The Two Ways Birds Actually Get Their Colours
Bird feathers get their colour from either pigments or physical structure. That's it. Most people think it's just one mechanism. It isn't. Understanding which is which matters if you're trying to identify species in the field, work in avian research, or just stop confusing iridescent hummingbird throat patches with actual pigment. Pigments are chemicals built into the feather as it grows. The bird's body produces them, transports them, and deposits them into the keratin matrix of the feather. The three main classes are carotenoids, melanins, and porphyrins. Carotenoids are the most common and the most misunderstood. Birds can't synthesize them. They have to eat them. A male Northern Cardinal gets his red from beta-carotene in his diet. Without it, he's pale. I spent a season tracking house finch populations in urban Arizona and noticed something that didn't make sense in the literature — males in neighborhoods near native scrub were significantly redder than those near feeders stocked with cheap seed blends. The seed had synthetic canthaxanthin added, which fluoresces under UV but doesn't oxidize the same way natural carotenoids do. Under direct sunlight, the feeder birds looked washed out. The scrub-feeders had the deep rust colour you see in field guides. This matters for mate selection studies and basically anything involving colour scoring in wild populations.
Melanins are different. The bird makes them internally. Eumelanin gives you black and grey. Pheomelanin gives you reddish-brown. They're far more stable than carotenoids. A starling's glossy black sheen is eumelanin. It doesn't fade in light the way a cardueline finch's red does. Melanin also strengthens feather structure, which is why heavily patterned species like crows tend to have tougher feathers in high-wear areas. Porphyrins are the rare one. They produce the greens and pinks in some pigeons, parrots, and the perching bird known as the trogon. They're breakdown products of heme. You'll almost never see them outside of specific taxonomic groups, and they're what gives a budgerigar's blue-green colour when combined with structural effects.
Structural Colour — When Physics Does the Work
Structural colour comes from the nanostructure of the feather itself. Light hits layers of keratin and air, gets scattered, reflected, and interfered with. The result is often iridescence — the colour changes depending on the angle you're viewing it from. A mallard drake's head looks green from one angle and purple from another. That's not pigment. That's physics. The mechanism involves melanosomes — tiny organelles that contain melanin — arranged in precise patterns inside the feather's barbules. When these are layered in regular sequences, they create constructive interference for specific wavelengths. Disrupt the layering even slightly and the colour shifts. This is how a hummingbird produces that metallic green that no pigment could ever replicate. Blue is almost entirely structural in birds. There is no blue pigment. A blue jay's blue comes from a layered structure that scatters short wavelengths and absorbs the rest. Remove the melanin underneath and you get white. The blue doesn't come from a blue chemical — it comes from the architecture of the feather.
Get the Full Details
Why This Matters in Practice
When you're doing colour analysis on specimens, you need to know which mechanism you're dealing with. Pigment-based colours degrade over time. Specimens in museum collections will fade, especially carotenoid-derived reds and yellows. Structural colours hold up better, but they can be damaged by humidity and handling. I've seen entire drawer rows of warblers where the yellow on the breast had faded to near-white while the structurally produced blue-grey on the wings looked fine decades later. If you're using spectrophotometry to measure plumage, you'll get dramatically different readings depending on whether you're measuring a pigment or a structural colour. Pigment reflects a broad band. Structural colour produces sharp peaks and troughs in the reflectance curve. If you don't account for this, your data is garbage. I learned this the hard way during a thesis project where I compared reflectance values across three passerine families without accounting for structural interference. My statistical model flagged what I thought was significant sexual dichromatism in a monomorphic species. It turned out the males and females had identical pigment loads — the apparent difference was purely angular, caused by how their feather microstructure scattered light differently at the measurement angle.
The Overlooked Complexity
Most birds combine both mechanisms. A European goldfinch has carotenoid yellow on its wings and structural blue in its tail. A peacock has pigment-based green-brown body feathers and structural iridescence on its train. The combination is what makes avian colour so hard to model accurately. There's also the question of oil depositing. Some birds, particularly herons and tropicbirds, have been shown to deposit carotenoid-rich oils onto their feathers after preening. This modifies the colour beyond what the feather matrix itself produces. It's a post-growth modification that most introductory texts completely skip over. The take-away is simple but rarely stated clearly: you cannot understand bird colour by looking at it alone. You need to know the taxonomic group, the feather type, the lighting conditions, and whether you're even measuring the same thing when you compare two birds. The colour you see is the end result of biochemistry, physics, behaviour, and ecology all happening at once.
