Cover of Phylogeny and Evolution of the Angiosperms
Phylogeny and Evolution of the Angiosperms Revised and Updated Edition
by Douglas Soltis, Pamela Soltis, Peter Endress, Mark W. Chase, Steven Manchester, Walter Judd, Lucas Majure and Evgeny Mavrodiev
University of Chicago Press, 2018
Cloth: 978-0-226-38361-3 | Electronic: 978-0-226-44175-7
DOI: 10.7208/chicago/9780226441757.001.0001

AVAILABLE FROM

University of Chicago Press (cloth, ebook)
ABOUT THIS BOOKAUTHOR BIOGRAPHYREVIEWSTABLE OF CONTENTS

ABOUT THIS BOOK

Although they are relative latecomers on the evolutionary scene, having emerged only 135‒170 million years ago, angiosperms—or flowering plants—are the most diverse and species-rich group of seed-producing land plants, comprising more than 15,000 genera and over 350,000 species. Not only are they a model group for studying the patterns and processes of evolutionary diversification, they also play major roles in our economy, diet, and courtship rituals, producing our fruits, legumes, and grains, not to mention the flowers in our Valentine’s bouquets. They are also crucial ecologically, dominating most terrestrial and some aquatic landscapes.

This fully revised edition of Phylogeny and Evolution of the Angiosperms provides an up-to-date, comprehensive overview of the evolution of and relationships among these vital plants. Incorporating molecular phylogenetics with morphological, chemical, developmental, and paleobotanical data, as well as presenting a more detailed account of early angiosperm fossils and important fossil information for each evolutionary branch of the angiosperms, the new edition integrates fossil evidence into a robust phylogenetic framework. Featuring a wealth of new color images, this highly synthetic work further reevaluates long-held evolutionary hypotheses related to flowering plants and will be an essential reference for botanists, plant systematists, and evolutionary biologists alike.

AUTHOR BIOGRAPHY

Douglas Soltis and Pamela Soltis are distinguished professors in the Florida Museum of Natural History at the University of Florida. Together, they are coeditors of multiple books, including Polyploidy and Genome Evolution and, with Jeff J. Doyle, MolecularSystematics of Plants. Peter Endress is professor emeritus of botany at the University of Zurich. He is the author of Diversity and Evolutionary Biology of Tropical Flowers and coeditor of Early Evolution of Flowers. Mark Chase is former director of the Jodrell Laboratory at the Royal Botanic Gardens, Kew. He is the author of Orchids: The Pictorial Encyclopedia of Oncidium. Steven Manchester is curator in the Division of Paleobotany at the Florida Museum of Natural History and one of the leading paleobotanists in the world. Walter Judd is distinguished professor emeritus in the Department of Biology at the University of Florida and affiliate curator in botany at the Florida Museum of Natural History. He is coauthor of Plant Systematics: A Phylogenetic Approach. Lucas Majure is a biologist of New World succulents at the Desert Botanical Garden in Arizona. Evgeny Mavrodiev is an associate scientist at the Florida Museum of Natural History.

REVIEWS

“[One of the] most significant books on flowering plant phylogeny and systematics to come out in recent years. . . . This book belongs on the bookshelf of any serious plant systematist—it will be a valuable resource for years to come.”
— BioScience, on the first edition

“Takes a unique but much needed approach to describe the phylogeny, systematics, and evolutionary patterns of flowering plants. The volume represents a major systematic contribution. . . . The book’s goals are to review the molecular phylogeny of all flowering plants and use this information to inform systematics and our understanding of major evolutionary trends; in this it succeeds admirably. . . . I find it challenging to criticize this book: it is a really well‐organized and well‐presented compendium, filling an important niche. . . . It will be a valuable reference for every botanist and many ecologists, and compelling reading for anyone who works on plant morphology, systematics, and evolution. It will be an important source of ideas and phylogenetic information for future higher‐level systematic and comparative studies of flowering plants.”
— Quarterly Review of Biology, on the first edition

“An important and very helpful book and a must for every plant systematist.”
— Plant Systematics and Evolution, on the first edition

“No endnote to a finished project, but rather, a dynamic and synoptic state-of-the-union of the ongoing effort by many botanists. . . . A necessary reference. . . . It is encyclopedic in its treatment of the subject matter and the lengthy list of works cited makes it an excellent source book for anyone hoping to begin broad phylogenetic study.”
— Cladistics, on the first edition

“Excellent . . . highly recommendable for anyone involved in plant systematics.”
— Plant Science Bulletin, on the first edition

“The authors present scholarly and lucid discussions of topics ranging from parallel and convergent evolution to the evolution of genome size and base chromosome number. . . . Fascinating reading, not only for the answers it provides to many . . . curiosities about plant evolution, but also for the scholarly discussion of those questions still awaiting answers.”
— Taxon, on the first edition

TABLE OF CONTENTS


DOI: 10.7208/chicago/9780226441757.003.0001
[seed plants;gnetales;origin of flower;anthophyte;caytoniales;bennetitales;glossopterids;glossopterids]
Relationships among the lineages of extant seed plants, as well as the relationships of living groups to fossil seed plant lineages, have been issues of longstanding interest and debate. At some point, nearly every living and fossil group of gymnosperms has been proposed as a possible ancestor of the angiosperms. Numerous studies (molecular and molecular + morphological) have tried to resolve the relationships among living seeds plants--cycads, Ginkgo, Gnetales, conifers, and angiosperms. Several features typically are recovered. Angiosperms are sister to extant gymnosperms, and among gymnosperms, cycads or cycads plus Ginkgo are sister to the remainder. The placement of Gnetales remains unclear, but most recent investigations recover various placements with conifers. The closest relatives of angiosperms remain enigmatic, but several extinct lineages consistently emerge as part of a clade with angiosperms: Glossopteridales, Caytonia, Pentoxylon, Bennettitales. The origin of the flower remains a mystery, but some molecular data support the “mostly male” hypothesis, which provides a mechanism by which a plant with separate male and female reproductive structures, a feature of all gymnosperms, could produce a bisexual structure such as a flower.


DOI: 10.7208/chicago/9780226441757.003.0002
[origin of angiosperms;Archaefructus;early angiosperm fossils;divergence times;molecular dating]
Although the origin of angiosperms remains unclear, numerous new insights into Darwin’s “abominable mystery” have been obtained via contributions from paleobotany and molecular phylogenetics. We overview the contributions of the paleobotanical record to the age of the angiosperms. The oldest unambiguous angiosperm fossils are pollen grains (~131.8 mya) from the the Early Cretaceous. Molecular dating methods have contributed to our understanding of the age and the diversification of angiosperms. These methods tend to overestimate ages; recent molecular dating efforts have converged on age estimates for angiosperms between 180 – 140 Ma (late Jurassic), predating the fossil dates by only 48 to 8 Myr. Extinct members of several extant basal angiosperm clades were present in the Early Cretaceous. Hence, there is congruence between the general framework of angiosperm phylogeny and the early fossil record. Few early fossils can be unambiguously assigned to any extant group. The small size of many Early Cretaceous flowers generally agrees with the floral characteristics of the basalmost extant angiosperms in having small or moderate-sized flowers with few or a moderate number of parts. Two very different views on the general habit and habitat of the earliest angiosperms are reviewed: woody and terrestrial or herbaceous and aquatic.


DOI: 10.7208/chicago/9780226441757.003.0003
[angiosperm overview;parsimony;maximum likelihood;Bayesian;supertree;supermatrix;mega phylogeny;phylogenomics;diversification]
There is enormous interest in angiosperm phylogeny in diverse fields, including systematics, ecology, and molecular biology. In this chapter we overview recent efforts to reconstruct angiosperm phylogeny using morphology, RNA (in early studies) and particularly DNA sequence data, as well as combined morphology and DNA datasets. Valuable lessons in data analysis have also been learned, including the importance of taxon and gene (genome) sampling and methods of data analysis; these have general implications that we also briefly review. In addition, we overview the best estimate of phylogeny now available for the angiosperms and compare these relationships with traditional, morphology-based classifications from the late 1900s. The molecular estimates of phylogeny also provided the stimulus for a revised classification of angiosperms (the most recent version of which is covered in Chapter 12). The broad DNA-based topologies also provide evidence for multiple episodes of rapid rapid radiation in the flowering plants and have facilitated estimation of the ages of the major diversification events.


DOI: 10.7208/chicago/9780226441757.003.0004
[basal angiosperms;Amborella;Nymphaeales;Austrobaileyales;angiosperm character evolution]
We review early cladistics analyses of angiosperms based on morphology as well as DNA. A major contribution of DNA systematics has been to elucidate the basalmost lineages of angiosperms. In most recent DNA analyses Amborellaceae, Nymphaeales, and Austrobaileyales (the ANA grade) are subsequent sisters to all remaining extant angiosperms (Mesangiospermae). We review the general features of ANA grade members. One interest in determining the basalmost branches of angiosperms has involved attempts to use that information to reconstruct hypothetical features of early angiosperms. For example, Amborellaceae, as well as most Nymphaeaceae, and Austrobaileyales have carpels that are not sealed by postgenital fusion but by secretion —the carpel is essentially glued shut. In contrast, most Mesangiospermae have carpels that are postgenitally fused. The realization that Amborella is the sister to all other living angiosperms offers the ability to ‘root’ analyses of all angiosperm features, from gene families to genome structure to morphology, chemistry, and pollination. The recently sequenced Amborella genome has provided crucial new insights into the earliest angiosperm genomes while serving as an evolutionary reference for comparisons with other angiosperms. It is also noteworthy that Amborella possesses vesselless wood—the plants have only tracheids for water conduction.


DOI: 10.7208/chicago/9780226441757.003.0005
[Magnoliales;Laurales;Canellaceae;Piperales]
Magnoliids (Magnoliidae sensu Cantino et al.) consist of Magnoliales plus Laurales, sister to Piperales plus Canellales. Some analyses indicate they are sister to Chloranthales. Magnoliids + Chloranthales occupy a pivotal position – they may be sister to the eudicots + Ceratophyllaceae plus monocots. Likely synapomorphies of the magnoliids include the phenylpropane compound asarone, the lignans galbacin and veraguensin, and the neolignan licarin. Magnoliales and Laurales have rich fossil histories that date to the early Cretaceous. Chloranthales comprise the family Chloranthaceae with four genera—the placement of the group has long been problematic. Significantly, some of the earliest known fossil angiosperm material is similar to modern Chloranthaceae.


DOI: 10.7208/chicago/9780226441757.003.0006
[ancestral angiosperm;habit;nodal anatomy;carpel;stamen;pollen;perianth differentiation;endosperm;phyllotaxis]
The framework of angiosperm relationships now available affords the opportunity, with caveats, to reconstruct ancestral character states for extant angiosperms and to investigate character evolution. We offer a reconstruction of early angiosperm motifs based on the features of extant species with input from the fossil record. Whether the first angiosperms were forest shrubs (dark-and-disturbed hypothesis) or aquatic herbs (wet-and-wild hypothesis) remains unclear. The earliest angiosperms may have lacked vessels in the xylem. Early terrestrial angiosperms may have been successful because they exploited wet, understory forest environments more effectively than ferns and other seed plants. Flowers of the hypothetical early angiosperms were likely small; the perianth was likely undifferentiated (sepals), it is unclear if parts were spiral or whorled. Ancestral merosity is unclear, but our reconstructions point to an early and significant role of trimery in angiosperms. Stamens likely were undifferentiated into anther and filament regions; that is they were leaf-like (laminar); pollen was uniaperturate. The early carpel was ascidiate and sealed by secretion. Major changes accompanied the evolution of the mesangiosperms (Mesangiospermae). Well-developed vessel elements were apparent. Perianth phyllotaxy was whorled, and our reconstructions favor a trimerous ancestral state for mesangiosperms. Carpels became more or less plicate in several groups.


DOI: 10.7208/chicago/9780226441757.003.0007
[monocots features;fossil history;origin of monocots]
The monocots – Monocotyledoneae – comprise 52,000 species (22% of angiosperm species diversity) and are united by a suite of synapomorphies: herbaceousness, an embryo with a single cotyledon, parallel leaf venation, sheathing leaf bases, stems with scattered vascular bundles, adventitious roots, trimerous flowers, sieve cell plastids with cuneate protein crystals, sympodial growth, and of course, DNA sequences. Both the age of the monocots and the phylogenetic placement of the clade are uncertain. Most molecular-based estimates place the crown clade between 130 and 150 mya, although few reliable fossils older than 80-90 mya have been recovered. Because many analyses suggest a sister relationship between monocots and eudicots (which date to 125 mya; Chapters 8-11), monocots must be approximately the same age. Within the monocots, Acorus (Acorales) is sister to the rest of the clade, which is well understood phylogenetically. Subsequently branching clades include Alismatales, Petrosaviales, Pandanales + Dioscoreales, Liliales, Asparagales, and the large clade of commelinids, which comprise Arecales + Dasypogonales sister to a clade of Poales + (Zingiberales + Commelinales). Despite many features uniting monocots, this clade exhibits tremendous diversity in size, habit, habitat, and floral diversity, in particular, and patterns of character evolution are discussed.


DOI: 10.7208/chicago/9780226441757.003.0008
[Eudicots;Ceratophyllaceae;basal eudicots;Ranunculales;fossils;character evolution;perianth evolution]
The eudicots – Eudicotyledoneae – comprise ~75% of angiosperm species diversity, with a fossil record that dates back 125 million years based on the single morphological synapomorphy for this diverse clade; triaperturate (and triaperturate-derived) pollen. Ceratophyllaceae, a small, enigmatic clade of aquatic species, often appear as sister to the eudicots. Within eudicots, a grade of ‘early-diverging’ lineages subtends the core eudicots (Gunneridae), which encompass the vast majority of eudicot species. Ranunculales, sister to all other eudicots, comprise seven families and extensive morphological diversity. Subsequently branching lineages of early diverging eudicots include Proteales, a morphologically disparate clade of Proteaceae, Platanaceae, and Nelumbonaceae, which, prior to molecular phylogenetics, had never been considered closely related; Sabiaceae, which may be sister to Proteales; and Trochodendrales, comprising Trochodendron and Tetracentron, each with a long fossil history and together forming the sister to Buxales (with two families) and the core eudicots, which are discussed in greater detail in Chapters 9-11. The phylogenetic placement of these early-diverging eudicots is crucial for understanding patterns of character evolution, especially floral features, throughout the eudicots as a whole, and reconstructions of several key characters are presented.


DOI: 10.7208/chicago/9780226441757.003.0009
[Pentapetalae;Gunneridae;Dilleniales;Gunnerales]
The core eudicots – Gunneridae – represent ~70% of angiosperm species diversity. Although no morphological or anatomical synapomorphies have been identified for this clade, strong molecular evidence supports its monophyly. Within Gunneridae, Gunnerales, consisting of two morphologically disparate families not considered close relatives until molecular phylogenetics, are sister to the large, species-rich Pentapetalae, which comprises the enigmatic Dilleniales and two large clades, Superrosidae and Superasteridae, each with over one-quarter of all angiosperm species diversity. The phylogenetic placement of the morphologically distinct Dilleniales remains unclear, despite intensive analysis, alternatively placed as sister to Superasteridae, Superrosidae, or Superasteridae + Superrosidae. Character transitions early in the evolution of Gunneridae are particularly labile, perhaps reflecting apparent rapid radiations that spawned most major clades of Pentapetalae (113-123 mya) and of Superrosidae (97-105 mya) and Superasteridae (104-117 mya). Shifts in chemistry, including the production of ellagic and gallic acids, occurred near the origin of the core eudicots, as did a pair of closely placed whole-genome duplications, although the precise locations of these events are uncertain. By contrast, the origin of the pentamerous, highly synorganized flower clearly arose in the common ancestor of Pentapetalae.


DOI: 10.7208/chicago/9780226441757.003.0010
[Superrosidae;Saxifragales;rosid;nitrogen fixing symbiosis;terrestrial dominant;co-speciation;glucosinolates]
Superrosidae, a recently recognized and defined clade, comprises Saxifragales, Vitales, and rosids (Rosidae), although the relationships among these component clades remain uncertain. Saxifragales are a diverse clade of 15 families that were considered distantly related prior to molecular phylogenetics. Despite the relatively small size of Saxifragales (~2500 species), this clade exhibits tremendous diversity in life form: trees, shrubs, lianas, annual and perennial herbs, succulents, and aquatics. Likewise, Saxifragales are highly diverse in floral characters. Vitales (Vitaceae, including Leeaceae) are sister either to Saxifragales or to the large rosid clade, which, with over 70,000 species, comprises nearly all of the species diversity of Superrosidae and represents ~one quarter of all angiosperm species. The 140 families of rosids form two clades, fabids (Fabidae) and malvids (Malvidae). The rosids are important ecologically, having radiated rapidly to form the majority of the world’s angiosperm-dominated forests and providing organically usable nitrogen through symbiotic relationships with nitrogen-fixing bacteria. Rosids also include many important crop plants and model organisms, such as Arabidopsis thaliana and other Brassicaceae.


DOI: 10.7208/chicago/9780226441757.003.0011
[Superasteridae;asterid;Santalales;Caryophyllales;Berberidopsidales;betalains]
Superasteridae, with over 90,000 species, represent minimum one-third of angiosperm species, assembled in four main clades: Santalales, Caryophyllales, Berberidopsidales, and Asteridae, the latter with nearly 80,000 species. Santalales are parasitic, either holoparasites (obligate parasites) andor hemiparasites (gaining water and minerals from their hosts), and are united by their lack of root hairs as well as several other features not related to parasitism. Berberidopsidales comprise two small families previously considered to be distantly related. Caryophyllales are now broadly defined, expanded from the Centrospermae of the mid-1900s to include 31 families divided into Caryophyllineae and Polygonineae. Synapomorphies for Caryophyllales include simple perforation plates and a specific sequence of anther wall development. The large Asteridae, recognized over two centuries ago on the basis of fused corollas, are further united by a single integument, a thin-walled megasporangium (nucellus), and the presence of iridoids and tropane alkaloids. Included in this clade are Cornales (seven families), Ericales (22 families), and the euasterids, the latter diagnosed by epipetalous stamens that equal the number of corolla lobes and a gynoecium of two fused carpels. The two clades of euasterids, lamiids (Lamiidae[PS1]) and campanulids (Campanulidae) comprise most of the species diversity of Asteridae and are clearly differentiated.


DOI: 10.7208/chicago/9780226441757.003.0012
[classification;angiosperm phylogeny group;phylocode;rank free classification]
Improved inferences of phylogeny provide opportunities for enhanced understanding of character evolution. Particularly valuable are re-evaluations of those characters considered to be either particularly labile or particularly stable: are hypotheses of character evolution based on previous classifications supported by current phylogenies, and what can our new phylogenies tell us about the processes that generate or constrain evolutionary novelty? In contrast to features such as glucosinolate production and the development of symbioses with nitrogen-fixing bacteria, each of which was considered to be highly labile but in fact has more restrictive origins, many characters are truly extremely labile. In this chapter, three characters that are prone to parallel or convergent evolution – parasitism, carnivory, and C4 photosynthesis – are considered to investigate what new inferences of phylogeny say about the lability of these features. Each character is described, its phylogenetic distribution is summarized, and clade-specific variants are noted.


DOI: 10.7208/chicago/9780226441757.003.0013
[parasitism;carnivory;C4 photosynthesis]
Clarification of phylogenetic relationships at all levels across the angiosperms tree of life has had broad implications, facilitating, for example, a more accurate assessment of character evolution. These topologies have provided insights into the evolution of key characters. Several morphological and chemical features long thought to have evolved many times following traditional classifications are now established as having arisen once or only a few times. Glucosinolate production, a chemical defense long thought to have evolved recurrently, was shown to have evolved only twice (in Brassicales and again in Putranjivaceae). Similarly, symbiosis with nitrogen-fixing bacteria was thought to have evolved many times, but molecular analyses revealed instead that some set of predisposing traits for this character evolved only once, in the ancestor of the nitrogen-fixing clade. However, within this one relatively small clade, nitrogen-fixing symbiosis may have evolved and/or been lost several times. Here we overview three features that are highly prone to parallel or convergent evolution: parasitism, carnivory, and C4 photosynthesis. Molecular phylogenetic investigations have helped to clarify the extent of parallelism or convergence of these traits and also have provided additional evolutionary insights.


DOI: 10.7208/chicago/9780226441757.003.0014
[floral diversification;merism;phyllotaxy;zygomorphy;symmetry;developmental genetics;canalization]
Angiosperms exhibit spectacular diversity in floral form, with profound implications for pollinator relationships and pollen, seed, and fruit dispersal. Recent achievements in studies of floral morphology, development, genetics, and angiosperm phylogeny have led to new appreciation of both the evolutionary origins and genetic bases of floral novelties. Key innovations – major evolutionary shifts that enable species diversification – in floral features have spurred angiosperm diversification, often in concert with associated pollinators. Prominent among such innovations are shifts to synorganization, the fusion or close association of floral structures, resulting in sympetaly, syncarpy, and adnation of stamens and petals. Floral features experience differential evolutionary constraints, and aspects of floral form related to floral organization, construction, and mode are decreasingly conserved, such that general structure of flowers (the positions of floral whorls, for example) are more conserved than inter- and intraspecific polymorphisms (such as features related to pollinator differences). The genetic underpinnings of many such features are now understood, allowing for synthetic views of floral morphology from both developmental and evolutionary perspectives. This chapter also reviews broad patterns of variation in major floral structures across angiosperms.


DOI: 10.7208/chicago/9780226441757.003.0015
[genome size;transposable elements;genome downsizing;concerted evolution;nuclear genomes;organellar genomes]
We overview the evolution of genome size, dynamics and mechanisms of genome size evolution, and resulting variation in genome content and structure. We also discuss the role of repetitive elements in genome size dynamics and horizontal gene transfer (HGT) as a source of variation in gene content. Polyploidy is briefly discussed as a basic component of genome size variation and evolution, as is genome downsizing. Although there is extensive variation in genome size, the ancestral genome size of angiosperms is very small. Ancestral genome sizes are also very small for most major clades of angiosperms. Genome size evolution across land plants and within angiosperms was dynamic; dramatic changes can be seen just within monocots. The grass and mustard families are used as models to illustrate tendencies in genome evolution. Extensive synteny across long periods of angiosperm evolutionary history is an important discovery from the comparison of sequenced plant genomes. Complete sequencing of the nuclear genome of Amborella, sister to all other angiosperms, has permitted reconstruction of the genome and gene content of the ancestral angiosperms. Increasing numbers of studies have indicated that all three plant genomes have experienced HGT; it may be much more common than previously appreciated.


DOI: 10.7208/chicago/9780226441757.003.0016
[polyploidy;genome doubling;whole genome doubling;methylation;genome shock;diploidization;fractionation]
Polyploidy or whole genome duplication (WGD) is now recognized as not just a process restricted largely to plants, but as a significant evolutionary force in all eukaryotes. Genomic tools have made it possible not only to obtain novel insights into polyploidy as a process, but also more precise estimates of the timing and frequency of ancient polyploidy in the ancestry of extant species. Genomic investigations revel that polyploidy is ubiquitous among angiosperms and indicate the presence of many major ancient WGD events. WGD is associated with the origin of the angiosperms, eudicots, monocots, Asteraceae, Brassicaceae, and Poaceae, The question is no longer what percent of angiosperms are polyploid, but how many WGD events occurred in any given lineage. A number of ancient WGD events are associated with a burst in species number. However, in many cases polyploidy was not immediately associated with a burst in species—this occurred several nodes later. An important genome processes that occurs following polyploidy is genome fractionation—the loss of genes from one or the other parental genome. Broad genomic comparisons indicate a consistent pattern of gene loss/retention following polyploidization. There may be basic rules that govern the fate of duplicated genes following WGD.

Book published by University of Chicago Press, 2018. Page maintained by BiblioVault.