Recent advances in molecular phylogenetics and a series of important palaeobotanical discoveries have revolutionized our understanding of angiosperm diversification. Yet, the origin and early evolution of their most characteristic feature, the flower, remains poorly understood. In particular, the structure of the ancestral flower of all living angiosperms is still uncertain. Here we report model-based reconstructions for ancestral flowers at the deepest nodes in the phylogeny of angiosperms, using the largest data set of floral traits ever assembled. We reconstruct the ancestral angiosperm flower as bisexual and radially symmetric, with more than two whorls of three separate perianth organs each (undifferentiated tepals), more than two whorls of three separate stamens each, and more than five spirally arranged separate carpels. Although uncertainty remains for some of the characters, our reconstruction allows us to propose a new plausible scenario for the early diversification of flowers, leading to new testable hypotheses for future research on angiosperms.

The diverse and species-rich order Ericales has found considerable interest among systematists in recent years. Molecular phylogenetic studies not only have convincingly demonstrated the monophyly of the order, comprising 23 families formerly placed in three different subclasses (Asteridae, Dilleniidae, and Rosidae), but have also resolved Ericales as sister to euasterids. Most ericalean families are well circumscribed and have been or are currently subject to intrafamilial phylogenetic studies. In spite of all the attention that Ericales have received recently, there remains a major challenge, the still largely unresolved deeper nodes in the ericalean phylogeny. This study aims to improve our current knowledge of the interfamilial relationships by expanding on gene and taxon sampling and to evaluate the evolution of important floral characters in light of the resulting phylogeny. We add a nuclear region (26s rDNA) to already published data sets (nuclear: 18s rDNA; mitochondrial: atp1, matR; chloroplast: atpB, ndhF, rbcL, matK, the rps16 intron, the trnT-trnF spacer, and the trnV-atpE spacer), for a total of 11 molecular markers that include nearly 20 kb of sequences. Our analyses, applying both maximum parsimony and Bayesian inference, resolve some of the deeper nodes in the phylogeny. Strongly supported groups, previously unrecognized or only weakly supported, include (1) a clade comprising all families except Balsaminaceae, Tetrameristaceae, Marcgraviaceae, Fouquieriaceae, Polemoniaceae, and Lecythidaceae; (2) a clade with Sapotaceae, Ebenaceae, and the primuloid families; (3) a clade with Symplocaceae, Styracaceae, and Diapensiaceae; and (4) a clade comprising the latter three families plus Theaceae, Roridulaceae, Actinidiaceae, Sarraceniaceae, Clethraceae, Cyrillaceae, and Ericaceae. At an analytical level, our results indicate that more data in the form of additional markers do improve resolution and branch support and should eventually lead to a fully resolved ericalean phylogeny. At the level of floral evolution, we demonstrate that sympetaly is a homoplasious character in the order, that a diplostemonous floral ground plan likely arose from haplostemonous flowers in Ericales, and that the combination of ovules with a single integument and cellular endosperm formation is characteristic for two of the major clades in the order.

Summary Pollination syndromes describe recurring adaptation to selection imposed by distinct pollinators. We tested for pollination syndromes in Merianieae (Melastomataceae), which contain bee‐ (buzz‐), hummingbird‐, flowerpiercer‐, passerine‐, bat‐ and rodent‐pollinated species. Further, we explored trait changes correlated with the repeated shifts away from buzz‐pollination, which represents an ‘adaptive plateau’ in Melastomataceae. We used random forest analyses to identify key traits associated with the different pollinators of 19 Merianieae species and estimated the pollination syndromes of 42 more species. We employed morphospace analyses to compare the morphological diversity (disparity) among syndromes. We identified three pollination syndromes (‘buzz‐bee’, ‘mixed‐vertebrate’ and ‘passerine’), characterized by different pollen expulsion mechanisms and reward types, but not by traditional syndrome characters. Further, we found that ‘efficiency’ rather than ‘attraction’ traits were important for syndrome circumscription. Contrary to syndrome theory, our study supports the pooling of different pollinators (hummingbirds, bats, rodents and flowerpiercers) into the ‘mixed‐vertebrate’ syndrome, and we found that disparity was highest in the ‘buzz‐bee’ syndrome. We conclude that the highly adaptive buzz‐pollination system may have prevented shifts towards classical pollination syndromes, but provided the starting point for the evolution of a novel set of distinct syndromes, all having retained multifunctional stamens that provide pollen expulsion, reward and attraction.

Computed tomography remains strongly underused in plant sciences despite its high potential in delivering detailed 3D phenotypical information because of the low X-ray absorption of most plant tissues. Existing protocols to study soft tissues display poor performance, especially when compared to those used on animals. More efficient protocols to study plant material are therefore needed. Flowers of Arabidopsis thaliana and Marcgravia caudata were immersed in a selection of contrasting agents used to treat samples for transmission electron microscopy. Grayscale values for floral tissues and background were measured as a function of time. Contrast was quantified via a contrast index. The thick buds of Marcgravia were scanned to determine which contrasting agents best penetrate thick tissues. The highest contrast increase with cytoplasm-rich tissues was obtained with phosphotungstate, whereas osmium tetroxide and bismuth tatrate displayed the highest contrast increase with vacuolated tissues. Phosphotungstate also displayed the best sample penetration. Furthermore, infiltration with phosphotungstate allowed imaging of all plants parts at a high resolution of 3 µm, which approaches the maximum resolution of our equipment: 1.5 µm. The high affinity of phosphotungstate for vasculature, cytoplasm-rich tissue, and pollen causes these tissues to absorb more X-rays than the surrounding tissues, which, in turn, makes these tissues appear brighter on the scan data. Tissues with different brightness can then be virtually dissected from each other by selecting the bracket of grayscale to be visualized. Promising directions for the future include in silico phenotyping and developmental studies of plant inner parts (e.g., ovules, vasculature, pollen, and cell nuclei) via virtual dissection as well as correlations of quantitative phenotypes with omics datasets. Therefore, this work represents a crucial improvement of previous methods, allowing new directions of research to be undertaken in areas ranging from morphology to systems biology.

No abstract