I assume that when light goes through matter, it doesnt really slow down, but the waveform is pushed back due to some resonance with the atoms. EDIT: Interference is probably a better word than resonance here

I also assume that the above effect is responsible for the refraction index of materials.

But according to these assumptions, light rays should curve more as they go deeper through matter shouldnt they ? In other words that effect should be cumulative with the thickness of matter the light does through?

However light doesnt bend at different angles if it goes through thicker glass. So where did I go wrong?

  • 2
    $\begingroup$ If light is travelling in a uniform medium after the transition, why would it keep bending? $\endgroup$ Mar 13, 2019 at 1:37
  • $\begingroup$ because of my first and second assumptions that are dependent on thickness $\endgroup$ Mar 13, 2019 at 1:54
  • 2
    $\begingroup$ Even if your assumptions are right it doesn't make sense. What breaks the symmetry to make it bend one way or the other as it continues through the new medium? $\endgroup$ Mar 13, 2019 at 2:01
  • $\begingroup$ excellent point $\endgroup$ Mar 13, 2019 at 2:04
  • $\begingroup$ Assumption is the mother of all failure. Instead of making your own assumptions, you should use existing knowledge. $\endgroup$
    – my2cts
    Mar 13, 2019 at 7:12

3 Answers 3


I also assume that the above effect is responsible for the refraction index of materials.

Yes. The index of refraction is a quantification of how the speed of light changes due to the medium it is propagating through. So whatever mechanism slows the light down is what is responsible for the index of refraction.

But according to these assumptions, light rays should curve more as they go deeper through matter shouldnt they ? In other words that effect should be cumulative with the thickness of matter the light does through ?

No. There is nothing that would break the symmetry to cause the light to be bending one way or the other once it was in the uniform second medium.

why do prism split light at angle instead of curving it?

The index of refraction has a wavelength dependency, so if we send in something like white light that is not monochromatic, the superposition gets bent at different angles as it crosses the interfaces of the different media. Within each medium the light travels in a straight line.

  • $\begingroup$ Why the down vote? $\endgroup$ Mar 13, 2019 at 2:26
  • 1
    $\begingroup$ I think you disregarded the OP assumptions without more explanations and your answers where circular. This is a hard problem! $\endgroup$ Mar 13, 2019 at 2:26
  • $\begingroup$ @IvánMauricioBurbano What needs more explanation? And I was pointing out the circular statement, not using it. Answers that point out flaws in the OPs reasoning are considered to be on topic here. And I address the title question. In other words, I point out what is wrong and I point out the right way to think about the question. $\endgroup$ Mar 13, 2019 at 2:28
  • 1
    $\begingroup$ To your first quote: why does the OPs assumption lead to your conclusion. I don't see how it follows. To your second quote: Every experiment, at a fundamental level, yields just a number. It doesn't mean that it doesn't point at a deeper thing. In physics we want to extract from numbers explanations to deeper things. In Tobias Osborne's terms, this is a nice way to compress information. To your fourth quote: you just cited what happens in the experiment. I imagine that the OP know this. This is the "fact". Why does it happen though? $\endgroup$ Mar 13, 2019 at 2:36
  • $\begingroup$ At the end, the OP has a very nice point. Things in physics, at the fundamental level, tend not to be discontinous. Discontinuities seem to only be observed at the level of macroscopic phenomena, i.e. phase transitions. I imagine that in the transition of light from one medium to another there is a curve. This curve should oscillate around the observed angle of difraction in an amortiguated fashion to yield our macroscopic observation. A simple lattice model of the material should allow us to better understand the problem "a la Feynman". Once again, this is a hard problem. $\endgroup$ Mar 13, 2019 at 2:42

At first all light components from different wavelengths move in air at same straight line.After they fall on the surface of the prism,they get splitted according to their wavelengths . Because prism has a constant cofficient of refraction, all the light from different wavelength move in the prism at different straight line. Again in air, they moves with constant velocities at different straight line with velocity $c_0$.Thus, the light components are splitted by the prism according to their wavelengths.


Light changes speed as it moves from one medium to another (for example, from air into the glass of the prism). This speed change causes the light to be refracted and to enter the new medium at a different angle (Huygens principle). The degree of bending of the light's path depends on the angle that the incident beam of light makes with the surface, and on the ratio between the refractive indices of the two media (Snell's law). The refractive index of many materials (such as glass) varies with the wavelength or color of the light used, a phenomenon known as dispersion. This causes light of different colors to be refracted differently and to leave the prism at different angles, creating an effect similar to a rainbow.

I believe the change in direction occurs immediately upon entering the glass which is why it does not bend more. (However I cannot prove this mathematically)


  • $\begingroup$ I know all that; however the speed change should be thickness dependent according to my 1st assumption $\endgroup$ Mar 13, 2019 at 1:55
  • $\begingroup$ I think it will change speed yes, but only immediately upon entry. If it were true that light would slow per thickness of material, you could theoretically stop light with enough material which I don't think is possible.... But as I said the math is beyond me... $\endgroup$
    – Rick
    Mar 13, 2019 at 2:00
  • $\begingroup$ This is all correct, but it doesn't address the specifics of the OP's question $\endgroup$ Mar 13, 2019 at 2:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.