Simões et al. () argued that large matrices are linked to the construction of "problematic" characters, and that those characters negatively affect tree topology. In their re-evaluation of two squamate datasets, however, Simões et al. () simply eliminated what they termed "problematic" characters, rather than recode them. This practice ignores potential sources of phylogenetic information and, if it were to be more widely followed, would inhibit the advancement of the field of systematics. Here, we defend the necessity and inevitability of large morphological (phenomic) datasets and discuss best practices for morphological data collection in contemporary phylogenetics.
© The Willi Hennig Society 2017.