The scaling of diversification rates with age is likely explained by sampling bias

Apparent timescaling of fossil diversification rates is caused by sampling bias

Negative scaling relationships between both speciation and extinction rates, on the one hand, and the age or duration of organismal groups on the other, are pervasive and recovered in both molecular phylogenetic and fossil time series.1,2,3,4 The agreement between molecular and fossil data hints at a universal cause and potentially at incongruence between micro- and macroevolution. However, the existence of negative rate scaling in fossil time series has not undergone the same level of scrutiny as in molecular data. Here, we analyze the marine animal fossil record across the last ∼538.8 Ma of the Phanerozoic to investigate the presence and strength of negative rate scaling. We find that negative rate scaling arises under commonly applied age range-based per capita rates, which do not control for sampling bias, but are severely reduced or absent when metrics are used that do correct for sampling. We further show by simulation that even moderately incomplete sampling of species occurrences through time may induce rate scaling. We thus conclude that there are no significant scaling relationships present in these fossil clades and that any apparent trend is caused by sampling artefacts and taxonomic practices. If rate scaling in molecular phylogenies is genuine, the absence of such a relationship in the fossil record will provide a valuable benchmark and constraint on what processes can cause it.

Reading Yule in light of the history and present of macroevolution

Yule's 1925 paper introducing the branching model that bears his name was a landmark contribution to the biodiversity sciences. In his paper, Yule developed stochastic models to explain the observed distribution of species across genera and to test hypotheses about the relationship between clade age, diversity and geographic range. Here, we discuss the intellectual context in which Yule produced this work, highlight Yule's key mathematical and conceptual contributions using both his and more modern derivations and critically examine some of the assumptions of his work through a modern lens. We then document the strange trajectory of his work through the history of macroevolutionary thought and discuss how the fundamental challenges he grappled with-such as defining higher taxa, linking microevolutionary population dynamics to macroevolutionary rates, and accounting for inconsistent taxonomic practices-remain with us a century later.This article is part of the theme issue '"A mathematical theory of evolution": phylogenetic models dating back 100 years'.

Plant species diversification in the Himalaya-Hengduan Mountains region: an example from an endemic lineage of Pedicularis (Orobanchaceae) in the role of floral specializations and rapid range expansions

The c. 270 endemic species of Pedicularis in the Himalaya-Hengduan Mountains (HHM) region exhibit high diversity in geographic distribution, elevational range and floral morphology. Many of these, including the species with the longest corolla tubes and beaked galeas, are monophyletic and represent a putative in situ radiation. In this study, we focus on the representative Clade 3 within the HHM region. We integrate the plastid phylogeny of this clade with environmental data and species distributions to infer environmental correlates of species diversity. We estimate macroevolutionary rates and reconstructed ancestral states for geographic ranges and corolla traits, and analyse patterns of range overlap and niche evolution to assess drivers of diversification in the HHM region. Our results show that the region from northwest Yunnan to southwest Sichuan is the centre of diversity for this clade of Pedicularis. Rates of diversification are associated with precipitation and multiple environmental factors. Multiple range expansions from the Sanjiang (Three Parallel Rivers) region, followed by allopatric speciation across the HHM region, contributed to early rapid diversification. Corolla traits are not significantly associated with species diversification. This study highlights the importance of integrated evidence for understanding species diversification dynamics and contributes to our understanding of the origins of the remarkable richness of plant species in the HHM region.

Commonly used Bayesian diversification methods lead to biologically meaningful differences in branch-specific rates on empirical phylogenies

Identifying along which lineages shifts in diversification rates occur is a central goal of comparative phylogenetics; these shifts may coincide with key evolutionary events such as the development of novel morphological characters, the acquisition of adaptive traits, polyploidization or other structural genomic changes, or dispersal to a new habitat and subsequent increase in environmental niche space. However, while multiple methods now exist to estimate diversification rates and identify shifts using phylogenetic topologies, the appropriate use and accuracy of these methods are hotly debated. Here we test whether five Bayesian methods-Bayesian Analysis of Macroevolutionary Mixtures (BAMM), two implementations of the Lineage-Specific Birth-Death-Shift model (LSBDS and PESTO), the approximate Multi-Type Birth-Death model (MTBD; implemented in BEAST2), and the Cladogenetic Diversification Rate Shift model (ClaDS2)-produce comparable results. We apply each of these methods to a set of 65 empirical time-calibrated phylogenies and compare inferences of speciation rate, extinction rate, and net diversification rate. We find that the five methods often infer different speciation, extinction, and net-diversification rates. Consequently, these different estimates may lead to different interpretations of the macroevolutionary dynamics. The different estimates can be attributed to fundamental differences among the compared models. Therefore, the inference of shifts in diversification rates is strongly method dependent. We advise biologists to apply multiple methods to test the robustness of the conclusions or to carefully select the method based on the validity of the underlying model assumptions to their particular empirical system.

The causes of species richness patterns among clades

Two major types of species richness patterns are spatial (e.g. the latitudinal diversity gradient) and clade-based (e.g. the dominance of angiosperms among plants). Studies have debated whether clade-based richness patterns are explained primarily by larger clades having faster rates of species accumulation (speciation minus extinction over time; diversification-rate hypothesis) or by simply being older (clade-age hypothesis). However, these studies typically compared named clades of the same taxonomic rank, such as phyla and families. This study design is potentially biased against the clade-age hypothesis, since clades of the same rank may be more similar in age than randomly selected clades. Here, we analyse the causes of clade-based richness patterns across the tree of life using a large-scale, time-calibrated, species-level phylogeny and random sampling of clades. We find that within major groups of organisms (animals, plants, fungi, bacteria, archaeans), richness patterns are most strongly related to clade age. Nevertheless, weaker relationships with diversification rates are present in animals and plants. These overall results contrast with similar large-scale analyses across life based on named clades, which showed little effect of clade age on richness. More broadly, these results help support the overall importance of time for explaining diverse types of species richness patterns.

Completing a molecular timetree of apes and monkeys

The primate infraorder Simiiformes, comprising Old and New World monkeys and apes, includes the most well-studied species on earth. Their most comprehensive molecular timetree, assembled from thousands of published studies, is found in the TimeTree database and contains 268 simiiform species. It is, however, missing 38 out of 306 named species in the NCBI taxonomy for which at least one molecular sequence exists in the NCBI GenBank. We developed a three-pronged approach to expanding the timetree of Simiiformes to contain 306 species. First, molecular divergence times were searched and found for 21 missing species in timetrees published across 15 studies. Second, untimed molecular phylogenies were searched and scaled to time using relaxed clocks to add four more species. Third, we reconstructed ten new timetrees from genetic data in GenBank, allowing us to incorporate 13 more species. Finally, we assembled the most comprehensive molecular timetree of Simiiformes containing all 306 species for which any molecular data exists. We compared the species divergence times with those previously imputed using statistical approaches in the absence of molecular data. The latter data-less imputed times were not significantly correlated with those derived from the molecular data. Also, using phylogenies containing imputed times produced different trends of evolutionary distinctiveness and speciation rates over time than those produced using the molecular timetree. These results demonstrate that more complete clade-specific timetrees can be produced by analyzing existing information, which we hope will encourage future efforts to fill in the missing taxa in the global timetree of life.

The Major Features of Macroevolution

Evolutionary dynamics operating across deep time leave footprints in the shapes of phylogenetic trees. For the last several decades, researchers have used increasingly large and robust phylogenies to study the evolutionary history of individual clades and to investigate the causes of the glaring disparities in diversity among groups. Whereas typically not the focal point of individual clade-level studies, many researchers have remarked on recurrent patterns that have been observed across many different groups and at many different time scales. Whereas previous studies have documented various such regularities in topology and branch length distributions, they have typically focused on a single pattern and used a disparate collection (oftentimes, of quite variable reliability) of trees to assess it. Here we take advantage of modern megaphylogenies and unify previous disparate observations about the shapes embedded in the Tree of Life to create a catalog of the "major features of macroevolution." By characterizing such a large swath of subtrees in a consistent way, we hope to provide a set of phenomena that process-based macroevolutionary models of diversification ought to seek to explain.

Conceptual and empirical bridges between micro- and macroevolution

Explaining broad molecular, phenotypic and species biodiversity patterns necessitates a unifying framework spanning multiple evolutionary scales. Here we argue that although substantial effort has been made to reconcile microevolution and macroevolution, much work remains to identify the links between biological processes at play. We highlight four major questions of evolutionary biology whose solutions require conceptual bridges between micro and macroevolution. We review potential avenues for future research to establish how mechanisms at one scale (drift, mutation, migration, selection) translate to processes at the other scale (speciation, extinction, biogeographic dispersal) and vice versa. We propose ways in which current comparative methods to infer molecular evolution, phenotypic evolution and species diversification could be improved to specifically address these questions. We conclude that researchers are in a better position than ever before to build a synthesis to understand how microevolutionary dynamics unfold over millions of years.

Arrested diversification? The phylogenetic distribution of poorly-diversifying lineages

Rapidly-diversifying lineages have been a major focus of modern evolutionary biology, with many hypotheses seeking to explain how they contribute to the uneven distribution of species in space and among taxa. However, an alternative view that is rarely explored is that some lineages evolve to become depauperate clades and show disproportionately low diversification, in a phenomenon we here call arrested diversification. In this study, we analyse several large-scale datasets including amphibian, squamate, mammal, and seed plant species to assess the extent to which poorly-diversifying lineages show distinct phylogenetic and spatial distributions in relation to other lineages. We found significant evidence that clades with low diversification rates tend to be more phylogenetically overdispersed than expected and show more idiosyncratic spatial distributions. These results suggest that arrested diversification is a real phenomenon that might play an important (yet largely overlooked) role in explaining asymmetries in the distribution of species across lineages.