Are geometric morphometric analyses replicable? Evaluating landmark measurement error and its impact on extant and fossil Microtus classification

Geometric morphometric analyses are frequently employed to quantify biological shape and shape variation. Despite the popularity of this technique, quantification of measurement error in geometric morphometric datasets and its impact on statistical results is seldom assessed in the literature. Here, we evaluate error on 2D landmark coordinate configurations of the lower first molar of five North American Microtus (vole) species. We acquired data from the same specimens several times to quantify error from four data acquisition sources: specimen presentation, imaging devices, interobserver variation, and intraobserver variation. We then evaluated the impact of those errors on linear discriminant analysis‐based classifications of the five species using recent specimens of known species affinity and fossil specimens of unknown species affinity. Results indicate that data acquisition error can be substantial, sometimes explaining >30% of the total variation among datasets. Comparisons of datasets digitized by different individuals exhibit the greatest discrepancies in landmark precision, and comparison of datasets photographed from different presentation angles yields the greatest discrepancies in species classification results. All error sources impact statistical classification to some extent. For example, no two landmark dataset replicates exhibit the same predicted group memberships of recent or fossil specimens. Our findings emphasize the need to mitigate error as much as possible during geometric morphometric data collection. Though the impact of measurement error on statistical fidelity is likely analysis‐specific, we recommend that all geometric morphometric studies standardize specimen imaging equipment, specimen presentations (if analyses are 2D), and landmark digitizers to reduce error and subsequent analytical misinterpretations.     

Patterns of Evolutionary Divergence and Convergence in the Bushy-Tailed Woodrat (Neotoma cinerea) Across Western North America

The degree of concordance between patterns of divergence in genetic and morphological characters provides insight into the process of evolutionary diversification. We analyzed species-wide morphological variation in a broadly distributed species of small mammal, the bushy-tailed woodrat (Neotoma cinerea), for comparison against a recently developed mitochondrial phylogeny. We focused our analyses on patterns of qualitative and quantitative craniodental morphological variation throughout the distribution of this taxon. We found strong concordance between the morphology of one prominent craniodental feature (sphenopalatine vacuity) and the major mitochondrial lineages, but only minimal differentiation in overall morphometric variation between these clades. Further, in contrast to the largely east–west subdivision of mitochondrial and sphenopalatine vacuity variation, cluster analyses of morphometric variation revealed a north–south morphological subdivision irrespective of mitochondrial clade. Our combined character sets are best explained by a deep history of evolutionary divergence within this taxon accompanied or followed by broad-scale convergent or parallel evolution.     

Molecular phylogenetics of slit‐faced bats (Chiroptera: Nycteridae) reveal deeply divergent African lineages

The bat family Nycteridae contains only the genus Nycteris, which comprises 13 currently recognized species from Africa and the Arabian Peninsula, one species from Madagascar, and two species restricted to Malaysia and Indonesia in South‐East Asia. We investigated genetic variation, clade membership, and phylogenetic relationships in Nycteridae with broad sampling across Africa for most clades. We sequenced mitochondrial cytochrome b (cytb) and four independent nuclear introns (2,166 bp) from 253 individuals. Although our samples did not include all recognized species, we recovered at least 16 deeply divergent monophyletic lineages using independent mitochondrial and multilocus nuclear datasets in both gene tree and species tree analyses. Mean pairwise uncorrected genetic distances among species‐ranked Nycteris clades (17% for cytb and 4% for concatenated introns) suggest high levels of phylogenetic diversity in Nycteridae. We found a large number of designated clades whose members are distributed wholly or partly in East Africa (10 of 16 clades), indicating that Nycteris diversity has been historically underestimated and raising the possibility that additional unsampled and/or undescribed Nycteris species occur in more poorly sampled Central and West Africa. Well‐resolved mitochondrial, concatenated nuclear, and species trees strongly supported African ancestry for SE Asian species. Species tree analyses strongly support two deeply diverged subclades that have not previously been recognized, and these clades may warrant recognition as subgenera. Our analyses also strongly support four traditionally recognized species groups of Nycteris. Mitonuclear discordance regarding geographic population structure in Nycteris thebaica appears to result from male‐biased dispersal in this species. Our analyses, almost wholly based on museum voucher specimens, serve to identify species‐rank clades that can be tested with independent datasets, such as morphology, vocalizations, distributions, and ectoparasites. Our analyses highlight the need for a comprehensive revision of Nycteridae.     

The taxonomic status of some Atlanto‐Mediterranean species in the subgenus Holothuria (Echinodermata: Holothuroidea: Holothuriidae) based on molecular evidence

Molecular and morphological data were used to evaluate the taxonomic status of the species Holothuria tubulosa Gmelin, 1790 , Holothuria stellati Delle Chiaje, 1823 , Holothuria mammata Grube, 1840 , and Holothuria dakarensis Panning, 1939, belonging to the nominate subgenus Holothuria (Holothuria) (family Holothuriidae) from the Mediterranean Sea and Atlantic Ocean. A 16S rRNA marker distinguished three well‐supported clades with clear genetic differentiation amongst them. The morphometric characters, although they reflected the clades, showed great variability, and some specimens from different clades overlapped. The morphological data and the literature suggest that the clades correspond to H. dakarensis (from Cape Verde Islands), H. mammata (from the Atlanto‐Mediterranean area) and H. tubulosa (from the Mediterranean Sea). Holothuria stellati is considered here to be a junior subjective synonym of H. tubulosa. Great morphological intraspecific variation within H. tubulosa and H. mammata explains the confusion in the literature. Holothuria tubulosa includes specimens with distinctive ossicles, but others are similar to H. mammata. In these cases, the presence or absence of Cuvierian tubules proved a reliable indicator to the identity of these species; unfortunately this character is difficult to assess in preserved material. According to the results of discriminant analysis we propose a set of ossicle morphometric variables that permit the optimum assignation of individuals to the clades. Our results present a new perspective on the taxonomic status of species in Holothuria (Holothuria), and show how a molecular approach, combined with a morphological approach, can solve taxonomic problems.     

How should gaps be treated in parsimony? A comparison of approaches using simulation

Simulation with indels was used to produce alignments where true site homologies in DNA sequences were known; the gaps from these datasets were removed and the sequences were then aligned to produce hypothesized alignments. Both alignments were then analyzed under three widely used methods of treating gaps during tree reconstruction under the maximum parsimony principle. With the true alignments, for many cases (82%), there was no difference in topological accuracy for the different methods of gap coding. However, in cases where a difference was present, coding gaps as a fifth state character or as separate presence/absence characters outperformed treating gaps as unknown/missing data nearly 90% of the time. For the hypothesized alignments, on average, all gap treatment approaches performed equally well. Data sets with higher sequence divergence and more pectinate tree shapes with variable branch lengths are more affected by gap coding than datasets associated with shallower non-pectinate tree shapes.     

Population and phylogenomic decomposition via genotyping‐by‐sequencing in Australian Pelargonium

Species delimitation has seen a paradigm shift as increasing accessibility of genomic‐scale data enables separation of lineages with convergent morphological traits and the merging of recently diverged ecotypes that have distinguishing characteristics. We inferred the process of lineage formation among Australian species in the widespread and highly variable genus Pelargonium by combining phylogenomic and population genomic analyses along with breeding system studies and character analysis. Phylogenomic analysis and population genetic clustering supported seven of the eight currently described species but provided little evidence for differences in genetic structure within the most widely distributed group that containing P. australe. In contrast, morphometric analysis detected three deep lineages within Australian Pelargonium; with P. australe consisting of five previously unrecognized entities occupying separate geographic ranges. The genomic approach enabled elucidation of parallel evolution in some traits formerly used to delineate species, as well as identification of ecotypic morphological differentiation within recognized species. Highly variable morphology and trait convergence each contribute to the discordance between phylogenomic relationships and morphological taxonomy. Data suggest that genetic divergence among species within the Australian Pelargonium may result from allopatric speciation while morphological differentiation within and among species may be more strongly driven by environmental differences.     

Intraspecific diversity and distribution of the cosmopolitan species Pseudo‐nitzschia pungens (Bacillariophyceae): morphology,genetics, and ecophysiology of the three clades

Three clades of Pseudo‐nitzschia pungens, determined by the internal transcribed space (ITS) region, are distributed throughout the world. We studied 15 P. pungens clones from various geographical locations and confirmed the existence of the three clades within P. pungens, based on ITS sequencing and described the three subgroups (IIIaa, IIIab, and IIIb) of clade III. Clade III (clade IIIaa) populations were reported for the first time in Korean coastal waters and the East China Sea. In morphometric analysis, we found the ultrastructural differences in the number of fibulae, striae, and poroids that separate the three clades. We carried out physiological tests on nine clones belonging to the three clades growing under various culture conditions. In temperature tests, only clade III clones could not grow at lower temperatures (10°C and 15°C), although clade I and II clones grew well. The estimated optimal growth range of clade I clones was wider than that of clades II and III. Clade II clones were considered to be adapted to lower temperatures and clade III to higher temperatures. In salinity tests, clade II and III clones did not grow well at a salinity of 40. Clade I clones were regarded as euryhaline and clade II and III clones were stenohaline. This supports the hypothesis that P. pungens clades have different ecophysiological characteristics based on their habitats. Our data show that physiological and morphological features are correlated with genetic intraspecific differentiation in P. pungens.     

A three‐dimensional geometric morphometric study of the development of sulcus versus shell outline in Permian neospiriferine brachiopods

The morphological variation of the sulcal development and shell outline in large Permian neospiriferine brachiopods including Fasciculatia Waterhouse, 2004 is investigated using geometric morphometrics. The sulcal tongues of spiriferide brachiopods can be, in a qualitative sense, categorized into three types according to the degree of their development: short sulcal tongue, long sulcal tongue and geniculated sulcal tongue. All three types have been noted within Fasciculatia striatoparadoxa, regardless of the nature of the substrate which they originally inhabited. To quantify its morphological variation both in sulcal development and shell outline, 51 brachiopod shells were scanned with a three‐dimensional (3‐D) surface imaging device, and their 3‐D models were reconstructed. Using two landmarks and 58 semilandmarks designated on the surface of the reconstructed 3‐D models, a landmark‐based morphometric analysis was performed. Our result demonstrates a significant intraspecific variation of sulcal development in F. striatoparadoxa and its relatives. Local environmental factors, especially the intensity of ambient water flow, are invoked as the most likely cause for this intraspecific variation. Additionally, this study also shows that there are considerable interspecific distinctions in shell outline among Fasciculatia species, independent of the high variation in the sulcal development. The strong stability of overall shell outline at species level implies a decoupled morphological development between sulcal tongue and whole shell outline. The 3‐D morphometric approach applied here demonstrates its great utility as a tool for quantifying and analysing the morphological variation of highly convex brachiopod shells.     

Morphological divergence of lake and stream Phoxinus of Northern Italy and the Danube basin based on geometric morphometric analysis

Minnows of the genus Phoxinus are promising candidates to investigate adaptive divergence, as they inhabit both still and running waters of a variety of altitudes and climatic zones in Europe. We used landmark‐based geometric morphometric methods to quantify the level of morphological variability in Phoxinus populations from streams and lakes of Northern Italy and the Danube basin. We analyzed body shape differences of populations in the dorsal, lateral, and ventral planes, using a large array of landmarks and semilandmarks. As the species identification of Phoxinus on morphological characters is ambiguous, we used two mitochondrial genes to determine the genetic background of the samples and to ensure we are comparing homogenous groups. We have found significant body shape differences between habitats: Minnow populations inhabiting streams had a deeper body and caudal peduncle and more laterally inserted pectoral fins than minnows inhabiting lakes. We have also found significant body shape differences between genetic groups: Italian minnows had deeper bodies, deeper and shorter caudal peduncles, and a shorter and wider gape than both groups from the Danube. Our results show that the morphology of Phoxinus is highly influenced by habitat and that body shape variation between habitats was within the same range as between genetic groups. These morphological differences are possibly linked to different modes of swimming and foraging in the respective habitats and are likely results of phenotypic plasticity. However, differences in shape and interlandmark distances between the groups suggest that some (though few) morphometric characters might be useful for separating Phoxinus species.     

Evaluating Diversification Hypotheses in the South American Cricetid <Emphasis Type="Italic">Thaptomys nigrita</Emphasis> (Lichtenstein, 1829) (Rodentia: Sigmodontinae): An Appraisal of Geographical Variation Based on Different Character Systems

Molecular surveys using mtDNA sequences have been used to identify cryptic species in sigmodontine rodents. However, where sampling is uneven, a critical appraisal of further evidence is pivotal to test whether genetic discontinuity represents different species. In order to investigate geographical variation in the Neotropical rodent Thaptomys nigrita, we analyzed patterns of morphological variation in qualitative and morphometric data, and compared our results with recent information on karyological and molecular diversity reported for the monotypic genus. Two subtle morphometric groups of populations, corresponding to karyomorphs 2n = 50 and 2n = 52, were revealed, but no qualitative aspect of craniodental morphology unambiguously distinguished them. A positive and significant association between geographical and both morphological and genetic distances suggest that the distinction between the two groups of population follows an isolation by distance model. This result, coupled with phylogeographic and karyotypic breaks coincident to a sampling gap extending for 540 km, and with the low phylogenetic resolution of molecular clades, does not allow rejecting the hypothesis that the divergent samples constitute polymorphic populations of a widely distributed species. We discuss possible determinants of these patterns and emphasize the need for an integrative approach in future efforts to disclose the evolutionary relationships of small mammals in situations of uneven sampling.     

Sharing is caring? Measurement error and the issues arising from combining 3D morphometric datasets

Geometric morphometrics is routinely used in ecology and evolution and morphometric datasets are increasingly shared among researchers, allowing for more comprehensive studies and higher statistical power (as a consequence of increased sample size). However, sharing of morphometric data opens up the question of how much nonbiologically relevant variation (i.e., measurement error) is introduced in the resulting datasets and how this variation affects analyses. We perform a set of analyses based on an empirical 3D geometric morphometric dataset. In particular, we quantify the amount of error associated with combining data from multiple devices and digitized by multiple operators and test for the presence of bias. We also extend these analyses to a dataset obtained with a recently developed automated method, which does not require human‐digitized landmarks. Further, we analyze how measurement error affects estimates of phylogenetic signal and how its effect compares with the effect of phylogenetic uncertainty. We show that measurement error can be substantial when combining surface models produced by different devices and even more among landmarks digitized by different operators. We also document the presence of small, but significant, amounts of nonrandom error (i.e., bias). Measurement error is heavily reduced by excluding landmarks that are difficult to digitize. The automated method we tested had low levels of error, if used in combination with a procedure for dimensionality reduction. Estimates of phylogenetic signal can be more affected by measurement error than by phylogenetic uncertainty. Our results generally highlight the importance of landmark choice and the usefulness of estimating measurement error. Further, measurement error may limit comparisons of estimates of phylogenetic signal across studies if these have been performed using different devices or by different operators. Finally, we also show how widely held assumptions do not always hold true, particularly that measurement error affects inference more at a shallower phylogenetic scale and that automated methods perform worse than human digitization.     

Host‐associated genomic differentiation in congeneric butterflies: now you see it,now you do not

Ecotypic variation among populations may become associated with widespread genomic differentiation, but theory predicts that this should happen only under particular conditions of gene flow, selection and population size. In closely related species, we might expect the strength of host‐associated genomic differentiation (HAD) to be correlated with the degree of phenotypic differentiation in host‐adaptive traits. Using microsatellite and Amplified Fragment Length Polymorphism (AFLP) markers, and controlling for isolation by distance between populations, we sought HAD in two congeneric species of butterflies with different degrees of host plant specialization. Prior work on Euphydryas editha had shown strong interpopulation differentiation in host‐adapted traits, resulting in incipient reproductive isolation among host‐associated ecotypes. We show here that Euphydryas aurinia had much weaker host‐associated phenotypic differentiation. Contrary to our expectations, we detected HAD in Euphydryas aurinia, but not in E. editha. Even within an E. aurinia population that fed on both hosts, we found weak but significant sympatric HAD that persisted in samples taken 9 years apart. The finding of significantly stronger HAD in the system with less phenotypic differentiation may seem paradoxical. Our findings can be explained by multiple factors, ranging from differences in dispersal or effective population size, to spatial variation in genomic or phenotypic traits and to structure induced by past histories of host‐adapted populations. Other infrequently measured factors, such as differences in recombination rates, may also play a role. Our result adds to recent work as a further caution against assumptions of simple relationships between genomic and adaptive phenotypic differentiation.     

Morphometric analysis of inter‐ and intraspecific variation in the Cambrian helcionelloid mollusc Mackinnonia

Phylogenetic relationships within the helcionelloid molluscs have been difficult to establish. One of the reasons for this is that qualitative approaches to investigating morphological variation in this group have struggled to identify clear patterns. An alternative method of identifying these patterns is to study these organisms quantitatively. Here this approach is exemplified by employing morphometric methods to investigate patterns of subtle morphological variation in two species of Mackinnonia Runnegar in Bengtson et al. from Cambrian Series 2, Stages 3–4. Specifically, a combination of elliptical Fourier and multivariate analyses were conducted to study intra‐ and interspecific variation in protoconch form as well as variation in ontogenetic trajectory of the teleoconch of two species of Mackinnonia. The material used consists of two assemblages of Mackinnonia rostrata (Zhou & Xiao), from the Shackleton Limestone of Antarctica and Ajax Limestone of Australia, and an assemblage of Mackinnonia taconica (Landing & Bartowski) from the Bastion Formation of Greenland. Results of this study show significant (p < 0.0001) differences in protoconch shape between all three groups. Ontogenetic sequences of outline curves truncated at successive rugae significantly (p < 0.05) discriminate between M. rostrata and M. taconica. These techniques uncovered significant intraspecific morphological variation of disparate assemblages of M. rostrata despite shared qualitative features and structure a conceptual framework for understanding such patterns of variation and put this in the context of the incipient species concept.     

Hypervolume concepts in niche‐ and trait‐based ecology

Hutchinson's n‐dimensional hypervolume concept for the interpretation of niches as geometric shapes has provided a foundation for research across different fields of ecology and evolution. There is now an expanding set of applications for hypervolume concepts, as well as a growing set of statistical methods available to operationalize this concept with data. The concept has been applied to environmental, resource, functional trait, and morphometric axes and to different scales, i.e. from individuals, species, to communities and clades. Further, these shapes have been variously interpreted as niches, ecological or evolutionary strategy spaces, or proxies for community structure. This paper highlights these applications’ shared mathematical framework, surveys uses of the hypervolume concept across fields, discusses key limitations and assumptions of hypervolume concepts in general, provides a critical guide to available statistical estimation methods, and delineates the situations where hypervolume concepts can be useful.     

Differentiation at a microgeographical scale within two species of ground beetle, Carabus auronitens and C. nemoralis (Coleoptera, Carabidae): a geometrical morphometric approach

New morphometric methods, the geometrical morphometrics, offer promising perspectives to appraise morphological variation among organisms and open up, to a large extent, the field of morphometrics for the study of systematics and evolution. Until now, however, few studies have explored the potential of these methods at a microgeographical scale. In the present work, we applied them to quantify morphological (size and shape) differentiation among populations of two forest species of ground beetles: Carabus auronitens and C. nemoralis . We found a significant shape variation among sites, as well as among sexes, for both species. Additionally, for C. auronitens , we found significant positive correlations in both sexes between morphological (shape) and geographical distances between populations. In contrast, significant size differences were found between sexes, but not between sites. We conclude that geometrical morphometric methods provide valuable tools for the study of morphological variation among populations and therefore offer, on the whole, interesting perspectives for the study of biodiversity patterns.     

High degree of genetic differentiation in marine three‐spined sticklebacks (Gasterosteus aculeatus)

Populations of widespread marine organisms are typically characterized by a low degree of genetic differentiation in neutral genetic markers, but much less is known about differentiation in genes whose functional roles are associated with specific selection regimes. To uncover possible adaptive population divergence and heterogeneous genomic differentiation in marine three‐spined sticklebacks (Gasterosteus aculeatus), we used a candidate gene‐based genome‐scan approach to analyse variability in 138 microsatellite loci located within/close to (<6 kb) functionally important genes in samples collected from ten geographic locations. The degree of genetic differentiation in markers classified as neutral or under balancing selection—as determined with several outlier detection methods—was low (F_ST = 0.033 or 0.011, respectively), whereas average F_ST for directionally selected markers was significantly higher (F_ST = 0.097). Clustering analyses provided support for genomic and geographic heterogeneity in selection: six genetic clusters were identified based on allele frequency differences in the directionally selected loci, whereas four were identified with the neutral loci. Allelic variation in several loci exhibited significant associations with environmental variables, supporting the conjecture that temperature and salinity, but not optic conditions, are important drivers of adaptive divergence among populations. In general, these results suggest that in spite of the high degree of physical connectivity and gene flow as inferred from neutral marker genes, marine stickleback populations are strongly genetically structured in loci associated with functionally relevant genes.     

Morphological and molecular variation in the least madtom Noturus hildebrandi (Siluriformes: Ictaluridae), a Mississippi Embayment endemic: evidence for a cryptic lineage in the Hatchie River

Sheared principal component analysis of 40 morphometric characteristics measured for 146 individuals and relative frequencies of pigmentation patterns scored for 980 individuals of the least madtom Noturus hildebrandi, a diminutive catfish endemic to eastern lowland drainages of the Mississippi Embayment region of North America, suggested a clinal pattern of morphological variation extending across the range from north to south. DNA sequence data representing 90 individuals from the mitochondrial gene cytochrome b (cytb) analysed using Bayesian phylogenetic methods recovered four major haplotype clades, suggestive of a high degree of isolation by drainage. Individual gene trees of cytb and four additional nuclear loci as well as trees based on concatenated datasets of these genes consistently recovered a cryptic lineage of individuals from the Hatchie River drainage that is morphologically indistinguishable from surrounding populations. Gene‐tree analyses failed to recover a monophyletic N. hildebrandi with respect to Noturus baileyi. A coalescence‐based species tree analysis, however, did recover N. hildebrandi monophyly with high support, suggesting that relationships reflected in individual gene trees and concatenated datasets are in part artefacts of incomplete lineage sorting or an ancient introgressive event. Results are consistent with the hypothesis of an ancient connection between the Hatchie and Tennessee River systems. Current subspecific designations are of limited utility as they reflect morphological variation and are not entirely consistent with phylogeny. Discrepancies between the pattern of variation observed in the morphological and molecular data may be explained by recent local adaptation to individual stream conditions that masks deeper evolutionary divergences.     

Morphological and Genetic Divergences in a Coral Reef Damselfish, Pomacentrus coelestis

Population differentiation is one of the main topics in evolutionary biology. Except the exploration of color variation, few studies focused on morphological divergences among populations of coral reef fishes. In this work, we studied morphological and genetic differentiation among populations of the damselfish, Pomacentrus coelestis, in the northwestern Pacific Ocean. The shapes of the mandible and the premaxilla were explored using geometric morphometric methods and the genetic structure was investigated using microsattelites. Various tests revealed significant shape variation among most P. coelestis populations for both skeletal units. Morphological variation of the mandible accompanies a genetic break between populations of mainland Japan and Okinawa-Taiwan. However, Mantel and Procrustes tests revealed no congruence between morphological and genetic structures. We illustrate that phenotypic plasticity and adaptive divergence are potential evolutionary mechanisms underlying shape difference among P. coelestis populations. An ecomorphological approach suggests that various diet could be related to shape variation of oral jaws.