Jared Edward Reser, Ph.D.
Article type: Hypothesis and research program
Date: September 2026
Abstract
Childhood blondness is usually treated as a weak or temporary version of adult hair pigmentation. This article develops a different possibility: light childhood hair may become visible when population-specific pigment-reducing variants act on an older, age-regulated program of follicular pigmentation. The model was first proposed as a comparison between human childhood blondness and primate natal coats (Reser, 2026a). It does not require blond children to possess a literal nonhuman-primate natal coat. Rather, it proposes a candidate developmental homology. Mammalian and primate ancestors already possessed mechanisms that altered hair pigmentation across life stages, and recent human variants may change the gain, threshold, or duration of this conserved system.
Several findings make the hypothesis testable. Longitudinal human studies show a reproducible, multiphasic hair-color trajectory during the first five years of life, accompanied by changes in shaft diameter and medullation. Twin data indicate strong genetic control of early color change. In Europeans, a causal blond-associated enhancer variant near KITLG weakens a LEF1 binding site and reduces hair-follicle enhancer activity. In Solomon Islanders, a recessive TYRP1 R93C variant produces blond hair through a distinct molecular route; published cross-sectional results suggest that age-related darkening is attenuated in homozygotes, although a formal genotype-by-age interaction has not been reported. Among nonhuman primates, contrasting natal coats are widespread. Limestone langurs provide an especially useful natural experiment because light orange infants later develop dark coats despite carrying an MC1R substitution with elevated basal signaling in vitro. Experimental work further shows that WNT signaling coordinates epithelial and melanocyte stem cells during pigmented hair regeneration and that a postnatal pulse of Kitl expression can have durable pigmentary effects.
No existing result proves that human childhood blondness and primate natal-coat transitions use the same orthologous regulatory program. The hypothesis nevertheless generates discriminating predictions. The decisive evidence would be a shared developmental follicle state in humans and a natal-coat primate, a human pigment allele that changes the slope or timing of that state, and an allele-swap experiment that changes developmental pigment output. This article separates that mechanistic question from competing explanations for why light-hair alleles spread, including sexual selection, ultraviolet adaptation, cold-related pleiotropy, and drift.
Keywords: childhood blondness; natal coat; hair pigmentation; ontogeny; KITLG; TYRP1; MC1R; melanocyte; primate evolution; developmental homology; heterochrony
1. Introduction
Many people who have brown hair as adults were conspicuously blond as children. The change is familiar from family photographs, yet familiarity has encouraged a shallow explanation. Childhood blondness is often described as if the adult pigment level were simply slow to arrive. That description may be correct at one level, but it leaves the central biological question unanswered: why is pigment output age-regulated, and what evolutionary history produced the regulatory system on which human variants now act?
In a previous essay, I proposed that childhood blondness might be related to the broader primate phenomenon of natal-coat coloration (Reser, 2026a). Many primates are born with pelage that differs sharply from the adult coat. White, orange, golden, dark, or patterned infant coats can persist for weeks, months, or, in some apes, several years. These coats demonstrate that primate follicles can occupy life-stage-specific pigmentary states. My earlier proposal was that human childhood blondness could be a reduced and prolonged expression of this ancient developmental capacity. This followed a broader argument that conspicuous human hair changes can act as life-stage cues and social signals (Reser, 2026b).
The comparison needs careful definition. Human childhood blondness is not identical in appearance, duration, body distribution, or known function to any particular monkey’s natal coat. The term “natal coat” properly refers to the species-typical pelage present around birth in a nonhuman animal. The human phenomenon considered here is better called juvenile-light hair coloration, defined as hair produced during infancy or childhood that is substantially lighter than hair produced by the same individual later in development. The proposed relationship is therefore not one-to-one phenotypic identity. It is a hypothesis about conserved developmental machinery.
The central claim is that population-specific hypopigmenting alleles can reveal, prolong, or attenuate an ancestral age-regulated follicle state. The alleles need not create an entirely new developmental program. They may lower pigment output enough for a normally subtle juvenile-adult difference to cross a visible threshold. The European KITLG blond-associated enhancer allele and the Oceanic TYRP1 R93C allele are important because they produce similar visible outcomes through different molecular routes. This convergence is expected if several genetic “dimmer switches” can act on the same age-sensitive pigment system.
This article evaluates the model as a falsifiable research program. It asks four separate questions:
1. Do humans possess an intrinsic age-regulated hair-pigmentation program?
2. Do particular human pigmentation variants change that program rather than merely lowering color at every age?
3. Is the human program developmentally homologous to a nonhuman-primate natal-coat transition?
4. If the mechanism is real, what evolutionary process caused the relevant alleles to spread?
The evidence for these questions is unequal. The first is well supported. The second has strong molecular plausibility and suggestive human data. The third remains untested. The fourth is likely to differ among populations and must not be inferred from mechanism alone.
2. From analogy to a formal developmental hypothesis
The refined hypothesis can be stated as follows:
> Human childhood blondness occurs when one or more population-specific pigment-reducing variants modify the gain, threshold, or duration of a conserved age-sensitive hair-follicle program. That program may be evolutionarily related to the regulatory architecture that produces natal-coat transitions in other primates.
This formulation makes three clarifications.
First, the ancestral feature need not have been blond hair. What may be ancestral is the capacity to change follicular pigment output with age. An early anthropoid or later hominoid ancestor could have possessed a modest age-dependent shift, a localized infant marking, or a more conspicuous natal coat. Existing evidence cannot identify which visible state was ancestral.
Second, modern light-hair alleles may be recent even if the machinery they modify is ancient. Evolution often changes the regulation or output of an existing developmental system. A variant that weakens a follicular enhancer, alters a melanosomal enzyme, or changes receptor signaling can make an otherwise inconspicuous juvenile state visible.
Third, developmental mechanism and selective history are different problems. Suppose a European allele delays childhood darkening through a conserved follicular pathway. That finding would not establish whether the allele spread through sexual selection, drift, linked selection, cold adaptation, or another process. Conversely, evidence of sexual selection would not show that the developmental mechanism is homologous to a primate natal-coat program.
A simple conceptual model makes the claim more precise. Let pigment output at age t be determined by an age-dependent program, A(t), multiplied or transformed by genotype-specific gain, G, plus structural and environmental effects:
[
P_i(t) = h{A(t),G_i} + S_i(t) + E_i(t).
]
Here, P is pigment measured in newly produced hair, S includes shaft diameter, medullation, and hair-cycle state, and E includes ultraviolet exposure, nutrition, and cosmetic treatment. Under a constant-reduction model, genotype shifts the curve up or down but does not alter its shape. Under the developmental-modifier model, genotype changes the curve’s slope, curvature, or transition age. This distinction is empirically testable.
3. Human hair color is a developmental phenotype
3.1 Longitudinal evidence
The best early-childhood study followed 232 healthy Prague children, 114 boys and 118 girls, from one month to five years of age. Hair was sampled at 1, 3, 6, 9, and 12 months and then at six-month intervals. Darker shades were relatively common during the first six months, lighter shades predominated from approximately nine months to two and a half years, and progressive darkening followed between ages three and five (Prokopec et al., 2000). This is not merely a linear increase in pigment from birth. It is a multiphasic trajectory that includes an early lightening phase followed by darkening.
The same study found that shaft diameter roughly doubled during the first year and that medullation changed with development. These results strengthen and complicate the hypothesis. They show that hair is undergoing coordinated postnatal maturation, but they also warn that apparent color is not a direct readout of melanocyte output. Thin, unmedullated fibers can scatter light differently from mature shafts. A definitive study must therefore measure eumelanin and pheomelanin per unit of shaft mass or volume while also measuring diameter, cross-sectional shape, medullation, and hair-cycle stage.
Longitudinal twin data provide a second line of evidence. Matheny and Dolan (1975) repeatedly scored hair color from three months to six years. Color changed substantially, yet monozygotic twins remained strongly concordant. Although the study preceded modern colorimetry and molecular genotyping, it makes an exclusively environmental account unlikely. The timing and extent of early hair-color change are heritable phenotypes.
3.2 Later-childhood darkening and a published-count reanalysis
Kukla-Bartoszek et al. (2018) tested 24 HIrisPlex pigmentation markers in 476 Polish children aged 6 to 13. Early hair color at ages two to three was available for a subset. Among 202 children recorded as blond at two to three years, 143, or 70.8 percent, were brown-haired by ages 6 to 13, while 59 remained blond.
The reported totals allow a limited aggregate reanalysis. The genotype model predicted later brown hair for 40 of the 143 children who darkened and for 8 of the 59 persistent blonds. The odds that the panel predicted brown were therefore approximately 2.48 times higher in darkeners than in persistent blonds (95 percent confidence interval 1.08 to 5.68; Fisher exact p approximately 0.030). This calculation uses group totals, not individual-level data, and cannot identify a causal locus. It nevertheless suggests that known pigmentation genotypes contain some information about persistence that is not fully expressed in the early-childhood phenotype.
Most darkening children were still predicted to be blond, so the panel did not capture the developmental process well. This is scientifically useful. Adult-trained pigment predictors tend to treat age as noise or a correction variable. The hypothesis instead treats change with age as the phenotype of interest.
A 2019 meeting abstract reported an even more direct analysis in 725 children of European ancestry from the Colorado Kids Sun Care Program. Hair color at ages 6 to 8 was compared with color at ages 14 to 16. Eighty-three percent of participants darkened. Eight of 32 tested variants were associated with color at one or both periods, and five were associated with color change (Tang et al., 2019). Unfortunately, the abstract did not name the five variants, provide effect sizes, or release the longitudinal data. This may be the most valuable existing dataset for the first decisive human test.
3.3 Childhood blondness is not a single genetic event
Adult hair color is highly polygenic in European-ancestry cohorts (Hysi et al., 2018). The natal-coat hypothesis does not predict a single “blondness gene.” It predicts that variants at several points in the pigment system can expose or prolong a low-output juvenile state. The most informative loci will be those for which causal function and age-dependent phenotype can both be measured.
The European KITLG enhancer
Guenther et al. (2014) functionally dissected rs12821256, located approximately 355 kb upstream of KITLG. The region acts as a hair-follicle enhancer. The derived blond-associated nucleotide weakens a conserved LEF1 binding site and reduces LEF1 responsiveness in human keratinocytes. Single-copy transgenic mice carrying the derived human enhancer produced less Kitl RNA in postnatal skin and showed lighter pigmentation than mice carrying the ancestral enhancer.
This is unusually strong evidence linking a human nucleotide to a tissue-biased regulatory element, altered transcription-factor response, altered gene expression, and visible pigmentation. It also supplies a plausible route by which a developmental signal could be attenuated. LEF1 is a transcriptional effector of WNT signaling, and WNT activity is tightly linked to hair cycling and pigment-cell activation.
The missing result is age specificity. Existing experiments show reduced enhancer output, but they do not show whether rs12821256 has a constant effect at every age or specifically weakens a juvenile-to-adult increase. A constant reduction would explain blond hair without supporting the core developmental-modifier claim. A significant genotype-by-age effect would be much more informative.
The Solomon Islander TYRP1 allele
Kenny et al. (2012) identified a recessive R93C substitution in TYRP1 that explains a large fraction of blond hair variation in Solomon Islanders. The derived allele had a frequency of approximately 26 percent, and a model including genotype, age, and sex explained 46.4 percent of spectrometrically measured hair-color variance. The allele was not a European import and acts through a different gene from the best-characterized northern-European enhancer.
The paper contains a particularly relevant age result. After adjustment for sex and geography, hair darkened significantly with age in R93/R93 homozygotes and R93/C93 heterozygotes, but not significantly in C93/C93 blond homozygotes. This pattern is consistent with the R93C genotype attenuating an age-related darkening process.
It is not yet proof of an interaction. A significant association in two genotype groups and a nonsignificant association in a third does not establish that the slopes differ. The data were cross-sectional, and severe reduction of TYRP1 function could produce a floor effect that makes further lightening or darkening difficult to detect. Reanalysis should fit a formal genotype-by-age term, preferably with a nonlinear age spline, and should then be followed longitudinally.
The broader Oceanic evidence reinforces genetic heterogeneity. The R93C allele is unevenly distributed across Island Melanesia and is nearly absent on Bougainville even though blondism occurs there (Norton et al., 2014, 2016). This implies additional light-hair alleles. A 2026 preprint identified a Denisovan-derived Alu insertion in OCA2 at high frequency on Bougainville and linked it to increased skin pigmentation and higher OCA2 expression in edited melanocytes (Kim et al., 2026). That insertion does not explain blond hair. Instead, it emphasizes that skin and hair pigmentation can be genetically dissociated, and that Oceanic pigmentation reflects several population-specific histories.
Historical research also documented fair-headedness, especially in children, among some Indigenous Australian groups (Abbie and Adey, 1953; Gates, 1960). These older studies used methods and population categories that require modern reevaluation, and no causal variant has been securely mapped. The observation should therefore be treated as a high-value target for community-led longitudinal genetics, not as established evidence for a shared Oceanic mechanism.
The key comparative point is modest but important. European, Solomon Islander, Bougainvillean, and some Indigenous Australian childhood light-hair phenotypes cannot safely be collapsed into one allele or one selective event. Their partial convergence is compatible with the idea that different pigment-reducing variants act on a broadly shared developmental substrate.
3.4 Scalp hair, body hair, and the scale of the phenotype
The original formulation emphasized that many blond children also have pale eyebrows, eyelashes, arm hair, leg hair, and fine body hair, which can darken with age (Reser, 2026a). If objectively confirmed, coordinated change across hair-bearing regions would be more consistent with a body-wide developmental state than with a scalp-specific ornament. At present, this is an undermeasured prediction rather than a well-quantified fact. Future cohorts should photograph and colorimetrically sample several body regions. Regional synchrony, or a reproducible order of darkening, could identify whether the signal is systemic, follicle-class-specific, or driven by local exposure.
4. The primate comparison
4.1 Natal coats are common and directionally diverse
Treves (1997) surveyed 138 primate species and found infant pelage contrasting with adult pelage in more than half. Across species with timing data, change began at an average of approximately 5.7 weeks and the natal coat disappeared at approximately 18 weeks. The distribution shows that age-limited coat states are not exceptional anomalies in primates.
The direction of change is crucial. Some infants are lighter than adults, such as white infant colobus monkeys and orange infant langurs that later become dark. Other infants are darker than adults, and still others differ mainly in hue or pattern. Human blondness can therefore be compared only with the subset in which infant or juvenile hair is pigment-reduced relative to adulthood. A generic similarity to any “distinctive” natal coat is not sufficient.
Caro et al. (2022) analyzed 286 primate species with phylogenetic controls. Distinctive natal coats were associated with reported infanticide and with shorter interbirth intervals, but not with allomothering or paternity confusion. These findings support a social and life-history context for natal coloration, but they do not establish that coloration itself prevents aggression. More importantly for the present hypothesis, the analysis grouped together lighter, darker, differently hued, and patterned infants. A directional reanalysis is needed to test whether specifically lighter-than-adult coats show a distinctive ecological or social distribution.
The social-signaling question should not be allowed to carry the mechanistic argument. Natal coats may elicit attention, tolerance, protection, or changes in maternal behavior, but childhood blondness could use related developmental machinery without retaining the same function. It could be a byproduct, a weak cue, a target of later sexual selection, or a neutral phenotype exposed by drift.
4.2 Limestone langurs as a natural experiment
Recent work on limestone langurs creates the strongest comparative clue. Liu et al. (2025) resequenced 48 individuals from 15 Trachypithecus species and surveyed 688 pigmentation genes. An MC1R E94D substitution was present in all sampled limestone langurs and absent from the sampled rainforest species. In HEK293T cells, the limestone-langur receptor showed higher basal cAMP signaling than rainforest-langur, rhesus macaque, and human receptors. This molecular background is consistent with the dark adult coats of several limestone species.
Yet these langurs are born light yellow-orange and darken later. The adult-promoting receptor allele is present from conception, but its dark phenotype is not expressed in the natal coat. This does not identify the developmental switch. It does show that genotype at a major melanocortin receptor cannot be read as a fixed color instruction independent of age. During infancy, another regulatory state must suppress, bypass, counterbalance, or overwhelm the receptor’s adult effect.
Nadler (2020) followed six captive-born Cat Ba langurs with serial photography during early development and supplemented these observations with wild individuals. Infants were uniformly light yellow-orange. Bare facial and extremity skin began to darken early in the second month, while dorsal pelage began darkening in the third month. Individuals varied substantially in transition speed, and adult coloration was not complete until roughly three years.
The sequence of skin darkening before pelage darkening is particularly informative. It suggests that a broader melanocyte or endocrine maturation signal may precede follicular coat replacement. It also prevents premature fixation on a single hair-specific enhancer. The transition could involve ligand-receptor balance, melanocyte availability, hair cycling, melanosome maturation, shaft structure, or coordinated changes across several tissues.
The langur evidence is still inferential because the genomic and longitudinal studies examined different samples and did not profile follicles during transition. Its value lies in the experiment it makes possible. In a lineage with a known pro-eumelanin adult background, investigators can ask which cell state keeps infant follicles light and which regulatory event releases the dark adult phenotype.
4.3 Shared genes are not enough
Pigmentation repeatedly recruits a limited toolkit, including MC1R, ASIP, KITLG, TYR, TYRP1, DCT, MITF, OCA2, and solute-carrier genes. Finding the same genes expressed in two color transitions would therefore be weak evidence of homology. Adult coat-color divergence in Sulawesi macaques, for example, involves differentiated variants in several loci, including TYR, MC1R, and ASIP (Yan et al., 2025). Similar color endpoints can evolve through different combinations of common pigment genes.
The homology claim requires a more specific match: the same orthologous regulatory element, the same cell-state transition, the same order of pathway activation, or the same stage-specific interaction among follicular cell types. Convergent use of TYR is expected. A conserved age-dependent rise in activity at the ortholog of the human KITLG hair enhancer would be much more probative.
5. A mechanistic bridge through WNT, LEF1, and KITLG
Hair pigmentation is a timed interaction among epithelial cells, melanocyte stem cells, differentiated melanocytes, dermal papilla cells, and the growing shaft. Pigment production is coupled to the anagen phase of the hair cycle rather than continuously applied to an inert fiber.
Rabbani et al. (2011) used cell-type-specific mouse genetics to show that WNT signaling is activated in epithelial and melanocyte stem cells at the onset of pigmented hair regeneration. WNT activity in melanocyte stem cells promoted differentiation, while epithelial WNT controlled follicle formation and melanocyte proliferation. This work establishes a coordinated regenerative state in which hair production and pigmentation are developmentally linked.
That result connects directly to the blond-associated human KITLG enhancer because the causal nucleotide weakens a LEF1 site. A plausible chain is therefore:
1. Age, endocrine state, or hair-cycle maturation changes WNT/LEF1 activity.
2. LEF1 changes enhancer output in follicular keratinocytes.
3. KIT ligand changes melanocyte survival, recruitment, self-renewal, or differentiation.
4. A population-specific enhancer allele changes the gain or threshold of that response.
Aoki et al. (2024) add an important timing result. In inducible mouse models, postnatal manipulation of Kitl altered melanocyte proliferation, differentiation, and stem-cell self-renewal. A single postnatal pulse produced long-lasting effects on melanocyte stem cells and pigmentation. This demonstrates that KIT ligand is not merely a static color-output factor. Transient postnatal signaling can leave durable pigmentary consequences.
The mechanistic convergence is compelling but incomplete. The mouse experiments did not model normal human childhood darkening, and the human enhancer has not been tested across juvenile and adult follicle states. The most coherent candidate mechanism is therefore also the clearest target for falsification.
6. Direct evidence, inference, and current evidential status
|Finding |Direct observation |Inference for the hypothesis |Status |
|———————————–|————————————————————————————————|————————————————————-|——————————————-|
|Prague longitudinal cohort |Hair follows a multiphasic trajectory from one month to five years; shaft structure also changes|Humans possess an intrinsic postnatal hair transition |Strong support, with structural confounding|
|Twin study |Monozygotic twins remain concordant while color changes |Timing and magnitude are heritable |Supportive but not locus-specific |
|Polish HIrisPlex cohort |Most early blonds darken; adult-oriented genotypes weakly distinguish darkeners |Some pigment genotypes may affect persistence |Suggestive |
|Solomon *TYRP1* R93C |Homozygotes are blond and show no significant cross-sectional age darkening |The allele may attenuate an age program |Strong clue; formal interaction absent |
|European *KITLG* enhancer |A causal nucleotide weakens LEF1 response and reduces follicular enhancer output |A developmental signal could be selectively dampened in hair |Strong mechanism; age specificity untested |
|Postnatal mouse *Kitl* pulse |Temporary postnatal expression has lasting pigment effects |KITLG can encode developmental timing or memory |Strong plausibility |
|Primate comparative studies |Contrasting natal coats are widespread and time-limited |Age-specific coat states are ancestral and recurrent |Strong broad context |
|Limestone langurs |Light infants become dark despite a high-basal-activity *MC1R* background |Juvenile state can override an adult pro-eumelanin genotype |Best comparative natural experiment |
|Cat Ba ontogeny |Skin darkens before pelage, with marked individual timing variation |A shared maturation signal and heritable timing can be tested|Valuable but based on a small sample |
|Shared orthologous follicle program|Not yet measured |Would establish developmental homology |Missing decisive evidence |
The present evidence supports a conservative conclusion: population-specific hypopigmenting variants can plausibly modify conserved, age-sensitive follicle-pigmentation machinery. It does not yet show that the human state and a particular primate natal coat descend from the same stage-specific regulatory program.
7. Alternatives and discriminating predictions
Several explanations can reproduce part of the phenotype. A useful hypothesis must make observations that its alternatives do not.
|Explanation |Main prediction |Result that would weaken it |
|———————————–|————————————————————————————————————-|——————————————————————————|
|Developmental-modifier model |Genotype changes slope, curvature, or transition age; stage-specific regulatory states overlap across species|Genotypes produce parallel age curves and no shared regulatory state is found |
|Constant pigment reduction |Genotype causes a stable color offset at all ages |Allele effect is confined to, or much stronger during, a juvenile state |
|Shaft maturation or replacement |Visible darkening tracks diameter, medullation, and replacement of infant hair |Melanin chemistry changes after controlling shaft structure |
|Sun bleaching |Distal hair is lighter than newly grown proximal hair; change follows season and exposure |Proximal, protected hair darkens with age under standardized exposure |
|Pubertal endocrine activation |Darkening aligns more closely with Tanner stage and hormones than chronological age |Major transition precedes puberty and tracks follicular state independently |
|Sexual selection |Allele frequency and persistence relate to mate preferences or sex-biased reproductive success |Developmental effect exists without evidence of sex-biased selection |
|Ultraviolet or vitamin D adaptation|Selected variants affect skin photoprotection and covary with ultraviolet environment |The causal effect is hair-specific and independent of skin pigmentation |
|Cold-related pleiotropy |Selected haplotypes affect thermogenesis as well as pigmentation |Fine-mapped hair enhancer effect segregates from thermogenic effects |
|Drift or founder effect |Frequencies fit demographic history without an adaptive benefit |Repeated selection signals and replicated functional age specificity are found|
These explanations are not mutually exclusive. A developmental mechanism could be correct even if an allele rose through drift. Sexual selection could favor adult retention of a juvenile-light state that originally emerged as a neutral threshold effect. A hair-specific enhancer could be selected directly while nearby variants affect metabolism. The research program should therefore separate three levels: proximal pigment mechanism, developmental history, and population-genetic cause.
8. The decisive research program
8.1 Human genotype-by-age reanalysis
The first priority is to obtain the Colorado data or another cohort with repeated, standardized hair measurements. For continuous colorimetry, an appropriate mixed model is:
[
Y_{it}=\beta_0+f(\text{age}{it})+\beta_GG_i+G_i\times f(\text{age}{it})+\boldsymbol{\gamma X}{it}+b{0i}+b_{1i}\text{age}{it}+\epsilon{it}.
]
The critical estimand is the genotype-by-age interaction. The covariate vector should include sex, ancestry principal components, season, ultraviolet exposure, hair products, pubertal stage, and, when possible, shaft properties. An intercept-only genotype effect supports constant pigment reduction. A genotype effect on slope, curvature, or transition age supports developmental modification.
The Solomon Islander data deserve the same analysis. Hair reflectance should be modeled with flexible age curves for all three TYRP1 genotypes, with an omnibus interaction test rather than separate within-genotype significance tests. A positive cross-sectional result should lead to repeat sampling of genotype-stratified individuals.
8.2 A prospective human cohort
A minimally invasive longitudinal cohort should recruit genotype-enriched participants at approximately 6 to 12 months, 2 to 3 years, 5 to 6 years, 9 to 10 years, and 14 to 16 years. Repeated sampling within individuals is preferable. Candidate loci should include rs12821256 near KITLG, TYRP1 R93C, MC1R, ASIP, SLC45A2, SLC24A4, HERC2/OCA2, IRF4, and any variants identified in the Colorado dataset.
At each visit, investigators should collect:
● standardized multispectral photographs with a color target;
● proximal and distal segments of newly cut hair;
● shaft diameter, cross-sectional shape, medullation, and hair-cycle state;
● chemical markers of eumelanin and pheomelanin normalized to shaft mass and volume;
● plucked anagen follicles for RNA or chromatin profiling when ethically appropriate;
● comparable measures from eyebrows and selected body-hair regions;
● season, ultraviolet exposure, nutrition, hair products, Tanner stage, and endocrine covariates.
This design can distinguish new pigment production from sun bleaching and optical maturation. It can also determine whether darkening occurs within successively produced anagen shafts or mainly when one hair class is replaced by another.
8.3 Directional primate phylogenetics
The Caro et al. dataset should be recoded along separate axes: infant lighter, darker, or similar in luminance to adult; hue change; pattern change; body region; onset, midpoint, and completion of transition; and timing relative to weaning and locomotor independence. Phylogenetic models should include infanticide rate, allomothering, ventral carriage, predation, habitat, adult coat luminance, sexual dichromatism, body size, and life-history pace.
The essential question is not whether “distinctive” coats have a social correlate. It is whether pigment-reduced infant coats repeatedly evolve in comparable developmental or social settings. Leave-one-clade-out analyses are necessary to ensure that any result is not driven entirely by colobines or another highly represented lineage.
8.4 Longitudinal langur phenotyping and pedigrees
Accredited conservation centers maintaining Cat Ba, Delacour’s, Hatinh, or François’ langurs may already possess years of serial photographs and studbook pedigrees. Calibrated analysis of these archives could estimate transition curves by body region and test whether timing is heritable. Animal models could separate additive genetic effects from maternal, nutritional, enclosure, sex, species, and cohort effects.
Naturally shed hair and samples collected during routine veterinary care could be compared across infant, transition, and adult stages. The initial assays need not be molecularly extravagant. Shaft colorimetry, melanin chemistry, diameter, and medullation would show whether the visible transition primarily reflects pigment synthesis or hair structure. Skin and pelage should be analyzed separately because their timing differs.
8.5 Orthologous enhancer assays
The most focused molecular experiment would compare the human rs12821256 enhancer with orthologous sequences from chimpanzees, macaques, Trachypithecus, gibbons, and other informative primates. Constructs should be inserted into the same genomic landing site and tested under graded WNT/LEF1 stimulation, juvenile-like and adult-like hormonal conditions, and keratinocyte-melanocyte co-culture.
The informative result would not merely be that the enhancer is active. The proposed smoking gun is a conserved nonlinear response to developmental state, combined with a human blond allele that reduces or delays that response. Editing the blond allele back to the ancestral nucleotide should restore the adult-like rise. Conversely, inserting the blond nucleotide should prolong the juvenile-like output.
8.6 Age-stratified single-cell multi-omics
Matched anagen follicles from infant, transitional, and adult langurs should be profiled with single-nucleus RNA sequencing and ATAC sequencing. If ethical human samples become available, comparable juvenile and adult follicles should be processed using the same platform. Cell types of interest include epithelial stem cells, matrix keratinocytes, dermal papilla cells, melanocyte stem cells, and pigment-producing melanocytes.
Candidate modules include WNT/LEF1, KITLG/KIT, MC1R/ASIP/POMC, MITF/TYR/TYRP1/DCT, EDN3/EDNRB, BMP signaling, melanosome transport, and shaft keratins. The analysis must match hair-cycle stage and homologous cell types before comparing age. A generic rise in melanogenesis genes would be expected in any darkening process. Evidence for homology requires a more distinctive shared regulatory sequence and temporal order.
9. What would count as a smoking gun?
The hypothesis should be considered strongly supported only if three observations converge:
1. The same orthologous enhancer or regulatory module changes activity during both human childhood darkening and a nonhuman-primate natal-coat transition.
2. A causal human blondness allele changes the slope, timing, or amplitude of that developmental response in a formal genotype-by-age analysis.
3. Editing or swapping the allele changes developmental pigment output in a controlled follicle model.
A particularly strong result would involve the KITLG hair enhancer. If the enhancer becomes more accessible or active during both transitions, the European blond allele weakens the stage-specific rise, and correction restores the response, the case for shared developmental architecture would be difficult to dismiss.
Several results would force revision. If blond-associated alleles produce parallel pigment offsets at all ages, childhood blondness would be better described as constant genetic hypopigmentation superimposed on a separate age program. If visible darkening disappears after adjustment for shaft diameter and hair replacement, the central mechanism may concern hair-class maturation rather than melanocyte regulation. If humans and langurs use different stage-specific enhancers, the visual similarity would be convergence rather than developmental homology.
10. Evolutionary implications
If supported, the natal-coat hypothesis would change how childhood blondness is interpreted. The phenotype would not be a wholly new human invention. It would be a population-specific exposure of a much older property of mammalian integument: the ability to produce different pigmentary states at different life stages.
The model also offers a heterochronic interpretation of adult blondness. In some individuals, a pigment-reducing genotype may allow the juvenile-light state to persist into adulthood. This resembles paedomorphosis at the phenotypic level, but the term should remain provisional until timing is measured directly. Adult blondness could also result from a stable low-output state unrelated to delayed transition.
The original proposal emphasized possible social signaling (Reser, 2026a), consistent with a broader account of human hair as a visible life-stage interface (Reser, 2026b). A light juvenile phenotype could make age and dependency more legible. Yet no evidence currently shows that blond hair altered caregiving, aggression, or survival in ancestral humans. Social function is therefore an optional evolutionary extension, not a premise of the developmental model.
Population differences may also have different causes. The European KITLG enhancer has a demonstrated hair-follicle effect, which makes a purely skin-based vitamin D account inadequate for that nucleotide. However, selection around the wider KITLG region may involve linked or pleiotropic effects, including thermogenesis (Yang et al., 2018). The Solomon Islander TYRP1 allele may reflect local selection, drift, or both in an island population. Bougainvillean blondism appears to require other alleles. The same developmental architecture can be modified independently and spread for different reasons.
11. Limitations and ethical considerations
The current literature was not designed to test this hypothesis. Human studies often use broad categorical hair colors, retrospective childhood reports, cross-sectional age comparisons, or adult-trained prediction models. Environmental exposure and shaft structure are inconsistently measured. The most direct pediatric genotype dataset remains available only as a conference abstract.
Comparative studies face different limitations. “Distinctive natal coat” is a heterogeneous category. Longitudinal primate samples are small, molecular tissue is rare, and hair-cycle stage can confound age comparisons. Even a shared molecular pathway would not by itself demonstrate common selective function.
Research involving children, Indigenous communities, and endangered primates requires unusually careful governance. Indigenous Australian and Oceanic projects should be community-led, with local control of biological samples, genomic data, interpretation, and publication. Older anthropological descriptions should not be treated as substitutes for contemporary consent or self-identification. Primate work should prioritize archived images, naturally shed hair, and samples already collected for clinical care. The central experiments do not justify harmful or invasive sampling.
12. Conclusion
The natal-coat hypothesis began with a visual and developmental analogy: blond children often become dark-haired adults, and many primates pass from light infant coats to darker adult coats (Reser, 2026a). Subsequent evidence makes the comparison more precise. Human hair color follows a heritable postnatal trajectory. Different populations reach light hair through different alleles. A causal European KITLG enhancer variant weakens LEF1-dependent follicular signaling. A Solomon Islander TYRP1 allele may attenuate age-related darkening. Light infant langurs can override a pro-eumelanin adult MC1R background, and postnatal KIT ligand signaling can durably alter pigment-cell behavior.
Together, these findings support a testable model, not a completed proof. Population-specific variants may act as dimmer switches on an ancestral age-sensitive follicle system, allowing a juvenile low-pigment state to become conspicuous or persist. The decisive evidence must connect human genotype to developmental trajectory and then connect that trajectory to the same orthologous regulatory state in another primate.
This is now a tractable research program. Existing human data can test genotype-by-age effects. Existing primate photographs can test the direction and timing of natal-coat transitions. Focused enhancer assays and age-stratified follicle profiling can determine whether the resemblance reflects deep developmental homology or convergent use of the pigment toolkit. Either outcome would be informative. The hypothesis succeeds scientifically by making the difference testable.
Data availability statement
No new participant-level dataset was generated for this hypothesis article. The odds ratio reported for the Polish childhood cohort is an aggregate calculation from counts published by Kukla-Bartoszek et al. (2018) and should not be interpreted as a substitute for individual-level reanalysis.
Conflict of interest statement
To be completed by the author before submission.
References
Abbie, A. A., and Adey, W. R. (1953). Pigmentation in a Central Australian tribe with special reference to fair-headedness. American Journal of Physical Anthropology, 11, 339-360.
Amyere, M., et al. (2011). KITLG mutations cause familial progressive hyper- and hypopigmentation. Journal of Investigative Dermatology, 131(6), 1234-1239. https://doi.org/10.1038/jid.2011.29
Aoki, H., Tomita, H., Hara, A., and Kunisada, T. (2024). Postnatal expression of Kitl affects pigmentation of the epidermis. Journal of Investigative Dermatology, 144(1), 96-105.e2. https://doi.org/10.1016/j.jid.2023.06.200
Caro, T., et al. (2022). On the evolution of distinctive natal coat coloration in primates. American Journal of Biological Anthropology, 177(3), 530-539. https://doi.org/10.1002/ajpa.24468
Gates, R. R. (1960). The genetics of the Australian aborigines. Acta Geneticae Medicae et Gemellologiae, 9(1), 7-50. https://doi.org/10.1017/S1120962300018424
Guenther, C. A., Tasic, B., Luo, L., Bedell, M. A., and Kingsley, D. M. (2014). A molecular basis for classic blond hair color in Europeans. Nature Genetics, 46, 748-752. https://doi.org/10.1038/ng.2991
Hysi, P. G., et al. (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. https://doi.org/10.1038/s41588-018-0100-5
Kenny, E. E., et al. (2012). Melanesian blond hair is caused by an amino acid change in TYRP1. Science, 336(6081), 554. https://doi.org/10.1126/science.1217849
Kim, K., et al. (2026). A Denisovan-derived Alu insertion in OCA2 contributes to pigmentation diversity in present-day Melanesians. bioRxiv [preprint]. https://doi.org/10.64898/2026.03.18.712481
Kukla-Bartoszek, M., et al. (2018). Investigating the impact of age-dependent hair colour darkening during childhood on DNA-based hair colour prediction with the HIrisPlex system. Forensic Science International: Genetics, 36, 26-33. https://pubmed.ncbi.nlm.nih.gov/29913343/
Liu, Z., et al. (2025). Living on the rocks: Genomic analysis of limestone langurs provides novel insights into adaptive evolution in extreme karst environments. Genomics, Proteomics & Bioinformatics, 23(1), qzaf007. https://doi.org/10.1093/gpbjnl/qzaf007
Matheny, A. P., Jr., and Dolan, A. B. (1975). Sex and genetic differences in hair color changes during early childhood. American Journal of Physical Anthropology, 42(1), 53-56. https://doi.org/10.1002/ajpa.1330420106
Nadler, T. (2020). The development of pelage coloration in Cat Ba langurs (Trachypithecus poliocephalus). Vietnamese Journal of Primatology, 3(2), 23-37. Full text
Norton, H. L., Friedlaender, J. S., Merriwether, D. A., Koki, G., Mgone, C. S., and Shriver, M. D. (2006). Skin and hair pigmentation variation in Island Melanesia. American Journal of Physical Anthropology, 130(2), 254-268. https://doi.org/10.1002/ajpa.20343
Norton, H. L., Correa, E. A., Koki, G., and Friedlaender, J. S. (2014). Distribution of an allele associated with blond hair color across Northern Island Melanesia. American Journal of Physical Anthropology, 153(4), 653-662. https://doi.org/10.1002/ajpa.22466
Norton, H. L., Hanna, M., Werren, E., and Friedlaender, J. S. (2016). The rs387907171 SNP in TYRP1 is not associated with blond hair color on the Island of Bougainville. American Journal of Human Biology, 28(3), 431-435. https://doi.org/10.1002/ajhb.22795
Prokopec, M., Glosova, L., and Ubelaker, D. H. (2000). Change in hair pigmentation in children from birth to 5 years in a Central European population (longitudinal study). Forensic Science Communications, 2(3). Official record
Rabbani, P., et al. (2011). Coordinated activation of Wnt in epithelial and melanocyte stem cells initiates pigmented hair regeneration. Cell, 145(6), 941-955. https://doi.org/10.1016/j.cell.2011.05.004
Reser, J. E. (2026a, May 11). Childhood blondness compared to primate natal coats: A conserved developmental pigmentation hypothesis. Observed Impulse. https://www.observedimpulse.com/2026/05/childhood-blondness-compared-to-primate.html
Reser, J. E. (2026b, January 20). An evolutionary explanation for hair loss and graying: Signaling de-escalation and reduced challenge. Observed Impulse. https://www.observedimpulse.com/2026/01/an-evolutionary-explanation-for-hair.html
Roos, C., et al. (2020). Mitogenomic phylogeny of the Asian colobine genus Trachypithecus with special focus on T. phayrei. Zoological Research, 41(6), 656-669. https://pmc.ncbi.nlm.nih.gov/articles/PMC7671912/
Tang, H., et al. (2019). 864 Genetics of pigmentation changes in a pediatric population. Journal of Investigative Dermatology, 139(5), S149. https://doi.org/10.1016/j.jid.2019.03.940
Treves, A. (1997). Primate natal coats: A preliminary analysis of distribution and function. American Journal of Physical Anthropology, 104(1), 47-70. https://pubmed.ncbi.nlm.nih.gov/9331453/
Yan, X., et al. (2025). Exome analysis reveals species divergence in TYR and identifies species genetic markers in five endemic Macaca species on Sulawesi Island. BMC Ecology and Evolution, 25, 66. https://doi.org/10.1186/s12862-025-02407-6
Yang, Z., et al. (2018). Darwinian positive selection on the pleiotropic effects of KITLG explain skin pigmentation and winter temperature adaptation in Eurasians. Molecular Biology and Evolution, 35(9), 2272-2283. https://doi.org/10.1093/molbev/msy136
Leave a Reply