Huygens' Principle

Drop a small pebble into the stepwell at Adalaj and watch the wavefront move. Circles grow outward from the splash, each one a perfect ring, each ring getting larger at the same steady speed. Now imagine you had to predict, without watching, where the ring will be one second from now. You could say: each point on the current ring is, in effect, sending out its own tiny ripple in every direction; the new ring is just where all those tiny ripples pile up constructively. If that sounds like a trick, it is — but it is a trick that works, and it is the trick that explains how light moves.

Christiaan Huygens proposed this in 1678, two centuries before anyone knew what light actually was. You do not need to know. You only need one idea: a wavefront is a population of in-phase oscillators; each one radiates; their combined forward radiation is the next wavefront. With this rule, reflection and refraction are theorems, diffraction is inevitable, and the door to Young's double-slit experiment — the experiment that finally settled whether light is a wave — is wide open.

Wavefronts — lines of equal phase

A wave is an oscillation that spreads. Stand anywhere in its path and you feel the medium go up and down (for a water wave), or the electric field swing back and forth (for a light wave). Two points in the wave's region are said to be in phase if they are doing the same thing at the same time — both at the top of their swing, or both passing through zero going up. The set of all points that are simultaneously in phase forms a surface in space. That surface is a wavefront.

For a pebble in a pond, the wavefronts are circles on the surface of the water, each circle representing a crest (all points on one circle are at their maximum height simultaneously). For a sound wave from a small speaker, the wavefronts are spheres. For sunlight at the Earth's surface, the Sun is so far away that the spherical wavefronts have expanded to be almost flat — you get plane wavefronts, thin flat sheets of in-phase oscillation travelling in one direction.

The direction a wave moves is always perpendicular to its wavefront. An imaginary straight line drawn from the source, perpendicular to every wavefront it crosses, is called a ray. Rays and wavefronts are dual descriptions: rays show the direction of energy flow, wavefronts show the geometry of phase. Ray optics works with rays; wave optics works with wavefronts.

Huygens' principle — the construction

Now the single idea of this chapter. Huygens asked: if I know the wavefront right now, how do I find the wavefront a little later?

His answer has two parts.

Part 1. Every point on the current wavefront is itself a source of secondary wavelets. Each of those wavelets is a tiny spherical wave centred at that point, expanding outward at the same speed v as the original wave in that medium.

Part 2. After a time \Delta t, each secondary wavelet is a sphere of radius v\,\Delta t centred on its origin point. The new wavefront is the forward envelope of all these spheres — the smooth surface that just touches all of them on the forward side.

That is it. Two sentences. Nothing more.

If the current wavefront is a plane, every wavelet is the same size, and the envelope on the forward side is another plane — parallel to the first, shifted forward by v\,\Delta t. A plane wavefront stays a plane.

If the current wavefront is a sphere of radius r centred on a point source, every point on the sphere sends out a wavelet of radius v\,\Delta t, and the forward envelope is a sphere of radius r + v\,\Delta t centred on the same point. A spherical wavefront stays spherical.

When the wavefront hits a boundary — a water surface, a glass slab, a slit cut in a screen — the construction still works, but something more interesting happens. That is where reflection, refraction, and diffraction come from.

Reflection from Huygens

A plane wavefront travels toward a plane mirror at angle \theta_i to the normal. As the wavefront meets the mirror, not every part of it arrives at the mirror at the same moment — the leading edge of the wavefront hits first, the trailing edge last. While the leading edge is waiting for the rest of the wavefront to catch up, it is sending out secondary wavelets that travel back into the original medium (the mirror blocks forward motion). By the time the trailing edge finally reaches the mirror, the leading edge's wavelet has had time to expand significantly. Draw the envelope of all the wavelets — only the ones above the mirror surface contribute — and you get a new wavefront reflecting away from the mirror at angle \theta_r.

The geometry is this. Call the incident wavefront AB (length \ell), with A just touching the mirror and B still above it. Let the wavefront travel at speed v. Point A sends out a wavelet; by the time B reaches the mirror (at the new point B′), a time \Delta t = \overline{BB'}/v has passed, and A's wavelet has grown to radius v\,\Delta t = \overline{BB'}. The reflected wavefront is the line from B′ tangent to A's wavelet — call its other endpoint A′.

Deriving \theta_i = \theta_r. In the figure, the incident wavefront AB is perpendicular to the incident ray, so the angle it makes with the mirror equals \theta_i (angles with perpendicular sides). Similarly the reflected wavefront A'B' makes angle \theta_r with the mirror. Now:

• Right triangle ABB' has hypotenuse \overline{AB'} and opposite leg \overline{BB'}. So \sin\theta_i = \overline{BB'}/\overline{AB'}.
• Right triangle AA'B' has hypotenuse \overline{AB'} and opposite leg \overline{AA'}. So \sin\theta_r = \overline{AA'}/\overline{AB'}.

By Huygens, \overline{AA'} = v\,\Delta t = \overline{BB'}. Therefore \sin\theta_i = \sin\theta_r, and since both angles are between 0 and 90°, \theta_i = \theta_r.

Why: the law of reflection is just the geometry of "A's wavelet expanded at the same speed for the same time as B travelled forward." Both legs equal v\Delta t, so the two right triangles are congruent, and congruent triangles have equal angles. Huygens turns an empirical law into a triangle proof.

Refraction from Huygens — and Snell's law

Now point the same incident wavefront at a water surface. Above the surface, light travels at speed v_1; below the surface, at a slower speed v_2 < v_1. Same game: A hits first, and now A's wavelet is a sphere inside the water, growing at the slower speed v_2. Meanwhile B is still in air, travelling at v_1, until it reaches B' after time \Delta t. At that moment A's wavelet has radius v_2 \Delta t, while \overline{BB'} = v_1 \Delta t. The refracted wavefront is the tangent from B' to A's smaller wavelet, bending closer to the normal.

Deriving Snell's law. In triangle ABB' (right angle at B): \sin\theta_1 = \overline{BB'}/\overline{AB'} = v_1 \Delta t / \overline{AB'}. In triangle AA'B' (right angle at A'): \sin\theta_2 = \overline{AA'}/\overline{AB'} = v_2 \Delta t / \overline{AB'}. Divide:

Now use the definition of refractive index, n = c/v. Then v_1 = c/n_1 and v_2 = c/n_2, so v_1/v_2 = n_2/n_1:

That is Snell's law — derived from Huygens' principle in about four lines of geometry.

Why: refraction is the geometry of "wavelets inside the slower medium have had less time to spread than the wavefront in the faster medium outside still travelling." The ratio of the wavelet radii is the ratio of the speeds, which by n = c/v is the inverse ratio of the refractive indices. Snell's law is not a postulate of wave optics — it is a theorem.

And this derivation tells you something ray optics cannot: light slows down in a denser medium. A ray diagram would let you draw refraction either way; Huygens tells you the bending is only consistent with a slower wave inside water, glass, diamond. Isaac Newton (who preferred a corpuscular picture) thought light should speed up in denser media; Huygens said no, it must slow down. In 1850, Foucault measured the speed of light in water and found it to be slower than in air — settling the question.

Why light bends around edges — diffraction in one gesture

Send a plane wave at a wall with a small hole in it. Everything except the part of the wavefront that reaches the hole is blocked. The only sources of secondary wavelets in the hole are the points inside the hole — a tiny patch of wavefront.

If the hole is much larger than the wavelength, the wavelets from the many points inside the hole add up to produce a nearly plane wave on the far side (with slight fringing at the edges). If the hole is comparable to the wavelength or smaller, only one or two wavelet sources fit inside, and the wave on the far side is almost spherical, spreading out in every direction on the transmitted side. The wave bends around the edge.

This is diffraction. Ray optics has no way to predict it — rays don't curve. Huygens does: when you have very few wavelet sources, there is no envelope that looks like a plane, and the natural shape of a few wavelets added together is a spherical front. Diffraction is not a correction to geometric optics; it is what you get when you do wave optics honestly and the geometric limit no longer applies. The details — fringe spacings, intensity patterns — come out in Diffraction from a Single Slit. What matters here is the gesture: a narrow aperture lets Huygens' wavelets spread.

You have seen this. On a smoggy Delhi evening, streetlamps have halos around them. Sound from another room reaches you even though you cannot see the speaker — long-wavelength sound diffracts around the doorway. Water ripples bending into a harbour mouth. All the same principle, applied to different waves.

Worked examples

Common confusions

• "Secondary wavelets spread in every direction. Why don't we get a backward wave?" Huygens' original formulation does not explain why the backward envelope is ignored — it has to be postulated. This was a known gap for two centuries, until Kirchhoff in the 19th century derived a corrected diffraction formula with an obliquity factor that suppresses the backward lobe. For the level of problems in JEE and class 12, the forward-only rule is taken as given, and it always gives the right answer.

• "If every point on a wavefront is a source, isn't the wavelet density infinite?" The principle is best read as a limiting statement: imagine the wavefront is covered by many point sources, and the field on a new wavefront is the superposition of their wavelets. In Kirchhoff's rigorous version, the "sum" becomes an integral, and the fact that there are infinitely many sources is handled by standard calculus. The pedagogical version with a handful of dots is just a finite approximation to the integral.

• "Wavefronts and rays — what's the difference?" A wavefront is a surface of equal phase (where the wave is, geometrically). A ray is a line that traces the direction of energy flow, perpendicular to every wavefront it crosses. They are dual descriptions of the same wave. For a plane wave, rays are straight parallel lines; for a spherical wave, rays are radii emanating from the source.

• "Does Huygens' principle apply to all waves, or only to light?" All waves. Water ripples, sound waves, seismic waves, and electromagnetic waves all obey a version of the principle (with the speed v being the appropriate wave speed in each medium). The famous cases in optics are only the most historically influential applications.

• "Why does the wavelength shrink in glass but not the frequency?" Frequency is set by the source — the oscillator that created the wave. A boundary cannot change it (the oscillation at the boundary has to be continuous). The speed changes because the medium responds differently. Since v = f\lambda and f is fixed, \lambda must change to match the new v. You can think of the wavelets inside the denser medium as being spaced closer together because they move more slowly in the same amount of time.

• "Huygens' principle says light slows in glass. But doesn't relativity forbid slower light?" No — relativity forbids anything carrying information faster than c in vacuum. In a medium, the phase of the wave moves at c/n, which is less than c. This is fine; the individual photons still travel at c between interactions with the medium's atoms, and the slowdown is an effective bulk behaviour from all the scattering and re-emission. The refractive index is an emergent quantity of a material, not a violation of relativity.

Where this leads next

• Young's Double Slit Experiment — the experiment that proved light is a wave, and the first quantitative triumph of Huygens' principle applied to two point sources.
• Coherence and Conditions for Interference — why two ordinary bulbs don't interfere even though both emit Huygens wavelets, and what "coherent" actually means.
• Diffraction from a Single Slit — the intensity pattern when the width of a slit is comparable to the wavelength, computed by integrating secondary wavelets across the slit.
• Refraction of Light — Snell's Law — the ray-optics companion to Huygens' refraction derivation, with worked examples on slabs and apparent depth.
• Properties of Electromagnetic Waves — the physical content of "what is actually oscillating in a light wave" and why the wave equation applies to \vec{E} and \vec{B}.