Comparative floral morphology and anatomy of Anacardiaceae and Burseraceae (Sapindales), with a special focus on gynoecium structure and evolution

Anacardiaceae and Burseraceae are traditionally distinguished by the number of ovules (1 vs. 2) per locule and the direction of ovule curvature (syntropous vs. antitropous). Recent molecular phylogenetic studies have shown that these families are sister groups in Sapindales after having been separated in different orders for a long time. We present a comparative morphological study of the flower structure in both families. The major clades, usually supported in molecular phylogenetic analyses, are well supported by floral structure. In Anacardiaceae, there is a tendency to gynoecium reduction to a single fertile carpel (particularly in Anacardioideae). The single ovule has a long and unusually differentiated funicle, which connects with the stylar pollen tube transmitting tract in all representatives studied. In Anacardiaceae–Spondiadoideae, there is a tendency to form an extensive synascidiate zone, with a massive remnant of the floral apex in the centre; these features are also present in Beiselia (Burseraceae) and Kirkiaceae (sister to Anacardiaceae plus Burseraceae) and may represent a synapomorphy or apomorphic tendency for the three families. In core Burseraceae, gynoecium structure is much less diverse than in Anacardiaceae and has probably retained more plesiomorphies. Differences in proportions of parts of the ovules in Anacardiaceae and Burseraceae are linked with the different direction of ovule curvature. © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 159 , 499–571.     

Caryophyllales: a key group for understanding wood anatomy character states and their evolution

Definitions of character states in woods are softer than generally assumed, and more complex for workers to interpret. Only by a constant effort to transcend the limitations of glossaries can a more than partial understanding of wood anatomy and its evolution be achieved. The need for such an effort is most evident in a major group with sufficient wood diversity to demonstrate numerous problems in wood anatomical features. Caryophyllales s.l., with approximately 12 000 species, are such a group. Paradoxically, Caryophyllales offer many more interpretive problems than other ‘typically woody’ eudicot clades of comparable size: a wider range of wood structural patterns is represented in the order. An account of character expression diversity is presented for major wood characters of Caryophyllales. These characters include successive cambia (more extensively represented in Caryophyllales than elsewhere in angiosperms); vessel element perforation plates (non‐bordered and bordered, with and without constrictions); lateral wall pitting of vessels (notably pseudoscalariform patterns); vesturing and sculpturing on vessel walls; grouping of vessels; nature of tracheids and fibre‐tracheids, storying in libriform fibres, types of axial parenchyma, ray anatomy and shifts in ray ontogeny; juvenilism in rays; raylessness; occurrence of idioblasts; occurrence of a new cell type (ancistrocladan cells); correlations of raylessness with scattered bundle occurrence and other anatomical discoveries newly described and/or understood through the use of scanning electron microscopy and light microscopy. This study goes beyond summarizing or reportage and attempts interpretations in terms of shifts in degrees of juvenilism, diversification in habit, ecological occupancy strategies (with special attention to succulence) and phylogenetic change. Phylogenetic change in wood anatomy is held to be best interpreted when accompanied by an understanding of wood ontogeny, species ecology, species habit and taxonomic context. Wood anatomy of Caryophyllales demonstrates problems inherent in binary character definitions, mapping of morphological characters onto DNA‐based trees and attempts to analyse wood structure without taking into account ecological and habital features. The difficulties of bridging wood anatomy with physiology and ecology are briefly reviewed. © 2010 The Linnean Society of London, Botanical Journal of the Linnean Society, 2010, 164 , 342–393.     

The distribution of circularly polarized light reflection in the Scarabaeoidea (Coleoptera)

The light reflected from many scarab beetles is strongly circularly polarized, a phenomenon that may be unique in nature. Approximately 1500 genera and more than 19 000 species/subspecies of scarab beetles (Scarabaeoidea) in the collection of the Natural History Museum, London, were surveyed with circularly polarizing filters to assess how widespread this effect is and how it relates to colour patterns in the individuals concerned. Over 1100 measurements were made of the strength of the circular polarization, in some cases registering 97%. Circular polarization had been previously described from five subfamilies of Scarabaeidae and, in the present study, was found in three further subfamilies: Phaenomeridinae, Dynastinae and Euchirinae, as well as in the subfamily Ceratocanthinae of the family Hybosoridae, comprising the first records outside the Scarabaeidae. The results may have implications for the taxonomy of the group, for visual discrimination, and for the behaviour of the species themselves. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100 , 585–596.     

Phylogeny of Veneroidea (Mollusca: Bivalvia) based on morphology and molecules

The largest Recent family of Bivalvia, the marine Veneridae with approximately 800 species, comprises one of the least understood and most poorly defined molluscan taxa, despite including some of the most economically important and abundant bivalves, for example quahog, Pismo clams, and Manila clams. A review of previous phylogenetic analyses including the superfamily Veneroidea (Veneridae, Petricolidae, Glauconomidae, Turtoniidae, Neoleptonidae) and within the Veneridae shows minimal taxon sampling leading to weak conclusions and few supported synapomorphies. New phylogenetic analyses on 114 taxa tested the monophyly of Veneroidea, Veneridae, and 17 nominal venerid subfamilies, using morphological (conchological, anatomical) data and molecular sequences from mitochondrial (16S, cytochrome oxidase I) and nuclear (28S, histone 3) genes. Morphological analyses using 45 exemplar taxa and 23 traditional characters were highly homoplastic and failed to reconstruct traditional veneroid classification. Full morphological analyses (31 characters) supported the monophyly of Veneroidea and Veneridae but only when certain taxa were excluded, revealing analytical difficulties caused by a suite of characters associated with neotenous or miniaturized morphology. Molecular analyses resulted in substantially higher clade consistency. The combined molecular data set resulted in significant support for a particular topology. The monophyly of Veneridae was supported only when Petricolidae and Turtoniidae were subsumed, and recognized as members with derived or neotenous morphologies, respectively. Morphological character mapping on molecular trees retained a high level of homoplasy, but revealed synapomorphies for major branch points and supported six subfamily groups (Dosiniinae, Gemminae, Samarangiinae, Sunettinae, Tapetinae, combined Chioninae + Venerinae). Glauconomidae and Neoleptonidae are provisionally maintained in Veneroidea pending further study; Petricolinae and Turtoniinae are placed in Veneridae. © 2006 The Linnean Society of London, Zoological Journal of the Linnean Society, 2006, 148 , 439–521.