MIAMI UNIVERSITY The Graduate School

Certificate for Approving the Dissertation

We hereby approve the Dissertation

of

Mirabai R. McCarthy

Candidate for the Degree:

Doctor of Philosophy

______Director (Dr. R. James Hickey)

______Reader (Dr. Michael Vincent)

______Reader (Dr. Richard Moore)

______Reader (Dr. Martin Henry Stevens)

______Graduate School Representative (Dr. Ellen Currano) ABSTRACT

MOLECULAR SYSTEMATICS AND MORPHOLOGY OF THE ADIANTUM PERUVIANUM GROUP (PTERIDACEAE)

by Mirabai R. McCarthy

The nearly cosmopolitan Adiantum (Pteridaceae) comprises approximately 200 species, most of which are tropical in distribution. Adiantum ferns are predominantly terrestrial forest taxa, but some have adapted to living along stream banks or moist, rocky habitats. Historically, Adiantum has been divided into morphological species groups, as a means to help taxonomists categorize species and interpret inter-specific relationships within the genus. However, group concepts and circumscriptions have changed extensively over time, due in part to different interpretations about the relative importance of various morphological characters, and subsequently, alternate group names have been used in the literature. Currently there is no clear consensus on how species should be grouped. The present study utilizes molecular phylogenetics to delineate species groups within Adiantum, and traditional morphology to revise and circumscribe one species group, the A. peruvianum Group. Our phylogenetic analyses revealed four primary clades (A, B, C, and D) within Adiantum. Clades A & B contain temperate species, clade C contains primarily paleotropical species, and clade D contains neotropical species. Neotropical clade D contains two sub-clades D1 and D2. Clade D1 contains species with visible venuloid idioblasts present between veins, whereas clade D2 contains species lacking visible venuloid idioblasts between veins. The Adiantum peruvianum Group is a sub- clade (clade D2.III) within clade D2, and contains at least eight species: A. anceps, ‘A. ecuadorianum’, ‘A. exuberans’, A. mathewsianum, A. pectinatum, A. peruvianum, A. polyphyllum, and A. trapeziforme. Members of this group can be identified by having clathrate rhizome scales with minutely denticulate margins; segment stalks that stop abruptly at segment bases; discreet sori that are depressed-ovate, reniform, or occasionally lunate; ultimate segments that are dimidiate, trapeziforme, deltate, or occasionally ovate-falcate; free, dichotomous veins that end in marginal teeth; and laminar tissue that lacks visible idioblasts between veins. A taxonomic account, including descriptions and distributions of the species within this group is presented. Two members of this group, ‘Adiantum ecuadorianum’ and ‘A. exuberans’, are described as new, as is A. mariposatum, a member of clade D2.I.

MOLECULAR SYSTEMATICS AND MORPHOLOGY OF THE ADIANTUM PERUVIANUM GROUP (PTERIDACEAE)

A DISSERTATION

Submitted to the Faculty of

Miami University in partial

Fulfillment of the requirements

For the degree of

Doctor of Philosophy

Department of Botany

by

Mirabai R. McCarthy

Miami University

Oxford, OH

2012

Dissertation Director: R. James Hickey

©

Mirabai R. McCarthy

2012

TABLE OF CONTENTS

Chapter 1. Introduction…………………………………………………………………..1

Chapter 2. The Adiantum peruvianum Group: a molecular circumscription……………8

Chapter 3. A taxonomic revision of the Adiantum peruvianum (Pteridaceae) Group…34

Chapter 4. Adiantum mariposatum (Pteridaceae, a new species from Ecuador………..83

Conclusion ……………………………………………………………………………..89

iii LIST OF TABLES

Chapter 2. Table 2.1: Taxa, vouchers, localities and Genbank accession numbers for all sequences analyzed……………………………………………….24

Table 2.2: Primers used for PCR and sequencing in the present study……………27

Chapter 3. Table 3.1: A comparison of characters that can be used to differentiate members of the Adiantum peruvianum Group……………………..66

Table 3.2: Our two putative A. trapeziforme groups showing characters with means that are significantly different from each other………….68

iv LIST OF FIGURES

Chapter 2. Figure 2.1: Consensus tree for 74 taxa, produced from analyses of rbcL dataset……………………………………………………………………..28

Figure 2.2: Consensus tree for 39 taxa, produced from analyses of combined atpA + rbcL datasets…………………………………………...……30

Figure 2.3: Phylogram resulting from combined atpA + rbcL reduced dataset, including only the A. peruvianum Group members……………...32

Chapter 3. Figure 3.1: Silhouettes of ultimate segments from species in the Adiantum peruvianum Group………………………………………….………69

Figure 3.2: Distribution maps of A. peruvianum, A. mathewsianum, A. anceps, A. polyphyllum…………………………………………………..…...... 70

Figure 3.3: Distribution maps of A. trapeziforme and A. exuberans………..…..…71

Figure 3.4: Scanning electron micrographs of spores…………………………..…72

Figure 3.5: Adiantum anceps………………………………………………………73

Figure 3.6: Adiantum ecuadorianum………………………………………..…..…74

Figure 3.7: Adiantum exuberans……………………………………………….…..75

Figure 3.8: Adiantum trapeziforme………………………………………………...76

Figure 3.9: Results from Principle Component Analyses…………………………77

Figure 3.10: Lectotype of Adiantum trapeziforme ………………………………...78

Figure 3.11: Adiantum mathewsianum………………………………………….…79

Figure 3.12: Adiantum pectinatum………………………………………………...80

Figure 3.13: Adiantum peruvianum…………………………………………….….81

Figure 3.14: Adiantum polyphyllum…………………………………………….…82

v Chapter 4. Figure 4.1: Holotype of Adiantum mariposatum………………..…………………87

Figure 4.2: Adiantum mariposatum adaxial & abaxial views of rachis and segment……………………………………………………………….88

vi ACKNOWLEDGEMENTS

There are many folks to whom I owe thanks… Starting with my committee, I’d like to thank the current members: Drs. Jim Hickey, Mike Vincent, Rich Moore, Hank Stevens, Ellen Currano, and the historical members: John Kiss, John Rakovan, Hardy Eshbaugh, Beth Schussler, and Linda Watson. Each of you has impacted my life in a unique way, and I am incredibly grateful for the many years of mentoring, scholarly wisdom, and friendship. It has been such a joy to know and work with you all – thank you, thank you, thank you!!! Jim Hickey, I couldn’t imagine having a better advisor. I will always remember our friendly debates, and banter. You have carved out a very special place in my heart. I also thank Barb Wilson and Vickie Sandlin for taking such great care of so many details over the years – you are wonderful! Eric Tepe, I am incredibly grateful for your companionship in Ecuador, and your assistance with statistics. My lab mates and friends, Laura Klein, Sushma Shrestha, Steve & Jessica Rybzinski, Melanie Link-Perez, Susannah Fulton, Tia Alhquist, Jennifer Brown, Peter Pelser, Li Zheng, and Aaron Kennedy you helped Miami feel like home. Thank you all! I would also like to thank the Department of Botany at Miami University for funding much of my research through the Academic Challenge, and Willard Sherman Herbarium Grants; the following herbaria for specimen loans: AAU, F, GH, MO, MU, NY, US; Matt Duley, and Richard Edelman in the SEM facility; Johnny Kou for help with imaging scales and spores; the Center for Bioinformatics and Functional Genomics at MU; and the sequencing facility at Yale University. Finally, I am endlessly thankful to my mom, and my partner in life Jeremiah Jonaitis, for their encouragement, support, and especially for helping to keep me balanced throughout this process. I love you.

vii 1

Introduction

The primary objectives of this dissertation were to explore the evolutionary relatedness among species of Adiantum, and determine whether the A. peruvianum Group is monophyletic. A molecular phylogeny was produced as a hypothesis of evolutionary relatedness, and comparative morphological data were employed in the formalization of taxonomic descriptions and keys.

Adiantum L. (Pteridaceae) The Pteridaceae is a large, diverse, and cosmopolitan family of leptosproangiate ferns of various ecological niches, including terrestrial, epiphytic, rupestral, and aquatic habitats. There are about 50 genera and over 1000 species in the Pteridaceae (Schuettpelz et al. 2007). Members of this family are characterized by sporangia born along veins, either superficially or along margins that are protected by a reflexed marginal flap (pseudoindusia), a vertical, interrupted annulus, trilete spores, and a chromosome number of x = 29 or 30 (Tryon & Tryon 1982). The Pteridaceae has experienced a great deal of scrutiny and debate about which genera should be included in this family and how they are related (Copeland 1947, Tryon & Tryon 1982, Tryon et al. 1990, Smith 1995, Prado et al. 2007, Schuettpelz et al. 2007). The most recent and robust molecular phylogenetic assessment of the Pteridaceae (Schuettpelz et al. 2007) divides the family into five monophyletic clades: the Pteridoid clade (17 genera, 400 species), Cryptogrammoid clade (3 genera, 23 species), Ceratopteridoid clade (2 genera, 6 species), Cheilanthoid clade (20 genera, 400 species), and Adiantoid clade (12 genera, 300 species). The latter Adiantoid clade includes the vittariod ferns (Adiantopsis, Anetium, Antrophyum, Haplopteris, Hecistopteris, Monogramma, Polytaenium, Radiovittaria,

1 Rheopteris, Vittaria) and Adiantum. A phylogenetic analysis of the Pteridaceae by Schuettpelz et al. (2007) found that the vittariod group and Adiantum form a well- supported monophyletic clade, but the genus Adiantum as currently circumscribed is paraphyletic. Prado et al. (2007) pointed out in his phylogentic assessment of this family that monophyly in Adiantum would be gained if two species; A. cuneatum and A. radianum were excluded from the genus. However, in a recent five-marker analysis by Lu et al. (2012) Adiantum did receive monophyletic support. It would seem that additional studies are needed to resolve the phylogenetic complexities of these species. Adiantum is distinctive in having a unique soral structure wherein the sporangia are attached on and along the reflexed membranous leaf margins - the false indusia, and by the presence of a black, shiny stipe. Leaves are usually monomorphic, entire, or 1-4 pinnate (occasionally 6-pinnate); ultimate segments are typically glabrous, but may be glaucous, pubescent or scaly. The veins can be free or reticulate, and if so, then lacking included free veinlets, and with or without visible venuloid idioblasts (false veins). Sori range in shape from orbicular to reniform or may fuse to form long, continuous coenosori. The false indusia are either glabrous, glaucous, or with indument; spores are trilete, and spore exine is mostly smooth (Tryon & Tryon 1982). The approximately 200 species of Adiantum have a cosmopolitan distribution, with nearly 50% of the species occurring in the Neotropics. Species occur in primary and secondary forests and range in altitude from sea level to about 5000m (Tryon & Tryon, 1982). Adiantum contributes considerably to species richness and composition in the understory of Amazonian rain forests (Tuomisto et al., 1998), and various species have been used for food (Jagtap et al. 2010), medicine (Pourmorad et al. 2006, Hammond et al. 1998, Singh et al. 2008, Komatsu et al. 1996, Mubashir & Shah 2011), and as ornamentals (Hoshizaki 1970). Despite the significance of this genus, there has never been a complete monograph of Adiantum developed and species relationships are poorly understood. A modern systematic and taxonomic study of Adiantum is much needed.

The Species Group In 1982, Tryon & Tryon attempted to organize Adiantum by dividing it into eight morphologically distinctive species groups: 1. ‘A. capillus-veneris’, 2. ‘A. patens’, 3. ‘A.

2 philippense’, 4. ‘A. reniforme’, 5. ‘A. pectinatum’, 6. ‘A. tetraphyllum’, 7. ‘A. platyphyllum’, and 8. ‘A. phyllitidis’. These assemblages provided a convenient way for taxonomists to categorize species and interpret inter-specific relationships within Adiantum. These species groups were not intended to be monographic in scope, and they are by no means complete; they include (perhaps) only about half of the total number of Adiantum species - those found primarily in the new world - and in some instances species were only tentatively placed into groups. However, Tryon’s circumscriptions remain the most detailed morphological account of Adiantum groups. Since Tryon’s classification of Adiantum, group concepts and circumscriptions have changed extensively (Moran et al. 1995, Lellinger & Prado 2001, Prado and Lellinger 2002, Prado 2003, Sundue & Prado 2005), due in part to different interpretations about the relative importance of various morphological characters, and subsequently alternate group names have been used in the literature. Currently there is no clear consensus on how species should be grouped. Only recently have molecular techniques been utilized in helping to delineate species groups; in 2007 Layne Huiet provided a list of about 15 species in the A. peruvianum Group based on preliminary molecular analyses (pers. com.). In 2012 Lu et al. constructed a molecular phylogeny of Adiantum that focused primarily on Chinese taxa; their data revealed six clades within the temperate taxa, and four clades within the pantropical taxa. The groups of Adiantum remain in flux and many species relationships are still poorly understood; yet molecular techniques have expanded our capacity to more accurately address phylogenetic questions. There has been no extensive work to establish phylogenetic relationships among Neotropical Adiantum species. Restricted morphological assessments have been proposed for several of these groups, and limited molecular analyses have contributed towards a broad-scope understanding of relatedness. A phylogenetic framework is needed to appropriately analyze the entire genus, and merging molecular and morphological characters would greatly enhance our comprehension of the more natural Adiantum assemblages and help us to determine which morphological characters are the most important in discriminating species groups.

3

Research Intentions & Questions The intention of the current research was to test the monophyly of the Adiantum peruvianum Group as identified by Huiet (pers. com.), and circumscribe and revise the species that fall within this clade. A combined morphological and molecular systematic approach was utilized to comprehensively define this species group and determine monophyly. In addition to examining the taxa mentioned by Huiet, I also examined other putatively related species from the A. platyphyllum, A. pectinatum, A. tetraphyllum, and A. phyllitidis Groups, as circumscribed by Tryon and Tryon (1982). The following questions inspired this research: 1. Is the Adiantum peruvianum Group, as identified by Huiet (pers. com.) monophyletic? 2. Is there a clearly recognizable clade associated with A. peruvianum, and does it reflect one of Tryon’s morphological clades? 3. What morphological characteristics define this clade? 4. What species are included in this clade and how are they related?

Outline of Chapters This dissertation represents the first investigation of Adiantum that utilizes both molecular and morphological techniques to evaluate the phylogeny of a species group. Chapter 2 consists of a molecular systematic study of the A. peruvianum Clade, which uses two chloroplast markers, atpA and rbcL, to resolve phylogentic relationships among species, interpret species-group boundaries and monophyly. Chapter 3 is a morphological study and revision of species in the revised A. peruvianum Group and includes full species descriptions, distribution maps, and a dichotomous key. Chapter 4 describes a new species, Adiantum mariposatum (McCarthy & Hickey, 2011), that was discovered in the process of this research. This description is included as a separate chapter because we do not consider Adiantum mariposatum part of the A. peruvinaum Group. This newly described species has segment stalks that continue into the base of the ultimate segments, and pubescent sori. These characters ally it to the closely related A.

4 urophyllum. This dissertation concludes with a summary about the major findings from this research, and direction for future studies.

5 LITERATURE CITED

Copeland, E. 1947. Genera Filicum. Chronica Botanica, Waltham.

Hammond, G., I. Fernandez, L. Villegas, and A. Vaisberg. 1998. A Survey of Traditional Medicinal Plants From the Callejon de Huaylas, Department of Ancash, Peru. Journal of Ethnopharmacology 61: 17-30.

Jagtap, S., S. Deokule, S. Mukherjee, A. Kuvalekar, S. Devkar, A. Harsulkar, and P. Pawar. 2010. Assessment of Nutritional Value of Some Wild Edible Plants from Satpura Hills of Maharashtra, India. Journal of Herbal Medicine and Toxicology 4: 77-82.

Komatsu K., K. Iida, S. Cai, M. Mikage, and T. Yoshizawa. 1996. Pharmacognostical studies on Adiantum plants. V. Classification based on spore morphology and distributional patterns of silicon and calcium in the ultimate pinnules. Journal of the Pharmaceutical Society of Japan. 116: 125-37.

Moran R., B. Zimmer, and A. Jermy. 1995. Adiantum. In: R.C. Moran & R. Riba (eds.). Flora Mesoamericana. Volumen 1. Psilotaceae a Salviniaceae. Pg. 106-117 Univ. Nac. Autonoma de Mexico. Mexico D. F., Mexico. Mubashir S., and W. Shah. 2011. Phytochemical and Pharmacological Review Profile of Adiantum venustum. International Journal of PharmTech Research 3: 827-830.

Hoshizaki B.J. 1970. The genus Adiantum in cultivation (Polypodiaceae). Baileya 17: 97-191.

Lellinger, D., and J. Prado. 2001. The group of Adiantum gracile in Brazil and environs. American Fern Journal 91: 1-8.

Pourmorad, F., S. Hosseinimehr, and N. Shahabimajd. 2006. Antioxidant Activity, Phenol and Flavinoid Contents of Some Selected Iranian Medicinal Plants. African Journal of Biotechnology 5: 1142-1145.

Prado, J. 2003. New species in Adiantum from Brazil. American Fern Journal 93: 76- 80.

Prado, J. and D. Lellinger. 2002. Adiantum argutum, an unrecognized species of the A. latifolium group. American Fern Journal 92: 23–29.

Schuettpelz, E.; H. Schneider; L. Huiet; M. Windham; and K. Pryer. 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.

6 Singh, M., N. Singh, P. Khare, and A. Rawat. 2008. Antimicrobial Activity of Some Important Adiantum Species Used Traditionally in Indigenous Systems of Medicine. Journal of Ethnopharmacology 115: 327-329.

Sundue, M. and J. Prado. 2005. Adiantum diphyllum, a rare and endemic species of the Bahia State, Brazil, and it’s close relatives. Brittonia 57: 123-128.

Tryon, R. M. and A. F. Tryon. 1982. Ferns and allied plants with special reference to tropical America. Pg. 319-320 Springer – Verlag, New York Inc.

Tryon R., A. Tryon, and K. Kramer. 1990. Pteridaceae. In: Kramer, K.U., Green, P.S. (Eds.), The Families and Genera of Vascular Plants. Vol. 1. Pteridophytes and Gymnosperms. Pg. 230-256 Springer-Verlag, Berlin.

Tuomisto, H., A. Dalberg Poulsen, and R. C. Moran. 1998. Edaphic distribution of some species of the fern genus Adiantum in western Amazonia. Biotropica 30: 392- 399.

7 2

The Adiantum peruvianum Group: a molecular circumscription

INTRODUCTION

The nearly cosmopolitan Adiantum comprises approximately 200 species, most of which are tropical in distribution. The various maidenhair ferns are predominantly terrestrial forest taxa, but some have adapted to living along stream banks or moist rocky habitats. Adiantum is distinct from other ferns by its unique soral structure wherein the sporangia are attached on and along the reflexed membranous leaf margins, the false indusia, and by the presence of a black, shiny stipe (Tryon & Tryon, 1982). Leaves are typically monomorphic, occasionally entire, or up to 6-pinnate, with variably shaped ultimate segments that can be glabrous or pubescent, and often glaucous. The veins are usually free, rarely reticulate, and with or without visible idioblasts (false veins). Along with the ceratopteroids, cheilanthoids, cryptogrammoids, and pteroids, the adiantoids constitute the five major clades in the Pteridaceae (Schuettpelz et al., 2007; Schuettpelz & Pryer, 2007). The adiantoids are sister to the cheilanthoids, and include the genus Adiantum and the vittarioid ferns (Vittaria etc.), although the relationship between these taxa remains unclear. Many studies on this large and diverse family have produced similar results suggesting that Adiantum is paraphyletic and the vittariods are derived from within Adiantum (Prado et al., 2007; Schuettpelz and Pryer, 2007, Schuettpelz et al., 2007, Bouma et al., 2010), but see Gastony and Rollo (1995). The most recent phylogenetic assessment of Adiantum utilized sequences from over 80, mostly Chinese taxa for five plastid markers (atpA, atpB, rbcL, trnL-F and rps4-trnS);

8

these analyses provide support for a monophyletic Adiantum (Lu et al., 2012). It appears as though results are sensitive to which taxa and genes are selected for sampling. Clarifying these relationships will take additional work. Relationships within Adiantum have never been satisfactorily established despite various attempts at classifying the genus based on regional studies. In 1957, Ching recognized six “series” encompassing 34 Chinese taxa. Nayer (1961) established four groups of Adiantum based on fifteen species in India. In 1977, Cooper-Driver & Swain segregated 58 new and old-world species of Adiantum into five sections based on flavonoid patterns, and morphology. And in 1980, Lin added an additional series to Ching’s series (1957) and made minor modifications. Tryon & Tryon (1982), working primarily with neotropical taxa, established eight morphological species groups: the ‘A. capillus-veneris’, ‘A. patens’, ‘A. philippense’, ‘A. reniforme’, ‘A. pectinatum’, ‘A. tetraphyllum’, ‘A. platyphyllum’, and ‘A. phyllitidis’ groups. Since Tryon’s circumscriptions there have been minor modifications of single groups done on a very regional basis (Moran et al. 1995; Lellinger & Prado, 2001; Prado and Lellinger, 2002; Prado, 2003; Sundue & Prado, 2005). These early attempts at elucidating relationships provide a foundation for more detailed studies and serve as first level hypotheses about relatedness. However, many, if not all, of these species groups are incomplete, particularly among the neotropical taxa, where most of the diversity exists, and currently there is no clear consensus on how species should be grouped. Advances in molecular techniques have greatly expanded our ability to address and explore questions regarding phylogenetics, but the vast majority of this type of work involving Adiantum has emphasized generic level relationships within the Pteridaceae (Gastony and Rollo, 1995; Prado et al., 2007; Schuettpelz and Pryer, 2007, Schuettpelz et al., 2007, Bouma et al., 2010). Surprisingly, there are very few molecular studies that have focused on species-level relationships within the genus, despite repeated efforts to establish morphological circumscriptions (Moran et al. 1995; Lellinger & Prado, 2001; Prado and Lellinger, 2002; Prado, 2003; Sundue & Prado, 2005). In 2004, Huiet and Smith assessed the phylogenetic relationships of some neotropical Adiantum species using the chloroplast gene rps4, and the intergenic rps4-trnS spacer; these results, however, remain unpublished. In 2007, Huiet (pers. comm.) established a list of species

9

in the Adiantum peruvianum Group (name established by Huiet) based on preliminary molecular analyses. Studies by Lu et al. (2012) represents one of the first publications to emphasize species relationships within Adiantum; these studies focused primarily on Chinese species, but additional, exemplar taxa from across its range were also included to broadly test the monophyly of the genus. Their results, using five plastid markers (atpA, atpB, rbcL, trnL-F and rps4-trnS), established a monophyletic Adiantum comprised of a temperate group, that includes almost all Asian species, and a pantropical grade that represents only a small portion of tropical taxa. The conclusions of Lu et al (2012) are incongruent with Lin’s (1980) morphological classification of Chinese Adiantum, and identified a couple of Tryon & Tryon’s (1982) morphological groups (e.g., the ‘A. capillus-veneris’ and ‘A. patens’ groups) as polyphyletic. These results suggest that morphology alone lacks the resolving power needed to establish an accurate classification for Adiantum, and that molecular data should be coupled with morphology. As previously mentioned, the neotropics contains the greatest amount of Adiantum diversity, and species relationships among this group have been largely overlooked in molecular studies. In the present research I utilized a molecular phylogenetic approach, using two plastid markers, atpA & rbcL, to provide a broader, preliminary assessment of new world species relationships and to resolve one species group, A. peruvianum of Huiet (pers. comm.). The principle objectives of this study were to determine (1) whether the A. peruvianum Group, as identified by Huiet (pers. comm.) is monophyletic, (2) to compare that clade, if real, with the morphological species groups of Tryon & Tryon (1982), (3) identify morphological characteristics that can be used to define this clade, and finally (4) establish which species are included within it and how they are related.

MATERIALS AND METHODS Taxon Sampling - 79 species of Adiantum (Table 2.1) were sampled for phylogenetic analyses. Sampling of taxa concentrated on species that were previously considered members of the A. peruvianum Group (as per Huiet pers. comm.), as well as species that were considered closely related to members of this group in Tryon & Tryon’s 1982 classification. Additional taxa were sampled from across the range of

10

morphological species groups (Tryon & Tryon, 1982) to help clarify species-group relationships among the neotropical taxa. Leaf tissue was collected from herbarium specimens, from fresh silica–dried material that was sent to us from collaborators, or that was collected during fieldwork in Puerto Rico and Ecuador. Vouchers and Genbank accession numbers for all sequences used in the analyses are listed in Table 2.1. Two chloroplast markers, atpA and rbcL, were amplified and sequenced. The rbcL region has been widely used in analyses of fern phylogenies in general, as well as, within the Pteridaceae (Hasebe et al., 1993 & 1994; Dubuisson, 1997; Pryer et al., 2001; Wolf et al., 2003; Pryer et al., 2004; Prado et al., 2007; Schuettpelz et al., 2007; Schuettpelz & Pryer, 2007; Bouma et al., 2010; Lu et al., 2012); its use in this study compliments those studies and provides an opportunity to establish a larger data set for future, more comprehensive analyses. The atpA region was selected because recent molecular studies (Schuettpeltz et al., 2006 & 2007) indicate that this gene has a higher resolving power than other genes, i.e. rbcL, atpB, rps4, 18S, at the species level and therefore was likely to provide a more robust and resolved phylogeny.

DNA extraction, amplification, and sequencing - Genomic DNA was isolated from herbarium leaf tissue, or when available from fresh (silica dried) tissue. Leaf tissue was weighed (10-20mg) and then pulverized using a Mini-BeadbeaterTM (Biospec Products). Total genomic DNA was extracted using the DNeasy Plant Mini kit (Qiagen, Valencia, California, USA), following, with minor modifications1, the manufacturers protocol. The atpA and rbcL regions were amplified following the protocol described in the IllustraTM, PuRe Taq, Ready-To-Go, PCR Beads product Booklet (GE Healthcare, Bio- Sciences, 2007) with an annealing temperature of 50˚C, and 40 cycles, and carried out using the atpA and rbcL primers listed in Table 2.2. Because DNA from herbarium specimens can be degraded, it was necessary to amplify smaller sections of the atpA and rbcL regions, using up to six primer pairs (forward and reverse) per sample (Table 2.2.). Amplification of PCR was prepared in 25 µl reactions: 1 PCR ready-to-go bead, 1 µl total

1AP1 buffer was increased from 400 to 450 uL in Step 7. Precipitation time was increased from 5 min to overnight in Step 9, in which case, samples were refrigerated on ice. Step 16 was repeated twice.

11

genomic DNA (full concentration), 25 pmol each of forward and reverse primers, and molecular grade water to volume. PCRs were preformed on a Dyad Peltier Thermal Cycler (BioRad). PCR products were electrophoresed on 1% agarose gels in sodium borate buffer and visualized with ethidium bromide. Gel images were viewed with an ALPHA Imager MultiImageTM Light Cabinet (Alpha Innotech Corporation) and images were captured with a Sony digital graphic printer. PCR products were purified by treatment with 0.5 µl Exonuclease 1, 0.5 µl Shrimp Alkaline Phosphotase, and 1 µl 1X SAP buffer (USB Products), for each 25 µl PCR reaction, and placed in a Dyad Peltier Thermal Cycler (BioRad) for 40 minutes at 37˚C, 15 minutes at 80˚C, and hold at 4˚C to deactivate enzymes. Purified PCR products were either sequenced at Miami University’s Center for Bioinformatics and Functional Genomics, labeled with BigDye v.3.1 (Applied Biosystems, Foster City, CA) and processed bi-directionally on ABI 310 or ABI 3100 Applied Biosystems capillary sequencers; or sequenced at Yale University DNA Analysis Facility labeled with BigDye v.3.1 (Applied Biosystems, Foster City, CA), sequencing reactions were processed bi-directionally on a 3730xl DNA Genetic Analyzer, and sequence data was post processed with Peaktrace (Software by Nucleics, Inc.). Forward and reverse sequences were obtained for all samples, and electropherograms were assembled and edited using Sequencher (Gene Codes, Ann Arbor, Michigan, USA). Sequences were manually aligned in Se-Al v2.0a11 (Rambaut, 2002).

Phylogenetic analyses - All accessions (46 for atpA and 74 for rbcL) were analyzed separately for each of the two markers, and in a combined data set. Trees were rooted using Vittaria graminifolia as the outgroup (Lu et al. 2012), except in a reduced 12-accession analysis that included only taxa in the Adiantum peruvianum Group, and its sister group, we used A. mariposatum as the outgroup. All analyses were carried out under Bayesian Inference (BI) and Maximum Parsimony (MP) optimality criteria. Bayesian Inference (BI) analyses were performed using MrBayes 3.1.2 (Huelsenbeck and Ronquist, 2001; Ronquist and Huelsenbeck, 2003; Altekar et al. 2004). Nucleotide substitution models were determined by MrModeltest 2.2 (Nylander, 2004), which

12

identified the GTR+I+G model as the most appropriate for all datasets analyzed. The Akaike Information Criterion (AIC) was used to choose the models because this method has been shown to perform better than the hierarchical Likelihood Ratio Test when comparing nested models (Posada and Buckley, 2004). Using random starting trees, MrBayes was run for 10,000,000 generations, and one tree was sampled every 1,000 generations. The analyses were performed using the parallel version of MrBayes 3.1.2 on the Bioportal cluster (Kumar et al. 2009) with 10 chains per run and with all other settings as the defaults. Post analysis was carried out in MrBayes to determine the number of trees to omit as burnin, and to compute the consensus tree and posterior probabilities (PP). Maximum parsimony (MP) analyses were performed using PAUP *4.0b10 (Swofford, 2003). All characters were weighted equally and gaps were treated as missing data. Analyses were run as full heuristic searches with 10 random addition sequence replicates, TBR swapping, and with all other settings kept as the defaults. The reduced data set including only the Adiantum peruvianum Group taxa and the outgroup, was analyzed using an Exhaustive search. Bootstrap (BS) values for nodes were estimated from full heuristic searches of 1,000 replicates with MaxTrees set at 10,000 and TBR branch swapping.

RESULTS DNA Sequences and alignment. - A total of 132 plastid sequences were utilized in the combined 2-marker alignment (Table 2.1). Of these, 80 were newly generated from the atpA & rbcL regions; additional sequences from 52 taxa were downloaded from GenBank. The combined data set consisted of 2,696 characters, of which, 822 were variable, and 429 were phylogenetically informative. The combined MP analysis yielded 998 steps in the shortest tree, a consistency index (CI) of 0.55, and a retention index (RI) of 0.74. Single gene datasets – Separate analyses were conducted for the atpA and rbcL regions to assess the level of congruence between these two data sets and to see if they produced similar topologies. The accompanying maximum parsimony rbcL tree (Fig. 2.1) demonstrates the presence of considerable hierarchy in our data set, as evidenced by

13

the nested sets of well-supported clades. Clade A (BS/PP = 1/100) consists solely of the temperate Adiantum pedatum. Clade B (BS/PP = 1/67) is a highly structured clade of temperate Old World species that, in our results, consists of two nested sub-clades, the first containing A. capillus veneris, A. ogasawarense, and A. flabellulatum (BS/PP = 1/97), and the second containing A. edgworthii, A. malesianum, A. capillus junonis, A. monochlamys, and A. venustum (BS/PP = 1/98). Clade C consists primarily of paleotropical species, except for Adiantum subcordatum, and A. ruizianum, which are neotropical. Clade C is only weakly supported (BS = 0.85), but if A. subcordatum is excluded from this clade, support increases substantially (BS/PP = 1/100). Clade D (1/100) is of particular interest relative to the questions being addressed here and consists solely of neotropical Adiantum species, with A. tenerum positioned as its sister. Neotropical clade D contains two sub-cades, D1 (BS/PP = 1/96) and D2 (BS/PP = 0.74/59). Within Clade D1 A. scalare and A. tetraphyllum are basal. Clade D2 contains additional nested clades, D2.I (BS/PP = 1/100), D2.II (BS/PP = 0.6/57), and D2.III (BS= 0.63); the D2.I clade containing A. mariposatum and A. urophyllum is sister to D2.II and D2III. These three nested clades within D2 received weak support in the single gene (rbcL) dataset, but strong support in our concatenated (rbcL & atpA) dataset (Figs. 2.2 and 2.3). The single-marker (atpA & rbcL) phylogenetic analyses produced congruent topologies, but the rbcL analyses (Fig 2.1) produced a more comprehensive topology than atpA (not shown) because there was a greater availability of taxa that could be included (74 rbcL accessions, as opposed to 46 accessions for atpA). Combined gene data sets – Our concatenated atpA and rbcL dataset contained fewer taxa (Fig. 2.2), because we had limited atpA accessions, but the bootstrap and posterior probability values were stronger within clade D than they were in the single- gene analyses, particularly among the nested clades in D2, where D2.I (BS/PP = 1/100), D2.II (BS/PP = 1/100), and D2.III (BS/PP = 1/100) all received strong support. In this clade, D2.I is basal and contains Adiantum urophyllum, and A. mariposatum. Clade D2.II contains A. ornithopodum and A. pentadactylon and is sister to the A. peruvianum Group (D2.III), which contains A. anceps, A. exuberans ined., A. mathewsianum, A. pectinatum, A. peruvianum, A. polyphyllum, and A. trapeziforme. Species relationships within the A. peruvianum Group were poorly resolved in the concatenated analyses, resulting in

14

polytomies (Fig. 2.2). In an attempt to better resolve the relationships among species in the A. peruvianum Group, we reduced our concatenated data set, including only one representative operational taxonomic unit from each species, its sister group, and an outgroup (Fig. 2.3). Species relationships were somewhat better resolved in these analyses, as evidenced by the phylogram in Fig. 2.3, but a limited number of changes between species resulted in short branch lengths.

DISCUSSION At a glance…the primary clades of Adiantum - Compared to atpA, our rbcL phylogenetic analysis (Fig. 2.1) provides the most comprehensive perspective on relationships within Adiantum. These data include 74 accessions, representing 44 out of (approximately) 200 total species in Adiantum. Clades A and B of the rbcL phylogeny (Fig. 2.1) have primarily temperate distributions; clade A, consisting of A. pedatum, is positioned as a sister group to the remaining Adiantum taxa. Clade B contains two strongly supported sub-clades, one clade with A. capillus veneris, A. ogasawarense and A. flabellulatum, and the other with A. edgworthii, A. malesianum, A. capillus junonis, A. monochlamys and A. venustum. The subgroups in clade B, although limited in number of representative species, correspond to “clades VII, VIII, and IX” of Lu et al., 2012, where “clade VII” contains A. edgworthii, A. malesianum, and A. capillus junonis, “clade VIII” contains A. flabellulatum, and “clade IX” contains A. capillus veneris. The paleotropical clade C (Fig. 2.1) corresponds to “clade I” of Lu et al., 2012, who further subdivided it into 3 sub-clades (I, II, III). Their “subclade 1” contains A. diaphanum, A. cunninghamii, A. fulvum, A. viridescens, A. aethiopicum, and A. hispidulum; “subclade II” contains A. raddianum; and “subclade III” contains A. formosum. Neotropical A. subcordatum and A. ruizianum were not included in the phylogeny by Lu et al., 2012, but in our results, A. subcordatum is positioned as sister to paleotropical clade C (Fig. 2.1), and the placement of A. ruizianum among this group is perplexing, given it’s neotropical distribution. Clade D (Fig. 2.1) represents the largest species group within Adiantum, those found in the neotropics. In the studies by Lu et al., 2012 this neotropical group, their “clade II”, was represented by A. peruvianum, and two “unknown” species. In our analyses, clade D consists of two sub-clades; clade D1 includes taxa with visible

15

idioblasts present between veins, whereas clade D2 includes taxa without visible idioblasts present between veins (Fig. 2.1). Clade D1 represents a large portion of tropical species, and may, with additional sampling, eventually be shown to contain additional subclades. Other taxa with visible idioblasts between veins, and potentially belonging in this group include: A. fovearum, A. incertum, A. argutum, A. seemanii, A. caryotideum, A. grossum, A. pulverulentum, A. pyramidale, A. gracile, A. nudum, A. alarconium, A. dolosum, A. poeppigianum, A. isthmicum, A. phyllitidis, A. petiolatum, A. lucidum, and A. vogelii. Clade D2 can be further divided into three subclades (D2.I, D2.II, D2.III), each of which received strong support in our concatenated atpA & rbcL analyses. In clades D2.I, and D2.II, stalk color continues into the segment bases, whereas in clade D2.III, stalk color stops abruptly at the segment base. Taxa in clade D2.I have pubescent segment stalks and indusia (see McCarthy & Hickey, 2011), whereas those in clade D2.II have glabrous indusia and, mostly, glabrous segment stalks (see Prado & Palacios-Rios, 1998). Clade D2III is the Adiantum peruvianum Group. The Adiantum peruvianum Group (clade D2.III) - Phylogenetic analyses of the combined atpA, and rbcL dataset clearly identify a strongly supported, monophyletic clade associated with Adiatum peruvianum. Species included in this clade are: Adiantum anceps, A. exuberans ined., A. mathewsianum, A. pectinatum, A. peruvianum, A. polyphyllum, and A. trapeziforme (Figs. 2.1, 2.2, and 2.3). These species all occur in South America, except A. trapeziforme and A. exuberans ined., which occur in Mexico and Central America. This clade can be identified by the presence of clathrate rhizome scales with minutely denticulate margins; segment stalks that stop abruptly at segment bases; discreet sori that are depressed-ovate, reniform, or occasionally lunate; ultimate segments that are dimidiate, trapeziforme, deltate, or occasionally ovate-falcate; free, dichotomous veins that end in marginal teeth; and laminar tissue that lacks visible idioblasts between veins. It is clear that the included phylogenetic estimation of Adiantum clades is largely incompatible with the schema proposed by Tryon & Tryon (1982). The A. peruvianum Group (clade D2.III) exemplifies this rather well. The current clade D2.III is comprised of members from Tryon & Tryon’s A. platphyllum Group (A. peruvianum and A. anceps)

16

and their A. pectinatum Group (A. mathewsianum, A. pectinatum, and A. trapeziforme). The other members of their A. platphyllum Group, A. subcordatum, A. platyphyllum, A. seemanii, A. pentadactylon and A. grossum are placed in our clades D1 and D2.II, or have not yet been sampled. Tryon & Tryon (1982) characterized species in the A. platyphyllum Group as having large, 1-4 pinnate lamina that are typically long stalked, with acute to acuminate segments, free veins, glabrous axes, and several to many indusia per segment. The morphological characters that Tryon & Tryon (1982) utilized in their circumscription of this group are widely inclusive, and appear to lack broad phylogenetic signal. Additionally, my results suggest that stalk length, lamina dissection, and size, which were considered by Tryon & Tryon (1982) to be defining group characters, vary among closely related species. Tryon & Tryon’s A. pectinatum Group also included: A. affine, A. formosum, A. fulvum, A. hispidulum, A. melanoleucum, A. pyramidale, A. brasiliense, A. macrocladum, and A. wilesianum. Of these, A. mathewsianum, A. pectinatum, and A. trapeziforme are representatives of clade D2.III, whereas A. fulvum and A. hispidulum (and perhaps A. formosum) fall within clade C. Tryon and Tryon’s A. pectinatum Group was characterized by species with 3-6 pinnate lamina, glabrous or scaly axes, ultimate segments that are acute-acuminate, with several indusia, veins that are free or reticulate, and sessile or short segment stalks. Again, this circumscription relies on lamina dissection, and stalk length as defining features, both of which vary across closely related species. Adiantum trapeziforme and A. mathewsianum were considered transitional between the A. platyphyllum and A. pectinatum Groups because of their larger, distinctly stalked segments. Current data (Figs. 2.1, and 2.3) indicate that the two groups defined by Tryon & Tryon, 1982 as A. platyphyllum, and A. pectinatum are unsupported by molecular characters. The group informally identified as A. peruvianum by Huiet (pers. comm.) includes several reticulate-veined species (A. adiantoides, A. krameri, A. leprieurii), A. anceps, A. mathewsianum, A. pectinatum, A. peruvianum, A. polyphyllum, A. trapeziforme, A. abscissum, A. pyramidale, as well as, another five species included in Tryon & Tryon’s (1982) ‘A. platyphyllum’ group. Our results indicate that these species fall among the presently defined clades D1, D2.II, D2.III, or have not yet been sampled.

17

In addition to Adiantum anceps, A. exuberans ined., A. mathewsianum, A. pectinatum, A. peruvianum, A. polyphyllum, and A. trapeziforme, we hypothesize that A. ecuadorianum ined. also belongs in this clade, but we were unable to obtain quality sequences from herbarium tissue. Adiantum ecuadorianum ined. lacks visible idioblasts, is once-pinnate, with large rhombic-flabellate shaped ultimate segments, has numerous discreet indusia and long segment stalks. All of these features are characteristics of the A. peruvianum Group. Adiantum exuberans ined. represents another new species that is well demarcated on the basis of morphology, and molecular results (Figs. 2.1, 2.2, & 2.3). In our analyses (Figs. 2.1, 2.2, and 2.3) Adiantum peruvianum collections from Ecuador and Bolivia came out in two separate places on the phylogeny, indicating that A. peruvianum likely consists of multiple species. The A. peruvianum accessions that group most closely with A. mathewsianum and A. anceps are from Ecuador, whereas the A. peruvianum accessions that group most closely with A. trapeziforme are from Bolivia. Upon closer examination of these taxa, we found only subtle morphological variations among these individuals (Chapter 3), and therefore do not feel comfortable segregating these until additional research is completed. Adiantum anceps, A. mathewsianum, A. polyphyllum, and A. pectinatum are morphologically distinct species; the polytomies in the A. peruvianum Clade (Figs. 2.1, 2.2, 2.3) are the result of insufficient phylogenetic signal. Little sequence variation among the taxa in this group, despite distinct morphological differences, suggests a recent and rapid radiation of the clade. Our data indicate that the atpA and rbcL genes work well for resolving species-group relationships among Adiantum, but are insufficient to clarify species-level relationships within groups. The atpA and rbcL genes are both actively coding regions, and may experience slower rates of evolution compared to other regions. According to Schneider and Schuettpelz 2006, the rbcL gene does not adequately discriminate among species. Intergenic spacers tend to be more variable, and therefore might be better at resolving species-level relationships. Different group relationships have been proposed for Adiantum using molecular (Lu et al., 2012; Huiet, (pers. comm.)) and morphological (Tryon & Tryon, 1982; Moran & Zimmer, 1995; Prado & Lellinger, 2002; Prado, 2003; Sundue & Prado, 2005) techniques. Here maximum parsimony analyses using atpA and rbcL sequences resulted

18

in 4 primary clades in Adiantum, each of which, presumably, contains additional structured hierarchy within it. The A. peruvianum Clade is monophyletic and includes at least eight taxa (A. anceps, exuberans ined., ecuadorianum ined., mathewsianum, pectinatum, peruvianum, polyphyllum, and trapeziforme). Our molecular analysis presents an enlarged, but still limited view of species relationships within Adiantum; it includes only 44 out of approximately 200 total species. Additional molecular and morphological studies are needed, particularly among the neotropical species of clade D (Figs. 2.1, & 2.2), to achieve a clear understanding of the genus. Of particular interest, and not addressed in the current study, is the placement of taxa with anastomosing veins (A. adiantoides, olivaceum, kramerii, leprieurii and wilsonii). Despite numerous attempts to extract DNA from herbarium tissue, I was unsuccessful at obtaining quality sequences from any of these species. Species in this group are mostly distributed in Guyana, and they lack visible idioblasts - except A. wilsonii, which is found in Ecuador and has visible idioblasts. For a more comprehensive morphological assessment of this group see Sundue & Prado, 2005, and Zimmer, 2007.

19

LITERATURE CITED

Altekar, G., Dwarkadas, S., Huelsenbeck, J.P., & Ronquist, F. 2004. Parallel Metropolis-coupled Markov chain Monte Carlo for Bayesian phylogenetic inference. Bioinformatics 20: 407-415. Bouma, W.L.M., Richie, P., & Perrie, L.R. 2010. Phylogeny and generic taxonomy of the New Zealand Pteridaceae ferns from chloroplast rbcL DNA sequences. Australian Systematic Botany 23: 143-415. Ching, R.C. 1957. On the genus Adiantum L. of China with notes on some related species from neighboring regions. Acta Phytotaxonomica Sinica 6: 301-354. Dubuisson, J. 1997. rbcL sequences: A promising tool for the molecular systematics of the fern Trichonomes (Hymenophyllaceae). Molecular Phylogenetics and Evolution 8: 128-138. Gastony, G.J., & Rollo, D.R. 1995. Phylogeny and generic circumscriptions of cheilanthoid ferns (Pteridaceae: Cheilanthoideae) inferred from rbcL nucleotide sequences. American Fern Journal 85: 341-360. Hasebe, M., Ito, M., Kofuji, R., Ueda, K. & Iwatsuki, K. 1993. Phylogenetic relationships of ferns deduced from rbcL gene sequence. Journal of Molecular Evolution 37: 476-482. Hasebe, M., Omori, T., Nakazawa, M., Sano, T., Kato, M. & Iwatsuki, K. 1994. rbcL gene sequences provide evidence for the evolutionary lineages of leptosporangiate ferns. Proceedings of the National Academy of Sciences 91: 5730-5734. Huelsenbeck, J.P., & Ronquist, F. 2001. MRBAYES: Bayesian inference of phylogeny. Bioinformatics 17: 754-755. Huiet, L., & Smith, A.R. 2004. Phylogenetic relationships in Adiantum inferred from chloroplast coding and non-coding sequences. In: Abstracts of Botany 2004, July 31-August 5, Botanical Society of America, Salt Lake City (Abstract). Kumar, S., Skjæveland, Å., Orr, R.S., Enger, P., Ruden, T., Mevik, B., Burki, F., Botnen, A. & Shalchian-Tabrizi, K. 2009. AIR: A batch-oriented web program

20

package for construction of supermatrices ready for phylogenomic analyses. BMC Bioinformatics. 10: 357. doi:10.1186/1471-2105-10-357. Lellinger, D., & Prado, J. 2001. The group of Adiantum gracile in Brazil and environs. American Fern Journal 91: 1-8. Lin, Y.X. 1980. New taxa of Adiantum L. in China. Acta phytotaxonomica sinica 18: 101-105. Lin, Y.X. 1990. Adiantaceae. In, Flora Reipublicae Popularis Sinicae Tomus. Vol. 3, issue 1. Science Press, Bejing. Lu, J.-M., Wen, J., Lutz, S., Wang, Y.-P., & Li, D.-Z. 2007. Phylogenetic relationships of Chinese Adiantum based on five plastid markers. Journal of Plant Research 125: 237-249. McCarthy, M.R. & Hickey, R.J. 2011. Adiantum mariposatum (Pteridaceae), a new species from Ecuador. American Fern Journal 101: 1-5. Davidse, G., Sousa, S.M., & Knapp, S. (general eds). 1995. Flora Mesoamericana. Vol. 1. Psilotaceae a Salviniaceae. Univ. Nac. Autonoma de Mexico, Mexico DF. Nayer, B.K. 1961. Ferns of India 1 Adiantum. Bulletin of the National Botanical Gardens 52: 1-38 Nylander, J.A. 2004. MrModeltest v2. Program distributed by the author. Evolutionary Biology Centre, Uppsala University. http://www.csit.fs.edu/~nylander/. Posada, D. & Buckley, T.R. 2004. Model selection and model averaging in phylogenetics: advantages of Akaike information criterion and bayesian approaches over likelihood ratio tests. Systematic Biology 53: 793-808. Prado, J. & Lellinger, D. 2002. Adiantum argutum, an unrecognized species of the A. latifolium group. American Fern Journal 92: 23–29. Prado, J. 2003. New species in Adiantum from Brazil. American Fern Journal 93: 76- 80. Prado, J., Rodrigues, C.N., Salatino, A., & Salatino, M.F. 2007. Phylogenetic relationships among Pteridaceae, including Brazilian species, inferred from rbcL sequences. Taxon 56: 355-368. Prado, J. & Palacios-Rios, M. 1998. Taxonomy and distribution of Adiantum

21

trapeziforme and A. pentadactylon. American Fern Journal 88: 145-149. Pryer, K. M., Schuettpelz, E., Wolf, P.G., Schneider, H., Smith, A.R., & Cranfill, R. 2004. Phylogeny and evolution of ferns (monilophytes) with a focus on the early leptosporangiate divergences. American Journal of Botany 91: 1582-1598. Pryer, K.M., Smith, A.R., Hunt, J.S., & Dubuisson, J.-Y. 2001. rbcL data reveal two monophyletic groups of filmy ferns (Filicopsida: Hymenophyllaceae). American Journal of Botany 88: 1118-1130. Rambaut, A. 2002. Se-Al Sequence Alignment Editor, Version 2.0aa11. Available at http://evolve.zoo.ox.ac.uk/software/SeAl/main.html. In. Ronquist, F. & Huelsenbeck, J.P. 2003. MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572-1574. Schneider H., & Schuettpelz, E. 2006. Identifying fern gametophytes using DNA sequences. Molecular Ecology Notes 6: 989-991. Schuettpelz, E., Schneider, H., Huiet, L., Windham, M.D., & Pryer, K.M. 2007. A molecular phylogeny of the fern family Pteridaceae: assessing overall relationships and the affinities of previously unsampled genera. Molecular Phylogenetics and Evolution 44: 1172-85. Schuettpelz, E., & Pryer, K.M. 2007. Fern phylogeny inferred from 400 leptosporangiate species and three plastid genes. Taxon 56: 1037-1050. Schuettpelz, E., Korall, P. & Pryer, K.M. 2006. Plastid atpA data provide improved support for deep relationships among ferns. Taxon 55: 897-906. Sundue, M. & Prado, J. 2005. Adiantum diphyllum, a rare and endemic species of the Bahia State, Brazil, and its close relatives. Brittonia 57: 123-128. Swofford, D.L. 2003. PAUP*: Phylogenetic Analysis Using Parsimony (* and Other Methods), Version 4.0b10. Sinauer, Sunderland. Tryon, R.M. & Tryon, A.F. 1982. Ferns and allied plants with special reference to tropical America. Springer – Verlag, New York. Wolf, P.G., Soltis, P.S. & Soltis, D.E. 1994. Phylogenetic relationships of dennstaedtioid ferns: evidence from rbcL sequences. Molecular Phylogenetics and Evolution 3: 383-392.

22

Zimmer, B. 2007. Adiantum krameri (Pteridaceae), a new species from French Guiana. Willdenowia 37: 557–562.

23

Table 2.1 Taxa, vouchers, localities and Genbank accession numbers for all sequences analyzed.

Taxa Voucher Genbank Previously published Locality Accession citations Number Adiantum aethiopicum L. WELT GU136781 Unpublished New Zealand P021136 EF469955 Perrie & Brownsey (2007) New Zealand LRP 2813

Adiantum anceps Maxon & Wurdack 1885 -- Current study Peru C.V. Morton

Adiantum capillus-junonis TNS 743693 AB574795 Ebihara et al (2010) Japan Rupr.

Adiantum capillus-veneris TNS 774832 AB574796 Ebihara et al (2010) Japan L. WELT GU136782 Unpublished Japan P022575 D14880 Hasebe et al. (1993) Japan -- DQ432659 Zhang et al. (2007) China -- Adiantum cajennense Willd. Prado et al. EF473675 Prado et al. (2007) Brazil ex Klotzsch 1201

Adiantum cuneatum Langsd. Prado & Yano EF473676 Prado et al. (2007) Brazil & Fisch. 1078

Adiantum cunninghamii WELT GU136783 Unpublished New Zealand Hook. P021138

Adiantum diogoanum Glaz. Araujo 2797 -- Current study Bolivia ex Baker

Adiantum diaphanum Blume TNS:777847 AB574797 Ebihara et al (2010) Japan WELT GU136784 Unpublished New Zealand Po21235 Adiantum edgeworthii Hook TNS:743692 AB574798 Ebihara et al (2010) Japan

Adiantum exuberans ined. Salas 3863 -- Current study Mexico McCarthy & Hickey Saunders 795 -- Current study Honduras 30135 -- Current study Mexico Monro 3464 -- Current study El Salvador Salas 3675 -- Current study Mexico McCarthy H120 -- Current study Honduras

Adiantum flabellulatum L. TNS:764008 AB574799 Ebihara et al (2010) Japan

Adiantum formosum R.Br. WELT GU136785 Unpublished New Zealand P022571 Adiantum fulvum M.Raoul WELT GU136786 Unpublished New Zealand P022572 Adiantum hispidulum Sw. WELT GU136787 Unpublished New Zealand P022570 Adiantum kalbreyerii C. Stolze 1746 -- Current study Ecuador Chr.

24

Table 2.1, continued. Adiantum latifolium Lam. Prado et al. 1294 EF473677 Prado et al. (2007) Brazil Sundue 1446 -- Current study Costa Rica Jansen 2061 -- Current study Guiana

Adiantum macrophyllum Sw. McCarthy H103 -- Current study Honduras McCarthy & -- Current study Ecuador Tepe 183 McCarthy & -- Current study Ecuador Tepe 133

Adiantum malesianum Huiet 111 (UC) EF452068 Schuettpelz et al. (2007) Unknown J.Ghatak Huiet 111 (UC) EF452132 Schuettpelz et al. (2007) Unknown

Adiantum mariposatum McCarthy & -- Current study Ecuador McCarthy & Hickey Tepe 141 Ollgaard 99574 -- Current study Ecuador

Adiantum mathewsianum Zardini 40959 -- Current study Paraguay Hook. Zardini 53826 -- Current study Paraguay Molina 305 -- Current study Bolivia

Adiantum monochlamys D.C. TNS:764001 AB574800 Ebihara et al (2010) Japan Eaton

Adiantum obliquum Willd. Prado et al. 1345 EF473678 Prado et al. (2007) Brazil Sundue et al. -- Current study Costa Rica 1451

Adiantum ogasawarense TNS:774843 AB574801 Ebihara et al (2010) Japan Tagawa

Adiantum ornithopodum C. Prado 1721 -- Current study Brazil Presl Prado 1764 -- Current study Brazil

Adiantum pedatum L. Huiet 117 (UC) EF452069 Schuettpelz et al. (2007) Unknown TNS:764338 AB574802 Ebihara et al (2010) Japan -- U05602 Hasebe et al. (1994) Unknown

Adiantum pectinatum Kunze Serrano 5890 -- Current study Bolivia ex Baker

Adiantum pentadactylon Prado et al. 1085 EF473679 Prado et al. (2007) Brazil Langsd. & Fisch. Labiak 3980 -- Current study Brazil Prado et al. NA -- Current study Brazil

Adiantum peruvianum Huiet 103 (UC) EF452070 Schuettpelz et al. (2007) Unknown Klotzsch Huiet 103 (UC) EF452133 Schuettpelz et al. (2007) Unknown Janovec 1383 -- Current study Ecuador Nee 34220 -- Current study Bolivia Sundue 523 -- Current study Bolivia

Adiantum platyphyllum Sw. Jiminez 2562 -- Current study Bolivia Sundue 881 -- Current study Bolivia

Adiantum polyphyllum Willd. Ortega 309 -- Current study Venezuela

25

Table 2.1, continued. Adiantum raddianum C. Wolf 717 (UTC) DQ390543 Schuettpelz et al. (2006) Unknown Presl Wolf 717 (UTC) EF452071 Schuettpelz et al. (2007) Unknown WELT P022574 GU136788 Unpublished New Zealand Wolf 244 U05906 Wolf et al. (1994) Hawaii Prado & Yano EF473680 Prado et al. (2007) Brazil 1077

Adiantum ruizianum Van Der Werff -- Current study Peru Klotzsch 16812

Adiantum Prado et al. 1545 EF473681 Prado et al. (2007) Brazil serratodentatum Humb. & Bonpl.

Adiantum scalare R.M. McCarthy & Tepe -- Current study Ecuador Tryon 192

Adiantum subcordatum Prado & Yano 107 EF473682 Prado et al. (2007) Brazil Sw. Schwartsberg -- Current study Brazil 1688 Cavalcanti 3172 -- Current study Brazil

Adiantum tenerum Sw. Huiet 107 (UC) EF452072 Schuettpelz et al. (2007) Unknown Huiet 107 (UC) EF452134 Schuettpelz et al. (2007) Unknown

Adiantum terminatum Prado et al. 1354 EF473683 Prado et al. (2007) Brazil Kunze ex Miq.

Adianum tetraphyllum Huiet 105 (UC) EF452073 Schuettpelz et al. (2007) Unnown Humb. & Bonpl. ex Huiet 105 (UC) EF452135 Schuettpelz et al. (2007) Unknown Willd. Sundue et al. 1432 -- Current study Costa Rica McCarthy & Tepe -- Current study Ecuador 182

Adiantum tomentosum Araujo 947 -- Current study Bolivia Klotzsch

Adiantum trapeziforme L. Morton 9786 -- Current study Cuba Acevedo 6474 -- Current study Cuba

Adiantum villosum L. McCarthy H105 -- Current study Honduras

Adiantum venustum D. Huiet 116 (UC) EF452136 Schuettpelz et al. (2007) Unknown Don

Adiantum viridescens WELT P022573 GU136789 Unpublished New Zealand Colenso

Outgroup

Vittaria graminifolia Schuettpelz 227 EF452128 Schuettpelz et al. (2007) Unknown Kaulf. (DUKE) Farrar CR312 VGU21295 Unpublished Unknown

26

Table 2.2 Primers used for PCR and sequencing in the present study

Target Primer name Primer sequence (5’ – 3’) BP Reference region Length atpA atpA138F AATGGCTGGCGAATTGGTGGAATC 24 Present study

atpA674F AAACTGCAAATTCTCCAGCCACCC 24 Present study

atpA792F ACAAGCCCAGGCTTATCGACAGAT 24 Present study

atpA806R TCTCTCCAAACAAGCCCAGGCTTA 24 Present study

atpA1454R AAAGGTGTCTACCGAACAAGCGGA 24 Present study

ESATPF412F GARCARGTTCGACAGCAAGT 20 Schuettpelz et al. (2006) ESATPA535F ACAGCAGTAGCTACAGATAC 20 Schuettpelz et al. (2006) ESATPA856F CGAGAAGCATATCCGGGAGATG 22 Schuettpelz et al. (2006) ESATPA557R ATTGTATCTGTAGCTACTGC 20 Schuettpelz et al. (2006) ESATPA877R CATCTCCCGGATATGCTTCTCG 22 Schuettpelz et al. (2006) ESTRNR46F GTATAGGTTCRARTCCTATTGGACG 25 Schuettpelz et al. (2006) rbcL rbcL1F ATGTCACCACAAACAGARACTAAAGC 26 Present study

rbcL825R TGCAGAAGTAACCCGTTGTCTCTGCA 26 Present study

rbcL1351R CTTCACAAGCAGCAGCTAGTTCAGGACTCC 30 Present study

rbcL660F AAC GTG AAT TCC CAA CCG TTT ATG 24 Gastony & Rollo, 1995 rbcL1080R ATCTTGGGTAAAATAGATGCC 21 Gastony & Rollo, 1995 J520F AATTATGGTAGAGCWGTTTA 20 Modified from Pryer et al. (2001) JYDS7(485)R AATTTTGGCTTGATTGTACATCC 23 Modified from Pryer et al. (2001) rbcL1379R TCACAAGCAGCAGCTAGTTCAGGACTC 27 Pryer et al. (2004)

27

Figure 2.1. Consensus tree for 74 taxa (including an outgroup, Vittaria graminifolia), produced from analyses of rbcL dataset. Numbers above are Bayesian posterior probabilities, and numbers below are maximum parsimony bootstrap support values. Branches lacking values are those with less then 50% support.

28 mathewsianum mathewsianum 0.76 anceps peruvianum peruvianum peruvianum 0.83/ peruvianum 67 peruvinaum trapeziforme D2.III 0.63 0.71/ trapeziforme 60 taxon 1 pectinatum polyphyllum D2 exuberans 0.99/62 exuberans exuberans exuberans exuberans 0.58/81 0.74/59 pentadactylon 0.6/57 pentadactylon D2.II ornothopodum ornithopodum Clade D 1/100 mariposatum mariposatum D2.I urophyllum latifolium 1/100 latifolium obliquum 0.92/91 obliquum 1/100 terminatum 1/96 serratodentatum 0.98/ 0.91/68 0.77 diogoanum 84 tetraphyllum 0.59 cajennense D1 villosum 0.67 1/100 0.87 platyphyllum platyphyllum 0.65 tetraphyllum 1/100 macrophyllum 1/96 macrophyllum macrophyllum 1/100 scalare tetraphyllum tenerum 1/100 hispidulum raddianum

0.96 raddianum 1/100 cuneatum raddianum ruizianum 1/96 fulvum 0.99 Clade C 1/99 1/100 viridescens cunninghamii 1/100 aethiopicum 1/100 aethiopicum 1/100 diaphanum 0.85 diaphanum formosum 1/100 subcordatum subcordatum capillus veneris 1/100 capillus veneris capillus veneris 1/97 capillus veneris ogasawarense Clade B 1/67 flabellulatum 1/94 edgworthii 1/100 malesianum 1/98 capillus junonis 1/100 monochlamys venustum 1/100 pedatum Clade A pedatum Vittaria graminifolia

29

Figure 2.2. Consensus tree for 39 taxa (including an outgroup, Vittaria graminifolia), produced from analyses of combined atpA + rbcL datasets. Numbers above are Bayesian posterior probabilities, and numbers below are maximum parsimony bootstrap support values. Branches lacking values are those with less then 50% support.

30 anceps mathewsianum

1/81 mathewsianum peruvianum peruvianum trapeziforme 0.63/88 trapeziforme 1/100 D2.III peruvianum exuberans exuberans D2 0.98/69 exuberans pectinatum polyphyllum ornithopodum 0.98/55 1/100 ornithopodum D2.II pentadactylon urophyllum 1/100 D2.I mariposatum Clade D mariposatum

1/100 diogoanum

1/100 tetraphyllum 1/100 1/100 latifolium

0.99/84 obliquum

1/100 platyphyllum

1/92 platyphyllum D1 1/93 0.7/54 tetraphyllum villosum 0.53 macrophyllum 1/96 1/100 macrophyllum macrophyllum

1/100 scalare tetraphyllum malesianum tenerum

1/100 raddianum 0.55 ruizianum subcordatum 31pedatum Vittaria graminifolia

Figure 2.3. Phylogram resulting from combined atpA + rbcL reduced dataset, including only the A. peruvianum Group members, sister group, and outgroup. Numbers above are Bayesian posterior probabilities, and numbers below are maximum parsimony bootstrap support values (only showing values greater then 50%).

32 0.51 anceps 1/92 mathewsianum 0.82/60 peruvianum polyphyllum 0.85/61 D2.III 1/100 peruvianum trapeziforme

1/100 exuberans pectinatum

1/100 ornithopodum D2.II pentadactylon urophyllum mariposatum 33 D2.I 5 changes 3

A taxonomic revision of the Adiantum peruvianum (Pteridaceae) Group

INTRODUCTION Adiantum L. (Pteridaceae) consists of approximately 200 species, with the majority occurring in the neotropics. Species occupy diverse habitats including, disturbed roadsides or riverbanks, as well as primary and secondary forests at elevations from sea level to 5000m (Tryon & Tryon, 1982). While most tend to grow in soil, it is not uncommon to find species growing on rocky outcrops or rock walls. The genus is morphologically distinctive within the Pteridaceae by its unique soral position, attached along reflexed membranous leaf margins that serve as false indusia, its black, shiny stipe, and its clathrate rhizome scales. The majority of species have veins with open dichotomous branching that end in, or between marginal teeth, but there are several species with reticulate venation. Venuloid idioblasts are common among genera in the Pteridaceae, and studies by Sundue (2009) indicate that these are also widespread in Adiantum. Venuloid idioblasts may occur between veins, in which case they are typically quite obvious, or they may be cryptically located on top of the veins. Certain patterns of distribution of these idioblasts may represent synapomorphies for clades within Adiantum (Sundue, 2009). The large number of species in Adiantum and their morphological diversity have been an impetus to the recognition of numerous, phenotype based species groups. These groups and their circumscriptions have changed extensively over the years due to different interpretations about the relative importance of morphological characters (Ching, 1957; Nayer, 1961; Tryon & Tryon, 1982; Prado & Palacios-Rios, 1998; Lin,

34 1990; Prado & Lellinger, 2002). Although there is no consensus on these groups, the proposal put forth by Tryon & Tryon (1982) is the most encompassing and provides a starting point for understanding the genus. In 2007, Huiet (pers. comm.) identified several species groups within Adiantum based on her initial molecular results (unpublished & unseen). The groups outlined by Tryon & Tryon (1982), combined with contributions from Huiet (pers. comm.), were utilized as a foundation to select species for this study, a goal of which was to help establish monophyletic species groups that could later be morphologically circumscribed. The results of our molecular analyses (Chapter 2) indicate that neotropical Adiantum taxa form a well supported clade (clade D), which contains two sub-clades, D1 and D2. Clade D2 can be further divided into three additional sub-groups, D2.I, D2.II, and D2.III. The present work addresses the taxonomy of the species in clade, D2.III (Chapter 2). To avoid confusion with other published group circumscriptions, clade D2.III will be referred to as the Adiantum peruvianum Group. As currently understood, the ‘Adiantum peruvianum’ Group includes A. anceps Maxon & C. V. Morton, ‘A. ecuadorianum’ ined. McCarthy & Hickey, ‘A. exuberans’ ined. McCarthy & Hickey, A. mathewsianum Hook., A. pectinatum Kunze ex Baker, A. peruvianum Klotzsch, A. polyphyllum Willd. and A. trapeziforme L. It is possible that other neotropical Adiantum species may yet prove to be included. The primary goal of the current chapter is to provide a taxonomic revision of the members of the Adiantum peruvianum Group, as it is currently understood.

MATERIALS AND METHODS Taxon sampling. Over 500 herbarium specimens of the taxa identified as belonging to the Adiantum peruvianum Group were borrowed from the following herbaria: New York Botanical Garden (NY), Missouri Botanical Garden (MO), Field Museum (F), University of Aarhus (AAU), Miami University (MU), Gray Herbarium (GH), and United States National Herbarium (US). Using locality data derived from these collections, distribution maps were developed using Google Earth (Version 6.1.0.4738 (beta)) as a means of locating and plotting the various collection sites

35 Morphology. Herbarium specimens were examined with a Heerbrugg Wild M3C dissecting scope, and over 50 discreet and continuous characters were measured from all taxa. Macromorphological characters were measured with a ruler, an ocular micrometer or vernier calipers. Sorus ratio was determined by dividing the number of sori along the acroscopic margins by the number of sori along the distal margins of the ultimate segments. Spores were collected from herbarium sheets with a dissecting needle, mounted on glass slides in Hoyer’s Mounting Medium (Anderson 1954, Barrington et al. 1986), and, after a period of 24 hours, were measured with the aid of an Olympus BH-2 compound microscope. Spores were also mounted directly onto metal stubs using double-sided sticky tabs, coated with gold, and viewed with a JOEL JSM – 840A scanning electron microscope. Rhizome scales were collected from herbarium specimens with a dissecting needle and mounted on glass slides in Hoyer’s Mounting Medium. Images of scales were captured using an Olympus SXZ12 stereo microscope, equipped with a Nikon D300 camera and imported for measurements into Image Pro Plus, version 6.3 (Media Cybernetics, Inc.). Descriptive morphology and photography was used to document variation between species and write species descriptions. The terminology for the descriptions is from Lellinger (2002), and the symmetric plane figures from the Systematics Association Committee for Descriptive Terminology (1962). Analyses –Statistical analyses were conducted on species in the A. trapeziforme complex using R software (version 2.12.0). Morphological characters were assessed for significant differences using an analysis of variance (ANOVA). Only those characters exhibiting significant differences (p-value ≤ 0.05) in means were used in the principle component analysis. In total, one qualitative character and seven quantitative characters showed significant differences. These were: presence/absence of rachis scales, ultimate segment stalk length, ultimate segment width, number of segments per terminal pinna, distance between segment stalks, sorus distribution ratio, rhizome diameter, and lamina height.

RESULTS Species known to be members of the Adiantum peruvianum Group are: A. anceps, ‘A. exuberans’, A. mathewsianum, A. pectinatum, A. peruvianum, A. polyphyllum, and A.

36 trapeziforme. In addition, one newly described species, ‘A. ecuadorianum’, is included. Although I was unable to obtain sequence data on this species, its morphology is consistent with other members of the clade and I tentatively assign it there. Members of this group can be identified by having clathrate rhizome scales with minutely denticulate margins; segment stalks that stop abruptly at segment bases; discreet sori that are depressed-ovate, reniform, or occasionally lunate; ultimate segments that are dimidiate, trapeziforme, deltate, or occasionally ovate-falcate (Fig. 3.1); free, dichotomous veins that end in marginal teeth; and laminar tissue that lacks visible idioblasts between veins (Table 3.1). The most informative characters that can be used to differentiate these eight species include, rhizome diameter, segment size and shape, segment stalk length, sorus ratio, number of pinnae, and lamina dissection (Tabe 3.1). All of these species occur in South America (Fig. 3.2), except Adiantum trapeziforme and ‘A. exuberans’, which occur in Mexico, Central America, and/or the Caribbean (Fig. 3.3). Rhizome and Rhizome Scale Characters – Rhizomes are decumbent, and in most species, short-to-medium-creeping, except for ‘Adiantum exuberans’ which has long- creeping rhizomes. Mean rhizome diameter typically ranges between 0.89-1.44 cm, except in ‘A. exuberans’ (mean=0.36 cm) and ‘A. ecuadorianum’ (0.4 cm) (Tables 3.1. and 3.2). Rhizome scales differ in length and width among species, but otherwise are quite uniform; all are triangular to narrowly-triangular, clathrate, and castaneous- stramineous, with minutely denticulate margins, especially distally. Frond & Pinnae Characters- All species in the Adiantum peruvianum Group display a pinnate laminar architecture, with the exception of ‘A. exuberans’, which occasionally tends toward a pedate architecture, particularly in specimens from Mexico. Frond height averages range between 70-101.5 cm, but ‘A. ecuadorianum’ is much smaller with leaves averaging only 41.5 cm (Table 3.1). Species are typically bipinnate to quadripinnate, except for the once pinnate ‘Adiantum ecuadorianum’ and the 4-6-pinnate A. mathewsianum. The number of pinnae per frond ranges between 3-11 in all species except A. polyphyllum and A. pectinatum, which tend to have more numerous pinnae (7-16, and 8-20 respectively) (Table 3.1). Species in this group have glabrous rachises, except for A. pectinatum, which has densely

37 pubescent-scaley rachises, and ‘A. exuberans’, which occasionally has weakly scaly rachises. Ultimate Segment and Segment Stalk Characters- The term ‘ultimate segment’ refers to the smallest division of the lamina. Species in this group have ultimate segment stalks that stop abruptly at the base of the ultimate segments. Ultimate segments are trapeziforme to dimidiate or deltate in all species, except Adiantum anceps, in which the ultimate segments tend more towards ovate-falcate. ‘A. ecuadorianum’, A. trapeziforme, A. peruvianum, and A. anceps have large ultimate segments (2.2-9.8 cm long, x 0.86-7.5 cm wide), with long segment stalks (0.2-3.4 cm) (Fig. 3.1 E-H). In ‘Adiantum exuberans’, A. mathewsianum, A. polyphyllum, and A. pectinatum the ultimate segments are smaller (0.74-6.37 cm long x 0.26-3.5 cm wide) with shorter segment stalks (0.01- 0.83 cm) (Fig. 3.1 A-D). Segment margins range from shallowly dentate in Adiantum anceps, A. peruvianum, A. mathewsianum, and A. polyphyllum, to biserrate in ‘A. ecuadorianum’, ‘A. exuberans’, and A. trapeziforme, to deeply incised in A. pectinatum. Spore Characters- Spore size and ornamentation are similar among the taxa in this group, which is not surprising in light of previous observations by A. Tryon (1991), who found little variation among the spores of closely related Adiantum species. Species in the A. peruviaum Group have trilete, tetrahedral-globose spores, with rugulate surfaces, irregular sheaths, and a size range of 29-53µm (Fig. 3.4).

TAXONOMIC TREATMENT

Species in the Adiantum peruvianum group can be recognized by having laminar tissue that lacks visible idioblasts between veins, clathrate rhizome scales with minutely denticulate margins; segment stalks that stop abruptly at segment bases; discreet sori that are depressed-ovate, reniform, or occasionally lunate; ultimate segments that are dimidiate, trapeziforme, deltate, or occasionally ovate-falcate; and free, dichotomous veins that end in marginal teeth.

38 KEY TO SPECIES

1. Stalks of medial ultimate segments more than 0.4 cm long………...………………....2

1a. Stalks of medial ultimate segments less than 0.4 cm long…………………...……….5

2. Mature lamina 1 pinnate; ultimate segments about as long as wide; sori orbicular- ovate, occasionally reniform, 0.14 – 0.46 cm long, 0.14 – 0.22 cm wide……… ………………………………………………………………………….A. ecuadorianum ined.

2a. Mature lamina 2-4 pinnate; ultimate segments mostly longer then wide; sori oblong-reniform, 0.23 – 0.74 cm long, 0.19 – 0.28 cm wide……….……………..…...3

3. Ultimate segments ovate to elliptic, often somewhat falcate, bases rounded, apices long tapering, attenuate-acuminate ……………………..…...... A. anceps

3a. Ultimate segments rhombic to dimidiate, bases acute, apices acuminate-acute, occasionally long tapering……………………………………………………....…..4

4. Ultimate segments elliptic to rhombic, weakly trapeziforme, 3.5 – 7.5 cm long, acroscopic corner rather distinctly rounded; stalks of medial ultimate segments 0.65 – 2.0 cm long………………………….…...... A. peruvianum

4a. Ultimate segments dimidiate to strongly trapeziforme, 2 – 5 cm long, acroscopic corner acute; stalks of medial ultimate segments 0.45 – 0.95 cm long.…………………………………………………...... A. trapeziforme

5. Rachis densely pubescent-scaly………………………...…………...... A. pectinatum

5a. Rachis entirely, or mostly glabrous………………………………………..………6 6. Rhizome less than 1cm diam, long-creeping; pinnae not or only slightly reduced in size toward apex…………………………...… ………………A. exuberans ined.

6a. Rhizome more than 1 cm diam, short-creeping; pinnae reduced in size toward apex………………………………………………………………………...…...…..7

7. Mature lamina 3-4 pinnate, ultimate segments 1.3 - 4.7 x 0.68 - 3.5 cm……………………………………………………………….A. mathewsianum

7a. Mature lamina 4-6 pinnate, ultimate segments 0.74 – 2.9 x 0.35 – 1.65 cm……………………………..………………………………………….A. polyphyllum

39 Adiantum anceps Maxon & C. V. Morton, Amer. Fern J. 24: 15. 1934. Type: Peru, Killip & Smith 24069 (holotype US!; isotype NY!).

Plants terrestrial. Rhizomes moderately long-creeping, 0.72-[1.16]-1.55 cm in diameter; scales dense, triangular to narrowly triangular, clathrate, concolorous, castaneous to stramineous, 0.4-[0.59]-0.76 X 0.05-[0.14]-0.25 mm at widest point, margins denticulate, apex attenuate to acuminate. Leaves monomorphic, triangular, 48.5- [80.5]-116 X 12-[28.3]-60 cm; stipes 40--70% the length of the fronds, atropurpureous to ebeneous, mostly glabrous, lustrous to weakly glaucous; stipe scales scattered proximally or absent, similar to rhizome scales; blades narrowly deltate-widely ovate, bipinnate- tripinnate, glabrous, occasionally glaucous; rachises atropurpureous to ebeneus, glabrous, occasionally glaucous; pinnae 3-8, alternate, ovate-lanceolate, basal pinnae 17-[27.6]- 40.5 x 8-[13.3]-25 cm, bearing 6--30+ ultimate segments, medial pinnae 16-[25]-38 x 6.5-[10.2]-18 cm, bearing 5--17 ultimate segments, apical pinnae 22-[27.4]-36 x 6-[11]- 17 cm, conform, bearing 8--12 ultimate segments; ultimate segments widely ovate to weakly falcate, articulate, much longer than wide, 3.9-[6.88]-9.78 X 2.43-[4.59]-7.51 cm, bases acute-obtuse, to round, sterile segments margins shallowly dentate, fertile segments crenate, apex concavely acuminate; terminal apical segment ovate-rhombiform/trullate, 5.5-[6.45]-7.9 X 2.5-[3.2]-4.3 cm; veins free, ending in marginal teeth, prominulous, lacking venuloid idioblasts between veins; segment stalks 0.56-[1.63]-3.4 cm long, shorter distally, darkened color stopping abruptly at segment bases; false indusia discrete, 9-[18]-37 per segment, depressed ovate, 0.16-[0.34]-0.68 X 0.08-[0.16]-0.2 cm, cinnamomeous when young, becoming atrocastaneous with age, margins erose, stramineous, becoming hyaline distally; spores yellow or yellow-brown, tetrahedral- globose, trilete, 36-[42.6]-51 µm; chromosome number unknown. Distribution and Habitat – This species occurs in the eastern lowlands of Peru, from northeastern Loreto, south through Huanuco with additional reports from Colombia, Ecuador (Tryon & Stolze, 1989), and Bolivia (Labiak & Prado, 2007) (Fig. 3.2 C). It grows from 300-1500 m in humid forests, along riverbanks, and occasionally on limestone cliffs.

40 Comments - Commonly known as the Double Edge Maidenhair, Adiantum anceps can be distinguished, by its widely ovate, and somewhat falcate segments with long, concavely acuminate apices and long segment stalks (Fig. 3.5). Similar, and geographically proximate species include, A. peruvianum, which has dimidiate segments, with blunt apices and acute to obtuse bases, and ‘A. ecuadorianum’, which is once pinnate with widely deltate, dimidiate segments. Adiantum trapeziforme also looks similar, but has considerably shorter segment stalks and smaller, strongly trapeziforme ultimate segments. Adiantum trapeziforme ranges from southeastern Mexico, eastern Central America, to Cuba, Jamaica, and the Lesser Antilles, well outside the range of A. anceps. Adianum platyphyllum and A. seemanii might be mistaken for A. anceps, but both have visible idioblasts between veins, and oblong rectangular indusia that are contiguous, whereas A. anceps has discrete, depressed ovate indusia. In A. platyphyllum, the dark color of the segment stalks passes into the segments, whereas in A. anceps the stalk color stops abruptly at the base of the segments.

Representative Specimens:

PERU: AMAZONAS REGION: Bagua. Terrestrial or on rocks, occasional near left bank of Rio Maranon opposite Campamento Stte, Montenegro, opposite Km. 278 of Maranon road, above Cascadas de Mayasi, 425 m, 7 September 1962, J. J. Wurdack 1885

(GH, MO, NY). CAJAMARCA REGION: Namballe, entre La Vega del Toro y Las Abejas. Bosque secundario con relictos de bosque primario, Bosque secundario, 04°58'46"S, 79°05'01"W, 800-1000 m, Peantes 4733, (F, NY). San Ignacio, Vega Del Toro, Bosque Primario, con tala selective, 04°58'46"S, 79°05'01"W, 800-1000 m, 7 December 1997, R.

Vasquez, R. Rojas, A. Pena & E. Chavez 25127 (MO, NY). CUSCO REGION:

Convencion, Quillouno, 850 m, 24 June 1961, C. Vargas 13556 (GH). HUANUCO

REGION: Pachitea, Honoria, Bosque Nacional de Iparia; region de "bosque seco tropical" (sensu Tosi, 1960) a lo largo del Rio Pachitea cerca del campamento Miel de Abeja, 1km arriba del pueblo de Tournavista o unos 20 km, ariba de la confluencia con el Rio Ucayali, alturna sobre el mar, 300-400 m, 25 October 1967, Jose Schunke 2262 (F, GH, NY, US). Tingo Maria in crevices of dry limestone cliffs, 680 m, 8 September 1956, R.

41 Tryon & A. Tryon 5288 (F, GH, MO, NY, US). JUNIN REGION: La Merced, Wooded valley, 700 m, 29 May 1929, E. Killip & A. Smith 23478 (NY, GH, F, US). Colonia Perene, dense forest, 680 m, 14 June 1929, E. Killip & A. Smith 25077 (F, NY, US). Chanchamayo, Rio Colorado, 23 km from San Ramon on San Ramon-Puente Paucartambo road, forest remnants, 75°30'W, 10°58'S, 760 m, 6 October 1982, D. Smith & R. Foster 2491 (MO). Tarma, 4 km north of La Merced, rich soil, wooded hillside,

750 m, 27 October 1956, R. Tryon & A. Tryon 5444 (F). LORETO REGION: Vicinity of

Aguaytia, 10 March 1972, T. B. Croat 20997, (MO). MADRE DE DIOS REGION: Cocha Nueva, Rio Manu, 10 km upstream from Coacha Cashu, 350 m, 25 January, 1987, P.

Nunez 6942 (F). PASCO REGION: Oxapampa, Iscozacin, near confluence of Rio Palcazu and Rio Iscosacin - Juan Frazen property, 75°13'W, 10°12'S, 430 m, 18 June 1982, D.

Smith 1957 (MO, NY). SAN MARTIN REGION: Lamas, along Rio Cumbaquiri, 548 m, No Collector Name 3408A (F, GH, MO, NY US). Tarapoto, San Antonio de Cumbaza, 800- 900 m, 15 April 1958, N. Vulgar & A. Sagastegui 0214 (GH). Lamas, Alonso de Alvarado, San Juan de Pacaizapa, km 72 carretera Tarapoto-Moyobamba, Al borde de la quebra da en bosque alto, 1000-1050 m, 9 June 1977, Jose Schunke 9670 (GH, MO, NY). Mariscal Caceres, quebrada de Cachiyacu de Lepuna, al borde de la quebrada debre recas humedas, 500-850 m, 21 July 1974, J. Schunke 7641 (MO, NY).

Adiantum ecuadorianum M. McCarthy & Hickey, ined. TYPE. ECUADOR. Pastaza province. Road - Puyo - Macas, km 79 at Rio Pastaza bridge, on sheltered rock wall, 77°50'W, 01°55'S, 600-650 m, 1 April 1998, B. Øllgaard & H. Navarrete 3035 (holotype: AAU!).

Plants terrestrial. Rhizomes decumbent, short-creeping, 0.4cm in diameter; scales abundant, triangular to narrowly triangular, clathrate, concolorous, castaneous to stramineous, 0.08-[0.21]-0.31 X 0.02-[0.05]-0.07 mm at widest point, margins denticulate, apex attenuate to acuminate, occasionally caudate. Leaves monomorphic, 25.5-[41.5]-56 X 10-[10.7]-12 cm; stipes 45--65% the length of the fronds, atropurpureous to ebeneous, mostly glabrous, lustrous to weakly glaucous; stipe scales

42 scattered proximally or absent, similar to rhizome scales; blades ovate-elliptic, pinnate, glabrous, occasionally glaucous; rachises atropurpureous to ebeneus, glabrous, occasionally glaucous; ultimate segments 9—10, widely deltate-dimidiate, articulate, about as long as wide, 3.65-[5.83]-7.38 X 2.64-[5.34]-7.2cm, membranaceous, bases acute-obtuse, sterile segment margins biserrate, fertile segments crenate, apex acute- cuspidate; terminal apical segment obtrullate, 3--5.75 X 2.6--3.25 cm; veins free, ending in marginal teeth, prominulous, lacking venuloid idioblasts between veins; segment stalks 0.2-[1.50]-2.35 cm long, shorter distally, darkened color stopping abruptly at segment bases; false indusia discrete, 2-[21.4]-36 per segment, depressed ovate, occasionally reniforme, 0.14-[0.26]-0.46 X 0.13-[0.18]-0.22 cm, atrocastaneous, margins undulate, stramineous, becoming hyaline distally; spores yellow or yellow-brown, tetrahedral- globose, trilete, 44-[47.9]-53 µm; chromosome number unknown. Distribution and Habitat – This species is known only from Pastaza Province in the eastern foothills of the Ecuadorian Andes. It grows on sheltered rock walls at 600-

650 m. Comments - Adiantum ecuadorianum ined. (Fig. 3.6) is distinguished by its short, pinnate fronds that reach about 60 cm, its large, widely deltate and dimidiate ultimate segments, and its long segment stalks. A. ecuadorianum ined. is a rare species; we have only seen one collection from Pastaza, Ecuador, where it was found growing on a sheltered rock wall. Adiantum ecuadorianum ined. looks similar to A. peruvianum, but can readily be distinguish by having short, once-pinnate laminae, widely deltate- dimidiate ultimate segments, thin creeping rhizomes, and small rhizome scales. Adiantum peruvianum has taller, bi-tripinnate laminae, with dimidiate ultimate segments, stouter rhizomes, and larger rhizome scales. Adiantum anceps looks similar but can be distinguished by having 1-3-pinnate fronds, with smaller, widely ovate, and somewhat falcate ultimate segments. Additionally, A. ecuadorianum ined. has thin, membranaceous ultimate segments, and short-creeping, thin rhizomes, whereas A. anceps and A. peruvianum have thicker, papyraceous ultimate segments with medium-creeping, stout rhizomes.

43 Adiantum exuberans M. McCarthy & Hickey, ined. TYPE: Nicaragua, Boaco department, rain forest on mountainside, 415 m, 24 January 1970, F.C. Seymour 3838 (holotype GH!; isotype MO!).

Plants terrestrial. Rhizomes decumbent, long-creeping, thin, 0.2-[0.36]-0.5 cm in diameter; scales dense, narrowly triangular to triangular, clathrate, concolorous- occasionally bicolorous, stramineous to castaneous, 0.15-[0.24]-0.3 X 0.09-[0.14]-0.22 mm at widest point, margins minutely denticulate becoming more pronounced towards the apex, apex attenuate to acute. Leaves monomorphic, 43-[70.08]-126 X 25-[33.5]-40 cm; stipes 40--60% the length of the fronds, atropurpureous to ebeneous, mostly glabrous, with scattered scales near the base, lustrous to weakly glaucous; stipe scales few at the base or absent, analogous to rhizome scales; blades ovate, to widely ovate, tripinnate-quadripinnate, sometimes pedate, glabrous, occasionally glaucous; rachises atropurpureous to ebeneus, glabrous, or occasionally with scales (Mexico taxa only), occasionally glaucous; pinnae 4-9, alternate, ovate-oblong, basal pinnae 15-[22.6]-29 x 15-[18]-21 cm, medial pinnae 14-[18.1]-23.5 x 6-[7.1]-9 cm, bearing 15--21 ultimate segments, apical pinnae 13.5-[16.7]-21.5 x 6.5-[7.25]-9 cm, bearing 12--23 ultimate segments; ultimate segments trapeziforme, occasionally tending towards falcate, articulate, much longer than wide, 3.4-[4.56]-6.37 X 0.5-[1.13]-1.63cm, bases acute- obtuse, acroscopically round, margins biserrate on sterile segments, crenate on fertile, apex long, concavely attenuate-acuminate; terminal apical segment rhombiform-trullate, 2.25-[3.64]-4.68 X 1.57-[2.38]-3.19 cm; veins free, ending in marginal teeth, prominulous, lacking venuloid idioblasts between veins; segment stalks 0.03-[0.23]-0.83 cm long, becoming nearly sessile distally, darkened color stopping abruptly at segment bases, glabrous, or occasionally with scales (Mexico taxa only); false indusia discrete, 2- [21.36]-36 per segment, reniforme-depressed obovate, 0.18-[0.27]-0.42 X 0.11-[0.17]- 0.22 cm, stramineous when young, becoming castaneous-atrocastaneous with age, margins undulate, becoming hyaline distally; spores yellow or yellow-brown, tetrahedral- globose, trilete, 29-[35.25]-41 µm, chromosome number unknown. Distribution and Habitat – This species occurs from Mexico, southward through Guatemala, Honduras, and Nicaragua to Costa Rica. It grows in moist, or dry tropical

44 forests, on riverbanks, rocky slopes, limestone ledges, or disturbed habitats, in shade, from sea level-2600m. Comments - Adiantum exuberans ined. (Fig. 3.7) is distinguished by its compact, decompound, imparipinnate-pedate fronds, trapeziforme, dimidiate ultimate segments with short segment stalks, and long-creeping, thin rhizomes. Adiantum exuberans ined. (Fig. 3.7) has long been mistaken for A. trapeziforme (Fig. 3.8), but morphology (Fig. 3.9; Table 3.1 and 3.2) and molecular data (Chapter 2) clearly indicate that these are two distinct taxa. Adiantum exuberans ined. is a mainland species occurring from Mexico to Costa Rica (Fig. 3.6 B). Adiantum trapeziforme is a species of Cuba and Jamaica, and in Mesoamerica from southern Mexico to Honduras (Fig. 3.6 A). The confusion surrounding these two species is likely due to their overlapping range in Mesoamerica, similar segment shape and morphological plasticity. In 1851, Hooker made the following note regarding A. trapeziforme “This fine species, from 2—4 feet high, has been much misunderstood, mainly owing to sufficient allowance not being made for those variations to which Ferns, in general, seem peculiarly liable”. In addition to displaying great morphological variation, interpretations of this species may be further confounded by the absence of an actual type specimen; Linnaeus based his description on an illustration (Fig. 3.10) by Sir Hans Sloane (1707: t. 59). The clear distinction between these two species is illustrated in the results from our analysis of variance (ANOVA), which showed significant differences (p-values ≤ .04) in the means of seven quantitative characters (Table 3.2), and evidence for two groups was also supported in the results from our principle component analysis (PCA) (Fig. 3.9). In our PCA the first principle component (PC1) accounts for 57% of the variation and PC2 accounts for 16%, for a total of 73% of the variation accounted for on these two components (Fig. 3.9). The characters with the greatest contribution to PC1 include, stalk length, rhizome diameter, distance between segment stalks, and ratio of sori; PC2 is most influenced by the number of segments per terminal pinna, lamina height, and segment width. In Adiantum exuberans ined. the ultimate segments are morphologically variable in size and shape, a potential source of confusion when differentiating it from A. trapeziforme. The latter taxon, however, tends to have wider ultimate segments, with longer ultimate segment stalks (Fig. 3.1). The single best

45 character for separating these two species is diameter of the rhizome: A. trapeziforme rhizomes are medium-creeping and stout, whereas those in A. exuberans ined. are long- creeping and much thinner. Additionally, A. trapeziforme has taller fronds, with more distantly positioned ultimate segments along the pinnae (Tables 3.1 and 3.2) Adiantum exuberans ined. is a rather variable taxon even after its isolation from A. trapeziforme. Ultimate segment shape and size varies widely, but its variation does not seem to coincide with other characters or geography. Some collections from the northern part of the range, Mexico and Guatemala, tend to have a weakly pedate laminar architecture, and/or pluricellular hairs and scales in the axils of the segment stalks; neither of these characters was observed in more southern Adiantum exuberans ined. collections. Additional research on the morphological variation in this species seems warranted. Adiantum mathewsianum looks similar to A. exuberans ined., but is restricted to South America, and has stout rhizomes, dimidiate ultimate segments with acute apices, and longer, more numerous pinnae (Table 3.1).

Representative Specimens:

BELIZE: CAYO DISTRICT: Vaca, on hillside, 375 m, 11 March 1938, Percy

Gentle 2342 (F, GH, MO, US). TOLEDO DISTRICT: Southern Maya mountains, Bladen nature reserve, dry, steep, rocky, limestone hill, 500-700 m, 15 May 1996, B. Holst 5342 (MO).

COSTA RICA: ALAJUELA PROVINCE: San Mateo, foret taillis, Surubres (pacific), 250 m, February 1906, P. Biolley 17 (GH, MO, NY). Ca. 7.5-8.5 km by road W of Ciudad Colon, ridge between Rio Virilla and Quebrada Micos, along road between Finca Micos and Llano Limon, in woods, 9°56’N, 84°18’W, 600 m, 16 January 1986, A. Smith,

T. Beliz, M. Grayum, P. Sleeper 1617 (AAU, MO, NY). CARTAGO PROVINCE: Turrialba,

09° 46'50"N, 83°23'45"W, 1150 m, 8 August, 1995, G. Herrera 8340 (F). GUANACASTE

PROVINCE: Garza, 15 m, March 1955, E. Scamman 7638 (GH, US). Canton de Bagaces, estacion biologica Palo Verde, junction of Cactus and Mirador trails, tropical dry forest, 10°21’00”N, 85°20’20”W, 5 m, 5 July 2002, Boyle, Holbrook, & Kirkpatrick 6457 (NY). Nicoya, Parque National Cabo Blanco, secondary forest, 09°36'00"N, 85°06' 00"W, 50

46 m, 30 April 1994, J. Bittner 2265 (AAU). In woods near Santa Cruz, 150 m, 14 February

1963, L. & T. Williams 24617 (F, US). PUNTARENAS PROVINCE: Near Lindora, Monte

Verde, 29 September 1970, H. Kennedy 569 (F). SAN JOSÉ PROVINCE: Rio Grande, 700 m, May 1901, A. Alfaro 46 (US).

EL SALVADOR: AHUACHAPÁN DEPARTMENT: Finca San Benito, nacimiento del rio Ahuachapillo, 750 m, 21 February 1979, R. Seiler 937 (F). Sierra de Apaneca, in the region of Finca Colima, 1220 m, 19 January 1922, P. Standley 20072 (US).

COJUTEPEQUE DEPARTMENT: Cerro de Los Pavos, 31 July, 1991, R. Seiler 1696 (MO).

CUSCATLAN DEPARTMENT: Ecocentro Cojutepeque (CESTA), 17 February 1999, S. Dar

3658 (MO). LA LIBERTAD DEPARTMENT: Laderas de la Laguna, Antiguo Cuscatlan, inner slopes of ancient volcanic crater, disturbed primary vegetation, 3 April, 1986,

Berendsahn 409 (MO). LA UNION DEPARTMENT: Cooperativa El Faro, Volcan

Conchagua, 29 November, 2000, A. Monro, J. Reyes, & J. Mauricio 3589 (MO). SAN

MIGUEL DEPARTMENT: San Rafael Chaletenango, common near stream, 350 m, 14

October 1977, R. Seiler 139 (F). SAN SALVADOR DEPARTMENT: Rio El Nacimiento near Canton la Fuente, Tonacatepeque, in deep shady barranca, 700 m, 18 December 1978, R. Seiler 785 (F). Vicinity of San Salvador, bank of Arenal, 750 m, 1 April 1922, P. Standley 22750 (GH, US). Republic del Salvador, 640 m, April 1906, L. V. Velaseo 8892

(US). SANTA ANA DEPARTMENT: Reserva las Lajas, disturbed forest on very steep inner slopes of volcano, canopy variable, in deep shade, 1030 m, 22 July, 2000, A. Monro & R. Douglas 3464 (MO, NY).

GUATEMALA: ALTA VERAPAZ DEPARTMENT: Tucuru, 5 April, 1939, P. Standley

70690 (F). ESCUINTLA DEPARTMENT: Medio monte, Palin, 30 September 1973, M.

Rivera 752 (F). JUTIAPA DEPARTMENT: 8 miles SW of San Cristobol along CA-2, near picnic area, along clear stream in canyon with tropical forest and large Ceiba trees, 640 m, 27 July 1979, D. Dunn, C. Dziekawanowski, S. Trott, & D. Thurm 23236 (NY).

PETÉN DEPARTMENT: Santa Eleana, about 8 km SE of village, 22 March 1961, E.

Contreras 2019 (F). RETALHULEU DEPARTMENT: Near Nueva Linda, halfway between

Retalhuleu and Champerico, 25 February 1941, P. Standley 88418 (F). SANTA ROSA

DEPARTMENT: Volcan Tecuamburro, 1224 m, February 1893, Heyde et Lux 4669 (F,

47 GH, MO, US). ZACAPA DEPARTMENT: Trail between Santa Rosalia de Marmol and Vegas, Sierra de las Minas, 1213 m, 19 January 1942, J. Steyermark 42962 (F, US).

HONDURAS: ATLÁNTIDA DEPARTMENT: 30 km south of Ceiba, 16 September

1972, M. Madison 707 (GH). BAY ISLANDS DEPARTMENT: Roatan Island, 1882, C. Peck

(GH). COLÓN DEPARTMENT: Trujillo, Rio Cristales Dam, lower slopes Puerto Arturo,

76 m, 3 January 1981, J. Saunders 795 (NY). COMAYAGUA DEPARTMENT:

Siguatepeque, 11 April 1987, M. Acosta 141 (F). COPAN DEPARTMENT: 2 miles SW of Rita, abundant on moist bank of Titoror Creek, 600 m, 27 August 1975, A. Molina 30733

(F). CORTÉS DEPARTMENT: En la foresta, bosque lluvioso de montana de Rio Piedras, zona de la presa del agua potable, 240 m, 4 December 1950, A. Molina 3539 (F, GH, US). Rio Lindo, dense tropical forest, 610 m, 27 September 1933, J.B. Edwards p-695

(F, GH). EL PARAÍSO DEPARTMENT: Between Las Mesas and Ojo de Agua, 700 m, 11

March 1951, C.V. Morton 6995 (US). FRANCISCO MORAZÁN DEPARTMENT: Drainage of the Rio Yeguare, 87°W, 14°N, 1000 m, 14 March 1951, A. Molina 3924 (F, GH).

MORAZAN DEPARTMENT: Along Rio Agua Amarilla, NW of El Zamorano, 27 July 1949,

P. Standley 21779 (F). OLANCHO DEPARTMENT: Cerro el Congrejo, 9 km O. de

campamento bosque humedo sub tropical, 21 September 1986, J. Euceda 55 (F). SANTA

BARBARA DEPARTMENT: Los Dragos SW of Quimistan, 14 April 1947, P. Standley & H.

Lindelie 7516 (F). VALLE DEPARTMENT: Slopes of El Tigre volcano, above Amapala, in

shady woods, volcanic soil, 16 September 1935, J. West 3545 (MO). YORO

DEPARTMENT: Near Progresso, 60 m, 8 March 1923, P. Ames 85 (US).

MEXICO: CHIAPAS STATE: Tuzantan, 15 km al NE de Huixtla, camino a Ejido J. M. Morelos, selva alta subperennifolia, 800 m, 9 February 1987, E. Martinez 19956 (F). Tonala, Cerro Bernal, ca. 25 km SE of Tonala, area of Microwave Station Quetzalapa, degraded deciduous forest on rocky slope, 500 m, 20 November 1984, G. Davidse 30135

(MO, NY). GUERRERO STATE: Acapulco and vicinity, 183 m, October 1894, Ed Palmer 519 (F, GH, MO). Montes de Oca, oak forest, 840 m, 14 October 1937, George Hinton

10802 (F, MO). HIDALGO-SAN LUIS POTOSI STATE: Jacala, km 343-344 on highway below Chapulhuacan, steep, wet, rocky slopes and limestone ledges, 640 m, June 1948,

H. H. Moore, C. E. Wood 3653 (GH). JALISCO STATE: Santa Cruz Vallarta, steep banks

of canyon, 300 m, 12 December 1926, Ynes Mexia 1297 (MO, F, GH). MÉXICO STATE:

48 Temascaltepec, Pantoja, dry barranca, 1790 m, 21 April 1935, G.B. Hinton 7644 (F, GH,

MO). MICHOACÁN DE OCAMPO STATE: Pinamo, 16 June 1892, M. Jones 501 (MO).

NAYARIT STATE: Ca. 1 km E-NE of La Bajada, 28 January 1994, T. & R. Devender 93-

49 (MO). OAXACA STATE: Pochutla, San Miguel del Puerto, rancho San Agustin, 740 m al SW sobre vereda a cerro la campana, selva mediana subperennifolia, cafetal, 15°59'24"N, 96°6'42"W, 740 m, 16 April, 2000, Feliciano Lopez 71 (NY). Pochutla, San Miguel del Puerto, finca el mamey, 1 km al W. de la casa, selva mediana subperennifolia,

cafetal, 15°58'27"N, 96°5'44"W, 870 m, 15 March 2001, Silvia Salas 3863 (NY). SAN

LUIS POTOSI SATE: 17 miles E of Tamanzunchale, ravine in tropical forest, 1830 m, 4 September 1948, Kenoyer and Crum 3983 (GH). Tamasopo, Sierra Madre Oriental, rocky limestone canyon west of Tamasopo, 450 m, 8 August 1934, Francis Pennell 17992 (GH). Tamasopo, coffee plantation, 527 m, 1 December 1891, C.G. Pringle 3960

(F, GH, MO, MU). TAMAULIPAS STATE: Sierra de Tamaulipas, region of Las Yucas, ca 40 km NW of Aldama, Los Cerritos, above Guadelupe Victoria (Laguna del Sapo), W of Las Yucas, mesic oak forest, 23°14'N, 98°10'W, 2636 m, 21 July 1957, Robert Dressler

1954 (GH, MO). VERACRUZ STATE: Roadside between Orizaba and Cordoba, 1010 m, 19 July 1950, David Potter 21374 (GH).

NICARAGUA: BOACO DEPARTMENT: Rain forest on mountainside, 415 m, 24

January 1970, F.C. Seymour 3838 (GH, MO). CARAZO DEPARTMENT: Jinotepe, 500 m,

4 November 1911, A. S. Hitchcock (US). CHINANDEGA DEPARTMENT: Chichigalpa, Volcan Casita, Filete El Venado, 5 February 2004, I. Coronado, N. Gutierrez, & M. Soto

491 (MO). CHONTALES DEPARTMENT: Cuapa, 9 December 1973, S. Marshall & D. Neill

6677 (MO, NY). ESTELI DEPARTMENT: Salto de Estanzuela, Rio Estanzuela ca 6 km S

of Esteli, 1 October 1979, W. Stevens, B. Krukoff 14415 (MO). GRANADA DEPARTMENT: Mombacho, extinct volcano, in lush growth on mountain, 800 m, 9 January 1969, D.

Dudley, & A. Moore (GH, MO). MADRIZ DEPARTMENT: Municipio de Somoto, Reserva Natural Tepesomoto, La Patasta, 17 May 2000, R. Rueda, D. Paguaga, & H. Mendoza

13594 (MO). MANAGUA DEPARTMENT: Sierra de Managua, A. Garnier 480 (GH).

JINOTEGA DEPARTMENT: Vicinity of Jinotega, 9 September 1947, P. Standley 10055 (F).

MASAYA DEPARTMENT: Sierra de Managua, region of Las Nubes, in moist, partly forested quebrada, 840 m, 14 May 1947, P. Standley 8060 (F, US). Sierra de Managua, E

49 slope, understory of tropical rain forest, 800 m, 6 January 1941, V. Grant 1033 (GH).

MATAGALPA DEPARTMENT: In ciliar forest along Rio Malacal, about 20 km NE of

Matagalpa, 350 m, 16 January 1963, L. O. Williams 23972 (F, US). NUEVA SEGOVIA

DEPARTMENT: Ca 5.2 km N of San Fernando, valley of Rio San Fernando, NE to Portillo Los Coyoles, SW up narrow quebrada to Cerro El Penascal, 13 August 1977, W. Stevens,

B. Krukoff 3199 (MO). RÍO SAN JUAN DEPARTMENT: La Pimienta, Siuna, 1 June 1984,

F. Ortiz 1965 (MO). RIVAS DEPARTMENT: Island Ometepe, Lake Nicaragua, January

1893, C. Smith 2134 (GH, MO). UKNOWN DEPARTMENT: Nicaragua, 1853-1856, C. Wright & Ringgold (GH).

Adiantum mathewsianum Hook. Species Filicum 2: 35. 1851. Type: Chachapoyas, Peru, Mathews 3296 (K - photo US!, fragment s.n. ex K - NY!).

Plants terrestrial. Rhizomes decumbent, medium-creeping, stout, 0.62-[1.35]-2.23 cm in diameter; scales dense, linear triangular, clathrate, concolorous-occasionally bicolorous, castaneous to stramineous, 0.25-[0.74]-1.07 X 0.1-[0.11]-0.13 mm at widest point, margins minutely denticulate, apex attenuate. Leaves monomorphic, 57-[101.5]- 150 X 22-[33.75]-51 cm; stipes 40--60% the length of the fronds, atropurpureous to ebeneous, mostly glabrous, with scattered scales near the base, lustrous to weakly glaucous; stipe scales few at the base or absent, analogous to rhizome scales; blades triangular-widely ovate, tripinnate-quadripinnate, glabrous, occasionally glaucous; rachises atropurpureous to ebeneus, glabrous, rarely with scattered scales, occasionally glaucous; pinnae 5-11, alternate, triangular-lanceolate, tapering apically, basal pinnae 21- [40]-59 x 7-[20.4]-36 cm, medial pinnae 16.5-[26.7]-44 x 4-[8.4]-25 cm, bearing 20-37 ultimate segments, apical pinnae 16.5-[21.25]-25 x 6-[6.3]-6.5 cm, bearing 21-36 ultimate segments; ultimate segments dimidiate-trapeziforme, articulate, longer than wide, 1.3-[2.89]-4.74 X 0.68-[1.25]-3.5cm, bases acute-obtuse, margins with minute teeth on sterile segments, crenate on fertile, apex acute-attenuate; terminal apical segment rhombiform-trullate, 2-[2.93]-4.05 X 1.12-[1.4]-2.1 cm; veins free, ending in marginal teeth, prominulous, lacking venuloid idioblasts between veins; segment stalks 0.07- [0.25]-0.55 cm long, shorter distally, darkened color stopping abruptly at segment bases;

50 false indusia discrete, 8-[13]-20 per segment, reniforme-lunate, 0.1-[0.25]-0.45 X 0.11- [0.16]-0.19 cm, stramineous-cinnamomeous when young, becoming atrocastaneous with age, margins undulate, stramineous, becoming hyaline distally; spores yellow or yellow- brown, tetrahedral-globose, trilete, 40-[44.9]-50 µm, chromosome number unknown. Distribution and Habitat – This species occurs from southeastern Brazil (Matto Grosso do Sul) westward into Paraguay and then north into central Bolivia with disjunct populations in southern Ecuador (Zamora-Chinchipe) and northwestern Peru (Amazonas and Cajamarca). It grows in humid forests, along riverbanks and on limestone outcrops, from 200-1700m. Comments - Adiantum mathewsianum (Fig. 3.11) can be distinguished by its rigid, dimidiate-trapeziforme ultimate segments with acute apices, and rigid, short segment stalks. Adiantum polyphyllum looks similar, but has 4-6-pinnate fronds (A. mathewsianum has 3-4-pinnate fronds), with smaller ultimate segments, shorter segment stalks, and occurs in Trinidad, Venezuela, and Colombia, outside the range of A. mathewsianum. Adiantum trapeziforme and ‘A. exuberans’ also look similar, but the former has longer segment stalks and the latter has long-creeping, thin rhizomes; both tend to have longer ultimate segments (Fig. 3.1) with concavely attenuate-acute apices and neither occur in South America. In our molecular analyses (Chapter 2), the Adiantum mathewsianum collection from Bolivia came out separately on the phylogeny, indicating that A. mathewsianum may consist of multiple species. Upon closer examination of these taxa, the Bolivian collections have ultimate segments that are generally smaller, imbricate, and compactly arranged along the pinna rachises, whereas the ultimate segments in the other collections from Brazil, Ecuador, Paraguay, and Peru are typically larger, discrete, and more distantly positioned along the pinna rachises. Other than this, we were unable to find any significant morphological differences, and therefore do not feel comfortable segregating these taxa currently; additional research on this species is certainly warranted. The accession from Bolivia is denoted on the phylogeny in Chapter 2 (Fig. 2.1) as taxon 1.

Representative Specimens:

51 BOLIVIA: SANTA CRUZ DEPARTMENT: Andres Ibanez, Monumento Natural Espejillos, en el bolson, siguiendo la ribera del rio, bosque chiquitano, 17°54'07"S, 63°25'56"W, 23 July 2007, A. Molina 305 (MO).

BRAZIL: MATO GROSSO DO SUL STATE: Bonito, Ilha do Padre, Rio Formoso, Regiao da Serra de Bodoquena, com grutas, cerrado, mata seca semi decidua de encosta e mata ciliar, terrestre, barranco umido sombreado, margem do rio, grande populacao, 56°28'W, 21°08’S, 500-600 m, 8 April 1994, M.R. da Silva, & C.E. Rodrigues 1332 (MO). Rodovia Bonito, Campos dos Indios, Fazenda Espora de Prata, barrancos de beira

do rio, 11 August 2002, G. & M. Hatschbach, & J. M. Silva 73878 (MO). SAO PAULO

STATE: Monte Alto, 12 June 1992, M. R. da Silva, & C. E. Rodrigues 258 (MO).

ECUADOR: ZAMORA-CHINCHIPE PROVINCE: Along road between Zumba and La Balsa, at Rio Blanco Canchis, along border with Peru, 04°58'32"S, 79°06'50" W, 27 July 2004, T. Croat 92355 (MO, NY).

PARAGUAY: CONCEPCION DEPARTMENT: Estancia Lapuri, between estancia Arrecife and road to Valle Mi Cerrado, in forest on granite, 300 m, 15 January 2000, E.

Zardini 53826 (NY). DE LA CORDILLERA DEPARTMENT: Estancia la Maria, 1 February

1898, J.D. Anists 2869 (US). UNKNOWN DEPARTMENT: In regione calcarea curses superioris fluminis Apa., 1 March 1913, E. Hassler 11045 (GH, NY, US).

PERU: AMAZONAS REGION: Bagua, rain forest on left bank of Rio Maranon opposite Quebrada Mirana, opposite km 277 of Maranon road above Cascadas de Myasi,

425-450 m, 16 September 1962, J. Wurdack 2024 (GH, NY, US). CAJAMARCA REGION: San Ignacio, Namballe, las melgas, bosque secundario, 04°58'46"S, 79°05'01"W, 1700 m, 8 December 1997, J. Campos, A. Pena & E. Chavez 4778 (MO, NY). San Ignacio, Namballe, entre La Vege del Toro y Las Abejas, bosque secundario con relictos de bosque primario, 04°58'46"S, 79°05'01"W, 800-1000 m, 12 July 1977, J. Campos, A. Pena & G. Pezantes 4733 (MO, NY).

52 Adiantum pectinatum Kunze ex Baker. Synopsis Filicum 120. 1867. Type: Peru, Burchell 7416 (syntype K digital photo!). Plants terrestrial. Rhizomes, medium-creeping, stout, 1.1--1.65 cm in diameter; scales dense, narrowly triangular to linear triangular, obscurely clathrate, concolorous- occasionally bicolorous, stramineous to castaneous, 0.12-[0.41]-0.6 X 0.03-[0.05]-0.09 mm at widest point, margins minutely denticulate, apex attenuate. Leaves monomorphic, 40--83 X 37--50 cm; stipes 40-60% the length of the fronds, ebeneous, lustrous to weakly glaucous, pubescent-scaly, scales most abundant near base; stipe scales ciliform or narrowly triangular; blades widely ovate, quadripinnate; rachises ebeneus, abundantly pubescent-scaley, hairs pluricellular, stramineous, scales similar to stipe scales; pinnae 8- 20, alternate, widely elliptic-narrowly oblong, basal pinnae 22--37 x 14--18 cm, medial pinnae 18--21 x 3-5 cm, bearing 43--50 ultimate segments, apical pinnae 12--14 x 3.5--5 cm, bearing 35--50 ultimate segments; ultimate segments dimidiate, articulate, longer than wide, 0.76-[1.49]-2.08 X 0.26-[0.6]-0.7 cm, bases acute, margins deeply incised, apex acute, blunt-rounded; terminal apical segment rhombiform-flabellate, 1.2-[1.25]- 1.32 X 0.7-[0.75]-0.8 cm; veins free, ending in marginal teeth, impressed, lacking venuloid idioblasts between veins; segment stalks 0.03-[0.07]-0.16 cm long, becoming sessile distally, darkened color stopping abruptly at segment bases; false indusia discrete, 3-[5.5]-8 per segment, ovate-lunate, 0.05-[0.1]-0.15 X 0.05 cm, stramineous- cinnamomeous, margins undulate; spores not seen; chromosome number unknown. Distribution and Habitat – This species occurs in Tarija, Bolivia in the premontane forests of Tucumano, to 900 m. Also from Costa Rica, Peru, Brazil, and Argentina (Lellinger, 1989). Comments - Adiantum pectinatum (Fig. 3.12) can readily be distinguished by its pubescent-scaley rachises, small, dimidiate ultimate segments with blunt apices, and nearly sessile segment stalks. Adiantum polyphyllum looks similar, but has larger and fewer ultimate segments, lacks pubescent rachises, and occurs in Trinidad, Venezuela and Colombia. ‘Adiantum exuberans’ might be confused with A. pectinatum, but it also has larger ultimate segments that are strongly trapeziforme, longer segment stalks, and long- creeping, thin rhizomes.

53 Representative Specimen:

BOLIVIA: TARIJA DEPARTMENT: Arce, Municipio Padcaya, Reserva Nacional de Flora y Fauna Tariquia, quebrada grande hacia Playa Ancha, bosque premontano tucumano boliviano, 22°26'22.32"S, 64°20'22.45"W, 900 m, 29 January 2005, M. Serrano, J. Villalobos, A. Lliully, I. Guachalla, & R. Leon 5890 (MO).

Adiantum peruvianum Klotzsch, Linnaea 18: 555. 1845. Type: Peru, Ruíz 25 & 27 (B, Frag. ex B, US!, NY!).

Plants terrestrial. Rhizomes 0.55-[0.89]-1.4 cm in diameter; scales dense, triangular to narrowly triangular, clathrate, concolorous, castaneous to stramineous, 0.32-[0.54]- 0.75 X 0.12-[0.18]-0.25 mm at widest point, margins denticulate, apex attenuate to acuminate. Leaves monomorphic, 36-[85.66]-122 X 17-[29.84]-39 cm; stipes 30--60% the length of the fronds, atropurpureous to ebeneous, mostly glabrous, lustrous to weakly glaucous; stipe scales scattered at base or absent, similar to rhizome scales; blades deltate-ovate, bipinnate-tripinnate, glabrous, occasionally glaucous; rachises atropurpureous to ebeneus, glabrous, occasionally glaucous; pinnae typically 4-8, alternate, oblong-ovate, basal pinnae 14.5-[25.7]-44.5 x 7.5-[10.3]-18 cm, bearing 9--30+ ultimate segments, medial pinnae 12.3-[24.8]-41 x 6-[7.9]-12 cm, bearing 8--23 ultimate segments, apical pinnae 17.8-[25.9]-40.6 x 6.35-[8.35]-11.4 cm, bearing 10--19 ultimate segments; ultimate segments dimidiate, articulate, slightly longer than wide, 4.2-[4.89]- 7.7 X 2.1-[3.81]-6.2 cm, bases acute-obtuse, sterile segment margins shallowly dentate, fertile segments crenate, apex acuminate-acute, blunt; terminal apical segment elliptic- rhombiform, 3.7-[5.15]-7.68 X 2.0-[3.21]-4.45 cm; veins free, ending in marginal teeth, prominulous, lacking venuloid idioblasts between veins; segment stalks 0.3-[0.86]-2.6 cm long, shorter distally, darkened color stopping abruptly at segment bases; false indusia discrete, 3-[18]-28 per segment, widely depressed ovate, 0.15-[0.36]-0.85 X 0.10-[0.19]- 0.26 cm, stramineous-cinnamomeous when young, becoming atrocastaneous with age, margins undulate, becoming hyaline distally; spores yellow or yellow-brown, tetrahedral- globose, trilete, 32--52 µm, chromosome number N= 30 (Mahabale & Kamble 1981)

54 Distribution and Habitat – This species occurs from southernmost Ecuador, through Peru to Bolivia. It grows in wet forests, disturbed habitats, and on limestone outcrops, from 500-2800m. Comments – Commonly known as the Silver Dollar Fern, Adiantum peruvianum (Fig. 3.13) can be distinguished by its large, dimidiate, to somewhat elliptic, ultimate segments with acute, blunt apices, acute bases, and long segment stalks. Other species that occur in overlapping ranges and look similar include Adiantum anceps (Fig. 3.5), which has somewhat falcate segments that are widely ovate, with long tapering apices, rounded bases, and longer segment stalks, and ‘A. ecuadorianum’ (Fig. 3.6), which is pinnate, with widely deltate-dimidiate segments. Adiantum trapeziforme (Fig. 3.8) also looks similar, but has strongly trapeziforme, rectangular ultimate segments, with slightly longer, concavely acuminate apices. The distribution of A. trapeziforme ranges from southeastern Mexico, eastern Central America, to Cuba, Jamaica, and the Lesser Antilles, well outside the range of A. peruvianum. Adianum platyphyllum and A. seemanii look similar as well, but both differ in having visible idioblasts present between veins. In A. platyphyllum the dark color from the segment stalks passes into the segments, and in A. seemanii the stalk color stops abruptly at the base of the segments. Both A. platyphyllum and A. seemanii have oblong rectangular indusia that are contiguous, whereas A. peruvianum has widely depressed ovate indusia that are discrete. In our molecular analyses (Chapter 2), our Adiantum peruvianum collections from Ecuador and Bolivia came out in two separate places on the phylogeny, indicating that A. peruvianum may consist of multiple species. Upon closer examination of these taxa we found some individuals with pinnae that are gradually reduced towards the apex, while in others, the pinnae are not obviously reduced towards the apex. This condition can be found among individuals from Ecuador, Bolivia and Peru, and may not be of any significance. Additionally, we encountered many aborted spores in some individuals from Ecuador, which are somewhat oddly shaped and black, suggesting perhaps, cryptic hybridization. Other than this, we were unable to find any morphological differences, and therefore do not feel compelled to segregate these taxa; additional research on this species is warranted.

55

Representative Specimens:

BOLIVIA: BENI DEPARTMENT: Florida, Cerro Herradura, 3 km NE de Bermejo, carretera Antigua Santa Cruz – Cochabamba, vegetacion arbustiva, terreno escarpado, pedregoso, crece en lugares humedos y sombredos, 18°07' S, 63°37" W, 1450 m, 25

September 1990, Israel Vargas 750 (MO). SANTA CRUZ DEPARTMENT: Andres Ibanez, Monumento Natural Espejillos, de la catarata principal hacia ariba siguindo la ribera del rio, bosque chiquitano, 17°54'07"S, 63°25'56"W, 525 m, 15 July 2007, A. Molina 272 (MO). Espejillos 20 km SO de Santa Cruz, foret de ravin bordant la riviere et les chutes, 17°59' S, 63°21'W, 550 m, 16 November 1993, F. Billiet & B. Jadin 6187 (MO). Andres Ibanez, 6.5 km NW of hwy, bridge over Rio Pirai, along quebrada salada and oil pipeline, deciduous dry forest, steep slopes of Andean foothills, 18°6' S, 63°30.7' W, 840 m, 27 March 2002, M. Sundue & M. Nee 523 (NY, US). Andres Ibanez, "Espejillas", a waterfall of a small tributary of the Rio Espejillas, ca. 12 km W of La Guardia, subtropical semideciduous forest on calcareous hills and around waterfall, 17°53'S, 63°23'W, 500-600 m, 21 February 1987, M. Nee 34220 (NY).

ECUADOR: MORONA-SANTIAGO PROVINCE. Road Patuca - Santiago, km 8.4 from Santiago. Rocky forested ravine, 78°05'W, 03°01'S, 480 m, 19 March 1997, Ben

Ollgaard, H. Navarrete 2468 (AAU). SANTIAGO-ZAMORA PROVINCE. ("Oriente"): Eastern slopes of the Cordillera, valley of the Rios Negro and Chupianza (on trail from Seville de Oro to Mendez). El partidero, 640-945 m, 14 December 1944, W. H. Camp

1510 (F, NY). ZAMORA-CHINCHIPE PROVINCE. Nearest town Shaime, canton nangaritza, pueblo Shaime on Rio Nangaritza, hill west of Shaime, 04°18'47"W, 78°29'59"S, 900 m, 20 September 2000, J. P. Janovec, A. K. Neil, W. Quizhpe, I. Toborg 1383 (MO, NY).

PERU: AMAZONAS REGION: Mendoza, along Mendoza-Arenal road, heavily disturbed vegetation, 1500 m, 12 April 2001, H. Van der Werff, R. Vasquez, B. Gray, R.

Rojas 16984 (F, MO, NY) & 17004 (MO, NY). CAJAMARCA REGION: San Ignacio, San Jose de Lourdes, El Crucero, bosque interviendo, 1200 m, 8 June 1999, C. Diaz & S.

Flores 10666 (MO). CUSCO REGION: Santa Rosa, Urubamba Valley, 1200 m, 8 July

1915, O. F. Cook, & G. Gilbert 1714 (F). HUANUCO REGION: Gorge of Rio Chinchao, 5 km above junction with Rio Huallaga, 60 km NE of Huanuco, in crevices of limestone

56 cliffs, 1000 m, 12 September 1956, Rolla & Alice Tryon 5313 (F, MO, US). Chinchao highway to Puente Durand, 2000 m, 12 December 1953, Pedro S. Coronado 81 (US). Monzon confluenciacon el Huallga, cerca a Tingo Maria, selva tropical, 700-750 m, 23 September 1954, Ramon Ferreyra 10237 (GH). Puente Durand, north of Huanuco, Valley of Chinchao Rio, deep jungle, rich black humus, "fern fronds reach 1 m; black shiny stems. Common here", 1100 m, 17 October 1938, H. E. Stroke, & O. B. Horton 9448 (F, GH, US). Tingo Maria, damp, shady limestone ravine, in humus, among rocks,

700 m, 30 August 1956, Tryon & Tryon 5236 (F, MO). JUNIN REGION: Tarma, cerca al puente San Felix, selva tropical, 1000 m, 28 October 1965, Emma Cerrate 2869 (GH). San Roman, along bank among fallen trees and humus, very damp, about 200 feet above road on left, 15 km from San Ramon towards Tarma, 1400 m, 14 August 1960, S. E. Saunders 574-A (F). Chanchamayo, about 5 km beyond San Luis de Shuaro towards

Oxapampa, 760 m, 14 August 1960, S.G.E. Saunders 575 (F). PASCO REGION: Oxapampa, Pozuzo District, Huampal, pastizal y chacra abandonada, helecho terrestre, soros marrol en obscuros, 10°11'S, 75°34'W, 1250 m, 21 September 2002, A. Manteagudo, C. Mateo, G. Ortiz 3965 (NY). Oxapampa, Pozuzo District, Parque Nacional Yanachaga chemmillen carretera cerca a la Quebrada Misho, bosque primerio, helecho soros marginales negros, 10°10"S, 75°34'W, 1210 m, 14 April 2003, A. Monteagudo, G. Ortiz, & R. Francis 5009 (NY). Oxapampa, along road Oxapampa-La Merced, Quebrada Honda, 10°42'S, 75°21'W, 1700 m, 26 June 2003, H. van der Werff, B.

Gray, R. Ortiz, & N. Davila 17781 (MO, NY). SAN MARTIN REGION: Along road Rioja- pedro Ruiz, in area with limestone, ferns on bare limestone, or nearly so. 05°40'27" S, 77°40'35"W, 1170 m, 23 March 1998, Henk Van der Werff, B Gray, R. Vasquez, R. Rojas 15474 (MO, NY) Rio Huallaga Canon, below Rio Santo Domingo, in crevice on vertical canyon wall, 1200 m, 3 June 1923, J. Francis Macbride 4215 (F, US). Tingo Maria, at 25 km, on road to Lima from Tingo Maria, along Huallaga River, 750 m, 30 October 1950, H.A. Allard 21511 (F, US).

57 Adiantum polyphyllum Willd. Species Plantarum. 5: 454. 1810. Type: Venezuela, Bredemeyer s.n. (holotype B digital photo!)

Plants terrestrial. Rhizomes decumbent, short-creeping, stout, 0.67-[1.44]-2.22 cm in diameter; scales dense, triangular to narrowly triangular, clathrate, concolorous- occasionally bicolorous, castaneous to stramineous, 0.26-[0.41]-0.72 X 0.02-[0.04]-0.06 mm at widest point, margins denticulate, apex attenuate. Leaves monomorphic, 52- [99.7]-147.5 X 27-[40]-49 cm; stipes 40--60% the length of the fronds, atropurpureous to ebeneous, mostly glabrous, with scattered scales near the base, lustrous to weakly glaucous; stipe scales few-many at the base, analogous to rhizome scales; blades triangular-widely ovate, quadripinnate-6-pinnate, conform, glabrous, occasionally glaucous; rachises atropurpureous to ebeneus, glabrous or with minute acicular hairs, occasionally glaucous; pinnae 7-16, alternate, lanceolate-ovate, basal pinnae 25-[39.5]-50 x 12.5-[21.3]-35 cm, medial pinnae 17.5-[29]-46 x 4-[12.7]-27.5 cm, bearing numerous ultimate segments, apical pinnae 13.7-[18.1]-22.5 x 3--5 cm, bearing 25--52 ultimate segments; ultimate segments dimidiate, articulate, longer than wide, 0.74-[1.8]-2.85 X 0.34-[0.78]-1.66 cm, bases acute, margins incised with minute teeth on sterile segments, crenate on fertile, apex acute-rounded; terminal apical segment rhombiform-trullate, 0.73-[1.47]-2.13 X 0.38-[0.76]-1.15 cm; veins free, ending in marginal teeth, impressed, lacking venuloid idioblasts between veins; segment stalks 0.01-[0.09]-0.23 cm long, becoming sessile distally, darkened color stopping abruptly at segment bases; false indusia discrete, 1-[8.4]-12 per segment, round-reneform, 0.09-[0.19]-0.36X 0.1-[0.16]- 0.22 cm, stramineous when young, becoming atrocastaneous with age, margins undulate, becoming hyaline distally; spores yellow or yellow-brown, tetrahedral-globose, trilete, 36-[39.9]-44 µm, chromosome number unknown. Distribution and Habitat – Adiantum polyphyllum is known from Trinidad, Venezuela, Colombia, and is also found in Guatelmala (Davidse et al., 1995). It grows in dry tropical forests, on rocky, forested slopes, cliffs, and riverbanks, from 200-2400 m. Comments - Adiantum polyphyllum (Fig. 3.14) can be distinguished by its tall, 4- 6-pinnate laminae, numerous, dimidiate ultimate segments with rounded apices, and short-sessile segment stalks. Adiantum mathewsianum (Fig. 3.11) looks similar, but is 3-

58 4-pinnate, has larger ultimate segments with longer segment stalks, and occurs from Ecuador to Paraguay. Adiantum pectinatum (Fig. 3.12) also looks similar, but can be distinguished by its smaller fronds, with pubescent, scaly stipes and rachises.

Representative Specimens:

COLOMBIA: GUAJIRA DEPARTMENT: Municipio Fonseca, corregimiento distraccion, borde de carretera entre Dos Caminos y Los Gorros, 1 September 1990, O.

Marulanda, J. Betancur 2174 (MO). MAGDALENA DEPARTMENT: About 7 km E of Codazzi, crevice in sandstone cliffs in forest, 300 m, 21 October 1943, Oscar Haught 3763 (GH, US). 8 km east of Codazzi, dry hillside, very sandy soil, 800 m, 19 November 1943, Oscar Haught 3856 (US). Forest near Las Vegas, finca El Recuerdo's (Santa Marta Mountains), 762 m, 6 February 1936, Doras Bennett 5 (NY, US). Cordillera Oriental, Sierra de Perija, Esperitu Santo Valley, 12 km E. of Codazzi, 13 km from the Venezuelan border, subtropical forest, 1250 m, 11 February 1945, Martin Grant 10894 (US). Santa Marta mountains, El Recuerdo, trail beyond falls, on slopes of damp shaded

ravine, 2 August 1926, Ernestine Niemeyer 39 (US). TOLIMA DEPARTMENT: 25 km E of Cajamarca (road to Ibague), forest remnants along steep creekside, 1 October 1974, K. Kramer 1812 (GH).

TRINIDAD: 17 September 1877, Fendler 71 (F, GH, NY).

VENEZUELA: BARINAS STATE: Along hwy 1 between Barinas and Merida, 8 km N of Barinitas, rocky gorge and steep rocky forested slope S of hwy, and 3.5 km N of gas station at NW edge of Barinitas, 8°47'N, 70°27'W, 450 m, 6 August 1982, Thomas Croat 54751 (MO). Pedraza, trail from Mesa de Canagua to Alto de la Aguada, ca. 23 km NW of Curbati, growing on road bank, 8°37'N, 70°40'W, 800-1400 m, 17 April 1988, L.

Dorr, L. Barnett, C. Aymard, N. Cuello, & G. Diggs 4778 (NY, US). PORTUGUESA

STATE: Guanare, desembocaderos, en la via a Biscucuy, terrestre en sitios humedos, 23

October 1981, F. Ortega & G. Aymard 1392 (NY). MIRANDA STATE: Quelbrado de Turumo near Guarenas, in shady places, 2 December 1923, H. Pittier 11273 (GH, US).

SUCRE STATE: El Guayabito, along the Rio Guayabo at its juncture with the Rio Zumbador (future basin of Represa Neveri), forested riverbank interspersed with clearings, 64°17'W, 10°10"N, 230-250 m, 20 November 1981, G. Davidse, & A.

59 Gonzalez 19110 (MO). TRUJILLO STATE: Bocono, La Morita, cafetal and forest above the Rio Saguaras, Trujillo-Portuguesa state line, 9°22'N, 70°02'W, 2300 m, 4 June 1988, L. Dorr, L. Barnett, C. Aymard & N. Cuello 5362 (NY).

Adiantum trapeziforme L. Species Plantarum 2: 1097. 1753. Lectotype: Tab. 59 in Sloane, Voy. Jamaica. 1707, designated by Lellinger (Proc. Bio. Soc. Wash. 89:704. 1977).

Plants terrestrial. Rhizomes decumbent, medium-creeping, stout, 0.75-[1.08]-1.5 cm in diameter; scales dense, narrowly triangular, clathrate, concolorous, ferrugineous, 0.36- [0.59]-0.8 X 0.2-[0.29]-0.42 mm at widest point, margins sparsely denticulate, becoming more numerous and pronounced apically, apex attenuate. Leaves monomorphic, 41- [96.7]-159 X 27-[31.1]-37 cm; stipes 40--60% the length of the fronds, atropurpureous to ebeneous, mostly glabrous, with scattered scales near the base, lustrous to weakly glaucous; stipe scales few at the base or absent, analogous to rhizome scales; blades triangular-widely ovate, bipinnate-quadripinnate, glabrous, occasionally glaucous; rachises atropurpureous to ebeneus, glabrous, occasionally glaucous; pinnae 3-10, alternate, oblong-ovate, basal pinnae 10-[22.2]-33.5 x 12-[22]-28 cm, medial pinnae 12.5-[18.7]-27 x 6.5-[8.3]-10, bearing 9-14 ultimate segments, apical pinnae 14.5-[23]-30 x 5.5-[7.7]-10.5cm, bearing 12-17 ultimate segments; ultimate segments trapeziforme- widely trullate, articulate, much longer than wide, 2.2-[4.48]-5.4 X 0.86[1.72]-2.57cm, bases acute-obtuse, margins biserrate on sterile segments, crenate on fertile, apex acuminate-acute; terminal apical segment trullate-widely obtrullate, 3.35-[5.13]-7.5 X 2- [3.27]-4.68 cm; veins free, ending in marginal teeth, prominulous, lacking venuloid idioblasts between veins; segment stalks 0.2-[0.71]-1.85 cm long, shorter distally, darkened color stopping abruptly at segment bases; false indusia discrete, 1-[13.2]-22 per segment, reniforme-transversely oblong, 0.14-[0.35]-0.67 X 0.11-[0.19]-0.24 cm, cinnamomeous when young, becoming atrocastaneous with age, margins undulate, becoming hyaline distally; spores yellow or yellow-brown, tetrahedral-globose, trilete, 31-[36.9]-40 µm, chromosome number unknown.

60 Distribution and Habitat – This species occurs in southeastern Mexico and eastern Central America, to Cuba, and Jamaica. Adiantum. trapeziforme is also widely cultivated, and as a result, has become naturalized in southern Florida, Dominica, Martinique, and Sri Lanka. Reports of A. trapeziforme in South America, cannot be substantiated, although, several collections identified as that species were in fact specimens of A. mathewsianum. This latter species is distinguished by its smaller - and more numerous, dimidiate ultimate segments with short segment stalks (Fig. 3.1), and its South American distribution (Fig. 3.2). Adiantum trapeziforme grows in rich forests, along riverbanks, or on shaded, rocky limestone hillsides from sea level to 1000 m. Comments – Commonly known as the Diamond Maidenhair, Adiantum trapeziforme (Fig. 3.8) can be distinguished by its large, long-stalked and strongly trapeziforme, ultimate segments (Fig. 3.1), with long segment stalks, and its medium- creeping, stout rhizomes. ‘Adiantum exuberans’ (Fig. 3.7) resembles A. trapeziforme, but can be distinguished from it by having smaller ultimate segments, with shorter segment stalks, and long-creeping, thin rhizomes.

Representative Specimens:

BELIZE: CAYO DISTRICT: Cohune Ridge, secondary forest, 535 m, June 1936, C. L. Lundell 6498 (US, GH). Chiquibul Forest, Monkey Tail, tall broadleaf forest over limestone, area within 50 m of river, close to trail, 16°44’26”N, 80°56’9” W, 375 m, 27 March 2001, A. D. Forrest & A. Ensoll 72 (NY).

CUBA: ARTEMISA PROVINCE: Near San Antonio, along shady bank, red soil, 8

April 1905, H. A. Van Hermann 811 (F). CIENFUEGOS PROVINCE: Trinidad mts, San

Blas, July 1941, R. A. Howard 6432 (MO, GH). GUANTÁNAMO PROVINCE: Vicinity of Baracoa, 10 m, 25 January 1902, C. pollard, & E. palmer 57 (F, GH, MO). Rio Macaguanigua, thick rocks, 350 m, 17 February 1910, J. A. Shafer 3974 (GH, US).

HOLGUÍN PROVINCE: Sierra de Nipe, Limestone massif in the foothills at the Rio Guaro, about 10 km southwest of Mayari in moist hardwood forest with poorly developed understory, common in deep shady woods, 20 m, 23 July 1951, Grady Webster 3921

(GH). LA HABANA PROVINCE: Calabazar, 66 m, October 1905, Van Herman 973 (US).

Mantanzas Province: Vento, April 1905, Ashburtiss (F, GH). SANTIAGO DE CUBA

61 PROVINCE: Forets de Yatera, 1844, J. Linden 1859 (F, GH). SANCTI SPIRITUS PROVINCE: Finca Cuba, alrededoras de mogote caburni, 650 m, P. Acevedo-Rdgz., R. Oviedo, M.

Fernandez & N. Vera 6474 (F). SANTA CLARA PROVINCE: Trinidad mountains, 600 m,

January 1927, Beirne Brues (GH). ORIENTE PROVINCE: sierra de mi cara, prope mayari arriba, ad vitam, 256 m, 3 March 1916, E. L. Ekman 6752 (GH). Near Loma del Gato, region of Alto Songo, 130 m, 16 September 1951, M. Lopez 143 (US). Prope villam monte verde dictam, cuba orientali, 570 m, January-July 1859, C. E. Wright 875 (F, GH,

MO, US). PINAR DEL RIO PROVINCE: Banos San Vicente, 46 m, 15 September 2010, Britton, & Gager 7348 (F, GH). In the mountains north of San Diego de Los Banos, common in rich low wood, 60 m, 11 April 1900, W. Palmer, J Riley 515 (US). Sierra de Anafe, shady place on rocky hillside, 110 m, 31 December 2011, P. Wilson 11586 (US). At the source of Arroyo Sumidero and to the top of Cuchillas de San Sebastian, ravine in shade, 305 m, 9 August 2012, J. A. Shafer, B. Leon 13694 (F, GH, US). El Valle de Anccon near San Vicente, 50 m, 4 February 1956, C.V. Morton 9786 (US).

GUATELMALA: ALTA VERAPAZ DEPARTMENT: Near the finca Sepacuite, road from Gajabon to Cherujija, Oxee, 1244 m, 25 April 1902, O. Cook, R. Griggs 675 (US). Finca Fransvaal, in virgin forest, 610 m, 15 March 1939, C. L. Wilson 314 (F).

HUEHUETENANGO DEPARTMENT: Cerro victoria, across river from Finca San Rafael, Sierra de los Cuchumatanes, "root medicinal , used for stomach troubles", 25 November,

1904, G. Goll 61 (F). IZABAL DEPARTMENT: Valley of Motagua River, along railroad between Bananera and 5 miles south to Sioux Station, thickets in lowland, 50 m, 6 April,

1940, J. Steyermark 38999 (F). PETÉN DEPARTMENT: Dolores, about 1 km west of village, in low forest, 433 m, December, 1961, E. Contreras 3158 (US).

HONDURAS: OLANCHO DEPARTMENT: Arroyadas del Rio Wampu, west of Gualaco, selvas humiedas de madras duras, 600 m, 15 March 1972, Nelson, & Clewell 415 (NY).

JAMAICA: CLARENDON PARISH: Chapelton to bull head, 18 September 1906,

L.M. Underwood 3242 (F). HANOVER PARISH: Near Green River on trail from Cinchona to Blue Mountain Peak, rocky banks, 22 April 1903, W. Maxon 1484 (US). SAINT ANN

PARISH: Upper slopes of Mount Diablo, 500-800 m, 28 March 1920, Maxon & Killip

501A (F). SAINT ANDREW PARISH: Off road from Kingston to Newcastle, common on

62 bank above stream, 381 m, 17 June 1963, M. Crosby 198 (F, GH, NY,). Hermitage dam and vicinity, wet shady woods along the water way, 500 m, 3 June 1926, W. Maxon 8812

(US). SAINT CATHERINE PARISH: Worthy Park, luidas vale, 1849, R. C. Alexander 50

(F). SAINT ELIZABETH PARISH: Wallingford, "rather rare", 1875, T. Moore 18 (GH).

MANCHESTER PARISH: Porus , 1892, Lloyd 1155 (F, MO). SAINT MARY PARISH: Grey's

Inn, 1 December 1927, C. Orcutt 4238 (F, MO). TRELAWNY PARISH: Near troy, cockpit country, 560 m, 10 May 2003, M. Underwood 2923 (US). Vicinity of Troy, shaded rocky bank, 630 m, 30 June 1904, W. Maxon 2953 (US). Cockpit country ca. 5 miles north of Quick Step, above Aberdeen P.O., rare, on a shaded hillside, 225 m, 9 March 1950, G. Proctor 4159 (US).

MEXICO: CHIAPAS STATE: Stream bank with montane rain forest and cafetal below Yajalon, 800 m, 11 November 1971, D.E. Breedlove, A. R. Smith 22262 (F). Forest 6-8 km north of Ocosingo along road to Bachajon, slopes and steep ravines with seasonal evergreen forest, 900 m, 9 December 1971, D. E. Breedlove, A. R. Smith 22120

(F, MO). Chiapas, 600 m, 1867, Dr. Ghiesbreght 399 (GH). VERACRUZ STATE: Hidalgotitlan, Pancho Villa, del km. 2-4 del Camino de Plan de Arroyos, selva alta perenifolia, 120 m, 7 June 1974, Jesus Dorantes 3103 (F, NY). Oaxaca, Minatitlan, Orientallas del Rio Grande, cerca del campamento Rio Grande, 16 km al E de la laguna por terraceria, selva raria, en area de selva perturbadas, zona caliza, helecho en pendientes humedas perturbadas, cerca del rio, 90 m, 29 November 1981, T. Wendt, & A. Villalobos 3532 (NY).

63 LITERATURE CITED

Anderson, L.E . 1954. Hoyer’s solution as a rapid permanent mounting medium for bryophytes. The Bryologist 57: 242 – 244.

Barrington, D.S., C.A. Paris, and T.A. Ranker. 1986. Systematic inferences from spore and stomate size in the ferns. American Fern Journal 76: 149 – 159.

Ching, R.C. 1957. On the genus Adiantum L. of China with notes on some related species from neighboring regions. Acta Phytotaxonomica Sinica. 6: 301-354.

Davidse, G., S.M. Sousa, S. Knapp. 1995. Flora Mesoamericana, Volumen 1. Psilotaceae a Salviniaceae. Mexico City: Instituto de Biolog√a, UNAM.

Google Earth (Version 5.1.3533.1731) [Software]. 2009. Mountain View, CA: Google Inc.

Hooker, W.J. 1851. Species Filicum, vol. 3. William Pamplin, London.

Labiak P.H., and J. Prado. 2007. New Records of Pteridophytes from Bolivia and Brazil. American Fern Journal 97(2): 113-123.

Lellinger, D.B. 1977. Nomenclatural and taxonomic notes on the pertidophytes of Costa Rica, Pananma, and Colombia. Proceedings of the Biological Society of Washington. 89: 703-732.

Lellinger, D.B. 1989. The Ferns and Fern allies of Costa Rica, Panama, and the Choco. Part 1: Psilotaceae through Dicksoniaceae). Pteridologia. 2A: 150-151.

Lellinger, D.B. 2002. A Modern Multilingual Glossary for Taxonomic Pteridology. Pteridologia. 3: 1-263.

Lin, Y.X. 1990. Adiantaceae. In, Flora Reipublicae Popularis Sinicae Tomus. Vol. 3, issue 1. Science Press, Bejing, pp 173-216.

Linnaeus, C. 1753. Species plantarum, vol. 2. Stockholm.

Mahabale, T.S., and S.Y. Kamble. 1981. Cytology of ferns and other pteridophytes of western India. Proceedings of the Indian National Science Academy. 2: 260-278.

McCarthy, M.R., and R.J. Hickey. 2011. Adiantum mariposatum (Pteridaceae), a new species from Ecuador. American Fern Journal. 101: 1-5.

Nayer, B.K. 1961. Ferns of India 1 Adiantum. Bulletin of the National Botanical Gardens. 52: 1-38

64

Prado, J., and M. Palacios-Rios. 1998. Taxonomy and distribution of Adiantum trapeziforme and A. pentadactylon. American Fern Journal. 88: 145-149.

Prado, J. and D. Lellinger. 2002. Adiantum argutum, an unrecognized species of the A. latifolium group. American Fern Journal. 92: 23–29.

R Development Core Team. 2010. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.

Sloane, H.G. 1707. A voyage to the islands Madera, Barbados, Nieves, S. Christophers and Jamaica with the natural history of the herbs and trees, four-footed beasts, fishes, birds, insects, reptiles, &c. of the last of those islands..., vol. 1. British Museum, London.

Systematics Association Committee for Descriptive Terminology. 1962. Terminology of simple symmetrical plane shapes (Chart 1). Taxon. 11: 145-156, 245-257.

Sundue, M. Silica bodies and their systematic implications in Pteridaceae (Pteridophyta). Botanical Journal of the Linnean Society. 161(4): 422–435.

Tryon, R.M., and R.G. Stolze. 1989. Pteridophyta of Peru. Part II. 13. Pteridaceae-15. Dennstaedtiaceae. Fieldiana, Botany. 22: 1-128.

Tryon, R.M., and A.F. Tryon. 1982. Ferns and allied plants with special reference to tropical America. Springer – Verlag, New York.

Tryon, A.F., and B. Lugardon. 1991. Spores of the pteridophyta. Pg. 179-185. Springer – Verlag, New York Inc.

65

A. A.

A. A. A. A. A. ‘A. ‘A. anceps pectinatum Character ex uberans’ peruvianum trapeziforme polyphyllum

ecuadorianum’ mathewsianum

(cm) 1.4 1.4 1.5 1.55 1.55 2.23 2.22

1.65

0.4 [0.36] 0.5 [1.16] [1.35] [1.44] [0.89] [1.08] sd=0.6 sd=0.06 sd=0.28 sd=0.53 sd=0.33 sd=0.27 - 0.55 - 0.75 - 1.1 - 0.72 - 0.62 - 0.67 rhizome - 0.2 diameter diameter

0.3 0.3 scale scale 0.8 0.8 0.6 0.75 0.75 0.31 1.07 0.72

0.76 0.76 (mm)

Adiantum peruvianum Group. [0.24] [0.54] [0.21] [0.74] [0.41] [0.59] [0.59] [0.41] sd=0.05 sd=0.15 sd=0.08 sd=0.31 sd=0.14 sd=0.14 sd=0.17 sd=0.14 - 0.4 - 0.36 - 0.12 – 0.15 - 0.32 - 0.08 - 0.25 - 0.26 length rhizome

0.22 0.22 0.25 0.25 0.25 0.07 0.06 0.09

0.42 0.42 0.13 (mm)

[0.18] [0.14] [0.05] [0.04] [0.05] [0.14] [0.29] [0.11] sd=0.04 sd=0.06 sd=0.01 sd=0.01 sd=0.02 sd=0.05 sd=0.07 sd=0.02 - 0.2 - 0.1 - 0.12 - 0.05 - 0.02 - 0.02 - 0.03 – 0.09 width width rhizome scale

56 116 83

122 159 126 150

147.5 147.5 40 - [41.5] [99.7] [80.5] [96.7] (cm) 36 - 41 - 43 - 57 - [85.66] [70.08] [101.5] - 25.5 sd=25.0 sd=15.3 sd=28.8 - 48.5 52 - sd=31.87 sd=26.68 sd=14.97 sd=32.46 frondheight

1- 4- 2-3- 2-3- 2-4- 3-4- 3-4- 4-6- pinnate pinnate pinnate pinnate pinnate pinnate pinnate pinnate dissection

10 11 16 20 4-8 3-8 4-9 NA 3- 5- 7- 8- pinnae number number

5.4 7.38 2.85 2.08

9.78 9.78 6.37 4.74 [1.8] [4.89] 7.7 [4.48] [5.83] [1.49] [6.88] [4.56] [2.89] (cm) - 2.2 sd=1.04 sd=0.63 sd=0.96 sd=0.58 sd=0.45 sd=1.43 sd=0.64 sd=0.76 - 3.9 - 3.4 - 1.3 - 3.65 - 0.74 - 0.76 - 4.2 segmentlength

7.2 3.5

[0.6] 0.7 sd= 1.19 [4.59] 7.51 [3.81] 6.2 [1.72] 2.57 [0.78] 1.66 [5.34] [1.25] (cm) [1.13] 1.63 sd=0.97 sd=0.14 sd=0.34 sd=0.23 sd=0.17 sd=1.34 sd=0.34 - 2.64 - 0.68 - 2.1 - 0.26 - 0.5 - 0.86 - 0.34 -2.43 segmentwidth

[0.86] 0.55 (cm)

2.35 sd= 0.57 [1.60] 3.4 2.6 2.6 [0.23] 0.83 [0.09] 0.23 [0.07] 0.16 [0.25] [1.50] [0.71] 1.85 sd=0.06 sd=0.05 sd=0.04 sd=0.13 sd=0.10 sd=0.51 sd=0.66 - 0.2 - 0.07 - 0.3 length - 0.56 - 0.2 - 0.03 - 0.01 - 0.03 segmentstalk Table 3.1. A comparison of characters that can be used to differentiate members the Values shown indicate the ranges, means, and standard deviations (sd).

66

spore size terminal terminal Character (µm) segment width segment continued… (cm) length (cm)

32-52 2.0-4.45 3.7-7.68 [5.0] A. [41.9] [2.93] sd=1.19 peruvianum sd=6.53 sd=0.87

36-51 2.5- 4.3 5.5- 7.9 A. [42.6] [3.2] [6.45] anceps sd=3.2 sd=0.68 sd=0.98

44-53 2.6-3.25 3.0-5.75 A. [47.9] ecuadorianu sd=2.62 m

67 31-40 2.0-4.68 3.35-7.5 A. [36.9] [3.27] [5.13] trapeziforme sd=2.54 sd=0.99 sd=1.48

29-41 [35.25] 1.57-3.19 2.25-4.68 A. sd=2.12 [2.38] [3.64] exuberans sd=0.57 sd=0.83

40-50 1.12-2.1 2.0-4.05 A. [44.9] [1.4] [2.93] mathewsianu sd=2.81 sd=0.30 sd=0.74 m

36-44 0.38-1.15 0.73-2.13 A. [39.9] [0.76] [1.47] polyphyllum sd=2.8 sd=0.21 sd=0.39

not seen 0.7-0.8 [0.75] 1.2-1.32 A. sd=0.05 [1.25] pectinatum sd=0.06

Table 3.2. Our two putative A. trapeziforme groups showing characters with means that are significantly different from each other (all p-values ≤ 0.001, except number of segments on terminal pinna, p-value = 0.038). Values shown indicate the ranges, means (in brackets), and standard deviations. *The sori ratio was determined by dividing the number of sori along the acroscopic ultimate segment margin, by the number of sori along the distal segment margin.

Character A. trapeziforme ‘A. exuberans’

stalk length (cm) 0.2-[0.71]-1.85 0.03-[0.23]-0.83 +/- 0.13 +/- 0.06

segment width (cm) 0.86-[1.72]-2.57 0.5-[1.13] 1.63 +/- 0.34 +/- 0.17

number of segments on 11-[14.93]-19.0 11-[17.16]-23 terminal pinna +/- 2.31 +/- 2.46

distance between ultimate 0.96-[1.32]-1.55 0.67-[0.87]-1.14 segment stalks (cm) +/- 0.15 +/- 0.13

sori ratio* 1.85-[2.24]-2.9 2.6-[3.76]-6.5 +/- 0.3 +/- 0.81

rhizome diameter (cm) 0.75-[1.08]-1.5 0.2-[0.36]-0.5 +/- 0.27 +/- 0.06

lamina height (cm) 41-[96.7]-159 43-[70.1]-126 +/- 28.8 +/- 14.97

68 A B C

D E F

G H

1.0 cm

Figure 3.1. Silhouettes of ultimate segments from species in the Adiantum peruvianum Group. A. Adiantum mathewsianum. B. Adiantum polyphyllum. C. Adiantum pectinatum. D. ‘Adiantum exuberans ’. E. Adiantum trapeziforme. F. Adiantum peruvianum. G. ‘Adiantum ecuadorianum’. H. Adiantum anceps. Segments A-D were illustrated from the 6th segment below the apex of the terminal pinna, segments E-H were illustrated from the 3rd segment below the apex of the terminal pinna.

69

A B

C D

Figure 3.2. Distribution maps of A. Adiantum peruvianum. B. Adiantum mathewsianum. C. Adiantum anceps. D. Adiantum polyphyllum.

70 A

B

Figure 3.3. Distribution maps of A. Adiantum trapeziforme and B. ‘Adiantum exuberans’.

71 A B

C D

E F

Figure 3.4. Scanning electron micrographs of spores showing fragmented irregular surfaces from species in the A. peruvianum Group. A. Adiantum anceps (N. Vulgar & A. Sagastegui 0214 (GH)), B. ‘Adiantum ecuadorianum’ (B. Øllgaard & H. Navarrete 3035 (AAU)), C. ‘Adiantum exuberans’ (Harriman 10743 (F)), D. Adiantum trapeziforme (Maxon & Killip 560A (GH)), E. Adiantum mathewsianum (J. Wurdack 2024 (US)), F. Adiantum peruvianum (H. E. Stroke, & O. B. Horton 9448 (GH)). 72

Figure 3.5. Adiantum anceps habit (Klug 2688 (US)).

73

B

A

1.0 mm

C

Figure 3.6. A. ‘Adiantum ecuadorianum’ habit (B. Øllgaard & H. Navarrete 3035 (AAU). B. Clathrate rhizome scale with denticulate margins (scale bar = 1mm). C. Silhouette of ultimate segment (scale bar = 3cm).

74

B

A

C

Figure 3.7. A. ‘Adiantum exuberans’ habit (Hitchcock (US)). B. Clathrate rhizome scale with denticulate margins (scale bar = 1 mm). C. Silhouette of ultimate segment (scale bar = 3 cm)

75

B

A

C

Figure 3.8. A. Adiantum trapeziforme habit (Maxon 8812 (US)). B. Rhizome scale with minutely denticulate margins (scale bar = 1 mm). C. Silhouette of ultimate segment (scale bar = 3 cm)

76

Figure 3.9. Results from Principle Component Analyses showing two distinct groups (Adiantum trapeziforme and ‘Adiantum exuberans’) within the current circumscription of Adiantum trapeziforme.

77

Figure 3.10. Lectotype by Sloane (1707: t. 59) represents the type specimen of Adiantum trapeziforme L.

78

Figure 3.11. Adiantum mathewsianum habit, showing a basal pinna (J. Wurdack 2024 (US)). 79

Figure 3.12. Adiantum pectinatum habit of a juvenile specimen (Serrano et al. 5890A (MO))

80

Figure 3.13. Adiantum peruvianum habit (A. Molina 272 (MO)).

81

Figure 3.14. Adiantum polyphyllum habit of a juvenile specimen (E. Niemeyer 39 (US)).

82 4

Adiantum mariposatum (Pteridaceae), a new species from Ecuador

MIRABAI R. MCCARTHY and R. JAMES HICKEY

Department of Botany Miami University, Oxford, OH 45056

ABSTRACT: Approximately 40 species of Adiantum are represented in Ecuador. A new species is described here from Pastaza province, Ecuador. It has been confused with A. anceps, but differs in having only once-pinnate blades and pubescent rachises, segment stalks, and indusia.

The genus Adiantum comprises about 200 species distributed worldwide (Mickel & Smith, 2004) and is particularly abundant in the New World tropics. Several new Adiantum species have been described from South America (Zimmer, 2007; Prado, 2003; Prado & Smith, 2002; Prado, 2000) including one from Ecuador (Smith & Prado, 2004). In this paper yet another new species of Adiantum, from Pastaza Province, is described.

Adiantum mariposatum M. McCarthy & Hickey, sp. nov. TYPE. Ecuador: Pastaza: c. 5 km E of Mera, on road to Shell-Mera, 78°5’W 1°28’S, rocky escarpment, road bank and riverside vegetation, 1050 m, 30 July 1980, B. Øllgaard, S. Roth, & C. Sperling 35582 (holotype: AAU!; isotypes: GH!, UC). Figs 1, 2.

Folia pinnata, 16--23 cm longa, 6--12 cm lata; stipites atropurpurei usque ebenei, longitudinaliter ½--2/3 folia aequantes, glabri; rhachides atropurpureae usque ebeneae, supra hirsutae; pinnae 1--8, dimidiatae usque trapeziformes, 52--74 mm longae, 34-47 mm latae; petioluli brevi, 0.5--2 mm longi, hirsuti; indusia discreta, 0--5 per pinnam, ovata usque late ovata, hirsuta.

83 Plants terrestrial. Rhizomes short-creeping, 4--5 mm in diameter, densely scaly; scales triangular to narrowly triangular, lustrous, rigid, bullate, darkened centrally, castaneous along margins and apices, bases auriculate, margins with spreading to recurved teeth, apices attenuate. Leaves monomorphic, 16--32 X 6--12 cm; stipes 50-- 60% the length of the fronds, atropurpureous to ebeneous, mainly glabrous, with scattered scales near the base, shining to weakly glaucous; stipe scales narrowly triangular, stramineous proximally, castaneous distally, margins with spreading to recurved teeth; blades oblong to broadly obovate, pinnate (entire when young), glabrous, pinna bases overlapping the rachis; rachises atropurpureous to ebeneus, abaxially glabrous, adaxially with minute scattered hairs 0.1--0.2 mm long, hairs castaneous proximally, stramineous distally, extending onto segment stalks; segments dimidiate to trapeziform, papyraceous, not articulate, 50--75 X 34--47 mm, stalks 0.5--2.1 mm long, darkened color passing into segment bases, bases basiscopically excavate, acroscopically truncate, margins shallowly dentate, apex acuminate; veins free, dilated distally, ending in marginal teeth or arcuate toward the nearest distal tooth, prominulous, markedly so adaxially, lacking venuloid idioblasts between veins; terminal blade segment broadly trullate, 80--100 X 65--80 mm; sori discontinuous, 2--16 per segment, widely depressed ovate, 2--3 X 2--3 mm, false indusia stramineous, transparent when young, becoming brittle and black with age, with small reddish-brown hairs along receptacle lines, margins erose; spores trilete, stramineous-gold, 23--44 µm.

This species is known only from Pastaza Province in the eastern foothills of the

Ecuadorian Andes. It grows in wet forests and on rocky riverbanks, in shade. The epithet for this new species makes reference to the large butterfly-shaped segments.

PARATYPE.-Ecuador. Pastaza. Mera-Shell Mera, ca. 2 km E of Mera, at bridge over Río Alpayacu, 78°06’W 01°28’S, 1100 m, 21 Jan 1992, B. Øllgaard et al. 99574 (AAU).

Adiantum mariposatum can be distinguished by having compact, pinnate fronds that reach about 30 cm, large glabrous segments that overlay the rachis, adaxial pubescence along the rachises and segment stalks, and by the broadly ovate, sparsely

84 pubescent indusia. It can be confused with A. anceps, which reaches 2 meters, has 1-3 pinnate fronds, and is completely glabrous along the rachises, segment stalks, and indusia. As currently circumscribed, Adiantum mariposatum falls within the A. tetraphyllum group as delineated by Tryon & Tryon (1982). This group has 1- or 2- pinnate blades, axes with scales or adaxial pubescence, sessile to short-stalked segments, and few to many indusia. A. humile Kunze., A. latifolium Lam., A. obliquum Kaulf., A petiolatum Desv., A. tomentosum Klotzsch, A. pulverulentum L., and A. tetraphyllum Humb. & Bonpl. ex Willd. are other members in this group. All of these species however, have conspicuous venuloid idioblasts (silica bodies) on laminar surfaces between veins (Sundue, 2009), a character that is lacking in A. mariposatum. This character was not mentioned in Tryon’s 1982 circumscription of the Adiantoid groups, but is now thought to be relevant in determining systematic relationships (Sundue, 2009). Adiantum mariposatum may be more closely allied to the more widespread A. urophyllum Hook., which lacks visible venuloid idioblasts, has 2-pinnate leaves (juvenile leaves may be 1-pinnate), more numerous and smaller segments with long tapering apices, densely pubescent stipes, rachises, and segment stalks, and sparsely pubescent to often glabrous indusia. The small red-brown hairs on the indusia of A. mariposatum tend to follow the receptacle lines, whereas the hairs of A. urophyllum appear randomly scattered across the indusia. No other Adiantum species observed during this study displayed a similar linear arrangement of pubescence along the indusia. This character appears to be unique to A. mariposatum. Additional Adiantum species with pubescent indusia include, A. terminatum Kze. ex Miq., A. trichochlaenum Mickel & Beitel, A. tricholepis Fée, and A. curvatum Kaulf. Adiantum terminatum and A. trichochlaenum are 2 pinnate, have indument abaxially along the rachises, and venuloid idioblasts between the veins on both surfaces of the pinnules. Adiantum tricholepis and A. curvatum are 3-4 pinnate, and lack venuloid idioblasts.

85 LITERATURE CITED

MICKEL, J. T. and A. R. SMITH. 2004. Pteridophytes of Mexico. Mem. New York Bot. Gard. 88: 21-45.

PRADO, J. 2000. A new species of Adiantum (Pteridaceae) from Bahia, Brazil. Brittonia. 52: 210-212.

PRADO, J. 2003. New species in Adiantum from Brazil. Amer. Fern J. 93: 76-80.

PRADO, J. and A. R. SMITH. 2002. Novelties in Pteridaceae from South America. Amer. Fern J. 92: 105-111.

SMITH, A. R. and J. PRADO. 2004. New species in Adiantum and Pteris (Pteridaceae) from the Andes. Brittonia 55: 310--316.

SUNDUE, M. A. 2009. Silica bodies and their systematic implications in Pteridaceae (Pterdiophyta). Bot. J. Linn. Soc. 161: 422–435.

TRYON, R. M. and A. F. TRYON. 1982. Ferns and allied plants with special reference to Tropical America. Springer-Verlag, New York.

ZIMMER, B. 2007. Adiantum krameri (Pteridaceae), a new species from French Guiana. Willdenowia. 37: 557-562.

86

Fig. 4.1 Holotype of Adiantum mariposatum (B. Øllgaard, S. Roth, & C. Sperling 35582, AAU). A. Habit showing entire and pinnate fronds (scale bar equals 5 cm). B. and C. Stipe and rhizome scales, respectively, showing margins with spreading and recurved spines (scale bar equals 1 mm).

87 B

Fig. 4.2 Holotype of Adiantum mariposatum (B. Øllgaard, S. Roth, & C. Sperling 35582, AAU). A. Adaxial view of rachis and segment with minute scattered hairs. B. Abaxial view of glabrous rachis and segment stalks. C. Sorus with minute red-brown hairs following receptacle lines. D. Segment margin with distally dilated veins.

88 Conclusion

This dissertation contributes to the systematics of Adiantum, a genus that has never been monographed, and for which few molecular phylogenetic studies exist (exception is Lu et al., 2012). The phylogeny generated here contributes towards a greater understanding of species relationships within Adiantum, particularly among the neotropical taxa, which account for the greatest amount of diversity in the genus. This dissertation also represents the first investigation into Adiantum utilizing both molecular and morphological techniques to evaluate the phylogeny of a species group. Specifically, molecular data were utilized to establish and test species group hypotheses, and morphological data were employed to circumscribe taxa in the neotropical ‘Adiantum peruvianum’ clade. The morphological species groups proposed by Tryon & Tryon (1982) provide a starting point for these systematic studies, but our molecular results indicate that several, and perhaps more, of these proposed groups are paraphyletic (Chapter 2). These same data clearly define four primary clades (A, B, C, and D) within Adiantum. Clades A & B contain temperate species, clade C contains primarily paleotropical species, and clade D contains neotropical species. Additional sub-clades within these four primary clades were also identified, but additional sampling is needed before their limits can be fully appreciated. Venuloid idioblasts, common among genera in the Pteridaceae, are also widespread in Adiantum (Sundue, 2009). Sundue suggested that the distribution of these idioblasts within leaves of Adiantum may represent synapomorphies; this study supports his findings. Neotropical clade D (Chapter 2) contains two included clades, D1 and D2. Clade D1 contains species with visible venuloid idioblasts between veins, whereas sub- clade D2 contains species lacking visible venuloid idioblasts between veins. The current study focused on circumscribing and revising the ‘A. peruvianum group’ (clade D2.III), a sub-group within clade D2. The A. peruvianum Clade contains at least eight species A. anceps, A. ecuadorianum ined., A. exuberans ined., A. mathewsianum, A. pectinatum, A. peruvianum, A. polyphyllum, and A. trapeziforme.

89 Members of this group can be identified by having clathrate rhizome scales with minutely denticulate margins; segment stalks that stop abruptly at segment bases; discreet sori that are depressed-ovate, reniform, or occasionally lunate; ultimate segments that are dimidiate, trapeziforme, deltate, or occasionally ovate-falcate; free, dichotomous veins that end in marginal teeth; and laminar tissue that lacks visible idioblasts between veins. A taxonomic account, including descriptions and distributions of the species within this clade is presented. Two members of this clade, Adiantum ecuadorianum ined. and A. exuberans ined., are described as new (Chapter 3), as is A. mariposatum (Chapter 4), a member of clade D2.I. The phylogeny generated in this study stimulates additional questions. First, what is the relationship of species within the remaining clades (A, B, C, and D1) and how does that compare to Tryon & Tryon’s, 1982 morphological circumscriptions? Lu et al., 2012 provides a detailed account of temperate Adiantum species (here called clades A & B), but there is limited molecular work (Bouma et al., 2010) on paleotropical species (clade C), and on the remaining neotropical species, those with visible idioblasts present between veins (clade D1). Second, where do the reticulate-veined Adiantum species (A. adiantoides, kramerii, leprieurii, olivaceum, and wilsonii) fit among the neotropical clade? Since we were unable to obtain quality sequences from any of these species, their placement is unknown. The absence of visible idioblasts in these reticulate-veined species, except for Adiantum wilsonii, hints of a possible relationship with clade D2; since reticulation is unique to this group, their placement is uncertain. Is reticulation a more primitive or derived character? Are these species basal, or more recently evolved compared to other neotropical species? The placement of these taxa is intriguing, and additional studies are certainly warranted. Finally, there is a need to clarify the interspecific relationships within the A. peruvianum Clade. The atpA and rbcL regions used in this study were effective at recovering species group relationships, but they were ineffective at recovering species level relationships. Additional sampling using a wider selection of genes, perhaps intergenic spacers, which tend to be more variable and might be better at resolving

90 species-level relationships, would help to clarify the phylogeny within the Adiantum peruvianum Group and other groups.

91 LITERATURE CITED

Bouma, W. L. M., Richie, P., and Perrie, L. R. 2010. Phylogeny and generic taxonomy of the New Zealand Pteridaceae ferns from chloroplast rbcL DNA sequences. Australian Systematic Botany 23: 143-415.

Lu, J. - M., Wen, J., Lutz, S., Wang, Y.-P., and Li, D.-Z. 2007. Phylogenetic relationships of Chinese Adiantum based on five plastid markers. Journal of Plant Research 125: 237-249.

Sundue, M. Silica bodies and their systematic implications in Pteridaceae (Pteridophyta). Botanical Journal of the Linnean Society. 161(4): 422–435.

Tryon, R. M. and A. F. Tryon. 1982. Ferns and allied plants with special reference to tropical America. Springer – Verlag, New York.

92 APPENDIX A -- SPECIMENS USED FOR MORPHOLOGICAL ANALYSES

Adiantum anceps

PERU: AMAZONAS REGION: J. J. Wurdack 1885 (GH, NY, MO). CAJAMARCA

REGION: Peantes 4733 (F, NY). R. Vasquez, R Rojas, A. Pena & E. Chavez 25127 (MO,

NY). HUANUCO REGION: J. Schunke 2262 (F, GH, NY). R. Tryon & A. Tryon 5288 (F,

GH, NY, MO). JUNIN REGION: E. Killip & A. Smith 23478 (F, GH, NY). E. Killip & A. Smith 25077 (F, NY). D. Smith & R. Foster 2491 (MO). R. Tryon & A. Tryon 5444 (F).

PASCO REGION: D. Smith 1957 (MO, NY). SAN MARTIN REGION: No Collector Name 3408A (F, GH, MO, NY). N. Vulgar & A. Sagastegui 0214 (GH). J. Schunke 9670 (GH, MO, NY). J. Schunke 7641 (MO, NY). Adiantum ecuadorianum ined.

ECUADOR: PASTAZA PROVINCE: B. Øllgaard & H. Navarrete 3035 (AAU). Adiantum exuberans ined.

BELIZE: CAYO DISTRICT: P. Gentle 2342 (F, GH, MO, US).

COSTA RICA: ALAJUELA PROVINCE: P. Biolley 17 (GH, MO, US). A. Smith, T.

Beliz, M. Grayum, & P. Sleeper 1617 (AAU, MO, NY). GUANACASTE PROVINCE: E. Scamman 7638 (GH, US). Boyle, Holbrook, & Kirkpatrick 6457 (NY). J. Bittner 2265

(AAU). L & T. Williams 24617 (F, US). SAN JOSÉ PROVINCE: A. Alfaro 46 (US).

EL SALVADOR: AHUACHAPÁN DEPARTMENT: R. Seiler 937 (F). P. Standley

20072 (US). SAN MIGUEL DEPARTMENT: R. Seiler 139 (F). SAN SALVADOR

DEPARTMENT: R. Seiler 785 (F). P. Standley 22750 (GH, US). Luis V. Velaseo 8892 (US).

GUATEMALA: JUTIAPA DEPARTMENT: D. Dunn, C. Dziekawanowski, S. Trott,

& D. Thurm 23236 (NY). SANTA ROSA DEPARTMENT: H. et Lux 4669 (F, GH, MO, US).

ZACAPA DEPARTMENT: J. Steyermark 42962 (F, US).

HONDURAS: COLÓN DEPARTMENT: J. Saunders 795 (NY). CORTÉS

DEPARTMENT: A. Molina 3539 (F, GH, US). J .B. Edwards p-695 (F, GH). EL PARAÍSO

DEPARTMENT: C. V. Morton 6995 (NAT). FRANCISCO MORAZÁN DEPARTMENT: A.

Molina 3924 (F, GH). YORO DEPARTMENT: P. Ames 85 (US).

93 MEXICO: CHIAPAS STATE: E. Martinez 19956 (F). G. Davidse 30135 (MO,

NY). GUERRERO STATE: E. Palmer 519 (F, GH, MO). G. Hinton 10802 (F, MO).

HIDALGO-SAN LUIS POTOSI STATE: H. H. Moore, & C. E. Wood 3653 (GH). JALISCO

STATE: Y. Mexia 1297 (F, GH, MO,). MÉXICO STATE: G. B. Hinton 7644 (F, GH, MO).

OAXACA STATE: F. Lopez 71 (NY). S. Salas 3863 (NY). SAN LUIS POTOSI SATE: Kenoyer and Crum 3983 (GH). F. Pennell 17992 (GH). C. G. Pringle 3960 (F, GH,

MO, MU). TAMAULIPAS STATE: R. Dressler 1954 (GH, MO). VERACRUZ STATE: D. Potter 21374 (GH).

NICARAGUA: BOACO DEPARTMENT: F. C. Seymour 3838 (GH, MO). CARAZO

DEPARTMENT: A. S. Hitchcock (US). GRANADA DEPARTMENT: D. Dudley, & A. Moore

(MO, GH). MASAYA DEPARTMENT: P. Standley 8060 (F, US). V. Grant 1033 (GH).

MATAGALPA DEPARTMENT: L. O. Williams 23972 (F, US). UKNOWN DEPARTMENT: C. Wright & Ringgold (GH). Adiantum mathewsianum

BOLIVIA: SANTA CRUZ DEPARTMENT: A. Molina 305 (MO).

BRAZIL: MATO GROSSO DO SUL STATE: M. R. da Silva, & C. E. Rodrigues 1332 (MO). G. & M. Hatschbach, & J. M. Silva 73878 (MO).

ECUADOR: ZAMORA-CHINCHIPE PROVINCE: T. Croat 92355 (MO, NY).

PARAGUAY: CONCEPCION DEPARTMENT: E. Zardini 53826 (NY). DE LA

CORDILLERA DEPARTMENT: J. D. Anists 2869 (US). UNKNOWN DEPARTMENT: E. Hassler 11045 (GH, NY, US).

PERU: AMAZONAS REGION: J. Wurdack 2024 (GH, NY, US). CAJAMARCA

REGION: J. Campos, A. Pena & E. Chavez 4778 (MO, NY). J. Campos, A. Pena & G. Pezantes 4733 (MO, NY). Adiantum peruvianum

BOLIVIA: BENI DEPARTMENT: I. Vargas 750 (MO). SANTA CRUZ

DEPARTMENT: A. Molina 272 (MO). F. Billiet & B. Jadin 6187 (MO). M. Sundue & M. Nee 523 (NY, US). M. Nee 34220 (NY).

ECUADOR: MORONA-SANTIAGO PROVINCE: B. Ollgaard, & H. Navarrete 2468

(AAU). SANTIAGO-ZAMORA PROVINCE: W. H. Camp 1510 (F, NY). ZAMORA-

94 CHINCHIPE PROVINCE: J. P. Janovec, A. K. Neil, W. Quizhpe, & I. Toborg 1383 (MO, NY).

PERU: AMAZONAS REGION: H. Van der Werff, R. Vasquez, B. Gray, & R. Rojas

16984 (F, MO, NY,) & 17004 (MO, NY). CAJAMARCA REGION: C. Diaz & S. Flores

10666 (MO). HUANUCO REGION: R. & A. Tryon 5313 (F, MO, US). P. S. Coronado 81 (US). R. Ferreyra 10237 (GH). H. E. Stroke, & O. B. Horton 9448 (F, GH, US). R.

Tryon & A. Tryon 5236 (F, MO). JUNIN REGION: E. Cerrate 2869 (GH). S. E. Saunders

574-A (F). S. G. E. Saunders 575 (F). PASCO REGION: A. Manteagudo, C. Mateo, & G. Ortiz 3965 (NY). A. Monteagudo, G. Ortiz, & R. Francis 5009 (NY). H. van der Werff,

B. Gray, R. Ortiz, & N. Davila 17781 (MO, NY). SAN MARTIN REGION: H. Van der Werff, B Gray, R. Vasquez, & R. Rojas 15474 (MO, NY). J. F. Macbride 4215 (F, US). H. A. Allard 21511 (F, US). Adiantum pectinatum

BOLIVIA: TARIJA DEPARTMENT: M. Serrano, J. Villalobos, A. Lliully, I. Guachalla, & R. Leon 5890 (MO). Adiantum polyphyllum

COLOMBIA: MAGDALENA DEPARTMENT: O. Haught 3763 (GH, US). O. Haught 3856 (US). D. Bennett 5 (NY, US). M. Grant 10894 (US). E. Niemeyer 39 (US).

TRINIDAD: Fendler 71 (GH, NY, F).

VENEZUELA: BARINAS STATE: T. Croat 54751 (MO). L. Dorr, L. Barnett, C.

Aymard, N. Cuello, & G. Diggs 4778 (NY, US). MIRANDA STATE: H. Pittier 11273

(GH, US). SUCRE STATE: G. Davidse, & A. Gonzalez 19110 (MO). TRUJILLO STATE: L. Dorr, L. Barnett, C. Aymard & N. Cuello 5362 (NY). Adiantum trapeziforme

BELIZE: CAYO DISTRICT: C. L. Lundell 6498 (GH, US). A. D. Forrest & A. Ensoll 72 (NY).

CUBA: GUANTÁNAMO PROVINCE: C. Pollard, & E. Palmer 57 (F, GH, MO,

US). J. A. Shafer 3974 (GH, US). HOLGUÍN PROVINCE: G. Webster 3921 (GH). LA

HABANA PROVINCE: V. Herman 973 (US). SANTA CLARA PROVINCE: B. Brues (GH).

ORIENTE PROVINCE: E. L. Ekman 6752 (GH). M. Lopez 143 (US). C. E. Wright 875 (F,

95 GH, MO, US). PINAR DEL RIO PROVINCE: Britton, & Gager 7348 (F, GH). W. Palmer, J. Riley 515 (US). P. Wilson 11586 (US). J. A. Shafer, & B. Leon 13694 (F, GH, US). C. V. Morton 9786 (US).

GUATELMALA: ALTA VERAPAZ DEPARTMENT: O. Cook, & R. Griggs 675

(US). C. L. Wilson 314 (F). IZABAL DEPARTMENT: J. Steyermark 38999 (F). PETÉN

DEPARTMENT: E. Contreras 3158 (US).

HONDURAS: OLANCHO DEPARTMENT: Nelson, & Clewell 415 (NY).

JAMAICA: HANOVER PARISH: W. Maxon 1484 (US). SAINT ANDREWS PARISH:

M. Crosby 198 (F, GH, NY). W. Maxon 8812 (US). TRELAWNY PARISH: M. Underwood 2923 (US). W. Maxon 2953 (US). G. Proctor 4159 (US).

MEXICO: CHIAPAS STATE: D. E. Breedlove, & A. R. Smith 22262 (F). D. E.

Breedlove, A. R. Smith 22120 (F, MO). Dr. Ghiesbreght 399 (GH). VERACRUZ STATE: Jesus Dorantes 3103 (F, NY). T. Wendt, & A. Villalobos 3532 (NY).

96