Botanical Journal of the Linnean Society, 2009, 161, 422–435. With 4 figures Silica bodies and their systematic implications in Pteridaceae (Pteridophyta) boj_1012 422..435 MICHAEL SUNDUE* The New York Botanical Garden, 200th St. and Southern Blvd., Bronx, NY 10458-5126, USA Received 12 August 2009; accepted for publication 15 October 2009 Wet ashing was used to study the occurrence of silica bodies in the fern family Pteridaceae. They were recovered in 48 of the 77 species examined. Silica bodies of Pteridaceae are elongate, ranging from 90–1320 ¥ 5–40 mm, linear to elliptic, with blunt or acute apices and smooth to sinuate sides. All previous records of silica bodies and venuloid idioblasts among Pteridaceae that were examined were confirmed by the results of this study, corroborating our assumptions regarding the presence of silica bodies. In contrast, assumptions regarding the absence of silica bodies were incorrect; in many species of Adiantum, for example, silica bodies are present but cannot be seen with the naked eye. Farris optimization demonstrates that the distribution of epidermal silica bodies is homoplastic within Pteridaceae, but that they act as a potential synapomorphies for several different groups within the family. These include the adiantoid clade: Adiantum and the 11 vittaroid genera, and in some pteridoid fern clades such as the sister pair Onychium and Actiniopteris and the genus Pityrogramma. They are also present in Pterozonium brevifrons and some species of the polyphyletic genus Pteris. Among cheilanthoid ferns, they were found in Mildella intramarginalis and two species of Aspidotis. Morphology of silica bodies differs between major lineages, reflecting their independent origins. © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435. ADDITIONAL KEYWORDS: Actiniopteris – Adiantum – Aspidotis – Onychium – phytoliths – Pityrogramma – Pteris – SiO2 – venuloid idioblasts – wet ashing. INTRODUCTION Silica (SiO2-nH2O) can be deposited throughout the shoot and root systems of plants, sometimes in large quantities (Kaufman et al., 1981; Hodson et al., 2005; Piperno, 2006). Plant silica has been demonstrated to occur within and upon epidermal cell walls (Davis, 1987) and as discrete isotropic cellular inclusions usually referred to as phytoliths or silica bodies. Silica bodies are known to occur at least sporadically within all major groups of vascular plants (Piperno, 1988, 2006), but have been most intensively studied in the monocotyledons where they occur within commelinids and Orchidaceae (Prychid, Rudall & Gregory, 2004). Epidermal silica bodies have proved useful for diagnostic and systematic purposes in monocotyledons (Dahlgren & Clifford, 1982; Prychid et al., 2004 and citations therein) but remain to be *E-mail: msundue@nybg.org 422 studied in many other groups. Silica bodies are evaluated here as a systematic character in the fern family Pteridaceae. Taxonomically diagnostic silica bodies have been reported in the following ferns and lycophytes: Selaginella P.Beauv. (Gibson, 1893; Bienfait & Waterkeyn, 1974; Piperno, 1988), Isoëtes L. (Iriarte & Alonso Paz, 2009) Equisetum L. (Kaufman et al., 1971), Marattiaceae (Rolleri, 2002, Lavalle, 2005; 2007), Trichomanes L. (Mettenius, 1864; Piperno, 2006), Anemia Sw. (Ribeiro et al., 2007) and Pteridaceae. The latter have a high incidence of epidermal silica bodies. Their presence has been cited as a synapomorphy for the vittarioid ferns (Stevenson & Loconte, 1996), but no systematic survey of their distribution has been conducted for either the vittarioids or Pteridaceae. Within Pteridaceae, silica bodies have been reported from Afropteris Alston (Tryon, Tryon & Kramer, 1990), Pteris L. (Tryon et al., 1990), Adiantum L. (Poirault, 1893) and the vittarioid © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 SILICA BODIES IN PTERIDACEAE (PTERIDOPHYTA) genera (often treated as Vittariaceae), including Ananthacorus Underw. & Maxon, Anetium Splitg., Antrophyum Kaulf., Haplopteris C.Presl, Hecistopteris J.Sm., Monogramma Comm. ex Schkuhr, Polytaenium Desv., Radiovittaria (Benedict) E.H.Crane, Rheopteris Alston, Scoliosorus T.Moore and Vittaria Sm. (Poirault, 1893; von Goebel, 1896; Benedict, 1911; von Goebel, 1924; Williams, 1927; Bower, 1928; Ogura, 1972; Chandra, 1976; Kramer, 1990). Several other authors have also reported thickened epidermal cells within Pteridaceae that may in fact be silica bodies, but they were not identified as such at the time. For example, Nayar (1962) illustrated ‘thickened epidermal’ cells in nine species of Adiantum and Wagner (1978) described ‘venuloid idioblasts’ from the epidermis of eight species of Pteris. Anatomical work has been carried out by Graçano, Azevedo & Prado (2001) who identified ‘idioblasts’ in two species of Pteris and 12 species of Adiantum using cleared leaves and thin sections to determine the position of idioblasts within the leaf tissue. Following the terminology of Wagner (1978), the visibility of elongate epidermal idioblasts has been used as a taxonomic character to distinguish species in Adiantum (Moran, Zimmer & Jermy, 1995; Prado & Lellinger, 2002; Prado & Smith, 2002; Mickel & Smith, 2004; Prado & Sundue, 2005; Sundue & Prado, 2005; Prado, 2006). Similarly, visible epidermal ‘spicular’ cells were cited as a distinguishing character in the protologue of Cheilanthes spiculatum Mickel (Mickel & Smith, 2004). Unfortunately, these studies did not employ a histological test to confirm that the cell contents are in fact silica, leaving the identity of spicular cells and idioblasts ambiguous and potentially confused with collenchymatous structures which are usually referred to as ‘false veins’ or ‘pseudoveins,’ in the literature (e.g. Asplenium L., Chaerle & Viane, 2004 and references therein). Recently, Kao et al. (2008) employed histochemical staining and scanning electron microscopy (SEM) with energy dispersive X-ray spectrometry (EDS) in Pteris grevilleana Wall. ex J.Agardh to confirm that Wagner’s idioblasts are indeed silica bodies. Thus, both directly and indirectly, a body of work has grown supporting the idea that silica bodies are valuable for distinguishing species and genera. However, the question of whether all ‘spicular cells’ and ‘venuloid idioblasts’ are in fact composed of silica remains to be addressed. Moreover, it is not clear whether silica bodies are present in plants when they cannot be seen because they are obscured by other tissues in the leaf. Rigorous histology and increased taxon sampling are needed to address these questions and to determine the distribution of silica bodies in Pteridaceae. Pteridaceae are a large cosmopolitan family that inhabit a broad spectrum of habitats from mangroves 423 at sea-level to alpine habitats at 5000 m (Prado et al., 2007). With 950 species distributed in c. 50 genera (Smith et al., 2006), the family constitutes about 10% of the extant diversity within leptosporangiate ferns (Schuettpelz et al., 2007) and exhibits a wide breadth of morphological diversity and growth habit. Pteridaceae include both very large ferns (e.g. Acrostichum L.) and some of the smallest (e.g. Hecistopteris, Monogramma) and they inhabit a wide range of ecological habitats including aquatic, terrestrial and epiphytic (Tryon et al., 1990). There are three morphological characters that may act as synapomorphies for the family: x = 30, the loss of a true indusium and having sporangia spread along the veins or submarginal commissure, but these remain to be examined thoroughly. Several molecular phylogenetic studies supported the monophyly of Pteridaceae (Hasebe et al., 1995; Gastony & Rollo, 1995; Zhang, Zhang & Chen, 2005; Prado et al., 2007; Schuettpelz et al., 2007), and two (Prado et al., 2007; Schuettpelz et al., 2007) resolved five major monophyletic clades within the family: a cryptogrammatoid clade (three genera/23 spp.), a ceratopteridoid clade (two genera/six spp.), a pteridoid clade (17 genera/400 spp.), an adiantoid clade (12 genera/300 spp.) and a cheilanthoid clade (20 genera/400 spp.). Morphological character support for these clades is lacking. Furthermore, the circumscription of genera is particularly problematic, with several genera known to be para- or polyphyletic (e.g. Cheilanthes Sw., Doryopteris J.Sm., Jamesonia Hook. & Grev., Pellaea Link, Notholaena R.Br., Pteris) (Gastony & Rollo, 1995; Sanchez-Baracaldo, 2004; Schuettpelz et al., 2007; Rothfels et al., 2008), suggesting that a re-evaluation of morphological characters will be crucial to circumscribe and distinguish monophyletic genera satisfactorily. To this end, the distribution of silica bodies is investigated as a novel source of character data for this morphologically heterogeneous family of ferns. MATERIAL AND METHODS SAMPLING The sampling included 148 species representing 39 out of the c. 50 genera of Pteridaceae (Table 1). Of these, 88 species representing 36 genera were subjected to wet ashing. Species not subjected to wet ashing were examined with a dissecting microscope. Each of the five major clades (sensu Schuettpelz et al., 2007) within the family was represented. Sampling was directed toward the adiantoid ferns in which a high incidence of silica bodies has been recorded. Seven of the 11 adiantoid genera were sampled and, within the genus Adiantum, 33 of the c. 200 species were subjected to wet ashing. An additional 32 species © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 424 M. SUNDUE Table 1. Species subjected to wet ashing in this study and corresponding voucher information. All vouchers are housed at NY. Length is reported for species where silica bodies are present Taxon Length Voucher: Collector Locality Acrostichum danaeifolium Langsd. & Fisch. Actiniopteris australis Link Adiantopsis radiata (L.) Fée Adiantum abscissum Schrad. Adiantum adiantoides (J.Sm.) C.Chr. Adiantum andicola Liebm. Adiantum capillus-veneris L. Adiantum capillus-junonis Rupr. Adiantum caryotideum H.Christ Adiantum caudatum L. Adiantum concinnum Humb. & Bonpl. ex Willd. Adiantum delicatulum Mart. Adiantum digitatum Hook. Adiantum formosum R.Br. Adiantum hispidulum Sw. Adiantum jordanii C.H.Mull Adiantum krameri B.Zimmer Adiantum leprieurii Hook. Adiantum lunulatum Burm.f. Adiantum macrophyllum Sw. Adiantum malesianum J.Ghatak Adiantum patens Willd. Adiantum peruvianum Klotzsch Adiantum poiretii Wikstr. Adiantum polyphyllum Willd. Adiantum pyramidale (L.) Willd. Adiantum raddianum C.Presl Adiantum raddianum C.Presl Adiantum reniforme L. Adiantum tenerum Sw. Adiantum tetraphyllum Humb. & Bonpl. ex Willd. Adiantum tomentosum Klotzsch Adiantum trapeziforme L. Adiantum tricholepis Fée Adiantum venustum D.Don Adiantum vogellii Mett. ex Keys Afropteris repens Alston Aleuritopteris argentea (S.G.Gmel.) Fée Aleuritopteris aurantiacea (Cav.) Ching Aleuritopteris farinosa (Forssk.) Fée Ananthacorus angustifolius (Sw.) Underw. & Maxon Anetium citrifolium (L.) Splitg. Anogramma leptophylla (L.) Link – 530–800 mm – 700 mm 175–700 mm 160 mm 880 mm 265–700 mm 660–1320 mm 265–1055 mm 880 mm 1230 mm – 90 mm 350 mm – 130–350 mm 175–615 mm 1230 mm 90–615 mm 880 mm 880 mm 220–615 mm – 1230 mm 265–880 mm 865 mm 705 mm – ~300 mm 530–1055 mm 615 mm 970 mm 350–700 mm 700 mm 1320 mm 175–660 mm – – – 790 mm 880 mm – Nee & Taylor 29158 Rodin 3664 Hernandez 1785 Wacket 212 Heald et al. 9 Mickel 7024 Kruckegeberg 3518 H. Smith 10256 Callejas et al. 4782 Maxwell 12 July 1972 s.n. Fay & Fay 4530 I. Vargas 3326 Culquicondor & Patino 678 Clemens s.n. 1947 Michel & Risler 1828 Montgomery & Huttleson 83-18 de Granville 6510 McDowell 2680 Mickel 1293 Campos & Lopez 4927 Topping 1619 Fay & Fay 3535 Camp 1510 Sundue 845 Paixão et al. 72 Watson et al. 1304 Molon 8458 Woythowski 6434 J.F. Casas 3050 Maas & Zanoni 6407 I. Vargas 4124 Prance et al. 8428 Hallberg 1626 Nee & Atha 47163 Kingdon-Ward 19016 Fay 1028 Balslev 641 Boufford et al. 28585 Tejero-Diaz 2205 Tejero-Diaz 2102 Irwin et al. 54781 Steyermark 87515 Mickel & Leonard 4710a Mexico South Africa Mexico Brazil French Guiana Mexico USA China Colombia Sri Lanka Ecuador Bolivia Peru Australia Australia Mexico French Guiana Guyana Mexico Peru Philippines Ecuador Ecuador Bolivia Brazil Dominican Republic Brazil Peru Canary Islands Dominican Republic Bolivia Bolivia Mexico Mexico India Sierra Leone Seychelles Islands China Mexico Mexico Suriname Venezuela Mexico Anopteris hexagona (L.) C.Chr. Antrophyum callifolium Blume Argyrochosma limitanea (Maxon) Windham Aspidotis californica (Hook.) Nutt. ex Copel. Aspidotis densa (Brack) Lellinger Astrolepis sinuata (Lag. ex Sw.) D.M.Benham & Windham Austrogramme decipiens (Mett.) Hennipman Bommeria hispida (Mett. ex Kuhn) Underw. Ceratopteris pteridoides (Hook.) Hieron. Ceratopteris richardii Brongn. Cheilanthes alabamensis (Buckley) Kunze Cheilanthes angustifolia Kunth Cheilanthes cuneata Kaulf. ex Link Cheilanthes decomposita (M.Martens & Galeotti) Fée Cheilanthes eatonii Baker Cryptogramma cripsa (L.) R.Br. ex Hook. Doryopteris ludens (Wall ex Hook.) J.Sm. Doryopteris sagittifolia J.Sm. Eriosorus hispidulus (Kunze) Vareschi Hecistopteris pumila (Spreng.) J.Sm. Hemionitis rufa (L.) Sw. Jamesonia verticalis Kunze – 280–660 mm – ~200 mm ~350 mm – – – – – – – – – – – – – – 1230 mm – – Sanchez et al. 4 K.N. Chun & C.L. Tso 43921 T.R. Pray 3229 Ross 6409 Ahart 10937 Diaz Vilchis 14 Franc 336 Holmgren 6777 S.R. Hill 13054 Proctor 38368 Siegler 960 Cowen 5629 Mickel & Leonard 6801 Fuentes 447 Washington 13274 Nakhutsrishvili 158 Williams 1501 H.D. Clarke 4909 Sanchez-Baracaldo & Cogollo 267 Boom 4403 Churchill 16772 Vasco & Sundue 649 Puerto Rico China USA USA USA Mexico New Caledonia USA Brazil Jamaica USA Mexico Mexico Mexico USA Georgia Philippines Guyana Colombia Bolivia Colombia Colombia © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 SILICA BODIES IN PTERIDACEAE (PTERIDOPHYTA) 425 Table 1. Continued Taxon Length Voucher: Collector Locality Llavea cordifolia Lag. Mildella intramarginalis (Kaulf. ex Link) Trevis. Neurocalis praestantissima Bory ex Fée Onychium japonicum Blume Onychium siliculosum (Desv.) C.Chr. Pentagramma triangularis (Kaulf.) Yatsk., Windham & E.Wollenw. Pityrogramma austroamericana Domin Pityrogramma calomelanos (L.) Link Pityrogramma ebenea (L.) Proctor Pityrogramma trifoliata (L.) R.M. Tryon Platyzoma microphyllum R.Br. Polytaenium guayanense (Hieron.) Alston Polytaenium lineatum (Sw.) J.Sm. Pteris arborea L. Pteris argyraea T.Moore Pteris mulitifida Poir. Pteris propinqua J.Agardh Pteris quadriaurita Retz. Pteris ryukyuensis Tagawa Pteris tremula R.Br. Pteris vittata L. – 565 mm – ~615 mm ~615 mm – Rzedowski 568 Soto Nunze, J.C. 9700 Diaz & Nino 237 Bracelin 1648 Topping 862 Christopher et al. 138 Mexico Mexico Venezuela Cultivated material Philippines USA ~630 mm ~430 mm ~200 mm ~270 mm – 350 mm 310–880 mm – – 360 mm – – 450 mm – – Bolivia Peru Peru Bolivia Australia Venezuela Dominican Republic Trinidad Cultivated material China Brazil Mexico Japan New Zealand China Pterozonium brevifrons (A.C.Sm.) Lellinger Scoliosorus ensiformis (Hook.) T.Moore Vittaria graminifolia Kaulf. ~330 mm 440 mm 175–880 mm Sundue et al. 595 Revilla 3054 Hutchinson 1387 Williams et al. 655 Brown & Doust s.n. 1992 Leisner 25464 Mickel 8361 Jenman s.n. 1898 Anon. 1874 X.Q. Wang 101 Daly et al. 11475 Mickel 7282 Mimono et al. 2769 Anon., Auckland June 1848 Sino American Guizhou Botanical Expedition 559 Redden 1326 Breedlove 29880 I. Vargas 2996 of Adiantum and 27 vittarioid ferns were examined with the dissecting microscope. Samples of Adiantum included both species with and without visible ‘venuloid idioblasts’ (sensu Prado & Lellinger, 2002; Prado & Smith, 2002; Prado & Sundue, 2005; Sundue & Prado, 2005; Prado, 2006). Among pteridoid ferns, 13 of the 17 genera were sampled, including eight species of the polyphyletic genus Pteris. Within Pteris, two of the species studied by Wagner (1978) were included to verify the identity of his ‘venuloid idioblasts’ as silica bodies. Sampling among cryptogrammatoid ferns included Cryptogramma R.Br. and Llavea Lag., two of the three genera. Both ceratopteridoid genera, Acrostichum and Ceratopteris Brongn., were sampled. Among cheilanthoid ferns, 20 species representing 12 out of the 20 currently recognized genera were sampled. All vouchers are housed at NY (Table 1). WET ASHING Silica bodies were extracted from leaf fragments that had been removed from herbarium specimens. Samples included a few square centimetres of laminar tissue, including veins, costae and leaf margins. Methods for wet ashing followed Piperno (1988) using sulphuric acid followed by chromium trioxide, or 30% hydrogen peroxide (H202), and a rinse in hydrogen chloride (HCl) to remove any carbonates. Silica bodies were transferred to a vial of 70% ethanol Guyana Mexico Bolivia for permanent storage. This method removes all plant tissues except silica, eliminating possible confusion of silica bodies with either calcium carbonate or calcium oxalate crystals. Consequently, the position of silica bodies within the leaf is lost and must be determined separately. POSITION After the presence of silica bodies was diagnosed through wet ashing, these same structures could often be relocated on live plants and herbarium specimens using a binocular dissecting microscope. This was attempted for the specimens used in this study. When present between the veins in the lamina, silica bodies were easily discernible at 60¥ magnification. In both living and dried leaves, intervenal silica bodies appeared either as dark or lustrous prominulous striations in the lamina surface. In dried laminae they often appear as prominulous striations between the veins; the drying and shrinking of surrounding tissue accentuates their presence. Silica bodies that have been determined to be present by wet ashing, but which are not intervenal, cannot be seen with a dissecting microscope. These are likely above the veins of the leaf as demonstrated by Graçano et al. (2001), but anatomical study is needed to determine their position. Consequently, silica body position is determined here simply as intervenal or not intervenal and it should not be assumed that plants with © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 426 M. SUNDUE intervenal silica bodies do not also have nonintervenal silica bodies residing elsewhere in the leaf. This is not revealed by morphology either, as silica bodies are usually monomorphic regardless of their position in the leaf. CHARACTER STATE RECONSTRUCTION The presence and position of epidermal silica bodies was scored as present or absent and optimized onto the three-gene phylogenetic tree of Schuettpelz et al. (2007) using Farris optimization (Farris, 1970), a parsimony algorithm for ordered state characters. Presence was determined either by wet ashing or visual inspection, whereas absence was determined only from the results of wet ashing. Position, interpreted as a binary character, was scored as being either intervenal or not, as determined by visual inspection. All characters were scored from specimens except those for Rheopteris cheesmanii Alston, which were determined from the literature (Kramer, 1990). Two characters were coded as follows: 1. Epidermal silica bodies (0) absent (1) present. 2. Position of silica bodies in lamina (0) not intervenal (1) intervenal. RESULTS Silica bodies were generally indurate, elongate, several times longer than wide (90–1320 ¥ 5–40 mm), with either smooth or undulate sides, flat or slightly canaliculate upper surfaces, flat or rounded lower surfaces and acute or blunt apices (Figs 1–3). Silica bodies were singular (Fig. 1B–L) or consisted of a few bodies joined together end to end or side by side forming sheets (Fig. 1A, Fig. 2B, E, G, Fig. 3B, C). Some of the thinner silica bodies became fragmented during the digestion and recovery process, obscuring the details of size and shape. Silicified 2- and 3-celled hairs were also recovered in Adiantum hispidulum Sw. Occasionally, some plant silica was also recovered that appeared to be from leaf parenchyma or vascular tissue, which was outside the scope of this study. Silica bodies were confirmed to be present and to occur between the veins in the laminae for 60 species by visual inspection (see also Supporting Information, Appendix); 55 of these species were not subjected to wet ashing. Of the samples subjected to wet ashing, silica bodies could be seen in 21 species. In other cases, the position of the silica bodies in the lamina could not be determined from herbarium specimens. Epidermal silica bodies were absent in the cryptogrammatoid and ceratopteridoid clades but recovered by wet ashing in 51 out of the 88 species representing the following taxa: cheilanthoid (Aspidotis (Nutt. ex Hook.) Copel., Mildella Trevis.), pteridoid (Actiniopteris Link, Afropteris, Onychium Kaulf., Pityrogramma Link, Pteris, Pterozonium Fée) and vittarioid (Adiantum, Ananthacorus, Anetium, Antrophytum, Hecistopteris, Polytaenium, Scoliosorus, Vittaria) (Table 1). ADIANTOID CLADE Silica bodies were recovered in 37 of the 41 adiantoid ferns that were subjected to wet ashing (Table 1). They are present in all of the sampled vittarioid ferns and absent in only five species of Adiantum. Silica bodies of Adiantum ranged 160–1320 ¥ 12–40 mm and were elongate, with sinuately lobed or shallowly sinuate-lobed sides. In most cases the silica bodies were singular, but in A. raddianum C.Presl (Fig. 1A) several silica bodies are joined side by side along the long axis. Silica bodies could be seen occurring between the veins on herbarium specimens in 37 species (see also Supporting Information, Appendix), of which A. caryotideum H.Christ, A. macrophyllum Sw. (Fig. 1B), A. tetraphyllum Humb. & Bonpl. ex Willd., A. tomentosum Klotzsch, A. venustum D.Don and A. vogellii Mett. ex Keys were also examined subjected to wet ashing. In the remaining species of Adiantum, silica bodies could not clearly be seen in the lamina tissue of herbarium specimens. Silica bodies of the vittarioid ferns (Fig. 1E–G, I–L) varied in length from 175 to 1230 mm and had sinuately lobed or shallowly lobed sides. They were recovered in all samples subjected to wet ashing. Vittarioid taxa included in this study have silica bodies occurring between the veins on both sides of the lamina. However, sometimes silica bodies are restricted toward the edge of the lamina, on the adaxial surface, in Polytaenium guayanense (Hieron.) Alston (Fig. 1J), and on both surfaces in Scoliosorus ensiformis (Hook.) T. Moore (Fig. 1G). Silica bodies were observed in the lamina of herbarium specimens in an additional 28 species (see also Supporting Information, Appendix). Although Ogura (1972: 395) reported that silica bodies were absent from Monogramma, they are clearly visible in herbarium specimens (see also Supporting Information, Appendix). Thus, it appears that silica bodies are present in all genera of vittarioid ferns. PTERIDOID CLADE Silica bodies were recovered in Afropteris, Actiniopteris, Onychium, Pityrogramma, Pterozonium and some species of Pteris. Silica bodies of Afropteris repens Alston (Fig. 2F) measured 176–660 mm long and had undulate sides with deep sinuses. These were clearly visible on herbarium specimens as lines © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 SILICA BODIES IN PTERIDACEAE (PTERIDOPHYTA) 427 Figure 1. Adiantoid silica bodies. A, Adiantum raddianum. B, Adiantum macrophyllum. C, Adiantum tetraphyllum. D, Adiantum hispidulum. E, Antrophyum callifolium. F, Hecistopteris pumila. G, Scolisorus ensiforme. H, Adiantum krameri. I, Ananthacorus angustifolius. J, Polytaenium guyanense. K, Polytaenium lineatum. L, Vittaria graminifolia. Scale bars, 100 mm. that occur between veins on both sides of the lamina, along the margin of the lamina and along veins on the abaxial surface. Silica bodies recovered from Onychium japonicum Blume (Fig. 2C) and O. siliculosum (Desv.) C.Chr. were elongate, up to 620 mm long, with undulate sides. The position of silica bodies within the epider- mis could not be determined with the methods used. Actiniopteris australis Link (Fig. 2G) yielded groups of elongate silica bodies, joined side by side and end to end, with either flat or papillate surfaces. In herbarium specimens, the epidermis of this species appears uniformly shiny and striate on green leaves, or whitish and opaque on dead or senescent © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 428 M. SUNDUE in herbarium specimens, but individual silica bodies could not be distinguished. Within Pteris, silica bodies were found in P. ryukyuensis Tagawa (Fig. 2A) (450 mm) and P. multifida Poir. (Fig. 2D) (360 mm), but not in P. arborea L., P. argyraea T.Moore, P. propinqua J.Agardh, P. quadriaurita Retz., P. tremula R.Br. or P. vittata L. Silica bodies of P. ryukyuensis and P. multifida have denticulate sides and blunt apices. Some silica bodies of P. ryukyuensis were attached to thin, nondescript jigsaw-shaped epidermal silica (Fig. 2A). In both of these species, the silica bodies were visible in herbarium specimens between veins and along the pale and scarious leaf margin. Silica bodies were absent from the other pteridoid taxa that were sampled (Anogramma leptophylla (L.) Link, Anopteris hexagona (L.) C.Chr., Eriosorus hispidulus (Kunze) Vareschi, Austrogramme decipiens (Mett.) Hennipman, Jamesonia verticalis Kunze and Platyzoma microphylla R.Br.). CHEILANTHOID Figure 2. Pteroid silica bodies. A, Onychium japonicum. B, Actiniopteris australis. C, Pityrogramma trifoliata. D, Pityrogramma calomelanos. E, Pteris multifida. F, Pteris ryukyuensis. G, Afropteris repens. Scale bars, 100 mm. leaves, but individual silica bodies could not be distinguished. Pityrogramma austroamericana Domin (c. 630 mm long), P. calomelanos (L.) Link (Fig. 2E) (c. 430 mm long), P. ebenea (L.) Proctor (c. 200 mm long) and P. trifoliata (L.) R.M.Tryon (Fig. 2B) (c. 270 mm long) have elongate silica bodies joined side by side and end to end, with smooth or slightly irregular sides and blunt apices. These were also joined to nondiagnostic, thin, jigsaw-puzzle-shaped epidermal silica (as described by Piperno, 2006: 42). Leaf surfaces of Pityrogramma appear shiny and striate CLADE Epidermal silica bodies were recovered in Mildella intramarginalis (Kaulf. ex Link) Trevis., Aspidotis californica (Hook.) Nuttall ex Copel. and A. densa (Brack) Lellinger. Silica bodies of M. intramarginalis (Fig. 3A) were elongate, 500–600 mm long, with undulate or entire margins. Those of Aspidotis californica and A. densa were joined together end to end and side by side but of two distinctly different types, one thin and smooth sided (Fig. 3B), the second thick with shallowly sinuate sides (Fig. 3C). On herbarium specimens, these silica bodies were located on the abaxial side of the leaf, below the veins and along margins of sterile leaf segments. Silica bodies were not found in Adiantopsis radiata (L.) Fée, Aleuritopteris argentea (S.G.Gmel.) Fée, A. aurantiaca (Cav.) Ching, A. farinosa (Forssk.) Fée, Argyrochosma limitanea (Maxon) Windham, Astrolepis sinuata (Lag. ex Sw.) D.M.Benham & Windham, Bommeria hispida (Mett. ex Kuhn) Underw., Cheilanthes alabamensis (Buckley) Kunze, C. angustifolia Kunth, C. cuneata Kaulf. ex Link, C. decomposita (M.Martens & Galeotti) Fée, Doryopteris ludens (Wall ex Hook.) J.Sm., D. sagittifolia J.Sm., Hemionitis rufa (L.) Sw. and Pentagramma triangularis (Kaulf.) Yatsk., Windham & E.Wollenw. DISCUSSION Silica bodies were recovered in all of the species where previously termed ‘idioblasts,’ ‘venuloid idioblasts’ or ‘spicular cells’ could be seen in the lamina of herbarium specimens, confirming that these variously termed epidermal structures are in fact plant silica. © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 SILICA BODIES IN PTERIDACEAE (PTERIDOPHYTA) 429 Figure 3. Cheilanthoid silica bodies. A, Mildella intramarginalis. B, Aspidotis californica. C, Aspidotis densa. Scale bars, 100 mm. Silica bodies were also recovered in some species where they could not be seen and were not anticipated, demonstrating that our assumptions about their absence in some species of Pteridaceae were incorrect and that the complete distribution of silica bodies cannot be determined by studying herbarium specimens alone. Optimization determines the hypothesis which best describes the data under the given criteria (Schuh, 2000). Here, silica characters are not tested against the other characters and not used to influence the topology, but simply mapped to reveal the most parsimonious hypothesis given our understanding of the phylogeny using molecular data. Based on the phylogenetic analysis of Schuettpelz et al. (2007), epidermal silica bodies have evolved at least three times in Pteridaceae (Fig. 4). Their presence acts as a synapomorphy for the adiantoid ferns (Fig. 4; as ‘adiantoid clade’), thus filling a critical gap in knowledge, as no morphological character has been previously demonstrated to unite these two clades. Silica bodies also act as a synapomorphy uniting several smaller clades of pteridoid ferns, in particular Pityrogramma, and the clade including the sister genera Actiniopteris and Onychium. Additional origins of the character are anticipated, given that some taxa found here to have silica bodies, such as Aspidotis and Mildella, are not included in that analysis. Furthermore, the character would become more homoplastic among pteridoid ferns when optimized on trees that more densely sample that clade, such as Prado et al. (2007) or the rbcL-only tree of Schuettpelz et al. (2007). The position of silica bodies, interpreted here as whether they occur between veins of the leaf or not, provides a second set of character data. Whereas the presence of silica bodies appears once in adiantoid ferns, the presence of silica bodies between veins occurs twice in that clade (Fig. 4). Silica bodies also © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 430 M. SUNDUE Doryopteris sagittifolia (0, --) Cheilanthes viridis Adiantopsis radiata (0, --) Hemionitis palmata (H. rufa) (0, --) Aleuritopteris argentea (0, --) Cheilanthes nitidula Pentagramma triangularis (0, --) Notholaena aschenborniana Pellaea intermedia Pellaea truncata Paragymnopteris marantae Astrolepis sinuata (0, --) Argyrochosma limitanea (0, --) Cheilanthes alabamensis (0, --) Cheilanthes eatonii (0, --) Bommeria hispida (0, --) Doryopteris ludens (0, --) Cheilanthoid clade Anetium citrifolium (1, 1) Polytaenium cajenense (P. guayanense, P. lineatum) (1, 1) Vittaria graminifolia (1, 1) Antrophyum latifolium (A. califolium) (1, 1) Haplopteris elongata (1, 1) Monogramma graminea (1, 1) Hecistopteris pumila (1, 1) Radiovittaria gardneriana (1, 1) Rheopteris cheesmaniae (1, 1) Adiantum capillus-veneris (1, 0) Adiantum malesianum (1, 0) Adiantoid clade Adiantum pedatum (1, 0) Adiantum tenerum (1, 0) Adiantum peruvianum (1, 1) Adiantum tetraphyllum (1, 1) Adiantum raddianum (1, 0) Pteridaceae Pteris cretica (0, --) Pteris multifida (1, 1) Pteris propinqua (P. tremula) (0, --) Ochropteris pallens Pteris arborea (0, --) Pteris argyraea (0, --) Pteris quadriaurita (0, --) Neurocallis praestantissima (0, --) Pteris tremula (0, --) Pteridoid Platyzoma microphyllum (0, --) Pteris vittata (0, --) Eriosorus cheilanthoides (E. hispidulus) (0, --) Jamesonia verticalis (0, --) Pterozonium brevifrons (1, 0) Pityrogramma austroamericana (P. calomelanos, P. trifoliata) (1, 0) Actiniopteris dimorpha (A. australis) (1, 0) Onychium japonicum (O. siliculosum) (1, 0) Acrostichum danaeifolium (0, --) Ceratopteris richardii (C. pteroides) (0, --) Coniogramme fraxinea Cryptogramma crispa (0, --) Llavea cordifolia (0 , --) clade Ceratopteridoid clade Coniogrammoid clade Blechnum occidentale Thelypteris palustris Cystopteris reevesiana Asplenium unilaterale Davallia solida Nephrolepis cordifolia Dryopteris aemula Didymochlaena truncatula Dennstaedtia punctilobula Microlepia platyphylla Monachosorum henryi Pteridium esculentum Figure 4. Parsimony character reconstructions as optimized onto the analysis of Schuettpelz et al. (2007). Numbers in parentheses after each terminal are characters states for each of the two characters scored. Presence of silica bodies is indicated by thickened branches. The presence of silica bodies occurring between the veins of the leaf is indicated by the boxes. Taxa in parentheses are placed on the tree based on their position in the single gene (rbcL) analysis of Schuettpelz et al. (2007). Taxa in grey were not examined in this study. © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 SILICA BODIES IN PTERIDACEAE (PTERIDOPHYTA) occur between the veins of Pteris multifida. In retrospect, the distribution of this character is equivalent to our assumptions about the presence of silica bodies in Pteridaceae prior to this study. Morphology of silica bodies appears to differ between lineages within Pteridaceae and may be an additional source of character data. The adiantoid ferns (Fig. 1A–L) can be distinguished from most other Pteridaceae (Fig. 2A–G, Fig. 3A–C) by silica bodies with elliptical shape, tapered ends and undulate sides. Pteridoid ferns generally have blunt ends and sides that are smooth, denticulate sparsely denticulate or shallowly lobed. The only known exception is Afropteris repens, which has deeply sinuately lobed sides that appear more similar to silica bodies of the adiantoid lineage. Silica bodies in cheilanthoid ferns are diverse and cannot be easily generalized at this time. ADIANTOID CLADE Comprising 12 genera and c. 300 species, the adiantoid ferns are a monophyletic group that includes Adiantum plus the 11 genera of vittarioid ferns which have been frequently treated at the family level as Vittariaceae. The presence of silica bodies in the vittarioid ferns has been noted by numerous authors dating back at least to Benedict (1911). Likewise, the similarity between silica bodies in Adiantum and those of the vittarioid ferns has been noted by several authors, but until now this has been considered homoplastic (Nayar, 1962; Wagner, 1978; Tryon & Tryon, 1982; Kao et al., 2008;). The failure to acknowledge silica bodies as a synapomorphy for this clade is probably because of two long-standing misconceptions: (1) that Adiantum is not closely related to the vittarioid ferns; and (2) that silica bodies are not widespread throughout Adiantum. Results here indicate that silica bodies are widespread in Adiantum, even in species where they are hard to see on the lamina surface (e.g. A. capillus-veneris L.). In such species, the silica bodies are probably located above the veins as demonstrated by Graçano et al. (2001) in Adiantum curvatum Kaulf., A. papillosum Handro and A. subcordatum Sw. Within Adiantum, silica bodies are highly uniform, with most species exhibiting elongate, linear silica bodies with sinuately lobed sides (Fig. 1A–E). A few species, such as A. krameri B.Zimmer (Fig. 1H) have shorter silica bodies and shallow lobes. Silica bodies of their sister group, the vittarioid ferns, are generally shorter and exhibit more variation (Fig. 1G, I–L), but those of Hecistopteris pumila (Spreng.) J.Sm. (Fig. 1F) are elongate and indistinguishable from the silica bodies of most Adiantum species. Thus, adiantoid ferns have silica bodies that can be distinguished as a group by their characteristics. 431 The position of silica bodies offers an additional synapomorphy for the vittarioid genera as well as a clade within Adiantum (Fig. 4). Silica bodies were present between veins in all of the vittarioid species examined. In Adiantum they are either present between veins or else presumably above the veins, as demonstrated for the few species subject to anatomical studies (Poirault, 1893; Graçano et al., 2001). PTERIDOID CLADE The pteridoid ferns are a monophyletic lineage that comprises c. 400 species in approximately 17 genera. Much of that diversity resides in Pteris, a polyphyletic genus of c. 250 species (Schuettpelz et al., 2007). Silica bodies were found in P. ryukyuensis (Fig. 2A) and P. multifida (Fig. 2E), but not in P. arborea, P. argyraea, P. propinqua, P. quadriaurita, P. tremula or P. vittata. Both P. ryukuensis and P. multifida were previously cited by Wagner (1978) as having ‘venuloid idioblasts.’ Wagner also cited P. angustipinnata Tagawa, P. cadieri H.Christ, P. grevilleana, P. kidoi Kurata and P. yamatensis Tagawa as bearing ‘venuloid idioblasts.’ These same species were recognized previously as an informal species group by Shieh (1966) under the name ‘subsection 2. cadieri’. Whether these species constitute a monophyletic group within Pteris should be investigated further. Pteris cretica, which was not sampled here, is resolved as sister to P. multifida in the analysis of Schuettpelz et al. (2007). Herbarium specimens of P. cretica L. do not appear to have silica bodies and were placed in a different subsection by Shieh (1966). The remainder of Pteris will require additional sampling to determine the systematic implications of silica bodies. Graçano et al. (2001) reported ‘idioblasts’ as present in P. denticulata Sw. and P. leptophylla Sw. They also reported them absent in P. propinqua, which is in agreement with the results presented here. Silica bodies were absent in Neurocallis praestantissima Bory ex Fée, a monotypic genus nested within Pteris (Schuettpelz et al., 2007) and they were present in Afropteris, which formed a clade with Pteris in the analysis of Sanchez-Baracaldo (2004). Actiniopteris is resolved as sister to Onychium in several molecular phylogenetic studies (Gastony & Johnson, 2001; Prado et al., 2007; Schuettpelz et al., 2007). Although silica bodies are present in samples subjected to wet ashing, they could not be located in herbarium specimens of these genera. The epidermis of all species in both genera is thick and lustrous, which may be because of the presence of silica, but anatomical work is needed to determine this. © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 432 M. SUNDUE Pityrogramma is a genus of c. 19 species. All four of the species studied here had silica bodies. Graçano et al. (2001) also reported ‘idioblasts’ in Pityrogramma calomelanos residing over the veins of the leaf. Pityrogramma is resolved as sister to a portion of the polyphyletic genus Anogramma in the analysis of Schuettpelz et al. (2007). Anogramma leptophylla was included here and found to lack epidermal silica bodies. Silica bodies were absent from the other pteridoid ferns that were sampled, including Anopteris hexagona, Eriosorus hispidulus, Austrogramme decipiens and Platyzoma microphylla. When presence and absence of epidermal silica are optimized onto the phylogenetic tree of Schuettpelz et al. (2007), the presence of silica bodies evolves two times within this clade, but taxon sampling needs to be considered when making this calculation. More densely sampled phylogenetic trees, such as the rbcL-only tree presented in Schuettpelz et al. (2007) which includes taxa not sampled here, and additional taxa found here that lack silica bodies (Anogramma lepotophylla), suggest that additional gains and losses are likely to have occurred within the pteridoid clade. CHEILANTHOID CONCLUSIONS All previous records of silica bodies and venuloid idioblasts were confirmed by the results of this study, corroborating our assumptions regarding the presence of silica bodies. In contrast, assumptions regarding the absence of silica bodies were incorrect; in many species of Adiantum, for example, silica bodies are present but cannot be seen without wet ashing. In these cases, wet ashing is helpful to detect their presence. From the survey conducted here, certain taxonomic implications can be drawn. Ceratopteridoid or cryptogrammatoid ferns appear to lack silica bodies. The presence of silica bodies is a potential synapomorphy for the adiantoid ferns, several smaller clades of pteridoid ferns, including Pityrogramma and the sister group of Actiniopteris and Onychium. They are present in the cheilanthoid ferns and should be investigated as possible synapomorphies for Aspidotis and Mildella. Morphological characters of silica bodies may also be a rich source of character data, particularly if discrete character states can be determined. ACKNOWLEDGEMENTS CLADE Cheilanthoid ferns comprise c. 400 species distributed among approximately 20 genera, several of which are in need of re-circumscription. Among the 14 cheilanthoid ferns sampled here, epidermal silica bodies were recovered in Mildella intramarginalis, Aspidotis californica and A. densa. On herbarium specimens of Mildella intramarginalis, silica bodies can be found on the abaxial side of the leaf, above the veins. They also appear to be present along margins of sterile leaf segments, suggesting that the ‘scarious leaf margin’ used to define Mildella by Hall & Lellinger (1967) may in fact be scarious because it is silicified. Mickel & Smith (2004) suggested that the New World species of Mildella are related to the Cheilanthes angustifolia complex based on shared characters including ‘prominulous veins,’ but silica bodies were not recovered from C. angustifolia in this study. Aspidotis is a tropical American genus of approximately seven species (Tryon et al., 1990), the monophyly of which has not been tested in a phylogenetic analysis. Herbarium specimens examined of both species, and those of A. meifolia D.C.Eaton, have laminae with a lustrous and striate adaxial surface. This was also noted by Lellinger (1968), who used that character as evidence to segregate the genus from Cheilanthes. The results presented here suggest that epidermal silica may be responsible for the lustre of these laminae. I thank Lisa Campbell and Dennis Stevenson for assistance in the Structural Botany Laboratory at The New York Botanical Garden, and Robbin Moran, Jefferson Prado and Alan Smith for discussions regarding venuloid idioblasts in adiantoid ferns. Vinson Doyle, Larry Kelly, Hilary Martin and Natalia Pabon-Mora read early drafts of this manuscript and made significant improvements. Thanks are also due to Dominick Basile who first suggested to me that Wagner’s idioblasts may be composed of silica. Initial work on this project was undertaken in his lab during a CUNY Graduate Center course in plant morphology held at Lehman College. Some of the results of this research were written while the author was a postdoctoral researcher on a grant to Robbin Moran from the US National Science Foundation (DEB 0717056). REFERENCES Benedict RC. 1911. The genera of the fern tribe Vittarieae: their external morphology, venation, and relationships. Bulletin of the Torrey Botanical Club 38: 153–190. Bienfait A, Waterkeyn L. 1974. Contribution a l’étude systématique des Selaginella. Spécificité des formations callosiques foliares observées en fluorescence. Bulletin du Jardin Botanique National de Belgique 44: 295–302. Bower FO. 1928. The ferns (Filicales), Vol. III. New York: Cambridge University Press. © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 SILICA BODIES IN PTERIDACEAE (PTERIDOPHYTA) Chaerle P, Viane RLL. 2004. Leaf anatomy and the occurrence of false veins in Asplenium (Aspleniaceae, Pteridophyta). Botanical Journal of the Linnaean Society 145: 187–194. Chandra S. 1976. Morphology of the sporophyte of the vittarioid fern Ananthacorus. Fern Gazette 11: 247–252. Dahlgren RMT, Clifford HT. 1982. The monocotyledons: a comparative study. New York: Academic Press. Davis RW. 1987. Ultrastructure and analytical microscopy of silicon in the leaf cuticle of Ficus lyrata Warb. Botanical Gazette 148: 318–323. Farris JS. 1970. Methods for computing Wagner trees. Systematic Zoology 19: 83–92. Gastony GJ, Johnson WP. 2001. Phylogenetic placements of Loxoscaphe thecifera (Aspleniaceae) and Actiniopteris radiata (Pteridaceae) based on analysis of rbcL nucleotide sequences. American Fern Journal 91: 197–213. Gastony GJ, Rollo DR. 1995. Phylogeny and generic circumscriptions of cheilanthoid ferns (Pteridaceae: Cheilanthoideae) inferred from rbcL nucleotide sequences. American Fern Journal 85: 341–360. Gibson RJH. 1893. On the siliceous deposit in the cortex of certain species of Selaginella. Annals of Botany 7: 356– 366. von Goebel E. 1896. Archegoniatenstudien. 8. Hecistopteris, ein verkannte Farngattung. Flora 82: 67–75. von Goebel E. 1924. Archegoniatenstudien. XVI. Vittariaceen und Pleurogrammaceen. Flora 117: 91–132. Graçano D, Azevedo AA, Prado J. 2001. Anatomia foliar das espécies de Pteridaceae do Parque Estadual do Rio Doce (PERD) – MG. Revista Brasileira de Botánica 24: 333–347. Hall CC, Lellinger DB. 1967. A revision of the fern genus Mildella. American Fern Journal 57: 113–134. Hasebe M, Wolf PG, Pryer KM, Ueda K, Ito M, Sano R, Gastony GJ, Yokoyama J, Manhart JR, Murakami N, Crane EH, Haufler CH, Hauk WD. 1995. [1996]. Fern phylogeny based on rbcL nucleotide sequences. American Fern Journal 85: 134–181. Hodson MJ, White PJ, Mead A, Broadley MR. 2005. Phylogenetic variation in the silicon composition of plants. Annals of Botany 96: 1027–1046. Iriarte J, Alonso Paz E. 2009. Phytolith analysis of selected native plants and modern soils from southeastern Uruguay and its implications for paleoenvironmental and archeological reconstruction. Quaternary International 193: 99–123. Kao TT, Chen S-J, Chiou W-L, Chuang Y-C, Kuo-Huang L-L. 2008. Various microscopic methods for investigating the venuloid idioblasts of Pteris grevilleana Wall. Taiwania 53: 394–400. Kaufman PB, Bigelow WC, Schmid R, Ghosheh NS. 1971. Electron microprobe analysis of silica of epidermal cells in Equisetum. American Journal of Botany 58: 309–316. Kaufman PB, Dayanandan P, Takeoka Y, Bigelow WC, Jones JD, Iler R. 1981. Silica in shoots of higher plants. In: Simpson TL, Volcani BE, eds. Silicon and silicaceous structures in biological systems. New York: Springer-Verlag, 409–449. Kramer KU. 1990. Vittariaceae. In: Kubitzki K, ed. The 433 families and genera of vascular plants: pteridophytes and gymnosperms. Berlin: Springer-Verlag, 230–256. Lavalle MC. 2005. Arquitectura foliar y otros caracteres del esporófito en especies neotropicales de Marattia Sw. (Marattiaceae-Pteridophyta). Acta Botanica Malacitana 30: 11–30. Lavalle MC. 2007. Characteristico diagnostico foliares en especies neotropicales de Marattia Sw. (MarattiaceaePteridophyta). Annals of the Missouri Botanical Garden 94: 192–201. Lellinger DB. 1968. A note on Aspidotis. American Fern Journal 58: 140–141. Mettenius GH. 1864. Über die Hymenophyllaceae. Abhandlungen bei Begründung der Königlich-Sächsischen Gesellschaft der Wissenschaften, Mathematisch-Physische Klasse 7: 403–504. Mickel JT, Smith AR. 2004. The pteridophytes of Mexico. Memoirs of the New York Botanical Garden 88: 210. Moran RC, Zimmer B, Jermy AC. 1995. Adiantum. In: Moran RC, Riba R, eds. for Pteridophytes. In: Flora Mesoamericana. Volumen 1, Pteridofitas, Psilotaceae a Salviniaceae. México, D.F.: Universidad Nacional Autónoma de México. Nayar BK. 1962. Studies in Pteridaceae V. Contributions to the morphology of some species of the maidenhair ferns. Botanical Journal of the Linnaean Society 58: 185–199. Ogura Y. 1972. Comparative anatomy of vegatative organs of the pteridophytes, 2nd edn. Berlin: Gebrüder Borntraeger. Piperno DR. 1988. Phytolith analysis: an archeological and geological perspective. San Diego, CA: Academic Press. Piperno DR. 2006. Phytoliths: a comprehensive guide for archaeologists and paleoecologists. Lanham, MD: Rowman Altamira. Poirault MG. 1893. Researches anatomiques sur les cryptogames vasculares. Annales des Sciences Naturelles, ser. 7, Botanique 18: 113–256. Prado J. 2006. Three new species of Adiantum (Pteridaceae) from Bolivia and Peru. Brittonia 58: 379–384. Prado J, Lellinger DB. 2002. Adiantum argutum an unrecognized species of the A. latifolium group. American Fern Journal 92: 23–29. Prado J, Rodrigues CDN, Salatino A, Salatino MLF. 2007. Phylogenetic relationships among Pteridaceae, including Brazilian species, inferred from rbcL sequences. Taxon 56: 355–368. Prado J, Smith AR. 2002. Novelties in Pteridaceae from South America. American Fern Journal 92: 105–111. Prado J, Sundue MA. 2005. Typification and identity of Adiantum tetragonum (Pteridaceae). American Fern Journal 95: 89–93. Prychid CJ, Rudall PJ, Gregory M. 2004. Systematics and biology of silica bodies in monocotyledons. Botanical Review 69: 377–440. Ribeiro MLR, Santos MG, Moraes MG. 2007. Leaf anatomy of two Anemia Sw. species (Schizeaceae– Pteridophyte) from a rocky outcrop in Niterói, Rio de Janeiro. Revista Brasileira de Botánica Brazil 30: 695– 702. © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 434 M. SUNDUE Rolleri CH. 2002. Caracteres diagnósticos y taxonomía en el género Angiopteris Hoffm. (Marattiaceae Bercht. & J. S. Presl): I, Los caracteres. Revista del Museo de La Plata, Sección Botánica 15: 23–49. Rothfels CJ, Windham MD, Grusz AL, Gastony GJ, Pryer KM. 2008. Toward a monophyletic Notholaena (Pteridaceae): resolving patterns of evolutionary convergence in xeric-adapted ferns. Taxon 57: 712–724. Sanchez-Baracaldo P. 2004. Phylogenetics and biogeography of the neotropical fern genera Jamesonia and Eriosorus (Pteridaceae). American Journal of Botany 91: 274–284. Schuettpelz E, Schneider H, Huiet L, Windham MD, Pryer KM. 2007. A molecular phylogeny of the fern family Pteridaceae: assessing overall relationships and the affinities of previously unsampled genera. Molecular Phylogenetics and Evolution 44: 1172–1185. Schuh RT. 2000. Biological systematics, principles and applications. Ithaca, NY: Cornell University Press. Shieh W-C. 1966. A synopsis of the fern genus Pteris in Japan, Ryukyu, and Taiwan. Botanical Magazine (Tokyo) 79: 283–293. Smith AR, Pryer KM, Schuettpelz E, Korall P, Schneider H, Wolf PG. 2006. A classification for extant ferns. Taxon 55: 705–731. Stevenson DW, Loconte H. 1996. Ordinal and familial relationships of pteridophyte genera. In: Camus JM, Gibby M, Johns RJ, eds. Pteridology in Perspective. Kew: Royal Botanic Gardens, 435–467. Sundue MA, Prado J. 2005. Adiantum diphyllum, a rare and endemic species of Bahia state, and its close relatives. Brittonia 57: 123–128. Tryon RM, Tryon AF. 1982. Ferns and allied plants, with special reference to tropical America. Berlin: SpringerVerlag. Tryon RM, Tryon AF, Kramer KU. 1990. Pteridaceae. In: Kubitzki K, ed. The families and genera of vascular plants: pteridophytes and gymnosperms. Berlin: Springer-Verlag, 230–256. Wagner WH. 1978. Venuloid idioblasts in Pteris and their systematic implications. Acta Phytotaxonomica et Geobotanica 29: 33–40. Williams S. 1927. A critical examination of the Vittarieae with a view to their systematic comparison. Transactions of the Royal Society of Edinburgh 55: 173–217. Zhang G, Zhang X, Chen Z. 2005. Phylogeny of cryptogrammoid ferns and related taxa based on rbcL sequences. Nordic Journal of Botany 23: 485–493. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Appendix S1. Species examined where silica bodies are clearly visible in the epidermis between veins of the leaf. All vouchers housed at NY. Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article. APPENDIX Taxon Adiantum Adiantum Adiantum Adiantum Adiantum Adiantum Adiantum Adiantum Adiantum Adiantum Adiantum Adiantum Adiantum Adiantum Adiantum Adiantum Adiantum argutum Splitg. caudatum L. cajennense Willd. cordatum Maxon dolosum Kunze fructuosum Poepp. ex Spreng. glaucescens Klotzsch humile Kunze incertum Lindman intermedium Sw. kaulfusii Kunze kendalii Jenman latifolium Lam. lucidum (Cav.) Spreng. macrophyllum Sw. mcvaughii Mickel & Beitel multisorum Sampaio Voucher: Collector Locality Nee & Moran 45225 Maxwell s.n. Maguire 22858 Pittier 4297 (US) Croizat s.n. A.R. Smith 1832 de Granville 7045 Sundue 306 Arroyo & Sims 1375 Underwood 100 C. Wright 1077 Jenman s.n. Fendler 2 Maxon 4644 Campos & Lopez 4927 McVaugh 19196 Prance et al. 475 Bolivia Sri Lanka Brittish Guiana Panama Venezuela Costa Rica French Guiana Belize Bolivia Jamaica Cuba Jamaica Trinidad Panama Peru Mexico Brazil © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435 SILICA BODIES IN PTERIDACEAE (PTERIDOPHYTA) 435 APPENDIX Continued Taxon Voucher: Collector Locality Adiantum oaxacanum Mickel & Beitel Adiantum obliquum Desv. Adiantum paraense Hieron. Adiantum petiolatum Desv. Adiantum phyllitidis J.Sm. Adiantum platyphyllum Sw. Adiantum poeppigianum (Kuhn) Hieron. Adiantum pulverulentum L. Adiantum scalare R.M.Tryon Adiantum seemanii Hook. Adiantum serratodentatum Willd. Adiantum terminatum Kunze ex Miq. Adiantum tetraphyllum Humb. Bonpl. ex. Willd. Adiantum tomentosum Klotzsch Adiantum trichochlaenum Mickel & Beitel Adiantum venustum D.Don Adiantum villosissimum Kuhn Adiantum villosum L. Adiantum vogellii Mett. ex Keys Adiantum wilsonii Hook. Antrophyum boryanum (Willd.) Spreng. Antryophum brasilianum (Desv.) C.Chr. Antrophyum latifolium Bl. Antrophyum lineatum (Sw.) Kaulf. Antrophyum mannianum Hook. Antrophyum minersum Bory Antryophym obovatum Baker Antrophyum plantagineum Cav. Antrophyum reticulatum (Forst.f.) Kaulf. Haplopteris anguste-elongata Hay Haplopteris elongata (Sw.) E.H.Crane Haplopteris ensiformis Sw. Haplopteris flexuosa Fée Haplopteris forestiana Ching Haplopteris zosterifolia Willd. Monogramma graminea (Poir.) Schkuhr Monogramma paradoxa (Fée) Bedd. Polytaenium cajenense (Desv.) Spreng. Polytaenium chlorosporum (Mickel & Beitel) E.H.Crane Polytaenium lanceolatum (L.) Desv. Polytaenium urbanii (Brause) Alain Pteris multifida Poir. Pteris ryukyuensis Tagawa Radiovittaria gardneriana (Fée) E.H.Crane Radiovittaria minima (Benedict) E.H.Crane Radiovittaria remota (Fée) E.H.Crane Vittaria flavicosta Mickel & Bietel Vittaria graminifolia Kaulf. Vittaria isoetifolia Bory Vittaria lineata (L.) J.E.Sm. Mickel 5775 Breedlove 33136 S.R Hill 12794 Mickel 7365 Steyermark 87166 van der Werff 11360 Maas et al. 4582 Mickel 7362 Moran 6255 Wagner 78584 G. Eiten & L. Eiten 9593 R.S. Williams 858 Steinbach 5195 Prance et al. 8428 Breedlove 26564 Kingdon-Ward 19016 Moran 7633 Sundue 329 Fay 1028 Breedlove 34532 Sieber s.n. W.W. Thomas et al. 10543 Anon., herb. E.B. Copeland Sundue & Matos 1677 Lewalle 3240 A.D.E. Elmer 11090 Sino-American Guizou botanical expedition 1491 anon. ex herb Redfield Coveny & Hind 7148 T.C. Huang 2262 Motley 2673 S.K. Lau 1458 J. Ohwi & K. Okamoto 1408 Hennipman 3403 Bouford & Bartholomew 25131 Lady Barcly s.n. Flynn 6243 Croat 68364 Rojas 5110 Sundue & Link-Perez 1448 D. Watt s.n. X.Q. Wang 101 Monono et al. 2769 J.Clarke 745 Skutch 5250 L. Arroyo et al. 3345 Mickel & Bietel 5944 A.N. Zafra 1642 Newman & Whitmore 462 Nee & Taylor 29183 Mexico Mexico Brazil Mexico Venezuela Ecuador Peru Mexico Ecuador Costa Rica Brazil Panama Bolivia Bolivia Mexico India Ecuador Guatemala Sierra Leone Mexico Mauritius Brazil Philippines Costa Rica Burundi Philippines China India Australia Taiwan French Polynesia, Rapa Iti China Japan Thailand Taiwan Mauritius Micronesia, Pohnpei Costa Rica Costa Rica Costa Rica Jamaica China Japan Ecuador Costa Rica Bolivia Mexico Mexico Mozambique Mexico © 2009 The Linnean Society of London, Botanical Journal of the Linnean Society, 2009, 161, 422–435