Terry Burke, Mark Blaxter, David Lank and colleagues report a reference genome sequence of the ruff and analysis of the three distinct male morphs of this bird species. They identify a ‘supergene’ consisting of a fixed inversion in two of the morphs and identify candidate reproductive trait genes in this region. Three strikingly different alternative male mating morphs (aggressive 'independents', semicooperative 'satellites' and female-mimic 'faeders') coexist as a balanced polymorphism in the ruff, Philomachus pugnax, a lek-breeding wading bird1,2,3. Major differences in body size, ornamentation, and aggressive and mating behaviors are inherited as an autosomal polymorphism4,5. We show that development into satellites and faeders is determined by a supergene6,7,8 consisting of divergent alternative, dominant and non-recombining haplotypes of an inversion on chromosome 11, which contains 125 predicted genes. Independents are homozygous for the ancestral sequence. One breakpoint of the inversion disrupts the essential CENP-N gene (encoding centromere protein N), and pedigree analysis confirms the lethality of homozygosity for the inversion. We describe new differences in behavior, testis size and steroid metabolism among morphs and identify polymorphic genes within the inversion that are likely to contribute to the differences among morphs in reproductive traits.

Avian studies are often interpreted using dual (e.g. ¹³C, ¹⁵N) isotope models, assuming turnover of both isotopes occur at similar rates, but only a few studies have quantified turnover rates for more than one of those isotopes simultaneously. To test the generality of previous turnover and fractionation estimates and assumption of synchronous C and N patterns of turnover rates, we captured Dunlin (Calidris alpina pacifica) wintering in the Fraser River Delta, British Columbia, and derived isotopic turnover rates and diet-tissue fractionation factors by experimentally manipulating diet. Birds (n = 15) were initially fed a terrestrially derived diet (mean δ¹³C: −24.7‰, mean δ¹⁵N: 3.5‰) for 54 days. A treatment group (n = 11) was then switched to a marine-derived diet (mean δ¹³C: −18.3‰, mean δ¹⁵N: 13.7‰); a control group (n = 4) was maintained on the terrestrial diet for a further 59 days. An exponential model described patterns of isotopic turnover for ¹³C and ¹⁵N, and turnover rates and half-lives of the two isotopes were correlated, confirming the assumption of synchronous patterns of turnover for those isotopes. The half-lives for ¹³C and ¹⁵N in Dunlin whole blood were 11.2 ± 0.8 days and 10.0 ± 0.6 days, respectively, and are among the lowest values obtained to date for wild birds. Variation in turnover rate among individuals was not related to indices of body condition. Les études portant sur la faune avienne sont souvent interprétées en utilisant des modèles isotopiques doubles (e.g. ¹³C, ¹⁵N), assumant que le renouvellement des deux isotopes se fait à des taux similaires. Seulement quelques études ont quantifié les taux de renouvellement pour plus d'un de ces isotopes simultanément. Pour étudier la généralisation de ce renouvellement, le fractionnement des estimés et l'hypothèse de patrons synchrones des taux de renouvellement en C et N, nous avons capturé des Calidris alpina pacifica hivernant dans le delta de la rivière Fraser, Colombie Britannique. Nous avons également dérivé les taux de renouvellement isotopiques ainsi que le fractionnement des facteurs diète-tissu en manipulant expérimentalement le régime alimentaire. Les oiseaux (n = 15) ont été tout d'abord nourris avec un régime dérivé de type terrestre (moyenne δ¹³C: −24.7‰, moyenne δ¹⁵N: 3.5‰) pendant 54 jours. Un groupe traitement (n = 11) a été ensuite soumis à un régime dérivé de type marin (moyenne δ¹³C: −18.3‰, moyenne δ¹⁵N: 13.7‰). Un groupe contrôle (n = 4) a été maintenu au régime de type terrestre durant 59 jours supplémentaires. Un modèle exponentiel décrivant les patrons de renouvellement isotopique en ¹³C et ¹⁵N, a été corrélé aux taux de renouvellement et aux demivies des deux isotopes, confirmant l'hypothèse de patrons synchrones de renouvellement des ces isotopes. Les demi-vies du ¹³C et du ¹⁵N dans le sang chez Calidris alpina pacifica étaient de 11.2 ± 0.8 jours et 10.0 ± 0.6 jours, respectivement, et sont parmi les valeurs les plus basses obtenues jusqu'à aujourd'hui pour des oiseaux sauvages. Les variations inter-individuelles du taux de renouvellement n'étaient pas reliées aux indices de conditions corporelles.

A balanced polymorphism in male mating behaviour exists in male ruffs, with no obvious parallel expression in females. Pedigree data of male phenotypes support an autosomal model of inheritance, in contrast to sex-linked inheritance patterns found in other taxa with sex-limited alternative mating strategy polymorphisms. We tested this model by inducing male courtship behaviour in gonad-intact female ruffs, using subcutaneous testosterone implants that produced physiological concentrations of testosterone. The implants rapidly induced in females both types of male mating behaviour, an increase in body mass typical of pre-breeding males, and the growth of normally male-limited breeding plumage. As predicted under an autosomal model, the distributions of induced male behaviour types in females paralleled those of their brothers and half-brothers, and were inconsistent with sex-linked models. Effects were reversible, and experimental females bred normally in subsequent years. Our results show that genotype-specific male characteristics can be induced by testosterone in female adults that have presumably not undergone neural organization for them early in life, showing direct use of genetic information in intra- and intersexual differentiation.

The adaptive significance of polyandry by female birds in the absence of direct benefits remains unclear. We determined the frequencies of polyandrous mating and multiple paternity in the ruff, a lekking shorebird with a genetic dimorphism in male mating behavior. More than half of female ruffs mate with, and have clutches fertilized by, more than one male. Individual females mate with males of both behavioral morphs more often than expected. Polyandrous mating was more likely following copulation interference, but interference was uncommon. The multiple paternity rate of ruffs is the highest known for avian lekking species and for shorebirds. The general hypothesis that pair-bond constraints are the major selective factor favoring multiple mating in birds does not predict our findings. Active genetic diversification, which has been widely dismissed as a functional explanation for polyandrous mating in birds, may apply with respect to the behavioral polymorphism in ruffs because of a Mendelian genetic basis for male behavioral morph determination and aspects of male—male cooperation and female choice. However, rates of multiple paternity in other species of lekking birds are higher than generally realized, and the potential benefits of diversification in general deserve further consideration.