When molecules and morphology clash: reconciling conflicting phylogenies of the Metazoa by considering secondary character loss

Molecular and morphological data sets have yielded conflicting phylogenies for the Metazoa. So far, no general explanation for the existence of this conflict has been suggested. However, I believe that a neglected aspect of metazoan cladistics has introduced a systematic and substantial bias into morphological phylogenetic analyses. Most characters used for metazoan cladistics are coded as binary absence/presence characters. For most of these characters, the absence states are assumed to be uninformative default plesiomorphies, if they are defined at all. This character coding strategy could seriously underestimate the number of informative apomorphic absences or secondary character losses. Because nodes in morphological metazoan phylogenies are typically supported by relatively small numbers of characters each with a potentially strong impact on tree topology, failure to distinguish between primary absence and secondary loss of characters before a cladistic analysis may mislead morphological cladistics. This may falsely suggest conflict with molecular phylogenies, which are not sensitive to this bias. To test the existence of this bias, I compare the phylogenetic placement of a variety of metazoan taxa in molecular and morphological trees. In all instances investigated here, phylogenetic conflict can be resolved by allowing for secondary loss of morphological characters, which were assumed to be primitively absent in cladistic analyses. These findings suggest that we should be cautious in interpreting the results of morphological metazoan cladistic analyses and additionally illustrate the value of a more functional approach to comparative morphology in certain circumstances.     

The role of character loss in phylogenetic reconstruction as exemplified for the Annelida

Annelid relationships are controversial, and molecular and morphological analyses provide incongruent estimates. Character loss is identified as a major confounding factor for phylogenetic analyses based on morphological data. A direct approach and an indirect approach for the identification of character loss are discussed. Character loss can frequently be found within annelids and examples of the loss of typical annelid characters, like chaetae, nuchal organs, coelomic cavities and other features, are given. A loss of segmentation is suggested for Sipuncula and Echiura; both are supported as annelid ingroups in molecular phylogenetic analyses. Moreover, character loss can be caused by some modes of heterochronic evolution (paedomorphosis) and, as shown for orbiniid and arenicolid polychaetes, paedomorphic taxa might be misplaced in phylogenies derived from morphology. Different approaches for dealing with character loss in cladistic analyses are discussed. Application of asymmetrical character state transformation costs or usage of a dynamic homology framework represents promising approaches. Identifying character loss prior to a phylogenetic analysis will help to refine morphological data matrices and improve phylogenetic analyses of annelid relationships.     

Phylogenetic placement of the Tasmanian spider Acrobleps hygrophilus (Araneae, Anapidae) with comments on the evolution of the capture web in Araneoidea

This paper studies the family‐level phylogenetic placement of the conflicting Tasmanian spider genus Acrobleps using both morphological and behavioral data. We also provide a formal taxonomic revision of Acrobleps, including information on its web architecture and natural history, as well as detailed morphological information for A. hygrophilus, its only species. Acrobleps hygrophilus lacks the typical mysmenid features. Furthermore A. hygrophilus does have all typical and diagnostic characteristics of Anapidae, except for the labral spur. We also discuss two noteworthy morphological features of Acrobleps: the pore bearing depressions of the carapace and the granulated cuticle of the spinnerets. Variation in the latter feature might provide a useful phylogenetic character. Based on the results of cladistic analyses we propose the transfer of Acrobleps from the Mysmenidae to its original placement within the Anapidae. We also propose a new lineage, informally labeled as the “clawless female clade”, which includes synaphrids, cyatholipids and “symphytognathoids.” The secondary absence of the female palpal claw provides support for the “clawless female clade.” We discuss the evolution of the orb web within anapids and other symphytognathoids based on the results of our cladistic analyses. The identical bi‐dimensional webs of the anapid Elanapis and of symphytognathids have evolved independently. Finally, we comment on the implications of one of our analyses regarding araneoid web evolution. We conclude that the taxon sample included in the previous orbicularian data matrix (modified and used in this study) is adequate to test the phylogenetic placement of Acrobleps in Anapidae but insufficient to significantly assess web evolution within Araneoidea. © The Willi Hennig Society 2007.     

Isozyme criteria in the testing of phyletic relationships between species of Gobius and related eastern Atlantic-Mediterranean genera (Teleostei: Gobüdae)

Horizontal starch gel electrophoresis was employed to investigate allozyme variation at 19 isozyme loci, in nine closely-related Atlantic-Mediterranean gobiid fish species, to assess genetic relationships, and to compare this with a phyletic hypothesis based on morphological apo-morphies. The species examined were Gobius niger, G. auratus, G. cruentatus, G. paganellus, Mauligobius maderensis, Zosterisessor ophiocephalus, Thorogobius ephippiatus, Padogobius martensii and P. nigricans . Various phenetic and cladistic analyses were performed on isozyme and morphological data. The phenetic and cladistic results from morphological data, and the cladistic results from isozyme data, were largely comparable, with P. martensü and P. nigricans forming a sister group to all the other taxa, species of Cobius forming a crown group within the latter, and stem lines formed by Mauligobius, Zosterisessor and Thorogobius , respectively. The isozymic estimates of genetic distance differed from these chiefly in the relative positions of Zosterisessor and G. auratus , which appear less similar to other species of Gohius. G. auratus shows many alleles which are unique within the genus Gobius ; possible explanations for this are put forward. The position of nigricans as congeneric with Padogobius martensü is confirmed by both phenetic and cladistic analyses of isozyme data.     

Unleashing the force of cladistics? Metazoan phylogenetics and hypothesis testing

The accumulation of multiple phylogenetic hypotheses for theMetazoa invites an evaluation of current progress in the field.I discuss three case studies from the recent literature to assesshow cladistic analyses of metazoan morphology have contributedto our understanding of animal evolution. The first case studyon cleavage cross patterns examines whether a decade of unanimouscharacter scoring across different cladistic studies can beconsidered a reliable indicator of accumulated wisdom. The tworemaining case studies illustrate how the unique strength ofcladistic analyses to arbitrate between competing hypothesescan be crippled when insufficient attention is directed towardsthe construction of the data matrix. The second case study discussesa recent morphological cladistic analysis aimed at providinginsight into the evolution of larval ciliary bands (prototrochs)in the Spiralia, and the third case study evaluates how foursubsequent morphological cladistic analyses have contributedto our understanding of the phylogenetic placement of a problematicum,the Myzostomida. I conclude that current phylogenetic analysesof the Metazoa have not fully exploited the power of cladisticsto test available alternative hypotheses. If our goal is togenerate genuine progress in understanding rather than stochasticvariation of opinions through time, we have to shift our attentionfrom using cladistics as an easy tool to generate "novel" hypothesesof metazoan relationships, towards employing cladistics morecritically as an effective instrument to test the relative meritof available multiple alternative hypotheses.     

Molecular systematics of Psocomorpha (Psocoptera)

Abstract.  Previous classification of the insect order Psocoptera has relied on morphological characters. Psocoptera are generally divided into three suborders: Trogiomorpha, Troctomorpha, and Psocomorpha. Traditional classification divides the Psocomorpha into four infraorders (Homilopsocidea, Caeciliusetae, Psocetae and Epipsocetae), but a recent morphological cladistic study removed Archipsocidae from Homilopsocidea and Hemipsocidae from Psocetae. We investigated the phylogenetic relationships within the suborder Psocomorpha using DNA sequences from the nuclear 18S and mitochondrial 16S, 12S and cytochrome oxidase I genes. Phylogenetic analyses of these gene sequences supported monophyly for Psocomorpha. In addition, monophyly of the traditional subgroups Caeciliusetae and Psocetae was generally supported. Monophyly of Homilopsocidea was not supported, and Archipsocus is removed from this group. Although the molecular phylogeny is generally consistent with recent cladistic studies of morphological characters, we found no evidence that Hemipsocidae should be removed from Psocetae.     

A cladistic analysis of the cockroach wasps based on morphological data (Hymenoptera: Ampulicidae)

The Ampulicidae are one of the most basal groups within the apoid wasps, the paraphyletic assemblage of wasps previously known as Sphecidae. The morphology and taxonomy of the Ampulicidae are poorly studied, and there is no strict cladistic approach on their phylogeny yet. Here we assemble morphological characters of phylogenetic significance and submit them to parsimony analyses using modern cladistic methods. This produces a well-resolved and firmly supported phylogenetic hypothesis on the generic relationships within the group. Based on our phylogenetic results a revised classification is provided, subdividing the Ampulicidae into two monophyletic subfamilies, Ampulicinae ( Ampulex and Trirogma ) and Dolichurinae, the latter comprising two monophyletic tribes, Dolichurini ( Dolichurus and Paradolichurus ) and Aphelotomini, new tribe ( Aphelotoma and Riekefella ). Based on the resulting cladogram, the historical biogeography and the fossil record of Ampulicidae are briefly discussed.
© The Willi Hennig Society 2009.     

Morphological phylogenetics of theAsteraceae-Asteroideae,with notes on character evolution

The relationships within theAsteraceae-Asteroideae are scrutinized by means of cladistic analyses of morphological and chemical data. New data are presented, and novel interpretations of features previously used in cladistic analyses are made. Examples are insertion of anther filaments in the corolla, and morphology of disc floret styles, respectively. The results include a polyphyletic or paraphyleticInuleae s. l., and a paraphyleticHeliantheae s.l. in which theEupatorieae constantly form a subclade. Comparisons are made with previous cladistic analyses of morphological data, and those acquired from molecular data. Speculations about the evolution of certain organs are presented.     

The grand game of metazoan phylogeny: rules and strategies

Many cladistic analyses of animal phylogeny have been published by authors arguing that their results are well supported. Comparison of these analyses indicates that there can be as yet no general consensus about the evolution of the animal phyla. We show that the various cladistic studies published to date differ significantly in methods of character selection, character coding, scoring and weighting, ground-pattern reconstructions, and taxa selection. These methodological differences are seldom made explicit, which hinders comparison of different studies and makes it impossible to assess a particular phylogeny outside its own scope. The effects of these methodological differences must be considered before we can hope to reach a morphological reference framework needed for effective comparison and combination with the evidence obtained from molecular and developmental genetic studies.     

Morphological phylogeny of the Tegulinae (Mollusca: Vetigastropoda) reinforces a Turbinidae position

A cladistic analysis of the Tegulinae (Turbinidae) is presented using 132 morphological characters and 41 taxa. Tegulinae is recovered and is sister to Prisogaster niger (Prisogasterinae) within the family Turbinidae. This scenario, with Tegulinae as a subfamily within Turbinidae, corroborates with the most molecular analyses. Tegulinae comprises >40 extant species, belonging to eight genera. Morphological studies have not resolved the placement of Tegulinae within Trochoidea sufficiently, and the systematic positions of the genera have never been investigated as a primary objective. The present morphology-based analysis of genus-level relationships within Tegulinae provides a robust, phylogenetic diagnosis of each group, rooted on a firm hypothesis of evolutionary relationships. An additional search was performed to include the tegulines Omphalius nigerrimus and Carolesia blakei terminals using unweighted and implied weighting. Our morphological data provide a solid foundation for ensuing systematic research on Tegulinae, as well as Trochoidea, and evidence facilitating the diagnosis of generic and suprageneric groups.     

Systematic affinities of Rhizophoraceae and Anisophylleaceae, and intergeneric relationships within Rhizophoraceae, based on chloroplast DNA, nuclear ribosomal DNA, and morphology

A cladistic analysis of sequences from the chloroplast gene rbcL was used to determine the systematic affinities of Rhizophoraceae and Anisophylleaceae. This analysis rejects close relationships of Rhizophoraceae with Celastraceae or Elaeocarpaceae, suggested previously, and identifies Erythroxylaceae as sister group within the Malpighiales, supported by several morphological and anatomical characters. Our molecular results also indicate that Anisophylleaceae are nested within Cucurbitales. Although this placement is novel, this affinity is also well supported by shared morphological characters. Tribal and generic relationships within Rhizophoraceae are evaluated with a combination of six molecular data sets (rbcL, atpB-rbcL intergenic spacer, trnL-trnF intergenic spacer, ITS1, ITS2, and 5.8S) and a morphological data set. These relationships are compared with results from previous morphological cladistic analyses. Against the background of the molecular results, we briefly discuss the evolution of morphological characters traditionally used for tribal subdivision as well as characters presumably significant for adaptation to mangrove habitats, namely, aerial stilt roots and vivipary.     

Phylogenetic analysis of Macrobiotidae (Eutardigrada, Parachela): a combined morphological and molecular approach

Combined analyses of morphological and molecular data were used to resolve phylogenetic relationships within Macrobiotidae (Eutardigrada). Morphological data were analysed using a cladistic approach with a matrix comprising 15 taxa with 17 characters to obtain a phylogenetic reconstruction. Molecular data were obtained by sequencing the cytochrome c oxidase subunit I gene in seven species of Macrobiotidae and one of Eohypsibiidae (used as outgroup). The morphological character defining the family, symmetrical claw on each leg, turns out to be plesiomorphic. Moreover, neither morphological nor molecular analyses supports a monophyletic clade for the subfamily Macrobiotinae, whereas both support a well defined evolutionary line (Murrayinae) within the family. We propose elevating the latter to family level (Murrayidae) while temporarily retaining as valid the family Macrobiotidae (amending its diagnosis, including within it only Macrobiotinae). Murrayidae opens an interesting evolutionary prospective, because the entire line has differentiated without sexual reproduction, constituting an example of evolution of asexual lineages.     

Morphological Evidence for the Phylogeny of Cetacea

A cladistic analysis of 54 extant and extinct cetacean taxa scored for 304 morphological characters supports a monophyletic Odontoceti, Mysticeti, Autoceta, and Cetacea. Forcing a sister-group relationship between Mysticeti and Physeteridae, as suggested by some, but not all, molecular studies, requires an additional 72 steps. In agreement with recent molecular studies, morphological data divide extant mysticetes into two clades: Balaenopteroidea (Eschrichtiidae + Balaenopteridae) and Balaenoidea (Balaenidae + Neobalaenidae). Cetotheriopsinae is removed from Cetotheriidae, elevated to Family Cetotheriopsidae, and placed within the Superfamily Eomysticetoidea. All extant mysticetes and all cetotheriids are placed in a new Parvorder Balaenomorpha, which is diagnosed by many morphological characters, including fusion of the anterior and posterior processes of petrosal to ectotympanic bulla, pronounced median keel on palate, and absence of ventral margin of sigmoid process of bulla. Many of the clades within Odontoceti in the most parsimonious trees of this study are at odds with recent phylogenetic analyses. For example, Platanistidae is not closely related to the extinct odontocete families Squalodontidae and Squalodelphinidae. Instead, it is more closely related to extant river-dwelling odontocetes (i.e., Lipotes, Inia), suggesting a single dispersal of odontocetes into freshwater habitats. We found several characters to support Physeteroidea (Physeteridae + Ziphiidae), a taxon considered paraphyletic by several molecular and some morphological analyses. Lack of agreement on the phylogeny within Odontoceti indicates that additional analyses, which include molecular and anatomical data as well as extant and extinct taxa, are needed.     

Phylogenetic implications of corolla morphology in subfamily Barnadesioideae (Asteraceae)

The previously insignificant and small South American subtribe Barnadesiinae of tribe Mutisieae has been shown recently to be ancient within Asteraceae. Due to absence of a 22 kb diagnostic chloroplast inversion, plus other features, this subtribe has been elevated to subfamilial status as Barnadesioideae, now containing 90 species in nine genera. Recent cladistic analyses at the generic and specific levels based on morphology and DNA sequences (ITS and trnL intron) have revealed different relationships among taxa of the subfamily. To better understand these conflicts, we analyze specific morphological features of corollas (shape and vascularization) to develop trends (trees) in these features based on minimal structural change (i.e., morphological parsimony), and to compare these with relationships among genera derived from the two recent cladistic analyses. We define six principal types of corollas in Barnadesioideae (tubular, split, double split, ligulate, subbilabiate and bilabiate) and six principal types of corolla vascularization patterns (combinations of presence or absence of central bundles, fusion of adjacent bundles and fusion of bundles at the apex of corolla lobes). In all features we assume character states in Calyceraceae to be ancestral. In corolla shape, from tubular ancestry, we hypothesize two general evolutionary trends within the subfamily: (1) splitting of the tube, and (2) flattening of the tube. In vascularization of corollas, from an ancestral condition of one central and one lateral bundle in each corolla lobe and all traces fused at the apex, we hypothesize four basic trends: (1) gain or (2) loss of lateral vascular traces in each lobe, (3) loss of the central vein, and (4) loss of fusion of traces in lobe apices. These morphological trends allow tests of the two previous phylogenetic hypotheses by (1) counting step changes (following steps in the morphological network) in the two characters on the two competing phylogenetic trees and (2) constructing an index of morphological advancement for each genus (based on morphological trees of the two characters) and correlating these with cladistic distances within the two phylogenies. Results of both tests reveal the molecular phylogeny to be more compatible with evolutionary inferences from the two morphological features. With Fulcaldea excluded, a highly significant correlation is seen between morphological advancement and cladistic distance in the molecular phylogeny.     

Morphology and angiosperm systematics in the molecular era

Several ways in which morphology is used in systematic and evolutionary research in angiosperms are shown and illustrated with examples: 1) searches for special structural similarities, which can be used to find hints for hitherto unrecognized relationships in groups with unresolved phylogenetic position; 2) cladistic studies based on morphology and combined morphological and molecular analyses; 3) comparative morphological studies in new, morphologically puzzling clades derived from molecular studies; 4) studies of morphological character evolution, unusual evolutionary directions, and evolutionary lability based on molecular studies; and 5) studies of organ evolution. Conclusions: Goals of comparative morphology have shifted in the present molecular era. Morphology no longer plays the primary role in phylogenetic studies. However, new opportunities for morphology are opening up that were not present in the premolecular era: 1) phylogenetic studies with combined molecular and morphological analyses; 2) reconstruction of the evolution of morphological features based on molecularly derived cladograms; 3) refined analysis of morphological features induced by inconsistencies of previous molecular and molecular phylogenetic analyses; 4) better understanding of morphological features by judgment in a wider biological context; 5) increased potential for including fossils in morphological analyses; and 6) exploration of the evolution of morphological traits by integration of comparative structural and molecular developmental genetic aspects (Evo-Devo); this field is still in its infancy in botany; its advancement is one of the major goals of evolutionary botany.     

Giant taxon‐character matrices: quality of character constructions remains critical regardless of size

Giant morphological data matrices are increasingly common in cladistic analyses of vertebrate phylogeny, reporting numbers of characters never seen or expected before. However, the concern for size is usually not followed by an equivalent, if any, concern for character construction/selection criteria. Therefore, the question of whether quantity parallels quality for such influential works remains open. Here, we provide the largest compilation known to us of character construction methods and criteria, as derived from previous studies, and from our own de novo conceptualizations. Problematic character constructions inhibit the capacity of phylogenetic analyses to recover meaningful homology hypotheses and thus accurate clade structures. Upon a revision of two of the currently largest morphological datasets used to test squamate phylogeny, more than one‐third of the almost 1000 characters analysed were classified within at least one of our categories of “types” of characters that should be avoided in cladistic investigations. These characters were removed or recoded, and the data matrices re‐analysed, resulting in substantial changes in the sister group relationships for squamates, as compared to the original studies. Our results urge caution against certain types of character choices and constructions, also providing a methodological basis upon which problematic characters might be avoided.     

Phylogenetic analysis of the African papionin basicranium using 3‐D geometric morphometrics: The need for improved methods to account for allometric effects

The basicranium has been argued to contain a strong phylogenetic signal in previous analyses of primate cranial morphology. Therefore, further study of basicranial morphology may offer new insights into controversial phylogenetic relationships within primate groups. In this study, I apply 3‐D geometric morphometric techniques in a phylogenetic analysis of the African papionin basicranium. The effects of allometry strongly influence African papionin basicranial morphology and, unless these size effects are controlled or eliminated, phylogenetic analyses suggest traditional phylogenetic groupings of small taxa (mangabeys) and large taxa (geladas, mandrills, drills, and baboons). When the effects of allometry are eliminated by excluding size‐correlated principal components (PCs) or by regression analysis with retention of residuals, phylogenetic analyses of African papionin basicranial morphology are incongruent with recent molecular and morphological studies. By contrast, a cladistic analysis of basicranial characters using the narrow allometric coding method suggests the same phylogenetic relationships as recent molecular and morphological studies. These results suggest that important phylogenetic information is contained within the size‐correlated data, and this information is being discarded during the attempt to eliminate the effects of body size. Future 3‐D morphometric studies of phylogeny should focus on the development of new methodologies to adjust for allometric effects, as current techniques appear to be ill‐equipped to deal with the case of a size‐disparate, lower‐level taxonomic group. Am J Phys Anthropol, 2010. © 2010 Wiley‐Liss, Inc.     

Preliminary morphological analysis of relationships between the spider wasp subfamilies (Hymenoptera: Pompilidae): revisiting an old problem

No qualitative cladistic analysis has been performed previously for the subfamily classification of Pompilidae (Hymenoptera). In 1994 Shimizu proposed six subfamilies, but their validity and relationships remain inconclusive. The objective of this study was to perform a quantitative analysis of phylogenetic relationships of the Pompilidae, with emphasis on testing the validity of proposed subfamilies. Two cladistic analyses were performed based on morphological evidence. First, a maximum-parsimony analysis of Shimizu's original morphological data matrix (72 taxa by 54 characters) was conducted, with the data subjected to a heuristic search for the first time with phylogenetic software. The resulting strict-consensus cladogram yielded a monophyletic Ceropalinae that was sister group to a large polytomy containing members of the remaining five subfamilies. In a second analysis, several of Shimizu's characters were re-examined, and new characters and more taxa were added to the data set. Terminal taxa were coded as species rather than as generic abstractions, and 20 additional morphological characters were introduced. The analysis was based on 77 morphological characters derived from the adults of 84 taxa. This second analysis suggested that Notocyphinae sensu Shimizu (1994) was nested within Pompilinae and that Epipompilinae sensu Shimizu (1994) was nested within Ctenocerinae; neither should retain their status as a separate subfamily. Lastly, Chirodamus s .s., which historically has been a member of the Pepsinae, is placed within the Pompilinae with reservations rather than erecting a new subfamily. After these allowances were made, a strict consensus tree gave the following relationships: (Ceropalinae + (Pepsinae + (Ctenocerinae + Pompilinae))).     

A phylogenetic supertree of the bats (Mammalia: Chiroptera)

We present the first estimate of the phylogenetic relationships among all 916 extant and nine recently extinct species of bats Mammalia: Chiroptera), a group that accounts for almost one-quarter of extant mammalian diversity. This phylogeny was derived by combining 105 estimates of bat phylogenetic relationships published since 1970 using the supertree construction technique of Matrix Representation with Parsimony (MRP). Despite the explosive growth in the number of phylogenetic studies of bats since 1990, phylogenetic relationships in the order have been studied non-randomly. For example, over one-third of all bat systematic studies to date have locused on relationships within Phyllostomidae, whereas relationships within clades such as Kerivoulinae and Murinae have never been studied using cladistic methods. Resolution in the supertree similarly differs among clades: overall resolution is poor (46.4%, of a fully bifurcating solution) but reaches 100% in some groups (e.g. relationships within Mormoopidae). The supertree analysis does not support a recent proposal that Microchiroptera is paraphyletic with respect to Megachiroptera, as the majority of source topologies support microbat monophyly. Although it is not a substitute for comprehensive phylogenetic analyses of primary molecular and morphological data, the bat supertree provides a useful tool for future phylogenetic comparative and macroevolutionary studies. Additionally, it identifies clades that have been little studied, highlights groups within which relationships are controversial, and like all phylogenetic studies, provides preliminary hypotheses that can form starting points for future phylogenetic studies of bats.     

Phenetic and cladistic studies help clarify species assemblages in <Emphasis Type="Italic">Passiflora</Emphasis> section <Emphasis Type="Italic">Dysosmia</Emphasis> (Passifloraceae)

One of the most enigmatic and unusual groups in the passionflower genus, Passiflora section Dysosmia (Passifloraceae), stands out as a group that is notoriously taxonomically complicated. Phenetic and cladistic analyses of Dysosmia were carried out with representatives from all 21 species currently recognized in the section, in order to delineate morphological groups above the species level. The study was based mainly on vegetative morphological characters and included principal coordinates, non-metric multidimensional scaling, cluster, and cladistic analyses. Results from each analysis reveals that three major species assemblages exist within Dysosmia, corresponding largely to vegetative pubescence and fruit color. The results presented here set the stage for future systematic and phylogenetic studies in Passiflora sect. Dysosmia.