adapted to recognize large shadows of distant predators rather than viewing objects in detail.

BACKGROUND

The nutria (Myocastor coypus; Molina, 1782) belongs to Myocastoridae, in the Caviomorpha superfamily of the Rodentiae (Upham and Patterson 2012). The nutria is native to South America, and five subspecies have been identified to date (Osgood 1943). In Japan, nutrias were introduced as a fur animal in the early 1900s as husbandry was being promoted (Miura 1976; Iwasa 2010). After World War II, the Japanese government promoted nutria breeding as a countermeasure to extreme food shortages; however, once the food problem was solved, demand for nutrias declined and escaped individuals became wild (Kobayashi and Oda 2016). Nutrias are now widely distributed, with large numbers in western Japan (Carter and Leonald 2002; Iori et al. 2013; Kawamura et al. 2018).

Nutrias are almost entirely herbivorous and eat animal material, and live predominantly near water. They eat plants, herbs and crops growing along waterside zones in the evening and night-time, severely damaging agriculture (Kinler et al. 1987; Gosling and Baker 1991; Reggiani et al. 1993; Verheyden and Abbas 1996; Guichón et al. 2003). Nutrias have webbed hind legs from the first to fourth toes, swim in rivers and ponds, and can dive for more than 5 minutes (Evans 1970). The webbed hind legs allow nutrias to catch shellfish and other food in the water (LeBlanc 1994; Ishida et al. 2015), leading them to be classified as a semi-aquatic rodent (Newson and Holmes 1968; Ehrlich 1967). In 2005, Japan’s Ministry of the Environment designated the species as invasive and began taking measures to control and prevent its diffusion. It is important to understand the behavioral ecology of this species to carry out these measures efficiently. Some researchers studying the ecology of nutria have suggested that the species is so invasive in part because of its ability to forage for food both on land and underwater.

Vision is an important sensory modality for most vertebrates, and a close relationship between functional properties and species-specific ecology has been reported (Ali 1978; Finlay and Sengelaub 1981; Fleishman 1992; Archer et al. 1999; Land and Nilsson 2002). The lens shape is a distinctive characteristic of aquatic adaptation in the eye; the lens is convex in terrestrial animals, whereas fish and cephalopods have spherical lenses. In the former, because the refractive index of the cornea is significantly different from that of the air, the cornea plays a role in refracting light and focusing images on the retina. However, for aquatic animals, because the refractive index of the cornea is almost equal to that of water, the cornea cannot sufficiently fulfill the function of refracting light. The surface of a spherical lens, having a high degree of curvature, plays a role in focusing images on the retina owing to its high refractive index. The lenses of aquatic mammals, such as dolphins and seals, are also approximately spherical.

The topographical distribution of ganglion cells in the whole retina has been examined in various animals and revealed that cells are densely distributed in the retinal area where the image is focused. For example, in fish, the coral cod Cephalopholis miniate—a predator that ambushes prey from within dark caves—possesses a high-density area in the temporal retina (Collin and Pettigrew 1988a). The horizontal streak, a high-density zone across the retinal meridian (dorso-temporal), is commonly found in fish that inhabit open water, suggesting that this feature helps these fish search for prey with an uninterrupted view of the sand-water horizon (Collin and Pettigrew 1988b). The distribution of ganglion cells in the retina has also been examined for marine mammals. In the retina of the bottlenose dolphin Tursiops truncates (Mass and Supin 1995), gray whale Eschrichtius gibbosus (Mass and Supin 1997) and beluga (Delphinapterus leucas) (Mass and Supin 2002), high-density areas have been reported in the temporal and nasal areas, with the highest density in the former. Mass and Supin (1995 1997 2002) suggested that the temporal area with high cell concentration is used for forward vision both underwater and in the air, whereas the nasal high cell area appears to be related to the marine mammals when they project onto an object of the postero-lateral part of the visual field. The Amazon river dolphin Inia geoffrensis reportedly has a high-density area only in the ventral retina, which is considered to be appropriate for viewing objects on the water’s surface, such as prey species in turbid water coming to the surface (Mass and Supin 1989). Two areas with high cellular density in the fovea and the temporal fossa found in the retina of diurnal bird species (Fite and Rosenfield-Wessels 1975; Reymond 1985 1987; Inzunza et al. 1991; Rahman et al. 2006) are considered to contribute to monocular vision and binocular visual field formation, respectively (Moore et al. 2015). In the rabbit retina, a horizontal high cellular density zone was reported from the rostral to the caudal region, which is thought to expand rabbits’ field of vision horizontally and protect it from predators (Hughes 1977).

Lens size is an index of how efficiently the eye can gather light. Nocturnal species have larger lenses relative to diurnal species (Ňemec et al. 2008; Schmitz and Wainwright 2011) and greater lens thickness (Van der Merwe et al. 2018). Thick lenses are rigid and unable to focus, but the lenses of nocturnal animals are thought to not require focusing (Baradia et al. 2010; Pardue et al. 2013).

The amount of light entering the eye is controlled by the pupil. Malmström and Kröger (2006) reported that animals with ring-shaped pupils are active even in low light conditions, and their lenses are thick and approximately spherical. In addition, the authors suggest that horizontal and vertical slit-type pupils are frequently observed in nocturnal animals, and that when the pupils are maximized, the lens adapts to multiple focal lengths because of chromatic aberrations, enabling the eyes to focus (Malmström and Kröger 2006). Banks et al. (2015) observed a striking correlation between the shape of terrestrial animal pupils, their foraging mode and the time of day they are active. The author suggests that species with elongated vertical pupils are more likely to be ambushed by predators and active day and night; in addition, the pupils of prey species extend horizontally and their eyes are positions horizontally to detect predators.

In the current study, we anatomically investigated the eyeball and crystalline lens of the nutria and histologically analyzed its retina. Using these analyses, we examined whether the nutria’s visual organ has properties that are useful for underwater activity and/or for the light environment to which the animal is adapted.

MATERIALS AND METHODS

Animals

Nutrias were captured by hunters engaged in controlling wild populations. This hunting was performed with official legal permission from a government office. Five cubs (A-C1–C5, Table 1) and five adults (A-A1–A5, Table 1) were captured in Aichi Prefecture from 2000 to 2001 and transferred to the School of Dentistry, Aichi Gakuin University, where their whole bodies were fixed in 10% formalin solution. Both eyeballs of each individual were enucleated from the head and measured for external characteristics. However, the retinas of these animals could not be analyzed because the tissue was degraded. In Okayama Prefecture, nutrias were captured using traps in 2016. Live individuals were transferred to an operating room in Okayama University of Science, euthanized, perfusion fixed with 10% formalin, and both eyeballs of each individual were enucleated. The external characteristics of eyeballs were measured and the retinas were histologically analyzed. The left eyeballs of O-A3 and O-A4 and the right eyeball of O-A5 were removed before formalin fixation and used for molecular experiments in another study. The time of death of the animals captured in Aichi is unknown. Surgery was performed during the day on the individuals captured in Okayama.

Our study was conducted in accord with the Declaration of Helsinki guidelines and was approved by the Ethics Review Committee for Animal Experimentation at Okayama University of Science.

Table 1.

General features of the nutria eyeball

Measurements of external characteristics

Muscle tissues outside of the eyeball were removed using fine point forceps and ophthalmology scissors under a stereomicroscope (Stemi 305, Zeiss). Using a digital caliper, the diameter of the eyeball at four bearing angles (ED1−ED4, Fig. 1A) and the axial length (Ax. L, Fig. 1B) were measured. The diameter of each eyeball was calculated as the average of these four values. The length and width of the pupil were measured on the surface of the cornea (PL; PW, Fig. 1A). The cornea and iris were then surgically separated at the retinal periphery, the lens was removed from the eyecup, and the diameter and thickness were measured (LD; LT, Fig. 1C). The specimens from Aichi Prefecture were slightly deformed and thus the axial lengths of the eyeball and pupil diameter were not measured. The other individuals’ formalin-fixed eyeballs were dissected, and paraffin sections of the eye-cup around the ciliary body at 5 μm thickness were prepared (using the method described below).

Fig. 1.

Frontal view (a) and lateral view (b) of nutria eyeball. The eyeball of the adult nutria is approximately spherical. The pupil seems to be vertically elongated (a) and the cornea surface is curved convexly (b). The lens is ellipsoidal shape compressed in the axial direction (c). Lateral view of the lens (c). Eyeball diameter was measured in four bearing angles (ED1~ED4). PL, pupil length; PW, pupil width; Ax. L, axial length; LD, lens diameter; LT, lens thickness. D, dorsa; V, ventral; A, anterior; P, posterior.

Retinal histology

The retina was flattened by making five cuts at the periphery of the eyecup, which was processed using the above method, then divided into small pieces (approximately 2 mm²) while taking notes so that the original position could be confirmed. The retinal pieces were dehydrated using an alcohol series, cleared with xylene, and embedded in paraffin. The retinas of O-A1, O-A2 and O-A5 were cut radially at 3 μm thickness and those of O-A3 and O-A4 were cut tangentially at 3 μm thickness. The sections were stained with hematoxylin and eosin and enclosed with glass cover slips.

Analysis of retinal thickness

Analysis of retinal thickness was conducted for the O-A2 retina. Radial sections—respective of the rostral, temporal, dorsal, ventral, and central retinal areas— were photographed using a digital camera (Axiocam, Zeiss) attached to a light microscope (Axio Imager A1, Zeiss). The retina thickness was measured for 10 regions on these digital photographs using microscope imaging software (Axiovision, Zeiss) with the light microscopy system.

Estimation of visual acuity

The histological visual acuity of the vertebrates was estimated from the maximum density of ganglion cells, which are bottleneck cells in the retinal neural circuit and the whole retina, and the distance from the center of the lens to the retina (posterior nodal distance; PND). When the ganglion cell layer of the nutria retina was observed under a light microscope, cell density appeared to be very sparse (0 to several cells per 100 μm2) and it was difficult to obtain a cell count value for each retinal piece. Therefore, we counted the number of photoreceptor nuclei, which included both rod and cone cells, and calculated visual acuity as described below.

Rod cells are highly sensitive to light and are associated with visual function at a very low lighting level (scotopic-vision), whereas cone cells are active under high light levels and enable color vision and spatial perception (photopic-vision). Scotopic vision enables far lower spatial perception than photopic vision, so visual acuity was evaluated using photopic vision of the eye under well-lit conditions. The proportion of cone cells among the total photoreceptors in nocturnal rodents is reported to be 1–3% (Szél and Röhlich 1992; Jeon et al. 1998; Peichl 2005; Volland et al. 2015), and the ratio of the numbers of cone cells to ganglion cells is approximately 1:1 in the fovea centralis (a small region of densely packed cones) in mice (Volland et al. 2015). The number of retinal ganglion cells (Y) in nutria was therefore estimated from the number of photoreceptor nuclei (X) using the following equation:

Y = X × 0.01 × 1 = 0.01 X

The visual acuity (cycles per degree) was estimated as follows based on the method reported by Collin and Pettigrew (1989).

PND is 0.6 times the eye axial length (Ax. L) in many vertebrates (Hughes 1977; Martin 1982 1985; Hall and Ross 2007).

PND = 0.6 Ax.L

Here, the angle (α, radian) subtending in 1 mm on the retina was calculated using the following equations:

tanα = 1/PND

α = arctan (1/PND)

When the highest density of ganglion cells D (cell/mm²) (= 0.01 X) is regularly arranged in a 1 mm square, the number of ganglion cells on a 1 mm straight line (cells/mm) is expressed as follows:

√D

Because the spatial resolution is based on the number of cells at one diopter, the spatial resolution is calculated as:

ganglion cells per degree = √D/α

At least two ganglion cells are needed to distinguish one cycle of the highest resolution light-dark grid. The visual acuity (cycles per degree) is therefore evaluated by half of the cells per degree and calculated using the following equation:

cycles per degree = 1/2 cells per degree

For the O-A3 and O-A4 retinas (cut tangentially), the number of photoreceptor nuclei in a 20 μm × 20 μm area was counted on the photographs for each of the small pieces. Regarding the retina of O-A5 (cut radially), nuclei located within a 20 μm width in the outer nuclei layer were counted on the images. The diameter of the nucleus was 3 μm or less, so it was multiplied by 6.7 to be equivalent to the tangential section counts. The obtained values were converted to cells/mm².

Measurement of visual overlap

The overlap between the eyes of nutrias was measured on a head skeletal specimen made previously. The visual overlap between the two eyes was measured using a skull specimen of another individual made previously. We used a photograph of the side of the skeleton’s mandible and drew straight lines on the obtained image connecting the orbit and tip of the rostrum; the angle between the two lines was measured using Image-J software.

RESULTS

Eyeball, cornea, lens and pupil

No significant differences were found between the eyeball diameter (ED) and axial length (Ax. L) measured in five adults (O-A1–O-A5; t-test, p = 0.65), indicating that the eyeball of the adult nutria is approximately spherical. The cornea surface was curved convexly (Fig. 1B). The lens diameter (LD) was 15–49% greater than the lens thickness (LT) in cubs and 17–57% greater in adults (t-test, p < 0.01), having an ellipsoidal shape (Fig. 1C). The LD was 46–59% of the ED in cubs and 54–67% of the ED in adults. The pupil seemed to be vertically elongated with an almond-shaped opening (Fig. 1A); however, no significant difference was found between the vertical and horizontal lengths (t-test, p = 0.09) (Table 1).

Comparing eyeball traits between cubs and adults, measured in specimens from Aichi Prefecture, both ED and LD were significantly greater in adults than in cubs (t-test, p < 0.01), as was LT (t-test, p = 0.10) (Table 1). When the percentage ratio of LD to LT was determined for each individual, the values ranged from 115–149% in cubs and 115–157% in adults, with no significant difference between the two groups (t-test, p = 0.34) (Table 1).

Ciliary muscle

In the section of the eyecup around the ciliary body, both the cornea and continuously connected sclera were observed to be stained with eosin (Fig. 2). Gathered ciliary portions were darkly pigmented, and the anterior chamber was identified as a space between the cornea and ciliary portions. The ciliary body was connected near the seam between the cornea and sclera, and the muscle—which would be expected to be stained with eosin—was not clearly observed.

Fig. 2.

Radial section of the nutria eyeball around ciliary body. In the ciliary body, muscles, which should stain with eosin, could not be observed. Ac, Anterior chamber; Cb, Ciliary body; Co, Cornea; Cp, Ciliary portion; Sc, Sclera. Scale bar = 100 μm.

Retina

In the radial retinal sections, the outer pigmented layer and inner neural layers (the typical retinal 10-layered structure) were observed (Fig. 3). The pigmented epithelial layer was distinguished by a brown color. Retinal thickness was 86.60–104.20 μm in the different regions of the nasal, lateral, ventral, dorsal, and central retina. There was no significant difference in thickness among these five retinal regions (analysis of variance, p = 0.43; post-hoc analysis by Tukey’s test, p > 0.46) (Fig. 4).

Fig. 3.

Light photomicrographs from a radial section of the nutria retina. The retina is a 10-layered structure. The pigmented epithelial layer is a brown color. The photoreceptor layer is pink with eosin staining, and the cell nuclei in the inner and outer nuclear layer and ganglion cell layer are stained purple with hematoxylin. Scale bar = 10 μm.

Fig. 4.

Comparison of retinal thickness in different areas of the nutria retina. No significant difference in thickness was observed among retinal regions. D, dorsal; N, nasal; V, ventral; T, temporal; C, center. Vertical bar indicates standard deviation.

Distribution of photoreceptor nuclei and visual acuity

The number of photoreceptor nuclei of O-A3 and O-A4 retina were counted in tangential sectional images and converted to densities. The maximum density of the O-A3 and O-A4 retina was 57.5 × 10³ and 70.0 × 10³ cells/mm², respectively (Table 2). Nuclei of O-A5 were counted in images of radial sections, and the maximum densities were 82.5 × 10³ cells/mm². Nuclei on the cutting plane may not have been counted in the radial histological images, meaning that the obtained values would have been underestimated. No localization of the high-density region was observed in any of the retinas (Fig. 5). Visual acuity was calculated at the highest density and greatest diameter of the lens, with O-A3, O-A4 and O-A5 yielding 1.44, 1.46 and 1.58 cycles/degree (cpd), respectively (Table 2).

Table 2.

Summary of photoreceptor cell counts and calculations of visual acuity in nutrias

Fig. 5.

Topographical maps of photoreceptor nuclei of the nutria retina. In the three retinas analyzed, cell density was not equal for the different retinal areas, and there was no localization of the high-density region. (a) O-A3 right retina; (b) O-A4 right retina; (c) O-A5 left retina. Numbers in parentheses are the cell densities for the O-A5 map. N, nasal; V, ventral. Scale bar = 2 mm.

Visual overlap

The visual overlap between the two eyes of the nutria was observed to be about 56 degrees (Fig. 6).

Fig. 6.

Photograph of a head of the nutria in dorsal view. The dotted line indicates the outline of eyeballs and lenses, and the dashed line indicates anterior-posterior direction (axial direction). The straight line in red connects the retinal periphery to the lens center. The visual overlap between the two eyes is measured to be approximately 56 degrees.

DISCUSSION

Eyeball, cornea and lens shapes

In the human eye, the cornea refracts light to display images of objects on the retina. The lens is an ellipsoidal sphere and the focus is finely adjusted by the ciliary muscle, which changes the thickness of the lens. In fish living in water, however, the refractive index of the cornea is almost equal to that of water. Therefore, the notably spherical crystalline lens of fish refracts light, and focus is adjusted via the movement of the lens by the retractor lentis muscle (Somiya 1987).

Like fish, marine mammals have approximately spherical lenses adapted to an aquatic environment; their ciliary muscle is immature or absent and the lenses do not focus (Sivak 1980; Bjerager et al. 2003; Mass and Supin 2007; Buono et al. 2012). Although the nutria swims in water, the lens shape in both young and adult nutrias was ellipsoid, with LD being 15–49% greater than LT among young nutrias and 17–56% greater than LT among adult nutrias. These results suggest that underwater vision in nutrias is less advantageous than vision on land. Erb-Eigner et al. (2015) and van der Merwe et al. (2018) reported that LD/LT values in mice were 111–119%. Thus, the lens of the nutria is thicker than that of the mouse and would be expected to have a higher refractive index than the mouse lens. Even in nutrias, the ciliary body did not contain muscle, suggesting a loss of accommodation. A visual focusing model by Doonan (1984) proposes that the extraocular muscles are essential for changing lens curvature and focal length. The thickness of the lens also changes with changes in intraocular pressure. Thus, the evidence that animals lacking ciliary muscles cannot adjust their focus is lacking.

The LD/LT value of the rat lens is reported to be 113.7% for young animals and 111.6% for adult animals, with the lens becoming more spherical with age (Sivak and Dovat 1983). In the nutrias examined in the present study, the LD/LT values of cubs and adults were 115–149% and 117–157%, respectively, with no significant differences between the two groups (t-test, p = 0.25). Although it will be necessary to conduct measurements in younger individuals than the specimens in the current study, the refractive index of the lens of young nutrias with a body weight of approximately 1 kg is likely to be similar to that of adult animals.

Rigid lenses are common among nocturnal animal species that do not require high visual resolution (Sivak et al. 1989; Hall and Ross 2007; Pardue et al. 2013). The LD/ED ratio of nocturnal animals is reported to be approximately 0.6–0.8, twice that of diurnal animals (0.3–0.4) (Animal eye p84). Van der Merwe et al. (2018) calculated LD/ED values for diurnal (Rhabdomys pumilio) and nocturnal (Micaelamys namaquensis) mice, reporting a ratio of 0.50 in the former and 0.73 in the latter. The nutria has an LD/ED value of approximately 0.6, similar to that of other nocturnal species. Thus, the current results suggest that the nutria’s ellipsoidal lens has the characteristics of nocturnal rather than aquatic adaptation. For nocturnal animals, the ability of the lens to gather light is more important than the ability to focus the retina. In humans, peripheral vision dominated by rods is superior to the central retinal region concerning movement perception. The possibility that the nutria eye has excellent light-gathering properties is also supported by the shape of the cornea, as shown below.

The shape of the cornea is also known to reflect visual adaptation in relation to an animal’s particular lifestyle. In the eyes of fish adapted to aquatic environments, the cornea does not bend incoming light, and the refractive function relies on the lens (Winkler et al. 2015). Thus the fish cornea is almost flat and the lens is approximately spherical, similar to that of seals, a semi-aquatic mammal (Land and Nilsson 2002; Land 2005; Winkler et al. 2015). Our observations of the nutria eyeball in the current study revealed that the cornea is rounded (Fig. 1B), suggesting that this feature helps the lens focus. It has been suggested, however, that highly curved surfaces are impacted by spherical aberrations, weakening the ability to focus (de la Cera et al. 2006). The cornea, which corrects for this problem, is reported to be dome-shaped in diurnal animals such as humans, with different curvatures at the center and periphery (Land 2005). There is a tendency for dome-shaped corneas to be smaller and round-shaped ones to be larger (Kirk 2004). The curved cornea observed in nutrias is considered to be typical in nocturnal-type eyes, suggesting that light is refracted and focused equally at all meridians. Therefore, although image quality should be considered (as discussed below), nutrias may have a large field of view.

Pupil shape

The pupil of the nutria was almond-shaped, with a greater vertical length than horizontal length, although this difference was not statistically significant. Schmitz (2009) evaluated eyeball structures among various vertebrate species and identified a relationship of Y = X^0.94 between the pupil diameter (Y) when fully dilated and the lens diameter (X). When applying the measured values of five samples from Okayama Prefecture to this formula, O-A1 and O-A3 did not appear to match this relationship (Table S1). The other three individuals were anesthetized with a gas in the laboratory. However, O-A1 and O-A3 were examined after being killed by a hunter. It is likely that the pupils of O-A1 and O-A3 did not dilate because these two individuals did not die in a state of relaxation.

Many rodents—such as rats, mice, beavers and squirrels—have circular pupils. However, the degu and chinchilla have vertically long pupils, similar to those of the nutria. Regarding the role of the vertical slit pupil, the following three hypotheses have been proposed (Malmström and Kröger 2006): (1) a vertically long pupil can reduce the diameter in the horizontal direction more quickly than a spherical pupil, resulting in instantaneous control of the amount of incident light to the eyeball; (2) different curvatures at the pupil edges make it easier for different wavelengths of light to focus on the same location, reducing chromatic aberration and compensating for focus adjustment; and (3) the field of view in the grass is more open in the vertical axis, matching the vertical slit pupil, allowing animals to inhabit grass that is taller than their own body height and objects through gaps in the grass.

Malmström and Kröger (2006) classified the pupil shapes of terrestrial vertebrates into three types: circular, vertical slit, and horizontal slit. The authors argued that, when the pupil contracts, the circular type loses image focus because the lens peripheral zones become shaded, whereas slit-shaped pupils enable the use of the full diameter of the lens. However, it should be noted that slit-shaped pupils in nocturnal monkeys (e.g., Galago senegalensis, Nycticebus coucang) and monochromats (e.g., the octopus) do not enable multi-focus optics.

Banks et al. (2015) examined the relationship between pupil shape and ecological niche in terrestrial species and noted that vertically elongated pupils are much more common in ambush predators than in other species. These authors conclude that the vertically elongated pupil of these animals facilitates stereoscopic vision for estimating the distance of objects perched on the ground, while at the same time allowing for an accurate estimation of the distances to the animal along the ground. Although the nutria is highly herbivorous, they also eat a wide range of invertebrates such as land snails and insects (U.S. Fish and Wildlife Service 2015). Their vertically long pupil may help visualize these invertebrates. The visual overlap between the two eyes in the nutria, however, was about 56 degrees, different from the frontal-eye characteristic of ambush predators (Banks et al. 2015).

The vertical slit-shaped pupil allows for a wide range of light intensity by varying the pupil diameter. For example, in humans with circular pupils, the diameter during dark adaptation is about 15 times larger than that during light adaptation. However, among cats and geckos with vertical slits, the pupil diameter during dark adaptation is 135 and 300 times larger, respectively (de Groot and Gebhard 1952; Wilcox and Barlow 1975; Roth et al. 2009). The latter animals can significantly increase the amount of light collected at night compared to daytime. Nutria has a vertically elongated pupil and may demonstrate sufficient visual accuracy in bright and dark environments.

As described above, the LD/LT ratio of the nutria was found to be larger than that of many nocturnal animals, although the LD/ED ratio was within the range of other nocturnal species. The color perception of nutrias was not examined in the current study. However, related species of degu, guinea pig and chinchilla are reported to exhibit middle-wave-sensitive and ultraviolet (UV)-sensitive opsins, indicating that these rodents have color vision. If the nutria also has color vision, then it may benefit from slit-based multifocal function in low light conditions. Thus, besides its relationship to other animal species-specific ecology, all three of the benefits of the vertical slit pupil mentioned above may apply to nutrias that live in bushy environments and are active at dawn and dusk. On the other hand, daytime UV radiance can damage the cornea and lenses (Meyer-Rochow 2000). UV is highly attenuated by water, so the nutria can reduce the UV damage to their eyes throughout their lives by searching for food in the water.

Distribution of photoreceptor nuclei and visual acuity

In the human retina, cone photoreceptor cells are most densely distributed in the central region, called the fovea. Human eyes have the best visual acuity in direction at which the axis connecting the fovea of each eye and the center of the lens intersect; this is known as the visual axis. In other vertebrates, such as fish and birds, the visual axes are estimated from the highest density of ganglion cells, which is the bottleneck of optical information in the retina (Collin and Pettigrew 1989; Dolan and Fernández-Juricic 2010; Gomi and Miyazaki 2015; Miyazaki and Kobayashi 2015; Martin 2017). This is because the ratio of cone cells to ganglion cells in the fovea is approximately 1:1 in humans and mice (Volland et al. 2015), unlike other vertebrates. The nutria retinal histology in the present study revealed that both cone and ganglion cells were sparsely distributed in the retina, and it was difficult to construct a topo-graphical density map from the counts per unit area of these cells. Therefore, a density map of the number of photoreceptor nuclei (both cone and rod nuclei) in the entire retina was prepared. There are no localized areas in which cell nuclei were densely concentrated in these maps, indicating that nutria do not have particularly precise vision. Similarly, no localized areas of thickening were observed in the retinal thickness. As for the retinal thickness, there were no localized thicker areas. The sensitivity of the photoreceptor depends on the thickness and length of its outer segment (which contains the visual pigment) and the light absorption efficiency (Land 1981). Even in the same retina, these aspects can differ from the area. In the future, it will be necessary to analyze the details of the shape and size of photoreceptor cells to clarify whether there are areas in the nutria’s retina that are especially effective at collecting visual information.

In fish, the cone cell to ganglion cell ratio differs in different regions, even in the same retina (O’Connell 1963; Wagner 1978; Shand 1997; Fritsches et al. 2003). Underwater, the distribution of light intensity and wavelengths is more complex than on the ground, so fish often have different light sensitivities and color vision depending on their visual angle. For example, the ventral retina receives ultraviolet to blue-violet light irradiated downward from the water surface, while the dorsal retina receives blue-green light reflected upward from the seafloor. Nutrias feed on plants and animals in the water and on land. At dawn and dusk, when they are actively ingesting/searching for these items, the changes in both light intensity and light wavelength are significant. To understand the mechanism by which the nutria's vision perceives these preys, we will need to analyze their color vision and differences in visual function depending on the visual angle.

The visual acuity of the nutria—calculated from the highest density of ganglion cells, which was estimated from photoreceptor nuclei counts and the lens diameter—ranged from 1.44 to 1.58 cpd. Previous studies reported that the histologically-determined rodent visual acuity was 1.0 cpd in rats (Dean 1981), 0.6 cpd in mice (Gianfranceschi et al. 1999) and 0.3–0.5 cpd in rodents (Ňemec et al. 2008). Prusky et al. (2000) found that the visual acuity of mice and rats was 0.5 and 1 cpd, respectively, based on behavioral responses to a visual target presented on a computer display. All of these rodents, including the nutria, are nocturnal. In contrast, the visual acuity reported for two diurnal rodents (the Nile grass rat and four-striped grass mouse) was approximately 1 cpd (Gaillard et al. 2008; van der Merwe et al. 2018). Regarding other mammals, visual acuity has been reported to be approximately 60 cpd in humans; 3 to 6 cpd in cats; 6 cpd in rabbits, capybara, and dolphins; and 2 cpd in bats (Hughes 1977). Larger eyeballs are associated with higher PND values (Kiltie 2000). Nutrias have slightly higher visual acuity than other small rodents, so this value is likely to be due to its larger lens. This result indicates that the nutria’s vision may be adapted to recognize large shadows of distant predators, rather than viewing objects in detail.

Although the nutria has a duplex retina, it is approximately 90–100 μm thick, thinner than that found in other rodents (100–250 μm, Rodriguez-Ramos and Dubielzig 2013), implying that the retina of the nutria is less developed compared with other rodents. On the other hand, observing the radial section view of the nutria retina, the pigment epithelium is densely pigmented and well-developed. Peichl et al. (2017) observed pigment concentrations in the retinal pigment epithelium in species with diverse daily activity patterns (diurnal, nocturnal and cathemeral), suggesting that nocturnal rodents lose pigment while diurnal species have a light brown color. If the retinomotor response (movement of pigment due to ambient brightness) of the nutria functions normally, then its cone photoreceptor cells may give it photopic and mesopic vision under bright conditions.

Nutrias are almost exclusively herbivorous, but they also hunt small terrestrial animals. In order to understand the extent to which they detect “moving animal prey,” it is necessary to analyze the shape of the ganglion cells and determine whether nutrias perceive movement.

CONCLUSIONS

The lens, which is spherical in aquatic adapted mammals, is elliptical in the nutria, meaning that living underwater does not offer any advantage over living on land. Its visual acuity is not so different from that of other small rodents, although the retina is rather thinner and the neural network was estimated to be low. Bigger nutrias have few natural enemies, and they only need enough vision to detect larger predators at long distances. In characteristic morphology, the diameter of the lens is large, occupying more than 50% of the eyeball diameter. The nutria’s eyes, which are excellent at capturing photons, should support their activities at twilight.

Supplementary materials

Table S1.

Calculations of a relationship of Y = X^0.94 between the pupil diameter (Y) when fully dilated and the lens diameter (X) (Schmitz 2009) for five samples from Okayama Prefecture.