[ Paleobotany of Angiosperm Origins ]
JOHN M. MILLER, PH.D.
Having discussed the origins of angiosperms from shrub-like Carboniferous or Permo-Triassic seed plant stock I outline and discuss the biodiversity and paleontology of extinct Paleozoic vascular plants and their phytophagous insect associates, which is necessary to solve the riddle of angiosperm beginnings within a coevolutionary and phylogenetic context.
The previous essay on the origin of angiosperms concluded that molecular evolution of invertebrate hemocyanin enzymes and their derived insect hexamerins was probably driven by the rise and fall of oxygen in the Earth's atmosphere (Burmester 2004) at two intervals during the Paleozoic Era. Phytophagous insects may have used oxygen-generating vegetation of hypoxic Paleozoic times as a source of food and for shelter from cold and ultraviolet radiation (Labandeira 2006).
Evidently, 200 to 300 million year old gene duplications in Carboniferous, Permian, or Triassic seed plants (Zahn et al. 2005, Soltis et al. 2007) were necessary as a building scaffold for developmental recombination and evolution of innovative morphologies such as developmentally labile bisexual cone axes (Baum and Hileman 2006, Theißen and Melzer 2007).
Further, I proposed that trampling and secreting insects on developmentally labile bisexual cone axes might have affected the evolution of floral homeotic protein dimers and quartets in certain Paleozoic plants. Finally I proposed that phytoecdysones secreted by Permo-Carboniferous and Permo-Triassic shrub lifeboats potentially affected body size and moulting time of phytophagous insect antagonists.
I also suggested that cladogenesis of flowering plants may be traced back in geologic time to the end-Permian extinction, and to surviving remnants of already divergent Permian seed plant lineages. Coevolving insect antagonists of monopodial Permo-Triassic seed plants potentially inhabited massive shoot-apical meristems (SAMs), cone axes, crevices among leaf bases and bark, and in wood.
Pteridosperms (seed ferns) are discussed in the next chapter as possible antecedents of angiosperms while drawing attention to possible coevolution of phytophagous insects with developmentally plastic Carboniferous and Permo-Triassic vascular plant stock.
As global atmospheric oxygen levels plummeted during the late Devonian-early Carboniferous ice-house (DeCARB), invertebrate hexamerin food storage and moulting proteins diverged in Plecopteran (stonefly) ancestors leading to several hemimetabolous and holometabolous clades of insects (Hagner-Holler et al. 2007).
The Carboniferous Period, well-known for its coal swamp flora and fossil insects, is also the time when extrabasinal habitats "may have served as cradles for major evolutionary innovations..." (p. 1077, Rothwell et al. 1996).
Innovative, monopodial seed plants of unstable extrabasinal paleoenvironments might have included gigantopterids (Mamay et al. 1988) and Vojnovskyales (Rothwell et al. 1996).
Is there convincing fossil evidence of the first flowering plants or their antecedents? No.
The image above is the distal end of the largest known specimen of Delnortea abbottiae (United States National Museum [USNM] specimen number 387473), photographed by the author in 1982 on the same day the fossil was unearthed from beds of the Lower Permian Road Canyon Formation of the Del Norte Mountains of southwestern North America. It is published as Figure 18 on page 1413 of Mamay et al. (1988).
A first clue to the mysterious origin of angiosperms comes from geochemical studies of oleanone triterpanes and other biomarkers (Moldowan and Jacobsen 2003). Oleananes are found in source bedding planes of Paleozoic rocks together with gigantopterid seed ferns. A weak oleanane signal was found in Carboniferous rocks but the plant that produced the molecular tracer is unknown (D. W. Taylor et al. 2006).
Gigantopterids are known from Carboniferous and Permian leaf compression floras. Permineralizations of reproductive material are rare and require detailed study. Whole plant morphology of gigantopterids is unknown. Surprisingly, oleananes are also found in Jurassic bennettitaleans and Cretaceous "dicotyledonous" flowering plants (D. W. Taylor et al. 2006).
Are oleonone triterpanes diagenetic products of phytoecdysones and terpenes including iridoids produced by phytophagous insect associates of gigantopterids, bennettitaleans, and angiosperm eudicots?
Despite these considerable gaps in our knowledge of Paleozoic gigantopterids and solid geochemical evidence that points toward a possible evolutionary connection with bennettitaleans and "dicotyledonous" flowering plants one recent paper analyzing leaf impressions and compressions by Glasspool et al. (2004) states:
"Additional similarities [to eudicot flowering plants] in the presence of water conductive vessels capable of sustaining similar physiological conditions, lead us to consider gigantopterids to be vegetative analogues of angiosperms."
The above quotation is from page 105 of Glasspool et al. (2004), Foliar physiognomy in Cathaysian gigantopterids and the potential to track Palaeozoic climates using an extinct plant group, Palaeogeography, Palaeoclimatology, Palaeoecology 205: 69-110. Remarks in brackets [] are mine.
Glasspool and coworkers state in a second paper defining the taxonomic concept of gigantopterid:
"These 'angiospermous' features [leaf size and shape, organization of the stele, and presence of vessels] have led to previous evolutionary scenarios suggesting angiosperm derivation from gigantopterid origins (Asama 1982) although these are now largely discounted (e.g. Doyle 2000) and most likely represent large-scale convergence in vegetative morphology and physiology."
The preceding statement is quoted from pages 1339 and 1340 of Glasspool et al. (2004), Defining the gigantopterid concept: a reinvestigation of Gigantopteris (Megalopteris) nicotianaefolia Schenck and its taxonomic implications, Palaeontology 47(6): 1339-1361. Remarks in brackets [] are mine.
I disagree with the opinions of Glasspool and coworkers. Simply put, detailed study of polished thin-sections of permineralized fertile gigantopterid plant material is required to justify considerations of the magnitude stated by these workers.
Why would Permian gigantopterids yield the same oleanone triterpanes as Cretaceous "dicotyledonous" angiosperms and bennettitaleans?
Paleobotanical Challenges and Evidence:
Now to the questions at hand which are critical but obviously unanswerable by studying the nucleic acid biochemistry, anatomy, morphology of extant flowering plants, researching the ecology of extant basal angiosperms, and drawing conclusions from statistically robust phylogenies based on analysis of possibly incorrectly scored, homoplasious ancestral seed plant characters.
What were the whole plant morphologies and architectural innovations of Carboniferous and Permo-Triassic seed plants?
Can we identify adaptations (anatomic, biochemical, physiologic, and morphologic) of Paleozoic gigantopterids and other enigmatic groups possibly exploited by Permo-Triassic animals and other organisms?
Based upon data to be gathered on the stratigraphic distribution of oleananes and plant fossils, did Paleozoic seed plants manufacture and use steroids, resins, and terpenoids as defense against herbivores?
If supported by studies of trace molecules in permineralizations of Paleozoic seed plants, could phytoecdysones potentially signal the hemocyanin, juvenile hormone (JH), and vitellogenin developmental tool kit of invertebrate antagonists residing in shrub lifeboats?
Are oleonone triterpanes diagenetic products of phytoecdysones and terpenes (including iridoids produced by phytophagous insect associates)
Do vojnovskyalian fossils yield oleananes?
As ascertained by analysis of geochemical/isotopic data from studies of carbonates and fossil soils, and biostratigraphic evidence, was it likely that oxygen-generating Permo-Triassic seed plants existed in hypoxic, carbon dioxide-, and methane-rich terrestrial paleoenvironments, coevolving with respiring colonies of intimate invertebrate mutualists and populations of physiologically-stressed tetrapods?
Could coevolving insect colonies inhabiting Permo-Carboniferous and Permo-Triassic shrub- or tree-like "lifeboats" environmentally or epigenetically affect the genomics of shoot apical meristems (SAM's) and accessory fertile meristems leading to developmental recombination via reproductive modules, genetic accommodation, and the eventual spread of adaptive phenotypes in pre-angiospermous populations?
What were the insect groups of Herbivore Expansion Phase 2 that exploited Carboniferous and Permian seed plants?
Can we compute dendrograms of phytophagous insects, which are calibrated and complimentary to phylogenetic reconstructions of specific seed plant host lineages?
Does a phylogenetic signal left behind by the original inhabitants of Permo-Triassic "shrub lifeboats" exist as ascertained through study of phytophagous insect antagonists of extant seed plants?
Pending discovery and paleobotanical study of additional permineralized reproductive material of Paleozoic gigantopterids and vojnovskyalians can we revise phylogenetic reconstructions of extinct seed plant lineages to better understand the Permo-Triassic roots of early angiosperms?
Reconstruction of the tree of life is an important goal of systematists, including paleontologists who study evolutionary relationships of insects and plants, among other groups of organisms. The reader is directed to Volume 91, Number 10 of the American Journal of Botany, which is devoted to this topic. Teachers, researchers, and students are directed to a detailed educational website: The Tree of Life Project, compiled by colleagues Christine Edwards, Doug Soltis, and Pam Soltis.
To aid research specialists in plant biology and paleontology (and writers of text books in botany), the Angiosperm Phylogeny Group has published an updated classification (APGII) for the flowering plants (Angiosperm Phylogeny Group 2003). Concomitantly, Professor Peter F. Stevens and the Missouri Botanical Garden maintains a web page, which is periodically updated based on published work by phylogenetic systematists.
There are many data sets available in the literature and anyone with standard software such as MACLADE or PAUP, among other programs, can reenter the data and generate phylogenetic trees using a personal computer. The data matrix on page 209 of J. A. Doyle (2006) is one such example. The graphic below represents the most parsimonious molecular constraints dendrogram of selected seed plant genera, families, and orders, including informal groups of species based on 339 steps, redrawn from an analysis published by J. A. Doyle (2006).
Several taxa depicted in J. A. Doyle's (2006) tree of seed plants (redrawn below) may be unfamiliar to the reader. The genus Elkinsia is a seed fern incertae cedis (Serbet and Rothwell 1992). Lyginopteris is classified in the seed fern order Lyginopteridales. The common name medullosans is a convenient way to group two extinct Paleozoic seed ferns: Quaestora and Medullosa.
Some of the seed fern groups are colored indigo brown on the dendrogram. Corystosperms and glossopterids appear in blue typescript. Bennettitaleans are denoted by the reddish-brown type-face. Pentoxylians are colored pink on the chart. Gnetophytes, once regarded as a sister group to flowering plants (J. A. Doyle and Donoghue 1986, 1987; Donoghue and J. A. Doyle 2000) are displayed as purple letters. Cycadales and Caytonia are shown on the graphic in green letters. Common groups of conifers and the ginkgos appear in brown type.
The preceding graphic is redrawn from Fig. 11 on page 186 of J. A. Doyle (2006), Seed ferns and the origin of angiosperms, Journal of the Torrey Botanical Society 133(1): 169-209.
Other details used in J. A. Doyle's 2006 phylogenetic studies, including application of statistically-based "bootstrap" confidence analysis and a discussion of taxa and characters, may be found on pages 202-208 of the original paper. The tree of seed plants shown above is one of the more precise and parsimonious trees calculated and displayed to date, but can only be regarded as preliminary.
Several possibly critical seed plant groups are missing from nearly all evolutionary trees published in the literature because fossilized reproductive material is unavailable or unstudied (D. W. Taylor et al. 2006). Detailed paleobotanical studies of the anatomy and morphology of obscure, poorly studied Paleozoic and Mesozoic gymnosperms are needed to fill-in the considerable gaps in existing data sets before phylogenetic analyses can be calibrated with minimum fossil ages.
The reader may navigate away from this page to other sections of the discussion, by following the underlined hot links in the bulleted list below. I now discuss the paleobotany of seed plant orders, which are relevant to the fossil history and early cladogenesis of angiosperms. Two of the seed plants groups listed below, Phasmatocycas and Sanmiguelia are not placed in a specific taxonomic order. The major taxonomic orders of seed plants (except flowering plants) are:
Bennettitales
Callistophytales
Caytoniales
Coniferales
Cordaitales
Corystospermales
Cycadales
Czekanowskiales
Gigantopteridales
Ginkgoales
Glossopteridales
Gnetales
Hydrospermales
Lagenostomales
Nilssoniales
Phasmatocycas
Peltaspermales
Pentoxylales
Petriellales
Sanmiguelia
Taxales
Trigonocarpales
Vojnovskyales
Voltziales
Callistophytales, Czekanowskiales, Ginkgoales, Nilssoniales, and Petriellales may be important to include in the essay at a later date: gymnosperm groups not underlined in the preceding list are omitted from the discussion. I now discuss the fossil history of certain seed plant orders needed to understand the origins of angiosperms.
Gigantopteridales:
One of the missing seed plant groups in the phylogenetic tree pictured above is an abundant and morphologically diverse order of Carboniferous and Permian seed ferns called gigantopterids. Little is known of the whole plant morphology and reproductive anatomy of these enigmatic Permian seed plants. Presence of woody leaf midribs and cuticles on large leaves up to a meter in length (Mamay et al. 1988), and sparse reproductive material (X. Li and Z. Yao 1983), suggests that certain gigantopterids are probably gymnosperms.
The place and time to begin a survey of gigantopterids is in Cathaysian rocks of the Permian Period of China (H. Xilin et al. 1996). The Upper Paleozoic Era consists of two geologic periods: Carboniferous (Mississippian and Pennsylvanian) and Permian. In Paleozoic times gigantopterids were a diverse group of probably unrelated vascular plants constituting one of several terrestrial vegetation types of the biogeographic provinces of Angara, Cathaysia, and North America (Read and Mamay 1964, X. Li and Z. Yao 1982, Mamay et al. 1984).
The kodachrome to the right is a nearly complete leaf of Delnortea abbottiae (USNM 364419). The image was captured on film the day the rock slab was unearthed and collected.
Gigantopterids had several morphologic features remarkably similar to certain bennettitaleans, gnetophytes, and certain modern "dicotyledonous" flowering plant genera (e.g. Guettarda and Psychotria [Rubiaceae, Rubiales, Asteridae]), including woody midribs, erect vernation, abscission layering at the base of petioles, vessels, and secondary xylary elements, including libriform fibers (X. Li and Z. Yao 1983, Mamay et al. 1988, H. Li and Tian 1990, H. Li et al. 1994, H. Li et al. 1996, H. Li and D. W. Taylor 1998, 1999). Reproduction in gigantopterids is incompletely known (X. Li and Z. Yao 1983).
Products of diagenesis of secondary plant defense compounds in some preserved gigantopterid material include oleananes (see above). Oleonone triterpanes have been isolated from several gigantopterid leaf specimens, fossilized bennettitalean foliage, pyrolized extant angiosperm plant material, and from leaf compressions of flowering plants (D. W. Taylor et al. 2006).
Were these seed plants forerunners of anthophytes (including flowering plants) or eudicot angiosperm analogs that diminished in numbers at the close of the Permian Period?
Known gross morphological characters of gigantopterids, and a list of phytophagous animal associates, references, and future research needs are summarized below.
WHOLE PLANT MORPHOLOGY--unknown, possibly shrub- and/or vine-like, stems possess a bifacial cambium with vessels (H. Li and D. W. Taylor 1998, 1999); additional paleontologic data are needed to reconstruct whole plants and nodal anatomy
REPRODUCTIVE MODULES--inadequate data, possibly phyllospermous (X. Li and Z. Yao 1983, Mamay et al. 1988), details of ovule attachment to laminar microsporophylls are needed, internal anatomy of ovules unknown, anatomy of microsporangia and pollen are unknown; anatomical and developmental details of sexual reproduction unknown, permineralizations are in need of discovery and study
LEAVES--dicot or Gnetum-like with simple with flaring petioles, four orders of reticulate venation, woody midribs, and abscission zones (Mamay et al. 1988); or compound pinnate with spines, reticulate venation, waxy cuticles, sclerenchyma, secretory ducts, and certain conducting tissues. Attachment details of leaves to stems are unclear (Z.-Q. Yao and P. R. Crane 1986, H. Li and Tan 1990, H. Li et al. 1994, H. Li et al. 1996, H. Li and D. W. Taylor 1998, H. Li and D. W. Taylor 1999, Z.-Q. Wang 1999, Glasspool et al. 2004, Z.-Q. Yao and Liu 2004).
Despite the abundance of Cathaysian gigantopterid compressions and impressions summarized by X. Li and Z. Yao (1982), X. Li and Z. Yao (1983), H. Xilin et al. (1996), Z.-Q. Wang (1999), and Glasspool et al. (2004), study of heretofore undiscovered permineralized leaf fossils would yield more useful information
PHYTOPHAGOUS ASSOCIATE(S)--arthropods belonging to the Caloneurodea, Orthoptera, Protorthoptera (Beck and Labandeira 1998); vertebrate coprolites require discovery and study, known leaf bite- and chew marks need analysis, more fossils are needed for study
PLANT IDENTIFICATION--Cathaysiopteris, Gigantopteridium, Zeilleriopteris (Labandeira 1998); Gigantonoclea hallei, Gigantonoclea lagrelii (Glasspool et al. 2003)
A taxonomic monograph of gigantopterids is needed pending a concerted multi-national funding effort to find, secure, and study permineralized reproductive material
HOST SEED PLANT ORGAN(S) BEING EATEN--leaves (Beck and Labandeira 1998, Glasspool et al. 2003), permineralizations of plant tissue, tetrapod coprolites, and insect frass require discovery and study
Delnortea and American gigantopterids. Permian rocks of North- and South America yield several species of gigantopterids including Cathaysiopteris yochelsonii, Delnortea abbottiae (Mamay et al. 1986), Evolsonia texana (Mamay 1989), Gigantonoclea sp. (Mamay 1986, 1988), Gigantopteridium americanum, and Zeilleropteris wattii. However, the late Paleozoic floral zones of the Permian of the western hemisphere including Alaska (Mamay and Read 1984), though dominated at least in part, by Gigantopteridaceae, are paleofloras without Gigantopteris (Mamay et al. 1988).
Large leaf compressions and permineralizations including some possibly fertile material of Lower Permian (Leonardian) plants were described about 20 years ago (Mamay et al. 1986, 1988). Delnortea abbottiae is now known from North American and South American sedimentary beds (Ricardi et al. 1999).
Mamay's suggestion that the stratigraphic occurrence of Delnortea in Upper Leonardian rocks may lead to a better understanding of Permian floral zones is supported by discovery of Delnortea from the Artinskian of northwestern South America (Ricardi et al. 1999).
The notion of Delnortea as a widespread and common Pangaean floristic element of the Lower Permian of North America (Mamay et al. 1984) is reinforced by a startling discovery of Delnortea leaf fragments and other plant fossils in core samples of Permian (Leonardian, Clear Fork) dolomite recovered from three wells drilled more than 2000 meters deep through the thick sediments of southwestern North America (DiMichele et al. 2000).
The image to the left is a nearly complete fossil leaf of the holotype of Delnortea abbottiae (USNM 364416), photographed by the author a few days after the fossil was unearthed from beds of the Lower Permian Road Canyon Formation, Del Norte Mountains, southwestern North America.
Following the discovery and anatomical studies of Delnortea abbottiae by Mamay and co-workers between 1984 and 1988, paleontologists reported additional anatomical findings that demonstrated dicot-like leaf anatomy and vessel elements in several Cathaysian gigantopterids (H. Li et al. 1994, H. Li et al. 1996, H. Li and D. W. Taylor 1998, H. Li and D. W. Taylor 1999).
Having leaves up to a meter long the form of the whole Delnortea abbottiae plant is a mystery. One specimen yielded ovoid lumps on the distal edge of the lamina, possibly ovules, but more permineralized material is needed to shed light on these structures. However, microscopic study of limonitic permineralizations of Delnortea abbottiae reveal a pattern of secondary growth from a vascular cambium; a developmental syndrome often seen in seed plants (Mamay et al. 1988).
Permian standard section: depositional environments. In the folded and overthrust mountain belts of southwestern North America are classic exposures of a nearly complete sequence of Permian rocks. Exposures consist of uplifted and overthrusted Paleozoic rocks of the Marathon Fold Belt, Del Norte, and Glass Mountains. Exposures in the Del Norte and Glass Mountains comprise the standard type section through rocks of Permian age.
While the invertebrate fauna of the uplifted and exposed Glass Mountains of western North America is well-known in the classic work of Cooper and Grant (1972), the terrestrial, transitional (deltaic), and marine depositional environments of the Del Norte Mountains, which contain the Permian (Leonardian) Delnortea florule are less well understood.
Cretaceous limestones overtopping a 200 m thick bed of limestone pebbles, dolomite, quartz, and shales comprising the Triassic Bissett Conglomerate (Jurassic rocks are eroded away); together with underlying Permian conglomerates, limestones, shales, and siltstones; and Paleogene volcanic intrusions, are prevalent in these rugged mountains, including significant deposits of lead and hematite that have been mined in the last century (Barnes 1982).
The late Sergius H. Mamay, Ph.D. of the United States National Museum is pictured to the left standing on the fossiliferous upper members of the Leonardian Road Canyon Formation of the Del Norte Mountains of southwestern North America.
Exposed Permian rocks of the Del Norte Mountains include marine and transitional, deltaic sediments of the Permian Wolfcampian, Leonardian, Wordian, Guadalupian, and Ochoan Ages (Wardlaw et al. 2000). The known delnortea beds are in the lowermost Road Canyon or uppermost Cathedral Mountain Formation (Rohr et al. 1987). Dating of the layers is supported by paleontological evidence from conodont fossils (Wardlaw et al. 1990, Wardlaw 2000) and fusulinids (Yang and Yancey 2000). The fifteen Upper Paleozoic North American floral zones are described by Read and Mamay (1964).
The image below-left is a rock slab exfoliating from the Upper Members of the Road Canyon Formation. A private mapping party in 1981 discovered this slab along a wild game trail in the Del Norte Mountains. The in situ slab pictured to the left shows overlapping leaf compressions of Delnortea abbottiae and Taeniopteris. The center-left kodachrome is of two fossilized seeds, assignable to Cordaicarpus or Samaropsis (Mamay et al. 1984).
The scanning electron micrographs on the right-hand side of this page show the arrangement of permineralized tracheids and ray parenchyma cells (far right image) of a fragment from the Dadoxylon log found exfoliating from graded conglomerates in the upper geologic section of the delnortea beds (center-right image).
Cathaysian gigantopterids. Cathaysian gigantopterids were distinct from glossopterids, extinct Permo-Triassic seed plants whose stratigraphic distribution has been used to support Wegener's Theory of Continental Drift. In 1982 X. Li and Z. Yao reviewed the work to that date on the Cathaysian flora in Asia. A more recent compilation is available in H. Xilin et al. (1996) that summarizes research on the Permian coal floras of Jiangxi Province, China.
Gigantopterids from Permian rocks of Asia were first described by Schenck as Megalopteris nicotianaefolia from poorly preserved fossil impressions (Glasspool et al. 2004). The 16 morphotype genera indigenous to Asian Paleozoic rocks are Aculeovinea, Cathaysiopteridium, Cathaysiopteris, Cardioglossum, Gigantonoclea, Gigantonomia, Gigantopteridium, Gigantopteris, Gigantotheca, Gothanopteris, Linophyllum, Neogigantopteridium, Palaeogoniopteris, Progigantopteris, Vasovinea, and Zeilleropteris (H. Li et al. 1994).
Fossilized connections of the leaf impressions with whole plants and reproductive structures of gigantopterids are exceedingly rare or undescribed. Permian Gigantopteris of China unearthed and studied by X.-G. Li and Z.-Q. Yao in 1983, yield a rare glimpse of fertile material: ovules and pollen-bearing organs were attached to leaves and leaf-midribs, but the anatomy and placement of the connections is indeterminate. A relatively recent report of reproductive structures found preserved in a bedding plane in close association with gigantopterid leaves only adds to the mystery of these plants and their gymnosperm associates (Mei et al. 1992).
Permineralized gigantopterid foliage and stems belonging to Aculeovinea yunguiensis, Gigantonoclea guizhouensis, and Vasovinea tianii (H. Li et al. 1994, Z.-Q. Wang 1999, H. Li and D. W. Taylor 1999), possess angiosperm-like vessels and libriform fibers. Leaves of Chinese gigantopterids with waxy cuticles have been described (Z.-Q. Yao and Crane 1986).
In 1992 Mei et al. described Eophyllogonium cathayense, an enigmatic seed plant from the Permian of China. Seed-bearing taeniopterid leaves with reticulate venation were found in the same bedding plane as sterile gigantopterid leaves assignable to Gigantonoclea acuminatiloba and Gigantopteris dictyophylloides. Was Eophyllogonium cathayense attached to a gigantopterid plant with dimorphic leaves?
Paleobotanists are better understanding the anatomy, morphology, and systematics of Cathaysian gigantopterids, but more work is needed.
Phasmatocycas:
Phasmatocycas is an emerging group of ovule-bearing taeniopterids from Carboniferous (Pennsylvanian) rocks (Axsmith et al. 2003) once thought to be allied with the cycads (Mamay 1976). Fertile material of taeniopterids from Pennsylvanian rocks of interior North America was first described as Spermopteris (Cridland and Morris 1960).
Spermopterids may be early cycadophytes, cycadeoids, nilssonialeans, or pentoxylalians. However, critical permineralizations that unambiguously demonstrate diagnostic anatomy of cuticles, stomata, and ovulate position on abaxial or adaxial leaf surfaces are unknown. Therefore, taxonomic assignment of spermopterids to a specific seed plant order is unsupported by lack of fossil evidence.
To the left is a photograph of a possible immature ectopic ovule attached to a megasporophyll (or the leaf was damaged before fossilization, and the object is a "tear") that we provisionally assign to Phasmatocycas.
What is a taeniopterid? It is an common name which refers to often abundant Paleozoic foliage that resembles the leaves of extant Calophyllum (Clusiaceae, Theales, Dilleniidae) or Musa (Musaceae, Zingiberales, Zingiberidae).
Fossilized leaf remains of Taeniopteris are generally not attached to a stem or rachis, thus, in at least some forms it is not known whether the fossil fragments represent pieces of simple or compound leaves, or one leaf-type of fossil plants having dimorphic leaves.
Taeniopterid leaf compressions and impressions are common in terrestrial and deltaic sedimentary deposits of the Paleozoic. Taeniopterids often co-occur with fossilized remains of detached net-veined leaves (or leaflets) of gigantopterids and glossopterids.
Pictured to the right is an example of a laminar microsporophyll that I provisionally assign to Phasmatocycas. Elongate rice-shaped structures on the adaxial leaf surface may be pollen-bearing sacs. This fossil is the only known microsporophyll of a spermopterid, and remains undescribed.
The images are 280 million year old permineralizations photographed by the author in 1982 a couple days after the fossils were excavated from the bedding plane of the section, lower Permian (Leonardian) Road Canyon Formation, Del Norte Mountains, North America. These specimens and others are deposited in the United States National Museum (Mamay et al. 1988). The fossilized leaf imaged to the right is actual size.
It is unclear whether megasporophylls and microsporophylls of Phasmatocycas were attached to the same plant or to two different female and male individuals.
Paleozoic spermopterids are a relatively unknown group. In Phasmatocycas bridwellii, ovules located on the lower (abaxial) surface of leaves were attached by stalks to leaf midribs but not to the leaf edges as suggested by Cridland and Morris (1960).
The image to the left is the distal portion of an undescribed Phasmatocycas ovulate leaf. The two lumps are probably ovules. Collected from the lower Permian (Leonardian) Road Canyon Formation, Del Norte Mountains, North America. This specimen is deposited in the United States National Museum (Mamay et al. 1988).
Howe and Cantrill (2001) describe paleosols from the Albian of Antarctica having lenses of fossilized, detached Taeniopteris daintreei leaves, Carnoconites crantwelli ovulate organs, and Pentoxylon stems. Did these organs belong to a shrub-like Pentoxylon plant?
Paleozoic spermopterids, and a list of phytophagous animal associates, references, and future research needs are summarized below.
WHOLE PLANT MORPHOLOGY--unknown, Phasmatocycas bridwellii was possibly shrub-like (Axsmith et al. 2003); additional paleontologic data are needed to reconstruct whole plants and nodal anatomy
REPRODUCTIVE MODULES--Phasmatocycas: the modules are phyllospermous (Cridland and Morris 1960, Mamay 1973, Axsmith et al. 2003), ovules attached to midribs on the lower (abaxial) surface of laminar megasporophylls
The only known spermopterid male specimen (illustrated above) suggests placement of rice-grained shaped pollen sacs on the upper (adaxial) surface of the microsporophyll; anatomical and developmental details of sexual reproduction unknown, permineralizations are in need of discovery and study
LEAVES--taeniopteroid with a stout multi-stranded midrib, the lateral veins parallel resembling Clusiaceae or Musaceae (see above), whole leaves unknown but probably simple (Cridland and Morris 1960, Mamay 1973, Axsmith et al. 2003); permineralizations with preserved leaf anatomy are needed for study
PHYTOPHAGOUS ASSOCIATE(S)--arthropods belonging to the Caloneurodea, Orthoptera, Protorthoptera (Beck and Labandeira 1998); vertebrate coprolites require discovery and study; fossilized invertebrate exoskeletons and guts are needed for study
PLANT IDENTIFICATION(S)--Taeniopteris (Labandeira 1998); definitive anatomical data are needed for precise taxonomic and nomenclatural placement of the form genus into a known seed plant order, family, and genus
HOST SEED PLANT ORGAN(S) BEING EATEN--leaves (Beck and Labandeira 1998); additional fossils require discovery and study
Cycadales:
A Paleozoic origin of cycads is demonstrable based on the work of T. N. Taylor (1969) and S. H. Mamay (1969, 1973), but the taxonomy of at least some early fossil forms (see Phasmatocycas section above) is open to reinterpretation (Axsmith et al. 2003).
The cycad literature is extensive covering the Permian, Triassic, Jurassic, Cretaceous, and Paleogene periods (Harris 1961, Delevoryas and Hope 1976, Smoot et al. (1985), Krassilov and Bugdaeva (1988), Zhu et al. (1994), Bremer et al. (2003), Klavins et al. (2003), Archangelsky and Villar de Seoane (2004), Krassilov and Doludenko (2004), Hermsen et al. (2006), and Watson and Ash (2006), among others.
A molecular phylogenetic analysis of extant cycadophytes is available (Hill et al. 2003). Volume 70, Number 2 of The Botanical Review (2004), compiles some of the research on these seed plants.
Pictured above is an ovule-bearing cycad leaf (megasporophyll) detached from a living specimen of Cycas revoluta and captured on film. When mature the erect or suberect leaves sheathing the monopodial SAM bear bright red seeds. The SAM's of some cycads contain colonies of beetles and weevils (Norstog et al. 1986, Norstog and Fawcett 1989).
Permian cycads are known from both China and North America (Mamay 1976, Zhi-Feng Gao and Thomas 1989, L. Liu and Z. Yao 2002). In many of the known fossil plant localities, cycad fossils (with preserved reproductive structures) are found with sterile leaf fragment impressions referable to the form genus Taeniopteris.
Triassic cycads have been reported (Ash 1985, Delevoryas and Hope 1971, 1976; Ash 2001, among others) including recent Antarctic fossil finds described as Antarcticycas schopfii (Smoot et al. 1985, Hermsen et al. 2006) and Delemaya spinulosa (Klavins et al. 2003). Hermsen et al. (2006) report on the anatomy of stems and leaves of these fossils together with a discussion of the evolution of the group.
Jurassic cycads are well-known principally of the work by Harris (1932, 1941, 1954, 1961, 1964). Harris (1961) and Tidwell (2002) are keys to the literature on the fossil history and paleobiology of cycads. Enigmatic fossils from the Jurassic Period, possibly transitional between earlier gymnosperms and later cycads include Baruligyna disticha from the Callovian of Georgia in western Asia (Krassilov and Doludenko 2004); a close relative of Semionogyna, from the Lower Cretaceous of Transbaikalia, Russia (Krassilov and Bugdaeva 1988).
The image to the right is a male cone of Cycas revoluta (Cycadaceae, Cycadophyta) from a plant in cultivation, photographed by the author.
Known gross morphological characters of some of cycads, and a list of phytophagous animal associates, references, and future research needs are summarized below.
WHOLE PLANT MORPHOLOGY--monopodial shrubs with the main stem sheathed in helically arranged cataphylls (Smoot et al. 1985, Hermsen et al. 2006)
REPRODUCTIVE MODULES--cones (Harris 1941, Zhu et al. (1994), Klavins et al. 2003), ovule-bearing leaves (megasporophylls) are incompletely known from compressions and impressions (Zhu et al. 1994); a bipinnate microsporophyll has been described as Androcycas santucci (Watson and Ash 2006)
Permineralized pollen cones with pollen are described as Delemaya spinulosa (Klavins et al. 2003); additional permineralizations of reproductive material are needed to better understand pollination biology and reproductive anatomy. Is Delemaya spinulosa the male cone of Antarcticycas schopfii?
LEAVES--cuticles (Harris 1932); simple and strap-shaped leaves which are associated with ovules have been described as Archaeocycas (Mamay 1976); compound pinnate leaves e.g. Zamites tidwellii (Ash 2001) are well-known from Mesozoic and Cenozoic rocks
PHYTOPHAGOUS ASSOCIATE(S)--arthropods belonging to the Caloneurodea, Orthoptera, Protorthoptera (Beck and Labandeira 1998), insect frass not assignable to a particular species is known from permineralizations of Antarcticycas (Hermsen et al. 2006); vertebrate coprolites require discovery and study; fossilized invertebrate exoskeletons and permineralized insect guts are needed for study
PLANT IDENTIFICATION(S)--Taeniopteris (Labandeira 1998); definitive anatomical data are needed for precise taxonomic and nomenclatural placement of the form genus into a family and genus belonging to the Cycadales
Antarcticycas schopfii is the host plant for unknown insect phytophagous associates (Hermsen et al. 2006); important insights have been gained as summarized by Hermsen et al. (2006), but much more paleobotanical research is needed and requires funding from granting agencies and foundations
HOST SEED PLANT ORGAN(S) BEING EATEN--leaves (Beck and Labandeira 1998) and cataphylls (Hermsen et al. 2006), more permineralizations require discovery and study
Key studies on the molecular systematics, evolutionary relationships, and phytophagous insect associates of extant cycads are published by Norstog et al. (1986), Norstog and Fawcett (1989), Azuma and Kono (2006), Wu et al. (2007), González et al. (2008), and D. A. Downie et al. (2008).
Glossopteridales:
One of the dominant vegetation types of the southern reaches of Pangaea during the Permian Period consisted of small to large trees belonging to a group of seed ferns known as glossopterids. Following the Late Carboniferous and Early Permian ice age, mesic forests of glossopterids spread poleward. Gradual warming at the south pole led to replacement of Glossopteris forests by Dicroidium. Change in the composition of overstory trees might have altered understory shrubs and herbs possibly contributing to a decline in dictyodont herbivore biodiversity (Tiffney 1992, Zavada and Mentis 1992).
Mary White (1986) reviews the fossil history of glossopterids. Several morphotype genera of glossopterids circumscribe detached fossil leaves (Eretmonia, Glossotheca, and Kendostrobus, among others), isolated pollen sacs (Arberiella, Lithangium, and Polytheca), ovule-bearing leaves (e.g. Scutum), compound ovulate structures (Lidgettonia etc.), detached seeds (Pterogospermum and Stephanostoma), fossil leaves (Belemopteris, Gangamopteris, Glossopteris, and Rhabtotaenia, among others), and underground parts (Vertebraria).
Plumstead (1973) presents an illustrated discussion of the Glossopteris flora, the paleogeography of Gondwana, and Wegener's Theory of Continental Drift. Glossopteridales and possible relationships with angiosperms are discussed by E. L. Taylor and T. N. Taylor (1992).
Key articles on glossopterids are published by Surange and Maheshwari (1970), Delevoryas and Gould (1971), Maheshwari (1972), Surange and Chandra (1972, 1973, 1975), Holmes (1973), Delevoryas and Person (1975), Chandra and Surange (1976), Schopf (1976), Pant and Choudhury (1977), Gould and Delevoryas (1977), White (1978), Rigby (1978), Pant and Nautiyal (1984), Pant and Nautiyal (1984), Pigg et al. (1987), Pigg (1990), Pigg and Taylor (1990), McLoughlin (1990), Berthelin et al. (2004), Nishida et al. (2004, 2007), and Tewari (2007).
Pigg and T. N. Taylor (1993), Iannuzzi (2000), and Pigg and Nishida (2006) compile particularly complete bibliographies on the fossil history of glossopterids.
The image to the left is a drawing of a detached ovule-bearing Lidgettonia africana leaf (megasporophyll) from a glossopterid tree. It is Figure 7B from Peter R. Crane (1985), Phylogenetic analysis of seed plants and the origin of angiosperms, Annals of the Missouri Botanical Garden 72: 716-796, reprinted with permission of the Missouri Botanical Garden and Peter Crane.
"Figure 7. Morphology of glossopterids. -B. Lidgettonia africana megasporophyll, based on Thomas (1958), Surange and Chandra (1975, text-fig. 1D), Schopf (1976, fig. 8D); X 1.5."
Known gross morphological characters of glossopterids, and a list of phytophagous animal associates, references, and future research needs are summarized below.
WHOLE PLANT MORPHOLOGY--trees and shrubs (Pant and Singh 1974, White 1986)
REPRODUCTIVE MODULES--ovule-bearing leaves (megasporophylls) are of two types: those of Section Megafructi produce either stalked or sessile ovules on upper surfaces of megasporophylls known as regular leaves (Crane 1985, White 1986, E. L. Taylor and T. N. Taylor 1992). In Microfructi, the ovules are attached to scale-leaves termed microphylls (White 1986).
Ovules of Glossopteris have been described as Homevaleia gouldii (Nishida et al. 2007). Preserved sperm, pollen tubes, and ovules provide evidence of zooidogamy in Glossopteris (Nishida et al. 2004)
Pollen-bearing (male) organs of glossopterids are attached to the upper surfaces of scales described as Eretmonia, Glossotheca, and Squamella (White 1986). All three form genera display Arberiella microsporangia that produce pollen (White 1986)
LEAVES--simple, lanceolate, of the Glossopteris and Gangamopteris-type (Crane 1985, White 1986, Tewari 2007)
PHYTOPHAGOUS ASSOCIATE(S)--wood-boring Coleoptera (Zavada and Mentis 1992, Weaver et al. 1997, Labandeira 1998). Caloneurids, orthopterans, and protorthopterans are external leaf-feeders on Glossopteris foliage (Labandeira 1998). Hypoperlids and grylloblattids feed on pollen inside of Protohaploxypinus microsporangia (Labandeira 1998)
PLANT IDENTIFICATION(S)--Glossopteris, Protohaploxypinus
HOST SEED PLANT ORGAN(S) BEING EATEN--bark, leaves, microsporangia, pollen, and wood
Gnetales:
The Gnetales is the only group of Paleozoic seed plants widely regarded as a sister group to the flowering plants (Arber and Parkin 1907, J. A. Doyle and Donoghue 1986, Cornet 1996, Krassilov 1997, Donoghue and J. A. Doyle 2000, Z.-Q. Wang 2004, J. A. Doyle 2006). Cones identifiable to Gnetales are well documented from Permian rocks (Z.-Q. Wang 2004).
Details of the fossil history of the group are summarized by Crane (1996).
The main body of published studies on the evolutionary relationships, biology, and phylogeny of the group appears in several papers contained in a 1996 supplementary volume of the International Journal of Plant Sciences (Carlquist 1996, Crane 1996, J. A. Doyle 1996, Endress 1996, Friedman 1996, Friedman and Carmichael 1996, Hufford 1996, and Price 1996).
Research on the molecular systematics and evolutionary-developmental biology of gnetophytes is published by van Konijnenburg-van Cittert (1992), Osborn et al. (1993), Mundry and Stützel (2004), Rydin et al. (2006), C.-S. Wu (2007), and H.-M. Lu et al. (2008).
Interestingly, despite research focus and debate on Gnetum and relatives and the origin of angiosperms, the Gnetales are no longer widely regarded as flowering plant antecedents.
The image to the right is an attached fossil flower-like organ of Eoantha zerikhinii, an anthophytic gnetalean from the Baisian Assemblage, early Cretaceous Period, Transbaikalia, Russia. Four ovuliphores are visible together with a perianth of linear bracts. Each ovule contain a pollen chamber filled with Ephedrites-type pollen (this picture is from an original image provided by Professor Valentin Krassilov, posted here with his permission).
Vojnovskyales:
In Carboniferous and Permian times another intriguing group of shrub-like seed plants with palm-like leaves, leafy short shoots, and bisexual cone axes appears in the stratigraphic column (Krassilov and Burago 1981, Rothwell et al. 1996, Naugolnykh 2001). Several species have been described from Permian rocks all over the world; perhaps the best known species are Sergeia neubergii (Rothwell et al. 1996) and Vojnovskya paradoxa (Mamay 1976). Vojnovskyalians bear close resemblance in many details of leaf, reproductive, and stem anatomy to Triassic Sanmiguelia and the later pentoxylalians (Crane 1985, Krassilov 1997, Naugolnykh 2001).
Mamay (1976) provoked interesting speculation on Maekawa's 1962 proposal casting Vojnovskya as a "presumable ancestor of angiosperms." It is equally interesting that some workers have drawn a connection between foliar material of Vojnovskya paradoxa and Sanmiguelia lewisii (Crane 1985, page 779).
Another view held by Mamay is that the Vojnovskyales might be remotely related to the Bennettitales or Cordaitales, or represent a "bizarre" evolutionary cul-de-sac (Mamay 1976).
Several groups of vojnovskyalians had architectural adaptations possibly exploited by Paleozoic insects. Intriguing fossilized evidence in Paleozoic rocks of whole plant organs of Sergeia neuburgii (Vojnovskyaceae, Vojnovskyales) is published on page 1072 of Rothwell et al. (1996). Unfortunately these fossils cannot be easily screened for invertebrate remains possibly intercalated between the helically arranged leaf permineralizations, because the fossil was transported with other turbidites some distance from the shoreline (Rothwell et al. 1996).
The image to the left consists of four drawings showing the fructification morphology of certain vojnovskyalians. It is Figure 1 from S. V. Naugolnykh (2001), Morphology and systematics of representatives of Vojnovskyales, Paleontological Journal 35(5): 545-556, reprinted with written permission of Pleiades Publishing, Inc. I thank Serge V. Naugolnykh and the Paleontological Journal for this contribution.
"Figure 1. Fructification morphology of representatives of Vojnovskyales: (a, c, d) Paravojnovskya (al. Gaussia) imbricata (Naug.) Naug. et Doweld, (a) specimen no. 3773(11)/326(92), (c, d) the specimen from the collection by Vaulev (Perm Regional Museum), (b) proposed arrangement of fructifications of P. imbricata on a fertile shoot, in axils of scale-form bracts of Nephropsis (Sulcinephropsis). The locality of Chekarda-1, the Kungurian, Lower Permian of the Middle Fore-Urals. Scale bar is 1 cm."
Known gross morphological characters of Vojnovskyales, and a list of phytophagous animal associates, references, and future research needs are summarized below.
WHOLE PLANT MORPHOLOGY--possibly shrub-like (Mamay 1975, 1976) with scaly, bracetose, and leafy short shoots often fertile in the developing SAM's (Krassilov 1997). Permineralizations are needed to better understand the anatomy of short shoots.
REPRODUCTIVE MODULES--globose head-like strobilus (e.g. Gaussia, Krassilov 1997) composed of basally attached ovule-bearing leaves (megasporophylls) and terminal microsporophylls each with two microsporangia attached to the distal end of each leaf; the fossils are incompletely known from compressions and impressions (Mamay 1976, Krassilov 1997). Shortened, bracteose, cone-like structures subtended by umbrella-shaped leaves that contain either pollen or stalked seeds were described by Naugolnykh (2001)
The reproductive biology of the Vojnovskyales is not clearly understood, but some research progress has been made (Rothwell et al. 1996, Naugolnykh 2001). Additional discoveries of permineralized sexual structures are needed to better understand these seed plants
LEAVES--simple and flabelliform in Sandrewia texana (Mamay 1975, 1976), and lanceolate in other genera (Krassilov 1997); the leaves of Permian Vojnovskya paradoxa resemble Sanmiguelia lewisii, a Triassic plant
PHYTOPHAGOUS ASSOCIATE(S)--unknown; existing specimens should be restudied for the presence of insect remains and traces
PLANT IDENTIFICATION(S)--not applicable
HOST SEED PLANT ORGAN(S) BEING EATEN--not applicable
Seed Plants of the Mesozoic Era:
One approach toward our better understanding of flowering plant cladogenesis from the lineage(s) containing gymnospermous seed plant ancestors is to list the known fossil seed plant groups of the Mesozoic Era and to compare and contrast them with respect to morphological characters most often associated with angiosperms.
The image to the right is a drawing of a detached spherical ovule-bearing Vardekloeftia sulcata head with densely packed ovules and interseminal scales, from a bennettitalean bush. It is Figure 9B from Peter R. Crane (1985), Phylogenetic analysis of seed plants and the origin of angiosperms, Annals of the Missouri Botanical Garden 72: 716-796, reprinted with permission of the Missouri Botanical Garden and Peter Crane.
"Figure 9. Morphology of the Vardekloeftia and Bennetticarpus plants. -B. V. sulcata, spherical head composed of ovules and interseminal scales, based on Harris (1932b, pl. 15, fig. 1, pl. 17, figs. 1, 2, pl. 18); X 2."
Did sauropods and smaller herbivorous reptiles feed on these fleshy Vardekloeftia seed heads? Were the ovules and scales of this plant filled with antiherbivory poisons such as polyacetylenes and toxic cycasins?
"In sum, from the plant's point of view, the age of dinosaurs was not an extension of Permian herbivory, nor a duplication of the present. While smaller herbivores duplicated some aspects of the preceding and following time, the immense herbivores imposed a unique selective force on the physiognomy and life history strategies of Mesozoic plants. Initially, these forces were met by a limited diversity of genetic lineages of plants, but by the end of the Mesozoic, gymnosperms with efficient vegetative growth and abilities to recover from damage (= angiosperms) had evolved. Much of the morphology and biology of Mesozoic plants should be considered in the light of this substantial herbivore pressure."
The preceding quotation is from page 94 of Tiffney (1992), The role of vertebrate herbivory in the evolution of land plants, The Palaeobotanist 41: 87-97.
I now present a brief survey of several enigmatic and sometimes spectacular seed plant finds.
Sanmiguelia and Enigmatic Seed Plants. The first occurrence of angiosperm-like palynomorphs in the stratigraphic record is from the Early Triassic Period (Hochuli and Feist-Burkhardt 2004), but the anatomy and morphology of the seed plant(s) which shed such pollen is a mystery. Fossilized remains of Triassic seed plants that have flowering plant characteristics are very rare and often poorly preserved.
Several enigmatic genera are known from Mesozoic rocks, including Baisia (Krassilov 1997), Cycandra (pollen cones resembling fossil cycads or Nilssonialeans), cupules of Dirhopalostachys (reminiscent of basal flowering plants such as Lesqueria elocata), Fraxinopsis (Axsmith et al. 1997), Furcula (Harris 1932), Irania (with both androclades and gynoclades), Leptostrobus (a czekanowskialean with stigmatic cupules), Pannaulika triassica (Cornet 1993), and Schmeissneria, an intriguing seed plant incertae cedis from the earliest Jurassic Era (Lias interval) of Europe, and Middle Jurassic of Asia (X. Wang et al. 2007).
Of all the enigmatic seed plants of the early Mesozoic Era, Sanmiguelia has attracted the most attention by paleobotanists. This rather common Triassic fossil of southwestern North America is remarkably similar to the Paleozoic Vojnovskyales (Crane 1985, page 779), another group of angiosperm-like seed plants from Asiatic and North American Paleozoic rocks (Mamay 1976).
Sanmiguelia lewisii is an innovative Triassic plant having palm-like leaves (Brown 1956, Ash 1976) with flowers and angiosperm-like reproductive modules not unlike the monocotyledonous angiosperm Veratrum (Cornet 1986, 1989). Detailed studies of the reproductive morphology of Sanmiguelia have been published (Cornet 1989). Additional permineralized fossil material of Sanmiguelia is probably needed to better understand the anatomy of reproduction and whole plant morphology.
Morphological details of Sanmiguelia, and a list of phytophagous animal associates, references, and future research needs are summarized below.
WHOLE PLANT MORPHOLOGY--Sanmiguelia lewisii was a woody herb (Cornet 1986, 1989), or small shrub. Better preserved fossilized material and more anatomical data are needed to better understand the anatomy and morphology of whole plants in relation to Vojnovskyales and the angiosperms Joinvillea and Veratrum
REPRODUCTIVE MODULES--ovuliferous inflorescences (first described as Axelrodia burgeri), polleniferous inflorescences (named Synangispadixis tidwellii), flowers with ovuliferous units and polleniferous units, megasporophylls as carpels, synangia as anthers, bracts, bitegmic ovules (Cornet 1986, 1989)
The reproductive biology of Sanmiguelia is not clearly understood, but some research progress has been made (Cornet 1986, 1989). Additional discoveries of permineralized sexual structures are needed to better understand the evolutionary position of these seed plants in relation to other anthophytes (including monocotyledonous flowering plants) and older vojnovskyalians
LEAVES--simple, alternate, clasping and flabelliform to lanceolate resembling extant Veratrum californicum (Liliaceae, Liliales, Liliidae) (Cornet 1989)
PHYTOPHAGOUS ASSOCIATE(S)--bite marks and trails of anther debris left by unknown insects (Cornet 1989); additional specimens await discovery to be studied for the presence of insect remains and traces
PLANT IDENTIFICATION(S)--Sanmiguelia lewisii (Cornet 1989)
HOST SEED PLANT ORGAN(S) BEING EATEN--leaf tissue and pollen? (Cornet 1989)
Bennettitales. By Triassic time, an intriguing group of monopodial seed plant shrubs, known as cycadeoids appears in the stratigraphic column (Crepet 1972, Crepet 1974, Crane 1986, Barbacka 2000, among others). These shrubs were cycad or palm-like in overall aspect. Reproductive structures were borne on modified leaves, clustered in flower-like strobili. Classified in the taxonomic order Bennettitales, cycadeoids were thought by Arber and Parkin (1907) as likely candidates as ancestors of modern flowering plants.
There are two or three main groups of cycadeoids often treated at the family level in the taxonomic hierarchy. Cycadeoidaceae studied by Wieland and Delevoryas include Cycadeoidea, fossilized remains of squat, shrub-like trunks bearing leaves, cones, and pollen recovered from Cretaceous rocks of North America and Monanthesia (see Delevoryas 1962). The other group classified in the Williamsoniaceae includes Williamsonia sewardiana, described from the Jurassic period of India.
Detailed studies of bennettitaleans by paleobotanists have been underway for several decades. Some of the more recent published work includes Delevoryas (1968), Harris (1969), Crepet and Delevoryas (1972), Crepet (1972, 1974), T. N. Taylor (1973), Harris (1974), Sharma (1974, 1976, 1977), Crane (1986), Pedersen et al. (1989), Delevoryas (1991), Osborn and Taylor (1995), Saiki and Yoshida (1999), Barbacka (2000), Rothwell and Stockey (2002), Stockey and Rothwell (2003), Watson and Lydon (2004), Boyd (2004), Pott et al. (2007), and Going et al. (2007).
Form genera representing detached fossilized organs of bennettitaleans include Anomozamites (leaves), Bucklandia (stems and leaf bases), Cycadeoidea (branches, cones, leaves, synangia, and trunks), Cycadolepis (cone bracts and scales), Exesipollenites (pollen), Ischnophyton (stems and fronds), Nilssoniopteris (leaves), Otozamites (leaves), Pseudocycas (leaves), Pterophyllum (leaves), Ptilophyllum (leaves), Weltrichia (pollen cones), Williamsonia (ovulate cones), and Zamites (leaves).
A reconstruction of a detached and cut bennettitalean flower as viewed from the side, which is classified as Williamsonia harrisiana appears above. It is Figure 10A from Peter R. Crane (1985), Phylogenetic analysis of seed plants and the origin of angiosperms, Annals of the Missouri Botanical Garden 72: 716-796, reprinted with permission of the Missouri Botanical Garden and Peter Crane.
"Figure 10. Morphology of Bennettitales. -A. Williamsonia harrisiana, longitudinal half-section of 'flower'"
Several potentially interesting points of discussion on the assumed relationship of phytophagous insect associates with cycadeioids may be found in Crepet (1974).
"It is possible that predation pressure was not only responsible for the phylogenetic 'closing' of the microsporophylls, but was also responsible for the position of the cones on the trunks of Cycadeoidea which offers additional protection."
The above quotation is from page 160 of Crepet (1974), "Investigations of the North American cycadeoids: the reproductive biology of Cycadeoidea," Palaeontographica Abt. B: 148.
Known gross morphological characters of Bennettitalians, and a list of phytophagous animal associates, references, and future research needs are summarized below.
WHOLE PLANT MORPHOLOGY--shrubs (Crepet 1974)
REPRODUCTIVE MODULES--flowers with ovule-bearing leaves (megasporophylls) and interseminal scales, flowers with pollen-bearing organs, nectar glands, ovules, pollen, seeds
LEAVES--simple and strap-shaped or compound pinnate and palm-like (Boyd 2000), xeromorphic with thick cuticles (Villar de Seoane 2001)
PHYTOPHAGOUS ASSOCIATE(S)--indeterminate beetles, helid and nemonychid weevils, possible pollinating parandrexid or protoceline chrysomelid beetles, and unidentified nectar-feeders (Crepet 1974, Dmitriev and Ponomarenko 2002)
PLANT IDENTIFICATION(S)--bisexual fructifications of Cycadeoidea with burrows (mines), insect traces in various Mesozoic fructifications (Crepet 1974, Dmitriev and Ponomarenko 2002)
HOST SEED PLANT ORGAN(S) BEING EATEN--leaf tissues, ovules, and pollen?
Caytoniales. The enigmatic seed fern order Caytoniales first appeared in the Triassic Period. By Jurassic time, caytonialians comprised an important floristic element indigenous to terrestrial biomes on the northern landmasses (Krassilov 1977, 1997).
The image to the right and the figure legend below in quotation marks is from page 753 of Peter R. Crane (1985), Phylogenetic analysis of seed plants and the origin of angiosperms, Annals of the Missouri Botanical Garden 72: 716-796, reprinted with permission of the Missouri Botanical Garden and Peter Crane.
"Figure 17. Morphology of the Caytonia plant. -A. Sagenopteris colpodes, based on Thomas (1925, pl. 15, fig. 50); x0.75.
-B. S. colpodes, detail of leaflet attachment and venation, based on Harris (1964, fig. 2H); x4.
-C. Caytonanthus arberi, based on Harris (1941, pl. 2, fig. 3); x7.
-D. Pollen from C. arberi, based on Townrow (1962b, fig. 3d, e); x1,200.
-E. Caytonia nathorstii megasporophyll, based on Harris (1964, fig. 10-A-C); x5.
-F. Caytonia "cupule" containing seeds, based on Reymanowna (1973, particularly text fig. 12E, F); x12.5.
-G. Caytonia "cupule," longitudinal section, based on Reymanowna (1973, particularly text fig. 12E, F); x12.5.
-H. C. nathorstii ovule longitudinal section, redrawn from Harris (1958, fig. 7); x110."
Several form genera were described by Harris (1940, 1941, 1951, 1964, 1971), including Amphorispermum (seeds), Caytonanthus (pollen-bearing organs), Caytonia (cupules), Sagenopteris (leaves), and Vitreisporites (pollen). Barbacka and Boka (2000) provide additional details.
General evolutionary relationships of caytonialians with other seed plant groups are discussed by Harris (1951) and Krassilov (1997). Krassilov (1997) draws a possible evolutionary connection between caytonialians and ranunculid flowering plants. Detailed discussion of the morphology and anatomy of this group of anthophytes, homology of characters, phylogenetic relationships with other seed plants, and the origin of angiosperms may be found in Krassilov (1977) and J. A. Doyle (2006).
Known gross morphological characters of the Caytoniales, and a list of phytophagous animal associates, references, and future research needs are summarized below.
WHOLE PLANT MORPHOLOGY--possibly shrub-like (Reymanowna 1974, Krassilov 1997). Coalifications, compressions, permineralizations, and petrifactions are needed to elucidate the form of whole plants
REPRODUCTIVE MODULES--cupules, bitegmic ovules, and saccate pollen
The reproductive biology of caytonialians is not clearly understood, but some research progress has been made (Krassilov 1997). Additional discoveries of permineralized sexual structures are needed to better understand these seed plants
LEAVES--compound palmate, the pinnae are lanceolate and glossopteroid in general aspect; leaves have been classified in the form genus Sagenopteris (Krassilov 1997)
PHYTOPHAGOUS ASSOCIATE(S)--indeterminate insect traces on foliage (Dmitriev and Ponomarenko 2002)
PLANT IDENTIFICATION(S)--Sagenopteris
HOST SEED PLANT ORGAN(S) BEING EATEN--leaves (Dmitriev and Ponomarenko 2002)
Corystospermales. Corystosperms were once a dominant forest vegetation type of Gondwana. The Dicroidium flora probably replaced the Glossopteris flora of the southern latitudes (White 1986) while many glossopterids were probably extirpated by the end-Permian extinction (EPE). The Corystospermales represented by the Paleogene Tasmanian fossil Komlopteris cenozoicus, is the only group of seed ferns that survived the K-T asteroid impact (McLoughlin et al. 2008).
The seed fern order Corystospermales is receiving more interest based upon recovery of permineralized reproductive material from Antarctica, the Middle East, and North America, including, for example, preserved pollen organs described as Pteruchus; cupules, pollen, and other detached remains (Pigg et al. 1993, Osborn and Taylor 1993, Yao et al. 1995, Axsmith et al. 2000, Kerp et al. 2006). Compression fossils of ovulate axes described as Umkomasia are known from the Triassic Molteno Formation of South Africa (Axsmith et al. 2000), and from Jurassic deposits.
Several form genera of corystosperms are known including Alisporites (pollen), Dicroidium (foliage), among other leaf morphotypes; Karibacarpon (detached ovules), Pilophorosperma (ovules and associated leaves), Pteroma (ovulate organs), Pteruchus (pollen organs and associated leaves), Rhexoxylon (permineralized wood), and Spermatocodon (ovulate organs).
The image to the left and figure legend is from page 755 of Peter R. Crane (1985), Phylogenetic analysis of seed plants and the origin of angiosperms, Annals of the Missouri Botanical Garden 72: 716-796, reprinted with permission of the Missouri Botanical Garden and Peter Crane.
"Figure 18. Morphology of corystosperms. -A. Pachypteris papillosa stem with leaves, based on Harris (1983a, fig. 2): X 0.25 [transposed caption letters G, H, and I were corrected by J. M. Miller in 2008].
-B. Pteroma thomasii synangium, abaxial view, redrawn from Harris (1964, fig. 66B); X 0.25.
-C. P. thomasii, lateral view, based on Harris (1964, fig. 66A-G, I); X 2.
-D. Dicroidium odontopteroides, redrawn from Thomas (1933, fig. 49a); X 0.75.
-E. Umkomasia macleanii, redrawn from Thomas (1933, fig. 1, pl. 26, fig. 56); X 2.5.
-F. Corystosperm ovule based on Thomas (1933, fig. 33c); X 4.
-G. [I]. Rhexoxylon, transverse section of stem, based on Archangelsky and Brett (1961, fig. 2A); X 0.5.
-H. [G]. Pollen of Pteruchus africanus, redrawn from Townrow (1962a, fig. 1A-D); X 2.5."
-I. [H]. P. africanus, based on Townrow (1962a, fig. 1A-D); X 2.5.".
Most of the form genera of corystosperms are now known to be attached to massive trees (Dicroidium odontopteroides) that once formed a prevalent forest type on southern portions of Pangaea (Axsmith et al. 2000). The phylogenetic position of corystosperms as a group, is problematic (Klavins et al. 2002), but a position within the anthophyte evolutionary clade is demonstrable.
Frohlich in his update of the mostly male theory (MMT) of flower origins raises the interesting suggestion that early Mesozoic corystosperms might be flowering plant ancestors (Frohlich 2002). However, despite his assertion that the morphology of Pteroma supports MMT, it is unlikely that corystosperms were angiosperm antecedents. In fact, the Paleogene Tasmanian fossil corystosperm, Komlopteris cenozoicus, survived the K-T asteroid impact and is found together with dicotyledonous angiosperm leaves (McLoughlin et al. 2008).
Pentoxylales. Taeniopteroid foliage, fragments of short-shoots, and reproductive structures were recovered from Jurassic sediments in the Rajmahal Hills of the Indian sub-continent and described as a new group of seed plants by Birbil Sahni in 1948.
Leaves attached to fragments of shoots of the whole plant Pentoxylon sahnii were first described as the form genus Nipaniophyllum.
Taeniopterid leaves which were first described from several Australasian Jurassic fossil plant localities as Taeniopteris spatulata, are now known to be attached to whole plants referable to Pentoxylon. Detached ovulate axes were initially described as Carnoconites but now too, belong to Pentoxylon (Sahni 1948, Wesley 1963, among others).
The image to the left is Figure 19 Peter R. Crane (1985), Phylogenetic analysis of seed plants and the origin of angiosperms, Annals of the Missouri Botanical Garden 72: 716-796, reprinted with permission of the Missouri Botanical Garden and Peter Crane.
"Figure 19. Morphology of Pentoxylon plants. -A. Carnoconites cranwelliae, ovulate heads, based on Harris (1962, text-fig. 2B, fig. 1): X 2.5 [transposed caption letters B and C were corrected by J. M. Miller in 2008].
-B. [C]. Carnoconites, longitudinal section through ovule, based on Sahni (1948, fig. 21); X 10.
-C. [B]. Nipaniophyllum raoi, redrawn from Sahni (1948, fig. 34 a,b); X 1.
-D. Pentoxylon sahnii, transverse section of stem showing vascular strands, based on Sahni (1948, fig. 9); X 8.
-E. Sahnia microsporangiate "flower," based on Vishnu-Mittre (1953, fig. 11) and Bose et al. (in press [1985]); X 2.5.
Following in the footsteps of Sahni (1948), Vishnu-Mittre (1953) described pollen-bearing "flowers" of the Pentoxylales. Rao (1976, 1981) provides additional discussion of the relationships of the Pentoxylales with other gymnosperms.
The paleoecology and taxonomy of pentoxylalians indigenous to the Antarctic polar forest of Cretaceous times was reviewed by Howe and Cantrill (2001). In Mesozoic times these plants were probably shrub-like not unlike some of the Bennettitalians. Additional discussion of the biodiversity and paleobiogeography of this enigmatic seed plant group may be found in Wesley (1973).
There are few, if any anecdotal accounts of phytophagous animal associates of pentoxylalians. Data are needed from study of existing museum specimens.
Table 3 summarizes the diagnostic anatomical and morphological characters of Mesozoic seed plants, excluding flowering plants. To maintain continuity with the essay on the origin of angiosperms, which has two tables, the following table becomes Table 3. Caytonia includes its component morphotype genera (Caytonia, Caytonanthus, and Sagenopteris). In a similar way the Dicroidium column incorporates the morphotype genus Umkomasia. Finally the column labeled as Sanmiguelia includes its morphotype genera (Axelrodia, Sanmiguelia, and Synangispadixis).
The source of information in Table 3 is Cornet (1989), Crane (1985), Klavins et al. (2002), Krassilov (1997), Stewart and Rothwell (1993), G. Sun et al. (2001), and X. Wang et al. (2007). The genera listed in the table header belong to (or are allied with) several of the major groups of seed plants, including Archaefructus, Caytonia (Caytoniales), Cycadeoidea (Bennettitales), Dicroidium (Corystospermales), Eoantha (Gnetales?), Furcula (Peltaspermales), Pentoxylon (Pentoxylales), Sanmiguelia (anthophyte incertae cedis), and Schmeissneria (anthophyte incertae cedis).
| 
Paleobotany of Angiosperm Origins