Towards a phylogeny of the Metazoa: evaluating alternative phylogenetic positions of Platyhelminthes, Nemertea, and Gnathostomulida, with a critical reappraisal of cladistic characters

This paper critically assesses all morphological cladistic analyses of the Metazoa that were published during the last one and a half decades. Molecular and total evidence analyses are also critically reviewed. This study focuses on evaluating alternative phylogenetic positions of the ‘acoelomate’ worms: Platyhelminthes, Nemertea, and Gnathostomulida. This paper consists of two parts. In Part I, all recently proposed sister group hypotheses and the supporting synapomorphies for these phyla are evaluated. Discrepancies in the treatment of corresponding characters in different cladistic analyses are identified, and where possible, resolved. In Part II, the overall phylogenetic significance across the Metazoa of all characters relevant for placing the ‘acoelomate’ worms is examined. The coding and scoring of these characters for other phyla are evaluated, and uncertainties in our understanding are pointed out in order to guide future research. The characters discussed in this paper are broadly categorized as follows: epidermis and cuticle, reproduction and sexual condition, development, larval forms, coeloms and mesoderm source, nervous system and sensory organs, nephridia, musculature, digestive system, and miscellaneous characters. Competing phylogenetic hypotheses are compared in terms of several criteria: 1) taxon sampling and the fulfillment of domain of definition for each character; 2) character sampling; 3) character coding; 4) character scoring and quality of primary homology; 5) quality of the proposed diagnostic synapomorphies as secondary homologies. On the basis of this study I conclude that a sister group for the Platyhelminthes has not yet been unambiguously established. A clade minimally composed of Neotrochozoa (Mollusca, Sipuncula, Echiura, Annelida) emerges as the most likely sister group of the Nemertea on the basis of morphological and total evidence analyses. Finally, morphological data currrently favor a sister group relationship of Gnathostomulida and Syndermata (probably plus Micrognathozoa). In contrast, molecular or total evidence analyses have not identified a reliable sister group of Gnathostomulida. Further progress in our understanding of metazoan phylogeny crucially depends on the improvement of the quality of currently adopted cladistic data matrices. A thorough reassessment of many of the more than 70 morphological characters discussed here is necessary. Despite the recent compilation of comprehensive data matrices, the power to test competing hypotheses of higher-level metazoan relationships is critically compromised due to uncritical data selection and poor character study in even the most recently published cladistic analyses.