Variation in the Dendrophylax Porrectus Species Complex (Vandeae

MISSING IMAGE

Material Information

Title:
Variation in the Dendrophylax Porrectus Species Complex (Vandeae Orchidaceae) Based Upon Morphological and Molecular Data
Physical Description:
1 online resource (69 p.)
Language:
english
Creator:
Molgo,Iwan E
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Botany, Biology
Committee Chair:
Williams, Norris H
Committee Members:
Judd, Walter S
Whitten, William M

Subjects

Subjects / Keywords:
dendrophylax -- its -- matk -- morphology -- orchidaceae -- porrectus -- vandeae -- ycf1
Biology -- Dissertations, Academic -- UF
Genre:
Botany thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Dendrophylax porrectus (Rchb.f.) Carlsward & Whitten, commonly known as the Jingle Bell orchid, is a leafless epiphytic orchid distributed throughout Florida (USA), the Greater Antilles, Mexico, Guatemala, and El Salvador. Previous phylogenetic analyses based on combined nuclear and plastid sequences (ITS, matK, and trnL-F) revealed considerable variation across its geographic range. This study aims to resolve the species circumscription and relationships within the Dendrophylax porrectus complex, as represented by populations in Florida (USA) and Yucat?n (Mexico) through morphometric, statistical, and phylogenetic analyses. Measurements of flower parts and roots were taken for comparative studies, and DNA sequences of ITS and plastid regions matK and ycf1 were used to estimate the phylogenetic relationships within Dendrophylax and D. porrectus. All morphometric measurements indicate a significance difference between the Florida (thin-rooted) and Yucat?n (thick-rooted) entities. Molecular analyses supports the monophyly of D. porrectus sensu lato and the hypothesis that the thin and thick-rooted taxa form sister clades. Morphological and molecular data support the recognition of the thin-rooted populations and the thick-rooted populations as two distinct species.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Iwan E Molgo.
Thesis:
Thesis (M.S.)--University of Florida, 2011.
Local:
Adviser: Williams, Norris H.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2011
System ID:
UFE0043411:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 V ARIATION IN THE DENDROPHYLAX PORRECTUS SPECIES COMPLEX (V ANDEAE : O RCHIDACEAE ) BASED UPON MORPHOLOGICAL AND MOLECULAR DATA By IWAN EDUARD MOLGO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR TH E DEGREE OF MASTERS OF SCIENCE UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 Iwan Eduard Molgo

PAGE 3

3 T o my f amily

PAGE 4

4 ACKNOWLEDGMENTS I thank my advisor Norris H. Williams who gave me the opportunity to start my graduate carrier in his lab I also thank W. M ark Whitten who has always been a great teacher Both Norris and Mark have contributed invaluable support, guidance and encouragement throughout my Masters of Science program. They introduced me to my thesis project, which turned out to be a great learning experience in molecular and morphological phylogenetics I thank Barbara Carlsward for giving me ac cess to her data and her advice in interpreting my anatomical results. Walter Judd has been an excellent sou rce of knowledge and enthusiasm, which contributed greatly to my education. I am grateful to David Oppenheimer and George Hecht who trained me in bright field and auto montage macro photography. I also want to thank Larry Winner for his assistance in choos ing the appropriate statistical methods for my research analyses. I thank all the staff of the University of Florida H erbarium, and the members of the U.F. Biology Department, for their assistance, friendship and encouragement, in particular Mario Blanco, Paul Corogin Lorena Endar a, G retchen I onta Trudy Lindle r, Kent Perkins and Kurt Neubig Kurt has not only been a fellow graduate student, but also a special ly good friend He was a great encourage ment and assisted me tremendously with man y lab techniques and analyse s. This research would not have been possible without the assistance and material provided by J ames Ackerman (University of Puerto Rico Rio Piedras Puerto Rico), Germn Carnevali (Centro de Investigacin Ci entfica de Yucatn, Mexico), Gerardo A. Salazar, (Universidad Nacional Autnoma de Mxico) Stephen Dickman (Hill s borough County Parks and Recreation Department), Robin Gardner and the Old Miakka United

PAGE 5

5 Methodist Church ( private landowners) I thank the Organization of American States for granting me the scholarship to study at the University of Florida. In addition, the cost of lab work and fieldtrips was supported by the generosity of No rris Williams and Mark Whitten. I thank my parents Edua rd and Jacqueline Molgo for their support and encouragement throughout my life. Finally, I thank my lovely wife Muriel Molgo and my beautiful daughter Isabella Molgo who have been patient and very support ive throughout my studies.

PAGE 6

6 TABLE OF CONTENTS pag e ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 11 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 15 Taxonomic History ................................ ................................ ................................ .. 15 Distribution ................................ ................................ ................................ .............. 16 Intraspecific Variation in D. porrectus ................................ ................................ ..... 16 2 MATERIALS AND METHODS ................................ ................................ ................ 21 Plant Material ................................ ................................ ................................ .......... 21 Morphological Analyses ................................ ................................ .......................... 21 General Morphology ................................ ................................ ......................... 21 Root Anatomy ................................ ................................ ................................ ... 22 Molecular Analyses ................................ ................................ ................................ 23 DNA Extraction ................................ ................................ ................................ 23 Amplification and Sequencing ................................ ................................ .......... 24 Data Analysis ................................ ................................ ................................ ... 25 Phenology ................................ ................................ ................................ ............... 26 3 RESULTS ................................ ................................ ................................ ............... 32 General Morphology ................................ ................................ ............................... 32 Root Thickness ................................ ................................ ................................ 32 Ovary Length ................................ ................................ ................................ .... 33 Sepal Measurements ................................ ................................ ....................... 33 Petal and Labellum Measurements ................................ ................................ .. 34 Anther Measurements ................................ ................................ ...................... 35 Root Anatomy ................................ ................................ ................................ ........ 35 Molecular Analysis ................................ ................................ ................................ .. 36 ITS Analyses of D. porrectus ................................ ................................ ............ 36 matK Analyses of D. porrectus ................................ ................................ ......... 37 ycf1 Analyses of D. porrectus ................................ ................................ ........... 37 Combined Plastid Analyses of D. porrectus ................................ ..................... 38

PAGE 7

7 Combined ITS and Plastid Analyses of D. porrectus ................................ ........ 38 Maximum Likelihood ................................ ................................ ......................... 39 Distribution of D. porrectus ................................ ................................ ............... 39 Phenology ................................ ................................ ................................ ............... 40 4 DISCUSSION and CONCLUSION ................................ ................................ .......... 60 Disc ussion ................................ ................................ ................................ .............. 60 Conclusion ................................ ................................ ................................ .............. 63 APPENDIX: SPECIMENS USED IN THIS STUDY WITH THEIR GENBANK ACCESSION NUMBER ................................ ................................ .......................... 64 LIST OF REFERENCES ................................ ................................ ............................... 66 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 69

PAGE 8

8 LIST OF TABLES Table page 1 1 List of herbarium specimens used to generate a distribution map of Dendrophylax porrectus ................................ ................................ .................... 18 2 1 Taxa used for molecular phylogenetic study. ................................ ..................... 28 2 2 Infiltration and polymerization of LR White resin with the root tissue .................. 30 2 3 Primer sequence of one nuclear and two plastid gene regions .......................... 30 2 4 Components of polymerase chain r eaction with Sigma brand reagents ............. 30 3 1 Comparison of tree statistics for each gene region and combinations of these gene regions for parsimony analyses. ................................ ................................ 41 A 1 Specimens used in this study with their GenBank accession number ................ 64

PAGE 9

9 LIST OF FIGURES Figure page 1 1 Distribution of Dendrophylax porrectus sensu lato ................................ ............ 20 3 1 Comparison of two representative individuals (Florida, A and Yucatn, B) of Dendrophylax porrectus illustrating measurements of different plant and floral p arts. ................................ ................................ ................................ .................. 42 3 2 The difference in root thickness within two populations of D. porrectus in Florida and Yucatn. ................................ ................................ .......................... 43 3 3 The difference in ovary length within two populations of D. porrectus in Florida and Yucatn. ................................ ................................ .......................... 44 3 4 Differences in the dorsal sepal of D. porrectus within the populations of Florida and Yucatn. ................................ ................................ .......................... 45 3 5 Differences in the lateral sepal of D. porrectus within the populations of Florida and Yucatn. ................................ ................................ .......................... 46 3 6 Differences in the petals of D. porrectus within the populations of Florida and Yucatn. ................................ ................................ ................................ ............. 47 3 7 Differences in the labellum of D. porrectus within the populations of Florida and Yucatn. ................................ ................................ ................................ ...... 48 3 8 The difference in callus length within two populations of D. porrectus in Florida and Yucatn. ................................ ................................ .......................... 49 3 9 Differences in the anther of D. porrectus within the populations of Florida and Yucatn.. ................................ ................................ ................................ ............ 50 3 10 Resin embedded root sections of D. porrectus representing Florida and Yucat n populations. ................................ ................................ .......................... 51 3 11 Resin embedded comparisons of cell wall thickness in the endodermis of two D. porrectus populations ................................ ................................ ..................... 52 3 12 One of the 4 equally parsimonious trees of ITS.. ................................ ................ 53 3 13 One of the 3 equally parsimonious trees of matK ................................ .............. 54 3 14 The most parsimonious tree of ycf1 ................................ ................................ ... 55 3 15 The most parsimonious tree of combined matK and ycf1 analysis. .................... 56 3 16 The most parsimonious tree of combined ITS, matK, and ycf1 analysis.. .......... 57

PAGE 10

10 3 17 Maximum Likelihood tree of combined nuclear and plastid data assuming the GTR+G model, log L = 7547.2470 based on100 replicates. ............................. 58 3 18 Distribution of D. porrectus with the three major clades shown in three different colors. ................................ ................................ ................................ ... 59

PAGE 11

11 LIST OF ABBREVIATION S C D egrees celcius AIC Akaike information criterion AMES Harvard University AMO Herbario AMO ANOVA Analysis o f Variance bps base pairs BS Bootstrap CICY Centro de Investiga cin Cientfica de Yucatn cm centimeter DELTRAN Delayed transformation DNA Deoxyribonucleic acid Fig Figure FLAS University of Florida Herbarium HTF Partition homogeneity test intF internal forward intR internal reverse ITS Internal Transcribed Spacer gene JK Jackknife M.S Masters of Science matK MaturaseK gene MEXU Universidad Nacional Autnoma de Mxico ml milliliter ML Maximum Likelihood mm millimeter

PAGE 12

12 NY New York Botanical Garden PCR Polymerase Chain Reaction rpm r evolutions per minute SEL Marie Selby Botanical Gardens SPR S ubtree pruning regrafting TBR tree bisection reconnection UPRRP University of Puerto Rico USF University of South Florida UVAL Universidad del Valle de Guatemala vs versus ycf1 hypothetical chloroplast open reading frame 1 L microliter m micrometer

PAGE 13

13 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for th e Degree of Master of Science VARIATION IN THE DENDROPHYLAX PORRECTUS SPECIES COMPLEX (VANDEAE: ORCHIDACEAE) BASED UPON MORPHOLOGICAL AND MOLECULAR DATA By Iwan Eduard Molgo August 2011 Chair: Norris H Williams Major: Botany Dendrophylax porrectus (Rchb.f.) Carlsward & Whitten commonly known as the Jingle Bell orchid is a leafless epiphytic orchid distributed throughout Florida (USA) the Greater Antilles, Mexico Guatemala and El Salvador. Previous phylogenetic analyses based on combined nuclear and plastid sequences (ITS, matK and trnL F ) revealed considerable variation across its geographic range T his study aims to resolve the species circumscription and relations hips within the Dendrophylax porrectus complex as represented b y populations in Florida (USA) and Yucatn (Mexico) th rough morphometric statistical and phylogenetic analyses Measurements of flower parts and roots were taken for comparative studies and DNA s equence s of ITS and plastid regions matK and ycf1 were used to estimate the phylogene tic relationships within Dendrophylax and D. porrectus All morphometric measurements indicate a significance difference between the Florida ( thin rooted ) and Yucatn ( thick rooted ) entities Molecular analyses supports the monophyly of D. porrectus sensu lato and the hypothesis that the thin and thick rooted taxa form sister clades. Morphological and

PAGE 14

14 molecular data support the recognition of the thin rooted populations and the thick rooted populations as two distinct spe cies.

PAGE 15

15 CHAPTER 1 INTRODUCTION Taxonomic History The genus Dendrophylax Rchb.f. (13 species) is sister to Campylocentrum Bent h. (35 species). These two genera are the only Neotropical member s of the African subtribe Angraecinae (Carlsward et al., 2003) Although Campylocentrum consists of primarily leafy taxa, these two genera both contain leafless species, with 13 leafless species known for Dendrophylax and 12 for Campylocentrum The leaflessness of these orchids is exceptional because they have a reduced shoot with non photosynthetic scales (Carlsward, 2004) As in all leafless orchids, Dendrophy lax porrectus (Rchb.f.) Carlsward & Whitten commonly known as Jingle Bell Orchid or Needle Root Orchid conducts photosynthesis in its greenish roots. Charles Wright first collected D. porrectus in Cuba and Reichenbach (1865) described it as Ae ranthus porrectus Rchb.f It was transferred to Campylocentrum as C. porrectum by Rolfe (1903) and six years later it was placed in its own genus Harrisella Fawc. & Rendle (1909 ) The name of this genus was chosen to com memorate William H Harris, 1860 1920, a British botanist and prolific collector of Jamaican plants (Ackerman, 1995) The molecular analyses of Carlsward (2003) clarified the placement and relationships o f H. porrecta and it was transferred to Dendrophylax as D porrectus (Rchb.f.) Carlsward & Whitten (2003) Other synonyms of D. porrectus are Harrisella amesiana Cogniaux and H. uniflora Dietrich (Ackerman, 1995)

PAGE 16

16 Distribution Based on vouche rs and database data acquired from various herbaria: FLAS, USF, SEL, AMES, NY, CICY, MEXU, AMO, and UPRRP (Holmgren et al., 1990) literature (Dix & Dix, 2000; Cansino et al., 1996) personal communication s G. Salazar & R. Jimnez Machorro 2011 (Table 1 1) D endrophylax porrectus is distributed through Florida th e West Indies (Cuba, Cayman Islands, Jamaica, Dominican Republic, and Puerto Rico) and Central America (Mexico, Guatemala, and El Salvador) ( Figure1 1 ) The elevation of D. porrectus ranges from 5 m 12 00 m above sea level (a.s.l.). These twig epiphytes d o not grow on a specific host. Based upon herbarium records, host plants (phorophytes) include Brya buxifolia (Murray) Urb, Taxodium distichum (L.) Rich Fraxinus caroliniana Mill, Juniperus virginiana var silicicola (Small) A.E. Murray Viburnum obovatum Walter Randia aculeata L Gymnopodium floribundum Rolfe Blepharidium guatemalense Standl. Citrus sp ., Psidium sp and Crescentia sp. Habitats include creek sides, river margins swamp s, cypress domes, hardwood hammocks (Brown, 2002) dry shrub forest, and high forest. Intraspecific V ariation in D. porrectus D endrophylax porrectus was recognized as a single species until Warford (1997) observed morphological differences in the flowers of populations in Jalisco, Mexico. When compared to populations of the Greater Antilles and Florida, the Jalisco populations have a narrower rostellar entry to the stigma, and the pollinia are flatted and ovoid, not globose (Warford, 1997) These preliminary data suggested that D. porrectus may be composed of mo re than one species. In addition p reliminary p hylogenetic analyses based on combined nuclear and plastid data sets ( ITS matK and trnL F ) from plants across the distributional range of D. porrectus showed high sequence

PAGE 17

17 divergence among populations (Carlsward et al., 2003) Differences in root thickness, habitat preferences (swamp vs. dry shrub ), and flower morphology among sites suggested that D. porrectus might be a complex of species that have gone unrecognize d by botanists because of their diminutive habit and uniformly inconspicuous flowers To test the hypothesis that cryptic species exist within our current concept of D. porrectus I have pursued statistical analyses on morphometric data and estimated the evolutionary relationships within D endrophylax by using phylogenetic analysis of DNA sequence data. The purpose of this study is to estimate the phylog enetic relationships within the D. porrectus complex by DNA sequencing of samples across its distrib utional range and correlating this phylogenetic information with patterns of morphological and anatomical variation, focusing on plants in the Yucatn and Florida. Determination of the appropriate species de limit ation within the D. porrectus complex will be based on the molecular and phenotype data considered in light of the evolutionary, diagnostic, morphological/phenetic and apomorphic species concepts (de Queiroz, 2007; Judd et al., 2007; de Queiroz & Good, 1997; Donoghue, 1985; Coyne & Orr, 2004)

PAGE 18

18 Table 1 1. List of herbarium specimens used to generate a distribution map of Dendrophylax porrectus The location from which each specimen was collected are listed. Vouchers are listed by c ollector and collector number; h erbaria are where the voucher is deposited. 1 : v oucher examined, 2 : database data ( specimen not examined) 3 : literature, 4 : photo material, 5 : unvouchered DNA sample Taxon Location : Country, State, County Voucher (Herbaria)/source D. porrectus Cuba Pinar del Ro, Sandina Ackerman 4514 (UPRRP) 5 D. porrectus Dominican Republic San Juan, Guanito Whitten 1950 (FLAS) 1 D. porrectus El Salvador, San Salvado r, La Libertad Hamer 506 (SEL) 1 D. porrectus Cayman Islands Raymond Tremblay s.n. 5 D. porrectus Guatemala El Progreso Dix 7502 (UVAL) 3 D. porrectus Guatemala Pet n Dix 7861 (UVAL) 3 D. porrectus Jamaica Carlsward 184 (FLAS) 1 D. porrectus Mexico, Campeche, Carnevali 4352 (CICY) 1 D. porrectus Mexico, Campeche, Municipio Carmen Carnevali 5822 (CICY) 1 D. porrectus Mexico, Campeche, Municipio Champoton Carnevali 4920 (CICY) 1 D. porrectus Mexico Chiapas Escuintla Matuda 18674 (MEXU) 2 D. aff. porrectus Mexico Jalisco La Huerta Salazar 8279 (MEXU) 2 D. aff. porrectus Mexico Jalisco Puerto Vallarta Warford 652 (SEL ) 1 D. aff porrectus Mexico Oaxaca San Carlos Yautepec Salas 32294 (MEXU) 2 D. porrectus Mexico, Quintana Roo Carnevali 6312 (FLAS) 1 D. aff. porrectus Mexico, Veracruz Xalapa Salazar 5204 (AMO) 4 D. aff. porrectus Mexico Yucatn Merida Carnevali 5907 (FLAS) 1 D. porrectus Mexico Yucatn Municipio Sotuta Ramirez 886 (CICY) 1 D. porrectus Puerto Rico Las Cobanitas Ackerman 3340 (UPRRP) 5 D. porrectus USA Florida, De Soto Shuey 1847 (USF) 1 D. porrectus USA Florida, Glades Carlsward 329 (FLAS) 1 D. porrectus USA Florida, Highlands Luer 100 (SEL) 1 D. porrectus USA Florida, Hillsborough Van Hoek 0594 (USF) 1 D. porrectus USA Florida, Lee Whitten 3745 (FLAS) 1 D. porrectus USA Florida, Manatee Shuey 1714 (USF) 1 D. porrectus USA, Florida, Orange Stoutamire s.n. (USF, 85314) 1 D. porrectus USA Florida, Pasco Genelle 1582 (USF) 1 D. porrectus USA Florida, Polk Wunderlin 9201 (USF) 1

PAGE 19

19 Table 1 1 (cont.) Taxon Location : Country, State, County Voucher (Herbaria)/source D. porrectus USA Florida, Glades Goldman 2271 (FLAS) 1 D. porrectus USA Florida, Hernando Lakela 28648 (USF) 1 D. porrectus USA Florida, Lee Carlsward 330 (FLAS) 1 D. porrectus USA Florida, Collier Woodmansee 1816 (USF) 1

PAGE 20

20 Figure 1 1. Distribution of Dendrophylax porrectus sensu lato Locations are based on h erbarium specimens literature (Dix & Dix, 2000) and photo material (pers. comm. G. Salazar & R. Jimnez Machorro 2011) Dendrophylax porrectus is distributed through out peninsular Florida the West Indies (Cuba, Cayman Islands, Jamaica, Dominican Republic, and Puerto Rico) and in Central America (Mexico, Guatemala, and El Salvador)

PAGE 21

21 CHAPTER 2 MATERIALS AND METHOD S Plant Material Specimens used in this investigation were obtained from wild collected plants and herbarium specimens. Most of the wild collected specimens are f rom Florida, USA and Yucatn and Jalisco in Mexico. The roots of the plants were placed in RNAlater (Bent & Taylor, 2010) or dried in silica gel (Chase & Hills, 1991) Samples from Florida were collected from private land s in Sarasota and Ft. Myers counties with permission of landowne rs. Collected plants were cultivated in the greenhouse of the University of Florida H erbarium (FLAS) In collaboration with Centro de Investigacin Cientfica de Yucatn in Mexico (CICY) plants were collected from the Yucatn peninsula These plants were cultivated, photographed and measured in Mexico by Germ n Carnevali (CICY) Herbarium specimens from USF, SEL, CICY, and FLAS (Holmgren et al., 1990) were examined for this study. Destructive sampling took place on some specimens from CICY. Since (Carlsward, 2004) DNA from her specimens was used to se quence additional DNA regions (T ab le 2 1). Morphological A nalyses General Morphology Morphometric measurements were taken from live or preserved plant material. Detailed measurements and observations of the roots and flowers were performed with a WILD Heerbrugg M3 dissecting stereo micr oscope. Five roots fr om five individuals of both thin and thick rooted plants were drawn with a Camera Lucida attachment (WILD Heerb r ugg type 256575)

PAGE 22

22 Within each root, five random parts of the roots were selected for measurements These recorded measurements were statistically analyzed with a nested ANOVA to test whether there is a significant difference between the thicknesses of the roots of both taxa Fifteen flowers of both thin and thick rooted taxa fresh or in 70% ethanol we re dissected and drawn with a Camera Lucida These drawing were used to measure the width and length of all the perianth parts (including the callu s), the ovary, and the anther. A Hotelling T 2 test (Johnson & Wichern, 1988) was performed to investigate if the perianth parts of D endrophylax porrectus populations are significant ly different A paired T test assuming equal variance was also performed on the remaining floral parts of the two putative taxa. Root Anatomy Roots were embedded for seven days in FAA (ethyl alcohol formaldehyde acetic acid) and vacuumed to remove air from the tissue and to ensure complete penetration of the solution T he roots were transferred to 70% ethanol a fter seven days R esin embedding was used to make cross sections of 8 in thickness T he resin embedded roots were selectively harvested 1 2 cm from the growing tip This procedure was followed to ensure that the roots are in the same developmental stage, which makes it suitable for compar ison studies. Resin embedding : Five roots of different individual plants per taxon were embedded in LR White resin (Electron Microscopy Sciences) using an infiltration process with LR White monomer/ethanol series (see steps in Table 2 1) based on rotocol (Ruzin, 1999) After the infiltration process, the roots were placed in Eppendorf tubes with monomer to polymerize at temperature of 60 65 C for 12 24

PAGE 23

23 hours. Th e roots were sectioned with a rotary microtome (AO Spencer 820) and sections were stained with 0.1 % aqueous Toluidine Blue O solution in 0.1M S phosphate buffer solution, pH 5.8 (Ruzin, 1999) Section s were then air dried mounted with Permount and photographed with a Zeiss Axiocam HRm camera mounted on a Zeiss Axioplan 2 Imaging microscope The randomly chosen cell walls of the endodermis in the roots were mea sured through the photographs with ImageJ (Rasband, 1997 2011) The measurements of both taxa and both protocols were statistically analyzed in a nested ANOVA to investigate if there is a significant difference between the two taxa. Molecular A nalyses DNA Extraction DNA material was obtained from wild collected plants, herbarium dried specimens, silica gel dried material (Chase & Hills, 1991) and cultivated plants (see T able 2 1). Fresh collected roots were field preserved in RNAlater (Bent & Taylor, 2010) Genomic DNA was extracted us ing the cetyl trimethylammonium bromide (CTAB) technique (Doyle & Doyle, 1987) modified and scaled to a 1 ml volume reaction. Approximately 2 cm of root tissue was ground in 1.2 ml CTAB 2X buffer and 5 L of Proteinase K using a mortar and pes tle. The homogenized mixture was transferred to Eppendorf tubes and incubated at 50 C for 2 hours. After incubation, 500 L of chloroform/isoamyl alcohol was added to the CTAB tissue mixture and vortexed until milky. The samples were centrifuged at 10,000 rpm for 4 minutes and 750 L of the supernatant was transferred to a new tube. In order to precipitate the DNA, 30 L of 3 M sodium acetate and 600 L of 100% isopropanol was added to the tube and gently mixed. The solution was allowed to chill

PAGE 24

24 overnight i n a 20 C freezer for maximum precipitation of DNA. After chilling, the solution was centrifuged at 13,000 rpm for 20 minutes to obtain a DNA pellet. The alcohol/ sodium acetate solution was poured off without disturbing the pellet and washed 3 times with 1ml of 70% ethanol In order to resuspend the DNA pellet, 200 L of 1X Tris EDTA (ethylenediaminetetraacetic acid) buffer (TE pH 8.0 ) w as added to the tube. After being incubated for 15 minutes at 65 C the samples were stored at 20 C. DNA extractions from herbarium material and impure DNA were cleaned with QIAquick PCR purification columns (Qiagen, Valencia, California, USA) and Buffer PE, then eluted with Buffer EB to remove inhibitory secondary compounds. The samples were resuspended EDTA (TE) buffer. Amplification and Sequencing Amplifications were performed using an Eppendorf Mastercycler EP Gradient S In this study three regions were amplified: o ne nuclear reg ion and two plastid regions The pr imers for these regions are in T able 2 3. The nuclear region ITS was amplified with the primers 17SE and 26SE (ITS 1 + 5.8S rDNA+ ITS 2) from Sun et al (1994) The plastid region matK trnK includes the entire matK trnK spacer that is ca 18 00 base pairs (bps) in length. This region was sequenced with the primers 19F (Molvray et al., 2000) and trnK 2R (Johnson & Soltis, 1994) The internal sequencing primers were matK intF and matK intR Some samples were amplified using the primers 56F and 1520R (Whitten et al., 2000) that yielded a shorter, but still nearly complete sequence of the matK exon (mi The second plastid region used in this study was ycf1 (Neubig et al., 2009) with The primers used were 3 720F and 5500R with the internal

PAGE 25

25 sequencing primers intF and intR. All the components used for the polymerase chain r eactions and the programs used for the thermocycler are shown in Table 2 4 and T able 2.5. The PCR products were sequenced at the Interdisciplinary Center for Biotechnology Research facility (ICBR) at the University of Florida and the sequenced data were ed ited and assembled using Sequencher 4.10.1 (GeneCodes, Inc, Ann Arbor, MI, 2010) All sequences were deposited in GenBank ( see Appendix ). Data Analysis The consensus files of each region for every taxon were used to assemble a manually aligned matrix using Se Al v2.0a11 (Rambaut, 1996) The matrix was then transferred to PAUP*4.0b10 (Swofford, 2003) for phylogenetic analyses. All nucleotide characters were weighed equally and unordered. The gaps were treated as missing data and indels were not coded as characters. Heuristic searches were performed with 1000 random a ddition replicates, saving 10 trees per replicate, with the tree bisection reconnection (TBR) algorithm. Deltran optimization was used for all analyses. Bootstrap analyses utilized 1000 replicates, with 10 random addition replicates (SPR swapping) per boot strap replicate. The partition homogeneity test (HTF) was used to investigate if the data were congruent with one another or if they were significant ly different Combined datasets with missing data were tested with PAUP*4.0b10 (Swofford, 2003; Johnson & S oltis, 1998) With the h euristic search, HTF tests were performed using 100 replicates and a TBR algorithm comparing all four combinations of the three genes (including plastid genes vs nuclear genes) Ten random addition replicates were performed per HTF

PAGE 26

26 replicate, holding 10 trees and saving no more than 10 trees per replicate The p robability values were greater than 0.05 The jModelTest (Posada, 2008) was used to carry out statistical selection of best fit models of nucleotide substit ution. The results were used to perform the Maximum Likelihood analysis in PAUP*4.0b10. Heuristic searches were performed with 100 random addition replicates, saving 10 trees per replicate, with the tree bisection reconnection (TBR) algorithm. Bootstrap an alyses utilized 100 replicates, with 10 random addition replicates (SPR swapping) per bootstrap replicate. Carlsward 's molecular data (Carls ward, 2004) provided sequence data of ITS and matK The trnL F sequence data were omitted from this study because of its lo w variability Campylocentrum micranthum (Rchb.f.) Rolfe was use d as an outgroup. The sequence data for this taxon were obtained f rom GenBank. An effort was made to amplify 46 specimens, but not all regions amplified successfully. Five individuals per population were sequenced for four populations using ITS to determine variation within populations Analyses were conducted with sever al data sets:(1) ITS data set containing 46 individuals (3 taxa missing data); (2) matK data set containing 25 individuals (6 taxa missing data); (3) ycf1 data set containing 26 individuals (5 taxa missing data); (4) combined plastid data sets ( matK and yc f1 ) with 27 taxa (with gaps for the missing plastid data); and (5) total combined data sets of nuclear and plastid regions, using 31 taxa (again with g aps for the missing plastid and nuclear data). Phenology In order to understand the reproductive mechani sms of Dendrophylax porrectus the plants were monitored during their flowering period. The flowering period was recorded and the flowers were labeled and observed throughout their lifespan (opening

PAGE 27

27 to wilting). All the wild collected plants were collected with young inflorescences developing, which eliminated the possibility that plants were influenced in their reproductive initiation by the greenhouse condition s Attempts were made to observe potential pollinators and the p lants were also tested to determine self compatibility Florida p lants with open flowers were completely covered with a piece of mesh to prevent potential pollinators having access to the flowers. The remaining plants were kept outside at the FLAS greenhouse to allow pollinators to visit the flowers The same procedure was implemented for the Yucatn plants in Mexico.

PAGE 28

28 Table 2 1. Taxa used for molecular phylogenetic study. The specimen numbers are indicated. Gene regions sampled include: I = ITS, M = matK and Y = ycf1 The location from which each spe cimen was collected are listed. Numbers linked to the locations indicate different individuals from one or more population. Vouchers are listed by collector and collector number; He rbaria are the herbarium where the voucher is deposited. Taxon Specimen number Gene region Location (County) Voucher (Herbaria) Campylocentrum micranthum B346 I, M, Panama Carlsward 315 (FLAS) C. micranthum B60 I, M, Y Mexico Carlsward 180 (FLAS) C. micranthum B12 I, M, Puerto Rico Ackerman 3341 (UPRRP) Dendrophylax alcoa Dod W778 I Dominican Republic Ackerman 2773 (UPRRP) D. barrettiae Fawc & Rendle B36 I, M, Y Jamaica 1 Carlsward 199 (FLAS) D. barrettiae W714 I, Y Jamaica 2 Whitten 1814 (FLAS) D. aff. porrectus B14 I, M, Y Mexico ( Yucatn 1 ) Carnevali 5907 (FLAS) D. aff. porrectus M1 I Mexico ( Yucatn 2 ) Carnevali 5907 (CICY) D. aff. porrectus M2 I, M, Y Mexico ( Yucatn3 ) Carnevali 5907 (CICY) D. aff. porrectus M3 I Mexico ( Yucatn4 ) Carnevali 5907 (CICY) D. aff. porrectus M4 I Mexico ( Yucatn5 ) Carnevali 5907 (CICY) D. aff. porrectus M5 I Mexico ( Yucatn6 ) Carnevali 5907 (CICY) D. aff. porrectus GS 8279 I, M, Y Mexico ( Jalisco ) Salazar 8279 (MEXU) D. fawcettii Rolfe W3265 Y Cayman Islands Whitten 3265 (FLAS) D. funalis (Sw.) Benth ex Rolfe B233 I, M, Y Jamaica 1 Carlsward 302 (FLAS) D. funalis W713 I, Y Jamaica 2 Whitten1935 (FLAS) D. lindenii (Lindl.) Benth ex Rolfe W730 I USA F L (Collier) Ward 5365 (FLAS) D. lindenii W716 I, M, Y Cuba Claude Hamilton s.n. D. porrectus JDA4514 I, M, Y Cuba ( Sandina) Ackerman 4514 (UPRRP) D. porrectus B359 I, M, Y Dominican Republic (Guanito) Whitten1950 (FLAS) D. porrectus B710 I, M, Y Cayman Islands no voucher D. porrectus B35 I, M, Y Jamaica Carlsward 184 (FLAS) D. porrectus C4468 I, Y Mexico (Campeche) Carnevali 4468 (CICY) D. porrectus B212 I, M, Y Mexico (Quintana Roo) Carnevali 6312 (FLAS) D. porrectus IR886 M, Y Mexico ( Yucatn 7 ) Ramirez 886 (CICY) D. porrectus B366 I, M, Y USA FL 1 (Glades ) Carlsward 329 (FLAS) D. porrectus DP1 I, M, Y USA FL 2 (Hillsborough 1 ) no voucher

PAGE 29

29 Table 2 1 (cont.) Taxon Specimen number Gene region Location (County) Voucher (Herbaria) D. porrectus DP3 I USA FL 4 (Hillsborough 1 ) no voucher D. porrectus DP4 I USA FL 5 (Hillsborough 2 ) Molgo 221 (FLAS) D. porrectus DP5 I USA FL 6 (Hillsborough 2 ) Molgo 221 (FLAS) D. porrectus D P 30 I USA FL 7 (Lee ) Whitten 3745 (FLAS) D. porrectus D P 31 I USA FL 8 (Lee ) Whitten 3745 (FLAS) D. porrectus D P 32 I USA FL 9 (Lee ) Whitten 3745 (FLAS) D. porrectus D P 33 I USA FL 10 (Lee ) Whitten 3745 (FLAS) D. porrectus W3745 I, M, Y USA FL 11 (Lee ) Whitten 3745 (FLAS) D. porrectus DP 40 I, M, Y USA FL 12 ( Sarasota) Molgo 222 (FLAS) D. porrectus DP41 I USA FL 13 ( Sarasota) Molgo 222 (FLAS) D. porrectus DP42 I USA FL 14 ( Sarasota) Molgo 222 (FLAS) D. porrectus DP44 I USA FL 15 ( Sarasota) Molgo 222 (FLAS) D. porrectus DP45 I USA FL 16 ( Sarasota) Molgo 222 (FLAS) D. porrectus DP 24 I USA FL 17 (Hillsborough 3 ) no voucher D. porrectus B70 I M, Y USA FL 18 (Glades ) Goldman 2271 (FLAS) D. porrectus B367 M, Y USA FL 19 (Ft. Myers) ) Carlsward 330 (FLAS) D. porrectus B 11 I, M, Y Puerto Rico Ackerman3340 (UPRRP) D. sallei (Rchb.f.) Benth ex Rolfe B360 I, M, Y Dominican Republic Whitten1945 (JBSD) D. varius (Gmel.) Urb B153 I Dominican Republic 1 Thompson 10683 (SEL) D. varius (Gmel.) Urb B362 I, M, Y Dominican Republic 2 Whitten1960 (JBSD) D. varius (Gmel.) Urb W779 I Puerto Rico Ackerman 2727 (UPRRP)

PAGE 30

30 Table 2 2. I nfiltration and polymerization of LR White resin with the root tissue Step # Solution Time 1 100% Ethanol 2 hours 2 50% monomer +50% 100%Ethanol 24 hours 3 70% monomer + 30 % 100% Ethanol 24 hours 4 100% monomer 24 hours 5 100% monomer 24 hours 6 Polymerase in100% monomer at 65 C 12 24 hours Table 2 3 Primer sequence of one nuclear and two plastid gene regions Primer Primer sequence ITS 17SE, forward ACGAATTCATGGTCCGGTGAAGTGTTCG 26SE, reverse TAGAATTCCCCGGTTCGCTCGCCGTTAC matK trnK 19F, forward CGTTCTGACCATATTGCACTATG 1520R, reverse CGGATAATGTCCAAATACCAAATA 56F, forward ACTTCCTCTATCCGCTACTCCTT trnK2R, reverse ACCTAGTCGGATGGAGTAG intF, forward TGAGCGAACACATTTCTATGG intR, reverse ATAAGGTTGAAACCAAAAGTG ycf1 3720F, forward TACGTATGTAATGAACGAATGG 5500R, reverse GCTGTTATTGGCATCAAACCAATAGCG intF, forward GATCTGGACCAATGCACATATT intR, reverse TTTGATTGGGATGATCCAAGG Table 2 4 Components of polymerase chain r eaction with Sigma brand reagents ITS matK trnK ycf1 Water (H 2 O) 11.0 14.5 18 14.5 18 10X buffer 2.5 2.5 5.0 2.5 5.0 25 mM MgCl 2 2.5 1 3 1 3 ) 0.5 0.5 0.5 ) 0.5 0.5 0.5 dNTPs (10 0.5 0.5 0.5 Taq polymerase 0.2 0.2 0.2 DNA template (~10 100 ng) 1.0 1.0 1.0 If the DNA templates are from a herbarium specimen the quantity of template will vary from 2

PAGE 31

31 Table 2 5 Thermocycler programs for polymerase chain reactions Steps # Temperature ( C ) Time N otes ITS 1 94 2 minutes 2 94 1 minute Step 2 4,15 times 3 76 1 minute Reducing 1C per cycle in step 3 4 72 1 minute 5 94 1 minute Step 5 7 21 times 6 59 1 minute 7 72 1 minute 8 72 3 minutes matK trnK 1 94 3 minutes 2 94 45 seconds Step 2 4 33 times 3 60 45 seconds 4 72 2 minutes 5 72 3 minutes ycf1 1 94 3 minutes 2 94 30 seconds Step 2 4 8 times 3 60 1 minute 4 72 3 minutes 5 94 30 seconds Steps 5 7 30 times 6 50 1 minute 7 72 3 minutes 8 72 3 minutes

PAGE 32

32 CHAPTER 3 RESULTS General M orphology The Florida and Yucatn populations differ significantly in numerous subtle morphological traits, including root thickness, ovary length, labellum width, callus length, anther size, pollinium shape, and rostellum shape. Figure 3 1 presents photographs of r epresentative individuals from Florida and Yucatn populations and highlights their differences These differences are summarized in Figures 3 2 to 3 9. Root T hickness A root thickness analysi s was performed with 250 measurements and 125 per taxa. The root s from the Yucatn population were thicker (mean diameter 1.90 mm ) than those of the Florida populations (mean diameter 1.23 mm). The nested ANOVA analysis teste d whether there is a plant location effect present and if there was a variation among root thick ness within locations The re is no location effect if all plants are equal and there is a location effect if not all of the plants are equal. The test resulted in a test statistic: F observed = 20.814, P value: (F 1,8 20.814) = 0.00185. These resul ts imply that the plants are significantly different between the F lorida and Yucatn locations There is no variation among root thickness within location if all thicknesses are equal and there is an effect if they are not equal. The test resulted in a test statist ic: F observed = 9.164, P value: (F 8 40 9.164) 0.000. These results imply that the root thickness variation is significantly different in both locations. In Figure 3 2, there is a box plot that represents the data of the root thickness. The boundary of th e box closest to zero indicates the 25 th percentile, a line within the box marks the median, a dashed line

PAGE 33

33 the mean, and the boundary of the box farthest from zero indicates the 75 th percentile. The error bars above and below the box indicate the 90 th and 10 th percentiles. (The symbols used in the box plots in Figures 3 3 and 3 8 are the same.) Ovary L ength Fifteen ovaries were measured of each population in Florida and Yucatn. The mean leng th of the ovary in Florida plants was 2.00 mm versus 1.50 mm for Y ucatn plants. A two sample T test was performed to test whether a significant difference was present. The test resulted in t = 4.088 with 28 degrees of freedom, P value = <0.001. These results imply that the ovary length is significantly different in the two populations. The box plot in Figure 3 3 shows the distribution of the ovary length data. Sepal M easurements The flowers of the Florida and Yucatn populations hav e bot h a width of 3 mm ( Fig 3 1) but when the perian th parts are uncurled, it is seen that they are significantly different The length and width of the dorsal sepal were measured in 15 flowers of the Florida and Yucatn populations A paired Hotelling T 2 test was performed to investigate if the length and width of the dorsal sepal of both populations are equal or not The mean dorsal sepal length and width for Florida and Yucatn populations were, 1.88 x 1.13 mm. vs. 2.27 x 1.15 mm. The test resulted in T 2 =70.267, test statistic: F observed = 33.878 P value :(F 2,27 33.878 These r esults imply that the dorsal sepal is significantly different in the two populations. The scatter plot in F igure 3 4 depicts the length of the dorsal sepal on the X axis and the width on the Y axis and visually demonstrates the difference between the Florida and Yucatn populations. The Hotelling T 2 test was also performed with the lateral sepal data to test whether the mean length and width of the lateral sepal are equal or not equal. The mean lat eral

PAGE 34

34 sepal length and width for Florida and Yucatn were, 2.16 x 1.03 mm. vs. 2.37 x 1.12 mm. The test resulted in T 2 = 10.75 test statistic: F observed = 5.18 P value : (F 2,27 5.18 ) =0.012. These results imply that the lateral sepal is significantly differen t in both populations The scatter plot in Figure 3 5 depicts the length of the lateral sepal on the X axis and the width on the Y axis and indicated that there is much overlap in the pattern of variation, although the populations are differentiated. Peta l and L abellum M easurements The average petal measurements wer e analyzed with the Hotelling T 2 test to investigate if the petals in the Florida and Yucatn populations were equal. The mean petal length and width for Florida and Yucatn were1.99 x 0.85 mm. vs. 2.36 x 0.86 mm. The test resulted in T 2 = 27.29 test statistic: F observed = 13.16 P value : (F 2,27 13.16 ) = 0.00103. These results imply that the petals are significantly different in both populations. The scatter plot in Figure 3 6 depicts the length of the pet al on the X axis and the width on the Y axis and illustrates the differences between these two populations. The data gathered from the labellum was a lso tested with the Hotelling T 2 to test whether the labella from the two populations are equal. The mean labellum length and width for Florida and Yucatn populations were, 1.99 x 0.85 mm vs. 2.36 x 0.86 mm. The test resulted in T 2 = 40.05 test statistic: F observed = 19.31 P value : (F 2,27 13.16 ) 0.000. These results imply that the labellum is sign ificantly different in both populations The scatter plot in Figure 3 7 depicts the variation in the labellum shape. The calli on the labella were measured for Florida and Yucatn plants The mean length of the callus of the Florida plants was 0.37 mm versus 0.15 mm for that of Yucatn plants A two sample T test was performed to investigate whether a significant

PAGE 35

35 difference was present. The test resulted in t = 7.843 with 24 degrees of freedom, P value = <0.001. These results imply that the callus is significantly different in Florida and Yucatn populations The box plot in Figure 3 8 depicts the distribution of the callus length data. Anther M easurements The Hotelling T 2 test was also performed with the anther data to test whether the mean length and width of the anther are equal or not equal. The mean anther length and width for Florida and Yucatn were, 0.49 x 0.34 mm. vs. 0.59 x 0.43 mm. The test resulted in T 2 = 82.99 test statistic: F observed = 3.976 P value : (F 2,2 3 3.976 ) 0.000. The se results imply that the anther is significantly different in both populations The scatter plot in Figure 3 9 depicts the length of the anther on the X axis and the width on the Y axis and shows a clear separation of the Florida and Yucatn populations. In Figure 3 1 t he pollin ia of the Yucatn individuals are slightly flattened and ovoid and the pollin ia of the Florida individuals are globose The stipes of the Florida pollinia appear a s a single structure in Figure 3 1, but they are not In fact, the s tipes represent two single structures attached to one another as they appear when attached to the rostellum. Even though the rostellum is the same in size, the rostellar entry is not The entry in the Florida individual s is 0.06 mm wider than the Yucatn e ntities Root A natomy Differences in the root section s were observed as co mparisons were made of resin embedded roots of the Florida and Yucatn populations ( Fig 3 10). The endodermis wall o f the Yucatn entities is smaller versus the wall of the Florida entities The data was normally distributed and equal in variance (Fig 3 11 ). The mean of the Florida cell wall was 3.645 m and the mean of the Yucat n was 2.004 m. The t

PAGE 36

36 t est of comparison resulted in a t = 9.329 with 48 degrees of freedom (P value = <0.001) These results imply that the endodermis wall of the roots is significantly different in the two populations. Molecular A nalysis In all analyses, Dendrophylax is monophyletic which concurs with the results of Carlswa rd et al (2003) Three outgroups were used, all Campylocentrum micranthum but with three accession s from different countries. The populations of Dendrophylax porrectus form a monophyletic group, which is comprised of two diagnosable subclades, here called the thick rooted cl ade and the thin rooted clade. Within the thin rooted clade ther e are two sister groups, which are identified as thin rooted clade s 1 and 2. Statistical comparisons of analyses for each gene region and combined gene regions are given in Table 3 1. ITS A nalyses of D. porrectus In the ITS analysi s 46 accessions were used and the analysis supported the monophyly of Dendrophylax ( 95% BS; 92% JK ; Fig 3 12 ) The D. porrectus clade is well supported ( BS 98% ; JK 93% ), with a thin rooted clade ( BS 98% ; JK 93% ) The thin rooted D. porrectus includes two subclades, clade 1 ( BS 86% ; JK 75 % ) and clade 2 ( BS 99 %; JK 96% ) Dendrophylax porrectus clade 2 shows a polytomy of all the various Florida populations and the ir replicates A polytomy is also seen in the thick rooted D. porrectus clade. This indicates that all the replicates in the populations have no sequence divergence and are genetically very similar. The thick rooted clade is strongly supported (BS 100 %; JK 100%) indicating that the Yucatn and Jalisco plants are sister to the Florida populations

PAGE 37

37 The other Dendrophylax species have the same topology as reported by Carlsward et al (2003) Dendrophylax barrettiae is sister to the rest of the Dendrophylax clade The sub clade with D funalis and D fawcettii is strongly supported (BS 99%; JK 97%) The sub clade with D varius D sallei D lindenii and D alcoa is moderate l y supported (BS 81% ; JK 77%) but within this monophyletic group there is strong suppo rt for groupings among the different species. matK A nalyses of D. porrectus The matK matrix had 25 taxa with only one representative per population The analysis supported the monophyly of the Dendrophylax clade ( BS 88 %; JK 85% ; Fig 3 13 ) The branch between D barrettiae and the rest of the Dendrophylax clade is poorly supported ( BS 60% ; JK 61% ) The D porrectus clade is strongly supported ( BS 99% ; JK 9 6% ) The D. porrectus clade has the same topology as in the ITS analysis The support for the thick rooted clade is still high (BS 98 % ; JK 92%), but the thin rooted D. porrectus clade is only weakly supported ( BS 63 % ) This does not mean that there is a conflict in the data but merely that a well supported clad e in ITS is poorly supported in matK d ue to lack of phylogenetic signal. The D funalis + D fawcettii subclade is strong ly support ed ( BS 100% ; JK 99% ) The D lindenii + D sallei + D varius subclade h as moderate support ( BS 87% ; JK 76% ) ycf1 A nalyses of D. porrectus There were 26 accessions in the ycf1 matrix and the analysis generated the same topology as the ITS and matK analyses ( Fig 3 14) T here is strong support for the monophyly of D. porrectus ( BS 100 % ; JK 100% ) The thin rooted clade 1 is supported by BS 94%; JK 89% For both BS and JK there is 100% support for thin rooted clade 2

PAGE 38

38 and the thick rooted clade is supported by BS 100%; JK 99% The other Dendrophylax species show a simi lar topology as in the matK analysis. Dendrophylax barrettiae is strongly supported ( BS; JK 100%) D endrophylax funalis and D fawcettii are a clade ( BS 99%; JK 96%), which was also reported in Carlsward et al (2003) a nd D lindenii D sallei and D varius form a well supported clade ( BS 90% ; JK 87% ) Combined Plastid A nalyses of D. porrectus In the combined plastid analyse s, there were 27 accessions and the analysis supported the monophyly of the Dendrophylax clade ( BS 92%; JK 87% ; Fig 3 15 ). The D. porrectus clade is also strongly supported ( BS 100% ; JK 100% ), as is the thick rooted clade (BS 100 %; JK 100%), but the thin rooted D. porrectus clade is only moderately supported (BS 79 %; JK 67%). The thin rooted D. porr ectus clade 1 is well supported (B S 95 %; JK 88%) and the thin rooted clade 2 also shows strong support (BS 100 %; JK 100%). In this tree, D barrettiae is sister to the rest of the Dendrophylax taxa. These results are congruent with the previously presented trees. Combined ITS and P lastid A nalyses of D. porrectus The combined ITS, matK and ycf1 matrix included 31 taxa. Most clades have strong BS support in this analysis and there were no conflicts between the regions ( Fig 3 16 ). A partition homog eneity test with heuristic search supported the congruence between these three datasets: for ITS and plastid (p value = 0.99), for ycf1 and matK (p value = 1.00), for ITS and matK (p value = 1.00) and for ycf1 and ITS (p value = 1.00). The Dendrophylax cl ade is strongly supported with BS 100%; JK 99%. The support for the D. porrectus clade is 99% for BS and JK and both thin rooted and thick rooted clades also are strongly supported. The thin rooted clade divides into two subclades: thin rooted 1 and 2. These subclades are also strongly supported ( BS; JK

PAGE 39

39 higher than 90% ) Dendrophylax barrettiae is sister to the rest of the taxa of Dendrophylax and the other species and species groups are well supported, which agree with the results of Carlsward et al (2003) The results of the comb ined plastid and ITS analysis are congruent with the trees of the separate gene regions and there were no contradictions when the combined plastid tree was compared with the three gene tree Maximum L ikelihood Only the three gene matrix with 31 accessions was used for the Maximum Likelihood (ML) analyses. The jModelTest selected the GTR+G model (Lanave et al., 1984) as the best fit statistical selection of nucleotide substitution with the following criteria: a negative log likelihood ( lnL) of 7547.2470, number of estimated parameters (K) = 69, Akaike Information Criterion (AIC) : 15232.4940, AIC difference weight: 0.0000 and a cumulative AIC weight of 0.5742. The ML tree has the same topology as th e three gene tree generated by parsimony analysis ( Fig 3 17 ). The Dendrophylax clade is 100% supported as well as the D. porrectus clade Strong support (BS 100%) was also recorded for the thin rooted and thick rooted clade s However, slight differences are noticeable in the trees. The bootstrap supports in the ML tree are slightly higher which led to suppor t of bra nches that were not supported under parsimony. Distribution of D. porrectus The three most important monophyletic groups are placed o n a distribution map to clarify biogeographical relationships among the different populations of Dendrophylax porre ctus ( Fig 3 18 ). Black circles indicate the thin rooted clade 1, blue circles indicate the thin rooted clade 2 and red circles show the location of collections of the thick rooted clade The red squares indicate voucher specimens/ photo material that are

PAGE 40

40 considered to represent the thick rooted clade (pers. comm. G. Salazar & R. Jimnez Machorro, 2011) The black squares are specimens that were examined with photo material, but a definite decision could not be made if they belong to the thin rooted or thick rooted clade due to limitations of the images The white squares indicate population s that were not determined a clade membership, due to the fact that specimens of these populations were not examined. Phenology Several plants from the Florida populations and the Yucatn populations were observed in cultivation during their flowering period. When the members of these populations where collected, inflo rescences were already developing The flowering period of the F lorida entity started on August 30 th and lasted until September 24 th The Yucatn entity started flowering on September 24 th and lasted until January 20 th The wilted Florida flowers fell off after a day if not pollinated and the Yucatn flowers remained a ttached to the inflorescence up to 5 days The dried up flowers then turn ed black on the inflorescence. In the populations of Ft. Myers ( FL), I observed that the plants flowered twice in one year. Plants flowered from August until September and flowered fo r the second time from December until January. This observation was not seen in the Sarasota (FL) population. The same patterns were seen in cultivation and in plants in their natural habitat s If the flowers of both populations were not pollinated, the li fe span of the flower of Florida populations was between 10 and 15 days and that of the Yucatn population was between 12 and 19 days. The flowers of both entities had a fragrance at night. The Florida entity had a strong sweet honey like fragrance but the Yucatn entity had a much milder fragrance that is difficult to describe.

PAGE 41

41 Table 3 1. Comparison of tree statistics for each gene region and combinations of these gene regions for parsimony analyses. Statistical inference ITS matK ycf1 matK & ycf1 ITS & matK & ycf1 Aligned characters 702 1355 1689 3044 3746 Uninformative characters 30 28 67 95 123 Informative characters 85 42 74 116 201 Number of trees saved 4 3 1 1 18 Tree length 161 78 168 239 397 Consistency index (CI) with uninformative characters 0.850 0.923 0.875 0.916 0.889 CI without uninformative characters 0.812 0.88 0.792 0.861 0.838 Retention index (RI) 0.952 0.949 0.936 0.955 0.942 Rescaled consistency index (RC) 0.810 0.876 0.819 0.875 0.837

PAGE 42

42 Figure 3 1 Comparison of two representative individuals (Florida, A and Yucatn B) of Dendrophylax porrectus illustrating measurements of d ifferent plant and floral parts. Ha = habit, Rt = root An = anther, Ca = callus, Ov = ovary, Ro = rostellum, and Po = pollinar ia.

PAGE 43

43 Figure 3 2. The difference in root thickness within two populations of D. porrectus in Florida and Yucatn

PAGE 44

44 Figure 3 3. The difference in ovary length within two populations of D. porrectus in Florida and Yucatn

PAGE 45

45 Figure 3 4. Differences in the dorsal sepal of D. porrectus within the populations of Florida and Yucatn The dark green represent Florida and the light green represent Yucatn. The length of the dorsal sepal is plotted on the X axis and the width on the Y axis

PAGE 46

46 Fi gure 3 5. Differences in the lateral sepal of D. porrectus within the populations of Florida and Yucatn The dark green represent Florida and the light green represent Yucatn. The average lengths of the lateral sepals are plotted on the X axis and the a verage width on the Y axis.

PAGE 47

47 Figure 3 6. Differences in the petal s of D. porrectus within the populations of Florida and Yucatn The dark green represent Florida and the light green represent Yucatn. The average lengths of the lateral sepals are plotte d on the X axis and the average width on the Y axis.

PAGE 48

48 Figure 3 7. Differences in the labellum of D. porrectus within the populations of Florida and Yucatn The dark green represent Florida and the light green represent Yucatn. The length of the labellum is plotted on the X axis and the width on the Y axis.

PAGE 49

49 Figure 3 8. The difference in callus length within two populations of D. porrectus in Florida and Yucatn

PAGE 50

50 Figure 3 9. Differences in the anther of D. porrectus within t he populations of Florida and Yucatn The dark green represent Florida and the light green represent Yucatn. The length of the labellum is plotted on the X axis and the width on the Y axis.

PAGE 51

51 Figure 3 10 Resin embedded root sections of D. porrectus rep resenting Florida and Yucat n populations. A) cross section showing a Florida root. B) close up of the Florida entity showing the stele with the thin walled endodermis cells. C.) cross section of the Yucatn root. D) close up of the Yucatn entity showing the stele with the thick walled endodermis cells.

PAGE 52

52 Figure 3 11 Resin embedded comparisons of cell wall thickness in the endodermis of two D. porrectus populations The black bars represent Florida and the gray bars Yucatn Every five root section numbers on the X axis represent one root of one plant

PAGE 53

53 Figure 3 12. One of the 4 equally parsimonious trees of ITS. The bootstrap values are listed above the branches and the jackknife percentages are listed below the branches. The triangles indicate th e nodes that collapsed in strict consensus.

PAGE 54

54 F igure 3 13 One of the 3 equally par simonious trees of matK Bootstrap value s are listed abo ve the branches and the jackknif e values are listed below the branches The topology of the three trees are identical as non e of the nodes collapsed in a strict consensus.

PAGE 55

55 Figure 3 14 The most parsimonious tree of ycf1 Bootstrap values are above branches; jackknife values below branches.

PAGE 56

56 F igure 3 15 The most parsimonious tree of combined matK and ycf1 analysis. Bootstrap values are above branches; jackknife values below branches.

PAGE 57

57 F igure 3 16 The most parsimonious tree of combined ITS, matK and ycf1 analysis. Bootstrap values are above branches; jackknife values below branches.

PAGE 58

58 Figure 3 17 Maximum Likelihood tree of combined nuclear and plastid data assuming the GTR+ G model, log L = 7547.2470 based on100 replicates Bootstrap support values are indicated above the branches

PAGE 59

59 Figure 3 1 8 Distribution of D. porrectus with the three major c lades shown in three different colors Black circles indicates the thin rooted clade 1, blue circles indicates the thin rooted clade 2 and red circles shows the location of the thick rooted clade The red squares indicate voucher specimens/ photo material that are morphologically similar to the thick rooted clade (pers. comm. G. Salazar & R. Jimnez Machorro, 2011). The black squares are specimens which were photographically examined but a definite decision could not be made if they belong to the thin rooted or thick rooted clade. The white squares indicate herbarium records not seen or unverified and literature reports (Dix & Dix, 2000)

PAGE 60

60 CHAPTER 4 DISCUSSION AND CONCLUSION Discussion The Florida populations differ significantly from the Yucatn populations in many morphological characters, including root thickness, ovary length, perianth shape, anther dimensions, callus length rostellar entry opening and po llinium shape. Although subtle, the difference in root thickness is readily apparent, especially if plants or herbarium specim ens are observed side by side. For plants in flower, the difference in length of the callus is the most reliable feature that dist inguishes the thick rooted Yucat n populations (thick rooted clade ) from the thin rooted populations ( thin rooted clades 1 and 2). The anatomical root results in Fig ure 3 10 shows that the endodermis cell walls of the Florida plants are thicker than the Yucat n plants, but these differences may be an artifact of the age of the root tissue. As the roots age, the endodermis cell wall thickens (Dickison, 2000) Since root growth and development are dependent on environmental factors (Dycus & Knudson, 1957) it is difficult to obtain roots that are in the same developmental stage, which then can be compared. I suggest that the root endodermis cell wall thickness it not a good character to distinguish these two entities. The Yucatn population s are the thick rooted plants and the results of this study revealed that the Yucatn entities are morphologically similar to the populations from Jalisco and Veracruz. The thin rooted plants (Florida entity) are similar to the populations in the Greater Antilles (Cayman Islands, Cuba, Dominican Republic, Jamaica, and Puerto Rico), Campeche and Quintana Roo.

PAGE 61

61 The differences between t he thin and thick rooted entities based on morpholog y correlate perfectly with the DNA based clades as the two entities show high divergence in both ITS and plastid molecular data. Together, these results suggest that these major clades should be recognized as two distinct species based on the phenetic / morphological apomorphic and evolutionary species concept s (de Queiroz & Good, 1997; de Queiroz, 2007; Coyne & Orr, 2004; Donoghue, 1985; Judd et al., 2007) D endrophylax porrectus sensu stricto is comprise d of the thin rooted c lade 1 which occur s in Mexico, the Cayman Islands and the Dominican Republic and the members of the thin rooted clade 2 which occur s in Jamaica, Puerto Rico, Cuba and Florida. Although the molecular data reveal tw o distinct clades of D. porrectus sensu stricto I am unable to find any morphological differences that would warrant their r ecognition as distinct species. They are thus retained in D. porrectus sensu stricto, at least until that clade is studied in more detail. These thin rooted plants retain the name D. porrectus because the type specimen is thin rooted. The thick rooted entity/cl ade, therefore, represents an un described species. The distribution map ( Fig 3 18) depi cts the three major clades and indicate s that the Yucatn thick rooted population s grows i n sympatry with the thin rooted population in this region These data imply that D. porrectus sensu stricto and the new species ( D aff. p orrectus ) are not interbreeding which suggests that the biological species concept (Mayr, 2000) could also be applied justifying their recognition as distinct species. The disjunction between Yucatn and Jalisco populations ( thick rooted clade ) is not novel as similar distribution s are also seen between sister taxa of Enriquebeltrania in the Euphorbiaceae (De Nova et al., 2006) D endrophylax porrectus has been

PAGE 62

62 reported from Oaxaca and Veracruz (Table 1 1) and photographs of herbarium spe cimens and living plants from the Veracruz and Oaxaca popula tions showed features similar to those of the Yucat n Jalisco entity (thick rooted clade ) and these are considered here as additional populations of this clade There is still a gap between the Veracruz Oaxaca population s and the Jalisco populations, but the distribution al gaps could be an art ifact of the few available collections According to Brown (2002) the flowering period of D. porrectus in Florida is from August to November. In this study, members of the Ft Myers population flowered t wice in a year with a gap of two months. This twice a year flower ing pattern was not obser ved in the Sarasota populations. Plants in the green house and in their natural habitat had the same flowering pattern The flowering dates also match with the collected specimens deposited in various herbaria. I t seems that there i s a variation in the phenology of D. porrectus in Florida, and that flowering plant s have been collec t ed from August until January. Members of the thin and thick rooted populations occurring in the Yucatn flowered from September until January, which means that interbreeding could take place between the D. porrectus and the new species as the plants grow in sympatry. Since we do not have any evidence of hybrids, I suggest that there is an interbreeding barrier present. This barrier could be the result of t he difference in rostellar entry op ening and the length of callus or there could be genetic or chromosomal barriers to hybridization Nothing is known about chromosome numbers in these species.

PAGE 63

63 I was not successful in determining the pollinator but, according to the flower morphology (apron like rostellum) the pollinator was sugg ested of a lepidopteran (Dressler, 1993) Conclusion The results of this study clearly support recognition of two distinct species within Dendrophylax porrectus ; these two species differ most conspicuously in root thickness, but also in various floral characters and in molecular sequence divergence. Since the type of D. porrectus is from Cuba where only the thin rooted are known to occur the thick rooted clade needs to be described as a new species. Within D. porrectus sensu s tricto (the thin rooted clade), molecular d ata reveal two subclades, but at present I am unable to find morphological characters that correlate with these molecular clades Additional studies especially focused on the morphology of the entities within D. porrectus s ensu strict o are needed in order to resolve the ir taxonomic status.

PAGE 64

64 APPENDIX SPECIMENS USED IN TH IS STUDY WITH THEIR GENBANK ACCESSION NU MBER Table A 1 Specimens u sed in this study with their GenB ank accession number Taxon Voucher Locality ITS matK ycf1 C. micranthum Carlsward 315 (FLAS) Panama AY147220 AY147235 N/A C. micranthum Carlsward 180 (FLAS) Mexico AF506298 AF506347 EU490725 C. micranthum Ackerman 3341 (UPRRP) Puerto Rico AY147219 AF506346 N/A D. alcoa Ackerman 2773 (UPRRP) Dominican Republic AF506307 N/A N/A D. barrettiae Carlsward 199 (FLAS) Jamaica AF506308 AF506353 JN176133 D. barrettiae Whitten 1814 (FLAS) Jamaica N/A N/A JN176134 D. aff porrectus Salazar 8279 (MEXU) Mexico (Puerto Vallarta) JN176093 JN176103 JN176111 D. aff porrectus Carnevali 5907 (CICY) Mexico (Yucat n) AF506314 AF506357 JN176112 D. aff porrectus Carnevali 5907 a (CICY) Mexico ( Yucatn ) JN176094 JN176104 JN176113 D. aff porrectus Carnevali 5907 b (CICY) Mexico ( Yucatn ) JN176095 N/A N/A D. porrectus Ramirez 886 (CICY) Mexico ( Yucatn ) N/A JN176105 JN176114 D. fawcettii Whitten 3265 (FLAS) Cayman Islands AF506309 AF506354 JN176135 D. funalis Carlsward 302 (FLAS) Jamaica AY147221 AY147229 JN176131 D. funalis Whitten 1935 (FLAS) Jamaica AF506310 N/A JN176132 D. lindenii Ward 5365 (FLAS) USA (Florida, Collier) AF506319 N/A N/A D. lindenii Claude Hamilton s.n. Cuba AF506318 AF506362 JN176130 D. porrectus Ackerman 4514 (UPRRP) Cuba JN176098 JN176106 JN176121 D. porrectus Whitten 1950 (FLAS) Dominican Republic AY147224 AY147238 JN176119 D. porrectus no voucher Cayman Islands JN176097 AF506361 JN176118 D. porrectus Carlsward 184 (FLAS) Jamaica AF506315 AF506358 JN176117 D. porrectus Carnevali 4468 (CICY) Mexico (Campeche) JN176096 N/A JN176115 D. porrectus Carnevali 6312 (FLAS) Mexico (Quintana Roo) AF506316 AF506360 JN176116 D. porrectus Goldman 2271 (FLAS) USA (Florida, Glades ) JN176102 AF506359 JN176122 D. porrectus Carlsward 330 (FLAS) USA (Florida, Ft Myers) N/A JN176110 JN176124 D. porrectus Molgo 221 (FLAS) USA (Florida, Hillsborough) JN176099 JN176107 JN176127 D. porrectus Molgo 222 (FLAS) USA (Florida, Sarasota) JN176101 JN176109 JN176126 D. porrectus Carlsward 329 (FLAS) USA (Florida, Glades ) AY147223 AY147237 JN176123 D. porrectus Ackerman 3340 (UPRRP) USA (Puerto Rico) AF506313 AF506356 JN176120

PAGE 65

65 Table A 1 (cont.) Taxon Voucher Locality ITS matK ycf1 D. sallei Whitten 1945 (JBSD) Dominican Republic AY147225 AY147239 JN176128 D. varius Whitten 1960 (JBSD) Dominican Republic AY147222 AY147236 JN176129 D. varius Thompson 10683 (SEL ) Dominican Republic AF506312 N/A N/A D. varius Ackerman 2727 (UPRRP) Puerto Rico AF506311 N/A N/A Indicates sequences previously published by Carlsward (2004) and Neubig et al (2009) Note. N/A = not sequenced

PAGE 66

66 LIST OF REFERENCES Ackerman JD. 1995. An orchid flora of Puerto Rico and the Virgin Islands Bronx, New York: The New York Botanical Garden. Bent E, Taylor DL. 2010. Direct amplification of DNA from fresh and preserved ectomycorrhizal root tips. Journal of Microbiological Methods 80 : 206 208. Brown PM. 2002. Wild orchids of Florida with references to the Atlantic and Gulf Coas tal Plains, Gainesville, FL: University Press of Florida. Cansino N, Clason HC, Gmez de Garca P. 1996. Orqudeas de El Salvador, San Salvador, El Salvador: Asociacin Salvadorea de Orquideologa; Fomento Cultural, Banco Agrcola Comercial de El Salvador Carlsward B. 2004. Molecular systematics and anatomy of Vandeae (Orchidaceae): the evolution of monopodial leaflessness, PhD Dissertation, University of Florida, Gainesville, FL. Carlsward BS, Whitten WM, Williams NH. 2003. Molecular phylogenetics of neo tropical leafless Angraecinae (Orchidaceae): reevaluation of generic concepts. International Journal of Plant Sciences 164 : 43 51. Chase MW, Hills HH. 1991. Silica gel: An ideal material for field preservation of leaf samples for DNA studies. Taxon 40 : 215 220. Coyne JA, Orr HA. 2004. Speciation, Sunderland, MA: Sinauer Associates. De Nova JA, Sosa V, Wurdack KJ. 2006. Phylogenetic relationships and the description of a new species of Enriquebeltrania (Euphorbiaceae s.s.): an enigmatic genus endemic to Mexi co. Systematic Botany 31 : 533 546. de Queiroz K. 2007. Species concepts and species delimitation. Systematic Biology 56 : 879 886. de Queiroz K, Good DA. 1997. Phenetic clustering in biology: a critique. The Quarterly Review of Biology 72 : 3 30. Dickison WC. 2000. Integrative plant anatomy, San Diego, California: Academic Press. Dix MA, Dix MW. 2000. Orchids of Guatemala, a revised annotated checklist, St. Louis: Missouri Botanical Garden Press. Donoghue MJ. 1985. A critique of the biological species conce pt and recommendations for a phylogenetic alternative. The Bryologist 88 : 172 181. Doyle JJ, Doyle JL. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19 : 11 15.

PAGE 67

67 Dressler RL. 1993. Phylogeny and class ification of the orchid family, Oregon: Dioscorides Press. Dycus AM, Knudson L. 1957. The role of the velamen of the aerial roots of orchids. Botanical Gazette 119 : 78 87. Fawcett JH, Rendle. 1909. Journal of botany, British and foreign 47 : 266. Holmgren P K, Holmgren NH, Barnett LC. 1990. Index herbariorum. part II: the herbaria of the world., New York: New York Botanical Garden Press. Johnson LA, Soltis DE. 1994. matK DNA sequences and phylogenetic reconstruction in Saxifragaceae s. str. Systematic Botany 19 : 143 156. Johnson LA, Soltis DE. 1998. Assessing congruence: empirical examples from molecular data. In: Soltis DE, Soltis PS, Doyle JJ eds. Molecular systematics of plants II: DNA sequencing. Boston: Kluwer Academic Publishers, 297 348. Johnson RA, Wic hern DW. 1988. Applied multivariate statistical analysis, Englewood Cliffs, N.J.: Prentice Hall. Judd WS, Campbell CS, Kellogg EA, Stevens PF, Donoghue MJ. 2007. Plant systematics: a phylogenetic approach, Sunderland, Massachusetts: Sinauer Associates. Lan ave C, Preparata G, Sacone C, Serio G. 1984. A new method for calculating evolutionary substitution rates. Journal of Molecular Evolution 20 : 86 93. Mayr E. 2000. The biological species concept. In: Wheeler QD, Meier R eds. Species concepts and phylogeneti c theory: a debate. New York: Columbia University Press, 16 19. Molvray M, Kores PJ, Chase MW. 2000. Polyphyly of mycoheterotrophic orchids and functional influences of floral and molecular characters. In: Wilson KL, Morrison DA eds. Monocots: systematics and evolution. Collingswood, Victoria, Australia: CSIRO Publishing, 441 448. Neubig K, Whitten W, Carlsward B, Blanco M, Endara L, Williams N, Moore M. 2009. Phylogenetic utility of ycf1 in orchids: a plastid gene more variable than matK Plant Systematics and Evolution 277 : 75 84. Posada D. 2008. jModeltest: phylogenetic model averaging. Molecular Biology and Evolution : 1253 1256. Rambaut A. 1996. Se Al: sequence alignment editor v2.0a11. University of Oxford, Oxford, UK. Available at website, http://tree.bio.ed.ac.uk/software/seal/ last accessed 10 March 2011

PAGE 68

68 Rasband WS. 1997 2011. ImageJ v1.45d2. U. S. National Institutes of Health, Bethesda, Maryland, USA. Available at website, http://imagej.nih.gov/ij/ last accessed 5 March 2011 Reichenbach HG. 1865. Flora. Flora oder Botanische Zeitung :welche Recensionen, Abhandlungen, Aufstze, Neuigkeiten und Nachrichten, die Botanik betreffend, enthlt /herausgegeben von der Knigl. Botanischen Gesellschaft in Regensburg 48 : 279. Rolfe RA. 1903. Orchid Review. 11 : 247. Ruzin S. 1999. Plant microtechnique and microscopy, New York: Oxford University Press. Sun Y, Skinner DZ, Liang GH, Hulbert SH. 1994. Ph ylogenetic analysis of Sorghum and related taxa using internal transcribed spacers of nuclear ribosomal DNA. Theoretical and Applied Genetics 89 : 26 32. Swofford DL. 2003. PAUP*: Phylogenetic analysis using parsimony (*and other methods), Sunderland, MA: S inauer Associates. Warford NM. 1997. Harrisella cf porrecta : leafless and lethal near Puerto Vallarta. Orchid Digest 61 : 148. Whitten WM, Williams NH, Chase MW. 2000. Subtribal and generic relationships of Maxillarieae (Orchidaceae) with emphasis on Stanhopeinae: combined molecular evidence. American Journal of Botany 87 : 1842 1856.

PAGE 69

69 BIOGRAPHICAL SKETCH Iwan Eduard Molgo was born in Paramaribo, Suriname to Eduard Stu art Molgo and Jacqueline Molgo in 1974. He graduated from the Henri Dalhberg High School in 1993 and started his undergraduate career in 1994 at the Anton de Kom University of Suriname (AdeKUS) In 2002, he graduated from AdeKUS with a Bachelor o f Science degree in a griculture and l ater that year he started working at the Foundati on of Nature C onservation of Suriname as Research Coordinator and Wildlife Supervisor. In 2006, he was hired as Assistant R esearcher at the National Herbarium of Suriname and was responsible for all orchid related research and the orchid collections. After working for three years Iwan decided to begin his graduate career at the University of Florida under the supervision of Dr. Norris Williams where he studied the evolutionary relationship s of Dendrophylax porrectus Iwan graduated with a M S. in b otany in August 2011 and hopes to continue his graduate career at the University of Florida in order to support h is countr y after he has finished his studi es