That competition is stronger among closely related species and leads to phylogenetic overdispersion is a common assumption in community ecology. However, tests of this assumption are rare and field-based experiments lacking. We tested the relationship between competition, the degree of relatedness, and overdispersion among plants experimentally and using a field survey in a native grassland. Relatedness did not affect competition, nor was competition associated with phylogenetic overdispersion. Further, there was only weak evidence for increased overdispersion at spatial scales where plants are likely to compete. These results challenge traditional theory, but are consistent with recent theories regarding the mechanisms of plant competition and its potential effect on phylogenetic structure. We suggest that specific conditions related to the form of competition and trait conservatism must be met for competition to cause phylogenetic overdispersion. Consequently, overdispersion as a result of competition is likely to be rare in natural communities.

No abstract

No abstract

The evolutionary history of the dracaenoid genera Dracaena and Sansevieria (Asparagaceae, Nolinoideae) remains poorly resolved, despite long-recognised issues with their generic boundaries and increased attention paid by both horticulturalists and taxonomists. In this study we aim to: (1) elucidate evolutionary relationships within and between Dracaena and Sansevieria using molecular phylogenetic inference of both nuclear (nDNA) and plastid (cpDNA) markers, (2) examine the infrageneric classifications of each genus, and (3) revise the circumscription of the dracaenoids in light of morphological and phylogenetic evidence. In total, we sampled 21 accessions of Dracaena (ca. 19 species), 27 accessions of Sansevieria (ca. 26 species), and six outgroup taxa. Phylogenetic analyses were based on nucleotide sequences of two non-coding plastid DNA regions, the trnL-F region (trnL intron and trnL-trnF intergenic spacer) and rps16 intron, and the low-copy nuclear region At103. Phylogenetic hypotheses were constructed using maximum parsimony, maximum likelihood, and Bayesian inference. Individual datasets were analysed separately and, after testing for congruence, as combined datasets. We recovered instances of soft incongruence between nDNA and cpDNA datasets in Sansevieria, but general trends in the dracaenoids were congruent, although often poorly supported or resolved. The dracaenoids constitute a strongly supported monophyletic group. Dracaena was resolved as a paraphyletic grade embedded with two clades of Sansevieria, a primary clade comprising most species, and a secondary clade including S. sambiranensis, a distinctive species from Madagascar. The backbone of our phylogeny was only resolved in nDNA analyses, but combined analyses recovered strongly supported species groups. None of the previous infrageneric classifications were supported by our phylogeny, and biogeographic groupings were frequently more significant than morphology. More work is needed to resolve internal relationships in the dracaenoids, but we support a recent proposal to recognise a broadened circumscription of Dracaena that includes Sansevieria. We provide a generic description for the recircumscribed Dracaena and new combinations for several species of Sansevieria in Dracaena.

Reconstructing the phylogenetic history of species in large taxonomic groups remains a challenge despite ever increasing access to molecular data. In subfamily Cercidoideae (Leguminosae), the large pantropical Bauhinia s.l. (~380 spp.) is never supported as monophyletic, but the generic boundaries and the number of genera that should be recognised remains a challenging taxonomic and phylogenetic question. Here we reconstruct the phylogeny of Bauhinia s.l. and related genera in the subfamily using sequences obtained from five loci (matK, trnL-F, Leafy, and two previously undetected duplicate copies of Legcyc), which we analyse individually and in a concatenated matrix (excluding Leafy). The individual and concatenated analyses generally support two distinct lineages in Bauhinia s.l.: (1) the Bauhinia clade, containing Bauhinia s.str., Brenierea, and Piliostigma; and (2) the Phanera clade, containing Barklya, Cheniella, Gigasiphon, Lasiobema, Lysiphyllum, Phanera s.str., Schnella, and Tylosema. Based on our analyses, we recognise 14 genera in Cercidoideae. We resurrect the genus Tournaya from Gigasiphon, and synonymise the Asian segregate genus Lasiobema back into Phanera s.str. Our analyses reveal a clear gene duplication event in CYCLOIDEA that is shared by all Cercidoideae excluding the sister lineage to the subfamily, Cercis, supporting recent hypotheses for a whole genome duplication in all Cercidoideae except this genus.

BackgroundPlukenetia is a small pantropical genus of lianas and vines with variably sized edible oil-rich seeds that presents an ideal system to investigate neotropical and pantropical diversification patterns and seed size evolution. We assessed the biogeography and seed evolution of Plukenetia through phylogenetic analyses of a 5069 character molecular dataset comprising five nuclear and two plastid markers for 86 terminals in subtribe Plukenetiinae (representing 20 of ~ 23 Plukenetia species). Two nuclear genes, KEA1 and TEB, were used for phylogenetic reconstruction for the first time. Our goals were: (1) produce a robust, time-dependent evolutionary framework for Plukenetia using BEAST; (2) reconstruct its biogeographical history with ancestral range estimation in BioGeoBEARS; (3) define seed size categories; (4) identify patterns of seed size evolution using ancestral state estimation; and (5) conduct regression analyses with putative drivers of seed size using the threshold model.ResultsPlukenetia was resolved into two major groups, which we refer to as the pinnately- and palmately-veined clades. Our analyses suggest Plukenetia originated in the Amazon or Atlantic Forest of Brazil during the Oligocene (28.7 Mya) and migrated/dispersed between those regions and Central America/Mexico throughout the Miocene. Trans-oceanic dispersals explain the pantropical distribution of Plukenetia, including from the Amazon to Africa in the Early Miocene (17.4 Mya), followed by Africa to Madagascar and Africa to Southeast Asia in the Late Miocene (9.4 Mya) and Pliocene (4.5 Mya), respectively. We infer a single origin of large seeds in the ancestor of Plukenetia. Seed size fits a Brownian motion model of trait evolution and is moderately to strongly associated with plant size, fruit type/dispersal syndrome, and seedling ecology. Biome shifts were not drivers of seed size, although there was a weak association with a transition to fire prone semi-arid savannas.ConclusionsThe major relationships among the species of Plukenetia are now well-resolved. Our biogeographical analyses support growing evidence that many pantropical distributions developed by periodic trans-oceanic dispersals throughout the Miocene and Pliocene. Selection on a combination of traits contributed to seed size variation, while movement between forest edge/light gap and canopy niches likely contributed to the seed size extremes in Plukenetia.Electronic supplementary materialThe online version of this article (10.1186/s12862-018-1308-9) contains supplementary material, which is available to authorized users.

Our well-resolved phylogeny supports most of the current species circumscriptions but not current generic circumscriptions. Additionally, our results are inconsistent with Iltis's hypothesis of species with elongated many-seed fruits giving rise to species with truncated few-seeded fruits. Instead, we find support for the reversion to elongated multiseeded fruits from a truncate few-seeded ancestor in Peritoma.

Abstract—We present a phylogenetic classification for Plukenetia (Euphorbiaceae, Acalyphoideae) based on morphology and molecular phylogenetic studies using nuclear (ETS, ITS, KEA1 introns 11 and 17, TEB exon 17) and plastid (matK, ndhF, psbA-trnH) DNA data. Plukenetia comprises 25 species divided into six sections, with three new sections and four new species described here. The circumscription of Plukenetia is unaltered from recent treatments and we continue to recognize Romanoa as distinct. The sections of Plukenetia correspond with the subclade system proposed by Cardinal-McTeague and Gillespie (2016): P1 = P. sect. Fragariopsis comb. et stat. nov.; P2 = P. sect. Penninerviae sect. nov.; P3 = P. sect. Plukenetia; P4 = P. sect. Angostylidium; and P5 = P. sect. Hedraiostylus + P. sect. Madagascarienses sect. nov. The sections are distinguished by a combination of leaf venation, staminate flower morphology, pistillate flower number, style morphology, fruit type, and seed size. Additionally, we describe three new species from South America belonging to sect. Penninerviae: Plukenetia brevistyla and Plukenetia megastyla from the Amazon basin and Plukenetia chocoensis from the Chocó Biogeographic Region of Colombia. The new Amazonian species are morphologically similar to P. brachybotrya but distinguished by their style shape and size. The new Colombian species is morphologically similar to P. penninervia but distinguished by its elongate basilaminar extrafloral nectaries, presence of abaxial laminar extrafloral nectaries, and longer inflorescences. We also describe a new species from sect. Plukenetia, Plukenetia sylvestris, which is found in central and southern Peru. This species is suggested to be the wild progenitor of the cultivated P. carolis-vegae, differing by its smaller seeds/fruits and fewer stamens. Molecular data, including a new ETS phylogeny sampling P. brevistyla, support our new taxa as distinct. Keys to the sections and species of Plukenetia are provided and we designate 12 new lectotypes for Plukenetia and Romanoa.