Short answer
Hair greys because follicles stop putting pigment into the fibre; there is no grey pigment, only pigmented hairs and unpigmented ones mixed, and an unpigmented hair looks white because it scatters light. In a survey of 4,192 people, 74% of those aged 45 to 65 had some grey, at an average of 27% of their hair. Measured on photographs, the contrast between facial features and the skin around them is also lower in older faces, and the eye's own lens passes less blue light with age. Together these shift the lightness relationships round a face. They are changes in optics, not faults, and nothing here is health advice.
A hair fibre is a translucent shaft of keratin. Its colour comes from melanin granules inside it, which absorb some of the light passing through; the more eumelanin, the less light comes back. Take the melanin away and nothing is left to absorb, so light entering the fibre is bent, reflected and scattered back out at every wavelength alike — and anything that scatters all wavelengths equally looks white, as clouds and snow do. No grey pigment exists. A head of hair that reads as grey is a mixture of fibres that still carry pigment and fibres that carry none, blended by the eye because individual hairs are too fine to resolve at ordinary distance. The beard page describes that blending; the point here is its arithmetic. The eye averages light, not lightness, and white fibres return many times more light than dark ones, so a modest share of white hairs moves the overall shade a long way, as the table shows.
Because white hair has no colour of its own to assert, it shows the colour of the light: warmer under tungsten-like lamps, cooler in open shade, tinted by a bright collar beneath it.
Each hair is coloured by pigment cells in its own follicle, renewed at every growth cycle from a small store of stem cells. Work on human follicles found a few active pigment cells remaining in grey hairs and none detectable in white ones, and work in mice and human samples linked greying to those stem cells failing to maintain themselves. A 2023 study that tracked the cells in living mouse skin refined the picture: the stem cells normally move between two regions of the follicle, switching between a reserve state and a pigment-making one, and with age more of them become stranded where they no longer supply pigment cells. That is mechanism research, mostly in mice. It does not say when any person will grey, and the US National Library of Medicine's summary is that almost everyone does eventually and that why is not fully understood. Colourwise makes no claim about slowing or reversing it.
The most useful numbers come from a survey published in the British Journal of Dermatology in 2012, in which trained assessors compared the natural hair of 4,192 healthy volunteers around the world with standard swatches. Among people aged 45 to 65, 74% had some grey hair, and on average 27% of their hair was grey. The saying that by fifty half of people are at least half grey did not hold anywhere: the share with at least half their hair grey at that age ran from 6% to 23% across the groups sampled. Men had more grey hair than women, and both the age it began and how fast it advanced differed between the populations surveyed and with natural hair colour. Those are descriptions of groups. They say nothing about an individual, and a person's hair colour is not evidence of their age or background.
The authors worked for a cosmetics company's research division. The survey's counts are descriptive and do not depend on any product.
Hair is only one edge of the frame. A study of standardised photographs of 763 women aged 20 to 80, living in France, the United States and South Africa, measured the difference in luminance and colour between the eyes, lips and brows and the skin around them. Most of those differences were smaller in older faces, in every one of the four groups. Add hair that has moved from dark towards white and the lightness pattern Colourwise's contrast pages describe is not fixed for life: someone whose colouring once gave a strong light-and-dark pattern may find it has become a softer, lighter one, or, with dark skin and white hair, a stronger one in a new place. The study was of women only and compared different people at one moment instead of following anyone over time. Colourwise uses it for the measurement and not for its framing of contrast as something to be put back.
Judging these changes in a mirror uses eyes that have aged alongside the hair. The lens absorbs more short-wavelength light with the years; on the average rates used on the ageing-eyes page, about 33% as much violet light reaches the retina at 60 as at 20. The brain re-centres its sense of white, so the world does not look yellow, but fine distinctions among blues, violets and near-whites get harder, especially in dim rooms. White hair against a pale cool top, or the difference between a cream and a blue-white near the face, is exactly that kind of distinction. The practical answer is the same one that page gives: make colour decisions in bright daylight, and if it matters, ask someone else what they see. Sudden changes in colour vision, glare or haloes are a reason to see an eye-care professional.
| Measure | Survey result | The saying |
|---|---|---|
| People aged 45–65 with some grey hair | 74% | Not addressed |
| Average share of hair that was grey, ages 45–65 | 27% | Not addressed |
| People with at least half their hair grey at 50 | 6–23%, depending on the group sampled | 50% |
| Share of white fibres | Mixed with dark brown: L* | Mixed with mid brown: L* |
|---|---|---|
| 0% | 18 | 36 |
| 10% | 38 | 47 |
| 27% | 55 | 60 |
| 50% | 70 | 72 |
| 75% | 82 | 83 |
| 100% | 91 | 91 |
Why: Unpigmented hair reflects whatever light reaches it: warm lamps in one room, cool daylight or shade in the other.
Fix: Judge its colour in daylight by a window; the hair has not changed between rooms.
Why: The lightness of the hair framing the face has changed, so the garment now contrasts with it differently.
Fix: Repeat the window comparison with a few lighter and darker versions of the colour and keep whichever is liked now.
Why: White fibres are far brighter than dark ones, and flash or overhead light picks them out; the camera may also clip them to pure white.
Fix: Use soft light from the front and lower the exposure slightly so the lightest hair keeps its detail.
Each statement is labelled by kind — established fact, a standard’s requirement, observed market data, a convention, or Colourwise’s own interpretation or analysis — with the strength of the evidence behind it.
FactModerate evidence
Grey and white hair result from little or no melanin in the fibre, not from a grey pigment; hair colour otherwise depends on the amounts of eumelanin and pheomelanin.
Source: MedlinePlus Genetics: Is hair color determined by genetics?; Human hair color
FactModerate evidence
Optical measurements model a hair fibre as a transparent cylinder with an absorbing interior and a scaly surface, so that light is both reflected from the surface and transmitted through the fibre before re-emerging.
Caveat: A scattering model built from measurements of pigmented fibres; that an unpigmented fibre looks white because nothing absorbs within it is Colourwise's inference from the model.
FactModerate evidence· Multiple countries
In a survey of 4,192 healthy volunteers, 74% of people aged 45 to 65 had some grey hair, with a mean intensity of 27%, and the proportion with at least 50% grey hair at age 50 ranged from 6% to 23% between the groups sampled.
Measured: 4,192 healthy men and women, assessed against standard hair swatches. Method: Expert visual assessment of natural hair colour by age, gender and origin
Caveat: One survey by a cosmetics company's research division; volunteers, not a random sample. Group figures cannot be applied to an individual.
FactModerate evidence
In mouse hair follicles, pigment stem cells move between follicle compartments and switch reversibly between stem and more mature states, and with ageing a growing number become stranded and no longer contribute pigment cells.
Caveat: Mouse study; it describes a mechanism and does not establish the timing or course of greying in people, or any way of changing it.
FactModerate evidence
In standardised photographs of 763 women aged 20 to 80 in four groups, most aspects of luminance and colour contrast between facial features and the surrounding skin were lower in older faces.
Measured: 763 women aged 20–80 living in Paris, the eastern United States and Pretoria.
Caveat: Women only; cross-sectional, so it compares different people of different ages; several authors were employed by a cosmetics company.
Colourwise analysisLimited evidence
If the eye simply averaged the light from a flat mixture of dark brown and white fibres, the survey's mean of 27% white would raise CIELAB lightness from about 18 to about 55, well over half the way to an all-white head at about 91.
Based on: Colourwise calculation: luminance of the mixture is the area-weighted average of two example fibre colours in linear light, converted to CIELAB L*. See the table.
Caveat: An upper bound from a flat mosaic. Real hair lies in layers that shade one another and white fibres are translucent, so the visible shift is smaller; example colours, not measurements of hair.
Source: Greying of the human hair: a worldwide survey, revisiting the '50' rule of thumb (Panhard, Lozano & Loussouarn, British Journal of Dermatology 167, 865–873, 2012); CIE 015:2018 Colorimetry, 4th edition
Colourwise interpretationModerate evidence
Greying and lower feature contrast change the lightness relationships between hair, skin and clothing without making any colour unsuitable; Colourwise treats them as a reason to look again, not as something to correct.
Based on: Colourwise's reading of the greying survey and the facial-contrast measurements against its own contrast-level descriptions, and its content policy for personal appearance.
Reviewed 1 October 2026. Colourwise summarises its sources in its own words and does not reproduce standards text or proprietary colour data. Spotted an error? Tell us.