Roan and brindle are not similar patterns. One is among the most common coat modifiers in horses; the other is so rare it spent decades classified as an anomaly. They get conflated anyway, mostly because both patterns mix two apparent colors in a single coat (the more statistically common stripe confusion is dun and the dorsal stripe, covered separately), and at a glance, or in a photograph, or in a registry dispute, that superficial resemblance is enough to cause a misidentification.
The confusion matters because classification is not cosmetic. A horse called roan when it is brindle loses the one thing that makes its record useful: precision. Registries rely on it; researchers need it; the 1997 archive catalogued on this domain documented brindle cases at a time when calling a brindle horse anything other than a freak was itself an act of precision. This page separates the two patterns and names the mechanism behind each. Other pattern-identity problems covered on this site include tobiano versus brindle (a pattern with a chromosomal inversion behind it) and dun dorsal stripe versus brindle.
What roan is
Roan is defined by an even mixture of white and colored hairs distributed across the body, with the head, mane, tail, and lower legs retaining the base coat color. Wikipedia’s roan article (Wikidata Q1520693) describes it as “an even mixture of colored and white hairs on the body” with white hairs “more scattered or absent on the horse’s points.” The pattern is congenital (“present at birth, though it may be hard to see until after the foal coat sheds out”) and it does not progress. Unlike gray, which systematically replaces pigmented hairs with white across the entire coat, roan is stable for the horse’s life: “grays lighten with age, while roans do not.” [Wikipedia, Roan (horse)]
Visually, the demarcation at the knee and hock is sharp: an inverted V of dark lower leg where the base coat persists without admixture. The body above shows the intermixed white at roughly uniform density. Roan appears to lighten slightly in summer coat and darken in winter coat as hair density changes, but the spatial distribution does not shift. [Wikipedia, Roan (horse)]
Roan varieties are named for their base: red roan (chestnut), bay roan (bay), blue roan (black base, reading blue-cast), strawberry roan (light chestnut). The underlying mechanism is the same across varieties; the apparent color difference reflects only the base coat. [Wikipedia, Roan (horse)]
The roan gene: well-located, not yet pinpointed
Roan is dominantly inherited: a single copy of the Rn allele produces the phenotype. The locus maps to equine chromosome 3 (ECA3) within the KIT gene sequence. Marklund et al. (1999) found “highly significant linkage disequilibrium between Rn and a KIT TaqI RFLP” and “a strong KIT-Rn association in most breeds.” [Marklund S et al., Mamm Genome, 1999;10(3):283-8; confirmed in OMIA:001216-9796, last updated 2026-05-31]
But that association has not resolved into a causal mutation. As of 2025, two haplotypes (RN1, RN2) together account for roughly 74% of phenotypically roan horses tested across multiple breeds; approximately 25% of roan horses lack both. Everts et al. (2025) state explicitly: “these haplotypes are based on association only and are not likely to include the causal mutation.” [Everts RE et al., Animals (Basel), 2025;15(12):1705] The causal variant for roan remains unidentified in the peer-reviewed literature. Commercial tests exist but detect only the known haplotypes, not the molecular cause.
A separate question about roan concerns homozygotes. Hintz & van Vleck (1979) proposed that Rn/Rn homozygosity was lethal in utero. Voss et al. (2020), studying Icelandic horses, found “no evidence of lethality” in homozygous roan horses; roan-x-roan matings produced 82% roan offspring, not consistent with strict lethality. [Voss K et al., Genes (Basel), 2020;11(6):680] The lethality hypothesis is now generally regarded as disproven, though the older literature still carries it.
What brindle is
Brindle is a different thing entirely. It is a pattern of irregular stripes, eumelanin (dark) on a phaeomelanin (lighter) base, running vertically along the body and horizontally around the legs, concentrated on the neck, shoulders, and hindquarters and generally sparing the head. [Wikipedia, Brindle (Wikidata Q1969557)] The stripes are not an intermixture of hairs the way roan is; they are clonal boundaries, zones where one population of pigment cells produces dark color and another produces light, meeting at an edge that reflects how those cell populations migrated during fetal development along pathways called Blaschko’s lines.
This domain has documented brindle horses since 1997; the archive precedes the genetic characterization of the pattern by nearly two decades. Brindle was formally recorded in the scientific literature by Lusis (1942/1943), who described a brindled Russian cab horse specimen in Genetica (a specimen later preserved at the Zoological Museum of the Academy of Science in Saint Petersburg). [Lusis JA, Genetica, 1942;23:31-62, confirmed via Springer; full text paywalled; Wikipedia cites this source in the Brindle article]
Three mechanisms, one appearance
Brindle in horses is not one condition. At minimum three distinct mechanisms produce the striped phenotype. They differ in whether the pattern is heritable and in what, if anything, a genetic test will find. [Wikipedia, Brindle; Kathman, Equine Tapestry, 2024-05-09]
1. Heritable Brindle 1 (BR1): the MBTPS2 variant
In 2016, Murgiano et al. identified the first heritable brindle in a family of American Quarter Horses and Paint Horses. The causal variant is intronic: c.1437+4T→C in MBTPS2 (membrane-bound transcription factor peptidase, site 2) on the X chromosome. This variant causes aberrant splicing, producing a transcript lacking exon 10 and parts of exon 11, deleting 32 codons encoding portions of the protein’s transmembrane domain. The variant was absent from 457 control horses across 17 breeds and co-segregated perfectly with the brindle phenotype across the pedigree. [Murgiano L et al., G3 (Bethesda), 2016;6(9):2963-2970; OMIA:002021-9796]
Inheritance is X-linked and semidominant. Heterozygous mares display the characteristic vertical stripe coat with altered hair texture; hemizygous stallions carrying the mutation show only sparse mane and tail without the pronounced striped coat. The MBTPS2 gene has a human orthologue associated with X-linked genodermatoses (IFAP syndrome, Olmsted syndrome, keratosis follicularis spinulosa decalvans); the equine BR1 mutation is a milder, coat-texture-only phenotype in comparison. [Murgiano et al., 2016] A commercial genetic test for BR1 is offered by the UC Davis Veterinary Genetics Laboratory (existence confirmed via OMIA record; direct page returned HTTP 403 during research for this article). [OMIA:002021-9796]
2. Chimeric and mosaic brindle: Blaschko’s lines without a germline mutation
Some brindle horses carry no identifiable BR1 variant and are instead chimeric or somatically mosaic. In chimeric cases, two fraternal embryos fuse during early development; the resulting individual carries two distinct genomes and expresses both in a pattern that follows the developmental migration pathways of pigment cells. In mosaic cases, a somatic mutation early in development produces two genetically distinct cell lineages within one animal. Both produce Blaschko-line patterning; neither is heritable because the variation is not in the germline. [Kathman, Equine Tapestry, 2024-05-09; Wikipedia, Brindle]
Blaschko’s lines were first described by dermatologist Alfred Blaschko, who mapped the predictable pathways along which clonal populations of pigment cells (melanocytes, derived from the neural crest) migrate during embryogenesis. Chimerism can be confirmed by DNA testing showing more than two alleles at multiple loci, a signal that two genomes are present. [OMIA:000393-9796, Tetragametic chimerism in Equus caballus] Wikipedia cites two genetically confirmed chimeric brindle horses; named examples in specialist equine genetics writing include Dunbar’s Gold and Sharp One, documented by equine tapestry sources. [Kathman, 2024] These individuals cannot pass the pattern to offspring.
3. Incontinentia pigmenti (IP): brindle as a symptom of disease
A third mechanism produces brindle-like streaking as one manifestation of a multi-system disease. Incontinentia pigmenti (IP) in horses is caused by a nonsense variant in IKBKG (c.184C→T; p.Arg62*) on the X chromosome. The same family of Quarter Horses studied by Murgiano et al. carried an IP variant alongside BR1; IP was first reported in horses by Towers et al. (2013). [OMIA:001899-9796; Towers RE et al., PLoS One, 2013;8(12):e81625] Affected mares develop progressive skin lesions following Blaschko’s lines, along with dental and hoof abnormalities. Hemizygous males are typically lethal in utero. The distinguishing feature from BR1 is systemic pathology: brindle-only horses (BR1 or chimeric) lack the hoof and dental signs characteristic of IP. IP is X-linked dominant.
The diagnostic split: what separates them in the field
The patterns look different once you know what to look for. Roan distributes its white hairs evenly across the body with uniform density; there are no boundaries, no zones, no stripes. The body is blended; the head and lower legs are not. Brindle has boundaries: visible stripes where one pigment population meets another, concentrated on the neck, shoulder, and hindquarters. Roan hairs are individually white mixed into the coat; brindle stripes are zones of pigment difference where the hair texture may also differ (in BR1 horses, striped hairs have a distinct, less straight texture alongside the color difference). [Murgiano et al., 2016]
Age behavior separates them definitively. Roan is stable: the pattern at one year is the pattern at fifteen, and its distribution does not change. A gray horse will progressively lighten; a roan will not. Brindle is also stable in a different sense: the stripes are present from birth and remain, though seasonal coat changes may affect their visibility. The key point is that roan does not stripe and brindle does not blend uniformly across the body.
Seasonally, roan appears to show more white in summer coat and darker in winter coat as overall hair density changes, a phenotypic observation documented in Wikipedia but mechanistically unexplained in the literature to date. Brindle does not show this whole-coat density shift.
Why roan cannot be confused for brindle at the genetic level
Roan maps to ECA3/KIT. Heritable brindle (BR1) maps to the X chromosome at MBTPS2. Chimeric brindle has no single locus. Incontinentia pigmenti maps to X/IKBKG. These are unrelated genes with unrelated inheritance patterns and unrelated cellular mechanisms. A roan horse tested for BR1 will be negative. A BR1 mare tested for roan’s haplotype markers will not return a roan-positive result. At the laboratory level, the confusion does not survive a genetic workup.
The confusion lives in photographs and registry records where color names are applied by visual assessment without molecular support. A horse with a light neck and darker body could be either, but the stripe versus blend distinction, and the presence or absence of the characteristic dark head and dark lower leg in roan, resolves most cases in the field without testing.
What remains unresolved
Roan’s causal mutation has not been identified as of 2025, despite strong localization to ECA3/KIT. Approximately 25% of phenotypically roan horses lack the known RN1 and RN2 haplotypes, meaning current commercial tests miss a material fraction of roan horses. [Everts et al., Animals, 2025] The homozygous lethality hypothesis for roan is now generally regarded as disproven following Voss et al. (2020), but the older 1979 literature still circulates.
For brindle: the BR1/MBTPS2 variant explains heritable brindle in the characterized Quarter Horse/Paint family and was absent from 457 controls. Whether additional heritable brindle loci exist in other breeds is an open question; Wikipedia’s brindle article notes that “one or more genes are responsible” but only one has been characterized. The precise boundary between non-IP, non-BR1, non-chimeric Blaschko-line brindle cases and the three confirmed mechanisms is not crisply delineated in the literature. The sooty-redistribution hypothesis for brindle, mentioned in some older reviews, has not been confirmed by a published genetic study and is not treated as fact here. [Wikipedia, Brindle]
References
- Wikipedia, “Roan (horse)” (Wikidata Q1520693). Verified 2026-06-03.
- OMIA:001216-9796: Coat colour, roan in Equus caballus. Last updated 2026-05-31.
- Marklund S, Moller M, Sandberg K, Andersson L. “Close association between sequence polymorphisms in the KIT gene and the roan coat color in horses.” Mamm Genome. 1999;10(3):283-8. PMID 10051325.
- Everts RE, et al. “Identification of three haplotypes associated with the roan coat color in horses using whole-genome sequencing.” Animals (Basel). 2025;15(12):1705. PMC12189688.
- Voss K, Tetens J, Thaller G, Becker D. “Genomic analyses reveal no evidence for the lethality of homozygous roan in Icelandic horses.” Genes (Basel). 2020;11(6):680. PMC7348759.
- Wikipedia, “Brindle” (Wikidata Q1969557). Verified 2026-06-03.
- Murgiano L, Waluk DP, Towers R, et al. “An Intronic MBTPS2 Variant Results in a Splicing Defect in Horses with Brindle Coat Texture.” G3 (Bethesda). 2016;6(9):2963-2970. doi:10.1534/g3.116.032433. PMC5015953.
- OMIA:002021-9796: Brindle 1 in Equus caballus. Last updated 2026-05-31.
- Towers RE, Murgiano L, Millar DS, et al. “A Nonsense Mutation in the IKBKG Gene in Mares with Incontinentia Pigmenti.” PLoS One. 2013;8(12):e81625. PMID 24324710.
- OMIA:001899-9796: Incontinentia pigmenti in Equus caballus. Verified 2026-06-03.
- OMIA:000393-9796: Tetragametic chimerism in Equus caballus. Verified 2026-06-03.
- Kathman L. “Mosaicism in Horses, Part 1.” Equine Tapestry. 2024-05-09. equinetapestry.com.
- Lusis JA. “Striping patterns in domestic horses.” Genetica. 1942;23:31-62. doi:10.1007/BF01763802. [Paywalled; Springer abstract confirmed; content cited via Wikipedia.]
Roan and brindle are both defined at the level of the gene, and the vocabulary of alleles, loci, and inheritance patterns that underpins this distinction is laid out plainly at horse-info.org’s gene entry. There is also a practical diagnostic note: sweet itch, a hypersensitivity to midge bites, produces diffuse hair loss and coat disruption across the topline and hindquarters that can temporarily create a mottled appearance in photographs. Sickhorses.com covers the full presentation at sweet itch and insect allergy, a condition worth ruling out before attributing a mixed-hair coat pattern to genetics.