The debate on whether to combine different data sets for simultaneous analysis has continued to the present day unabated. We have studied the effects of combining one morphological data set with four molecular data sets (two mitochondrial gene sequences and two nuclear gene sequences) for a group of butterflies belonging to the tribe Nymphalini using partitioned Bremer support. We particularly focus our attention on a group of species belonging to the genera Aglais, Inachis, Roddia, Nymphalis, Kaniska, and Polygonia. We find that, despite significant incongruence between most data partitions, all data partitions contribute positively to the support of most nodes in the most parsimonious trees found for the combined data set. We also find that the morphological data set resolves one particular node (Kaniska basal to Polygonia) with good support, while the molecular data sets are ambiguous about the existence of this node. We suggest that partitioned Bremer support allows one to critically appraise the robustness of each node in a given tree and thereby identify nodes that may change with the addition of new data and nodes that are likely to remain unchanged with new data. We also suggest that morphological data are still crucial to our being able to understand the relationships of extant organisms, despite published views to the contrary. Based on our results we suggest that Inachis should be synonymized with Aglais, Roddia with Nymphalis, and Kaniska with Polygonia.