UvA-DARE (Digital Academic Repository) Macroevolution of Animal Body plans Jenner, R.A. Publication date 2002 Link to publication Citation for published version (APA): Jenner, R. A. (2002). Macroevolution of Animal Body plans. Grafische Producties, UvA. General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible. UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl) Download date:30 Sep 2021 Chapterr 7 Bilateriann Phylogeny and Uncritical Recyclingg of Morphological Data Sets s Ronaldd A. Jenner Publishedd in Systematic Biology 50: 730-742 (2001) 141 1 142 2 SjB-f.. BW. 5«5>:73O-7«.20Ol Bilateriann Phylogeny and Uncritical Recycling of Morphological Dataa Sets RONALDD A. JENNER InstituteInstitute for Biodiversity and Ecosystem Dynamics, University of Amsterdam. P.O. Box 94766,1090GT, Amsterdam, TheThe Netlierlands; E-mail: jcnner9science.uva.nl DATAA QUALITY, DATA RECYCLING, AND clusivelyy based on the morphological data of DATAA MATRIX COMPILATION Zrzavyy et al. (1998); and Sorensen et al. (2000),, adopting a modified version of the Thee vitality of systematic biology as a sci- dataa set of Nielsen et al. (1996). Indeed, encee depends on our invested efforts to en- observingg that previously compiled data suree the quality of the input data of our anal- setss are recognized as valuable sources of yses.. When insufficient attention is directed phylogeneticc information is encouraging towardss the construction of the data matrix, andd could signify the continual refinement properr interpretation of the results of cladis- andd consolidation of an increasingly ma- ticc analyses becomes impossible. Neverthe- turingg morphological data set. However, less,, some systematists have argued that severall critical comments are in order. modemm systematics is increasingly charac- Inevitably,, any morphological phyloge- terizedd by a shift of emphasis from observa- neticc analysis of higher-level animal taxa tionn as the source of the data matrix to what hass to rely largely on information assembled cann be inferred from the data matrix after its fromm the published literature. One should be constructionn (e.g., see Patterson and Johnson, cautious,, however, in extracting character in- 1997;; Grande and Bemis, 1998; Rieppel and formationn from the literature, especially from Zaher,, 2000). The current availability of an dataa sets previously compiled by different unprecedentedd amount of comparative data authors.. Yet, the recent phylogenetic analy- andd ever-increasing computing powers may sess listed above did not always attempt to havee largely fostered such a shift. explicitlyy and critically evaluate or reevalu- Itt may therefore not be unexpected to atee the recycled data matrices. Of course, a discernn a tendency among recent phylo- practicall problem is apparent. The vastness geneticc studies of higher-level animal re- off the ever-expanding field of invertebrate lationshipss (both morphological and total zoologyy makes it increasingly difficult and evidencee approaches) to use morphologi- timee consuming to obtain a complete first- call characters garnered from character lists handd overview and appreciation of compar- previouslyy compiled by different authors as ativee data for higher-level phylogenetic stud- raww data for new phylogenetic studies. A iess of the Metazoa. Nevertheless, insufficient partiall list might include Eemisse et al. attentionn to the quality of the data set may (1992),, strongly relying on Brusca and Brusca stronglyy impair the quality of the resulting (1990)) and Meglitsch and Schram (1991); phylogenies. Kimm et al. (1996), relying principally on Thee morphological matrix of Zrzavy et al. dataa from Brusca and Brusca (1990), Eemisse (1998;; ZEA) is especially important, compris- ett al. (1992), and Wheeler et al. (1993); Sund- ingg 276 characters and representing the most bergg et al. (1998), relying on Brusca and comprehensivee morphological data matrix Bruscaa (1990); Zrzav^ et al. (1998), chiefly forr the Metazoa compiled to date. Their mor- relyingg on data derived from Meglitsch phologicall data set has been incorporated andd Schram (1991), Schram (1991), Eemisse intoo several new phylogenetic studies of the ett al. (1992), Backeljau et al. (1993), Schram Metazoaa (Giribet, 1999; Giribet et al. 2000), andd Ellis (1994), Nielsen (1995), Rouse and andd the phylogenetic results of this latter Fauchaldd (1995), Nielsen et al. (1996), Ax often-citedd study have already been used as (1996),, Haszprunar (1996a, b), Wallace et al. aa framework for interpreting different as- (1996),, and Gilbert and Raunio (1997); Giribet pectss of animal evolution (de Queiroz, 1999; (1999)) and Giribet et al. (2000), which are ex- Vermeijj and Lindberg, 2000). Determining 143 3 thee quality of this matrix thus becomes criti- att integration or quality assessment of pri- cal.. Below I will demonstrate legitimate rea- maryy homologies hinders the discrimina- sonss for questioning the quality of this ma- tionn of valuable and problematic data. The trixx and show that its uncritical recycling se- matrixx is chiefly based on syntheses, espe- riouslyy compromises several important re- ciallyy other phylogenetic analyses and text- sultss from the latest phylogenetic analysis of bookss (see listing above). It is impossible to bilateriann relationships (Giribet et al., 2000; tracee the exact source of character informa- GEA),, recently published in this journal. I tionn in the matrix, and none of the charac- willl refer to particular characters from the terr definitions are discussed. ZEA (p. 251) matrixx of ZEA as ZX, where X is the number write,, "The original data were not reanalyzed off the character in the matrix of ZEA. priorr to analysis...." This is a problem be- Insufficientt attention to data matrix con- causee various authors have shown that sev- structionn and uncritical use of data matrices sera ll of the sources of the matrix suffer from off other authors is especially problematic for aa range of difficulties (e.g., see Rouse and phylogeneticc analyses of higher-level taxa. Fauchald,, 1995; Nielsen et al., 1996; Jenner Browerr (2000:15) writes about "characters as andd Schram, 1999). Nevertheless, ZEA's ma- observations"" to highlight the link between trixx was adopted without change by Giribet dataa matrix entry and its empirical support in (1999)) and GEA. phylogeneticc analyses. However, for phylo- Givenn the impossibility of assessing the geneticc studies of higher-level taxa, to equate qualityy of a morphological matrix by ap- characterss with observations is an oversim- pealingg to general principles, I will offer a plification.. Observations on characters are fil- broadd range of examples for diverse taxa and teredd through multiple layers, often implicit, characters.. Nevertheless, I will propose sev- off interpretation in their transformation to a erall general and related categories of prob- definitivee data matrix entry. The scoring of lemss associated with this large morpholog- aa single 0 or 1 for a given supraspecific ter- icall data matrix, the recognition of which minall taxon in the matrix may encapsulate willl be essential for a proper evaluation of aa variety of information and interpretation, thee phylogenetic results of ZEA and GEA. includingg synopses of the extent and impor- II will specifically illustrate the problems in- tancee of character variation within terminal volvedd by briefly discussing several of the taxa,, the internal phylogenetic relationships majorr conclusions from the recent study withinn the terminals along with their bearing off GEA. onn the reconstructed ground patterns that are scoredd in the matrix, the density and relia- bilityy of original observations for the termi- Theory-DrivenTheory-Driven Scoring Without nall units (e.g., recent or old literature), the EmpiricalEmpirical Support adoptedd strategy for character selection, and Thesee problems refer to data matrix en- thee many difficult decisions of character cod- triess that are largely or wholly determined ingg of complex features that may be function- byy a priori assumptions of character evolu- allyy and structurally linked. Recycling of any tion.. For example, among the taxa scored morphologicall data matrix for use in a new presentt for Z93 (tripartite body and coelom) studyy would supposedly logically imply a aree the chordates. The trimeric, archimeric, fulll agreement with all the decisions that fed orr oligomeric organization of the ances- intoo construction of the matrix. When these trall deuterostome has played an important issuess are not explicitly confronted, however, rolee in hypotheses about the evolution of thee data matrix is reduced to a black box. If thee chordate body plan (e.g., Nielsen, 1995; ourr goal is