The eyes of mammals reveal a dark past

A popular hypothesis suggests that the mammalian eye developed in the shadow of the dinosaurs.
Emily Sohn is a freelance journalist in Minneapolis, Minnesota.

Search for this author in:

Artist's impression of the early mammal-like animal Kayentatherium

Kayentatherium (centre), an early mammal-like animal, lived in the shadow of dinosaurs such as Dilophosaurus during the early Jurassic, a period that occurred 201 million–174 million years ago. Credit: Mark Witton

For more than 100 million years, dinosaurs dominated the daylight hours, with warmth from the Sun enabling them to thrive. The night, therefore, might have provided small, mostly insect-eating mammals with the best opportunity to hunt without becoming a dinosaur’s next meal.

Early mammals, which emerged in the shadow of the dinosaurs, are thought to have lived mainly nocturnal lives. Mammals evolved keen senses of smell and hearing and, at the expense of some ability to see in colour and at higher resolution, their eyes developed adaptations that improved their vision in the dark: larger pupils that allow more light to enter the eye; greater numbers of rod cells, the photoreceptors required for seeing in dim light; and a reflective layer of tissue called the tapetum lucidum, which increases light absorption by the retina.

By studying the eyes and genes of modern animals, as well as peeking into the fossil record, scientists have begun to piece together how long mammals were confined to the night and how this period shaped the visual systems of today’s mammals. Such work is also offerings insight into why human eyes sometimes falter. Although the details remain up for debate, understanding the evolution of mammalian vision during an extended period of nocturnality could also help scientists to predict how modern mammals might adapt as increasing light pollution and human activity pushes some diurnal animals to reclaim the night.

“What we have with mammals today are a radiation of organisms that evolved the quintessential features of their sensory systems in a nocturnal setting,” says Christopher Kirk, a biological anthropologist at the University of Texas at Austin. “One hundred million years of nocturnality — I’ve said it before and I’ll say it again — it’s got consequences.”

Birth of a hypothesis

In his 1942 monograph, The Vertebrate Eye and its Adaptive Radiation, Gordon Walls, a biologist at the then Wayne University in Detroit, Michigan, proposed that there had been a prolonged period of nocturnal living during the early evolution of mammals, to explain why mammals’ eyes tend to differ from those of other vertebrates. The idea became known as the ‘nocturnal bottleneck’ hypothesis.

Many mammalian quirks could be explained by a history of nocturnality. As well as having large corneas and pupils to maximize the amount of light that can enter the eye, mammals’ eyes have less distance between the lens and retina than do many vertebrates’ eyes, which helps the lens to project a bright image onto the retina in dim light. There are also fewer types of photoreceptor cell for detecting colours (known as cone cells). And the eyes of most mammals — although notably not those of humans or certain other primates — lack a fovea, an area of the retina rich in cone cells that provides sharp and detailed vision to fish, birds and reptiles that hunt during the day. “Most mammalian eyes are not up to snuff as they are for birds or lizards, or even reef fishes,” says Lars Schmitz, an evolutionary biologist at Claremont McKenna College in California.

In the past decade, researchers have used new tools and approaches to gather data to help bolster the nocturnal bottleneck hypothesis. In 2012, Kirk and his colleagues analysed the eye morphologies of 266 species of mammal from 23 orders1. They found that there was little variation in shape, regardless of whether the species were active during the day, the night or both. That distinguished mammals from birds and lizards, the eye shapes of which differ more between nocturnal and diurnal species. Mammals’ eyes tended to resemble those of nocturnal birds and lizards, with the exception of humans and closely related monkeys and apes. Kirk suggests that this is because some mammals’ eyes are likely to have started to ‘re-evolve’ features useful for diurnal living after they abandoned a nocturnal lifestyle.

Researchers are developing a clearer idea of when this period of nocturnality occurred, and how long it lasted. In 2017, a group in the United States suggested that the ancestors of mammals were diurnal. They used fossil evidence to show that, around 385 million years ago (just before vertebrates emerged from the water onto land2), vertebrates’ eyes tripled in size and moved from the sides to the tops of their heads. These adaptations enabled early vertebrates to see much farther and to hunt with their eyes peeking above the water, as crocodiles do. This would have been most advantageous in strong light, says Roi Maor, a mammal ecologist at Tel Aviv University in Israel and University College London — suggesting that early vertebrates were active in the day.

Something must have changed for mammals to adopt nocturnal behaviour. In late 2017, a group of researchers including Maor presented evidence for what that might be. The team used data on patterns of daily activity for 2,415 living species of mammal to reconstruct two potential family trees — one spanning 166 million years of mammalian evolutionary history and the other 218 million years3. The tree with the shorter timeline suggested that mammals began to occupy a daytime niche after the reign of non-avian dinosaurs was ended by the Cretaceous/Palaeogene extinction, triggered when an asteroid struck the Yucatán penninsula in what is now Mexico, around 66 million years ago. That study was the first to pin mammalian activity patterns to an explicit timeline, says Maor, and supports the hypothesis that early mammals were driven to nocturnal lifestyles by the dinosaurs’ dominance of daylight hours.

The Cretaceous/Palaeogene extinction “was the best thing that happened for mammals”, Kirk says. “Without this 10-kilometre-wide chunk of rock smashing into the Yucatán, who knows what mammals would look like today — probably still just a bunch of little shrew-like things trying not to become dinner.”

Grey zone

Not all evidence supports this version of the nocturnal bottleneck hypothesis, however. Some research, for instance, hints that there might have been more overlap between the waking hours of mammals and dinosaurs than the hypothesis suggests.

For example, there is evidence that early mammals had variants of genes that encode light-sensitive proteins called opsins, which are tuned for colour vision. In 2012, this evidence was used to suggest that mammals’ activity might not have been confined to the hours of darkness4. The combination of opsins in such mammals resembled that in modern snakes that are active in dim light. Mammals might therefore have hunted by twilight rather than become fully nocturnal.

Some dinosaurs were probably also out and about after the sun went down, says Schmitz. Together with Ryosuke Motani, a palaeobiologist at the University of California, Davis, he studied the fossils of 33 archosaurs, a group that includes dinosaurs and flying reptiles known as pterosaurs. The pair focused their analysis on the animals’ eye sockets (orbits) and scleral rings5, which are bones that surround the eyes of many vertebrates.

They started by comparing the sizes of scleral rings with their corresponding orbits in more than 150 modern animals, and correlating the resulting ratios with the animals’ patterns of activity. They then used this information to infer the activity of dinosaurs from the size of their eyes, leading them to suggest that certain dinosaurs actually prowled the dark. “Most dinosaurs were active during the day and most mammals were active during the night,” Schmitz says. “But it’s probably not clear cut.” Such an overlap could help to explain the emergence of an almost 1-metre-long, dinosaur-eating mammal around 130 million years ago.

There is also evidence that, rather than mammals retreating into the night, it might have been their ancestors that did so, Schmitz says. In 2014, he and Kenneth Angielczyk, a palaeobiologist at the Field Museum of Natural History in Chicago, Illinois, analysed the orbits and scleral rings of 24 species of non-mammal synapsids — mammal-like reptiles that preceded true mammals. By extrapolating eye shape from the fossils’ dimensions, the pair found evidence of nocturnality occurring more than 300 million years ago — more than 100 million years before the emergence of mammals6. “They may have entered the nocturnal phase even earlier than we thought,” Schmitz says.

But Kirk and several other researchers, including Margaret Hall, an evolutionary biologist and anatomist at Midwestern University in Glendale, Arizona, have questioned whether the sizes of orbits and scleral rings in modern animals provide enough information to reconstruct the behaviour of fossilized creatures. Hall and Kirk suggest that predicting how animals see requires knowledge of the size of the cornea and the diameter of the eyeball, measurements that are uncertain for long-extinct animals. “The story we all tell ourselves and each other — that somehow dinosaurs forced mammals into a nocturnal niche — could be true,” Hall says. “It’s the most likely just-so story we can come up with. But there is no way to know.”

Return to darkness

Some researchers think that the nocturnal bottleneck could be the root of some of the problems that affect human eyes. For example, while adapting to nocturnal living, mammals seem to have lost the mechanism by which modern lizards and birds squeeze their lenses to focus. This makes people more vulnerable to presbyopia, or long sightedness, as they age. Similarly, age-related macular degeneration, in which a small region of the retina degrades, resulting in the loss of central vision, has such a grave effect because of how humans and other primates acquired high-acuity vision: by cramming cone cells into the fovea. That structure probably arose as an adaptation to the development of a network of blood vessels over the base of the retina after the nocturnal bottleneck. To avoid this network interfering with high-acuity vision, cone cells were forced to occupy a small — and therefore vulnerable — area, says Kirk. “When you ask why we have to worry about these disease processes in the first place, a lot of it goes back to lost functionality in the bottleneck and everything that happened to compensate for that,” he says.

A European beaver stands on the sandy bank of river flowing through an urban area at night

Modern mammals such as the European beaver (Castor fiber) are often active at night in urban areas.Credit: Geslin Laurent

That process of compensation offers a glimpse of how animals might adapt to life in environments that are becoming brighter and more densely populated by humans. In a 2018 meta-analysis, researchers in the United States examined 76 studies of the activity of 62 species of mammal on 6 continents7. Mammals that lived in areas with a high level of human disturbance (or in quieter areas at times of increased human activity, such as hunting season) were 36% more nocturnal than those that lived relatively undisturbed.

Kaitlyn Gaynor, a wildlife ecologist at the University of California, Berkeley, who worked on the meta-analysis, says that it is difficult for researchers to predict every ecological consequence of animals switching niches. In the Santa Cruz Mountains of California, coyotes living near popular hiking areas have become more nocturnal and, in turn, their diet is now more likely to include animals that are active at night. That will have consequences for their new prey, as well as the night-time predators, including foxes, with which they are now in competition. And it is unclear how successful other diurnal mammals will be at communicating, feeding and avoiding predation after switching to nocturnality.

As levels of human activity continue to increase, people might force a further nocturnal bottleneck. “Mammals have been enjoying the sunlight since the dinosaurs went extinct, but now we’re driving them back into the night-time,” Gaynor says. “We’re a ubiquitous, terrifying force on the planet — much like the dinosaurs were.”

doi: 10.1038/d41586-019-01109-6

This article is part of Nature Outlook: The eye, an editorially independent supplement.


  1. 1.

    Hall, M. I., Kamilar, J. M. & Kirk, E. C. Proc. Biol. Sci. 279, 4962–4968 (2012).

  2. 2.

    MacIver, M. A., Schmitz, L., Mugan, U., Murphey, T. D. & Mobley, C. D. Proc. Natl Acad. Sci. USA 114, E2375–E2384 (2017).

  3. 3.

    Maor, E., Dayan, T., Ferguson-Gow, H. & Jones, K. E. Nature Ecol. Evol. 1, 1889–1895 (2017).

  4. 4.

    Davies, W. I., Collin, S. P. & Hunt, D. M. Mol. Ecol. 21, 3121–3158 (2012).

  5. 5.

    Schmitz, L. & Motani, R. Science 332, 705–708 (2011).

  6. 6.

    Angielczyk, K. D. & Schmitz, L. Proc. Biol Sci. 281, 20141642 (2014).

  7. 7.

    Gaynor, K. M., Hojnowski, C. E., Carter, N. H. & Brashares, J. S. Science 360, 1232–1235 (2018).

Download references

Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up