Light as a wave

Two pictures of light competed for a long time. Descartes, in 1637, treated light as a stream of particles (the corpuscular model) and obtained Snell's law from it. Newton developed the idea in his book Opticks, and the model is usually credited to him because that book was so widely read.

In 1678 the Dutch physicist Christiaan Huygens proposed instead that light is a wave. Both models could account for reflection and refraction; they disagreed about the speed of light after refraction.

For a ray that bends towards the normal, the particle model needs light to go faster in the second medium, while the wave model needs it to go slower.

Foucault settled the question in 1850 by measuring light to be slower in water than in air, as the wave model says.

Thomas Young's interference experiment of 1801 put the wave nature of light on firm ground. For about forty years after it, interference and diffraction experiments kept turning up results that only waves could explain.

The wavelength of visible light is tiny: yellow light has a wavelength of roughly 0.6 μm, that is 6 × 10⁻⁷ m. Next to the size of ordinary mirrors and lenses this is so small that light seems to move in straight lines.

Geometrical (ray) optics is the limit in which the finite wavelength is ignored altogether. A ray is then the path along which energy travels as the wavelength tends to zero.

One objection remained: a wave was thought to need a medium, yet light crosses empty space. Maxwell answered it. His equations predicted electromagnetic waves travelling at a speed very close to the measured speed of light, so light is an electromagnetic wave: changing electric and magnetic fields that sustain each other and need no medium.