TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England PA L A E O B O TA N Y Prumnopitys anglica sp. nov. (Podocarpaceae) from the Eocene of England David R. Greenwood,1 Christopher R. Hill2 & John G. Conran3 1 Department of Biology, Brandon University, 270-18th Street, Brandon, Manitoba, Canada, R7A 6A9 2 Department of Palaeobotany and Palynology, Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, 39 Beijing Donglu, Nanjing 210008, P.R. China 3 ACEBB & SGC, School of Earth and Environmental Sciences, Benham Bldg, DX 650 312, The University of Adelaide, SA 5005, Australia Author for correspondence: David R. Greenwood, greenwoodd@brandonu.ca Abstract Leaves from the Eocene Bournemouth Freshwater Beds of southern England assigned originally to the genus Podocarpus L’Hér. ex Pers. (Podocarpaceae) as Podocarpus eocenicus Unger—currently treated as Amentotaxus gladifolia (R. Ludwig) D.K. Ferguson, Jähnichen & Alvin—are re-examined to assess their generic and familial status. Well-preserved cuticle on one specimen and its gross morphology, together with the morphology of other associated specimens suggests that they should be placed in Prumnopitys Philippi (Podocarpaceae), the long, lanceolate leaves resembling Pr. amara (Blume) de Laub. (= Sundacarpus amarus (Blume) C.N. Page). Because the fossil taxon is consistent with the genus Prumnopitys and not with Amentotaxus gladifolia (Taxaceae), a new species, Prumnopitys anglica D.R. Greenw., C.R. Hill & Conran sp. nov., is proposed. Phylogenetic comparison with other Prumnopitys species suggests that Pr. anglica is close to the extant Australasian species, Pr. amara but is still relatively isolated. The presence of Prumnopitys in the earliest Palaeocene of Asia and the Eocene of southern England supports the view that Podocarpaceae were an uncommon but distinctive component of Northern Hemisphere Late Cretaceous and Palaeogene floras before becoming largely extinct there in the Neogene. Keywords Amentotaxus; Bournemouth Freshwater Beds; Branksome Formation; Eocene; fossil; Podocarpaceae; Podocarpus; Prumnopitys Supplementary Material The Electronic Supplement (Table S1) is available in the Supplementary Data section of the online version of this article (http://www.ingentaconnect.com/content/iapt/tax). Received: 2 Nov. 2012; revision received: 12 Mar. 2013; accepted: 21 Apr. 2013 INTRODUCTION There are numerous reports of Mesozoic and Cenozoic Podocarpaceae megafossils and palynomorphs from the Northern Hemisphere, but Florin (1963) discounted most earlier reports and considered the Podocarpaceae to be an exclusively Southern Hemisphere family. However, later reports have tended to favour the opposing view: that the family formerly was widespread in North America, Europe and eastern Asia during the Mesozoic and Cenozoic (e.g., Mädler, 1939; Ferguson, 1967; Dilcher, 1969; Taggart, 1973; Krassilov, 1974; Harris, 1979; Nosova & Kiritchkova, 2008; Reichgelt & al., 2013). Recent molecular phylogenetic analyses (Conran & al., 2000; Sinclair & al., 2002; Biffin & al., 2011b; Knopf & al., 2012) support recognition of the major genera in the family and corroborate the fossil evidence suggesting that extant genera are the last remnants of formerly much more diverse and widely distributed lineages (R.S. Hill, 1995; Crisp & Cook, 2011; Jordan & al., 2011; Leslie & al., 2012). Biffin & al. (2011b) and Leslie & al. (2012) also proposed timings of major branching points for clades within the Podocarpaceae, as well as the time of origin of the genera, the latter authors as part of a broader analysis of conifers. Leslie & al. (2012) concluded from their analysis of living and fossil species that northern conifer clades reflected Neogene speciation, whereas southern clades, including the Podocarpaceae, were more ancient, reflecting persistence in the Southern Hemisphere of mild wet habitats, and the loss of these in the Northern Hemisphere during the Neogene. The Eocene Bournemouth Freshwater Beds of southern England are of considerable interest palaeobotanically although they are now only very rarely collectable because of coastal erosion protection measures. They contain a large and diverse flora principally composed of leaves (Gardner, 1882, 1895), some with preserved cuticle, whereas the systematically better known Early Eocene London Clay Flora is predominantly composed of reproductive structures (Collinson, 1996; McElwain, 1998). The Bournemouth Freshwater Beds are now considered part of the Middle Eocene Branksome Formation of the Hampshire Basin, a complex of fluvial and marginal-marine sediments, and the flora is considered Lutetian (48.6–40.4 Ma) in age (Bristow & al., 1991; Collinson, 1996; McElwain, 1998; King, 2006). Version of Record (identical to print version). 565 TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England A large number of plant fossil specimens from the Bournemouth Freshwater Beds is housed in the Natural History Museum, London (BM), including a cuticle collection (McElwain, 1998). Amongst them are leaves originally attributed by Gardner (1886) to the species Podocarpus eocenicus Unger (1850), which is currently treated as Amentotaxus gladifolia (R. Ludwig) D.K. Ferguson & al. (Taxaceae). Bandulska (1923) examined material from this flora with cuticle, describing some coniferous leaves, but did not comment on any Podocarpaceae. According to McElwain (1998) little systematic research on Bandulska’s cuticle collection from the Bournemouth Freshwater Beds has been performed since they were collected. Most of the Podocarpus-like leaves from the Bournemouth Freshwater Beds are impressions in mudstone with an organic stain. However, examination of the collections revealed a single specimen with well-preserved cuticle, suggestive of a placement in Prumnopitys Phil., a genus of nine or ten living species mainly from the Southern Hemisphere (de Laubenfels, 1978; Page, 1989, 1990; Stockey & Frevel, 1997; Farjon, 2001, 2010; Kelch, 2002; Mabberley, 2008), with a fossil record possibly extending back to at least the earliest Palaeocene in eastern Asia (Krassilov, 1974) and the Eocene of North America (Dilcher, 1969) (Fig. 1). In particular, the fossil closely resembles the large-leaved Pr. amara (Blume) de Laub., which is sometimes segregated as the monospecific genus Sundacarpus (J. Buchh. & N.E. Gray) C.N. Page (Page, 1989, 1990; Farjon, 2001, 2010), or even placed into a separate order (Podocarpales) from Prumnopitys (Prumnopityales) (Melikian & Bobrov, 2000). However, as several morphological (de Laubenfels, 1978; Stockey & Frevel, 1997), as well as many molecular studies of Podocarpaceae include Pr. amara within Prumnopitys (Conran & al., 2000; Kelch, 2002; Sinclair & al., 2002; Biffin & al., 2011b; Leslie & al., 2012), it is placed within Prumnopitys for this study. Despite the suggestion by Knopf & al. (2012) that cuticular characters in Podocarpaceae are only really useful at the species level, long-standing work on Australian and New Zealand Eocene and Miocene Podocarpaceae has highlighted the utility of cuticle for the identification of Cenozoic representatives of the family at the generic level (e.g., Greenwood, 1987; Wells & Hill, 1989; Pole 1992a, b, 1997a, b, 2007; Carpenter & al., 2011; Jordan & al., 2011). Similarly, authors such as Stockey & Ko (1988), Wells & Hill (1989) and Pole (1992a, 1997b, 2007) have demonstrated the greater detail possible for cuticular studies of conifers generally through the use of scanning electron microscopy (SEM). Accordingly, this study aims to place the Bournemouth fossil phylogenetically through morphological comparison with living Podocarpaceae including Prumnopitys species, as well as with comparable, large-leaved Podocarpus L’Hér. ex Pers. and the superficially similar but anatomically distinct genus Amentotaxus Pilg. (Taxaceae). MATERIALS AND METHODS Fossil materials. — Compression fossils of leaves assigned to Podocarpus sp. ? and Po. eocenicus Unger (some erroneously labelled “P. eocenica Gardner”) in the collections of the British Museum of Natural History (Natural History Museum, BM) from the Eocene Bournemouth Freshwater Beds were surveyed in 1990 to find specimens with cuticle. A single specimen collected by Bandulska (specimen no. V.46883) was found to have cuticle and was sampled. In order to assess the generic status of the Bournemouth material, cuticle samples were taken for analysis by both transmitted light microscopy (TLM) and SEM. TLM comparisons with modern herbarium and field-collected material (representing 73 of the approximately 170–200 species in the Podocarpaceae) and published accounts of the cuticle morphology of extant and fossil Prumnopitys species (Stockey & Ko, 1988; Stockey & Frevel, 1997) were undertaken. In addition, fossil material of Australian (Greenwood, 1987: Po. platyphyllus D.R. Greenw.) and New Zealand (Pole, 1993: Po. travisiae Pole) Fig. 1. Map showing the Bournemouth Freshwater Beds (Branksome Formation) fossil collection site (□) and the distribution of living Prumnopitys (enclosed by thick black outline) and of other fossil Prumnopitys (■) or putative Prumnopitys-like species (). 566 Version of Record (identical to print version). TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England broad-leaved Podocarpus fossils was studied to assess their affinities to the English material. Similarly, the Prumnopitys-like earliest Palaeocene (Danian 65.5–61.7 Ma, but listed originally as Latest Cretaceous) Po. tzagajanicus Krassilov from the Tzagajan beds in the Bureya River region of south-eastern Russia (Krassilov, 1974) was also included, based on the description and illustrations in Krassilov’s paper. Fossil and living Amentotaxus were also examined given the close superficial similarity of their leaves to the fossil and their presence in the Bournemouth Beds. Leaf fragments from the Bournemouth fossil were cleaned overnight in 40% HF at room temperature, followed by maceration with Schulze’s solution (70% HNO3 saturated with KClO3) and clearing with 5% NH4OH. At all stages the material was washed thoroughly with several changes of distilled deionized water. For TLM the cuticles were mounted unstained on microscope slides in phenol-glycerin jelly and examined using both bright field and Nomarski differential interference contrast optics. Comparative TLM sides of extant taxa were prepared and mounted in a similar manner but were either stained in 1% aqueous safranin O, or 0.1% crystal violet. For scanning electron microscopy, cuticle fragments of fossil and extant taxa were dried from 50% ethanol onto cover slips pre-mounted on standard stubs with epoxy resin, sputter coated with gold/ palladium and examined in a Hitachi S800 field emission SEM or Cambridge S360 Stereoscan SEM at 8 kV. Phylogenetic analysis. — Phylogenetic analysis to determine the relationship of relevant fossil taxa to modern genera was undertaken using the data for 23 OTUs and 57 characters (Table 1). Extant taxa in Prumnopitys (including Sundacarpus) Table 1. Characters and character states used for phylogenetic analysis of extant and fossil Prumnopitys and outgroup taxa. 1. Leaf arrangement; 0: alternate helical; 1: alternate 2-ranked; 2: opposite; 3: pseudo-verticillate 28. Epidermal papillae; 0: absent; 1: present 2. Leaf fall; 0: single; 1: branchlets 30. Epidermal periclinal walls pitted; 0: absent; 1: present 3. Leaf maximum length [mm]; 0: > 50; 1: 1–50 4. Leaf maximum width [mm]; 0: > 5; 1: 1–5 29. Epidermal periclinal walls rugose; 0: absent; 1: present 31. Epidermal periclinal walls with globules; 0: absent; 1: present 32. Epidermal periclinal walls granular; 0: absent; 1: present 5. Leaf symmetry; 0: symmetrical; 1: asymmetrical 6. Leaf curvature; 0: straight; 1: falcate 33. Epidermal periclinal wall sculpturing stronger abaxially; 0: absent; 1: present 7. Leaf apex; 1: acute; 2: mucronate; 3: bluntly truncate 34. Transverse mesophyll fibre ridges; 0: absent; 1: present 8. Petiole length [mm]; 0: > 3; 1: 1–3; 2: ~absent 35. Adaxial stomatal band number; 0: absent; 1: single per side; 2: >1 per side 9. Leaf vein number; 0: > 1 vein per leaf; 1: 1 vein per leaf 10. Midrib (adaxial); 0: plane; 1: grooved; 2: raised 11. Midrib (abaxial); 0: plane; 1: grooved; 2: raised 12. Abaxial leaf surface strongly white-waxy; 0: absent; 1: present 13. Adaxial epidermal cell shape; 0: isodiametric; 1: irregular; 2: rectangular 36. Abaxial stomatal bands discontinuous; 0: absent; 1: present 37. Stomatal rows uniseriate; 0: absent; 1: present 38. Adaxial stomatal length [µm]; 0: absent; 1: 1–40; 2: >40 39. Adaxial stomatal width [µm]; 0: absent; 1: 1–35; 2: >35 40. Abaxial stomatal length [µm]; 0: 1–40; 1: >40 14. Adaxial epidermal cell length [µm]; 0: 1–40; 1: 41–80; 2: > 80 41. Abaxial stomatal width [µm]; 0: 1–35; 1: >35 15. Adaxial epidermal cell width [µm]; 0: 1–20; 1: 21–40; 2: > 40 42. Stomatal papillae; 0: absent; 1: present 16. Adaxial epidermal anticlinal walls beaded; 0: absent; 1: present 43. Stomatal plug components; 0: solid layers; 1: rods; 2: globular 17. Adaxial epidermal anticlinal walls shape; 0: straight to curved; 1: sinuous (Wilkinson (1979) type 3+) 44. Stomata amphicyclic; 0: absent; 1: present 18. Abaxial epidermal cell shape; 0: isodiametric; 1: irregular; 2: rectangular 45. Subsidiary cells radially elongated in a star-like pattern; 0: absent; 1: present 46. Two subsidiary cells; 0: absent; 1: occur; 2: common 19. Abaxial epidermal cell length [µm]; 0: 1–40; 1: 41–80; 2: > 80 47. Four subsidiary cells; 0: occur; 1: common 20. Abaxial epidermal cell width [µm]; 0: 1–20; 1: 21–40; 2: > 40 48. More than 4 subsidiary cells; 0: occur; 1: common; 2: always 21. Abaxial cell L/W ratio; 0: < 2/1; 1: 2/1 to 5/1; 2: > 5/1 49. Polar cells; 0: usually absent; 1: usually present 22. Abaxial stomate band epidermal cell length [µm]; 0: absent; 1: 1–20; 2: 21–40; 3: > 40 50. Stomatal subsidiary cell papillae; 0: absent; 1: present 23. Abaxial stomate band epidermal cell width [µm]; 0: absent; 1: 1–20; 2: 21–40; 3: > 40 52. Stomatal subsidiary cell flanges; 0: absent; 1: present 51. Stomatal subsidiary cell striations; 0: absent; 1: present 24. Abaxial epidermal anticlinal walls beaded; 0: absent; 1: present 53. Guard cell cuticular flanges; 0: thin; 1: wide 25. Abaxial epidermal anticlinal walls buttressed; 0: absent; 1: present 54. Polar extensions; 0: short, thick; 1: ribbon-like 26. Abaxial epidermal anticlinal walls shape; 0: straight to curved; 1: sinuous (type 3+) 27. Abaxial epidermal cell flanges; 0: absent; 1: present 55. Inter-guard/subsidiary cell flanges; 0: slightly rolled; 1: prominently rolled 56. Florin ring groove; 0: absent; 1: slight; 2: prominent 57. Florin ring groove surface; 0: smooth; 1: striated Version of Record (identical to print version). 567 TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England and examples of some fossil Podocarpus species with leaves superficially similarly to the Bournemouth fossil (including Po. tzagajanicus) and one extant Podocarpus species each from eight of the infrageneric sections recognised by de Laubenfels (1985) were included in the analysis (Electr. Suppl.: Table S1): Podocarpus sect. Podocarpus (Po. latifolius (Thunb.) R. Br. ex Mirb.); sect. Acuminatus de Laub. (Po. dispermus C.T. White); sect. Capitulatus de Laub. (Po. salignus D. Don); sect Crassiformis de Laub. (Po. smithii de Laub.); sect. Foliolatus de Laub. (Po. neriifolius D. Don); sect. Longifoliolatus de Laub. (Po. longifoliolatus Pilg.); sect. Polystachyus de Laub. (Po. elatus R. Br.); and sect. Scytopodium (Po. henckelii Stapf ex Dallim. & Jackson). Sciadopitys verticillata Siebold & Zucc. (Sciadopityaceae) was used as the outgroup taxon and two living Amentotaxus species (A. argotaenia (Hance) Pilg., A. formosana H.L. Li) and the fossil A. gladifolia were also included (Electr. Suppl.: Table S1), because of the presence of the latter in the Bournemouth deposit. Classification and nomenclature for extant taxa largely follow Farjon (2010), Christenhusz & al. (2011), and Earle (2011). Nevertheless, because of its consistent molecular placement, Sundacarpus was treated here as part of Prumnopitys. In addition, both formerly recognised varieties of Pr. montana (Humb. & Bonpl. ex Willd.) de Laub. were included in view of reported leaf character state differences between them (Stockey & Frevel, 1997), but with “var. meridensis” treated as an informal name, as the combination in Prumnopitys has not been made formally, with the usage of the name in Stockey & Frevel (1997) constituting a nomen invalidum. The data were analysed in ASADO v.1.89 (Nixon, 2004) with the ratchet (island hopper) option, using NONA v.2.0 (Goloboff, 1999), 1000 iterations per replicate, 10 sequential ratchet runs, 10 simultaneous threads, 10% random constraints and 20 trees to hold per iteration; and six characters to resample per run. All characters were coded as non-additive and uninformative characters were deactivated prior to analysis. Branch support was calculated by bootstrapping using 1000 replicates, 10 search replicates and one tree to hold per iteration (TBR off). Bremer support was also determined. The phylogenetic position of the fossils was determined using a strict consensus tree produced from the analysis, with character evolution explored by mapping character state changes onto the branches of a randomly selected most parsimonious tree derived from the analysis. RESULTS Fig. 2. Strict consensus tree derived from 5602 most-parsimonious trees (length 188 steps; CI 44; RI 72) from a phylogenetic analysis of extant and fossil Podocarpaceae, principally members of Prumnopitys, with Sciadopitys as the outgroup, using the characters and states listed in Tables 1 and S1 (Electr. Suppl.). Numbers at nodes represent bootstrap percentages > 50% and Bremer support values, respectively. 568 The phylogenetic analysis resulted in 5602 most parsimonious trees (length 188, consistency index 44, retention index 72). There were three main lineages common to all trees, corresponding to Amentotaxus (99% bootstrap support; Bremer decay 6), Podocarpus (BS 79%; BD 2) and Prumnopitys (BS 93%; BD 2), with the Bournemouth fossil placed in an unsupported polytomy (BS < 50%) with Pr. amara from sect. Sundacarpus (J. Buchholz & N.E. Gray) de Laub., Pr. ferruginoides (Compton) de Laub. and an unsupported clade representing the members of Sect. Prumnopitys (Fig. 2). Podocarpus tzagajanicus was placed in a poorly supported (BS 60%; BD 1) trichotomy with the Podocarpus and Prumnopitys clades. Character state mapping of unambiguous character state changes (to avoid possible issues of parsimony ancestral character reconstruction caused by DELTRAN or ACCTRAN raised by Agnarsson & Miller 2008) onto a randomly selected tree from the most-parsimonious solutions (Fig. 3), showed that based on the character set used here, leaf anatomical and morphological features are important at the family and generic level, supporting the conclusions of numerous palaeontological conifer studies. Podocarpaceae were defined, albeit with weak support, by the hypothetical ancestral character (HAC) states of absence of uniseriate stomatal rows (Character 37/ state 0, as per Table 1), presence of amphicyclic stomata (44/1), and a prominent Florin ring groove (56/2) with a striated surface (57/1). The Podocarpus clade shared the synapomorphies of > 1 stomatal abaxial band per side (35/2), abaxial stomatal length ≤ 40 μm (40/0), and always > 2 subsidiary cells (48/2); whilst the Prumnopitys (incl. Sundacarpus) clade was defined Version of Record (identical to print version). TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England Fig. 3. Randomly selected tree from the 5602 most-parsimonious trees (tree 1; length 188 steps; CI 44; RI 72) resulting from a phylogenetic analysis of extant and fossil Podocarpaceae, principally members of Prumnopitys, with Sciadopitys as the outgroup, mapping character evolution using the characters and states listed in Tables 1 and S1 (Electr. Suppl.). Unique synapomorphies are indicated by filled circles, and homoplasious character states by open circles. by the HAC states of no abaxial epidermal flanges (27/0), presence of subsidiary cell wall striations (51/1), ribbon-like polar extensions (54/1) and prominently rolled inter-guard/subsidiary cell flanges (55/1). Within Prumnopitys s.str., the sect. Prumnopitys clade was supported by the HAC states of leaves arranged in branchlet units (2/1) and each leaf ≤ 50 mm long (3/1), ≤ 5 mm wide (4/1) and asymmetrical (5/1), and with a petiole 1–3 mm long (8/1). Although some specimens referred to Pr. anglica (but lacking cuticle) appear to possess a petiole (Fig. 4H–I), most specimens including the holotype lacked this feature. Interestingly, the two forms of Pr. montana were separated from each other in the analysis (and were not sister to each other in any of the most-parsimonious trees), supporting the observations of Stockey & Frevel (1997) that they are apparently distinct from each other and indicating that further work is needed to determine their taxonomic status. TAXONOMIC TREATMENT Order: Araucariales Family: Podocarpaceae Genus: Prumnopitys Philippi Prumnopitys anglica D.R. Greenw., C.R. Hill & Conran, sp. nov. – Holotype: England, Clay lenses, Bournemouth Freshwater Beds (Branksome Fm.), coastal exposure at Bournemouth, England, Bandulska s.n. (BM no. V.46883). — Figures 4C, J–K; 5A–I. Version of Record (identical to print version). 569 TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England Diagnosis. – Leaf simple, entire, hypodromous, linear to lanceolate, 3–6 mm wide, more than 5 cm long, hypostomatic. Stomatal bands narrow and well defined, individual stomata free and in discontinuous rows. Stomata paratetracytic, with rectangular pores and a Florin ring; polar subsidiary cells not shared and generally extending past their contact with lateral subsidiary cells by more than half their length. Anticlinal walls of epidermal cells pitted, generally straight, rarely undulate. Form of the stomata (Fig. 5F–H) like that of the family Podocarpaceae (Figs. 6, 7), and especially of Prumnopitys (Figs. 6A–F, 7). Differs from Pr. andina (Poepp. ex Endl.) de Laub., Pr. exigua de Laub., Pr. ferruginea (D. Don.) de Laub., Pr. ladei (F.M. Bailey) de Laub., Pr. montana (Humb. & Bonpl. ex Willd.) de Laub. “var. meridensis”, and Pr. taxifolia (Sol. ex. D. Don.) de Laub. by presence of well-developed Florin rings. The character of the wax plug associated with the stomatal aperture under SEM (Fig. 5D–E) is especially similar to that seen in the extant species Pr. amara (Fig. 6E–F), but differs from that seen in other extant or fossil species of Prumnopitys. The general form of the leaves with a solitary, large narrow lamina (Fig. 4A–I) is also comparable with Pr. amara. Differs from the fossil and extant species of Prumnopitys with the exception of Pr. amara, Pr. andina and Pr. taxifolia in the possession of singly abcissing leaves vs. distichous shoots shed as a unit. Differs from Pr. andina, Pr. exigua, Pr. ferruginea, Pr. harmsiana (Pilg.) de Laub., Pr. montana (Humb. & Bonpl. ex Willd.) de Laub., Pr. standleyi (J. Buchholz & N.E. Gray) de Laub. and Pr. taxifolia in the possession of epidermal cells with straight anticlinal walls vs. sinuous walls. Differs from all Australian and New Zealand fossil species of Prumnopitys in possessing comparatively large leaves (> 40mm length) and well developed Florin rings. Description. – Leaves typically isolated, simple, lamina linear to narrowly lanceolate, at least 5–8 cm long and 3–6 mm wide, narrowing to an acuminate apex, basally acute-tapering, margins entire (Fig. 4A–I); hypodromous, often with two narrow prominent lighter bands, one on either side of the midvein; attachment (seen only in V.521 for a single leaf, see Fig. 4A) slightly twisted, narrow (Fig. 4A); phyllotaxis unknown. Epidermis hypostomatic, stomatiferous surface with a diffuse stomatal band on either side of a wide stomate-free area over the midvein, 34–41 cells wide (Figs. 4J–K; 5A, F). Stomata solitary in diffuse rows, adjacent stomata rarely in the same row, typically irregularly arranged and widely spaced (Figs. 4K, 5F); paratetracytic (occasionally with lateral subsidiary cells divided) with both lateral and polar subsidiary cells distinct (Fig. 5G–H); stomatal pores prominent, rectangular, parallel to long axis of leaf. Polar subsidiary cells more or less square to rectangular with square end walls, generally not shared with adjacent stomata of the same row; always projecting beyond contact with lateral subsidiary cells. Florin ring present, not strongly developed, visible over stomatal apparatus under TLM, obscured by a wax plug under SEM (Fig. 5D). Subsidiary cell lateral anticlinal walls generally well developed, sometimes not discernible. Non-stomatal epidermal cells in the stomatal bands and of adaxial cuticle characteristically square to rectangular, sometimes trapezoid or slightly hexagonal, with square to sometimes angular end walls angled at up to 45 degrees, rarely 570 more (Figs. 4K, 5A–B, F), rarely isodiametric; anticlinal walls smooth to slightly beaded with well-developed pits. Epidermal cells of stomate-free areas of abaxial cuticle also characteristically elongate-rectangular, greatly elongated near the margin (Fig. 4J) and on some parts of the midvein. Etymology. – The specific epithet refers to the fossil coming from England. Specimens investigated. – Prumnopitys anglica: V.521 (Fig. 4A), V.529, V.15107 (Fig. 4I), V.15108 (Fig. 4E), V.15109 (Fig. 4H), V.15111 (Fig. 4G), V.15112 (Fig. 4B), V.15113 (Fig. 4F), V.45091, V.46114, V.46117, V.46118 (Fig. 4D), and V.46883 (holotype, Figs. 4C, J–K, 5). Paratypes, V.15108 (Fig. 4E) and V. 46118 (Fig. 4D). DISCUSSION Comparison with living Podocarpaceae. — Recent molecular phylogenies of Podocarpaceae generally speaking support the traditional morphology-based classification of the family but differ in the placement of individual genera into clades (Conran & al., 2000; Sinclair & al., 2002; Biffin & al., 2011b; Leslie & al., 2012). The cuticles of leaves from modern Podocarpaceae have a number of characteristics that together distinguish them from other families, contrary to one recent opinion (Knopf & al., 2012). The stomata of Podocarpaceae are paratetracytic or amphicyclic and the stomatal pores are characteristically aligned parallel to the long axis of the leaf (e.g., Figs. 6B, 7I), although the exact number of subsidiary cells surrounding the stomatal pore may range from two to as many as seven by division of the lateral cells (Dilcher, 1969; Greenwood, 1987; Stockey & Ko, 1988; Stockey & Frevel, 1997). In many genera the subsidiary cells are not clearly defined, but in Podocarpus sensu de Laubenfels (1969, 1978, 1985) they are well defined and the polar cells typically do not, or only scarcely project beyond the border of the lateral subsidiary cells (Greenwood, 1987; Stockey & al., 1998; as in the fossil species Po. travisiae Pole [Pole, 1993] shown in Fig. 6H). In Prumnopitys, the polar subsidiary cells (if present) are typically rectangular and extend beyond their contact with the lateral subsidiary cells by more than half their length (Greenwood, 1987; Pole, 1992a, 1997a, 2007; Stockey & Frevel, 1997; e.g., Figs. 7D, H, I). However, Stockey & Frevel (1997) found that polar cells are sometimes to usually absent in some species, always absent in Pr. harmsiana, and are only normally present in Pr. ferruginea, Pr. ladei and Pr. taxifolia (Electr. Suppl.: Table S1). Pole (2007) also considered the tendency for the subsidiary cells in the stomatal complexes for stomata in adjacent rows to “bulge out from their common walls” to be diagnostic for the genus. However, Stockey & Ko (1988) and Stockey & Frevel (1997) found that Pr. ferruginoides (Fig. 7E–F) is an exception, typically having stomata of the form found in Podocarpus s.str., suggesting that it is plesiomorphic. The leaves of Amentotaxus Pilger (Taxaceae) and some modern species of Podocarpus and Prumnopitys can be confused due to their shared large size and linear to lanceolate shape. However, according to Ferguson & al. (1978), fresh or Version of Record (identical to print version). TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England Fig. 4. Prumnopitys anglica sp. nov., including the holotype A–I, fossil specimens: A, H, and I, adaxial views; B, D–G, abaxial; C, orientation uncertain. A, fragment of a shoot (at bottom) bearing a nearly complete leaf, previously figured as Pl II, fig. 15 by Gardner (1886), V.521; B, V.15112; C, holotype, V.46883; D, paratype, V.46118; E, paratype, V.15108; F, V.15113; G, V.15111; H, V.15109; I, V.15107. J–K, transmission light micrographs (TLM) of cuticle of holotype, V.46883: J, upper (adaxial) cuticle plus, at left, the leaf margin and marginal region of the lower cuticle; K, lower (abaxial) cuticle with midrib region at left. — Scale bars/magnification: A, B, D, F, I = 10 mm approx.; C, E, H = 8.4 mm approx.; G = ×2.45; J–K = 0.45 mm. Version of Record (identical to print version). 571 Greenwood & al. • An Eocene Prumnopitys from England TAXON 62 (3) • June 2013: 565–580 Fig. 5. Prumnopitys anglica, cuticle of holotype, V.46883. A–C, inner views of upper (adaxial) cuticle, showing typically elongated cells with prominent and minutely interrupted anticlinal cuticular flanges: A, TLM; B–C, SEM. D–E, exterior views using SEM of a stomatal plug (arrow in D), composed of soluble cuticular lipids. F, inner view of lower (abaxial) cuticle under SEM, showing typical cell outlines and general distribution of stomata; G, Optical Nomarski differential interference contrast micrograph of inner view of a stomatal apparatus (arrow); H, inner view by SEM of a stomatal apparatus (arrow); I, cuticle as seen in cross section (arrow) under SEM. — Scale bars: A–B = 100 μm; C = 40 μm; D = 20 μm; E = 1 μm; F = 200 μm; G–H = 50 μm; I = 5 μm. 572 Version of Record (identical to print version). TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England Fig. 6. Cuticle of the extant species Prumnopitys amara (A–F) and fossil Podocarpus travisiae Pole (G–I) stained with crystal violet under TLM (transmitted light microscopy) and SEM (scanning electron microscopy). A, TLM view of adaxial cuticle of Pr. amara showing sinuous anticlinal walls; B, TLM view of abaxial cuticle showing files of stomata in interrupted rows; C, TLM view showing close up view of stomata with Florin rings; D–F, SEM views of stomata displaying cuticular plugs (E–F), Florin rings (D–E), plug area arrowed in E and detail of stomatal plug rods in F; G, TLM view of adaxial cuticle of Po. travisiae showing sinuous anticlinal walls; H, SEM view of inner cuticle showing stomata in rows and form of stomatal complex with polar subsidiary cells within the ring of cells, I, SEM view of abaxial outer cuticle surface showing Florin rings and absence of stomatal plugs. — Scale bars: A = 100 μm; B, D = 500 μm; C, G–H = 50 μm; E = 20 μm; F = 1 μm; I = 25 μm. Version of Record (identical to print version). 573 Greenwood & al. • An Eocene Prumnopitys from England TAXON 62 (3) • June 2013: 565–580 Fig. 7. Selection of TLM views of extant Prumnopitys cuticles (A–B and O–P, stained with crystal violet, the remainder with Safranin O). A–B, Pr. andina; C–D, Pr. ferruginea; E–F, Pr. ferruginoides; G–H, Pr. harmsiana; I–J, Pr. ladei; K–L, Pr. montana; M–N, Pr. standleyi; O–P, Pr. taxifolia. — Scale bars = 100 μm. 574 Version of Record (identical to print version). TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England dried leaves of Amentotaxus can be distinguished from those of Podocarpus s.l. on the basis of their possession of a whitish wax-like covering on the stomatal bands and by the unusual star-like arrangement of numerous subsidiary cells around the stomatal pore. The waxy coating may preserve as a darker or lighter band on carbonaceous fossils, but whilst some of the specimens from the Bournemouth Freshwater Beds exhibit broad light coloured bands like Amentotaxus (Fig. 4B, D–E, G), the stomatal subsidiary cell arrangement is typical of Podocarpaceae (Fig. 5F–H). A characteristic of some modern species of Prumnopitys is in fact the restriction of the stomata into narrow bands (in hypostomatic species) which can appear as two broad lighter areas running either side of the relatively dark midvein. The lanceolate leaves of the Bournemouth specimens (Fig. 4A–I) suggest a placement in Podocarpus; however, the stomata (Fig. 5F–H) are of a form commonly seen in Prumnopitys (Fig. 7). There are ten living species of Prumnopitys, of which nine are currently placed in sect. Prumnopitys and a single species, Pr. amara, in sect. Sundacarpus (de Laubenfels, 1978). The species in sect. Prumnopitys are found as follows: in South and Central America, Pr. andina, Pr. exigua, Pr. harmsiana, Pr. montana and Pr. standleyi; New Caledonia Pr. ferruginoides; New Zealand, Pr. ferruginea, Pr. taxifolia; and Pr. ladei from north-eastern Queensland. Prumnopitys amara is also found in north-eastern Queensland, but extends to New Guinea, parts of Indonesia and the southern Philippines (de Laubenfels, 1978). Page (1989) transferred Pr. amara to a separate genus, Sundacarpus and Stockey & Frevel (1997), although retaining it in Prumnopitys, noted that its cuticular anatomy differed from other species in the genus. However, molecular phylogenetic analyses of Podocarpaceae demonstrate that Pr. amara is nested within Prumnopitys (Conran & al., 2000; Sinclair & al., 2002; Biffin & al., 2011b; Leslie & al., 2012), a placement we follow here. Prumnopitys amara has broadly lanceolate to linear leaves, similar to Pr. anglica, whereas other extant Prumnopitys have much narrower leaves than either Pr. anglica or Pr. amara, often arranged in two ranks on small, leaf-like deciduous twigs each representing a single growth event. The above discussion and our phylogenetic analysis (Figs. 2–3) indicate that some of the material identified by Gardner (1886) as Podocarpus or Po. eocenicus from the Bournemouth Freshwater Beds (Branksome Formation) belongs to Prumnopitys sect. Sundacarpus. This material (i.e., the specimens listed above in the Taxonomic Treatment) is accordingly described as Prumnopitys, recognising the molecular evidence for placing the extant species Pr. amara within that genus. However, it is important to note that cuticle could be recovered from only one of these specimens and that although some cuticular characters resemble those of Pr. amara (Fig. 6B, D–F), some are more like other species of Prumnopitys (Fig. 7). However, both conditions are consistent with the unresolved phylogenetic placement of Pr. anglica within the Prumnopitys s.l. clade (Fig. 2). In the absence of cuticle or the light coloured (? stomatal) bands on the leaves, the remaining Bournemouth specimens, including perhaps some of those listed above, are of doubtful identity. Modern species of Prumnopitys typically have epidermal cells with well-developed sinuous anticlinal walls (Florin, 1931; Greenwood, 1987; Stockey & Frevel 1997; e.g., Figs. 6A, C, 7A–B, G, O); a feature lacking in the fossil. However, sinuosity amplitude is variable in P. amara, both within and between leaves on a single plant. Stockey & Frevel (1997) reported that it possessed straight anticlinal walls in the adaxial cuticle (cf. Fig. 6A), whereas both Pr. ferruginoides and Pr. ladei (Fig. 7E, I) lack sinuous anticlinal walls altogether, i.e., on both adaxial and abaxial cuticles (Greenwood, 1987; Stockey & Ko, 1988; Stockey & Frevel, 1997). About half of the extant species of Prumnopitys are hypostomatic and Pr. ferruginea may have both hypostomatic and amphistomatic leaves (Townrow, 1965; de Laubenfels, 1978). There is also an approximate correspondence between hypostomatic leaves and well-developed Florin rings: the rings of thickened cuticle surrounding the stomatal pore (Florin, 1931, 1963) (Fig. 7). The leaves of Pr. anglica are intermediate in size between Pr. amara and the other extant species and are generally much narrower than those of Pr. amara. Leaves of many Prumnopitys species are borne in distichous shoots which are shed as a moreor-less intact unit at abscission (Table 1), like the deciduous branches of many conifers, but in Pr. amara they are arranged more or less helically and abscise singly (a feature it shares with Pr. andina and Pr. taxifolia, with which it consistently forms a clade in recent molecular analyses). The occurrence of mainly isolated Pr. anglica leaves, only one being attached (Fig. 4A), implies that they were also borne and shed singly and this, together with their size also suggests affinity with Pr. amara. Although the apparent lack of a petiole on the holotype might therefore seem novel, two of the other specimens referred to Pr. anglica but lacking cuticle probably do show a petiole. Comparison with other Prumnopitys-like fossil Podocarpaceae. — In Australia, several Eocene Prumnopitys species have been described, including Pr. tasmanica (Townrow) D.R. Greenw. ex R.R. Mill & R.S. Hill (2004) from Tasmania and Victoria (Townrow, 1965; Greenwood, 1987) and Pr. portensis Pole and Pr. sp. “cf. Pr. montana” of Pole from the Eocene of Tasmania (Pole, 1992b). A further species described originally as Pr. lanceolata D.R. Greenw. (Greenwood, 1987) is now placed in the extinct Retrophyllum-like genus Smithtonia R.S. Hill & Pole (R.S. Hill & Pole, 1992). However, Pr. anglica is quite different from the other Australian and New Zealand fossil species of Prumnopitys, all of which are small-leaved (< 40 mm long) and amphistomatic with poorly developed Florin rings. In New Zealand, Pr. opihiensis Pole, Pr. limaniae Pole and Pr. sp. “Mt Somers” were described from several Palaeocene localities (Pole, 1997a, 1998) and a Miocene fossil was referred by Pole (1997b) to the extant species Pr. taxifolia. Similarly, members of superficially Prumnopitys-like extinct genera such as the Falcatifolium-like Sigmaphyllum R.S. Hill & L.J. Scriven (R.S. Hill & Scriven, 1999) and plurinerved, Retrophyllum-like Smithtonia and Willungia R.S. Hill & Pole (R.S. Hill & Pole, 1992) are all small-leaved (e.g., Smithtonia and Willungia 3–33 mm long × 1–5 mm wide) and differ from the Bournemouth fossils in a range of morphological, Version of Record (identical to print version). 575 TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England cuticular and leaf arrangement characteristics (Blackburn, 1981; R.S. Hill & Pole, 1992), making them highly unlikely to have been close relatives of Pr. anglica or even Prumnopitys. A small-leaved Eocene Podocarpus described from North America by Dilcher (1969) may also represent Prumnopitys (= Podocarpus sect. Stachycarpus sensu Buchholz & Gray, 1948) and shows some correspondence with Pr. anglica in cuticle morphology; both species possess well-developed Florin rings and lack the strongly sinuous walls of most extant Prumnopitys (Fig. 7). This is of interest given the observation by Fowler & al. (1973) that the Eocene-aged Barton Leaf Bed (Hampshire) shows phytogeographic affinities to southern Florida. However, the leaf size of these two fossils is vastly different and the North American material is also amphistomatic. The exact status of the North American taxon is the subject of ongoing research. Nomenclature and status of Podocarpus eocenicus. — The correct nomenclature for previously described Podocarp-like taxa from southeastern England requires some discussion. Because the name Po. eocenicus is based on Early Cenozoic material from Europe (Unger, 1850, 1860) and does not refer to a member of the Podocarpaceae, it is therefore not conspecific with the Bournemouth specimens examined here. Gardner (1886) referred specimens of “Podocarpus” from the Bournemouth Freshwater Beds to Po. eocenicus Unger; however, several records of Podocarpus-like leaves from Eocene to Miocene localities in central and eastern Europe have also been reported as Po. eocenicus (summarized in Ferguson & al., 1978 and Jongmans & Dijkstra, 1974), as were leaves from the Eocene flora of Alum Bay, England (Ettingshausen, 1879). However, according to Ferguson & al. (1978), some, if not all of these records represent A. gladifolia, as they found no appreciable differences between the Palaeocene Ardtun Head and Miocene European specimens. Whilst suggesting that the time differences implied they may be separate species, they nevertheless included them together under the one species A. gladifolia. The Palaeocene Po. campbellii J.S. Gardner from Ardtun Head, Scotland was also transferred to A. gladifolia by Ferguson & al. (1978) and Boulter & Kvaček (1989), as were Podocarpus specimens from three German sites: Kreuzau (Middle Miocene), Salzhausen (Upper Miocene) and Eichelskopf, Holzhausen (Middle Miocene). Two of these German records had also been assigned originally to Po. eocenicus by Unger (1860) and Schindehütte (1907). Hence a new specific name is required for the podocarpaceous Bournemouth specimens. Status of Podocarpus tzagajanicus and similar fossils. — Podocarpus tzagajanicus Krassilov (Krassilov, 1974) from the earliest Palaeocene (Danian: 65.5–61.7 Ma) of south-eastern Russia was placed in sect. Stachycarpus by Krassilov (1974) and thus might be a possible relative of Pr. anglica. Based on Krassilov’s (1974) description, the gross vegetative morphology of Po. tzagajanicus is consistent with placement in Prumnopitys, but although most members of Podocarpaceae have rectangular stomatal pores (Greenwood, 1987 and unpub. data; e.g., Fig. 7), those of Pr. tzagajanicus are elliptical (Krassilov, 1974). Nevertheless, our phylogenetic analysis placed it in a polytomy at the base of the Prumnopitys line, with which it 576 shares the synapomorphies of a raised midrib (11/2), discontinuous abaxial stomatal bands (36/1) and absence of uniseriate stomatal rows (37/0), although differing from the rest of the clade in numerous features (Fig. 3). The presence of prominent papillae overarching the stomata and also on the epidermal cells in the stomatal rows (Krassilov, 1974) is not a characteristic of any living Prumnopitys and is indeed an unusual feature amongst extant members of Podocarpaceae as a whole. Stomatal subsidiary cell papillae are characteristic of extant Taxus L. and some species of Amentotaxus, such as A. formosana (Taxaceae), Torreya Arn. (Cephalotaxaceae), Sciadopitys (Sciadopityaceae) and some Cupressaceae (Florin, 1931; Dilcher, 1969; Keng, 1969; Lemoine, 1972; R.S. Hill & Carpenter, 1989), but there is variability within Amentotaxus and some species (including A. gladifolia) appear to lack them. Stomatal papillae are also widespread amongst fossil conifers in general, for example in the extinct family Cheirolepidiaceae (Alvin, 1982), such as Pseudofrenelopsis (Nathorst) Watson (e.g., Yang & al., 2009; C.R. Hill & al., 2012). They also occur in some fossil Podocarpaceae, such as Coronelia Florin from the Eocene of Chile (Florin, 1940) and Tasmania (Townrow, 1965) and Phyllocladus elongatus G.J. Jord. & al. from the Oligo–Miocene of New Zealand (Jordan & al., 2011). They have also been reported on Mataoraphyllum Pole & P.R. Moore (Taxaceae ?) from the Miocene of New Zealand (Pole & Moore, 2011), the Triassic Antarctic genus Notophytum Meyer-Berthaud & Taylor (Voltziales) and the widespread, possibly congeneric Triassic Gondwanan voltzialean genus Heidiphyllum Retallack (Axsmith & al., 1998). Overarching papillae and elliptical stomata are also known from some fossils of the extinct Mesozoic (Jurassic–Cretaceous) Northern Hemisphere conifer family Miroviaceae, some of which were also podocarp- or Amentotaxus-like in leaf form (Nosova & Wcisło-Luraniec, 2007; Nosova & Golovneva, 2010), suggesting strongly that the relationships of Po. tzagajanicus require further study, particularly given its isolated position in our rather limited analysis. Similarly, the identity and relationships of the large-leaved, papillate Po. harrisii Krassilov (Krassilov, 1967) fossil from the Aptian (125–112 Ma) Lipovtsy Formation deposits at Primorye in far eastern Russia which was related to Prumnopitys by Krassilov (1967) and Kelch (1997) also need detailed examination. Podocarpus harrisii has also been suggested to be synonymous with Mirovia orientalis (Nosova) Nosova (Miroviaceae; formerly Oswaldheeria orientalis Nosova; Bugdaeva & Markevich, 2009), which is also present in the Lipovtsy Formation (Volynets, 2009), despite the apparent lack of overarching papillae in the latter taxon (Nosova & Wcisło-Luraniec, 2007). However, Miroviaceae resembled extant Sciadopitys in possessing a median abaxial stomatal zone and in at least one species, two veins per leaf, but were apparently a distinct lineage and not closely related to either Podocarpaceae/Araucariaceae or Sciadopityaceae (Gordenko, 2007). Age and origins of Prumnopitys. — Recent dated molecular phylogenetic analyses for Podocarpaceae suggest that the major clades within the family differentiated during the Cretaceous Version of Record (identical to print version). TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England and Palaeogene, most modern genera being in existence by the late Palaeogene. Biffin & al. (2011b) suggest a mean divergence date of 101 (67–135) Ma for a prumnopityoid clade consisting of Halocarpus C.J. Quinn, Lagarostrobos C.J. Quinn, Manoao Molloy, Parasitaxus de Laub. and Prumnopitys. The dates for the divergence of Prumnopitys suggest a mean stem age of 80 (64–121) Ma and crown age of 64 (40–91) Ma, giving a probable Late Cretaceous origin for the genus—a result corroborated by Leslie & al. (2012). The sect. Sundacarpus clade was dated at 26 (22–62) Ma, with the sect. Prumnopitys clade at 42 (17–51) Ma. However, as the mean date for the divergence of Prumnopitys is much older than Po. tzagajanicus, it suggests that the latter (if actually a member of Podocarpaceae) could be part of an early, now extinct lineage within the genus, or a close relative. Florin (1963) dismissed Mesozoic and Cenozoic Podocarpaceae records from the Northern Hemisphere and some European Cenozoic Podocarpus records were subsequently transferred to Amentotaxus (Ferguson, & al., 1978). Miller (1977) similarly listed only Gondwanan fossil records for the family, whereas R.S. Hill & Brodribb (1999) suggested that although present even today in the Northern Hemisphere, the family is essentially southern. Nevertheless, dispersed Mesozoic pollen, if correctly identified as podocarp (e.g., Reichgelt & al., 2013), and Cretaceous foliar remains attributed to Podocarpaceae indicate that the family was in the Northern Hemisphere prior to the Cenozoic, persisting there until at least the Miocene (Krassilov, 1967, 1974; Dilcher, 1969; Taggart, 1973). Nageia Gaertn., for example, has a long fossil history extending back to the Lower Cretaceous of Japan and eastern Russia, leading Jin & al. (2010) to suggest an Asian origin for the genus. Eocene podocarpaceous shoots from Tennessee (Dilcher, 1969) and Miocene pollen from Oregon and Idaho, if correctly identified as podocarp (Taggart, 1973), similarly demonstrate that the family was present in the Northern Hemisphere of the New World until the early Neogene, though the proximity of South American Prumnopitys and Podocarpus species might indicate evidence for a southern immigrant origin. Although at present based solely on vegetative features, Prumnopitys anglica is demonstrated by our analysis to be a member of Podocarpaceae, confirming that the family— and Prumnopitys in particular—was well established across the Northern Hemisphere in the Eocene. This has important implications regarding the biogeography of the Pinales. As indicated above, the conventional view has been that the Podocarpaceae is a predominantly Southern Hemisphere family with only recent and/or minor incursions into the Northern Hemisphere (e.g., Florin, 1963; R.S. Hill & Brodribb, 1999). It has been postulated that the apparent separation of the Pinales into northern and southern families is instead a result of geographically selective Cenozoic extinctions (Ferguson, 1967; Krassilov, 1967, 1974; Dilcher, 1969). In contrast, Mao & al. (2012) demonstrated biogeographic divergence patterns between the subfamilies of Cupressaceae that are consistent with the breakup of Pangea, rather than resulting from factors such as global climate change, so it is likely that different factors operated for different groups of organisms. Harris (1979) noted that certain conifer morphotaxa from as early as the Jurassic of Yorkshire more or less closely resemble Podocarpaceae; particularly the mutually associated female strobilus Scarburgia Harris together with pollen-bearing male cones of Pityanthus scarburgensis van Konijnenburg-van Cittert, which are also known attached to scale-leaved shoots named Cyparissidium blackii (T.M. Harris) T.M. Harris. In addition, the occurrence of two of these three morphotaxa in Lower Cretaceous rocks from north-east China indicates that these plants were distributed widely in the Mesozoic of the Northern Hemisphere (Sun & al., 2001). There are old records of Podocarpaceae and Podocarpus in particular from the Eocene of Europe, especially France (e.g., de Saporta, 1862–1863, 1865–1866). Most of these fossils have in the past been dismissed as improbable, but in the light of our new find from England, some of these early records need re-investigation to test their credibility. In addition to the macrofossil record of Podocarpaceae, dispersed pollen attributed to Podocarpus has also been recorded. For example, Cavagnetto & Anadón (1996) recorded what they identified as Podocarpus pollen from Lower Oligocene formations in north-east Spain and Leopold & al. (2008) found what they regarded as Podocarpus pollen at the Florissant Late Eocene locality in North America. Similarly, Podocarpaceae-like palynomorphs from the Norian (Upper Triassic) Sonsela Member of the Chinle Formation at Petrified Forest National Park, Arizona (Reichgelt & al., 2013), if correctly attributed, would further support the idea of an ancient and widespread family. Many Tertiary dispersed pollen grains attributed to Podocarpus are however the stratigraphic palynomorph Podocarpidites I.C. Cookson ex R.A. Couper, which may be a member of the Podocarpaceae, but may also be Pinaceae (e.g., Leffingwell (1970) treats Podocarpidites as a “Podocarpaceae-Pinaceae complex”). Nichols & Brown (1992: F18) stated that they “regard Podocarpidites as a form genus having no implication of botanical affinity and the species P. maximus as a fossil pinaceous species”. It is our view, therefore, that in the absence of a comprehensive and detailed comparative reference database including SEM study of pollen in situ from both fossil and living podocarps (and from the often very similar Pinaceae) many if not all of the present records of Podocarpus based solely on light microscopy of dispersed pollen must be regarded as provisional. Molecular studies of conifer evolution place the family in the Cupressales, as sister order to Pinales (Rydin & al., 2002; Rai & al., 2008), with the timing of the Podocarpaceae–Araucariaceae split in the latest Permian or Triassic and the divergence of a prumnopityoid lineage from related genera in the mid-Cretaceous (Biffin & al., 2011b; Leslie & al., 2012). This suggests that the family should be more common in the Northern Hemisphere, at least historically, but Brodribb (2011) and Leslie & al. (2012) theorised that the generally poor response of podocarps to climate change, particularly cold, ultimately led to their extinction there. Similarly, Biffin & al. (2011a) postulated that competition with angiosperms in the Cretaceous led to significant changes in podocarp distributions worldwide, as well as the multiple evolution of broad-leaved forms within the family. Version of Record (identical to print version). 577 TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England The few other good macrofossil records for Podocarpaceae in the Northern Hemisphere (e.g., Dilcher, 1969; Harris, 1979; Krassilov, 1974) do not mean that they were as common there as in the Eocene of the Southern Hemisphere (e.g., Greenwood, 1987; R.S. Hill, 1994 and associated references), but do demonstrate that the family was a persistent although now extinct element, a pattern repeated in other gymnosperm groups (Crisp & Cook, 2011). In conclusion the presence of Pr. anglica in the Eocene of England shows that Podocarpaceae were clearly cosmopolitan in the Early Tertiary, albeit with an apparent centre of greatest diversity (both past and present) in the Southern Hemisphere. ACKNOWLEDGEMENTS DRG would like to thank Margaret Collinson and Jerry Hooker for hospitality and assistance with preparation of the text during visits to the U.K., Chris King and Ian West for advice on the geology of the “Bournemouth Freshwater Beds”, and Ed Biffin (University of Adelaide) for advice on podocarp phylogeny. Jennifer Bannister (University of Otago) is thanked for generously preparing the cuticles and SEMs of Podocarpus travisiae, as is Liz Girvan from the Otago Centre for Electron Microscopy, University of Otago for assistance and access to facilities. David Dilcher kindly loaned cuticle preparations of the Eocene North American material of Podocarpus. We thank Mr. C.H. Shute, Natural History Museum, London, for taking the optical micrographs of Pr. anglica. This project was initiated in 1990 while DRG was recipient of a NSERC International Fellowship at the University of Saskatchewan. Its conclusion was supported by a Chinese Academy of Sciences Visiting Professorship for Senior International Scientists, Grant No. 2009S1-40 awarded to CRH at the Nanjing Institute of Geology and Palaeontology, and through a Grant (DG 311934) to DRG from the Natural Sciences and Engineering Research Council of Canada. LITERATURE CITED Agnarsson, I. & Miller, J.A. 2008. Is ACCTRAN better than DELTRAN? Cladistics 24: 1–7. http://dx.doi.org/10.1111/j.1096-0031.2008.00229.x Alvin, K.L. 1982. Cheirolepidiaceae: Biology, structure and paleoecology. Rev. Palaeobot. Palynol. 37: 71–98. http://dx.doi.org/10.1016/0034-6667(82)90038-0 Axsmith, B.J., Taylor, T.N. & Taylor, E.L. 1998. Anatomically preserved leaves of the conifer Notophytum krauselii (Podocarpaceae) from the Triassic of Antarctica. Amer. J. Bot. 85: 704–713. http://dx.doi.org/10.2307/2446541 Bandulska, H. 1923. A preliminary paper on the cuticular structure of certain dicotyledonous and coniferous leaves from the Middle Eocene Flora of Bournemouth. Bot. J. Linn. Soc. 46: 241–269. http://dx.doi.org/10.1111/j.1095-8339.1923.tb00488.x Biffin, E., Brodribb, T.J., Hill, R.S., Thomas, P. & Lowe, A.J. 2011a. Leaf evolution in Southern Hemisphere conifers tracks the angiosperm ecological radiation. Proc. Roy. Soc. London, Ser. B, Biol. Sci. 279: 341–348. http://dx.doi.org/10.1098/rspb.2011.0559 Biffin, E., Conran, J.G. & Lowe, A.J. 2011b. Podocarp evolution: A molecular phylogenetic perspective. Pp. 1–20 in: Turner, B.L. & Cernusak, L.A. (eds.), Ecology of the Podocarpaceae in tropical 578 forests. Smithsonian Contributions to Botany 95. Washington, D.C.: Smithsonian Institution Scholarly Press. Blackburn, D.T. 1981. Tertiary Megafossil flora of Maslin Bay, South Australia: Numerical taxonomic study of selected leaves. Alcheringa 5: 9–28. Boulter, M.C. & Kvacek, Z. 1989. The Palaeocene flora of the Isle of Mull. London: Palaeontological Society. Bristow, C.R., Freshney, E.C. & Penn, I.E. 1991. Geology of the country around Bournemouth: Memoir for 1 : 50 000 geological sheet 329 (England and Wales). London: H.M.S.O. Brodribb, T.J. 2011. A functional analysis of podocarp ecology. Smithsonian Contr. Bot. 95: 165–173. http://dx.doi.org/10.5479/si.0081024X.95.165 Buchholz, J.T. & Gray, N.E. 1948. A taxonomic revision of Podocarpus. I. The sections of the genus and their subdivisions with special reference to leaf anatomy. J. Arnold Arbor. 29: 49–63. Bugdaeva, E.V. & Markevich, V.S. 2009. The coal-forming plants of Rhabdopissites in the Lipovtsy coal field (Lower Cretaceous of Southern Primorye). Paleontol. J. 43: 1217–1229. http://dx.doi.org/10.1134/S0031030109100049 Carpenter, R.J., Jordan, G.J., Mildenhall, D.C. & Lee, D.E. 2011. Leaf fossils of the ancient Tasmanian relict Microcachrys (Podocarpaceae) from New Zealand. Amer. J. Bot. 98: 1164–1172. http://dx.doi.org/10.3732/ajb.1000506 Cavagnetto, C. & Anadón, P. 1996. Preliminary palynological data on floristic and climatic changes during the Middle Eocene–Early Oligocene of the eastern Ebro Basin, northeast Spain. Rev. Palaeobot. Palynol. 92: 281–305. http://dx.doi.org/10.1016/0034-6667(95)00096-8 Christenhusz, M.J.M., Reveal, J., Farjon, A., Gardner, M.F., Mill, R.R. & Chase, M.W. 2011. A new classification and linear sequence of extant gymnosperms. Phytotaxa 19: 55–70. Collinson, M.E. 1996. Plant macrofossils from the Bracklesham Group (Early & Middle) Eocene, Bracklesham Bay, West Sussex, England: Review and significance in the context of coeval British Tertiary floras. Tert. Res. 16: 175–202. Conran, J.G., Wood, G.A., 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 CP gene rbcL. Austral. J. Bot. 48: 715–724. http://dx.doi.org/10.1071/BT99062 Crisp, M.D. & Cook, L.G. 2011. Cenozoic extinctions account for the low diversity of extant gymnosperms compared with angiosperms. New Phytol. 192: 997–1009. http://dx.doi.org/10.1111/j.1469-8137.2011.03862.x De Laubenfels, D.J. 1969. A revision of the Malesian and Pacific rainforest conifers, I. Podocarpaceae, in part. J. Arnold Arbor. 50: 274–369. De Laubenfels, D.J. 1978. The genus Prumnopitys (Podocarpaceae) in Malesia. Blumea 24: 189–190. De Laubenfels, D.J. 1985. A taxonomic revision of the genus Podocarpus. Blumea 30: 251–278. De Saporta, L.C.J.G. 1862–1863. Études sur la végétation du sud-est de la France à l’époque tertiaire, Première partie. Paris: Victor Masson. De Saporta, L.C.J.G. 1865–1866. Études sur la végétation du sud-est de la France à l’époque tertiaire, Deuxième partie. Paris: Victor Masson. Dilcher, D. 1969. Podocarpus from the Eocene of North America. Science 164: 299–301. http://dx.doi.org/10.1126/science.164.3877.299 Earle, C.J. 2011. The Gymnosperm Database, version 20 Mar. 2011. http://www.conifers.org/index.php (accessed 14 Apr 2011). Ettingshausen, C.B. von 1879. Report on phyto-palaeontological investigations of the fossil flora of Alum Bay. Proc. Roy. Soc. London 30: 228–236. http://dx.doi.org/10.1098/rspl.1879.0111 Farjon, A. 2001. World checklist and bibliography of conifers, ed. 2. London: Royal Botanic Gardens, Kew. Version of Record (identical to print version). TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England Farjon, A. 2010. A handbook of the World’s conifers. Leiden: E.J. Brill. http://dx.doi.org/10.1163/9789047430629 Ferguson, D.K. 1967. On the phytogeography of Coniferales in the European Cenozoic. Palaeogeogr. Palaeoclimatol. Palaeoecol. 3: 73–110. http://dx.doi.org/10.1016/0031-0182(67)90007-7 Ferguson, D.K., Jähnichen, H. & Alvin, K.L. 1978. Amentotaxus Pilger from the European Tertiary. Feddes Repert. 89: 379–410. http://dx.doi.org/10.1002/fedr.19780890702 Florin, R. 1931. Untersuchungen zur Stammesgeschichte der Coniferales und Cordaitales. Erster Teil: Morphologie und Epidermisstruktur der Assimilationsorgane bei den rezenten Koniferen. Kongl. Svenska Vetensk. Acad. Handl., ser. 3, 10: 1–588. Florin, R. 1940. The Tertiary fossil conifers of South Chile and their phytogeographical significance (with a review of the fossil conifers of southern lands). Kongl. Svenska Vetensk. Acad. Handl., ser. 3, 19: 1–87. Florin, R. 1963. The distribution of conifer and taxad genera in time and space. Acta Horti Berg. 20: 1–312. Fowler, K., Edwards, N. & Brett, D.W. 1973. In situ coniferous (Taxodiaceous) tree remains in the Upper Eocene of southern England. Palaeontology 16: 205–217. Gardner, J.S. 1882. Description and correlation of the Bournemouth Beds – Part II. Lower or Freshwater Series. Quart. J. Geol. Soc. London 38: 1–15. http://dx.doi.org/10.1144/GSL.JGS.1882.038.01-04.02 Gardner, J.S. 1895. The Bournemouth leaf-beds. Proc. Dorset Nat. Hist. Antiq. Field Club 16: 178–184. Gardner, J.S. 1886. A monograph of the British Eocene Floras, vol. 2, Gymnospermae. London: Palaeontological Society. Goloboff, P.A. 1999. NONA (No Name), version 2.0. Published by the author, Tucamán, Argentina. Gordenko, N.V. 2007. A new species of the conifer genus Oswaldheeria with well-preserved leaf anatomical elements from the Bathonian of the Kursk Region. Paleontol. J. 41: 319–326. http://dx.doi.org/10.1134/S0031030107030112 Greenwood, D.R. 1987. Early Tertiary Podocarpaceae: Megafossils from the Eocene Anglesea locality, Victoria, Australia. Austral. J. Bot. 35: 111–133. http://dx.doi.org/10.1071/BT9870111 Harris, T.M. 1979. The Yorkshire Jurassic flora, vol. 5, Coniferales. London: Trustees of the British Museum (Natural History). Hill, C.R., Yang, X., Zhou, Z. & Doyle, J.C. 2012. Exceptionally preserved conifer twigs of Pseudofrenelopsis from the marine Lower Cretaceous of Yorkshire, England. Acta Palaeontol. Sin. 51: 395–410. Hill, R.S. (ed.) 1994. History of the Australian vegetation: Cretaceous to Recent. Cambridge: Cambridge University Press. Hill, R.S. 1995. Conifer origin, evolution, diversification in the Southern Hemisphere. Pp. 10–29 in: Enright, N.J. & Hill, R.S. (eds.), Ecology of the southern Conifers. Melbourne: Melbourne University Press. Hill, R.S. & Brodribb, T.J. 1999. Turner Review No. 2: Southern conifers in time and space. Austral. J. Bot. 47: 639–696. http://dx.doi.org/10.1071/BT98093 Hill, R.S. & Carpenter, R.J. 1989. Tertiary gymnosperms from Tasmania: Cupressaceae. Alcheringa 13: 89–102. http://dx.doi.org/10.1080/03115518908619044 Hill, R.S. & Pole, M.S. 1992. Leaf and shoot morphology of extant Afrocarpus, Nageia, and Retrophyllum (Podocarpaceae) species, and species with similar leaf arrangement, from Tertiary sediments in Australasia. Austral. Syst. Bot. 5: 337–358. http://dx.doi.org/10.1071/SB9920337 Hill, R.S. & Scriven, L.J. 1999. Falcatifolium (Podocarpaceae) macrofossils from Paleogene sediments in south-eastern Australia: A reassessment. Austral. Syst. Bot. 11: 711–720. http://dx.doi.org/10.1071/SB97014 Jin, J., Qiu, J., Zhu, Y. & Kodrul, T. 2010. First fossil record of the genus Nageia (Podocarpaceae) in south China and its phytogeographic implications. Pl. Syst. Evol. 285: 159–163. http://dx.doi.org/10.1007/s00606-010-0267-4 Jongmans, W.J. & Dijkstra, S.J. 1974. Fossilium Catalogus II: Plantae, pars 85, Gymnospermae (Ginkgophyta et Coniferae) VII Pinus quenstedti–Pseudolarixylon (Fossilium Catalogus Plantae). Dordrecht: Kluwer Academic Publishers. Jordan, G.J., Carpenter, R.J., Bannister, J.M., Lee, D.E., Milden­ hall, D.C. & Hill, R.S. 2011. High conifer diversity in Oligo–Miocene New Zealand. Austral. Syst. Bot. 24: 121–136. http://dx.doi.org/10.1071/SB11004 Kelch, D.G. 1997. Phylogeny of the Podocarpaceae based on morphological evidence. Syst. Bot. 22: 113–131. http://dx.doi.org/10.2307/2419680 Kelch, D.G. 2002. Phylogenetic assessment of the monotypic genera Sundacarpus and Manoao (Coniferales: Podocarpaceae) utilising evidence from 18S rDNA sequences. Austral. Syst. Bot. 15: 29–35. http://dx.doi.org/10.1071/SB01002 Keng, H. 1969. Aspects of morphology of Amentotaxus formosana with a note on the taxonomic position of the genus. J. Arnold Arbor. 50: 437–445. King, C. 2006. Paleogene and Neogene: Uplift and a cooling climate. Pp. 395–428 in: Brenchley, P.J. & Rawson, P.F. (eds.), The geology of England and Wales, 2nd ed. London: The Geological Society. Knopf, P., Schulz, C., Little, D.P., Stützel, T. & Stevenson, D.W. 2012. Relationships within Podocarpaceae based on DNA sequence, anatomical, morphological, and biogeographical data. Cladistics 28: 271–299. http://dx.doi.org/10.1111/j.1096-0031.2011.00381.x Krassilov, V.A. 1967. Early Cretaceous flora from south Primorye and its significance for stratigraphy. Moscow: Nauka. [in Russian] Krassilov, V.A. 1974. Podocarpus from the Upper Cretaceous of eastern Asia and its bearing on the theory of conifer evolution. Palaeontology 17: 365–370. Leffingwell, H.A. 1970. Palynology of the Lance (Late Cretaceous) and Fort Union (Paleocene) Formations of the type Lance area, Wyoming. Special Pap. Geol. Soc. Amer. 127: 1–64. Lemoine, S.C. 1972. Structures épidermiques chez Sciadopitys et interprétation des organes. Bull. Soc. Bot. France 119: 61–74. Leopold, E.B., Manchester, S.R. & Meyer, H.W. 2008. Phytogeography of the Late Eocene Florissant flora reconsidered. Pp. 53–70 in: Meyer, H.W. & Smith, D.M. (eds.), Paleontology of the Upper Eocene Florissant Formation, Colorado. Geological Society of America Special Paper 435. Boulder, Colorado: Geological Society of America. 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. Proc. Natl. Acad. Sci. U.S.A. 109: 16217– 16221. http://dx.doi.org/10.1073/pnas.1213621109 Mabberley, D.J. 2008. Mabberley’s plant-book: A portable dictionary of plants, their classifications, and uses, 3rd ed. Cambridge: Cambridge University Press. Mädler, K. 1939. Die pliozäne Flora von Frankfurt am Main. Abh. Senckenb. Naturf. Ges. 446: 1–202. Mao, K., Milne, R.I., Zhang, L., Peng, Y., Liu, J., Thomas, P., Mill, R.R. & Renner, S.S. 2012. Distribution of living Cupressaceae reflects the breakup of Pangea. Proc. Natl. Acad. Sci. U.S.A. 109: 7793–7798. http://dx.doi.org/10.1073/pnas.1114319109 McElwain, J.C. 1998. Do fossil plants signal palaeoatmospheric CO2 concentration in the geological past? Philos. Trans., Ser. B 353: 83–96. http://dx.doi.org/10.1098/rstb.1998.0193 Melikian, A.P. & Bobrov, A.V. 2000. Morfologiya shchenskikh reproduktivi’kh organov i op’t rostroenniya filogeneticheskoj sistemy poryadkov Podocarpales, Cephalotaxales i Taxales. Bot. Zhurn. (Moscow & Leningrad) 85: 50–68. Mill, R.R. & Hill, R.S. 2004. Validations of the names of seven Podocarpaceae macrofossils. Taxon 53: 1043–1046. http://dx.doi.org/10.2307/4135571 Miller, C. 1977. Mesozoic conifers. Bot. Rev. (Lancaster) 43: 217–280. http://dx.doi.org/10.1007/BF02860718 Nichols D.J. & Brown, J.L. 1992. Palynostratigraphy of the Tullock Version of Record (identical to print version). 579 TAXON 62 (3) • June 2013: 565–580 Greenwood & al. • An Eocene Prumnopitys from England Member lower Paleocene of the Fort Union Formation in the Powder River Basin Montana and Wyoming. Bull. U.S. Geol. Surv. 1917: F1–F35, 10 pl. Nixon, K.C. 2004. ASADO, version 1.89. Program and documentation distributed by the author, Ithaca, New York. Nosova, N.V. & Golovneva, L.B. 2010. A new conifer genus Sachatenia gen. nov. from the Upper Cretaceous deposits of the Lena-Vilyui depression (Eastern Siberia). Paleontol. J. 44: 1332–1338. http://dx.doi.org/10.1134/S0031030110100096 Nosova, N.V. & Kiritchkova, A. 2008. A new species and a new combination of the Mesozoic genus Podocarpophyllum Gomolitzky (Coniferales). Paleontol. J. 42: 665–674. http://dx.doi.org/10.1134/S0031030108060129 Nosova, N.V. & Wcisło-Luraniec, E. 2007. A reinterpretation of Mirovia Reymánowna (Coniferales) based on the reconsideration of the type species Mirovia szaferi Reymánowna from the Polish Jurassic. Acta Palaeobot. 47: 359–377. Page, C.N. 1989. New and maintained genera in the conifer families Podocarpaceae and Pinaceae. Notes Roy. Bot. Gard. Edinburgh 45: 377–396. Page, C.N. 1990. Podocarpaceae. Pp. 332–346 in: Kramer, K.U. & Green, P.S. (eds.), The families and genera of vascular plants, vol. 1, Pteridophytes and Gymnosperms. Berlin: Springer . Pole, M.S. 1992a. Eocene vegetation from Hasties, northeastern Tasmania. Austral. Syst. Bot. 5: 431–475. http://dx.doi.org/10.1071/SB9920431 Pole, M.S. 1992b. Early Miocene flora of the Manuherikia Group, New Zealand. 2. Conifers. J. Roy. Soc. New Zealand 22: 287–302. http://dx.doi.org/10.1080/03036758.1992.10420822 Pole, M.S. 1993. Early Miocene flora of the Manuherikia Group, New Zealand. 9. Miscellaneous leaves and reproductive structures. J. Roy. Soc. New Zealand 23: 345–391. Pole, M.S. 1997a. Paleocene plant macrofossils from Kakahu, South Canterbury, New Zealand. J. Roy. Soc. New Zealand 27: 371–400. http://dx.doi.org/10.1080/03014223.1997.9517544 Pole, M.S. 1997b. Miocene conifers from the Manuherikia Group, New Zealand. J. Roy. Soc. New Zealand 27: 355–370. http://dx.doi.org/10.1080/03014223.1997.9517543 Pole, M.S. 1998. Paleocene gymnosperms from Mount Somers, New Zealand. J. Roy. Soc. New Zealand 28: 375–403. http://dx.doi.org/10.1080/03014223.1998.9517571 Pole, M.S. 2007. Early Eocene dispersed cuticles and mangrove to rainforest vegetation at Strahan–Regatta Point, Tasmania. Palaeontol. Electronica 10(3): 15A. http://palaeo-electronica.org/2007_3/126/126.pdf Pole, M.S. & Moore, P.R. 2011. A late Miocene leaf assemblage from Coromandel Peninsula, New Zealand, and its climatic implications. Alcheringa 35: 103–121. http://dx.doi.org/10.1080/03115518.2010.481829 Rai, H.S., Reeves, P.A., Peakall, R., Olmstead, R.G. & Graham, S.W. 2008. Inference of higher-order conifer relationships from a multi-locus plastid data set. Botany 86: 658–669. http://dx.doi.org/10.1139/B08-062 Reichgelt, T., Parker, W.G., Martz, J.W., Conran, J.G., Van Konij­ nenburg­van Cittert, J.H.A. & Kürschner, W.M. 2013. The palynology of the Sonsela Member (Late Triassic, Norian) at Petrified 580 Forest National Park, Arizona, USA. Rev. Palaeobot. Palynol. 189: 18–28. http://dx.doi.org/10.1016/j.revpalbo.2012.11.001 Rydin, C., Källersjö, M. & Friis, E.M. 2002. Seed plant relationships and the systematic position of Gnetales based on nuclear and chloroplast DNA: Conflicting data, rooting problems, and the monophyly of conifers. Int. J. Pl. Sci. 163: 197–214. http://dx.doi.org/10.1086/338321 Schindehütte, G. 1907. Die Tertiärflora des Basalttuffes vom Eichelskopf bei Homberg (Bez. Kassel). Abh. Preuss. Geol. Landesanst. 54: 1–81. Sinclair, W.T., Mill, R.R., Gardner, M.F., Woltz, P., Jaffré, T., Preston, J., Hollingsworth, M.L., Ponge, A. & Möller, M. 2002. Evolutionary relationships of the New Caledonian heterotrophic conifer, Parasitaxus ustus (Podocarpaceae), inferred from chloroplast trnL–F intron/spacer and nuclear rDNA ITS2 sequences. Pl. Syst. Evol. 233: 79–104. http://dx.doi.org/10.1007/s00606-002-0199-8 Stockey, R. & Frevel, B.J. 1997. Cuticle micromorphology of Prumnopitys Philippi (Podocarpaceae). Int. J. Pl. Sci. 158: 198–221. http://dx.doi.org/10.1086/297431 Stockey, R.A. & Ko, H. 1988. Cuticle micromorphology of some New Caledonian podocarps. Bot. Gaz. 149: 240–252. http://dx.doi.org/10.1086/337712 Stockey, R.A., Frevel, B.J. & Woltz, P. 1998. Cuticle micromorphology of Podocarpus, subgenus Podocarpus, section Scytopodium (Podocarpaceae) of Madagascar and South Africa. Int. J. Pl. Sci. 159: 923–940. Sun, G., Zheng, S., Dilcher, D.L., Wang, Y. & Mei, S. 2001. Early angiosperms and their associated plants from western Liaoning, China. Shanghai: Shanghai Scientific and Technical Publishing House. Taggart, R.E. 1973. Additions to the Miocene Sucker Creek flora of Oregon and Idaho. Amer. J. Bot. 60: 923–928. http://dx.doi.org/10.2307/2441074 Townrow, J.A. 1965. Notes on some Tasmanian pines. 1. Some lower Tertiary podocarps. Pap. & Proc. Roy. Soc. Tasmania 99: 87–107. Unger, F. 1850. Die fossile Flora von Sotzka. Vienna: Kaiserlichkönigliche Hof- und Staatsdruckerei. Also published as Denkschr. Kaiserl. Akad. Wiss. Math.-Naturwiss. Kl. 2: 135–197 (1851). Unger, F. 1860. Sylloge plantarum fossilium. Sammlung fossiler Pflanzen besonders aus der Tertiär-Formation. Denkschr. Kaiserl. Akad. Wiss. Math.-Naturwiss. Kl. 19: 1–48. Volynets, E.B. 2009. Diversity of Cretaceous gymnosperms in the Alchan Depression of Primorye. Paleontol. J. 43: 1339–1350. http://dx.doi.org/10.1134/S0031030109100153 Wells, P.M. & Hill, R.S. 1989. Fossil imbricate-leaved Podocarpaceae from Tertiary sediments in Tasmania. Austral. Syst. Bot. 2: 387– 423. http://dx.doi.org/10.1071/SB9890387 Wilkinson, H.P. 1979. The plant surface (mainly leaf). Pp. 97–165 in: Metcalfe, C.R. & Chalk, L. (eds.), Anatomy of the dicotyledons, 2nd ed. Oxford: Clarendon Press. Yang, X.­J., Guignard, G., Thévenard, F., Wang, Y.­D. & Barale, G. 2009. Leaf cuticle ultrastructure of Pseudofrenelopsis dalatzensis (Chow et Tsao) Cao ex Zhou (Cheirolepidiaceae) from the Lower Cretaceous Dalazi Formation of Jilin, China. Rev. Palaeobot. Palynol. 153: 8–18. http://dx.doi.org/10.1016/j.revpalbo.2008.06.002 Version of Record (identical to print version).