The how and why of branch support and partitioned branch support, with a new index to assess partition incongruence

An introduction and step-by-step guide to the calculation of branch support (BS) and partitioned branch support (PBS) is provided, along with a new summary measure of conflict among data partitions, the partition congruence index (PCI). Patterns of PCI vs. BS are compared for two data matrices from butterflies. PCI is a close mirror of BS unless BY is relatively low and there is a lot of conflict among data partitions, in which case PCI becomes negative. Some theoretical concerns relating to PBS are noted. The Willi Hennig Society 2006. There are a lot of phylogenetic trees around these days. Everything from cladograms to directly optimized alignments to phenograms based on distance algorithms are presented as ‘‘phylogenies’’ of this or that taxon. A question that most systematists would like to answer, at least in a heuristic way, is, ‘‘How good is my tree?’’ (e.g., Wenzel, 1997). Two aspects of clade quality that are frequently mentioned are support and stability. Support for a given branch in a tree is a measure of the net amount of evidence that favors the appearance of that branch in a most-parsimonious topology. Stability is the persistence of a given branch in the face of the addition, deletion or reweighting of characters, taxa, or both from the data matrix (Judd, 1998). Both of these attributes have proven to be particularly tricky to measure in a direct manner, due to the complexity of character interactions in homoplastic data (Goloboff and Farris, 2001). Nevertheless, having some measure of the stability of and support for particular groups in an inferred topology serves as a means to discern groups that are plausible from those that are dubious, and can act as a guide to the generation of additional data to refine and improve the hypothesis. There are numerous methods to infer the plausibility of particular branches in a phylogenetic tree, and the pros and cons of many of these have been discussed at length in the literature (e.g., Kluge and Wolf, 1993; Davis, 1995; Sanderson, 1995). The purpose of this paper is to provide an accessible review of branch support (BS) and partitioned BS (PBS), including a simple worked example based on a contrived data matrix. Towards the end of the paper, a new index of congruence among data partitions is discussed and illustrated with contrived and empirical data sets.