The genomic architecture of continuous plumage colour variation in the European barn owl (Tyto alba).

The maintenance of colour variation in wild populations has long fascinated evolutionary biologists, although most studies have focused on discrete traits exhibiting rather simple inheritance patterns and genetic architectures. However, the study of continuous colour traits and their potentially oligo- or polygenic genetic bases remains rare in wild populations. We studied the genetics of the continuously varying white-to-rufous plumage coloration of the European barn owl (Tyto alba) using a genome-wide association approach on the whole-genome data of 75 individuals. We confirmed a mutation at the melanocortin-1-receptor gene (MC1R) is involved in the coloration and identified two new regions, located in super-scaffolds 9 and 42. The combination of the three regions explains most of the colour variation (80.37%, 95% credible interval 58.45-100%). One discovered region, located in the sex chromosome, differs between the most extreme colorations in owls sharing a specific MC1R genotype. This region may play a role in the colour sex dimorphism of this species, possibly in interaction with the autosomal MC1R. We thus provide insights into the genetic architecture of continuous colour variation, pointing to an oligogenic basis with potential epistatic effects among loci that should aid future studies understanding how continuous colour variation is maintained in nature.

Landscape and Climatic Variations Shaped Secondary Contacts amid Barn Owls of the Western Palearctic.

The combined actions of climatic variations and landscape barriers shape the history of natural populations. When organisms follow their shifting niches, obstacles in the landscape can lead to the splitting of populations, on which evolution will then act independently. When two such populations are reunited, secondary contact occurs in a broad range of admixture patterns, from narrow hybrid zones to the complete dissolution of lineages. A previous study suggested that barn owls colonized the Western Palearctic after the last glaciation in a ring-like fashion around the Mediterranean Sea, and conjectured an admixture zone in the Balkans. Here, we take advantage of whole-genome sequences of 94 individuals across the Western Palearctic to reveal the complex history of the species in the region using observational and modeling approaches. Even though our results confirm that two distinct lineages colonized the region, one in Europe and one in the Levant, they suggest that it predates the last glaciation and identify a secondary contact zone between the two in Anatolia. We also show that barn owls recolonized Europe after the glaciation from two distinct glacial refugia: a previously identified western one in Iberia and a new eastern one in Italy. Both glacial lineages now communicate via eastern Europe, in a wide and permeable contact zone. This complex history of populations enlightens the taxonomy of Tyto alba in the region, highlights the key role played by mountain ranges and large water bodies as barriers and illustrates the power of population genomics in uncovering intricate demographic patterns.

Binocular field configuration in owls: the role of foraging ecology.

The binocular field of vision differs widely in birds depending on ecological traits such as foraging. Owls (Strigiformes) have been considered to have a unique binocular field, but whether it is related to foraging has remained unknown. While taking into account allometry and phylogeny, we hypothesized that both daily activity cycle and diet determine the size and shape of the binocular field in owls. Here, we compared the binocular field configuration of 23 species of owls. While we found no effect of allometry and phylogeny, ecological traits strongly influence the binocular field shape and size. Binocular field shape of owls significantly differed from that of diurnal raptors. Among owls, binocular field shape was relatively conserved, but binocular field size differed among species depending on ecological traits, with larger binocular fields in species living in dense habitat and foraging on invertebrates. Our results suggest that (i) binocular field shape is associated with the time of foraging in the daily cycle (owls versus diurnal raptors) and (ii) that binocular field size differs between closely related owl species even though the general shape is conserved, possibly because the field of view is partially restricted by feathers, in a trade-off with auditory localization.

Density-dependent selection and the maintenance of colour polymorphism in barn owls.

The capacity of natural selection to generate adaptive changes is (according to the fundamental theorem of natural selection) proportional to the additive genetic variance in fitness. In spite of its importance for development of new adaptations to a changing environment, processes affecting the magnitude of the genetic variance in fitness-related traits are poorly understood. Here, we show that the red-white colour polymorphism in female barn owls is subject to density-dependent selection at the phenotypic and genotypic level. The diallelic melanocortin-1 receptor gene explained a large amount of the phenotypic variance in reddish coloration in the females ([Formula: see text]). Red individuals (RR genotype) were selected for at low densities, while white individuals (WW genotype) were favoured at high densities and were less sensitive to changes in density. We show that this density-dependent selection favours white individuals and predicts fixation of the white allele in this population at longer time scales without immigration or other selective forces. Still, fluctuating population density will cause selection to fluctuate and periodically favour red individuals. These results suggest how balancing selection caused by fluctuations in population density can be a general mechanism affecting the level of additive genetic variance in natural populations.

Unexpected post-glacial colonisation route explains the white colour of barn owls (Tyto alba) from the British Isles.

The climate fluctuations of the Quaternary shaped the movement of species in and out of glacial refugia. In Europe, the majority of species followed one of the described traditional postglacial recolonization routes from the southern peninsulas towards the north. Like most organisms, barn owls are assumed to have colonized the British Isles by crossing over Doggerland, a land bridge that connected Britain to northern Europe. However, while they are dark rufous in northern Europe, barn owls in the British Isles are conspicuously white, a contrast that could suggest selective forces are at play on the islands. Yet, our analysis of known candidate genes involved in coloration found no signature of selection. Instead, using whole genome sequences and species distribution modelling, we found that owls colonised the British Isles soon after the last glaciation, directly from a white coloured refugium in the Iberian Peninsula, before colonising northern Europe. They would have followed a hitherto unknown post-glacial colonization route to the Isles over a westwards path of suitable habitat in now submerged land in the Bay of Biscay, thus not crossing Doggerland. As such, they inherited the white colour of their Iberian founders and maintained it through low gene flow with the mainland that prevents the import of rufous alleles. Thus, we contend that neutral processes probably explain this contrasting white colour compared to continental owls. With the barn owl being a top predator, we expect future research will show this unanticipated route was used by other species from its paleo community.

Genomic consequences of colonisation, migration and genetic drift in barn owl insular populations of the eastern Mediterranean.

The study of insular populations was key in the development of evolutionary theory. The successful colonisation of an island depends on the geographic context, and specific characteristics of the organism and the island, but also on stochastic processes. As a result, apparently identical islands may harbour populations with contrasting histories. Here, we use whole genome sequences of 65 barn owls to investigate the patterns of inbreeding and genetic diversity of insular populations in the eastern Mediterranean Sea. We focus on Crete and Cyprus, islands with similar size, climate and distance to mainland, that provide natural replicates for a comparative analysis of the impacts of microevolutionary processes on isolated populations. We show that barn owl populations from each island have a separate origin, Crete being genetically more similar to other Greek islands and mainland Greece, and Cyprus more similar to the Levant. Further, our data show that their respective demographic histories following colonisation were also distinct. On the one hand, Crete harbours a small population and maintains very low levels of gene flow with neighbouring populations. This has resulted in low genetic diversity, strong genetic drift, increased relatedness in the population and remote inbreeding. Cyprus, on the other hand, appears to maintain enough gene flow with the mainland to avoid such an outcome. Our study provides a comparative population genomic analysis of the effects of neutral processes on a classical island-mainland model system. It provides empirical evidence for the role of stochastic processes in determining the fate of diverging isolated populations.

Genomic basis of insularity and ecological divergence in barn owls (Tyto alba) of the Canary Islands.

Islands, and the particular organisms that populate them, have long fascinated biologists. Due to their isolation, islands offer unique opportunities to study the effect of neutral and adaptive mechanisms in determining genomic and phenotypical divergence. In the Canary Islands, an archipelago rich in endemics, the barn owl (Tyto alba), present in all the islands, is thought to have diverged into a subspecies (T. a. gracilirostris) on the eastern ones, Fuerteventura and Lanzarote. Taking advantage of 40 whole-genomes and modern population genomics tools, we provide the first look at the origin and genetic makeup of barn owls of this archipelago. We show that the Canaries hold diverse, long-standing and monophyletic populations with a neat distinction of gene pools from the different islands. Using a new method, less sensitive to structure than classical FST, to detect regions involved in local adaptation to insular environments, we identified a haplotype-like region likely under selection in all Canaries individuals and genes in this region suggest morphological adaptations to insularity. In the eastern islands, where the subspecies is present, genomic traces of selection pinpoint signs of adapted body proportions and blood pressure, consistent with the smaller size of this population living in a hot arid climate. In turn, genomic regions under selection in the western barn owls from Tenerife showed an enrichment in genes linked to hypoxia, a potential response to inhabiting a small island with a marked altitudinal gradient. Our results illustrate the interplay of neutral and adaptive forces in shaping divergence and early onset speciation.

Elder Barn Owl Nestlings Flexibly Redistribute Parental Food according to Siblings' Need or in Return for Allopreening.

Kin selection and reciprocation of biological services are distinct theories invoked to explain the origin and evolutionary maintenance of altruistic and cooperative behaviors. Although these behaviors are not considered to be mutually exclusive, the cost-benefit balance of behaving altruistically or cooperating reciprocally and the conditions promoting a switch between such different strategies have rarely been tested. Here, we examine the association between allofeeding, allopreening, and vocal solicitations in wild barn owl (Tyto alba) broods under different food abundance conditions: natural food provisioning and after an experimental food supplementation. Allofeeding was performed mainly by elder nestlings (hatching is asynchronous) in prime condition, especially when the cost of forgoing a prey was small (when parents allocated more prey to the food donor and after food supplementation). Nestlings preferentially shared food with the siblings that emitted very intense calls, thus potentially increasing indirect fitness benefits, or with the siblings that provided extensive allopreening to the donor, thus possibly promoting direct benefits from reciprocation. Finally, allopreening was mainly directed toward older siblings, perhaps to maximize the probability of being fed in return. Helping behavior among relatives can therefore be driven by both kin selection and direct cooperation, although it is dependent on the contingent environmental conditions.

Glucocorticoid levels are linked to lifetime reproductive success and survival of adult barn owls.

Glucocorticoid hormones, such as corticosterone, are crucial in regulating daily life metabolism and energy expenditure, as well as promoting short-term physiological and behavioural responses to unpredictable environmental challenges. Therefore, glucocorticoids are considered to mediate trade-offs between survival and reproduction. Relatively little is known about how selection has shaped glucocorticoid levels. We used 15 years of capture-recapture and dead recovery data combined with 13 years of corticosterone and breeding success data taken on breeding barn owls (Tyto alba) to investigate such trade-offs. We found that survival was positively correlated with stress-induced corticosterone levels in both sexes, whereas annual and lifetime reproductive success (i.e. the sum of young successfully fledged during the entire reproductive career) was positively correlated with both baseline and stress-induced corticosterone levels in females only. Our results suggest that, in the barn owl, the stress-induced corticosterone response is a good proxy for adult survival and lifetime reproductive success. However, selection pressure appears to act differently on corticosterone levels of males and females.

C-terminal tails mimicking bioactive intermediates cause different plasma degradation patterns and kinetics in neuropeptides γ-MSH, α-MSH, and neurotensin.

Peptides are attractive drugs because of their specificity and minimal off-target effects. Short half-lives are within their major drawbacks, limiting actual use in clinics. The golden standard in therapeutic peptide development implies identification of a minimal core sequence, then modified to increase stability through several strategies, including the introduction of nonnatural amino acids, cyclization, and lipidation. Here, we investigated plasma degradations of hormone sequences all composed of a minimal active core peptide and a C-terminal extension. We first investigated pro-opimelanocortin (POMC) γ2/γ3-MSH hormone behavior and extended our analysis to POMC-derived α-melanocyte stimulating hormone/adrenocorticotropic hormone signaling neuropeptides and neurotensin. We demonstrated that in all the three cases analyzed in this study, few additional residues mimicking the natural sequence alter both peptide stability and the mechanism(s) of degradation of the minimal conserved functional pattern. Our results suggest that the impact of extensions on the bioactivity of a peptide drug has to be carefully evaluated throughout the optimization process.

Lack of statistical power as a major limitation in understanding MHC-mediated immunocompetence in wild vertebrate populations.

Disentangling the sources of variation in developing an effective immune response against pathogens is of major interest to immunoecology and evolutionary biology. To date, the link between immunocompetence and genetic variation at the major histocompatibility complex (MHC) has received little attention in wild animals, despite the key role of MHC genes in activating the adaptive immune system. Although several studies point to a link between MHC and immunocompetence, negative findings have also been reported. Such disparate findings suggest that limited statistical power might be affecting studies on this topic, owing to insufficient sample sizes and/or a generally small effect of MHC on the immunocompetence of wild vertebrates. To clarify this issue, we investigated the link between MHC variation and seven immunocompetence proxies in a large sample of barn owls and estimated the effect sizes and statistical power of this and published studies on this topic. We found that MHC poorly explained variation in immunocompetence of barn owls, with small-to-moderate associations between MHC and immunocompetence in owls (effect size: .1 ≥ r ≤ .3) similar to other vertebrates studied to date. Such small-to-moderate effects were largely associated with insufficient power, which was only sufficient (>0.8) to detect moderate-to-large effect sizes (r ≥ .3). Thus, studies linking MHC variation with immunocompetence in wild populations are underpowered to detect MHC effects, which are likely to be of generally small magnitude. Larger sample sizes (>200) will be required to achieve sufficient power in future studies aiming to robustly test for a link between MHC variation and immunocompetence.

Female-biased dispersal and non-random gene flow of MC1R variants do not result in a migration load in barn owls.

Non-random gene flow is a widely neglected force in evolution and ecology. This genotype-dependent dispersal is difficult to assess, yet can impact the genetic variation of natural populations and their fitness. In this work, we demonstrate a high immigration rate of barn owls (Tyto alba) inside a Swiss population surveyed during 15 years. Using ten microsatellite loci as an indirect method to characterize dispersal, two-third of the genetic tests failed to detect a female-biased dispersal, and Monte Carlo simulations confirmed a low statistical power to detect sex-biased dispersal in case of high dispersal rate of both sexes. The capture-recapture data revealed a female-biased dispersal associated with an excess of heterozygote for the melanocortin-1 receptor gene (MC1R), which is responsible for their ventral rufous coloration. Thus, female homozygotes for the MC1RWHITE allele might be negatively selected during dispersal. Despite the higher immigration of females that are heterozygote at MC1R, non-random gene flow should not lead to a migration load regarding this gene because we did not detect an effect of MC1R on survival and reproductive success in our local population. The present study highlights the usefulness of using multiple methods to correctly decrypt dispersal and gene flow. Moreover, despite theoretical expectations, we show that non-random dispersal of particular genotypes does not necessarily lead to migration load in recipient populations.

Baseline and stress-induced corticosterone levels are heritable and genetically correlated in a barn owl population.

The hypothalamic-pituitary-adrenal (HPA) axis is responsible for the regulation of corticosterone, a hormone that is essential in the mediation of energy allocation and physiological stress. As a continuous source of challenge and stress for organisms, the environment has promoted the evolution of physiological adaptations and led to a great variation in corticosterone profiles within or among individuals, populations and species. In order to evolve via natural selection, corticosterone levels do not only depend on the strength of selection exerted on them, but also on the extent to which the regulation of corticosterone is heritable. Nevertheless, the heritability of corticosterone profiles in wild populations is still poorly understood. In this study, we estimated the heritability of baseline and stress-induced corticosterone levels in barn owl (Tyto alba) nestlings from 8 years of data, using a multivariate animal model based on a behavioural pedigree. We found that baseline and stress-induced corticosterone levels are strongly genetically correlated (r = 0.68-0.80) and that the heritability of stress-induced corticosterone levels (h2 = 0.24-0.33) was moderate and similar to the heritability of baseline corticosterone levels (h2 = 0.19-0.30). These findings suggest that the regulation of stress-induced corticosterone and baseline levels evolves at a similar pace when selection acts with the same intensity on both traits and that contrary to previous studies, the evolution of baseline and stress-induced level is interdependent in barn owls, as they may be strongly genetically correlated.

Expression of glucocorticoid and mineralocorticoid receptor genes co-varies with a stress-related colour signal in barn owls.

Glucocorticoid hormones are important intermediates between an organism and its environment. They enable an organism to adjust its behavioural and physiological processes in response to environmental changes by binding to mineralocorticoid receptors (MR) and glucocorticoid receptors (GR) expressed in many tissues, including the integument. The regulation of glucocorticoids co-varies with melanin-based colouration in numerous species, an association that might result from pleiotropic effects of genes in the melanocortin system and evolve within a signalling context. Most studies have focused on the circulating levels of glucocorticoids disregarding the receptors that mediate their action, and that might partly account for the covariation between the regulation of stress and melanin-based colouration. We investigated the association of the expression levels of GR and MR genes with melanin-based colouration in the growing feathers of nestling barn owls (Tyto alba). We also explored the association between GR and MR expression levels and the expression of genes related to the melanocortin system and melanogenesis to better understand the origin of the link between the expression of receptors to which corticosterone binds and melanin-based colouration. Nestling barn owls displaying larger eumelanic black feather spots expressed GR and MR at lower levels than smaller-spotted individuals. However, we found that the expression of the GR and MR genes was positively rather than negatively correlated with the expression of genes involved in the deposition of melanin pigments at the time we sampled the nestlings. This provides mixed evidence of the association between melanin-based traits and MR and GR gene expression. The finding that the expression of GR and MR was positively associated with the expression of the PCSK2 gene (encoding one of the protein convertase responsible for the production of hormone peptide ACTH and α-MSH) suggests that the melanocortin system may be implicated in the establishment of the covariation between melanin-based colour and the expression of receptors to which glucocorticoids bind. However, further studies investigating the expression of melanin-based traits with stress-related endpoints at different time points of feather development will be necessary to understand better the proximate mechanism linking melanin-based traits with stress.

Toward Understanding the Repeated Occurrence of Associations between Melanin-Based Coloration and Multiple Phenotypes.

Melanin is the most widespread pigment in organisms. Melanin-based coloration has been repeatedly observed to be associated with the same traits and in the same direction in different vertebrate and insect species. However, whether any factors that are common to different taxa account for the repeated evolution of melanin-phenotype associations remains unclear. We propose to approach this question from the perspective of convergent and parallel evolution to clarify to what extent different species have evolved the same associations owing to a shared genetic basis and being subjected to similar selective pressures. Our current understanding of the genetic basis of melanin-phenotype associations allows for both convergent and parallel evolution, but this understanding is still limited. Further research is needed to clarify the generality and interdependencies of the different proposed mechanisms (supergenes, pleiotropy based on hormones, or neural crest cells). The general ecological scenarios whereby melanin-based coloration is under selection-protection from ultraviolet radiation, thermoregulation in cold environments, or as a signal of social status-offer a good opportunity to study how melanin-phenotype associations evolve. Reviewing these scenarios shows that some traits associated with melanin-based coloration might be selected together with coloration by also favoring adaptation but that other associated traits might impede adaptation, which may be indicative of genetic constraints. We therefore encourage further research on the relative roles that selection and genetic constraints play in shaping multiple melanin-phenotype associations. Placed into a phylogenetic context, this will help clarify to what extent these associations result from convergent or parallel evolutionary processes and why melanin-phenotype associations are so common across the tree of life.

Comprehensive molecular phylogeny of barn owls and relatives (Family: Tytonidae), and their six major Pleistocene radiations.

The owl family Tytonidae comprises two genera: Phodilus, limited to the forests of central Africa and South-East Asia, and the ubiquitous Tyto. The genus Tyto is majorly represented by the cosmopolitan Common Barn Owl group, with more than 30 subspecies worldwide. Discrete differences in body size and plumage colouration have led to the classification of this family into many species and subspecies, but the taxonomic status and phylogenetic relationships between taxa remain unclear, and in some groups controversial. Although several previous studies attempted to resolve this problem, they have been limited in their taxonomic and geographical coverage, or have relied on restricted molecular evidence and low sample sizes. Based on the most comprehensive sampling to date (16 out of 17 Tyto species, and one out of three Phodilus species), a multi-locus approach using seven mitochondrial and two nuclear markers, and taking advantage of field data and museum collections available worldwide, our main questions in this study were: (1) what are the phylogenetic relationships and classification status of the whole family; (2) when and where did the most important speciation events occur? We confirm that the Common Barn Owl, Tyto alba is divided into three main evolutionary units: the American Barn Owl, T. furcata; the Western Barn Owl, T. alba; and the Eastern Barn Owl, T. javanica, and suggest a Late Miocene (ca. 6 mya) Australasian and African origin of the group. Our results are supported by fossil age information, given that the most recent common ancestor between the Tytonidae genera Phodilus and Tyto was probably from the Oligocene (ca. 28 mya) of Australasia. We finally reveal six major Pleistocene radiations of Tyto, all resulting in wide-range distributions.

A melanin-based trait is more strongly related to body size in the tropics than in temperate regions in the globally distributed barn owl family.

Life history traits differ between organisms living in the tropics, Northern and Southern Hemispheres, and sexual selection is thought to be stronger close to the equator than in temperate regions. Although birds are often supposed to be more brightly coloured in the tropics, the current evidence of geographic variation in the intensity of sexual selection and sex-specific natural selection is equivocal. Whether sex-specific traits signal aspects of individual quality better in the tropics than in the temperate regions of the Northern and Southern Hemispheres therefore remains an open question. We examined predictions of this hypothesis in the Tytonidae family (barn owls and their relatives) because females, on average, display larger black spots on the tip of their ventral body feathers than males, and this trait is associated with aspects of individual quality. We measured the size of melanic spots and the wing length of 7893 Tytonidae skins collected worldwide and preserved in natural history museums. The covariation between spot size and wing length was stronger in females than in males, in large- than small-spotted Tyto taxa and close to the equator than in temperate regions. This suggests that selection for spot size, which can be used by owls as an additional cue to assess individual body size and other aspects of phenotypic quality, is stronger in females than in males, particularly near the equator.

Lack of evidence for selection favouring MHC haplotypes that combine high functional diversity.

High rates of gene duplication and the highest levels of functional allelic diversity in vertebrate genomes are the main hallmarks of the major histocompatibility complex (MHC), a multigene family with a primordial role in pathogen recognition. The usual tight linkage among MHC gene duplicates may provide an opportunity for the evolution of haplotypes that associate functionally divergent alleles and thus grant the transmission of optimal levels of diversity to coming generations. Even though such associations may be a crucial component of disease resistance, this hypothesis has been given little attention in wild populations. Here, we leveraged pedigree data from a barn owl (Tyto alba) population to characterize MHC haplotype structure across two MHC class I (MHC-I) and two MHC class IIB (MHC-IIB) duplicates, in order to test the hypothesis that haplotypes' genetic diversity is higher than expected from randomly associated alleles. After showing that MHC loci are tightly linked within classes, we found limited evidence for shifts towards MHC haplotypes combining high diversity. Neither amino acid nor functional within-haplotype diversity were significantly higher than in random sets of haplotypes, regardless of MHC class. Our results therefore provide no evidence for selection towards high-diversity MHC haplotypes in barn owls. Rather, high rates of concerted evolution may constrain the evolution of high-diversity haplotypes at MHC-I, while, in contrast, for MHC-IIB, fixed differences among loci may provide barn owls with already optimized functional diversity. This suggests that at the MHC-I and MHC-IIB respectively, different evolutionary dynamics may govern the evolution of within-haplotype diversity.

MC1R variants affect the expression of melanocortin and melanogenic genes and the association between melanocortin genes and coloration.

The melanocortin-1 receptor (MC1R) gene influences coloration by altering the expression of genes acting downstream in the melanin synthesis. MC1R belongs to the melanocortin system, a genetic network coding for the ligands that regulate MC1R and other melanocortin receptors controlling different physiological and behavioural traits. The impact of MC1R variants on these regulatory melanocortin genes was never considered, even though MC1R mutations could alter the influence of these genes on coloration (e.g. by decreasing MC1R response to melanocortin ligands). Using barn owl growing feathers, we investigated the differences between MC1R genotypes in the (co)expression of six melanocortin and nine melanogenic-related genes and in the association between melanocortin gene expression and phenotype (feather pheomelanin content). Compared to the MC1R rufous allele, responsible for reddish coloration, the white allele was not only associated with an expected lower expression of melanogenic-related genes (TYR, TYRP1, OCA2, SLC45A2, KIT, DCT) but also with a lower MC1R expression and a higher expression of ASIP, the MC1R antagonist. More importantly, the expression of PCSK2, responsible for the maturation of the MC1R agonist, α-melanocyte-stimulating hormone, was positively related to pheomelanin content in MC1R white homozygotes but not in individuals carrying the MC1R rufous allele. These findings indicate that MC1R mutations not only alter the expression of melanogenic-related genes but also the association between coloration and the expression of melanocortin genes upstream of MC1R. This suggests that MC1R mutations can modulate the regulation of coloration by the pleiotropic melanocortin genes, potentially decoupling the often-observed associations between coloration and other phenotypes.

Circulating testosterone and feather-gene expression of receptors and metabolic enzymes in relation to melanin-based colouration in the barn owl.

Knowledge of how and why secondary sexual characters are associated with sex hormones is important to understand their signalling function. Such a link can occur if i) testosterone participates in the elaboration of sex-traits, ii) the display of an ornament triggers behavioural response in conspecifics that induce a rise in testosterone, or iii) genes implicated in the elaboration of a sex-trait pleiotropically regulate testosterone physiology. To evaluate the origin of the co-variation between melanism and testosterone, we measured this hormone and the expression of enzymes involved in its metabolism in feathers of barn owl (Tyto alba) nestlings at the time of melanogenesis and in adults outside the period of melanogenesis. Male nestlings displaying smaller black feather spots had higher levels of circulating testosterone, potentially suggesting that testosterone could block the production of eumelanin pigments, or that genes involved in the production of small spots pleiotropically regulate testosterone production. In contrast, the enzyme 5α-reductase, that metabolizes testosterone to DHT, was more expressed in feathers of reddish-brown than light-reddish nestlings. This is consistent with the hypothesis that testosterone might be involved in the expression of reddish-brown pheomelanic pigments. In breeding adults, male barn owls displaying smaller black spots had higher levels of circulating testosterone, whereas in females the opposite result was detected during the rearing period, but not during incubation. The observed sex- and age-specific co-variations between black spottiness and testosterone in nestling and adult barn owls may not result from testosterone-dependent melanogenesis, but from melanogenic genes pleiotropically regulating testosterone, or from colour-specific life history strategies that influence testosterone levels.