Idea

The Fourth Color

Most people see color with three kinds of cone. Some women may have four, birds and fish do, and a rock band wrote a song about wanting one. What a fourth color channel really adds.

Arts

Overview

Human color vision runs on three kinds of cone cell in the retina, each tuned to a different band of light. Every color a typical person sees comes from comparing the signals of those three. Tetrachromacy means having four cone types working at once, which in principle adds a whole extra dimension of color.1

Many animals have it. Most birds carry a fourth cone sensitive to ultraviolet or violet light, and behavioral tests on wild hummingbirds show them telling apart colors that mix ultraviolet with red, green or yellow, colors no human has a name for.2 Goldfish pass color-mixing tests that need four channels to explain.3

In people the question is harder. The genes for the red and green cone pigments sit on the X chromosome, so a woman can carry two different versions and grow four kinds of cone. About 12 percent of women carry such a variant.14 The extra cone does not guarantee extra colors. When the Newcastle psychologist Gabriele Jordan and colleagues tested 24 carriers, one woman, known in the paper only as cDa29, behaved as a true tetrachromat on every test.4

A rock band named an album's closing song after the idea, one stop on a long thread of eyes and altered sight that runs through its catalog.

Deep Dive: A color and vision thread through King Gizzard

King Gizzard & the Lizard Wizard return again and again to eyes, sight and altered perception, and fans tie much of the catalog into a shared story world they call the Gizzverse. Stu Mackenzie has said on radio that the albums exist in a parallel universe, though the band has never confirmed a plot.9

Read across the catalog, the thread runs through song titles, album concepts, tunings and even the cover art. Nonagon Infinity (2016) introduced the cyborg Han-Tyumi in its opener "Robot Stop."9 Five albums followed in 2017 alone, and four of them come up below. Gumboot Soup, the last of that year, returns to tetrachromacy in "All Is Known."10 Later, Omnium Gatherum (2022) named a song for magenta, a color with no wavelength of its own.11

Eyes and color in King Gizzard
  1. 2014 "The Fourth Colour" written A jam from the I'm In Your Mind Fuzz sessions.
  2. 2017 Polygondwanaland "Tetrachromacy" and "The Fourth Colour" close the album.
  3. 2021 Butterfly 3000 The cover is a cross-eyed autostereogram.
  4. 2026 Field of Vision II "The Fourth Colour" in Buena Vista, Colorado.

Polygondwanaland, 2017. The fourth of five albums the band released in 2017, written in shifting odd meters, with no song entirely in 4/4.5 The cultural theorist Benjamin Kirbach reads it as a three-act myth: civilization faces ecological collapse, a coup fails, and survivors enhance themselves with machines and take over an information world.5 The closing suite starts with "Tetrachromacy," in which the narrator, after the deaths of "Horology," hears of a color beyond human sight and resolves to find it; the band's fan archive notes that "All Is Known" on Gumboot Soup later returns to the idea.1210 "Searching..." follows, and "The Fourth Colour" delivers the payoff, the narrator seeing everything and losing the body. The fan archive reads the screamed ending as a hint that what the narrator finally sees may be uglier than expected.68

The song's own history. "The Fourth Colour" is the oldest song on the record. Mackenzie said in a 2020 Reddit AMA that it was written during the I'm In Your Mind Fuzz sessions in 2014 as a lyric-less jam in standard tuning.6 It debuted live on 18 February 2018 in Melbourne, usually paired with "Crumbling Castle," and from May 2024 gained a long jam before the second chorus. The studio outro has never been played live.6 By the band's archive it had been played 123 times as of late 2026.13

Han-Tyumi and the machine body. On Murder of the Universe (June 2017) the cyborg Han-Tyumi gains consciousness and longs for the two things a machine cannot do, to vomit and to die; he builds a machine that only vomits, merges with it, and drowns the universe.14 His synthetic voice saying "hello" at the end of Polygondwanaland links the two albums.69 Both records ask what a body can perceive and what lies past it, from opposite ends: Han-Tyumi wants a body's crude sensations, and the Polygondwanaland narrator gets a sense no human body has.

Microtones as more colors. Flying Microtonal Banana (February 2017), subtitled "Explorations into Microtonal Tuning, Volume 1," uses guitars refretted for quarter tones, 24 steps to the octave instead of 12.15 The comparison is loose, but the move is the same one as a fourth cone: add channels between the familiar ones and the space of possible differences grows.

Butterfly 3000, 2021. Mackenzie called the butterfly "the central motif" of the record, and Jason Galea's cover is a cross-eyed autostereogram, a picture that only resolves when the viewer's eyes change how they combine their inputs.16

Field of Vision. The band's own festival takes its name from the same sense. At Field of Vision II, at Meadow Creek in Buena Vista, Colorado, on 15 August 2026, "The Fourth Colour" came out of "Crumbling Castle" and ran into "Organ Farmer"; the official bootleg puts it at eight and a half minutes.1317 The show is in the Releases catalog as Live at Field of Vision II '26.

In context

Color vision is old. The common ancestors of vertebrates probably had four cone types. Early mammals, small and active at night in the age of dinosaurs, lost two of them, so most mammals today are dichromats with two cones. Around 35 million years ago the ancestors of Old World monkeys and apes split one X-linked cone gene into two, and gained the red and green pigments that make human trichromacy.18 Stoddard and colleagues, writing about birds, note that since tetrachromacy seems to have evolved early in vertebrates, rich color perception of that kind is probably widespread.2

That history explains the human case. Because the red and green genes are on the X chromosome, men have one copy and women two. A woman whose two copies differ may grow a fourth cone sensitive in the yellow-green, between the usual red and green.1 New World monkeys show the same trick: in most species only females with two different versions of their single X-linked cone gene are trichromats, and the rest are dichromats.18

The fourth color over time
  1. 1914 Bees see color Karl von Frisch trains honeybees to colored cards.
  2. 1969 Basic Color Terms Berlin and Kay rank color words across languages.
  3. 1983 Reddish green Crane and Piantanida report "forbidden" colors under a stabilized image.
  4. 2010 cDa29 Jordan and colleagues find one functional tetrachromat among 24 carriers.
  5. 2025 Olo Berkeley's Oz laser stimulates green-sensitive cones alone.

Language has its own "fourth color." In Basic Color Terms (1969) Brent Berlin and Paul Kay argued that languages add basic color words in a fixed order. Every language separates dark from light; the third term is red; and the fourth is either green or yellow, with the other following next, then blue.19 Critics have questioned the method and the evolutionary reading, and the authors themselves loosened the sequence in later work.19

Research

Human tetrachromats

The X chromosome makes the extra cone possible. It does not guarantee anything. Jordan's group at Newcastle, with Samir Deeb, Jenny Bosten and John Mollon, tested 24 women who were obligate carriers of deuteranomaly, a mild form of red-green color weakness. Using a forced-choice version of the Rayleigh color-matching test, a scaling task that maps how many dimensions a person's color space has, and genetic analysis, they found that most carriers showed ordinary three-dimensional color vision. One, cDa29, behaved as a tetrachromat on all their tests. Her genes coded for three well-separated long-wave cone pigments alongside the short-wave one.4

Deep Dive: How a tetrachromat is tested

Owning a fourth cone and using it are separate questions, and the tests for human tetrachromacy are built to pull them apart. Most of them grow out of one old experiment with a yellow light.

The Rayleigh match. In 1881 Lord Rayleigh built an apparatus that mixed a narrow band of red light with a narrow band of green to match a yellow. Most observers settled on nearly the same mixture, but a few chose mixtures far from everyone else's, and those few turned out to have unusual color vision.20 The test works because of where the cones sit. A pure yellow near 589 nanometers and a mixture of red near 670 and green near 545 can look identical, since the short-wave blue cone barely responds to any of them and the red-sensitive and green-sensitive cones can be made to respond the same way to both. A person whose red or green pigment is shifted accepts a different mixture, or a wider range of them.20 The physiologist Willibald Nagel built the idea into a clinical instrument and named it the anomaloscope in 1907. More than a century later it is still the reference instrument for sorting out red-green variants.20

Why the match suits the question. A woman who carries an anomalous gene may have an extra pigment between the usual red and green. In the narrow red-to-yellow range of the Rayleigh test, where the blue cone is silent, she would then have three working signals where a typical observer has two. A mixture that balances two cones need not balance three, so a true tetrachromat should reject matches everyone else accepts. That makes the Rayleigh match the natural place to look for her.21

First hints. In 1981 the vision scientists Nagy, MacLeod, Heyneman and Eisner tested the rule that color matches add up: if two lights match two others, their sums should match too. Most observers obeyed it. Some carriers did not, after their rod cells had been bleached by bright light, which suggested a fourth pigment was changing their matches. Yet all of them still made acceptable Rayleigh matches with only two lights.22 Jordan and Mollon later called this "weak tetrachromacy": a fourth pigment that alters the matches without giving its owner a new dimension of color she can use.21

The 1993 study. Gabriele Jordan and John Mollon then went looking for the strong kind. They tested 43 women aged 30 to 59, 31 of them obligate carriers, mothers of color-deficient sons, and classified each carrier by testing her son on the anomaloscope. On a large-field Rayleigh match, eight of fourteen candidates refused the matches that normal observers accepted, and two performed on a ratio-matching task the way a tetrachromat should. The authors called this preliminary evidence and stopped short of a claim.21

The 2010 tests. The Newcastle study that found cDa29 combined three lines of evidence. The Rayleigh test was given in forced-choice form, where the observer must pick between alternatives on every trial, so the result does not depend on how strict she is about calling two lights a match. A multidimensional scaling task had her judge how different many pairs of colors looked, and the judgments were then fitted to color spaces of different dimensions to see how many were needed. Genetic analysis estimated where each of her cone pigments peaked.4 cDa29 passed all three. Most carriers did not, and the authors concluded that a color advantage for carriers is unlikely to be what keeps these variant genes common.4

Testing the fourth cone
  1. 1881 Rayleigh's match Red plus green against yellow.
  2. 1907 The anomaloscope Nagel's clinical instrument.
  3. 1981 Weak tetrachromacy Nagy and colleagues find carriers' matches shift.
  4. 1993 Jordan and Mollon Eight of fourteen candidates refuse normal matches.
  5. 2010 cDa29 One carrier passes every test.

What does not work. The online quizzes that ask how many bands a viewer can see in a rainbow strip cannot test for tetrachromacy. A screen makes every color from three primaries, so it cannot produce the pairs of lights that look the same to three cones and different to four. The Newcastle project and the Cleveland Clinic both say there is no valid online test.23 Real testing still needs narrow-band light sources, careful calibration and, ideally, a look at the genes.

Popular accounts tell how the search went. Jordan switched in 2007 to a test where women compared colored circles, one of them a computer-made mix of red and green light; cDa29, a doctor in northern England, got every answer right.24 Some articles from that time describe her result as unpublished, though the 2010 paper reports it.424

The project at Newcastle is careful about numbers. It states that about 12 percent of women carry the altered gene, and that nobody yet knows how many of them can use the extra cone to tell apart colors others cannot.1 The press figure of "100 million colors" comes from extrapolating the extra dimension; nobody has measured it.

The best-known self-described tetrachromat is a painter.

Animals with four cones

Birds. Most birds have four cone types, one sensitive to ultraviolet or violet light, and their lens and cornea let ultraviolet through.2 In a study at the Rocky Mountain Biological Laboratory in Colorado, Mary Caswell Stoddard and colleagues trained wild hummingbirds to feeders marked by lights. The birds told apart nonspectral colors, colors made by stimulating cone types that sit far apart on the spectrum, including ultraviolet plus red, ultraviolet plus green, ultraviolet plus yellow and purple. From about 3,300 plumage and plant spectra, the team estimated that birds see many natural colors as nonspectral that humans would not.2

For humans, purple is the clearest nonspectral color: it needs the blue and red cones active together, and no single wavelength of light, nothing in a rainbow, can produce it.2 A tetrachromat bird has many more such combinations available.

One of the band's color songs is named for a nonspectral hue.

Fish. Goldfish trained to choose between lights show a sharp ability to tell wavelengths apart near 410 to 420 nanometers, which points to an ultraviolet cone, and Christa Neumeyer's color-mixing experiments in the 1980s and 1990s support full tetrachromacy.3

Insects. Honeybees are trichromats with a different set: ultraviolet, blue and green receptors, peaking near 344, 436 and 544 nanometers. They have no red receptor.27 Karl von Frisch's 1914 experiments, training bees to colored cards with sugar water, were among the first convincing proofs of color vision in any animal besides humans.27 Bees show that the number of cones matters less than where they sit: three channels shifted toward ultraviolet give a different world from human three.

The song "Tetrachromacy" names a real Arctic finding.

The mantis shrimp

The mantis shrimp became the internet's emblem of super color vision, helped by a much-shared 2013 comic from The Oatmeal.30 Its eyes do hold an unusual set of receptors: twelve color types, each tuned to a narrow band from deep ultraviolet at 300 nanometers to far red at 720.31

Deep Dive: Inside the mantis shrimp's eye

Mantis shrimp, or stomatopods, are reef and burrow predators, and more than 520 living species are known. They come in two working types: spearers, which impale soft prey, and smashers, which hit hard-shelled prey with a mineralized club. A smasher's strike can reach more than 20 meters per second from a standing start and is fast enough to tear the water into cavitation bubbles, whose collapse delivers a second blow and can even give off light.32 An animal that hunts like that needs to judge distance and identify targets quickly, and its eyes look built for it.

An eye in three parts. Each eye sits on a mobile stalk and moves independently of the other. Each is a compound eye of tens of thousands of facets, and a band of two to six rows of specialized facets runs across its middle. The midband divides the eye into upper and lower halves that handle form and motion, and because all three regions can look at the same point, a single eye can judge depth on its own.32 The color and polarization receptors sit in the midband, which sees only a strip about 5 to 10 degrees wide at any moment. To build up a picture the shrimp swings and rotates its eyes, sweeping the strip across the scene.32

How the count grew. In 1989 Thomas Cronin and Justin Marshall reported in Nature a retina with at least ten spectral types of photoreceptor, far beyond anything known in another animal.33 Later work raised the number. Many species have twelve receptor types for color, and some have sixteen in all once the polarization receptors are counted.32 Up to six of the receptor types work in the ultraviolet. In 2014 Michael Bok and colleagues showed that in one species those ultraviolet receptors rest on only two visual pigments; filters in the eye, made from mycosporine-like amino acids, compounds that reef animals usually use as sunscreen, split the two pigments' response into separate bands.34 Other species tune their long-wave receptors to the color of the water they live in.32

Light with a twist. Mantis shrimp also read polarization, the direction in which light waves vibrate. Many animals see linear polarization. In 2008 Chiou, Marshall and colleagues showed that the shrimp Odontodactylus sees circular polarization too, light whose vibration spirals as it travels. Two rows of the midband are modified so their receptors respond to right-handed or left-handed spirals, and parts of the animals' own bodies reflect circularly polarized light, which may make it a private signal in courtship.35 No other animal is known to do this, and the ability has since been found in six stomatopod species.32

What all those channels buy. With twelve color receptors, a human-style system that compared every receptor against its neighbors would give an enormous color space. The test described next shows that the shrimp do worse than humans at telling close colors apart, which is why Thoen and Marshall proposed that they skip the comparisons and recognize colors by which receptor fires, something like reading a barcode.3136 That idea is still argued over. A 2022 study found no clear evidence for a pure barcode system: some species could tell high-saturation colors from low-saturation ones, and several mechanisms may work together.32 Either way, the shrimp shows that the number of receptors and the richness of color are different measures. More channels can serve speed, polarization and depth as easily as they serve extra hues.

Reading the mantis shrimp eye
  1. 1989 Ten receptor types Cronin and Marshall in Nature.
  2. 2008 Circular polarization Chiou and colleagues.
  3. 2014 Discrimination test Thoen's shrimp need gaps of 15 to 25 nm.
  4. 2022 Saturation Some species tell vivid from pale.

In 2014 Hanne Thoen, working with Justin Marshall at the University of Queensland, tested how well the shrimp actually discriminate. Shrimp were trained to strike the right one of two lit optical fibers for food, and the colors were moved closer together until the animals failed.36 They needed a gap of 15 to 25 nanometers. A human in the same kind of test can manage about 5.36 The authors proposed that mantis shrimp do not compare receptor outputs the way humans and bees do. They may read the pattern across all twelve receptors at once, perhaps while scanning their eyes, and recognize colors quickly without fine comparison.31 Marshall compared the system to a satellite that builds images from look-up tables.36

So twelve receptors buy speed of recognition, probably useful for fast decisions on a busy reef, at the cost of fine discrimination.

Colors that should not exist

Beyond extra cones, there is a second route to a "new" color: trick the three you have. The opponent-process account holds that the brain codes color as three contrasts, red against green, blue against yellow and dark against light. That is why nobody normally sees a reddish green.37

In 1983 Hewitt Crane and Thomas Piantanida reported in Science that this rule can bend. An eye tracker kept the border between red and green stripes fixed on one spot of the retina. The border faded and some observers reported a color they had never seen, which they could not name.37 Later work disputed and refined the result. Billock, Gleason and Tsou in 2001 matched the brightness of the stripes and got reports of reddish greens and bluish yellows, while Hsieh and Tse in 2006 argued the percepts were ordinary in-between colors.37

In 2025 a laser made the most direct attempt yet.

Deep Dive: Olo

The red-sensitive and green-sensitive cones respond to almost the same range of light. About 85 percent of the light that excites a green-sensitive cone also excites its red-sensitive neighbors, so in nature no light can stimulate the green-sensitive cones on their own.38 Every green anyone has ever seen is a mixture of the two signals, diluted by the red.

Building Oz. A team at UC Berkeley led by Ren Ng and Austin Roorda set out to remove that dilution. Their system, Oz, was begun in 2018 by J. Fong as an undergraduate project; Fong who said the name stood for a journey to the land of Oz to see a color never seen before.38 It works in two stages. First an imaging system developed by Ramkumar Sabesan and Vimal Prabhu Pandiyan at the University of Washington maps a small patch of a volunteer's retina and identifies every cone in it as short-, medium- or long-wave.38 Then a single-color laser, the green of an ordinary laser pointer, scans that patch and fires a microdose of light only at the instant the beam crosses a cone it has chosen. Oz can steer up to 1,000 cones at once in this way.38 The displayed square is small, 0.9 degrees across, about twice the apparent width of the Moon. Roorda said that if they could, they would have filled the whole visual field like an IMAX screen.3839

From one cone to a thousand. Oz grew out of earlier work in Roorda's lab on stimulating single cones. In a 2016 study, Sabesan, Roorda and colleagues flashed tiny spots of light onto 273 individual cones in each of two volunteers and asked what each flash looked like. Many cones, including red-sensitive and green-sensitive ones, produced only a sensation of white, and fewer than expected produced a color.40 One reading is that color is assembled further along, from the pattern across many cones. Oz scales the single-cone method up to a whole population, so that the pattern itself can be written.

What they saw. Aiming only at green-sensitive cones gave a color the team named olo, after its coordinates in the standard map of cone responses, zero for the long-wave cones, one for the medium and zero for the short.39 The five people who saw it, including Ng and Roorda, described a blue-green or peacock green far more saturated than any color in nature, and more saturated than the nearest pure spectral light.38 When the laser deliberately hit random cones instead, the color vanished and they saw the laser's ordinary green.38 The paper appeared in Science Advances in April 2025.41

The argument. Not everyone accepts that olo is a new color. John Barbur of City, University of London, said it is a very saturated green rather than a new color, and its existence "open to argument"; the study is also tiny.3839 The disagreement turns partly on words. If a new color means a new hue, olo may not qualify; if it means a cone signal nobody's eye has produced before, it plainly does.

What the method could do. The method may matter more than the hue. By writing signals to chosen cones, Oz could imitate the loss of cones in healthy volunteers, to study diseases that destroy them. The team has suggested it might one day help people with color blindness, and it opens an experiment directly relevant to this page: whether a brain given a fourth independent cone signal would learn to see a new dimension of color.3839 It also shows that color is a construction of the brain from cone signals: change the signals in a pattern nature never produces, and the brain reports something it has no word for.

Where it stands

Tetrachromacy is well established in animals. Birds, many fish and other vertebrates carry four cone types and use them, and hummingbirds have now shown in the wild that they see mixtures humans cannot.23 In humans the genetic route is real and common, but working tetrachromacy has been shown in very few people, and the size of the gain is modest.1425 The mantis shrimp is a lesson that receptor counts and color richness are different things.3136 Forbidden colors and olo come at the question from the other side: what the brain will report when its usual inputs are pushed out of their normal combinations.3738

Footnotes

  1. The Tetrachromacy Project, "The Science," Newcastle University. research.ncl.ac.uk ↩ ↩2 ↩3 ↩4 ↩5

  2. M. C. Stoddard, H. N. Eyster, B. G. Hogan, D. H. Morris, E. R. Soucy and D. W. Inouye, "Wild hummingbirds discriminate nonspectral colors," PNAS 117, 2020 (abstract and significance statement read). doi.org ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7

  3. C. Neumeyer and colleagues on goldfish color vision, Journal of Comparative Physiology A (abstracts through a search summary; full texts not read). link.springer.com ↩ ↩2 ↩3

  4. G. Jordan, S. S. Deeb, J. M. Bosten and J. D. Mollon, "The dimensionality of color vision in carriers of anomalous trichromacy," Journal of Vision 10(8):12, 2010 (abstract read on PubMed). pubmed.ncbi.nlm.nih.gov ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7

  5. Wikipedia, "Polygondwanaland" (secondary source; cites Benjamin Kirbach's reading of the album). en.wikipedia.org ↩ ↩2 ↩3 ↩4

  6. "The Fourth Colour: History," KGLW.net (fan-run song archive). kglw.net ↩ ↩2 ↩3 ↩4 ↩5 ↩6

  7. King Gizzard & the Lizard Wizard, Polygondwanaland, Bandcamp (free download; "BOOTLEG THIS ALBUM"). kinggizzard.bandcamp.com ↩

  8. "The Fourth Colour: Lyrics," KGLW.net (paraphrased here). kglw.net ↩ ↩2

  9. Aimee Ferrier, "What is the Gizzverse?," Far Out, 1 July 2023. faroutmagazine.co.uk ↩ ↩2 ↩3

  10. "All Is Known: History," KGLW.net. kglw.net ↩ ↩2

  11. Wikipedia, "Omnium Gatherum (album)" (secondary source). en.wikipedia.org ↩ ↩2

  12. "Tetrachromacy: History," KGLW.net. kglw.net ↩ ↩2

  13. "The Fourth Colour: Every Time Played," KGLW.net (123 plays as of the 15 August 2026 show). kglw.net ↩ ↩2

  14. Wikipedia, "Murder of the Universe" (secondary source). en.wikipedia.org ↩

  15. Wikipedia, "Flying Microtonal Banana" (secondary source). en.wikipedia.org ↩

  16. Wikipedia, "Butterfly 3000" (secondary source). en.wikipedia.org ↩

  17. King Gizzard & the Lizard Wizard, Live at Field of Vision II '26, official bootleg on Bandcamp, released 8 October 2026. bootleggizzard.bandcamp.com ↩

  18. Wikipedia, "Evolution of color vision in primates" (secondary source). en.wikipedia.org ↩ ↩2

  19. Wikipedia, "Basic Color Terms" (secondary source on Berlin and Kay, University of California Press, 1969, and its critics). en.wikipedia.org ↩ ↩2

  20. Wikipedia, "Anomaloscope" (secondary source, for Rayleigh's 1881 apparatus, the 589, 670 and 545 nanometer lights and Nagel's 1907 instrument), with the National Academies' Procedures for Testing Color Vision (1981) on its standing as the standard test. en.wikipedia.org ncbi.nlm.nih.gov ↩ ↩2 ↩3

  21. G. Jordan and J. D. Mollon, "A study of women heterozygous for colour deficiencies," Vision Research 33(11), 1993, 1495–1508 (through search summaries and later reviews; the PDF not read in full). sites.socsci.uci.edu ↩ ↩2 ↩3

  22. A. L. Nagy, D. I. A. MacLeod, N. E. Heyneman and A. Eisner, "Four cone pigments in women heterozygous for color deficiency," Journal of the Optical Society of America 71, 1981, 719–722 (through later papers' summaries; not read). doi.org ↩

  23. Snopes, fact check on online "fourth cone" tests, quoting Newcastle University's Tetrachromacy Project, and Cleveland Clinic, "Tetrachromacy" (both through a search summary). snopes.com my.clevelandclinic.org ↩

  24. Veronique Greenwood, "The Humans With Super Human Vision," Discover, 2012 (through a search summary). discovermagazine.com ↩ ↩2

  25. "This woman sees 100 times more colors than the average person," UC Irvine School of Social Sciences, 13 October 2014 (an excerpt of a Popular Science story). socsci.uci.edu ↩ ↩2 ↩3 ↩4

  26. K. A. Jameson and A. D. Winkler, preliminary report on Concetta Antico's color vision testing and opsin genotype, 2014 (through a search summary; the report not read). sites.socsci.uci.edu ↩

  27. Review of honeybee color vision on PubMed Central (through a search summary). pmc.ncbi.nlm.nih.gov ↩ ↩2

  28. "Tetrachromacy: Lyrics," KGLW.net (paraphrased here). kglw.net ↩

  29. N. J. C. Tyler, K.-A. Stokkan, C. Hogg and colleagues, including G. Jeffery, "Ultraviolet Vision and Avoidance of Power Lines in Birds and Mammals," Conservation Biology 28(3), 2014, 630–631 (through a search summary; cites Hogg et al., Journal of Experimental Biology 214, 2011). pmc.ncbi.nlm.nih.gov ↩ ↩2

  30. The Oatmeal, "Why the mantis shrimp is my new favorite animal" (comic; page title read, the image text not). theoatmeal.com ↩

  31. H. H. Thoen, M. J. How, T.-H. Chiou and J. Marshall, "A Different Form of Color Vision in Mantis Shrimp," Science 343(6169), 2014, 411–413 (abstract read on PubMed). pubmed.ncbi.nlm.nih.gov ↩ ↩2 ↩3 ↩4

  32. Wikipedia, "Mantis shrimp" (secondary source, citing Cronin et al. 2014, Patek and Caldwell 2005, Chiou et al. 2008, Templin et al. 2017 and Streets et al. 2022; first two-thirds read). en.wikipedia.org ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7

  33. T. W. Cronin and N. J. Marshall, "A retina with at least ten spectral types of photoreceptors in a stomatopod crustacean," Nature 339, 1989, 137–140 (reference only). nature.com ↩

  34. M. J. Bok, M. L. Porter, A. R. Place and T. W. Cronin, "Biological Sunscreens Tune Polychromatic Ultraviolet Vision in Mantis Shrimp," Current Biology 24(14), 2014, 1636–1642 (abstract through a search summary). research-information.bris.ac.uk ↩

  35. T.-H. Chiou, S. Kleinlogel, T. Cronin, R. Caldwell, B. Loeffler, A. Siddiqi, A. Goldizen and J. Marshall, "Circular Polarization Vision in a Stomatopod Crustacean," Current Biology 18(6), 2008, 429–434 (abstract and press summaries). mdsoar.org ↩

  36. Ed Yong, "The Mantis Shrimp Sees Like A Satellite," National Geographic Phenomena, 23 January 2014. nationalgeographic.com ↩ ↩2 ↩3 ↩4 ↩5

  37. Wikipedia, "Impossible color" (secondary source summarizing H. D. Crane and T. P. Piantanida, Science 221, 1983, 1078–1080; Billock, Gleason and Tsou, JOSA A 18, 2001; Hsieh and Tse, Vision Research 46, 2006). en.wikipedia.org ↩ ↩2 ↩3 ↩4

  38. Kara Manke, "Scientists trick the eye into seeing new color 'olo'," UC Berkeley News, 22 April 2025, with later press coverage for John Barbur's comment (through search summaries). news.berkeley.edu ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7 ↩8 ↩9 ↩10

  39. Wikipedia, "Olo (color)" (secondary source, for the 0.9 degree display, the name and Barbur's comment). en.wikipedia.org ↩ ↩2 ↩3 ↩4

  40. R. Sabesan, B. P. Schmidt, W. S. Tuten and A. Roorda, "The elementary representation of spatial and color vision in the human retina," Science Advances 2(9), 2016, with Science News coverage (through a search summary). pmc.ncbi.nlm.nih.gov sciencenews.org ↩

  41. J. Fong and colleagues, including W. S. Tuten, A. Roorda and R. Ng, "Novel color via stimulation of individual photoreceptors at population scale," Science Advances 11(16), 2025 (publisher page blocked direct reading). science.org ↩