Questions About Light

A biologist friend of mine recently emailed me with questions about the nature of light, so I figured I'd share his questions and my response. My favorite part is part 3 about the color magenta.

Hi Anne,

I'm teaching my kids in freshman bio about spectrophotometry tomorrow. To really understand it well you need to know how light wavelengths and filters work. I understand it well enough to teach it, but looking into color theory led me to some questions I am utterly confused about - we thought maybe you could help. Ok, here we go:

1. Supposed facts:
- White light is actually a combination of all light wavelengths from ~400nm to ~800nm
- White light can be made by shining blue light (450-475nm), green light (495nm-570nm) and red light (620-750nm)
Problem I have:
- If I make white light from red, blue and green light, how can I then derive yellow light (570-590nm) or cyan light (476-495nm) from it? How can white light be BOTH a combination of all wavelengths AND be successfully formed from three specific wavelength bands? If I use only three bands there are several holes in the spectrum.

2. Red light (620-750nm) plus yellow light (570-590nm) makes orange light (590-620nm).
- Does this mean that we perceive the light as orange as an average of the red and yellow - OR do the wavelenghts in the range 590-620 ACTUALLY exist - if so, from where did they origniate? Did the red wavelenths somehow shorten and did the yellow lengthen? If so how? Did the red wavelentghs somehow cause the yellow to lengthen? Likewise did the yellow cause the red to lengthen?

3. Similar to problem 2 - if I mix blue light (450-475nm) and red light (620-750nm) I get violet (magenta) light (380-450nm).
- How can mixing these two lights result in a wavelenth BELOW the lowest wavelength of origin?


Clearly I don't know much about light theory and it seems to me that it isn't a simple matter of adding and subtracting.

Anytime I do an internet search about "how to make magenta light" all I get is information telling me to mix red and blue - yes, but that doesn't tell me HOW the light (i.e. the wavelength range) is formed.

Any insight you have would be very welcomed!!!

Thanks so much!


My response:
Hey!

Wow. Those are _really_ good questions, and I had to do some poking around to fill in my own gaps... ok, big gaps in my understanding. Here is what I've got for you, not total answers, but I hope it helps:

1. Since "white" is not really a color, you could say that the color white is just a construct in our minds. I wouldn't even say it's always made up of "all the colors". It's just the "color" we see when each of the 3 kinds of cones in our eyes are stimulated with an equivalent amount of red, green, and blue light. Though you could also get white from simply mixing cyan, magenta, and yellow monochromatic lights. So... yes! There can be holes. In fact, I'm pretty sure that most "white" light is full of imperceptible holes anyway, known as absorption and/or emission lines. They're not the same thing, but basically amount to the same idea: white light is not necessarily always made of "all the colors".
Here's more info on absorption/emission spectra:
http://physics.ucr.edu/~wudka/Physics7/Notes_www/node107.html

2. I think it's good here to keep in mind that light is just the superposition (or overlapping) of many frequencies, so while I think orange can exist on its own (say as it's emitted from some glowing element), if you had orange light, composed of a red source and a yellow source, you could break it down through a prism, into those components. The yellow is not longer and the red is not shorter. It's a question of these two being superimposed and perceived as one color in this case - similar to white light that's composed of some set of other colors (red, green, blue, or magenta, cyan, yellow). This is governed by something called chromaticity, which is, as you suggested, a weighted average of the wavelengths of the ingredient wavelengths.
More here: http://www.handprint.com/HP/WCL/color5.html

3. Ok, so here's where things get super weird. I don't think it's really fair to say that magenta has a wavelength. Magenta: not specifically included in the electromagnetic spectrum. You _can't_ get magenta without mixing two colors. I know. That's weird. Magenta on the color wheel is that special place where the two ends of the spectrum meet. I would think that our brains would just take an average like every other superposition, but no, in this case, we mix and get something entirely new. But then, why should I be surprised? When we mixed red, blue, and green we got white - also not in the spectrum.
more here:
http://www.null-hypothesis.co.uk/science/strange-but-true/colour_spectrum_magenta_complimentary_bizarre

I hope this has been helpful. I certainly had fun digging into this stuff a little deeper.
Cheers!

*** not included in the email: I suspect that all the non-spectral colors (including magenta, brown, grey, white) are physical examples of emergence. Emergence is one of my favorite concepts in the universe.