Where's the evidence?

Science is in large part the art of careful measurement, and a fixed measurement scale is the sine qua non of this art. It is obvious to us that measurement devices lacking fixed units and constancy of scale across applications are problematic, yet we seem oddly laissez faire in our approach to measurement of one critically important quantity: statistical evidence. Here I reconsider problems with reliance on p values or maximum LOD scores as measures of evidence, from a measure-theoretic perspective. I argue that the lack of an absolute scale for evidence measurement is every bit as problematic for modern biological research as was lack of an absolute thermal scale in pre-thermodynamic physics. Indeed, the difficulty of establishing properly calibrated evidence measures is strikingly similar to the problem 19th century physicists faced in deriving an absolute scale for the measurement of temperature. I propose that the formal relationship between the two problems might enable us to apply the mathematical foundations of thermodynamics to establish an absolute scale for the measurement of evidence, in statistical applications and possibly other areas of mathematical modeling as well. Here I begin to sketch out what such an endeavor might look like.     

Are phylogenies resolved at the genus level appropriate for studies on phylogenetic structure of species assemblages?

Phylogenies are essential to studies investigating the effect of evolutionary history on assembly of species in ecological communities and geographical and ecological patterns of phylogenetic structure of species assemblages. Because phylogenies well resolved at the species level are lacking for many major groups of organisms such as vascular plants, researchers often generate a species-level phylogenies using a phylogeny well resolved at the genus level as a backbone and attaching species to their respective genera in the phylogeny as polytomies or by using a megaphylogeny well resolved at the genus level as a backbone and adding additional species to the megaphylogeny as polytomies of their respective genera. However, whether the result of a study using species-level phylogenies generated in these ways is robust, compared to that based on phylogenies fully resolved at the species level, has not been assessed. Here, we use 1093 angiosperm tree assemblages (each in a 110 × 110 km quadrat) in North America as a model system to address this question, by examining six commonly used metrics of phylogenetic structure (phylogenetic diversity and phylogenetic relatedness) and six climate variables commonly used in ecology. Our results showed that (1) the scores of phylogenetic metrics derived from species-level phylogenies resolved at the genus level with species being attached to their respective genera as polytomies are very strongly or perfectly correlated to those derived from a phylogeny fully resolved at the species level (the mean of correlation coefficients is 0.973), and (2) the relationships between the scores of phylogenetic metrics and climate variables are consistent between the two sets of analyses based on the two types of phylogeny. Our study suggests that using species-level phylogenies resolved at the genus level with species being attached to their genera as polytomies is appropriate in studies exploring patterns of phylogenetic structure of species in ecological communities across geographical and ecological gradients.     

Where's the ecology in molecular ecology?

Molecular techniques have had a profound impact in biology. Major disciplines, including evolutionary biology, now consistently utilize molecular tools. In contrast, molecular techniques have had a more limited impact in ecology. This discrepancy is surprising. Here, we describe the unexpected paucity of ecological research in the field colloquially referred to as 'molecular ecology.' Publications over the past 15 years from the journals Ecology , Evolution and Molecular Ecology reveal that much of the research published under the molecular ecology banner is in fact evolutionary in nature, and that comparatively little ecological research incorporates molecular tools. This failure to more broadly utilize molecular techniques in ecology is alarming because several promising lines of ecological inquiry could benefit from molecular approaches. Here we summarize the use of molecular tools in ecology and evolution, and suggest several ways to renew the ecological focus in 'molecular ecology'.     

What is the phylogenetic placement of Dipteronia dyerana Henry? An example of plant species placement based on nucleotide sequences

Abstract

DNA sequence data have been widely used to evaluate species delimitations and examine infraspecific relationships. However, species placements inferred from different nucleotide sequences are frequently in conflict. As an example of plant species placement based on nucleotide sequences, the phylogenetic placement of Dipteronia dyerana Henry (Aceraceae) was analyzed in the present study. The study species included eight Acer species (from different sections of Acer), two Dipteronia species, and two outgroup taxa. Phylogenetic trees based on five datasets (ITS, trnL‐F, trnD‐trnT, psbM‐trnD, and rpl16 regions) as well as their combined datasets were generated by using maximum parsimony (MP) and maximum likelihood (ML) analyses. Further analyses were conducted to compare the strict consensus trees based on single regions and the combination of different regions. The results revealed a significant discrepancy among the phylogenetic placements of D. dyerana, inferred from various sequences. Phylogenetic trees using MP analysis based on trnD‐trnT, rpl16, and the four chloroplast combined sequences supported the genus Dipteronia as a monophyletic group, while in the other trees D. dyerana was positioned either in parallel with D. sinensis and Acer species or within the genus Acer. In ML analysis, only rpl16 and the four chloroplast combined sequence datasets supported the genus Dipteronia as a monophyletic group. We concluded that, although significant genetic differentiation occurred between D. dyerana and D. sinensis, D. dyerana was more advanced than D. sinensis. However, whether Dipteronia is monophyletic remains to be further investigated, e.g., by using more closely related taxa and more sequences. Furthermore, in addition to internal transcribed spacer sequences, more chloroplast gene sequences should be used for phylogenetic analyses of species.     

Feeding performance of fleas on different host species: is phylogenetic distance between hosts important?

We asked if and how feeding performance of fleas on an auxiliary host is affected by the phylogenetic distance between this host and the principal host of a flea. We investigated the feeding of 2 flea species, Parapulex chephrenis and Xenopsylla ramesis, on a principal (Acomys cahirinus and Meriones crassus, respectively) and 8 auxiliary host species. We predicted that fleas would perform better (higher proportion of fleas would feed and take larger bloodmeals) on (a) a principal rather than an auxiliary host and (b) auxiliary hosts phylogenetically closer to a principal host. Although feeding performance of fleas differed among different hosts, we found that: (1) fleas did not always perform better on a principal host than on an auxiliary host; and (2) flea performance on an auxiliary host was not negatively correlated with phylogenetic distance of this host from the principal host. In some cases, fleas fed better on hosts that were phylogenetically distant from their principal host. We concluded that variation in flea feeding performance among host species results from interplay between (a) inherent species-specific host defence abilities, (b) inherent species-specific flea abilities to withstand host defences and (c) evolutionary tightness of association between a particular host species and a particular flea species.     

Species concepts and floras: what are species for?

In 1992, in a special paper in the American Journal of Botany, Ernst Mayr attempted to ‘prove’ the biological species concept (BSC) worked as well in plants as it did in animals by analyzing the flora of the Concord region of northern Massachusetts. He concluded that there were minimal difficulties when applying the BSC for the plants of this particular area, and concluded that botanists were misguided in not accepting the BSC. He suggested that what he called ‘typological’ thinking was prevalent in the taxonomic community, and that this was a factor in botanical resistance to the BSC. Typology, as defined by Mayr in his 1992 foray into botany, is to a certain extent a straw‐man and, by the late 20th Century, no longer a way of thinking in widespread use in the taxonomic community in any organismal group. Here, I examine his analysis in the light of current interest in plant diversity. Species can be characterized as hypotheses about the distribution of variation in nature, subject to test with new data of many kinds. Species concepts like the BSC, although of interest philosophically and to researchers looking at mechanisms of speciation, may actually get in the way of achieving a baseline understanding of the diversity of life on Earth. © 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 95 , 17–25.     

Where's the evidence that active learning works?

Calls for reforms in the ways we teach science at all levels, and in all disciplines, are wide spread. The effectiveness of the changes being called for, employment of student-centered, active learning pedagogy, is now well supported by evidence. The relevant data have come from a number of different disciplines that include the learning sciences, cognitive psychology, and educational psychology. There is a growing body of research within specific scientific teaching communities that supports and validates the new approaches to teaching that have been adopted. These data are reviewed, and their applicability to physiology education is discussed. Some of the inherent limitations of research about teaching and learning are also discussed.     

Host-specialization and species diversity in fish parasites: phylogenetic conservatism?

The pattern of parasite species diversification and specialization, appreciated by host range, is investigated in fish parasites. We test whether host range is linked with phylogeny at a high taxonomic level, and if there is a relationship between host range and host species diversification. For this purpose we used two sets of data, one on macro-parasites of marine fishes of the Mediterranean Sea and the other on macro-parasites of marine and freshwater fishes of Canada. Similar patterns of host range among parasitic groups were found. Our findings suggest that habitat (marine vs freshwater) and geographic localization (Canada vs Mediterranean region) play little role in determining the observed patterns of host range. We highlight the potential influence of phylogeny (high-taxonomic level) on the level host range in parasites. We find that parasites with free-swimming larval stages and with direct life cycles have a narrower range of host species than do parasites with indirect life cycle, even if we cannot control for phylogenetic effects because of the lack of variation of life cycles within each parasitic group. Finally, a positive relationship was found between the number of known hosts and parasite species diversity in the case of Mediterranean parasite species. The relationship between host range and species diversification should be related to the mechanism of cospeciation.     

Do major species concepts support one, two or more species within Cryptococcus neoformans?

Cryptococcus neoformans, the agent of cryptococcosis, had been considered a homogeneous species until 1949 when the existence of four serotypes was revealed based on the antigenic properties of its polysaccharide capsule. Such heterogeneity of the species, however, remained obscure until the two morphologically distinct teleomorphs of C. neoformans were discovered during the mid 1970s. The teleomorph Filobasidiella neoformans was found to be produced by strains of serotype A and D, and Filobasidiella bacillispora was found to be produced by strains of serotype B and C. Ensuing studies revealed numerous differences between the anamorphs of the two Filobasidiella species with regard to their ecology, epidemiology, pathobiology, biochemistry and genetics. At present, the etiologic agent of cryptococcosis is classified into two species, C. neoformans (serotypes A and D) and Cryptococcus gattii (serotypes B and C). Intraspecific genetic diversity has also been revealed as more genotyping methods have been applied for each serotype. As a result, the number of scientifically valid species within C. neoformans has become a controversial issue because of the differing opinions among taxonomists as to the appropriate definition of a species. There are three major species concepts that govern classification of organisms: phenetic (morphologic, phenotypic), biologic (interbreeding) and cladistic (evolutionary, phylogenetic). Classification of the two C. neoformans species has been based on the phenetic as well as the biologic species concept, which is also supported by the cladistic species concept. In this paper, we review and attest to the validity of the current two-species system in light of the three major species concepts.     

Does phylogenetic distance aid in detecting environmental gradients related to species composition?

Questions: How should we evaluate the success of new distance measures combining community abundance and phylogenetic information? How do we interpret ordinations using these metrics? Methods: We generated synthetic data along a known environmental gradient with two hypothetical underlying phylogenetic structures: niche phylogenetically conserved or dispersed along a gradient. We also examined tree species composition associated with gradients in elevation and longitude in Oregon, USA. NMS ordinations of plots in species space from phylogenetic (PD) and Sørensen distance (SD) matrices allowed comparison of the use of PD in different scenarios. Results: PD caused plots to cluster based on the clades that they contained, reducing stress with the synthetic data but not with the real example. Phylogenetic distance highlighted clades related to gradients when these were associated. When phylogeny was not conserved along a gradient, that gradient was less strong. Regardless of phylogenetic conservation, NMS using SD consistently extracted the strongest gradients in species composition. Conclusions: The success of PD should be evaluated on how well it extracts gradients in species composition and allows community ecologists to determine which gradients are partially explained by phylogeny and not based on its ability to reduce ordination stress. PD ordinations can help community ecologists interpret niche conservation but may obscure gradients related to species composition when niches are not conserved along the gradient of interest at the scale of the study.     

Is high species richness at intermediate disturbance level valuable for phylogenetic conservation? A test on bird species in South‐East Botswana

The intermediate disturbance hypothesis (IDH) is one of the most debated theories in ecology. However, even when evidence is provided to support the hypothesis, its relevance for phylogenetic conservation has rarely been tested. Here, we investigated this question on birds in the South‐East district of Botswana along a disturbance gradient across three types of landscapes. We first reconstructed the phylogeny for all species recorded. Next, we assessed the relationship between dissimilarity measures and habitat types using the permutational MANOVA. Finally, we tested the IDH by fitting a generalized linear mixed effect model to account for errors due to spatial pseudo‐replications of our collection design. We found that, although species richness and phylogenetic diversity (PD) follow the prediction of the IDH, the evolutionary component of PD (i.e. PSV, phylogenetic species variability) contributes little to the prediction, suggesting that the correlation between PD and disturbance level is driven by the richness component of PD (i.e. PSR, phylogenetic species richness). However, the increased richness at the intermediate disturbance level does not result in phylogenetically diverse bird communities, indicating that the IDH contributes little to phylogenetic diversity. Our study adds to the body of literature questioning the relevance of IDH in ecology.     

Is phylogenetic relatedness to native species important for the establishment of reptiles introduced to California and Florida?

Aim Charles Darwin posited that introduced species with close relatives were less likely to succeed because of fiercer competition resulting from their similarity to residents. There is much debate about the generality of this rule, and recent studies on plant and fish introductions have been inconclusive. Information on phylogenetic relatedness is potentially valuable for explaining invasion outcomes and could form part of screening protocols for minimizing future invasions. We provide the first test of this hypothesis for terrestrial vertebrates using two new molecular phylogenies for native and introduced reptiles for two regions with the best data on introduction histories. Location California and Florida, USA. Methods We performed an ordination of ecological traits to confirm that ecologically similar species are indeed closely related phylogenetically. We then inferred molecular phylogenies for introduced and native reptiles using sequence data for two nuclear and three mitochondrial genes. Using these phylogenies, we computed two distance metrics: the mean phylogenetic distance (MPD) between each introduced species and all native species in each region (which indicates the potential interactions between introduced species and all native species in the community) and the distance of each introduced species to its nearest native relative – NN (indicating the degree of similarity and associated likelihood of competition between each introduced species and its closest evolutionary analogue). These metrics were compared for introduced species that established and those that failed. Results We demonstrate that phylogenetically related species do share similar ecological functions. Furthermore, successfully introduced species are more distantly related to natives (for NN and MPD) than failed species, although variation is high. Main conclusions The evolutionary history of a region has value for explaining and predicting the outcome of human‐driven introductions of reptiles. Phylogenetic metrics are thus useful inputs to multi‐factor risk assessments, which are increasingly required for screening introduced species.     

Species as family resemblance concepts: The (dis‐)solution of the species problem?

The so-called "species problem" has plagued evolutionary biology since before Darwin's publication of the aptly titled Origin of Species. Many biologists think the problem is just a matter of semantics; others complain that it will not be solved until we have more empirical data. Yet, we don't seem to be able to escape discussing it and teaching seminars about it. In this paper, I briefly examine the main themes of the biological and philosophical literatures on the species problem, focusing on identifying common threads as well as relevant differences. I then argue two fundamental points. First, the species problem is not primarily an empirical one, but it is rather fraught with philosophical questions that require-but cannot be settled by-empirical evidence. Second, the (dis-)solution lies in explicitly adopting Wittgenstein's idea of "family resemblance" or cluster concepts, and to consider species as an example of such concepts. This solution has several attractive features, including bringing together apparently diverging themes of discussion among biologists and philosophers. The current proposal is conceptually independent (though not incompatible) with the pluralist approach to the species problem advocated by Mishler, Donoghue, Kitcher and Dupré, which implies that distinct aspects of the species question need to be emphasized depending on the goals of the researcher. From the biological literature, the concept of species that most closely matches the philosophical discussion presented here is Templeton's cohesion idea.     

Does multi‐level environmental filtering determine the functional and phylogenetic composition of wild bee species assemblages?

A central goal in ecology is to develop theories that explain the diversity and distribution of species. The evolutionary history of species and their functional traits may provide mechanistic links between community assembly and the environment. Such links may be hierarchically structured such that the strength of environmental filtering decreases in a step‐wise manner from regional conditions through landscape heterogeneity to local habitat conditions. We sampled the wild bee species assemblages in power‐line strips transecting forests in south‐eastern Norway. We used altitude, landscape diversity surrounding sites and plant species composition, together with total plant cover as proxies for regional, landscape and local environmental filters, respectively. The species richness and abundance of wild bees decreased with altitude. The reduction in species richness and abundance was accompanied by a phylogenetic clustering of wild bee individuals. Furthermore, regional filters followed by local filters best explained the structure of the functional species composition. Sites at high altitudes and sites with Ericaceae‐dominated plant communities tended to have larger bees and a higher proportion of social and spring‐emerging bees. When Bombus species were excluded from the analysis, the proportion of pollen specialists increased with the dominance of Ericaceae. Furthermore, we also found that the taxonomic, phylogenetic and functional compositional turnover between sites was higher in the northern region than in the southern part of the study region. Altogether, these results suggest that regional filters drive the species richness and abundance in trait‐groups whereas local filters have more descrete sorting effects. We conclude that the model of multi‐level environmental filters provides a good conceptual model for community ecology. We suggest that future studies should focus on the relationship between the biogeographical history of species and their current distribution, and on the assumption that closely related species do indeed compete more intensely than distantly related species.