The question of why the sky is blue is one of the older questions in the history of human curiosity. Aristotle attempted an answer in the fourth century BCE. Newton, in the seventeenth century, suggested it had something to do with the reflection of light from the ocean. Ancient civilisations across multiple cultures proposed mystical, religious, and aesthetic explanations of varying poetic merit and zero physical accuracy. The actual answer turned out to be substantially weirder than any of these traditional accounts, and was not fully worked out until the late nineteenth and early twentieth centuries — when the British physicist John William Strutt, the third Baron Rayleigh, derived the mathematical formula that bears his name, and Albert Einstein, in 1911, supplied the final theoretical underpinning. The blue of the sky, it turns out, is not the colour of air. Air, in any small quantity, has no colour. The blue is the colour of light being bent by the molecules of air, in a particular pattern that depends on wavelength, and that produces the specific shade of blue that humans see overhead on any clear day.
According to a reference summary of the physics maintained by the mathematician John Baez at UC Riverside, the first crucial step toward the modern understanding was taken by John Tyndall in 1859. Tyndall demonstrated, experimentally, that when light passes through a clear liquid containing small particles in suspension — a tank of water with a little milk mixed in, for example — the shorter blue wavelengths are scattered more strongly than the longer red ones. From the side of the tank, the beam appears bluish. Looking through the long axis of the tank, the transmitted light appears reddish. The geometry, Tyndall recognised, was identical to the geometry of the daytime sky versus the sunset. Lord Rayleigh derived the precise mathematical relationship governing the scattering in 1871, showing that the intensity of scattered light is inversely proportional to the fourth power of the wavelength. Both Tyndall and Rayleigh, however, initially attributed the sky’s blue to small particles of dust and droplets of water vapour suspended in the atmosphere, rather than to the gas molecules themselves. Subsequent scientists realised that this dust-and-water-vapour framework predicted more variation of sky colour with humidity and haze conditions than was actually observed, and that the molecules of nitrogen and oxygen in the air must be sufficient scatterers on their own. The case was definitively settled by Albert Einstein in 1911, who calculated the detailed formula for the scattering of light by individual molecules and showed that it agreed precisely with observation.
What Rayleigh scattering actually does
The Rayleigh scattering formula, which has been a fixture of undergraduate physics curricula for more than a century, states that the intensity of scattered light is inversely proportional to the fourth power of the wavelength. The practical implication is straightforward but striking. Blue light, with a wavelength of approximately 450 nanometres, is scattered roughly 5.5 times more strongly than red light, with a wavelength of approximately 700 nanometres. As sunlight enters the upper atmosphere and travels downward toward an observer on the ground, every gas molecule it encounters scatters a small fraction of the incoming light in random directions — but the fraction is heavily weighted toward the blue end of the spectrum. By the time the light reaches the observer, the direct beam from the Sun has been depleted of some of its blue content, and the indirect light arriving from all other directions in the sky is dominated by the scattered blue.
Per the UK Met Office’s reference explanation of the phenomenon, the human eye perceives this combination as a uniformly blue sky overhead, even though the underlying sunlight is white and the underlying air is colourless. The blue is, in a literal sense, a property of the geometry — of the angle between the observer, the Sun, and the column of atmosphere through which the light has travelled. Looking straight up on a clear day, the observer sees blue because they are looking through a column of atmosphere whose molecules have been preferentially scattering blue light toward them all day. Looking directly at the Sun (which one should not do), the observer sees the residual yellowish-white of sunlight that has been partially depleted of its blue. Looking at the Sun near the horizon at sunset, the observer sees red — because at that geometry, the sunlight has had to travel through approximately 38 times more atmosphere than it would at noon, and almost all of the blue has been scattered out before the light reaches the eye.
Why the sky is not violet
A logical question, given the wavelength dependence, is why the sky is not violet rather than blue — since violet light, with an even shorter wavelength than blue, should be scattered even more strongly. As reported by a physics-questions page maintained by Christopher Baird at West Texas A&M University, three factors combine to produce the observed blue rather than violet. First, the Sun emits less violet light than blue light to begin with — solar emission peaks closer to the green-yellow part of the spectrum and tapers off toward both red and violet. Second, the human eye is substantially less sensitive to violet than to blue; the retina’s cone cells are tuned to a colour space in which blue is more salient than violet. Third, some violet light is absorbed by molecules in the upper atmosphere before it can be scattered downward. The combined effect is that the sky appears blue rather than violet, even though violet is being scattered slightly more strongly than blue.
The same physics also explains a related puzzle: the actual colour of the Sun itself. From space, the Sun is essentially white — its emission spectrum, integrated across all visible wavelengths, produces a colour that the human visual system reads as white with a very slight bluish tint. From the surface of Earth, the Sun appears yellowish-white because some of the blue light has been scattered out of the direct beam during its passage through the atmosphere. The yellow colour of the Sun, in other words, is not a property of the Sun. It is a property of looking at the Sun through a Rayleigh-scattering medium. From space, astronauts see a white Sun.
The clouds and the other planets
As discussed in a 2024 explainer at When Notes Fly on the history and physics of sky colour, the absence of Rayleigh-style colour selectivity also explains why clouds appear white. Clouds are made of water droplets and ice crystals that are substantially larger than the wavelength of visible light — typically tens to hundreds of micrometres across, compared to the ~0.3-nanometre size of an air molecule and the ~500-nanometre wavelength of visible light. At those sizes, the scattering is governed by a different formula, called Mie scattering, which does not strongly favour any particular wavelength. The result is that clouds scatter all wavelengths of light roughly equally, which the eye perceives as white.
The Rayleigh-scattering explanation also predicts that the sky on other planets will have different colours, depending on what their atmospheres are made of and how thick they are. Mars, with its thin carbon-dioxide atmosphere and substantial dust load, has a brownish-red sky during the day and a blue tint near the Sun at sunset — essentially the reverse of Earth, where the everyday sky is blue and the sunset is red. Titan, Saturn’s largest moon, has a thick orange haze of hydrocarbons that produces an orange-yellow sky. Venus, with its dense and chemically complex atmosphere, has a yellowish-orange sky. The blue overhead on Earth is, in cosmic terms, a specific property of a specific combination of atmospheric chemistry, atmospheric depth, and human visual physiology — produced by the interaction of light with molecules so small that, taken individually, they are invisible, and that collectively produce the colour everyone above the cloud line has been looking at for the entire history of the species.
