English translation of the original Chinese article. Publication dates and the extent of recovered text are preserved. Figures retain their original labels. Read the Chinese original.

Institute of Mechanics, Chinese Academy of Sciences
Keywords: vision, light, colour, wavelength
Back in childhood, when you picked up pencils and drew to your heart’s content, did you ever wonder why the world contains so many brilliant colours? Let us explore the science behind the colours around us.
Figure 1: Coloured pencils
Colour is the sensation produced when our eyes receive light. You have probably heard that light is an electromagnetic wave. The wavelengths that produce a visual response in humans lie between 400 nm and 700 nm, where nm means nanometres. We experience such varied colours because light at different wavelengths in this range carries different energies—in other words, its spectral composition differs. Colour perception is a sensation created when the brain processes the light received by the eyes, then classifies it as red, orange, yellow, army green, space grey, flashy gold and many other colours. To discuss how light is received, we need to introduce cone cells in the retina. Human eyes generally have three types, S, M and L, shown in Figure 2, which respond to short, medium and long wavelengths of visible light, respectively. When relatively long-wavelength light at 650 nm enters the eye, the visual signal received by the brain comes mainly from L cones, producing a sensation of red. Light at 590 nm stimulates both L and M cones, producing a sensation of yellow.
Figure 2: Sensitivity curves for the three types of cone cell [1]
The discussion so far concerns monochromatic light, which contains only one spectral component. Light with more than one component is called polychromatic light. Ordinary colour screens contain only red, green and blue pixels. As Figure 3 shows, switching on red and green at equal brightness produces a sensation of yellow. This mixed yellow light looks very similar to monochromatic light at 590 nm. Switching on other pairs of red, green and blue produces magenta and cyan; switching on all three produces white. Combining the three at different brightness levels can, of course, produce millions of colours.
Figure 3: How the red, green and blue primaries work [2]
The coloured region in Figure 4 represents all the colours the human eye can see at equal brightness. It is bounded by a curve and a straight line. Colours on the curve correspond to monochromatic light. The straight line spans the transition from red through magenta to blue. Because these colours involve simultaneous stimulation of S and L cones, light representing any point on this line, apart from its endpoints, must be polychromatic. Colours near the edge are vivid; those closer to the centre are paler, with white at the very centre. Moving from the lower-right corner towards the centre, for example, takes you through bright red, pink and white. The diagram does not represent differences in lightness or darkness, so it contains no colours such as dark red or black. The sRGB triangle is a colour space used by many advanced displays. In theory, such a screen can show every colour within that triangle; its actual performance depends on how “pure” its red, green and blue primaries are.
Figure 4: Colours in nature and their relationship to different colour spaces [2]
If you have experience with painting, you may say that red, yellow and blue pigments can be mixed to make every colour. That is approximately true, with one correction: the three pigments are magenta, yellow and cyan—the colours formed by combining pairs of primaries in Figure 3. Like sRGB, the CMYK colour space, based on cyan, magenta, yellow and black pigments, cannot represent every colour. Its range is more like the pentagonal region in Figure 5.
Figure 5: The magenta, yellow and cyan primaries used in printing [2]
Now let us discuss white. Sunlight is white, a mixture of red, orange, yellow, green, blue, indigo and violet. But in what proportions are these colours mixed? Figure 6 shows the Sun’s spectrum from infrared to ultraviolet. The outer envelope is the solar emission spectrum. After absorption and scattering by the atmosphere, the spectrum reaching the ground is the inner one. Its highest-energy region happens to fall in the green part of the spectrum. Beyond visible light, the solar spectrum also contains appreciable energy in the near-infrared and ultraviolet bands.
Figure 6: The solar spectrum [3]
Many objects around us are white, yet placing them together reveals that some look bluish and others yellowish. Among lighting sources, incandescent bulbs are yellowish, while fluorescent lamps appear whiter. Another, more specific description of white is colour temperature: the colour of light radiated by a hot black body. You have probably seen footage of steelworks. Hot steel just out of the furnace glows red, corresponding to a relatively low colour temperature, perhaps around 2000 K, or kelvin. Liquid steel at about 3000 K is white with a slight yellow tint, similar to an incandescent bulb. The enormous stars shown in science programmes can be pale blue, with surface temperatures above 7000 K—higher than the Sun’s. For white light, a yellowish tint indicates a lower colour temperature and a bluish tint a higher one, the reverse of our everyday definitions of warm and cool colours.
Of the white-light sources just mentioned, an incandescent bulb emits light because electric current heats its filament. Its colour temperature equals the filament’s temperature. A fluorescent lamp works differently: it is a cold-light source. Interactions between electrons and mercury atoms first produce ultraviolet radiation, which strikes fluorescent material that emits red, green and blue light, combining to produce white. Many LED lamps also first generate ultraviolet light and then use fluorescent materials of three colours to produce white. Although these sources all look white, missing spectral components can make objects’ colours appear unnatural. Suppose a vivid yellow object reflects only yellow light near 590 nm. Under a fluorescent lamp, which contains little light at 590 nm, it will appear dull and unnatural. Taking photographs under fluorescent lamps, or using a phone screen as fill light for selfies, is therefore not a particularly good idea.
From now on, if someone mentions white, you may ask: which kind of white? Likewise, something that looks black to us is not necessarily entirely dark. You have probably seen infrared surveillance cameras, important security equipment in many places. They illuminate a scene with infrared radiation and record that radiation, even though we can hardly see it. Similarly, because human eyes are insensitive to ultraviolet light, a spectrum containing only ultraviolet appears completely dark. Ultraviolet differs from infrared, however, in that interactions with materials readily convert it into visible light. Ultraviolet photons have high energies; after absorption, fluorescence can produce lower-energy visible photons that our eyes detect. In most fluorescence processes, emitted photons have no more energy than absorbed photons, so lower-energy infrared photons generally cannot be converted by this process into higher-energy visible photons. For example, the light from an ultraviolet banknote detector looks dim because its spectrum is mainly invisible ultraviolet with a small amount of blue-violet light. When ultraviolet is converted into visible light by fluorescence in an anti-counterfeiting feature, that area appears particularly bright. Some animals, including birds, have infrared or ultraviolet vision and see a world very different from ours.
Why do ordinary humans have this kind of vision? The solar spectrum in Figure 6 shows that sunlight is strongest in the human visible range, making vision in that range most effective. Colour vision depends on the types of cones present. Many animals have four cone types and tetrachromatic vision, allowing them to perceive a richer range of colours. These include most reptiles, birds and insects. During evolution, however, most mammals lost some cone types and some ability to distinguish colours. Most land mammals other than primates have only two cone types, while most marine mammals cannot even distinguish colours. Why, then, do humans have three? Flowering plants are closely involved. You may have seen apes roaming forests, eating fruit and casually throwing away the stones after a meal. Plants exploit this behaviour to disperse their seeds farther away and reduce competition with their own kind. But having unripe fruit picked is not worthwhile for the plant. Plants therefore guide animals towards ripe fruit in two ways. First, fruit changes colour: most fruit is green before ripening, then turns yellow or red, standing out vividly among green leaves as though advertising itself. Figure 7 compares a fruit tree under human trichromatic vision with a simulation of dichromatic vision. An ordinary human can immediately spot the ripe peaches in the left-hand image. Distinguishing red from green is certainly useful. Second, unripe fruit tastes sour and astringent; once ripe, hormones rapidly turn it sweet, encouraging eaters to tell whether it is ready. Under this natural selection, the few individuals with three cone types had a competitive advantage. Over time, most of our ancestors acquired three types. A substantial number of people today still lack the genes for trichromatic vision and have red–green colour blindness. A very small number are even said to have tetrachromatic vision!
Figure 7: The difference between trichromatic and dichromatic vision
After the example of primates and fruit trees, you may wonder how ultraviolet vision benefits other animals. The answer again involves flowering plants. Colourful flowers are pleasing to human eyes, but insects that feed on pollen and nectar need to locate their food precisely. Ultraviolet-reflecting petals act like the guiding lights of an airport runway. For birds, ultraviolet vision can help reveal traces of prey.
Humans do not normally perceive ultraviolet light directly. This depends on the transmission properties of the eye’s optical media and the response of its light-sensitive system. Light passes through the cornea, lens and vitreous before reaching the retina; the lens absorbs much near-ultraviolet light. Research has found that observers without a natural lens can have different sensitivity to near-UV. Cataract surgery usually removes a clouded lens and implants an artificial lens to restore vision. It should not be described as replacing the vitreous to add ultraviolet vision. UV can also damage eye tissues, so directly exposing the eyes to UV is not a way to experience this difference.
We have seen that colour cannot fully describe a spectrum. Conversely, the colours we name in everyday life cannot be fully explained by spectra alone. Visit a car dealership or a smartphone launch and you will discover just how many colours the world seems to have. Colour perception results from the brain’s combined processing of visual signals, including not only the light intensities detected by the three cone types, but also the overall image. The spectral compositions of silver and grey-white do not differ greatly. A grey object, however, reflects light diffusely, while a silver one reflects in a way between diffuse and mirror-like reflection, giving it a metallic lustre. Speaking of lustre, we should mention pearls. Those in Figure 8 shine quite differently from ordinary objects and have a colourful sheen. This arises from diffraction and is closely linked to their surface structure. Light reaching a pearl leaves not only at the usual reflection angle but also at other angles because of surface diffraction, creating more than one highlight under a single light source. It resembles the colourful reflections of a CD, but the less regular structure of pearls makes their iridescence softer and more natural. Diffraction effects are common in nature: some beetle shells, butterfly wings and peacock feathers produce unusual flashing colours or rainbow hues. The bright fluorescent colours made by marker pens result from conversion of ultraviolet into visible light.
Figure 8: Pearls of different colours and the surface structure of a pearl [4]
The richness of the colours we see comes from combinations of endlessly varied spectra and different surface textures. As technology advances, people may manipulate light in even more remarkable ways, and perhaps new colours will be defined.
*This article expresses the author’s personal views, not those of this website. Other media outlets, websites or individuals who reproduce it must credit the source and bear their own responsibility for copyright and other legal matters. Authors who do not wish their work to be reproduced, or who wish to discuss reproduction fees, should contact us.
References:[1] Color, Wikipedia
[2] Color Spacd, Wikipedia
[3] Julia Rosen And, Anne E. Egger, Factors-that-Control-Climate. https://www.visionlearning.com/en/library/Earth-Science/6/Factors-that-Control-Climate/234
[4] Pearl, Wikipedia
Sources for this correction: National Eye Institute · Cataracts; Griswold and Stark (1992), human near-UV sensitivity; NEI · Protecting your eyes from UV light.
Editorial note: On October 10, 2026, statements about the lens, ultraviolet vision and cataract surgery were corrected using NEI information and the original research. The original author, date and figures are retained.


