Controversy over the origins and evolution of social behaviour in the major groups of social bees (the corbiculate bees) has fuelled arguments over different approaches for building evolutionary trees. However, the application of different analytical methodologies does not explain why molecular and morphological data suggest strikingly different hypotheses for the evolution of eusociality in bees. Determining the phylogenetic root is expected to help resolve the question of the social evolution of corbiculate bees. However, this requires that the long branch attraction problem is overcome. This phenomenon affects both molecular and morphological data for corbiculate bees.