What colors do bees, birds, and dogs see the world in?

Short answer

Each species sees the world in as many colors as the number and tuning of the light receptors in its own eyes allow. Dogs have two cone types and struggle to tell red from green; bees can barely see red at all but see the ultraviolet we cannot; and birds' four cone types open up color differences for them that we don't even have names for.

The number question: how many cones, how many color axes

The light receptors the eye uses for color are called cones. Most people have three cone types, each most sensitive to a different part of the spectrum. The brain produces color by comparing the responses of the three cones; the sensitivity curves of the cone types are the subject of BI-09. The key idea here is simple: the number of cone types available for comparison sets the number of color "axes" a species can tell apart.

So when you ask about another species' color world, the first question is: "how many cone types are there, and where are they tuned?" The answer varies widely from species to species, and each one carries a clue about what that species needs to see to survive.

Dogs: two cones, one color axis

Dogs' retinas have two cone types. One is most sensitive to the blue-violet end of the spectrum, the other to the yellow-green region. This doesn't mean dogs don't see color; they tell blue and yellow apart easily. But red, orange, and green are, to them, largely different brightness levels of the same yellowish tone. A red ball on the grass stands out to a dog not by its color but, at most, by its lightness or its movement.

Humans have a similar situation: some people see with only two cone types and confuse red and green. This is a good window into a dog's world, but it's a separate topic covered in BI-12. One interesting side effect: because a blush is a shift along the red-green axis, it most likely goes unnoticed by someone with dichromatic vision. Whatever the skin tone, a blush is, for them, only a very small change in brightness.

Bees: blind to red, open to ultraviolet

The honeybee also has three types of light receptor, but they're placed in a different slice of the spectrum. One is most sensitive to ultraviolet, one to blue, one to green. The bee's visible range is therefore shifted toward shorter wavelengths than ours: it sees ultraviolet, but it can barely see the long-wavelength red at all.

Birds: four cones and colored droplets

Most birds have four cone types: ones sensitive to red, green, blue, and violet or ultraviolet. What's more, inside their cones are colored oil droplets that filter the light. These droplets narrow the range each cone is sensitive to and reduce the overlap between cones; as a result, colors that are very close to each other are easier to tell apart.

A fourth color axis means a bird can see two separate colors in what we take to be a single color. If one of two feathers that look identically black to our eyes reflects ultraviolet, a bird tells them apart easily. The evolutionary story behind this difference, and why mammals come from a dichromatic past, is the subject of DO-01.

Example — the hidden target on a yellow daisy

The black-eyed Susan, a yellow flower common on North American prairies, looks to us like a flat, uniform yellow from end to end. Seen through an ultraviolet-sensitive camera, though, the base of the petals darkens and the flower turns into a target: a bright ring on the outside, a dark patch at the center. This pattern comes from pigments called flavonols that build up at the base of the petals and absorb ultraviolet. For a bee, this target is an arrow pointing to where the nectar is.

It's no accident that the flower doesn't carry this pattern for us. The flower isn't speaking to us, it's speaking to its pollinator. A sign written in a channel we cannot see shows its real reader exactly where to go.

Common misconception

Summary

1. A species' color world depends on the number of cone types in its eyes and which parts of the spectrum they're tuned to.
2. Dogs see with two cones and confuse red and green; bees can't see red but can see ultraviolet.
3. Birds' four cones and colored oil droplets open up color differences for them that we cannot see; flowers carry patterns in these same channels.

Test yourself

1. (Easy) How many cone types does a dog's retina have?
2. (Easy) Which color can humans see that honeybees cannot?
3. (Medium) Why would a flower carry patterns that are invisible to us?
4. (Medium) What do the colored oil droplets in bird cones do?
5. (Hard) How can a bird tell apart two feathers that look identically black to us?
6. (Hard) In what sense is the sentence "Dogs are colorblind" wrong, and in what sense is it partly true?

Answers

1. Two; their peak sensitivities are at approximately 429 nm and 555 nm ("Dogs: two cones, one color axis").
2. Red; the bee has no receptor sensitive to red ("Bees: blind to red, open to ultraviolet").
3. The flower carries its pattern to show its pollinator, not us; because bees see ultraviolet, a nectar guide drawn in ultraviolet directs them to the nectar. The fact that our eyes can't see that channel doesn't make the pattern useless ("Example — the hidden target on a yellow daisy").
4. Acting as a filter, they narrow the range each cone is sensitive to, reduce the overlap between cones, and make it easier to tell close colors apart ("Birds: four cones and colored droplets").
5. If one feather reflects ultraviolet and the other doesn't, the bird's ultraviolet-sensitive cone picks up that difference; because we have no such cone, the two feathers look the same to us ("Birds").
6. It's wrong in the sense that they don't see color at all, because they do tell blue and yellow apart. It's partly true in the sense of "can't make a red-green distinction," because they do confuse red and green ("Common misconception").

Sources

• Neitz, J., Geist, T., and Jacobs, G. H., "Color vision in the dog," Visual Neuroscience, 1989.
• Kevan, P. G., Chittka, L., and Dyer, A. G., "Limits to the salience of ultraviolet: lessons from colour vision in bees and birds," Journal of Experimental Biology 204(14), 2001.
• Olsson, P., Lind, O., and Kelber, A., "Bird colour vision: behavioural thresholds reveal receptor noise," Journal of Experimental Biology 218, 2015.
• "Flavonols: Pigments Responsible for Ultraviolet Absorption in Nectar Guide of Flower," Science 177, 1972.

Next step

DO-01 deepens this lesson's biology: it tells the story, as an evolutionary narrative, of mammals' dichromatic past, how trichromatic vision evolved in primates, birds' fourth cone and oil droplets, and flowers' ultraviolet patterns. DO-02, meanwhile, tests a popular claim: using color-discrimination experiments, it shows why the mantis shrimp's large number of receptors doesn't mean it sees richer color.