⇐ PTC tasting |
Toe length ⇒
Table of Contents
Genetics of red hair
Summary
If you have red hair (like I do), you've probably heard that it's caused by a recessive allele. That's not true! There is one gene (MC1R) that has a major effect on the amount of red pigment in hair, but there are other genes that also have an effect, plus genes that affect the amount of brown pigment. Don't be alarmed if your family has two red-haired parents and a brown or blond-haired child, and if you're a teacher, don't use hair color as an example of a simple genetic trait.
Can hair colors be divided into distinct categories?
The amount and type of the pigment melanin determines your hair color. Your body makes two kinds of melanin, brown melanin (eumelanin) and red melanin (phaeomelanin). The amount of melanin determines the intensity of the hair color, which can range from pale blond to deep black or deep red. The ratio of red to black melanin determines the color. People with a high ratio of red to brown melanin have red hair, which can range from pale red ("strawberry blond") to bright red, depending on the total amount of melanin. People with a low ratio of red to brown melanin have blond, brown, or black hair. People with an intermediate ratio of red to brown melanin have brownish-red ("auburn") or reddish-brown ("chestnut") hair.
Most studies divide hair color into a small number of categories, such as blond, red, and brown. Reed (1952) criticized the subjectiveness of this approach and used a reflectance spectrophotometer to measure the amount of light reflected by hair at different wavelengths. He found a formula that used the reflectance at three wavelengths to estimate the ratio of brown to red melanin, and found that the ratio did not fall neatly into two categories; instead, some individuals had with intermediate brown/red ratios that could not easily be classified as red or non-red.
Naysmith et al. (2004) used chemical methods (Wakematsu and Ito 2002) to measure the amounts of red and brown melanin in people from northern England, with about half of the people subjectively classified as having red hair. They also found that the ratio of brown to red pigments varied smoothly; hair couldn't be divided into clearly distinct categories of red vs. non-red.
Family studies
The variety of hair colors makes it difficult to summarize the results of family studies in detail, and they're confusing and contradictory, so I'm not even going to try. Neel (1943) and Reed (1952) reviewed the various hypotheses that had been proposed, including that red was recessive; that red was dominant; that red was dominant, but could be masked by brown; or that red was usually recessive but could sometimes be dominant. Both concluded that the genetic basis for red vs. non-red hair was more complicated than a simple one-gene, two-allele model.
Genome studies
Hair color is an interesting, conspicuous trait, and forensic geneticists in particular would like to determine the hair color of a suspect using DNA from a crime scene. For that reason, there have been many studies of the molecular genetics of hair color. Over 100 genes associated with hair color have been identified, and they explain less than half the variation (Hysi et al. 2018), so it will be a long time before anyone can look at DNA data and predict the hair color of a person with perfect accuracy. For example, in a sample of 585 Danish high school students (Cabrejas-Olalla et al. 2025), analysis of their DNA predicted that 28 of them had red hair, but 9 of the 28 had blond hair and 9 had brown hair. In addition, 6 redheads were predicted to have blond or brown hair.
The ratio of red to brown melanin, however, is mostly affected by just one gene, the gene for melanocortin 1 receptor (MC1R). Valverde et al. (1995) surveyed DNA sequence variation in MC1R and found several amino acid variants that were common in red-haired people but rare in non-red people. Other researchers have followed up on this, and there are now at least 200 variant alleles of MC1R known (Garcia-Borron et al. 2014). Nine alleles have a major effect on the likelihood of having red hair, while three other alleles have a minor effect (Beaumont et al. 2007, Zorina-Lichtenwalter et al. 2019); the effect of most of the other, very rare variants on hair color is unknown.
Sulem et al. (2007) surveyed genetic variation throughout the genome of a large sample of Icelanders and found that MC1R is the only gene with a strong association with red hair. However, knowing the genotype of an individual at the MC1R locus is not enough to predict whether they have red hair with 100% accuracy. Box et al. (1997) found 13 pairs of dizygotic (non-identical) twins in which one twin had red hair and the other one didn't. For 5 of the pairs, the twins had identical MC1R sequences, so some other genes must have caused their difference in hair color. Forensic models using MC1R alleles to predict whether someone has red hair work fairly well, but they do give inaccurate predictions for some individuals (Soechtig et al. 2015, Zorina-Lichtenwalter et al. 2019). So while red vs. non-red hair is mostly determined by one gene, it doesn't fit the simple model of one-gene, two-allele inheritance; there are a lot more than two alleles of MC1R, and other genes do have some effect on red vs. non-red hair.
Red hair and pain
There have been several studies of pain tolerance that compared either redheads vs. nonreadheads, or compared MC1R genotypes (Zorina-Lichtenwalter et al. 2020, Augustinsson et al. 2024). Some studies have shown that red-haired women are more sensitive to pain than non-red-haired women, while other studies, with very similar experimental designs, have not found a difference. There doesn't seem to be good evidence for a difference in pain tolerance between red-haired and non-red-haired men. Some studies have shown that opioids are less effective for pain relief in red-haired women, while other studies have not found a difference. Some studies have shown that redheads need more anaesthetic than non-readheads, while other studies have not. Overall, there isn't anything that's definitely known about pain and red hair.
Evolution
The MC1R gene is unusual in two ways. One is that European populations show more DNA sequence variation than African populations (Rana et al. 1999, Harding et al. 2000). Most human genes show the opposite pattern, greater variation in African populations, due to prehistoric population sizes being much larger in Africa. The other weird thing about MC1R is that there are fairly large differences in allele frequency between European and African populations (Gerstenblith et al. 2007, Savage et al. 2008); most genes show fairly similar allele frequencies among human populations, because human populations have not been separate for very long. Red hair is only common in northern European populations, which has led to speculation that the alleles for red hair were favored by the lower amounts of ultraviolet radiation there, since red hair is associated with pale skin (Jablonski and Chaplin 2010). Harding et al. (2000) applied several statistical tests for DNA sequence data and found no evidence that the large amount of amino acid variation at MC1R resulted from positive selection; in particular, the large number of amino acid sites that vary within human populations are comparable to the large number of amino acid differences between human and chimp MC1R. Harding et al. (2000) did not find any amino acid variation within African populations, so they concluded that there is strong negative selection there (new amino acid mutations are selected against). They concluded that the variation outside of Africa reflects a relaxation of negative selection allowing new alleles to drift in frequency, rather than new alleles being favored. In contrast, other authors (Rana et al. 1999, Savage et al. 2008, Martinez-Cardenas 2013) concluded that MC1R was affected by positive selection, based on the greater geographic variation in allele frequency than most human genes, greater levels of polymorphism, and an unusually large number of low frequency polymorphisms.
DNA sequencing of the MC1R gene in two Neanderthal individuals found one amino acid difference between the Neanderthals and modern humans (Lalueza-Fox et al. 2007). Laboratory studies showed that the Neanderthal sequence had lower MC1R function, just like MC1R alleles in modern humans that cause red hair, so at least some Neanderthals may have had red hair. Subsequent Neanderthal sequences have not found this allele, however, so it was either rare in Neanderthals, or a sequencing error. While modern European and Asian populations have about 2% Neanderthal ancestry, the Neanderthal allele of Lalueza-Fox et al. (2007) has never been found in modern humans. One MC1R allele (Val92Met) in modern humans did come from Neanderthals; it is at a frequency of 5% in Europeans, 30% in east Asians, and 60 to 70% in indigenous Tiawanese. However, this allele has very little association with hair color, and may be more relevant to skin color and tanning (Ding et al. 2014). The MC1R region in Europeans with non-red hair has about 10% Neanderthal ancestry (including Val92Met and other alleles), an unusually high percentage, while Europeans with red hair have only 1.4% Neanderthal ancestry for the MC1R region (Danneman and Kelson 2017). So modern humans evolved red hair independently, we didn't get it from our Neanderthal ancestors.
Further research
Most studies of the effects of different MC1R alleles on hair color lump together people of different ages and sexes (both of which can affect hair color) and divide people into a small number of hair-color categories. The small number of studies that have measured red and brown melanin, using either light reflectance or chemical methods, have analyzed the ratio of the two kinds of melanin, rather than analyzing the amounts of red and brown melanin separately. It might be interesting to take a sample of people of one sex and similar ages (such as male college undergraduates), use chemical methods to measure the amount of red melanin, and see if that gives a clearer picture of the effects of the different MC1R genotypes. Or, that might be a difficult, expensive, and pointless experiment (I know it's not something I'm going to do).
References
Augustinsson, A., E. Franze, M. Almqvist, M. W. Stomberg, C. Sjöberg, and P. Jildenstå. 2024. Red-haired people's altered responsiveness to pain, analgesics, and hypnotics: myth or fact?--A narrative review. Journal of personalized medicine 14: 583.
Beaumont, K. A., S. L. Shekar, R. A. Newton, M. R. James, J. L. Stow, D. L. Duffy, and R. A. Sturm. 2007. Receptor function, dominant negative activity and phenotype correlations for MC1R variant alleles. Human Molecular Genetics 16: 2249-60.
Box, N., J. Wyeth, L. O'Gorman, N. Martin, and R. Sturm. 1997. Characterization of melanocyte stimulating hormone receptor variant alleles in twins with red hair. Human Molecular Genetics 6:1891-1897.
Cabrejas-Olalla, A., F.G. Jørgensen, J.Y. Cheng, P.C. Kjærgaard, M.H. Schierup, T. Mailun, and G. Athanasiadis. 2025. Genetic predictions of eye and hair colour in the Danish population. Forensic Science International: Genetics 78: 103267.
Dannemann, M., and J. Kelso. 2017. The contribution of Neanderthals to phenotypic variation in modern humans. American Journal of Human Genetics 101: 578-589.
Ding, Q., Y. Hu, S. Xu, C.C. Wang, H. Li, R. Zhang, S. Yan, J. Wang, and L. Jin. 2014. Neanderthal origin of the haplotypes carrying the functional variant Val92Met in the MC1R in modern humans. Molecular Biology and Evolution 31: 1994-2003.
Garcia-Borron, J.C., Z. Abdel-Malek,and C. Jimenez-Cervantes. 2014. MC1R, the cAMP pathway, and the response to solar UV: extending the horizon beyond pigmentation. Pigment Cell and Melanoma Research 27: 699-720.
Gerstenblith, M. R., A. M. Goldstein, M. C. Fargnoll, K. Peris, and M. T. Landi. 2007. Comprehensive evaluation of allele frequency differences of MC1R variants across populations. Human Mutation 28: 495-505.
Harding, R. M., et al. (11 co-authors). 2000. Evidence for variable selective pressures at MC1R. American Journal of Human Genetics 66: 1351-1361.
Hysi, P.G., et al. (45 co-authors). 2018. Genome-wide association meta-analysis of individuals of European ancestry identifies new loci explaining a substantial fraction of hair color variation and heritability. Nature Genetics 50: 652-656.
Jablonski, N.G., and G. Chaplin. 2010. Human skin pigmentation as an adaptation to UV radiation. Proceedings of the National Academy of Sciences 107: 8962-8968.
Lalueza-Fox, C., et al. (17 co-authors). 2007. A melanocortin 1 receptor allele suggests varying pigmentation among Neanderthals. Science 318: 1453-1455.
Martinez-Cadenas, C., et al. (15 co-authors). 2013. Simultaneous purifying selection on the ancestral MC1R allele and positive selection on the melanoma-risk allele V60L in south Europeans. Molecular Biology and Evolution 30: 2654-2665.
Naysmith, L., K. Waterston, T. Ha, N. Flanagan, Y. Bisset, A. Ray, K. Wakamatsu, S. Ito, and J.L. Rees. 2004. Quantitative measures of the effect of the melanocortin 1 receptor on human pigmentary status. Journal of Investigative Dermatology 122: 423-428.
Neel, J. V. 1943. Concerning the inheritance of red hair. Journal of Heredity 34: 93-96.
Rana, B.K., et al. (12 co-authors). 1999. High polymorphism at the human melanocortin 1 receptor locus. Genetics 151:1547-1557.
Reed, T. E. 1952. Red hair colour as a genetical character. Annals of Eugenics 17: 115-139.
Savage, S. A., M. R. Gerstenblith, A. M. Goldstein, L. Mirabello, M. C. Fargnoll, K. Peris, and M. T. Landi. 2008. Nucleotide diversity and population differentiation of the melanocortin 1 receptor gene, MC1R. BMC Genetics 9: 31.
Söchtig, J., et al. (12 co-authors). 2015. Exploration of SNP variants affecting hair colour prediction in Europeans. International Journal of Legal Medicine 129: 963-975.
Sulem, P., et al. (25 co-authors). 2007. Genetic determinants of hair, eye and skin pigmentation in Europeans. Nature Genetics 39: 1443-1452.
Valverde, P., E. Healy, I. Jackson, J. L. Rees, and A. J. Thody. 1995. Variants of the melanocyte-stimulating hormone receptor gene are associated with red hair and fair skin in humans. Nature Genetics 11: 328-330.
Wakamatsu, K., and S. Ito. 2002. Advanced chemical methods in melanin determination. Pigment Cell Research 15:174-183.
Zorina-Lichtenwalter, K., R.N. Lichtenwalter, D.V. Zaykin, M. Parisien, S.Gravel, A. Bortsov, and L. Diatchenko. 2019. A study in scarlet: MC1R as the main predictor of red hair and exemplar of the flip-flop effect. Human Molecular Genetics 28: 2093-2106.
OMIM entries for red hair and MC1R
Go to John McDonald's home page
This page was last revised October 26, 2026. ©2026 by John H. McDonald. Content is available under the Creative Commons Attribution-Noncommercial license.