Journal of Biogeography (J. Biogeogr.) (2015) ORIGINAL ARTICLE Molecular and fossil evidence disentangle the biogeographical history of Podocarpus, a key genus in plant geography Mar
ıa Paula Quiroga1,2,*, Paula Mathiasen1,2, Ari Iglesias2, Robert R. Mill3 and Andrea C. Premoli1,2 1 Laboratorio Ecotono, Centro Regional Universitario Bariloche, Universidad Nacional del Comahue, Quintral 1250, Bariloche, Rıo Negro 8400, Argentina, 2Instituto de Investigaciones en Biodiversidad y Medioambiente, CONICET, Quintral 1250, Bariloche, Rıo Negro 8400, Argentina, 3Royal Botanic Garden Edinburgh, 20A Inverleith Row, Edinburgh EH3 5LR, UK ABSTRACT Aim The genus Podocarpus (Podocarpaceae) provides an opportunity to contrast biogeographical hypotheses within and among continents, and to analyse divergence between disjunct tropical and temperate forests of South America. We developed a calibrated phylogeny of Podocarpus to reconstruct the ancestral areas and potential expansion routes within Podocarpaceae. Location Podocarpus consists of two extant subgenera: Foliolatus from Asia and Oceania, and Podocarpus located in Gondwanan continents and north to the Caribbean. The paper focuses mainly on the area occupied by the latter subgenus. Methods We combined previously published and novel DNA sequences with fossil records. New species sequenced are members of Podocarpus subgenus Podocarpus from South and Central America. We assembled DNA sequences of the chloroplast (matK and rbcL) and nuclear (ITS1 and ITS2) to analyse phylogenetic relationships within Podocarpus subgenus Podocarpus by Bayesian methods, which were calibrated using macrofossils that could be confidently identified as modern genera. Ancestral areas were inferred using the dispersal– extinction–cladogenesis model. Results The phylogenetic reconstruction inferred a minimum age for the origin of Podocarpus s.l. in the late Cretaceous–early Palaeogene (63 Ma) and strongly supported monophyly of the genus Podocarpus and of subgenera Podocarpus and Foliolatus. Subgenus Podocarpus consists of two monophyletic, latitudinally structured clades. One clade consists of temperate American species while the other includes species from tropical-subtropical Africa and South America. Main conclusions The history of the subgenera within Podocarpus is older than previously reported: they can be traced back to late Cretaceous–early Palaeocene biogeographical connections between Australasia and South America through Antarctica. Latitudinally disjunct lineages within South America most probably diverged from widespread ancestors as a result of a persistent arid barrier that was established prior to the late Palaeogene. The calibrated age for the Tropical–Subtropical clade suggests an Atlantic–subtropical biogeographical corridor between South America and Africa long after the breakup of Gondwana and the stabilization of the circum-Antarctic current. *Correspondence: Mar
ıa Paula Quiroga, Laboratorio Ecotono, Centro Regional Universitario Bariloche, Universidad Nacional del Comahue, Quintral 1250, Bariloche, R
ıo Negro 8400, Argentina. E-mail: emepequ@gmail.com ª 2015 John Wiley & Sons Ltd Keywords Bayesian inference, Caribbean, Gondwana, historical biogeography, molecular dating, phylogeography, Podocarpaceae, Podocarpus, South America, vicariance http://wileyonlinelibrary.com/journal/jbi doi:10.1111/jbi.12630 1 M. P. Quiroga et al. INTRODUCTION Taxa within families of southern origin, such as Podocarpaceae (podocarps) and Nothofagaceae (southern beech), are considered key sources of information in plant geography (Couper, 1960; van Steenis, 1972). Couper (1960: p. 492) stated: ‘A study of the past distribution of Podocarpaceae and Nothofagus should then provide fundamental data for phytogeographical and palaeogeographical theories’ and ‘. . .specific relationship (in the strict sense) is not inferred between fossil and recent plants, but simply that in our present state of knowledge the fossil forms are the most likely ancestral forms of the recent species’. The biogeographical relevance of Podocarpaceae relies on the fact that it is the second largest conifer family (Farjon, 2010). In contrast to taxa within Nothofagaceae that are restricted to temperate areas, widespread genera within Podocarpaceae also reach subtropical and tropical areas. This is the case for Podocarpus, the most species-rich (c. 100 species) and widely distributed genus, which is found in Southeast Asia and almost all continents of the Southern Hemisphere that originated from Gondwana. Phylogenetic and phylogeographical studies of such a wide-ranging genus may prove valuable in addressing as yet largely unresolved biogeographical questions of long-distance dispersal and –vicariance within and among the southern continents. The family Podocarpaceae was very diverse and broadly distributed during the Mesozoic and Cenozoic. However, few fossil records can be assigned to modern genera. The earliest accepted podocarp macrofossils are from the Triassic of Gondwana (Townrow, 1967; Morley, 2011), yet recent molecular dating studies and fossil data suggest a Jurassic origin of Podocarpaceae (Biffin et al., 2011; Leslie et al., 2012; Rothwell et al., 2012; Escapa et al., 2013). In addition, the divergence of the two Podocarpus subgenera from their common ancestor has been suggested to have occurred in the Palaeogene (Biffin et al., 2011; Leslie et al., 2012). A fair correspondence was shown between pollen types and major genera within Podocarpaceae (Morley, 2011), whereas those pieces of evidence were restricted to Old World Neogene podocarps and thus did not include all genera. Also, the lack of diagnostic features of Podocarpus pollen precludes identification of distinct species or subgenera (Pocknall, 1981; Hooghiemstra et al., 2006; Morley, 2011). Furthermore, in the absence of specific synapomorphic characters, reproductive structures or whole-plant reconstructions, the phylogenetic placement of fossil conifers stay unresolved (e.g. Hill & Brodribb, 1999; Wilf et al., 2009; Escapa et al., 2010, 2013; Wilf, 2012; Escapa & Catalano, 2013). Besides preservation biases and the lack of organic connection that fossils usually have, another problem that arises in fossil calibration relates to the precise chronology of the data source (Wilf & Escapa, 2015). Most historical fossil records are poorly time constrained, and the range of dates can be so wide that a defined age remains uncertain. Such records are therefore not useful for phylogenetic node calibration 2 (Sauquet et al., 2012). Therefore, hypotheses about the historical biogeography of Podocarpus based only on fossils continue to be extremely uncertain. A morphological description of reproductive and vegetative organs of extant and fossil Podocarpaceae was given by Florin (1940a). The most recent complete revision of Podocarpus is found in a long series of papers by Buchholz and Gray, the most pertinent to our work being Buchholz & Gray (1948) and Gray (1953, 1955, 1956, 1958). Later, de Laubenfels (1985) revised the genus, listing 95 species divided between two subgenera Foliolatus and Podocarpus that were based on characters of the female cone, foliar anatomy and cuticle; 18 sections were also recognized (nine per subgenus). A new revision of Podocarpus is in progress (Mill, 2014, 2015a,b) but meantime de Laubenfels (1985) and Farjon (2010) are the standard references. Earlier phylogenetic analyses of the Podocarpaceae (Chaw et al., 1995; Kelch, 1997, 1998; Conran et al., 2000) did not include all genera until Sinclair et al. (2002). In addition, more recent phylogeographical studies have shown that chloroplast regions, previously used in phylogenetic studies of Podocarpaceae, are polymorphic within species, such as trnL–trnF for Podocarpus matudae (Ornelas et al., 2010) and P. parlatorei (Quiroga et al., 2012). Omission of such intraspecific variation may bias phylogenetic reconstructions if not considered. While subgenera Podocarpus and Foliolatus have been confirmed as monophyletic by molecular analyses (Conran et al., 2000; Knopf et al., 2012; Leslie et al., 2012), most of the sections recognized by de Laubenfels (1985) have not been supported by any phylogenetic reconstructions (Stark, 2004; Knopf et al., 2012; Leslie et al., 2012). Therefore, internal relationships of clades within subgenera remain to be explained. A detailed analysis within subgenus Podocarpus may help to elucidate relevant biogeographical questions such as transcontinental disjunctions between temperate South America and Australia–New Zealand, as well as tropical Africa and South America-Central America. Phylogenetic reconstructions may elucidate alternative hypotheses that Podocarpus s.l. originated in the Palaeogene (Biffin et al., 2011; Leslie et al., 2012) and diversified to reach its present wide distribution as a consequence of long-distance dispersal, or that it consists of ancient (i.e. Cretaceous) widespread lineages that evolved within Gondwanan continents by vicariance (Warren & Hawkins, 2006). While the latter hypothesis, of a Gondwanan relict distribution, was used by Verboom et al. (2014) to explain the presence of Australasian lineages in the Greater Cape flora of South Africa, they said that it was much more difficult to explain the South American connections with that flora because of uncertainties over identifying sister lineages as well as the paucity of divergence time estimates. In addition, no biogeographical inferences have yet been made concerning the diversity of Podocarpus in the Neotropics, where the genus is the principal conifer element of montane forests. It has been hypothesized that Andean uplift stimulated the entrance of Austral–Antarctic elements (such as Podocarpus) into the temperate forests of the Journal of Biogeography ª 2015 John Wiley & Sons Ltd Podocarpus key genus in plant geography Neotropics, and that new Neotropical taxa, adapted to the montane conditions caused by the Andean upheaval, were in turn derived from these (Van der Hammen & Hooghiemstra, 2000). We here present a fossil-calibrated molecular dating analysis, including new DNA sequences and macrofossil data of South American Podocarpus species that were not included in previous studies, in order to: (1) elucidate phylogenetic relationships among South American Podocarpus species using nuclear internal transcribed spacers 1 and 2 (ITS1 and ITS2) and conserved chloroplast DNA regions (rbcL and matK); (2) estimate the divergence times of clades within subgenus Podocarpus, in order to detect whether naturally disjunct Neotropical species are the product of recent diversification from ancestors of austral origin or are relicts from widespread ancient lineages; and (3) reconstruct ancestral areas of Podocarpus lineages. MATERIALS AND METHODS Podocarpus species and their geographical distribution Podocarpus is one of the largest extant conifer genera (Mill, 2014). We assembled a data set comprising the molecular sequences of species representing a near-complete sample of the world-wide distribution of the genus. To avoid possible synonymy, species names follow the Missouri Botanical Garden database (http://www.tropicos.org) and The Plant List (http://www.theplantlist.org). We used the Global Biodiversity Information Facility (http://www.gbif.org) as the most complete list of Podocarpus species and distribution range of Podocarpus (see Appendix S1 in Supporting Information). Molecular data Molecular data consisted of 108 sequences downloaded from GenBank and novel sequences of 15 American species (see Appendix S1 for accession numbers). Two of these 15 species (P. glomeratus and P. ingensis) have never been previously sequenced for any gene, while we provide sequences for additional genes for five species (P. oleifolius, P. parlatorei, P. purdieanus, P. sellowii and P. trinitensis) that were sequenced for other genes by Knopf et al. (2012) and/or Leslie et al. (2012). The remaining seven species (P. angustifolius, P. coriaceus, P. hispaniolensis, P. lambertii, P. matudae, P. nubigenus, P. salignus) were also studied by Biffin et al. (2011) and/or Leslie et al. (2012) but our sequences are newly generated. The polymerase chain reaction conditions for these fifteen species are described in Stark (2004) and Quiroga et al. (2012). In an effort to avoid missing taxa we requested samples from various herbaria; however, the resulting DNA products were too degraded for further analyses. The complete data set contained 71% of the total number of Podocarpus species. For the ingroup, we assembled four data sets comprising DNA sequences of Podocarpus species for each of two chloroplast [54 and 73 Journal of Biogeography ª 2015 John Wiley & Sons Ltd operational taxonomic units (OTUs) for matK and rbcL respectively] and two nuclear (36 and 42 OTUs for ITS1 and ITS2 respectively) regions separately. These data sets were also combined as follows: for each type of marker separately consisting of 78 and 53 OTUs for chloroplast and nuclear regions, respectively, and all markers combined (49 OTUs) (see Table S2 in Appendix S2). The amount of missing data for the four-marker combined data set was: matK = 9%, rbcL = 9%, ITS1 = 40% and ITS2 = 19%. The outgroup consisted of at least one species of each of the other 18 genera of Podocarpaceae, plus Araucaria araucana (Araucariaceae). We used Bayesian inference to estimate phylogenetic relationships for each marker and for the combined data set using mr bayes 3.1.2 (Ronquist & Hulsenbeck, 2003). The parameters used to run the analyses are described in Appendix S2. We found that the variation in the numbers of missing data among markers did not affect branch length, and the topology of the phylogenies was independent of the numbers of taxa or DNA regions used (see Appendix S3). Fossil calibration and molecular dating The phylogeny of Podocarpus was calibrated based on wellknown, precisely dated macrofossils that could be confidently assigned to modern genera (Table 1, Fig. 1). Macrofossils were used according to their earliest appearance in the fossil record; this does not mean that the age of the fossil corresponds to a point of divergence but rather indicates the confirmed presence of a taxon at a given point in geological time. We revised recent literature on fossil podocarps (Hill & Brodribb, 1999; Wilf, 2012) and used several new fossil age data, which come from isotope geochronological dating or that can be well correlated with a specific geological time period (Table 1, Fig. 1). Fossil calibrations at generic nodes (Table 1) are described in Appendix S2. Few fossil records can be certainly assigned to Podocarpus, the great majority of them restricted to the Palaeogene (66– 23 Ma). To date, the oldest reliable fossil corresponds to P. andiniformis from the R
ıo Pichil
eufu flora of Patagonia (USNM40384: Fig. 1e–g; Berry, 1938; Florin, 1940a). As noted in Table 1, recent radiomagnetic dates give a well-constrained age of 47.74  0.18 Ma for the R
ıo Pichil
eufu fossiliferous strata (Wilf et al., 2005; Wilf, 2012). Podocarpus andiniformis was later found in the Laguna del Hunco flora (Wilf et al., 2005; Wilf, 2012), which is slightly older with a well-constrained age of 52.22  0.29 Ma (early Eocene; Wilf, 2012). This pushes back the oldest reliably dated fossil evidence for Podocarpus by 5 Myr. Species divergence times were estimated using the combined data set by two different Bayesian approaches as implemented in beast2 1.8.0 (Bouckaert et al., 2014). The settings to run beast2 and the calibration nodes used are described in Appendix S2. 3 M. P. Quiroga et al. Table 1 Fossil calibration points used for divergence time estimation in Ma. Node numbers correspond to those in Fig. 2. Node Fossil species Epoch Root Araucariales* Araucaria spp. Late Triassic 214  10* 1. Podocarpaceae* Rissikia media Triassic?-Jurassic 160  10* 2. Acmopyle Acmopyle florinii Acmopyle engelhardtii Dacrycarpus sp. Dacrydium sp. Retrophyllum sp. Nageia hainanensis Podocarpus andiniformis Late Palaeocene 57.5  1.5 3. 4. 6. 7. 8. Dacrycarpus Dacrydium Retrophyllum Nageia Podocarpus Early Palaeocene Middle Eocene Early Eocene Eocene Early Eocene Minimum age (Ma) 64.48  44  52.22  34–55 52.22  0.59 4.0 0.22 0.22 Localization Antarctica1,2 Australia3 Argentina, India, Australia, Antarctica3,4,5 Australia6 Argentina7,8,9 Salamanca Fm., Argentina10,11 Australia12 Laguna del Hunco, Argentina8 China13 Laguna del Hunco, Argentina8,14,15 *The Araucariales node was set with a maximum age estimation based on transitional conifers. For Podocarpaceae node we used minimal age estimation based on recent literature. We also ran analyses without using these values, obtaining similar results that do not substantially modify other node values. 1 Escapa et al. (2010); 2Escapa & Catalano (2013); 3Rothwell et al. (2012); 4Townrow (1967); 5Escapa et al. (2013); 6Hill & Carpenter (1991); 7 Florin (1940b); 8Wilf (2012); 9Fig. 1d; 10Iglesias (2007); 11Fig. 1a–c; 12Carpenter & Pole (1995); 13Jin et al. (2010); 14Florin (1940a); 15Fig. 1e–g. Figure 1 Images of precisely dated fossil materials (Table 1) used for phylogenetic calibrations and confidently identified as modern genera: (a) Dacrycarpus sp. from Salamanca Fm. (early Danian) from Iglesias (2007). White arrow indicates leaf magnified in Fig. 1c. (b) Counterpart of material in Fig. 1a. (c) Closer view of a leaf from Fig. 1a, note typical Dacrycarpus leaf characters: fine-needled bilaterally flattened leaves; acuminate apex with a long fine mucro; and typical linear pair of stomata bands (white arrow) deployed at equal distance from the mid-vein (Wilf, 2012). Scale = 1 mm. (d) Acmopyle engelhardtii (Berry) Florin, type specimen (USNM 40385b); note distinctive distichous bilaterally flattened leaves, larger and wider leaf shape (Wilf, 2012). (e) Podocarpus andiniformis Berry, type specimen (USNM 40384) from the Eocene Pichileuf
u flora, Patagonia. (f–g): closer view of leaves in (e), note the large adpressed leaf region, wider base shape, straight apex shape, and strong middle vein that characterize the leaves in the genus (Florin, 1940a). All scales represent 1 cm (except in c). 4 Journal of Biogeography ª 2015 John Wiley & Sons Ltd Podocarpus key genus in plant geography Ancestral area reconstruction Six broad geographical areas were delimited based on the current distribution range of Podocarpus. These areas were: Austral (southern South America, Antarctica, Australia); Oceania (Malaysia, Micronesia, New Zealand, New Caledonia, Papuasia); Asia (China, Indonesia, Japan, Malaysia, Philippines, Taiwan); Tropical America (Central America and Caribbean); Subtropical America (South America Neotropical); and Africa (see Appendix S2). To reconstruct ancestral areas of Podocarpus we used the dispersal–extinction–cladogenesis (DEC) model implemented in lagrange 20130526 (Ree & Smith, 2008), modifying the adjacency matrix, the number of time slices, and the dispersal probabilities according to geological epoch since the late Cretaceous and the corresponding climatic and geological settings (see Appendix S2). RESULTS Phylogenetic analysis Nuclear and chloroplast DNA yielded two monophyletic clades with maximum branch support that matched the distribution of extant Podocarpus subgenera. The Foliolatus clade included only species from Asia, Oceania and Australasia while the Podocarpus subgenus clade contained species from South and Central America, Africa, New Zealand, New Caledonia and eastern Australia (Fig. 2). The results of molecular dating suggest that both subgenera may have diverged in the early Eocene or even before (82–52 Ma; Table 2). Within subgenus Podocarpus, two latitudinally structured, well-supported major clades can be distinguished, Austral and Tropical-Subtropical (Fig. 2), that may have diverged in the Palaeogene (42 Ma; Table 2). The Austral clade consists Table 2 Bayesian relaxed molecular clock age estimates in million years (Ma) for different species of Podocarpus phylogeny, using two Bayesian approaches in beast: birth/death and Yule. 95% highest posterior density (HPD) values are given in parentheses, *fossil calibrated. Journal of Biogeography ª 2015 John Wiley & Sons Ltd of species from temperate latitudes of southern South America, New Zealand, New Caledonia and Australia. The Tropical-Subtropical clade is composed of species from tropical and subtropical America and Africa. The Tropical-Subtropical clade in turn includes three well-supported subclades: the subtropical South American subclade, which is sister to the subclade of subtropical African species, and a third monophyletic group that includes all species with tropical distributions in Central America, Caribbean and South America (north of Amazonia, as well as the tropical and subtropical Andes) (Fig. 2). Biogeography and divergence time The minimum divergence time of Dacrycarpus and Dacrydium (Table 2, Fig. 2) was 91 and 54 Ma (late Cretaceous) respectively. The estimated minimum divergence times obtained for the sister group of Podocarpus (Retrophyllum, Nageia and Afrocarpus) were also in the late Cretaceous (85 Ma) (Table 2). The estimated minimum age for Podocarpus s.l. was 63 Ma (Table 2), and 52.22  0.29 Ma based on the evidence of Patagonian fossils (Tables 1 & 2). Thus, Podocarpus may have already been differentiated from the rest of Podocarpaceae by the latest Cretaceous (Fig. 2). Subgenera within Podocarpus may have diverged prior to the late Palaeogene, although the range of minimum age estimations for the Foliolatus node indicate that this is much younger (23 Ma) than subgenus Podocarpus (42 Ma) (Table 2). Based on these results and modern distribution, the ancestral area of subgenus Foliolatus was restricted to East Gondwana, while subgenus Podocarpus was widely distributed in West Gondwana, including the tropical climatic belt and Antarctica (DEC model P = 0.385; Fig. 2). Two nearly synchronic diversifications of subgenus Podocarpus can be recognized: (1) the C1 Austral clade (25 Ma) within the cool climatic belt in Gondwana, including temperate species distributed in Node Clade Birth/death 95% HPD (Ma) Yule 95% HPD (Ma) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. Podocarpaceae* Acmopyle* Dacrycarpus* Dacrydium* Sister group to Podocarpus Retrophyllum* Nageia* Podocarpus* P. subgenus Foliolatus P. subgenus Podocarpus Austral clade Tropical–Subtropical clade SA–African subclades SA Subtropical subclade African subclade C-SA Tropical subclade 230.25 141.84 90.73 52.63 82.62 64.94 48.78 60.01 20.31 38.78 22.27 30.51 28.97 17.88 8.42 5.02 (199.15, 258.15) (114.80, 168.18) (69.29, 113.41) (44.00, 66.93) (67.99, 98.48) (53.00, 77.88) (44.5, 56.77) (52.22, 71.80) (13.18, 28.48) (27.97, 51.05) (13.96, 31.46) (21.27, 41.14) (18.59, 40.21) (11.91, 24.26) (3.78, 13.84) (1.60, 9.28) 226.89 141.68 91.62 54.49 87.46 68.34 50.58 65.21 26.24 46.12 27.82 37.00 35.11 22.57 10.93 6.58 (196.70, (114.13, (67.63, (44.00, (69.40, (53.40, (44.50, (52.22, (15.78, (32.54, (16.89, (24.71, (21.00, (14.49, (4.53, (1.76, 257.21) 170.37) 115.07) 71.67) 108.06) 84.98) 61.92) 81.83) 38.58) 62.46) 39.75) 50.28) 49.05) 32.07) 19.07) 12.92) 5 M. P. Quiroga et al. Figure 2 Chronogram indicating the evolutionary relationships among Podocarpaceae genera, and Podocarpus species by means of matK, rbcL and ITS1–2 consensus tree calculated with Yule estimation. The numbers above the branches indicate the highest posteriori probability values (95% highest posterior density) for Bayesian inference analysis. Bold numbers correspond to the nodes in Table 2. Asterisks in nodes indicate fossil calibration points. Ancestral areas inferred in Lagrange as indicated with squares as S, Austral SA; O, Oceania; A, Asia; T, Tropical; B, Subtropical SA; F, Africa. The geochronological scale is based on ICS (2014). southern land masses: southern South America, Australia, New Zealand, New Caledonia and (2) the C2 Tropical–Subtropical clade (34 Ma), comprising species distributed in tropical and subtropical latitudes of America and Africa 6 (Fig. 2). Clade C2 contained three subclades (Fig. 2): (1) subclade C2a (SA Subtropical: subtropical South America), which was sister to (2) subclade C2b, the African subclade (species from subtropical Africa) and (3) subclade C2c (CJournal of Biogeography ª 2015 John Wiley & Sons Ltd Podocarpus key genus in plant geography SA Tropical: species with tropical distributions in northern Amazonia, Antilles, Caribbean Islands and Central America, as well as species from tropical and subtropical Andes). Species in subclade C2c occupy the tropical and warm climatic belt in Central and South America (Fig. 3). Results section. Our fossil-calibrated relationships and ancestral area reconstruction suggest an Atlantic subtropical biogeographical corridor and also that latitudinally disjunct lineages within South America probably diverged from widespread ancestors resulting from the persistent arid barrier. DISCUSSION Fossil inference We aimed to evaluate biogeographical hypotheses regarding the intercontinental geographical distribution of the conifer genus Podocarpus (Podocarpaceae) in the Southern Hemisphere mainly using a molecular, fossil-calibrated phylogenetic hypothesis, and to reconstruct the ancestral areas and dispersal routes among continents using the DEC model. We found that the minimum age of Podocarpus ranged between 82 Ma and 52 Ma, 20 Myr older than previously reported (Biffin et al., 2011; Leslie et al., 2012). We obtained a strongly supported monophyly for Podocarpus divided into two main clades, Podocarpus and Foliolatus, which were widely distributed within Gondwana. The Podocarpus clade was further divided into two subclades (C1 Austral, C2 Tropical–Subtropical). Subclade C2 Tropical-Subtropical could itself be subdivided into three subclades (C2a SA Subtropical, C2b African, C2c C-SA Tropical), whose compositions and relationships to one another are defined in the Our results suggest that the origin of Podocarpus could be traced back to the late Cretaceous. The improvement in both the updated fossil ages and well-recognized fossil records provides older minimum ages for several nodes in the Podocarpaceae phylogeny. Because fossil records estimate minimum ages, they may push back the ages for nodes when adding older fossil calibrations. For instance, when using the Patagonian P. andiniformis macrofossil to calibrate modern South American species (i.e. the Podocarpus subgenus node), the Podocarpus node was extended 37–34 Myr back (data not shown, although performed as separate analysis). In Patagonia, the conifer fossil record is abundant through the Jurassic and Cretaceous periods (Archangelsky & Romero, 1974; Iglesias et al., 2011; Wilf et al., 2013). Currently, in temperate Austral and Neotropical forests of South America, conifers are reduced to three families: Cupressaceae (three monotypic genera); Araucariaceae (one genus and two (a) (b) Figure 3 Modern distribution of Podocarpus subgenus Podocarpus in South America and Africa superimposed to (a) Early Palaeogene (65–55 Ma) palaeoclimatic and palaeogeographical reconstruction based on Scotese (2010). Note the broad subtropical arid climatic belt at the region where the SA and African Subtropical subclades (C2a and C2b respectively) are distributed today. The solid black arrow indicates probable direction of movement for both the warm temperate climatic belt and the SA Subtropical subclade C2a. The white arrow indicates probable direction of movement for both the cool temperate climatic belt and the Austral clade C1. The black dashed arrow indicates probable direction of dispersion for the African subclade C2b within the warm temperate climatic belt; (b) Modern geography and climate. Note the possible subtropical connection between the SA and African Subtropical subclades (C2a and C2b respectively) that may have occurred after climatic belt migration during the Oligocene (black dashed arrow). Journal of Biogeography ª 2015 John Wiley & Sons Ltd 7 M. P. Quiroga et al. species); and Podocarpaceae (five genera). However, during the Mesozoic and early Palaeogene, a significant number of modern conifer taxa occupied Patagonia that now are extinct in South America, including Acmopyle, Agathis, Dacrycarpus and Papuacedrus (Wilf et al., 2009, 2014; Wilf, 2012). Our estimated minimum age for divergence indicates that the ancestor of Podocarpus may have diverged from its sister group (Retrophyllum, Nageia and Afrocarpus) earlier than the latest Cretaceous. Re-calculated ages based on fossil-calibrated Bayesian molecular dating gives a minimum age of 85 Ma for the sister group of Podocarpus (Table 2). Afterwards, in the Palaeogene, Podocarpus (63 Ma) had already diverged into the two geographically structured subgenera: Foliolatus (23 Ma) and Podocarpus (42 Ma). Therefore, the ancestral lineage from which both subgenera are presumed to have originated may have survived the Cretaceous–Palaeogene global extinction event. Ecological tolerance Although the widespread subgenus Podocarpus has been reported as having originated in the Palaeogene–Neogene transition (Biffin et al., 2011), our results show that the ancestors of the Austral and Tropical–Subtropical clades may have differentiated earlier, during the warm Palaeogene in western Gondwana. According to the current biogeography and temperature tolerances, Podocarpus species present a highly conservative niche. Today, the Austral clade mainly occurs south of latitude 20° S in South America, Australia and New Caledonia. The only exception is P. smithii (Australia), which is distributed in montane rain forest between 15° S and 17° S. Following Biffin et al. (2012), the species of the Austral clade may occupy environments with high rainfall and average minimum temperatures below 0 °C of the coldest month. The ancestors of the Tropical–Subtropical clade probably remained confined to warmer climates in tropical South America, while cold-tolerant lineages of the Austral clade adapted to novel cooling trends that developed after the Eocene–Oligocene boundary (Zachos et al., 2001; Bowen & Zachos, 2010). This biogeographical pattern is similar to that suggested for cold-hardy lineages of the Nothofagaceae that also evolved once cold climates began in southern South America (Premoli et al., 2012). The establishment of cool trends at high latitudes was synchronic with the development of the Antarctic Circumpolar Current when Antarctica separated from South America and Australia and with the onset of glaciation Antarctica during the late Palaeogene (Zachos et al., 2001; Iglesias et al., 2011; Lawver et al., 2011). These events probably deepened clade differentiation within subgenus Podocarpus, although the overall diversification of extant taxa within the Austral clade has been poor. Phylogeny and biogeography of subgenus Podocarpus The reciprocal monophyly of the Austral clade containing species of South America and Oceania, and the Tropical8 Subtropical clade including South American and African taxa, suggests that these latitudinally divergent lineages probably originated from a widespread common ancestor during the late Cretaceous. The ancestral area reconstruction, molecular and fossil analyses support the hypothesis of a Gondwanan wide-range distribution of ancestral clades during or previous to the Eocene (Fig. 2). Further data from Oceania are needed to test the alternative hypothesis that subgenus Podocarpus developed in South America and then migrated to Oceania through Antarctica, a connection that seems to have persisted up to the middle Eocene (Iglesias et al., 2011; Wilf et al., 2013). Our results challenge the hypothesis of Van der Hammen & Hooghiemstra (2001) who proposed that tropical and subtropical South American Podocarpus were derived as a result of northward migration by cold-tolerant ancestors of Austral-Antarctic origin (i.e. the Austral clade). The early divergence of the Austral and Tropical–Subtropical lineages reflect their independent evolution on either side of the arid belt (Fig. 3) as suggested by the ancestral area analysis. Therefore, the evidence presented here supports a vicariance hypothesis within South America that closely matches tectonic and climatic events such as the latest Cretaceous–earliest Palaeocene marine ingressions and the longlasting arid (dry and hot) belt at mid-latitudes of South America (Iglesias et al., 2011; Woodburne et al., 2014; Fig. 3). These probably acted as a persistent, effective barrier in South America that reinforced the south–north isolation of the Austral and Tropical-Subtropical clades (see below). Nonetheless, the fact that South American temperate Podocarpus taxa within the Austral clade are closely linked to those from Australia, New Zealand and New Caledonia strengthen the southern South America–Oceania connection, as suggested for widespread ancestral Nothofagus lineages (Hill, 1991, 2001). Climate affinity Our data support a long-lasting persistence of lineages of subgenus Podocarpus within the South American Plate. The ancestors of the SA Subtropical lineage were probably located south of its current distribution under the warm temperate climatic belt (Fig. 3a, black arrow). Similarly, the ancestral lineage of the Austral clade was probably located under the Palaeogene cool temperate climatic belt (Fig. 3a, white arrow), which may have been further south (e.g. Antarctica). In addition, the Tropical-Subtropical clade, with a minimum estimated divergence age prior to the Neogene, and probably earlier than the late Eocene (Table 2), shows that the African and SA Subtropical subclades fall within the same palaeolatitudinal belt, which in turn were isolated from those with tropical distributions (Fig. 3). We propose two alternative hypotheses for the relationship between the African and South American subtropical subclades: (1) an early migration of Podocarpus from South America to Africa within the southern warm temperate climatic belt (Fig. 3a, black dashed arrow) during the uppermost Cretaceous–early Palaeogene Journal of Biogeography ª 2015 John Wiley & Sons Ltd Podocarpus key genus in plant geography (minimum age in the Eocene–Oligocene transition; Table 2); or (2) a later migration of Podocarpus from South America to Africa once the subtropical climatic belt was established during the Oligocene, when the humid belt shifted towards lower latitudes (Fig. 3b, black dashed arrow). The subtropical subclades seem to have originated before the late Palaeogene, as evidenced by highly supported long branch lengths (Fig. 2). However, short branch lengths within the SA Subtropical and African subclades probably indicate more recent species radiations during the Neogene (Fig. 2), most likely at the middle Miocene climatic optimum (Zachos et al., 2001). Although members of the Podocarpaceae were probably present in ecologically suitable areas of eastern Africa (i.e. humid; Fig. 3a) since the late Cretaceous and early Palaeogene, they migrated into subtropical regions when the climate cooled in the Pliocene (Morley, 2011). Subtropical South American species of the TropicalSubtropical subclade constitute a well-supported subclade (C2a) that unites species from the Andes and the Atlantic coast. This subclade is sister to the African species (subclade C2b) and is clearly phylogenetically separated from South American species of northernmost latitudes (subclade C2c, C-SA Tropical). Dry seasonal climates of northern South America during the uppermost Cretaceous and Palaeocene (Dino et al., 1999; Hoorn et al., 2010) may have constituted an effective long-lasting barrier that isolated the SA Subtropical and C-SA Tropical subclades. Afterwards, during the Miocene, shallow marine ingressions over tropical lowlands in north-western Brazil, Colombia and Peru (e.g. Pebas System: Rossetti & Netto, 2006; Hovikoski et al., 2010; Hughes et al., 2013; Boonstra et al., 2015) may have also kept these subclades isolated. Meanwhile the arid diagonal was expanding geographically in South America, broadening in subtropical areas (Uba et al., 2005), thus maintaining the divergence between the Austral and Tropical-Subtropical clades. Divergence times of lineages within subgenus Foliolatus occurred at the Palaeogene–Neogene boundary (Table 2), shortly after the separation of South America and Australia from Antarctica (Zachos et al., 2001). Further molecular studies are required to better elucidate biogeographical and phylogenetic relationships within subgenus Foliolatus. CONCLUSION The phylogenetic reconstruction of Podocarpus presented here shows a strong geographical control. The use of new DNA sequences helps to resolve some inconsistencies that appeared in previous studies, particularly among South American and Caribbean species. Similarly, the inclusion of novel fossil data from Patagonia made an invaluable contribution to this work. If additional diagnostic subgeneric characters were considered in future analyses, this could lead to better resolution of the phylogeny. The estimated minimum age of Podocarpus s.l. gives origination times in the late Cretaceous–early Palaeogene (63 Ma), much older than that estimated previously. The new minimum age estimates also suggest that all lineages Journal of Biogeography ª 2015 John Wiley & Sons Ltd (subclades) within Podocarpus were already present by the Eocene. Thus, the biogeographical patterns of extant Podocarpus clades are the result of vicariance events related to palaeoclimatic changes and tectonic events that have affected the latitudinal distribution of such ancient lineages almost since the late Cretaceous–early Palaeogene. The recent diversification of African and northern South American species occurred as a result of dispersal events during the Neogene. Finally, our results suggest a biotic connection between South America and Africa, which may be related to palaeoclimatic and palaeo-biogeographical belts at subtropical latitudes. The evidence presented here, in addition to previous morphological, phylogenetic, and phylogeographical studies, highlight the need to revisit the taxonomy of Podocarpus. ACKNOWLEDGEMENTS We dedicate this paper to the memory of Darian Stark (1980–2011) who obtained the Caribbean sequences used here during his M.Sc. research supervised in 2004 by Robert Mill. We thank Edward Biffin who provided sequences of the ITS region of Podocarpus published in Biffin et al. (2011). We are very grateful to Peter Wilf, Raymond Carpenter and National Science Foundation grant DEB-0919071 for providing images of fossil-types. Research for this contribution was possible thanks to Argentinean grant: Agencia Nacional de Promoci
on Cient
ıfica y Tecnol
ogica (PICT2010-N430 to M.P.Q.). All authors except Robert Mill are members of the National Research Council of Argentina CONICET. The Royal Botanic Garden Edinburgh is supported by the Scottish Government’s Rural and Environment Science and Analytical Services Division. REFERENCES Archangelsky, S. & Romero, E.J. (1974) Polen de Gimnospermas (Con
ıferas) del Cret
acico Superior y Pale
ogeno de Patagonia. Ameghiniana, 11, 217–236. Berry, E.W. (1938) Tertiary flora from the R
ıo Pichileuf
u, Argentina. Geological Society of American Special Paper, 12, 1–149. Biffin, E., Conran, J.G. & Lowe, A.J. (2011) Podocarp evolution: a molecular phylogenetic perspective. Ecology of the Podocarpaceae in tropical forests (ed. by B.L. Turner and L.A. Cernusak), pp. 1–19. Smithsonian Institution Scholarly Press, Washington, DC. Biffin, E., Brodribb, T.J., Hill, R.S., Thomas, P. & Lowe, A.J. (2012) Leaf evolution in Southern Hemisphere conifers tracks the angiosperm ecological radiation. Proceedings of the Royal Society B: Biological Sciences, 279, 341– 348. Boonstra, M., Ramos, M.I.F., Lammertsma, E.I., Antoine, P.O. & Hoorn, C. (2015) Marine connections of Amazonia: evidence from foraminifera and dinoflagellate cysts (early to middle Miocene, Colombia/Peru). Palaeogeography, Palaeoecology, Palaeoclimatology, 417, 176–194. 9 M. P. Quiroga et al. Bouckaert, R., Heled, J., K€ uhnert, D., Vaughan, T.G., Wu, C.-H., Xie, D., Suchard, M.A., Rambaut, A. & Drummond, A.J. (2014) BEAST2: a software platform for Bayesian evolutionary analysis. PLoS Computational Biology, 10, e1003537. Bowen, J.G. & Zachos, J.C. (2010) Rapid carbon sequestration at the termination of the Palaeocene-Eocene Thermal Maximum. Nature Geoscience, 3, 866–869. Buchholz, J.T. & Gray, N.E. (1948) A taxonomic revision of Podocarpus IV. The American species of section Eupodocarpus, Subsections C and D. Journal of the Arnold Arboretum, 29, 123–151. Carpenter, R.J. & Pole, M. (1995) Eocene plant fossils from the Lefroy and Cowan Paleodrainages, Western Australia. Australian Systematic Botany, 8, 1107–1154. Chaw, S.M., Sung, H.M., Long, H., Zharkikh, A. & Li, W.H. (1995) The phylogenetic positions of the conifer genera Amentotaxus, Phyllocladus, and Nageia inferred from 18S rRNA sequences. Journal of Molecular Evolution, 41, 224– 230. Conran, J.G., Wood, G.M., Martin, P.G., Dowd, J.M., Quinn, C.J., Gadek, P.A. & Price, R.A. (2000) Generic relationships within and between the gymnosperm families Podocarpaceae and Phyllocladaceae based on an analysis of the chloroplast gene rbcL. Australian Journal of Botany, 48, 715–724. Couper, R.A. (1960) Southern Hemisphere Mesozoic and Tertiary Podocarpaceae and Fagaceae and their palaeogeographic significance. Proceedings of the Royal Society of London Series B, Biological Sciences, 152, 491–500. Dino, R., Silva, O.B. & Abrah~ao, D. (1999) Caracterizacß~ao palinol
ogica e estratigr
afica de estratos cret
aceos da Formacß~ao Alter do Ch~ao, Bacia do Amazonas. Simposio sobre o Cretaceo do Brasil, 5, Rio Claro, Atas, 557–565. Escapa, I.H. & Catalano, S.A. (2013) Phylogenetic analysis of Araucariaceae: integrating molecules, morphology, and fossils. International Journal of Plant Sciences, 174, 1153– 1170. Escapa, I.H., Decombeix, A.L., Taylor, E.L. & Taylor, T.N. (2010) Evolution and relationships of the conifer seed cone Telemachus: evidence from the Triassic of Antarctica. International Journal of Plant Sciences, 171, 560–573. Escapa, I.H., Rothwell, G. & C
uneo, N.R. (2013) Calibrating the ages of conifer clades. Botany 2013 (Botanical Society of America Annual Meeting, New Orleans, July 27–31, 2013), Abstract 364. Available at: http://2013.botanyconference.org/engine/search/index.php?func=detail&aid=364 Farjon, A. (2010) A handbook of the world’s conifers. E. J. Brill, Leiden. Florin, R. (1940a) The tertiary conifers of south Chile and their phytogeographical significance. Kungliga Svenska Vetenskapsakademiens Handlingar, 19, 1–107. Florin, R. (1940b) Die heutige und fr€ uhere Verbreitung der Koniferengattung Acmopyle Pilger. Svensk Botanisk Tidskrift, 34, 117–140. 10 Gray, N.E. (1953) A taxonomic revision of Podocarpus VIII. The African species of section Eupodocarpus, subsections A and E. Journal of the Arnold Arboretum, 34, 163–175. Gray, N.E. (1955) A taxonomic revision of Podocarpus IX. The South Pacific species of section Eupodocarpus, subsection F. Journal of the Arnold Arboretum, 36, 199–206. Gray, N.E. (1956) A taxonomic revision of Podocarpus X. The South Pacific species of Section Eupodocarpus, subsection D. Journal of the Arnold Arboretum, 37, 160–172. Gray, N.E. (1958) A taxonomic revision of Podocarpus XI. The South Pacific species of Section Podocarpus, subsection B. Journal of the Arnold Arboretum, 39, 424–477. Hill, R.S. (1991) Tertiary Nothofagus (Fagaceae) macrofossils from Tasmania and Antarctica and their bearing on the evolution of the genus. Botanical Journal of the Linnean Society, 105, 73–112. Hill, R.S. (2001) Biogeography, evolution and palaeoecology of Nothofagus (Nothofagaceae): the contribution of the fossil record. Australian Journal of Botany, 49, 321–332. Hill, R.S. & Brodribb, T.J. (1999) Southern conifers in time and space. Australian Journal of Botany, 47, 639–696. Hill, R.S. & Carpenter, R.J. (1991) Evolution of Acmopyle and Dacrycarpus (Podocarpaceae) foliage as inferred from macrofossils in south-eastern Australia. Australian Systematic Botany, 4, 449–479. Hooghiemstra, H., Wijninga, V.M. & Cleef, A.M. (2006) The paleobotanical record of Colombia: implications for biogeography and biodiversity. Annals of the Missouri Botanical Garden, 93, 297–325. Hoorn, C., Roddaz, M., Dino, R., Soares, E., Uba, C., Ochoa-Lozano, D. & Mapes, R. (2010) The Amazonian Craton and its influence on past fluvial systems (Mesozoic–Cenozoic, Amazonia). Amazonia: landscape and species evolution, a look into the past (ed. by C. Hoorn and F.P. Wesselingh), pp. 103–122. Blackwell Publishing, Oxford, UK. Hovikoski, J., Wesselingh, F.P., R€as€anen, M., Gingras, M. & Vonhof, H.B. (2010) Marine influence in Amazonia: evidence from the geological record. Amazonia: landscape and species evolution, a look into the past (ed. by C. Hoorn and F.P. Wesselingh), pp. 143–161. Blackwell Publishing, Oxford, UK. Hughes, C.E., Pennington, R.T. & Antonelli, A. (2013) Neotropical plant evolution: assembling the big picture. Botanical Journal of the Linnean Society, 171, 1–18. ICS (2014) International chronostratigraphic chart v2014/02. International Commission on Stratigraphy. Available at: http://www.stratigraphy.org/ICSchart/ChronostratChart2014-02.pdf. Iglesias, A. (2007) Estudio paleobotanico, paleoecologico y paleoambiental en secuencias de la Formacion Salamanca, Paleoceno Inferior en el sur de la provincia de Chubut, Patagonia, Argentina. PhD Thesis, Universidad de La Plata. Iglesias, A., Arbate, A.E. & Morel, E.M. (2011) The evolution of Patagonian climate and vegetation, from the Mesozoic Journal of Biogeography ª 2015 John Wiley & Sons Ltd Podocarpus key genus in plant geography to the present. Botanical Journal of the Linnean Society, 103, 409–422. Jin, J., Qiu, J., Zhu, Y. & Kodrul, T.M. (2010) First fossil record of the genus Nageia (Podocarpaceae) in south China and its phytogeographic implications. Plant Systematic and Evolution, 285, 159–163. Kelch, D.G. (1997) The phylogeny of the Podocarpaceae based on morphological evidence. Systematic Botany, 22, 113–131. Kelch, D.G. (1998) Phylogeny of Podocarpaceae: comparison of evidence from morphology and 18S rDNA. American Journal of Botany, 85, 986–996. Knopf, P., Schulz, C., Little, D.P., St€ utzel, T. & Stevenson, D.W. (2012) Relationships within Podocarpaceae based on DNA sequence, anatomical, morphological, and biogeographical data. Cladistics, 28, 271–299. de Laubenfels, D.J. (1985) A taxonomic revision of the genus Podocarpus. Blumea, 30, 251–278. Lawver, L.A., Gahagan, L.M. & Norton, I. (2011) Palaeogeographic and tectonic evolution of the Arctic region during the Palaeozoic. Geological Society of London, Memoir, 35, 61–77. Leslie, A.B., Beaulieu, J.M., Rai, H.S., Crane, P.R., Donoghue, M.J. & Mathews, S. (2012) Hemisphere-scale differences in conifer evolutionary dynamics. Proceedings of the National Academy of Sciences USA, 109, 16217–16221. Mill, R.R. (2014) A monographic revision of the genus Podocarpus (Podocarpaceae): I. Historical review. Edinburgh Journal of Botany, 71, 309–360. Mill, R.R. (2015a) A monographic revision of the genus Podocarpus (Podocarpaceae): II. The species of the Caribbean Bioregion. Edinburgh Journal of Botany, 72, 61–185. Mill, R.R. (2015b) A monographic revision of the genus Podocarpus (Podocarpaceae): III. The species of the Central America and Northern Mexico Bioregions. Edinburgh Journal of Botany, 72, 243–341. Morley, R.J. (2011) Dispersal and paleoecology of tropical Podocarps. Ecology of the Podocarpaceae in tropical forests (ed. by B.J. Turner and L.A. Cernusak), pp. 21–41. Smithsonian Institution Scholarly Press, Washington, DC. Ornelas, J.F., Ruiz-S
anchez, E. & Sosa, V. (2010) Phylogeography of Podocarpus matudae (Podocarpaceae): pre-Quaternary relicts in northern Mesoamerican cloud forests. Journal of Biogeography, 37, 2384–2396. Pocknall, D.T. (1981) Pollen morphology of the New Zealand species of Dacrydium Solander, Podocarpus L’H
eritier, and Dacrycarpus Endlicher (Podocarpaceae). New Zealand Journal of Botany, 19, 67–95. Premoli, A.C., Mathiasen, P., Acosta, M.C. & Ramos, V.A. (2012) Phylogeographically concordant chloroplast DNA divergence in sympatric Nothofagus s.s. How deep can it be? New Phytologist, 193, 261–275. Quiroga, M.P., Pacheco, S., Malizia, L. & Premoli, A. (2012) Shrinking forests under warming: evidence of Podocarpus parlatorei (pino del cerro) from the subtropical Andes. Journal of Heredity, 103, 682–691. Journal of Biogeography ª 2015 John Wiley & Sons Ltd Ree, R.H. & Smith, A.S. (2008) Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesis. Systematic Biology, 57, 4–14. Ronquist, F. & Hulsenbeck, J.P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics, 19, 1572–1574. Rossetti, D.F. & Netto, R.G. (2006) First evidence of marine influence in the Cretaceous of the Amazonas Basin, Brazil. Cretaceous Research, 27, 513–528. Rothwell, G.W., Mapes, G., Stockey, R.A. & Hilton, J. (2012) The seed cone Eathiestrobus gen. nov.: fossil evidence for a Jurassic origin of Pinaceae. American Journal of Botany, 99, 708–720. Sauquet, H., Ho, S.Y.W., Gandolfo, M.A., Jordan, G.J., Wilf, P., Cantrill, D.J., Bayly, M.J., Bromham, L., Brown, G.K., Carpenter, R.J., Lee, D.M., Murphy, D.J., Sniderman, M.K. & Udovicic, D.F. (2012) Testing the impact of calibration on molecular divergence times using a fossil-rich group: the case of Nothofagus (Fagales). Systematic Biology, 61, 289–313. Scotese, C.R. (2010) PALAEOMAP, Earth history and climate history. Available at: http://www.scotese.com/. (Accessed: 2015 february). Sinclair, W.T., Mill, R.R., Gardner, M.F., Woltz, P., Jaffr
e, T., Preston, J., Hollingsworth, M.L., Ponge, A. & M€ oller, M. (2002) Evolutionary relationships of the New Caledonian heterotrophic conifer, Parasitaxus usta (Podocarpaceae), inferred from chloroplast trnL-F intron/spacer and nuclear rDNA ITS2 sequences. Plant Systematics and Evolution, 233, 79–104. Stark, D. (2004) Taxonomic studies of Caribbean and Central American species of Podocarpus subgenus Podocarpus: a multi-disciplinary approach. Unpublished MSc Thesis, Royal Botanic Garden Edinburgh/University of Edinburgh, UK. van Steenis, C.G.G. (1972) Nothofagus, key genus to plant geography. Blumea, 19, 65–98. Townrow, J.A. (1967) On Rissikia and Mataia, podocarpaceous conifers from the Lower Mesozoic of southern lands. Papers and Proceedings of the Royal Society of Tasmania, 101, 103–136. Uba, C.E., Heubeck, C. & Hulka, C. (2005) Facies analysis and basin architecture of the Neogene Subandean synorogenic wedge, southern Bolivia. Sedimentary Geology, 180, 91–123. Van der Hammen, T. & Hooghiemstra, H. (2000) Neogene and Quaternary history of vegetation, climate, and plant diversity in Amazonia. Quaternary Science Review, 19, 725–742. Van der Hammen, T. & Hooghiemstra, H. (2001) Historia y paleoecologia de los bosques montanos andinos neotropicales. Bosque nublados del Neotropico (ed. by M. Kappelle and A.D. Brown), pp. 63–84. Instituto Nacional de Biodiversidad, Santo Domingo de Heredia. Verboom, G.A., Linder, H.P., Forest, F., Hoffmann, V., Bergh, N.G. & Cowling, R.M. (2014) Cenozoic assembly of 11 M. P. Quiroga et al. the Greater Cape Flora. Fynbos ecology, evolution, and conservation of a megadiverse region (ed. by N. Allsopp, J.F. Colville and G.A. Verboom), pp. 93–118. Oxford University Press, Oxford, UK. Warren, B.H. & Hawkins, J.A. (2006) The distribution of species diversity across a flora’s component lineages: dating the Cape’s ‘relicts’. Proceedings of the Royal Society B: Biological Sciences, 273, 2149–2158. Wilf, P. (2012) Rainforest conifers of Eocene Patagonia: attached cones and foliage of the extant Southeast Asian and Australasian genus Dacrycarpus (Podocarpaceae). American Journal of Botany, 99, 562–584. Wilf, P. & Escapa, I. (2015) Green Web or megabiased clock? Plant fossils from Gondwanan Patagonia speak on evolutionary radiations. New Phytologist, 207, 283–290. Wilf, P., Johnson, K.R., C
uneo, N.R., Smith, M.E., Singer, B.S. & Gandolfo, M.A. (2005) Eocene plant diversity at Laguna del Hunco and R
ıo Pichileuf
u, Patagonia, Argentina. The American Naturalist, 165, 634–650. Wilf, P., Little, S.A., Iglesias, A., Zamaloa, M.C., Gandolfo, M.A., C
uneo, N.R. & Johnson, K.R. (2009) Papuacedrus (Cupressaceae) in Eocene Patagonia, a new fossil link to Australasian rainforests. American Journal of Botany, 96, 2031–2047. Wilf, P., C
uneo, N.R., Escapa, I.H., Pol, D. & Woodburne, M.O. (2013) Splendid and seldom isolated: the paleobiogeography of Patagonia. Annual Review of Earth and Planetary Sciences, 41, 561–603. Wilf, P., Escapa, I.H., C
uneo, N.R., Kooyman, R.M., Johnson, K.R. & Iglesias, A. (2014) First South American Agathis (Araucariaceae), Eocene of Patagonia. American Journal of Botany, 101, 156–179. Woodburne, M.O., Goin, F.J., Bond, M., Carlini, A.A., Gelfo, J.N., L
opez, G.M., Iglesias, A. & Zimicz, A.N. (2014) Pale- 12 ogene land mammal faunas of South America; a response to global climatic changes and indigenous floral diversity. Journal of Mammalian Evolution, 21, 1–73. Zachos, J., Pagani, M., Sloan, L., Thomas, E. & Billups, K. (2001) Trends, rhythms, and aberrations in global climate 65 Ma to present. Science, 292, 686–693. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Appendix S1 List of species, distribution and accession numbers of Podocarpus and outgroup taxa. Appendix S2 Extended Materials and Methods section. Appendix S3 Phylogenetic trees for each independent DNA region. BIOSKETCH M. Paula Quiroga conducts research in phylogeny, phylogeography and population genetics of trees in South America. She has a special interest in interpreting the biogeographical history in disjunct genera and species on the continent. Author contributions: M.P.Q. led the research, analyses and writing, P.M. analysed divergence time and DEC model. A.I. contributed with the fossil evidence of several Podocarpaceae genera. A.C.P. conceived the ideas. R.R.M. provided expertise on Podocarpus biogeography. Editor: Malte Ebach Journal of Biogeography ª 2015 John Wiley & Sons Ltd