The three-spined stickleback (Gasterosteus aculeatus) is an important model system for the study of parallel evolution in the wild, having repeatedly colonized and adapted to freshwater from the sea throughout the northern hemisphere. Previous studies identified numerous genomic regions showing consistent genetic differentiation between freshwater and marine ecotypes, but these had typically limited geographic sampling and mostly focused on the Eastern Pacific region. We analysed population genomic data from the three-spined stickleback marine and freshwater ecotypes covering the entire species' range to detect loci involved in parallel evolution at different geographic scales. Most signatures of parallel evolution were unique to the Eastern Pacific and trans-oceanic 2 marine-freshwater differentiation was restricted to a limited number of shared genomic regions, including three chromosomal inversions. Based on simulations and empirical data, we demonstrate that this could result from the stochastic loss of freshwater-adapted alleles during the invasion of the Atlantic basin and selection against freshwater-adapted variants in the sea, both of which can reduce standing genetic variation available for freshwater adaptation outside the Eastern Pacific region. Moreover, the elevated linkage disequilibrium associated with marine-freshwater differentiation in the Eastern Pacific is consistent with secondary contact between marine and freshwater populations that evolved in isolation from each other during past glacial periods. Thus, contrary to what earlier studies from the Eastern Pacific region have led us to believe, parallel marinefreshwater differentiation in sticklebacks is far less prevalent and pronounced in all other parts of the species global distribution range.

Stickleback fishes in the family Gasterosteidae have become model organisms in ecology and evolutionary biology. However, even in the case of the most widely studied species in this family - the three-spined stickleback (Gasterosteus aculeatus) - the worldwide phylogenetic relationships and colonization history of the different populations and lineages remain poorly resolved. Using a large collection of samples covering most parts of the species distribution range, we subjected thousands of single nucleotide polymorphisms to coalescent analyses in order to reconstruct a robust worldwide phylogeny of extant G. aculeatus populations, as well as their ancestral geographic distributions using Statistical-Dispersal Vicariance and Bayesian Binary MCMC analyses. The results suggest that contemporary populations originated from the Pacific Ocean in the Late Pleistocene, and the Atlantic was colonized through the Arctic Ocean by a lineage that diverged from Pacific sticklebacks ca 44.6 Kya. This lineage contains two branches: one that is distributed in the Mediterranean area, from the Iberian Peninsula to the Black Sea ('Southern European Clade'), and another that is comprised of populations from northern Europe and the east coast of North America ('Trans-Atlantic Clade'). Hence, the results suggest that the North American East Coast was colonized by trans-Atlantic migration. Coalescence-based divergence time estimates suggest that divergence among major clades is much more recent than previously estimated.

Hybridization and convergent evolution are phenomena of broad interest in evolutionary biology, but their occurrence poses challenges for reconstructing evolutionary affinities among affected taxa. Sticklebacks in the genus Pungitius are a case in point: evolutionary relationships and taxonomic validity of different species and populations in this circumpolarly distributed species complex remain contentious due to convergent evolution of traits regarded as diagnostic in their taxonomy, and possibly also due to frequent hybridization among taxa. To clarify the evolutionary relationships among different Pungitius species and populations globally, as well as to study the prevalence and extent of introgression among recognized species, genomic data sets of both reference genome‐anchored single nucleotide polymorphisms and de novo assembled RAD‐tag loci were constructed with RAD‐seq data. Both data sets yielded topologically identical and well‐supported species trees. Incongruence between nuclear and mitochondrial DNA‐based trees was found and suggested possibly frequent hybridization and mitogenome capture during the evolution of Pungitius sticklebacks. Further analyses revealed evidence for frequent nuclear genetic introgression among Pungitius species, although the estimated proportions of autosomal introgression were low. Apart from providing evidence for frequent hybridization, the results challenge earlier mitochondrial and morphology‐based hypotheses regarding the number of species and their affinities in this genus: at least seven extant species can be recognized on the basis of genetic data. The results also shed new light on the biogeographical history of the Pungitius‐complex, including suggestion of several trans‐Arctic invasions of Europe from the Northern Pacific. The well‐resolved phylogeny should facilitate the utility of this genus as a model system for future comparative evolutionary studies.

Knowledge on species’ breeding biology is the building blocks of avian life history theory. A review for the current status of the knowledge at a global scale is needed to highlight the priority for future research. We collected all available information on three critical nesting parameters (clutch size, incubation period and nestling period) for the close to 10 000 bird species in the world and identified taxonomic, geographic and habitat gaps in the distribution of knowledge on avian breeding biology. The results show that only one third of all extant species are well known regarding the three nesting parameters analyzed, while the rest are partly or poorly known. Most data deficient taxonomic groups are tropical forest nesters, particularly from the Amazon basin, southeast Asia, Equatorial Africa and Madagascar – the places that harbor the world's highest bird diversity. These knowledge gaps could be hampering our understanding of avian life histories. Ornithologists are encouraged to pay more efforts to explore the breeding biology of those poorly‐known species.

Population genetic theory predicts that small effective population sizes (Ne) and restricted gene flow limit the potential for local adaptation. In particular, the probability of evolving similar phenotypes based on shared genetic mechanisms (i.e. parallel evolution), is expected to be reduced. We tested these predictions in a comparative genomic study of two ecologically similar and geographically co-distributed stickleback species (viz. Gasterosteus aculeatus and Pungitius pungitius). We found that P. pungitius harbours less genetic diversity and exhibits higher levels of genetic differentiation and isolation-by-distance than G. aculeatus. Conversely, G. aculeatus exhibits a stronger degree of genetic parallelism across freshwater populations than P. pungitius: 2996 vs. 379 SNPs located within 26 vs. nine genomic regions show evidence of selection in multiple freshwater populations of G. aculeatus and P. pungitius, respectively. Most regions involved in parallel evolution in G. aculeatus showed increased levels of divergence, suggestive of selection on ancient haplotypes. In contrast, haplotypes involved in freshwater adaptation in P. pungitius were younger, and often associated with reduced diversity. In accordance with theory, the results suggest that connectivity and genetic drift play crucial roles in determining the levels and geographic distribution of standing genetic variation, providing evidence that population subdivision limits local adaptation and therefore also the likelihood of parallel evolution.