Smilax (Smilacaceae) From The Miocene Of Western Eurasia With Caribbean Biogeographic Affinities

Dimitrios Velitzelos; tuncay guner; Thomas Denk

The Pliocene flora of Frankfurt am Main described by Karl Mädler during the first half of the twentieth century is a key flora for the European Pliocene. In the present study, we revised the leaf fossil taxa described by Mädler and investigated plant material collected after Mädler's publication. The revised and augmented floral list comprises seven new species and some new combinations of taxa described by Mädler. In total, 16 gymnosperm species in 15 genera and 73 angiosperm species (of which 15 could not be assigned to a genus) in 40 genera are recognised in the leaf record. Main characteristics of the flora are the high diversity of conifers, the diverse assemblage of exclusively deciduous Fagaceae, including six species of oaks, and the high diversity of Rosaceae. These features indicate cool temperate climatic conditions (comparable to Lugano in southern Switzerland). Angiosperm genera that are today confined to North America and/or East Asia (Eucommia, Magnolia and Sassafras) also are deciduous, whereas evergreen taxa are shrubs typical of the understorey (Buxus, Ilex, Pachysandra, Prunus lusitanica type) and Viscum. Eighteen taxa recorded in the Pliocene of Frankfurt am Main are today absent from western Eurasia and eastern North America, and 25 taxa are absent from western North America. This shows (i) a strong biogeographic link of the Pliocene flora of Frankfurt am Main with East Asia, (ii) surprisingly high levels of speciation (Pliocene endemisms) and (iii) that the European flora was more diverse in woody species shortly before the onset of major Pleistocene glaciations than today.

The early Burdigalian (MN3) plant assemblage of the Güvem area (northwestern Central Anatolia) is preserved in lacustrine sediments of the Dereköy pyroclastics. Its age is well constrained by radiometric dates of basaltic rocks bracketing the pyroclastics, making the Güvem flora one of the extremely few precisely dated early Miocene floras in the Mediterranean region. The rich assemblage of impression fossils comprises ferns and fern allies (2 species), gymnosperms (12 spp.) and angiosperms (129 spp.). Ilex miodipyrena sp. nov. is described as a new fossil-species. The most diverse families in the assemblage are the Fagaceae with 12 taxa and the Fabaceae with 12 leaf morphotypes and one fruit taxon. Aquatic plants are represented by seven taxa, riparian (including palms) and swamp forest elements by >35 taxa, and lianas by three taxa (Smilax spp., Chaneya). The relatively large number of aquatic and riparian/swamp elements is congruent with the rich fish, amphibian and reptile record of the Güvem area. Another characteristic feature of the plant assemblage is the presence of various lobed leaves which show similarities with modern species of different families (e.g. Alangium, various Malvales). Trees and shrubs growing on well-drained soils and forming closed-canopy and open-canopy forests are the most diversified group (>70 taxa). In terms of number of specimens in the collection and based on field observations, by far the most abundant leaf fossils belong to evergreen oaks of Quercus drymeja and Q. mediter-ranea and to various types of foliage that cannot be assigned to a particular extant or extinct genus of Fagaceae. These sclerophyllous trees must have covered vast areas surrounding the wetlands that developed during the early Miocene in the Güvem Basin. Based on a recent reassessment of the ecology and taxonomic affinity of these trees, they are considered to reflect humid temperate climatic conditions but with a brief drier season during the winter months. These forests are more similar to the laurel forests of the southeastern United States and those stretching in a narrow belt south of the Himalayas to eastern central China. The large number of Fabaceae may indicate the presence of warm subtropical environments but this is difficult to assess, as they are known for having wide ecological ranges today and in the past. All in all, a larger part of the plant taxa point to forested vegetation. This is in agreement with previous palynological studies which detected only small amounts of her-baceous and grass pollen. Open patches of vegetation may have been restricted to river banks and to rocky areas in a volcanic landscape. The biogeographic patterns detected for the early Miocene of the Güvem assemblage are manifold; most taxa are widespread Northern Hemispheric elements. A substantial part of the species migrated from Asia into Europe during the (late) Paleogene and reached Anatolia during the early Miocene (Fagus, Paliu-rus, Chaneya, Ailanthus, Quercus kubinyii, Davallia haidingeri, Acer angustilobum, A. palaeosaccharinum). Fewer taxa may have been in Anatolia before they migrated to Europe (e.g. Nerium, Smilax miohavanensis, Quercus sosnowskyi). Finally, very few taxa are Anatolian endemics (e.g. Ilex miodipyrena).

The remarkable collections of fossil leaves of Messinian age (ca. 5,7 Ma) from two sites of the Vena del Gesso Fm. (Monte Tondo and Tossignano, Bologna and Ravenna provinces, Italy) are studied morphologically to assess the taxonomic composition of the assemblages and carry out a palaeoenvironmental analysis. The flora of Tossignano includes so far 42 taxa of vascular plants, with 10 conifers and 32 angiosperms (29 dicots and 3 monocots). In the plant material from Monte Tondo 72 taxa of vascular plants were recognized, with 9 conifers and 63 angiosperms (59 dicots, 3 monocots, 1 plant incertae sedis). Several fossil-species, well known in the Miocene of central Europe and Italy, have been detected. The most frequent ones are, at Monte Tondo, Pinus cf. rigios, Daphnogene polymorpha, Fagus gussonii, Platanus leucophylla and Quercus pseudocastanea, at Tossignano: Sequoia sp., Quercus roburoides, Ailanthus pythii. Other species are frequent in both sites: Taxodium dubium, Laurophyllum sp. A and B, Leguminosae gen. et sp. 1. The general palaeovegetation structure of the Monte Tondo and Tossignano plant assemblages corresponds to the vegetation transect of the Evaporitic Messinian with swamp, riparian vegetation, and zonal “subtropical humid forest”. For the last one, the results of IPR vegetation analysis indicate a transitional (ecotone) vegetation type between “Broad-leaved Evergreen Forest” and “Mixed Mesophytic Forest". The Coexistence Approach (CA) results indicate that both floras existed under comparable climatic conditions , with a MAT 14-16.2 °C for Tossignano and 16-16.5 °C for Monte Tondo, hence warmer than at present (MAT around 13.5 °C).

GALLEY PROOF — AJBD1400495 RESEARCH ARTICLE A M E R I C A N J O U R N A L O F B OTA N Y SMILAX (SMILACACEAE) FROM THE MIOCENE OF WESTERN EURASIA WITH CARIBBEAN BIOGEOGRAPHIC AFFINITIES1 THOMAS DENK2,6, DIMITRIOS VELITZELOS3, H. TUNCAY GÜNER4, AND LILIAN FERRUFINO-ACOSTA5 2Swedish Museum of Natural History, Department of Palaeobiology, Box 50007, 10405 Stockholm, Sweden; 3Athens University, Department of Geology and Geoenvironment, Section of Historical Geology and Paleontology, Panepistimiopolis, Athens 15784 Greece; 4Istanbul University, Faculty of Forestry, Department of Forest Botany, 34473 Bahceköy, Istanbul, Turkey; and 5Universidad Nacional Autónoma de Honduras, Facultad de Ciencias, Departamento de Biología, Laboratorio de Histología Vegetal y Etnobotánica, Boulevard Suyapa, Tegucigalpa, Honduras • Premise of the study: Recent molecular studies provide a phylogenetic framework and some dated nodes for the monocot genus Smilax. The Caribbean Havanensis group of Smilax is part of a well-supported “New World clade” with a few disjunct taxa in the Old World. Although the fossil record of the genus is rich, it has been difficult to assign fossil taxa to extant groups based on their preserved morphological characters. • Methods: Leaf fossils from Europe and Asia Minor were studied comparatively and put into a phylogenetic and biogeographic context using a molecular phylogeny of the genus. • Key results: Fossils from the early Miocene of Anatolia represent a new species of Smilax with systematic affinities with the Havanensis group. The leaf type encountered in the fossil species is exclusively found in species of the Havanensis group among all modern Smilax. Scattered fossils of this type from the Miocene of Greece and Austria, previously referred to Quercus (Fagaceae), Ilex (Aquifoliaceae), and Mahonia (Berberidaceae) also belong to the new species. • Conclusions: The new Smilax provides first fossil evidence of the Havanensis group and proves that this group had a western Eurasian distribution during the Miocene. The age of the fossils is in good agreement with the (molecular-based) purported split between the Havanensis and Hispida groups within Smilax. The Miocene Smilax provides evidence that all four subclades within the “New World clade” had a disjunct intercontinental distribution during parts of the Neogene involving trans-Atlantic crossings (via floating islands or the North Atlantic land bridge) and the Beringia land bridge. Key words: biogeography; bird dispersal; disjunct distribution; evolution; floating islands; North Atlantic land bridge; Smilacaceae; Smilax Havanensis group; transatlantic crossing. molecular studies identified four major clades within Smilax (including Heterosmilax Kunth; Fig. 1). (A) The EurasianAfrican-Indian Smilax aspera L. is sister to the remainder of the genus albeit with low support. (B) A clade of American species (the New World clade) consists of five subclades. Notably, three of these subclades have a single or very few Old World representatives that are sister to the American species (Qi et al., 2013; Fig. 2). One subclade, B3, comprises species from South America (Brazil, Venezuela), the Caribbean, and southern Florida. This subclade corresponds to the Smilax Havanensis group according to Ferrufino-Acosta (2010), which includes ten species with a distinctly dentate leaf margin, and partly to the Smilax Schomburgkiana group according to Ferrufino-Acosta (2010). Finally, C and D, the remaining two clades, which include chiefly Old World species plus one clade of North American herbaceous species. Molecular dating studies have suggested a stem node age of 58 ± 9.9 to 46 ± 8.2 Ma (million years) for the Smilacaceae (Vinnersten and Bremer, 2001) or 90 ± <16 Ma (Janssen and Bremer, 2004) corresponding to Late Cretaceous to early Cenozoic. Leaf morphology in Smilax is highly variable within a species and among clades and using patterns of leaf venation and shape The genus Smilax L. (Smilacaceae) comprises ca. 210 species of lianas, shrubs, and herbs occurring mainly in the subtropics and tropics of both hemispheres with some extensions into temperate regions (Ferrufino-Acosta, 2010; Qi et al., 2013). Smilax is a monocot genus in the order Liliales; it is sister to a clade comprising Philesiaceae with two species in South America and Ripogonaceae with six species in Australia, New Zealand, and New Guinea (Kim et al., 2013). Recent comprehensive 1 Manuscript received 12 November 2014; revision accepted 11 February 2015. The authors thank: Diane Erwin, UCMP, Berkeley, California, USA; Martin Gross and Reinhold Niederl, Universalmuseum Joanneum, Graz, Austria; Irene Zorn, Geologische Bundesanstalt, Vienna, Austria; and Andreas Kroh, Naturhistorisches Museum, Vienna, Austria for facilitating work in the collections. Benjamin Bomfleur, Stockholm, Sweden is thanked for help with nomenclatural issues. This work was supported by a grant of the Swedish Research Council to TD. Two reviewers provided helpful comments on the manuscript. 6 Author for correspondence (e-mail: thomas.denk@nrm.se) doi:10.3732/ajb.1400495 American Journal of Botany 102(3): 1–16, 2015; http://www.amjbot.org/ © 2015 Botanical Society of America 1 GALLEY PROOF — AJBD1400495 2 • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • A M E R I C A N J O U R N A L O F B O TA N Y Fig. 1. Simplified phylogenetic relationships of Smilax (based on Qi et al., 2013). The distribution of characteristic leaf types across the main clades A to D is shown. Clades A and B share ovate to lanceolate leaves with a conspicuous cordate-hastate base (gray). Note that this leaf type is restricted to Smilax aspera (clade A) and Smilax bona-nox and allied species (clade B). White arrowhead indicates presence of spinose-ciliate leaf margin in these species. Ovate to oblong leathery leaves with a spiny margin (dark gray) are confined to clade B (Havanensis group). Narrowly elliptic to lanceolate leaves (white) occur in clades B, C, and D. The most common leaf type in clades B, C, and D is ovate (to elliptic and oblong) with a rounded or cordate base (background shading). Drawings of leaves are based on herbarium specimens (herbaria E, K, P, S) and on Flora of China Editorial Committee (2000), Flora of North America Editorial Committee (2003), and FerrufinoAcosta (2010). Leaves are not to scale. alone makes it difficult to discriminate infrageneric groups of Smilax or even to distinguish Smilax from other monocot genera (Daghlian, 1981; Wilde, 1989; Mai, 1995; Ferrufino-Acosta, 2010). This makes the fossil record of foliage resembling Smilax difficult to interpret. The oldest leaf remains and reproductive structures assigned to Smilax date back to the Cretaceous in both hemispheres (e.g., Berry, 1911; Filippova, 1994) but all these records are in strong need of revision (Daghlian, 1981; Greenwood and Conran, 2000). Unambiguous fossils of Smilax date back at least to the early middle Eocene (Wilde, 1989) and of Ripogonum to the early Eocene (Conran et al., 2009). From the middle Eocene and Eocene/Oligocene of North America and Europe several leaf remains with strong similarity to modern Smilax were reported (Cockerell, 1914; Chaney and Sanborn, 1933; Wilde, 1989; Dilcher and Lott, 2005); in some cases, leaf epidermal features provide strong support for the generic assignment of the leaf remains (Wilde, 1989). All the leaf remains recovered conform to two basic leaf types: 1) ovateelliptic; and 2) (wide) ovate and distinctly cordate-hastate. In view of several New World-Old World disjunctions within clades of Smilax (Fig. 2) it will be important to link fossil taxa with specific infrageneric clades as they might provide crucial biogeographic information. In most, if not all cases, the previous fossil record of Smilax is difficult to put into a phylogenetic and historically biogeographic context because the basic leaf types in Smilax have evolved independently in distantly related infrageneric groups (Fig. 1). In this study, we describe a new species of Smilax from the Miocene of western Eurasia with a highly diagnostic leaf morphology. We refer the fossil species to the modern Havanensis group and discuss the implications of the new fossil find for the evolution of modern disjunct distribution patterns in various subgroups of the New World clade of Smilax. MATERIALS AND METHODS The fossil plant material investigated for the current study originates from the following collections: University of California Museum of Paleontology, Berkeley, California, USA (UCMP); Universalmuseum Joanneum [formerly Landesmuseum Joanneum], Graz, Austria (LMJ); Geologische Bundesanstalt, Vienna, Austria (GBA); Naturhistorisches Museum Wien, Vienna, Austria (NHMW); and Istanbul University, Faculty of Forestry, Istanbul, Turkey (ISTO-F). High resolution scans of herbarium material of modern Smilax were provided from the herbaria at Berlin, Germany (B; Röpert, 2000), Edinburgh, UK (E; Royal Botanic Garden Edinburgh, 2014), Kew, UK (K; Royal Botanic Gardens, Kew, 2014), and Paris, France (P; http://science.mnhn.fr/institution/ mnhn/search). The terminology for the morphological description of the fossil leaves mainly follows Dilcher (1974). The specimens investigated for this study and specimen details are listed in Appendix S1 (see Supplemental Data with the online version of this article). The plant fossils originate from lower to middle Miocene sedimentary formations GALLEY PROOF — AJBD1400495 D E N K E T A L . — E U R A S I A N M I O C E N E S M I L A X • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • 3 Fig. 2. Phylogeny of the “New World Clade” of Smilax showing three New World-Old World disjunctions, two achieved including the North Atlantic land bridge (NALB), one including the Beringian land bridge (BER). E: E. North America, Az: Azores, BA: Balkans, Asia Minor, M: Mesoamerica, CM: Canary Islands, Madeira, M/S: Mesoamerica, South America, W: W. North America, E/W: E. North America, W. North America. NALB: North Atlantic Land Bridge, PP: Bayesian posterior probability. Asterisk denotes fully supported clades. Phylogenetic framework for “New World clade” of Smilax from Qi et al. (2013). Inferred divergence age between Hispida group and remainder of New World clade from Zhao et al. (2013). Light gray shading denotes group for which the fossil species described here provides a minimum age of ca. 20 Ma. Note that the Glauca and Hispida groups share closely similar, undiagnostic, leaf types. Morphologically recognized infrageneric “groups” of Smilax from Ferrufino-Acosta (2010). from Turkey (Güvem, Ankara; Soma-Deniș, Manisa; Yatağan Basin, Muğla), Greece (Kymi, Euboea), and Austria (Parschlug, Styria; Fig. 3). the Soma Formation (lower coal seam) indicate a Burdigalian age (MN3). Plant fossils are from the marls above the lower coal seam and thus are of late Burdigalian to early Langhian age (early to middle Miocene). Güvem area— The plant-bearing sediments are fossil-rich diatomites and lacustrine claystones of the Dereköy pyroclastics forming the basal part of the Güvem Formation (Wilson et al., 1997; Tankut et al., 1998; Yavuz-Ișık, 2008). The Dereköy pyroclastics unconformably overlie the Çukurviran dacite, for which a Potassium-Argon (K-Ar) age of 19.7 ± 0.6 Ma was obtained (Wilson et al., 1997). Above the Dereköy pyroclastics, the Bakacak andesite yielded a K-Ar age of 17.9 ± 0.5 Ma (Wilson et al., 1997). Hence, the plant fossils from the Güvem area (Beș Konak, Kısilcahaman, Keseköy) are of early to middle Burdigalian age (early Miocene). Yatağan Basin, Muğla— The early to early late Miocene Eskihisar Formation is the lowermost in the Neogene succession of the Yatağan Basin. The fossil-bearing sediments (Eskihisar, Tınaz, Salipașalar) are deposited above the lignite seam of the Sekköy member of the Eskihisar Formation. Based on palynological data, radiometric dating, vertebrate fossils and lithostratigraphic correlation, a Langhian to Serravallian (middle Miocene) age is suggested (see Güner and Denk, 2012). Soma-Deniş— Volcanic ash on top of the lower coal seam of the Soma Formation has been radiometrically dated as 17.3 ± 0.4 Ma (Becker-Platen et al., 1977). The small mammal fossils recovered in the basal coal-bearing part of Kymi, Euboea— Plant fossils in the Aliveri-Kymi Basin originate from the Marmarenia Formation, which is part of the Prinias Group. Small mammal fossils [AQ1] and palynology suggest a Burdigalian age (Velitzelos, 2002; Velitzelos et al., 2014). GALLEY PROOF — AJBD1400495 4 • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • A M E R I C A N J O U R N A L O F B O TA N Y Fig. 3. Map showing localities from which Smilax miohavanensis is known and their stratigraphic positions. Miocene stages follow Cohen et al. (2013). Parschlug, Styria— In the Parschlug basin the main lignite seam is overlain by claystone and marls. Plant fossils occur in the light claystone and typically in the hard, reddish-blackish marlstone (ironstone; Kovar-Eder et al., 2004). Based on biostratigraphic correlation, a Karpatian-early Badenian age was suggested (Kovar-Eder et al., 2004). This corresponds to Burdigalian to Langhian (late early to early middle Miocene; Harzhauser and Piller, 2007). Etymology— The specific epithet denotes the morphological similarity of the Miocene species with extant members of the Smilax Havanensis group according to Ferrufino-Acosta (2010; corresponding to clade B3 of Qi et al., 2013, the Central and South America lineage). Holotype— Specimen UCMP Kasapligil 6914 (Fig. 4, A-C). SYSTEMATIC PALEOBOTANY Family— Smilacaceae Vent., Tabl. Règn. Vég. 2: 146. 1799; nom. cons. Genus— Smilax L., Sp. Pl. 2: 1028. 1753. Species—Smilax miohavanensis Denk, D.Velitzelos, T.Güner et Ferrufino-Acosta, nom. nov. Basionym— Quercus aspera Unger (in Chloris protogaea: 108. 1847). Synonym (replaced)— Smilax aspera (Unger) Denk, D. Velitzelos, T.Güner et Ferrufino-Acosta, comb. nov. The new combination is a junior homonym of the extant Smilax aspera L. (in Sp. Pl. 2: 1028. 1753). In accordance with Articles 6.10, 6.11, and 41 of the International Code of Nomenclature for Algae, Fungi, and Plants (Melbourne Code, 2011), the replacement name Smilax miohavanensis is proposed. 1847 Quercus aspera Unger, p. 108, pl. 30, figs. 1 (right), 2, 3 (right). ?1847 Ilex sphenophylla Unger, p. 148, pl. 50, fig. 9. 1864 Ilex sphenophylla Unger, p. 12, pl. 3, figs. 1-6. 1864 Ilex cyclophylla Unger, p. 13, pl. 3, figs. 7, 8. 1864 Ilex neogena Unger, p. 13, pl. 3, figs. 9, 10, 13 (? non 11, 12). 1867 Ilex cyclophylla Unger, p. 76, pl. 13, fig. 14. 1867 Ilex neogena Unger, p. 76, pl. 13, figs. 16, 17 (non 15, 18). 2004 Mahonia (?) aspera Kovar-Eder et Kvaček, p. 57, pl. 13, figs. 3, 4, 6, 8 (?1, 2, 5). Paratypes—Specimens UCMP Kasapligil 5522b, 5524, 5720, 5861, 6072, 6927b, n.n. [Fig. 4, G, H] ; GBA 2002_01_0043, GBA 2002_01_0690, GBA 2002_01_0082, GBA BOT 2805; LMJ 76529, LMJ 76532 ; NHMW (Ettingshausen 7137) B.1878 VI 9140, NHMW (Ettingshausen 7173) 1878 VI 9140, NHMW (Ettingshausen 7473) B. 1878 VI 9476, NHMW (Ettingshausen) 1876 XVI 80; ISTO-F 01009. (1) Diatomites of the Dereköy pyroclastics, lower Miocene (Burdigalian; see Güner and Denk, 2012), Güvem area (Beș Konak, Kısilcahaman, Keseköy), Anatolia, Turkey. Collection Baki Kasapligil, Museum of Paleontology, University of California (UCMP), Berkeley. (2) Fine-grained lacustrine sediments of the Marmarenia Formation, lower Miocene (Burdigalian; see Velitzelos et al., 2014), Kymi (Kimi), Euboea. Collections GBA, NHMW. (3) Marls of the Soma Formation, Soma coal basin, and Sekköy Member of the Eskihisar Formation, Yatağan Basin, western Turkey (see Güner and Denk, 2012). Newly collected material housed at ISTO-F. (4) Reddish marlstone (ironstone), Parschlug basin, middle Miocene; see Kovar-Eder et al. (2004), Parschlug, Styria, Austria. Collections GBA, LMJ, NHMW. Specific diagnosis— Leaves simple, petiolate, elliptic to ovate, base obtuse, acute or cordate, apex obtuse to acute, mucronate, primary venation acrodromous (3-veined), secondary veins connecting central and lateral primary veins, leaf margin hyaline, deeply spinose or nearly entire margined, dentate axes approximately perpendicular to the tangent of the margin. Description— Leaves simple, elliptic to ovate to rounded, symmetric; petiolate, petiole flattened and twisted when inserted GALLEY PROOF — AJBD1400495 D E N K E T A L . — E U R A S I A N M I O C E N E S M I L A X • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • 5 into the lamina, 3 to 12 mm long; lamina 15 to 75 mm long, 9 to 40 mm wide, base obtuse, acute or cordate, apex obtuse to acute, mucronate; primary venation acrodromous (3-veined), secondary veins connecting the central and lateral primary veins, departing from the central vein at angles of 15° to 45°; secondary veins departing from the lateral primary veins at angles of 15° to 45° and forming irregularly angular loops toward the margin, from which veinlets branch off into the teeth; leaf margin conspicuously hyaline, dentate, deeply spiny, dentate axes approximately perpendicular to the tangent of the margin or slightly inclined to the tangent, number of teeth variable ranging from 0 to 30+ (0 to 18+ in specimens from Parschlug; to 30+ in specimens from Güvem; Figs. 4–7; Appendix S2, see Supplemental Data with the online version of this article). Remarks— The material from Anatolia (Güvem) is considered to represent the same species as slightly younger material from the historical collections of Greece (Kymi [Kumi]) and Austria (Parschlug) described by Unger in various papers between 1847 and 1867 (Unger, 1847, 1850a, b, 1864, 1867). Unger used various names to describe such leaves (Quercus aspera, Ilex sphenophylla, Ilex cyclophylla, Ilex stenophylla, Ilex aspera nomen nudum, Ilex neogena, and partly Smilax schmidtii nomen nudum; Appendix S1, see Supplemental Data with the online version of this article). The earliest names used by Unger were Quercus aspera and Ilex sphenophylla. Quercus aspera from the middle Miocene of Austria and the specimens from the early Miocene of Turkey described here are highly similar warranting inclusion in a single species. However, the epithet “aspera” is occupied by the extant species Smilax aspera L. Thus the new combination “Smilax aspera” would be an illegitimate junior homonym to the extant species. Ilex sphenophylla described at the same time (Unger, 1847) is based on much less material, the small size and preservation of which make an unambiguous determination impossible (Appendix S3, see Supplemental Data with the online version of this article). On the original lithographic plate (plate 50, fig. 9 in Unger, 1847), three specimens of I. sphenophylla are shown on a single slab. The actual specimens housed at LMJ are on individual slabs and only in one case (LMJ76515) do they provide a close match to the illustration (Appendix S3, A-C). The second specimen referred to fig. 9 by its original label (Appendix S3, D) resembles the two specimens shown on the left half of the illustration but does not exactly match either the one or the other. The third specimen could not be located in the collections of LMJ, NHMW or GBA. Noteworthy, the original specimens of Ilex sphenophylla to Unger’s publication from 1864 were also labeled as ”83. Ilex sphenophylla Ung. Chlor. Protog. p. 148. t. 50. f. 9”. Ilex sphenophylla was also described and figured by Unger in later publications. The specimen figured from Gleichenberg (Unger, 1850b) probably is not Smilax, while the leaves figured in Unger (1864) most likely belong to Smilax miohavanensis. DISCUSSION Generic affinity of the fossil species— The foliage described here has previously been attributed to various genera. Of these, Quercus and Ilex never have acrodromous primary veins. Mahonia Orientales group typically has leaflets with a primary venation that approaches an acrodromous pattern. However, the venation in Mahonia is better described as festooned brochidodromous with secondary veins forming loops followed by additional lateral loops (Güner and Denk, 2012). In contrast, the primary and lateral primary veins in Smilax are connected by secondary veins that depart from the primary veins at relatively low angles (20 to 55°). The highly variable degree of spiny dentition, from many teeth to no teeth at all on the same branch, in Smilax Havanensis group (Appendices S4, S5, see Supplemental Data with the online version of this article) is also seen in the fossil specimens but never in Mahonia. Furthermore, leaflets of Mahonia, except for the terminal ones, lack a distinct petiole and usually have an asymmetric base. A similar leaf dimorphism (entire to conspicuously dentate) as in Smilax Havanensis group is seen in modern species of Ilex but these cannot be confused with Smilax on account of their different leaf venation. The character combination encountered in the fossil leaves described here matches exactly the one in modern members of the Smilax Havanensis group: 3-5 (−7) acrodromous primary veins, central and lateral primary veins originating in one point, deeply spiny leaf margin, dental axes typically perpendicular to the tangent of the leaf margin, co-occurrence of leaves with densely spiny and entire margin. The lateral primary veins in the fossil species commonly are not clearly visible in the upper portion of the lamina (appearing as imperfect-developed acrodromous). Only few specimens (e.g., Fig. 7B; Appendix S6, see Supplemental Data with the online version of this article) show that the lateral veins actually go all the way up to the apical part of the leaf lamina. This may be in part due to the preservation of the leaves. However, modern members of Smilax including species of Smilax Havanensis group can also have relatively thin lateral major veins that are prominent only on the upper or lower leaf surface (Smilax aquifolium Ferrufino et Greuter, S. cristalensis Ferrufino et Greuter, S. cuprea Ferrufino et Greuter, S. ilicifolia Desv. ex Ham.; Ferrufino-Acosta, 2010) and species with chartaceous leaves commonly have lateral primary veins that do not reach all the way to the leaf apex but instead form a loop toward the primary vein (Appendix S7, see Supplemental Data with the online version of this article). Within the Havanensis group according to Ferrufino-Acosta (2010), Smilax miohavanensis resembles most closely Smilax aquifolium, S. coriacea Spreng., S. cristalensis, S. havanensis, S. ilicifolia, and S. populnea Kunth, all of which share ovate to elliptic leaves and the deeply spinose leaf margin. In addition, Smilax campestris Griseb. and S. spinosa Mill. of the Spinosa group may have leaves similar to Smilax miohavanensis, but the leaf margin commonly is less deeply spinose. In addition, these two species have highly polymorphic leaves including ovate and elliptic dentate leaves (approaching the Havanensis morphotypes), larger, long ovate ones with entire margin (Smilax Weberi group morphotype), and narrowly elliptic to lanceolate leaves with entire margin (Petiolata group morphotype). This may suggest that different morphotypes encountered in a fossil assemblage originated from the same species. For example, it cannot be ruled out that the foliage of Smilax miohavanensis from the early Miocene of Kimi (Fig. 7) and S. weberi from the same locality (Appendix S8, right, see Supplemental Data with the online version of this article), may actually have been produced by the same species. However, the lanceolate leaves commonly co-occurring with ovate-elliptic ones in modern species of the Spinosa group have never been found together with foliage of Smilax miohavanensis. GALLEY PROOF — AJBD1400495 6 • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • A M E R I C A N J O U R N A L O F B O TA N Y Fig. 4. Smilax miohavanensis nom. nov. from the early Miocene of Anatolia. (A) Leaf with distinct petiole and acrodromous venation. UCMP, Kasapligil 6914. (B, C) Details of A showing hyaline spiny margin and leaf base. (D) Small leaf. UCMP, Kasapligil 5524. (E) Detail of (D). (F) Imprint of an elliptic leaf. UCMP, Kasapligil 6072. (G, H) UCMP, Kasapligil n.n. (I) Basal half of a small leaf. UCMP, Kasapligil 6927b. Scale bar = 3 cm in A, F, G; 2 cm in B, C, D, and 1 cm in E, H, I. GALLEY PROOF — AJBD1400495 D E N K E T A L . — E U R A S I A N M I O C E N E S M I L A X • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • 7 Fig. 5. Smilax miohavanensis nom. nov. from the early Miocene of Anatolia. (A) Large leaf with distinct petiole. UCMP, Kasapligil 5861. (B) Detail of (A) showing hyaline spiny margin. (C) UCMP, Kasapligil 5522b. (D) Detail of (C) showing secondary veins departing from lateral primary vein. (E) Leaf fragment. UCMP, Kasapligil 5720. (F, G) Details of (E) showing hyaline, spiny leaf margin and secondaries connecting the central and lateral primary veins. Scale bar = 3 cm in A, B, C, E, and 2 cm in D, F, G. GALLEY PROOF — AJBD1400495 8 • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • A M E R I C A N J O U R N A L O F B O TA N Y Fig. 6. Smilax miohavanensis nom. nov. from the middle Miocene of Austria. (A) Coarsely dentate leaf. GBA 2002_01_0090. (B) LMJ 76532, as Quercus aspera, Unger 1847, pl. 30, fig. 1. (C) Broadly ovate leaf. NHMW (Ettingshausen 7137), B. 1878 VI 9140. (D) Entire-margined leaf. NHMW (Ettingshausen 7473), B. 1878 VI 9476. Scale bar = 1 cm in A-D. GALLEY PROOF — AJBD1400495 D E N K E T A L . — E U R A S I A N M I O C E N E S M I L A X • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • 9 Fig. 7. Smilax miohavanensis nom. nov. Broadly ovate leaf types. (A) Middle Miocene of the Yatağan Basin, western Turkey. ISTO-F 01009. (B) Early Miocene of Kymi, Euboea, Greece. NHMW (Ettingshausen), 1876 XVI 80. Scale bar = 1 cm in A, B. Smilax in the Cenozoic of Eurasia and North America— A large number of species of Smilax have been described from Cenozoic sediments of Eurasia and North America comprising a range from ovate to distinctly hastate foliage. The latter has commonly been compared to the modern Eurasian-African-Indian Smilax aspera and the eastern North American Smilax bona-nox L. Oldest reliable fossils of Smilax are from the Paleocene and Eocene of Europe and North America. Paleocene records are difficult to assign to a particular genus with certainty. For example, Heer (1859, p. 106, plate 133, fig. 24) figured the basal part of a small leaf from the middle Paleocene of Ménat (Selandian, ca. 61 Ma) which he referred to Smilax sagittifera Heer. This leaf fragment may belong to Smilax but is insufficient for a conclusive generic assignment. A leaf fragment from the Paleocene of western Greenland (Heer, 1874) assigned to Smilax lingulata Heer lacks the basal and apical lamina and cannot be assigned to any modern genus. Another leaf from the Paleocene of Atanikerluk assigned by Heer (1871) to Smilax grandifolia Unger is closely similar to modern Smilax by its leaf shape and venation. Lesquereux (1888) and Cockerell (1914) described Smilax carbonensis (Lesq.) Cockerell from the Paleocene/Eocene of Wyoming. The lamina lacks the apical part but might well belong to Smilax. Foliage with preserved epidermal features from the early Eocene of Australia (Conran et al., 2009) unequivocally belongs to Ripogonium and ovate and deeply cordate leaves with pre- served epidermal features from early middle Eocene (ca. 47 Ma) sediments of Messel, Germany, belong to Smilax (Wilde, 1989). From the middle Eocene of Hungary, Erdei and Rákosi (2009) described ovate foliage with obtuse leaf base and from the middle Eocene Clairborne Formation, Tennessee, Sun and Dilcher (1988) and Dilcher and Lott (2005) described ovate and cordate-hastate Smilax foliage, the latter of which they compared to the modern Smilax bona-nox Noteworthy, the two types of foliage encountered in the Clairborne Formation are morphologically very similar to the foliage from Messel. From East Asia, Ding et al. (2011) list a single Paleocene leaf record that probably needs to be reconfirmed (leaf dimensions are 1.6 × 1.1 cm) and one Eocene record. Overall, this suggests that Smilax was established at the latest by the early Eocene (late Ypresian, ca. 47 Ma) and possibly already by the middle Paleocene (Selandian, ca. 61 Ma; Ménat) or early/middle Paleocene (Danian/Selandian, ca. 62 Ma; Greenland), and would be in good agreement with the inferred Cretaceous stem age for the Smilacaceae (Vinnersten and Bremer, 2001; Janssen and Bremer, 2004). From Oligocene and Miocene sediments many species of Smilax were described from North America and Europe (Table. 1). All these records can be attributed to four morphotypes (“Formenkreis” according to Weyland, 1937). The “Sagittifera morphotype” comprises foliage with ovate to lanceolate leaves with a conspicuous cordate-hastate base and corresponds to the leaf type found in species belonging to clades A and B of modern GALLEY PROOF — AJBD1400495 10 • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • A M E R I C A N J O U R N A L O F B O TA N Y TABLE 1. Origin EU NALB NALB [AQ18] AM The Cenozoic record of Smilax. Age Paleocene Paleocene Paleocene Paleocene/Eocene Reference EU EU Early middle Eocene Early middle Eocene Heer, 1859 Heer, 1871 Heer, 1874 Lesquereux, 1878; Cockerell, 1914 Wilde, 1989 Wilde, 1989 EU Early middle Eocene Wilde, 1989 N Eur Middle Eocene Heer, 1869 N Eur N Eur N Eur EU EU AM AM Middle Eocene Middle Eocene Middle Eocene Middle Eocene Middle Eocene Middle Eocene Middle Eocene Heer, 1869 Heer, 1869 Heer, 1869 Erdei and Rákosi, 2009 Erdei and Rákosi, 2009 Dilcher and Lott, 2005 Dilcher and Lott, 2005 [AQ19] EU EU AM Late Eocene Late Eocene Late Eocene Knobloch et al., 1996 Knobloch et al., 1996 Cockerell, 1914 [AQ20] AM Late Eocene Knowlton, 1899 AM Late Eocene EA AM Eocene Eocene/Oligocene Chaney and Sanborn, 1933 Ding et al., 2011 Meyer, 2003 EU Oligocene EU Early Oligocene Kvaček and Walther, 1995 Saporta, 1863 EU Early Oligocene Saporta, 1863 EU EU Early Oligocene Early Oligocene AM EU Early Oligocene Middle Oligocene Saporta, 1863 Walther and Kvaček, 2007 Becker, 1969 Mai and Walther, 1978 EU Middle Oligocene Saporta, 1888 EU Middle Oligocene Saporta, 1888 EU EU EU Late Oligocene Late Oligocene Late Oligocene Wessel and Weber, 1856 Wessel and Weber, 1856 Wessel and Weber, 1856 EU Late Oligocene Saporta, 1862; 1873 EU Late Oligocene Saporta, 1865a EU Late Oligocene Saporta, 1865a EU Late Oligocene Kovar-Eder, 1982 EU Early Miocene Brongniart, 1828 EU Early Miocene Heer, 1855 EU Early Miocene Saporta, 1865b EU Early Miocene Saporta, 1865b Leaf shape and base Morphotype (MT) according to Weyland (1937) amended by present study Ovate, deeply cordate Ovate, cordate Lanceolate (fragment) Broadly ovate, cordate “Sagittifera MT” No MT assigned “Petiolata MT” No MT assigned Ovate, obtuse-decurrent Narrowly ovate-elliptic to broadly ovate, shallowly cordate-hastate Elliptic to lanceolate, obtuse to acute Broadly ovate, obtuse-hastate Ovate, obtuse Ovate, otuse-decurrent Narrowly ovate-elliptic Ovate, obtuse Ovate, obtuse Ovate, obtuse-decurrent Ovate, deeply cordate-hastate Ovate-elliptic, acute Elliptic, acute Ovate, truncate-decurrent “Weberi MT” “Petiolata MT” to widespread unspecific type Broadly ovate-elliptic, shallowly cordate Ovate, acute No MT assigned Ovate, round Broadly ovate, truncate No MT assigned No MT assigned Ovate, oblong, slightly hastate Narrowly ovate, shallowly cordate Narrowly ovate, shallowly sagittate Lanceolate, auriculate Ovate, oblong, slightly hastate Ovate, shallowly cordate Narrowly ovate-elliptic Ovate, deeply cordate-hastate Narrowly ovate, shallowly cordate Ovate-elliptic to oblong Ovate-elliptic Ovate, deeply cordate-hastate Roundish, deeply cordate and notched Broadly ovate, hastate-truncate Lanceolate, auriculate Long ovate, deeply cordate Narrowly ovate, deeply cordate Broadly ovate, shallowly cordate Narrowly ovate, deeply cordate ? Taxon name Smilax sagittifera Heer Smilax grandifolia Heera Smilax lingulata Heer Smilax carbonensis Cockerell cf. Smilax sp. 1 cf. Smilax sp. 2 No MT assigned ? Smilax sp. 3 “Weberi MT” Smilax paliformis Heer “Weberi MT” No MT assigned “Petiolata MT” “Weberi MT” “Weberi MT” No MT assigned “Sagittifera MT” Smilax reticulata Heera Smilax convallium Heer Smilax lingulata Heer Smilax sp. 1 ? Smilax sp. Smilax sp. 1 Smilax sp. 2 No MT assigned No MT assigned No MT assigned “Weberi MT” Smilax sp. 1 Smilax sp. 2 Smilax labidurommae Cockerell Smilax lamarensis Knowlton Smilax goshensis Chaney and Sanborn Smilax sp. Smilax labidurommae Cockerell Smilax sp. “Petiolata MT” Smilax linearis Saporta “Sagittifera MT” Smilax sagittiformis Saporta Smilax elongata Saporta Smilax weberi Wessel No MT assigned No MT assigned “Weberi MT” No MT assigned “Petiolata MT” “Sagittifera MT” Smilax trinervis Morita Smilax petiolata Weyland Smilax coquandi Saporta “Sagittifera MT” Smilax philiberti Saporta “Weberi MT” “Weberi MT” “Sagittifera MT” Smilax weberi Wessel Smilax ovata Wessela Smilax remifolia Wessel No MT assigned Smilax rotundilobus Saporta Smilax garguieri Saporta “Weberi MT” No MT assigned “Sagittifera MT” “Sagittifera MT” Smilax abscondita Saporta Smilax sp. “Weberi MT” Smilacites hastata Brongniarta Smilax grandifolia Heera “Sagittifera MT” Smilax hastata Saportaa No MT assigned Smilax appendiculata Saporta GALLEY PROOF — AJBD1400495 D E N K E T A L . — E U R A S I A N M I O C E N E S M I L A X • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • 11 TABLE 1. Origin Continued. Age Reference EU Early Miocene Saporta, 1865b EU Early Miocene EU Early Miocene Knobloch and Kvaček, 1976 Hably, 1983 EU Early Miocene Hably, 1983 EU Early Miocene Hably, 1983 EU Early Miocene Bůžek et al., 1996 EU Early Miocene EU Early Miocene Knobloch and Kvaček, 1996 This study AM Early Miocene Chaney, 1920 EA EA EU Early/middle Miocene Early/middle Miocene Middle Miocene Huzioka,1963 Huzioka,1963 This study EU Middle Miocene Unger, 1847 EU Middle Miocene Unger, 1847 EU EU EU Middle Miocene Middle Miocene Middle Miocene Heer, 1855 Heer, 1855 Heer, 1855 EU Middle Miocene Heer, 1855 EU Middle Miocene Heer, 1859 EU Middle Miocene Heer, 1859 EU Middle Miocene Unger, 1869 EU Middle Miocene Hantke, 1954 [AQ22] EU Middle Miocene EU [AQ21] Leaf shape and base Broadly ovate, shallowly cordate Broadly ovate, shallowly cordate Ovate, acute-decurrent to truncate Ovate-oblong, cordate-hastate ovate-elliptic, obtuse Broadly ovate, deeply cordate Broadly ovate, deeply cordate-hastate Ovate-elliptic, acute, obtuse to cordate Ovate, shallowly cordate-hastate Ovate, cordate Ovate, acute-obtuse Ovate-elliptic, acute, obtuse to cordate Broadly ovate, deeply cordate Narrowly ovate, deeply cordate Ovate, deeply cordate Broadly ovate, hastate Broadly ovate, shallowly cordate Narrowly ovate, shallowly cordate Ovate, deeply cordate-hastate Ovate, deeply cordate-saggitate ovate-elliptic, obtuse Morphotype (MT) according to Weyland (1937) amended by present study Taxon name No MT assigned Smilax asperula Saporta “Weberi MT” Smilax weberi Wessel “Weberi MT” Smilax weberi Wessel No MT assigned Smilax aspera L. fossilis No MT assigned “Sagittifera MT” Smilax borsodensis Andreánszky Smilax sagittifera Heer “Sagittifera MT” Smilax sagittifera Heer “Havanensis MT” Smilax miohavanensis nom. nov. Smilax magna Chaney “Sagittifera MT” No MT assigned No MT assigned “Havanensis MT” “Sagittifera MT” “Weberi MT” No MT assigned Smilax minor Morita Smilax trinervis Morita Smilax miohavanensis nom. nov. Smilacites grandifolia Ungera Smilacites sagittata Ungera Smilax sagittifera Heer Smilax obtusiloba Heer Smilax parvifolia Heerb “Sagittifera MT” Smilax angustiloba Heer “Sagittifera MT” Smilax obtusangula Heer Smilax auriculata Heer “Sagittifera MT” “Sagittifera MT” No MT assigned No MT assigned No MT assigned Smilax hyperborea Unger Smilax sagittifera Heer emend. Hantke Smilax borsodensis Andreánszky Smilax hyperborea Unger Smilacites grandifolia Ungera Smilax praeaspera Andreánszky Smilax cf. oldhamii Miq. No MT assigned Smilax grandifolia Heera No MT assigned “Petiolata MT” “Weberi MT” “Sagittifera MT” Taxon LXXII Taxon LXXIII Smilax weberi Wessel Smilax wardii Lesquereux Smilax reticulata Heera Smilax (cf. S. magna Chaney) Smilax sp. Smilax trinervis Morita Smilax petiolata Weyland Smilacites cocchiana Massalongo Smilacites spadaeana Massalongo “Sagittifera MT” Andreanszky, 1959 Broadly ovate, deeply cordate-hastate ovate-elliptic, obtuse Middle Miocene Andreanszky, 1959 ovate-elliptic, obtuse No MT assigned EU Middle Miocene Andreanszky, 1959 “Sagittifera MT” EU Middle Miocene Andreanszky, 1959 EU Middle Miocene Andreanszky, 1959 EU Middle Miocene Iljinskaja, 1964 EU EU N Eur AM Middle Miocene Middle Miocene Middle Miocene Middle Miocene Ferguson, 1971 Ferguson, 1971 Christensen, 1975 Lesquereux, 1888 AM AM Middle Miocene Middle Miocene Hollick, 1936 Smiley et al., 1975 Broadly ovate, deeply cordate Broadly ovate, deeply cordate Narrowly ovate, obtuse-acute Broadly ovate, deeply cordate Ovate, acute-obtuse Narrowly ellipitc Ovate, obtuse-decurrent Narrowly ovate, deeply cordate Ovate, obtuse Ovate, obtuse-decurrent NALB EA EU Middle Miocene Middle Miocene Late Miocene Denk et al., 2005 Tanai and Suzuki, 1963 Weyland, 1957 Ovate, obtuse-decurrent Ovate-roundish, obtuse Narrowly ovate-elliptic No MT assigned No MT assigned “Petiolata MT” [AQ23] EU Late Miocene Massalongo, 1859 Elliptic No MT assigned EU Late Miocene Massalongo, 1859 Ovate-oblong, slightly hastate “Weberi MT” No MT assigned “Sagittifera MT” “Weberi MT” “Weberi MT” GALLEY PROOF — AJBD1400495 12 • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • A M E R I C A N J O U R N A L O F B O TA N Y TABLE 1. Origin Continued. Age Reference EU Late Miocene Massalongo, 1859 EU Late Miocene EU Leaf shape and base Morphotype (MT) according to Weyland (1937) amended by present study “Weberi MT” Massalongo, 1859 Broadly ovate, shallowly cordate Ovate, cordate Late Miocene Massalongo, 1859 Ovate, deeply cordate “Sagittifera MT” EU Late Miocene Massalongo, 1859 Ovate, deeply cordate “Sagittifera MT” EU Late Miocene Massalongo, 1859 No MT assigned EU Late Miocene Laurent, 1908 EU Late Miocene Berger, 1957 EU EU Late Miocene Late Miocene Berger, 1957 Kolakovski, 1964 EU Late Miocene Knobloch, 1969 Broadly ovate, slightly hastate Broadly ovate, cordate-hastate Ovate, hastate- to saggitate-cordate Elliptic, acute Ovate to ovate-oblong, truncate, cordate to hastate Ovate, cordate-hastate AM Late Miocene Brown, 1937 AM Late Miocene EA EA Late Miocene Late Miocene Chaney and Axelrod, 1959 Uemura, 1988 Ding et al., 2011 AM Early Pliocene Axelrod, 1980 EU Early Pliocene Pop, 1936 Ovate, shallowly cordate-hastate Ovate, shallowly cordate-hastate Ovate, elliptic Ovate, acute-obtuse Broadly ovate, deeply cordate Broadly ovate to ovate-oblong, hastate No MT assigned No MT assigned “Sagittifera MT” No MT assigned No MT assigned No MT assigned Taxon name Smilacites orsiniana Massalongo Smilacites nestiana Massalongo Smilacites pulchella Massalongo Smilacites sagittifera Massalongo Smilacites proxima Massalongo Smilax mauritanica Desf. fossilis Smilax hastata Brongniarta Smilax cf. ovata Wessel Smilax aspera L. fossilis “Sagittifera MT” Smilax hastata Brongniarta Smilax magna Chaney “Sagittifera MT” Smilax magna Chaney No MT assigned No MT assigned Smilax trinervis Morita Smilax tiantaiensis Ding et al. Smilax remingtonii Axelrod Smilax aspera L. No MT assigned No MT assigned Notes: aInvalid homonym of: Smilax auriculata Walter 1788; S. grandifolia Buckley 1843; S. hastata Jacq. 1760; S. ovata Duhamel 1803; S. reticulata Desv. 1825; S. sagittata Desv. ex Ham. 1825. Taxonomic treatment based on World Checklist of Monocotyledons, website http://apps.kew.org/wcsp/home.do. Plant name not in italics = probably not belonging to Smilax. EU = Europe and Asia Minor, NALB = involving the North Atlantic land bridge, AM = North America, N Eur = North Europe, EA = East Asia. Smilax (Fig. 1). The “Weberi morphotype” encompasses ovate (to elliptic and oblong) foliage with a rounded or cordate to slightly hastate base and corresponds to the most widespread leaf type in modern Smilax found in all four main clades of the genus. These two morphotypes were also recognized in previous work (e.g., Weyland, 1937; Christensen, 1975). A further morphotype detected in the fossil record comprises distinctly narrowly elliptic to lanceolate leaves, hereafter called “Petiolata morphotype” based on the foliage of Smilax petiolata Weyland (Fig. 1, Table 1). This morphotype, again, is also recognized among modern species of Smilax and occurs in clades B, C, and D both in New World and Old World species. The last morphotype is here called “Havanensis morphotype” referring to the early and middle Miocene Smilax miohavanensis and corresponding to the Caribbean to South American species of the Smilax Havanensis group according to Ferrufino-Acosta (2010; Fig. 2). While the first three morphotypes are not diagnostic for a particular modern clade of Smilax, the Havanensis morphotype is restricted to clade B (Qi et al., 2013; Fig. 1). For the remaining (fossil) morphotypes, the high degree of parallel evolution of leaf morphology observed in the modern species of Smilax is also seen in the fossil record. Nevertheless, it is striking that identical leaf types are recorded both from North American and from European sedimentary formations. For example, broadly ovate foliage with leaf bases ranging from hastate to hastate-cordate was reported from the Eocene/ Oligocene Florissant beds of Colorado (Cockerell, 1914; Meyer, 2003) and from the Oligocene flora of Suletice-Berand of Bohemia (Kvaček and Walther, 1995). A leaf referred to as Smilax labidurommae Cockerell from the Eocene/Oligocene Florissant Formation (Meyer, 2003, fig. 117) is virtually identical to specimens described as Smilacites garguieri Saporta from the lower Oligocene of France (Saporta, 1865a). To our knowledge, the fossil record of Smilax from East Asia is not as rich as that for Europe and North America. Miocene leaf remains described by Morita (1931), Chaney and Axelrod [AQ2] (1959), Tanai and Suzuki (1963), Huzioka (1963), and Ding et al. (2011) all belong to the general leaf type found in all four major clades of Smilax. Also from Central and South America little is known about the fossil history of Smilax, which may to some extent explain why the Smilax Havanensis group has no fossil record in the New World. According to Graham (2010), pollen referred to as Smilacites has been reported from late Eocene to Pliocene sedimentary formations of Mexico and Argentina. In Parschlug, Smilax miohavanensis (Havanensis group) cooccurs with S. sagittifera (Sagittifera group morphotype; Kovar-Eder et al., 2004). In Kymi, Smilax miohavanensis cooccurs with S. weberi (as S. schmidtii Unger nom. nud. in Un- GALLEY PROOF — AJBD1400495 D E N K E T A L . — E U R A S I A N M I O C E N E S M I L A X • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • 13 ger, 1867; Appendix S8, see Supplemental Data with the online version of this article). Integrating plant fossils in time-calibrated phylogenetic studies— In a recent study, Chen et al. (2014) used fossils to constrain nodes of intrageneric groups/species of Smilax. Two fossil species, Smilax trinervis Morita and S. tiantaiensis Ding et al. were used to constrain the ages of a S. davidiana A.DC., -S. china L. clade and of S. glaucochina Warb. ex Diels. Chen et al. (2014) indicated an age of 6 Ma for S. trinervis based on material from the late Miocene Takamine flora (Honshu, Japan; Uemura, 1988). This is problematic as Smilax trinervis (including S. minor Morita) was a common element in the late early and middle Miocene of Japan (e.g., late early Miocene Utto flora, Huzioka, 1963; see also Uemura, 1988). Hence, a node age constrained with S. trinervis would more appropriately be around 17-16 Ma. Furthermore, the leaf morphology of Smilax trinervis cannot be used to place this fossil taxon close to the living S. china. Instead, the fossil resembles several species of the clade comprising the species S. glabra Roxb. to S. ferox Wall. ex Kunth (fig. 3a in Chen et al., 2014; see also Morita, [AQ3] 1933). The second fossil, Smilax tiantaiensis, resembles the modern S. china, S. davidiana, S. glaucochina, S. cyclophylla Warb., and S. stans Maxim. according to the original paper by Ding et al. (2011), and therefore cannot be used to constrain the age of a single species. A third fossil species used to constrain an intrageneric node in the study by Chen et al. (2014) was S. magna Chaney described from early and middle Miocene deposits of western North America (Table 1; Chaney, 1920; Chaney and Axelrod, 1959; Smiley et al., 1975). Smilax magna was used to constrain the age of the S. hispida lineage to 18.5 Ma. Smilax magna shows a characteristic variability in leaf morphology (Weberi morphotype, Sagittifera morphotype, and unspecific leaf morphology) that includes leaf types encountered in various species of clade B of Qi et al. (2013; e.g., S. bona-nox, S. glauca Walter, S. hispida Raf., and S. rotundifolia L., cf. Chaney and Axelrod, 1959). Therefore, the fossil species could just as well have been used to constrain the age of the clade comprising S. glauca to S. rotundifolia, or the entire New World Smilax-clade of Chen et al. (2014, fig. 3a). Alternatively, fossils displaying very similar leaf variability to S. magna are also known from Eocene and Oligocene sediments of North America and Central Europe (Dilcher and Lott, 2005; Wessel and Weber, 1856). In other words, the three fossil taxa selected by Chen et al. (2014) cannot convincingly be assigned to a particular in-group node or their stratigraphic age was misinformed. This calls for caution when fossil are included in time-calibrated phylogenetic studies. Biogeographic patterns-multiple intercontinental disjunctions—The molecular phylogeny of Smilax can be used as a phylogenetic framework for assessing the evolution of basic leaf types in Smilax and the importance of fossils for biogeographic reconstructions. As discussed in the previous section most leaf types in Smilax are shared by all major clades (Figs. 1, 2). Therefore, leaf fossils assigned to Smilax may be of limited value for biogeographic reconstructions. However, the contemporaneous presence of highly similar leaf types in European and North American sediments in the Eocene and possible already in the Paleocene (see above) suggests that the North Atlantic might have played an important role for range expansion in the early Cenozoic. This is in agreement with the phylogenetic framework, in which the first diverging clades A and B of Qi et al. (2013) represent (western) Eurasian and American taxa. In this study, the focus is on the New World clade of Smilax (clade B). This clade comprises three New World—Old World disjunctions that must have been achieved independently in the Neogene (Qi et al., 2013; Zhao et al., 2013). Zhao et al. (2013) inferred a divergence age between the Hispida group and the remainder of the New World clade of Smilax of ca. 25 Ma (Fig. 2). Hence, the divergence ages of the three New World—Old World disjunctions in clades B4 and B5, and in the Smilax Havanensis group (B3) must be younger. From the early and middle Miocene of Central Europe, northern Europe and North America closely similar leaf types have been referred to as Smilax weberi and S. cf. magna, respectively (Knobloch and Kvaček, 1976; Christensen, 1975; Smiley et al., 1975). These leaves clearly belong to the Weberi morphotype and possess the same type of epidermal characteristics. Although it cannot be determined whether these similarities evolved in parallel it is noteworthy that the Miocene fossils are highly similar to the modern species of clade B5 and B4 involving North Americanwestern Eurasian disjunctions (Fig. 2). The same is true for the fossil taxon pair Smilax sp. from the middle Miocene of Iceland (Denk et al., 2005, 2011) and leaf fossils from the Miocene of Alaska assigned to S. lingulata Heer by Hollick (1936). The paleobotanical records and modern trans-Atlantic sister group relationships are strongly suggestive of migration across the North Atlantic. The Greenland-Scotland Transverse Ridge consisting of a chain of islands might have acted as a corridor (the so-called North Atlantic land bridge) for temperate plants with different dispersal modes until the latest Miocene (Grímsson and Denk, 2007; Denk et al., 2010, 2011). The modern species of clades B4 and B5 displaying New World-Old World disjunctions thrive in humid warm temperate climates (Denk et al., 2001; Flora of North America Editorial Committee, 2003) comparable to the middle Miocene conditions in the northern North Atlantic (Denk et al., 2011). This may imply that the modern distribution of some members of these clades in tropical regions (e.g., Smilax velutina Killip et C.V.Morton) represents a more recent adaptation. The fossil Smilax described here provides evidence that also clade B3 of Qi et al. (2013) that exclusively consists of New World species today had a wider distribution during the Neogene. Given the modern range of species of clade B3 and the divergence ages reconstructed by Zhao et al. (2013), a more southern route than the North Atlantic land bridge could be assumed. Trans-Atlantic dispersal at tropical latitudes has been invoked for numerous angiosperm lineages (Thorne, 1973; Renner, 2004) and for animals (platyrrhine monkeys, rodents; Houle, 1999; Poux et al., 2008; Rowe et al., 2010) based on fos- [AQ4] sil data and molecular phylogenetic studies. Dispersal across the tropical Atlantic could have happened in both directions based on the available wind and sea currents (cf. Fratantoni, 2001; Renner, 2004). The most likely means for crossing of the tropical Atlantic during the Early Cenozoic until the Miocene appears to be so-called floating islands (Houle, 1998). Based on the geographical and stratigraphic distribution of Smilax miohavanensis (Appendix S9, see Supplemental Data with the online version of this article), two dispersal scenarios can be inferred. First, migration across the Atlantic could have been achieved via Africa south of 30°N and crossing of the Atlantic by means of floating islands. Alternatively, as in the more tem- GALLEY PROOF — AJBD1400495 14 • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • A M E R I C A N J O U R N A L O F B O TA N Y perate members of clade B4 and B5, crossing could have been via the North Atlantic land bridge, although temperatures at higher latitudes probably were too cool for subtropical-tropical plants (see Denk et al., 2011, for an assessment of the paleotemperatures across the North Atlantic land bridge during the Neogene). If the present distribution of the Smilax Havanensis group is the result of a niche shift similar to the one seen in Smilax velutina (see above in previous paragraph) then its (temperate) ancestors may have been able to cross the North Atlantic via the North Atlantic land bridge. Smilax is mainly dispersed by birds (Riddley, 1930). Important dispersers of seeds of Smilax are crows (Corvidae), thrushes (Turdidae; e.g., Turdus, Sialis), woodpeckers (Picidae), and ducks (Anatidae). A dated phylogeny of Corvus (crows) suggested that the genus originated in the Palearctic in the early Miocene and that the Eurasian-North American genus Pica (magpies) is their modern sister group. Within crows, early phylogenetic splits involve Palearctic-Nearctic/ Caribbean disjunctions dated to middle Miocene (Jønsson et [AQ5] al., 2012). These authors explained the modern distribution pattern by trans-Atlantic dispersal. This scenario, if correct, would be closely similar to the pattern seen in the MioceneRecent distribution of the Smilax Havanensis group. Various Caribbean-Nearctic sister group relationships in Corvus may also support the statement by Qi et al. (2013) that Caribbean members of the Smilax Havanensis group might be derived from mainland lineages. Similarly, thrushes appear to have crossed the Atlantic several times during the late Neogene (Voelker et al., 2009) and ducks underwent a global diversification also during the Miocene (Gonzalez et al., 2009). Although it is clear that these birds disperse Smilax, it is not clear how the fruits traveled across the Atlantic. For example, fruit-eating birds are unlikely to retain the seeds across the Atlantic (Renner, 2004). The most likely explanations for the trans-Atlantic crossings of Smilax and other plants during the Neogene are a combination of bird dispersal and transport on floating islands (southern North Atlantic) or of bird dispersal and island hopping (via the northern North Atlantic). LITERATURE CITED AXELROD, D. I. 1980. Contributions to the Neogene Paleobotany of Central California. University of California Publications in Geological Sciences 121: 1–212. BECKER, H. F. 1969. Fossil plants of the Tertiary Beaverhead Basins in southwestern Montana. Palaeontographica B 127: 1–142. BECKER-PLATEN, J. D., L. BENDA, AND P. STEFFENS. 1977. Litho- und biostratigraphische Deutung radiometrischer Altersbestimmungen aus dem Jungtertiär der Türkei. Geologisches Jahrbuch B 25: 139–167. BERGER, W. 1957. Untersuchungen an der obermiozänen (Sarmatischen) Flora von Gabbro (Monti Livornesi) in der Toskana. Palaeontographia Italica 51: 1–96. BERGER, W., AND F. ZABUSCH. 1953. Die obermiozäne (sarmatische) Flora der Türkenschanze in Wien. Neues Jahrbuch für Geologie und Palaontologie. Abhandlungen 98: 226–276. [AQ6] BERRY, E. W. 1911. The flora of the Raritan Formation. Geological Survey of New Jersey, Bulletin 3: 1–233. BRONGNIART, M. A. 1828. Notice sur les plantes d’Armissan, pres Narbonne. Annales des Sciences Naturelles 15: 43–51. BROWN, R. W. 1937. Additions to some fossil floras of the western United States. U.S. Geological Survey Professional Paper 186J: 163–206. BŮŽEK, C., F. HOLÝ, AND Z. KVAČEK. 1996. Early Miocene flora of the Cypris Shale (western Bohemia). Acta Musei Nationalis Pragae B 52: 1–72. CHANEY, R. W. 1920. The flora of the Eagle Creek Formation. Contributions from Walker Museum, University of Chicago 2: 115–181. CHANEY, R. W., AND D. I. AXELROD. 1959. Miocene floras of the Columbia Plateau. Part 2: Systematic considerations. Carnegie Institution of Washington Publication 617, 135–237. Carnegie Institution of Washington, Washington, District of Columbia, USA. CHANEY, R. W., AND E. I. SANBORN. 1933. The Goshen Flora of west central Oregon. Carnegie Institution of Washington Publication 439. Carnegie Institution of Washington, Washington, District of Columbia., USA. CHEN, C., Z.-C. QI, X.-H. XU, H. P. COMES, M. A. KOCH, X.-J. JIN, C.-X. FU, AND Y.-X. QIU. 2014. Understanding the formation of Mediterranean-African-Asian disjunctions: Evidence for Miocene climate-driven vicariance and recent long-distance dispersal in the Tertiary relict Smilax aspera (Smilacaceae). New Phytologist 204: 243–255. CHEN, S. C., X. P. ZHANG, S. F. NI, C. X. FU, AND K. M. CAMERON. 2006b. The systematic value of pollen morphology in Smilacaceae. Plant Systematics and Evolution 259: 19–37. CHRISTENSEN, E. F. 1975. The Søby Flora: Fossil plants from the Middle Miocene delta deposits of the Søby-Fasterholt area, Central Juland, Denmark. Part I. Danmarks Geologiske Undersøgelse 2: 1–41. COCKERELL, T. D. A. 1914. Two new plants from the Tertiary rocks of the west. Torreya 18: 135–137. COHEN, K. M., S. C. FINNEY, P. L. GIBBARD, AND J.-X. FAN. 2013 (updated). The ICS international chronostratigraphic chart. Episodes 36: 199–204. Website http://www.stratigraphy.org/ICSchart/ ChronostratChart2014-02.pdf. CONDIT, C. 1938. The San Pablo flora of western Central California. Carnegie Institution of Washington Publication 476, 217–286. Carnegie Institution of Washington, Washington, District of Columbia, USA. CONRAN, J. G., R. J. CARPENTER, AND G. J. JORDAN. 2009. Early Eocene Ripogonium (Liliales: Ripogonaceae) leaf macrofossils from southern Australia. Australian Systematic Botany 22: 219–228. DAGHLIAN, C. P. 1981. A review of the fossil record of monocotyledons. Botanical Review 47: 517–555. DENK, T., N. FROTZLER, AND N. DAVITASHVILI. 2001. Vegetational patterns and distribution of relict taxa in humid temperate forests and wetlands of Georgia (Transcaucasia). Biological Journal of the Linnean Society. Linnean Society of London 72: 287–332. DENK, T., F. GRÍMSSON, AND Z. KVAČE K. 2005. The Miocene floras of Iceland and their significance for late Cainozoic North Atlantic biogeography. Botanical Journal of the Linnean Society. Linnean Society of London 149: 369–417. DENK, T., F. GRÍMSSON, AND R. ZETTER. 2010. Episodic migration of oaks to Iceland: Evidence for a North Atlantic “land bridge” in the latest Miocene. American Journal of Botany 97: 276–287. DENK, T., F. GRÍMSSON, R. ZETTER, AND L. A. SÍMONARSON. 2011. Late Cainozoic floras of Iceland. 15 Million years of vegetation and climate history in the northern North Atlantic. Topics in geobiology, vol. 35. Springer-Verlag, Amsterdam, Netherlands. DILCHER, D. L. 1974. Approaches to the identification of angiosperm leaf remains. Botanical Review 40: 1–157. DILCHER, D. L., AND T. A. LOTT. 2005. A middle Eocene fossil plant assemblage (Powers Clay Pit) from western Tennessee. Bulletin of the Florida Museum of Natural History 45: 1–43. DING, S.-T., B.-N. SUN, J.-Y. WU, AND X.-C. LI. 2011. Miocene Smilax leaves and associated epiphyllous fungi from Zhejiang, East China and their paleoecological implications. Review of Palaeobotany and Palynology 165: 209–223. OSI. 2009. The middle Eocene flora of Csordakút (N ERDEI, B., AND L. RÁK Hungary). Geologica Carpathica 60: 43–57. FERGUSON, D. K. 1971. The Miocene flora of Kreuzau, western Germany. 1. The leaf remains. Verhandelingen der Koninklijke Nederlandsche Akademie van Wetenschappen. Afdeeling natuurkunde, tweed sectie 60: 1–297. [AQ7] [AQ8] [AQ9] [AQ10] GALLEY PROOF — AJBD1400495 D E N K E T A L . — E U R A S I A N M I O C E N E S M I L A X • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • 15 FERRUFINO-ACOSTA, L. 2010. Taxonomic revision of the genus Smilax (Smilacaceae) in Central America and the Caribbean Islands. Willdenowia 40: 227–280. FILIPPOVA, G. G. 1994. Coniacian flora of the northern part of the Pekulney Range. Kolyma 3: 13–21 (in Russian). FLORA OF CHINA EDITORIAL COMMITTEE. 2000. Flora of China. Volume 24: Flagellariaceae through Marantaceae. Science Press, Beijing, China and Missouri Botanical Garden Press, St. Louis, Missouri, USA. FLORA OF NORTH AMERICA EDITORIAL COMMITTEE. 2003. Flora of North America: North of Mexico. Volume 26: Magnoliophyta: Liliidae: Liliales and Orchidales. Oxford University Press, New York, New York, USA. FRATANTONI, D. M. 2001. North Atlantic surface circulation during the 1990s observed with satellite-tracked drifters. Journal of Geophysical Research 106: 22067–22093. GONZALEZ, J., H. DÜTTMANN, AND M. WINK. 2009. Phylogenetic relationships based on two mitochondrial genes and hybridization patterns in Antidae. Journal of Zoology 279: 310–318. GRAHAM, A. 2010. Late Cretaceous and Cenozoic history of Latin American vegetation and terrestrial environments. Missouri Botanical Garden Press, St. Louis, Missouri, USA. GREENWOOD, D. R., AND J. G. CONRAN. 2000. The Australian Cretaceous and Tertiary monocot fossil record. In K. L. Wilson, and D. A. Morrison [eds.], Monocots: Systematics and evolution, 52-59. CSIRO, Melbourne, Australia. GRÍMSSON, F., AND T. DENK. 2007. Floristic turnover in Iceland from 15 to 6 Ma—extracting biogeographic signals from fossil floral assemblages. Journal of Biogeography 34: 1490–1504. GÜNER, T., AND T. DENK. 2012. The genus Mahonia in the Miocene of Turkey: Taxonomy and biogeographic implications. Review of Palaeobotany and Palynology 175: 32–46. [AQ11] HABLY, L. 1983. Early Miocene plant fossils from Ipolytarnóc, N Hungary. Geologica Hungarica. Series Palaeontologica 45: [AQ12] 77–255. HANTKE, R. 1954. Die fossile Flora der obermiozänen Oehninger-Fundstelle Schrotzburg. Denkschriften der Schweizerischen Naturforschenden Gesellschaft 80: 27–118. HARZHAUSER, M., AND W. E. PILLER. 2007. Benchmark data of a changing sea—Palaeogeography, Palaeobiogeography and events in the Central Paratethys during the Miocene. Palaeogeography, Palaeoclimatology, Palaeoecology 253: 8–31. HEER, O. 1855. Flora Tertiaria Helvetiae. Vol. 1. J. Wurster & Compagnie, Winterthur, Switzerland. HEER, O. 1859. Flora Tertiaria Helvetiae. Vol. 3. J. Wurster & Compagnie, Winterthur, Switzerland. HEER, O. 1869. Miocene baltische flora. W. Koch, Königsberg, Germany. HEER, O. 1871. Flora fossilis Arctica. Vol. 2. Contributions to the fossil flora of North-Greenland. J. Wurster & Compagnie, Winterthur, Switzerland. HEER, O. 1874. Nachträge zur miocenen Flora Grönlands. Kongliga Svenska Vetenskaps-Akademiens Handlingar 13: 1–29. (included in Flora fossilis Arctica, vol. 3. 1875. J. Wurster & Compagnie, Winterthur, Switzerland) HOLLICK, A. 1936. The Tertiary floras of Alaska.U.S. Geological Survey Professional Paper 182, 1–185. HOULE, A. 1998. Floating islands: A mode of long-distance dispersal for small and medium-sized terrestrial vertebrates. Diversity & Distributions 4: 201–216. HOULE, A. 1999. The origin of platyrrhines: An evaluation of the Antarctic scenario and the floating island model. American Journal of Physical Anthropology 109: 541–559. HUZIOKA, K. 1963. The Utto flora of northern Honshu. In R. W. Chaney [ed.], Tertiary floras of Japan, Miocene floras, 151-216. The Collaborating Association to commemorate the 80th anniversary of the geological survey of Japan. Geological Survey of Japan, Tokyo, Japan. ILJINSKAJA, I. A. 1964. The Tortonian flora of Swoszowice. In A. L. Takhtajan [ed.], Palaeobotanica 5, 113-144. Komarov Botanical Institute, Russian Academy of Sciences, St. Petersburg, Russia. JANSSEN, T., AND K. BREMER. 2004. The age of major monocot groups inferred from 800+ rbcL sequences. Botanical Journal of the Linnean Society. Linnean Society of London 146: 385–398. KAYA, O., E. ÜNAY, F. GÖKTAȘ, AND G. SARAÇ. 2007. Early Miocene stratigraphy of Central and West Anatolia, Turkey: Implications for the tectonic evolution of the Eastern Aegean area. Geological Journal 42: 85–109. KIM, J. S., J.-K. HONG, M. W. CHASE, M. F. FAY, AND J.-H. KIM. 2013. Familial relationships of the monocot order Liliales based on a molecular phylogenetic analysis using four plastid loci: matK, rbcL, atpBand atpF-H. Botanical Journal of the Linnean Society. Linnean Society of London 172: 5–21. KNOBLOCH, E. 1969. Tertiäre Floren von Mähren. Moravske Museum, Brno, Czech Republic. KNOBLOCH, E., AND Z. KVAČE K. 1976. Miozäne Blätterfloren vom Westrand der Böhmischen Masse. Rozpravy Ústředního Ústavu Geologického 42: 1–131. KOLAKOVSKI, A. A. 1964. A Pliocene flora of the Kodor River [Pliotsenovaia flora Kodora]. Georgian Academy of Sciences. Institute of Systematic Botany Monographs [Izd-vo Akademii nauk Gruzinskoi SSR, Monografii / Sukhumskii botanicheskii sad] 1: 1–209 [In Russian]. KOVAR-EDER, J. 1982. Eine Blätter-Flora des Egerien (Ober-Oligozän) aus marinen Sedimenten der Zentralen Paratethys im Linzer Raum (Österreich). Beiträge zur Paläontologie von Österreich 9: 1–209. KOVAR-EDER, J., Z. KVAČE K, AND M. STRÖBITZER-HERMANN. 2004. The Miocene flora of Parschlug (Styria, Austria) - revision and synthesis. Annalen des Naturhistorischen Museums in Wien. Serie A 105: 45–159. KVAČE K, Z., AND H. WALTHER. 1995. The Oligocene volcanic flora of Suletice-Berand near Ústí Nad Labem, North Bohemia—a review. Acta Musei Nationalis Pragae, Series B 50: 25–54. LAURENT, L. 1908. Flore Plaisancienne des argiles cinéritiques de Niac (Cantal) [avec une introduction géologique par Pierre Marty]. Annales de Musée d’histoire naturelle de Marseille 12: 1–88. LAURENT, L. 1912. Flore fossile des schistes de Menat. Annales de Musée d’histoire naturelle de Marseille 14: 1–246. LESQUEREUX, L. 1888. Recent determinations of fossil plants from Kentucky, Lousiana, Oregon, California, Alaska, Greenland, etc., with descriptions of new species. Proceedings of the United States National Museum 11: 11–38. MAI, D. H. 1995. Tertiäre Vegetationsgeschichte Europas. Gustav Fischer Verlag, Jena, Germany. MAI, H. D., AND H. WALTHER. 1978. Die Floren der Haselbacher Serie im Weisselster-Becken (Bezirk Leipzig, DDR). Abhandlungen des Staatlichen Museums für Mineralogie und Geologie zu Dresden 28: 1–200. MASSALONGO, A., AND G. SCARABELLI. 1859. Studii sulla flora fossile e geologia stratigrafica del Senigalliese. Ignatio Galeati e figlio, Imola, Italy. MEYER, H. W. 2003. The fossils of Florissant. Smithsonian Institution Books, Washington, District of Columbia, USA. PALAMAREV, E., V. BOZUKOV, K. UZUNOVA, A. PETKOVA, AND G. KITANOV. 2005. Catalogue of the Cenozoic plants of Bulgaria. Phytologica Balcanica 11: 215–364. POP, E. 1936. Flora pliocenică dela Borsec. Tipografia Natională S.A., Cluj, Romania. POUX, C., P. CHEVRET, D. HUCHON, W. W. DE JONG, AND E. J. P. DOUZERY. 2006. Arrival and diversification of caviomorph rodents and platyrrhine primates in South America. Systematic Biology 55: 228–244. QI, Z., K. M. CAMERON, P. LI, Y. ZHAO, S. CHEN, G. CHEN, AND C. FU. 2013. Phylogenetics, character evolution, and distribution patterns of the greenbriers, Smilacaceae (Liliales), a near-cosmopolitan family of monocots. Botanical Journal of the Linnean Society. Linnean Society of London 173: 535–548. RENNER, S. 2004. Plant dispersal across the tropical Atlantic by wind and sea currents. International Journal of Plant Sciences 165 (S4): S23–S33. RIDDLEY, H. N. 1930. The dispersal of plants throughout the world. L. Reeve & Co., Ashford, UK. [AQ13] [AQ14] [AQ15] [AQ16] GALLEY PROOF — AJBD1400495 16 • V O L . 1 0 2 , N O. 3 M A R C H 2 0 1 5 • A M E R I C A N J O U R N A L O F B O TA N Y RÖGL, F. 1998. Palaeogeographic considerations for Mediterranean and Paratethys seaways (Oligocene to Miocene). Annalen des Naturhistorischen [AQ17] Museums in Wien. Serie B, Botanik und Zoologie 99: 279–310. RÖPERT, D. 2000 onward (continuously updated). Digital specimen images at the Herbarium Berolinense. Website http://ww2.bgbm.org/ herbarium/ (Barcode: B 10 0247561 / ImageId: 254084) [accessed 9 October 2013]. ROWE, D. L., K. A. DUNN, R. M. ADKINS, AND R. L. HONEYCUTT. 2010. Molecular clocks keep dispersal hypotheses afloat: Evidence for transAtlantic rafting by rodents. Journal of Biogeography 37: 305–324. ROYAL BOTANIC GARDEN EDINBURGH. 2014. Herbarium Catalogue. Website http://elmer.rbge.org.uk/bgbase/vherb/bgbasevherb.php [accessed 24 September 2014] ROYAL BOTANIC GARDENS, KEW. 2014. Kew Herbarium Catalogue. Website http://www.kew.org/herbcat [accessed 24 September2014]. SAPORTA, G. 1862. Études sur la végétation du sud-est de la France à l'époque Tertiaire. Première partie, III-IV. Annales des Sciences Naturelles Botanique. Sér. 4 17: 191–. SAPORTA, G. 1863. Études sur la végétation du sud-est de la France à l'époque Tertiaire. Première partie, V. Annales des Sciences Naturelles Botanique. Sér. 4 19: 159–178. SAPORTA, G. 1865a. Études sur la végétation du sud-est de la France à l'époque Tertiaire. Deuxième partie, I-II. Annales des Sciences Naturelles Botanique. Sér. 5 3: 5–152. SAPORTA, G. 1865b. Études sur la végétation du sud-est de la France à l'époque Tertiaire. Deuxième partie, III. Annales des Sciences Naturelles Botanique. Sér. 5 4: 5–264. SAPORTA, G. 1873. Études sur la végétation du sud-est de la France à l'époque Tertiaire. Supplement I, 2. Annales des Sciences Naturelles Botanique. Sér. 5 17: 5–44. SAPORTA, G. 1888. Dernières adjonctions a la flore fossile d’Aix -enProvence. Annales des Sciences Naturelles Botanique, Sér. 7 7: 5–104. SMILEY, C. J., J. GRAY, AND L. M. HUGGINS. 1975. Preservation of Miocene fossils in unoxidized lake deposits, Clarkia, Idaho. Journal of Paleontology 49: 833–844. SUN, Z., AND D. DILCHER. 1988. Fossil Smilax from Eocene sediments in western Tennessee. American Journal of Botany 75 (6, part 2 [Program with abstracts of papers presented at the annual meetings of the Botanical Society of America with other affiliated societies at the University of California, Davis, Davis, California, USA,14-18 August 1988]): 118. TANAI, T., AND N. SUZUKI. 1963. Miocene floras of southwestern Hokkaido, Japan. In R. W. Chaney [ed.], Tertiary floras of Japan, Miocene floras, 9-149. The Collaborating Association to commemorate the 80th anniversary of the geological survey of Japan. Geological Survey of Japan, Tokyo, Japan. TANKUT, A., W. WILSON, AND T. YIHUNIE. 1998. Geochemistry and tectonic setting of Tertiary volcanism in the Güvem area, Anatolia, Turkey. Journal of Volcanology and Geothermal Research 85: 285–301. THORNE, R. F. 1973. Floristic relationships between tropical Africa and tropical America. In B. J. Meggers and W. D. Duckworth [eds.], Tropical forest ecosystems in Africa and South America: A comparative review, 27-47. Smithsonian Institution, Washington, District of Columbia, USA. UEMURA, K. 1988. Late Miocene floras in northeast Honshu, Japan. National Science Museum, Tokyo, Japan. UNGER, F. 1847. Chloris protogæa. Beiträge zur Flora der Vorwelt. Engelmann, Leipzig, Germany. UNGER, F. 1850a. Genera et species plantarum fossilium. Wilhelm Braunmüller, Vienna, Austria. UNGER, F. 1850b. Die fossile Flora von Sotzka. Denkschriften der mathematisch-naturwissenschaftlichen Classe der kaiserlichen Akademie der Wissenschaften 2: 127–197. UNGER, F. 1864. Sylloge Plantarum Fossilium. Pugillus Secundus. Sammlung fossiler Pflanzen besonders aus der Tertiär-Formation. Karl Gerold’s Sohn, Vienna, Austria. UNGER, F. 1867. Die fossile Flora von Kumi auf der Insel Euboea. Karl Gerold’s Sohn, Vienna, Austria. UNGER, F. 1869. Die fossile Flora von Szántó in Ungarn. Denkschriften der mathematisch-naturwissenschaftlichen Classe der kaiserlichen Akademie der Wissenschaften 30: 1–20. VELITZELOS, D. (Ed.). 2002. Field guide to the Neogene of the island of Evia—Early Miocene flora of Kymi. Field guide, 6th European palaeobotany-palynology conference, University of Athens, Greece. VELITZELOS, D., J. M. BOUCHAL, AND T. DENK. 2014. Review of the Cenozic floras and vegetation of Greece. Review of Palaeobotany and Palynology 204: 56–117. VINNERSTEN, A., AND K. BREMER. 2001. Age and biogeography of major clades in Liliales. American Journal of Botany 88: 1695–1703. VOELKER, G., S. ROHWER, D. C. OUTLAW, AND R. C. K. BOWIE. 2009. Repeated trans-Atlantic dispersal catalyzed a global songbird radiation. Global Ecology and Biogeography 18: 41–49. WALTHER, H., AND Z. KVAČE K. 2007. Early Oligocene flora of Seifhennersdorf (Saxony). Acta Musei Nationalis Pragae, Series B 63: 85–174. WESSEL, P., AND O. WEBER. 1856. Neuer Beitrag zur Tertiärflora der niederrheinischen Braunkohleformationen. Palaeontographica 4: 111–168. WEYLAND, H. 1937. Beiträge zur Kenntnis der rheinischen Tertiärflora. II. Erste Ergänzungen und Berichtigungen zur Flora der Blätterkohle und des Polierschiefers von Rott im Siebengebirge. Palaeontographica B 83: 67–122. WEYLAND, H. 1957. Kritische Untersuchungen zur Kutikularanalyse tertiärer Blätter III. Monocotylen der rheinischen Braunkohle. Palaeontographica B 103: 34–74. WILDE, V. 1989. Untersuchungen zur Systematik der Blattreste aus dem Mitteleozän der Grube Messel bei Darmstadt (Hessen, Bundesrepublik Deutschland). Courier Forschungsinstitut Senckenberg 115: 1–213. WILSON, M., A. TANKUT, AND N. GULEÇ. 1997. Tertiary volcanism of the Galatia Province, NW Central Anatolia, Turkey. Lithos 42: 105–121. YAVUZ-IȘIK, N. 2008. Palaeovegetational and palaeoclimatic investigations in the Early Miocene lacustrine deposits of the Güvem Basin (Galatean Volcanic Province), NW Central Anatolia, Turkey. Review of Palaeobotany and Palynology 150: 130–139. ZHAO, Y. P., Z. C. QI, W. W. MA, Q. Y. DAI, P. LI, K. M. CAMERON, J. K. LEE, ET AL. 2013. Comparative phylogeography of the Smilax hispida group (Smilacaceae) in eastern Asia and North America—implications for allopatric speciation, causes of diversity disparity, and origins of temperate elements in Mexico. Molecular Phylogenetics and Evolution 68: 300–311. GALLEY PROOF — AJBD1400495Q Author: Read proofs carefully. This is your ONLY opportunity to make changes. NO further alterations will be allowed after this point. Author Queries [AQ1]: Confirm that this is the case for determining the age. [AQ2]: The in-text citation "Morita (1931)" is not in the reference list. Please correct the citation, add the reference to the list, or delete the citation. [AQ3]: The in-text citation "Morita, 1933" is not in the reference list. Please correct the citation, add the reference to the list, or delete the citation. [AQ4]: The in-text citation "Poux et al., 2008" is not in the reference list. Please correct the citation, add the reference to the list, or delete the citation. [AQ5]: The in-text citation "Jønsson et al., 2012" is not in the reference list. Please correct the citation, add the reference to the list, or delete the citation. [AQ6]: Reference "Berger, Zabusch, 1953" is not cited in the text. Please add an in-text citation or delete the reference. [AQ7]: There is no 2006a. [AQ8]: Reference "Chen, Zhang, Ni, Fu, Cameron, 2006b" is not cited in the text. Please add an in-text citation or delete the reference. [AQ9]: Reference "Condit, 1938" is not cited in the text. Please add an in-text citation or delete the reference. [AQ10]: Reference appears to have more than 100 pages. Please verify page numbers (in reference "Dilcher, 1974"). [AQ11]: Please confirm the date and pagination of this cited article. Other sources give the date as 1985, or pagination as 44-46: 79-175. GALLEY PROOF — AJBD1400495Q [AQ12]: Reference appears to have more than 100 pages. Please verify page numbers (in reference "Hably, 1983"). [AQ13]: Reference "Kaya, Ünay, Göktaş, Saraç, 2007" is not cited in the text. Please add an in-text citation or delete the reference. [AQ14]: Reference "Laurent, 1912" is not cited in the text. Please add an in-text citation or delete the reference. [AQ15]: Reference "Palamarev, Bozukov, Uzunova, Petkova, Kitanov, 2005" is not cited in the text. Please add an in-text citation or delete the reference. [AQ16]: Reference "Poux, Chevret, Huchon, de Jong, Douzery, 2006" is not cited in the text. Please add an in-text citation or delete the reference. [AQ17]: Reference "Rögl, 1998" is not cited in the text. Please add an in-text citation or delete the reference. [AQ18]: The in-text citation "Lesquereux, 1878" is not in the reference list. Please correct the citation, add the reference to the list, or delete the citation. [AQ19]: The in-text citation "Knobloch et al., 1996" is not in the reference list. Please correct the citation, add the reference to the list, or delete the citation. [AQ20]: The in-text citation "Knowlton, 1899" is not in the reference list. Please correct the citation, add the reference to the list, or delete the citation. [AQ21]: The in-text citation "Knobloch and Kvaček, 1996" is not in the reference list. Please correct the citation, add the reference to the list, or delete the citation. [AQ22]: The in-text citation "Andreanszky, 1959" is not in the reference list. Please correct the citation, add the reference to the list, or delete the citation. [AQ23]: Massalongo and Scarabelli, 1859?

RELATED PAPERS

RELATED TOPICS

Log In

Smilax (Smilacaceae) From The Miocene Of Western Eurasia With Caribbean Biogeographic Affinities

Smilax (Smilacaceae) From The Miocene Of Western Eurasia With Caribbean Biogeographic Affinities

Related Papers

RELATED PAPERS

RELATED TOPICS