# 1145: Sky Color

 Sky Color Title text: Feynman recounted another good one upperclassmen would use on freshmen physics students: When you look at words in a mirror, how come they're reversed left to right but not top to bottom? What's special about the horizontal axis?

## Explanation

The point of this comic is that often, curious children ask their parents simple questions about understanding how the world works. Often, although the question is simple, the answer is not. "Why is the sky blue?" is a common example, since most parents are not familiar with Rayleigh scattering, and thus are unable to answer the question.

Randall's hobby is to make those questions even harder, so that the parents who are familiar with the subject (scientists, for example) will be stumped.

Another point of this comic is that we often think that we understand a scientific phenomenon (e.g., "Why is the sky blue?"). However, a certain simple question (e.g., "Why isn't the sky violet?") can often uncover large gaps in our actual understanding.

Rayleigh scattering is the phenomenon that explains the color of the sky, where light of every wavelength gets scattered in the air by the inverse quartic (fourth power) of its wavelength as given in the comic. In the visible spectrum, blue light has a wavelength of 450–495 nm while violet has a shorter wavelength of 380–450 nm. Violet light does indeed get scattered more than blue light, however the lower portion of the spectrum for sunlight consists of blue light and eyes are much more sensitive to blue light than violet light. Furthermore, the sunlight contains more blue than violet to begin with as a result of the surface temperature of the sun. This leaves the impression of a blue sky. A good explanation, including why blue and not violet, can be found in Usenet Physics FAQ :: Why is the sky blue?, but note that human color perception is more complicated than described there. xkcd later tackles the same question in 1818: Rayleigh Scattering.

The title text refers to a mirror image and is discussed by the famous American theoretical physicist Richard Feynman in a famous BBC documentary [1], as one of the problems which he used to have fun with first years (British English for first year student or fresher).

A mirror image is a virtual image produced by the reflection of light on a mirror. It's common to think of images in mirrors as being reversed left-to-right, as any text held in front of us will appear flipped. This is actually an issue of perception. In a plane mirror, images are reflected directly: the left side of your body will be reflected in the left side of the mirror, and vice-versa. The source of confusion is that people tend to think of a mirror image the way we would think of a person facing us. When another person faces us, they turn around the vertical axis, placing their right hand on our left side, so seeing our left hand on our left side in a reflection seems like an inversion, even though it's a direct representation. By the same token, in order to hold text up to a mirror, we generally flip it around the vertical axis, so that the start of the text is on right, and the end on the left (in English, at least). When the mirror reflects this, we see the text as backward, but the mirror hasn't reversed it, we reverse it when we turn it toward a mirror.

In other words, the vertical axis is only "special" because we're used to objects turning around it, so we come to expect that reversal, instead of a reflection.

You can induce a mirror to reverse left and right only --- by standing next to it instead of in front of it, facing along the plane of the mirror itself. If you lift your right arm, you can clearly see your image's left arm raising, without having to adjust for frame of reference. Similarly, you can induce a mirror to reverse top and bottom only by holding it flat above your head or laying it flat on the ground and standing on it (or perhaps standing under a suitably equipped bedroom ceiling). See this video for a demonstration.

## Transcript

[Jill and her mother, Megan, but with her hair up. Megan is at a desk and facing the girl.]
Jill: Mommy, why is the sky blue?
Megan: Rayleigh scattering! Short wavelengths get scattered way more (proportional to 1/λ4). Blue light dominates because it's so short.
Jill: Oh.
Jill: So why isn't the sky violet?
Megan: Well, because, uh... ...hmm.
[Caption Below the panel:]
My hobby: Teaching tricky questions to the children of my scientist friends.

# Discussion

I think most explanations of the the mirror issue overlook an even simpler explanation -- Things appear to be flipped such that left and right are reversed. However, that is only because you are used to things, such as people, rotating about a vertical axis, with top and bottom staying in the same position. If a clone of you stood on its head and you faced each other, your right arms would be on the same side (e.g. "closer to the door"). Now, if you look in a mirror, it is the same as seeing the clone flipped top to bottom. 24.41.66.114 01:06, 6 September 2013 (UTC)

I have read that when you are in front of the mirror, the reason that it seems to reverse the horizontal or x-axis (left-right) view but not the vertical or y-axis view, is that the mirror actually reverses the z-axis front-back. The part nearest to the mirror reverses. Example, if you hold your right gloved hand in front of the mirror with the 5 fingers pointing to the mirror, the image that you see in front of the mirror is not similar to the left-hand glove, but it is the right glove turned inside-out. To test, wear a right latex glove, write something on it, then remove the glove, making the right glove inside out. If you can read what you had just wrote, it had reversed horizontally, but not vertically, assuming same frames of reference. 188.114.106.101 (talk) (please sign your comments with ~~~~)

Of course with vertical mirror vertical axis is selected: perceived switching of left and right (really close with far to mirror surface). When standing on horizontal mirror we will perceive switching bottom from top. --JakubNarebski (talk) 09:09, 10 December 2012 (UTC)

You're certainly correct, but I think that the original question is not really asking about text (or other things) which are perpendicular to the mirror, but rather text which is parallel to it (and thus the close vs. far doesn't come into it). For example, when reading signs in your rear view mirror or holding a book in front of your chest while looking in a mirror. I've added a little bit to the explanation to attempt to help clarify what's happening in that situation. I'm not sure if it really helps or not. KeithyIrwin (talk) 10:00, 10 December 2012 (UTC)

If the part you added was "mirrors don't reverse things, turning around does", then I think that that is the best description. If you can see text in a mirror, then that means that you have turned away from the text. And normally, humans turn around the z-axis. If, rather than turning away from a sign to look at a mirror on the opposite wall, you instead somersaulted backward into a handstand, then you would see the sign reversed vertically instead of horizontally. 108.162.249.76 22:33, 5 November 2017 (UTC)

Easier way to describe it: Imagine you hold a piece of glass. Write on the glass and hold it in front of the mirror, so that you can see both the original text and the mirrored text. Both versions of the text will look identical. So the mirror doesn't change anything. 62.220.2.194 11:10, 10 December 2012 (UTC)

Another way: draw a line between the real object and its reflection. Things are reflected around that line. If that line is going up & down (relative to your eyes), then things are reflected left/right (relative to your eyes). If that line is horizontal (again relative to your eyes), then things are reflected top/bottom. So it's not so much whether the mirror is horizontal or vertical, but rather what direction you are looking into the mirror (although that can be influenced a lot by the mirror's orientation).CityZen (talk) 04:17, 11 December 2012 (UTC)

I always wonder: Since the sky goes from red to blue to red and the optical spectrum goes from red to green to blue. How come the sky is never green?

Because of human color perception. You only perceive green in polychromatic light when said light is stronger in the middle wavelengths than the low or high wavelengths; in other words, you would need a process in the sky that removed both the high and low wavelengths from white light. As the sun sets, only the lower wavelengths are removed, so you perceive yellows and reds -- this perception of color is "one-sided", i.e. it is not interfered with by even longer wavelengths. By the way, sometimes you do see green briefly in the sky, it's called a Green Flash. --Prooffreader (talk) 16:41, 10 December 2012 (UTC)
I used to go outside after a rain storm during the day, and sometimes the sky would seem very green. The effect could last for hours. 76.122.5.96 12:15, 12 December 2012 (UTC)
The sky is green, at times. Growing up in the Upper Midwest (USA), I quickly learned that green sky means it's time to watch out for tornadoes. I don't know the actual connection between the two situations--I would guess from the previous comment that whatever atmospheric conditions create tornado conditions also "edit out" both high and low wavelengths, at least to a degree. 129.176.151.14 14:44, 12 December 2012 (UTC)

This sentence doesn't make sense: "(from "his" right to left instead of from "his" left to right)" Trek7553 (talk) 15:15, 10 December 2012 (UTC)

Repeat Character Watch: The girl has appeared previously in 842: Mark, 892: Null Hypothesis, 1058: Old-Timers, and 1104: Feathers (A similar looking character also appears in 635: Locke and Demosthenes but this is actually the character Valentine from the book Ender's Game). The mother is seen in comics 806: Tech Support and 813: One-Liners. lcarsos_a (talk) 18:12, 10 December 2012 (UTC)

About this edition: 1/(x^4) does not look like a root to me. IMHO the forth root of x would be more like x^(1/4) but it's not the formula from the comic. (I'm too lazy to try to type lambda). Lmpk (talk) 19:00, 10 December 2012 (UTC)

You are correct. It's been fixed. The editor that made that edit was probably confusing 1/x4 with x1/4, the latter of which would indeed be the fourth root. lcarsos_a (talk) 19:53, 10 December 2012 (UTC)

This page, linked from the explanation says that "the most strongly scattered indigo and violet wavelengths stimulate the red cones slightly as well as the blue, which is why these colours appear blue with an added red tinge." -- this seems rather strange. Assuming the cones are simulated based on frequency/wavelength, ultra-blue colors shouldn't stimulate the red cones because the electromagnetic spectrum is linear, not circular, despite the appearance of similarity between violet and red. Or am I missing something? --Waldir (talk) 16:14, 11 December 2012 (UTC)

If you look at the response curve (middle of cited page) you'll see that red receptors have two peaks, one in the red wavelengths, and another (very tiny one) in the violet. That's why purple (which is red + blue) looks so similar to violet, and why the "color wheel" works. 207.225.239.130 21:59, 11 December 2012 (UTC)
PS: "first years" is an idiom. Wouldn't that be "first year students" to be proper English? 207.225.239.130 22:05, 11 December 2012 (UTC)
Idioms are "proper English" too. There is no doubt about what is meant here (or at least, I hope there isn't, but perhaps there are regional differences that mean some English speakers don't say "first years" to talk about students in their first year), and the register is not unduly colloquial for this kind of a site. 14:00, 12 December 2012 (UTC)
The easiest way to explain mirrors is: they don't change left and right, they change forward and backward. What is farther from the mirror appears farther in the mirror. If you look at yourself, your nose and the nose of your reflection are the closest parts of the body together (at least in a bathroom mirror :-)), so *if there were* another person standing where the mirror *simulates* it, that person would wave it's right arm when you wave your left. But in a "absolute reference frame", both image and original wave their arm nearest to the door. Interestingly automobile drivers don't make this error: if you see a car in the rear mirror blinking left you don't assume they want to turn right...

There was a hilarious Get Fuzzy strip where Rob tried to explain why the sky is blue to Satchel, but I can't find it. This one?98.174.41.183 00:49, 31 December 2012 (UTC) Yes that's the one. --Smartin (talk) 04:23, 1 January 2013 (UTC)

I would also just like to add that, as I understand it, the Sun puts out a lot more blue light than violet light, so it would make sense for blue to dominate. After green light, where the Sun's output peaks, the intensity of the light starts dropping dramatically. 71.104.183.59 (talk) (please sign your comments with ~~~~)

I don't think that woman is Megan. She looks like a black-haired version of Ponytail. 108.162.219.203 (talk) (please sign your comments with ~~~~)

Have to agree. Doesn't look like Megan. 108.162.210.177 23:53, 22 June 2015 (UTC)

The mirror explanation is convoluted. The question tricks you into thinking there's right/left symmetry being mixed up, when the symmetry is actually across the surface of the mirror. That's it. 108.162.215.94 01:41, 28 September 2014 (UTC)

Clifford Stoll, in his book "Cuckoo's Egg", reports being asked why the sky was blue in an oral examination by Astronomy professors. Every time he thought he'd explained in sufficient detail, they would say "go on…" until he'd been talking for an hour on the topic. AmbroseChapel (talk) 01:22, 4 September 2017 (UTC)

My quick explanation for the mirror thing is that a mirror doesn't show the world reversed, it shows it inside-out. AmbroseChapel (talk) 01:22, 4 September 2017 (UTC)