From Wikipedia, the free encyclopedia
Jump to: navigation, search
edit·history·watch·refresh Stock post message.svg To-do list for Rainbow:

Here are some tasks awaiting attention:
  • Expand : Polarization of rainbow light;[1] rainbows on Titan;[2] infrared rainbow;[3] (PDF) derivation of the 40-42° angle.[4]
  • Verify : more inline citations are needed.
Former featured article Rainbow is a former featured article. Please see the links under Article milestones below for its original nomination page (for older articles, check the nomination archive) and why it was removed.
Main Page trophy This article appeared on Wikipedia's Main Page as Today's featured article on March 20, 2004.
Article milestones
Date Process Result
January 19, 2004 Featured article candidate Promoted
January 21, 2006 Featured article review Kept
February 11, 2007 Featured article review Demoted
Current status: Former featured article
          This article is of interest to the following WikiProjects:
WikiProject Meteorology (Rated C-class, Low-importance)
WikiProject icon This article is within the scope of WikiProject Meteorology, a collaborative effort to improve the coverage of Meteorology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Low  This article has been rated as Low-importance on the project's importance scale.
WikiProject Physics (Rated C-class, Low-importance)
WikiProject icon This article is within the scope of WikiProject Physics, a collaborative effort to improve the coverage of Physics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Low  This article has been rated as Low-importance on the project's importance scale.
Wikipedia Version 1.0 Editorial Team / v0.5
WikiProject icon This article has been reviewed by the Version 1.0 Editorial Team.
Taskforce icon
This article has been selected for Version 0.5 and subsequent release versions of Wikipedia.
C-Class article C  This article has been rated as C-Class on the quality scale.

Graphic not aligned with text and nature[edit]


The above graphic shows red as the inner most color whereas in nature and in the text red has the smaller angle i.e. less than 42 degrees, and is on the outside as can be seen on pictures of rainbows.

This is a common confusion, but the graphic is accurate. This video explains the concept nicely (starting from 1:20). Drabkikker (talk) 19:06, 14 June 2015 (UTC)

While the graphic may be accurate for a single ray of incoming sunlight, somewhere near 60° incident on the drop, the graphic (1) does not accurately explain anything about the formation of rainbows, (2) is only indirectly referenced in the article, and (3) that reference is not only hard to understand, it (4) does not explain the formation of rainbows in any way, only higher-order complications.

There are two thoughts about what causes rainbows in popular literature - and one is wrong. It sounds intuitive and doesn't require effort to understand, which is why it is popular, but it is completely invalid. It says that, after one internal reflection, any red light that exits the raindrop is at about 42° and any violet light that exits the raindrop is at about 40°. This is what the graphic in question suggests, and it needs to be removed from the article.

The real cause of rainbows is stated in the start of the paragraph that indirectly references the graphic, but it is stated incorrectly: "The reason the returning light is most intense at about 42° is that this is a turning point – light hitting the outermost ring of the drop gets returned at less that 42°." The incorrect part is "the outermost ring of the drop" - as I said, the turning point is light that is incident at about 60°. The graphic above the one in question here demonstrates this. I'm intending to re-write this paragraph, but I want to get prior cooperation before I do so we don't get into edit wars. The paragraph will explain that exiting light is brightest near the turning point, which is why colored bands appear. It will then explain that the turning point - not the exit angle for a single incoming ray - of each color is different due to dispersion. And finally, it will say that the turning point for violet is less than that for red, which is why violet appear below read in the rainbow. (This is actually not the best - the angle should be measured from the direction of the original ray, not the observation direction. The deflection angles are 140° for violet and 138° for red. But it is much easier to understand.)

This new paragraph will not try to explain effects due to the finite width of the sun (which the current one is correct about) or luminance not going to infinity (which it is not - luminance from any single point on the sun does go to infinity). JeffJor (talk) 12:12, 18 June 2015 (UTC)

Sounds like an appropriate improvement. I agree this topic should be addressed properly on the page. (Incidentally, the video I linked to above addresses it near the end, albeit in laymen's terms.) By all means go ahead! Drabkikker (talk) 20:57, 19 June 2015 (UTC)


Anaxagoras could be mentioned. — Preceding unsigned comment added by (talk) 17:16, 7 March 2015 (UTC)

According to this the only surviving fragment from Anaxagoras that deals with the rainbow is "We call the reflection of the sun in the clouds the rainbow". That's not enough. Ceinturion (talk) 13:46, 8 March 2015 (UTC)

Colors of rainbow count[edit]

Is the colors of the rainbow is 6 colors: red, orange, yellow, green, blue and violet (purple)? — Preceding unsigned comment added by (talk) 09:57, 9 March 2015 (UTC)

Color order in secondary rainbows[edit]

The article currently says "Secondary rainbows are caused by a double reflection of sunlight inside the raindrops, and appear 10° outside of the primary rainbow at an angle of 50–53°. As a result of the second reflection, the colours of a secondary rainbow are inverted compared to the primary bow, with blue on the outside and red on the inside." This is incorrect. The number of reflections has nothing whatsoever to do with the order in which see the colors.

I've twice tried to correct it, with understandable text. Twice it has been reverted, with the claim that it was unintelligible. I suspect that it is because it contradicts commonly held, but incorrect, beliefs about how rainbows are formed. Specifically, So I'm going to try, once again. If you still think it is unintelligible, please tell me hear what it is you think is unclear so we can make it clearer to you. JeffJor (talk) 22:51, 25 April 2015 (UTC)

I'm going to revert for now. Please provide a source for your additions. Thanks. Gandydancer (talk) 01:11, 26 April 2015 (UTC)
Thanks, Gandydancer. Sorry to say, JeffJor, but your edit (at least the second attempt) is not so much unintelligible as it is plain incorrect. I understand what you're trying to say with the whole "wrapping around the zenith" reasoning, but no such process is involved here. Please see Drabkikker (talk) 09:14, 26 April 2015 (UTC)

Unfortunately, the formation of rainbows is a subject where some incredibly obvious-sounding explanations turn out to be inaccurate at best, and sometimes outright wrong. Expert explanations aimed at non-experts have to compromise between giving inaccurate impressions that match what people expect without being explicitly incorrect, and losing them in more exact detail. And Wikipedia editors need to understand how and when such compromises are made.

The site Drabkikker linked to does not say what you both see to think it says, it says what I did. And I used it as a reference in my first attempt - for the second, I was in a hurry. For example, the specific page that was linked only seems to say that each color of light in an Nth order rainbow follows only the path shown. What it omits is that these angles represent the minimum deviation angle in a wide range of angles, as properly described on and There are even sliders so you can show most of these paths.

I said "most" because the sliders don't let you take the original ray down to an incident angle of 0°. In the primary bow, that ray reflects straight back to the sun. This is a deflection angle of 180°, but the non-expert usage in the Wikipedia article calls it 0° ("The overall effect is that part of the incoming light is reflected back over the range of 0° to 42°"). The more detailed page in our reference calls it 180° down to "about 137.5°," the minimum deviation. This is where the colors are seen, but light from one reflection is seen at all of those angles. So the primary bow is a complete disk, not just the rim, and it's center is seen at a deflection angle of 180° (the anti-solar point).

The secondary bow's slider doesn't let you go anywhere near an incident angle of 0°; or even angles that end up away from the sun. But if you manipulate it you should be able to convince yourself that smaller incident angles do (it probably couldn't be programmed as easily). In fact, the deviations for the secondary bow go from 360° degrees (technically not 0°, it actually reverses twice) to a minimum of about 230 to 233°, as the reference describes in various places. It is also a disk, with its center on the sun. It wraps past the zenith (deviation 270°, just beyond the range of the slider), all the way to deviations 230° to 233°. Red deviates least (230°, seen at 50° when looking away from the sun),which is on the OUTSIDE of the disk. It just seems to be on the inside. JeffJor (talk) 14:46, 26 April 2015 (UTC)

Alright. It seems we're both looking at the same phenomenon in very different ways. I follow your reasoning and I'm willing to be proven wrong, but could you provide any sources, other than the page at Atmospheric Optics, which we both appear to interpret differently? Drabkikker (talk) 20:12, 26 April 2015 (UTC)

With no disrespect intended, I'm looking at it from the point of correct physics, and you seem to be looking at it from the point of folklore that you don't want to accept is incorrect. Many of my references are textbooks, which don't try to explain it the same way - just teach. So they don't spend much time describing what is seen, but you can get that from the equations. The one I like is Or

The summary is that when the original incident angle is A (all angles from 0 to 90 are present), light enters the drop at angle B=asin(sin(A)/k) where k is the index of refraction. It deflects (A-B) as it enters, (180-2B) at each reflection, and another (A-B) as it exists. This makes the net deflection after N reflections D(A,N) = 180N + 2A - 2(N+1)B. Try plotting it, if you want - just don't truncate it to a range of 360 degrees.

When A is 0, D=180N which is the "center" of all the reflected light. It is the anti-solar point when N is odd, and the sun when N is even. As A increases, D decreases for a while until it hits a minimum. The light becomes highly concentrated at the minimum deflection angle. Since each color has a different minimum deflection angle, colored bands appear. Red is always the "lowest" minimum, so it is always "inside" and is the only monochrome band. All the others are mixes of lower-frequency colors, that are perceived as the color of the concentrated light. Inside the violet band, it doesn't stop - it simply fades to white. This means that the rainbow exists "inside" the colored bands, as white light. Alexander's Band is the absence of this white light. JeffJor (talk) 23:51, 26 April 2015 (UTC)

Thanks for your reply. No disrespect meant here either, and I apologize for calling your explanation incorrect earlier. It took me a while to wrap my head around the concept, but I now understand what you are saying. Yes, I was aware of the fact that rainbows are disks rather than mere bands, it just never dawned on me that the secondary rainbow is a "negative disk", so to speak. That's actually very interesting, and I thank you for broadening my mind! Perhaps it's useful to add the references you gave to the Wikipedia page. Even so, couldn't we still maintain that the color reversal (or the apparent reversal, if you wish) of the secondary bow is the (indirect) result of two internal reflections? Drabkikker (talk) 04:53, 27 April 2015 (UTC)

"Wrap my head around the concept" ? Pun intended?

There are many "folklore" explanations of rainbows: "Colors are created by the physical separation (implying isolation) of colors." No, they are caused by the concentration of each color in a different place, causing it to dominate over others. "Think of the raindrops as prisms." A lens, or a flashlight's parabolic mirror, is a much better analogy. "They occur at the specific angles where total internal reflection occurs." Not only is TIR unrelated, it is impossible. "Colors are reversed in the secondary because the additional reflection inverts the order." Again, with no disrespect intended, these incorrect statements seem so obviously correct, that they are almost universally accepted as true. Most explanations that recognize the differences will still use what I call "fuzzy language" that allows those who believe these untruths to continue doing so. If they didn't, they would be disbelieved just like you did. I even expected it here. So colors are explained by "separation of colors," which is true, but not the way people will assume the term implies.

Can we still maintain that the color reversal is due to two reflections? Almost. We can also maintain that day and night are caused by the sun moving in a circle around the Earth. In fact, that is a better explanation, since motion is relative and the true cause - Earth rotation - can always be translated into equivalent orbits. The same isn't true for color order. Find a liquid whose indices of refraction for violet and red light, respectively, are 1.15 and 1.13. The primary rainbow will be between 79.2° (violet) and 85.2° (red). The secondary will be in the same direction between 13.7° and 23.9°. The order is the same, because the secondary disk will wrap around the anti-solar point as well. And while it is hypothetical, for the indices I use for water (1.344 and 1.331), the odd/even=inside/outside ordering changes around the 12th or 14th order rainbow. JeffJor (talk) 12:00, 27 April 2015 (UTC)

Understood. Thanks once more! (As for the pun, it wasn't so much intended as serendipitous ;)) Drabkikker (talk) 19:02, 27 April 2015 (UTC)

No axial symmetry[edit]

The text mentions that there is an axial symmetry along the parallel sun rays. This is obviously wrong. Fixing any particular axis parallel to a sun ray clearly shows a lack of rotational symmetry. (talk) 11:30, 13 July 2015 (UTC)

I'm not quite sure if I understand what you mean. Care to elaborate? Drabkikker (talk) 06:21, 14 July 2015 (UTC)