The article by Scott V. Edwards discusses the emergence of a new paradigm in molecular systematics, emphasizing the primacy of species trees over gene trees. Species trees, which represent the evolutionary history of lineages, populations, and species, are argued to be a more comprehensive and accurate representation of the Tree of Life compared to gene trees, which focus on the evolutionary history of genes. The author highlights the limitations of current phylogenetic methods, particularly the concatenation approach, which combines multiple genes into a single supermatrix, often leading to incorrect phylogenetic inferences due to gene tree heterogeneity and deep coalescence. Edwards suggests that the current paradigm, characterized by the accumulation of many genes and their concatenation, is a "local optimum" in the space of possible phylogenetic models. He argues that species trees, by explicitly modeling gene tree heterogeneity and the coalescent process, provide a more robust framework for understanding evolutionary relationships. The article reviews recent advances in methods for estimating species trees, including likelihood-based and Bayesian approaches, and discusses the challenges and benefits of these methods. The author also addresses the issue of confidence in phylogenetic trees, suggesting that species tree approaches may yield lower statistical confidence, especially for long-diverged clades, due to the sensitivity of species tree methods to missing data. Despite these challenges, Edwards predicts that species tree methods will ultimately provide a more accurate and comprehensive representation of evolutionary history, clarifying many issues related to polytomies and the star tree paradox. In conclusion, the article advocates for a shift towards species tree approaches in molecular systematics, emphasizing the need to better align phylogeography and phylogenetics, and to address the challenges posed by gene tree heterogeneity and deep coalescence.The article by Scott V. Edwards discusses the emergence of a new paradigm in molecular systematics, emphasizing the primacy of species trees over gene trees. Species trees, which represent the evolutionary history of lineages, populations, and species, are argued to be a more comprehensive and accurate representation of the Tree of Life compared to gene trees, which focus on the evolutionary history of genes. The author highlights the limitations of current phylogenetic methods, particularly the concatenation approach, which combines multiple genes into a single supermatrix, often leading to incorrect phylogenetic inferences due to gene tree heterogeneity and deep coalescence. Edwards suggests that the current paradigm, characterized by the accumulation of many genes and their concatenation, is a "local optimum" in the space of possible phylogenetic models. He argues that species trees, by explicitly modeling gene tree heterogeneity and the coalescent process, provide a more robust framework for understanding evolutionary relationships. The article reviews recent advances in methods for estimating species trees, including likelihood-based and Bayesian approaches, and discusses the challenges and benefits of these methods. The author also addresses the issue of confidence in phylogenetic trees, suggesting that species tree approaches may yield lower statistical confidence, especially for long-diverged clades, due to the sensitivity of species tree methods to missing data. Despite these challenges, Edwards predicts that species tree methods will ultimately provide a more accurate and comprehensive representation of evolutionary history, clarifying many issues related to polytomies and the star tree paradox. In conclusion, the article advocates for a shift towards species tree approaches in molecular systematics, emphasizing the need to better align phylogeography and phylogenetics, and to address the challenges posed by gene tree heterogeneity and deep coalescence.