No abstract

Ranunculus s.l. shows a considerable species diversity and degree of endemism in the Mediterranean region and occurs with various life forms from the lowlands to the highest mountains. Based on a sampling from all continents, sequences of the ITS of nrDNA, the plastid matK, and the adjacent trnK regions were analysed using maximum parsimony and Bayesian inference. Both separate and combined analyses of the two datasets yielded a large core clade of Ranunculus excluding Ficaria, Coptidium, and the extraeuropean genera Beckwithia, Callianthemoides, Halerpestes, and Peltocalathos. The Ceratocephala‐Myosurus‐clade is sister to the core Ranunculus in the plastid and the combined datasets on very long branches, thus supporting a classification of Ceratocephala and Myosurus as separate genera. Within Ranunculus s.s., eight well supported and highly consistent clades correspond either to widespread ecological groups (wetlands, high altitudes/latitudes) or to regional (mainly European) geographical groups. Alpine Mediterranean buttercups belong to orophytic clades, most species of which also occur in the European alpine system; others show widespread northern hemisphere distributions. Only one Mediterranean clade is restricted to the Iberian Peninsula and adjacent regions. Present distribution patterns and molecular data support a hypothesis of an origin of alpine buttercups from lowland ancestors of the same geographical region. At lower altitudes, the predominant life forms, i.e., therophytes and geophytes, evolved multiple times suggesting parallel adaptations to the Mediterranean climate. Geophytes differentiated into an eastern and western Mediterranean group, and are most closely related to the subalpine, non‐monophyletic “R. montanus” group, thus supporting a hypothesis of a common lowland ancestor. Tentative estimates for divergence times of the major clades in Ranunculus s.l. were made based on an age calibration for the Ranunculus‐Xanthorhiza‐split, using matK sequences and penalized likelihood analyses. The results from this study suggest that the split of allied genera from Ranunculus s.s. occurred during the Eocene and Oligocene, with the core clade of Ranunculus being c. 24.0 Myr old. Diversification of Ranunculus s.s. into main ecological/geographical clades took place in the late Miocene, and speciation within the Mediterranean groups during the Pliocene and Pleistocene. Diversification of life forms at lower altitudes occurred mainly during or after the establishment of the Mediterranean climate. Island endemics of Macaronesia and Crete are probably rather young descendents of neighbouring geographical groups. Diversification of alpine groups took place at different geological times, but is in general correlated with periods of colder climate. The high diversity of buttercups is likely a consequence of the broad spectrum of different habitats in the Mediterranean region.

Ranunculeae represent a highly diverse and cosmopolitan tribe within Ranunculaceae. Because of the great diversity of morphological features and lack of molecular phylogeny for the tribe, the classification of its genera has always been controversial. We report here molecular phylogenetic analyses based on nuclear and plastid markers (nrITS, matK, trnK, psbJ-petA) that provide a framework for understanding relationships and character evolution within the tribe. Maximum parsimony analyses suggest a weakly supported basal dichotomy, while Neighbor Net analysis indicates strong support for five distinct lineages. Both methods revealed several well-supported, small terminal clades which correspond to previously described genera, characterised by unique morphological features and character combinations. Anatomical structures of the achenes suggested relationships with greatest concordance to those in the molecular phylogeny. Macroscopic analysis of achene morphology often indicated parallel evolution of structures related to certain dispersal mechanisms. Characters of perianth, androecium, gynoeceum and pollen are highly homoplasious, but several features characteristic of small terminal clades and terminal branches can be observed. Geographic isolation and adaptions may have triggered the evolution of morphological characters. We conclude that a classification accepting several small genera (Arcteranthis, and a large genus Ranunculus s.str. (including Batrachium, Aphanostemma and Gampsoceras) reflects best the molecular phylogeny and morphological diversity of the tribe.

No abstract

Rensch's rule refers to a pattern in sexual size dimorphism (SSD) in which SSD decreases with body size when females are the larger sex and increases with body size when males are the larger sex. Many animal taxa conform to Rensch's rule, but it has yet to be investigated in plants. Using herbarium collections from New Zealand, we characterized the size of leaves and stems of 297 individuals from 38 dioecious plant species belonging to three distantly related phylogenetic lineages. Statistical comparisons of leaf sizes between males and females showed evidence for Rensch's rule in two of the three lineages, indicating SSD decreases with leaf size when females produce larger leaves and increases with leaf size when males produce larger leaves. A similar pattern in SSD was observed for stem sizes. However, in this instance, females of small-stemmed species produced much larger stems than did males, but as stem sizes increased, SSD often disappeared. We hypothesize that sexual dimorphism in stem sizes results from selection for larger stems in females, which must provide mechanical support for seeds, fruits, and dispersal vectors, and that scaling relationships in leaf sizes result from correlated evolution with stem sizes. The overall results suggest that selection for larger female stem sizes to support the weight of offspring can give rise to Rensch's rule in dioecious plants.