Recently, there have been a lot of intense debates about the acceptance/rejection of paraphyletic groups in biological classification. On the one hand, evolutionary classification states that similarity and common descent are two criteria for biological classification and paraphyletic groups are natural units of biological classification. On the other hand, cladistic classification considers that common descent is the only criterion in biological classification and monophyly should be strictly adhered to. Holcoglossum is used herein as a case to illustrate this problem. Although Holcoglossum is a small orchid genus of less than 20 species, there is little consensus about its generic circumscription since it was established, which leads to confusion in taxonomic treatments in the Aerides-Vanda group. Based on the analyses of molecular and morphological evidence, our results suggest that the clade comprising Holcoglossum s.s., Ascolabium, Penkimia and Ascocentrum himalaicum is strongly supported as a monophyly, and that the three taxa are nested within different subclades of Holcoglossum s.s. Thus, it is reasonable to recognize a monophyletic circumscription of Holcoglossum, which is also well supported by some vegetative and floral characters. The Holcoglossum s.l. would facilitate a better understanding of pollinator-driven floral divergence and vegetative stasis than a paraphyletic and narrowly defined genus.