Skip to main content
Dryad

Phylogenetic position of Centroglossa and Dunstervillea (Ornithocephalus clade: Oncidiinae: Orchidaceae) based on molecular and morphological data

Cite this dataset

Royer, Carla Adriane; Toscano de Brito, Antonio Luiz Vieira; Mauad, Anna Victoria; Smidt, Eric de Camargo (2022). Phylogenetic position of Centroglossa and Dunstervillea (Ornithocephalus clade: Oncidiinae: Orchidaceae) based on molecular and morphological data [Dataset]. Dryad. https://doi.org/10.5061/dryad.tmpg4f51j

Abstract

Even though the monophyly of the Ornithocephalus clade (Oncidiinae) is currently well defined, the systematic positioning of Centroglossa and Dunstervillea remains obscure in the clade due to the absence in previous phylogenetic studies. Centroglossa has a very similar habit and is indistinguishable from Zygostates, whereas Dunstervillea has as its main characteristic the calcarate labellum, also found in Centroglossa. We clarify the systematic and phylogenetic positioning of Centroglossa and Dunstervillea in the Ornithocephalus clade (OC) through analysis of maximum likelihood, Bayesian inference, and maximum parsimony from molecular data (nrITS and matK cpDNA) and morphology. Our results indicate that Dunstervillea is phylogenetically close to Eloyella; both genera have a psigmoid habit, single-sided and flattened leaves, floral perianth with the same coloring, petals with entire margins, and a short rostellum. Centroglossa appears as a subclade within Zygostates. In addition to several homoplastic features, these two genera have the dorsal position of viscidium as a synapomorphy. The calcarate labellum, common to Centroglossa and Dunstervillea, originated more than once in the OC. Based on the phylogenetic results, we propose the nomenclatural changes to include Dunstervillea in Eloyella and Centroglossa in Zygostates. Lectotypes are indicated to Centroglossa macroceras and C. glaziovii.

Methods

Leaf samples fresh or preserved in saturated NaCl-CTAB solution (Rogstad 1992) were obtained from collections in the regions of occurrence of the species and public and private collections. Unfortunately, some species were not found (such as Dunstervillea mirabilis), or we were not successful in amplifying and sequencing the studied regions. In these cases, we used data from GenBank (Appendix 1) or, if it is unavailable from this database, only its morphological data was used. For the morphological analysis, we included 54 species from Zygostates (23), Centroglossa (5), Dunstervillea mirabilis, and another 20 taxa of the OC. Five taxa of phylogenetically close clades inserted in the molecular analyses were used as the external group (Smidt et al. 2018). The morphological analysis was based on the protologues of the species, herbarium specimens deposited at the following herbaria: B, BR, BHCB, ESA, FUEL, FURB, GUA, HB, HEID, HUCP, HUEFS, HUPG, ICN, M, MBM, MBML, R, RB, SP, SPF, SPPS, UEC, UPCB, W (acronyms according to Thiers 2019, continuously updated), complemented with field-collected material data (Appendix 2), and the studies of Toscano de Brito (1994, 2001, 2007) and Royer et al. (2017a, b). We used characters found in vegetative (rhizome, stem, and leaf) and reproductive organs (inflorescence, flowers, sepals, petals, labellum, gynostemium, and pollinarium). The terminology used in the description of the characters and their respective states were extracted from Toscano de Brito (1994, 2001), Harris and Harris (1999), and Stearn (2004) (Table 2). The matrix (Appendix 3 and 4) was built with Mesquite program version 3.31 (Maddison and Maddison 2017). We selected 72 qualitative characters, being ten vegetative characters (nine binaries and one multistate), and 62 reproductive characters (38 binaries and 24 multistate), according to the following cumulative criteria: (1) their potential to distinguish taxa through studied genera; (2) the stability of their states between individuals of the same taxon; (3) the viability of the associated observations; and (4) independence from each other (Emerson and Hastings 1998; Sereno 2007). All characters were treated with equal weight and unordered states (Fitch's parsimony; Fitch 1971).

Funding

Coordenação de Aperfeicoamento de Pessoal de Nível Superior, Award: 001

National Council for Scientific and Technological Development, Award: 314642/2020-0