Plastid atpB/rbcL intergenic spacer sequences were obtained from 21 selected taxa and one hybrid of Anemoninae (Anemone, Pulsatilla, Hepatica) and compared with Ficaria (Ranunculinae) as an outgroup. From the resulting matrix (1226 bp) a single most parsimonious tree was obtained (Fig. 1 ). The branching of this tree is confirmed by many informative indels and appears largely congruent with past plastid restriction analyses. Several new taxa are added. The monophyly of the Anemoninae and their early split into two major clades is supported: clade I with the chromosome base number x = 8, clade Ⅱ with the reduced x = 7. Clade Ⅰ is made up of the basal Pulsatilla and the Rivularis + Vitifolia groups. The Multifida group links to the crown groups Coronaria, Blanda and Nemorosa. Clade Ⅱ consists of the basal Dichotoma group, followed by Hepatica, and finally by the N. Hemisphere Narcissiflora and the S. Hemisphere Antucensis groups as sisters. The problems of the Anemoninae ancestry, phylogenetic differentiation, and recent attempts for systematic classification are critically discussed. In view of the still incomplete sampling of DNA data, aconservative and informal approach to classification problems is recommended.

( 1 ) Nine species and 4 varieties are described as new, and 2 new ranks, 1 new combination, and 1 new name are proposed. (2) Clematis dioica L. ssp. virginiana (L.) Kuntze var. bahamica Kuntze, C. bahamica (Kuntze) Britton, C. orbiculata Correll, C. brasiliana DC. var. laxa St. Hilaire, C. perulata Kuntze, C. barrancae Jones, C. discolor Gardn., C. laxiflora Baker, C. bathiei Lévl., and C. mauritiana Lam. var. sulfurea Viguier & Perrier are reduced into synonymy. (3) A new classification for the varieties of C. chinensis Osbeck is proposed. (4) The specific status of C. flukenetii DC. is discussed and confirmed. (5) The differences between C. acapulcensis Hook. & Arn., the South American population of which has been misidentified as C. affinis St. Hilaire, and its close ally, C. affinis St. Hilaire, are given. (6) The taxonomical confusion made by Viguier and Perrier about the species belonging to sect. Meclatis subsect. Wightianae distributed in Madagascar and adjoining islands is clarified, and an enumera-tion of them with a key is given.

The phylogenetie relationships of Triplostegia Wall. ex DC., comprising two species of perennial herbs from southeastern Asia, have long been in dispute. This genus was placed in either Dipsacaceae or Valerianaceae or in a family of its own, Triplostegiaceae. In this paper, the chloroplast DNA (cpDNA) trn L-F regions of 21 species in the Dipsacales s. l. (including Valerianaceae, Dipsacaceae, Triplostegia, Morina, Caprifoliaceae s. l. and Adoxaceae) and an outgroup Panax schin-seng Nees. were amplified and sequenced. The phylogenetic relationships among these 22 species were constructed based on trn L-F sequences. The results demonstrated that Valerianaceae, Dipsacaceae, Triplostegia, Morina and four genera from the Caprifoliaceae s. l. form a monophyletic group with a strong support (100% bootstrap). Triplostegia, a sister group to Dipsacaceae, is close enough to be placed in the Dipsacaceae as a subfamily. The traditional Caprifoliaceae s.l. are polyphyletic, and relationships of Morina among the groups within Dipsacales s. l. are uncertain. Key words Triplostegia; Caprifoliaceae s. l.; Morina; Dipsacales s. l.; trnL-F sequences; Sys-tematic position

The Chinese species of Selaginella subgenus Tetragonostachys are revised. Three species are distinguished in the present study, i.e.S. sibirica, S. indica and S. vardei. Illustrations of the leaves and strobili, distribution maps of the subgenus and the species occurring in China are made to facilitate easy identification of the species. The distribution of the species in Chinashows an interesting southwestern China and northeastern China distribution pattern.

Polyploidy is widely acknowledged as a major mechanism of adaptation and speciation in plants. Recent estimates suggest that 70% of all angiosperms have experienced one or more episodes of polyploidization. Interdisciplinary approaches combining phylogenetic and molecular genetic perspectives have enhanced our awareness of the myriad genetic interactions made possible by polyploidy. In this paper, cytological mechanisms of polyploid formation and processes and mechanisms of gene and genome evolution in polyploids are reviewed. In many cases, spontaneous polyploids have cytotypes that appear to have been formed by the union of reduced and unreduced gametes. Recent studies demonstrate that most polyploid species have formed recurrently from different populations of their progenitors. Genes duplicated by polyploidy may retain their original or similar function, undergo diversification in protein function or regulation, or one copy may become silenced through mutational or epigenetic means. Duplicated genes also may interact through inter-locus recombination, gene conversion, or concerted evolution. Extensive and rapid genome restructuring can occur after polyploidization. These include inter-genomic chromosomal exchanges, non-Mendelian genomic evolution in nascent polyploids, inter-genomic invasion, and cytonuclear stabilization. Continued application of molecular genetic approaches to questions of polyploid genome evolution holds promise for producing lasting insight into processes by which novel genotypes are generatedand ultimately into how polyploidy facilitates evolution and adaptation.

The genus Oryza L. is one of the most important plant groups in the grass family (Poaceae), which includes more than 20 species and is distributed in tropics and subtropics of the world. The future breakthrough of rice breeding relies greatly on the exploration and utilization of rich germplasm in the rice genepool, particularly the wild rice species. Because of its significant importance in the agricultural production, the genus Oryza has attracted much attention of many taxonomists, geneticists, breeders, and molecular biologists for various kinds of research. During the last two hundred and fifty years since the first description of the genus Oryza by Linnaeus, great changes have taken place in the genus in terms of number of species and taxonomic status. Many taxonomists have made extensive researches on species circumscription and taxonomic ranking at the subgenus level, which was essential for the establishment of the modern taxonomic systems of Oryza. The taxonomic system established by Roschevicz (1931) based on his comprehensive and detailed studies on plant specimens and literature provided an important foundation for the modern taxonomy of the genus Oryza. Sharma & Shastry (1965) offered a taxonomic system of Oryza, which was essentially influenced by that of Roschevicz. This system treated subgenus rank properly, but their definition of Oryza was in a much wider sense and several species in this system have been excluded from the current Oryza. Vaughan (1989) extensively studied and compared Oryza samples from all over the world and provided a taxonomic system including updated data such as morphological variation, geographic distribution, and genome constitution of each species. This system not only adopted a reasonable generic definition, but also provided a good reference of species relationship. Unfortunately, the subgenus rank “complex” used by Vaughan (1989) does not have any legitimate standing in the International Code of Botanic Nomenclature (ICBN). Lu (1999) summarized the major taxonomic studies of previous authors including new species published in Oryza over the past ten years, and proposed an Oryza taxonomic system with 3 sections, 7 series and 24 species. We provided in this paper the most updated studies of the Oryza species and suggested a revised version of Oryza taxonomy with a morphological key to species. We also discussed the existing problems in the taxonomy ofthe genus Oryza.