Ecologists frequently use a supertree method to generate phylogenies in ecological studies. However, the robustness of research results based on phylogenies generated with a supertree method has not been well evaluated. Here, we use the angiosperm tree flora of North America as a model system to test the robustness of phylogenies generated with a supertree method for studies on the relationship between phylogenetic properties and environment, by comparing the relationship between phylogenetic metrics and environmental variables derived from a phylogeny reconstructed with a supertree method to that derived from a phylogeny resolved at species level. North America was divided into equal area quadrats of 12 100 km². Nine indices of phylogenetic structure were calculated for angiosperm tree assemblages in each quadrat using two phylogenies resolved at different levels (one resolved at the family level and the other resolved at the species level). Scores of phylogenetic indices were related to two major climatic variables (temperature and precipitation) using correlation and regression analyses. Scores of phylogenetic indices resulting from the two phylogenies are perfectly or nearly perfectly correlated. On average, there is no difference in the variation explained by the two climatic variables between scores of phylogenetic indices derived from the two phylogenies. Our study suggests that a phylogeny derived from a well resolved family-level supertree as backbone with genera and species attached to the backbone as polytomies is robust for studies investigating the relationship between phylogenetic structure and environment in biological assemblages at a broad spatial scale.

Rubia L. is the type genus of the coffee family Rubiaceae and the third largest genus in the tribe Rubieae, comprising ca. 80 species restricted to the Old World. China is an important diversity center for Rubia, where approximately half of its species occur. However, its internal phylogenetic relationships are still poorly understood. The objective of the present study is to contribute to the phylogenetic relationships within Rubia, using the nuclear internal transcribed spacer and six plastid markers and focusing on species from China. Twenty-seven species of Rubia were sampled to infer their phylogeny using Maximum parsimony, Maximum likelihood, and Bayesian analyses. The monophyly of Rubia is supported, provided that R. rezniczenkoana Litv. is excluded from Rubia and transferred to Galium as a new combination: G. rezniczenkoanum (Litv.) L. E Yang & Z. L. Nie. Within Rubia, two clades are clearly supported. They correspond to the traditional sect. Rubias.l. (A) and sect. Oligoneura Pojark. (B), and are morphologically mainly separable by their pinnate (A) versus palmate (B) leaf venation. Plesiomorphic features are the pinnate leaf veining in sect. Rubia s.l. and the occurrence of some species with opposite leaves and true stipules in sect. Oligoneura. In sect. Oligoneura one can assume an evolution from species with opposite leaves and true stipules (as in the R. siamensis Craib group) to those with whorls of two leaves and two leaf-like stipules (as in ser. Chinenses and the R. mandersii Collett & Hemsl. group) and finally with whorls of 6 or even more elements (as in ser. Cordifoliae). The correlation between clades recognized by DNA analyses and available differential morphological features is partly only loose, particularly in the group of R. cordifolia with 2×, 4×, and 6× cytotypes. This may be due to rapid evolutionary divergence and/or hybridization and allopolyploidy.

A close relationship between Androsace and related genera (Douglasia, Vitaliana, and Pomatosace) has long been recognized. Recent molecular studies have provided abundant evidence that Douglasia, Vitaliana, and Pomatosace are nested within Androsace and together constitute the monophyletic “Androsace group”. We investigated the pollen morphology of 80 taxa representing all sections of Androsace s.s. as traditionally construed, as well as Douglasia, Vitaliana, and Pomatosace,to see whether they are congruent with phylogenetic relationships. We uncovered subtle variation in pollen morphology within the group. The shape of pollen grains ranges from spheroidal to perprolate. Pollen size ranges from 9.37 µm in Androsace sect. Samuelia to 20.68 µm in Douglasia. Exine ornamentation includes microreticulate, microechinate, perforate, and rugulate types. The polar view varies from circular, triangular and planaperturate, to triangular and angulaperturate. Various pollen morphological characters support the monophyly of major clades, including /Septentrionalis, /Pomatosace, /Orthocaulon, and /Megista, which were recognized previously based on molecular evidence.

The southwestern Iberian Peninsula is an important biogeographic region, showing high biodiversity levels and hosting several putative glacial refugia for European flora. Here, we study the genetic diversity and structure of the Mediterranean, thermophilous plant Cheirolophus sempervirens (Asteraceae) across its whole distribution range in SW Iberia, as a tool to disentangle some of the general biogeographic patterns shaping this southern refugia hotspot. Null genetic diversity was observed in the cpDNA sequencing screening. Nonetheless, AFLP data revealed high levels of among-population genetic differentiation correlated to their geographic location. Our results suggest longer species persistence in southern Iberian refugia during glacial periods and subsequent founder effects northwards due to colonizations in warmer stages (i.e., the southern richness to northern purity pattern). Additionally, our phylogeographic analyses indicate the presence of two separate genetic lineages within Ch. sempervirens, supporting the hypothesis of multiple minor refugia for SW Iberia in agreement with the refugia within refugia model.

The tribe Iphigenieae (Colchicaceace, Liliales) includes two genera, viz. Camptorrhiza and Iphigenia, which are distributed in Africa, India, and Australasia. Iphigeniais represented by 12 species, of which six occur in India while Camptorrhiza comprises one species each in Africa (C. strumosa) and India (C. indica). The genusCamptorrhiza possesses a knee-shaped tuber attached to the corms, filaments with a thick bulge in the middle and styles with single stigma. Iphigenia on the other hand lacks knee-shaped tuber, bears linear filaments and has styles with three stigmas. Camptorrhiza indica possesses ovoid corms, linear filaments and styles with a single stigma. These characters are intermediate between Iphigenia and Camptorrhiza and hence we studied the cytogenetics and phylogenetic placement of this species to ascertain its generic identity. Somatic chromosome count (2n = 22) and karyotypic features of C. indica are very similar to that of Iphigenia species. Molecular phylogenetic studies based on atpB-rbcL, rps16, trnL, and trnL-F regions showed that C. indica is nested within a lineage of Indian Iphigenia species. Thus, C. indica was reduced to a species of Iphigenia, i.e., I. ratnagirica. Camptorrhiza is now a monotypic genus restricted only to southern Africa. A key to the IndianIphigenia species is provided. In addition, a new combination Wurmbea novae-zelandiae is proposed for Iphigenia novae-zelandiae.

DNA barcodes have proved to be efficient for plants species discrimination and identification using short and standardized genomic regions. The genus Sinosenecio(Asteraceae) is used for traditional medicinal purposes in China. Most species of the genus occur in restricted geographical regions and exhibit a wide range of morphological variations within species, making them difficult to differentiate in the field. Previously, taxonomic revisions have been made based on morphological and cytological evidence. In the present study, barcoding analysis was performed on 107 individuals representing 38 species in this genus to evaluate the performance of four candidate barcoding loci (matK, rbcL, trnH-psbA and internal transcribed spacer [ITS]) and detect geographical patterns. Three different methods based on genetic distance, sequence similarity, and the phylogenetic tree were used. Comparably high species discrimination power was detected in species-level taxonomic process by the ITS dataset alone or combined with other loci, which was suggested to be the most suitable barcode for Sinosenecio. Our results are congruent with previous taxonomic studies concerning the monophyly of the S. oldhamianus group. The present study provides an empirical paradigm for the identification of medicinal plant species and their geographical patterns, ascertaining the congruence between taxonomical studies and barcoding analysis inSinosenecio.