Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Extra-pair paternity rates vary markedly across avian taxa, but patterns of variation in this trait have been obscured by a paucity of data on closely related species, especially those spanning broad environmental gradients. Here we compare variation in extra-pair paternity rates among five species in the widespread swallow genus Tachycineta. Rates of extra-pair paternity vary widely in this group, ranging from 13 to 87% of nests having extra-pair young. The inter-specific variation in extra-pair paternity within this small group of closely related swallows has a range equivalent to that found among all Hirundinidae and is close to the range of variation across all birds. Despite theory that predicts extra-pair paternity rates to be explained by latitudinal variation in breeding synchrony our results show that extra-pair paternity rates in this genus do not closely track a latitudinal gradient, as predicted by studies of other life-history traits, and are not explained by differences in breeding synchrony as previously suggested. The genetic mating systems of birds, described by the rates of extra-pair paternity, are connected to all other life-history traits through a complex network of trade-offs with organismal (phylogenetic) and ecological (environmental) factors. Disentangling each of these interactions to understand latitudinal patterns in any given life-history trait remains a daunting task.
The focus of our study is to analyse geographic variation in EPP rates in a group of closely related birds and test the breeding synchrony hypothesis. We explore here the genetic mating system of five species in the swallow genus Tachycineta (Tree swallow T. bicolor, Violet-green swallow T. thalassina, Mangrove swallow T. albilinea, White-rumped swallow T. leucorrhoa, and Chilean swallow T. meyeni) that span a wide breeding distribution in North, Central and South America. Previous studies on four species in this genus have shown that Tachycineta swallows have high variation in their rates of EPP, ranging from 13 to 87% of nests having extra-pair young18,19,20,21. In this study, we more fully document inter-specific variation in EPP among Tachycineta swallows by (i) characterizing for the first time the genetic mating system of Violet-green swallows; (ii) describing the rate of EPP in additional populations of Mangrove and White-rumped swallows; (iii) examining the geographic pattern of variation in EPP rates and testing the latitudinal variation in this trait; and (iv) testing the breeding synchrony hypothesis along a latitudinal gradient. This analysis of variation among closely related Tachycineta species allows us to simultaneously evaluate diversification in EPP rates from a historical and a contemporary ecological standpoint. To our knowledge this is the first and most comprehensive study of genetic mating system of several members in a taxon with a widespread New World distribution, spanning Northern, tropical and Southern latitudes.
Map of the ranges of the nine species of Tachycineta swallows. (A) Bars on the right represent per cent of nests with extra-pair young for the colonies sampled, including the ones reported for the first time in this study and those taken from the literature. (B) Bars on the right represent per cent of nests with extra-pair young for the colonies sampled and the per cent of breeding synchrony in the colony for the years sampled (note that EPP rates for Tree swallows differ in this respect between the two maps, this is because the study used for the reported synchrony showed a somewhat lower EPP rate for other years studied). Bars are coded by species (see colour codes) and are located at the approximate latitudes of the populations sampled.
Understanding patterns of variation in life-history traits (e.g., EPP) requires looking at within-population variation as well as variation between species and populations. Most studies on avian EPP rates to date have been restricted to single populations and single species within most genera, limiting our understanding of how evolutionary and environmental changes can affect genetic mating systems6,22. The North temperate species Tree swallow has long been the focus of research of behavioural ecologists, in part because of its extremely high rates of EPP23 (Table 4). The contrasting finding of low rates of EPP in the tropical Mangrove swallow and the southernmost species, Chilean swallow (Table 4), makes a compelling case for studying variation in EPP rates between closely related species in this genus, providing us with a unique dataset for studies of variation in paternity.
We sampled a population of Violet-green swallows breeding in Lee Vining, California. We found very high rates of EPP in this population (67% of nests and 56% of nestlings), comparable to those of White-rumped swallows breeding at a similar latitude in the South (Table 1). Despite the many studies on Tree swallows in North America, it is surprising that their congener, the Violet-green swallow, had been largely neglected in studies of mating systems, especially since both species overlap in part of their range24,25. This is, to our knowledge, the first study to describe the genetic mating system of this species. We used this information on the analyses of latitudinal variation in EPP and SI.
Our findings might be strongly driven by data from the Chilean swallow population sampled and/or by the scarcity of sites at intermediate latitudes in both hemispheres (see Fig. 1). The population of Chilean swallows breeding in Tierra del Fuego may be an outlier to a general latitudinal trend20. Other life-history traits in this population (i.e., unexpectedly small clutch size, high adult survival29,33) do not fit the pattern of geographic variation described in multi-species studies34 and many aspects of the breeding biology of this population may be responses to the distinctively extreme climate of the Fuegian breeding site. One of the challenges of the latitudinal hypothesis is that it does not specify exactly which environmental features act proximally to cause a latitudinal pattern of variation. As mentioned above, the other important caveat is the lack of studies from sites at intermediate latitudes in both hemispheres, which might respond mainly to the concentration of researchers at other latitudes. Further field work, concentrated on sites with extreme sets of environmental variables as well as those at intermediate latitudes, could help elucidate which of the many factors that change with latitude are likely to be most important. One consequence of such further work might be that Fuegian Chilean swallows will come to be seen as only one of many exceptions to a latitudinal trend, or that Chilean swallows will serve as indicators of a relationship between EPP and some presently unappreciated aspect of their biology. Indeed, it may be that studies at a few more sites will be sufficient to erase any suggestion of a general latitudinal trend. We currently do not have enough information and sites sampled to tease apart these two alternatives.
We extracted DNA from blood and muscle samples using DNA purification kits by Eppendorf and Qiagen. Extracted DNA was diluted 1:10 in ultrapurified H2O and then amplified at a panel of highly polymorphic microsatellite loci43,44,45,46, which differed somewhat across species (see the Supplementary Table S1 for combination of loci used for each species, primer concentrations, details on polymerase chain reaction (PCR) concentrations, conditions and cycling profiles). PCR products were then genotyped on an ABI PRISM 3100 Genetic Analyzer (Applied Biosystems), and the sizes of the microsatellite alleles estimated using GeneScan-500 LIZ size standard (Applied Biosystems) and the software GeneMapper (v3.7 Applied Biosystems).
We used the program Cervus 3.047,48 to generate allele frequencies and population genetic parameters, and assess paternity for the populations studied. The combined exclusion probability for all loci used for each species was >0.9999. Species-specific details on the maximum likelihood assessment of paternity can be found in the Supplementary Methods.
We used population-wide rates of EPP and average SI to map these variables relative to latitude, but the statistical analyses of latitudinal variation in SI and EPP were limited to those populations for which we had individual female information. All statistical analyses were done in R, version 3.5.151.
To assess variation in SI among the different latitudes we used a generalized linear model (GLM52), and considered latitude a categorical variable in this analysis. We assumed a beta binomial distribution because we obtained overdispersion when modelling with a binomial error distribution53,54. The response variable was modelled with a logit link function, using the R package and function glmmTMB55.
We also assessed variation in EPP, considered a dichotomous variable in our analyses (i.e., females that had extra-pair young in their nests, and females that did not have any extra-pair young in their nests), with respect to the explanatory variables latitude and SI. We used a GLM, with a binomial error distribution and a logit link function52, and used the glm function of the stats package in R51.
V.F. conceived and designed the study, collected field data for White-rumped swallows and Mangrove swallows, carried out all the molecular work and statistical analyses, and drafted the manuscript. V.M. collected field data for White-rumped swallows and helped edit the manuscript. M.L. collected field data for Chilean swallows. M.V.L. carried out the statistical analyses. D.W.W. collected field data for Violet-green swallows and helped edit the manuscript. All authors gave final approval for publication.
c80f0f1006