A fern from Madagascar is described and illustrated as a new species, Elaphoglossum nidusoides. We investigated its phylogenetic relationships within Elaphoglossum based on analyses of three non-coding plastid loci (trnL-trnF, rps4-trnS, atpB-rbcL) in 30 species representative of the diversity of the sections in the genus. Our results show that E. nidusoides belongs to section Squamipedia. This is an unexpected phylogenetic position because the new species has a short-creeping rhizome, phyllopodia, fronds glabrous or nearly so and more than 15 cm long, and lacks the characteristic peg-like aerophores on the rhizome. Nevertheless, its position as sister to another Malagasy species, E. marojejyense, is supported by the succulent fronds.

Illiciaceae comprise the single genus, Illicium, and approximately 30–40 species distributed in southern and eastern Asia, the southeastern United States, Mexico, and the Greater Antilles. On the basis of perianth morphology, current taxonomic treatments recognize two sections that do not correspond to geographic distributions. A recent phylogenetic study using ITS sequences found strong support for separate Old and New World clades, but did not include all New World species. We elaborate on this previous study by including all known New World species of Illicium, as well as additional Asian species, using ITS and the plastid loci trnL, the trnL-F spacer, and trnG. We estimated divergence times within Illicium using penalized likelihood and multiple fossil calibration points. Our results support separate New and Old World clades, and we re-circumscribe current sectional groupings based on these data. The Illicium crown group appears to have arisen during the Cretaceous, with New World extant taxa diversifying as recently as the late Miocene or early Pliocene. These dates are consistent with recent fossil evidence and molecular estimates for minimum ages of the angiosperms, and for timing of diversification events in other ancient groups, such as Nymphaeales and Chloranthaceae.

This study presents the first phylogenetic analysis of tribe Chamaedoreeae (Arecaceae), using parsimony and Bayesian analyses of plastid DNA sequences (matK, rps16 intron, 3′ region of ndhF, and trnD-trnT). The tribe includes more than 115 species, and has a disjunct distribution with four genera in Central and South America and one genus in the Mascarene Islands. While the placement of Chamaedoreeae within Arecaceae has been controversial, the monophyly of this tribe is well supported by plastid DNA sequence data. All genera in Chamaedoreeae are resolved as monophyletic with high support, but relationships among genera are not fully resolved. The placement of Hyophorbe and the monotypic Wendlandiella as sisters to the remaining genera indicates that solitary flowers and dioecy arose at least twice within this tribe, once in Wendlandiella and once in Chamaedorea. Although a low substitution rate of palm plastid DNA has been widely noted, the results of this study show high resolution at the species level, especially within the largest genus, Chamaedorea, indicating that plastid DNA is useful for the inference of relationships at low taxonomic levels in some groups of palms.

A cladistic analysis of Lymania was conducted using morphology and sequences from three chloroplast DNA regions: the matK coding region and the psbA-trnH and trnL-trnF intergenic spacers. The monophyly of the genus and the phylogenetic relationships among related genera were examined. Of the nine Lymania species, eight are endemic to southern Bahia, Brazil. Lymania is the first genus in Bromeliaceae subfamily Bromelioideae to be subjected to a combined morphological and molecular analysis. The genera of Bromelioideae have been particularly difficult to classify and there has been disagreement about their interrelationships and monophyly. Morphological data show better resolution than molecular data alone. The partition homogeneity test supported a combined analysis of the two data sets, yielding a single most parsimonious tree. In the combined analysis, monophyly of Lymania is moderately supported, and the genus is closely related to species of Aechmea subg. Lamprococcus and subg. subg. Ortigiesia. The morphological distinctiveness coupled with low molecular divergence indicates relatively recent and rapid speciation within Lymania. The combined analysis of morphological and molecular data as done in this study provides a framework for future research on other Bromelioideae genera that could foster better taxonomic rearrangements.

Proteoideae (27 genera, 655 species) are one of the largest clades in the early-diverging eudicot family Proteaceae and are distributed in Australia, New Caledonia, and Africa. New palynological observations using scanning and transmission electron microscopy are reported for 15 genera and 49 species, which complement previous work on the subfamily. These are integrated with an extensive review of all literature data available to provide a summary of pollen morphology and variation in each genus of Proteoideae, using a new rigorous method of character scoring. The results are discussed in the context of recent phylogenetic analyses of Proteaceae. The monophyletic Leucadendreae, comprising two Australian genera and a large clade of 10 southern African genera, appear to be relatively homogenous in pollen morphology, especially in ultrastructure of the apertural region. In addition, two other African genera, Faurea and Protea, share very similar pollen characters, some of which appear to be apomorphic, consistent with strong phylogenetic evidence for their sister group relationship. Beauprea, Eidothea, Faurea, Protea, and Stirlingia share a distinct pattern of foot layer disintegration at the apertural region and complete lack of endexine. The occurrence of the same apertural type in the early-diverging Persoonioideae suggests that it is plesiomorphic in Proteoideae and Proteaceae as a whole. This study highlights the phylogenetic potential of ultrastructural characters in Proteaceae and high levels of homoplasy and polymorphism in other characters including pollen shape and tectum sculptural patterns.

Most floral diversity in the core eudicots is based on a groundplan that includes among its most distinctive elements a bipartite perianth and pentamerous merosity combined with a bicarpellate gynoecium. To obtain insights into the origins of this groundplan, the evolution of its structural components in Hamamelidoideae (Hamamelidaceae, Saxifragales, core eudicots) is explored. Phylogenetic hypotheses were obtained from independent analyses of morphological and molecular data, and from two alternative approaches to integrating morphological and molecular data from living and fossil taxa. In spite of the large numbers of missing data cells, fossils were unambiguously placed within particular clades in the well resolved but poorly supported trees obtained from morphological data. Phylogenetic analyses of individual and combined molecular data sets using different optimization criteria resulted in congruent trees that share most strongly supported nodes. No well-supported incongruence existed among the morphological and molecular trees. A total evidence analysis including only complete taxa was congruent with molecular trees, but resolution decreased substantially as incomplete taxa and incomplete characters were included. A parsimony analysis of morphological data constrained with the strongly supported nodes from the total evidence tree resulted in the same backbone relationships as the molecular trees and in the unambiguous placement of fossil taxa; however, relationships within some clades were unique. The floral characters of ancient fossils and their nested placement in the phylogeny indicate stasis in floral structure since the origin of major lineages of Hamamelidoideae. Individual elements of floral structure display independent evolutionary histories, suggesting that the core eudicot floral groundplan evolved through the accretion of components which originated at different points during angiosperm history: some outside the eudicots, and others, namely the bipartite perianth and the pentamerous merosity, most likely within core eudicots.

Phylogenetic relationships among the tribes of Onagraceae were studied using sequences of PgiC, a nuclear gene encoding cytosolic phosphoglucose isomerase. Fragments of 21 genes were examined from plants of 15 species representing 11 of the 17 genera in the family and seven of the eight tribes (the newly proposed Gongylocarpeae was not studied). Maximum likelihood analysis provided a phylogenetic tree with 100% bootstrap support for all but one of the clades that indicate tribal relationships, and corroborated results of recent studies based on chloroplast genes and nuclear ITS sequences. PgiC appears to have a greater rate of divergence than the chloroplast genes and provides much phylogenetically useful data. Two PgiC genes were found in each of the two species sampled in Epilobium; they formed two sister clades, with each species having one gene of each type, showing that the species descended from a single ancestral tetraploid. The two species, E. brachycarpum and E. canum, are not closely related, and previous studies showed that the smallest clade including both contained six of the seven sections of Epilobium. Thus, all species of Epilobium except those in section Epilobium (which was not studied) share a common tetraploid origin, refuting the “Stebbins/Raven” model of the origin of tetraploidy in this genus.

A long history of hybridization and introgression in the evolution of Betula has made the use of chloroplast genes to infer interspecific relationships prone to errors. Single and low copy nuclear genes, however, may offer some hope for accurately reconstructing phylogenetic relationships of Betula and inferring the origin of polyploid species. In this study, sequences of the third intron of nuclear nitrate reductase (Nia) were used to estimate relationships of known diploid species of Betula. Two distinct types of the Nia gene were observed in Betula. Sequences of one type were gathered from all sampled species; those of the second type, obtained from only B. nana, were not used in phylogenetic analyses. Cloned sequences of the Nia gene varied little within individuals or species. Maximum parsimony and likelihood analyses generated robust phylogenies with similar relationships. Betula alnoides and B. maximowicziana of subgenus Betulaster formed a well supported clade, which is characterized by the racemose pistillate inflorescence. Shrubby species (B. nana and B. michauxii) of a traditionally recognized group did not form a clade. Instead, B. nana was most closely related to white-barked species (e.g., B. pendula, B. platyphylla, and B. populifolia) and together they formed a clade with B. nigra. In contrast, B. michauxii is closely allied with B. lenta. Our results suggest that subgenera Betulaster and Betula are monophyletic, while subgenera Chamaebetula and Neurobetula are polyphyletic.

The genus Rosa (roses) comprises approximately 190 shrub species distributed widely throughout the temperate and subtropical habitats of the northern hemisphere. Despite numerous recent studies examining phylogenetic relationships in the genus, relationships remain obscure due to problems such as poor identification of garden specimens, hybridization in nature and in the garden, and low levels of chloroplast and nuclear genome variation. Phylogenetic analyses of non-coding chloroplast sequences from the trnL-F region and psbA-trnH intergenic spacer for 70 taxa show slightly more variation than previous analyses of the genus. Bayesian and parsimony analyses suggest that subg. Rosa can be divided into two large clades, each with low internal resolution. One comprises species from sections Carolinae, Cinnamomeae and Pimpinellifoliae p.p., whilst the other consists of all of the remaining sections of subg. Rosa (Banksianae p.p., Bracteatae, Caninae, Indicae, Laevigatae, Rosa, Synstylae and Pimpinellifoliae p.p.). A fairly complete sampling of field-collected North American taxa has been incorporated in this analysis. Analyses indicate that migration into North America occurred at least twice within this primarily Old World genus. Most North American taxa, except R. setigera and R. minutifolia, fall into a single clade that includes Asian and European taxa. Analyses also are consistent with the notion that cultivated commercial roses have a relatively narrow genetic background. Six of the seven primary taxa believed to be involved in the creation of domesticated roses are found within the same large clade that mostly includes Asian and European taxa.