For plains zebras to lead the migrations across the African savannah ecosystem, as we propose (link to collars with migration page), they must be able to determine precisely where they are and where they want to go. We suggest that the sensory system used to guide these cognitive functions is primarily visual. Visual information of specific landmarks, and how they might have changed from previous migrations, is central to the migrations, not only for the plains zebra, but also for all other species that co-migrate with them. To understand how the plains zebra might use visual recognition of landmarks to navigate, what these landmarks might be, and from what distance they are visible to the plains zebra (and other co-migrating species), we need to estimate the visual capacities of the plains zebra. This will allow us to gain insights on how the plains zebra and other co-migrating species see the world. To achieve this, we plan to examine three important aspects of the visual system: (1) the spatial resolving power of the eye, (2) how the plains zebras spatially sample their surroundings with high resolution, and (3) the how the density of colour-sensing cells is distributed in the retina to potentially enable colour vision across the visual space.

The first aspect refers to the ability of the eye to extract fine spatial information from the surroundings – the spatial resolving power of the eye. This measure of visual resolution depends on the density of neural receptors in the retina (i.e., cones and ganglion cells) and the optical properties of the eye. Conventionally, spatial resolving power is measured as the frequency of cycles (one cycle = a black + a white bar) per degree of visual angle. The estimates of spatial resolving power vary considerably among species and appear to reflect their specific needs to extract fine details from the visual environment (Hughes, 1977). For example, humans (with normal vision) have an average spatial resolving power of approximately 60 cycles per degree, whereas the estimates for horses are about one third (20-25 cycles/degree) of that of humans (Curcio et al., 1990; Timney and Keil, 1992; Evans and McGreevy, 2007). To date, no estimate of the spatial resolving power of the zebra eye is available. Thus, we plan to measure the spatial resolving power of the plains zebra eye and other co-migrating species. This will allow us to estimate whether they see the world like horses or whether their eyes afford enhanced spatial discrimination, that would aid visual communication and orientation during migrations.

The second aspect to be examined is the distribution of neural cells in the retina. The density of neurons across the retina is non-uniform, which creates areas of high and low neuronal density. Areas of high neuronal density afford greater resolution to discriminate fine detail compared to areas of low neuronal density. These densities can be mapped across the retina (Hughes, 1977; Collin, 1999), giving insights into how the eyes of an animal sample their surroundings. Two general patterns of neuronal distribution are normally found in the mammal retina: concentric and elongated (Hughes, 1977; Collin, 1999). Humans, for example, have a concentric pattern with a maximum density of cells in the fovea – a small pit in the retina that affords high spatial-resolution (Curcio et al., 1990). In contrast, the neural densities in horses are arranged in an elongated pattern across the eye equator – the horizontal visual streak (Evans and McGreevy, 2007). The horizontal visual streak enables horses to sample more visual information across the horizon. This is particularly relevant for obtaining a panoramic view of the surroundings which may aid in the early detection of predators and continuous sampling of visual landmarks. It is likely that the plains zebra retina also has a horizontal visual streak, but it is unknown whether the plains zebra retina is equipped with higher cell densities than seen in the horse, that may afford zebras enhanced capacities for detection of predators and potential landmarks compared to horses, and indeed co-migrating species.

The third aspect is the mapping of the colour-sensing cells, the cones, that are maximally sensitive to distinct portions of the colour spectrum. Humans and other diurnal primates are trichromats as they have three different types of cones mostly sensitive to blue, red and green light. In contrast, most other mammals have only two types of cones sensitive to blue and green light, giving them the potential for dichromatic colour vision (Ahnelt and Kolb, 2000; Peichl, 2005). In horses and other equids, including the Grevy’s zebra, the presence of blue and green-sensitive cones has been reported (Sandman et al., 1996); however, detailed mapping of their topographic distribution across the retina is lacking. Measuring the number, density distribution, and proportions of these spectrally distinct cone types in the plains zebra retina will allow us to gain insights on how their retinas extract colour information from the environment. These findings would be valuable for predicting the role of colour vision in plains zebras and their potential use for detecting landmarks during migrations.

Given these goals, in this part of the project we hope to understand how the eye of plains zebra sees the world in terms of landmarks for navigation and predator avoidance. We plan to compare this to (1) phylogenetically related species such as mountain zebra (Equus zebra), domestic horse (Equus caballus), and domestic donkey (Equus asinus); (2) sympatric migrating species including the blue wildebeest (Connochaetes taurinus), Grant’s gazelle (Nanger granti), Thomson’s gazelle (Eudorcas thomsonii), white-eared kob (Kobus leucotis), tiang antelopes (Damaliscus lunatus tiang), and Mongalla gazelles (Eudorcas albonotata); and (3) non-migrating sympatric species including impala (Aepyceros melampus), waterbuck (Kobus ellipsiprymnus), nyala (Tragelaphus angasii), kudu (Tragelaphus strepsiceros), and eland (Taurotragus oryx).

Hypothesis: The retina of the plains zebra will be organized in similar fashion to the domestic horse, but is likely to show variations in cell density that will lead to enhanced visual discrimination of potential landmarks during the migrations.

Aim: To estimate whether, in comparison to phylogenetic and ecologically related species, the eye of the plains zebra shows specialisations that will enhance their capacity to recognise landmarks used for navigating the migratory paths.