This item is only available as the following downloads:
1 THE EVOLUTION AND SYSTEMATICS OF THE Opuntia h umifusa COMPLEX By LUCAS C. MAJURE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012
2 2012 Lucas C. Majure
3 To my amazing and ever supportive parents, Terrence and Diana Majure, my incredible wife Mariela Pajuelo and beautiful son Gabriel
4 ACKNOWLEDGMENTS I thank my advisors, Drs. Douglas E. and Pam S. Soltis, and Walter S. Judd for their utmost support, enthusiasm, critical guidance, and encouragement throughout my PhD program. I thank my committee member Marc Branham for his help and ideas with my project I also thank current and former members of the Soltis Lab (Monica Arakaki, Samuel Brockington, Charlotte Germain Aubrey, Maribeth Latvis, Nic olas Miles, Michael J. Moore, Stein Servick, Victor Suarez) the herbarium FLAS (Richard Abbott, Paul Co rogin, Lorena Endara, Mark Whitten, Kurt Neubig Kent Perkins, Norris Williams), and the Department of Biology for their support and help throughout my degree. I thank my collaborators, Raul Puente, M. Patrick Griffith, and Donald J. Pinkava for their expe rtise. I also thank those institutions and peo p le who provided me with specimens for use in this work and/or aided with fieldwork : Desert Botanical Garden (DBG), Eastern Kentucky University herbarium (EKY), Huntington Botanical Garden (HBG), Illinois Natur al History Survey (ILLS ), Louisiana State University herbarium (LSU), Miami University Herbarium (MU), Missouri Botanical Garden (MO), New York Botanical Garden (NY), Rancho Santa Ana Botanical Garden, Smithsonian Institution (US), Troy University herbariu m (TROY), University of Alabama (UNA), University of Miami herbarium (MU), University of Michigan herbarium (MICH), University of North Carolina (UNC), University of Tennessee herbarium (TENN), University of Wisconsin (WIS). Ron Altig, Frank Axelrod, Marc Baker, Bryan Connol l y, Tony Frates, Ty Harrison, Jovonn Hill, George Johnson, Terry Majure, Tom Mann, Ivan Marino, Michael J. Moore, Bill Nichols, Brent Patenge, Kevin Philley, George Phillips, Michael Powell, Kenneth Quinn, Chris Reid, Eric Ribbens, Barry Snow, Dean Spallone, Heather Sullivan, Blake Wellard, Theo Witsell, Dorde Woodruff. I thank Opuntia in the beginning and for his continued support and encouragement I thank the late Dr. Lyman Benson for all of
5 the work he carried out in Opuntia This work would not have been possible otherwise. Financial support for this work was provided in part by the Cactus and Succulent Society of America, the Botanical Society of America, American Society of Plant Taxonomists, the New England Botanical Club, the Florida Division of Forestry, and an NSF Dissertation Improvement Grant (DDIG DEB #1011270).
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 9 LIST OF FIGURES ................................ ................................ ................................ ....................... 10 ABSTRACT ................................ ................................ ................................ ................................ ... 12 CHAPTER 1 GENERAL INTRODUCTION ................................ ................................ .............................. 14 2 PHYLOGENY OF Opuntia S.S. (CACTACEAE): CLADE DELINEATIONS, GEOGRAPHIC ORIGINS, AND RETICULATE EVOLUTION ................................ ......... 18 Background ................................ ................................ ................................ ............................. 18 Material and Methods ................................ ................................ ................................ ............. 23 Taxon Sampling ................................ ................................ ................................ ............... 23 DNA Extraction, PCR, Sequencing, Sequence Editing, and Alignment ......................... 23 Phylogenetic Analyses ................................ ................................ ................................ ..... 25 Biogeographic Analysis and Divergence Time Estimation ................................ ............. 27 Results ................................ ................................ ................................ ................................ ..... 29 Relationships in Opu ntieae ................................ ................................ .............................. 30 Opuntia s.s. ................................ ................................ ................................ ...................... 31 Interclade Allopolyploids and Hybrids ................................ ................................ ........... 31 Intraclade Allopolyploids ................................ ................................ ................................ 34 Biogeography and Divergence Time Estimation of Opuntia s.s. ................................ .... 35 Discussion ................................ ................................ ................................ ............................... 36 Consolea ................................ ................................ ................................ .......................... 36 Opuntia lilae and Opuntia schickendantzii ................................ ................................ ..... 37 Nopalea ................................ ................................ ................................ ............................ 38 South North American Disjunction in Opuntia ................................ .............................. 38 The North American Radiation ................................ ................................ ....................... 41 Reticulate Evolution in Opuntia ................................ ................................ ...................... 41 Summary ................................ ................................ ................................ ................................ 44 3 Opuntia lilae ANOTHER Tacinga HIDDEN IN Opuntia S.L. (CACTACEAE) ................. 55 Background ................................ ................................ ................................ ............................. 55 Materials and Methods ................................ ................................ ................................ ........... 57 Taxon Sampling and Phylogenetic Analysis ................................ ................................ ... 57 Ancestral State Reconstruction ................................ ................................ ........................ 58 Results ................................ ................................ ................................ ................................ ..... 58 Phylogenetic and Morphological Analysis ................................ ................................ ...... 58
7 Ancestral State Reconstruction ................................ ................................ ........................ 59 Discussion ................................ ................................ ................................ ............................... 61 4 A CASE OF MISTAKEN IDENTITY, Opuntia abjecta LONG LOST IN SYNONYMY UNDER THE CARIBBEAN SPECIES, O.triacantha AND REASSESSMENT OF THE ENIGMATIC O. cubensis ................................ ........................ 67 Background ................................ ................................ ................................ ............................. 67 Materials and Methods ................................ ................................ ................................ ........... 69 Result s ................................ ................................ ................................ ................................ ..... 70 Phylogeny ................................ ................................ ................................ ........................ 70 Morphology O. abjecta vs. O. triacantha ................................ ................................ .. 71 Morphology O. militaris vs. O. triacantha ................................ ................................ 72 Morphology O. cubensis vs. O. ochrocentra ................................ ............................. 73 Key to the Species ................................ ................................ ................................ ........... 74 Discussion ................................ ................................ ................................ ............................... 75 Opuntia abjecta vs O. triacantha ................................ ................................ ................... 75 O. militaris vs. O. triacantha ................................ ................................ ........................... 77 The Opuntia cubensis Enigma ................................ ................................ ......................... 78 Summary ................................ ................................ ................................ ................................ 79 5 CYTOGEOGRAPHY OF THE Humifusa CLADE OF Opuntia S.S. MILL. 1754 (CACTACEAE, OPUNTIOIDEAE, OPUNTIEAE): CORRELATIONS WITH PLEISTOCENE REFUGIA AND MORPHOLOGICAL TRAITS IN A POLYPLOID COMPLEX ................................ ................................ ................................ ............................. 83 Background ................................ ................................ ................................ ............................. 83 Material and Methods ................................ ................................ ................................ ............. 88 Chromosome Counts ................................ ................................ ................................ ....... 88 Taxonomy ................................ ................................ ................................ ........................ 88 Cytogeographic Analysis ................................ ................................ ................................ 89 Phyl ogenetic Analysis ................................ ................................ ................................ ..... 89 Results ................................ ................................ ................................ ................................ ..... 89 Discussion ................................ ................................ ................................ ............................... 91 Diploid Refugia and Polyploidy Formation ................................ ................................ .... 92 Agamospermy ................................ ................................ ................................ .................. 96 Autopolylploidy vs. Allopolyploidy ................................ ................................ ................ 97 Morphological Correlations with Polyploids ................................ ................................ .. 98 Summary ................................ ................................ ................................ ................................ 99 6 PHYLOGENY OF THE H umifusa CLADE ( Opuntia S.S.): WHAT DIPLOIDS CAN TELL US ABOUT THE EVOLUTIONARY HISTORY OF THE GROUP ....................... 109 Background ................................ ................................ ................................ ........................... 109 Material and Methods ................................ ................................ ................................ ........... 112 Taxon and Marker Sampling ................................ ................................ ......................... 112 Phylogenetic Analysis ................................ ................................ ................................ ... 114
8 Results ................................ ................................ ................................ ................................ ... 115 P hylogenetic Analysis (Diploid Taxa) ................................ ................................ .......... 115 P hylogenetic Analyses (Polyploid Taxa) ................................ ................................ ...... 115 Discussion ................................ ................................ ................................ ............................. 117 Opuntia abjecta and O. pusilla ................................ ................................ ...................... 118 O puntia humifusa s.l. ................................ ................................ ................................ ..... 118 Opuntia macrorhiza s.l. ................................ ................................ ................................ 119 Opuntia pottsii ................................ ................................ ................................ ............... 120 Opuntia cymo chila ................................ ................................ ................................ ......... 120 Morphological Characters of the SE and SW Clades ................................ .................... 121 Summary ................................ ................................ ................................ ............................... 122 7 TAXONOMIC REVISION OF THE Opuntia humifusa COMPLEX ( Opuntia : CACTACEAE) OF THE EASTERN UNITED STATES ................................ .................... 130 Background ................................ ................................ ................................ ........................... 130 Hybridization, Polyploidy, and Morphological Variability ................................ .......... 132 Taxonomic History of the O. humifusa Complex ................................ ......................... 134 Species Concept ................................ ................................ ................................ ............. 137 Description of the Opuntia humifusa Complex ................................ ................................ .... 139 Key to the Members of the Humifusa Complex ................................ ................................ ... 140 Key to Subspecies of O. humifusa ................................ ................................ ................. 175 8 GENERAL CONCLUSIONS ................................ ................................ ............................... 212 APPENDIX A VOUCHERS USED WITH GENBANK NUMBERS ................................ ......................... 216 B ACCESSIONS USED WITH GENBANK NUMBERS ................................ ...................... 227 C SPECIMENS EXAMINED for chapter 4 ................................ ................................ ............. 229 D ACCESSIONS USED FOR CHROMOSOME COUNTS ................................ ................... 231 LIST OF REFERENCES ................................ ................................ ................................ ............. 237 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 255
9 LIST OF TABLES Table page 2 1 DNA regions and associ ated primers used in this study ................................ ................... 46 2 2 Statistics of regions sequenced in this study based on the diploid data sets ...................... 46 2 3 Interclade derived taxa recovered in our analyses ................................ ............................. 47 5 1 Synonyms of O. humifusa s.l. and O. macrorhiza s.l. sampled during this study ........... 101 5 2 Selected taxa of O. humifusa s.l. and O. macrorhiza s.l. with morphological char acters and corresponding ploidy ................................ ................................ ............... 102 5 3 Taxa used in phylogenetic analyses of ITS sequence data given with their GenBank accession numbers ................................ ................................ ................................ ........... 103 6 1 Synonyms of O. humifusa s.l. and O. macrorhiza s.l. used in our analyses .................... 124 6 2 Polyploid taxa of the Humifusa clade sampled in our analys es of nuclear and plastid data ................................ ................................ ................................ ................................ ... 125
10 LIST OF FIGURES Figure page 2 1 Diploid phylogeny of Opuntia s.s ................................ ................................ ...................... 48 2 2 Diploid phylogeny of Opuntia s.s. (adapted from Fig. 2 1) with interclade reticulate taxa mapped on their put ative diploid progenitor clades ................................ ................... 49 2 3 Intraclade phylogeny of Opuntia s.s. (total evidence phylogeny exc luding interclade derived taxa) ................................ ................................ ................................ ....................... 50 2 4 Ancestral area reconstruction and putative dispersal pathways of Opuntia clades ........... 51 2 5 Plastid (left) and ITS (right) phylogeny including all diploid and polyploid species of Opuntia as well as the genus Consolea ................................ ................................ ............. 52 2 6 Diploid phylogeny including the genus Consolea. Consolea is resolved as sister to the Tacinga, Brasiliopuntia, Opuntia s.s. clade ................................ ................................ 53 2 7 Cronogram from r8s analysis showing an early Pliocene origin of the North American clade of Opuntia ................................ ................................ ................................ 54 3 1 Phylogram of Tacinga and other members of Opuntieae from a combined analysis of nuclear and plastid loci ................................ ................................ ................................ ...... 64 3 2 ML character state reconstructions ................................ ................................ .................... 65 3 3 Morphological characters of Tacinga lilae from the type collection ................................ 66 4 1 Putative diploid ML phylogeny (most likely topology) of Opuntia s.s. using South American species ( O. macbridei, O. retrorsa ) as outgroups ................................ ............. 80 4 2 Most likely topologies (from RAxML) from ITS (A), ppc (B), and (C) plastid phylogenies ................................ ................................ ................................ ........................ 81 4 3 Morphological characters of O. abjecta, O. triacantha, O. cubensis, O. ochrocentra O. dillenii and O. repens ................................ ................................ ................................ ... 82 5 1 Selected taxa in the Humifusa clade with associated chromosome squashes .................. 104 5 2 Cytogeography of O. humifusa s.l., O. macrorhiza s.l., O. pottsii, and O. tortispina ..... 105 5 3 Cytogeography of O. pusilla Diploids are represented by black circles, triploids by gray circles, a nd tetraploids by white circles ................................ ................................ ... 106 5 4 Majority rule consensus topology from 10000 ML bootstrap pseudoreplicates using RAxML based on the nrITS region ................................ ................................ ................ 107
11 5 5 Hypothetical origin and subsequent dispersal of poly ploid taxa from diploid refugia .... 108 6 1 Diploid phylogeny of the Humifusa clade using co mbined plastid and nuclear data ...... 126 6 2 Plastid phy logeny including polyploid taxa ................................ ................................ ..... 127 6 3 ITS phylogeny including polyploids ................................ ................................ ................ 128 6 4 The isi1 phylogeny including polyploid taxa ................................ ................................ ... 129 7 1 Phylogeny of the O. humifusa complex ................................ ................................ ........... 196 7 2 Morphological features of O. abjecta ................................ ................................ .............. 197 7 3 Geographic distribution of O. abjecta ................................ ................................ ............. 198 7 4 Morphological features of O. ochrocentra ................................ ................................ ...... 19 9 7 5 Morphological features of O. austrina ................................ ................................ ............. 200 7 6 Distribution of O. austrina ................................ ................................ .............................. 201 7 7 Morphological features of O. cespitosa ................................ ................................ ........... 202 7 8 Distribution of O. cespitosa Note: Essex Coun ty, Ontario is represented by .............. 203 7 9 Mophological features of O. drummondii ................................ ................................ ........ 204 7 10 Distribution of O. drummondii ................................ ................................ ........................ 205 7 11 Distribution of O. humifusa ................................ ................................ ............................. 206 7 12 Morphological features of O. humifusa subsp. humifusa ................................ ................ 207 7 13 Morphological features of O. humifusa subsp. pollardii ................................ ................. 208 7 14 Morphological features of O. humifusa subsp. lata ................................ ......................... 209 7 15 Morphological features of O. nemoralis ................................ ................................ .......... 210 7 16 Distribution of O. nemoralis ................................ ................................ ............................ 211
12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE EVOLUTION AND SYSTEMATICS OF THE O puntia humifusa COMPLEX By Lucas C. Majure August 2012 Chair: Douglas E. Soltis Major: Botany Relationships among major clades of Opuntieae and the circumscription of Opuntia s.s. were unresolved prior to this study. Opuntia s.s., as currently circumscribed, comprises 120 200 species, occurring natively throughout the Americas. The Opuntia humifusa species complex (OHC) is taxonomically misunderstood due to high morphological variability, frequent hybridization, and polyploidy. There is no comprehensive phylogeny of either the genus or the O humifusa complex. The goal of this study was to reconstruct the phylogeny of Opuntieae to elucidate major clade relationships, divergenc e times, and the biogeographic history of Opuntia s.s. Evolutionary relationships and ploidal levels of the Humifusa clade [HC (including the OHC)] were assessed to provide the foundation for a taxonomic revision of the OHC Based on sequence data, Opuntia s.s. forms a well supported clade, including the genus Nopalea and is sister to a clade containing Tacinga and Brasiliopuntia Opuntia s.s. originated in the late Miocene in southern South America and then dispersed to North American deserts. Numerous ta xa originating through reticulate evolutionary processes were discovered in Opuntia s.s. The HC originated in northeastern Mexico/southwestern United States in the late Pliocene or early Pleistocene. Opuntia lilae was resolved in Tacinga and transferred to that genus. Although placed in synonymy with O triacantha, O abjecta is not closely related, and was recognized as
13 a separate species. Opuntia cubensis and O ochrocentra were found to be of hybrid origin derived from different parental taxa, and were t hus considered distinct from one another. Chromosome counts of the HC revealed that 66% of 277 accessions were polyploid and displayed a much larger distribution (from the southern United States to Canada) than diploid members [restricted to the southweste rn (SW) and southeastern (SE) United States in presumed Pleistocene refugia]. The SW and SE diploids each formed strongly supported clades; however, many polyploids formed as a result of the union of members of the SE and SW clades. Phylogenetic, cytologic al, and morphological analyses showed that the most widespread member of the OHC, O humifusa s.l., was polyphyletic and is now recognized as several distinct taxa. The taxonomic revision presented here includes seven species of the OHC
14 CHAPTER 1 GENERAL INTRODUCTION The Cactaceae are a well supported clade endemic to the New W orld and consist of between 1438 (Hunt et al. 2006) and 1850 species (Nyeffler and Eggli 2010) of mostly stem succulents that produce characteristic short shoots, embedded within the long shoot (i.e., areoles), modified leaves in the forms of spines, and o varies deeply embedded in stem tissue or perica rpels (Mauseth 2006). Two early diverging clades, Pereskia and Rhodocactus retain ancestral features (e.g., large photosynthetic leaves, lack of succulent stems, cymose inflorescences, basal placentation, sup erior ovaries) of the family (Edwards et al. 2005). Of primary importance, in terms of species diversity, are the two major subfamilies, Opuntioideae (349 species) and Cactoideae (1498 species) ( Nyffeler and Eggli 2010 a ), which represent the most iconic gr owth forms within Cactaceae, exhibiting succulent photosynthetic stems, with a drastic reduction or even loss of long shoot leaves (although, there are exceptions to this). Subfamily Opuntioideae is unique in producing small, hair like retrorsely barbed sp ines (i.e., glochids; Mauseth 2006) and seeds with a hard funicular girdle and funicular envelope covering the seed (Stuppy 2002). Tribe Opuntieae of Opuntioideae consists mostly of species with flattened stems and sympodial growth ( Opuntia s.s., Tunilla ), although, Brasiliopuntia, Consolea and Tacinga demonstrate indeterminate growth to some extent (Anderson 2001, Taylor et al. 2002). The Opuntieae consists of seven genera, Brasiliopuntia, Consolea, Miqueliopuntia, Opuntia s.s., Salmiopuntia Tacinga and Tunilla (Majure et al. 2012 a ). Although Nyffeler and Eggli (2010) concluded that Consolea should be considered a synonym of Opuntia s.s., Majure et al. (2012 a ) demonstra ted that the genus forms a well supported clade and is evolutionarily divergent from Opuntia s.s. Evolutionary relationships among the genera of
15 Opuntieae are still unresolved and the actual circumscription of the culturally, economically, and medicinally important genus, Opuntia s.s., is undetermined. Opuntia s.s. is one of the largest g enera in Cactaceae, with around 180 200 species (Anderson 2001; Nyffeler and Eggli 2010 a ), and exhibits the widest distribution of any genus in Cactaceae, as it occurs from Canada to Argentina (Anderson 2001) in habitats ranging from tropical to subtropica l dry forests, moderate deserts, and even temperate forests (Benson 1982). Opuntia also has been introduced throughout the world for use as a foodstuff for humans and animals and as ornamentals (Anderson 2001 ; Inglese et al. 2002 ; Nefzaoui and Salem 2002). Opuntia is renowned for hybridization and polyploidy (Benson 1982 ; Pinkava 2002 ; Majure et al. 2012 a,b; Majure et al. in review) and also for its morphological variability, wherein certain morphological characters expressed within an individual are close ly linked to environmental factors (Benson 1982 ; Rebman and Pinkava 2001 ; Majure 2007). Species of Opuntia also are notoriously difficult to work with from herbarium specimens, as methods used to collect specimens are typically inefficient, leading to poor specimen preservation, and the complete loss of most taxonomically useful characters as a result of the succulent nature of the plants ( Reyes Agero et al. 2007). Opuntia species also are poorly collected, as a consequence of the difficulties in specimen preparation (Rebman and Pinkava 2001), and their highly bothersome glochids and spines ( Reyes Agero et al. 2007). One poorly understood group within Opuntia is the O. humifusa complex of the eastern United States. This group is distributed over a wide ra nge from Ontario, Canada, south to the Florida Keys, and west to Wisconsin, Iowa, Missouri, Arkansas, and Texas (Benson 1982 ; Pinkava 2003 ; Majure et al. 2012 b ). Species within the O. humifusa complex are known to hybridize (Benson 1982), contain numerous polyploid entities (Majure et al. 2012 b ), and are
16 poorly represented in herbaria (Majure and Ervin 2008), which has provided for a taxonomically complex history and nebulous species limits. T hus, the O. humifusa complex presents an opportunity to explore species boundaries, consequences of polyploidiz ation and hybridization, and evolutionary history. The primary goals of this study were: to determine the circumscription, date of origin, and biogeographic history of Opuntia s.s., as well as the limits of the Humifusa clade (including the O. humifusa complex); to clarify the phylogenetic placement of O. lilae and the morphological synapomorp hies of the genus Tacinga ; to clarify the phylogenetic placement and taxonomic status of O. abjecta and O. triacantha and determine the origin of putative hybrids, O. cubensis and O. ochrocentra ; to carry out chromosome counts for members of the O. humif usa complex and relate ploidy to historical biogeography and the formation of polyploid taxa; to reconstruct the phylogeny of the Humifusa clade, with an emphasis on determining the origins of the many polyploid taxa in the group; to produce a taxonomic re vision of the O. humifusa species complex. These goals are discussed in the following seven chapters. In C hapter 2, I reconstruct the phylogeny of Opuntieae, determine relationships among the genera of Opuntieae, provide a circumscription of Opuntia s.s. and the Humifusa clade, and use the phylogeny to test the origin of the many polyploids in the genus, as we ll as the biogeographic history and divergence dates of Opuntia s.s. In C hapter 3, I use previously gathered data to build a phylogeny of members of the Opuntieae and determine the e volutionary placement of O. lilae I then use the phylogeny to determine which morphological characters may be synapomorphic for the genus Tacinga and formally transfer O. lilae to Tacinga
17 In C hapter 4, I reconstruct the phylogeny of several clades of Opuntia s.s. This work show s that O. triacantha is not monophyletic, as currently circumscribed, and consists of several taxa: O. abjecta, O. militaris and O. triacantha Opuntia cubensis and O. ochrocentra are shown to be of hybrid origin, but they are derived from different parental species and should not be considered synonymous. In C hapter 5, I report chromosome numbers for 277 accessions of members of the Humifusa clade and determine that the origin of many polyploids in the group is most likely the result of hybridization between the two diploid clades of the Humifusa clade at the end of the Pleistocene. In C hapter 6, I reconstruct the phylogeny of the Humifusa clade and use the diploid topology to di scover hybrid, polyploid derivatives from the union of diploid members of the two subclades, SE and SW, of the Humifusa clade. The widespread O. humifusa s.l. also is determined to be polyphyletic and should be recognized as several taxa. In C hapter 7, I present a taxonomic revision of the O. humifusa complex of eastern North America, in which I recognize seven species (i.e., O. abjecta, O. austrina, O. cespitosa, O. drummondii, O. humifusa, O. nemoralis and O. ochrocentra ) and three infraspecific taxa wi thin O. humifusa (var. humifusa var. lata and var. pollardii ). In C hapter 8, I provide general conclusions about the phylogenetic structure of Opuntieae, Opuntia s.s., the Humifusa clade, and the O. humifusa complex, as well as information discovered (th rough this study) about reticulate evolution and polyploidy in these groups.
18 CHAPTER 2 PHYLOGENY OF Opuntia S.S. (CACTACEAE): CL ADE DELINEATIONS, GE OGRAPHIC ORIGINS, AND RETICUL ATE EVOLUTION Background Cactaceae, comprising a well supported clade (Hershkovitz and Zimmer 1997; Applequist and Wallace 2001; Nyffel er, 2002 2007 ; Edwards et al. 2005 ) apparently sister to Anacampserotaceae (Nyffel er and Eggli 2010b ) are endemic to the New World except for the occurrence of one species, Rhipsalis bacci fera (Mill.) Stearn in the Old World tropics (Benson 1982). Other Cactaceae have been introduced, however, to locations around the world (Britton and Rose 1920; Anderson 2001). Although no reliable fossils have yet been found, the clade is suggested to rep resent a young radiation that evolved as a result of aridification in the Americas at the end of the Eocene through the beginning of the Miocene, ca. 30 million years ago (Ma) (Hershkovitz and Zimmer 1997). This date has been corroborated by the phylogenom ic analyses of Arakaki et al. (2011) who estimated an age of ca. 35 Ma for the origin of Cactaceae. Arakaki et al. (2011) also suggested that many of the major radiations within Cactaceae were initiated at the end of the Miocene (ca. 10 5 Ma), concomitan t with increased atmospheric CO 2 and aridity in the Americas. Cactaceae comprise ca. 1500 1800 species (Anderson 2001), which have been divided variously into 3 6 subfamilies (Crozier 2004). Pereskioideae were generally considered to be sister to the r est of the family, but Edwards et al. (2005), Brcenas et al. (2011) and Hernndez Hernndez et al. (2011) have shown that this subfamily is paraphyletic, forming two separate Reprinted with permission from the Botanical Society of America. Original publication: Majure, L.C., R. Puente, M.P. Griffith, W.S. Judd, P.S. Soltis, and D.S. Soltis. 2012. Phylogeny of Opuntia s.s. (Cactaceae): reticulate evolution, geographic origins, and clade delineation. American Journal of Botany 99: 847 864.
19 clades that are the successive sisters to the rest of the family (Edwards et al 2005 ) Currently, very reduced leaves and primarily rely on stem photosynthesis: sensu Mauseth 2006), Cactoideae and Opuntioideae (Edwards et al. 2005). Opu ntioideae encompass Opuntia Mill. s.l. and four associated genera ( Cumulopuntia F. Ritter s.l., Maihueniopsis Speg. s.l., Pterocactus K. Schum., Puna R. Kiesling s.l.; [Griffith and Porter 2009 ]), although, Opuntia s.l. (e.g., Benson 1982) was shown throu gh molecular phylogenetic studies to be polyphyletic (Wallace and Dickie 2002; Griffith and Porter 2009). Thus, Opuntia (hereafter Opuntia s.s.) has been reduced drastically in size with many segregate genera [e.g., Austrocylindropuntia Backeb., Brasiliopu ntia (K. Schum.) A. Berger, Cylindropuntia (Engelm.) F. M. Knuth] now recognized (Anderson 2001;Wallace and Dickie 2002; Hunt 2006; Griffith and Porter 2009 ) Currently, five tribes (Wallace and Dickie 2002 ) and 15 (Anderson 2001 ), 16 (Stuppy 2002 ) or 18 ( Hunt 2006 ) genera are recognized within Opuntioideae. Tribe Opuntieae (platyopuntioids) is a well supported clade within Opuntioideae (Wallace and Dickie 2002 ; Griffith and Porter 2009 ; Hernndez Hernndez et al. 2011 ) that consists of Brasiliopuntia (K Schumann) A. Berg., Consolea Lemaire, Miqueliopuntia Nopalea Salm Dyck, Opuntia s.s., Salmiopuntia ( Guiggi 2010 ) Tacinga Britton & Rose, and Tunilla Hunt and Illiff. The platyopuntioids were so named by Britton and Rose (1920) for the flat, photosynthetic stem segments (i.e., cladodes) characteristic of most members, although they did not include Miqueliopuntia Tacinga Tunilla Nopalea or Salmiopunt ia in the group. Species of Maihueniopsis s.l. were also recovered in Opuntieae (Griffith and Porter 2009), but this genus is often placed in tribe Cumulopuntieae (Hunt 2002 )
20 DNA studies have provided conflicting results regarding the placement of Consol ea (outside of Opuntia s.s. or nested within Opuntia s.s.), but the morphologically distinct genus Nopalea has consistently been nested within Opuntia However, due to low resolution and/or insuffi cient taxon sampling, the circumscription of Opuntia s.s. remains unclear (Wallace and Dickie 2002; Griffith and Porter 2009; Brcenas et al. 2011; Hernndez Hernndez et al. 2011). Opuntia s.s. (nopales, prickly pears; excluding Consolea ) is the largest genus in Opuntioideae and the most widespread genus in Cact aceae, distributed natively from Canada to Argentina (Anderson 2001). There are 150 (Stuppy 2002) to 180 recognized species (including Nopalea ; Anderson 2001; Hunt 2006 ) within the genus, which is suggested to have originated as recently as 5.6 ( 1.9) mya (Arakaki et al. 2011). Members of Opuntia s.s. are cultivated worldwide as fruit and vegetable crops (Inglese et al. 2002) and are increasingly used as forage and fodder for livestock in arid areas of the world, such as parts of Brazil, Mexico, western As ia, and northern and southern Africa (Nefzaoui and Salem 2002 ) Medicinally, Opuntia polysaccharides have been shown to protect brain tissue from glucose and oxygen deprivation (Huang et al. 2008). Opuntia ficus indica (L.) Mill. has been used to protect t he liver from harmful organophosphorous pesticides (Ncibi et al. 2008), and various Opuntia species have shown hypoglycemic effects in diabetic patients, returning blood glucose to normal levels ( Trejo Gonzlez et al. 1996; Laurenz et al. 2003 ) Opuntia st reptacantha Lem. has even been used as a bioaccumulator in lead contaminated waters ( Miretzky et al. 2008 ) Species of Opuntia are also known as some of the most highly invasive species in arid areas of their nonnative range such as Australia (Freeman 1992 ) the Mediterranean region (Vil et al. 2003 ) and Africa. Millions of hectares invaded by Opuntia stricta (Haw.) Haw. (Dodd
21 1940) were eventually brought under control in Australia using a well known biological control agent, Cactoblastis cactorum Be rg ( Zimmermann et al. 2000 ) This moth is now wreaking havoc in the native range of prickly pears in North America (Simonsen et al. 2008). The nutritive tissues and high production rates of O. stricta, introduced into Kruger National Park (South Africa), m ake it irresistible to the native fauna, primarily baboons and elephants; thus, this species is easily dispersed, increasing its invasion in the park (Reinhardt and Rossouw 2000 ; Foxcroft et al. 2004 ; Foxcroft and Rejmanek 2007 ). In its native range, Opunt ia s.s. provides food for numerous herbivores, including tortoises, iguanas, birds, rabbits, deer, bats, sloths, squirrels, coyotes, bears, pigs, and bison (Mellink and Riojas L pez 2002 ) ; this also clearly underscores the ecological importance of prickly pear. Opuntia also is culturally important. In Mexico, where species of Opuntia have been cultivated fo r at least the last 14 000 yr ( Casas and Barbera 2002 ) they represent an iconic national fi g ure, illustrated on the country flag. The large, tree like Opuntia species, O. megasperma O. echios and O. galapaegia, are some of the most conspicuous species of the Gal pagos Islands. Even Charles Darwin could not resist the intrigue of Opuntia when he collected the fi rst specimen of O. galapaegia (later described by Henslow 1837 ) Polyploidy is a common phenomenon throughout tribe Opuntieae, which has be en well studied cytologically ( Pinkava 2002 ; Majure et al. 2012 b ; L. C. Majure et al. unpublished manuscript). In fact, diploids (2 n = 2 x = 22) are rel atively rare in the tribe making up only 26.2% of the 164 species with reported chromosome counts (L. C. Majure et al. unpublished manuscript). Polyploid taxa within Opuntia range from triploid (2 n = 3 x = 33) to octoploid (2 n = 8 x = 88), and many specie s have multiple ploidal levels ( Pinkava 2002 ; Majure et al. 2012 b ; L. C. Majure et al. unpublished manuscript ). Species limits are still poorly understood, as a result
22 of the high frequency of polyploid taxa, morphological variability, poor representation in herbaria, and freque nt interspecific hybridization in Opuntia s.s. ( Cota and Philbrick 1994 ; Rebman and Pinkava 2001 ; Pinkava 2002 ; Majure et al. 2012 b ) Furthermore, there is no comprehensive phylogeny of Opuntia s.s., so limits of major clades are largely unknown. Numerous morphological and cytological studies have been conducted on large groups of taxa and species complexes (e.g., Doyle 1990 ; Parfi tt 1991 ; Leuenberger 2001 ; Majure et al. 2012 b ) but Opuntia s.s. has n ot been studied comprehensively using molecular data. Griffi th and Porter (2009) included 28 species of Opuntia s.s. in their molecular phylogeny of Opuntioideae but were unable to resolve relationships within Opuntia s.s. using ITS and the plastid interge nic spacer trnL F Hernndez Hernndez et al. (2011) and B rcenas et al. (2011) recovered South American Opuntia s.s. species and South American species of Opuntia plus Tunilla erectoclada (Backeb.) Hunt & Illiff, respectively, as sister to the rest of Op untia s.s. However, Hernndez Hernndez et al. (2011) only surveyed seven species of Opuntia and B rcenas et al. (2011) had no resolution among clades. In addition, although a number of Opuntia s.l. species hav e been shown to be interspecifi c hybrids using molecular data ( Mayer et al. 2000; Griffi th 2003 ) the prevalence of reticulation in this group has not been extensively surveyed. We broadly sampled species in tribe Opuntieae using nuclear and plastid sequence data and produced a phylogeny of the clade to (1) determine the circumscription of Opuntia s.s. and the major clades within it, (2) resolve the placement of the problematic genera Consolea and Nopalea (3) investigate the geographic origin and subsequent spread of Opuntia s.s., and (4) survey for potential reticulate evolution.
23 Material and Methods Taxon S ampling We sampled 112 taxa (98 species) of Opuntia nine species of Nopalea six species of Consolea four species of Tacinga, and Brasiliopuntia brasiliensis (Willd.) Berg. Our sampling includes members from all 29 series of subgenus Platyopuntia recognized by Britton and Rose (1920) and thus represents a broad sampling of the most likely members of Opuntia s.s. Other members of Opuntieae, Maihueniopsis cf. ova ta (Pfeiffer) F. Ritter, Miqueliopuntia miquelii (Monville) F. Ritter, Salmiopuntia salmiana (J. Parmentier ex Pfeiffer) Guiggi, and Tunilla corrugata (Salm Dyck) Hunt and Illiff were used as outgroups based on Gri ffith and Porter (2009) and Hernndez Her n ndez et al. (2011 ). GenBank accession numbers and voucher data are given in Appendix A DNA Extraction PCR, Sequencing, Sequence Editing and Alignment Total genomic DNA was extracted using a modifi ed CTAB method ( Doyle and Doyle, 1987 ) Although cacti have highly mucilaginous tissues, we successfully extracted high quality DNA from live plants, silica dried material, or herbarium specimens using this method. When possible, we used the small, ephemeral leaves, which are produced as new cladodes develop. This produced the highest quality and cleanest DNA of any samples used. Otherwise we used epidermal tissue with the cuticle removed (cf. Gri ffith and Porter, 2003 ) We sampled four plastid intergenic spacers ( atpB rbcL ndhF rpl32 psbJpetA and trnL F following Mavrodiev et al.  M. J. Moore, Oberlin College [unpublished data], Shaw et al.  and Taberlet et al.  respectively), the plastid gene matK ( http://www.kew.org/barcoding/upda te .html), ca. 900 bp from the 5 ycf1 (K. Neubig, Florida Museum of Natural History, unpublished data), the nuclear gene ppc ( Her nn dez Her nn dez et al. 2011 ) and the nuclear ribosomal internal transcrib e d spacers (ITS; following W h ite et al. 1990 ) We designed new
24 primers for atpB rbcL, ndhF rpl32 psbJ petA spacer, ycf1 and ppc after the initial sequencing of those PCR products ( Table 2 1 ) A sequence of matK for Tacinga funalis Britton & Rose was downloaded from GenBank (Appendix A ). Mixtures for 25 amplifi cation reactions were as follows: 0.5 nd Taq polymerase (produced in the Soltis lab from E. coli producing the Taq gene). PCR cycling conditions for the plastid intergenic spacers and matK followed Shaw et al. (2007) although the initial annealing temperature was modifi ed to 55 C and the number of cycles was increased to 35. PCR cycling conditions for ITS were an initial denaturation at 95 C for 2 min; followed by 5 cycles of 95 C for 1 min, 53 C for 1 min, and 72 C for 2 min; followed by 40 cycles of 95 C for 1 min, 48 C for 1 min, and 72 C for 2 min; with a fi nal extension step at 72 C for 12 min. PCR cycling conditions for ppc were 95 C for 5 min; followed by 44 cycles of 94 C for 1 min, 55 C for 1 min increasing 0.3 C /cycle, and 72 C for 2.5 min; with a fi nal extensi on of 72 C for 10 min. PCR cycling conditions for ycf1 followed N eubig et al. (2008) with modifi cation of the initial annealing temperature from 60 C to 63 C Plastid ycf1 and nuclear ppc were only sequenced for diploid Opuntia taxa. All PCR products were initially sequenced directly, except for presumed hybrids and polyploid taxa surveyed from each clade (discussed later). We searched for nucleotide polymorphisms in sequence chromatograms of ITS, especially in polyploid Opuntia and cloned those produ cts using the TOPO TA (Invitrogen, Carlsbad, California, USA) or Stratagene cloning kit (Stratagene, La Jolla, California). We also cloned at least one polyploid member from each any taxa thought to be of hybrid origin. Eight clones per accession were directly sequenced at the Interdisciplinary
25 Center for Biotechnology Research at the University of Florida using bacterial primers (T3 T7) from the kits. A subset of polyploid taxa was cloned and sequenced for ppc to ascertain the degree of nucleotide polymorphism among taxa. However, the use of ppc for analysis of polyploids was discontinued, as sequence divergence in this gene was less than that of ITS. Sequences were edited eithe r in the program Sequencher 4.2.2 (Gene Codes, Ann Arbor, Michigan, USA) or Geneious Pro 5.1 (Biomatters Ltd., Auckland, New Zealand) and automatically aligned using the program Muscle (Edgar 2004 ) ; this alignment was then adjusted manually in the program Se Al v2.0 (Rambaut 2007 ) All gaps introduced during alignment were coded as missing data. Phylogenetic Analyses Opuntia has been well studied cytologically (see Pinkava 2002 ) and we have made extensive chromosome counts, adding 31 new counts of previous ly uninvesti gated taxa (L. C. Majure et al. unpublished manuscript). Using this cytological information, we established multiple data sets: (1) nuclear data for diploids, (2) ITS for all cytotypes, (3) plastid data for diploids, (4) plastid data for all cytotypes, (5) combined nuclear and plastid data for diploids, and (6) combined nuclear and plastid data for all cytotypes (total evidence). We conducted separate analyses of diploids only (1) because allopolyploids do not arise via cladogenesis, and their inclusion in phylogenetic analyses can result in misleading results (Rieseberg et al. 1996; Soltis et al. 2008 ) and (2) to test the parentage of potential allopolyploids using phylogenetic methods (Mavrodiev et al. 2008; Soltis et al. 2008 ) All data set s were analyzed separately using maximum parsimony (MP) in the program PAUP* 4.0 (Swofford 2002 ) Maximum likelihood (ML) using the program RAxML (Stamatakis 2006 ) and Bayesian methods (BI) in the program MrBAYES (Huelsenbeck and Ronquist 2001 ) The MP an alyses were conducted on all data sets with 10 000 random addition sequence replicates, and support was evaluated by running 1000
26 nonparametric bootstrap (bs) pseudoreplicates, each with 10 random addition sequence replicates. The ML analyses were carried out in RAxML by partitioning each region under 25 rate categories using the GTR model of molecular evolution and carrying out 10 000 nonparametric rapid bootstrap pseudoreplicates for the separate and combined data sets. For BI analyses, models of molecula r evolution for each marker were determined using the program ModelTest ( Posada and Crandall 1998 ) and the Akaike information criterion (AIC). Analyses were carried out by partitioning the data by marker, each with its corresponding model of molecular evol ution, and using four heated chains for 20 million generations, sampling a tree every 1000 generations. We determined stationarity and thus the number of generations Tracer v. 1.5 (http://tree.bio.ed.ac.uk/software/ tracer/). Incongruence length difference (ILD) tests (Farris et al. 1995 ) between plastid and nuclear data sets were carried out in PAUP* (Swofford 2002 ) We initially ran analyses using plastid and nuclear data separately with only known Opuntia diploids. Visual inspection of tree topologies of separate nuclear vs. plastid data analyses (MP, ML, BI) also was used to determine whether any strong incongruence existed between nuclear and plastid data sets that justi fied not combining data (Johnson and Soltis 1998 ; Fishbein et al. 2001 ) Due to the lack of resolution along the backbone of the phylogenies using either plastid or nuclear data alone and the resolution of many of the same clades using the data sets separately, hard inco ngruence (sensu Seelanan et al. 1997 ) using a bootstrap nuclear data sets for further MP, ML, and BI analyses. We then ran separate plastid and nuclear analyses using all of the aforementioned phylogenetic method s with all taxa sampled, including polyploids, to determine from which putative progenitors (at the clade level) many of the polyploid taxa within Opuntia s.s. may have originated. We also analyzed ITS haplotypes from
27 the combined diploid/polyploid data se t in the program TCS v1.21 (Clement et al. 2000 ) to take into account potential incomplete lineage sorting in ITS and inherent problems with the inclusion of reticulate taxa in a bifurcating phylogeny. Polyploid taxa that were recovered in disparate clades using nuclear or plastid data alone in phylogenetic analyses or that were found to have ITS haplotypes from more than one putative progenitor or the same haplotype of a taxon whose relationshi s placement in plastid phylogenet ic analyses were considered interclade allopolyploids. Morphological characters of the putative interclade hybrids and distributions of taxa also were compared with members of putative progenitor clades to provide further evidence for their hypothesized pa rentage We then removed interclade allopolyploids from further analyses. Polyploid taxa inferred to be intraclade polyploids (i.e., polyploids derived from within a given clade) were not removed from our total evidence phylogenetic analyses (i.e., intracl ade phylogeny), because we were interested primarily in clade delimitation and not necessarily species delimitation, which may be obscured by the inclusion of intraclade allopolyploids when employing both nuclear and plastid markers in a combined analysis Biogeographic Analysis and Divergence Time Estimation We used the programs Mesquite v. 2.73 (Maddison and Maddison 2010 ) and RASP (Yu et al. 2011 ) to infer the geographic origin of Opuntia s.s. and major clades by coding all diploid taxa for geographic distribution based on literature (Britton and Rose 1920 ; Anderson 2001 ) and personal experience. We coded seven geographic areas for diploid Opuntia taxa and outgroups based on generalized distributions of the diploid taxa. Those geographic areas were (1 ) s outhwestern South America (western central Chile, Chaco + Monte regions), (2 ) eastern South America (Caatinga), (3 ) western South America (Central Andean valleys), (4 ) northern South America (Caribbean region), (5 ) Central America (including tropical dry f orest of southern
28 Mexico and the Caribbean), (6 ) North American desert region, (7 ) and the southeastern United States. Geographic areas for South America are based on Sarmiento (1975 ) In Mesquite v. 2.73, we implemented the maximum likelihood Mk1 model (u sing our diploid ML topology), which is a Markov k state 1 parameter model that allows for an equally probable change from one character state to the next (Lewis 2001 ; Maddison and Maddison 2010), but without allowing polymorphic states for taxa. In RASP, we used the Bayesian binary Markov chain Monte Carlo (MCMC ) analysis method (Olsson et al. 2006 ; Sanmartn et al. 2008 ; Yu et al. 2011 ) by implementing the JC model with equal rates (Sanmartn et al. 2008 ) and 50 000 MCMC cycles with 10 chains using the tr ees from our Bayesian analysis of diploid taxa as input. We built a condensed (consensus ) tree from those BI input trees to use as a fi nal tree for ancestral area reconstruction. We also used RASP to perform a DIVA (Ronquist 1996 ) analysis and infer dispe rsal scenarios based on our Bayesian trees. Divergence time estimates were obtained using the program r8s v.1.71 (Sanderson 2003 ) and implementing the penalized likelihood method (Sanderson, 2002 ) using the TN algorithm. We calculated smoothing using the c ross validation technique (Sanderson 2003). No fossils are known in Cactaceae (e.g., Hershkovitz and Zimmer 1997), so we used a fi xed age of 5.6 ( 1.9 ) Myr for the crown node of Opuntia s.s. based on dates proposed by Arakaki et al. (2011 ) which coincid es with an inferred late Miocene increase in lineage diversifi cation rates in the clade (Arakaki et al. 2011). We fi xed the age of our outgroup node at 15 ( 2.9 ) Myr, which is the inferred age of the crown node of Opuntioideae, to test the effect of tha t calibration on subsequent age estimates within Opuntia s.s. We also constrained the divergence time of the North American clade with a minimum age of 3 Myr based on the proposed timing for the
29 closure of the Isthmus of Panama (Marshall et al. 1979), whic h would support migration rather than long distance dispersal of the most recent common ancestor of the North American clade into North America. Results We observed very low sequence divergence among the plastid and nuclear sequences in diploid data sets ( Table 2 2) and very little nucleotide polymorphism was observed in directly sequenced ITS products from polyploid taxa. Neither nuclear nor plastid data for diploid taxa alone fully resolved relationships among major clades, but many of the major clades were recovered using either data set separately, although our ILD tests showed a signifi cant difference between all nuclear compared to all plastid sequences ( P = 0.01). It is well known, however, that the ILD test is extremely sensitive and used alone sho uld not be an indicator of data set combinability (e.g., Yoder et al. 2001 ) Rate heterogeneity among sites and small numbers of parsimony informative characters may result in r ejecting congruence among data sets (Darlu and Lecointre 2002 ) There was no ha rd incongruence based on comparison of the nuclear vs. plastid trees using a bootstrap cut off of 70% using either MP or ML. Combining the diploid data sets resulted in well supported clades in the diploids only analysis (Fig. 2 1) Well supported clades a re named based on the series recognized by B ritton and Rose (1920 ) Engelmann (1856 ) or a morphological feature of a given clade. Our analysis of diploids and polyploids placed many polyploid taxa in different clades in the separate ITS and plastid trees (e.g., Opuntia tomentosa is in the Nopalea clade with ITS and the Basilares clade with plastid data ; Fig. 2 5 and B, see Supplemental Data with the online version of this article). Those taxa also were recovered in disparate locations in our analysis of IT S haplotypes using TCS. However, many taxa sharing ITS haplotypes were not resolved in clades together in our phylogenetic analysis of ITS due to the lack of synapomorphies for certain clades. We inferred
30 these taxa to be interclade derived allopolyploids ( Fig. 2 2) Those interclade allopolyploids also reduced clade support when analyzed with the combined nuclear/plastid data set (data not shown). The intraclade phylogeny exhibits well supported clades (bootstrap ) and agrees with the diploid topology, but species relationships within subclades are generally poorly supported ; Fig. 2 3) BI, ML, and MP topologies are virtually identical except for reduced clade support and resolution among clades with MP. Relationships in Opuntieae Subgenus Platyopuntia as recognized by Britton and Rose (1920 ) was paraphyletic, given that most of Tacinga and Nopalea are not included in this subgenus in their classifi cation. Consolea formed a clade with both plastid and ITS data as shown in Fig. 2 5 However, plastid data resolved Consolea outside of Opuntia s.s. (bs = 53%), and ITS data placed Consolea within Opuntia s.s. (bs = 75% ; Fig. 2 5 ), placements that have been found in previous studies (Wallace and Dickie 2002 ; Griffi th and Porter 2009 ) using plastid and ITS data, respectively. However, Consolea was well supported (bs = 86% ) as sister to a clade containing Brasiliopuntia, Tacinga and Opuntia s.s. in a diploids only analysis using combined nuclear and plastid data ( Fig. 2 6 ). Tacinga formed a well supported clade (bs = 81% ) that included Opuntia lilae Trujillo and Ponce, and Brasiliopuntia and Opuntia schickendantzii F.A.C. Weber formed a clade (bs = 87% ) sister to Tacinga The Brasiliopuntia Tacinga clade was not recovered in MP analyses. The Brasiliopuntia, O. schickendantzii, Tacinga clade was resolved as sister to the well supported Opuntia s.s. clade (bs = 84%). Nopalea was nested within Opuntia s.s., as in other st udies (e.g., Wallace and Dickie 2002 ; Wallace and Gibson 2002 ; Griffith and Porter 2009 ; B rcenas et al. 2011 ; Hernndez Hern ndez et al. 2011 ) Altogether, our phylogenetic analyses recovered 10 major clades of Opuntia s.s. ( Figs. 2 1, 2 3 ) which are recognized based on high support values. These 10 major cla des were recovered in BI, MP, and ML analyses.
31 Opuntia s.s. In the ML analyses, the Elatae and Macbridei clades of South America (Argentina Bolivia and central Peru, respectively ) were successive sisters to North A merican Opuntia which comprised two species rich and morphologically diverse clades ( Fig. 2 1 ) However, the sister to the North American clade was unresolved with BI or MP analyses. The more morphologically extreme of the two large North American clades consists of the Nopalea and Basilares sister clades. For example, the Nopalea clade contai ns species with flowers modifi ed for hummingbird pollination. Subclades of the Basilares clade have dry fruited species (subclade Xerocarpa ) rhizomatous taxa (subclade Rhizomatosa ) dioecious species, such as O. stenopetala (Parfitt 1985 ) and the iconic and deceivingly harmless O. microdasys (bunny ear prickly pear ) of the Microdasys subclade. The other of the two large North American clades consists of three subclades ( Scheerianae, Macrocentra and Humifusa ) all containing taxa that, despite extensive vegetative morphological diversity, are fairly homogeneous in their fl oral an d fruit morphology, all with fl eshy fruits and open entomophilous fl owers. Of the 29 series of subgenus Pl atyopuntia of ( Britton and Rose 1920 ) 26 series roughly conformed to Opuntia s.s. (i.e., excluding Brasiliopuntia Consolea and Tacinga inamoena ) Of those 26 series, no single series corresponds exactly to any clade recovered in our topology ; however, there was often general agreement between clades and series composition. For example, series Basilares (Britton and Rose 1920 ) includes O. basilaris O. rufi da and O. microdasys which formed part of the Basilares clade in our phylogeny ( Fig. 2 1 ) Inter clade Allopolyploids and Hybrids We recovered 24 interclade derived taxa. Of these, 20 are inferred to be allopolyploids (4 x 5 x 6 x 8 x and 9 x ) and one is an interclade homoploid hybrid ( Table 2 3 ) We have not yet determined ploidy in O. bella Britton & Rose, O. pittieri Britton & Rose, or O. schumannii
32 F.A.C. Weber ex A. Berger, but they also are inferred to be of interclade origin. Twenty of these taxa are derived from within Opuntia s.s., but four taxa were determined hybrids based on current taxonomy. Opuntia acaulis Ekman & Werdermann, O. bahamana Britton & Rose, and O. lucayana Britton are derived from Consolea and Opuntia s.s., and O. bella is apparently derived from Tacinga and Opuntia s.s. It was not possible to determine the parental species of all of these allopolyploids using ITS, possibly as a result of complete concerted evolution in ITS (lvarez and Wendel 2003 ; Kovarik et al. 2005 ; Kim et al. 2008 ; Soltis et al. 2008 ) Concerted evolution in ITS has also been infer r ed in polyploid species of Gal pagos Opuntia ( Helsen et al. 2009 ) reducing the ability to determine relationships among those species. Furthermore, we have not sampled all extant taxa, and some parental diploids may be extinct. We discovered two or more IT S haplotypes in most cloned accessions, and certain haplotypes were not represented in any other taxa. Although, we recovered O. leucotricha as an interclade allopolyploid, we are uncertain about its placement, given that ITS data place the species (althou gh poorly supported ; bs = 53% ) in the Humifusa clade, with which O. leucotricha neither shares morphological characters nor is sympatric ( Table 2 3 ; Fig. 2 2) Opuntia acaulis O. bahamana and O. lucayana are all derived from hybridization between members of Consolea and a member of the Scheerianae clade, most likely O. dillenii (Ker Gawler ) Haw., which occurs sympatrically with Consolea species throughout their range. Morphology provides support for this interclade hybrid ization. Opuntia acaulis has the indeterminate cladode growth form of Consolea but O. bahamana and O. lucayana possess the determinate cladode growth form of Opuntia s.s. All three taxa show strongly tuberculate areoles, which characterize certain species of Consolea but generally have mostly yellow spines and a shrubby growth form like O. dillenii ; these three hybrids are mosaics, with some
33 morphological traits from each parent, and can be distinguished from both of their putative progenitors. Opuntia bol dinghii Britton & Rose and O. sp. nov. (R. Puente, unpublished data ) were recovered as interclade products between the Nopalea clade and the Scheerianae clade. Nopalea was recovered as the maternal donor and the Scheerianae clade as the paternal donor. Bot h taxa have fl oral characters that combine the morphologies of Nopalea (erect, reddish pink tepals ) and O. dillenii (entomophilous fl owers with spreading tepals). Opuntia cubensis Britton & Rose has long been considered a hybrid derived from O. militaris Britton & Rose (currently a synonym of O. triacantha ) and O. dillenii (Britton and Rose 1920 ) Cloned products of ITS suggest that O. cubensis is an interclade allopolyploid between O. abjecta (currently treated as a synonym of O. triacantha ) of the Humifusa clade and a member of the Scheerianae clade, likely O. dillenii with which it is sympatric. Opuntia cubensis has a combination of yellow, smooth, fl attened spines like O. dillenii and whitish, retrorsely barbed, cylindrical spines that turn gray in age li ke O. abjecta The overall growth form and size of O. cubensis is more similar to O. dillenii but O. cubensis demonstrates disarticulating cladodes like O. abjecta Opuntia bakeri E. Madsen, O. bisetosa Pittier, O. bravoana E. M. Baxter, O. eichlamii Ros e, O. fi cus indica (L. ) Mill., O. megacantha Salm Dyck, O. pillifera F.A.C. Weber, O. pittieri O. schumannii and O. tomentosa Salm Dyck arose from hybridizations between the Nopalea and the Basilares clades ( Fig. 2 2 ) However, it is possible that additional clades from our diploids analysis, not recovered with our data for interclade allopolyploids, may have been involved in these allopolyploidization events given that many of these taxa are hexa and octoploids ( Table 2 3 )
34 Intraclade Allopolyploi ds Determining parentage of allopolyploids derived from within a given subclade of Opuntia s.s. was diffi cult because of sequence similarity among close relatives. However, certain cases were straightforward and are noted here. Hexaploid O. aurea McCabe ex E. M. Baxter and octoploid O. pinkavae Parfi tt (Parfitt 1991 1997 ) are likely intraclade allopolyploids of the Xerocarpa clade, both involving O. basilaris, and members of the O. polyacantha complex. Parfi tt (1991 ) suggested this relationship for O. aure a but not O. pinkavae Plastid data place both of these taxa with high support in the O. polyacantha complex ( O. pinkavae is strongly supported as sister to O. erinacea Engelm. & J. M. Bigelow, and Benson included O. pinkavae in his concept of O. erinacea var. utahensis (Engelm. ) L. D. Benson ; Parfi tt, 1997 ) but ITS sequence data do not support this relationship and rather suggest, according to haplotype analysis, a relationship with O. basilaris Engelm. & J. M. Bigelow. Combined plastid and ITS a nalyses place O. basilaris and O. aurea as subsequent sisters to the O. polyacantha complex ( Fig. 2 3 ) and O. pinkavae as sister to O. erinacea ( Fig. 2 3 ) Both O. aurea and O. pinkavae display numerous morphological characters that are mosaics between O. basilaris and members of the Polyacantha clade. Opuntia pinkavae exhibits pubescent cladodes and pink fl owers like O. basilaris and O. aurea exhibits pubescent cladodes like O. basilaris but the green stigmas, mostly yellow fl owers, seeds with a broad fun icular girdle, and pollen morphology similar to members of the O. polyacantha complex (Parfitt 1991 ) Opuntia aurea and O. pinkavae are found where the geographic distributions of diploid O. basilaris and polyploid members of the O. polyacantha complex overlap (Parfitt 1997 ; Pinkava 2002 ) Opuntia carstenii O. depressa and O. robusta were recovered within the Basilares clade with plastid data and a grade containing mostly members of the Basilares clade with ITS data
35 (Fig. 2 5, 2 6 ), but it was not possible to determine parentage of those taxa from among the four clades (i.e., Excelsa Microdasys Rhizomatosa Xerocarpa ) Biogeography and Divergence Time Estimation of Opuntia s.s. Our biogeographic analysis supports a southwestern South American origin for Opuntia s.s. with subsequent dispersals to the Central Andean Valleys of Peru and the western North American desert region (Fig. 2 4) The most recent common ancestor of Brasiliopuntia and Tacinga also appears to have dispersed from southwestern South America, and one lineage, O. lilae dispersed to the Caribbean region of Venezuela (Fig. 2 4) Both ML (Mesquite ) and Bayesian (RASP ) results support an origin of the North American Opuntia radiation in the deserts of western North America. From the North American desert region, the Nopalea clade dispersed into the tropical dry forest of Mexico, Central America, and the Caribbean. Other North American clades continued to radiate in the North American desert region and in some cases signifi cantly incr eased their ranges via the formation of polyploid taxa. For example, O. fragilis of the Xerocarpa clade moved from the southwestern United States into Canada and the upper Midwest ( Parfi tt 1991 ; Majure and Ribbens, in press ) after formation, and the Humifu sa clade migrated from the west into the southeastern United States forming a small radiation in the Gulf Coastal Plain. Divergent diploid members of the Humifusa clade from the west and east eventually formed contact zones, and allopolyploid taxa expanded north after the last glacial maximum, far surpassing the distributions of diploid taxa ( Majure et al. 2012 b ) Our divergence time estimates suggest that the North American clade originated 5.12 ( 1.6 ) Ma (Fig. 2 7 ), which according to our ancestral area reconstruction would place the North American clade in the western North American desert region before the presumed closure of the Isthmus of Panama at 3 Ma ( Marshall et al. 1979 ) Constraining the North American clade at 3 Ma had no effect on divergence t ime estimates. Subclades within the North American clade
36 subsequently originated from 5 1.5 Ma (i.e., from the early Pliocene through the early Pleistocene) ; however, the majority of those subclades originated during the middle Pliocene ( Fig. 2 7 ). Discussion Consolea The Caribbean genus Consolea consists only of hexaploid and octoploid species (L. C. Majure et al. unpublished manuscript), and the clade could have originated via an allopolyploidization event between other members of tribe Opuntieae ( Negrn Ortiz 2007 ; Griffith and Porter 2009 ) The confl icting placement based on ITS vs. plastid sequence data of species of Consolea certainly support this possibility. Consolea is supported as monophyletic with either ITS or plastid sequence data ( Fig. 2 5 ). Consolea is not resolved as sister to any clade of Opuntia in analyses of ITS data alone (ITS is insuffi ciently variable to illuminate relationships among clades within Opuntia s.s., as shown in Griffith and Porter 2009 ) and plastid data place Consol ea as sister to the Tacinga Brasiliopuntia Opuntia clade ( Fig. 2 5 ). Furthermore, combined analyses of nuclear and plastid diploid data sets place Consolea with strong support (bs = 86% ) as sister to the Tacinga Brasiliopuntia Opuntia clade ( Fig. 2 6 ), so Consolea should not Opuntia s.s., as suggested by Nyffeler and Eggli (2010b ) If Consolea is a result of ancient reticulation, concerted evolution and subsequent ITS divergence may obscure progenitor discovery, or the putative progenitors may have since gone extinct. On the contrary, the placement of Consolea within Opuntia s.s. may represent incomplete lineage sorting or homoplasy in ITS data. Further work will be necessary to resolve the placement of Consolea C onsolea shares morphological characters with numerous taxa. These include monopodial trunks, as in Brasiliopuntia hairy seeds as in Brasiliopuntia Tacinga and some members of
37 Opuntia s.s. (Stuppy 2002 ) hook shaped embryos as in Tacinga (Stuppy 2002 ) a nd expanded floral nectaries for hummingbird pollination as in Tacinga ( Taylor et al. 2002 ) and several Opuntia species (e.g., O. quimilo, Nopalea ; Daz and Cocucci 2003 ; Puente 2006). However, members of Consolea also demonstrate unique characters, which do not appear elsewhere in the Opuntieae, except in interclade allopolyploids with Consolea (e.g., reticulate epidermis and cryptic dioecy ; Strittmatter et al. 2008 ) Consolea has diversified into at least nine species (Areces Mallea 2001 ; Negrn Ortiz 200 7 ) and should not be regarded as synonymous with Opuntia s.s., as proposed by Nyffeler and Eggli (2010b ) Opuntia lilae and Opuntia schickendantzii Previous analyses have shown that one of our outgroups, previously regarded as Opuntia Salmiopuntia salmian a is resolved outside of Opuntia s.s. (Griffith and Porter 2009 ) Our analyses indicated that O. lilae and O. schickendantzii also are not members of Opuntia s.s. (Fig. 2 1) The placement of these two species outside of Opuntia s.s. was unexpected given that Trujillo and Ponce (1990 ) considered O. lilae to be a member of Opuntia series Tunae of Britton and Rose (1920 ) and O. schickendantzii has traditionally been considered a member of Opuntia series Aurantiacae (Britton and Rose 1920 ) Our sequence data here and morphological analyses (L. C. Majure and R. Puente unpublished manuscript ) indicate that O. lilae is a Venezuelan Caribbean member of the mostly Brazilian Tacinga clade. The dis jct. of Cactaceae congeners from the Caatinga of eastern Brazil to th e Caribbean region of northern South America has been observed previously (Sarmiento 1975 ) More research is essential to determine how O. schickendantzii should be treated taxonomically, given that it does not share obvious morphological characters with B rasiliopuntia (Nyffeler and Eggli 2010a ) its sister taxon in our analyses.
38 Nopalea Our results indicate that the hummingbird pollinated Nopalea is nested within Opuntia s.s., in agreement with other analyses (Wallace and Dickie 2002 ; Griffith and Porter 2 009 ; B rcenas et al. 2011 ; Hern ndez Hern ndez et al. 2011 ) Hence, Nopalea should not be recognized at the generic level but does form a clade and could still be recognized within Opuntia s.s. In our combined diploid analysis, Nopalea forms a well supported clade (bs = 96% ) that also includes insect pollinated O. caracassana O. guatemalensis O. jamaicensis O. sanguinea and O. triacantha Shifts from insect pollination to hummingbird pollination have occurred several times in Opuntieae (e.g. Tacinga O. quimilo O. stenopetala Nopalea ; data not shown). In Nopalea this shift resulted in pronounced fl oral morphological changes (e.g., short, erect tepals, and exerted stamens and styles). Such pollinator shifts are common in angiosperms and often result in major morphological changes (e.g., Grant 1994 ; Fenster et al. 2004 ; Penneys and Judd 2005 ; Crepet and Niklas 2009 ) South North American Dis j un ct ion in Opuntia The North American Opuntia clade is nested within the South American Opuntia cla des (Fig. 2 1); the ancestral area reconstruction for the Macbridei (Andean Valleys of Peru and Ecuador ) + North American clade suggests that their most recent common ancestor was from southwestern South America (66% proportional likelihood ; Fig. 2 4) Thus, our data suggest that the most recent common ancestor of North American Opuntia migrated north or was dispersed long distance from South America to western North America. Our DIVA analysis agrees with the long distance dispersal scenario, although wi th a low probability (0.50). The dis jct. of North and South American Opuntia has not been proposed previously, presumably because species of Opuntia exist throughout the Americas from Argentina to Canada (Anderson 2001 ) Similar patterns of dis jct. s betwee n South America and North America can be seen in Cactoideae
39 ( Hern ndez Hern ndez et al. 2011 ) elsewhere in Opuntioideae (Griffith and Porter 2009 ) as well as in the close relatives of cacti, Grahamia (Nyffeler 2007 ) and Portulaca ( Hershkovitz and Zimmer 2000 ) There are other we ll known examples of similar fl oristic dis jct. s between southern South America and the southwestern United States/northern Mexico (Johnston 1940 ; Axelrod 1948 ; Raven 1963 ; Solbrig 1972 ; Lia et al. 2001 ; Simpson et al. 2005 ; Bessega et al. 2006 ; Moore et al. 2006 ; Hawkins et al. 2007 ) although there is still speculation as to why such dis jct. s occur (Solbrig 1972 ) Many of these disjuncts also appear to have their origins in South America (Johnston 1940 ) Most analyses sugges t that these North South American dis jct. s must have formed via long distance dispersal events (Raven 1963 ; Simpson et al. 2005 ; Bessega et al. 2006 ; Moore et al. 2006 ) since ver y few species of the overall fl oras are shared between the two areas (e.g., 2% ; Raven 1972 ) many of these disjunct taxa are not host to the same insect faunas, and the same vertebrates often are not found in the two geographic locations (Raven 1963 1972 ) Further supporting long distance dispersal in Opuntia the cactophagous m oth, Cactoblastis cactorum Berg (Pyralidae), which occurs naturally in southern South America, our proposed geographic origin of Opuntia does not occur naturally in North America. In fact, as aforementioned, introduced populations of C. cactorum are used as a biocontrol agent to destroy introduced populations of North American Opuntia which have not evolved to cope with its gregarious feeding habits (Stiling 2000 ; Stiling and Moon 2001 ; Marisco et al. 2010 ) Likewise, cactophagous moths in North America ( e.g., Melitara Walker ) are in a different clade from C. cactorum suggesting that the internal feeding behavior of these pyralid moths evolved several times within this lineage after the initial evolution of cactophagy in the Pyralidae ( Simonsen, 2008 )
40 I t is presumed that African, Malagasy, Sri Lankan, and Indian populations of the epiphyte, Rhipsalis (Cactoideae), originated via long distance dispersal by birds from their native range in South America (Thorne 1973 ; Benson 1982 ; Barthlott 1983 ; Anderson 2 001 ) Long distance dispersal of Didiereaceae from South America to Africa also has been postulated (Applequist and Wallace 2001 ) Birds (e.g., species of Geospiza ) are also known to disperse the seeds of Gal pagos Opuntia at least for short distances (Gra nt and Grant 1981 ) Numerous other species of birds have been recorded eating fruits and seeds of Opuntia in other areas as well (Dean and Milton 2000 ; Mellink and Riojas Lpez 2002 ) so there may be a link between birds and the long distance dispersal of Opuntia seeds in South and North America. Species of Opuntia s.s. currently exist throughout the neotropics, and it is possible that ancestral populations of the North American clade once occurred in local refugia throughout Central America, a scenario tha t also has been proposed for other arid adapted disjunct taxa (Solbrig 1972 ) It has been established that a desertified environment did not exist throughout the neotropics from the Miocene through the Pliocene (Axelrod 1948 ; Raven 1963), although isolated habitats may have existed (Solbrig 1972 ) These local refugia from South America to western North America (Raven 1972 ; Solbrig 1972 ) with northward migrating populati ons eventually going extinct in more southerly locations. Regardless, the Isthmus of Panama did not create an impassible barrier for the continued northern migration of Opuntia s.s. considering that the closure of the Isthmus of Panama is proposed to have taken place 3 mya ( Marshall et al. 1979 ) and divergence time estimates for the North American radiation (5.12 1.6 M a) place the origin of the clade before that time.
41 The North American Radiation Our phylogeny suggests that Opuntia s.s. radiated rapidly with substantial morphological diversifi cation after its movement into North America. The modern day Sonoran and Chihuahuan deserts were hotspots for the formation of new clades and rampant speciation, as evidenced by the great diversity of Opuntia in thes e regions (G mez Hinostrosa and Hern ndez 2000 ; Hern ndez et al. 2001 ; Powell and Weedin 2004 ) Our dating analysis indicates that the North American clade originated 5.12 ( 1.6 ) Ma. All subclades of the North American clade originated from 5 1.5 Ma, su ggesting that diversifi cation of the clade was facilitated by the expansion of arid habitats during the mid Pliocene through the early Pleistocene (Axelrod 1948 ) and possibly coinciding with the middle Pliocene warm period (Axelrod 1948 ; Haywood et al. 200 1 ; Haywood and Valdes 2004 ) Speciation within and among North American clades was further increased by hybridization and subsequent allopolyploidy, which are common in Opuntia s.s. In contrast, there is little evidence for interclade allopolyploids betwee n the South American clade and other clades, suggesting that those clades remained isolated until modern times with the human introduction of North American taxa into South America or naturally southward migrating taxa (Kiesling 1998 ; Novoa 2006 ) Reticula te Evolution in Opuntia Hybridization between species and subsequent polyploidization (i.e., allopolyploidy ) is a common speciation process in plants (Stebbins 1950 1971 ; de Wet 1971 ; Grant 1981 ; Gibson and Nobel 1986 ; Ramsey and Schemske 1998 ; Soltis and Soltis 2009 ) In Opuntia, the production of allopolyploid species is very common and has led to the origin of many new species ( Pinkava 2002 ) These polyploids often are not completely reproductively isolated from other species ( Grant and G rant 1982 ) How ever, these new genomic combinations often result in
42 morphologically distinct entities, which may propagate themselves indefi nitely via agamospermy, vegetative apomixis, or sexual reproduction ( Rebman and Pinkava 2001 ) Most crosses leading to the formati on of interclade allopolyploids are natural ; however, a few appear to have been human mediated (Kiesling 1998 ; Griffith 2004 ; Reyes Agero et al. 2005 ) Evidence for the use of Opuntia in central Mexico as a foodstuff by Native Americans, where many of the se polyploid taxa occur, has been found dating to at least 14 000 yr ago (Casas and Barbera 2002 ) Kiesling (1998 ) suggested an 8000 9000 yr old date for the domestication of the polyploid, O. fi cus indica, a species still cultivated and used widely as a foodstuff today ( Inglese et al. 2002 ; Felker et al. 2005 ) Many of the shrubby to arborescent allopolyploid taxa, most of which are octoploids, occurring from central Mexico through northern South America, are derivatives of the Nopalea clade, which conta ins the arborescent Nopalea members, and one or more of two other clades (e.g., Basilares Scheerianae ; Fig. 2 2) Baker (2002 ) noted the possible relationship between the Ecuadorian Peruvian octoploid, O. soderstromiana, and the introduced central Mexican octoploid, O. fi cus indica. Berger considered O. schumannii to be intermediate between Nopalea and Opuntia ( Britton and Rose, 1920 ) These taxa have putative progenitors in common from the Nopalea clade and the Basilares clade ( Fig. 2 2) This was unexpec ted, as several South American taxa (e.g., O. bisetosa O. boldinghii O. pittieri O. schumanii ) are actually derived from the North American clade, suggesting that they originated from species of Opuntia migrating south from North America or those being dispersed south by humans or other fauna (e.g., O. fi cus indica ) The common consumption of the fruit of Opuntia by humans and many other animals would allow for the facile dissemination of seeds and thus dispersal by migrating frugivores.
43 Sixty nine speci es of vertebrates (not including Homo sapiens ) have been recorded eating the fruits and/or seeds of Opuntia species ( Mellink and Riojas Lpez 2002 ) Davis et al. (1984 ) found seeds of Opuntia in wooly mammoth ( Mammuthus ) dung, which confi rms the use of Opu ntia s.s. by Pleistocene megafauna and further emphasizes potential long distance dispersal via migrating herbivores. Interclade taxa involving the Scheerianae clade consistently have a member of the Scheerianae clade as the paternal donor and the other cl ade involved as the maternal donor (e.g., O. acaulis O. bahamana O. boldinghii O. cubensis O. lucayana Opuntia sp. nov. 1). This is most likely the result of specialized pollination syndromes (primarily bird pollination ) in Consolea and Nopalea since hummingbirds presumably rarely visit the entomophilous fl owers of Scheerianae However, insects occasionally visit hummingbird pollinated taxa, such as O. quimilo (Daz and Cocucci 2003 ) and Nopalea (Puente 2006 ) In the case of the allopolyploid O cubensis the putative paternal progenitor O. dillenii of the Scheerianae clade is much larger than the putative maternal progenitor O. abjecta and may thus be more easily accessible to insect pollinators, leading to higher transfer rates of pollen from O. dillenii to receptive stigmas of O. abjecta Alternatively, genetic interactions may determine whether reciprocally formed polyploids are both viable. The precise origins of those species designated intraclade polyploids are not clear for several reaso ns. First, limited sequence divergence among closely related species precludes determination of the specifi c origins of true intraclade polyploids. Second, concerted evolution of ITS in an allopolyploid may conceal one of the putative progenitors (lvarez and Wendel 2003 ; Kovarik et al. 2005 ; Kim et al. 2008 ; Soltis et al. 2008 ) such that a true allopolyploid (interclade or intraclade ) would not be detected as such. Finally, autopolyploidy, rather than
44 allopolyploidy, could explain a pattern of shared sequences between diploids and polyploids. Some taxa included in our analyses are composed of more than one ploidal level (e.g., O. macrocentra O. pusilla O. strigil ; Pinkava 2002 ; Powell and Weedin 2004 ; Majure et al. 2012 b ); samples of different cytoty pes are sometimes morphologically similar and form clades (e.g., O. pusilla ; Fig. 2 3) suggesting autopolyploidy. Autopolyploids have been found elsewhere in Cactaceae, although the best documented are restricted to subfamily Cactoideae ( Sahley 1996 ; Hamr ick et al. 2002 ; Nassar et al. 2003 ) Autopolyploids may play a much larger role in plant speciation than is currently recognized ( Judd et al. 2007 ; Soltis et al. 2007 ) and may have been infl uential in the diversifi cation of Opuntia s.s. as well. Determin ing the o rigins of all intraclade polyploids thus would be especially informative. Summary Opuntia s.s. is a well supported clade, which originated in southwestern South America and quickly diversified after a northern migration or long distance dispersal into the arid regions of western North America. Reticulate evolution and polyploidization have played a major evolutionary role in the clade by producing novel phenotypes and increasing species richness. The complexity of phylogenetic relationships among s pecies and major clades is increased by polyploids, so determining the ploidy of all taxa is imperative to the construction of an accurate evolutionary history of the clade. Detailed phylogenetic, morphological, and fi eld studies of taxa within each clade will be necessary to fully understand relationships and biogeographic patterns at the species level. Given the proposed recent ages for Opuntia s.s. (5.6 1.9 Ma ; Arakaki et al. 2011 ) and its subclades given here, Opuntia s.s. shows the signature of a cl ade that has undergone a rapid radiation (i.e., broad distribution, high morphological and species diversity, and low molecular marker divergence ; Malcomber 2002 ) The nuclear and plastid data do not fully resolve species
45 relationships within clades, and s everal nodes along the backbone of the phylogeny lack high bootstrap support, although the major clades of Opuntia s.s. are generally well supported. Rapid radiations are often constrained by the lack of support for clade relationships (e.g., Fishbein et a l. 2001 ; Malcomber 2002 ; Valente et al. 2010 ) Increased taxon and marker sampling is an important next step in determining relationships among all species of Opuntia s.s. Species delimitation will require development of appropriate markers to allow for th e discovery of intraspecifi c variation, using multiple accessions from each potential species described within that clade. This work will also allow for the potential discovery of morphologically cryptic species within taxa composed of multiple ploidal lev els and for illuminating the origins and evolutionary role of the abundant polyploids in the clade.
46 Table 2 1. DNA regions and associated primers used in this study. Region Primer name: sequence or reference atpB rbcL GTAAACTATGTCGAAATTCTTTGC GTAAACTATGTCGAAATTCTTTGC matK matKx: ( http://www.kew.org/barcoding/update.html) matK5: ( http://www.kew.org/barcoding/update.html) ndhF rpl32 TGCTGAATAGACAGCTTCA TGGTCAAACGAATCTTTG psbJ petA psbJ: (Shaw et al. 2007) CAACATCAAGTTCGTAACAAG trnL F trnE: (Taberlet et al. 1991) trnF: (Taberlet et al. 1991) ycf1 CTTATCTCTTACTTCTCCAAGCTC GCGGCTAAACTAGGTGGATGTG nrITS ITS4: (White et al. 1990) ITS5: (White et al. 1990) ppc GAGATGAGGGCAGGGATGAGTTACTTCC CTAGCCAACAAGCAAACATC Table 2 2. Statistics of regions sequenced in this study based on the diploid data sets The length (bp) of aligned sequences includes gaps introduced during alignment. Region Length (bp) No. pars. infor. characters Model selected atpB rbcL 861 20 HKY matK 905 27 F81+I+G ndhF rpl32 1699 43 GTR+I+G psbJ petA 1169 72 K91uf+l trnL F 441 14 K81uf ycf1 873 51 K81uf+l+G ITS 599 39 TVM+G ppc 469 37 HKY+G cpDNA combined 5948 227 nuclear combined 1068 76 All Combined 7016 303
47 Table 2 3. Interclade derived taxa recovered in our analyses. Ploidy is given for each species where known based on Majure et al. (submitted). Species Putative progenitors Source O. acaulis (8 x ) O. bahamana (6 x ) O. lucayana (4 x ) Consolea Scheerianae clade Plastid data ITS data O. bella (unknown) Basilares clade Nopalea clade Tacinga Plastid data ITS data ITS data O. bakeri (9 x ) O. bisetosa (6 x ) O. bravoana (6 x ) O. eichlamii (6 x ) O. ficus indica (8 x ) O. fuliginosa (8 x ), O. megacantha (8 x ) O. pilifera (8 x ) O. pittieri (unknown) O. schumannii (unknown) O. soederstromiana (8 x ), O. tomentosa (8 x ) Basilares clade Nopalea clade Basilares clade Nopalea clade O. boldinghii (6 x ) O. sp. nov. 1 (2 x ) Nopalea clade Scheerianae clade Plastid data ITS data O. durangensis (4 x ) O. oricola (6 x ) O sp. nov. 2 (6 x ) Basilares clade Scheerianae clade Plastid data ITS data O. cubensis (5 x ) Humifusa clade Scheerianae clade Plastid data ITS data O. phaeacantha (6 x ) Scheerianae clade Macrocentra clade Plastid data ITS data O. leucotricha (4 x ) Basilares clade Humifusa clade ? Plastid data ITS data Consolea Sister to Tacinga, Brasiliopuntia, Opuntia s.s. clade Opuntia s.s.? Plastid data ITS data
48 Figure 2 1. Diploid phylogeny of Opuntia s.s. Most likely topology from our RAxML run with 10 000 bootstrap pseudoreplicates using our combined nuclear (ITS and ppc ) and plastid data set for diploid taxa only (i.e., all presumably polyphyletic taxa excluded). Opuntia schickendantzii is resolved as sister to Brasiliopuntia brasiliensis, and O. lilae is resolved as sister to Tacinga palmadora The Brasiliopuntia Tacinga clade is sister to Opuntia s.s. in which Nopalea is deeply nested. Well supported clades are named based on series recognized by B ritton and Rose (1920), Engelmann (1856), or a orphological feature of a given clade. Bootstrap values are given to the left above branches and posterior probabilities (right) are denoted as + for values of 1.0 and ies < 0.95 are not given.
49 Figure 2 2. Diploid phylogeny of Opuntia s.s. (adapted from Fig. 2 1) with interclade reticulate taxa mapped on their putative diploid progenitor clades. We did not discover any interclade taxa derived from the South American Elatae or Macbridei clades. Instances where putative progenitors of inferred interclade all opolyploids could not be verifi ed are denoted as ? (e.g., Consolea ). Interclade reticulate evolution is always associated with members of the North American Opuntia radiation.
50 Figure 2 3. Intraclade phylogeny of Opuntia s.s. (total evidence phylogeny excluding interclade derived taxa). The 50% majority rule consensus tree from a RAxML analysis of 10 000 rapid bootstrap pseudoreplicates using our combined nuclear and plastid data set for all diploid taxa (blue) and intraclade polyploids (red). Taxa lacking ploidy information are left black. Bootstrap values are shown above branches; posterior represented below branches by a plus sign ( + ).
51 Figure 2 4. Ancestral area reconstruction and putative dispersal pathways of Opuntia clades. Ancestral reconstructions are represented as (A) southwestern South America, (B) Caatinga, (C) Central Andean valleys, (D) northern South A merica (Caribbean Region), (E) tropical dry forests (Mexico, Central America, Caribbean), (F) western North American desert region, and (G) southeastern United States. Proportional likelihoods are given for each node in the phylogeny. Dispersal probabilites are given along a given pathway on the map. Opuntia s.s. originated in southern South America (A), and then expanded to the Central Andean Valleys (C) and the western North American desert region (F) from where it expanded in distribution and diversifi ed into eight subcl ades From the southwestern North American desert region, the Nopalea clade dispersed into the tropical dry forests of Mexico, Central America, and the Caribbean (E), and the Humifusa clade dispersed into the southeastern United States (G). The ancestor t o the Tacinga Brasiliopuntia clade (B), an eastern Brazilian clade of the Caatinga, also originated in southwestern South America. One lineage, O. lilae, dispersed to the northern South American Caribbean from the Caatinga region (D).
52 Figure 2 5. Plastid (left) and ITS (right) phylogeny including all diploid and polyploid species of Opuntia as well as the genus Consolea
53 Figure 2 6. Diploid phylogeny including the genus Consolea. Consolea is resolved as sister to the Tacinga, Brasiliopuntia, Opuntia s.s. clade.
54 Figure 2 7. C h ronogram from r8s analysis showing an early Pliocene origin of the North American clade of Opuntia
55 CHAPTER 3 Opuntia l ilae ANOTHER Tacinga HIDDEN IN Opuntia S.L. (CACTACEAE) Background Tribe Opuntieae within subfamily Opuntioideae of Cactaceae consists of Brasiliopuntia A. Berger Consolea Lem. Miqueliopuntia Fri ex F. Ritter Opuntia schickendantzii F. A. C. Weber, Opuntia Mill. s.s., Salmiopuntia Fri ex Guiggi Tacinga Britton & Ros e, and Tunilla D. R. Hunt & Iliff (Wallace and Dickie 2002; Griffith and Porter 2009; Majure et al. 2012a ). Most of the aforementioned genera were historically included in Opuntia s.l. (Britton and Rose 1920), but have recently been segregated based on morphological and molecular data (Stuppy 2002; Wallace and Dickie 2002). Britton and Rose (1920), however, separated the distinctive genus Tacinga from their broadly circumscribed Op untia s.l. At the time of its description by Britton and Rose (1920) Tacinga included only one species, T. funalis Britton & Rose, from Baha, Brazil. This species was discovered in the characteristic dry caatinga (spelled catinga in Britton and Rose 1920 ) vegetation and was thus named for it using the anagram Tacinga (Britton and Rose 1920). Molecular phylogenetic analyses have increased the size of the genus, showing that species previously considered part of Opuntia s.l. are more closely related to mem bers of Tacinga than to other species of Tacinga (e.g., T. inamoena (K. Schum) N. P. Taylor & Stuppy T. palmadora (Britton & Rose) N. P. Taylor & Stuppy T. saxatilis (F. Ritter) N. P. Taylor & Stuppy T. werneri (Eggli) N. P. Taylor & Stuppy; Wallace an d Dickie, 2002). Taylor et al. (2002) subsequently transferred those species to Tacinga which now includes seven species (including T. braunii Esteves and T. subcylindrica (M. Machado & N. P. Taylor) M. Machado & N. P. Taylor; see Lambert 2009 and Menezes et al. 2011) and one hybrid taxon, T. x quipa (F. A. C. Weber) Taylor & Stuppy (Taylor et al. 2002). Morphological analyses of those taxa have also illuminated potential synapomorphies for
56 members of Tacinga (Stuppy 2002), which distinguish members of this clade from Opuntia s.s. and other genera in Opuntieae. As found in some other members of Opuntieae (e.g., Consolea, Nopalea of Opuntia s.s.), species of Tacinga produce hummingbird pollinated flowers with tepals v arying in color from green (greenish white) to red or orange. The tepals are spreading to recurved in some species (e.g., T. funalis, T. inamoena ), but may be erect and form a tube in other members of the clade (e.g., T. palmadora, T. werneri ). Stamens ar e athigmonastic, and in some taxa one or two rows of staminodia are produced (Stuppy 2002; Lambert 2009). The fruits of Tacinga have a characteristic deep and narrow umbilicus (Stuppy 2002). Members of Tacinga also develop a hook shaped embryo, unlike the coiled embryo of Opuntia s.s., and have relatively reduced perisperm formation relative to the small embryo size, as compared to the very reduced perisperm and large embryo size in Opuntia s.s. (Anderson 2001; Stuppy 2002). Species of Tacinga produce see ds with a hairy funicular envelope, and some species display indeterminate growth, two characters shared by other members of Opuntieae ( Brasiliopuntia, Consolea ). Trujillo and Ponce (1990) and Fern ndez (1995) considered O. lilae Trujillo & Ponce to repr esent part of Opuntia series Tunae of Britton and Rose (1920), although Trujillo and Ponce (1990) mention that the species does not have the easily disarticulating cladodes of the other members of that series. The geographic location of O. lilae in the northeastern portion of Venezuela and the proximity to the Caribbean, where other members of series Tunae occur (e.g., O. caracassana Salm Dyck O. triacantha (Willd.) Sweet), presumably were influential in decisions regarding relationships to othe r taxa. Opuntia bella Britton & Rose from Colombia, which was placed in series Tunae by Britton and Rose (1920) and shares characters with O. lilae (erect tepals, included stamens at anthesis), is suggested to be an intergeneric hybrid between
57 Opuntia s.s. and Tacinga (Majure et al. 2012a ). The morphological similarity of O. bella and O. lilae may also have encouraged Trujillo and Ponce to include O. lilae in the series Tunae as no other members of the series share obvious morphological characters with O. lilae Recent molecular analyses of Opuntia s.s. using plastid and nuclear sequence data revealed that O. lilae is most closely related to members of the Tacinga clade (Majure et al. 2012a). Here we reconstruct the phylogeny showing the relationship of O. lilae with Tacinga and use the phylogeny to search for putative synapomorphies of the Tacinga clade. We then describe morphological apomorphies of O. lilae (based on the type collection of O. lilae ), which are shared with other members of the Tacinga clade Materials and Methods Taxon Sampling and Phylogenetic Analysis We performed a phylogenetic analysis including four species of T acinga (T. funalis, T. inamoena, T. palmadora, T. saxatilis), Brasiliopuntia, Opuntia lilae, O. schickendantzii, and three species of South American Opuntia s.s ., O. arechevalatae Spegazzini O. macbridei Britton & Rose, and O. quimilo K. Schum. Miqueliopuntia miquelii (Monv.) F. Ritter Tunilla corrugata (Salm Dyck) D. R. Hunt & Iliff, and Salmiopuntia salmiana (J. Parm. ex Pfeiff.) Guiggi were used as outgroups. Our molecular data and morphological observations of O. lilae were based upon the holotype ( see also Appendix B ). We used previously gathered data from Majure et al. ( 2012a ); the plastid intergenic spacers, atpB rb cL, ndhF rpl32, psbJ petA, trnL F plastid genes, matK and ycf1 the nuclear gene ppc and the nuclear ribosomal internal transcribed spacer s (ITS). A sequence of matK for T. funalis was downloa ded from GenBank (see Appendix B ). We performed a maximum likelihood analysis of combined plastid and nuclear data sets (see Majure et al. 2012a), however, with the reduced taxon data set described above, in RAxML (Stamatakis
58 2006) implementing 10000 rapid bootstrap (bs) pseudoreplicates under 25 rate categories using the GTR+ model of molecular evolution. Ancestral State Reconstruction We coded the following 11 morphological characters for those taxa used in our phylogenetic analysis: 1) growth determinate vs. indeterminate, 2) pollinat ion syndrome, i.e., insect vs. hummingbird, 3) embryo shape coiled vs. hooked, 4) funicular envelope with or without hairs, 5) stomata placement at the epidermal surface or sunken vs. raised above the surface, 6) pollen exine condition bullate, reticula te, punctate, or tectate punctate, 7) perisperm production i n relation to embryo size, greatly reduced vs. reduced 8) umbi licus shallow and broad vs. deep and narrow, 9) bud shape acute vs. compressed, 10) stamen movement absent vs. present, and 11) stami nodia absent or present. We performed an ancestral state reconstruction in Mesquite v. 2.73 (Madison and Madison 2010) using maximum likelihood (ML) and maximum parsimony (MP) methods with unordered states to determine which characters exhibited by O. lil ae may represent likely synapomorphies of the Tacinga clade. Under ML we implemented the Mk1 model of evolution, which is a Markov k state 1 parameter model that allows for an equally probable change from one character state to the next ( Lewis 2001; Maddis on and Maddison 2010). We also cross compared morphological characters of the type collection of O. lilae with other members of the Tacinga clade to support likely species relatio nships with regard to the phylogeny. Results Phylogenetic and Morphological Analysis Opuntia lilae is well supported (bs = 92) as a member of the Tacinga clade. The phylogeny suggests a close relationship of O. lilae with T. palmadora (Fig. 3 1). Unlike members of the hummingbird pollinated Opuntia clade, Nopalea which develop acute flower
59 buds (Fig. 3 2A), O. lilae has the typical compressed flower bud apex (Fig. 3 2B) of most Tacinga species (e.g., T. inamoena, T. palmadora, T. saxatilis, T. werneri ). However, T acinga braunii and T. funalis have relatively lon g, acute flower buds (see photos in Taylor et al. 2002; Lambert 2009). The flowers of O. lilae have a tubular, orange red corolla (Fig. 3 2B D), much like that of T. palmadora or T. werneri (Ande rson 2001; Taylor et al. 2002; Lambert 2009). Analysis of th e fruit shows the characteristic deep, narrow, umbilicus of other Tacinga species (Fig. 3 2F), hairy seeds, a strongly hooked embryo, and reduced perisperm development (Fig. 3 2H), of other Tacinga species, as compared to Opuntia s.s. (Fig. 3 2G). Opuntia lilae also shows the characteristic tectate punctate pollen (not shown) of other Tacinga species (Taylor et al. 2002). Ancestral State Reconstruction Of the 11 characters used in our analysis, four of those appear to represent synapomorphies for the Tacin ga clade under both ML and MP methods (Fig. 3 2). Species of the Tacinga clade have: 1) hook shaped embryos with 2) reduced perisperm, 3) raised stomata (as evident in Opuntia lilae, Tacinga inamoena, T. palmadora and T. saxatilis ), and 4) a deep, narrow umbilicus. These features contrast with the coiled embryos with greatly reduced perisperm, and superficial or sunken stomata of Opuntia s.s., Brasiliopuntia and Opuntia schickendanztii In general, other members of Opuntieae have relatively shallow and broader umbilici than those of Tacinga (Fig. 3 3). Although, all species of Tacinga studied thus f ar have tectate punctate pollen, members of Opuntia s.s. also have punctate pollen, which may either be finely tectate punctate, as i n members of the Nopalea clade (sensu Majure et al. 2012a ), or punctate with large slit like apertures, as in O. macbridei and O. stenopetala (see Leuenberger 1976). The switch from reticulate to punctate pollen in Opuntia s.s. appears to be related to sh ifts to hummingbird
60 pollination, as those Opuntia exhibiting punctate p ollen are generally hummingbird pollinated (e.g., O. stenopetala, members of the Nopalea clade). Interestingly, the finely tectate punctate pollen of some species of Tacinga (e.g., T. funalis among those studied) and of the Nopalea clade of Opuntia s.s. is strongly associated with hummingbird pollination and the possession of flowers with greatly exerted stamens and stigmas. However, other species of Tacinga ( O lilae, T. inamoena, T. palmadora, T. werneri ) and Opuntia s.s. (e.g., O. macbridei, O. stenopetala ) exhibit coarsely tectate punctate pollen and have flowers that may be more commonly visited by both hummingbirds and insects. This variational pattern highlights the fact that p ollen characters in Opuntieae are homoplasious and likely dependent on pollination s yndrome. Seeds with a hairy funicular envelope are pleisiomorphic in Tacinga as, according to our analyses, the most recent common ancestor of Opuntieae is reconstructed as having hairy seeds. Hairy seeds have subsequently been lost in other groups (certain Opuntia s.s., Tunilla ) but were retained in Tacinga Bud shape, growth fo rm, pollination syndrome, the presence of staminodia, and athigmonastic stamens are apparently homoplasious characters, as they have been derived separately in other lineages of Opuntieae. Even the Tacinga clade shows a shift in bud shape: from generally c ompressed in most members to acute in T. funalis and T. braunii Tacinga also exhibits variation in growth form, with some species showing determinate growth (e.g., O. lilae T. inamoena, T. saxatilis ) and others indeterminate growth (e.g., T. braunii, T. funalis, T. palmadora ), and staminodia are lacking in most members of the genus, but have been acquired in T. braunii, T. funalis and T. werneri (Taylor et al. 2002). Considering our phylogeny, indeterminate growth and staminodia evolved twice in Tacinga However, a more complete taxon sampling (i.e., including T. braunii T. subcylindrica and T. werneri ) and a well resolved phylogeny are necessary to fully explore character shifts in the clade.
61 Discussion It is clear not only from our molecular phyloge netic analyses, but also from our morphological investigation that O. lilae from northeastern Venezuela, is a member of the Tacinga clade Opuntia lilae has a hook shaped ovary (Fig. 3 2A, Fig. 3 3H) with reduced perisperm relative to emb ryo size (Fig. 3 2B, Fig. 3 3H) and a very deep umbilicus (Fig. 3 2D, Fig. 3 3F), as compared to most species of Opuntia s.s.; all of these are synapomorphic for the Tacinga clade. Opuntia lilae has a characteristic epidermis, which feels like sand paper to the touch. Fernndez (1995) showed this to be the result of raised stomata (see Figs. 1 5 in Fern ndez 1995). Tacinga inamoena, T. palmadora, and T. saxatilis also have raised stomata but we have not examin ed this character in other members of the Tacinga clade (Fig. 3 2C). Stomata are superficial or slightly sunken in Brasiliopuntia Opuntia schickendanztii Salmiopuntia and species of Opuntia s.s. that have been surveyed (Eggli 1984, Majur e unpubl. data). Thus, at least based on the taxon sampling of our analysis, this distinctive anatomical condition is also an apomorphy of Tacinga Opuntia lilae also possesses erect, orange red tepals forming a tube, included stamens (unlike hummingbird pollinated Nopalea of Opuntia s.s., which has exserted stamens), and a hairy funicular envelope, as do other species of Tacinga (Stuppy 2002; Taylor et al. 2002). Brasiliopuntia, Consolea and some species of Opuntia s.s. also have a hairy funicular envel ope (Stuppy 2002), so although characteristic of Tacinga this character is not sy napomorphic for the genus (see R esults). Opuntia lilae appears to be most closely related to T. palmadora, and likely also T. werneri, considering vegetative morphology, floral characters, and the close relationship of T. palmadora and O. lilae in our phylogeny. The yellowish spines of O. lilae (Fig. 3 2I) suggest a closer relationship with T. palmadora than to T. werneri The hexaploid chromosome count of O.
62 lilae (Majure et al. in review), however, differs from the morphologically similar but diploid T. palmadora (de Castro 2008) suggesting that O. lilae and T. palmadora are likely not conspecific. Also, O. lilae does not show the inde terminate growth displayed by T. palmadora (Taylor et al. 2002). Ploidy of T. werneri also needs to be determined, and detailed morphological comparisons of all three taxa need to be undertaken. Whether or not O. lilae is derived from hybridization and su bsequent genome duplication or merely intraspecific genome duplication also requires further study. Tacinga had been considered to be endemic to Brazil (Taylor et al. 2002; Lambert 2009); however, the distribution of this clade must now be extended to nort hern Venezuela, a dis junction of ca. 3, 285 km from the northwestern most population of Tacinga (T. palmadora ) in Pernambuco, Brazil, based on distributions given by Taylor et al. (2002). Divergence time estimates and ancestral area reconstructions suggest that Tacinga originated in Brazil during the Pliocene and subsequently dispersed to northern South America (Majure et al. 2012a ). Tacinga may have once possessed a contiguous distribution from eastern Brazil to at least the western cordillera of Colombia, where the inter clade hybrid between Opuntia s.s. and Tacinga O. bella is found (Britton and Rose 1920; Majure et al 2012a ). Alternat ively, Tacinga dispersed to Venezuela and Colombia, becoming an apoendemic taxon (sensu Stebbins 1971) in Venezuela and hybridizing with Opuntia in Colombia (Majure et al. 2012a ). Sarmiento (1975) noted numerous floristic similarities (i.e., shared genera) between the caatinga and Caribbean dry region of Venezuela and suggested that these two areas may have once been closely linked by a broader dry region spanning the two areas or that the vegetation of both areas may be derived from a common origin. The p resence of O. lilae in Venezuela would suggest a common origin scenario for certain vegetational elemen ts in these two areas, although most Opuntia s.s. species
63 found in Venezuela (e.g., O. boldinghii, O. caracassana ) are derived from North American clades (Majure et al. 2012a ) and not those from South America. Trujillo and Ponce (1990) stated that O. lilae is naturally rare in Venezuela and is known from only two states, Lara and Sucre. Opuntia lilae is known to produce vegetative propagules (i.e., to d emonstrate proliferous growth) even from mature fruit (Trujillo and Ponce 1990) that contain apparently viable seeds, so the species regularly produces vegetative clones from parent plants. The role of sexual reproduction, however, is not known, but is sug gested to be low (Trujillo and Ponce 1990). More research is warranted to determine whether or not apomixis may play a role in seed formation in this uncommon, highly clonal, hexaploid species. Also, better taxon sampling will be necessary to determine how closely related O. lilae is to T. palmadora T. werneri, or other members of the Tacinga clade. Phylogenetically based morphological analyses, including all Tacinga species, will be necessary to fully evaluate morphological character shifts within the cla de. Because as discussed above, our molecular and morphological phylogenetic analyses strongly support the transfer of Opuntia lilae to the genus Tacinga we provide the following new combination. Tacinga lilae (Trujillo and M. Ponce) Majure and R. Puente comb. nov. Basionym: Opuntia lilae Trujillo and M. Ponce, Ernstia 58 60: 1 1990.
64 Figure 3 1. Phylogram of Tacinga and other members of Opuntieae from a combined analysis of nuclear and plastid loci in RAxML carrying out 10000 bootstrap pseudoreplicates under 25 rate categories and implementing the GTR+ model of molecular evolution. Bootstrap values are given above the branches. Opuntia lilae is resolved as sister to Tacinga palmadora in the well supported Tacinga clade. Low bootstrap support and resolution for the subclade of Tacinga containing T. funalis likely is the result of lack of data (i.e., only matK was available for T. funalis ) and taxon sampling.
65 Figure 3 2 ML character state reconstructions for A) embryo shape, B) perisperm development, C) stomatal placement, and D) umbilicus structure. Hook shaped embryos, reduced perisperm development, raised stomata, and deep, narrow umbilici are putative synapomorphies of the Tacinga clade as shown in our reconstructions.
66 Figure 3 3 Morphological characters of Tacinga lilae from the type collection, Trujillo & Ponce 18643 A) acute flower bud of the hummingbird pollinated Opuntia cochenillifera contrasting with B) the dorsally compressed flower bud of O lilae C D) red flowers with erect tepals forming a tube and included stamens typical of certain members of Tacinga (i.e., T. palmadora, T. werneri ), E) shallow umbilicus of Opuntia macrorhiza ( LCM 3510 ) as compared to F) the de ep, narrow umbilicus of O lilae (scale = 1cm) G ) cross section of a seed of O. macrorhiza showing greatly reduced perisperm development (arrow) with relation to embryo size and a coiled embryo compared to H) O lilae with moderately reduced perisperm development (arrow) with relation to embryo size and a hook shaped ovary, (scale = 3mm), and I) cladode of O lilae showing spine production resembling T. palmadora and T. werner i.
67 CHAPTER 4 A CASE OF MISTAKEN I DENTITY, Opuntia a bjecta LONG LOST IN SYNONYMY UNDER THE CARIBBEAN SPECIES, O. t riacantha AND REASSESSMENT O F THE ENIGMATIC O. c ubensis Background John Kunkell Sm all, a plant systematist and Curator of the New York Bot anical Garden fro m 1898 1934, wrote a flora for the Southeastern United States for his Ph.D. dissertation ( http://sciweb.nybg.org/science2/libr/finding_guide/small.asp ). He produced three ed itions of his treatment of the Southeastern F lora from 1903 1933 (Small 1903, 1913, 1933) in which he paid special attention to the cacti of the southeastern United States. Small described 16 species of Opuntia from Florida alone. Two species, Opuntia ab jecta Small and O. ochrocentra Small, were described from Big Pine Key, Florida (in Britton and Rose 1920). The population of O. abjecta at Big Pine Key was the only population that Small mentioned when he described the species and was the only population of the taxon known until recently (K. Bradley, Institute for Regional Conservation, pers. comm.). Opuntia ochrocentra apparently was kno wn from the type locality and fa rther north (135 km) at Cap e Romano (Small 1933), although only specimens from Big Pin e Key and Big Munson Island h ave ever been seen (Benson 1982; Majure et al. 2012b ). O. abjecta and the Cuban O. militari s Britton and Rose (Britton and Rose 1919) in synonymy with the Greater and Lesser Antillean species, O. triacantha (Willd.) Sweet. Since that publication, the name O. triacantha has been used, mostly without question, for material from the Florida Keys ( Doyle 1990 ; Pinkava 2003 ; Wunderlin and Hansen 2003, 2011). Inter estingly, Anderson (2001) treated O. abjecta as a synonym of O. triacantha but did not include the Florida Keys within the geographic distribution of that species. Opuntia abjecta (under O triacantha ) is considered an endangered species in Florida
68 (Coile and Garland 2003) and thought to be the northernmost population of O. triacantha in North America, occurring as a northern disjunct from the nearest population of O. triacantha in southeas tern Cuba (i.e., O. militaris ; Benson 1982). Opuntia cubensis Britton and Rose was originally described from the Guantanamo Bay area of Cuba (Britton and Rose 1912), as a putative hybrid between O. militaris and O. dillenii (Ker Gawl.) Haw. (Britton and R ose 1920). Benson (1982) later determined that another species, O. ochrocentra Small, also described by J.K. Small from Big Pine Key Florida, was synonymous with O. cubensis although Britton and Rose (1920) had considered O. ochrocentra to be a close relative of O. dillenii included O. ochrocentra within O. cubensis (Anderson 2001 ; Pinkava 2003 ; Wunderlin and Hansen 2003, 2011). Phylogenetic analyses of Opuntia (Majure et al. 2012a ) an d morphological studies of Opuntia for the monograph of the Humifusa clade ( Chapter 7 ) suggest that O. abjecta is a different species and evolutionar il y divergent from O. triacantha and another of its synonyms, O. militaris. Majure et al. (2012b) determin O. cubensis was likely derived from hybridization between O. abjecta and most probably O. dillenii instead of O. militaris from Cuba. We expand upon those previous analyses here with the inclusion in our phylog eny of O. cubensis and O. militaris from Cuba. We also present a detailed morphological examination of O. abjecta, O. cubensis, O. ochrocentra and O. triacantha to provide a clear understanding of why O. abjecta, O. triacantha, O. cubensis, and O. ochroc entra should not be considered conspecific. We also discuss the relationship of O. militaris to O. triacantha from a morphological and phylogenetic perspective
69 Materials and Methods Previously gathered data from the plastid intergenic spacers atpB rbcL ndhF rpl32, psbJ petA, trnL F, the plastid genes ycf1 and matK the nuclear ribosomal internal transcribed spacers (ITS; White et al. 1990) and the nuclear gene ppc (Majure et al. 2012a ) were used for our phylogenetic analyses. However, we enhanced our sa mpling to include O. cubensis and O. militaris from Cuba. Live material of Opuntia cubensis was obtained from field collected (Cuba; Areces s.n .) material now grown at Gemini Botanical Garden, Florida. Although no recent specimens of O. militaris exist, to our knowledge, we were able to extract and amplify DNA from a specimen collected in 1951 ( R.N. Jervis 1033 ; MICH) from the Guantnamo Bay area (see Appendi x C ). Both tepal and epidermal tissu e produced usable DNA, although tepal tissue was supe rior in quality (i.e., DNA was less degraded) to the epidermal tissue used. Opuntia triacantha also was sampled from an herbarium sample ( Mori et al. 22693 ; NY), as we did not have live material of that species. We cloned ITS and ppc PCR products of O. c ubensis and O. ochrocentra using the Stratagene cloning kit (Stratagene, La Jolla, CA) and sequenced eight clones of each using bacterial primers (T3 T7) from the kits. We sampled the type (diploid) population of O. abjecta (Majure et al. 2012b ) and O. oc hrocentra (pentaploid) from Big Pine Key, as well as available herbarium material for morphological work, including the type specimens. We also included diploid members of the Humifusa clade, the closely related Macrocentra and Scheerianae clades, and me mbers of the Nopalea clade (sensu Majure et al. 2012a ), to which O. triacantha is morphologically most similar (e.g., O. caracassana, O. guatemalensis, O. jamaicensis ). South American species of Opuntia, O. retrorsa and O. macbridei, were used as outgroup s based on results from Majure et al. (2012b) (Table 1).
70 For reaction specifications for each DNA region used see Majure et al. (2012a ). Sequences were edited either in Sequencher 4.2.2 TM ( Gene Codes, Inc., Ann Arbor, MI USA ) or Geneious Pro TM 5.1 (Bi omatters Ltd., Auckland, NZ) and the alignment was adjusted manually in Se Al v2.0 (Rambaut, 2007). All gaps introduced during alignment were coded as missing data. Combined nuclear and plastid regions were analyzed for all putative diplo id taxa (see Ma jure et al. 2012a ) using maximum likelihood (ML) in RAxML (Stamatakis 2006) carrying out 10000 nonparametric rapid bootstrap pseudoreplicates under 25 rate categories and implementing the GTR model of molecular evolution. Separate plastid and nuclear data sets containing O. cubensis and O. ochrocentra then were analyzed using the same methods. Morphological characters (e.g., cladode shape, flower color, glochid color, growth form, spine color/development pattern) were observed and measurements were taken from herbarium specimens of O. abjecta, O. cubensis, O. militaris O. ochrocentra and O. triacantha and live material of O. abjecta, O. cubensis, and O. ochrocentra As mentioned above, n o live material of O. militaris or O. triacantha was available for study. We also compared O. militaris and O. triacantha to herbarium specimens of a closely related Caribbean species, O. repens (see Majure et al. 2012a ). Results Phylogeny Opuntia abjecta and O. triacantha are resolved in disparate cl ades. Opuntia abjecta is nested in the southeastern United States subclade of the Humifusa clade, and O. triacantha is closely related to the Caribbean and Central American taxa, O. caracassana, O. jamaicensis, and O. guatemalensis of the Nopalea clade (F ig. 4 1; see also Majure et al. 2012a ). Opuntia militaris
71 also is nested within the Nopalea clade and is not closely related to O. triacantha but is resolved as closely related to O. caracassana Opuntia cubensis s.l. is recovered in three places; O. oc hrocentra s.s. (from the Florida Keys) is nested within the Humifusa clade using plastid data. The sample of O. cubensis s.s. is nested in the Nopalea clade using plastid data (Fig. 4 2A). This suggests that the maternal parent of O. ochrocentra is a mem ber of the Humifusa clade and the maternal parent of O. cubensis is a member of the Nopalea clade. Two ITS copy types were discovered for both O. cubensis and O. ochrocentra, after excluding putative recombinants. One ITS haplotype of O. ochrocentra was recovered in the Humifusa clade and another is unresolved in a grade containing members of the Scheerianae clade, which contains one of the putative parents of O. ochrocentra (based on morphology), O. dillenii One ITS haplotype of O. cubensis is re solved within the Nopalea clade, as closely related to O. militaris (i.e., one haplotype is nearly identical to O. militaris ) and the other haplotype is unresolved within a grade containing members of the Scheerianae clade (Fig. 4 2B). Two copy types als o were found in ppc clones of O. ochrocentra which were placed in the Humifusa clade and in a grade of other taxa ( ppc data provide very poor resolution at the clade level), respectively; however, only one copy type of ppc was found from O. cubensis whic h shared synapomorphies only with the Nopalea clade (Fig. 4 2C). Morphology O. abjecta vs. O. triacantha Opuntia abjecta is strikingly different from O. triacantha in growth form, spine color and production, flower bud shape, flower color, and color of a reolar trichomes and glochids. Opuntia abjecta is a small spreading ascending shrub with basally disposed, radiating branches that reach up to 30 cm in height. Opuntia triacantha is a small erect to semi erect shrub generally with a central, semi cylind rical trunk much like that of its close relative O. repens Bello, and reaches heights of up to 40 cm or more. The spines of O. abjecta are strongly
72 retrorsely barbed like those of O. triacantha but they are a lustrous, dark reddish brown during developme nt, instead of dull yellow as in O. triacantha (Fig. 4 3), and they mature to a bright white instead of pale white color. The spines of both taxa become dark gray in age. Up to 3 spines are produced from the areoles of terminal cladodes of O. abjecta, and these a re usually all in the same plane of symmetry (e.g., all spreading, all reflexed, etc.). Up to 6 spines are produced from the areoles of O. triacantha, and they are in two planes of symmetry with the central spine typically divergent (porrect a 4 3), as in the closely related species, O. repens and O. caracassana The spines of O. triacantha are also shorter on average than those of O. abjecta (3.7 cm vs. 4.4 cm). Opuntia abjecta has a rounded flower bud apex, while O. triacantha has an acute flower bud. Opuntia abjecta has completely dark yellow inner tepals, while O. triacantha has sulfur yellow inner tepals that are often tinged pink along the midrib. Tepals are obovate in O. abjecta with a rounded to flat apex with a mucronate tip, and oblong to obovate in O. triacantha with a rounded apex, often without a mucro. The areolar trichomes of O. triacantha, O. militaris and O. repens are yellowish, while the areolar trichomes of O. abjecta ar e white. Opuntia abjecta has stramineous colored glochids on younger cladodes, while O. triacantha has bright yellow glochids on younger cladodes. In general, O. abjecta may be differentiated from O. militaris by the same features as used to distinguish it from O. triacantha because, as indicated in the next section, O. militaris and O. triacantha are morphologically very similar. Morphology O. militaris vs. O. triacantha Opuntia militaris is strikingly similar to O. triacantha although in general O. militaris is smaller than O. triacantha Like O. triacantha O. militaris grows erect with one central trunk eventually producing a small, branching shrub to 30 cm high (Britton and Rose 1920). Flower color of O. militaris and O. triacantha is similar, with both having sulfur yellow inner tepals that
73 may be tinged pink along the midvein and do not have a strong mucronate tip. Flower buds in both species are acute, as in other species of the Nopalea clade. The average cladode length and width of O. mili taris contrasts with O. triacantha (6.2 x 2.8 cm for O. militaris and 7.8 x 3.9 cm for O. triacantha ). Spine lengths and diameters also are smaller in O. militaris, as compared to O. triacantha (2.5 cm long x 0.5 mm in diameter vs. 3.7 cm long x 0.76 mm i n diameter). Opuntia triacantha may have up to 6 spines per areole, and O. militaris may have up to 4 spines per areole, although difference in spine number needs to be explored further in the field, as it can be a highly variable character. Opuntia mili taris also exhibits the porrect spines of O. caracassana, O. repens and O. triacantha Morphology O. cubensis vs. O. ochrocentra Opuntia ochrocentra from the Florida Keys and O. cubensis from Cuba share morphological features suggesting that O. dillenii could be one of the parents of both. This similarity likely led Britton and Rose (1920) to include these taxa in the same series as O. dillenii i.e., Opuntia series Dillenianae In both taxa the spines are produced in a star pattern from the areoles (Fig. 4 3); they also produce radial spines that are basally flattened, as in O. dillenii Most developing radial spines of O. ochrocentra are lustrous yellow as in O. dillenii bu t central spines are produced that are mottled to banded red brown, as in the developing spines of O. abjecta Although the spines of O. cubensis are produced from the areole as in O. dillenii (i.e., in a star pattern), the young developing spines are dul l yellow to creamy white as in O. militaris and O. triacantha Cladodes of O. ochrocentra are on average larger than those of O. cubensis (15.6 x 7.5 cm vs. 12.3 x 4.8 cm) and produce longer central spines (5.3 vs. 3.1 cm long). Average spine diameters are nearly the same for both taxa (1.05 vs. 1.01 mm). The central spines of both O. ochrocentra and O. cubensis are generally round in cross section and may or may not be twisted at the base, as in O. abjecta and O. militaris Mature spines of O.
74 ochroce ntra turn dark gray in age and become strongly deflexed, while mature spines of O. cubensis apparently turn light brown in age and do not deflex. Both species also have eas ily disarticulating cladodes, like those of their putative parental species, O. abj ecta and O. militaris Below we provide a key to distinguish O. abjecta, O. militaris O. triacantha, O. cubensis, and O. ochrocentra We also include the widespread species O. repens a close relative of and morphologically similar to O. triacantha as s hared characters of those two taxa often lead to misidentifications. This artificial key is based on both live material and herbarium specimens Key to the Species 1. Spines disposed from the areoles in a star like pattern, radiating in all directions, radial spines strongly flattened dorsiventrally central spines round in cross section. 1. Spines disposed from the areoles in the same plain, or with some spin es porrect, spines (including radials and centrals, if both present) round in cross section or merely twisted at 2. Developing spines dull yellow to cream or dull light brown in color, spines stou t to 3.1 (2.2 4.2) cm long, with one central spin e (round in cross section) O. cubensis 2. Developing spines bright, lustrous yellow to orange red or mottled yellow and reddish brown banded in color, spines delicate to 5. 3 (4.7 5.8) cm long, with multiple central spines (round or twisted in cross section) O. ochrocentra 3. Mature plants cespitose (with numerous branches arising from the base), inner tepals entirely yellow, developing spines d ark red brown to mottled red brown and white, spines mostly spread ing from the areoles in one pla n e cladodes rotund to obovate in outl .................... O. abjecta
75 3. Mature p lants with solitary stems (although these may form dense patches from the disarticulation of terminal cladodes), inner tepals yellow to yellow green, often tinged with pink along the tepal midvein, developing spines dull to lustrous yellow or creamy white, usually spreading with one to two large porrect spines and 1 to numerous deflexed obovate, glochids yellow. .......................................................... .............. ........................... 4 4. Cladodes sub cylindrical to flat, narrow, on average 5 cm long, 1.8 cm wide, developing spines lustrous yellow, spines flexible (delicate), on average 0.44 mm in diameter, 3.3 cm lo O. repens 4. Cladodes flat, wider, on average 6.2 7.8 cm long, 2.8 3.8 cm wide, developing spines dull yellow, spines stout, on average 0.5 0.76 mm in diameter, 2.5 3.7 cm long, ..... ............. .... 5 5. Cladodes 7.8 (5 10.9) cm long, 3.8 (2.4 5.8) wide, spines 3.7 (2.3 6.2) cm long O. triacantha 5. Cladodes 6.2 (4.7 8.5) cm long, 2.8 (2.3 3.5) wide, spines 2.5 (1.4 3.1) cm long and .............. ...... ... ....... O. militaris Discussion Opuntia abjecta vs. O. triacantha Chromosome counts reveal that the type population of O. abjecta is diploid, while other material from the Florida Keys is te traploid (Majure et al. 2012 b ). Also, material cultivated at ( LCM 3318 ; Majure et al. 2012b ), suggesting that another population of O. abjecta may exist somewhere i n the lower keys. The population of O. abjecta on Long Key is morphologically identical to that of other tetraploid material and thus is most likely tetraploid as well. Opuntia
76 ochrocentra has been recorded as pentaploid (2 n = 55) from three accessions ( Majure et al. 2012b ), and O. cubensis was tetraploid (2 n = 44) from one count ( Areces s.n. ) made by Majure et al. (in review). Spencer (1955) reported a diploid count for O. triacantha from Puerto Rico; however, we have not been able to confirm this count No chromosome counts are available for O. militaris Benson (1982) likely placed O. abjecta in synonymy with O. triacantha because these taxa share several morphological features, such as disarticulating cladodes, and terminal cladodes that often exhi bit 2 3 spines per areole. Spines of both species overlap in length and diameter, and cladode shapes and sizes slightly overlap, as does the height of both species. Opuntia abjecta is only found in the Florida Keys and was considered merely a northern ex tension of the Caribbean O. triacantha (Benson 1982 ; Pinkava 2003). This southern Florida/Caribbean dis junction is very common for a wide variety of taxa (Wunderlin and Hansen 2003). In Cactaceae alone, Acanthocereus, Consolea, Harrisia, and Pilosocereus are shared with neighboring Caribbean Islands (Acevedo Rodriguez 1996). Opuntia triacantha is and Rose 1920). Opuntia abjecta likewise is found on limestone outcrops (Key Largo Limestone) within 0.5 km or less o f the ocean ( Benson 1982; Majure pers. obsv.). Additionally, the misidentification of the interspecific hybrid presumably involving O. triacantha O. ochrocentra (as O. cubensis ), added further evidence for the northern disjunct distribution of O. triacantha in the Florida Keys (Benson 1982). Coincidentally, O. austrina Small, another endemic species to Florida is much more similar morphologically to true O. triacantha than is O. abjecta Opuntia austrina forms treelets (i.e., the ammophila enti ty) to large shrubs and generally is characterized by a single, cylindrical stem, which may be copiously spiny as in O. triacantha (Fig. 4 3). Opuntia austrina also is a
77 member of the Humifusa clade, so these morphologically similar characters are merely c onvergent between O. austrina and O. triacantha Opuntia austrina is morphologically similar to O. abjecta as well, having similar spine characters, glochid and flower colors, and cladode shapes. Consequently, Benson (1982) also misidentified some materi al of O. abjecta from the Florida Keys as O. austrina Although Anderson (2001) included O. abjecta in synonymy with O. triacantha, as mentioned above, he did not include the Florida Keys within the distribution of the species, although his photo of O. tri acantha is actually of O. abjecta from the Florida Keys! O. militaris vs. O. triacantha Considering the limited data here and poor resolution in the diploid phylogeny (Fig. 4 1), it is still premature to determine whether or not O. militaris and O. triacan tha should be considered the same species. Opuntia militaris shares numerous morphological features with O. triacantha although it is generally less robust and has fewer spines, characters that may be influenced as a result of different environmental con straints across the distribution of the two taxa. However, even with the limited data presented here, it is obvious that O. triacantha and O. militaris are not genetically identical (e.g., O. militaris is more closely related to O. caracassana in our diploid phylogeny; Fig. 4 1). Thus, our phylogenetic data suggest that O. triacantha and O militaris represent distinct lineages Opuntia militaris is also disjunct from the nearest population of O. triacantha on Desecheo Island, Puerto Rico, by ca. 765 km. It will be necessary to study morphological characters and ploidal levels of living material of O. militaris, O. triacantha and other closely related species within the Greater and Lesser Antilles (e.g., O. caracassana, O. jamaicensis, O. re pens, O. taylori Britton and Rose) to determine species limits within this group. However, O. militaris is tentatively c onsidered specifically distinct and is included in the above key.
78 The Opuntia cubensis Enigma Benson (1982) considered O. ochrocentra S mall described from Big Pine Key to be a synonym of O. cubensis described from Guant namo, Cuba (Britton and Rose 1912). Opuntia cubensis has generally been considered a hybrid between O. dillenii and O. militaris (Britton and Rose 1920), and molecular da ta support the hybrid origin of O. cubensis from the Florida Keys (i.e., O. ochrocentra ) between a member of the Humifusa clade and O. dillenii (Majure et al. 2012a ; see Fig. 4 2 A B in this study). However, this study suggests that O. militaris is not conspecific with O. abjecta and may be more closely related to, although likely not conspecific with O. triacantha Therefore, O. cubensis in Cuba is derived from different putative progenitors, O. militaris and O. dillenii than that of the Flor ida Keys material, which is derived from a member of the Humifusa clade ( O. abjecta ) and a member of the Scheerianae clade ( O. dillenii ). Thus O. cubensis ified to by Majure et al. (2012a ) should be referred to as O. oc hrocentra, given that O. cubensis and O. ochrocentra are not synonymous as shown here. Characters of O. abjecta and O. militaris O. cubensis mistaken for any of those putative progenitors, because those characters diffe rentiating O. cubensis from O. dillenii are spine color and shape, the smaller growth form, and cladode disarticulation, all characters shared to some degree by the other putative maternal progenitors ( O. abjecta and O. militaris ). Identifying these speci es is made more difficult when using only herbarium material, as most identifying characters of these stem succulents, other than spine characters, are mostly lost during the drying process ( Reyes Agero 2007). Opuntia cubensis and O. ochrocentra, however are morphologically separable. Opuntia ochrocentra shares the mottled yellow to reddish brown colored young spines of O. abjecta and O. cubensis has dull yellow young spines, as does O. triacantha and O. militaris The spine
79 pattern s of O. ochrocentra and O. cubensis are slightly different, with O. cubensis always having one strong, porrect, cylindrical central spine and O. ochrocentra with several weaker, cylindrical or basally twisted central spines. The spines in O. ochrocentra deflex and become ap pressed along the stem in age, a character apparently not demonstrated by O. cubensis Opuntia cubensis generally has shorter spines and smaller cladodes than O. ochrocentra, as mentioned above. The cladodes of O. cubensis are typically elliptical in outline, while cladodes of O. ochrocentra may be either elliptical or obovate Summary The true identities of the Floridian species O. abjecta and O. ochrocentra were long obscured as a result of incorrect assumptions made reg arding phytogeographic relationships of Opuntia from the Florida Keys and the Caribbean region (Benson 1982). Opuntia abjecta and O. triacantha are distinct species morphologically and phylogenetically. Thus, material from Florida should not be referred to as O. triacantha but rather represents a species endemic to the state, which should be recognized as O. abjecta Opuntia cubensis is a Cuban taxon that does not occur in Florida and which originated via hybridization between O. militaris of the Nopale a clade and likely O. dillenii of the Scheerianae clade, as suggested by Britton and Rose (1920). O. cubensis O. ochrocentra which is of hybrid origin most likely between O. abjecta of the Hu mifusa clade and O. dillenii More research will be necessary to determine whether or not O. militaris is distinct from O. triacantha but given the limited morphological and phylogenetic data presented here, I suggest that O. militaris should be regarded as a separate species.
80 Figure 4 1. Putative diploid ML phylogeny (most likely topology) of Opuntia s.s. using South American species ( O. macbridei, O. retrorsa ) as outgroups. It is well supported that O. triacantha is in a different clade (i.e., the Nopalea clade) than O. abjecta ( Humifusa clade). Opuntia militaris although nested within the Nopalea clade, is not resolved as sister to O. triacantha with which it is currently placed in synonymy. Opuntia abjecta O. militaris and O. triacantha are denoted by asterisks. Bootstrap values are indicated above branches.
81 Figure 4 2. Most likely topologies (from RAxML) from ITS (A), ppc (B), and (C) plastid phylogenies including the putative hybrid taxa O. cubensis and O. ochrocentra (indicated with asterisks). ITS clones of O. ochrocentra are resolved in the Humifusa clade and unresolved with O. dillenii and O. ellisiana of the Scheerianae clade, while ITS clones of O. cubensis are resolved in the Nopalea clade, as well as unresolved with O. dilleni i and O. ellisiana ppc clones of O. ochrocentra are also resolved with the Humifusa clade and unresolved with members of the Scheerianae clade, while only one copy type of ppc was discovered for O. cubensis that is unresolved with a member of the Nopalea clade. Opuntia ochrocentra is resolved in the Humifusa clade, and O. cubensis is resolved in the Nopalea clade in the plastid phylogeny. D) represents the putative hybrid scenario, where O. ochrocentra was derived from O. abjecta (maternal lineage) and O. dillenii (paternal lineage), while O. cubensis was derived from O. militaris (maternal lineage) and O. dillenii (paternal lineage). Bootstrap values are given above branches in A C.
82 Figure 4 3. Morphological characters of O. abjecta, O. triacantha, O. cubensis, O. ochrocentra, O. dillenii and O. repens A) O. abjecta ( LCM 3908 ) showing reddish brown developing spines, white mature spines and stramineous glochids, B) O. triacantha ( P Duss 3071 ) showing solitary, mostly erect trunk C) O. triacantha ( A.C. Smith 10442 ) showing dull yellow developing spines, pale white mature spines, and yellow glochids, D E) O. ochrocentra ( LCM 3907 ) showing one and two year old cladodes with the younger cladodes showing yellow spines as in O. dillenii and the older c ladodes showing spines turning white in age; in E) young spines of ca. 6 mo. in age showing mottled color (banding) typical of O. abjecta F G) O. cubensis ( Areces s.n. ) showing yellow glochids and young developing spines that are pale yellow initially agi ng to white; both O. ochrocentra and O. cubensis have spine growth patterns similar to O. dillenii H) O. dillenii (Buckaneer State Park, FL), and I) O. repens ( LCM 3839 ), a close relative of and morphologically similar to O. triacantha showing porrect cen tral spines, as described above for certain Caribbean members of the Nopalea clade.
83 CHAPTER 5 CYTOGEOGRAPHY OF THE Humifusa CLADE OF Opuntia S.S. MILL. 1754 (CACTACEAE, OPUNTIOI DEAE, OPUNTIEAE): CO RRELATIONS WITH PLEI STOCENE REFUGIA AND MORPHOLO GICAL TRAITS IN A PO LYPLOID COMPLEX Background Ploidy has a long tradition of utility for illuminating species boundaries, hybrid zones, and interspecific relatio nships among plants (e.g., Stace 2000). Knowing the ploidal levels of taxa used in phylogenetic analyses can also aid in detecting potential hybridization events through incongruence in reconstructions using biparentally inherited n uclear loci (Ionta et al 2007; Soltis et al. 2008). Researchers have frequently used cytological data to help understand species evolution and delimitations in the nopales or prickly pear cacti, i.e., the genus Opuntia (Pinkava and McLeod 1971; Pinkava et al. 1973, 1977, 1985; W eedin and Powell 1978; Pinkava and Parfitt 1982; Weedin et al. 1989; Pinkava et al. 1992; Powell and Weedin 2001, 2004). Subfamily Opuntioideae ( Opuntia s.l., as previously recognized; Benson 1982) is known to have the highest number of polyploids in Cac ta ceae (Cota and Philbrick 1994; Pinkava 2002), and Opuntia s.s. is well known for interspecific hybridization (e.g., Grant and Grant 1982; Griffith 2003) and subsequent g enome duplication (Pinkava 2002; L.C. Majure (LCM), R. Puente (RP), P. Griffith (PG), W .S. Judd (WSJ), P.S. Soltis (PSS), D.E. Soltis (DES) unpubl. data). The significance of polyploidy in plant evolution and speciation has long been recognized (Stebbins 1940, 1950, 1971; Swanson 1957; DeWet 1971; Harlan and DeWet 1975 ; Grant 1981 ; Leitch an d Bennett 1997 ; Ramsey and Schemske 1998 ; Adams and Wendel 2005 ; Tate et al. 2005 ; Doyle et al. 2008 ; Soltis and Soltis 2009 ; Jiao et al. 2011). As stated by Stebbins (1950), p. Reprinted with permission from the authors. Original publication: Majure, L.C., D.E. Soltis, P.S. Soltis, and W.S. Judd. 2012b. Cytogeography of the Humifusa clade of Opuntia s.s. (Cactaceae: Opuntioideae): Correlations with geographic distributions and morphological differentiation of a polyploid complex. Comparative Cytogenetics 6: 53 77.
84 radically different, but nevertheless vigorous and well unequivocal means of true sympatric speciation (Futuyma 1998 ; Otto and Whitton 2000) and is considered to be common in plants (Stebbins 194 0 ; DeWet 1971 ; Ramsey and Schemske 1998 ; Tate et al. 2005). For example, virtually all major clades of angiosperms have undergone one or more episodes of genome duplication (Soltis and Soltis 2009). Likewise, polyploidy is very important throughout Cactace ae (Pinkava 2002), and within Opuntia s.s., polyploids previously have been recorded in Opuntia humifusa (Raf.) Raf., 1820, and relatives (Bowden 1945a, b ; Pinkava et al. 1985 ; Powell and Weedin 2004 ; Baker et al. 2009a, b ; Majure and Ribbens in press) of the Humifusa clade (sensu LCM, RP, PG, WSJ, PSS, DES unpubl. data). There are currently six species recognized in the Humifusa clade, O. abjecta Small, 1923, O. humifusa O. macrorhiza Engelm., 1850, O. pottsii Salm Dyck, 1849, O. pusilla (Haw.) Haw., 1812 and O. tortispina Engelm. & J.M. Bigelow, 1856 (Pinkava 2003; LCM unpubl. data) The Humifusa clade is distributed widely from the western U.S. and northern Mexico (represented by O. macrorhiza s.l. O. pottsii and O. tortispina ) and throughout the east ern U.S. including the upper Midwest (e.g., Michigan, Minnesota, Wisconsi n) and southern Ontario (Benson 1982; represented by O. abjecta O. humifusa s.l. O. macrorhiza s.l. and O. pusilla ). Opuntia humifusa s.l. is composed of numerous morphological ent ities that have been recognized in certain taxonomic treatments as different species (see Small 1933). Throughout its range, O. humifusa s.l. has been divided into as many as 14 taxa (Britton and Rose 1920 ; Small 1933 ; Benson 1982 ; Majure and Ervin 2008). Thus, O. humifusa s.l. is occasionally referred to as a species complex (Doyle 1990). Currently, two taxa are recognized in O. humifusa s.l. ( O. humifusa var. ammophila (Small) L.D. Benson and O. humifusa var. humifusa ; Pinkava 200 3).
85 Likewise, Opuntia macrorhiza has been divided into as many as 11 taxa (see Benson 1982). Opuntia macrorhiza was previously considered a variety of O. humifusa (see Benson 1962; see Table 5 1 for synonyms of O. humifusa s.l. and O. macrorhiza s.l. sampl ed in this study), O. pottsii was considered a variety of O. macrorhiza and O. tortispina was placed in synonymy with O. macrorhiza (Benson 1982). Opuntia pusilla has been divided into several species: O. drummondii Graham, 1841 O. frustulenta Gibbes, 18 58, O. impedita Small, 1923 O. pes corvi LeConte, 1857 and O. tracyi Britton, 1911 (Britton and Rose 1920 ; Small 1933); however, Benson (1982) placed them in synonymy under the name O. pusilla Opuntia triacantha (Willd.) Sweet, 1826, also has been divid ed into several species, i.e., O. abjecta of the Florida Keys, O. militaris Britton & Rose, 1919, of Cuba, and O. triacantha from different parts of the Greater and Lesser Antilles (Britton and Rose 1920), but all of these have since been placed in synonym y within O. triacantha (Benson 1982). Phylogenetic and morphological studies have indicated that O. abjecta is not even in the same clade as O. triacantha (LCM, WSJ unpubl. data) and so here is treated as O. abjecta Contributing to the confusing taxonomic history of this clade is the high degree of morphological variation exhibited by most taxa, the lack of complete sampling throughout the range of the clade, the absence of cytological and phylogenetic evidence, reliance on poorly prepared and sparse herba rium collections (Majure and Ervin 2008 ; LCM unpubl. data), and hybridization and polyploidy (Benson 1982 ; Rebman and Pinkava 2001). Careful examination of morphological characters across the geographic range of the widely distributed O. humifusa s.l. and O. macrorhiza s.l. reinforces the hypothesis that hybridization may have preceded the origin of geographical morphotypes, because morphological characters displayed by certain taxa appear to be introgessive between O. humifusa s.l. and O. macrorhiza s.l. ( Table 5 2). For instance, O.
86 cespitosa Raf., 1830, from the eastern U.S. and recently recognized by Majure and Ervin (2008), has yellow tepals that are basally tinged crimson to orange red, a characteristic typical of O. macrorhiza and occasionally O. tor tispina from western North America (Benson 1982 ; Pinkava 2003 ; Powell and Weedin 2004), but the spine characters of O. cespitosa are typical of O. humifusa s.l (see Majure and Ervin 2008). Although chromosome counts have been reported for many of the Opun tia taxa from the southwestern U.S. and other areas (Stockwell 1935 ; Spencer 1955 ; Pinkava and McLeod 1971 ; Pinkava et al. 1973, 1977; Weedin and Powell 1978 ; Pinkava and Parfitt 1982 ; Pinkava et al. 1985 ; Weedin et al. 1989 ; Pinkava et al. 1992 ; Powell an d Weedin 2001 ; Pinkava 2002 ; Negr n Ortiz 2007 ; Segura et al. 2007 ; Baker et al. 2009a, b), few chromosome counts have been reported for taxa of Opuntia in the eastern and midwestern U.S. (Majure and Ribbens in press), and most of those taxa belong to the Humifusa clade. Bowden (1945a, b), Hanks and Fairbrothers (1969), Doyle (1990), and Baker et al. (2009 a, b) have all made counts of members of the Humifusa clade from the eastern U.S. Bowden (1945a, b), Doyle (1990), and Baker et al. (2009a) recorded dipl oid (2 n = 22) and tetraploid (2 n = 44) material of O. humifusa from the eastern U.S., and Bowden (1945a) recorded tetraploid (2 n = 44) material of O. impedita (currently syn. of O. pusilla ). Hanks and Fairbrothers (1969) recorded an aneuploid number for O. humifusa (2 n = 17, 19) likely in error, since aneuploids are very rare in Cactaceae (Pinkava 2002). Majure and Ribbens (in press) recorded tetraploids of O. humifusa s.l. and O. macrorhiza s.l. from the Midwest, suggesting that the northernmost population s of those taxa are polyploid. Opuntia macrorhiza, O. pottsii, and O. tortispina have all been counted extensively in the southwestern U.S. (Pinkava and McLeod 1971; Pinkava et al. 1973 ; Pinkava et al. 1977 ; Pinkava et al. 1992 ; Pinkava et al. 1998 ; Powell and Weedin 2001 ; Powell and Weedin 2004), where O.
87 macrorhiza and O. pottsii have been recorded exclusively as tetraploids, and O. tortispina has been recorded as either tetra or hexaploid. Chromosome counts reported for species in the Humifusa cl ade do not encompass all of the taxa within the range of the clade nor the wide distributions exhibited by several of the more common species. To further our understanding of species complexes and the evolution of polyploids within those complexes, cytolog ical data are needed from the entire distribution of a given species (Babcock and Stebbins 1938 ; Stebbins 1942 ; Stebbins 1950). Thus, an in depth study of the distribution of cytotypes and correlations between cytotypes and morphology is desperately needed in order to aid in the delimitation of potentially unrecognized and cryptic species and to elucidate relationships in the Humifusa clade. Here we present chromosome counts for all taxa considered to be part of the O. humifusa complex and all taxa of the H umifusa clade (LCM, WSJ, PSS, DES, unpubl. data) and provide counts throughout most of the known ranges of all taxa to determine the geographic structure of ploidy and differences in ploidy among morphologically distinct taxa. We also reconstruct a phyloge ny of diploid and polyploid members of the Humifusa clade based on nrITS data to investigate the relationship between geographic distribution and evolutionary relationships. We provide counts for another common species in the southeastern U.S., O. stricta (Haw.) Haw., 1812, because it has been hypothesized to hybridize with members of the Humifusa clade (Benson 1982). In addition, ploidy of the putative hybrid between O. abjecta and O. stricta i.e., O. ochrocentra Small, 1923, was analyzed. Ploidy determin ations of the Humifusa clade, coupled with morphological character analysis and further molecular phylogenetics, will aid in the delimitation of species in the group and in determining the origin and evolutionary significance of polyploidy in this clade.
88 M aterial and Methods Chromosome Counts Methods follow those of Majure and Ribbens (in press). Briefly, root tips were collected from early morning throughout early afternoon and placed in 2mM 8 hydroxyquinoline (Solti s 1980) for up to 8 hours at 4 C or in N 2 O (Kato 1999) for 1 hour and then fixed in a 3:1 solution of absolute ethanol: glacial acetic acid for 2 to 24 hours. Root tips then were placed in 70% ethanol for at least 2 hours and digested in 40% HCl for 5 10 minutes (depending on the size of the roo t) at room temperature. Squashes were performed in 60% acetic acid and stained with 1% aceto orcein dye and viewed on a Zeiss Photomicroscope III (Carl Zeiss, Oberkochen, Germany). To confirm each count, at least three to five metaphase cells were counted per specimen. These multiple counts per sample alleviated concerns regarding endomitosis, which has been reported in the allopolyploid (4 x ), Opuntia spinosibacca M.S. Anthony, 1956, (Weedin and Powell 1978), tetraploid O. pusilla (Bowden 1945b), as well as in many other angiosperms (e.g., Barrow and Meister 2003, Tate et al. 2009, I. Jordan Thaden, pers. comm.). We counted chromosomes of 277 individuals of the Humifusa clade, 14 individuals of O. stricta s.l., three samples of the putative hybrid O. ochroce ntra and two individuals of the putative hybrid O. alta Griffiths, 1910. Generally, only one accession per population was counted. Taxonomy Taxa used for ploidy a nalysis are listed in Appendix D Species delimitations within O. humifusa s.l. and O. macror hiza s.l. are problematic, so we recognize both O. humifusa and O. macrorhiza as broadly circumscribed (Table 5 1). Thus, we have arranged our counts of plants within t hese two species (see Appendix D ) according to their various segregates to determine whether the morphological variation of these segregate entities (Table 5 2) is correlated with cytotype and/or geographical and phylogenetic patterns.
89 Cytogeographic Analysis We mapped the localities for all of the individuals for which we determined ploid y (277 in number) and incorporated previous counts (n = 41) (Bowden 1945a ; Pinkava and McLeod 1971 ; Pinkava et al. 1973 ; Weedin and Powell 1978 ; Pinkava and Parfitt 1982 ; Pinkava et al. 1985 ; Weedin et al. 1989 ; Doyle 1990 ; Pinkava et al. 1992 ; Pinkava et al. 1998 ; Powell and Weedin 2001 ; Baker et al. 2009a, b ; Majure and Ribbens in press) to cover the majority of the geographic distribution of each taxon. This allowed us to explore the geographic boundaries of the different ploidal levels encountered in th is clade and construct hypotheses regarding polyploid formation and speciation. Phylogenetic Analysis We generated sequences from the nuclear ribosomal internal transcribed spacer (nrITS: White et al. 1990) for a sample of diploid (n = 6) and polyploid taxa (n = 8) of the Humifusa clade from the eastern and western U.S. (Table 5 3). Opuntia basilaris Engel m. & J.M. Bigelow, 1856, was used as an outgroup based on previous analyses of Opuntia (LCM unpubl. data). A phylogenetic analysis of these data was carried out to determine whether the geographic distribution of ploidy (as determined here) was correlated with the evolutionary history of the clade. We carried out a Maximum Likelihood analysis using RAxML (Stamatakis 2006) running evolution. Results The base chromosome number for Cactaceae has been well established as x = 11 (Remski 1954 ; Pinkava and McLeod 1971 ; Lewis 1980 ; Pinkava et al. 1985 ; Pinkava 2002), and we saw no deviation from this in our counts (Appendix D). Out of 318 counts of the Humifusa clade, includin g 41 from the literature, 210 (66%) were polyploid and 108 (34%) were diploid. Diploid
90 (2 n = 2 x = 22) and tetraploid (2 n = 4 x = 44) O. humifusa s.l. and O. macrorhiza s.l. were discovered (Fig. 5 1A D, I J, L). Diploid O. humifusa s.l. is restricted entire ly to the southeastern U.S., whereas diploid O. macrorhiza s.l. is restricted entirely to the southwestern U.S. (eastern Texas (see Appendix D) and southeastern New Mexico (M. Baker and D.J. Pinkava pers. comm.)). Tetraploid members of O. humifusa s.l. and O. macrorhiza s.l. are much more widely distributed throughout the U.S. than are their diploid relatives (Fig. 5 2). Tetraploids of O. humifusa s.l. are found from Massachusetts south to the southeastern U.S. where they abut the distribution of di ploid taxa and throughout the eastern and midwestern U.S. Tetraploid O. macrorhiza s.l. is distributed throughout parts of the Great Plains through the midwestern U.S., most of the southwestern U.S., parts of the Rocky Mountains, and the upper Sierra Madre Occidental in Sonora, Mexico (Fig. 5 2). Diploid, triploid, and tetraploid populations of O. pusilla were discovered (Fig. 5 1E G) throughout its restricted range in the southeastern U.S. (Fig. 5 3). Interestingly, with the exception of two populations, p olyploid individuals (3 x and 4 x ) were mostly confined to the coastline, although diploid populations were much more widespread throughout the interior part of the distribution of the species (Fig. 5 3). Of the three examples of O. abjecta sampled from the Florida Keys, one was diploid (Fig. 5 1H), and two were tetraploid. Opuntia tortispina (southwestern U.S.) was hexaploid in six and tetraploid in one of the populations examined (see Fig. 5 2 for hexaploid distribution). Individuals of O. stricta sampled f rom the southeastern U.S. were all hexaploid. Samples included members of the taxa considered by some (Anderson 2001) to be O. dillenii (Ker Gawl.) Haw., 1819, and O. stricta Three individuals of the putative hybrid O. ochrocentra from two
91 localities in t he Florida Keys were pentaploid (Fig. 5 1K), and the putative hybrid O. alta was hexaploid. Maximum likelihood analysis of ITS data reveals that the Humifusa clade is made up of two well supported subclades. One is restricted to the southeastern U.S. and includes polyploid members of O. pusilla and O. abjecta and the other includes southwestern diploid O. macrorhiza and all other polyploids pertaining to O. humifusa s.l., O. macrorhiza s.l., and O. tortispina There is no further resolution within the tre e at the species level using ITS (Fig. 5 4). Species relationships within these two clades are further resolved with the addition of other loci (LCM unpubl. data), however, that is beyond the scope of this study. Discussion Opuntia macrorhiza has only been recorded previously as tetraploid (Pinkava et al. 1971 ; 1973, 1977, 1992, 1998; Powell and Weedin 2001 ; 2004; Pinkava 2003). These are the first reports of diploid O. macrorhiza and likely represent descendants of those progenitors from which tetraploid O macrorhiza s.l. and other polyploids arose. Likewise, this is the first report of diploid and triploid O. pusilla which was formerly known only from tetraploid counts (Bowden 1945a). Diploid members of O. humifusa s.l. (e.g., represented by the segregat e taxa O. ammophila Small, 1919 O. austrina Small, 1903 O. lata Small, 1919, in this study; see also Appendix D) exhibit high levels of morphological variability but each is diagnosable morphologically, which suggests that these segregate taxa may need t o be recognized at the species level. Likewise, diploid material of O. macrorhiza s.l. from eastern Texas (e.g., O. xanthoglochia Griffiths, 1910, in this study; see also Appendix D) and southeastern New Mexico is morphologically distinct from tetraploid m aterial of O. macrorhiza s.l., which may also justify the recognition of O. xanthoglochia and O. macrorhiza as separate species.
92 Our hexaploid counts of O. stricta are consistent with those of P inkava et al. (1992) and Negr n Ortiz (2007). In contrast, Spencer (1955) reported O. stricta from Puerto Rico to be m ore recent counts (e.g., Negr n Ortiz 2007 for Consolea Lem., 1862). Ou r three pentaploid counts of O. ochrocentra support the proposed hybrid origin of this species between hexaploid O. stricta (2 n = 66) and diploid O. abjecta (2 n = 22) through unreduced gametes of O. abjecta Opuntia ochrocentra also exhibits intermediate m orphological characters (e.g., growth form, spine characters) that further supportits hybrid origin (LCM unpubl. data). Diploid Refugia and Polyploidy Formation Polyploidy is very common within the Humifusa clade, occurring in 66% of the samples reported h ere. Most researchers that have studied Opuntia cytologically have found polyploid taxa (e.g., Bowden 1945a ; Weedin and Powell 1978 ; Pinkava et al. 1985 ; Doyle 1990 ; Segura et al. 2007 ; Baker et al. 2009a, b ; Majure and Ribbens in press ; but see Spencer 19 55). All diploids in our analysis were restricted to either the southeastern or southwestern (eastern Texas and southeastern New Mexico) U.S., and the polyploid individuals were found nearly everywhere in Humifusa clade and in other studies between the southeastern U.S. and the southwestern U.S. is thought to have occurred as a result of the disruption of a semi arid zone along the Gulf Coast region during the mid Pleistocene (Webb 1990; Althoff and Pellmyr 2002). These two areas likely served as glacial refugia for a variety of animals and plants (e.g., Remington 1968 ; Davis and Shaw 2001 ; Al ; Althoff and Pellmyr 2002 ; Soltis et al. 2006 ; Waltari et al. 2007 ; Whittemore and Olsen 2011) and may have promoted current species richness and genetic diversity in southern populations (Hewitt 2000). Specifically, Swenson and Howard (2005) identified southeastern Texas and northern Florida as Pleistocene refugia for
93 animal and plant species. Species from these regions subsequently came into contact following the last glacial maximum and formed hybrid zones at contact areas expand ing out from these refugia. Swenson and Howard (2005) also hypoth these propose d diploid refugia (e.g., Fig. 1, G & H in Swenson and Howard 2005). Those post glacial routes and diploid contact zones are consistent with the current distributions of polyploid taxa within O. hum ifusa s.l. and O. macrorhiza s.l. The restricted diploid and widespread polyploid distribution pattern has been recorded in many other plants and is a common pattern seen in polyploidy complexes (Babcock and Stebbins 1938 ; Stebbins 1950, 1971 ; DeWet 1971 ; Lewis 1980 ; Grant 1981 ; Parfitt 1991). The seemingly disjunct southeastern New Mexico diploid population of O. macrorhiza s.l. may represent a mere extension of the eastern Texas diploid refugium, which has since been mostly replaced by polyploid taxa. Alt ernatively, a diploid extension may still exist but was not detected due to the lack of cytological data for populations from east Texas to southeastern New Mexico (Fig. 5 2). Diploid taxa of other clades (e.g., O. polyacantha Haw. var. arenaria (Engelm.) Parfitt, 1819) are coincidentally found near the same region (Pinkava 2002, 2003), however, suggesting that a third diploid refugium, i.e., in southeastern New Mexico western Texas, may need to be recognized. Pinkava (2003) suggested that an O. humifusa O. macrorhiza O. pottsii complex originated along the east coast of the U.S. and spread westward to Arizona, where it came into contact and hybridized with O. polyacantha and formed the mostly hexaploid O. tortispina From our data, this scenar io is plausible in that O. tortispina has morphological characters representative of both O. polyacantha and O. macrorhiza and is found where populations of diploid and tetraploid O. macrorhiza s.l. and diploid O. polyacantha come into contact.
94 However, co nsidering the two diploid refugia suggested by our analyses and what is known about the historical biogeography of the southeastern U.S. (e.g., Webb 1990), it is likely that the Humifusa clade originated in the southwestern U.S. and adjacent northern Mexic o, then dispersed eastward into the southeastern U.S. The arid habitat along the coast of the Gulf of Mexico during the mid Pliocene to early Pleistocene would have been interrupted during the mid Pleistocene, creating the dis jct. and promoting the genetic divergence among diploid populations we see today (Fig. 5 4). Taxa from these two diploid refugia would have come back into contact and formed the widely successful polyploids of the Midwest and eastern U.S. (Fig. 5 5). This scenario is further corroborat ed by phylogenetic analyses, where eastern U.S. polyploids of O. humifusa s.l. are resolved in a clade with the southwestern diploid O. macrorhiza (Fig. 5 4). The lower frequency of diploids encountered in western populations of the Humifusa clade also sug gest that those diploid populations may be older (see Stebbins 1971, p. 157) than those of the southeastern U.S.; however, this could merely be a bias resulting from more limited sampling of western populations. The various morphotypes of tetraploid O. mac rorhiza in the western U.S. likely arose from southwestern diploid populations but subsequently spread in all directions after formation. Tetraploid O. macrorhiza appears to have arisen numerous times, given that several morphotypes exist throughout its ra nge. However, only two diploid mor photypes are known to exist (eastern Texas and southeastern New Mexico), suggesting that other ancestral diploids may have since gone extinct or have not yet been found, or that polyploid taxa exhibiting unique, derived c haracters were partly responsible for the origin of certain morphotypes, which have no diploid counterparts. Stebbins (1971) suggested that there are several degrees of maturation of polyploidy complex formation (i.e., initial, young, mature, declining, re lictual), which may be
95 deduced by comparing the relative geographic distribution of polyploids versus diploids. By these criteria, Opuntia humifusa s.l. and O. macrorhiza s.l. may represent a mature polyploid complex. The diploid taxa are less common than polyploids and are largely restricted in distribution, whereas the polyploid taxa are much more widespread. Stebbins (1971) also proposed that mature polyploid complexes are relatively young, derived during the Plio or Pleistocene epochs. This scenario w ould place polyploid formation in the Humifusa clade at the same time as Pleistocene megafauna. Thus, frequent environmental disturbances associated with glacial and interglacial cycles could have mediated the repeated contact of divergent diploid taxa lea ding to polyploid formation. Migrating herbivores would have then dispersed those polyploidy products over large geographic areas (Jansen 1986). Divergence time estimation of the Humifusa clade places the origin of the clade in the late Pliocene to early P leistocene (LCM, RP, PG, WSJ, PSS, DES unpubl. data), in agreement with this scenario. The occurrence of only polyploid individuals in previously glaciated areas of the U.S. provides further evidence for their subsequent spread into those available niches following the last glacial maximum. Many polyploid populations of O. humifusa s.l. and O. macrorhiza s.l., especially in the eastern U.S., are largely isolated from one another and from diploid populations, suggesting that polyploid formation is not ongoin g, at least on such a large scale as during the Pleistocene or immediately after the last glacial maximum. In contrast, polyploids in O. pusilla are mostly sympatric with diploids in the Gulf of Mexico region and are represented by triploids and tetraploid s. Polyploids of O. pusilla also do not share the wide geographic distribution of those polyploids derived from O. humifusa s.l. and O. macrorhiza s.l. These observations suggest that the polyploids of O. pusilla may have formed only recently, do not share comparable dispersal
96 agents, or lack the obvious adaptive advantages of those polyploids derived from O. humifusa s.l. and O. macrorhiza s.l. Many polyploid populations of O. humifusa s.l. and O. macrorhiza s.l. occupy northerly distributions and thus hav e a very high tolerance to cold temperatures. The hexaploid Opuntia fragilis (Nutt.) Haw., 1819 (not in the Humifusa clade) similarly inhabits areas of northern North America (Parfitt 1991 ; Loik and Nobel 1993 ; Ribbens 2008 ; Majure and Ribbens in press), w ith diploid relatives (e.g., O. polyacantha var. arenaria ) restricted to the southwestern U.S. (Parfitt 1991 ; Pinkava 2002). Thus, certain polyploid taxa appear to be more cold resistant than their southerly diploid relatives (and presumed progenitors). Op untia humifusa s.l. from northern areas of its distribution can withstand temperatures of 20C (Nobel and Bobich 2002). However, the cold tolerance of diploid taxa has not been tested. Certain polyploid taxa of the Humifusa clade may therefore be better adapted to adverse environmental conditions than their diploid progenitors, which may partly explain their wide distribution relative to their diploid counterparts. Agamospermy The tetraploid O. cespitosa (an entity within O. humifusa s.l.; see Table 5 1) produces viable seed in the absence of outcrossing (Majure pers. obsv.), so this taxon is either self compatible, which is common in Cactaceae (Rebman and Pinkava 2001), or agamospermous. Agamospermy is commonly associated with polyploidy (Stebbins 1950 ; D eWet and Stalker 1974 ; Harlan and DeWet 1975 ; Lewis 1980 ; Grant 1981 ; Whitton et al. 2008) and has been reported in numerous polyploidy Opuntia ; Felker et al. 2010), including O. humifusa s.l. and O. stricta (Naumova 1993). Agamospermy would account for the high level of morphological variation observed among polyploid populations, as a result of the maintenance of a specific genotype within a given population through the lack of
97 recombination (DeWet and Stalk er 1974). Some agamic complexes also have wider distributions than their diploid progenitors (Babcock and Stebbins 1938 ; Stebbins1950), as do certain polyploid taxa in this study. Autopolylploidy vs. Allopolyploidy The mechanism by which Opuntia polyploids are formed (auto vs. allopolyploidy) is unclear. Unreduced gametes have frequently been found in meiotic analyses of Cactaceae (e.g., Pinkava et al. 1977 ; Pinkava and Parfitt 1982 ; Pinkava et al. 1985). Unreduced gamete formation coupled with interspecif ic hybridization (allopolyploidy) likely is a major factor in polyploid formation within the genus, given that Opuntia is renowned for hybridization (Benson 1982 ; Grant and Grant 1982 ; Pinkava 2002 ; Griffith 2004 ; LCM, RP, PG, WSJ, PSS, DES unpubl. data). It is probable that unreduced gamete formation within a single species (autopolyploidy) also plays a role in the formation of polyploids. Autopolyploids have been discovered in Cactaceae (Pinkava et al. 1985; Sahley 1996 ; Hamrick et al. 2002) and may be mo re common than is suspected. Opuntia humifusa as currently circumscribed consists of numerous morphological entities, which are either diploid or tetraploid; those populations differing in ploidy are generally geographically well separated from one another It is evident from our phylogenetic analysis (Fig. 5 4) that O. humifusa is polyphyletic. Considering morphological and genetic data, it is likely that tetraploid O. humifusa is of allopolyploid origin. However, the pattern in O. pusilla is different, wi th populations of diploids found in close proximity to populations of triploids and tetraploids (Fig. 5 3). This evidence, plus morphological similarity among ploidal levels, suggests possible formation of autopolyploids. This same pattern is seen in other autopolyploid taxa (Lewis 1967 ; Nesom 1983), although there are exceptions to this pattern (Stebbins 1950 ; Soltis 1984 ; Husband and Schemske 1998). Molecular phylogenetic analysis (Fig. 5 4) and
98 morphological characters (LCM, RP, PG, WSJ, PSS, DES unpubl. data; see Fig. 5 1E G) of O. pusilla also do not support an interspecific hybrid origin for the different ploidal levels herein observed for this species, although more variable molecular markers, cytogenetic work, and more detailed morphological analyses are needed to appropriately address this question. Morphological Correlations with Polyploids Some polyploid taxa in the Humifusa clade share morphological characters with diploids and other polyploids, suggesting that they may be derived from hybridizati on (Table 5 2). Opuntia nemoralis Griffiths, 1913, (Fig. 5 1J; an entity within O. humifusa s.l.; see Table 5 1) shares spine color and orientation, cladode color, and glochid color of tetraploid O. macrorhiza (from Arkansas), although, it possesses small and easily disarticulating cladodes, retrorsely barbed spines, and the pile forming growth form and yellow flowers of O. pusilla (Fig. 5 1E G). Opuntia cespitosa (Table 5 1), as mentioned above, exhibits the red centered flowers, glaucous gray cladodes, an d dark glochids (Fig. 5 1I) of tetraploid O. macrorhiza (Fig. 5 1D), as well as the spine characters of diploid O. humifusa s.l. (= O. ammophila O. austrina O. lata ; Table 5 2). Throughout the distribution of the most common polyploid taxa, there also ar e polyploid populations that appear to be introgessive products of hybridization with other polyploids. For instance, in Michigan, Wisconsin, and western Illinois, certain populations display characters of both O. cespitosa and tetraploid O. macrorhiza (see Majure 2010, Fig. 5 1). In Bibb County, Alabama, populations appear to be interme diate between O. cespitosa and O. pollardii Britton & Rose, 1908, (tetraploids of O. humifusa s.l.; see Table 5 1), with the red centered flowers and rotund cladodes of O.cespitosa but the yellowish glochids and light green cladode color of O. pollardii. In Fayette County, Tennessee, plants appear intermediate between O. humifusa s.s. (i.e., tetraploid O. humifusa represented by the type collection) and O. cespitosa hav ing the
99 yellowish glochids of tetraploid O. humifusa s.s. and the spine characters of O. cespitosa Each one of the areas in which these intermediate plants occur appears to be a region of secondary contact, where polyploid taxa have introgressed to form n ew polyploidy morphotypes that exhibit characters of both of the putative parents. In the eastern U.S., most populations are represented by only one morphotype and thus appear to be morphologically stable (except for typically variable characters such as spine number; see Rebman and Pinkava 2001), indicating that hybridization is not ongoing among genomically distinct polyploid taxa. In contrast, in central Arkansas and populations farther west, more than one species and/or morphotype may be encountered wi thin a given population. Also, in many coastal populations throughout the southeastern U.S., more than one species may be encountered, and putative hybrid taxa are sometimes observed. Summary Members of the Humifusa clade are found throughout most of the c ontinental U.S., with no obvious breaks or dis jct. s in distribution patterns until detailed analyses of chromosome number were carried out. Our analyses indicate that diploid taxa in the Humifusa clade are presently confined to the southwestern and the sou theastern U.S., which likely represent Pleistocene refugia for these taxa. Polyploid taxa of O. humifusa s.l. and O. macrorhiza s.l. were likely formed when diploids from these two refugia came into contact during interglacial cycles of the Pleistocene. Th is scenario is supported further by phylogenetic analyses, in which two clades correspond to these two diploid refugia, and polyploid taxa are found in either clade. Polyploid taxa likely also contributed to the diversity of polyploid morphotypes through s econdary contact and introgression with other polyploids. After the end of the last glacial maximum, open niches would have been readily available for colonization by polyploid taxa produced towards the leading edge of the expansion and distribution of the Humifusa clade. These polyploids
100 subsequently dispersed throughout most of the continent and occupied all suitable habitats available after glacial retreat, accounting for the distribution that we see today. Distributional success was enabled by the extre me cold tolerance displayed by many of the polyploid taxa, which allowed them to colonize more northern areas presumably unsuitable for diploid taxa.
101 Table 5 1. Synonyms of O. humifusa s.l. and O. macrorhiza s.l. sampled during this study. Opuntia humifu sa s.l. Opuntia macrorhiza s.l. Opuntia allarei Opuntia fusco atra Opuntia ammophila Opuntia grandiflora Opuntia austrina Opuntia xanthoglochia Opuntia cespitosa Opuntia cespitosa Opuntia lata Opuntia lata Opuntia nemoralis Opuntia nemoralis Opuntia pollardii Opuntia pollardii
102 Table 5 2. Selected taxa of O. humifusa s.l. and O. macrorhiza s.l. with morphological characters and corresponding ploidy. Polyploids often exhibit characters from more than one diploid taxon or characters of other polyploids, although certain characters (e.g., red glochids) have not been observed in any diploids analyzed thus far. Taxon (ploidy) Flower color Cladode color Spine barbedness / Cladode disarticulation Glochid color O. ammophila (2 x ) Yellow Dark green Not barbed/no Stramineous O. austrina (2 x ) Yellow Dark green Barbed/yes Stramineous O. cespitosa (4 x ) Red centered Glaucous green Not barbed /no Red O. lata (2 x ) Yellow Dark green Barbed /yes Stramineous O. humifusa (4 x ) Yellow Dark green Not barbed /no Stramineous O. macrorhiza (4 x ) Red centered Glaucous green Not barbed/no Red/yellow O. nemoralis (4 x ) Yellow Glaucous green Barbed/yes Yellow O. pollardii (4 x ) Yellow Dark green Barbed/yes Stramineous O. xanthoglochia ( 2 x ) Red Centered Glaucous green Not barbed/no Yellow
103 Table 5 3. Taxa used in phylogenetic analyses of ITS sequence data given with their GenBank accession numbers. Accession Locality GenBank accession # Opuntia basilaris (outgroup) Inyo Co., CA R. Altig s.n. JF786913 Opuntia abjecta (2 x ) Monroe Co., FL LCM 3908 JF787021 Opuntia abjecta (4 x ) Monroe Co., FL LCM 3318 JQ245716 Opuntia ammophil a (2 x ) Marion Co., FL LCM 2826 JF786904 Opuntia austrin a (2 x ) Highlands Co., FL LCM 3450 JF786911 Opuntia cespitos a (4 x ) Scott Co., MO LCM 2441 JQ245717 Opuntia humifusa (4 x ) Warren Co., VA LCM 3800 JQ245718 Opuntia lat a (2 x ) Irvin Co., GA LCM 3785 JF786949 Opuntia macrorhiza (4 x ) Kerr Co., TX LCM 3510 JF786960 Opuntia nemoralis (4 x ) Garland Co., AR LCM 2196 JQ245720 Opuntia pusilla (2 x ) Lowndes Co., MS LCM 843 JQ245721 Opuntia pusilla (3 x ) Baldwin Co., AL LCM 1091 JF786985 Opuntia pusilla (4 x ) Jackson Co., MS LCM 1920 JF786986 Opuntia tortispina (6 x ) Hutchinson Co., TX LCM 3533 JF787020 Opuntia xanthoglochi a (2 x ) Bastrop Co., TX LCM 1982 JQ245719
104 Figure 5 1. Selected taxa in the Humifusa clade with associated chromosome squashes A) diploid O. humifusa ( O. lata ) LCM 4106 B) tetraploid O. humifusa s.s. LCM 3810 C) diploid O. macrorhiza ( O. xanthoglochia ) LCM 1983 D) tetraploid O. macrorhiza LCM 3510 E) diploid O. pusilla LCM 753 F) triploid O. pusilla LCM1033 G) tetraploid O. pusilla LCM 3700 H) diploid O. abjecta LCM 3908 I) tetraploid O. humifusa (O. cespitosa) LCM 2610 J) tetraploid O. humifusa ( O. nemoralis) LCM 4204 K) pentaploid O. ochrocentra LCM 3907 and L) tetraploid O. humifusa ( O. pollardii) LCM 769
105 Figure 5 2. Cytogeography of O. humifusa s.l., O. macrorhiza s.l., O. pottsii, and O. tortispina Diploids are represented with black circles, tetraploids by white circles, and hexaploids are represented by gray circles. Opuntia humifusa diploids are confined to t he southeastern U.S., and O. macrorhiza diploids are located in eastern Texas and southeastern New Mexico
106 Figure 5 3. Cytogeography of O. pusilla Diploids are represented by black circles, triploids by gray circles, and tetraploids by white circles. Note that most polyploids are restricted to coastal areas.
107 Figure 5 4. Majority rule consensus topology from 10000 ML bootstrap pseudoreplicates using RAxML based on the nrITS region. The western diploid O. macrorhiza s.l. ( O. xanthoglochia ) forms a well supported clade with polyploid O. macrorhiza, O. tortispina and the eastern polyploid morphotypes of O. humifusa s.l. ( O. cespitosa, O. humifusa and O. nemoralis ). The southeastern diploid morphotypes of O. humifusa s.l. ( O. ammophila, O. austrina, O. lata ) and diploid O. abjecta and O. pusilla form a wellsupported clade with polyploid members of O. pusilla and O. abjecta
108 Figure 5 5. Hypothetical origin and subsequent dispersal of polyploid taxa from diploid refugia. Diploid refugia are represented by A southeastern O. humifusa s.l. diploids B C eastern Texas and southeastern New Mexico O. macrorhiza s.l. diploids D I represent polyploid formation where D represents O. humifusa E represents O. cespitosa F represents O. pollardii G represents O. nemoralis H represents tetraploid O. macrorhiza (showing likely multiple formations), and I represents tetra and hexaploid O. tortispina
109 CHAPTER 6 PHYLOGENY OF THE Humifusa CLADE ( Opuntia S.S.): WHAT DIPLOIDS CAN TEL L US ABOUT THE EVOLUTIONA RY HISTORY OF THE GR OUP Background Opuntia s.s. (nopales, prickly pear cacti) is a well supported clade of shrubs and trees in subfamily Opuntioideae of Cactaceae (Majure et al. 2012a). Flat, succulent, photosynthetic stem segments with determinate growth characterize species within the clade (Pinkava 2003) These species may be either hummingbird or insect pollinated (Diaz and Cocucci 2003; Puente 2006; Reyes Agero et al. 2006; Majure et al. 2012a). The clade is suggested to have originated in southern South America with subsequent expansion into North America (Majure et al. 2012a) and is considered to have the widest geographical range of any genus within Cactaceae (Anderson 2001 ; Wallace and Dickie 2002 ). The Humifusa clade is the result of a small radiation of insect pollinated species within Opuntia s.s., which is proposed to have originated in western North America with subsequent migration into the eastern United States at the end of the Pliocene or beginning of the Pleist oc ene (Majure et al. 2012a, b). The clade currently consists of six recognized species, O. abjecta Small O. humifusa (Raf.) Raf. O. macrorhiza Engelm. O. pottsii Salm Dyck O. pusilla (Haw.) Haw., and O. tortispina Engelm. ex Bigelow (see Pinkava 2003 ; Majure et al. 2012a, b) although, the use of the name O. tortispina (Pinkava 2003) is mostly based on the misinterpretation of O. cymochila (Pinkava pers. comm.), so the name O. cymochila will be used throughout the rest of this study for that taxon. Mo rphological and cytological data suggest that the recognition of additional species in the clade may be warranted (Majure and Ervin 2008 ; Majure et al. 2012b). As shown in Majure et al. (2012b) with ITS data, the Humifusa clade consists of two subclades, 1) the southwestern O. macrorhiza s.l. subclade (SW), which includes diploid and tetraploid O. macrorhiza s.l., tetraploid O. humifusa s.l., tetraploid O. pottsii and tetra and
110 hexaploid O. cymochila and 2) the southeastern United States O. humifusa s.l subclade (SE), which includes diploid O. humifusa s.l., diploid and tetraploid O. abjecta and diploid, triploid, and tetraploid O. pusilla This suggests that the widely distributed taxon O. humifusa s.l. is not monophyletic and may actually be compose d of several morphologically cryptic species. For example, the widespread, tetraploid taxon O. cespitosa which is currently in synonymy with O. humifusa has yellow flowers with red centers, a character typical of the SW sub clade but exhibits a growth form and spine characters that are more typical of certain members of the SE sub clade Morphologically, t his suggests that O. cespitosa may have originated from hybridization between the two subclades but also that it constitutes a different entity that should be recognized separately from members of its putative progenitor subclades Athough, O. cymochila was resolved in the SW sub clade in previous ana lyses (Majure et al. 2012a, b), it is suggested, morphologically and cytologically, to have originated via hybridization between O. macrorhiza (of the SW sub clade) and O. polyacantha (Pinkava 2003) of the Polyacantha clade (Majure et al. 2012a). This is supported by spine patterns, flower color, and tetra and hexaploid chromosome counts, but has not been v erified with DNA sequence data (Majure et al. 2012a, b). Polyploidy is very common throughout Opuntia s.s. (Majure et al. in review) and is also quite widespread in the Humifusa clade. Out of 318 counts reported for the Humifusa clade, roughly two thirds (66%) were polyploid and 34% were diploid (Majure et al. 2012b). Polyploid taxa in this group are much more widespread than diploid members of the clade, extending from the southeastern United States, as far north as Ontario, Can ada, but diploid taxa are restricted to two presum ed glacial refugia in the southwestern and southeastern United States (Majure et al. 2012b). The pattern of widely distributed polyploids and geographically restricted diploids is a
111 common observation in p olyploid complexes (Stebbins 1950; Grant 1981). The wide distributions exhibited by certain polyploid taxa may be facilitated by their higher cold tolerance, as compared to their southern diploid counterparts (Majure and Ribbens 2012 ; Majure et al. 2012b). For example, Nobel and Bobich (2002) report that O. humifusa s.l. from the northern United States (i.e., part of the polyploid distribution of the species; Majure and Ribbens 2012; Majure et al. 2012b) is able to survive temperatures as low as 25C. Tole rance to more extreme environmental conditions by polyploid taxa in contrast to their diploid relatives is a common feature in many polyploid complexes (Stebbins 1950 ; 1971 ; Grant 1981; Levin 1983 ) Harsh environmental conditions have even been suggested to increase the frequency of polyploidy (Stebbins 1950; Grant 1981; see also review by Soltis and Soltis 2009) Most polyploid taxa in the Humifusa clade are thought to have arisen as a result of secondary contact with divergent diploid taxa from the sout hwestern (SW clade) and southeastern United States (SE clade) during and after the Pleistocene (Majure et al. 2012b) Newly formed polyploids between these two clades would have subsequently occupied open, available niches northward concomitant with glacia l retreat after the last glacial maximum (LGM). This scenario is supported by divergence time estimation of the Humifusa clade, polyploid distribution patterns, morphology, and a phylogenetic analysis using ITS sequence data (Majure et al. 2012a, b). Spe cies limits in the Humifusa clade are unresolved partly as a result of presumed hybridization among species resulting in individuals or populations demonstrating combinations of characters of putative progenitors, which may obscure clear morphological syna pomorphies for species. Also, species of Opuntia are rarely collected, and when they are, poor collection methods of these succulents generally result in low quality specimens (Reyes Agero et al. 2007)
112 that lack much if any useful taxonomic information, a s their three dimensional structure is typically lost. Lastly, Opuntia are inherently morphologically variable, wherein morphological characters exhibited by an individual may depend on microclimatic conditions ( e.g ., numbers of spines produced, cladode si zes, etc.; Benson 1982; Rebman and Pinkava 2001; Majure 2007). Thus, species determinations may be virtually impossible from herbarium specimens unless the collector sampled morphological diversity from throughout a given population and made note of those morphological characters lost in the collection process (e.g., cladode thickness, epidermis color, flower color, growth form, etc.). In this study we aim to reconstruct the phylogeny of the diploid members of the Humifusa clade to aid in the determination of species boundaries, as well as to test the origin of polyploid taxa (especially O. humifusa s.l.) within the clade, using maternally inherited plastid and bi parentally inherited nuclear data. We further test t he proposed hypothesi s of the origin of po lyploids via hybridization between the two diploid clades which has been proposed based on diploid glacial refugia polyploid distributions and morphological characters (Majure et al. 2012b). Material and Methods Taxon and Marker Sampling We sampled all six recognized species within the Humifusa clade (see above) from throughout their ranges, including diploids and polyploids of those species, when applicable (Majure et al. 2012b). We also sampled the different morphotypes of diploid and tetraploid O. hu mifusa s.l. (e.g., O. ammophila (2 x ), O. austrina (2 x ), O. cespitosa (4 x ) O. humifusa s.s. (4 x ) O. lata (2 x ) O. nemoralis (4 x ) O. pollardii (4 x ) and O. macrorhiza s.l. (e.g., O. allairei (4 x ) O. fusco atra (4 x ) O. grandiflora (4 x ) O. macrorhiza s.s. (4 x ) O. xanthoglochia (2 x )) (see Table 1). Opuntia polyacantha was used as an outgroup based on (Majure et al. 2012a) and to test the
113 origin of O. cymochila, as O. polyacantha is suggested to be one of the parents of O. cymochila (Pinkava 2003). We sam pled the plastid intergenic spacers ndhF rpl32, psbJ petA, trnL F the plastid genes, ycf 1 and matK the low copy nuclear gene ppc the nuclear ribosomal internal transcribed spacers (ITS) following Majure et al. (2012a), and the low copy nuclear gene, is i1 (Rook et al. 2006). See Majure et al. (2012b) for primers and reaction specifications for ndhF rpl32, psbJ petA, trnL F matK, ycf1 ppc and ITS. After initial amplification, cloning, and sequencing of isi1 products derived from primers designed by (Franck et al. in press), we discovered two copies of isi1 isi1 as analysis of the long copy revealed molecular synapomorphies for both the SW an d SE clades Hence, this copy was deemed useful as a marker for uncovering potentia l reticulations between the two clades. Reaction specifications for isi1 are the same for markers used in Majure et al. (2012a). PCR cycling conditions for isi1 were as follows: 95C for 5 min; followed by 44 cycles of 94C for 1 min, 55C for 1 min incre asing 0.3C/cycle, and 72C for 2.5 min; with a final extension of 72C for 10 min. We cloned a subset of polyploid taxa for ITS and isi1 using the Stratagene cloning kit (Stratagene, La Jolla, CA) to search for multiple copies derived from the union of divergent genomes through allopolyploidy. The gene ppc was uninformative for this purpose and was not cloned for polyploid taxa. Cloning w as focused on those polyploid taxa that were resolved in different locations using plastid and directly sequenced ITS products and the multiple polyploid taxa of O. humifusa s.l. (Table 1). We cloned one accession each of four tetraploid taxa of O.
114 humifus a s.l.: O. cespitosa, O. humifusa s.s., O. nemoralis and O. pollardii We also cloned ITS for tetraploid O. macrorhiza s.l., the tetraploid O. pottsii, and hexaploid O. cymochila We cloned isi1 products of tetraploid O. macrorhiza s.l. (including the ta xa O. allairei, O. macrorhiza s.s., and O. macrorhiza from AR), and a segregate of O. humifusa s.l. i.e., O. nemoralis (We had only marginal success amplifying isi1 for many of the polyploid taxa). We sequenced eight clones of each accession using bact erial primers (T3 T7) from the kits. Sequences were edited either in Sequencher 4.2.2 TM ( Gene Codes, Inc., Ann Arbor, MI ) or Geneious Pro TM 5.1 (Biomatters Ltd., Auckland, NZ) and the alignment was adjusted manually in Se Al v2.0 (Rambaut, 2007). Any obvio us recombinant sequences were excluded from phylogenetic analyses. Phylogenetic Analysis Maximum likelihood (ML) analysis was conducted using RAxML (Stamatakis 2006) undertaking 1000 nonparametric rapid bootstrap (bs) pseudoreplicates under 25 rate categories using the GTR+ model of molecular evolution for sequencing data and the BINGAMMA model of evolution for binary data (see below). We firs t performed a combined analysis of plastid and nuclear loci of only diploid taxa, as ploidy for all taxa under study here has been documented (Majure et al. 2012a) and the addition of allopolyploid (i.e., reticulate) taxa may lead to topological incongruen ce among data sets (Majure et al. 2012b), which is likely not the result of incomplete lineage sorting or other biological processes that could lead to incongruence ( see Wendel and Doyle 1998 ). We separated our diploid dataset into 1) sequence data, and 2) sequence data plus binary data of 7 coded indels from the combined plastid and nuclear dataset. Indels coded were those that were most likely homologous among ingroup taxa based on the outgroup (i.e., basal most taxa; Graham et al. 2000). Polyploid taxa w ere added to both plastid and nuclear datasets and analyzed separately, after initial analyses of diploid taxa, to test for
115 topological incongruence between nuclear and plastid phylogenies. Indel coding was not used for phylogenetic analyses with polyploid s included. Topological incongruence for a given polyploid taxon among resultant plastid and nuclear phylogenies was taken as evidence for allopolyploidy (i.e., hybrid origin among divergent parental genomes). Results P hylogenetic Analysis (Diploid Taxa ) A s in Majure et al. (2012b), the Humifusa clade was composed of two subclades, the southwestern O. macrorhiza clade (SW) and the southeastern O. humifusa clade (SE). The combined analysis of DNA sequence data along with indel codings, as well as analysis o f DNA sequence s alone, provide support for the two subclades (bs = 100/100 and 99/83, respectively; Fig. 6 1). Very little sequence divergence is evident in the resulting topology within the SE clade. Opuntia pusilla is resolved as sister to the rest of the clade and diploid O. humifusa ( lata entity ) is supported by indel coding (bs = 79/) as sister to a clade containing O. abjecta and O. humifusa ( austrina and ammophila entities). The three accessions of diploid O. macrorhiza are resolved in a well su pported clade (as noted above; bs = 100/100), however, the diploid O. macrorhiza entity referred to as O. xanthoglochia from eastern Texas does not form a clade with the other accession of O. xanthoglochia from eastern Texas but rather forms a well support ed clade (bs = 84/87) with diploid material from New Mexico. P hylogenetic Analyses (Polyploid Taxa ) Plastid data resolve O. humifusa s.l. and O. macrorhiza s.l. in several places. The O. humifusa s.l. taxa (i.e., O. cespitosa (from MI, MS, and TN) O. nemoralis (3 accessions, AR=1 and LA=2) and O. pollardii (3 accessions, AL, GA, and MS )) and one unnamed taxon of O. macrorhiza s.l. (1 accession, AR) are resolved in a well supported clade within a grade of SE diploid clade members. Likewise, Opunt ia humifusa s.s. (3 accessions, MA, MD, and MS) is
116 unresolved with other members of the SE diploid clade. Triploid and tetraploid O. pusilla and tetraploid O. abjecta are also resolved with diploid members of the SE clade. Members of both O. humifusa s.l. ( i.e., Opuntia allairei 1 accession TX, O. cespitosa 2 accessions, MI, WI, and O. nemoralis, 1 accession LA) and O. macrorhiza s.l. [i.e., O. fusco atra 1 accession TX O. grandiflora 2 accessions, MS, TX O. macrorhiza s.s., 3 accessions, TX=2, NM =1, and O. macrorhiza (unnamed taxa), 2 accessions, AR, UT ] as well as O. pottsii and O. cymochila are resolved in the diploid SW clade (Fig. 6 2). Directly sequenced PCR products of ITS for polyploid taxa virtually never exhibited polymorphisms in chro matograms. Directly sequenced ITS products of Opuntia macrorhiza s.s. and its segregate taxa, formed a well supported clade with the SW diploids along with O. pottsii and O. cymochila As well, most eastern taxa belonging to O. humifusa s.l. also were re covered in the SW clade (e.g., O. cespitosa, O. humifusa s.s., O. nemoralis ), except for O. pollardii and one accession of O. nemoralis from LA, which were recovered within the SE clade (Fig. 6 3). ITS clones of O. humifusa s.s. and O. nemoralis were rec overed in both the SW and SE clades, while ITS clones of O. pollardii were only recovered in the SE clade and clones of O. cespitosa, O. macrorhiza s.l. and O. pottsii were only recovered in the SW clade. Clones of O. cymochila were recovered in the SW clade and with the outgroup, O. polyacantha one of its putative progenitors (Fig. 6 3). Only one copy type was recovered for isi1 clones for O. pollardii which was resolved again with the SE clade. Likewise, only one copy type was recovered for O. nemoralis which was resolved in the SW clade. One accession each of O. macrorhiza s.l. ( O. macrorhiza unnamed entity, AR) and O. humifusa s.l. ( O. allairei ) was resolved in the SW clade and a subclade of the SW clade with a clone of Opuntia pottsii The same accession of O. macrorhiza
117 (unnamed entity) from AR also was resolved in the SE clade. Only one copy type of isi1 was found for O. macrorhiza s.s., which was resolved in the SW clade (Fig. 6 4). Discussion T he recent origin of th e Humifusa clade (from the late Plio to early Pleistocene; Majure et al. 2012a) most likely has not allowed sufficient time for notable sequence divergence among diploid members of the SE clade using the markers implemented in this study. However taxon relationships among t he diploid members of the SE clade are mostly resolved with DNA sequence data only, and are further supported with the addition of binary data from indel coding (see Fig. 6 1). Diploid taxa within the SE clade are morphologically diver se, ranging from small, prostrate species with disarticulating cladodes, mostly of the coastal zone of the southeast (excluding the Florida peninsula; e.g., O. pusilla ), to large, robust shrubs or small tree like taxa of the interior Florida peninsular scr ub (e.g., O. ammophila and O. austrina entities, both elements within O humifusa s.l. ) and ascending to slightly erect, shrubs of the Florida Keys (e.g., O. abjecta ) M orphological and phylogenetic data suggest that several diploid members of O. humifusa s.l. should be recognized as separate from tetraploid O. humifusa s.s., especially considering the paraphyly of O. humifusa s.l. in our diploid phylogeny (i.e., O. ammophila and O. austrina entities vs. the O. lata entity; Fig. 6 1) Members of the diploid SW clade are not notably morphologically divergent from one another. The two accessions of the O. xanthoglochia entity are more similar to one another, morphologically, than with the one diploid accession from NM, however, they are n ot sister taxa in our phylogeny (Fig. 6 1). So, t hese diploid accessions likely represent one species considering morphological observations and phylogenetic data. The w ide genetic divergence between the SW and SE clades was further increased as those tw o clades were most likely separated during the early mid Pleistocene by the proposed
118 disruption of the Gulf Coast arid zone (Webb 1990) as suggested by Majure et al. (2012b). The genetic discordance among members of both clades was thus influential in the production of allopolyploids when members of the SE and SW clades came back into contact with one another (see below Opuntia humifusa s.l.), although, the production of autopolyploids is likely a factor in the evolution of the diversity exhibited by both the SW and SE clades. Opuntia abjecta and O. pusilla Polyploid members of both O. abjecta and O. pusilla were always resolved in the SE clade, suggesting that those taxa were only derived from SE clade members (Table 2). Polyploids of these two species are nearl y identical to diploid individuals suggesting possible autopolyploid formation (Stebbins 1950 ; Soltis et al. 2007 ). However, one tetraploid accession of O. pusilla and one tetraploid accession of O. abjecta were of different haplotypes than their put ative diploid counterparts (and other polyploid accessions of both species), suggesting that they could have arisen through hybridization with another member of the SE clade, although, this will need to be tested further with population genetic level appro aches Genetic differences are also known to occur between autopolyploid taxa and their diploid progenitors (Soltis et al. 1989; Judd et al. 2007; Soltis et al. 2007). O puntia humifusa s.l. Our results indicate that O. humifusa s.l. is polyphyletic with the polyploid taxa of O. humifusa s.l. being derived from separate crosses, mostly between the SW and SE clades and the SE diploid taxa forming a paraphyletic assemblage (see above) The taxon Opuntia humifusa s.s. was derived from hybridization between the SW and SE clades, with the SE clade as the maternal lineage and the SW clade as the paternal lineage. The taxa referred to as O. cespitosa and O. nemoralis were each derived from two way crosses, with the SE clade and SW clades serving as both materna l and paternal lineages. Opuntia pollardii appears to have been derived
119 solely from the SE clade (based on plastid, ITS, and isi1 data), and O. allairei most likely is derived from the SW clade only (Table 6 2) and thus should not be considered synonymous with O. humifusa The clade formed from the tetraploid taxon of O. humifusa s.l., O. pollardii and close relatives in the plastid phylogeny consisted only of polyploid taxa (Fig. 6 2), so the diplo id counterpart to this clade either was not sampled or simply no longer exists, although, the diploid, O. lata entity of O. humifusa s.l. is very similar morphologically to O. po l lardii Autopolyploid formation of O. pollardii cannot be ruled out. The clo se relationship of the putative SE derived O. pollardii to those taxa derived from both the SE and SW clades ( O. cespitosa, O. macrorhiza AR, and O. nemoralis ), suggests that O. pollardii is one of the putative parents of those taxa, at least in some cros ses leading to those morphotypes. Opuntia macrorhiza s.l. Opuntia macrorhiza s.s. and several other polyploid taxa ( O. allairei, O. fusco atra, O. grandiflora ) were only resolved in the SW clade, suggesting that they originated via members of that clade on ly. Whether or not those polyploids were formed as the result of autopolyploidy or allopolyploidy is still to be determined. Diploid members of O. macrorhiza (e.g., entity O. xanthoglochia ; Fig. 6 1), are morphologically very similar to tetraploid O. macrorhiza s.s., although, they have more tenuous spines and tend to be smaller plants, so the production of autopolyploids in this group is possible. Very few diploids appear to exist in the primary range of these taxa (e.g., southwestern United States), and it is likely that most putative progenitors of these polyploid taxa could be extinct or that some polyploid taxa were actually derived from crosses among other polyploid taxa. The production of fertile hybrids is most effective among taxa with the same chromosome number (Lewis 1967), and this could also account for the morphological diversity in polyploid taxa, which is not seen in the diploids.
120 One accession of O. macrorhiza from AR, however, is clearly an allopolyploid derived from the SW and SE clad e. However, this individual is not typical morphologically for O. macrorhiza as it produces flowers with completely yellow tepals. Typical flowers of O. macrorhiza s.s. have yellow tepals that are basally tinged red. Opuntia pottsii Opuntia pottsii also was resolved completely within the SW clade and exhibited plastid, ITS, and isi1 sequences that were unique to this species Opuntia pottsii is the strangest member of the SW clade, being the only species that commonly produces pink flowers and that has a single, stout trunk (although diminutive) much like the more robust taxon, O. austrina, of the SE clade. Determining the origin of this tetraploid will most likely require broader sampling of the species throughout its range, which extends into the C hihuahuan and Sonoran deserts (Powell and Weedin 2004). This will also require a search for putative diploid pro genitors, if any still exist. It may also be possible that O. pottsii is of autopolyploid origin, or its putative diploid progenitors are extin ct, and thus no morphologically similar taxa have been discovered for comparison with the species. Opuntia cymochila Although Opuntia cymochila a mostly hexaploid species, has been recovered in the SW clade using ITS and plastid data (Majure et al. 2012a, 2012b), morphology has long suggested that O. polyacantha of the Xerocarpa clade (sensu Majure et al. 2012a), may also be one of the putative progenitors (Pinkava 2003). ITS haplotypes recovered here also support a close relationship with O. polyacantha and the SW clade, implicating an interclade origin for this species. It is most likely that O. cymochila arose through hybridizations between a member of the SW clade and O. polyacantha at the boundary of diploid and tetraploid populations of O. macrorhiz a s.l. and diploid populations of O. polyacantha as suggested by Pinkava (2003). The
121 formation of O. cymochila likely has occurred numerous times, as both tetraploid and hexaploid individuals have been reported (Pinkava 2003; Powell and Weedin 2004; Maju re et al. 2012b). Recurrent formation of polyploid species is not uncommon (Soltis and Soltis 1991, 1999 ; Soltis et al. 2007 ). Morphological Characters of the SE and SW Clades The SE and SW clades are morphologically distinct. Diploid members of the SE clade exhibit stramineous colored glochids, spines that are typically retrorsely barbed to some degree, and flowers with completely yellow inner tepals. Diploid members of the SW clade, on the other hand, exhibit bright yellow glochids, smooth spines lacking noticeable retrorse barbs (at least to the touch), and flowers with yellow inner tepals that are basally tinged red, reddish brown, red orange, or reddish pink. Both clades con tain members that exhibit tuberous roots, a character attributed mostly to O. macrorhiza s.l. (Benson 1982) of the SW clade Although, numerous species of Opuntia have been recorded exhibiting more than one flower color (e.g., O. macrorhiza, O. pottsii ; P inkava 2003), it is clear from our analyses that differences in flower color are directly related to differential crosses leading to the origin of the taxon (or morphotype; see O. macrorhiza from AR above). This is easily exhibited in O. humifusa s.l., wh ich is often reported as having yellow flowers or yellow flowers with red centers (Britton and Rose 1920 ; Small 1933 ; Kalmbacher 1976 ; Ferguson 1987 ; Doyle 1990 ). For example, the tetraploid taxon of O. humifusa s.l., O. cespitosa has yellow flowers with red centers and was partially derived from the polyploid O. pollardii clade of the SE clade and partially derived from the SW clade (Table 6 2). Tetraploid O puntia humifusa s.s., on the other hand, has completely yellow flowers and was derived from other members of the SE clade (not the polyploid O. pollardii clade) and the SW clade (Table 6 2). Thus, more research into
122 different flower colors exhibited by species of Opuntia may reveal that many of those morphotypes are of distinct origins from one another Spine characters also may be analyzed in this manner. Those spines produced by O. cespitosa resemble O. pollardii in length, diameter, and their development from the cladode, whereas O. humifusa s.s. is mostly spineless, as are some diploid SE populatio ns of O. humifusa s.l. ( O. ammophila, O. austrina, and O. lata entities ). Spines produced by O. nemoralis are strikingly similar to those of O. macrorhiza s.l. (of the SW clade) in color and development from the areoles, while the growth form and flower color of O. nemoralis is suggestive of characters seen in O. pusilla (of the SE clade) Hence morphological characters also often are indicative of the cro sses leading to the formation of those taxa. Summary The Humifusa clade is composed of two well supported diploid subclades, the SE and SW clades, which diverged from one another most likely as a result of a break in the arid zone along the Gulf Coast of s outheastern North America during the Pleistocene. Members of both clades eventually formed contact zones primarily in eastern North America, where they formed numerous allopolyploid entities, several of which appear to represent cryptic species. These all opolyploid taxa exhibit morphologically unique combinations of characters derived from their progenitor clades Several of these polyploid taxa undoubtedly arose multiple times, as shown by bidirectional gene flow (i.e., from plastid and nuclear data) lea ding to the formation of those taxa (e.g., O. cespitosa, O. nemoralis ; see Table 6 2 ). Opuntia humifusa s.l. as currently circumscribed is highly polyphyletic, consisting of different ploidal levels, and a wide array of morphological diversity. Diploid me mbers of O. humifusa s.l., according to our phylogeny, form a paraphyletic assemblage and thus should be recognized as separate taxa.
123 Consequently, o ur concept of the species that occur in the eastern United States must be reevaluated to take into account their evolutionary history, as revealed through cytological, morphological, and phylogenetic data, if we intend to incorporate the biological processes involved in species formation in this clade into an informative and predictive, phylogenetically accura te system of classification However, if we are to regard different morphotypes of distinct origins as species, it will also require careful analysis of morphological characters and ploidy over the entire distribution of the taxon where possible, to gene rate a practical system of classification based on cohesive morphological characters for a given species
124 Table 6 1. Synonyms of O. humifusa s.l. and O. macrorhiza s.l. used in our analyses. Synonyms are listed on the left (based on Benson 1982, Pinkava 2003, Powell and Weedin 2004) and are given with their ploidy as reported by Majure et al. (2012a). Synonyms Currently Recognized Opuntia allarei Griffiths (4x) Opuntia humifusa s.l. Opuntia ammophila Small (2x) Opuntia austrina Small (2x) Opuntia cespitosa Raf. (4x) Opuntia lata Small (2x) Opuntia humifusa (Raf.) Raf. s.s. (4x) Opuntia nemoralis Griffiths (4x) Opuntia pollardii Britton (4x) Opuntia fusco atra Griffiths (4x) Opuntia macrorhiza s.l Opuntia grandiflora Griffiths (4x) Opuntia macrorhiza Engelm. s.s. (4x) Opuntia xanthoglochia Griffiths (2x)
125 Table 6 2. Polyploid taxa of the Humifusa clade sampled in our analyses of nuclear and plastid data. Taxa are listed with their inferred maternal lineage based on plastid data and inferred paternal lineage based on nuclear data. Taxon Maternal lineage (cp) Paternal lineage (nuclear) O. abjecta SE Clade SE Clade O. allairei SW Clade SW Clade O. cespitosa SE Clade/SW Clade SE Clade/SW Clade O. cymochila SW Clade Polyacantha Clade O. fusco atra SW Clade SW Clade O. grandiflora SW Clade SW Clade O. humifusa s.s. SE Clade SW Clade O. macrorhiza AR SE Clade/SW Clade SE Clade/SW Clade O. macrorhiza s.s. SW Clade SW Clade O. nemoralis SE Clade/SW clade SE Clade/SW Clade O. pollardii SE Clade SE Clade O. pottsii SW Clade SW Clade O. pusilla SE Clade SE Clade
126 Figure 6 1. P hylogeny of diploid taxa of the Humifusa clade using combined plastid and nuclear data. In the SE clade, O. pusilla is sister to the rest of the taxa, and O. humifusa s.l. is made paraphyletic by O. abjecta The diploid entity O. xanthoglochia, of the SW clade ( O. macrorhiza s.l.), is not sister to another accession of the same morphotype. Bootstrap values are given above branches. Bootstrap values on the right are for the sequence data plus indel coding dataset, and those on the left represent just the s equence dataset (see Materials and Methods).
127 Figure 6 2. Plastid phylogeny including polyploid taxa. The SE clade of the diploid phylogeny is un resolved here as a grade, although, an entirely polyploid clade (the pollardii clade) is well supported (bs=75) within the SE grade. The SW clade is resolved, as in the diploid phylogeny. Members of Opuntia humifusa s.l. and O. macrorhiza s.l. are found in both the SW clade and the SE grade. Polyploid O. pusilla and O. abjecta are on ly recovered in the SE grade.
128 Figure 6 3. ITS phylogeny including polyploids. The SE clade of the diploid phylogeny is resolved here as a grade, as in the plastid phylogeny, although, one clade within the SE clade is resolved, albeit poorly (bs=50). The SW clade is again resolved and is well supported. Members of O. humifusa s.l. a nd O. macrorhiza s.l. are found in both the SW clade and SE grade, as well as the SE subclade. Opuntia pollardii is once again resolved with SE taxa. Clones of Opuntia cymochila are recovered within the SW clade and as sister to O. polyacantha
129 Figure 6 4. The isi1 phylogeny including polyploid taxa. The SE and SW clades are well supported (bs=99 and 100, respectively). Once again members of both O. humifusa s.l. and O. macrorhiza s.l. are resolved within both the SE and SW clades. Opuntia pollardii is resolved with SE clade members, as in the ITS and plastid phylogenies.
130 CHAPTER 7 TAXONOMIC REVISION O F THE Opuntia humifusa COMPLEX ( Opuntia : CACTACEAE) OF THE EASTERN UNITE D STATES Background Opuntia Mill. is native throughout the Americas, ranging from southern Argentina to Canada (Anderson 2001); the genus and occupies many habitats, from seasonally dry tropical and subtropical deciduous forests and scrub, to moderate desert environments, to temperate prairi es, coastlines, and forest openings (Benson 1982). Opuntia is considered to be the most widespread genus in Cactaceae (Anderson 2001). Opuntia exhibits very interesting morphological characters, which include longitudinally flattened stem segments, or cla dodes, that take over the photosynthetic function of the small, ephemeral long shoot leaves that are produced as the cladode develops. Cladodes may be glabrous or pubescent and may be a number of different colors. All species of Opuntia have glochids, or r etrorsely barbed and deciduous hair like spines that are produced from specialized short shoots (areoles), which are mostly included within the stem tissue. These often become exserted and conspicuous as the cladode develops and form a formidable armament against herbivores. Long spines are also produced in most species. These can be strongly retrorsely barbed or smooth. Some species form one type of spine, while others may develop both central (those produced from the center of the areole) and radial spine s (those produced from around the periphery of the areole). The development of spines from the areole can be a useful taxonomic character. Spine color changes through time but can also be diagnostic at the specific level. Virtually all species of Opuntia s trongly produce betalain pigments under stressful conditions, so water or cold stressed plants often become reddish, pinkish, or purplish around the areoles. Opuntia can form shrubs and small or large trees. Most tree like taxa are found in tropical or sub tropical areas. In temperate areas, smaller shrubby taxa, which commonly sprawl or trail
131 along the ground, are more frequently found. Although not a synapomorphy of Opuntia, the seeds are characteristic in having a bony funicular girdle that surrounds a bo ny funicular envelope, which covers the embryo. The funicular envelope may be glabrous or hairy and surface features of the funicular envelope may be taxonomically useful for delimiting species. Opuntia originated in the late Miocene in southern South Ame rica and from there dispersed north into the North American desert region (modern day central and northern Mexico and southwestern United States), where the clade diversified and expanded through to the Caribbean Islands, and throughout the rest of the con tinental United States. A small clade, the Humifusa clade, eventually migrated to the eastern United States (Majure et al. 2012a), where it experienced an additional, small radiation there as well. The Humifusa clade consists of two subclades, a southweste rn subclade (SW) including the widespread taxon O. macrorhiza s.l., and the southeastern subclade (SE), which includes the widespread taxon, O. humifusa s.l. and several other species (Chapter 6). The diploids of the SW subclade are characterized by procum bent species with yellow glochids, non retrorsely barbed (smooth) spines, and yellow flowers with red centers. Red centered flowers are also seen in diploid members of the sister clade to the Humifusa clade (i.e., Macrocentra clade), and so likely represe nts an ancestral state in the Humifusa clade. Diploids of the SE subclade are characterized by procumbent, trailing, and erect species that have stramineous glochids, retrorsely barbed spines (to some degree), and entirely yellow flowers. Numerous polyploi d taxa have formed within the Humifusa clade. Several of those taxa were shown to be the products of hybridization between the SE and SW subclades and demonstrate characters of both of those clades (Majure et al. 2012b; Chapter 6). The SE clade and polyplo id derivatives occurring in the eastern United States are here referred to as the O. humifusa complex.
132 Hybridization, Polyploidy and Morphological Variability Hybridization in Opuntia is common and plays into polyploid formation and oftentimes the origin of new species (Pinkava 2002; Majure et al. 2012a). The ability for taxa to readily hybridize and produce nearly fertile offspring would suggest the breakdown of species boundaries by those biologists following a strict biological species concept (Mayr 1942). Hybridization in Opuntia s.s occurs even among members of widely divergent clades and with other closely related genera, such as Consolea (Majure et al. 2012a). Hence, as in many p lant groups, the ability to hybridize and form viable offspring is meaningless regarding species boundaries (Soltis and Soltis 2009). In Opuntia, hybridization between different species is frequently associated with polyploidization (allopolypl oidy; see M ajure et al. 2012a), so reproductive barriers likely exist among divergent diploid species. However, polyploidization of those hybrid derivatives presumably aid s in overcoming sterility barriers (Stebbins 1950, 1971; Grant 1981; Levin 1983) The true mec hanism behind polyploidization in this group needs further study, however, unreduced gametes are commonly found in Opuntia which are likely the primary cause of the formation of polyploids (Pinkava 2002) both within (i.e., autopolyploidy) and among specie s (i.e., allopolyploidy). Although, allopolyploidy appears to be the most common type of polyploidy in Opuntia autopolyploidy may also be relatively common. Several taxa of Opuntia have been suggested to produce autopolyploids [e.g., O. abjecta O. drumm ondii, O. humifusa (subsp. pollardii ) O. macrocentra, O. strigil ; Majure et al. 2012a, b; Chapter 6], however, this needs to be investigated further. Polyploidy serves as a bridge for species formation and, in many cases, the combination of different gen omes, which may also lead to adaptations to extreme environmental conditions, as
133 in northern temperate members of the Humifusa clade (Nobel and Bobich 2002; Majure and Ribbens 2012; Majure et al. 2012b) and the Polyacantha clade (e.g., Nobel and Bobich 200 2; Majure and Ribbens 2012). It is very well known that Opuntia can be incredibly variable morphologically, in which cladode size, spine production, tuberous root production, among other features, are in many instances phenotypically plastic (Britton and Rose 1920; Benson 1982; Rebman and Pinkava 2001; Majure 2007; Majure and Ervin 2008). Thus, aside from hybridization and polyploidy, species delimitation in the group is made much more difficult, as populations of a species may show polymorphisms that res ult from growth under divergent environmental conditions (Majure 2007; Majure and Ervin 2008). Also, because of their succulence and spine production, opuntias are rarely collected, or the resulting specimens are improperly processed leading to scarce and very poor representation in herbaria ( Reyes Agero et al. 2007). Phenologically, species of Opuntia within the Humifusa complex are highly variable in flowering time, which seems to be directly related to changes in temperature regimes. For example, O. austrina in Florida alone may begin to flower in south Florida in mid March but in the same year may bloom in north Florida at the end of March or beginning of April. The same individual, if moved to cooler climates, will further alter its flowering time. Material of O. austrina from Florida, which typically blooms around the beginning of April, blooms around the first or second week of May in central Mississippi (Majure, pers. obs.). The same phenomenon can be seen in O. cespitosa and O. humfusa Southern populations start to flower before more northerly populations. Individuals taken from northern populations and transplanted to more southerly locations alter their flowering times within one or two growing seasons to nearly match those of the local inhabit ants (Majure, pers. obs.).
134 T axonomic History of the O. humifusa Complex Opuntia humifusa (Raf.) Raf. was described in 1820, albeit with no type locality (Rafinesque 1820; as Cactus humifusus ), as a low growing, yellow flowered, spineless (except for the glochids) species. Rafinesque described the range of the species as from New York to Kentucky and west to Missouri. The majority of the distribution given for O. humifusa by Rafinesque (Kentu cky west to Missouri) actually is inhabited by a red centered flowered Opuntia (see O. cespitosa below), so it is apparent that Rafinesque did not have a clear idea of the distribution of the species he was describing. Rafinesque (1820, 1830) noted the con fusion of Opuntia humifus a with that of Cactus opuntia L. (basionym of O. opuntia (L.) Karst., nom. illeg.) of the Atlantic coast. Nonetheless O. humifusa was again synonymized in later treatments under the tautonym Opuntia opuntia (see Britton and Rose 1920, Leuenberger 1993). Rafinesque (1830) described two more species, O. cespitosa Raf. from Kentucky and Tennessee (Rafinesque 1832), and O. mesacantha Raf. from west Kentucky to Louisiana (Rafinesque 1832), which also were subsequently placed in synonymy with O. humifusa (see Britton and Rose 1920). Engelmann (1856) proposed another name O. rafinesquei apparently in honor of Rafinesque, which he used to replace all three previously described species, O. cespitosa, O. humifusa and O. mesacantha At this time, O. vulgaris Mill. was accepted instead of O. opuntia and thus two species were recognized in the eastern United States, O. vulgaris of the Atlantic coast a nd O. rafinesquei ranging in distribution from the Mississippi Valley from Kentucky to Missouri and north to Minnesota (Engelmann 1856). Notably, although Rafinesque (1820) gave nearly the same distribution for the species (his O. humifusa ), he described the flowers as yellow, while Engelmann (1856) described them as being mostly yellow with red centers, demonstrating that Engelmann at least had a clear idea of the morphology of the species that grew throughout the range given with his description. Opuntia macrarthra Gibbes was later described by Gibbes
135 (1859) for low growing, yellow flowered material from South Carolina. Britton and Rose (1908) described yet another species of low growing, yellow flowered, spiny Opuntia from the coast of Biloxi, Mississip pi, O. pollardii Britton & Rose. Wherry (1926) described the yellow flowered O. calcicola from West Virginia, a species apparently restricted to circumneutral soils. John Kunkel Small began his exploration of Florida in the early 1900s, where he described 10 species from the O. h umifusa complex, O. abjecta Small, O. austrina Small (Small 1903), O. ammophila Small, O. lata Small (Small 1919), O. eburnispina Small, O. impedita Small, O. pisciformis Small, O. turgida Small (Britton and Rose 1923), O. atrocapensis Small, O. cumulicola Small, O. nitens Small, and O. polycarpa Small (Small 1933), most of which Benson (1982) later placed in synonymy with O. humifusa or merely considered them hybrid derivatives of O. stricta (Haw.) Haw. and O. humifusa (exc ept for O. abjecta ). Benson (1982) placed O. abjecta of the Florida Keys in synonymy with the Caribbean species O. triacantha (Willd.) Sweet. Benson (1982) recognized three varieties of O. humifusa, O. humifusa var. ammophila (Small) L.D. Benson, O. humif usa var. austrina (Small) Dress, and O. humifusa var. humifusa although, Pinkava (2003) did not recognize O. humifusa var. austrina and Wunderlin and Hansen (2003, 2011) did not recognize any varieties within O. humifusa Oddly, Benson (1982) concluded that O. humifusa is strictly a yellow flowered species, as further demonstrated in his key, although, his figure 438 of O. humifusa (Benson 1982; p. 439), is a typic al specimen of what Majure and Ervin (2008) referred to as O. cespitosa that has yellow flowers with red centers. It is thus apparent that Dr. Benson did not have a clear idea of the delimitation of O. humifusa a problem that likely developed from his use of herbarium specimens to interpret morphological variability across such a large range, and the fact that many
136 such specimens lose diagnostic features. Pinkava (2003) likewise suggested that O. humifusa has completely yellow flowers and used red centered flowers to separate O. macrorhiza from O. humifusa However, the majority of the distribution given for O. humifusa by both Benson (1982) and Pinkava (2003) is of the red centered taxon referred to here as the tetraploid, O. cespitosa (see below). Leuenberger (1993) recognized that O. humifusa although now widely accepted as the correct name of a widely distributed species in eastern North America, had not been formally typified. Thus, he neotypified O. humifusa based on material from Pennsylvania, as no type specimens for the species described by Rafinesque exist (Leuenberger 1993). Opuntia drummondii Graham was described from Appalachicola, Florida (Maund 1846). Subsequent researchers described numerou s taxa for the same type of material from the Atlantic and Gulf coasts, i.e., O. pes corvi LeConte ex Engelmann ( Engelmann 1856), O. frustulenta Gibbes (Gibbes 1859), and O. tracyi Britton (Britton 1911). Benson (1982) later placed all of these taxa, incl uding O. drummondii in synonymy under an ambiguous species of unknown origin and with no known type specimen, O. pusilla (Haw.) Haw. The name has sin c e been accepted by subsequent researchers (Doyle 1990 ; Pinkava 2003 ; Wunderlin and Hansen 2003 2011). O puntia nemoralis Griffiths was described from Longview, Texas by Griffiths (1913) and has since been placed in synonymy both with O. drummondii (Weniger 1967, 1970) and O. humifusa (Benson 1982). More recently, Majure and Ervin (2008) suggested that O. hu mifusa is composed of several taxa and used the name O. cespitosa for material of O. humifusa s.l. with red centered flowers. Cytological (Majure et al. 2012b) and phylogenetic (Majure et al. 2012a; Chapter 6) work has
137 provided further evidence, clearly in dicating that O. humifusa is not monophyletic and actually consists of several taxa. Those taxa are treated here. Seven species are recognized in this treatment of the O. h umifusa complex. These are Opuntia abjecta Small, O. austrina Small, O. cespitosa Raf., O. drummondii Graham, O. humifusa (Raf.) Raf., O. nemoralis Griffiths and O. ochrocentra Small. Three subspecies of O. humifusa are recognized: O. humifusa subsp. humifusa subsp. lata (Small) Majure, and subsp. pollardii (Raf.) Majure. Opuntia ces pitosa, O. humifusa subsp. humifusa and O. nemoralis are allop olyploid derivatives of the southeastern (SE) and southwestern (SW) subclade s of the Humifusa clade. Opuntia humifusa subsp. pollardii is a tetraploid, and apparently has been derived solely from the SE clade, while O. humifusa subsp. lata is a diploid member of the SE clade (Chapter 6). Opuntia ochrocentra is an allopolyploid derived from a member of the southeastern subclade and O. dil lenii (Ker Gawl) Haw. (Majure et al. 2012a; Chapter 4). Species outside of the O. humifusa complex that occur in the eastern United States, either as ornamentals or naturally, are not covered in this treatment (e.g., O. engelmannii, O. fragilis, O. leucotr icha, O. macrorhiza, O. monacantha, O. stricta ). In addition, this revision does not include members of the Humifusa clade that belong to the SW subclade, i.e., O. macrorhiza, O. pottsii and relatives, which are species primarily distributed throughout t he western United States and northern Mexico. Species Concept I apply a combined approach using phylogenetic, evolutionary, ecological, and morphological species concepts to delimit species in the Humifusa clade (Donoghue 1985; de Queiroz 2007). Species re lationships and boundaries in Opuntia are obscured by the paucity of morphological characters and frequently also by the inadvertent loss of the few that exist in the process of preparing herbarium specimens (although with effort taxonomically useful speci mens
138 can be prepared; see Reyes Agero et al. 2007). In addition, the succulence of these plants inhibits collectors, and the resulting lack of herbarium material, and especially those with useful habitat and morphological data, make specific and infraspec ific delineation exclusively through the use of herbarium specimens virtually impossible in many instances. Thus the time consuming process of collecting and growing plants for use in assessing morphological variability (and correlating this variabilit y with geography) is the only means to study the group in a relatively unbiased manner. The scarcity of detailed biological data, especially regarding variation in chromosome number, and the lack of an understanding of phylogenetic relationships also long has impeded proper species delimitation in this clade. Those data coupled with observations based on live material greatly enhance the ability to make accurate estimates of species boundaries. Undoubtedly, some researchers may find the species circumscript ion employed here to be too finely drawn, while others may wish that even more species had been recognized. I have taken a relatively conservative approach to species delimitation, underscoring the evolutionary history of these organisms, as well as their morphological cohesiveness and ploidy levels. The taxa here recognized are believed to be both biologically meaningful (reflecting the complex evolutionary history of the group) and diagnosable using accepted/traditional systematics methods (and thus appr opriate for recognition in Floras and ecological investigations). The following key was generated through the use of living specimens, supplemented by herbarium material, and so is most useful for identifying living individuals. In addition, knowledge of t he range of morphological variation within a population is often necessary to accurately identify the species, as individuals within a population may or may not display characters essential for the identification of a given species (as the result of phenot ypic plasticity,
139 age of the plant, or other factors). As a result, this dichotomous key is best used to identify a species when there is information about morphological variation within a given population and also when the entire plant, in living condition is available for observation. Description of the Opuntia humifusa Complex Small to large shrubs or treelets, erect, decumbent, or trailing, 0.1 2 m tall, branching profusely or sparingly; with tuberous or fibrous roots. Cladodes elliptical, rotund, oblong, or obovate, 0.8 29.5 cm long, 0.6 11.3 cm wide, 4 19.9 mm thick, dark or yellow green, or glaucous, gray green, margins smooth or scalloped, remaining turgid or cross wrinkling during the winter. Leaves green or glaucous, gray green, 2.2 13.8 mm lo ng, ascending parallel to the cladode or spreading, tips reflexed or not. Glochids conspicuous, exserted, or inconspicuous, included within the areole, red, reddish brown, yellow, or stramineous when young, aging dark brown, light brown, or amber. Spines a bsent or 1 18 per areole, 0.9 10.3 cm long, 0.2 1.3 mm in diameter, dark brown, reddish brown, yellow, brown and white or brown, white, and yellow mottled during development, turning white with age and later gray, cylindrical, flattened, or twisted at the base, only central spines present or radial and central spines present, retrorsely barbed or smooth to the touch. Flowers: outer tepals green, yellow green or red with light green margins, ovate, triangular, or triangular subulate, inner tepals yellow or y ellow with red bases, 7 10, obovate, or obtriangular to emarginate, 2.2 5.5 cm long, generally with a mucronate apex, stamina filaments yellow or yellow with yellow green, or red bases. stigmas white, cream, or green, 3 10 lobed. Berries clavate or barrel shaped, 1.8 5.0 cm long, pink, purple, red, orange red, or green at maturity. Seeds 3.1 5.9 mm long, with the funicular envelope smooth, or only moderately elevated by the cotyledons and hypocotyl of the embryo, or bumpy, greatly elevated by the cotyledon s and hypocotyl of the embryo, funicular girdle 0.4 1.3 mm wide, regular, smooth, or irregular, bumpy.
140 Key to the Members of the Humifusa Complex O. ochrocentra 1. Radial spines 0 1, cylindrica 2 2 Plants forming small trees, large shrubs, or sub shrubs; stems ascending or erect, 0.3 2 m tall; .. 3 2 Plants forming small shrubs in clumps or mat s; stems ascending, decumbent or trailing; 0.1 to 0 .5 m tall; inner tepals entirely yellow or yellow with red bases 3 Plants developing from a single flat or terete stem (or trunk), usually erect or strongly ascending, cladodes not easily d isarticulating, spines barbed to the touch, outer tepals ascending, incurved, or recurved, in bud, Peninsular FL... ..... .... O. austrina 3 Plants branching from the base, thus forming clumps, stems strongly ascending, cladodes easily disarticulating, spines strongly retrorsely barbed to the touch, outer tepals incurved in b ud Florida Key O. abjecta 4 Cladodes easily disarticulating, flat or cylindrical, spines strongly retrorsely barbed to the touch 5 4 6 5. Cladodes glaucous, gray green, developing spines yellow or bright white, glochids yellow or dull brown, inner tepals yellow or r arely yellow with pinkish bases ... ........ O. nemoralis 5. Cladodes not glaucous, dark green, developing spines dark reddish brown, or brown and white 6. Cladodes n ot noticeably glaucous, dark green, inner tepals entirely yellow O. humifusa 6. Cladodes glaucous, gray green or lead green, inner tepals entirely yellow, or yellow with ..
141 7 Plants small, sometimes even diminutive, cladodes 3.6 (0.8 11.1) cm long, 1.8 (0.6 3.4) cm wide, 10.4 (5.3 14.8) mm thick, elliptical, oblong, or rounded in shape, terminal cladodes mostly cylindrical in cross section, with 1 2 areoles per diagonal row at midstem, coastal southeastern United States............................................................................. O. drummondii 7 Plants larger, not diminutive, cladodes 7.6 (3.2 13.5) cm long, 4.5 (2.4 6.7) cm wide, and 10.2 (6.5 15.8) mm thick, elliptical, or rotund, terminal cladodes not cylindrical in cross section, with 2 3 areoles per diagonal row at midstem, Florida Keys O. abjecta 8. Plants small, cladodes 6.3 (4.5 8.4) cm long, 3.9 (2.8 5.8) cm wide, 11.2 (8.1 14.2) mm wide, oblong, elliptical, or obovate, easily disarticulating, spines barbed to the touch, 2.3 (1.4 3.0) cm long, inner tepals almost always yellow, rarely fain tly pink at the base, W of O. nemoralis 8. Plants larger, cladodes 10.5 (3.8 18.7) cm long, 8.0 (3.2 11.3) cm wide, 10 (4 19.2) mm thick, mostly elliptical, obovate, or more commonly rotund, not disarticulatin g, spines smooth to the touch, 2.9 (1.5 4.3) cm long, inner tepals basally tinged crimson red, orange red, reddish brown, or pinkish red, widespread, eastern United States O. cespitosa 1. Opuntia abjecta Small in Britton and Rose, The Cactaceae, pp 102 & 226c. 1923. T YPE : United States. Florida, Monroe Co.: hammock, southeaste rn tip of Big Pine Key, 12 Apr 1921, J.K. Small s.n. with G.K. Small, P. Matthews (holotype: NY!; see Fig. 7 2A). Shrubs to 0.3 m tall, usually with multiple stems arising fr om the base, stems strongly ascending and rigid; roots commonly forming tubers in older individuals. Cladodes disposed mostly with margins parallel to the soil surface, thus the cladode disposed with the broad (flat) side perpendicular to soil surface, not becoming cross wrinkled during the winter (as in other
142 non erect species, such as O. drummondii and O. humif usa ) Cladodes easily disarticulating from the nodes, generally dark green, not glaucous, and with slightly raised podaria, cladodes round to obovate (or more typically elliptical in tetraploids) in outline with 2 3 areoles per diagonal row at midsection o f cladode, cladodes 7.6 (3.2 13.5) cm long, 4.5 (2.4 6.7) cm wide, and 10.2 (6.5 15.8) mm thick. Leaves dark green, ascending, parallel to the cladode surface, 5.2 (3.5 7) mm long. Glochids straw yellow (stramineous), areolar trichomes white. Spines mostly 2 per areole, but oftentimes 3 on terminal cladodes, but generally more on basal cladodes (up to 6), which continue to produce new spines, when 3 spines on terminal cladodes, 2 long and 1 short, the spines dark reddish brown when young, turning white when mature and gray in age, strongly retrorsely barbed, twisted to cylindrical in cross section, most spines twisted at least at the base, 4.01 (2.5 5.6) cm long, 0.66 (0.33 0.99) mm in diameter. Flowers: out er tepals dark green, ovate, tepal tips erect to in curved in bud, apex of bud rounded to acute (Fig. 7 2E), inner tepals 8, dark yellow (Fig. 7 2G H), 23.8 (21 26) mm long, stamens with yellow filaments turning orange red as flower ages (Fig. 7 2G), stigma cream colored with 6 lobes. (The diploid populatio n at Big Pine Key produces mutant flowers with inner tepals producing anthers at their tips and finally with normal stamens in the center of the flower surrounding the gynoecium (Fig. 7 2G; this aberrant flower type has also been seen in O. austrina and O. drummondii ). The two tetraploid populations have completely normal flowers (Fig. 7 2H).) Berries barrel shaped, dark purple or yellow green (Fig. 7 2I), 2.7 (2.1 3) cm long. (Tetraploids appear to only produce sterile fruit.) Seeds 3.3 (3.1 3.6) mm long, funicular gi rdle 0.80 (0.66 0.88) mm wide, funicular envelope smooth, i.e., with no impression of the embryo apparent on seed surface. Phylogenetic placement Opuntia abjecta is sister to O. austrina (Majure et al. 2012a; Fig. 7 1).
143 Ploidy Opuntia abject a is diploid, 2n=22, and tetraploid, 2n=44 (Majure et al. 2012b). The diploid population occurs at the type locality of the species (Majure et al. 2012b). There are cryptic morphological differences among the diploid and tetraploid populations (cladode sha pe, spine length), although, these minor differences are not suggestive of species boundaries. Based on phylogenetic studies (Chapter 6) and morphological similarity, it appears most likely that the tetraploid populations are autopolyploid in origin. Phenology Opuntia abjecta blooms in early spring (late March mid April) in southern Florida, although, individuals transplanted further north bloom later (e.g., early May), demonstrating the plasticity in blooming time relative to climate. Distribution As far as is known, O puntia abjecta is restricted to the Florida Keys, Monroe Co. (Fig. 7 3) and has only been recorded from three populations. Habitat. Opuntia abjecta is restricted to Key Largo limestone of the lower Florida Keys where it can be found gr owing in depressions in the limestone containing enough humus to support root establishment. Notes Benson (1982) considered this species to be synonymous with the Caribbean taxa, O. triacantha (Willd.) Sweet and O. militaris Britton and Rose. However, it is clear from morphology and DNA sequence data that O. triacantha is more closely related to other Caribbean taxa, such as O. caracassana, O. jamaicensis and O. repens Bello (Majure et al. 2012b; Chapter 4) rather than members of the Humifusa clade. Britton and Rose (1920) also considered O. triacantha to be more closely related to other Caribbean taxa, and even included the species in Opuntia Series Tunae which includes O. caracassana and O. jamaicensis Opuntia militaris is closely related to O. caracassana, O. jamaicensis and O. triacantha but is likely not conspecific with O. triacantha (Chapter 4).
144 Opuntia triacantha is typically erect with a single, well defined trunk, whereas O. abjecta has numerous ascending stems produced from the base of the plant but never produces an erect main trunk. Opuntia triacantha produces chalky yellow spines when immature, which mature chalky white. The spines of O. abjecta are darker reddish brown when immature and mature bright white, not chalky white. Cladodes of O. triacantha are oblong to obovate or narrowly elliptic, while cladodes of O. abjecta are mostly rounded, obovate or broadly elliptic. Opuntia triacantha also has large tufts of yellow glochids associated with yellowish clear trichomes, which are more pronounced than the stramineous glochids and white clear trichomes of O. abjecta Additional s pecimens e xamined United States. Florida. Monroe Co. : Big Pin e Key 12 May 1919 P. Barrtsch s.n. (US); Long Key, r ocky, open, low ground, 23 Apr 1966 C. Byrd s.n. (FLAS); Big Pine Key, 4 May 1951 E.P. Killip 41332 (US); ibid, 10 Jan 1952, E.P. Killip 41708 (US); S E end of Big Pine Key, Cactus Hammock, National Ke y Deer Refuge, 6 Mar 2010, L. C. Majure 3908 (FLAS) ; Big Pine Key 22 Feb 1935 G.S. Miller, Jr. 1710 (US) ; Crawl Key, Jul 2008 K. Sauby s.n. (FLAS) ; Big Pine Key 17 May 1922 J.K. Small s.n. (NY, US) 2. Opuntia ochrocentra Small in Britton and Rose, The Cactaceae, p. 262. 1923. T YPE : Florida, Monroe Co.: Big Pine Key, h ammock, southern end of Big Pine Key, 11 Dec 1921, J.K. Small s.n. with G.K. Small, P. Matthews (holotype: NY!; see Fig. 7 4A). L arge, scrambling to slightly erect shrub from 0.4 0. 5 m tall, usually with one main trunk, although, branching heavily above; roots fibrous. Cladodes mostly elliptical or rarely obovate in outline, with slightly scalloped margins, terminal cladodes disarticulating with only slight force, cladodes light green, 15.6 (11.6 19) cm long, 7.5 (5.9 8.9) cm wide, 14.3 (13.2 16.2) mm thick, with 3 4 areoles per diagonal row. Leaves light to dark green, small, 3.6 (3.2 3.9) mm
145 long. Glochids bright yellow (as in O. dillenii ) conspicuous. Spines developing from th e areoles in a stellate pattern (as in O. dillenii ), 1 5 spines per areole, central spines delicate, 5.3 (4.7 5.8) cm long, 1.04 (0.86 1.3) mm in diameter, cylindrical in cross section or basally twisted, radial spines flattened at the base and deflexed al ong the face of the cladode in age, immature spines yellow aging white or mottled cream and brown and then gray. Flowers: outer tepals broadly ovate or triangular ovate, yellow green or reddish with light green margins, inner tepals 8, entirely yellow or y ellow green, obovate or emarginate, with a mucronate tip, the abaxial surface often reddish down the center, 3.5 (2.8 3.7) cm long, stamens with yellow filaments, stigma white or light yellow green, 6 lobed. Berries clavate or barrel shaped, although, matu re fruit not been seen in cultivated material from Big Pine Key or Big Munson Island, immature fruit 3 (2.8 3.3) cm long, and mature fruit reported to be red and to 2 cm long (Small 1923). Seeds not seen (and not present on lectotype or any other material available for study), but described as 2.5 3 mm long, and numerous (Small 1933). Phylogenetic Placement This species is a pentaploid (Majure et al. 2012b) of interclade hybrid origin most likely between O. abjecta and O. dillenii (Ker Gawl.) Haw. (Majure et al. 2012a, Chapter 4), with which it is largely sympatric on Big Pine Key. Ploidy Opuntia ochrocentra was reported as pentaploid, 2 n =55, from three individuals that have been analyzed (Majure et al. 2012 b ). Phenology Opuntia ochrocentra flowers i n late spring to early summer (early April May) growing in cultivation in north Florida. Distribution Opuntia ochrocentra is only known from the lower Florida Keys and co occurs with O. abjecta on Big Pine Key, where it has nearly been extirpated thr ough attack by Cactoblastis cactorum Berg. (Majure 2010; Majure pers. obs.) and anthropogenic disturbance
146 (Benson 1982, Majure pers. obs.). It is also known from Big Munson Island, just west of Big Pine Key (see collections L.C. Majure 3968 69 ), where it h as also been seen under attack by C. cactorum (Majure pers. obs.). It has been recorded from Cape Romano as well (Small 1933), but no specimens have been seen from that locality (Benson 1982; Majure pers. obs.). Habitat Opuntia ochrocentra occurs essent ially in the same habitat as O. abjecta one of its putative parents (Majure et al. 2012b). Notes This species was placed in synonymy with O. cubensis Britton and Rose by Benson (1982). Molecular, morphological, and cytological data show that O. ochroce ntra is not conspecific with O. cubensis and thus should not be considered synonymous with that Cuban species (Chapter 4). As noted by Britton and Rose (1923), Opuntia ochrocentra most closely resembles O. dillenii one of its putative progenitors, although, its spines are more delicate and age gray as in O. abjecta (its other putative progenitor; Majure et al. 2012a; Chapter 4). Opuntia ochrocentra also forms a smaller, more delicate shrub compared to the more erec t and robust growth form of O. dillenii Additional specimens examined. United States. Florida Monroe Co. : Big Pine Key, 12 18 Feb 1935 E.P. Killip 31423 (US); Big Pine Key, hammock, 2 Mar 1936, E.P. Killip 31712 (US); Big Pine Key, SE hammock, 19 Mar 19 52, E.P. Killip 42026 (US); S end of Big Pine Key, 6 Mar 2010, L.C. Majure 3907 (FLAS); Big Munson Island, 8 Mar 2010 L.C. Majure 3968 69 (FLAS) ; h ammock, S front of Big Pine Key, 17 May 1922, J.K. Small s.n. (US). 3 Opuntia austrina Small Fl. S E. US. p. 816. 1903. Opuntia compressa (Salisbury) J.F. Macbride var. austrina (Small) L.D. Benson, Cact. Succ. J. 41: 125. 1969. Opuntia humifusa (Raf.) Raf. var. austrina (Small) Dress, Baileya 19 (4): 164. 1975. T YPE : United States. Florida, [Miami Dade C o.:] Miami, in pinelands, 28 Oct 28 Nov 1903,
147 J.K. Small 1216, with J.J. Carter (lectotype designated by L.D. Benson ( 1982 ) : US! ; isolectotype: NY!; see Fig. 7 5A). Opuntia ammophila Small, J. New York Bot. Gard. 20: 29. 1919 Opuntia compressa (Salisbury) J.F. Macbride var. ammophila (Small) L.D. Benson, Cact. Succ. J. 41: 124. 1969. Opuntia humifusa (Raf.) Raf. var. ammophila (Small) L. D. Benson, Cact. Succ. J. 48: 59. 1976. T YPE : United States. Florida. [St. Lucie Co.:] h ammock on sand d une, St. Lucie Sound, 6 mi S of Ft. Pierce, 20 Dec. 1917, J.K. Small 8456 (holotype: two sheets, NY!). O puntia pisciformis Small in Britton and Rose, Cactaceae 4: 258. 1923. T YPE : United States. Florida. [ Duval Co.:] d unes, Pilot Island, 26 April 1921, J.K. Small s.n. (holotype: NY!). O puntia turgida Small in Britton and Rose, Cactaceae 4: 265. 1923. T YPE : United States. Florida. [Volusia Co.:] a bout 5 mi S of Daytona, 30 Nov 1919, J.K. Small s.n. (holotype: NY!, two sheets; isotype: US!). Opuntia atrocapensis Small Man. S E. Fl. 905. 1933. T YPE : not found, and therefore a neotype is designated here : Unite d States. Florida. Monroe Co.: s and dunes; Middle Cape Sable, 28 Nov 1916, J.K. Small s.n. (US!). Opuntia cumulicola Small Man. S. E. Fl. 907. 1933. T YPE : United States. Florida. [Miami Dade Co.:] b each. Bull Key, opposite Lemon City, 6 Nov. 1903, J.K. Small 970 with J.J. Carter (holotype: NY!; isotype s : NY!; US!). Opuntia nitens Small Man. S. E. Fl. 906. 1933. T YPE : United States. Florida. [Volusia Co.: ] h ammock, 5 mi S of Daytona, Florida, 23 Aug 1922, J.K. Small s.n. with G.K. Small, J.B. DeWinkler (holotype: NY!; isotype: US!).
148 Opunti a polycarpa Small Man. S. E. Fl. 905. 1933. Type: United States. Florida. [Collier Co.:] s and dunes, Caxambas Island, 11 May 1922, J.K. Small s.n. (holotype: NY!, isotype s : NY!; US!). Small to large shrubs or small treelets, 0.2 1.2 ( 2) m tall, usually erect but in some cases merely ascending, but with a central trunk, which may be cylindrical or flattened (Fig. 7 5B, D E), but plants damaged at the base of the trunk (e.g., burned, cut off, scarred, damaged by insects) often producing numerous branches from the base, and in age basal most cladodes often strongly fused and appearing as a single unit (Fig. 7 5D; instead of several stem segments), the plants typically heavily branched towards the apex and frequently semaphore like; roots commonly tuberous (Fig. 7 5C) or fibrous, the tubers more commonly produced in very well drained, deep sands. Cladodes h ighl y variable, gene rally elliptic but commonly obovate or rarely completely round, dark or light green, sometimes slightly glaucous, never cross wrinkling unless under severe drought stress, 14.5 (6.5 29.5) cm long, 6.5 (3.7 9.5) cm wide, thin 8.2 (6.4 10.9) mm thick, mostly with slightly scalloped margins, but margins sometimes non scalloped, from 2 6 (mostly 4) areoles per diagonal row cladodes occasionally easily disarticulating during winter months (in the polycarpa entity, see below), but generally with cladodes not easily detaching (the ammophila entity, see also below). Leaves dark green or sometimes glaucous, 9.3 (6.7 13.8) mm long, ascending (parallel to the cladode surface; Fig. 7 5F) or commonly spreading with the tips recurved. Glochids conspicuous, exserted from the areole, stramineous, forming adaxial crescent in older cladodes from the compression of the areole, trichomes mostly clear or appearing clear white. Spines mostly 1 2 per areole on terminal cladodes, although up to 3, or plants occasionally spineless, the trunks occasionally with up to 18 spines per areole, round in cross section or commonly twisted longitudinally, the spines highly variable in length, 6.1 (2
149 10.3) cm long, 0.9 (0.6 1.2) mm in diameter, strongly retr orsely barbed or relatively smooth to the touch, developing spines dark reddish brown or mottled (banded) brown yellow and white, turning white after maturity and finally gray in age, often deflexed upon maturation. Flowers: outer tepals dark green, triang ular or triangular subulate, tips ascending, incurved or commonly recurved in bud (Fig. 7 5G), inner tepals 8, dark yellow to light sulfur yellow (Fig. 7 5H I), obovate 3.8 (3.4 4.2) cm long, with a mucronate tip, staminal filaments yellow or greenish yell ow, stigmas white with generally 6 lobes. Berries clavate or barrel shaped (Fig. 7 5J K), dark purple, red, pink, or yellow green when mature, 3.8 (2.8 5.0) cm long. Seeds 4.2 (3.9 4.7) mm long, funicular girdle 0.96 (0.67 1.26) mm wide, funicular envelope smooth with the cotyledon and hypocotyl region of the embryo only moderately raised. Phylogenetic placement Opuntia austrina is sister to O. abjecta, as shown in Majure et al. (2012b Chapter 6 ) (Fig. 7 1) Ploidy Opuntia austrina is diploid 2 n =22 throughout its range (Majure et al. 2012 b ). Phenology Opuntia austrina begins flowering in southern Florida during late March early April. However, plants grown in more northern areas (e.g., central M ississippi ) typically produce flowers around the be ginning of May. Thus, flowering time appears to be strongly correlated with changes in climate. Distribution. Opuntia austrina is mostly restricted to the Florida peninsula (Fig. 7 6). One specimen from Gads den Co. Florida has been tentatively identifie d as O. austrina and one specimen from Lowndes Co. Georgia (UNC), was described with essentially the same growth form as O. austrina but the specimen is insufficient to confirm the its specific identity.
150 Habitat. Opuntia austrina is most common in pen insular Florida scrub habitat dominated by scrub oaks, Quercus chapmannii, Q. geminata, Q. myrtifolia and sand pine, Pinus clausa as well as sandhills dominated by Pinus palustris or Pinus elliottii Notes. Opuntia austrina is the most common species in the Florida peninsula and is most often found in remnant scrub habitats. Opuntia austrina is a highly polymorphic species and has by some workers been divided into a number of other taxa that are here placed in synonymy : O. ammophila, O. nitens, O. polycar pa, and O. turgida Of those four taxa, O. ammophila and O. polycarpa are quite distinctive and easily recognizable in parts of their range s and are here informally O. austrina The O. ammophila entity is most common from the Ocala National Forest in Lake, Marion, and Putnam co unties south to St. Lucie Co. where it was first described (Small 1903). Opuntia ammophila can form relatively large shrubs or treelets up to 1.2 m tall with a large diameter, cylindrical trunk (up to 40 cm in circumference). J ohn K. Small recorded individuals up to nearly 2 m tall (Small 1919, 1933) but no such individuals have been found since. The O. polycarpa entity is primarily found in ividuals have also been seen from Lee County. The O polycarpa entity is recognized by its extremely long spines, sometimes up to 10 cm long that are strongly retrorsely barbed, easily disarticulating cladodes, and generally strongly recurved tips of the tepals when in bud. The O polycarpa entity may form relatively large shrubs or even small treelets to 1 m tall. Although, both the O. ammophila and O. polycarpa entities are strikingly distinct in certain populations, they form a gradation of morphological characters that overlap with other populations of O. austrina includ ing growth form, spine production and color of spines, the degree of spine barbedness, cladode shape and size, and tepal shape. Hence, morphological variation within most populations of both of these
151 entities directly overlaps with those characters seen in typical O. austrina and for this reason as well as the lack of phylogenetic structure these taxa are treated as part of O. austrina Another entity of O. austrina which in contrast to the O. polycarpa and O. ammophila entities has not been formally nam ed, is noteworthy because it forms erect shrubs, which are basically miniature forms of the O. ammophila entity ranging in height from 20 30 cm tall. This entity produces copious spines and in certain specimens resembles an erect form of O. drummondii (se e below) The spines are usually strongly barbed, tuberous roots are produced, and a cylindrical trunk is also a common feature of this entity. I have collected it in Osceola and Orange counties and have seen another specimen from Lee Co Typical Opuntia austrina forms erect shrubs from 40 60 cm tall, although, with a relatively flat trunk. Plants may or may not be heavily covered with spines and the spines are slightly retrorsely barbed to the touch or oftentimes smooth. Cladodes do not disarticulate eas ily and plants are generally smaller and less robust than the O. ammophila and O. polycarpa entities. Benson (1982) included O. austrina, at the infraspecific level, in his broad concept of O. humifusa, however, phylogenetic analyses have shown that O. austrina and O. humifusa are not synonymous (Majure et al. 2012a, Chapter 6 ). Benson (1982) cited O. humifusa var. austrina (here O. austrina ) from Big Pine Key, although, the photo presented (p. 442; Fig. 443) is actually of O. abjecta not O. austrina B enson (1982) also concluded that O. pisciformis (included here under synonymy with O. austrina ) was of hybrid origin between O. humifusa and O. stricta However, characters possessed by the type specimen of O. pisciformis fall completely within the bounds of O. austrina as circumscribed here. So a hybrid origin of O. pisciformis appears dubious.
152 Additional specimens examined. United States. Florida. Brevard Co.: off of A1A SW of Jetty Park, Cape Canaveral; 17 0539480N 3141126E 16 Mar 2007 L.C. Majure 208 7 ( MISSA ). Charlotte Co.: Port Charlotte Beach State Recreation Area, N portion of park, bayside, Manasota Key 7 Mar 1991 S. Erickson PC0031 ( USF ). Citrus Co.: FL 491, ca. 1 mi E of Holder 26 Mar 1965 J. Beckner 662 ( FLAS ). Collier Co.: 1 mi E of North Naples on FL 846 7 Aug 1967 O. Lakela 30902B ( USF ). Duval Co.: Florida Beach 17 Nov 1929 H.N. Moldenke 5233 ( NY ). Flagler Co.: Flagler Beach: off of Hwy. A1A S, just E of Silver Lake 17 May 2008, L.C. Majure 3222 ( FLAS ). Gadsden Co.: Bear Creek Educa tional Forest (E of Rt. 267 and ca. 10 air mi SSW of Quincy); 30.47650N 84.62331W 28 Jun 2011 L.C. Anderson 25542 ( FSU ). Glades Co.: 7.4 km E of Charlotte Co. line, 9.7 km S of Highlands Co. line, NE corner of C 731 and FL 74 8 May 2010 A.R. Franck 2131 ( USF ). Highlands Co.: off of Hwy. 27N, ca. 21 km S of the town of Lake Placid; lower portion of Lake Wales Ridge 20 Jul 2008 L.C. Majure 3450 ( FLAS ). Indian River Co.: off of Hwy. 1S, jct. of 65th St. and Old Dixie Hwy. at Winter Beach 11 Feb 2011 L.C. Majure 4182 ( FLAS ). Lake Co.: Ocala National Forest, off of Hwy. 40 W ca. 4 km W of Aster Park and 2 km E of jct. with Hwy. 19; along roadside; 24 May 2008 L.C. Majure 3246 ( FLAS ). Lee Co.: Wulfert, w estern Sanibel 28 Mar 1973 W.C. Brumbach 8290 ( FLAS ). Manatee Co.: South Fork State Park 4 Apr 1992 R. Owens SF0050 ( USF ). Marion Co.: Ocala National Forest, Salt Springs off of Hwy. 19S at jct. with Hwy. 314 24 May 2008 L.C. Majure 3244 ( FLAS ). Miami Dade Co.: Miami 28 Oct 28 Nov 1903 J.K. Small s.n. ( NY ); Miami 1 Feb 1911 J.K. Small s.n. ( NY ). Okeechobee Co.: 0.25 km S of the Okeechobee County line, off of Hwy. 441W 11 Feb 2011 L.C. Majure 4185 ( FLAS ). Orange Co.: 3.5 mi SE of Hwy. 528 along Hwy. 520 under powerline; 17 0504805N 3142822 E 16 Mar 2007 L.C. Majure 2086 ( MISSA ). Osceola Co.: off of Hwy. 441S (192) S of St. Cloud at Harmony 28 Mar 2009 L.C.
153 Majure 3702 ( FLAS ). Palm Beach Co.: Near Boca Raton Airport, E of Interstate 95; NW 40th St. & 6th Way 9 Mar 2010 L.C. Majure 3970 ( FLAS ). Pinellas Co.: sand dunes, Long Key 28 Nov 1921 J.K. Small s.n. ( NY ). Polk Co.: S of Frostproof along Rt. 27 ca. 3.6 mi N of Highlands Co. line 1 Nov 1980 W.S. Judd 2841 ( NY ). Putnam Co.: Ocala National Forest, Delancy Lake, along FR 75 2, just W of Hwy. 19 24 May 2008 L.C. Majure 3248 ( FLAS ). Seminole Co.: off of Hwy. 419S, W of Mills Lake Park; 17 0487398N 3166757E 16 Mar 2007 L.C. Majure 2085 ( MISSA ). St. Johns Co.: Crescent Beach 1 Jan 1942 H. Kurz 279 ( MICH ). St. Lucie Co.: Ancient Dunes, near Ft. Pierce 6 Sep 1922 J.K. Small s.n. ( NY ). St. Lucie Co.: v icinity of Ft. Pierce, off of Hwy. A1A, E of Jack Island 28 Mar 2009 L.C. Majure 3705 ( FLAS ). Volusia Co.: off of SR 40 ca. 3 km W of jct. with Hwy. 11N 18 May 2008 L.C. Majure 3 232 ( FLAS ) 4. Opuntia cespitosa Raf. Bull. Bot. Seringe. 216. 1830. TYPE : United States. Kentucky Woodford County, Hwy. 60 N at jct. of Hwy. 62; just N of Versailes, L.C. Majure 3275 with B. Patenge 9 Jun 2008 ( neotype, here designated: FLAS!; isoneotype, US!; see Fig. 7 7A ). Opuntia rafinesqueii Engelm. var. microsperma Engelm. Proc. Amer. Acad. 3: 295. 1856. Opuntia mesacantha Raf. var microsperma (Engelm.) J.M. Coult. Contr. U.S. Natl. Herb. 3: 429. 1896. Opuntia humifusa (Raf.) Raf. var. microsperma (Engelm.) A. Heller. Cat. N. Amer. Pl., ed. 2. 8. 1900. Opuntia compressa (Salisb.) MacBride var. microsperma (Engelm.) L.D. Benson, Proc. Calif. Acad. Sci. 4th ser. 25: 250. 1944 T YPE : United States. Missouri. Cultivated in Missouri Botanic a Garden 1854 (lectotype designated by Benson ( 1982 ) : MO!). Opuntia rafinesqueii Engelm. var. minor Engelm. Proc. Amer. Acad. 3: 295. 1856.
154 Opuntia mesacantha Raf. var. parva J.M. Coult. Contr. U.S. Natl. Herb. 3: 429. 1896. nom. superfl. Opuntia humifusa (Raf.) Raf. var. parva (Coult.) A. Heller. Cat. N. Amer. Pl., ed. 2. 8. 1900. nom. superfl. Opuntia humifusa (Raf.) Raf. subsp. minor (Engelm.) R. Crook & Mottram. Bradleya 16: 135. 1998. T YPE United States. Missouri. Sandstone rock in southern Missrouri. Engelmann s.n. (lectotype designated by Benson (1982): MO!). Sprawling shrub, to 0.3 m tall, with chains of up to 2 6 cladodes, the cladodes generally produced with the flat (broad) surface parallel to the ground surface; roots fibrous or tuberous, apparently depending on the substrate. Cladodes mostly obovate, rotund, or ell iptical in outline, margins not scalloped, with 4 6 (generally 5) areoles per diagonal row, cladodes strongly glaucous green (gray green) when developing, aging dark green or light gray green, cross wrinkling during the winter months, 10.5 (3.8 18.7) cm lo ng, 8.0 (3.2 11.3) cm wide, 10 (4 19.2) mm thick. Leaves glaucous, gray green, ascending parallel to the cladode surface or slightly spreading, 6.0 (5.5 6.8) mm long. Glochids dark red, crimson red, or dark amber, aging light to dark brown. Spines robust o r delicate, smooth to the touch, 1 2 (3) per areole (most common ly 1), 2.9 (1.5 4.3) cm long, these castaneous at the base during development but maturing bony whit e, and finally dark gray in age, typically spreading in one plain from the areoles (i.e., in line with one another) with primarily 1 spine, or occasionally 2 of roughly the same length or 1 long and 1 short and slight l y deflexed (these characters can be seen in individuals in the same population or even on the same plant ), rarely 3 spines produce d from the areoles, but in this case the central spine t ypically not porrect ( as in O. macrorhiza ); i n age the mid cladode and especially the basal cladode spines tend to deflex. Flowers: outer tepals triangular to ovate, inner tepals 9 10, 3.0 (2.5 5.5) c m long, basally tinged dark red, crimson, orange red, or reddish pink,
155 obovate with a mucronate tip, glaucous green, staminal filaments yellow, reddish basally, stigmas white to cream, lobes 6 10. Berries d ark red, or orange red, 3.9 (2.7 4.5) cm long. See ds 5.1 (4.9 5.4) mm long, funicular girdle 1.1 (0.95 1.3) mm wide, funicular envelope bumpy from the enlargement of the cotyledons and hypocotyl, the funicular girdle also tends to be slightly irregular or bumpy. Phylogenetic placement Opuntia cespitosa is an allopolyploid derivative of the southwestern O. macrorhiza species complex (SW clade) and the O. humifusa species complex (SE clade) of the southeastern United States. The southeastern progenitor of O. cespitosa was most likely the tetraploid, O. hu mifusa subsp. pollardii or an ancestor thereof, which was derived solely from the SE clade ( Chapter 6 ). Phenology. Flowering time for O. cespitosa appears to be directly related to latitude, with more southerly populations blooming before more no r therly one s. For instance, plants growing in central Mississippi generally begin flowering around the first or second week of May, while material from Michigan and Wisconsin flowers around late June or early July (see Introduction). Distribution Opuntia cespito sa is the most common species in the eastern United States occurri ng mostly west of the Appalachia n Mountain s west to Wisconsin, Iowa, Missouri, Arkansas, and eastern Texas, south to Mississippi and Alabama, and north to Michigan in the United States, and also in southeastern Ontario, Canada. Populations are occasionally found in the eastern Appalachia n s as well (Fig. 7 8). Habitat. Opuntia cespitosa is most commonly found in sandy or blackland prairies, juniper glades, or growing on rock outcrops (genera lly limestone or sandstone). It is commonly associated with Juniperus virginiana, Ratibida pinnata, Rhus aromatica, Xanthoxylum clava herculis among many other species.
156 Ploidy Opuntia cespitosa is tetraploid 2 n =44, throughout its range (Majure et al. 2 012 b ). Notes Engelmann (1856) was the first to truly recognize the difference between O. cespitosa and O. humifusa although, he recognized O. cespitosa under the superfluous name, O. rafinesquei, apparently in an attempt to reconcile the taxonomic confu sion surrounding the Opuntia humifusa at the time was recognized as Opuntia vulgaris. Central United States populations (in Arkansas, Iowa, Illinois, Michigan, Missouri, and nearly all populations in Wisconsin) often show e vidence of introgression with the eastern flank of O. macrorhiza as they have spreading spines in more than one plain and occasionally one small, bristle like radial spine produced at the base of the areole (e.g., MI, Musekegon Co.: L.C. Majure 3259 ; WI, Dane Co.: D. Ugent 60 11J ). There also is apparent introgression with O. humifusa at the eastern boundary of the two species (e.g ., eastern NY Orange Co.: H.M. Dunslow s.n. Nantucket Island, MA ), and populations in Bibb County, Alabama (e.g., L.C. Majure 2042 ) are nearly identical to O. humifusa subsp. pollardii except for the red centered flowers, lack of strong barbs on the spines, and the typical rotund cladodes of O. cespitosa Additional specimens examined. Canada. Ontario. Essex Co.: Pelee Island, W side of Fish Point, near N end 9 May 1981 A.A. Reznicek 6230 ( MICH ). Kent Co.: Harwich Township, Bethel Cemetery, UTM 107042, map 40J/8, square 17MT10 6 Sep 1986 M.J. Oldham 6867 (note: most likely planted, originating from Pelee Island) ( MICH ). United States Alabama. Bibb Co.: o ff of Hwy. 219N from Hwy. 5 N at jct. with Schultz Cr. Rd.; 16 0486461E 3653643N 7 Mar 2007 L.C. Majure 2042 ( MISSA ). Colbert Co.: off of Natchez Trace, just S of jct. with Hwy. 72, 34.7525N 88.0269W 25 Jul 2007 L.C. Majure 2610 ( MISSA ). Franklin Co.: along Spruce Pine Hwy., ca. 4 mi S of Russelville 30 Aug 1966 R.C. Clark 8004 ( UNC ). Jackson Co.:
157 Hwy. 79 0.2 mi N of jct. with US Hwy. 72 8 Jul 1966 R.C. Clark 4582 ( UNC ). Lawrence Co.: Prairie Grove Glades,3.6 km NE of Mt. Hope, 34.4859N 87.5005W 25 Jul 2007 L.C. Majure 2609 ( MISSA ). Limestone Co.: Elkmont 14 Jul 1913 E.G. Holt 13 64 ( NY ). Marshall Co.: 7 mi W of Guntersville on Georgia Mt. 9 Apr 1966 D.H. Brown 7 ( UNA ). Morgan Co.: off CR 55 in Massey east of Emmanuel Church, Moulton Valley district of the Highland Rim section; 3422'12''N 8701'12"W 6 Jul 2003 D.D. Spaulding 11977 ( UNA ). Perry Co.: Uniontown 1 Jan 1912 A.H. Howell 12 47 ( NY ). Arkansas. Bradley Co.: Warren 21 May 1937 D. Demaree 15046 ( NY ). Garland Co.: Ouachita Nat'l Forest, N of FS r d. 130 and Cedar Fourche Landing of Lake Ouachita; 34.6660N, 93.2834W 6 Apr 2007 L.C. Majure 2198 ( MISSA ). Grant Co.: 5 Jun 1 940 D. Demaree 21180 ( MO ). Hempstead Co.: near Tokis, 22 Oct 1932 D. Demaree 10024 ( US ). Hot Springs Co.: Magnet Cove 10 Oct 1937 D. Demaree 16493 ( MO ). Independence Co.: along Pine Hollow Rd. where it crosses Lafferty Creek, 3.5 mi W of Cushman, Sec. 14, T14N, R8W 6 Jun 1968 R.D. Thomas 8023 ( TENN ). Izard Co.: Guion 1 6 Aug 1913 W.H. Emig 187a ( MO ). Marion Co.: opposite Cotter (in Baxter Co.) 15 Sep 1960 G.N. Jones 31045 ( ILL ). Miller Co.: Stateline Rd. S of jct. with Hwy. 134 on W side of Miller County Sandhills Natural Area, 33 11.137N, 94 2.569W, 2 Oct 2008 B. Snow 2062 ( FLAS ). Pulaski Co.: Levy 4 Nov 1931 D. Demaree 8849 ( NY ). Saline Co.: Just N of Detonti; E of Bauxite Cutoff Rd.; 34.5300N, 92.5043W 6 Apr 2007 L.C. Majure 2194 ( MISSA ). Washington Co.: White River, Fayetteville G. Engelmann 931 1 Jun 1835 ( MO ). Yell Co.: 1.95km NNW of Dardanelle, just W of Arkansas River 1 Oct 2011 G.P. Johnson s.n. ( FLAS ). Connecticut New Haven Co.: M ilford, exposed ledges 8 July 1892, E.H. Eames s.n. ( ILL ). Illinois. Adams Co.: Mississippi Bottom SE of Quincy 15 Jul 1943 R. Brinker 2824 ( ILLS ). Calhoun Co.: Cap au Gris Hill Prairie, 2.5 mi SE of Batchtown, T15N, R2W, Sec. 29 4 Jun
158 1987 K. Robertson 4514 ( ILLS ). Cass Co.: Chandlerville 7.5 min topo map 6 Jul 1994 L.R. Phillipe 24911 ( ILLS ). Crawford Co.: SE of Palestine 26 Nov 1949 R.A. Evers 22044 ( ILLS ). Fayette Co.: bluffs of Dismal Creek NE of Laclede 23 Jul 1947 R.A. Evers 5587 ( ILLS ). Gallatin Co.: sandstone outcrop NE of The Pounda, SW of Gibsonia 24 Sep 1947 R.A. Evers 8550 ( ILLS ). Hardin Co.: above Ohio River, 1 mi E of Rosiclare 29 May 1949 G.S. Winterringer 1963 ( ILL ). Henderson Co.: N of Oquawka R.A. Evers 14786 18 Sep 1948 ( ILLS ). Jackson Co.: cliff summits in Giant City State Park, SE Jackson Co.: 1 Oct 1931 H.S. Pepoon s.n. ( ILLS ). Jersey Co.: Riehl Station A.H. Horrell 09 168 30 May 1909 ( US ). Jo Daviess Co.: Savanna Ar my Depot, Green Island 7.5 min q uad E of Building D 107 18 Jul 1996 L.R. Phillipe 27862 ( ILLS ). Johnson Co.: Marion quad, R2E T11S 1.75 mi S of Goreville, E of Dunntown school 27 Jul 1931 J. Schopf 858 ( ILLS ). La Salle Co.: SW of Naplate 30 Aug 1974 R.A. Evers 113834 ( ILLs ). Lake Co.: N of Waukegan and E of the glacial Glenwood Ridge 1 Jul 1908 F.C. Gates 2802 ( ILL ). Lee Co.: near Amboy 8 Jul 1956 J.B. Long 293 ( ILL ). Mason Co.: 0.2mi S of Batts, on St. Rt. 78 on 600N hea ding W turn on 1400E 7 Jun 2007 R. Altig s.n. ( MISSA ). Menard Co.: Athens E. Hall s.n. 1 Jan 1862 ( NY ). Mercer Co.: SE of Keithsburg 2 Jul 1964 R.A. Ev ers 80855 ( ILLS ). Monroe Co.: r ock ledges, 3 mi S of Valmeyer 24 May 1950 R.A. Evers 23057 ( ILLS ). Morgan Co.: S of Meredosia 31 May 1947 R.A. Evers 3521 ( ILLS ). Perry Co.: N of Pinckneyville 24 Sep 1 953 R.A. Evers 41629 ( ILLS ). Pike Co.: rock ledge of Kinderhook 7 Sep 1949 R.A. Evers 20829 ( ILLS ). Pope Co.: Brownfield Quad R5E, T12S, Pine Hollow 2.5 mi E of Dixon Springs, 18 Aug 1931 J. Schopf 1360 ( ILLS ). Putnam Co.: 1 mi S of Hennepin 15 Jul 1955 R.A. Evers 7661 ( ILLS ). Randolph Co.: r ock ledges, 1 mi N of Prairie du Rocher 24 May 1950 R.A. Evers 23137 ( ILLS ). Scott Co.: 3.5 mi W of Winchester T14N R13W Sec. 27 NW 1/4 23 Jun 1981 K.R. Robertson 2607
159 ( ILLS ). St. Clair Co.: on bluff top 2.5 mi S of Falling Springs near Dupo 31 Aug 1 947 J. Neill 1427 ( ILLS ). Tazewell Co.: near Spring Lake, 10 mi SW of Pekin 6 Jun 1948 H.R. Hoehn s.n. ( ILL ). Union Co.: r ocky slopes of Pine Hills, SE of Aldridge 2 Oct 1948 R.A. Evers 15475 ( ILLS ). Whiteside Co.: along CB&Q & CMSP&P RRs, N of Fulton 29 Jan 1968 C.J. Sheviak 206 ( ILL ). Will Co.: Kankakee River Watershed, Sand Ridge Savanna Nature Preserve 2.7 mi E of Braidwood (jct. o f Rt. 53 and 113) along Rt. 113, in the Kankakee Sand Area Section of the Grand Prairie Natural Division, 41.26025N 88.1662W 13 Aug 2007 L.R. Phillipe 40000 ( ILLS ). Indiana. Adams Co.: Sec. 20, T3S, R8W 22 Jun 1943, R.A. Evers 1205 ( NY ). Clark Co.: 1 mi E of Charlestown 9 May1953, F.B. Buser 3047 ( ILL ). Fountain Co.: Covington 1 Aug 1953, F.B. Buser 3167 ( ILL ). Jasper Co.: Walker TP and SJ Wheatfield, 8 Jul 1924, W. Welch 619 ( ILL ). Jefferson Co.: just N of Madison, 22 Jun 1913, C.C. Deam 13412 ( MO ). Lake Co.: Long Lake 20 Jul 1927, W.B. Welch 5616 ( NY ). Porter Co.: just S of Lake Michigan off of Hwy. 12 8 Jun 2008, L.C. Majure 3274 ( FLAS ). Tippecanoe Co.: SW of Lafayette 17 Jun 1941, C.M. Ek ( NY ). Iowa. Muscatine Co.: E of Cedar River R.R. bridge, W of Bayfield 3 Jul 1915, B. Shimek ( NY ). Kentucky. Anderson Co.: off of Hwy. 127S, ca. 1 mi S of jct. with Bluegrass Parkway 9 Jun 2008, L.C. Majure 3276 ( FLAS ). Boyle Co.: just SW of Perryville 27 Aug 1959 C.F. Reed 45203 ( MO ). Breckinridge Co.: W of Cloverport, Rt. US 60, Breckinridge 17 Aug 1961 C.F. Reed s.n. ( MO ). Caldwell Co.: Pennyrile St. Park, 22 Jun 1966 G.E. Hunter 1676 ( UNC ). Clark Co.: 1.5 mi SE of Indian Fields 31 Aug 1939 H.A. Gleason, Jr. 115 ( MICH ). Crittendon Co.: Rt. 60, 1 mi N of Mattoon 23 Jun 1974 C.F. Reed 138388 ( MO ). Cumberland Co.: ca. 0.25 mi W of KY Hw y. 704; ca. 1 mi S of Adair Cumberland county line 15 Apr 1999 R.C. Clark 24276 ( EKY ). Edmonson Co.: Mammoth Caves 7 Jun 1949 C.F. Reed 15304 ( MO ). Franklin Co.: Dad's farm, 10 Oct 1980 S. Rice FR 109 ( EKY ). Garrard Co.: W bank Paint Lick
160 Creek, and directly W of Geo. Caldwell farm, thi s on KY 21, 1.7 mi E of jct. r d. and KY 52, Blue Grass Province 5 Oct 1963 E.M Browne 8150 ( EKY ). Hancock Co.: USGS Topo, Cloverport 3786 86, 450 Jeffry Cliff entered from US 60 on the gravel rd. 1.35 km SE of US 60 and KY 1406 jct. 18 Jun 1980 R. Hannan 4259 ( EKY ). Henry Co.: 1.4 mi SW of Lockport, 6 Jun 1962 J.L. Gentry, Jr. 301 ( NY ). Jefferson Co.: Goose Creek, 30 Apr 1947 Davies s.n. ( UNC ). Jessamine Co.: Jessamine Cr.; Blue Grass Province 17 Jun 1961 E.M. Browne 4221 ( EKY ). Johnson Co.: C. Ferguson farm near Flat Gap, Mud Lick, rd. to Paintsville 22 Jun 1949 O. McKenzie s.n. ( MO ). Lyon Co.: western KY, Kuttawa 2 18 Jun 1909 W.W. Eggleston s.n. ( NY ). Madison Co.: G Caldwell farm, 1.7 mi E of Paint Lick on KY 21 5 Oct 1963 E.M Browne 8130 ( EKY ). Owen Co.: Gilbert Tract WMA; Brown Bottom. 3 Oct 2003 R.L. Jones 9542 ( EKY ). Pike Co.: Old US 460, 4.6 mi E of jct. t his rd. and KY 80 near Fishtrap Dam 9 Jun 1964 E.M. Browne 8641 ( EKY ). Pulaski Co.: Pumpkin Hollow, Burnside Q., E of Burnside, Rte N side of Lake Cumberland in Williams Bend in Pumpkin Hollow 11 Apr 1979 R. Hannan 1190 ( EKY ). Warren Co.: Bowling Green 1 Jan 1899 S.F. Price s.n. ( MO ). Wayne Co.: Cooperville Rd., 15 Apr 1940 E.L. Brau n 2814 ( US ). Maryland. Baltimore Co.: Factory Rd., 0.5 mi N of Harford Rd. 10 Dec 1981, C.F. Reed 121328 ( MO ). Washington Co.: Kemps Mills N of Williamsport 14 Jun 1952, C.F. Reed 29115 ( MO ). Massachusetts. Nantucket Co.: Nantucket Island, Coatue Point 1 Sep 1964, F.C. MacKeever ( NY ). Michigan. Allegan Co.: 3 mi W of Allegan, summit of bluffs above Kalamazoo River, 15 May 1950 R. McVaugh 11260 ( MICH ). Manistee Co.: S end of Maple Grove Township Cemetery immediately E of Kaleva on the N side of Nine Mile Rd; SE quarter of Sec. 21, T23N, R14W 30 Jul 1978 A.B. Johnsen 1596 ( MICH ). Monroe Co.: S side of Tunnicliffe Rd., ca. 5 mi SE of Pet ersburg 7 May 1992 A.A. Reznicek 8931 ( MICH ). Muskegon Co.: off of Hwy. 31N, 6.2 KM SE of Whitehall, 2 km
161 NW of Lakewood 2 Jun 2008 L.C. Majure 3259 ( FLAS ). Newaygo Co.: off of Hwy. 31N; ca. 8.3 km NE of Newaygo and 6.7 km SE of White Cloud 2 Jun 2008 L.C. Majure 3260 ( FLAS ). Oceana Co.: off of Hwy. 20, W of jct. with 132 Ave. 2 Jun 2008 L.C. Majure 3262 ( FLAS ). Van Buren Co.: 0.03 mi N of 28th Ave. and 0.39 mi W of 77th St., 4218'57.795N 8617'57.904W 14 Jul 2008 T.L. Walters 11972 ( MICH ). Mississippi. Carroll Co.: Holly Property adj. to Hwy 82 W 8 Mar 2005 L.C. Majure 799 ( MISSA ). Clay C o.: off of Herman Shirley Rd., ca. 0.25 mi S of Hwy. 50 16 May 2006 L.C. Majure 1442 ( MISSA ). Holmes Co.: off of Hebron Rd., Loess Hills, 33.07420 90.15976 15 May 2007 L.C. Majure 2365 ( MISSA ). Lee Co.: Tombigbee State Park, W side of lake shore 30 Dec 2005 L.C. Majure 1292 ( MISSA ). Lowndes Co.: Old West Point Rd., ca 0.5 mi E of Catalpa Cr. 18 Dec 2004 L.C. Majure 736 ( MISSA ). Madison Co.: Natchez Trace Parkway, W.B. McDougall 1651 20 May 1948 ( US ). Montgomery Co.: Sam Marter prope rty, N of C R 404 13 Jan 2005 L.C. Majure 768 ( MISSA ). Noxubee Co.: off Hwy. 14 behind St. John's Church 25 Jun 2005 L.C. Majure 1543 ( MISSA ). Oktibbeha Co.: property of Anne Daniels, just W of Hwy. 389, near Trim Cane Creek, N of Starkville 6 May 2006 L.C. Majure 1380 ( MISSA ). Pontotoc Co.: v icinity of Troy, off of Shannon Rd. 7 Jun 2006 L.C. Majure 1519 ( MISSA ). Scott Co.: S of town of Forest off of Hwy. 501S, ca. 1km NW of Norris, 32.3000N, 89.4448W 25 Jun 2007 L.C. Majure 2563 ( MISSA ). Tishomingo Co .: J.P. Coleman State Park, E of boat launch at end of Steel Br. Rd., 24 May 2007 H. Sullivan s.n. ( MISSA ). Missouri. Adair Co.: terrace on E side of Chariton River, ca. 2.5 mi N of State Hwy. 6 on W side of dirt CR, ca 5 mi NW of Kirksville ; T63N, R16W, S15/22 boundary 29 Jun 1994 D. Ford 719 ( MO ). Barry Co.: Eagle Rock B.F. Bush 614 25 Sep 1896 ( MO ). Cape Girardeau Co.: South of Delta 19 May 1926 R.E. Woodson s.n. ( MO ). Carter Co.: along Hwy. 103, just W of Ozark National Scenic Riverways entrance gate, 3 May
162 1993 B. Summers 5607 ( MO ). Crawford Co.: Savanna Ridge Glade, T40N R2W sec. 23 NE1/4 5 Jun 1997 C.E. Darigo 2888 ( MO ). Greene Co.: vicinity of Willard 30 Aug 1912 P.C. Standley 9634 ( US ). Harrison Co.: T66N, R26W, S10 7 May 1985 P. Delozier 1722 ( MO ). Henry Co.: bluffs of Grand River, 3 mi NE of Piney, near Benton Co. line 8 Oct 1934 J.A. Steyermark 15977 ( MO ). Howell Co.: Tingler Lake Conservation Area, ca. 7 mi S of West P lains on S Fork of Spring River, T22N R08W S06 N1/4 23 Ma y 1997 B. Summers 8132 ( MO ). Jasper Co.: Joplin, barrens on Turkey Creek 19 May 1909 E.J. Palmer 2278 ( MO ). Jefferson Co.: 5 mi SE of Catawissa 22 Jun 29 J.A. Steyermark 1279 ( MO ). Laclede Co.: Sweet Hollow Creek, 3 mi W of Eldridge, off Hwy. NN, T36N R17W S28 9 Oct 1991 B. Summers 4759 ( MO ). Lincoln Co.: west of bridge on Cuivre River 23 Oct 1982 M.R. Crosby 14618 ( MO ). Monroe Co.: near Victor 27 Jun 1933 E.J. Palmer 40740 A ( MO ). New Madrid Co.: off of I 55N at mi 61.2, S of S ikeston, 36.80077 89.53167 25 May 2007 L.C. Majure 2435 ( MISSA ). Newton Co.: Reding's Mill, 12 Aug 1908 B.F. Bush 5072 ( MO ). Phelps Co.: Duke 1 Aug 1913 W.H. Emig 187 ( MO ). Scott Co.: off of I 55N beside Best Western Hotel, Miner and N Sikeston; 36.8883N, 89 .5320W 26 May 2007 L.C. Majure 2441 ( MISSA ). Shannon Co.: Ozark National Scenic Riverways, Jerktail Mt., ca. 8 mi des NE of Eminence, 3713'30"N, 9118'00"W 25 Oct 1996 C. Dietrich 461 ( MO ). Taney Co.: Mark Twain National Forest, Ozark T23N R18W, along Blair Ridge Rd. 2 Jun 1978 J.L. Hicks 988 ( MO ). Washington Co.: 7 mi N of Potosi on st ate rd. F at Mineral Fork Creek, 38 00'N, 904.9'W 19 Apr 1990 J.S. Miller 4879 ( MO ). New York. Columbia Co.: Hudson, NY 25 Aug 1904, G.T. Hasting s ( NY ). Orange Co.: Coronham 7 Aug 1912, H.M. Dunslow ( NY ). O hio. Adams Co.: Hwy. 52E along the Ohio River (N of river and Hwy. 52) in graveyard of Sand Springs Church 1 Jun 2008 L.C. Majure 3251 ( FLAS ). Brown Co.: Huntington Township, above Aberdeen, Ohio station # 1. 1
163 Apr 1975 J. Bryant s.n. ( MU ). Erie Co.: Cedar Point, Sandusky, Ohio; in the sand near the new Lake Laboratory 1 Jul 1903 A. Wetzstein s.n. ( MU ). Gallia Co.: off of Hwy. 141 E just (ca. 1 km) E of Raccoon Creek, 1 Jun 2008 L.C. Majure 3252 ( FLAS ). Hamilton Co.: Ft. Dennison, E of Cincinnati 10 Sep 1922 E.T. Wherry s.n. ( NY ). Jackson Co.: dry cliff 0.5 mi west of Jackson 14 Jun 1936 Bartley 59 ( NY ). Lucas Co.: b eside Swan Creek, along Oak Openings Preserve, just E of Wilkins Rd., S of Reed Rd., and W of Berkeley Southern Rd. 1 Jun 2008 L.C. Majure 3254 ( FLAS ). Ottawa Co.: Cedar Point 23 Jun 1894 W. Worra llo s.n. ( MU ). Woods Co.: off of Zepernick Rd. from Hwy. 6 E ca. 8 mi E of Bowling Green, 1 Jun 2008 L.C. Majure 3253 ( FLAS ). Pennsylvania Lancaster Co.: Peach Bottom 26 Jun 1893, Castez s.n. ( NY ). Tennessee. Bledsoe Co.: Off of Lowes Gap Mt. Rd., vicinity of Litton, 21 Dec 2006 L.C. Majure 1938 ( MISSA ). Blount Co.: Cave Springs near Townsend 14 Aug 1942 A.J. Sharp S 4247 ( TENN ). Cannon Co.: jct. o f Poplar Bluff Rd. (TN hwy. 96) and Hurricane Cr. Rd.; 16 583120N 3980169E 9 Mar 2007 N. Sondermann s.n. ( MISSA ). Cheatam Co.: just W of Nashville along Hwy. 24W, ca. 0.5km E of exit 40 (Old Hickory Blvd.) 10 Jun 2008 L .C. Majure 3280 ( FLAS ). Davidson Co.: Knapp Farm about 500 yards E of Mills Creek 13 Jun 1937 N.H. Woodruff s.n. ( TENN ). De Kalb Co.: 36.05907N, 85.80175W, Edgar Eva ns State Park, Center Hill Dam q uad. 15 Sep 2002 L.R. Phillippe 34887 ( TENN ). Decatur Co.: eastern shore of TN River, 13 mi above Perryville 22 Aug 1907 W. Clark 456 ( US ). Fayette Co.: Gordon Hill area, S of Lagrange 4 Oct 1980 V. Bates, Jr 1925 ( TENN ). Fentress Co.: 19 Apr 1949 N.C. Fassett 27936 ( TENN ). Franklin Co.: along Old Stage R d., SW of Cowan about 4 mi 2 Jun 2001 M. Rhinehart s.n. ( TENN ). Giles Co.: S W of Pulaski, on north side of Hwy. 11, 100 yards N of Circle Rd. 14 May 2001 D. Estes 1626 ( TENN ). Hamilton Co.: off of Hwy. 321, 21 Dec 2006 L.C. Majure 1937 ( MISSA ). Haywood Co.: SE corner of the Hillville Loop Rd. 20 Mar 1984 P.
164 Lewis 1769 ( TENN ). Jackson Co.: Flynn Creek Rd., about 9 mi W of Rt. 56 11 Jun 1992 V.E. McNeilus 92 545 ( TENN ). Jefferson Co.: Friends Station, 1 mi W of New Market 17 Jun 1934 M. Weaver 1586 ( TENN ). Knox Co.: 4 mi from Knoxville, on old Sevierville Rd. 5 Jun 1934 S.A. Cain 555 ( TENN ). Lewis Co.: Natchez Trace Parkway, Jacks Branch pulloff; 35 24' 50''N 87 30' 59''W 19 Jun 2009 J.G. Hill s.n. ( FLAS ). Marion Co.: High limestone bluff s N of Lee Highway Bridge; Cedar Mtn. 1 Jan 2003, J. Beck 4398 ( TENN ). Marshall Co.: j ust W of Henry Horton State Park, South side of Caney Springs Rd., at a point ca. 0.7 mi W of Hwy. 31A; 3536'10''N, 86 42'43"W 6 Jun 2005 M. Rhinehart s.n. ( TENN ). Maury Co.: off of I 65 N; mi marker no. 32 51 20 Jul 2007 N. Sondermann s.n. ( MISSA ). Roane Co.: Pisgah Ridge, facing Watts Bar Lake at riv er mi 570.8 (Bacon Gap q uad) 14 Jun 1 984 B.E. Wofford 84 40 ( TENN ). Rutherford Co.: Rt. 99, between Murphreesboro and Rockvale 11 Oct 1958 A.J. Sharp 25493 ( TENN ). Shelby Co.: S of Cordova, 20 Aug 1947 A.J. Sharp 6605 ( TENN ). Smith Co.: Caney Fork; Upchurch Rd., ca. 4.9 mi W of jct. with Horseshoe Bend Rd., due E of Carthage, TN 8 May 1999 T.J. Weckman 4888 ( EKY ). Sumner Co.: 5 mi N of Gallatin 13 Oct 1968 K.E. Rogers 42903 ( TENN ). Wayne Co.: 2.2 mi E of Clifton Junction, N side of Hwy. 64; 35 18.416N, 87 54.276W 3 Aug 2008 B. Snow 2061A (FLAS). Wilson Co.: S end of Lebanon 8 Aug 1968 K.E. Blum 2863 ( FL AS ). Virginia. Fredrick Co.: off of Hwy. 50 W at Hayfield; jct. of N Hayfield St. and Hwy. 50 W 30 May 2009, L.C. Majure 3806 ( FLAS ). Page Co.: Luray 20 Sep 1926, E.T. Wherry s.n. ( US ). Wythe Co.: Barren Springs, Rt. 100 C.F. Reed 97725 6 Mar 1975 ( MO ). West Virginia. Cabell Co.: sandy field, near Milton 9 Oct 1935, L. Williams 366 ( MO ). Wisconsin. Columbia Co.: sandy roadside near Lake Wisconsin 24 Sep 1955 R.C. Koeppen s.n Dane Co.: off of Hwy. 78S just N of Mt. Horeb, (T7N, R6E, Sec. 26) 21 Jul 1960 D. Ugent 60 11J ( WISC ). Grant Co.: sec. 5, Woodman Township, T7N, R4W; W side of
165 Hwy. 132 N of Mt. Hope 4 Oct 1968 L.J. Musselman 2291 ( WISC ). Green Co.: 1 mi off county Hwy. on Sawmill Rd. (T4N, R6E, Sec. 20) 4 Dec 1970 J. Cram 27 ( WISC ). Green Lake Co.: Granite Knobs, Marquette 18 Sep 1929 N.C Fassett 9203 ( WISC ). Jackson Co.: near Rezin Marsh, 23 mi SE of Black River; Falk (Sec. 25, T26N, R1W) 20 Aug 1947 D.F. Grether 6618 ( WISC ). La Crosse Co.: Trunk quad, T18N, R6E, Sec. 9; Farmington Tw p. 28 Jun 1956 T.G. Hartley 923 ( WISC ). Marquette Co.: Budsin Corner (T17N, R10E, Sec. 28) 9 Aug 1960 D. Ugent 60 16 ( WISC ). Monroe Co.: Dalton Ave. 0.75 mi E of hwy. 71; T16N, R3W, Sec. 3 4 Sep 1988 J.M. Graber 310 ( WISC ). Richland Co.: ca. 1 mi NW of Hub City, off of Hwy. 80S 8 Jun 2008 L.C. Majure 3273 ( FLAS ). Sauk Co.: vic i n ity o f Troy (T9N, R5E, Sec. 35, SW1/4) 17 Jul 1960 D. Ugent 60 8c 8a ( WISC ). Waushara Co.: T18N, R9E, Se c. 30, SE of Coloma and SW of Richford W of Hwy. 22 1 Sep 1959 R. F. Pochmann 15152 ( WISC ). 5 Opuntia drummondii Graham in Maund., Botanist 5: 246.1846. T YPE : United States. Florida. St. Johns Co.: FL, dunes 5 mi S of Ponte Verde, 2 Sep 1954, L. & R.L. Benson 15388 (neotype designated by L.D. Benson (1982); POM!; see Fig. 7 9A). Opuntia pes corvi LeConte ex Engelmann, Proc. Amer. Acad. 3: 346. 1856. T YPE : United States. Florida. [Franklin Co.:] Apalachicola, FL, April, July, Nov., 1860, Chapman s.n. (neotype designated by L.D. Benson (1982); MO!). Opuntia frustulenta Gibbes Proc. Elliott Soc. Nat. Hist. 1: 273. 1859. T YPE : United States. South Carolina. Charleston Co.: Folly Island, near Charleston, 15 Feb 1916, J.K. Small s.n. (neotype designated by L.D. Benson (1982); US!). Opuntia tracyi Britton Torreya 11: 152. 1911. T YPE : United States. Mississippi. [Harrison Co.:] Coast, Biloxi, 11 May 1911, S.M. Tracy s.n. (holotype: NY!).
166 Small shrubs 0.2 0.3 m tall, often forming large mounds as a result of disarticulating stems coupled with a high degree of branching, typically consist of numerous (3 5 or more) radiating branches (Fig. 7 9B) from a thick but shallow rootstock; older stems with a thin scaly bark; roots are mostly fibrous but commonly expand in girth for a short distance proximally (Fig. 7 9E). Cladodes typically cylindrical, but flattened as well, especially in larger basal cladodes, dark green, or yellow green, not glaucous, sma ll relative to other species, 3.6 (0.8 11.1) cm long, 1.8 (0.6 3.4) cm wide, 10.4 (5.3 14.8) mm thick, elliptical, oblong, or rounded in shape, with 1 2 areoles per diagonal row at midstem, the terminal cladodes easily disarticulating at the nodes, leading to frequent vegetative dispersal. Leaves green, 2.8 (2.3 3.5) mm long, spreading or ascending parallel to cladode. Glochids stramineous, usually exserted and conspicuous. Spines 3.0 (1.5 4.9) cm long, 0.6 (0.2 0.9) mm in diameter, dark brown or mottled br own and white during development aging white and finally gray, the b asal cladodes usually producing spines throughout their lifetime with up to 5 spines per areole, the t erminal cladodes usually with 2 3 spines per areole; a ll spines are strongly retrors ely barbed, but spines on the terminal, easily disarticulating cladodes with more pronounced barbs, presumably aiding in vegetative dispersal. Flowers: outer tepals green or yellow green, triangular or triangular ovate, erect and generally incurved in bud, generally small, inner tepals 8, dark ye llow or occasionally light sulf ur yellow, obovate with a mucronate tip, 2.6 (2.2 3.2) cm long, staminal filaments yellow or yellow toward the apex and greenish yellow at the base, stigmas white, with 3 6 lobes. Berr ies small, barrel shaped or clavate (Fig. 7 9G), 2.6 (1.8 3.5) cm long, purple, pink, reddish pink, or green at maturity. Seeds 4 .7 (4 5.4) mm long, funicular girdle 0.7 (0.4 0.9) mm wide, funicular envelope smooth (with no prominent expansion from the emb ryo).
167 Phylogenetic placement. Opuntia drummondii is sister to the rest of the diploid species in the SE clade of the Humifusa clade ( Chapter 6 ; see Fig. 7 1). Phenology Opuntia drummondii flowers from mid April through mid May with occasional flowers pr oduced through June depending on environmental conditions. Distribution Opuntia drummondii is found in coastal areas from North Carolina to western coastal Mississippi and can be found substantially far inland in Alabama and Mississippi (Majure and Ervi n 2008). This species is slightly disjunct from the Gulf of Mexico to the Atlantic Coast (i.e., contiguous populations have not been found stretching across the Florida peninsula to the Atlantic Coast; Fig. 7 10). Interestingly, disjunct mountain populatio ns and introgressive forms produced from hybridization with O. humifusa have been found in Georgia and South Carolina suggesting a distribution pattern coincident with changing sea levels during interglacial cycles with the subsequent extinction of popula tions of the species in parts of the outer coastal plain. Habitat Opuntia drummondii is most commonly found in coastal strand vegetation of the Gulf of Mexico and the Atlantic Coast commonly associated with O. humifusa subsp. pollardii or O. humifusa sub sp. lata It is most common in non shifting sands behind primary dunes, although, the species is also very common in certain parts of its range along major river systems with open, sandy habitats (see Majure and Ervin 2008). Opuntia drummondii is occasiona lly found on rock outcrops, as well, almost always associated with O. humifusa subsp. pollardii Ploidy Opuntia drummondii has been recorded as diploid (2 n =22) triploid (2 n =33) and tetraploid (2 n =44) (Majure et al. 2012a). There are very minor morphological differences associated with cytotype however, sufficient differences have not been observed that would suggest different ploidal levels should be recognized as separate species. Phylogenetic analy ses reveal that these ploidal levels are also most closely related to one another suggesting that the
168 polyploids may be autopo lyploids (Chapter 6 ). Polyploids are mostly limited to coastal areas, under presumable harsher environmental conditions, whereas diploid members of the species are more widespread. Notes Benson (1982) placed this taxon in synon y my with O. pusilla with no clarification as to why he thought this southeastern United States species belonged within O. pusilla Benson (1982) designated the neotype of O. pusilla as the line drawing by Pfeiffer & Otto of O. foliosa (see Britton and Rose 1920, p. 106), which although somewhat conforming to the morphology of O. drummondii does not show a sufficient number of diagnostic characters to be ide ntified to species Thus, it is unreasonable to use this name for the southeastern United States material, since no type locality was ever given for O. pusilla (Haworth 1803, 1812), and the actual identity of O. pusilla is ambiguous. Britton and Rose (1920 ) mentioned that the species was typically assigned to South America and may even belong within Tephrocactus, another genus with in subfamily Opuntioideae. Haworth (1812) also thought that the species may have been from South America and in his monograph d escribed it alongside O. curassavica a species from the Lesser Antilles (Venezuelan Caribbean). It is highly likely that O. pusilla could have been confused with the southeastern United States material, as many European cactus collectors often traded and sold O. drummondii under the mistaken identity of O. pusilla (Britton and Rose 1920). As far as is cur rently known, no material closely related to the southeastern US species has been found in the West Indies or South America, although, the closely related O. abjecta is found in the Florida Keys, which shares some Caribbean taxa with Antillean islands. However, Opuntia drummondii has never been recorded from the West Indies and was described from the southeastern United States ( Appalachicola, Florida ; Graha m 1846), so this name should be used
169 for the southeastern United States material instead of the ambiguous and un determinable taxon O. pusilla Further study is needed to determine the correct identity and affinity of O. pusilla Although, O. drummondii was described from Appalachicola, Florida ( along the Gulf coast of Florida ) Benson (1982) designated a neotype for the species from the Atlantic coast in St. O. pes corvi, a synonym of O. drummondii d escribed from South Carolina, from Appalachicola, Florida Anderson (2001) placed O. drummondii, O. pes corvi, O. pisciformis, and O. tracyi (all taxa described from the southeastern US) under synonymy with O. pusilla and further sta ted that the species i s found in the West Indies By his placement of these other taxa in synonymy with O. pusilla it is clear that Anderson (2001) underst ood neither where these other taxa were actually native, nor from where they were originally described. Opuntia drummondi i is listed for Louisiana (under O. pusilla ; USDA, NRCS, 2012 ), but the collections actually represent O. nemoralis Griffiths. Benson (1982) placed O. macateei under synony my with O. pusilla which also conforms to O. nemoralis. Likewise, Weniger (1967) e ncountered what he identified as O. drummondii on Galveston Island, Texas. His collection also is O. nemoralis not of O. drummondii Weniger described the glaucous color of the stems, as well as a slightly reddish hue of the inner tepals, both characters e xhibited by some populations of O. nemoralis Opuntia drummondii never exhibits reddish coloring of the inner tepals, as O. drummondii is derived solely from the yellow flowered southeastern subclade of the Humifusa clade ( Chapter 6 ). Additional specimens examined. United States. Alabama. Baldwin Co.: Bon Secour National Wildlife Refuge, off of Mobile St. 29 May 2006 L.C. Majure 1512 ( MISSA ). Butler Co.: Hwy. 7, 2.5 mi N of Butler Co. Hwy. 54, 3156'50.38''N 8651'25.19''W 6 Sep 2007 A.R. Diamond 18045 ( TROY ). Coffee Co.: CR 43, mi marker 3 K.E. Childree 7 ( TROY ). Conecuh
170 Co.: US Hwy. 31 at N side of the Sepulga River, W of the rd, 3127'16.5''N 8647'8.5''W 17 May 2008 A.R. Diamond 19255 ( TROY ). Mobile Co.: Dauphin Island off of Hwy. 163 S; 30.2645N, 88.1156W 8 Jul 2007 L.C. Majure 2570 ( MISSA ). Florida. Baker Co.: along FL 2, at Breakfast Branch, 3.5 mi NW of Eddy Tower, 4.2 mi SE of Georgia St. line 11 Jul 1984 B. Hansen 9964 ( USF ). Bay Co.: off of Hwy. 98W just W of Bay County Canal & W of Pt. St. Joe at Highland View 6 Nov 2011 L.C. Majure 4223 ( FLAS ). Columbia Co.: off of Hwy. ( CR ) 246W, 7.7km E of White Springs 15 Feb 2011 L.C. Majure 4191 ( FLAS ). Duval Co.: at Jacksonville Beach, ca. 1.9 km S of Neptune Beach, 28 Mar 2009 L.C. Majure 3700 ( FLAS ). Escambia Co.: Near Pensacola, S of Navy Blvd., W of Pace Blvd., just E of RR tracks and bridge 3 Jun 1979 J.R. Burkhalter 6405 ( FLAS ). Flagler Co.: off of Hwy. A1A S, just S of Sumner; roadside 17 May 2008 L.C. Majure 3221 ( FLAS ). Franklin Co.: jct. of NE 12th St. and Hwy. 98W at Carrabelle 6 Nov 2011 L.C. Majure 4222 ( FLAS ). Gulf Co.: KS 325 FLAS; Hamilton Co.: East side of Alapaha River, 6 mi W of Jasper off of Hwy. 41W, at jct. with Hwy. 6 15 Feb 2011 L.C. Majure 4192 ( FLAS ). Nassau Co.: Fernandina 21 Aug 1922 J.K. Small s.n. ( NY ). Okaloosa Co.: just S of Botanical Facility, ca. 3 mi S of Valparaiso 9 May 1 968 R.R. Smith 2395 ( FLAS ). St. Johns Co.: just S of St. Augustine, off of Hwy. A1A S, E side, vicinity of Anastacia Beach area; off of Magnolia Ave. 17 May 2008 L.C. Majure 3218 ( FLAS ). Wakulla Co.: Panacea Springs 28 Jul 1942 H. Kurz 290 ( MICH ). Walton Co.: off of Hwy. 30A, Grayton Beach State Park 25 Jun 2005 L.C. Majure 1066 ( M ISSA ). Georgia. Camden Co.: sand dunes, St. Marys 20 Aug 1922, J.K. Small s.n. ( NY ). DeKalb Co.: off of Hwy. 124 N from exit 75 off of Interstate 285N 27 May 2009, L.C. Majure 3788 ( FLAS ). Glynn Co.: Jeckyll Island, Atlantic Coast side in center of the island 1 Sep 2008, Tom Mann s.n. ( FLAS ). Mississippi. Clarke Co.: off of Hwy. 45, just N of jct. with Hwy. 512 22 Oct 2005 L.C. Majure 1270
171 ( MISSA ). Forrest Co.: H attiesburg off of Edwards Rd. 5 Jan 2005 L.C. Majure 756 ( MISSA ). George Co.: Charles Deaton Nature Preser ve, Nature Conservancy Property off of US Hwy 98; v icinity of Merrill; Hog Island 22 Jan 2005 L.C. Majure 771 ( MISSA ). Greene Co.: Charles Deaton Nature Preserve, Nature Conservancy Property off of US Hwy 98; v icinity of Merrill 22 Jan 2005 L.C. Majure 772 ( MISSA ). Hancock Co.: v icinity of Ansley, along Clairborne Rd., peninsula between tidal marshes 9 Jun 2005 L.C. Majure 1033 ( MISSA ). Harrison Co.: Biloxi 1 Aug 1896 C.L. Pollard s.n. ( NY ). Jackson Co.: Belle Fontaine Beach, W end 18 May 2005 L.C. Majure 955 ( MISSA ). Jasper Co.: Hwy 503 S; S of Paulding, across from Willam Chapel 9 Jan 2005 L.C. Majure 766 ( MISSA ). Kemper Co.: off of Hwy. 39S b e tw een Daleville and Dekalb 32.71215N 88.66920W 16 Jun 2007 L.C. Majure 255 5 ( MISSA ). Lauderdale Co.: p ipeline off Point Wanita Lake Rd. 9 Jan 2005 L.C. Majure 763 ( MISSA ). Lowndes Co.: off of I 82, ca 0.5 mi W of Columbus 12 Ap r 2005 L.C. Majure 843 ( MISSA ). Newton Co.: Chunky River; ca. 0.75 mi SE of Hwy 80 5 Feb 2005 L.C. Majure 776 ( MISSA ). Noxubee Co.: Gholson, off of Hwy. 21 20 Aug 2005 L.C. Majure 1155 ( MISSA ). Perry Co.: Mars Hill, Camp Shelby, DeSoto National Forest 31 Dec 2004 L.C. Majure 757 ( MISSA ). Smith Co.: Ainsworth property, S of Hwy 18 29 Dec 2004 L.C. Majure 753 ( MISSA ). North Carolina. Brunswick Co.: Smith Island 3 Apr 1918 McCorm s.n. ( NY ). Carteret Co.: Ft. Macon, opposite Beaufort 18 Aug 1922 J.K. Small s.n. ( NY ). Currituck Co.: 1.9 mi S of Mamie on US 158 25 Jun 1958 H.E. Ahles 44470 ( UNC ). Dare Co.: Town of Kitty Hawk off of Hwy. 158 W between Wachovia Bank and The Marketplace in front of shopping center 2 Jun 2009 L.C. Majure 3826 ( FLAS ). Hyde Co.: Ocracoke T.H. Kearney 2275 13 17 Oct 1898 ( US ). New Hanover Co.: Wrightsville Beach at the end of Hwy. 76 E 3 Jun 2009 L.C. Majure 3830 ( FLAS ). Onslow Co.: at end of NC 210 near New River Inlet 28 Apr 1969 S.W. Leonard 2396 ( UNA ). South Carolina.
172 Cha rleston Co.: Porcher's Bluff, Christ Church Parish 7 May 1911 E.A. Mearees s.n. ( US ). Clarendon Co.: Rt. 15 301, E Santee 26 Mar 1988 C.F. Reed 126895 ( MO ). Horry Co.: a t the S end of Folly Island County Park (Ashley Ave.) 3 Jun 2009 L.C. Majure 3833 ( FLAS ). York Co.: ca. 3 mi NE of Clover off of Hwy. 321 N then off of Old Carriage Rd. 28 May 2009 L.C. Majure 3792 ( FLAS ). 6 Opuntia humifusa (Raf.) Raf. Med. Fl. U.S 2: 247. 1830. Cactus humifusus Raf. Annals Nat. 15. 1820 Sprawling, decumbent shrubs, forming large patches, branching from the base forming chains of 1 4 cladodes; roots typically fibrous or proximally thickened. Cladodes light to dark green, 8.9 (3.1 17.7) cm long, 5.3 (2.0 9.0) cm wide, and 10.2 (3.6 19.9) mm thick, cross wrinkli ng during the winter, with 3 4 areoles per diagonal row at midstem. Leaves dark green, ascending parallel to developing cladode or slightly spreading, 7.4 (4.9 9.6) mm long. Glochids conspicuous, exserted or inconspicuous, included within the areole, stram ineous when young, aging brown. Spines ab sent, or 1 2 per areole, dark brown, brown and white mottled, or brownish yellow and white mottled during development, aging white and then gray, relatively smooth to the touch or strongly retrorsely barbed, delicat e or robust, 2.5 (0.9 4.9) cm long, 0.9 (0.7 1.3) mm in diameter. Flowers: outer tepals ovate, or triangular, dark green, or light green, erect or incurved in bud, inner tepals 8, entirely yellow, 3.4 (2.3 4.3) cm long, obovate with a mucronate tip. Berrie s clavate or barrel shaped, red, pink, purple, or green at maturity, 3.4 (2.1 4.9) cm long. Seeds 5.2 (4 5.9) mm long, funicular girdle 0.9 (0.6 1.2) mm wide, funicular envelope smooth or bumpy, with or without protrusion from the cotyledons and hypocotyl region of the embryo.
173 Opuntia humifusa is most commonly found in the eastern United States, east of the Appalachian Mountains to the Atlantic Coast, south to Florida, and east to Louisiana (Fig. 7 11). The distribution for the species given here is much r educed from that of Benson (1982) or Pinkava (2003), as O. cespitosa is recognized as distinct from O. humifusa as well as other less spiny forms of O. macrorhiza Opuntia austrina also is consi dered specifically distinct from O. humifusa and not at the varietal level of O. humifusa as in Benson (1982). Opuntia humifusa as circumscribed here consists of three subspecies i.e., Opuntia humifusa subsp. humifusa O. humifusa subsp. pollardii a nd O. humifusa subsp. lata which are considered genetically and morphologically distinct (Chapter 6) A dditionally O. humifusa subsp. lata is diploid, while the other two subspecies are tetraploid. Although, these three subspecies are generally morp hologically recognizable certain populations may exhibit morphological characters that make identification very problematic without other sources of data (i.e., molecular genetic or ploidy data). Thus, it is considered most appropriate to recognize these distinct taxa at the subspecific level within a broadly circumscribed O. humifusa A listing of county records for O. humifusa that cannot be identified to the level of subspecies follows, but the specimens that can be completely identified are listed aft er each of the subspecies treatments. Addi tional specimens e xamined subspecies undertermined United States. Alabama Crenshaw Co.: off of hwy. 331S ca. 9 mi S of Brantly along powerline, 16 0572674N 3489826E 7 Mar 2007, L.C. Majure 2044 ( MISSA ). Picken s Co.: Sipsey River at Al 14 crossing, 6.6 mi SSE of Aliceville 2 Aug 1967, R.C. Clark 17281 ( UNC ). Georgia Gwinnett : Co.: Yellow River, near McGuire's Mill 20 July 1893 J.K. Small s.n. ( NY ). Marion Co.: 2 mi SE of Juniper, Fall Line Sandhill 25 Aug 2010 J. Hill s.n. ( FLAS ). Walton Co.: Hard Labor Creek State Park: 2 mi
174 N of Rutledge, 23 Sep 1979 J.W. Hill 1206 ( NY ). Mississippi Calhoun Co.: Calhoun Wildlife Management Area, in cemetery 10 Oct 2010 J.G. Hill s.n. ( FLAS ). Marion Co.: E Marion County, just N of Hwy. 13 E of Lamar County, Lower Little River 18 Jun 2009 T. Mann s.n. ( FLAS ). North Carolina Alexander Co.: SW base of Roc ky Face Mt. along CR 1426 4 Jun 1973 R.M. Downs 13671 ( UNC ). Bertie Co.: 8.2 mi S W of Woodard on rd. paralleli ng Roanoke River 9 Jul 1958 H.E. Ahles 46275 ( UNC ). Bladen Co.: Rt. 701, 4 mi N of White Lake 25 Dec 1978 C.F. Reed 103135 ( MO ). Buncombe Co.: outcrop above Parkway near Bull Gap, 2 mi N of Riceville 3 May 1953 A.E. Radford 6968 ( UNC ). Cabarrus Co.: N side of NC 49, 4.5 mi E of Harrisburg 7 Oct 1970 L.T. Musselmann 3963 ( UNC ). Cumberland Co.: near Cedar Creek 28 Sep 1963 R.C. Clark E216 ( EKY ). Dare Co.: Nag's Head, beside Blackman St. beach access parking area, off of S Virginia Dare Tr. (Rt. 12) 13 Jun 2001 M. A. Vincent 9386 ( MU ). Lee Co.: 1 mi S of Juniper Spring Church 17 Jun 1958 S. Stewart 696 ( UNC ). Madison Co.: Hot Springs, on steep slopes of French Broad River on CR 1304, 3.6 mi W of Hot Springs 5 Jun 1981 D. Sather 1284 ( UNC ). Montgomery Co.: E shore of Yadkin River at Falls Dam, Uwharrie WMA, off NC 109 2 mi NW of Uwharrie 15 Sep 1969 E.F. Wells 2193 ( UNC ). Moore Co.: w ildlife refuge E of Rose wood, 19 Oct 1957 C. J. Burk 37 10 ( MU ). New Hanover Co.: S of Masonboro, cut off on rd. to Myrtle Grove 12 Jun 1958 C. Ritchie Bell 12932 ( UNC ). Swain Co.: Ellen School 1 Jul 1956 Salloway s.n. ( UNC ). Wake Co.: Rt. 55, 6.2 mi E of Mt. Olive; Rt. 117, 6 Oct 1982 J. Doyle 319 ( UNC ). Wayne Co.: NC 55, 2.5 mi E of NC 403; ca. 10 mi E of Mt. Olive; S side of rd. 26 May 1986 J. Doyle 802 ( UNC ). South Carolina Abbeville Co.: near SC 81, 3 mi S of Calhoun Falls 13 May 1957 A.E. Radford 22826 ( UNC ). Chester Co.: Fishing Cr. pond dam 13 May 1957 C.R. Bell 7437 ( UNC ). Dorchester Co.: 0.8 mi SE of jct. CR 138 and 148 on CR 138 (SW of Reevesville) H.E. Ahles 26267 27 May 1957 ( UNC ). Edgefield Co.:
175 near US 25, 9 mi SSW of Trenton 13 May 1957 A.E. Radford 22825 ( UNC ). Fairfield Co.: 1.2 mi NE of Strother 12 May 1957 C.R. Bell 7104 ( UNC ). Florence Co.: 2 mi N of Oak Grove school (E of Claussen) 24 May 1957 C.R. Bell 7521 ( UNC ). Hampton Co.: US Hwy. 601, ca. 2 mi NW of Miley 11 May 1956 C.R. Bell 2565 ( UNC ). Kershaw Co.: Rt. 601 of Rt. 20, near Lugoff 23 Mar 1986 C.F. Reed 125526 ( MO ). Lancaster Co.: Forty Acre Rock 4 Apr 1977 B Jacobs 14 ( MU ). Marion Co.: ca. 10 mi S of US Hwy. on CR 49 (S of Britton Neck) 26 May 1957 C.R. Bell 7859 ( UNC ). Marlboro Co.: 2 mi NW of Drake 10 Jun 1956 A.E. Radford 12543 ( UNC ). Richland Co.: SC 12, 2 mi E of Ft. Jackson entrance 13 May 1958 Duke 531 ( UNC ). Saluda Co.: SC 392, 2 mi S of Ridge Spring 26 May 1957 A.E. Radford 23214 ( UNC ). Union Co.: near Farming Cr. of Robat 5 Jun 1957 C.R. Bell 8547 ( UNC ). Tennessee. Fayette Co.: Ames Plantation, 19 Oct 1972, H.R. Deselm s.n. (TENN). Hardeman Co .: NE of Newcastle, 21 Aug 1947 A.J. Sharp 6629 (TENN). Key to Subspecies of O. humifusa 1. Cladodes spineless, cladodes mostly elliptical, glochids mostly inconspicuous (incl uded within the areole), except for older cladodes areoles generally 4 per diagonal at midstem ... O. humifusa subsp. humifusa 1. Cladodes with spines, cladodes elliptical, obovate, or rotund, glochids usually conspicu ous (exser 2. Seeds with funicular envelope smooth, only moderate, if any, protrusion of the cotyledons and hypocotyl cladodes typically scalloped margined, elliptical or rotund, spines delicate 0.8 (0.7 O. humifusa subsp. lata 2. Seeds with funicular envelope bumby, cotyledons and hypocotyl noticeably protruding, c ladodes typically smooth margined, obovate or rotund, spines robust 1 (0.95 1.3) mm in .......................................... ........... ............... O. humifusa subsp. pollardii
176 6 a Opuntia humifusa (Raf.) Raf. subsp. humifusa Med Fl U S 2: 247. 1830. Cactus humifusus Raf. Annals Nat. 15. 1820 Opuntia rafinesquei Engelm. Proc. Amer. Acad. 3: 295. 1856. nom. superfl. T YPE : Berks Co.: PA, 0.75 mi southwest of New Jerusalem, 10 July 1927 C.T. Wherry s.n. ( neotype designated by Leuenberger (1993); US!; see Fig. 7 12A). Opuntia calcicola Wherry Jour. Wash. Acad. Sci. 16: 12. 1926. T YPE United States. West Virginia. Jefferson Co.: growing on limestone edges, exact locality along B. & O. RR track about 2 mi N of Harper's Ferry RR station 10 Jun 1925 E.T. Wherry s.n. ( holotype: US isotype: NY! ). Although, Wherry (1926) cited his specimen as collected, 9 June 1925, only collections from the 10 June, 1925 exist at the two repositories listed in his description (i.e., US and NY) and are here designated as the holotype and isotype based on the inter pretation that Wherry merely mistakenly altered the date of collection in the protologue. These type specimens replace the lectotype designation by Benson (1982) of a specimen collected by Wherry in 1935. S prawling or slightly ascending shrub, during warm er months, forming large, often dense colonies, or cespitose clumps, the cladodes produced in chains of 1 4, often branching towards the tips of the plant and from the base; roots fibrous. Cladodes elliptical or rotund, dark green, not glaucous, cross wrin king during the winter, 12.6 (9 15) cm long, 7.2 (5.7 8.3) cm wide, 11.5 (9.6 15.7) mm thick, 3 4 (mostly 4) areoles per diagonal row at midstem. Leaves dark green, 8.2 (6.2 9.6) mm long, triangular ovate, to lanceolate, ascending (parallel to the cladode surface). Glochids inconspicuous, generally only exserted in older, basal stems, stramineous, but turning light brown or amber in age. Spines absent. Flowers: outer tepals dark green to slightly gray green, ovate or long triangular, erect or incurved inne r tepals 8, entirely yellow, 8 9, 3.9 (3.7
177 4.0) cm long, obovate, stamina filaments yellow or yellow green, stigma white, 6 7 lobed. Berries green, red, or orange red at maturity, 4.4 (4.2 4.8) cm long. Seeds 4.4 (4.0 4.6) mm long, funicular girdle 0.8 (0. 6 0.9) mm wide, often bumpy or irregular, funicular envelope raised along the margin from the increase in size of the cotyledons and hypocotyl bumpy portion of the funicular envelope surrounding the radical not evidently raised Phylogenetic placement Opuntia humifusa subsp. humifusa is an allotetraploid derivative of the southeastern and southwestern diploid subclades of the Humifusa clade ( Chapter 6 ). Phenology Opuntia humifusa subsp. humifusa flowers from early May through June and July depending on latitude. Plants al ready in flower in northern Vir ginia in late May may just be developing flower buds on Cape Cod, Massachusetts. Distribution Opuntia humifusa subsp. humifusa is most common alo ng the eas t ern edge of the Appalachian Mountains to the Atlantic seaboard. It also occurs sporadically in the southeastern United States ( see Additional specimens examined ). Although, not recor ded from Alabama here, this subspecies certainly should occur t here. Habitat Opuntia humifusa subsp. humifusa is most commonly found on rock outcrops (commonly slate) on the eastern slopes of the Appalachian Mountains or sandy soils of the Atlantic Coast. In other parts of its range, it is often found in sandy or cl ayey soils on xeric hilltops. This subspecies appears to be more tolerant of mesic conditions than the other two subspecies of O. humifusa Ploidy Opuntia humifusa subsp. humifusa is tetraploid throughout its range (Majure et al. 2012b). Notes Opuntia humifusa subsp. humifusa geographically, is most often found between regions dominated by O. humifusa subsp. pollardii and O. cespitosa sug gesting those two taxa,
178 or their ancestors could have been involved in the origin(s) of O. humifusa subsp. humifusa which is derived from the SW (paternal lineage) and SE clade (maternal lineage), as is O. cespitosa Although, O. cespitosa is suggested to have been derived from O. humifusa subsp. pollardii (maternal lineage), backcrosses of O. cespitosa to O. humifusa subsp. pollardii could have resulted in the formation of O. humifusa subsp. humifusa Crossing studies need to be performed to further test this hypothesis Additional specimens examined. United States. Conneticut. Fairfield Co.: n ear Stratford 25 Aug 1894 C.L. Pollard 256 ( US ). New Haven Co.: Leete's Island 20 Jul 1958 J.J. Neale s.n. ( FLAS ). Middlesex Co.: Madison 12 Sep 1970 F.C. Seymour s.n. ( M O ). Delaware. Sussex Co.: at Frankford off of Hwy. 113 S at jct. with Catmans Rd. 2 Jun 2009, L.C. Majure 3824 ( FLAS ). District of Columbia Washington Co.: Bare rocks, i sland in Potomac 22 Jun 1897, E.S. Steele s.n. ( MU ). Maryland. Alleghany Co.: b etween Interstate 68 and Hwy. 144 E just E (ca. 1 km) of the town of Flintstone 30 May 2009 L.C. Majure 3810 ( FLAS ). Anne Arundel Co.: along Chesapeake Bay, Sandy Pt. St. Park 17 Jul 1959 C.F. Reed 43690 ( MO ). Baltimore Co.: edge of field at Loch Raven C.F. Reed 5362 10 Jul 1946 ( MO ). Calvert Co.: S of Upper Marlboro 4 Jul 1946 C.F. Reed 6075 ( M O ). Calvert Co.: S of Upper Marlboro 4 Jul 1946 C.F. Reed 6075 ( MO ). Harford Co.: Cedar Church Rd. off Rt. US #1, near Dublin 27 Feb 1982 C.F. Reed 121327 ( MO ). Montgomery Co.: Plummer's Island 12 Sep 1910 W.L. MacAtee s.n. ( NY ). Worchester Co.: along Rt. 388, 8 12 mi W of Snow Hill 3 Nov 1990 C.F. Reed 132951 ( MO ). Dorchester Co.: Wet Swag, 1 mi N of Eldorado, Rt. 313 17 Jun 1970 C.F. Reed 93538 ( MO ). Prince George Co.: Rt. US 50 and 301, Priests Bridge 3 May 1960 C.F. Reed 46305 ( MO ). Massachusett s. Barnstable Co.: o ff of Pilgrims Springs Rd., ca. 1.3 km S of WellFleet Harbor 31 May 2009, L.C. Majure 3814 ( FLAS ). Hampden Co.: Southwick 26 Apr 1910, Gillette 15057
179 ( US ). Mississippi. Attala Co.: off of rd. 5217, SE of Ethel from Kings R d., 33.08230 89.42710 15 May 2007 L.C. Majure 2364 ( MISSA ). Choctaw Co.: 6.5 km E of Stewart 1 Jun 2009 K. Philley 499 ( FLAS ). Grenada Co.: Gore Springs, off of Betterton Dr. from Gore Springs Rd. 17 Sep 2006 L.C. Majure 1833 ( MISSA ). Webster Co.: 1.6 km SW of Tomnolen 1 Jun 2009 K. Philley 498 ( FLAS ). New Hampshire. Hampshire Rockingham Co.: at Seabrook 1 Aug 2011, B. Nichols s.n. ( FLAS ). New Jersey. Atlantic Co.: ca. 0.25 km SE of Nesco Rd. (Hwy. 542), E of Nescohague Lake, P.E. Marucci Center for Cranberry and blueberry Research, Hammonton 1 Sep 2008 P. Oudemans s.n. ( FLAS ). Burlington Co.: ca. 1.5 km SE of Pemberton, along Hwy. 530 N 1 Jun 2009 L.C. Majure 3821 ( FLAS ). Camden Co.: without definitely locality, 25 Jun 1874 J.H. Redfield 245 9 ( MO ). Gloucester Co.: Wenonah 1 Oct 1904 G.T. Hastings ( NY ). Ocean Co.: Tom's River, 14 Jul 1919 F.W. Hunnewell 6465 ( NY ). Passaic Co.: 5 mi W of Boardville 14 Jul 1907 K.K. Mackenzie 2738 ( NY ). Somerset Co.: Rock Cliffs, near Biltmore 1 Jun 1896 Anonymous s.n. ( US ). New York New York Co.: NY, NY, Van Cortlandt Park and vicinity 1 Jan 1925 E.P. Bicknell s.n. ( NY ). Orange Co.: West Point 14 Mar 1905 E.A. Means s.n. ( NY ). Queens Co.: Hunters Point 28 Jun 1865 W.H.Leggett s.n. ( NY ). Richmond Co.: Cooke's Point, Staten Island 16 Jul 1914 Photo ( NY ). Rockland Co.: Rocky Knoll, S end of Palisades Interstate Park, sec., Iona Island 30 Jun 1953 J.H. Lehr 283 ( NY ). Suffolk Co.: at Ashrokan, Huntington, Long Island 16 Oct 1926 H.J. Banker 3844 ( NY ). Westchester Co.: White Plains, O.R. Willis s.n. ( NY ). Worchester Co.: Pelham 9 July 1882 C.H. Day s.n. ( NY ). North Carolina. Hertford Co.: 0.8 mi S of Barretts Crossroads 8 Jul 1958, H.E. Ahles 46028 ( UNC ). Jones Co.: Island Creek, E of the jct. of Trent River and Island Cr., 5 mi NE of Pollocksville 3 Oct 1965, M.N. Sears 6841 ( UNC ). Pennsylvania. Susquehanna Co.: Round Top Island, East of Harrisburg 1 Jan 1912, E. Gillett s.n. ( NY ). Virginia. Accomac Co.: just S of
180 Wachopreague on CR 605 5 M ar 1966 F.C. James 3843 ( UNC ). Amherst Co.: Rt. 130, 1.3 mi W of Rt. 635, NW of Elon 25 Apr 1983 J.Doyle 444 ( UNC ). Augusta Co.: j ct. of CR 1212 and CR 608, ca. 2 mi N of Vesuvius 31 May 1986 J.Doyle 827 ( UNC ). Craig Co.: Rt. 311 ca. 0.1 mi W of Rt. 611 near Johns Creek 5 Sep 1982 J. Doyle 310 ( UNC ). Culpeper Co.: near Cedar Hill, Rt. 522 3 May 1969 C.F. Reed 87781 ( MO ). Franklin Co.: Bald Knob, Rocky Mount 26 Jun 1976 C.E. Stevens 13064 ( MO ). Fredrick Co.: off of Hwy. 50 W at Hayfield jct. of N Hayfield St. and Hwy. 50 W 30 May 2009 L.C. Majure 3807 ( FLAS ). Gloucester Co.: G.P. Coleman Memorial Building on US 17 (S point of county) 5 Jul 1966 F.C. James 4926 ( UNC ). Goochland Fluvanna Co.: cliffs, Rt. 6, at Columbia 30 Jun 1970 C.F. Re ed 97330 ( MO ). Isle of Wight Co.: near Franklin 7 28 Jun 1893 A.A. Heller 916 ( NY ). King & Queen Co.: Rt. 360, just S of Stephen's Church 25 N ov 1974 C.F. Reed 96621 ( MO ). Page Co.: off of Hwy. 34 N, 5.9 km N of town of Shenandoah and 0.10 km N of Shenandoah River 29 May 2009 L.C. Majure 3798 ( FLAS ). Pittsylvania Co.: jct. of CR 734 and 730 (1 mi S of Ringgold) 2 Oct 1965 F.C. James 3227a ( UNC ). Powhatan Co.: 0.2 mi NW of j ct. CR 711 & CR 641 on 641 (NE of Powhatan) 8 Jun 1967 F.C. James 6417 ( UNC ). Warren Co.: off of Hwy. 11 W just E of Middletown ca. 0.1 km at Cedar Creek Battleground 29 May 2009 L.C. Majure 3800 ( FLAS ). Wes t Virginia. Hampshire Co.: At the town of j ct. just W (ca. 0.25 km) along shaly slopes of Hwy. 50 W 30 May 2009 L.C. Majure 3808 ( FLAS ). Mineral Co.: ca. 3 km W of the town of j ct. along Hwy. 50 W 30 May 2009 L.C. Majure 3809 ( FLAS ). Pendleton Co.: ca. 1.2 km S of Brandywine 1 Jun 2011, E. Ribbens s.n. ( FLAS ). 6b. Opuntia humifusa (Raf.) Raf. subsp. pollardii (Britton and Rose) Majure comb. nov. Opuntia pollardii Britton and Rose, Smithsonian Misc. Coll. 50: 523. 1908 T YPE :
181 United States. Mississippi. Harrison Co.: Biloxi, 1 Aug 1896, C.L. Pollard 1138 ( holotype: NY!; isotypes: MO! US! see Fig. 7 13A). Opuntia mesacantha Raf. Bull. Bot. Seringe. 216. 1830. T YPE : no type designated by author, United States. Virginia. Hampton, 31 May 1878, J.W. Chickering, Jr. s.n. (neotype designated here US!). Opuntia macrarth a Gibbes, Proc. Elliott Soc. Nat. Hist. 1: 273. 1859. T YPE : within a few miles of Charleston (presumably destroyed in the US Civil War), United States. South Carolina. Charleston Co.: Isle of Palms, near Charleston, 1 4 Feb 1916 J.K. Small s.n. (neotype designated here, US! ). Sprawling shrubs, often s lightly ascending, forming large colonies, sometimes several meters in diameter; roots typically fibrous, although generally thickening proximally. Cladodes mostly frequently obovate, but also elliptical, or rotund, dark green to light yellow green, not gl aucous, cross wrinkling during the winter, 8.5 (3.1 17.7) cm long, 5.2 (2 9) cm wide, 10 (3.6 18.6) mm thick, occasionally cladodes disarticulating with ease in summer months, although, generally not disarticulating without force, areoles 3 4 (generally 3) per diagonal row at midstem. Leaves light or dark green, ascending parallel to the cladode or slightly spreading, triangular or ovate, 4.9 (4.8 5) mm long. Glochids conspicuous, exserted or inconspicuous, included within the areole, stramineous, aging lig ht brown, or light amber. Spines relatively long or short, conspicuous in many specimens, robust, 2.1 (0.9 3.2) cm long, 1.0 (0.95 1.3) mm in diameter, strongly retrorsely barbed when young to several years old, this often being lost in older spines, dark brown white mottled, yellow brown, or brown yellow white mottled during development, white when mature, aging gray. Flowers: outer tepals green, broadly triangular, erect or commonly incurved in bud, inner tepals 8, entirely yellow, 2.7 (2.3 3.0) cm long,
182 obovate to obtriangular, apex margins often moderately lacerate, stamen filaments yellow or yellow green, stigma white, 6 lobed. Berries green, red, orange red, clavate to barrel shaped. Seeds 5.5 (5.0 5.9) mm long, funicular girdle 1.0 (0.7 1.3) mm wide, often bumpy or irregular, funicular envelope raised along the margin from the increase in size of the cotyledons and hypocotyl, bumpy, portion of the funicular envelope surrounding the radical not evidently raised. Phylogenetic placement Opuntia humifusa subsp. pollardii is a tetraploid, apparently derived solely from the SE clade, and it is very closely related to O. cespitosa and O. nemoralis according to plastid DNA sequence data, and likely is was one of the progenitors of both species. Phenology. Th is subspecies begins to flower at the end of April or beginning of May. Distribution. Opuntia humifusa subsp. pollardii is mostly confined to the coastal plain of the eastern United States ( see specimen s examined ), however this subspecies covers the broade st distribution of the three recognized with in O. humifusa Opuntia humifusa subsp. pollardii is one of the most common taxa of Opuntia along the Gulf Coast of Alabama, Mississippi, and along the panhandle of Florida. Habitat. Opuntia humifusa subsp. poll ardii is most common in the eastern United States pine belt in sandy soils in Pinus palustris sandhills or mixed Pinus Quercus sandhills although, it is frequently encountered on granitic outcrops in Georgia, South Carolina, and North Carolina. In Alabama Mississippi, and the Florida panhandle it is common in non shifting dunes behind primary dunes, similar to O. drummondii, with which it is commonly sympatric. Ploidy Opuntia humifusa subsp. pollardii is tetraploid 2 n =44, throughout its range (Majure e t al. 2012b). Notes. The polyploid Opuntia humifusa subsp. pollardii apparently originated solely from the SE clade, although, the nature of its formation has not been determined. Considering
183 morphological characters as compared to its diploid relative, O. humifusa subsp. lata it seems likely that Opuntia humifusa subsp. pollardii could have arisen via autopolyploidy, although this needs further study. Opuntia pollardii was elevated to subspecific rank over the two earlier names, O. mesacantha and O. macra rtha as O. pollardii was described with great detail by Britton and Rose (1908). There also are photos of live material of the type specimen and, and the type specimen itself was available for study. Although, the other two taxa, O. mesacantha and O. macr artha apparently belong here, there were no type specimens for them, and the original descriptions of the taxa were greatly lacking in detail. Additional specimens examined. United States. Alabama. Baldwin Co.: Bon Secour National Wildlife Refuge, off of Hwy. 180, E of Ft. Morgan 29 Jun 2005 L.C. Majure 1082 ( MISSA ). Marion Co.: 6 mi SW of Hackelburg 22 Aug 2007 J. Hill s.n. ( FLAS ). Mobile Co.: Mobile, Springhill 12 May 1888 C. Mohr s.n. ( UNA ). Randolph Co.: granite outcrop (Bald Rock) Almond Community, 3.5 mi W of Wadley 2 Aug1 966 R.C. Clark 6526 ( UNC ). St. Clair Co.: Pottsville sandstone outcrops E end of mt. 4 Jun 1963 P.E. Bostick 284 1 ( UNC ). Winston Co.: 4.8 mi N of Haleyville 10 May 1967 R.C. Clark 13102 ( UNC ). Florida. Bay Co.: W of Panama City 7 Jun 1938 F.N. Young s.n. ( FLAS ). Escambia Co.: W end of Santa Rosa Island at Gulf Islands National Seashore, S of Ft. Pickens Rd., ca. 0.7 mi W of Battery Langdon 5 May 1979 J.R. Burkhalter 6397 ( FLAS ). Levy Co.: T14S, R13E, Sec. 9, NW of SE 28 Aug 1982 K.A. Kron 1036 ( FLAS ). Okaloosa Co.: v icinity of Ft. Walton Beach, off of Hwy. 98 W 25 Jun 2005 L.C. Majure 1075 ( MISSA ). Santa Rosa Co.: j ct. of US 98 and Hwy. 399 25 Jun 2005 L.C. Majure 1081 ( MISSA ). Walton Co.: off of Hwy. 30A, Grayton Beach State Park 25 Jun 2005 L.C. Majure 1067 ( MISSA ). Georgia. DeKalb Co.: off of Hwy. 124 N from exit 75 off of Interstate 285N 27 May 2009, L.C. Majure 3787 ( FLAS ). Emanuel Co.: Ohoopee Dunes SNA,
184 32 32' 15" 82 27' 40" 16 May 2007, J. Hill s.n. ( MISSA ). Jackson Co.: o ff of Interstate 85 S; ca. 6.7 km NW of Commerce, 0.3 km NE of Oconee River 27 May 2009, L.C. Majure 3789 ( FLAS ). Taylor Co.: along side of GA 96, 3 mi W of Butler; 7 mi W of Howard 20 May 1972, H. Register s.n. ( UNC ) Louisiana. Allen Co.: N side of RR tracks; S of LA 190; ca. 5 mi W of Kinder 16 Apr 1983 J. Doyle 358 ( UNC ). Washington Parish Co.: N of LA 437 via logging roads, ca. 1.5 mi W of Bogue Chitto River; ca. 1.5 air mi WSW of Enon 1 Sep 2008 Chris Reid s.n. ( FLAS ). Natchitoches Co.: 3.25 mi W of Natchitoches on Hwy. 6, 8 Aug 1980 W.C. Holmes 3951 ( NY ). Maryland Anne Arundel Co.: Rt. 4 near Waysons Corner 4 Jun 1969, C.F. Reed 112805 ( MO ). St. Marys Co.: Piney Point 26 Apr 1958, C.F. Reed 40649 ( MO ). Mississippi. Forrest Co.: j ct of Hwy 49 S with Hwy 13 E; Vic of Maxie 17 Mar 2005 L.C. Majure 806 ( MISSA ). Hancock Co.: St. Joseph's Cemetery, vicinity of Diamondhead 9 Dec 2006 L.C. Majure 1924 ( MISSA ). Harrison Co.: Little Florida, DeSoto Natio nal Forest 9 Jul 2006 L.C. Majure 1603 ( MISSA ). Jackson Co.: Greenwood Island at Bayou Casotte 14 Jan 2006 L.C. Majure 1297 ( MISSA ). Lafayette Co.: off of Hwy. 7, S of Oxford, 34.35054 N 89.51003 W 27 May 2007 L.C. Majure 2448 ( MISSA ). Marshall Co.: off of Hwy. 78E ca. 1 mi NE of Wall Doxey State Park 30 Dec 2005 L.C. Majure 1293 ( MISSA ). Neshoba Co.: off of Hwy. 15 N, adjacent to Pearl River, ca. 0.25 mi S of bridge 18 Sep 2005 L.C. Majure 1201 ( MISSA ). Noxubee Co.: Gholson, off of Hwy. 21 20 Aug 2005 L.C. Majure 1156 ( MISSA ). Winston Co.: Tombigbee National Forest, o ff of Sturgis Rd. 16 Jan 2005 L.C. Majure 769 ( MISSA ). Yalobusha Co.: trailside off of Co. rd 221 13 Jan 2005 L.C. Majure 767 ( MISSA ). New Jersey Burlington Co.: Atsion al ong NJCRR 29 Aug 1951, W.A. Stern s.n. ( ILL ). North Carolina. Brunswick Co.: C.C.C. Camp, Southport 10 May 1935 A.C. Matthews s.n. ( UNC ). Chowan Co.: 4 mi W of Small's Crossroads 24 Jun 1958 H.E. Ahles 44231 ( UNC ). Cleveland Co.: along
185 the Broad River, ca. 4.5 mi S of Boiling Springs on NC Rt. 150 22 Jun 1956 H.E. Ahles 15375 ( UNC ). Currituck Co.: o ff of Hwy. 158 E just N of Kitty Hawk 2 Jun 2009 L.C. Majure 3825 ( FLAS ). Dare Co.: a long Hwy. 12 S, S of the town of Kitty Hawk at jct. with Palmetto St. 2 Jun 2009 L.C. Majure 3827 ( FLAS ). Davidson Co.: 1 mi E of Linwood Southmont Rd., 0.25 mi N of Rockcrusher Rd., 19 Jun 1966 S.W. Leonard 192 ( UNC ). Franklin Co.: Rt. 98; 0.2 mi E of Jct. 1001; 1.8 mi W of Bunn, S side 7 Jun 1983 J. Doyle 472 ( UNC ). Granville Co.: 0.7 mi W of Franklin Co. line on NC 56 and 1.5 mi NW of dirt rd. 28 Sep 1956 H.E. Ahles 20204 ( UNC ). Halifax Co.: 3.3 mi S and E of Halifax on NC 561 19 Jul 1956 H.E. Ahles 17155 ( UNC ). Hoke Co.: 0.1 mi S of CR 1101 on 401; ca. 8 mi S of Raeford 28 May 1986 J. Doyle 814 ( UNC ). Onslow Co.: o ff of Hwy. 17 N ca. 1.8 mi N of jct. with Hwy. 172 E 2 Jun 2009 L.C. Majure 3829 ( FLAS ). Pender Co.: sandhill, 1 mi E of jct. of US 421 and NC 53 on NC 53 (W of Burgaw) 13 Jun 195 7 H.E. Ahles 28102 ( UNC ). Robeson Co.: 4.3 mi SE of Red Springs near ACLRR 2 Jun 1958 R.F. Britt 1972 ( UNC ). Rowan Co.: t own of Granite Quarry off of Du nn Mt. Rd. at Dunn Mt. (summit) 28 May 2009 L.C. Majure 3793 ( FLAS ). Scotland Co.: Rt. 401, 2 mi S of Lumber River 19 Apr 1967 C.F. Reed 79087 ( MO ). Wake Co.: Mitchell's Mill, 3.6 mi E of 40 1 on 96; 0.2 mi SE on 2224, N 0.2 mi on 2300 17 May 1983 J. Doyle 447 ( UNC ). South Carolina. Anderson Co.: dry field 3 mi ESE of Fair Play 31 May 1956, H.E. Ahles 13415 ( UNC ). Bamberg Co.: 0.7 mi S of jct. CR 27 and 22 on CR 27 (SW of Govan) 26 May 1957 H.E. Ahles 26037 ( UNC ). Beaufort Co.: 1 mi E of Co. Rt. 76 on US Rt. 21 10 May 1956 H.E. Ahles 12357 ( UNC ). Charleston Co.: Isle of Palms, Charleston 1 Fe b 1916 J.K. Small s.n. ( NY ). Chesterfield Co.: 0.5 mi S on CR 20 from jct. with US 1, S of Cheraw ; 3434'N 7953'W 16 May 1976 J.C. Solomon 1945 ( MO ). Greenville Co.: slopes of Cesar's Head, North Cove 3 Sep 1876 G. Engelmann s.n. ( MO ). Orangeburg Co.: Santee Club 1 Jun 1910 L.A. Beekauree s.n.
186 ( NY ). Pickens Co.: Glassy Mt., ca. 3 mi NE of Pickens 28 May 2009 L .C. Majure 3790 ( FLAS ). York Co.: ca. 3 mi NE of Clover off of Hwy. 321 N then off of Old Carriage Rd. 28 May 2009 L.C. Majure 3791 (FLAS). Virginia. Amelia Co.: Rock Sable, SW of Deatonville; 0.7 mi S on 1st dirt rd. to left, ca. 0.4 mi W of jct. 618 and 617; near Saylers Cr. St. Battlefield 31 May 1986 J. Doyle 815 ( UNC ). Brunswick Co.: off Rt. 626 and Rt. 705, near Gasburg 19 Aug 1978 C.F. Reed 103249 ( MO ). Goucester Co.: sandy fields, Goucester Point 17 Apr 1983 C.F. Reed 117312 ( MO ). Hampton Co.: Ft. Monroe, Hampton 7 May 1977 C.F. Reed 102057 ( MO ). Madison Co.: Rt. 29 at Robinson Run, N of Madison 27 Apr 1981 C.F. Reed 114429 ( MO ). Richmond Co.: Richmond D. Chalmot s.n. ( US ). Suffolk Co.: Nansemond, ca. 1 mi E of Blackwater River and 6 mi N of VA NC state line 22 Jun 1963 H.E. Ahles 58238 ( UNC ). Virginia Beach Cape Henry, Rt. 6, 3 Sep 1940 F.E. Egler 40 370 ( NY ). 6c. Opuntia humifusa subsp. lata (Small) Majure comb. nov. Opuntia lata Small Jour. N. Y. Bot. Gard. 1919. T YPE United States. Florida. [Alachua Co.:] pine woods, 12 mi west of Gainesville, 13 Dec 1917, J.K. Small s.n. (holotype: NY! on two sheets; see Fig. 7 14A) Opuntia eburnispina Small in Britton and Rose, The Cactaceae 1: 24. 260. 1923. T YPE : United States. Florida [ Collier Co.:] s and dunes, Cape Romano, 10 May 1922, J.K. Small s.n. (holotype: NY!). Opuntia impedita Small in Britton and Rose The Cactaceae 4: 257. 1923. T YPE United States. Florida. Duval Co.: Atlantic Beach, east of Jacksonville, 26 April 1921, J.K. Small s.n. (holotype: NY!; isotype: US!). Small shrubs, procumbent or slightly ascending during warmer months, often with 1 nu merous cladodes arising from the base; roots typically fibrous, although oftentimes thickened
187 proximally. Cladodes dark green, not glaucous, cross wrink l ing during the winter, generally not easily disarticulating, often heteromorphic (i.e., a single plant or population may have widely different cladode shapes), mostly elliptical, but also rotund or oblong (Fig. 7 14B E), margins typically scalloped, 8.4 (4.1 13.0) cm long, 4.7 (3.6 5.9) cm wide, 11.6 (6.2 19.9) mm thick. Leaves green, 7.5 (7.2 7.8) mm long, ascending parallel to the cladode or slightly spreading, ovate. Glochids conspicuous, exserted, stramineous aging light brown. Spines 1 5 per areole (generally 1), moderately or occasionally strongly retrorsely barbed, this being lost with age of the spine, brown and white mottled during development, aging white after the first year, and then gray, 3.7 (2.4 4.9) cm long, 0.8 (0.7 0.9) mm di ameter. Flowers: outer tepals green, triangular to ovate, erect or incurved in bud, inner tepals 8, entirely yellow, obovate with a mucronate tip, 3.9 (3.4 4.3) cm long, stamen filaments yellow or yellow green, stigmas white to cream, 6 lobed. Berry clavat e, red, pink, or green at maturity, 3.4 (2.1 4.9) cm long. Seeds 5.0 (4.7 5.3) mm long, funicular girdle 0.8 (0.6 1.1) mm wide, regular, generally not bumpy, funicular envelope smooth, usually not raised from the expansion of the cotyledons or hypocotyl, i f slightly raised then generally not bumpy. Phylogenetic placement Opuntia humifusa subsp lata is sister to the clade containing O. abjecta and O. austrina (Fig. 7 1). Phenology Opuntia humifusa subsp. lata begins flowering in southern Florida during mid March. More northern populations soon follow and are in full f lower typically in early to mid April in northern Florida. Distribution. Opuntia humifusa subsp. lata is distributed through the outer coastal plain of the southeastern United States from N orth Carolina south to Florida and west to Mississippi.
188 Habitat. Opuntia humifusa subsp. lata is most common in the southeastern United States in Pinus palustris or P. elliottii sandhills, or mixed Quercus geminata, Q. incana, Q. laevis, P. palustris xeri c sandhills. Ploidy Opuntia humifusa subsp. lata is diploid 2 n =22, throughout its range (Majure et al. 2012b). Notes. Morphologically, Opuntia humifusa subsp. lata is the diploid version of O. humifusa subsp. pollardii Both taxa have the same growth form and O. humifusa subsp. lata can be easily confused with O. humifusa subsp. pollardii Opuntia humifusa subsp. lata tends to have non uniform cladodes that are often scallop margined unlike O. humifusa subsp. pollardii that mostly has smooth margined cladodes. Opuntia lata also tends to have seeds with a smooth funicular envelope, which contrasts with the bumpy funicular envelope of O. humifusa subsp. pollardii Benson (1982) considered O. eburnispina to be an intersp ecific hybrid between O. humifusa and O. stricta based on the numerous spines produced from the areole. Here O. eburnispina is considered synonymous with Opuntia humifusa subsp. lata. The numerous spines per areole produced by O. eburnispina material are seen in all members of the SE diploid clade (i.e., O. abjecta, O. austrina, O. drummondii, O. humifusa subsp. lata ) and therefore do not signify hybridization with O. stricta Also the spines are not produced in a stellate pattern as in O. stricta and re lated taxa or hybrids (see O. ochrocentra above). Additional specimens examined. United States. Ala bama. Autauga Co.: off of Hwy. 82E just N of Vida j ct. 16 0529179N 3604707E 7 Mar 2007 L.C. Majure 2043 ( MISSA ). Butler Co.: Hwy. 77, 0.93 mi N of Butler Co. Hwy. 62, 3153'40.5''N 8632'30.8''W A.R. Diamond 19258 17 May 2008 ( TROY ). Crenshaw Co.: off of H wy. 331S ca. 9 mi S of Brantly 16
189 0572674N 3489826E 7 Mar 2007 L.C. Majure 2044 ( MISSA ). Dale Co.: Dale County Lake, 6 Jun 2000 M. Woods 8106 ( TROY ). Dallas Co.: bluff on Mulberry Cr., vicinity of USGS stream gauge station, 2.9 mi ENE of Valley Creek jct. 12 Aug 1967 R.C. Clark 18088 ( UNC ). Elmore Co.: near Good Hope Church, 5 mi SE of Wetumpka 27 Jul 1967 R.C. Clark 17129 ( UNC ). Henry Co. : 7.5 air mi SE of Abbeville, Co. Rt. 65 off of State Rt 95, 0.25 WSW of Hardwicksburg, 3129'56''N 859'17''W 23 Jul 2003 R.R. Hayes 10413 ( TROY ). Marengo Co.: Myrtlewood E. Holt 1 Jan 1912 ( NY ). Mobile Co.: Mile marker 27, off of Hwy. 45N 6 Mar 2011 L.C. Majure 4194 ( FLAS ). Pike Co.: N of Ozark off of Hwy. 231N; near Bama Nut Shop, ca. 3 mi S of Brundidge 6 Jul 2007 L.C. Majure 2569 ( MISSA ). Sumter Co.: ca. 1 mi NE of Woodford on CR 23 5 Aug 1966 R.C. Clark 6743 ( UNC ). Wilcox Co.: along AL 41, ca. 5 mi SW of Camden, just N of Pebble Hill Community 6 J un 1967 R.C. Clark 13941 ( UNC ). Florida. Alachua Co.: Micanopy, off of Hwy. 234, just E of 441N 11 Apr 2010, L.C. Majure 3991 ( FLAS ). Calhoun Co.: open oak woods, cultivated at M issouri B otanical Garden, 31 May 1978 J.C. Solomon 3836 ( MO ). Citrus Co.: along S41 and the railr oad, 9 mi N of Inverness 15 Apr 1976 L.M. Baltzell 8272 ( FLAS ). Clay Co.: off of Hwy. 301 E (heading N), ca. 1.6 km S of CR 218 28 Mar 2009 L.C. Majure 3699 ( FLAS ). Collier Co.: TP Scrubs (Coll28), Secs 2, 3 and 10, T48S, R25E (Bonita Springs quad.) 30 Mar 1986 R.B. Huck 3954 ( FLAS ). Columbia Co.: off of US Hwy. 27 W, ca. 2 mi NW of Ft. White; 2.2 mi E of Itchatucknee State Park entrance 24 Apr 2008 L.C. Majure 3089 ( FLAS ). DeSoto Co.: Deep Creek, 800 ha on W side of Peace River, ca. 4.5km NE of jct. I 75 & Kings Hwy. (CR 769), 31 Jul 2008 A. Franck 751 ( USF ). Dixie Co.: W of Suwanee River and M anatee Springs off of Hwy. 349S, 17 0306724E 3264465N 8 M ar 2007 L.C. Majure 2050 ( MISSA ). Duval Co.: St. John's Bluff 1 Jan 1942 H. Kurz 274 ( MICH ). Franklin Co.: St. Teresa H. Kurz 288 28 Jul 1942 ( MICH ). Gilchrist Co.: 5
190 mi W of Ft. White 28 Apr 1961 G.R. Cooley 8190 ( UNC ). Hamilton Co.: j ust S of Crossroads, off of Hwy. 141S 24 Sep 2011 L.C. Majure 4217 ( FLAS ). Hardee Co.: ca. 1 mi (by air) S of Ft. Green Springs 9 Nov 1993 B. Hansen 12444 ( USF ). Hernando Co.: off of Hwy. 93S, ca. 0.5 mi S of Withlacoochee River 14 Feb 2010 L.C. Majure 3948 ( FLAS ). Highlands Co.: Off of Hwy. 64W, just E of Avalon Park, corner of Dodd's Rd., empty lot 11 Mar 2010 L.C. Majure 3977 ( FLAS ). Hillsborough Co.: Little Manatee River State Park, E side of main park drive, 250 ft, S of Ranger Station 27 Apr 1999 J. Myers 360 ( USF ). Lafayette Co.: off of Hwy. 27 W; ca. 4 km NW of jct. with Hwy. 349 28 Oct 2007 L.C. Majure 2795 ( FLAS ). Leon Co.: along Tram Rd., 20 mi east of Tallahassee 12 May 1976 M. Blaker 40 ( FSU ). Levy Co.: o ff of Hwy. 24W ca. 1 mi W of Alachua Co. line 6 Dec 2008 L.C. Majure 3645 ( FLAS ). Lake Co.: Palatakaha Park, off of Hull Rd., just S of Clermont 11 Jun 2010 L.C. Majure 4092 ( FLAS ). Liberty Co.: along SR 12, NE of Bristol 9 May 1977 L. Rosen 10 ( FSU ). Manatee Co. : Jct. of CR 675 and Jennings Rd., 0.6 km S of the east arm of Lake Manatee 23 May 2010 L.C. Majure 4065 ( FLAS ). Marion Co.: off of Interstate 75N, ca. 0.5 mi S of reststop, ca. 6.4 km S of Ocala Exit 28 Mar 2009 L.C. Majure 3709 ( FLAS ). Marion Co.: off of Interstate 75N, ca. 0.5 mi S of reststop, ca. 6.4 km S of Ocala Exit 28 Mar 2009 L.C. Majure 3709 ( FLAS ). Okeechobee Co.: off of Hwy. 441N, 0.5km N of Ft. Dunn 11 Feb 2011 L.C. Majure 4188 ( FLAS ). Orange Co.: ca. 2.5 km W of Oakland along Hwy. 438, 14 Nov 2010 L.C. Majure 4174 ( FLAS ). Osceola Co.: off of Hwy. 441S, ca. 5 mi S of jct. with Hwy. 192 28 Mar 2009 L.C. Majure 3703 ( FLAS ). Pasco Co.: along C 587, 6.2 mi SW of jct. with FL 52; ca. 4.5 mi ENE of New Port Richey 31 May 1984 B. Hansen 9907 ( USF ). Putnam Co.: off of Hwy. 310 just W of Hwy. 19 just N of Rodman Reservoir, 24 May 2008 L.C. Majure 3249 ( FLAS ). Santa Rosa Co.: along Choctaw OLF (Dillon Field) Rd., ca. 0.5 mi W of FL 87 and ca. 5 mi N of Holley 4 May 1977
191 K.D. Perkins 2 23 ( FLAS ). Sarasota Co.: Longboat Key, along Gulf of Mexico Dr. ca. 0.1 mi NW of golf course entrance 2 Apr 1981 R.P. Wunderlin 8915 ( USF ). St. Johns Co.: Rt. 204, just W of US Rt. 1 15 Apr 1982 D.S. Correll 53660 ( NY ). Sumter Co.: off of Hwy. 301N along railroad tracks ca. 2.5km S of Bushnell 18 May 2008 L.C. Majure 3238 ( FLAS ). Suwanee Co.: ca. 3 mi N of Beachville off of Hwy. 247 W, 17 0323021E 3322521N 8 Mar 2007 L.C. Majure 2049 ( MISSA ). Wakulla Co.: at the jct. of Hwy. 98W and Hwy. 365 6 Nov 2011 L.C. Majure 4221 ( FLAS ). Walton Co.: Nekuse Preserve off of Hwy. 81, 30.53163N 85.94189W 14 Jul 2007 L.C. Majure 2589 ( MISSA ). Georgia. Bullock Co.: ca. 8.5 mi S of G.S.C. off of US 301 at Lower Lotts Creek Church 16 Jun 1965 C.B. Oneal 10 ( UNC ). Candler Co.: SE of Stillmore, 3.7 km SE of Emanuel Co. line on Stillmore Rd., 30 May 1988 D.E. Boufford 23886 ( NY ). Charlton Co.: off of Hwy. 121N, just S of St. George 15 Feb 2011 L.C. Majure 4190 ( FLAS ). Chatham Co.: Savannh 1 Apr 1927 F.D Heyward s.n. ( MICH ). Crawford Co.: 2.6 mi SE of Knoxville, 10 Aug 2009 J.G. Hill s.n. ( FLAS ). Emanuel Co.: 5 mi S of Swainsboro on Hwy. 1 4 May 1974 C.L. Rodgers 74081 ( FLAS ). Houston Co.: off of Interstate 75, ca. 1 mi SW of Perry at jct. with I 75 N 27 May 2009 L.C. Majure 3786 ( FLAS ). Irwin Co.: off of Hwy. 32/125, ca. 0.15 km W of Alapaha River and Big Creek, ca. 2 km W of Irwinville 27 May 2009 L.C. Majure 3785 ( FLAS ). Johnson Co.: E of Kite on Georgia Rte 57, just W of the Emanue l Co. line, elev. 80 m 30 May 1988 D.E. Boufford 23890 ( MO ). Macon Co.: 6.4 mi E of Montezuma on Hwy. GA 224 8 May 1976 J.E. Taylor, Jr. s.n. ( UNC ). Randolph Co.: off of Hwy. 82W, just W of Springvale, 16 0698524N 3522499E 10 Mar 2007 L.C. Majure 2053 ( MISSA ). Sumter Co.: 9 mi WSW of Americus 21 Mar 1966 E. Parker 140 ( LSU ). Tatnall Co.: Big Hammock, 2.7 km SE of Birdsville 19 Jul 2007 J.J. Hill s.n. ( FLAS ). Mississippi. Greene Co.: Palesti nian Gardens off of US Hwy 98, p roperty of James Kirkpatrick
192 22 Jan 2005 L.C. Majure 773 ( MISSA ). Jasper Co.: off of Hwy. 503 just N of Paulding 5 Mar 2005 L.C. Majure 795 ( MISSA ). Lauderdale Co.: E of Lost Gap, N of Hwy. 80 32.3465N, 88.7887W 3 Mar 2007 L.C. Majure 2035 ( MISSA ). Newton Co.: of f of Goodhope Deactur Rd, ca 3.5 mi NE of Decatur, 26 Mar 2005, L C M ajure 828 ( MISSA ). Wayne Co.: Gopher Farm, off of Brewerton Rd., off of Hwy. 63 25 Nov 2005 L.C. M ajure 1290 ( MISSA ). North Carolina. Cumberland Co.: NC 87, 12 mi S of Fayetteville 21 Jun 1958, J.A. Duke 1204 ( UNC ). South Carolina Aiken Co.: a long Interstate 10 W, 10 mi NW of Aiken 23 Jul 2008 Jovonn Hill s.n. ( FLAS ). Bamberg Co.: ca. 3 mi S of Branchville on H wy 21, 29 Oct 1997 B. Summers 8475 ( MO ). Calhoun Co.: ca. 2 mi NNW of Lone Star on SC 267 and 0.9 mi ENE on paved rd. 19 May 1957 H.E. Ahles 25588 ( UNC ). Colleton Co.: Waterboro on US 15, ca. 5.5 mi W of 15 34 14 May 1982 J. Doyle 148 ( UNC ). Darlington Co.: Lake at Hartsville 20 May 1932 B.E. Smith s.n. ( UNC ). Georgetown Co.: 9 mi N of Georgetown 7 Jul 1939 R.K. Godfrey 305 ( MO ). Horry Co.: Myrtle Beach off of Hwy. 17 S (Business) at jct. with 82 Ave. 3 Jun 2009 L.C. Majure 3832 ( FLAS ). Jasper Co.: edge of Ridgeland on US Hwy. 17 12 May 1956 C.R. Bell 2579 ( UNC ). Lexington Co.: just S of I 20 at exit 44, side of SC 34 to Gilbert (ca. 30 mi W of Columbia), 3 May 1996, C.M. Christy 2745 ( US ). W illiamsburg Co.: 5 mi S of Kingstree 10 Jul 1939 R.K. Godfrey 375 ( NY ). 7. Opuntia nemoralis Griffiths M onass tch. Kakteenk pp 133 134. 1913. T YPE : United States. Texas. Gregg County, Longview, Oct 1911, D. Griffiths 10480 ( holotype: US !; see Fig. 7 15A). Opuntia macatei Britton and Rose The Cactaceae 1: 113. 1923. T YPE : United States. Texas.Aransas Co.: Rockport, 28 Dec 1910, W.L. McAtee 1992 (holotype: US!).
193 Plants forming small, spreading shrubs (Fig. 7 15B), oftentimes these forming masses (piles) of cladodes resulting in large patches, mounds, or clones with cladodes ascending to 30 cm tall in the summer; roots typically forming tubers (Fig. 7 15E), but this depends on substrate, and sometimes roots fibrous. Cladodes small, gray green, glaucous, 6.3 (4.5 8.4) cm long, 3.9 (2.8 5.8) cm wide, 11.2 (8.1 14.2) mm wide, obl ong, elliptical, or obovate, the terminal cladodes easily detaching, becoming strongly cross wrinkled during the winter. Leaves glaucous, gray green, ascending parallel to the cladode or slighty spreading, 5.4 (3.7 7.7) mm, ovate. Spines 1 6 produced per a reole (typically 2), white or yellowish during development, aging bright white when mature and then gray in age, strongly retrorsely barbed when developing and into maturity, 2.3 (1.4 3.0) cm long, 0.6 (0.5 0.8) mm in diameter. Glochids bright yellow when young turning a dull brown in age. Flowers: outer tepals triangular to ovate, glaucous, gray green, incurved in bud, inner tepals 7 8, yellow (or rarely tinged pink basally), obovate with a mucronate tip, 3.0 (2.7 3.5) cm long, staminal filaments yellow or greenish yellow, stigmas creamy white or more commonly light green, lobes 4 9. Berries clavate, 3.0 (2.3 4.2) cm long, dark red to pink, or occasionally light green at maturity. Seeds 4.8 (4 .2 5.1) mm long, funicular girdle 1.0 (0.9 1.4) mm wide, funicula r envelope only moderately raised by the hypocotyl and cotyledons, not smooth, bumpy (or rough), funicular girdle irregular, bumpy. Phylogenetic placement Opuntia nemoralis is an allopolyploid derivative of the southeastern and southwestern subclades of the Humifusa clade (Fig. 7 1). Opuntia nemoralis appears to be derived partially from an ancestor of O. humifusa subsp. pollardii ( Chapter 6 ) Phenology. Flowering mid to late April and into May. Distribution Opuntia nemoralis is found from the Oachita M ountains of Arkansas south to southwestern Louisiana in Cameron Parish and through parts of eastern Texas. In Louisiana O.
194 nemoralis is found in saline barrens, and in the Oachitas, the species is found mostly on shale barrens. I have not seen live materia l from Texas. One specimen from Missouri has been tentatively identified as O. nemoralis More fieldwork is needed and l ikely the distribution of the species is much greater than that shown here. Habitat. Opuntia nemoralis commonly occurs on saline or sod ic prairies in Louisiana and on rock outcrops in the Ouachita Mountains of Arkansas. In the Ouachitas, it is commonly associated with O. cespitosa In southern Louisiana populations are in sandy prairies or sandhill communities. Ploidy This species is te traploid 2 n =44, throughout its range in Arkansas and Louisiana (Majure et al. 2012b ), although material from Texas and Missouri has not been counted. Notes. This species has long been placed in synonymy with either O. macrorhiza or O. humifusa (Benson 1982), but the small size of the cladodes, retrorsely barbed spines, easily disarticulating cladodes, and green stigmas set this species apart from the two aforementioned species. As a result of the disarticulating clad od es, Britton and Rose (1920) placed this species in Opuntia series Curassavicae along with O. pes corvi and O. drummondii a mostly synthetic series whose members are from various evolutionarily divergent clades ( Majure et al. 2012a ). Weniger (1967) described plants of O. drummondii from Ga lveston, Texas but that material is referable to O. nemoralis with its sometimes faintly orange centered flowers, and greenish stigma lobes, as well as glaucous gray cladodes Additional specimens examined. United States. Arkansas Garland Co.: Ouachita National Forest, shale barrens off of FR 11, 9.3 km N of Possum Kingdom and Ouachita Lake, 9 March 2011, L.C. Majure 4204 (FLAS). Hempstead Co.: Fulton, sandy soil, 23 May 1909, B.F. Bush 5718 (MO ). Pulaski Co.: Hwy 65/I 540 about 0.9 miles north of Dixon Road exit, 34 40'
195 57.96", 92 16' 09.09" 5 Sep 2010, B.L. Snow 2130 (FLAS). Yell Co.: S side of the Arkansas River, 0.75 mi S of the Hwy. 7 bridge at Dardanelle; 35.21494N 93.14752W, 4 Sept 2007, T. Witsell s .n. (FLAS). Lousiana Beauregard Parish: off of Mouth of the Creek Rd., 24.5 km W of Deridder, 8 March 2011 L.C. Majure 4197 (FLAS) Caddo Parish: Barron Road/Boggy Bayou Saline Prairie, T16N R14W, S27; N of Barron Rd. south of Boggy Bayou, 20 April 2006 B.R. and M.H. MacRoberts 7396 (LSU) Cameron Parish: off of Hwy. 27N, at Johnson Bayou, 25 km E of Sabine Pass, 7 March 2011 L.C. Majure 4196 (FLAS); DeSoto Parish, off of Hwy. 152S, ca. 4 km NNW of Kingston, 8 March 2011 L.C. Majure 4198 (FLAS) Vernon Parish: Rt. 111, 9 mi N of Merriville, just past Bayou River on right, 16 April 1983 J. Doyle 358 (UNC) Winn Parish: between Coldwater and Goldonna, near jct. of Hwy. 156 and Parish Rd. 882, 0.6 mi W of jct. of Hwy. 1233 and Hwy. 153, 27 Apr 2008, B.L. Snow 2053 (FLAS). Missouri Wayne Co.: Greenville q uad, 37.20678N 90.49950W, Corps access rd., 23 Sep 2003, K. Pocklington 445 (MO). Texas Galveston Co.: Galveston Bay, Weniger 687 688 (UNM) (Weniger 1967).
196 Figure 7 1. Phylogeny of the O. h umifusa complex. This is a diploid phylogeny of the Humifusa clade, which consist s of a SE and SW subclade. The O. humifusa complex is represented by the SE clade, as well as the reticulate taxa shown here ( O. cespitosa, O. humifusa subsp. humifusa and O. nemora lis ), and the putative autotetraploid Opuntia humifusa subsp. pollardii (not shown here).
197 Figure 7 2. Morphological features of O. abjecta A) type specimen, J.K. Small s.n. Monroe Co., FL (NY), growth form of 2 x, L.C. Majure 3908 (B) and 4 x L.C. Majure 3318 (C) O. abjecta D) spine development of a terminal cladode of 2 x O. abjecta showing 1 3 spines produced per areole, E) flower bud of O. abjecta F) slighty tuberous roots developing on a specimen planted for almost 2 years, flowers of 2 x (G) an d 4 x (H) O. abjecta and (I) barrel shaped fruit of 2 x O. abjecta
198 Figure 7 3. Geographic distribution of O. abjecta
199 Figure 7 4. Morphological features of O. ochrocentra A) type specimen of O. ochrocentra J.K. Small s.n. Monroe Co., FL (NY), B C) growth forms of O. ochrocentra from B) Big Pine Key, L.C. Majure 3907 and C) Big Munson Island, L.C. Majure 3968 D) spine characters and flower buds of L.C. Majure 3968 E) spine characters of L.C. Majure 3907 showing numerous r adial spines flattened at the base and several central spines twisted or cylindrical at the base, F) flower bud, G) flower, and (H) immature fruit of L.C. Majure 390 7.
200 Figure 7 5. Morphological features of O. austrina A) isotype specime n of O. austrina J.K. Small s.n. Miami Dade Co., FL (US), B) example of O. austrina entity polycarpa, L.C. Majure 3975 with spines up to 10 cm long, C) tuberous roots of O. austrina ( L.C. Majure 4184 left and L.C. Majure 4189 right), growth forms of O. austrina ammophila entity D) L.C. Majure 2754 Marion Co., FL, E) L.C. Majure 4184 Indian River Co., FL. F) long shoot leaves of O. austrina G) flower buds of O. austrina H I) color variation in flowers of O. austrina and J K) fruit color and shape variation of O. austrin a.
201 Figure 7 6. Distribution of O. austrina
202 Figure 7 7. Morphological features of O. cespitosa A) neotype specimen of O. cespitosa L.C. Majure 3275 Woodford Co., KY (FLAS), B) spre ading growth form of O. cespitosa C D) pad shape variation showing glaucous color and cladodes either spiny or spineless, E) occasional tuberous roots of O. cespitosa F) flower and G) fruit of O. cespitosa
203 Figure 7 8. Distribution of O. cespitosa Note: Essex County, Ontario is represented by *.
204 Figure 7 9. Mo r phological features of O. drummondii A) type specimen of O. drummondii L.D. Benson 15388 St. Johns Co., FL (POM), B) spreading/trailing growth form of O. drummondii C) young cladodes, showing long shoot leaves and reddish brown developing spines, D) flower bud of O. drummondii E) fibrous roots of O. drummondii showing proximal thickenings, flower F) and fruit G) of O. drummondii
205 Figure 7 10. Distribution of O. drummondii
206 Figure 7 11. Distribution of O. humifusa
207 Figure 7 12. Morphological features of O. humifusa subsp. humifusa A) type specimen of O. humifusa, C.T. Wherry s.n., Berks Co., PA (US), B) clumping/spreading growth form of O. humifusa subsp. humifusa C D) flower, mature cladodes showing inconspicuous glochids, and immature cladode with leaves, flower E) and fruit F) of O. humifusa subsp. humifusa
208 Figure 7 13. Morphological features of O. humifusa subsp. pollardii. A) type specim en of O. pollardii C.L. Pollard 1138 Harrison Co., MS (NY), B) growth form of O. humifusa subsp. pollardii C E) cladode and spine production variation in O. humifusa subsp. pollardii F) flower and G) fruit of O. humifusa subsp. pollardii
209 Figure 7 1 4. Morphological features of O. humifusa subsp. lata A) type specimen of O. lata J.K. Small s.n. Alachua Co., FL (NY), B E) growth form, cladode shape, and spine production variation in O. humifusa subsp. lata F) flowers and G) fruits of O. humifusa su bsp. lata
210 Figure 7 15. Morphological features of O. nemoralis A) type specimen of O. nemoralis, D. Griffiths 10480 Greggs Co., TX (NY), B) growth form of O. nemoralis C D) cladode shape variation, glaucous, gray green color, and spine variation in O. nemoralis E) tuberous roots of O. nemoralis, F) flower bud, G) flower, and H) red fruit of O. nemoralis
211 Figure 7 16. Distribution of O. nemoralis
212 CHAPTER 8 GENERAL CONCLUSIONS The overall goal of this work was to clarify the phylogenetic limits of Opuntia s.s. and the Humifusa clade (including the O. humifusa complex), as well as to provide an understanding of the distribution of ploidy in the group and the influence of reticulate evolution in species formation with the ultimate purpose of providing a taxonomic revision of the O. humifusa complex. The first o bjective was to determine the phylogenetic relationships among genera of Opuntieae, the circumscription of Opuntia s.s. and the Humifusa clade, as we ll as the biogeographic history and divergence date of Opuntia s.s. Tacinga and Brasiliopuntia along with O puntia lilae and O. schickendantzii form a well supported clade sister to Opuntia s.s. Nopalea is deeply nested within Opuntia s.s., and Consolea is resolved within Opuntia s.s. using ITS data and as sister to the Tacinga + Brasiliopuntia + Opuntia clade using plastid data. Thus, Nopalea is nothing m ore than a group of hummingbird pollinated species of Opuntia Consolea either evolved from a hybrid derived ancestor that originated from Opuntia and a member of some other clade of Opuntieae (regarding plast id and ITS data placements), or ITS data could be confounded by homoplasy or incomplete lineage sorting. Regardless, although the position of Consolea within Opuntieae needs to be tested further, Consolea is monophyletic and should be recognized as a genu s. Opuntia s.s. originated in southern South America in the late Miocene and then subsequently spread to the dry Central Andean Valleys of northwestern South America and the North American Desert region, where it further diversified and formed numerous spe cies via reticulate evolution. Most of those species also are polyploid. The Humifusa clade is well supported. It apparently evolved in the western North American desert re gion at the end of the Pliocene and then migrated to eastern North America, leading to the evolution of the O. humifusa complex.
213 Opuntia lilae was resolved outside of Opuntia s.s. as sister to Tacinga palmadora To resolve the position of O. lilae I reconstructed a phylogeny of the genera of Opuntieae, including O. lilae and mapped morphological characters o n the phylogeny to determine the synapomorphies of Tacinga and to discover those characters shared with O. lilae and Tacinga Tacinga exhibits deep umbilici, reduced perisperm relative to embryo size, hook shaped embry os, and raised stomata, all characters shared with O. lilae I subsequently transferred O. lilae to the genus Tacinga Next, I reconstructed the phylogeny of various clades of Opuntia s.s. to determine the position of O. abjecta, O. militaris and O. tria cantha The polyploid taxa and putative hybrids O. cubensis and O. chrocentra were included to de termine their origins. Although O. militaris and O. triacantha are closely related, they are not sister taxa, suggesting that they may represent different spec ies. Opuntia abjecta is placed in the Humifusa clade, completely unrelated to O. triacantha Opuntia cubensis and O. ochrocentra are resolved in different progenitor clades as well, suggesting that they are derived from separate crosses. Thus, O. abjecta a nd O. militaris are not synonymous with O. triacantha ; O. cubensis and O. ochrocentra are not the same and should be recognized as distinct species. Chromosome counts of the Humifusa clade were carried out to determine the geographic distribution of ploid al levels of members in the group. Diploids were confined to the southwestern (SW) and southeastern (SE) United States, while polyploids were distributed much more broadly: from the southern United States north to Canada. Many of the polyploids display mor phological features from both the SW and SE diploids and are suggested to have originated via hybridization from members of both groups. An ITS phylogeny resolves the diploids in either a SW or a SE subclade, with the polyploids found in either subclade, s upporting the hypothesis
214 that hybridization between members of these subclades led to the formation of certain polyploid taxa. This hybridization and polyploidy most likely occurred after the last glacial maximum, and polyploid taxa subsequently occupied o pen niches northward following glacial retreat. Next, I reconstructed the phylogeny of the Humifusa clade to aid in the determination of species limits and to explore further the origin of polyploid taxa. Diploids again were resolved in two clades (i.e. SW and SE) and numerous polyploids were found to be of allopolyploid origin between the two clades, supporting the hypothesis of polyploid formation and subsequent increase in distribution after the last glacial maximum of the Pleistocene. Opuntia humif usa s.l. was found to be highly polyphyletic and inferred to represent several separate species. I then presented a taxonomic revision of the O. humifusa complex. In the taxonomic treatment, I recognized seven species: O. abjecta, O. austrina, O. cespito sa, O. drummondii, O. humifusa (including three subspecies), O. nemoralis and O. ochrocentr a provided a key for their identification, nomenclatural information, detailed species descriptions, and distribution maps (based on detailed assessment of living populations and herbarium material). Evolution and species delimitation in Opuntia are very complex. F requent hybridiz ation and polyploidy, morphological variation and phenotypic plasticity and the high distortion of dried material, coupled with the lack of basic biological data (e.g., lack of knowledge regarding variation in ploidy, sparse collections due to difficulty in preparing high quality herbarium material) present obstacles in undertaking systematic studies. However, the occurrence of reticulate evolution, often with subsequent polyploidy, offers a very interesting system in which to study the consequences of their important evolutionary events. In addition, the ease of propagation of these plants allows a di stinctive advantage over many other groups in studying
215 developmental patterns in a given species and for readily providing material (e.g., dividing roots tips) for chromosome counts. Future studies will consist of resolving specific relationships among al l major clades in Opuntia s.s. and providing systematic treatments (at the species level) of those taxa. The vast array of morphological diversity also is of interest, and more work will be carried out to determine the phylogenetic pattern of morphological character states in Opuntia s.s. and other clades of Opuntieae. Specifically, shifts in pollination syndrome are of interest as several clades in Opuntieae, including certain members of Opuntia s.s., have switched to hummingbird pollination. Resolving the placement of the problematic genus Consolea is also of special interest.
216 APPENDIX A VOUCHERS USED WITH G ENBANK NUMBERS Taxa, voucher information (collector, herbarium acronym or botanical garden), and Genbank accessions used in our study ( ndhF r pl32, psbJ petA atpB rbcL, trnL F, matK, ycf 1 ppc nrITS). Missing data for a given region is listed as: Material obtained from living collections is cited as: DBG (Desert Botanical Garden, Phoenix, AZ), HBG (Huntington Botanical Garden, San Marino, C A), KEW (Royal Botanic Gardens, UK), MBC (Montgomery Botanical Center, Coral Gables, FL), and SRSC (Sul Ross State University, Alpine, TX). Brasiliopuntia brasiliensis (Willd.) A. Berger DBG 1990 0559 02 Zimmerman 2606 Cult. ( DES) JF787309, JF787469, JF787155, JF712685, JF786712, JN387143, JN387207, JF786876 ; Consolea corallicola Small; DBG 1997 0397 01, L.C. Majure 3321 United States, FL, ( FLAS ) JF787310 11, JF787470 71, JF787156 57, JF712686 87, JF786713 14, JQ676987 86, JF786877 78; Consolea fal cata (Ekman and Werdermann) F.M. Knuth, DBG 1997 0394 Dominican Republic, Bayajibe (DES) JF787312, JF787472, JF712688, JF786715, JQ676988, JF786879; Consolea moniliformis (L.) A. Berger, L.C. Majure 3909 United States, FL, ( FLAS ) JF787313, JF78747 3, JF787158, JF712689, JF786716, JQ676989, JF786880; Consolea nashii (Britton) A. Berger, DBG 1996 0257 C. Fleming s.n, Turks & Caicos Is., South Caicos Island ; DBG 1999 0025, JF787314 15, JF787474 75, JF787159 60, JF712690 91, JF786717 18, JQ676991 90, JF786881 82; Consolea rubescens (Salm Dyck ex A.P. deCandolle) Lem., DBG 1997 0390 cult., MBC, L.C. Majure 3323, United States, FL, Cult., ( FLAS ) JF787316 17, JF787476/JF787606, JF787161 62, JF712692 93, JF786719 20, JQ676992 93, JF786883 84/JF787 126 32; Consolea spinosissima (P.Mill.) Lem., DBG 1995 0389 Jamaica, Hellshire Hills; MBC, L.C. Majure 3322 United States, FL, Cult., ( FLAS ) JF787318 19, JF787477 78, JF787163, JF712694 95, JF786721 22, JQ676995 94,
217 JF786885 86; Maihueniopsis cf. ovata (Pfeiffer) F. Ritter, DBG 2001 0101, DBG 2001 0102, JF787320 21, JF787479 (01), JF712696 97, JF786723 24, JN387144 JN387208 JN387209, JF786887 88; Miqueliopuntia miquelii (Monville) F. Ritter, DBG 1997 0129 E. F. Anderson 6306 Chile, Huasco B ajo Regin. (DES) JF787322, JF787480, JF787164, JF712698, JF786725, JN387145, JN387210, JF786889; Nopalea auberi (Pfeiffer) Salm Dyck, M.P. Griffith 175 ( SRSC ) JF787323, JF787481, JF787165, JF712699, JF786726, JN387146, JN387211, JF786890; Nopalea coche nillifera (L.) Salm Dyck, DBG 1997 0395 Costa Rica, San Jose, D. Lancaster s.n. L.C. Majure 2789 United States, FL, ( FLAS ) JF787324 25, JF787482 83, JF787166 67, JF712700 01, JF786727 28, JN387147, JN387212, JF786891 92; Nopalea dejecta (Salm Dyck) S alm Dyck, DBG 2002 0342 0101 R. Puente 1614 Mexico, Valles. (DES, ASU) JF787326, JF787168, JF786729, JN387148, JN387213, JF786893; Nopalea gaumeri Britton & Rose, DBG 1997 0367 0101 Mexico, Yucatan JF787327, JF787484, JF787169, JF712702, JF786730, JN387149, JN387214, JF786894; Nopalea hondurensis (Standley) Rebman, DBG 1996 0554, DBG 1990 0544 0201 0201 A. D. Zimmerman 2626 Honduras, Olanchito ( DES ) JF787329 30, JF78748 6 87, JF787171 72, JF712704 05, JF786732 33, JF786896 97; Nopalea inaperta Schott ex Griffiths, DBG 1997 0367 Mexico, Yucatan JF787331, JF787488, JF787173, JF712706, JF786734, JN387151, JN387216, JF786898; Nopalea karwinskiana (Salm Dyck) K. Schumann, DBG, JF787332, JF787489, JF787174, JF712707, JF786735, JN387152, JN38 7217, JF786899; Nopalea lutea Rose, DBG 1997 0368 0102 C ult., JF787333, JF787490, JF787175, JF712708, JF786736, JN387153, JF786900; Nopalea nuda Backeberg, M.P. Griffith 171 ( SRSC ) JF787334, JF787491, JF787176, JF712709, JF786737, JF786901; Opun tia abjecta Small, L.C. Majure 3908 United States, FL, ( FLAS ) JF787455, JF787598, JF787300, JF712838, JF786865, JN387199,
218 JN387264, JF787021; Opuntia acaulis Ekman & Werdermann, DBG 1997 0360, JF787335, JF787492, JF787177, JF712710, JF786738, JF786 902/JF787078 83; Opuntia ammophila Small, L.C. Majure 2753, 2826 United States, FL, ( FLAS ) JF787336/JF787463, JF787493 94, JF787178 79, JF712711 12, JF786739 40, JN387154 / JQ676984 JN387218 19, JF786903 04; Opuntia x andersonii H.M.Hernndez, G mez Hin. Brcenas, Puente 1239, San Luis Potosi, Mexico (ASU) JF787337, JF787495, JF787180, JF712713, JF786741, JF786905 ; Opuntia arechavalatae Spegazzini, DBG, JF787338, JF787496, JF787181, JF712714, JF786742, JN387155, JN387220, JF786906; Opuntia arenaria Engelm., R.D.Worthington 36390 United States, TX, ( SRSC ) JF787339, JF787182, JF712715, JF786743, JN387155, JN387220, JF786907; Opuntia assumptionis K. Schumann, DBG, R. Puente 2010 (3) JF787441, JF787586, JF787286, JF712824, JF786846, JF787007; Opuntia atrispina Griffiths, B.L. Snow 2106 United States, TX, ( FLAS ) JF787340, JF787497, JF787183, JF712716, JF786744, JN387155, JN387220, JF786908; Opuntia aurea McCabe ex E.M. Baxter, D. Woodruff 111A United States, UT, ( FLAS ) JF787341, JF 787498, JF787184, JF712717, JF786745, JF786909; Opuntia aureispina (S. Brack & K.D. Heil) Pinkava & B.D. Parfitt, M.P. Griffith 73 ( SRSC ) JF787342, JF787607, JF787185, JF712718, JF786746, JN387158, JN387223, JF786910; Opuntia austrina Small, L.C. M ajure 3450, United States, FL, ( FLAS ) JF787343, JF787499, JF787186, JF712719, JF786747, JQ676985 JN387224, JF786911; Opuntia bahamana Britton & Rose, DBG 1996 0298, JF787344, JF787500, JF787187, JF712720, JF786748, JF787032 37; Opuntia bakeri J.E. Madsen, DBG 1985 0571, JF787345, JF787501, JF787188, JF712721, JF786749, JF786912/ JF787092 97; Opuntia basilaris Engelm. & Bigelow var. basilaris R. Altig s.n., United States, CA, ( FLAS ) JF787346, JF787502, JF787189, JF712722, JF786750, JN387159, JN387225, JF786913;
219 Opuntia bella Britton & Rose, DBG 1997 0400 Colombia, Venicas Del Dagna JF787347, JF787503, JF787190, JF712723, JF786751, JF786914; Opuntia bisetosa Pittier, DBG 1997 0396, JF787348, JF787504, JF787191, JF712724, JF7867 52, JF786915; Opuntia bolding h ii Britton & Rose, DBG 1997 0391, JF787349, JF787505, JF787192, JF712725, JF786753, JF786916; Opuntia bravoana E.M. Baxter, DBG 1939 0094 01 H. Gates s.n. Mexico, Baja California Sur ( DES ) ASDM 2005 0280 01 R. Felg er s.n. Mexico, Sonora (DES) JF787350 51, JF787506 07, JF787193 94, JF712726 27, JF786754 55, JF787038 45; Opuntia camanchica Engelm., J.F. Weedin 374 United States, TX, ( SRSC ) L.C. Majure 3514 United States, TX, ( FLAS ) JF787352/JF787409, JF7875 08/JF787556, JF787195/JF787253, JF712728/JF712788, JF786756/JF786816, JF786917/JF786973; Opuntia caracassana Salm Dyck, DBG 1993 0667, JF787464, JF787509, JF787196, JF712729, JF786757, JN387159, JN387225, JF786918; Opuntia x carstenii R. Puente & C. Hamann, DBG R. Puente 2901 Coahuila, Mexico. (Holotype DES) JF787353, JF787510, JF712730, JF786758, JF786919/JF787111 18; Opuntia cespitosa Raf., L.C. Majure 1380 United States, MS, L.C. Majure 1938 United States, TN, ( MISSA ) JF787354 55, JF787511 12, JF787197 98, JF712731 32, JF786759 60, JF786920 21; Opuntia chaffeyi Britton & Rose, DBG 1990 0238, JF787356, JF787513, JF787199, JF712733, JF786761, JF786922; Opuntia chisosensis (M. Anthony) D.J. Ferguson, DBG 1999 0040, JF787357, JF787514, JF787200, JF712734, JF786762, JN387159, JN387225, JF786923; Opuntia chlorotica Engelm. & Bigelow, DBG 1977 1021, JF787358, JF787608, JF787201, JF712735, JF786763, JN387162, JN387228, JF786924; Opuntia cochabambensis Crdenas, R. Puente 2010 (2), DBG, JF787359, JF787609, JF787202, JF712736, JF786764, JF787046 53; Opuntia cubensis Britton and Rose, L.C. Majure 3907, 3968 United States, FL, ( FLAS ) JF787360 61, JF787515 16, JF787203,
220 JF712737 38, JF786765 66, F786925/JF787054 60/JF786926; Opuntia cymochila Engelm. ex. Bigelo, L.C. Majure 3530 United States, TX, ( FLAS ) JF787362, JF787517, JF787204, JF712739, JF786767, JF786927; Opuntia decumbens Salm Dyck, M.P. Griffith 177, ( SRSC ) JF787363, JF787518, JF787205, JF712740 JF786768, JF786928/JF787133 40; Opuntia dillenii (Salm Dyck.) Ker Gawl., L.C. Majure 3220 United States, FL, MBC, L.C. Majure 3319 United States, FL, ( FLAS ) JF787444 45, JF787588 89, JF787289 90, JF712827 28, JF786854 55, JF787010 11; Opun tia drummondii Graham, L.C. Majure 2094 United States, MS, ( MISSA ) JF787365, JF787520, JF787207, JF712742, JF786770, JN387163, JN387229, JF786930; Opuntia durangensis Britton & Rose, DBG 1988 0166 0201, JF787366, JF787521, JF787208, JF712743, JF786771, JF786931; Opuntia echios J.T. Howell, DBG 1994 0009 E. F. Anderson 2533 Ecuador, Galapagos Is. JF787367, JF787522, JF787209, JF712744, JF786772, JF786932; Opuntia eichlamii Rose, DBG 2011 0005 01 C. Hamann s.n. Guatemala JF787368, JF787610, JF787210, JF712745, JF786773, JF786933; Opuntia elata Link & Otto ex Salm Dyck, R. Puente s.n., United States, AZ, cult., DBG, JF787369, JF787211, JF712746, JF786774, JN387164, JN387230, JF786934; Opuntia ellisiana Griffiths B.L. Snow 1083 United States, TX, ( FLAS ) DBG 1999 0040 0103 cult., JF787370 71, JF787523 24, JF787212 13, JF712747 48, JF786775 76, JN387166 65, JN387232 31, JF786935 36; Opuntia engelmannii Salm Dyck ex Engelm. var. engelmannii L.C. Majure 3586 United States, TX, ( FLAS ) A .M. Powell 6009 United States, TX, ( SRSC ) JF787372 73, JF787525 26, JF787214 15, JF712749 50, JF786777 78, JF786937 38; Opuntia engelmannii Salm Dyck ex Engelm. var. lindheimeri (Engelm.) B.D. Parfitt & Pinkava, L.C. Majure 3506 United States, TX, ( FLAS ) JF787374, JF787527, JF787216, JF712751, JF786779, JF786939; Opuntia engelmannii Salm Dyck ex Engelm. var.
221 linguiformis (Griffiths) B.D. Parfitt & Pinkava, L.C. Majure 3947 United States, NM, ( FLAS ) JF787375, JF787528, JF787217, JF712752, JF786780, JF786940; Opuntia erinacea Engelm. & Bigelow, M. H. 658, RSA; D. Woodruff s.n., United States, UT, ( FLAS ) JF787376 77, JF787529 (658), JF787218 19, JF712753 54, JF786781 82, JF786941 (658); Opuntia excelsa Snchez Mejorada, DBG 1986 0546 1001, JF787378, JF787530, JF787220, JF712755, JF786783, JN387167, JN387233, JF786942; Opuntia ficus indica (L.) P.Mill., L.C. Majure 3225 United States, FL, ( FLAS ) M.P. Griffith 326 ( SRSC ) JF787379 80, JF 787531 32, JF787221 22, JF712756 57, JF786784 85, JF786943 44/JF787101 03; Opuntia fragilis (Haw.) Nutt., E. Ribbens 612 United States, WI, ( MWI ) JF787381, JF787533, JF787223, JF712758, JF786786, JF786945; Opuntia fuliginosa Griffiths, DBG 19 86 0027 1005, JF787382, JF787534, JF787224, JF712759, JF786787, JF786946; Opuntia galapageia Henslow, DBG 1994 0012 01 E. F. Anderson 2540 Galapagos Is., Ecuador JF787383, JF787535, JF787225, JF712760, JF786788, JF786947; Opuntia gosseliniana F.A.C.Weber, R. Puente 3273 B Sierra Mazatan, Sonora (DES, USON) JF787384, JF787611, JF787226, JF712761, JF786789, JN387169, JN387234, JF786948; Opuntia guatemalensis Britton & Rose, DBG 1990 0534 Zimmerman 2609 La Paz, Honduras ( DES ) JF787328, JF787485, JF787170, JF712703, JF786731, JN387150, JN387215, JF786895; Opuntia humifusa (Raf.) Raf., L.C. Majure 37 85, United States, GA, ( FLAS ) ; L.C. Majure 1833, United States, MS, ( MISSA ) JF787385 86, JF787536 37, JF787227 28, JF712762 63, JF786790 91, JN387169 JN387234 JF786949 50; Opuntia hystricina Engelm. & Bigelow, L.C. Majure 3529 United States, NM, ( FLAS ) JF787538, JF787229, JF712764, JF786792, JF786951; Opuntia jamaicensis Britton & Harris, DBG 1997 0357, JF78723 0, JF712765, JF786793, JN387169, JN387234, JF786952; Opuntia keyensis Britton ex Small, L.C.
222 Majure 4156 United States, FL, ( FLAS ) JF787387, JF787235, JF712766, JF786794, ; Opuntia leucotricha A.P. deCandolle, L.C. Majure 3953 United States, F L, ( FLAS ) ; DBG 1987 0448, JF787388 89, JF787539 40, JF787231 32, JF712767 68, JF786795 96, JF786953 54; Opuntia lilae Trujillo & Ponce, DBG 1997 0369 01 Trujillo & Ponce 18643 Venezuela, Sucre JF787390, JF787612, JF787233, JF712769, JF786797, JN387171, JN387237, JF786955; Opuntia lucayana Britton, DBG 1997 0398, JF787391, JF787541, JF787234, JF712770, JF786798, JF786956; Opuntia macbridei Britton & Rose, HBG, DBG 1990 0601, L.C. Majure 3848 United States, FL, C ult., ( FLAS), JF787392 93/J F787423, JF787542 43/JF787616, JF787236 37/JF787269, JF712771 72/JF712806, JF786799 00/JF786833 JN387172 73/84, JN387238 39/49, JF786957 58/JF786990; Opuntia macrocentra Engelm., United States, L.C. Majure 3516 United States, NM, ( FLAS ) JF787394, JF7875 44, JF787238, JF712773, JF786801, JN387174, JN387240, JF786959; Opuntia macrorhiza Engelm., United States, L.C. Majure 3510 United States, TX, ( FLAS ) ; M.H. Baker 15682 United States, NM, (FLAS) JF787395 96, JF787545 46, JF787239 40, JF712774 75, JF78680 2 03, JQ676983 JN387241, JF786960 61; Opuntia magnifica Small, L.C. Majure 3451 United States, FL, cult., ( FLAS ), JF787397, JF787613, JF787241, JF712776, JF786804, JF786962; Opuntia martiniana (L.D. Benson) B.D. Parfitt, DBG 1984 0579, JF787398, JF787547, JF787242, JF712777, JF786805, JF787061 66; Opuntia megacantha Salm Dyck, M.P. Griffith 1288 ( SRSC ) JF787399, JF787548, JF787243, JF712778, JF786806, JF786963/JF787098 100; Opuntia megarhyza Rose, Puente 1884 A Rio Verde, SLP, Mexi co (ASU), JF787400, JF787549, JF787244, JF712779, JF786807, JN387175, JN387242, JF786964; Opuntia megasperma J.T. Howell, DBG 1994 0075, JF787401, JF787550, JF787245, JF712780, JF786808, JF786965; Opuntia microdasys (Lehmann) Pfeiffer, L.C. Majure 3519
223 United States, NM, cult., ( FLAS ) JF787402, JF787551, JF787246, JF712781, JF786809, JN387175, JN387242, JF786966; Opuntia monacantha (Willd.) Haw., L.C. Majure 3847 United States, FL, cult., ( FLAS ) JF787403, JF787552, JF787247, JF712782, JF786810, JF786967; Opuntia orbiculata Salm Dyck ex Pfeiffer, C. Hamann s.n., cult, ( DES ) JF787404, JF787248, JF712783, JF786811, JF786968; Opuntia oricola Philbrick, DBG 1994 0178, JF787405, JF787553, JF787249, JF712784, JF786812, JF786969; Op untia pachyrrhyza H. M. Hernndez, C. Gmez Hinostrosa & R. T. Brcenas, Puente 601 Mexico, San Luis Potosi, (ASU, DES); Puente 1260 Queretaro, Mexico (DES), JF787406 07, JF787554 55, JF787250 51, JF712785 86, JF786813 14, JN387178 79, JN387178, JF78 6970 71 ; Opuntia pailana Weingart, R. Puente 3371 Coahuila, Mexico (DES) JF787408, JF787614, JF787252, JF712787, JF786815, JF786972; Opuntia phaeacantha Engelm., M.P. Griffith 214 United States, ( SRSC ) JF787410, JF787557, JF787254, JF712789, JF786817, JF786974; Opuntia pilifera F.A.C. Weber, DBG, JF787411, JF787558, JF787255, JF712790, JF786818, JF786975; Opuntia pinkavae B.D.Parfitt, D. Woodruff 118A United States, UT, ( FLAS ) JF787559 JF787256, JF712791, JF786819, JF786976, Opuntia pittieri Britton & Rose, DBG 1995 0319, JF787412, JF787560, JF787257, JF712792, JF786820, JF786977/JF787104 110; Opuntia pollardii Britton & Rose, L.C. Majure 1921 United States, MS, ( MISSA ) JF787413, JF787561, JF787258, JF712793, JF786821, JF786978; Opuntia polyacantha Engelm., L.C. Majure 3526, United States, NM, ( FLAS ) ; D. E. Soltis 2902 United States, WY, ( FLAS ) JF787465/ JF787562/ JF787259/ JF712794 95, JF786822 23, / JN3 87180, /JN387245, JF786979/ ; Opuntia pottsii Salm Dyck, A.M. Powell 6897 United States, TX, (SRSC, FLAS ) JF787414, JF787563, JF787260, JF712796, JF786824, JF786980; Opuntia puberula Pfeiffer, DBG 1993 0887 1003,
224 JF787415, JF787615, JF787261, JF7 12797, JF786825, JF786981; Opuntia pubescens Wendland ex Pfeiffer, M.P. Griffiths 300 ( SRSC ) JF712798, JF786982; Opuntia pumila Rose, R. Puente 2297 Mexico, Oaxaca, (DES), JF787416, JF787564, JF787262, JF712799, JF786826, JF786983/JF787141 46; Opuntia pusilla (Haw.) Haw., L.C. Majure 753 United States, MS, ( MISSA ) ; L.C. Majure 1091 United States, AL, ( MISSA ) ; L.C. Majure 1920 United States, MS, ( MISSA, MMNS ) JF787417 19, JF787566 68, JF787263 65, JF712800 02, JF786827 2 9, JN387181 JN387246 JF786984 86; Opuntia pycnantha Engelm., DBG 1987 0916 01 Baja California Sur, Mexico JF787420, JF787565, JF787266, JF712803, JF786830, JN387182, JN387247, JF786987; Opuntia quimilo K. Schumann, DBG 2003 0111 0101 Argentina, cult JF787421, JF787569, JF787267, JF712804, JF786831, JN387183, JN387248, JF786988; Opuntia quitensis F.A.C. Weber, DBG 1988 0262 0201, cult., JF787422, JF787570, JF787268, JF712805, JF786832, JF786989; Opuntia rastrera F.A.C. Weber, DBG, JF787424, J F787571, JF787270, JF712807, JF786834, JF786991; Opuntia repens Bello, L.C. Majure 3837 United States, VI, ( FLAS ) ; L.C. Majure 3838 39, United States, PR, ( FLAS ) JF787425 27, JF787572 74, JF787271 73, JF712808 10, JF786835 37, JF786992 94/JF 787147 54; Opuntia retrorsa Spegazzini, J.R. Abbott 16248, Bolivia, Santa Cruz, ( FLAS ) JF787428, JF787575, JF787274, JF712814, JF786839, JN387185, JN387250, JF786995; Opuntia robusta Wendland, M.P. Griffith 327 ( SRSC ) JF787429, JF787576, JF787275, JF712811, JF786838, JF786996/ JF787119 25; Opuntia rufida Engelm., DBG 199 0 0343 0202 United States TX, Big Bend ; Manning s.n. TX, ( FLAS ) JF787430 31, JF787577/ JF787276 77, JF712812 13, JN387186 87, JN387251 52, JF786840 41, /JF786997; Opuntia sanguinea Proctor, DBG 1996 0297 0101, JF787434, JF787580, JF712817 JF786844, JN387190, JN387255, JF787000; Opuntia santa rita
225 (Griffiths & Hare) Rose, DBG 1940 1421 0103W, JF787435, JF787617, JF787280, JF712818, JF7 86845, JN387191, JN387256, JF787001; Opuntia scheeri F.A.C. Weber, R.Puente s.n., DBG, JF787436, JF787581, JF787281, JF712819, JF786847, JN387192, JN387257, JF787002; Opuntia schickendantzii F.A.C. Weber, DBG 2010 0049 01 Cult. JF787437, JF787582, JF7872 82, JF712820, JF786848, JN387192, JN387257, JF787003; Opuntia schumannii F.A.C.Weber ex A.Berger, DBG 1997 0362, JF787438, JF787583, JF787283, JF712821, JF786849, JF787004; Opuntia setispina Engelm. Ex Salm Dyck, Puente 3656, Cosihuariachi, Chihuahua Mexico (DES), JF787439, JF787584, JF787284, JF712822, JF786850, JF787005; Opuntia soederstromiana Britton & Rose, DBG 1985 0569 0101, JF787440, JF787585, JF787285, JF712823, JF786851, JF787006; Opuntia stenopetala Engelm., M.P. Griffith s.n., DBG, JF787442, JF787618, JF787287, JF712825, JF786852, JN387192, JN387257, JF787008; Opuntia stricta (Haw.) Haw., L.C. Majure 1922, United States, MS, ( MISSA ) JF787443, JF787587, JF787288, JF712826, JF786853, JF787009; Opuntia strigil Engelm. A.M. Powell 6008 ( SRSC ) L.C. Majure 3515 United States, TX, ( FLAS ) Puente 3359 United States, TX (DES) JF787446 48, JF787590 91, JF787291 93, JF712829 31, JF786856 58, JN387195 97, JN387260 62, JF787012 14; Opuntia sulphurea Don, DBG 1995 0372, JF7 87449, JF787592, JF787294, JF712832, JF786859, JF787015; Opuntia tapona Engelm., DBG 1939 0093 0101 Comondu, Baja California, Mexico JF787450, JF787593, JF787295, JF712833, JF786860, JN387198, JN387263, JF787016; Opuntia tomentosa Salm Dyck, M.P. Gr iffith 181 ( SRSC ) ; DBG 1996 0371 0101; DBG 1978 0326 0101, JF787451 53, JF787594 96, JF787296 98, JF712834 36, JF786861 63, JF787017 19/JF787067 69; Opuntia tortispina Engelm., L.C. Majure 3533 United States, TX, ( FLAS ) JF787454, JF787597, JF7872 99, JF712837, JF786864, JF787020; Opuntia triacantha
226 (Willd.) Sweet., Mori et al. 26693 Netherlands Antilles, Saba, ( NY ) JN676104 JN676105 JN676101 JN676103 JN387200, JN387265, JN676102; Opuntia sp. nov. 1 DBG 2003 0155 0102 Puente 1614 Valles, San Luis Potosi, Mexico (DES) JF787456, JF787599, JF787301, JF712839, JF786866, JF787022/JF787070 77; Opuntia sp. nov 2, A.L. Reyna 97 292 Sonora, Mexico (ASU, ARIZ) JF787457, JF787600, JF787302, JF712840, JF786867, JF787023; Opun tia vaseyi (Coult.) Britton & Rose, DBG 1987 0049 0201, JF787458, JF787601, JF787303, JF712841, JF786868, JF787024; Opuntia cf. wilcoxii Britton & Rose, S. Friedman 94 148 Mesiaca, Sonora (ASU, ARIZ) JF787466, JF787602, JF787304, JF712842, JF786869 JF787025; Salmiopu ntia salmiana ( Parmentier ex Pfeiffer ) Guiggi HBG 18366, RBG 2000 1099, JF787432 33, JF787578 79, JF787278 79, JF712815 16, JF786842 43, JN387188 89, JN387253 54, JF786998 99; Tacinga funalis Britton & Rose, AY042660, ; Tacinga inamoena (K.Schumann) Stuppy & Taylor, L.C. Majure 3849 United States, FL, cult., ( FLAS ) ; DBG 1997 0017, JF787467/JF787459, JF787619/ JF787305 06 /JF712843 44, JF786870 71, JN387201 02, JN387201 02, JF787026 2 7; Tacinga palmadora (Britton & Rose) Stuppy & Taylor, DBG 1997 0392 01 Brazil JF787460, JF787603, JF787307, JF712845, JF786872, JN387203, JN387267, JF787028; Tacinga saxatilis (F.Ritter) Stuppy & Taylor, C. Hamann s.n., cult, DBG, JF787468, JF787620, JF 787308, JF712846, JF786873, JN387204, JN387268, JF787029; Tunilla corrugata (Salm Dyck) Hunt & Illiff, DBG 2001 0005 ; Hunt 66371 ( DES ) JF787461 62,, JF787604 05 JF712847 48 JF786874 75 JN387205 06 JN387269 70 JF787030 31
227 APPENDIX B ACCESSIONS USED WITH GENBANK NUMBERS Specimens used in our phylogenetic analysis given with their GenBank accession numbers ( ndhF rpl32, psbJ petA atpB rbcL, trnL F, matK, ycf1, ppc nrITS). Material obtained from botanical gardens is given with the garde n acronym as follows: DBG (Desert Botanical Garden, Phoenix, AZ), HBG (Huntington Botanical Garden, San Marino, CA). Voucher information is provided following the garden acronym where applicable. Brasiliopuntia brasiliensis (Willd.) A. Berger : DBG 1990 055 9 02 Zimmerman 2606 Cult. ( DES) JF787309, JF787469, JF787155, JF712685, JF786712, JN387143, JN387207, JF786876 ; Miqueliopuntia miquelii (Monville) F. Ritter: DBG 1997 0129 E. F. Anderson 6306 Chile, Huasco Bajo Regin. (DES) JF787322, JF787480, JF787164, JF712698, JF786725, JN387145, JN387210, JF786889; Opuntia arechavalatae Spegazzini: DBG, JF787338, JF787496, JF787181, JF712714, JF786742, JN387155, JN387220, JF786906; Opuntia lilae Trujillo & Ponce: DBG 1997 0369 01, Truj illo & Ponce 18643 Venezuela, Sucre (MY), JF787390, JF787612, JF787233, JF712769, JF786797, JN387171, JN387237, JF786955; Opuntia macbridei Britton & Rose : L.C. Majure 3848 United Sta tes, FL (cult., FLAS), JF787423 JF787616 JF787269 JF712806 JF786833, JN387184, JN3872 49, JF786990; Opuntia quimilo K. Schumann : DBG 2003 0111 0101 Argentina, cult. JF787421, JF787569, JF787267, JF712804, JF786831, JN387183, JN387248, JF786988; Opuntia schickendantzii F.A.C. Weber, DBG 2010 0049 01 Puente s.n. Cult. (DES), JF787437, JF787582, JF787282, JF712820, JF786848, JN387192, JN387257, JF787003; Salmio puntia salmiana (Parmentier ex Pfeiffer) Guiggi : HBG 18366, JF787433, JF787579, JF7872 79, JF71 2816, JF7868 43, JN387189, JN3872 54, JF7869 99; Tacinga funal is Britton & Rose: AY042660, ; Tacinga inamoena (K.Schumann) Taylor & Stuppy : L.C. Majure 3849 United States, FL,
228 cult., ( FLAS ) ; DBG 1997 0017, JF787467/JF787459, JF787619/ JF787305 06 /JF712843 44, JF786870 71, JN387201 02, JN387201 02, JF787026 27; Tacinga palmadora (Britton & Rose) Taylor & Stuppy : DBG 1997 0392 01 Brazil (DES), JF787460, JF787603, JF787307, JF712845, JF786872, JN387203, JN387267, JF787028; Tacinga saxatilis (F.Ritter) Taylor & Stuppy: D BG, C. Hamann s.n., cult, (DES), JF787468, JF787620, JF787308, JF712846, JF786873, JN387204, JN387268, JF787029; Tunilla corrugata (Salm Dyck) Hunt & Illiff : DBG 2001 0005, JF787461, JF787604, JF712847, JF786874, JN387205, JN387269, JF787030
229 APPENDIX C SPECIMENS EXAMINED FOR CHAPTER 4 Opuntia abjecta Small: Florida Monroe Co., Big Pine Key, P. Barrtsch s.n., 12 May 1919 (US); E.P. Killip 41332 4 May 1951 (US), E.P. Killip 41708 10 Jan 1952 (US); L.C. Majure 3908 with I. Marino, M. Pajuelo 6 Mar 2010 (FLAS); G.S. Miller, Jr. 1710 22 Feb 1935 (US); J.K. Small s.n. w/ P. Matthaus, 12 April 1921 (type: NY); J.K. Small s.n., 17 May 1922 (NY US); Crawl Key, K. Sauby s.n., July 2008 (FLAS); Long Key, small area of rocky, open, low ground, C. Byrd s.n. 23 April 1966 (FLAS). Opuntia cubensis Britton and Rose: Cuba, Guant namo Bay, Oriente, dry sand, valley near coast, N.L. Britton 2064 17 30 March 1909 (type: NY); Areces Mallea s.n. (FLAS). Opuntia militaris Britton and Rose: Cuba Guantnamo Bay Oriente, coastal hills, N.L. Britton 1957 17 30 March 1909 (Type: NY); Guantnamo Bay, R.N. Jervis 246 4 Sept 1950 (MICH); Schist hills and fringing fossil coral terrace from Escondido Bay to US Naval Operating Base, R.N. Jervis 1033 7 Jan 1951 (MICH) Opuntia ochrocentra Small: Florida Monroe Co., Big Pine Key, Hammock, southern end of Big Pine Key, J.K. Small s.n. with G.K. Small, P. Matthews 11 Dec. 1921 (type: NY); Hammock, southern front of Big Pine Key, J.K. Small s.n. 17 May 1922 (US); Big Pine Key, E.P. Killip 31423 12 18 Feb, 1935 (US); Big Pine Key, hammock, E.P. Killip 31712 2 Mar 1936 (US); Big Pine Key, southeast hammock, E.P. Killip 42026 19 Mar 1952 (US); southern end of Big Pine Key, L.C. Majure 3907 with I. Marino, M. Pajuelo, 6 Mar 2010 (FLAS); Big Munson Island, L.C. Majure 3968 69 with I. Marino, 8 Mar 2010 (FLAS). Opuntia repens Bello: Puerto Rico Punta Melones, coastal rocks, N.L. Britton s.n. 26 Feb 1915 (NY); Lajas, ca. 5km NW of La Parguera, of f of Hwy. 116, L.C. Majure 3838 with T. Majure, F. Axelrod 15 June 2009 (FLAS); Cabo Rojo, Refugio de vida Silvestre, Salinas de
230 Cabo Rojo, L.C. Majure 3839 with T. Majure, F. Axelrod 15 June 2009 (FLAS); St. Thomas off of Hwy. 32E at Red Hook, ca. 1 km NE of interisland ferry, L.C. Majure 3837 with T. Majure 13 June 2009 (FLAS). Opuntia triacantha (Willd.) Sweet: Antigua, H.E. Box 1455 Galley Bay, xerophytic coastal areas, 21 May 1938 (US); J.N. Rose 3304 w/ W.R. Fitch, P.G. Russell, 4 16 Feb 1913 (N Y US); English Harbour, A.C. Smith 10442 4 Apr 1956 (NY US); Guadeloupe, Mornes Basaltiques secs., Isle les Saintes; Terre de Haut, Morne Charreau, H. Stehl 1726 15 May 1917 (NY) ; Raquette volante, Isle les Saintes, Dsirade (Leproserie), P re Duss 3071 1904 (NY US) ; Montserrat rocky cliffs, J.A. Schafer 543 13 Feb 1907 (NY US); Puerto Rico, Desecheo Island, coastal rocks, N.L. Britton 1565 w/ J.F. Cowell, W.E. Hess, 18 19 Feb 1914 (NY); Saba, SW corner of island; Giles Quarter Trail; 17 36' 55''N 63 14' 45''W S.A. Mori 22693 w/ C.A. Gracie, W.R. Buck, P.C. Hoetjes, M. Hoetjes, H. Sipman, J. den Dulk, 4 Mar 2007 (NY); St. Barthelemy A. Anested 924 1939 (US); St. Kitts, Basseterre, J.N. Rose 3241 w/ W.R. Fitch, P.G. Russell, 2 Feb 1913 (NY); St. Martins N.L. Britton 1901 (US); St. Thomas, Buck Island, coastal rocks, N.L. Britton 1388 w/ J.A. Schafer, 25 Feb 1913 (NY).
2 31 APPENDIX D ACCESSIONS USED FOR CHROMOSOME COUNTS Currently recognized Opuntia species investigated are listed (1 6). Syn onyms of recognized species (sensu Benson 1982, Pinkava 2003, and Powell et al. 2008 in part; see Table 1) and their respective ploidy are given below the recognized species name. Recognized species are split by ploidy, where species have more than one cyt otype. Their somatic chromosome number is given along with locality, collector, and repository according to Index Herbariorum (Thiers 2011). Taxa counted for the first time or cytotypes not previously recorded for a species are delimited with an asterisk ( *). All counts were made by L.C. Majure. 1) Opuntia abjecta Small Opuntia abjecta Small; 2 n = 22 Florida, Monroe Co., LCM 3908 (FLAS). Opuntia abjecta Small; 2 n = 44, Florida, Monroe Co., LCM 3318 (FLAS), Monroe Co., KS s.n. (FLAS). 2) Opuntia humifusa (Raf.) Raf Opuntia humifusa (2 x ) taxa: Opuntia ammophila Small ; 2 n = 22, Florida, Brevard Co., LCM 2087 (MISSA), Broward Co., KS 62 (FLAS), Flagler Co., LCM 3222 (FLAS), Indian River Co., LCM 4182 (FLAS), Indian River Co., LCM 4183 (FLAS), India n River Co., LCM 4184 (FLAS), Lake Co., LCM 3246 (FLAS), Lake Co., LCM 4093 (FLAS), Marion Co., LCM 2753 (FLAS), Marion Co., LCM 2754 (FLAS), Marion Co., LCM 2826 (FLAS), Marion Co., LCM 3247 (FLAS), Okeechobee Co., LCM 4185 (FLAS), Okeechobee Co., LCM 418 6 (FLAS), Orange Co., LCM 2086 (MISSA), Orange Co., LCM 3962 (FLAS), Osceola Co., LCM 3702 (FLAS), Osceola Co., LCM 4181 (FLAS), Osceola Co., LCM 4189 (FLAS), Putnam Co., LCM 3248 (FLAS), St. Johns Co., K.S. s.n. (FLAS), St. Lucie Co., LCM 3704 (FLAS), St. Lucie Co., LCM 3705 (FLAS), St. Lucie Co., LCM 3708 (FLAS), Seminole Co., LCM 2085 (MISSA), Volusia Co., LCM 3224 (FLAS), Volusia Co., LCM 3232 (FLAS). Opuntia austrina Small ; 2 n
232 = 22, Florida, Charlotte Co., KS 45 (FLAS), Highlands Co., FL KS 64 (FLAS), Highlands Co., LCM 3450 (FLAS), Highlands Co., LCM 3975 (FLAS), Highlands Co., LCM 3976 (FLAS), Highlands Co., LCM 3978 (FLAS), Okeechobee Co., KS 29 (FLAS), Okeechobee Co., KS 42 (FLAS), Palm Beach Co., LCM 3970 (FLAS), Palm Beach Co., LCM 397 3 (FLAS), Polk Co., KS s.n. (FLAS), Polk Co., LCM 3979 (FLAS). Opuntia lata Small ; 2 n = 22, Alabama, Autauga Co., LCM 2043 (MISSA), Mobile Co., LCM 4194 (FLAS), Florida, Alachua Co., LCM 3991 (FLAS), Alachua Co., LCM 4061 (FLAS), Alachua Co., LCM 4064 (FLAS), Hernando Co., LCM 3948 (FLAS), Highlands Co., LCM 3977 (FLAS), Lafayette Co., LCM 2795 (FLAS), Lake Co., KS 15 (FLAS), Lake Co., LCM 4117 (FLAS), Levy Co., LCM 3645 (FLAS), Manatee Co., LCM 4065 (FLAS), Okaloosa Co., LCM 3954 (FLAS), Okeechobee Co ., LCM 4187 (FLAS), Okeechobee Co., LCM 4188 (FLAS), Orange Co., LCM 4174 (FLAS), Palm Beach Co., LCM 3971 (FLAS), Putnam Co., LCM 4106 (FLAS), Sumter Co., LCM 3238 (FLAS), Sumter Co., LCM 4066 (FLAS), Georgia, Charlton Co., LCM 4190 (FLAS), Crawford Co., JH s.n. (FLAS), Irwin Co., LCM 3785 (FLAS), Houston Co., LCM 3786 (FLAS), Tatnall Co., JH s.n. (FLAS), Mississippi, Newton Co., LCM 938 (MISSA), Wayne Co., LCM 1290 (MISSA), South Carolina, Aiken Co., LCM 3588 (FLAS), Horry Co., LCM 3832 (FLAS). Opuntia humifusa (4 x ) taxa : Opuntia allairei Griffiths ; 2 n = 44, Texas, Liberty Co., LCM 3504 (FLAS). Opuntia cespitosa Raf .; 2 n = 44, Alabama, Bibb Co., LCM 2042 (MISSA), Colbert Co., LCM 2610 (MISSA), Lawrence Co., LCM 2609 (MISSA), Arkansas, Garland Co., LCM 2198 (FLAS), Garland Co., LCM 4203 (FLAS), Garland Co., LCM 4205 (FLAS), Saline Co., LCM 2194 (MISSA), Yell Co., GPJ s.n. (FLAS), Illinois, Cass Co., IL ER s.n. (FLAS), Jo Daviess Co., IL ER s.n. (FLAS), Kentucky, Anderson Co., LCM 3276 (FLAS), Louisiana, Caddo Parish, LCM 4200 (FLAS), Caddo Parish, LCM 4201 (FLAS), Caddo Parish,
233 LCM 4202 (FLAS), Massachusetts, Dukes Co., BC s.n. (FLAS), Mississippi Lee Co., MS JH s.n. (FLAS), Lowndes Co., LCM 755 (MISSA), Oktibbeha Co., LCM 1380 (MISSA), Scott Co., LCM 2563 (MISSA), Tennessee, Bledsoe Co., LCM 1938 (MISSA), Cannon Co., LCM 2072 (MISSA), Davidson Co., JH s.n. (FLAS), Fayette Co., LCM 1956 (MISSA; note O cf. cespitosa ), Fayette Co., JH s.n. (note O. cf. cespitosa FLAS), Franklin Co., BLS 2061 ( FLAS), Lewis Co., JH s.n. (FLAS), Marshall Co., JH s.n. (FLAS), Rutherford Co., JH s.n. (FLAS), Texas, Lamar Co., BS 2069 (FLAS), Virginia, Fredrick Co., LCM 3806 (FLAS). Opuntia humifusa (Raf.) Raf .; 2 n = 44, Alabama, Marion Co., AL JH s.n. (FLAS), Delaware, Sussex Co., LCM 3824 (FLAS), Georgia Dekalb Co., GA LCM 3787 (FLAS), Jackson Co., LCM 3789 (FLAS), Marion Co., JH s.n. (FLAS), Maryland, Alleghany Co., LCM 3810 (FLAS), Massachusetts Barnstable Co., MA LCM 3814 (FLAS), Mississippi, Calhoun Co., MS JH s.n. (FLAS), Carroll Co, LCM 799 (MISSA), Choctaw Co., KP 499 (MMNS), Grenada Co., LCM 1833 (MISSA), Marion Co., J. Hill s. n. (FLAS), Marshall Co., LCM 1293 (MISSA), Montgomery Co., LCM 768 (MISSA), Stone Co., TM s.n. (FLAS), Webster Co ., KP 498 (MMNS), Yalobusha Co., LCM 767 (MISSA), New Hampshire, Rockingham Co., B.Nichols s.n. (FLAS), New Jersey, Atlantic Co., V. Doyle s.n. (FLAS), Burlington Co., LCM 3821 (FLAS), North Carolina, Bladen Co., JH s.n. (FLAS), Currituck Co., LCM 3825 (FL AS), Dare Co., LCM 3827 (FLAS), Onslow Co., LCM 3829 (FLAS), Rowan Co., LCM 3793 (FLAS), Surry Co., JH s.n. (FLAS), South Carolina, Pickens Co., LCM 3790 (FLAS), York Co., LCM 3791 (FLAS), Virginia, Fredrick Co., LCM 3807 (FLAS), Page Co., LCM 3799 (FLAS), Warren Co., LCM 3800 (FLAS), West Virginia, Hampshire Co., LCM 3808 (FLAS), Mineral Co., LCM 3809 (FLAS), Pendleton Co., ER s.n. (FLAS). Opuntia nemoralis Griffiths 2 n = 44, Arkansas, Garland Co., LCM 2192 (MISSA), Garland Co., LCM 2196 (MISSA), Garland Co., LCM 4204
234 (FLAS); Louisiana, Beauregard Parish, CR s.n. (FLAS), Cameron Parish, LCM 4196 (FLAS), DeSoto Parish, LCM 4198 (FLAS), DeSoto Parish, LCM 4199 (FLAS), Winn Parish, BLS 2053 (FLAS). Opuntia cf. nemoralis Griffiths 2 n = 44, Arkansas, Pulaski Co., BLS 2131 (FLAS), Yell Co., TW s.n. (FLAS). Opuntia pollardii Britton & Rose ; 2 n = 44, Alabama, Baldwin Co., LCM 1082 (MISSA), Florida, Santa Rosa Co., LCM 1075 (MISSA), Walton Co., LCM 1067 (MISSA), Walton Co., LCM 1070 (MISSA), Lou isiana, Washington Parish, CR s.n. (FLAS), Mississippi, Forrest Co., LCM 806 (MISSA), Hancock Co., LCM 748 (MISSA), Jackson Co., LCM 1921 (MISSA), Jackson Co., LCM 1297 (MISSA), Jackson Co., LCM 4057 (FLAS), Jackson Co., LCM s.n. (MMNS), Neshoba Co., LCM 1201 (MISSA), Noxubee Co., LCM 1156 (MISSA), Stone Co., TM s.n. (FLAS), Winston Co., LCM 769 (MISSA). 3) Opuntia macrorhiza Engelm Opuntia macrorhiza (2x) taxa: Opuntia xanthoglochia Griffiths 2 n = 22, Texas, Bastrop Co., LCM 1982 (MISSA), Bastrop Co., MJM 949 (FLAS), Fayette Co., LCM 1983 (MISSA), Harris Co., BLS 2089 (FLAS), Milam Co., TX MJM 947 (FLAS), Smith Co., BLS 2082 (FLAS). Opuntia macrorhiza (4 x ) taxa: Opuntia fusco atra Engelm .; 2 n = 44, Texas, Fayette Co., LCM 3505 (FLAS). Opuntia grandiflora Engelm. ; 2 n = 44, Arkansas, Miller Co., BLS 2062 (FLAS), Mississippi, Bolivar Co., LCM 1680 (MISSA), Holmes Co., HS s.n. (FLAS), Yazoo Co., LCM 2366 (MISSA), Texas, Anderson Co., BLS 2077 (FLAS), Austin Co., BLS 2091 (FLAS), Henderson C o., BLS 2081 (FLAS), Jack Co., LCM 3536 (FLAS), Leon Co., BLS 2074 (FLAS), Marion Co., BLS 2086 (FLAS), Smith Co., LCM 3540 (FLAS), Van Zandt Co., BLS 2083 (FLAS). Opuntia macrorhiza (4 x ) taxa: Opuntia fusco atra Engelm .; 2 n = 44, Texas, Fayette Co., LCM 3505 (FLAS). Opuntia grandiflora Engelm. ; 2 n = 44, Arkansas,
235 Miller Co., BLS 2062 (FLAS), Mississippi, Bolivar Co., LCM 1680 (MISSA), Holmes Co., HS s.n. (FLAS), Yazoo Co., LCM 2366 (MISSA), Texas, Anderson Co., BLS 2077 (FLAS), Austin Co., BLS 2091 (FLAS), Henderson Co., BLS 2081 (FLAS), Jack Co., LCM 3536 (FLAS), Leon Co., BLS 2074 (FLAS), Marion Co., BLS 2086 (FLAS), Smith Co., LCM 3540 (FLAS), Van Zandt Co., BLS 2083 (FLAS). 4) Opuntia pusilla (Haw.) Haw. Opuntia pusilla 2 n = 22, Alabama, Lamar Co., JH s.n. (FLAS), Florida, Alachua Co., LCM 4003 (FLAS), Bay Co., KS 307 (FLAS), Bay Co., KS 309 (FLAS), Columbia Co., LCM 4191 (FLAS), Escambia Co., KS 328 (FLAS), Franklin Co., KS 301 (FLAS), Franklin Co., KS 330 (FLAS), Gulf Co., KS 325 (FLAS), Hamilton Co., LCM 4192 (FLAS), Hamilton Co., FL LCM 4193 (FLAS), Levy Co., LCM 2819 (FLAS), Mississippi, Clarke Co., LCM 1270 (MISSA), Forrest Co., LCM 756 (MISSA), Jasper Co., LCM 766 (MISSA), Lamar Co., LCM 1548 (MISSA), Lauderdale Co., LCM 2094 (MISSA) Lauderdale Co., LCM 3919 (MISSA), Lowndes Co., LCM 843 (MISSA), Newton Co., LCM 828 (MISSA), Newton Co., LCM 937 (MISSA), Newton Co., LCM 4211 (FLAS), Perry Co., LCM 757 (MISSA), Smith Co., LCM 753 (MISSA), Wayne Co., TM s.n. (FLAS), Wayne Co., TM s.n. ( FLAS). Opuntia pusilla 2 n = 33, Alabama, Baldwin Co., LCM 1091 (MISSA), Florida, Flagler Co., LCM 3221 (FLAS), St. Johns Co., LCM 3219 (FLAS), Walton Co., LCM 1066 (MISSA), Mississippi, Hancock Co., LCM 1033 (MISSA), South Carolina, Horry Co., JH s.n. (FLAS), Horry Co., LCM 3833 (FLAS). Opuntia pusilla 2 n = 44, Florida, Duval Co., LCM 3700 (FLAS), Nassau Co., CJ s.n. (FLAS), St. Johns Co., LCM 3218 KS 9.4.10 (FLAS), Georgia, Dekalb Co., LCM 3788 (FLAS), Glynn Co., TM s.n. (FLAS ), Mississippi, Jackson Co., LCM 955 (MISSA), Jackson Co., LCM 1920 (MISSA), North Carolina, Dare Co., LCM 3828 (FLAS),
236 Dare Co., LCM 3836 (FLAS), New Hanover Co., LCM 3830 (FLAS), South Carolina, York Co., LCM 3792 (FLAS). 5a) Opuntia stricta (Haw.) Haw. Opuntia dillenii (Ker Gawl.) Haw., 2 n = 66, Florida, Charlotte Co., LCM 3949 (FLAS), Flagler Co., LCM 3220 (FLAS), Monroe Co., LCM 3319 (FLAS), Hillsborough Co., LCM 3952 (FLAS), Puerto Rico, Cabo Rojo, LCM 3843 (FLAS). Opuntia stricta (Haw.) Haw., 2 n = 66, Alabama, Mobile Co., LCM 823 (MISSA), Florida, Clay Co., LCM 3701 (FLAS), Levy Co., LCM 2820 (FLAS), Monroe Co., LCM 3320 (FLAS), St. Johns Co., LCM 3217 (FLAS), Seminole Co., LCM 2083 (MISSA), Mississippi, Jackson Co., LCM 1922 (MISSA). 5b) Putativ e hybrids involving Opuntia stricta Opuntia alta Griffiths 2 n = 66, Louisiana, Cameron Parish, LCM 4195 (FLAS), LaFourche Parish, CR s.n. (FLAS). Opuntia ochrocentra Small 2 n = 55, Florida, Monroe Co., LCM 3907 (FLAS), Monroe Co., LCM 3968 (FLAS), Monroe Co., LCM 3969 (FLAS). 6) Opuntia tortispina Engelm. & J.M. Bigelow. Opuntia tortispina 2 n = 44, New Mexico, Quay Co., LCM 3531 (FLAS), Opuntia tortispina 2 n = 66 New Mexico, Benalillo Co., LCM 3528 (FLAS), Sierra Co., LCM 3521 (FLAS), Ok lahoma, Cimarron Co., ER s.n. (FLAS), Texas, Carson Co., LCM 3532 (FLAS), Hutchinson Co., LCM 3533 (FLAS), Hutchinson Co., LCM 3535 (FLAS).
237 LIST OF REFERENCES Acevedo Rodriguez, P. 1996. Flora of St. John: U.S. Virgin Islands. New York Botanical Garden Press, Bronx, NY. Adams, K. L. and J. F. Wendel. 2005. Polyploidy and genome evolution in plants. Current Opinion in Plant Biology 8: 135 141. Al nuclear micros atellites. Forest Ecology and Management 163: 263 271. Althoff, D. M. and O. Pellmyr. 2002. Examining genetic structure in a bogus yucca moth: sequential approach to phylogeography. Evolution 56: 1632 1643. lvarez, I. and J. F. Wendel. 2003. Ribosomal IT S sequences and plant phylogenetic inference. Molecular Phylogenetics and Evolution 29:417 434. Anderson, E. F. 2001. The cactus family. Timber Press, Portland, OR. Applequist, W. L. and R. S. Wallace. 2001. Phylogeny of the Portulacaceous Cohort based o n ndhF sequence data. Systematic Botany 26:406 419. Arakaki, M., P A. Christin, R. Nyffeler, A. Lendel, U. Eggli, R. M. Ogburn, E. Spriggs, M. J. major succulent plant lin eages. Proceedings of the National Academy of Sciences 108: 8379 8384. Areces Mallea, 2001. A new opuntioid cactus from the Cayman Islands, B. W. I., with a discussion and key to the genus Consolea Lemaire. Brittonia 53: 96 107. Axelrod, D. I. 1948. Clima te and evolution in western North America during middle Pliocene time. Evolution 2: 127 144. Babcock, E. B. and G. L. Stebbins. 1938. The American species of Crepis : their relationships and distribution as affected by polyploidy and apomixis. Carnegia Inst itution of Washington, Publication No. 504. Baker, M. A., J. P. Rebman, B. D. Parfitt, D. J. Pinkava, C. Christy, A. Salywon, R. Puente Martinez, A. Zimmerman, and J. H. Cota. 2009 a Chromosome numbers of miscellaneous angiosperm taxa. Journal of the Botan ical Research Institute of Texas 3: 279 283. Baker, M. A., J. P. Rebman, B. D. Parfitt, D. J. Pinkava, and A. Zimmerman. 2009 b Chromosome numbers in some cacti of Western North America VIII. Haseltonia 15: 117 134. Baker, M. A. 2002. Chromosome numbers an d their significance in some Opuntioideae and Cactoideae (Cactaceae) of mainland Ecuador and Per. Haseltonia 9: 69 77.
238 Brcenas, R. T., T. Yasson, and J. A. Hawkins. 2011. Molecular systematics of the Cactaceae. Cladistics 27: 1 20. Barrow, M. and A. Meister. 2003. Endopolyploidy in seed plants is differently correlated to systematics, organ, life strategy and genome size. Plant, Cell and Environment 26: 571 584. Barthlott, W. 1983. Biogeography and evolution in neo and paleotropical Rhipsalinae. Pp 2 41 248 in Dispersal and Distribution, ed. K. Kubitzki. Verlag Paul Parey, Hamburg, Germany. Benson, L. 1962. Plant Taxonomy. The Ronald Press Company, New York. Benson, L. 1982. The cacti of the United States and Canada. Stanford University Press. Bergs ten, J. 2005. A review of long branch attraction. Cladistics 21: 163 193. Bessega, C., J. C. Vilardi, and B. O. Saidman. 2006. Genetic relationships among American species of the genus Prosopis (Mimosoideae, Leguminosae) inferred from ITS sequences: evide nce for long distance dispersal. Journal of Biogeography 33: 1905 1915. Bowden, W. M. 1945 a A list of chromosome numbers in higher plants. I. Acanthaceae to Myrtaceae. American Journal of Botany 32: 81 92. Bowden, W. M. 1945 b A list of chromosome numbers in higher plants. II. Menispermaceae to Verbenaceae. American Journal of Botany 32: 191 201. Britton, N.L. 1911. Shorter Notes. Torreya 11: 152. Britton, N.L. and J.N. Rose. 1908. A preliminary treatment of the Opuntioideae of North America. Smithsonian M iscellaneous Collection 1786. Britton, N. L. and J. N. Rose. 1912. Undescribed species of Cuban cacti. Torreya 12: 13 16. Britton, N. L. and J. N. Rose. 1920. The Cactaceae. Vol. 1. Carnegie Institute of Washington, Washington, DC. Britton, N. L. and J. N Rose. 1923. The Cactaceae Vol. 4. Carnegie Institute of Washington, Washington, DC. Casas, A. and G. Barbera. 2002. Mesoamerican domestication and diffusion. In P.S. Nobel, [ed], Cacti: biology and uses, 143 162. University of California Press, Berkeley CA. Clement, M., D. Posada, and K. A. Krandall. 2000. TCS: a computer program to estimate gene genealogies. Molecular Ecology 9: 1657 1659. Department of Agriculture and Consumer Services, Gainesville, FL.
239 Cota, J. H. and C. T. Philbrick. 1994. Chromosome number variation and polyploidy in the genus Echinocereus (Cactaceae). American Journal of Botany 81:1054 1062. many angiosperm species? American Journal of Botany 96: 366 381. Crozier, B.S. 2004. Subfamilies of Cactaceae Juss., including Blossfeldioideae subfam. nov. Phytologia 86:52 65. Darlu, P. and G. Lecointre. 2002. When does the incongruence length difference test fail? Molecular Biology and Evolution 19: 432 437. Davis, M. B. and R. G. Shaw. 2001. Range shifts and adaptive responses to Quaternary climate change. Science 292: 673 679 doi: 10.1126/science.292.5517.673 Davis, O. K., L. Agenboard, P. S. Martin, and J. I. Mead. 1984. The Pleistocene blanket of Bechan Cave, Utah. Special Publication of Carnegie Museum of Natural History 8: 267 282. de Castro, J. P. 2008. Nmeros cromoss micos em espcies de Cactaceae ocorrentes no nordeste do Brasil. Dissertao. Universidade Federal da Paraba, Paraba. de Queiroz, K. 2007. Species concepts and species delimitation. Systematic Biology 56: 879 886. Dean, W. R. J. and S. J. Milton. 2000. Directed dispersal of Opuntia species in the Karoo, South Africa: are crows the responsible agents? Journal of Arid Environments 45: 305 314. deWet, J. M. J. 1971. Polyploidy and evolution in plants. Taxon 20: 29 35. deWet, J. M. J. and H. T. Stalker. 1 974. Gametophytic apomixis and evolution in plants. Taxon 23: 689 697. Daz, L. and A. A. Cocucci. 2003. Functional gynodioecy in Opuntia quimilo (Cactaceae), a tree cactus pollinated by bees and hummingbrids. Plant Biology 5: 531 539. Dodd, A. P., 1940. The biological campaign against prickly pear. Commonwealth Prickly Pear Board, Brisbane, Australia. Donoghue, M.J. 1985. A critique of the biological species concept and recommendations for a phylogenetic alternative. Bryologist 88: 172 181. Doyle, J. J. L. E. Flagel, A. H. Paterson, R. A. Rapp, D. E. Soltis, P. S. Soltis, and J. F. Wendel. 2008. Evolutionary genetics of genome merger and doubling in plants. Annual Review of Genetics 42: 443 461. doi: 10.1146/annurev.genet.42.110807.091524 Doyle, J. J. a nd J. L. Doyle. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11 15.
240 Doyle, J.D. 1990. Systematics of the Opuntia humifusa complex. Ph.D. dissertation. University of North Carolina, Chapel Hill North Carolina, USA. Edgar, R.C. 2004. MUSCLE: multiple sequence alignment with high accuracy and throughput. Nucleic Acid Research 32:1792 1797. Edwards, E. J., R. Nyffeler, and M. J. Donoghue. 2005. Basal cactus phylogeny: implications of Pereskia (C actaceae) paraphyly for the transition to the cactus life form. The American Naturalist 92:1177 1188. Eggli, U. 1984. Stomatal types of Cactaceae. Plant Systematics and Evolution 146: 197 214. Engelmann, G. 1856. Synopsis of the Cactaceae of the territor y of the United States and adjacentregions. Proceedings of the American Academy 3: 259 346. Farris, J. D., M. Kallersjo, A. G. Kluge, and C. Bult. 1995. Constructing a significance test for incongruence. Cladistics 10:315 319. Felker, P., R. Zapata, X. Wa ng, D. Medina, R. Bunch, and A. Paterson. 2010. Fruit characters among apomicts and sexual progeny of a cross of the Texas native Opuntia lindheimeri (1250) with a commercial fruit type Opuntia ficus indica (1281). Journal of the Professional Association o f Cactus Development 12: 48 66. Felker, P., S. del C. Rodriguez, R. M. Casoliba, R. Filipini, D. Medina, and R. Zapata. 2005. Comparison of Opuntia ficus indica varieties of Mexican and Argentine origin for fruit yield and quality in Argentina. Journal of Arid Environments 60: 405 422. Fenster, C. B., W. S. Armbruster, P. Wilson, M. R. Dudash, and J. D. Thomson. 2004. Pollination syndromes and floral specialization. Annual Review of Ecology, Evolution, and Systematics 35: 375 403. Ferguson, D.J. 1987. Opu ntia cymochila Eng. and Big.: a species lost in the shuffle. Cactus and Succulent Journal 59: 256 260. Fernndez, A. 1995. Anatoma en especies de Opuntia (Cactaceae) de Venezuela. I. caracteres epidermicos de Opuntia lilae Trujillo et Ponce. Ernstia 5: 49 72. Fishbein, M., C. Hibsch Jetter, D.E. Soltis, and L. Hufford. 2001. Phylogeny of Saxifragales (Angiosperms, Eudicots): analysis, of a rapid, ancient radiation. Systematic Biology 50: 817 847. Foxcroft, L. C. and M. Rejmanek. 2007. What helps Opunt ia stricta invade Kruger National Park, South Africa: baboons or elephants? Applied Vegetation Science 10: 265 270. Foxcroft, L. C., M. Rouget, D. M. Richardson, and S. MacFayden. 2004. Reconstructing 50 years of Opuntia stricta invasion in the Kruger Nat ional Park, South Africa: environmental determinants and propagule pressure. Diversity and Distributions 10: 427 437.
241 Franck, A.R., B.J. Cochrane, and J.R. Garey. In press. Low copy nuclear primers and ycf1 primers in Cactaceae. American Journal of Botany Primer Notes Freeman, D. B. 1992. Prickly pear menace in eastern Australia 1880 1940. Geographical Review 82: 413 429. Futuyma, D. J. 1998. Evolutionary Biology. Third Edition. Sinauer Associates, Inc., Massachusetts. Gibbes, L.R. 1859. Contributions t o the library. Proceedings of the Elliott Society of Natural History 1: 272 273. Gibson, A. C. and P. S. Nobel. 1986. The Cactus Primer. Harvard University Press, MA. Gmez Hinostrosa, C. and H. M. Hernndez. 2000. Diversity, geographical distribution, and conservation of Cactaceae in the Mier y Noriega region, Mexico. Biodiversity and Conservation 9:403 418. Graham, S.W., Reeves, P.A., Burns, A.C.E., and R.G. Olmstead. 2000. Microstructural changes in noncoding chloroplast DNA: interpretation, evolution and utility of indels and inversions in basal angiosperm phylogenetic inference. International Journal of Plant Sciences 161: S83 S96. Grant, V. and K. A. Grant. 1982. Natural pentaploids in the Opuntia linheimeri phaeacantha group in Texas. Botanical G azette 143: 117 120. Grant, B. R. and P. R. Grant. 1981. Exploitation of Opuntia cactus by birds on the Galapagos. Oecologia 49: 179 187. Grant, V. 1981. Plant Speciation. 2nd edition. Columbia University Press, NY. Grant, V. 1994. Modes and origins of mechanical and ethological isolation in angiosperms. Proceedings of the National Academy of Sciences 91: 3 10. Grant, V. and K. A. Grant. 1982. Natural pentaploids in the Opuntia linheimeri phaeacantha group in Texas. Botanical Gazette 143: 117 120. Griffith, M. P. 2003. Using molecular data to elucidate reticulate evolution in Opuntia (Cactaceae). Madroo 50:162 169. Griffith, M. P. 2004 The origins of an important cactus crop, Opuntia ficus indica (Cactaceae): new molecular evidence. American Journal of Botany 91: 1915 1921. Griffith, M. P. and M. Porter. 2003. Back to the basics: a simple method of DNA extraction for mucilaginous cacti Bradleya 21:126 128. Griffith, M.P. and M. Porter. 2009. Phylogeny of Opuntioideae (Cactaceae). International Journal of Plant Sciences 170: 107 116.
242 Griffiths, D. 1913. Einige neue Opuntioideen. Monatsschrift fr Kakteenkunde 23: 130 140. Guiggi, A. 20 10. Genera nova et combinations novae in cactaceis austroamericanis ad subfamiliam Opuntioideae K. Schumann spectantibus. Supplement to Cactology 2: 1 4. Hamrick, J. L., J. D. Nason, T. H. Fleming, and J.M. Nassar, 2002. Genetic diversity in columnar cact i. Pp 122 123 in Columnar Cacti and Their Mutualists: Evolution, Ecology, and Conservation, eds. H. Fleming and A. Valiente Banet. University of Arizona Press, Tucson, AZ. Hanks, S. and D. E. Fairbrothers. 1969. Diversity of populations of Opuntia compres sa (Salisb.) Macbr. in New Jersey. Bulletin of the Torrey Botanical Club 96: 641 652. Harlan, J. R. and J. M. J. deWet. 1975. On Winge and a prayer: the origins of polyploidy. Botanical Review 41: 361 390. Hawkins, J. A., N. Boutaoui, K. Y. Cheung, R. D van Klinken, and C. E. Hughes. 2007. Intercontinental dispersal prior to human translocation revealed in a cryptogenic invasive tree. New Phytologist 175: 575 587. Haworth, A.H. 1803. Miscellanea Naturalia, sive Dissertationes Variae ad Historiam Natura lem Spectantes. London. Haworth, A.H. 1812. Synopsis plantarum succulentarum. Richard Taylor, London. Haywood, A. M. and P. J. Valdes. 2004. Modelling Pliocene warmth: contribution of atmosphere, oceans, and cryosphere Earth and Planetary Science Lett ers 218: 363 377. Haywood, A. M., P. J. Valdes, B. W. Sellwood, J. O. Kaplan, and H. J. Dowsett. 2001. Modelling middle Pliocene warm climates of the USA. Palaeontologia Electronica 4: 1 21. Helsen, P., R. A. Browne, D. J. Anderson, P. Verdyck, and S. V Opuntia (prickly pear) cacti: extensive morphological diversity, low genetic variability. Biological Journal of the Linnean Society 96: 451 461. Henslow, J. S. 1837. Description of two new species of Opuntia ; with remarks on t he structure of the fruit of Rhipsalis Magazine of Zoology and Botany 1: 466 469. Hernndez Hernndez, T., H. Hernndez J. A. de Nova, R. Puente, L. E. Eguiarte, and S. Magalln. 2011. Phylogenetic relationships and evolution of growth form in Cactacea e (Caryophyllales, Eudicotyledoneae). American Journal of Botany 98: 44 61. Hernndez, H. M., C. Gmez Hinostrosa, and R. T. Brcenas. 2001. Diversity, spatial arrangement, and endemism of Cactaceae in the Huizache area, a hot spot in the Chihuahuan Deser t. Biodiversity and Conservation 10:1097 1112.
243 Hershkovitz, M. A. and E. A. Zimmer. 1997. On the evolutionary origin of the cacti. Taxon 46: 217 232. Hershkovitz, M. A. and E. A. Zimmer. 2000. Ribosomal DNA evidence and disjunctions of western American Por tulacaceae. Molecular Phylogenetics and Evolution 15: 419 439. Hewitt, G. 2000. The genetic legacy of the Quaternary ice ages. Nature 405: 907 913. Huang, X., Q. Li, Y. Zhang, Q. Lu, L. Guo, L. Huang, and Z. He. 2008. Neuroprotective effects of cactus poly saccharide on oxygen and glucose deprivation induced damage in rat brain slices. Cell and Molecular Neurobiology 28: 559 568. Hulsenbeck, J. P. and F. R. Ronquist. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17:754 755. Hunt, D Opuntia that was! Succulent Plant Research 6: 245 248. Hunt, D.R. 2006. The New Cactus Lexicon. dh books, The Manse, Chapel Lane, Milbourne Port, England. Husband, B. C. and D. W. Schemske. 1998. Cytotype distribution at a diploid tetra ploid contact zone in Chamerion ( Epilobium ) angustifolium (Onagraceae). American Journal of Botany 85: 1688 1694. Inglese, P., F. Basile, and M. Schirra. 2002. Cactus pear fruit production. Pp 163 183 in Cacti: biology and uses, ed. P. S. Nobel. Universit y of California Press, Berkeley, CA. Ionta, G. M., W. S. Judd, N. H. Williams, W. M. Whitten. 2007. Phylogenetic relationships in Rhexia (Melastomataceae): from DNA sequence data and morphology. International Journal of Plant Sciences 168: 1055 1066. Jansen, D. H. 1986. Chihuahuan Desert nopaleras: defaunated big mammal vegetation. Annual Review of Ecology and Systematics 17: 595 636. Jiao, Y., N. J. Wickett, S. Ayyampalayam, A. S. Chanderbali, L. Landherr, P. E. Ralph, L. P. Tomsho, Y. Hu, H. Liang, P S, Soltis, D. E. Soltis, S. W. Clifton, S. E. Schlarbaum, S. C. Schuster H. Ma, J. Leebens Mack, C. W. dePamphilis. 2011. Ancestral polyploidy in seed plants and angiosperms. Nature doi: 10.1038/nature09916 Johnson, L. A. and D. E. Soltis. 1998. Asses sing congruence: empirical examples from molecular data. Pp 297 348 in Molecular Systematics of Plants II, eds. D.E. Soltis, P.S. Soltis, and J. Doyle. Kluwer, Boston. Johnston, I. M. 1940. The floristic significance of shrubs common to North and South Ame rican areas. Journal of the Arnold Arboretum 21: 356 363.
244 Judd, W.S., Soltis, D.E., Soltis, P.S., and G. Ionta. 2007. Tolmiea diplomenziesii : A new species from the Pacific Northwest and the diploid sister taxon of the autotetraploid T. menziesii (Saxifragaceae). Brittonia 59: 217 225. Kalmbacher, G. 1976. Correct name for the Opuntia of the eastern U.S. Cactus and Succulent Journal 48: 27 28. Kato, A. 1999. Air drying method using nitrous oxide for chromosome counting in maize. Biotechnic and His tochemistry 74: 160 166. Kiesling, R. 1998. Origen, domesticacin, y distribucin de Opuntia ficus indica Journal of the Professional Association for Cactus Development 3. Online at http://www.jpacd.org/contents1998.html. Kim, S T., S. E. Sultan, and M. J. Donoghue. 2008. Allopolyploid speciation in Persicaria (Polygonaceae): insights from a low copy nuclear region. Proceedings of the National Academy of Science 105: 12370 12375. Kovarik, A., J. C. Pires, A. R. Leitch, K. Y. Lim, A. M. Sherwood, R. Matay asek, J. Rocca, D. E. Soltis, and P. S. Soltis. 2007. Rapid concerted evolution of nuclear ribosomal DNA in two Tragopogon allopolyploids of recent and recurrent origin. Genetics 169: 931 944. Lambert, S. M. 2009. Tacinga, the hummingbird pollinated prick ly pear. Cactus and Succulent Journal 81: 156 161. Laurenz, J. C., C. C. Collier, and J. O. Kuti. Hypoglycaemic effect of Opuntia lindheimeri Engelm. in a diabetic pig model. Phytotherapy Research 17: 26 29. Leitch, I. J. and M. D. Bennett. 1997. Polyploid y in angiosperms. Trends in Plant Science 2: 470 476. Leuenberger, B. E. 1976. Pollen morphology of the Cactaceae. The Cactus and Succulent Journal of Great Britain 38: 79 94. Leuenberger, B. E. 2001. The South American Opuntia ser. Armatae (= O ser. Ela tae ) (Cactaceae). Botanische Jahrbucherfur Systematik, Pflanzengeschichte und Pflanzengeographie 123: 413 439. Leuenberger, B.E. 1993. Interpretation and typification of Cactus opuntia L., Opuntia vulgaris Mill., and O. humifusa (Rafin.) Rafin. (Cactaceae) Taxon 42: 419 429. Levin, D.A. 1983. Polyploidy and novelty in flowering plants. The American Naturalist 122: 1 25. Lewis, H. 1967. The taxonomic significance of autopolyploidy. Taxon 16: 267 271. Lewis, W. H.1980. Polyploidy in species populations. Pp 1 03 144 in Polyploidy: biological relevance, ed. W. H. Lewis. Plenum Press, New York.
245 Lewis, P.O. 2001. A likelihood approach to estimating phylogeny from discrete morphological character data. Systematic Biology 50: 913 925. Lia, V. V., V. A. Confalonieri C. I. Comas, and J. H. Hunziker. 2001. Molecular Phylogeny of Larrea and Its Allies (Zygophyllaceae): Reticulate Evolution and the Probable Time of Creosote Bush Arrival to North America. Molecular Phylogenetics and Evolution 21: 309 320. Loik, M. E. a nd P. S. Nobel. 1993. Freezing tolerance and water relations of Opuntia fragilis from Canada and the United States. Ecology 74: 1722 1732. Maddison, W. P. and D. R. Maddison. 2010. Mesquite: a modular system for evolutionary analysis. Version 2.73 http://mesquiteproject.org Majure, L. C., R. Puente, and D. J. Pinkava. In Review. Miscellaneous chromosome counts in Opuntieae (Cactaceae) with a compilation of counts for the group. Haseltonia Majure, L.C. 2007. The ecology and morphological variation of Opuntia species in the mid Majure, L.C. 2010. Towards an evolutionary understanding of the Opuntia humifusa complex. Cactus and Succulent Journal 82:156 163. Majure, L.C. and G. N. Ervin. 2008. The Opuntia (Cactaceae) of Mississippi. Haseltonia 14: 111 126. Majure, L.C. and E. Ribbens. 2012. Chromosome counts of Opuntia (Cactaceae), prickly pear cacti, in the Midwestern United States and environmental factors restricting the distribution of Opuntia fragilis Haseltonia 17: 58 65. Majure, L.C., R. Puente, M.P. Griffith, W.S. Judd, P.S. Soltis, and D.E Soltis. 2012 a Phyl ogeny of Opuntia s.s. (Cactaceae): reticulate evolution, geographic origins, and clade delineation. American Journal of Botany 99: 847 864. Majure, L.C., D.E. Soltis, P.S. Soltis, and W.S. Judd. 2012 b Cytogeography of the Humifusa clade of Opuntia s.s. ( Cactaceae: Opuntioideae): Correlations with geographic distributions and morphological differentiation of a polyploid complex. Comparative Cytogenetics 6: 53 77. Malcomber, S. T. 2002. Phylogeny of Gaertnera Lam. (Rubiaceae) based on multiple DNA markers: evidence of a rapid radiation in a widespread, morphologically diverse genus. Evolution 56: 42 57. Marsico, T. D., L. E. Wallace, G. N. Ervin, C. P. Brooks, J. E. McClure, and M. E. Welch. 2010. Geographic patterns of genetic diversity from the native rang e of Cactoblastis cactorum (Berg) support the documented history of invasion and multiple introductions for invasive populations. Biological Invasions DOI 10.1007/s10530 010 9874 9.
246 Marshall, L. G., R. F. Butler, R. E. Drake, G. H. Curtiss, and R. H. Tedfo rd. 1979. Calibration of the great American interchange. Science 204: 272 279. Maund, B. 1846. The Botanist: containing accurately colored figures of tender and hardy ornamental plants; with descriptions, scientific and popular; intended to convey both m oral and intellectual gratification. 5: pl. 246. Mauseth, J.D. 2006. Structure function relationships in highly modified shoots of Cactaceae. Annals of Botany 98: 901 926. Mavrodiev E. V., D. E. Soltis, and P. S. Soltis. 2010. Plastid sequence data sugg ests that the genus Rohrbachia (Typhaceae) is sister to Typha s. str. (Typhaceae): reply to A. Zernov. Bulletin of Moscow Society of Naturalists 115: 72 74. Mavrodiev, E., D. E. Soltis, and P. S. Soltis. 2008. Parentage of six old world polyploids in Trag opogon L. (Asteraceae: Scorzonerinae) based on ITS, ETS and plastid sequence data. Taxon 57: 1215 1232. Mayer M. S., L. M. Williams, and J. P. Rebman. 2000. Molecular evidence for the hybrid origin of Opuntia prolifera (Cactaceae). Madroo 47:109 115. May r, E. 1942. Systematics and the origin of species. Columbia University Press, New York. Mellink, E. and M. E. Riojas Lpez. 2002. Consumption of platyopuntias by wild vertebrates. Pp 109 123 in Cacti: biology and uses, ed. P. S. Nobel. University of Calif ornia Press, Berkeley, CA. Menezes, M.O.T., N.P. Taylor, M.C. Machado, P.J.A. Coelho, and D. Correia. 2011. Diversity and distribution of Cactaceae in Cear state, north eastern Brazil. Bradleya 29: 13 42. Miretzky, P., C. Muoz, and A. Carrillo Chvez. 2 008. Experimental binding of lead to a low cost on biosorbent: Nopal ( Opuntia streptacantha ). Bioresource Technology 99: 1211 1217. Moore, M. J., A. Tye, and R. K. Jansen. 2006. Patterns of long distance dispersal in Tiquilia subg. Tiquilia (Boraginaceae) : implications for the origins of amphitropical disjuncts and Galapagos Islands endemics. American Journal of Botany 93: 1163 1177. Nassar, J. M., J. L. Hamrick, and T. H. Fleming. 2003. Population genetic structure of Venezuelan chiropterophilous columna r cacti (Cactaceae). American Journal of Botany 90: 1628 1637. Naumova, T.N. 1993. Apomixis in angiosperms: nucellar and integumentary embryony. CRC Press, Inc., Boca Raton, FL. Ncibi, S., M. B. Othman, A. Akacha, M. N. Krifi, and L. Zourgui. Opuntia ficu s indica extract protects against chlorpyrifos induced damage on mice liver. Food and Chemical Toxicology 46: 797 802.
247 Nefzaoui, A. and H. B. Salem. 2002. Forage, fodder, and animal nutrition. Pp 199 210 in Cacti: biology and uses, ed. P. S. Nobel. Unive rsity of California Press, Berkeley, CA. Negrn Ortiz, V. 2007. Chromosome numbers, nuclear DNA content, and polyploidy in Consolea (Cactaceae), an endemic cactus of the Caribbean Islands. American Journal of Botany 94:1360 1370. Nesom, G. L.1983. Galax (D iapensiaceae): geographic variation in chromosome number. Systematic Botany 8: 1 14. Neubig, K. M., W. M. Whitten, B. S. Carlsward, M. A. Blanco, L. Endara, N. H. Williams, and M. Moore. 2008. Phylogenetic utility of ycf1 in orchids: a plastid gene more va riable than matK. Plant Systematics and Evolution 277: 75 84. Nobel, P.S. and E.G. Bobich. 2002. Environmental Biology. Pp 57 74 in Cacti: biology and uses, ed. P. S. Nobel. University of California Press, Berkeley, CA. Novoa, S. 2006. Sobre el origen de l a tuna en el Per: algunos alcances. Zonas ridas 10: 174 181. Nyffeler, R. 2002. Phylogenetic relationships in the cactus family (Cactaceae) based on evidence from trnK/matK and trnL trnF sequences. American Journal of Botany 89:321 326. Nyffeler, R. 2007. The closest relatives of cacti: insights from phylogenetic analyses of chloroplast and mitochondrial sequences with special emphasis on relationships in the tribe Anacampseroteae. American Journal of Botany 94:89 101. Nyffeler, R. and U. Eggli. 2010a A farewell to dated ideas and concepts: molecular phylogenetics and a revised suprageneric classification of the family Cactaceae. Schumannia 6: 109 149. Nyffeler, R. and U. Eggli. 2010b Disintegrating Portulacaceae a new familial classification of the suborder Portulacineae (Caryophyllales) based on molecular and morphological investigations. Taxon 59: 227 240. Olsson, U., P. Alstrom, M. Gelang, P.G. Ericson, and P. Sundberg. 2006 Phylogeography of Indonesian and Sino Himalayan bush warblers (Cettia, Aves). Molecular Phylogenetics and Evolution 41: 556 561. Otto, S. P. and J. Whitton. 2000. Polyploidy incidence and evolution. Annual Reviews in Genetics 34: 401 437. Parfitt, B. D. 1985. Dioecy in North American Cactaceae: A review. Sida 11: 200 206. Parfitt, B. D. 1991. Biosystematics of the Opuntia polyacantha (Cactaceae) complex of Western North America. Ph.D. Dissertation. Arizona State University, Tempe.
248 Parfitt, B. D. 1997. Opuntia pinkavae (Cactaceae), a new species from Arizona and Utah. Rhodora 99:223 228. Pennys, D. S. and W. S. Judd. 2005. A systematic revision and cladistics analysis of Charianthus (Melastomataceae) using morphological and molecular data. Systematic Botany 30: 559 584. Pinkava, D. J., M. A. Baker, B. D. Parfitt, and M. W. Mohlenbrock. 1985. Chromosome numbers in some cacti of western North America V. Systematic Botany 10: 471 483. Pinkava, D. J., L. A. McGill, T. Reeves, and M. G. McLeod. 1977. Chromo some numbers in some cacti of western North America III. Bulletin of the Torrey Botanical Club 104: 105 110. Pinkava, D. J. and M. G. McLeod. 1971. Chromosome numbers in some cacti of western North America. Brittonia 23: 171 176. Pinkava, D. J., M. G. McLe od, L. A. McGill, and R. C. Brown. 1973. Chromosome numbers in some cacti of western North America II. Brittonia 25: 2 9. Pinkava, D. J. and B. D. Parfitt. 1982. Chromosome numbers in some of the cacti of western North America IV. Bulletin of the Torrey Bo tanical Club 109: 121 128. Pinkava, D. J., B. D. Parfitt, M. A. Baker, and R. D. Worthington. 1992. Chromosome numbers in some cacti of western North America VI, with nomenclatural changes. Madroo 32: 98 113. Pinkava, D. J., J. P. Rebman, and M. A. Baker. 1998. Chromosome numbers in some cacti of western North America VII. Haseltonia 6: 32 40. Pinkava, D. J. 2002. On the evolution of the North American Opuntioideae. Pp 59 98 in Studies in the Opuntioideae (Cactaceae), eds. Hunt and Taylor. David Hunt, The Manse, Chapel Lane, Milborne Port Sherborne, England. Pinkava, D.J. 2003. Opuntia Pp 123 148 in Flora of North American North of Mexico, eds. Flora of North America Editorial Committee 1993+ 12+ vols., New York and Oxford. Posada, D. and K. A. Crandall 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817 818. Powell, A. M. and J. F. Weedin. 2001. Chromosome numbers in Chihuahuan Desert Cactaceae. III. Trans Pecos Texas. American Journal of Botany 88: 481 485. Powell, A. M. an d J. F. Weedin. 2004. Cacti of the Trans Pecos and Adjacent Areas. Texas Tech University Press, Lubbock, TX. Powell, A. M., J. F. Weedin, S. A. Powell. 2008. Cacti of Texas: a field guide. Texas Tech Univer sity Press, Lubbock, TX.
249 Puente, R. 2006. Humming birds and pricky pears: flower adaptations in the genus Nopalea The Sonoran Quarterly 60: 4 6. Rafinesque, C. S. 1820. Annals of nature, Annual synopsis of new genera and species of animals,plants, etc., discovered in North America. Iowa City. Rafinesque, Rafinesque, C.S. 1832. A tlantic journal, and friend of knowledge in eight numbers: containing about 160 original articles and tracts on natural and historical sciences, the description of about 150 new plants, and 100 new animals or fossils; many vocabularies of. Vol.1, Philadelp hia. Rambaut, A. 2007. Se Al v. 2.0. Website: http://tree.bio.ed.ac.uk/software/seal/. Ramsey, J. and D. J. Schemske. 1998. Pathways, mechanisms, and rates of polyploid formation in flowering plants. Annual Review of Ecology and Systematics 29: 467 501. Raven, P. H. 1963. Amphitropical relationships in the floras of North and South America. The Quarterly Review of Biology 38: 151 177. Raven, P. H. 1972. Plant species disjunctions: a summary. Annals of the Missouri Botanical Garden 59: 234 246. Rebman, J.P., and D.J. Pinkava. 2001. Opuntia of North America: an overview. The Florida Entomologist 4: 474 483. Reinhardt, C. F. and L. Rossouw. 2000. Ecological adaptation of an alien invader plant ( Opuntia stricta ) determines management strategies in the Kruger National Park. Zeitschrift fur Pflanzenkrankheiten und Pflanzenschutz sonderheft 17: 77 84. Remington, C. 1968. Suture zones of hybrid interaction between recently joined biotas. Evolutionary Biology 2: 321 428. Remski, M. F. 1954. Cytological investigations in Mammillaria and some associated genera. Botanical Gazette 116: 163 171. Reyes Agero, J. A., Rivera, J. R. A., and J. L. Flores. 2005. Variacin morfolgica de Opuntia (Cactaceae) en relacin con su domesticacin en la altiplanicie meridional de Mxico. INCI 30: 476 484. Reyes Agero, J.A., F. Carlin Castelan, J.R. Aguirre R., and H. M. Hernandez. 2007. Preparation of Opuntia herbarium specimens. Haseltonia 13: 76 82. Reyes Agero, J.A., J.R. Aguirr e R., and A. Valiente Banuet. 2006. Reproductive biology of Opuntia : a review. Journal of Arid Environments 64: 549 585.
250 Ribbens, E. 2008. Opuntia fragilis : Taxonomy, distribution, and ecology. Haseltonia 14: 94 110. Rieseberg, L. H., J. Whitton, and C. R. Linder. 1996. Molecular marker incongruence and phylogenetic trees. Acta Botanica Neerlandica 45: 243 262. Rook, F., Corke, F., Baier, M., Holman, R., May, A.G., and M.W. Bevan. 2006. Impaired sucrose induction1 encodes a conserved plant specific protein that couples carbohydrate availability to gene expression and plant growth. The Plant Journal 46: 1045 1058. Sahley, C. T. 1996. Bat and hummingbird pollination of an autotetraploid columnar cactus, Weberbaureocereus weberbaueri (Cactaceae). American Jou rnal of Botany 83: 1329 1336. Sanderson, M. J. 2002. Estimating absolute rates of molecular evolution and divergence times: a penalized likelihood approach. Molecular Biology and Evolution 19: 101 109. Sanderson, M. J. 2003. r8s: inferring absolute rates of molecular evolution and divergence times in the absence of a molecular clock. Bioinformatics 19: 301 302. Sanmartn, I., P. V D Mark, and F. Ronquist. 2008. Inferring dispersal: a Bayesian approach to phylogeny based island biogeography, with special reference to the Canary Islands. Journal of Biogeography 35: 428 449. Sarmiento, G. 1975. The dry plant formation of South America and their floristic connections. Journal of Biogeography 2: 233 251. Seelanan, T., A. Schnabel, and J. F. Wendel. 1997. Cong ruence and consensus in the cotton tribe (Malvaceae). Systematic Botany 22: 259 290. Segura, S., L. Scheinvar, G. Olalde, O. Leblanc, S. Filardo, A. Muratalla, C. Gallegos, and C. Flores. 2007. Genome sizes and ploidy levels in Mexican cactus pear species Opuntia (Tourn.) Mill. series Streptacanthae Britton et Rose, Leucotrichae DC., Heliabravoanae Scheinvar and Robustae Britton et Rose. Genetic Resources of Crop Evolution 54: 1033 1041. Shaw, J., E. B. Lickey, E. E. Schilling, and R. L. Small. 2007. Compar ison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in angiosperms: the tortoise and the hare III. American Journal of Botany 94: 275 288. Simonsen, T. J. 2008. Phylogeny of the cactus feeding phycitines and the ir relatives (Lepidoptera, Pyralidae) based on adult morphology: evaluation of adult character systems in phycitine systematics and evidence for a single origin of Cactaceae feeding larvae. Insect Systematics and Evolution 39: 303 325. Simonsen, T. J., R. L. Brown, and F. A. H. Sperling. 2008. Tracing an invasion: Phylogeography of Cactoblastis cactorum (Lepitoptera: Pyralidae) in the United States based on mitochondrial DNA. Annals of the American Entomological Society 101: 899 905.
251 Simpson, B. B., J. Tate and A. Weeks. 2005. The biogeography of Hoffmannseggia (Leguminosae, Caesalpinioideae, Caesalpinieae): a tale of many travels. Journal of Biogeography 32: 15 27. Small, J. K. 1903. Flora of the southeastern United States. Privately Printed, New York, NY Small, J. K. 1913. Flora of the southeastern United States. Privately Printed, New York, NY. Small, J. K. 1919. The prickly pears of Florida. Journal of the New York Botanical Garden 20: 21 39. Small, J. K. 1933. Manual of the Southeastern Flora. J.K. Sm all, published by the author, New York. Annals of the Missouri Botanical Garden 59: 218 223. Soltis, D. E. 1980. Kary otipic relationships among species of Boykinia, Heuchera, Mitella, Sullivantia, Tiarella, and Tolmiea (Saxifragaceae). Systematic Botany 5: 17 29. Soltis, D. E. 1984, Autopolyploidy in Tolmiea menziesii (Saxifragaceae). American Journal of Botany 71: 1171 1174. Soltis, D.E. and P.S. Soltis. 1989. Genetic consequences of autopolyploidy in Tolmiea menziesii (Saxifragaceae). Evolution 43: 586 594. Soltis, P. S. and D. E. Soltis. 1991. Multiple origins of allopolyploidy in Tragopogon mirus (Compositae): rDNA evidence. Systematic Botany 16: 407 413 Soltis, D.E. and P.S. Soltis. 1999. Polyploidy: recurrent formation and genome evolution. Trends in Ecology and Evolution 14: 348 352. Soltis, D.E. and P.S. Soltis. 2009. The role of hybrid ization in plant speciation. Annual Review of Plant Biology 60: 561 588. Soltis, D. E., E. V. Mavrodiev, J. J. Doyle, J. Rauscher, and P. S. Soltis. 2008. ITS and ETS sequence data and phylogeny reconstruction in allopolyploids and hybrids. Systematic Bot any 33: 7 20. Soltis, D. E., A. B. Morris, J. B. MaClachlan, P. S. Manos, and P. S. Soltis. 2006. Comparative phylogeography of unglaciated eastern North America. Molecular Ecology 15: 4261 4293. Soltis, D. E., P. S. Soltis, D. W. Schemske, J. F. Hancock, J. N. Thompson, B. C. Husband, and W. S. Judd. 2007. Autopolyploidy in angiosperms: have we grossly underestimated the number of species? Taxon 56: 13 30.
252 Spencer, J. L.1955. A cytological study of the Cactaceae of Puerto Rico. Botanical Gazette 117: 33 3 7. Stace, C. 2000. Cytology and cytogenetics as a fundamental taxonomic resource for the 20th and 21st centuries. Taxon 49: 451 477. Stamatakis, A. 2006. RAxML VI HPC: Maximum Likelihood based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22:2688 2690. Stebbins, G. L. 1940. The significance of polyploidy in plant evolution. The American Naturalist 74: 54 66. Stebbins, G. L. 1942. Polyploid complexes in relation to ecology and the history of floras. The American Naturalist 76 : 36 45. Stebbins, G. L. 1950. Variation and evolution in plants. Columbia University Press, New York. Stebbins, G. L. 1971. Chromosomal evolution in higher plants. London, Addison Wesley. Stilling, P. 2000. A worm that turned. Natural History 109: 40 43. Stilling, P. and D. Moon. 2001. Protecting rare Florida cacti from attack by the exotic cactus moth, Cactoblastis cactorum (Lepidoptera: Pyralidae). Florida Entomologist 84: 506 509. Stockwell, P. 1935. Chromosome numbers of some of the Cactaceae. Bota nical Gazette 96: 565 570. Strittmater, L. I., J. Hickey, and V. Negrn Ortiz. 2008. Heterochrony and its role in sex determination of cryptically dioecious Consolea (Cactaceae) staminate flowers. Botanical Journal of the Linnean Society 156: 305 326. Stu ppy, W. 2002. Seed characters and the generic classification of the Opuntioideae (Cactaceae). Pp 25 58 in Studies in the Opuntioideae (Cactaceae), eds. Hunt and N.P. Taylor. David Hunt, The Manse, Chapel Lane, Milborne Port Sherborne, England. Swanson, C P. 1957. Cytology and cytogenetics. Prentice Hall, Inc., New Jersey. Swenson, N. G. and D. J. Howard. 2005. Clustering of contact zones, hybrid zones, and phylogeographic breaks in North America. The American Naturalist 166: 581 591. Swofford, D. L. 2002 PAUP*: phylogenetic analysis using parsimony (*and other methods), version 4. Sinauer Associates, Sunderland, Massachusetts. Taberlet, P.G., P. Ludovic, P. Guy, and J. Bouvet. 1991. Universal primers for amplification of three non coding regions of chl oroplast DNA. Plant Molecular Biology 17:1105 1109. Tate, J. A., D. E. Soltis, and P. S. Soltis. 2005. Polyploidy in plants. In: Gregory TR (Ed) The Evolution of the Genome. Elsevier Academic Press, San Diego, 371 426.
253 Tate, J. A., V. Symonds, A. N. Doust, R. J. A. Buggs, E. Mavrodiev, L. C. Majure, P. S. Soltis, and D. E. Soltis. 2009. Synthetic polyploids of Tragopogon miscellus and T. mirus American Journal of Botany 96: 979 988. Taylor, N. P., W. Stuppy, and W. Barthlott. 2002. Realignment and revision of the Opuntioideae of eastern Brazil. Pp 99 132 in Studies in the Opuntioideae (Cactaceae), eds. D.R. Hunt and N.P. Taylor. David Hunt, The Manse, Chapel Lane, Milborne Port Sherborne, England. Thiers, B 2011. Index Herbariorum: A global directory of public herbaria and associated staff. 18 Oct 2011. Thorne, R. F. 1973. Floristic relationships between tropical Africa a nd tropical America. Pp 27 47 in Tropical Forest Ecosystems in Africa and South America: a comparative review, eds. B. J. Meggars, E. S. Ayensu, and W. D. Duckworth. Smithsonian Institute Press, Washington, D.C. Trejo Gonzlez, A., G. Gabriel Ortiz, A. M. Puebla Prez, M. D. Huzar Contreras, M. R. Mungua Mazariegos, S. Meja Arregun, and E. Calva. 1996. Extract from prickly pear cactus ( Opuntia fuliginosa ) controls experimentally induced diabetes in rats. Journal of Ethnopharmacology 55: 27 33. Trujill o, B. and M. Ponce. 1990. Nueva Opuntia (Cactaceae) de Venezuela y algunos aspectos ecolgicos de la misma. Ernstia 58 60: 1 7. USDA, NRCS. 2012. The PLANTS Database ( http://plants.usda.gov 6 June 2012). National Plant Data Team, Greensboro, NC 27401 4901 USA. Valente, L. M., V. Savolainen, and P. Vargas. 2010. Unparalleled rates of species diversification in Europe. Proceedings of the Royal Society B doi:10.1098. Vil, M., J. A. Burriel, J. Pino, J. Chamizo, E. L lach, M. Porterias, and M. Vives. 2003. Association between Opuntia species invasion and changes in land cover in the Mediterranean region. Global Change Biology 9: 1234 1239. Wallace, R. S. and S. L. Dickie. 2002. Systematic implications of chloroplast DNA sequences variation in the Opuntioideae. Pp 9 24 in Studies in the Opuntioideae (Cactaceae), eds. Hunt and Taylor. David Hunt, The Manse, Chapel Lane, Milborne Port Sherborne, England. Wallace, R. S. and A. C. Gibson. 2002. Evolution and systematics. In Park S. Nobel [ed], Cacti, biology and uses, 1 21, University of California Press, Berkeley and Los Angeles, CA. Waltari, E., R. J. Hijmans, A. T. Peterson, A S Ny ri S. L. Perkins, and R. P. Guralnick. 2007. Locating Pleistocene refugia: comparing phylogeographic and ecological niche model predictions. PLOS ONE 2: e563. doi: 10.1371/journal.pone.0000563
254 Webb, S.D. 1990. Historical biogeography. Pp 70 100 in Ecosystems of Florida, eds. R. L. Myers and J. J. Ewel. University Press of Florida, Gainesville, FL. Weedin, J. F. and A. M. Powell. 1978. Chromosome numbers in Chihuahuan Desert Cactaceae. Trans Pecos, Texas. American Journal of Botany 65: 531 537. Weedin, J. F., A. M. Powell, and D O. Kolle DO. 1989. Chromosome numbers in Chihuahuan Desert Cactaceae. II. Trans Pecos, Texas. The Southwestern Naturalist 34: 160 164. Wendel, J. and J.J. Doyle. 1998. Phylogenetic incongruence: window into genome history and molecular evolution. Pp 265 296 in Molecular Systematics of Plants II: DNA sequencing, eds. D. E. Soltis, P. S. Soltis, and J.J. Doyle. Kluwer Academic Publishers, Norwell, MA. Weniger, D. 1967. Opuntia drummondii a new record for Texas. Cactus and Succulent Journal 39:112 114. W eniger, D. 1970. Cacti of the southwest: Texas, New Mexico, Oklahoma, Arkansas, and Louisiana. University of Texas Press, Austin. Wherry, E.T. 1926. A new circumneutral soil prickly pear from the middle Atlantic states. Journal of the Washington Academy of Sciences 16: 11 14. White, T. J., T. Bruns, S. Lee, and J. Taylor. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. Pp 315 322 in PCR protocols: a guide to methods and applications, eds. M. A. Innis, D. H. Gelfan d, J. J. Sninsky, and T. J. White. Academic Press, San Diego, CA. Whittemore, A. T. and R. T. Olsen. 2011. Ulmus americana (Ulmaceae) is a polyploid complex. American Journal of Botany 98: 754 760. Whitton, J., C. J. Sears, E. J. Baack, and S. P. Otto. 20 08. The dynamic nature of apomixis in the angiosperms. International Journal of Plant Sciences 169: 169 182. Wunderlin, R.P. and B.F. Hansen 2003. Guide to the vascular plants of Florida. Second edition, University Press of Florida, Gainesville, FL. Wunde rlin, R.P. and B.F. Hansen 2011. Guide to the vascular plants of Florida. Third edition, University Press of Florida, Gainesville, FL. Yoder, A. D., J. A. Irwin, and B. A. Payseur. 2001. Failure of the ILD to determine data combinability for slow loris p hylogeny. Systematic Biology 50:408 424. Yu Y., A. J. Harris, and X. J. He. 2011. RASP (Reconstruct Ancestral State in Phylogenies) 1.1. Available at http://mnh.scu.edu.cn/soft/blog/RASP Zimmerman, H. G., V. C. Moran, and J. H. Hoffmann. 2000. The renowned cactus moth, Cactoblastis cactorum : its natural history and threat to native Opuntia floras in Mexico and the United States of America. Diversity and Distributions 6: 259 269.
255 BIOGRAPHICAL SKET CH Lucas C. Majure was born in Jackson, MS and lived in Pisgah, MS until the age of four, when he and his family moved to Decatur, MS. H e attended Newton County School, where he graduated in 1999. He then attended East Central Community College, where he r eceived his a ssociate s degree in Spring 2001 He completed undergraduate studies at Mississippi State University (MSU) in Starkville in the fall of 2003 with a major in biology and minor in S panish. After graduation he worked at the Mississippi Museum of Natural Science for one year as an herbarium assistant and also performed work with the Natural Heritage Program locating and monitoring rare and endangered plant populations in the state of M S In spring 2004, he started m Dr. Gary N. Ervin, where he focused on the ecology and morphological variation of Opuntia in the mid south United States. In the fall of 2007, he began doctoral work at the University of Florida under the direction of Drs. Doug and Pam Soltis and Dr. Walter Ju dd, where he studied the evolution and systematics of the Opuntia humifusa complex, as well as broad scale evolution of tribe Opuntieae. Lucas graduated in the summer of 2012 with a degree in b otany and began post doctoral research in August 2012 with Walt er Judd studying the species limits of a clade in the genus Miconia (Miconeae: Melastomataceae).