作者: Harold N. Bryant
DOI: 10.1007/978-1-4020-2330-9_17
关键词:
摘要: The construction of supertrees using matrix representation with parsimony (MRP) is equivalent operationally to the cladograms cladistic analysis character data. However, validity MRP as a phylogenetic method has been questioned because data used construct are topologies trees rather than consistency following principles evaluated: 1) only synapomorphies provide evidence for relationships, 2) ad hoc hypotheses be minimized in generation hypotheses, and 3) inference relationships must independent each other. To consistent these principles, based on source that were generated analyses data, weight input account relative support individual nodes eliminate inappropriate biases associated variation tree size, high indices, 4) relationships. Achieving criteria extremely difficult, all published fail meet one or more conditions. Although might justified pragmatic grounds, should considered heuristic synthesis hierarchical information, rigorous included taxa.