Pseudanthia in angiosperms: a review

BACKGROUND

Pseudanthia or 'false flowers' are multiflowered units that resemble solitary flowers in form and function. Over the last century the term 'pseudanthium' has been applied to a wide array of morphologically divergent blossoms, ranging from those with easily noticeable florets to derived, reduced units in which individual flowers become almost indistinguishable. Although initially admired mostly by botanists, the diversity and widespread distribution of pseudanthia across angiosperms has already made them a fascinating topic for evolutionary and developmental comparative studies.

SCOPE

This review synthesizes historical and current concepts on the biology of pseudanthia. Our first aim is to establish a clear, operational definition of pseudanthium and disentangle common terminological misconceptions surrounding that term. Our second aim is to summarize knowledge of the morphological and developmental diversity of pseudanthia and embed it within a modern phylogenetic framework. Lastly, we want to provide a comprehensive overview on the evolution and ecological importance of pseudanthia and outline perspectives for future studies.

CONCLUSIONS

The understanding of pseudanthia has changed multiple times and reflects three different interpretations of their 'flower-like' qualities: developmental (similarity in structure), figural (similarity in form and function) and phylogenetic (homology between angiosperm flowers and monoecious reproductive shoots in gymnosperms). Here, we propose to narrow the term pseudanthium to multiflowered blossoms resembling zoophilous flowers in form, i.e. in being structurally subdivided in a showy periphery and a reproductive centre. According to this definition, pseudanthia sensu stricto evolved independently in at least 41 angiosperm families. The recurrent acquisition of pseudanthia sensu stricto in all major lineages of flowering plants indicates repeated interactions between developmental constraints (smallness of flowers, meristematic conditions) and selective pressures, such as demands of pollinators and/or environmental conditions.

Wood and bark structure in Buddleja: anatomical background of stem morphology

Bark (all tissues outside of the vascular cambium) has been extensively studied in recent years, especially its anatomy and physiology. Macromorphological bark characters can be important taxonomically for many plant groups, including the genus Buddleja (Scrophulariaceae). However, the relationship between macroscopic bark appearance and its microscopic structure remains obscure, hampering the use and interpretation of bark traits in plant taxonomy and phylogenetics as well as in other fields of botany. We studied micro- and macrostructure of bark in the species of Buddleja representing wide taxonomic and geographic diversity to identify general relationships between bark anatomy and morphology. We also examined Buddleja xylem and discussed the importance of anatomical traits for understanding the relationships between clades in this genus. The smooth bark surface in sect. Gomphostigma and the outgroup (Freylinia spp.) relates to the small number of periderms of superficial origin and limited sclerification. This allows for the retention of visible lenticels. In the rest of Buddleja, bark sloughs off and division of labour is present: collapsed phloem undergoes sclerification and acts as a protective layer, while thin-walled phellem forms the separation layers. A similar pattern is found in some groups (e.g. Lonicera), but in others (e.g. Vitis and the species of Eucalyptus with stringy bark), the pattern is inversed. Wood and bark anatomy supports a sister relationship between the southern African section Gomphostigma and the rest of Buddleja but is taxonomically uninformative among remaining clades. Limited development of periderms and sclerification allows for the retention of a smooth bark surface and conspicuous lenticels. Sloughing off of bark requires division of labour into a lignified protective layer and a thin-walled separation layer. These two functions are never served by a single tissue but are rather divided between phloem and periderm. How more subtle features (e.g. size and shape of fissures) are determined requires further study. Simultaneously, bark anatomy could be a useful source of data to complement molecular phylogenetic studies in a total evidence approach for systematics.

Exceptional evolutionary lability of flower-like inflorescences (pseudanthia) in Apiaceae subfamily Apioideae

PREMISE

Pseudanthia are widespread and have long been postulated to be a key innovation responsible for some of the angiosperm radiations. The aim of our study was to analyze macroevolutionary patterns of these flower-like inflorescences and their potential correlation with diversification rates in Apiaceae subfamily Apioideae. In particular, we were interested to investigate evolvability of pseudanthia and evaluate their potential association with changes in the size of floral display.

METHODS

The framework for our analyses consisted of a time-calibrated phylogeny of 1734 representatives of Apioideae and a morphological matrix of inflorescence traits encoded for 847 species. Macroevolutionary patterns in pseudanthia were inferred using Markov models of discrete character evolution and stochastic character mapping, and a principal component analysis was used to visualize correlations in inflorescence architecture. The interdependence between net diversification rates and the occurrence of pseudocorollas was analyzed with trait-independent and trait-dependent approaches.

RESULTS

Pseudanthia evolved in 10 major clades of Apioideae with at least 36 independent origins and 46 reversals. The morphospace analysis recovered differences in color and compactness between floral and hyperfloral pseudanthia. A correlation between pseudocorollas and size of inflorescence was also strongly supported. Contrary to our predictions, pseudanthia are not responsible for variation in diversification rates identified in this subfamily.

CONCLUSIONS

Our results suggest that pseudocorollas evolve as an answer to the trade-off between enlargement of floral display and costs associated with production of additional flowers. The high evolvability and architectural differences in apioid pseudanthia may be explained on the basis of adaptive wandering and evolutionary developmental biology.

Parallel evolution of arborescent carrots (Daucus) in Macaronesia

PREMISE

Despite intensive research, the pathways and driving forces behind the evolution of derived woodiness on oceanic islands remain obscure. The genus Daucus comprises mostly herbs (therophytes, hemicryptophytes) with few rosette treelets (chamaephytes) endemic to various Macaronesian archipelagos, suggesting their independent evolution. To elucidate the evolutionary pathways to derived woodiness, we examined phylogenetic relationships and the habit and secondary xylem evolution in Daucus and related taxa.

METHODS

Sixty taxa were surveyed for molecular markers, life history, and habit traits. Twenty-one species were considered for wood anatomical characters. A dated phylogeny was estimated using Bayesian methods. The evolution of selected traits was reconstructed using parsimony and maximum likelihood.

RESULTS

Daucus dispersed independently to the Canary Islands (and subsequently to Madeira), Cape Verde, and the Azores in the late Miocene and Pleistocene. Life span, reproductive strategy, and life form were highly homoplastic; the ancestor of Daucus was probably a monocarpic, biennial hemicryptophyte. Rosette treelets evolved independently in the Canarian-Madeiran lineage and in Cape Verde, the latter within the last 0.13 Myr. Treelets and hemicryptophytes did not differ in wood anatomy. Pervasive axial parenchyma in wood occurred more often in polycarpic rather than monocarpic species.

CONCLUSIONS

Life span and life form in Daucus are evolutionarily labile and may change independently of wood anatomy, which is related to plant reproductive strategy rather than to life form. Insular woodiness may evolve rapidly (as demonstrated in D. bischoffii), and in Daucus, it does not seem to be an adaptation to lower the risk of xylem embolism.