Where feather colors come from.

February 5, 2009 7:52 AM

"Unlike virtually every other feather color, no pigment turns feathers blue. We've known that for decades. Instead, it's long been thought that a layer of cells on blue birds' feathers reflected light at blue wavelengths, similar to the phenomenon that makes the sky blue." Now, however, scientists have another explanation.

An interesting article about where birds' colors come from. See also interference.

posted by metastability (13 comments total)

As a total bird nerd, I found this really interesting. I wonder if I feed the goldfinches near me some purple seeds I can make them change color.
posted by jessamyn at 7:57 AM on February 5

"Now"? Interference has long been an explanation of, for instance, peacock feathers and other "shimmery" ones. My college physics text (my edition published no later than about 2000 but probably more like 1995) invoked this explanation for bird feathers in the chapter on light waves. It's the same thing that makes an oily puddle have a rainbow.

And I don't even know what this is supposed to mean: a layer of cells on blue birds' feathers reflected light at blue wavelengths, similar to the phenomenon that makes the sky blue. What "layer of cells" makes the sky blue? And what's the difference between a blue pigment and a layer of some substance X such that it "reflects light at blue wavelengths"?
posted by DU at 8:06 AM on February 5

Just from reading that article, its not a substance that reflects blue light, but the structure of the feather:

"In the cloudy zone of blue feathers, the melanin and air cavities are so close that the distance between them is shorter than a wavelength of light, according to research by Richard O. Prum of Yale University and his colleagues. When scattering elements are this small, they interact with light through a process called constructive interference."

They go on to explain that the feather would look gray in the shade. I'm pretty sure a pigment would retain it's color. I'm no light or color expert; maybe it's really not so unique as they describe. But I'm inclined to think so. Can't speak to the recentness of the discovery.
posted by Mister Cheese at 8:37 AM on February 5

Just from reading that article, its not a substance that reflects blue light, but the structure of the feather:

Right, I get that. The post mentions 3 theories:

1) Pigment
2) "layer of cells that reflect light at blue wavelengths"
3) This "new" theory about interference

I'm asking what the difference is between #1 and #2 and how #2 is at all related to the sky.
posted by DU at 8:45 AM on February 5

2) is Rayleigh scattering which is a different mechanism than pigmentation.

This article here, much more clearly written, may explain it better to you, DU.

I don't know about your prescient college physics text, but the physical mechanism behind coloration in bird feathers has only been recently understood (see this 2003 paper)
posted by vacapinta at 8:51 AM on February 5

Ah, sorry, I misread you. Yeah, I have no idea what they're talking about. I guess you could read it as that there's something about the cells that is not pigment based that reflects blue light. But who knows what that is. I agree that the sky analogy is really out there. Not a good way to explain things.
posted by Mister Cheese at 8:56 AM on February 5

Hey vacapinta, am I reading that the difference with bird's feathers is that it's a coherent scattering of light rather than incoherent? So the old theory was that a layer of cells scattered light like the sky: "the reflecting surfaces" were "randomized positions relative to the incident light." But they've recently discovered details that lead them to believe that the reflecting surface is in fact highly structured, leading to a coherent scattering. I'm going to start reading more about this stuff!
posted by Mister Cheese at 9:07 AM on February 5

I don't have my 5th edition Halliday & Resnick (1996) here but I do have the 7th (2005). It mentions butterflies and "color shifting ink" but not peacocks. I also remember the former from the 5th but not the latter. Possibly peacocks were never in there, possibly it was replaced with ink, possibly I just extrapolated it myself (seems pretty obvious if you know about butterflies) or possibly I learned about peacocks later and conflated it with butterflies.
posted by DU at 9:16 AM on February 5

Really great article. Thanks for posting.
posted by painquale at 9:23 AM on February 5

I went back to Frank Gill's Ornithology, the textbook I had in my ornithology class in 1998, and it says that particles of melanin in the surface cells on the feather barb are the significant factor in creating blue. This, according to the old theory, caused longer wavelengths of light (red and yellow) to pass through the surface layer to an absorbent melanin layer below, leaving the blue wavelengths to be reflected. The theory then was that the size of the melanin particles determined whether the feather was green or blue, with larger melanin particles shifting the wavelength toward greener blues.

And now it sounds like they think the 'absorbent melanin layer' itself is responsible for the blue coloring. Neat.
posted by winna at 10:38 AM on February 5

I don't have my 5th edition Halliday & Resnick (1996)

I could probably find this for you in the conference room down the hall from me at Wiley if you want.
posted by spicynuts at 8:06 PM on February 5

Didn't some company come out with makeup that was based on this principle a few years ago? Part of the marketing was that the makeup would look different on each individual. From what I recall of it, it looked weird. I think it was eyeshadow and lipstick.

I seem to recall a National Geographic article about this in the last couple of years as well. [Googles...] Yup! But more than a couple of years ago.
posted by Xoebe at 8:42 PM on February 5 [1 favorite has favorites]

This is one of those totally great and absolutely underrated posts.
posted by iamkimiam at 11:01 PM on February 5