Abstract

Nuclear DNA ITS and ETS sequences of 71 representatives from nine genera and 11 sections of the core Laureae were combined with a matrix of morphological characters, analyzed using maximum parsimony with both equally and successively weighted characters, and analyzed for Bayesian inference, minimum evolution by neighbor joining, and maximum likelihood inference for molecular data alone. The large genera Actinodaphne Nees, Lindera Thunb., and Litsea Lam. were polyphyletic, as were Lindera sect. Aperula (Blume) Benth. and Litsea sections Conodaphne (Blume) Benth. & Hook. f., Cylicodaphne (Nees) Hook. f., and Tomingodaphne (Blume) Hook. f. In contrast, Neolitsea (Benth.) Merr. was monophyletic and terminal in a larger monophyletic lineage above an Actinodaphne grade. A major disparity exists between these molecular results and traditional morphology-based classifications within Lauraceae. These results suggest that the use of two- versus four-celled anthers for Laureae generic delimitation has resulted in polyphyletic or paraphyletic genera, and the character of dimerous versus trimerous flowers is of only limited phylogenetic value. Several of the major lineages in Laureae are supported by inflorescence morphology and ontogeny, with Laureae defined by short shoots with a vegetative terminal bud, splitting into thyrsoid (Actinodaphne and Neolitsea) versus racemose (Laurus, Litsea s. str., and Lindera s. str. and Lindera sect. Aperula), although there appear to be at least two different pathways to form the Laureae pseudo-umbel. Similarly, imbricate, early deciduous inflorescence basal involucral bracts defined an Actinodaphne–Neolitsea–Parasassafras–Sinosassafras clade, although within it, Neolitsea was defined in part by decussate, persistent bracts. Accordingly, our study indicates the need for caution in the use of morphology for assessing affinity in Laureae, as virtually all traditional morphological characteristics show high levels of homoplasy and/or reversal, but future research may help to resolve whether this indicates problems of homology or ontogenetic convergence.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call