Poor taxon sampling, poor character sampling, and non-repeatable analyses of a contrived dataset do not provide a more credible estimate of insect phylogeny: a reply to Kjer

The wealth of data available for phylogenetic analysis of the insect orders, from both morphological and molecular sources, is steadily increasing. However, controversy exists among the methodologies one can use to reconstruct ordinal relationships. Recently, Kjer (2004) presented an analysis of insect ordinal relationships based exclusively on a single source of information: 18S rDNA sequence data. Kjer claims that his analysis resulted in a more "credible" phylogeny for the insect orders and strongly criticized our previous phylogenetic results. However, Kjer only used a subset of the data that are currently available for insect ordinal phylogeny, misrepresented our analyses, and omitted other analyses we have published on insect ordinal phylogeny. In our estimation, Kjer did a poor job of representing the current state of affairs in insect ordinal phylogenetics. Furthermore, we examine a number of analytical issues that are relevant not only for insect phylogeny, but systematics as a science, such as: repeatability and objectivity, locating alignment boundaries, secondary structure, goodness of fit measure, epistemological coherence, practicality and homology. (c) The Willi Hennig Society 2005.