Taxonomic Sampling, Phylogenetic Accuracy, and Investigator Bias
of authors use several different approaches to examine the effects of taxonomic sam-pling on phylogenetic analysis. This topic is receiving increasing attention, in part because recent studies have reached a con-fusing diversity of conclusions about the effects of taxonomic sampling. For in-stance, contrast the conclusions reached in two recent papers on this topic: If the evolutionary question of interest does not re-quire a large number of taxa, it seems best to use fewer taxa because larger trees are more likely to contain inconsistent branches. (Kim, 1996:372) Including large numbers of taxa in an analysis may be the best way to ensure phylogenetic accuracy. (H illis, 1996:131) These recommendations, taken at face val-ue, appear to be in direct con ¯ ict w ith re-gard to advice on taxon sampling. The pa-pers in this issue extend these studies and modify these recommendations on the ba-sis of analyses of real data sets (Soltis et al., 1998; Poe, 1998), simulations (Graybeal, 1998), and theoretical considerations (Kim, 1998). One conclusion from reading these papers is that whether increased taxonom-ic sampling helps or hinders the process of accurate phylogenetic estimation depends to a great extent on how accuracy is eval-uated and what is meant by ``taxonomic sampling. ’ ’