The products of ecological speciation rarely engage in wholesale hybridization with their progenitors, because the ecological barriers between them are unusually strong. This strength may simply be the result of directional selection, or they may arise from the differential survival of incipient species with the strongest barriers. A case is made for the latter. Those incipient species most similar in niche space to their progenitors would be most subject to hybridization with them. This would lead to a reduction in population growth and/or mongrelization of their gene pools associated with such, and ultimately the demise of the derivative. The most ecologically divergent derivatives would suffer less from this process referred to as isolate selection, and have the greatest chance of becoming a thriving new species. Long-distance hybridization has been demonstrated in crop and wild populations. The incidence of long-distance hybridization is inversely correlated with population size. The ecological derivatives struggling to invade new habitats would be especially vulnerable to hybridization.

Isothecium myosuroides is an abundant and taxonomically problematic moss that occurs in Europe and on the west and east coasts of North America. It has sometimes been split into two taxa, I. myosuroides on the east coast of North America and in Europe, and I. stoloniferum on the west coast of North America. Isothecium stoloniferum has four distinct morphological types known as the ordinary, julaceous, coarse, and pinnate morphs. The degree to which the morphotypes are genetically distinct versus ecologically determined has been debated as has the distinction between I. stoloniferum and I. myosuroides, resulting in conflicting taxonomic treatments. This study uses phylogenetic analyses of the trnL-trnF and trnG regions of the chloroplast genome to examine relationships among the two species and four morphotypes. Molecular phylogenetic analyses reveal that there is some genetic basis to the morphs of I. stoloniferum, and that I. stoloniferum is paraphyletic. Accessions of I. myosuroides form a trans-Atlantic monophyletic group. The remaining samples form two distinct groups that each include samples of two morphotypes within I. stoloniferum. Populations of julaceous and ordinary morphotypes form a monophyletic group. Populations of coarse and pinnate morphotypes form an unresolved polytomy at the base of the clade that includes accessions of I. myosuroides.

A comprehensive taxonomic revision of the genus Mitrephora (Annonaceae) in Cambodia (Kampuchea), Laos, and Vietnam is presented. Eight species are recognized, including five that are newly described, viz. M. calcarea sp. nov., M. macclurei sp. nov., M. pallens, M. petelotii sp. nov., M. phanrangensis sp. nov., M. poilanei sp. nov., M. teysmannii, and M. tomentosa (inclusive of M. thorelii). Previous taxonomic confusion regarding the delimitation of M. calcarea is clarified and the name is validated for the first time. The correct application of the name M. maingayi is clarified: although the name has been used to refer to a species that is distinct within the region, the name itself is a synonym of M. teysmannii. Most of the Indochinese collections determined as “M. maingayi” are shown here to represent a new species, M. macclurei.

Satyrium is a morphologically anomalous genus. Although clearly a member of the terrestrial orchid subfamily Orchidoideae, its phylogenetic relationships within the subfamily are uncertain. Morphologically it has always been placed in tribe Diseae, albeit associated with different subtribes, but recent molecular studies suggest a closer relationship to Orchideae. The infrageneric classification of Satyrium is equally problematic, and several different classifications are available. The only infrageneric classification based on a phylogenetic analysis of morphological characters resulted in recognition of the three subgenera Brachysaccium, Bifidum, and Satyrium. DNA sequence data from nuclear (ITS1, 5.8S, and ITS2) and plastid (trnL intron, trnL-F intergenic spacer, and part of the matK gene and trnK intron) genome are used to test the monophyly of these subgenera. Topologies of cladograms resulting from parsimony analysis of separate datasets show several cases of incongruence, some of which are well supported. Combined analysis is performed on a dataset from which two problematic taxa are pruned. Parametric bootstrap, as well as Bayesian posterior probability, rejects monophyly of all three subgenera and alternative groupings are suggested.

Phylogenetic analyses were performed on six genera and 46 species of the Neotropical palm tribe Geonomeae (subfamily Arecoideae) and outgroups composed of 14 species from the Arecoideae, two from the Ceroxyloideae, and one from the Phytelephantoideae. The analyses were based on two low copy nuclear DNA sequences from the genes encoding phosphoribulokinase and RNA polymerase II. The two data sets did not appear to conflict and were analyzed separately and in combination. Our results added support to the previously reported monophyly of the tribe. The basal node of the tribe was polytomous containing the genera Welfia and Pholidostachys and a clade consisting of the rest of the Geonomeae. Pholidostachys formed a monophyletic group. The currently accepted genera Calyptronoma and Calyptrogyne formed a well-supported clade with Calyptronoma resolved as paraphyletic to Calyptrogyne. Geonoma formed a strongly supported monophyletic group consisting of two main clades. Geonoma section Taenianthera was not monophyletic but a group of three high-elevation Geonoma species was resolved.

Recently, advocates of DNA taxonomy have complained that there is inadequate control of the taxonomy in databases such as GenBank. This is correct, but the uncertainty may be extended to the sequences themselves. The present study shows that as long as vouchers are available neither problem is fatal, but if no voucher exists, bad sequences and bad taxonomy may be forever linked. Previous phylogenetic analyses of rbcL sequences have indicated that the small, south hemisphere, genus Oxychloë (Juncaceae) surprisingly either is embedded within or is a sister group to the Cyperaceae. This is not in accordance with traditional or current morphological data. By studying five new accessions, representing four species of Oxychloë, and re-examining the two vouchers of O. andina that were used in previous phylogenies, it has been possible to show that these two sequences are erroneous. One is a chimeric sequence and the other is most likely “a contaminant.”

Our studies on the enigmatic genus Bienertia (Chenopodiaceae), with its C₄ photosynthesis and lack of Kranz anatomy, led us to the discovery of a second species of this previously-supposed monotypic genus. The new species is named Bienertia sinuspersici after its main range around the Persian Gulf countries and the northern side of the Gulf of Oman. Bienertia sinuspersici occurs in hot climates and is a vicariant of Bienertia cycloptera, which is found at higher latitudes and elevations in temperate and cold deserts of the region. Like Bienertia cycloptera, the new species has unique chlorenchyma cells with dimorphic chloroplasts and single cell C₄ photosynthesis. However, it differs anatomically by having mostly one to two layers of chlorenchyma cells, versus two to three layers in Bienertia cycloptera. Furthermore, the new species has longer cotyledon leaves, larger seeds, larger flowers, and larger chromosomes, and differs in a set of micro-morphological features. All of this supports our conclusion that this widely distributed, novel plant is an overlooked new species. Bienertia sinuspersici grows well in very hot climates, under conditions which most species can barely tolerate. Its wide distribution indicates that its novel C₄ photosynthesis may confer advantages for CO₂ fixation in these habitats not found in C₄ species having conventional Kranz anatomy.

Phylogenetic analyses of two chloroplast and five putatively unlinked nuclear DNA regions were used to explore the evolutionary relationships of a circumpolar arctic polyploid species complex in Silene. Gene phylogenies inferred from introns in the low copy nuclear genes RPA2, RPB2, RPD2a, and RPD2b, and ITS1 and ITS2 from the nuclear ribosomal DNA region, indicate two consecutive polyploidization events. The first involved the diploid arctic/subarctic S. uralensis lineage as the cytoplasmic donor, as indicated by chloroplast rps16 and psbE-petL sequences, and highly unexpected from a morphological perspective, the diploid Siberian/northeast Asian S. ajanensis lineage as pollen donor. The hybridization and polyploidization resulted in the tetraploid S. involucrata lineage. A second hybridization and polyploidization with the S. ajanensis lineage as pollen donor, and the tetraploid S. involucrata lineage as cytoplasmic donor, resulted in the hexaploid lineages of S. sorensenis and S. ostenfeldii. In general, two paralogous sequences were identified from the tetraploids and three paralogues from the hexaploids in the low copy nuclear genes. Interlocus concerted evolution appears to have homogenized the ITS regions, because only the sequences corresponding to the paternal lineage were recovered in the polyploids. The power, and neccessity, of using several potentially unlinked regions is revealed by the fact that the gene phylogenies had small deviations from the general pattern explained by alloploidy. These deviations are better explained as gene duplication and/or lineage sorting events, or simply lack of information.

DNA sequence data from nuclear ribosomal internal transcribed spacer, including the 5.8s coding region, and plastid trnK/matK were used to reconstruct the phylogeny of Montieae (Portulacaceae), and a revised classification for the group based on monophyly is proposed. Montieae consists of the sister clades Claytonia and Montia. In Claytonia, there is strong support for the following clades: section Limnia, all annuals except the perennials C. sibirica and C. palustris; section Rhizomatosae, high elevation or high latitude perennials with rhizomatous or caudicose growth habits; and section Claytonia, species with spheric or obconic underground perennation structures. All analyses recover clades of Montia circumscribed as sections Montiastrum, Australiensis, and Montia. Heenan's Australasian Neopaxia is placed robustly in Montia section Australiensis, a clade limited primarily to Australia and New Zealand. DIVA and MacClade reconstruct the ancestral area for Montieae as western North America. Beringian taxa of Claytonia section Rhizomatosae are derived from low latitude western North American ancestors adapted to persistently wet, alpine habitats. Although we infer that section Claytonia had multiple shifts to northern high latitudes and potentially to eastern North America from western North America, relationships among clades in this section have limited support. Habitat reconstructions show that the pleisiomorphic condition for moisture regime in section Claytonia is largely maintained over shifts in elevation/latitude.

Previous classifications based on morphology recognized a distinctive but poorly delimited group of species from western North America as Oenothera subg. Pachylophus. In earlier papers, I split this group into four sections more comparable to others in the genus, but nevertheless assumed the whole group was monophyletic based on morphological analyses. Recent molecular analyses indicate that the morphological analysis was incorrect, however, and suggest instead that the sections are part of a basal grade within the genus. The two yellow flowered sections are shown by the molecular study to be sister taxa, yet because of highly distinctive morphology, especially of the capsules and seeds, I maintain them as separate sections, sect. Contortae (O. xylocarpa) and sect. Eremia (O. primiveris) comparable to others in the genus. The two white flowered sections, sect. Pachylophus (O. brandegeei, O. cespitosa, O. cavernae, O. harringtonii, and O. psammophila) and sect. Ravenia (O. muelleri, O. riskindii, and O. tubifera with two subspecies, tubifera and macrocarpa), are not so closely related and the shift to white petals appears to have been independent. Detailed taxonomic descriptions are given for the species of sects. Contortae, Eremia, and Ravenia. Taxonomic treatment of the species of sect. Pachylophus was given elsewhere.

A new genus and species of Euphorbiaceae s.s., Brasiliocroton mamoninha, is described from two disjunct areas of lowland forest remnants in eastern and northeastern Brazil. It is a member of tribe Crotoneae and was previously confused with Croton and Micrandra. The resemblance to Micrandra is based on the branched inflorescences and terminal position of the pistillate flowers, trigonous capsules, and ovate leaves with basal glands, but the stamens are more numerous and the pollen is inaperturate (like Croton). Brasiliocroton has a stellate-rosulate indumentum like some Croton species, but its anthers are erect in bud, and the position of the pistillate flowers is anomalous in Croton. Molecular data place Brasiliocroton as the closest sister group to Croton, once Astraea has been removed from Croton and placed sister to Acidocroton and Ophellantha. Wood anatomical features support the generic status of Brasiliocroton.

Phylogenetic relationships within Sapindaceae sensu lato are assessed using sequence data from two plastid genes, analyzed separately and together. A total of 46 rbcL sequences (31 of which are new), and 89 matK sequences (75 new) representing 64 genera were subjected to parsimony and Bayesian analysis. The results support three major lineages, relationships between which are only weakly supported. Xanthoceras sorbifolium is not clearly a member of any of these lineages, and there is some support for it being the first lineage to diverge within the entire assemblage. Support for a broadly defined Sapindaceae incorporating Aceraceae, Hippocastanaceae, and Xanthoceras is very robust. A division into four subfamilies is proposed: Sapindoideae (including Koelreuteria and Ungnadia); Hippocastanoideae (including taxa previously referred to Aceraceae and Hippocastanaceae, plus Handeliodendron); a more narrowly defined Dodonaeoideae; and a monotypic Xanthoceroideae. Tribal groupings are critically evaluated in light of the analyses. Although many of the current tribes appear paraphyletic or polyphyletic, there is support for the monophyly of some core groups of genera that suggest realignments of tribal concepts that would render them more informative of relationships.

This study represents the outstanding revision of the Chilean C. integrifolia s.l. complex. In the monograph of the Chilean species of Calceolaria, some taxa of doubtful identity or unclear taxonomy but clearly closely allied to C. integrifolia s.str. were placed under the name C. integrifolia s.l. The taxa, all united by rugose leaves, well developed inflorescences, and overall very similar flower shape, seemed to comprise a close aggregation of innumerable different populations that were impossible to differentiate into well defined species. Extensive fieldwork and cultivation of more than 80 populations collected from almost the entire distribution area together with a broad selection of herbarium collections revealed characters for species delimitation. Main characters are well defined types of indumentum. In combination with other characters like habit and shape of leaves, the taxonomic revision of the C. integrifolia s.l. complex now recognizes nine species: C. andina, C. angustifolia, C. auriculata, C. georgiana, C. integrifolia s.str., C. rubiginosa (described here as a new species), C. talcana, C. verbascifolia and C. viscosissima. Every species has its own characteristic distribution pattern, which is correlated with ecological and climatic factors. Distribution maps as well as figures of main characters, a key to the taxa, synonymies, descriptions, and representative specimens are included.

Members of Veroniceae are some of the most prominent members of the alpine vegetation of New Guinea. Generic classification suggests at least three origins because the endemic species of Veroniceae comprise two species of Veronica, twelve species of Parahebe, and the monotypic genus Detzneria. DNA sequence data (nr ITS region, plastid trnL-F region) have been analyzed using parsimony and maximum likelihood to test the hypothesis of three different introductions to New Guinea and their origin. The results largely confirm previous suggestions based on morphology, although lack of sufficient sequence variability in the Hebe complex does not allow definite conclusions. Veronica archboldii is closely related to the widespread V. serpyllifolia from the Northern Hemisphere. Parahebe vandewateri and P. albiflora are sisters and probably derived from New Zealand species, although not necessarily from species of Parahebe there. Finally, Detzneria tubata also belongs to the Hebe complex but has no clear affinities in this group. It may be a relict from the first immigration of Veronica species to the Southern Hemisphere. Reasons for different success of the three groups in speciating in New Guinea are discussed in the light of the known vegetation history of New Guinea.

Solanum arcanum and S. huaylasense, two new wild tomato species segregated from Solanum peruvianum sensu lato, are described and illustrated. These two new species are placed in a key with two other segregates of S. peruvianum sensu lato: S. peruvianum sensu stricto and S. corneliomulleri, and the morphologically similar species S. chilense. We also present a list of all 13 species of wild tomatoes we recognize, and their equivalent former names in Lycopersicon.

The nuclear gene SAMT—salicylic acid methyltransferase—is a novel phylogenetic marker that was isolated using RT-PCR in order to estimate intergeneric relationships within the Solanaceae. SAMT catalyzes the formation of the volatile compound methyl salicylate, which is an important secondary metabolite for both pathogen defense and floral fragrance in this family. Relationships inferred with SAMT cDNA sequences are mostly congruent with previous cpDNA estimates, which not only increases our confidence in Solanaceae phylogeny, but indicates that paralogy and lineage sorting appear not to be severe for this nuclear gene at this phylogenetic level. Relationships observed with SAMT include Schizanthus as sister to the rest of the family. The second branch was Schwenckioideae Cestroideae. The final two lineages were Petunioideae sister to the x=12 clade, which includes Nicotianoideae Solanoideae. Although mostly congruent, one important phylogenetic difference between previous cpDNA and nuclear SAMT phylogenetic estimates is noted. Nicandra and Exodeconus are not supported as sister taxa; rather, Exodeconus is near the base of the Solanoideae while Nicandra appears to belong to a lineage that includes Physalis, Capsicum, Solanum, and Datura. The results obtained in this study indicate that expanded sampling of taxa for SAMT and the study of additional nuclear gene sequences in conjunction with cpDNA will be valuable towards improving our understanding of Solanaceae phylogeny.

A new species endemic to Mitiaro, Cook Islands, Tetramolopium mitiaroense, is described and illustrated. The species is most closely related to Hawaiian species of Tetramolopium section Tetramolopium. Tetramolopium mitiaroense shares an upright dwarf shrub habit with T. remyi and T. capillare but differs in possessing semi-succulent, linear-oblanceolate leaves versus sclerophyllous, involute, linear leaves for the latter two taxa. Tetramolopium sylvae also has succulent leaves but differs from T. mitiaroense in having a prostrate, rosette habit and spatulate leaves. A key is provided for all eastern Pacific species of Tetramolopium. Published morphological, molecular, and geological data are reviewed to assess relationships amongst Tetramolopium taxa in the eastern Pacific and elucidate the biogeographic pattern of dispersal. The data support the hypothesis that Tetramolopium was first dispersed to the Hawaiian Archipelago from New Guinea with a secondary dispersal event from the Hawaiian Islands to the Cook Islands. Available evidence indicates that the Pacific Golden-Plover is the probable agent of dispersal from the Hawaiian Islands to the Cook Islands.