Comparison of the trees obtained using the different treatments of the morphological quantitative data and the tree obtained by the Bayesian analysis of the molecular data. ‘Morph data in four or more partitions’ refers to the homoplasy-based partitioning (Rosa et al. 2019). A Quartets in the morphological tree shared with the molecular tree (blue), which are different (orange), and which are unresolved (gray); the white line results from the Bayesian analysis without the 14 quantitative characters and was drawn to facilitate comparison. B Percentage of ‘accurately’ resolved quartets by the number of resolved quartets. The line is the best fit exponential regression calculated from the data. The colors represent the different treatments (i.e., blue, analyses without the 14 characters; red, analyses with the binary codings; orange, analyses with the additive coding; green, analyses with the non-additive multistate coding; purple, analyses with the 14 characters decomposed into binary sub-characters). — Abbreviations: BI, Bayesian inference; EW, equal weights parsimony; IW, implied weights parsimony.