The Evolution of Color Vision
Your three-cone color vision is not the default - it's a comeback story. Early vertebrates saw the world in four colors; mammals, hiding in the dark through the age of dinosaurs, lost most of it and went nearly colorblind. Then one branch of primates duplicated a gene and clawed a third cone back, probably to spot ripe fruit in green leaves. This is the interactive guide to how we came to see in three, and why the fix left color blindness written into the X chromosome.
A comeback, not a default
It's tempting to think of trichromacy - three cone types - as the natural, advanced state of color vision. It isn't. Birds, reptiles, and many fish are tetrachromats with four cone types (often including ultraviolet), inherited from a distant vertebrate ancestor. By that measure humans are color-impoverished: we run on three cones, and most other mammals get by on just two.
The reason is history. During the nocturnal bottleneck - the ~100+ million years mammals spent as small, night-active animals under dinosaur-dominated daytimes - color vision was nearly useless, and two of the four ancestral cone opsins were lost. Mammals became dichromats. Only later, in one primate lineage, did a lucky gene duplication rebuild a third channel. Our color vision is a partial recovery from an ancient loss.
The fruit-finding advantage
Why would a third cone be worth keeping? The leading answer is food. A dichromat, seeing only a blue-yellow axis, struggles to tell reddish from greenish - so ripe red or orange fruit and tender young leaves blend into the surrounding green foliage. Add the L/M distinction and those items leap out. Toggle between a dichromat and a trichromat looking at the same tree.
The same tree, two kinds of eye
A patch of green foliage scattered with ripe reddish fruit. The trichromat view is what you see; the dichromat view simulates a red-green-blind eye (like most mammals). Switch to the dichromat and the fruit sinks into the leaves - the exact problem a third cone solves, and the pressure that likely kept primate trichromacy.
Four, then two, then three
Cone count is not a ladder of progress - it zig-zags across the tree of life. The vertebrate ancestor had four cone opsins. Fish, reptiles, and birds largely kept them (and many added UV). Mammals dropped to two in the nocturnal bottleneck. Old World primates - including us - climbed back to three. Pick a lineage and see its cones and roughly how it carves up the spectrum.
How many channels, and what they buy
Each lineage's cone set, shown as sensitivity peaks along the spectrum, with a strip beneath showing roughly how finely it can distinguish hues. More cones (birds, our reptile-line ancestors) means a richer carve-up, often into the ultraviolet; the mammalian dichromat sees the world on essentially one color axis.
The L/M gene duplication
Here's the clever bit. Old World primates didn't invent a new cone from scratch - they duplicated the existing long-wavelength opsin gene, and the two copies drifted apart into separate L (long, ~564 nm) and M (medium, ~534 nm) pigments. That's why L and M peaks sit so close together compared with the far-off S (short, ~420 nm) cone: they're recent siblings. And both L and M genes sit in tandem on the X chromosome - a fact with big consequences. See how close L and M really are.
L and M: a recent split on the X chromosome
The three cone sensitivity curves. Notice L (red) and M (green) overlap heavily and peak just ~30 nm apart, while S (blue) sits far to the left - because L and M are a recent gene duplication, not ancient neighbors. The schematic shows the L and M genes side by side on the X chromosome, with S on a separate autosome.
Why color blindness is X-linked
Because the L and M genes live on the X chromosome - and sit right next to each other, inviting copying errors - red-green color vision is sex-linked. Men have a single X: one faulty L/M copy and there's no backup, so they're color blind. Women have two X chromosomes, so a good copy on the other usually rescues them. The result is stark: about 8% of men but under 0.5% of women. Cross some parents and watch the odds.
How the trait passes to sons and daughters
Choose each parent's status for the X-linked red-green gene and see the offspring probabilities, split by sex. A carrier mother passes the affected X to half her children; it shows up in sons (one X) but is usually masked in daughters (two X). This asymmetry is exactly why the condition is far more common in men.
A useful diversity
If dichromacy is a handicap, why hasn't evolution scrubbed it out? Because color vision is more of a trade-off than a straight ranking.
Takeaways and caveats
Test your understanding
Six questions on the evolution and genetics of color vision. Instant feedback, no scores recorded - a wrong answer comes with a short explanation.
Quick check
Continue your journey
The evolutionary story sets up the cones, the deficiencies, and the range of color vision across animals and people.
Human Color Vision: Cones, Opponent Signals, and the Brain
The three cones this history built, and how the brain uses them.
Vision · 6Color Blindness and Color Vision Deficiency
The modern result of that X-linked genetics, up close.
Vision · 34Animal and Non-Human Color Vision
The tetrachromats and beyond that kept what mammals lost.
Vision · 64Tetrachromacy, Anomalous Trichromacy, and the Range of Human Color Vision
The spectrum of human color vision, including a possible fourth cone.
Vision · 7Color Constancy, Adaptation, and Why Colors Change with Context
What the evolved system does with the signals it gets.
Foundations · 1What Color Is and How Humans See It
The big picture this evolutionary chapter fits into.