The transfer of Dryandra R. Br. to Banksia L. f. was based on the use of holophyly (monophyly s. str.) as an essential criterion for recognition of taxa. The transfer was significant in scope and focuses on two iconic genera of plants in Western Australia. It has been accepted by some and rejected by others. It is one of many examples in a debate that pits recent genetic analysis against centuries of field and herbarium studies, and cladists against classical taxonomists. I argue that: (1) there are sound morphological characters distinguishing Dryandra from Banksia and they should be maintained as genera; (2) paraphyly should be accepted in biological classification; (3) scientifically, and for a morphologically complex genus of 137 specific and infraspecific taxa, the use of 11 taxa for the molecular analysis of Dryandra was insufficient; (4) some morphological data, mapped onto the cladogram a posteriori, were incorrect; (5) molecular cladistic approaches should complement rather than override pre-existing and extensive classifications based on phenotypic traits; (6) the acceptance of the transfer for the Australian Plant Census was premature according to guidelines published by Australian herbaria.