ben = erie JOURNAL OF THE BOTANICAL RESEARCH INSTITUTE OF TEXAS J. Bot. Res. Inst. Texas ISSN 1934-5259 VOLUME 3 NUMBER 1 15 JULY 2009 COPYRIGHT 2009 Botanical Research Institute of Texas (BRIT) 500 East 4th Street Fort Worth, Texas 76102-4025, USA © Hugh H. Itis EDITOR: Barney Lipscomb ASSISTANT EDITOR: Brooke Byerley Botanical Research Institute of Texas 500 East 4th Street Fort Worth, Texas 76102-4025, USA 817-332-7432; 817-332-4112 fax Electronic mail: barney@brit.org; jbritebrit.org HISTORY AND DEDICATION 1962— Lloyd H. Shinners (left), à member of the Southern Methodist University (SMU) faculty and a prolific researcher and writer, published the first issues of Sida, Contributions to Botany (now J. Bot. Res. Inst. Texas) CONTRIBUTING SPANISH EDITOR 1971—William F. Mahler (right), professor of botany at SMU and director emeritus of BRIT, inherited editorship and copyright. 1993—BRIT becomes publisher/copyright holder. 2007—First issue of J. Bot. Res. Inst. Texas. MISSION AND GOALS The BRIT Press seeks innovation and excellence in preparation, manufacture, and distribution of botanical research and scientific discoveries for the twenty-first century. The BRIT Press—bringing out the best in botanical Science for plant conservation and education. DIRECTION AND COVERAGE The BRIT Press considers original research papers concerned with classical and modern systematic botany, sensu lato, for publication in J. Bot. Res. Inst. Texas. All submissions are peer-reviewed. Guidelines for submissions are available from the BRIT Press website, http://www.britpress.org. BIBLIOGRAPHICAL Citation abbreviation for the Journal of the Botanical Research Institute of Texas is J. Dot. Res. Inst. Texas following the principles of B.P.H. (informally JBRIT). International Standard Serial No. (ISSN 1934-5259) FREQUENCY OF PUBLICATION J. bot. Res. Inst. Texas is published semiannually (summer/fall) as one volume by the Botanical Research Institute of Texas. Félix Llamas Dpto. de Botánica, Facultad de Biologia Universidad de Léon E-2471 Léon, SPAIN SUBSCRIPTION PRICES (2009) 543. Personal (Individual/Family) $95. USA (Institutional) $105. Outside USA (Institutional) SUBSCRIPTIONS/BACK VOLUMES J. Bot. Res. Inst. Texas and Sida, Contr. Bot. Publications Assistant Botanical Research Institute of Texas 500 East 4th Street Fort Worth, Texas 76102-4025, USA Electronic mail: orders@brit.org COMPOSITION rhorngraphics, Plano, Texas; rlrchorn@verizon.net PRINTING Prepress production and printing in the United States of America by Millet the Printer, Dallas, Texas www.millettheprinter.com COVER ILLUSTRATION Electronically tinted botanical illustration of Liatris aestivalis originally used on BRIT's anniversary poster 2001. Summer gayfeather flowers mid Jul-Aug(-Sep) and is endemic to Oklahoma and Texas. Sida 19:768. 2001. Botanical illustration by Linny Heagy €2001. BRIT PRESS TABLE OF CONTENTS COMMENTARY Learning about other species: an updated component of a liberal arts education Guy L. NESOM AND ÁLAN S. WEAKLEY SYSTEMATICS Orobanche riparia (Orobanchaceae), a new species from the American Midwest L. TURNER COLLINS, ALISON E.L. COLWELL, AND GEORGE YATSKIEVYCH Bristlecone pine dwarf mistletoe: Arceuthobium microcarpum subsp. aristatae (Viscaceae), a new subspecies of western spruce dwarf mistletoe from northern Arizona JARED M. SCOTT AND ROBERT L. MATHIASEN A new cliff-dwelling species of Zamia (Zamiaceae) from Belize MICHAEL CALONJE A new species of Zamia (Zamiaceae) from the Maya Mountains of Belize MICHAEL CALONJE, JAN MEERMAN, PATRICK GRIFFITH, AND GEOFFREY HOESE What is Zamia prasina (Zamiaceae: Cycadales)? MICHAEL CALONJE AND JAN MEERMAN Festuca aloha and F. molokaiensis (Poaceae: Loliinae), two new species from Hawai'i PILAR CATALÁN, ROBERT J. SORENG, AND PAUL M. PETERSON A new genus, Mexotis, for five Mexican species of Hedyotideae (Rubiaceae) EDWARD E. TERRELL AND HAROLD ROBINSON Capsicophysalis: a new genus of Solanaceae (Physaleae) from Mexico and Central America JOHN E. AvERETT AND MAHINDA MARTÍNEZ Synopsis of Gonolobus s.s. (Apocynaceae: Asclepiadoideae) in Trinidad and Tobago ALEXANDER KRINGS An overview of the Diospyros campechiana complex (Ebenaceae) and description of three new species MITCHELL C. PROVANCE AND ÁNDREW C. SANDERS Novedades y notas misceláneas en las Bromeliaceae de Mesoamérica J. FRANCISCO MORALES Una nueva especie y notas misceláneas en el género Oreopanax (Araliaceae) en Centro América J. FRANCISCO MORALES AND ALVARO IDARRAGA Five new species of Myrtaceae from Ecuador MARIA Lúcia KAWASAKI AND BRUCE K. HOLst Folia taxonomica 12. Paradrymonia (Gesneriaceae: Episcieae) from the Guiana Shield: P. magu- irei, a new species from Amazonas, and distribution and floral morphology of P. maculata CHRISTIAN FEUILLET New species of Baccharis (Asteraceae: Astereae) from Rio de Janeiro state, southeastern Brazil G. HEIDEN, J.F.A. BAUMGRATZ, AND R.L. ESTEVES Diplycosia indica (Ericaceae): a new species and a new generic record for India M.R. DEBTA AND H.J. CHOWDHERY Amorpha confusa, a new name for an old Amorpha (Fabaceae: Amorpheae) SHANNON C.K. STRAUB, BRUCE A. SORRIE, AND ALAN S. WEAKLEY A new combination in North American Tephrosia (Fabaceae) Guy L. NESOM AND JAMES L. ZARUCCHI 23 31 43 51 59 71 77 85 113 117 123 133 139 147 151 157 A recombination for varieties of Anticlea elegans (Melanthiaceae) WENDy B. ZOMLEFER AND WALTER S. JUDD Taxonomic overview of Eurybia sect. Herrickia (Asteraceae: Astereae) Guy L. Nesom The phylogeny of Selenia (Brassicaceae) inferred from chloroplast and nuclear sequence data JAMES B. BECK : Preliminary insights into the phylogeny and speciation of Scalesia (Asteraceae), Galápagos Islands JEREMY D. BLASCHKE AND RoGER W. SANDERS Introduced lesser celandine (Ranunculus ficaria, Ranunculaceae) and its putative subspecies in the United States: a morphometric analysis ANGELA R. Post, ALEXANDER KRINGS, WADE A. WALL, AND JOSEPH C. NEAL The types of Astragalus section Diphysi (Fabaceae), a complex endemic to Western North America, Part I: lectotypifications, epitypifications, and new combinations of several taxa J. ANDREW ÁLEXANDER The typification of Crotalaria rotundifolia and Crotalaria maritima (Fabaceae) DaNiEL B. WARD Lectotypification of Cardamine flexuosa (Brassicaceae) ANGELA R. Post, ALEXANDER KRINGS, QUI-YUN (JENNY) XIANG, BRYON R. SOSINSKI, AND JOSEPH C. NEAL An investigation of morphological evidence supports the resurrection of Pyrrocoma scaberula (Asteraceae: Astereae) Curtis R. BJORK AND MARK DARRACH Miscellaneous typifications, one new series, and one new varietal combination in Crataegus (Rosaceae) J.B. Pipers What is Ranunculus gelidus (Ranunculaceae)? ALAN T. WHITTEMORE Infrageneric taxonomy of Astrophytum (Cactaceae), with remarks on the status of Digitostigma RICHARD R. MONTANUCCI Mycological literature on Texas fleshy Basidiomycota, two new combinations, and new fungal records for Texas Davip P. Lewis AND CLARK L. OVREBO Twenty-eight new three-letter family acronyms for vascular plants (with comprehensive listings on-line) NEIL SNOW CHROMOSOME NUMBERS Chromosome numbers of miscellaneous angiosperm taxa Marc BAKER, Jon REBMAN, BRUCE PARFITT, DONALD PINKAVA, CHARLOTTE CHRISTY, ANDREW SALYWON, RAUL PUENTE-MARTINEZ, ALLAN ZIMMERMAN, AND J. Huso COTA ANATOMY AND MORPHOLOGY Anatomia de la epidermis foliar en las especies Mexicanas del género Muhlenbergia (Poaceae) OCTAVIO ROSALES CARRILLO AND YOLANDA HERRERA ARRIFTA FLORISTICS, ECOLOGY, AND CONSERVATION Primer reporte de Cystopteris reevesiana (Aspleniaceae) para México José A. VILLARREAL-Q. AND CELESTINO FLORES-L. 159 161 169 177 193 211 219 227 231 239 245 251 257 273 279 285 307 Floristic composition, abundance, and distribution pattern of woody plants in a tropical savanna in northern Ghana ALEX ASASE, PATRICK K. EKPE, AND JOHN Y. AMPONSAH New state records of vascular plants for the Southern Rocky Mountain region PAMELA F. SMITH, DONALD L. HAZLETT, AND NEIL SNOW Pyracantha koidzumii (Rosaceae) new to the Arkansas flora BRETT E. SERVISS Medicago rigidula and M. truncatula (Fabaceae): new to the California flora RICHARD E. RIEFNER, JR. AND ERNEST SMALL New and noteworthy plants from Florida Jonn M. KUNZER, RICHARD P. WUNDERLIN, LORAN C. ANDERSON, AND JAMES R. BURKHALTER Vascular flora of Churchill Ranch, Sarasota County, Florida ALAN R. FRANCK AND RICHARD P. WUNDERLIN Vascular flora of a saline prairie in Winn Parish, Louisiana BARBARA R. MACROBERTS, MICHAEL H. MACROBERTS, CHRISTOPHER S. REID, AND PATRICIA L. FAULKNER Vascular flora of Morse Clay prairies in northwestern Louisiana BARBARA R. MACROBERTS, MICHAEL H. MACROBERTS, CHRISTOPHER S. REID, AND PATRICIA L. FAULKNER Floristics of upland shortleaf pine/oak-hickory forest in northwestern Louisiana BARBARA R. MACROBERTS AND MICHAEL H. MACROBERTS A revision to the vascular flora of Calcasieu Parish, Louisiana Ray NEYLAND Additions and emendations to the vascular flora of Caddo Parish, Louisiana BARBARA R. MacRosrnrs, MICHAEL H. MACROBERTS, CHRISTOPHER S. REID, AND PATRICIA L. FAULKNER Mentha suaveolens and M. xrotundifolia in North Carolina: a clarification of distribution and taxonomic identity MICHAEL W. DENSLOW AND DERICK B. POINDEXTER Castilleja angustifolia var. dubia (Scrophulariaceae), a new record for South Dakota GRACE KOSTEL AND LYNN HETLET Glyceria declinata (Poaceae) new to the flora of Texas CHARLES M. ALLEN, PATRICIA LEWIS, AND Davip P. Lewis Vegetation areas of Texas: concept and commentary Ray C. TELFAIR II Vascular plant species richness of the Big Thicket, East Texas MicHAEL H. MACROBERTS AND BARBARA R. MACROBERTS A classification of geographic elements and analysis of the flora of Big Bend region of Texas ANNA SAGHATELYAN Annotated checklist of the vascular flora of the Menard Creek Corridor Unit of the Big Thicket National Preserve, Polk, Liberty, and Hardin counties, Texas Larry E. Brown, BARBARA R. MACROBERTS, MICHAEL H. MACROBERTS, AND WARREN W. PRUESS Cyperus pilosus (Cyperaceae) new to the flora of Texas RICHARD CARTER, CHARLES M. ALLEN, PATRICIA LEWIS, AND Davip P. Lewis Cryptostegia grandiflora (Apocynaceae: Asclepiadoideae), a new non-native weed for Texas THOMAS F. PATTERSON AND Guy L. NESOM Tournefortia hirsutissima (Boraginaceae) new to the flora of Texas ALFRED RICHARDSON AND W. KEN KING 309 317 319 323 331 339 349 355 367 375 379 383 391 393 395 443 457 461 Book Reviews and Notices 12, 22, 30, 42, 50, 76, 84, 122, 146, 156, 168, 192, 210, 226, 244, 272, 278, 284,308, 322,330, 0239197804390, 400, 406, 442, 456, +00: 404 Announcement 468 INDEX to new names and new combinations in J. Bot. Res. Inst. Texas 3(1), 2009 Amorpha confusa (Wilbur) S.C.K. Straub, Sorrie @ Weakley, comb. et stat. nov.—154 Anticlea elegans var. glaucus (Nutt.) Zomlefer € Judd, comb. nov.—159 Arceuthobium microcarpum subsp. aristatae J.M. Scott & Mathiasen, subsp. nov.—14 Astragalus lentiginosus var. bryantii (Barneby) J.A. Alexander, comb. nov.—215 Astragalus lentiginosus var. iodanthus (S. Watson) J.A. Alexander, comb. nov.—215 Astragalus lentiginosus var. pseudiodanthus (Barneby) J.A. Alexander, comb. nov.—216 Baccharis altimontana Heiden, Baumgratz & Esteves, sp. nov.—140 Baccharis friburgensis Heiden, Baumgratz & Esteves, sp. nov.—143 Calyptranthes compactiflora M.L. Kawasaki & B.K. Holst, sp. nov.—123 Calyptranthes fusca M.L. Kawasaki & B.K. Holst, sp. nov.—125 Calyptranthes sparsiflora M.L. Kawasaki & B.K. Holst, sp. nov.—125 Capsicophysalis (Bitter) Averett & M. Martínez, stat. nov.—72 Capsicophysalis potosina (B.L. Rob. & Greenm.) Averett & M. Martínez, comb. nov.—72 Crataegus series Altaicae J.B. Phipps, ser. nov.—243 Crataegus chrysocarpa var. subrotundifolia (Sarg.) J.B. Phipps, comb. et stat. nov.—240 Diospyros camposii M.C. Provance & A.C. Sanders, sp. nov.—93 Diospyros crotalaria M.C. Provance & A.C. Sanders, sp. nov.—95 Diospyros haberi M.C. Provance & A.C. Sanders, sp. nov.—98 Diplycosia indica M.R. Debta @ H.J. Chowdhery, sp. nov. —147 Eurybia aberrans (A. Nels.) G.L. Nesom, comb. nov.—166 Eurybia kingii (D.C. Eaton) G.L. Nesom, comb. nov.—166 Eurybia kingii var. barnebyana (Welsh & Goodrich) G.L. Nesom, comb. nov.—166 Festuca aloha Catalán, Soreng & PM. Peterson, sp. nov.—51 Festuca molokaiensis Soreng, PM. Peterson & Catalán, sp. nov.—54 Hygrocybe chamaeleon (Cibula) Lewis & Ovrebo, comb. nov.—264 Hygrocybe mississippiensis (Cibula) Lewis & Ovrebo, comb. nov.—264 Mexotis Terrell & H. Rob., gen. nov.—60 Mexotis galeottii (M. Martens) Terrell & H. Rob., comb. nov.—61 Mexotis kingii (Terrell) Terrell & H. Rob., comb. nov.— 64 Mexotis latifolia (M. Martens & Galeotti) Terrell & H. Rob., comb. nov.—61 Mexotis lorencei Terrell & H. Rob., sp. nov.—67 Mexotis terrellii (Lorence) Terrell & H. Rob., comb. nov.—64 Myrcia aequatoriensis M.L. Kawasaki € B.K. Holst, sp. nov. —128 Myrcia verticillata M.L. Kawasaki € B.K. Holst, sp. nov. —130 Oreopanax paramicolus J.F Morales & A. Idárraga, sp. nov.—118 Orobanche riparia L.T. Collins, sp. nov. —7 Paradrymonia maguirei Feuillet, sp. nov.—134 Paradrymonia subg. Pagothyra (Leeuwenb.) Feuillet, comb. et stat. nov. —134 Tephrosia xintermedia (Small) G.L. Nesom & Zarucchi, comb. nov.—157 Werauhia luctuosa J.F Morales, sp. nov. —114 Zamia decumbens Calonje, Meerman, M.P Griff. & Hoese, sp. nov —31 Zamia meermanii Calonje, sp. nov.—23 COMMENTARY LEARNING ABOUT OTHER SPECIES: AN UPDATED COMPONENT OF A LIBERAL ARTS EDUCATION Guy L. Nesom Alan S. Mid 2925 Hartwood Drive Herbarium/NC B Fort Worth, Texas 76109, U.S.A. Univ North Carolina, Coker Hall, E 2 guynesom@sbcglobal.net Chapel Hill, North Carolina 27599-3280, U.S.A. weakley@unc.edu Some aspects of a liberal arts education—critical thinking, respect for truth and clarity of expression—are unchanging in value. Knowledge of humans and human culture remains central. But if the goal of the cur- riculum is to prepare students for broad engagement in the world, an understanding of the Earth and all of life calls for equal time. Humanity is, for now at least, bound to Earth. Here, unbridled growth of the human population and profligate use of natural resources threaten our own existence and put at risk the well-being and continuance of all life and the environment. An appreciation of the planet and preparation toward committed stewardship should be among the primary goals of ae E arts education can not afford to be self-centered, in the sense of study of human cult l context and without attention to the thousands of species that are our evolutionary associates. We propose a course that guides students towards an appreciation of the Earth as home to many spe- cies. At the end of the essay, the course is integrated within an updated liberal arts curriculum. Learn to know other species. Perhaps the most effective path toward an appreciation of life on Earth is through first-hand knowledge of species other than our own. The primary goal of the proposed course is to learn the identities of commonly encountered species. The course need not be heavily laden with deep academic underpinning—the value lies in learning to recognize species by sight, as they are encountered in everyday life, and to identify them by name. A know-the-species course provides foundation for one of the most basic lessons in humility—humans are one species among hundreds of thousands. So many of us are hardly aware that others exist, except in an abstract way. Daily acquaintance with other species on a "first-name" basis greatly broadens one's view of the natural world and engenders a closeness, a sense of unity, with it. Even if names of species eventually are forgotten, at least the memory remains that each is unique, distinguished by features that most have previously been unaware of. One can quickly learn to identify a half-dozen species (hollies, oaks, maples, etc.) and to see that some of the species are more similar among themselves while others are more unique. Such an appreciation of variation and diversity, even in a simple system, is the quickest way possible toward real insight into the process of evolution— patterns of similarity among species result from common ancestry. Each species on Earth is the product of long evolutionary lineage, interrelated among the others, each unique and beauti- ful. “Lam strongly induced to beli hat, as in music, tl ho understands every note will, if he also possesses a proper taste, more thoroughly enjoy the whole, so ene que examines a part of a aoe view, n c puo iia the full d be a botanist, for in all I "Charles Darwin, 1836, Voyage i the E. Nature of the course. Plants are the most conspicuous and abundant expression of non-human life, even in urban settings, and we believe they are the best focus for the proposed course. It's joyful to walk along a J. Bot. Res. Inst. Texas 3(1): 1 — 2. 2009 sidewalk, through a park, or through the woods and recognize species of oaks, elms, and maples by name. Even in a cityscape, dozens of species are commonly planted and usually many more are scrabbling out their existence, unassisted, seeding themselves in unlikely places. Awareness of other species also could be gained from an introduction to birds, beetles, fish, or fungi, or practically any other kind of organism, but none of these is as easily accessible as plants. Do other courses in biology (e.g., physiology, genetics, molecular biology, cell biology) offer the same potential insights? Simply, no—most are based on study of a single species or of no species in particular. Some courses in ecology re- quire a basic knowledge of species diversity, but the focus is different. Studies of geology, meteorology, and astronomy may be doorways to recognition of the diversity of everyday extra-human experience and could be analogous to identification of plants, but diversity, or at least every-day accessibility, is much lower. Residents of the United States and other highly urbanized areas are so removed from other species that it is easy to be unaware that plants underpin our existence. Do undergraduates recognize that our diet is mostly of roots, stems, leaves, fruits, and seeds? Do typical students know that most of the energy expended by human individuals comes directly from starch in corn, rice, cassava, wheat, and potatoes? Beans, ba- nanas, and sorghum are the world's dietary staples. Chocolate, coffee, sugar—the plant foods that tempt us. Cotton, flax, hemp, mulberry—the plants of fashion. Even the meat in our diet is but one step away from plant nutrition. Plants are responsible for nearly all our oxygen, our clothes (directly or indirectly), and most of our non-food energy (fossil fuels —oil, natural gas, coal). These topics deserve a place in a liberal arts education. Since the suggested course is not an abstruse undertaking, basic knowledge of other species should be taught not only as part of a liberal arts collegiate education but in elementary through high school as well. Of course, this is one among many basic topics, but lack of knowledgeable teachers is a constraint and the responsibility usually comes to rest at college level. Suggested course outline. (A) identification of 100 species, concentrating on first-hand recognition of those most easily at hand and abundant; (B) a brief overview of plant vegetative features and reproduction (flower and fruit structure usually are critical in identification) and common families encountered among the species; (C) perspectives on evolution, the interrelatedness of life, and global patterns of diversity; and (D) a brief introduction to ecological and economic values of plants tered in everyday life (e.g., urban forests, invasive species, food, clothes, construction, fuel). Simply learning to know the species is the first priority. Goals of a liberal arts education. In the broad view, we believe that a liberal arts curriculum should be aimed (I) toward continued existence of our own species and all others and (2) toward maximizing the potential for freedom and happiness (including creativity, appreciation of beauty, productivity, service, and increasing understanding of the world). Overview of an updated liberal arts curriculum. We divide a suggested curriculum into four major, overlap- ping categories. 1. The Earth and all things non-human. Life and life processes (humans as animals, other species, ecology, evolution), geology, weather and climatology. Our proposed course would figure as a central focus in this category. 2. Intellectual outlook. Reasoning and critical thinking (including scientifi hod), respect for truth, clarity of expression, philosophy and ethics. 3. Humans and human culture. Beauty (visual art, music, perception), cultural diversity and history (history, language, philosophy, ethics, and religion). 4. The Universe and universal principles. Astronomy and cosmology, physics and chemistry, mathematics. OROBANCHE RIPARIA (OROBANCHACEAB), A NEW SPECIES FROM THE AMERICAN MIDWEST L. Turner Collins Alison E.L. Colwell Science and Technology Department Resources Management and Science vangel University Yosemite National Par 111 N Glenstone PO. Box 700, 5083 Foresta Road Springfield, Missouri 65802, U.S.A. El Portal, California 95318, U.S.A. collinst@evangel.edu alison_colwell@nps.gov George Yatskievych Missouri Botanical Garden PO. Box St. Louis, Missouri 63166, U.S.A. george.yatskievychemobot.org ABSTRACT Orobanche dedi mE oe in the n Plains, is tl idely distributed species i t. Nothaphyllon. It was first described by T in 181 ted at Ft. Mandan, North Dakota. During the foll and a half, most botanists had HQ field Ee "ih living plants of the g ] very few herbarium peci ith e to a ke newly collected eae! bl fusion al he ci ion of thi i lits g N hyllon ( M hiza) by B (1890, 1930) aud 1930) 1 led | i l species name too broadly to par whose mn was s discordant with Nuttall's original protologue, a the dinde ofa an Bn. broad led hi O. ludoviciana DO pn (Collins 1973) 1d his problem, including f l gnition of logical itia O. ludoviciana sensu stricto with different h inhabi land d tl I fined to I land iated with major river systems. Recent additional investigations have revealed fu further evid pporting gnition of tl listi ic entiti : oe logical distincti i 1 : p i f the upland f O. ludoviciana ] gnition the | land f distinct species, O. riparia, sp. nov. RESUMEN Orobanche ludoviciana, cuya extensión abarca mayormente la zona de Great Plains, es la especie de mayor distribución en la Sección Nothaphyllon. El o que n una ee fue pas Nuttall en 1818, que se K en una muestra NAME en Ft. Mandan, Dakota del N D te el g Į i p g y con ejemplares de a con las due comparar los e obtenidos igui hubo! fusión en torno a la circun- scripción de esta especie y su distribución geográfica. Los estudios ROM con la Sección Nothaphyllon (sec. Myzorrhiza) hechos por i ie pid y Munz Coa hacen que los botánicos apliquen de modo demasiado amplio los nombres de las plantas cuya uttall. El problema que permanece es el de un concepto demasiado amplio de E 8 la E Estudios doctoral ducid por el autor pip durante los primeros años de la década de los setenta resultaron en TRE d 1 I O. ludoviciana (Collins 1973), incluso un reconocimiento formal de dos tro de O. ludovici j distribución algo dif éspedes. Una de las plantas era s tierras altas y es s retenida aquí como O. ludoviciana y l finada a ti baj iadas principal i Jecerincion: Reci ] lido a 1a] asd dich t a. En r trabajo se comentan las diferencias morea entre estos dos ecotipos y se propone reconocer les adams de las tierras bajas ta O. riparia, sp. nov. COLL p Key Wonps: Orobanche, Orobanchaceae, parasitic plants, broomrape, Flora of North America Orobanche L. sect. Nothaphyllon (A. Gray) Heckard (sect. Myzorrhiza Phil.) comprises of an array of ca. 13 species that account for most of the taxonomic diversity within New World holoparasitic Orobanchaceae. It has been considered a monophyletic group by most botanists based on morphological and molecular data (Olmstead et al. 2001; Schneeweiss et al 2004; Bennett & Matthews 2006). Within sect. Nothaphyllon, two J. Bot. Res. Inst. Texas 3(1): 3 — 11. 2009 major clades exist whose relationships have not yet been fully resolved: the O. californica Cham. & Schltdl. complex (Heckard 1973), which has diversified mainly in the far-western United States; and the O. ludoviciana Nutt. complex (Collins 1973), which is widespread in the remainder of the United States and Canada, and southward into Mexico (also disjunct in southern South America). Orobanche ludoviciana Nutt., with a range primarily in the Great Plains, is the most widely distributed species in sect. Nothaphyllon. It was first acude. by Thomas Nuttall in 1818 from a specimen collected at Ft. Mandan, North Dakota. B ists during the following century had limited field experience with living plants of the genus and had very few herbarium specimens with which to compare newly collected materials, the species name was applied too broadly to include plants whose morphology is discordant with Nuttall's original protologue. The result has been considerable confusion about the circumscription of this species and its geographic distribution. All those who have treated Section Nothaphyllon (sect. Myzorrhiza) (Beck 1890, 1930; Munz 1930; Collins 1973; Heckard 1973), have at times recognized various infraspecific taxa, some of which are no longer accepted and therefore excluded from this discussion. Nevertheless, the lingering problem has been that of an overly broad species concept. Even Philip Munz (1930), who provided good insights into taxonomic variation within O. ludoviciana in his otherwise excellent taxonomic revision of sect. Myzorrhiza, overlooked significant morphological variation. Subsequent authors of floristic works who relied on his revision to determine specimens continued to some degree to overlook the taxonomic heterogeneity of plants treated by him as O. ludoviciana var. genuina Beck. Doctoral studies by the senior author led him to suggest a number of ludoviciana complex (Collins 1973), one of which was formal recognition of two ecological races within O. ludoviciana sensu stricto with differing host preferences: one inhabiting only uplands and the other confined to bottomlands associated with major river systems. Recent discoveries of new populations and examina- tion of additional existing herbarium specimens of the bottomland plants have added evidence in support of this proposal. The present paper therefore discusses the morphological distinctions between these two ecotypes, and we propose to retain the upland form as O. ludoviciana and to recognize the bottomland form as a distinct species, O. riparia. It should be noted that, based on Collins's dissertation, the epithet O. riparia was discussed far in advance of the present publication by Musselman (1982, on Orobanchaceae in Virginia) and by Kartesz (1994, in a North American plant checklist), but neither of these usages affects the validity of the present description. Bringing this new taxon to the botanical literature requires a brief review of the existing nomenclature and morphology of O. ludoviciana sensu stricto (O. ludoviciana var. genuina, sensu Beck). Several species of Orobanche in North America are commonly misidentified as O. ludoviciana, but are currently recognized as distinct species. They include O. multiflora Nutt., O. cooperi (A. Gray) A. Heller, O. multicaulis Brandegee, and O. valida Jepson. Of these, the two with the most morphological similarity and name association are O. ludoviciana and O. cooperi, and must therefore be contrasted with O. riparia. Collins (1973) found that Munz's 1930 description and illustration of the corolla lobes of O. ludoviciana did not match the morphology of Nuttall's type specimen. The corolla lobes of the type specimen are rounded on both the upper and lower lips, but Munz's illustration, based on a specimen from White County, Illinois, shows a corolla with triangular-pointed lobes. Because Munz used material that matched his illustration in assembling his description of O. ludoviciana, subsequent regional and state floras continued to overlook the mu gical eub pee Mp i to i euer ated 1 within the O. range has revealed that the character of founded e ibe is consistent throughout for UN PR (Fig. 1). In contrast, the corolla lobes of bottomland plants, hereafter referred to as O. riparia, are consistently triangular and pointed (Fig. 1). The populations of O. riparia share several additional characteristics (Table 1): D They are found almost exclusively in river bank and sandbar habitats in floodplains; 2) The native hosts are annual members of Ast tribe Heliantheae, subtribe Ambrosiinae, primarily Ambrosia trifida L. and Xanthium strumarium L., and occasionally Ambrosia artemisiifolia L. and Dicoria canescens A. Gray 3) Collins et al., Orobanche riparia, a new species y as karl Ae, 7 VL x2 D^ re Af ncn A à - "FE pl 4. E Fic. 1. Comp ludoviciana. 6 Journal of tt IR h Institute of Texas 3(1) Tape 1. C | id f Orot | ludoviciana, and cooperi Character Orobanche riparia Orobanche ludoviciana Orobanche cooperi Calyx len 7-11 mm 8-14 mm 7-12 mm Corolla length 15-22 mm 14-20 mm 15-32 mm ape and size Triangular-acute, 4-5mm, Rounded-obtuse, 4-8mm, Triangular-acute, 5-8 mm, of corolla erect or slightly reflexed erect or ascending apiculate apex, ascending lips/lobes or reflexed Corolla tube Strongly curved, tube Slightly curved, tube ascending, Strongly curved, tube Corolla color horizontal slightly flared distally Pallid, or upper lobe apex purple/lavender slightly flared distally Purple, lavender, rose, or pallid, tube exterior whitish ae nalig: (alee ay Dar | half with T eeii and throat Pubescence pubescent with glandularand Pubescent with "uad Pubescent with glandular eglandular trichomes or trichomes except in trichomes on tube, eglandular glabrate lobe surface on inner surface of lobes Flowering August-October April-August December-April Inflorescence Open raceme, flowers in Compact raceme, flowers in Open or compact raceme with arrangemen loose, regular spiral on dense, often irregular spiral on flowers in loose, regular spiral Primary hosts Habitat and floral axis Annuals; Ambrosia, Xanthium, icoria River banks, sandbars eastern floral axis Perennials; Grindelia, Artemisia, Heterotheca Prairie, desert, dunes; central and western U.S. on floral axis Paranniale Amhrocin + Sonoran Desert, ravines, dry stream beds, dunes; main range U.S. and prairie provinces southwestern U.S, The flowering period is from mid-August to early October; 4) Inflorescence color is typically pallid with only the interior tips of the corolla lobes tinted purple; 5) The corolla tube is strongly curved with the distal portion of the tube horizontal, slightly flared distally, and palatal folds that are either glabrous or densely pubescent. Compared with O. riparia, O. ludoviciana has slightly shorter corollas with a flared throat and rounded corolla lobes (Table 1). It occurs mainly through the Great Plains (and eastward along the Prairie Peninsula) in various habitats, including sandy prairies, sand dunes, dry washes, calcareous (caliche) soils, and lower slopes of the Rocky Mountains. Its hosts are almost exclusively perennial members of Asteraceae tribes As- tereae and Anthemideae, including Grindelia squarrosa (Pursh) Dunal, Heterotheca spp., Artemisia spp., and occasionally other genera. The principal flowering period is from late April in the southern plains through mid-August in the northern plains, although sporadic flowering has been observed rarely at other times in marginal habitats. The corolla tube is slightly curved at the constriction, ascending, and flared distally, with the palatal folds sparsely pubescent. Floral pigmentation varies from purple to pale rose and occasionally yellow, usually with dark purple upper lobes (Table 1). Additionally, populations of O. riparia are ecologically and somewhat geographically distinct, and the habitat distinction is maintained in the western populations, where the two ecological races are geographi- cally somewhat sympatric. Populations east of the Mississippi River (but including ell are allopatric with O. ludoviciana. West of the Mississippi River, sympatry of m eastern Nebraska and Kansas to eastern Colorado, southward into New Mexico and the Texas Panhandle. A small overlap in flowering times in early August could provide an opportunity for hybridization between the two species. Although no suspected cases dix uen have been addas based on morphology or field observa- tic st ded to answer this question. Molecular +} + ors tions of populations, data from g studies by Colwell are currently underway. Munz (1930) treated O. cooperi as a variety of O. ludoviciana, but subsequent authors mostly have ac- cepted it as a separate, but closely related species (Shreve & Wiggins 1964; Munz 1974). Orobanche cooperi is Collins et al., Orobanche riparia, a new species 7 distinguished by its dark purple color, larger, di lla tube, and erect or strongly reflexed, triangular lobes with apiculate apices. The corolla tube has a whie exterior and is slightly curved and flared distally, with palatal folds pubescent at the mouth and becoming glabrous in the throat. The habitat is primarily Sonoran Desert scrub, often associated with washes that remain dry except in monsoon season. Its hosts are shrubs in various tribes of Asteraceae, principally species of Ambrosia and Encelia (all Heliantheae, Am- brosiinae), but also reported on Gutierrezia (Astereae). The flowering period is mainly December through April (Table 1). Collins (1973) found that the cross-sectional stem anatomy of O. cooperi and O. riparia are very similar and that O. ludoviciana differs significantly from both of the other taxa. Several characters suggest that O. riparia is more closely related to O. cooperi than to O. ludoviciana. Orobanche riparia and O. cooperi share an affinity for riparian or water-eroded habitats and hosts in the sub- tribe Ambrosiinae. They differ in that O. cooperi occurs only in seasonally arid habitats, whereas O. riparia occurs in both mesic and semi-arid habitats, and that the former parasitizes shrubby hosts, whereas the host taxa of the latter are annuals. The pubescence of the corolla lobes in both species is eglandular on the inner surface with vestiture considerably longer than the short glandular trichomes typical of the exterior surface. Likewise, the palatal folds are invested with eglandular trichomes, differing between the species only in their distribution on the folds. This shared pattern of trichomes is more evident in the western populations of O. riparia than in the eastern populations. Similarly, the eastern populations have reduced indumentum as compared with the western populations. One can speculate that O. riparia and O. cooperi share a recent common ancestor. A possible scenario is that a progenitor’s range became bisected, with populations that became O. riparia adapting and dispersing into new habitats eastward across the American Midwest and populations now referable to O. cooperi attaining a distribution in the southwestern U.S. and adjacent Mexico. If this is the case, O. riparia adapted quickly to annual hosts and a late summer to early fall flowering period, in contrast with the shrubby hosts and late winter to early spring flowering period of O. cooperi. Of this character set, O. ludoviciana shares only the semi-arid habitat and perennial hosts with O. cooperi, although the host taxa of O. ludoviciana often are herbaceous. Orobanche riparia L.T. Collins, sp. nov. (Fig. 1). ter: UNITED STATES: Inpiana: Gibson Co.: common on bank of slough .5 mi N of Griffin, 16 Aug 1931, Deam 50941 (noLorvee: IND; isorvees: A, E GH, IND, MINN, WIS). Ab O. ludoviciana differt lobis corollae apice acutis, puberulis vel cs eure horizontali, tricto flexoq per ovarium, inflorescentia laxe racemosa. Ab O. cooperi differt corolla 15-22 mm longa, lo! I piculati flexis, inflorescentia pallide purpure Annual root parasite, lacking chlorophyll. Coralloid roots few, rudimentary, stout, fleshy, usually forming a single host attachment. Stem stout or slender, fleshy, 5—35 cm long, mostly subterranean, solitary or clus- tered, Aoi or more ee ae from near base with a few to rarely numerous branches, each beari larly enlarged or not, 5-20 mm in diameter; glabrous, creamy siue: yellow, or tan, drying Piraeus: dark bone. or black. Leaves reduced to scales, 7-10 mm long, cauline, spirally alternate, appressed, broadly ovate to ovate triangular, becoming narrower distally. Inflorescence an open spike-like raceme, glandular pubescent; bracts mostly ca. 1 cm long, the distal ones shortened, narrowly oblong-lanceolate, acute at the tip, pale lavender to dark purple. Flowers sessile or the proximal ones on short pedicels, evenly and spirally inserted on the axis, subtended by 2 small, linear-subulate bracteoles; calyces 7-11-13) mm long, purple, deeply 5-lobed, the lobes lance-linear to linear-subulate, about equal in length, densely glandular pubescent; corollas (13—)15-22 mm long, constricted above the ovary, strongly curved at the constriction, the tube nearly horizontal, exterior surface glandular puberulent or pubescent, persistent, the tube narrow or slightly flared distally, cream-white sometimes tinted with purple, the throat with yellow palatal folds, glabrous or pubescent on folds and in grooves with eglandular trichomes; upper lip ca. 5 mm long, erect, divided about half its length, the 2 lobes triangular-acute, lavender or purple, puberulent or densely pubescent with eglandular trichomes; lower lip 3-4 mm long, divided to its base into 3 linear lobes with an acute apex, ca. 1 mm wide, pubescent with eglandular trichomes, pale, often with 1-3 purple veins. Stamens 4, inserted on the corolla tube at its constriction, generally glabrous or with a few scattered hairs, anthers ca. 1 mm long, white, glabrous or sparsely pubescent, equalling or shorter than style. Ovary somewhat asymmetrically ovoid, 2-carpellate, nectary not evident; placentation parietal, with 4 lateral placentae; style filiform, equaling the corolla tube or shorter, stigma peltate, discoid- crateriform or rarely somewhat 2-parted Fruits 2-valved capsules, 0.7-1 cm long; seeds numerous, 0.3-0.5 mm long, ovoid to broadly ellipsoid, reticulate-pitted, light tan to dark brown. 2n = 48 (Collins 1973). Flowering mid-August to early October. The species ranges (Fig. 2) from the Potomac and Shenandoah River valleys of Virginia to southern Ohio and Indiana along the Ohio River to Central Illinois along the Wabash and Sangamon Rivers, to the Meramec River in eastern Missouri (near St. Louis), southward to a tributary of the Mississippi River in western Ten- nessee, and westward along the Platte River in Nebraska and Eastern Colorado. Some apparently isolated populations occur in the Kansas, Oklahoma, Texas, and New Mexico along the Canadian, Cimarron, and Rio Grande rivers and their tributaries. Only historical collections are currently known from the District of Columbia, Virginia, and West Virginia. The species is found entirely on mixed sand and silt deposits on stream banks of eastern rivers or on sandbars or sandy banks of western rivers. Hosts are Ambrosia trifida and Xanthium strumarium, and occasionally Ambrosia artemisiifolia and Dicoria canescens. Rarely reported alternate hosts are Nicotiana tabacum L. and N. glauca Graham. Representative specimens: UNITED STATES. COLORADO: Crowley Co.: cultiv: m ae near nee 22 Je 1948, ld: s.n. (CS). Jackson Co.: N sandhills E of Cow uA 1 Oct Mine Wi pd 13922 is ).I ] near Ster- == anonymous, 8 Sep 1965 (CS). P ,6mi E of Two Buttes Peak, 24 Aug 1964, Martin s.n. (COLO, CS). Weld Co.: sandhills 3 mi NE of Roggen, 1 Sep 197, pud s.n. He CS). DISTRICT OF COLUMBIA: E end of New Long Bridge, 2 Sep 1905, Ward s.n. (US). ILLINOIS: nim: ~ s of Embarras River, 2 mi E of Greenup, 18 Sep 1980, Collins 1618 (MO). Mason Co.: bank of Salt Creek and edg field, 5 mi N per cati - nis in os 1615 (MO). Menard Co.: bottoms, Athens, 1860, 1861, 1866, 1867, E. Hall s.n. (F, GH, IL, MO, POM, WIS). V mel along banks of Wabash River and Greathouse Creek, 18 Sep 1980, Collins 1619 (MO). INDIANA: Harrison Co.: W of ig at ER 25 Aug 1945, Deam 63566 (IND). Jefferson Co.: tob field 0.5 mi E of Brooksburg, 15 Sep 1935, Banta s.n. (IND). Knox Co.: along ditch 2 mi N x ud 9 Oct 1938, Deam 59255 Ends vieni Co.: E bank of Wabash River at New Harmony, 28 AE 1971, Collins 1555 (MO, WIS). k, 6.5 mi SW of Carlisle, 1 Sep 1939, Deam 59424 (IND). Vandenburgh Co: river slough, 17 Oct 1941 Zeiner s.n. (IND), Warrick Co.: banks of Ohio River at Newburg, 28 Aug 1971, Collins 1543 (UWM). KANSAS: Allen Co.: Iola, Sep 1925, anonymous (KSC). Greenwood Co.: potato patch near Eureka, 24 Sep 1935, Farmer s.n. (KS). Morton Co.: Cimarron River bed N of Wilburton, Aug 1929, Gates 16205 (F, KSC, MO). KENTUCKY: Pendleton Co.: Ohio River bank at Ivor, 27 Nov 1942, Chicoine 4745 (MO, US). MISSOURI: St. Louis Co.: a pes a bar on S bank of Meramec River, 2 mi N of Crescent, 1 Sep 1991, Yatskievych & Colwell 91-195 (MO). NEBRASKA: B Platte River, 2 mi S of Odessa, 8 Oct 1999, Collins 1620 (MO, s Co.: sandbar in Platte River, 2 mi NE of Louisville, 15 Sep D. Morrison 1342 (NEB). Dawson Co.: sandbar in Platte River, S of Lexington, 8 Oct 1999, Collins 1621 (MO). Kearney Co.: sandbars of Platte River, Minden, 15 Sep 1922, Hapeman s.n. (PH, POM/RSA, WTU). Lincoln Co.: sandbar in Platte River, Sutherland, 13 in 1936, anonymous (NEB). Merrick Co.: sandbar in Platte River, near Central City, 22 Aug 1938, I. Mueller s.n. (NEB). NEW MEXICO: Doña Ana Co.: sand, od e 25 s 1906, Wooten & Stanley s.n. (US). Sandoval Co.: sandy soil, Arroyo Chico, 22 Sep 1973, ee 1758 (ARIZ). San River, 30 m S of Farmington, pipeline ee 7 on BIA Road 2, 4 sep 2007, Collins & Heil 2032 (MO). Socorro Co.: sand along Rio Cani San Marcial, 1 Aug 1897, H k 82 ). Union Co.: idle field, 17.5 mi SW of Clayton, 1 Sep 1936, Cogdell 125 (GH). OHIO: 5 mi E of Chilo, 20 Oct 1949, Bardy s.n. (OS, US). Hamilton Co.: moni a ae Miami River, 25 pace wee Haller s.n. (US). Scioto Co.: Nile Bae a Ohio River, 29 Aug 1949, zn s.n. An f Canadian River, 10 mi NW of N , 8 Sep 1946, Goodman 4078 (MO). T dale C hofLower Forked Deer Creek, ca.2 m NE of in 16 Oct 1972, Piehl s.n. (TENN). pois Brewster Co.: »: Boquillas Given. Chisos Mts., sand near Rio Grande River bank, 5 Aug 1937, Warnock 838 (ARIZ, TEX, US). D dy soil, Rita Blanca Creek N of Dalhart, Aug 1949, York & Rogers 233 (OSU, TTC). El Paso Co.: vicinity of El Paso, 1911, Stearns 40 (MO, US). Hartley Co.: deep sand on creek bank, Punta del Agua Creek 30 mi SW of Dalhart, 26 Sep 1965, Collins 821 (OSU, TTC). Wheeler Co.: 5 mi 5 & 7 mi E of Miami, Jul 1967, Baggerman s.n. (TTC, WIS, UWM). VIRGINIA: Campbell Co.: 8 mi from Lynchburg, 13 Oct 1933, Freer s.n. (US). Page Co.: banks of Shenandoah River, Luray, 22 Aug 1912, Holtzman s.n. (US). Warren Co.: sandy field, Front Royal, 18 Sep 1897, Miller s.n. (US). WEST RGINIA: Summers Co: New River, above Shanklin's Ferry, 23 Aug 1946, McNeill s.n. (WVA) Eastern and western populations have an apparent disjunction in distribution between eastern Missouri and east central Nebraska, a distance of ca. 650 km. A full list of specimens examined is available upon request to the authors. It is uncertain whether this is a true disjunction or simply the lack of intensive field work in this region to locate obscure populations (many botanists avoid botanizing in the midst of dense Collins et al., Orobanche riparia, a new species 11000 100*0'0"w 90*go"W 80'00"W TO*OO"W 50”0'0"N 1 20*00W EAS M HA se ap ui a EL ul 40"0'0"NH LFFEPEED ATE H LEN [E Sr Poen ret had H Y L1 [e Ces HH erry REA: H ESTEE er 2 iss EY Di. Sere) r CL EHEER a EA A e Yl X CERO je uade e dal ez 1] WE hr TAN 30"0'0"N+ son MEER EOS - ITA zy 5 PA E SS) dA LAD (1 vi te: d^ rdg sa EA ed i wi zt Bs el tating. ee RO, W "vs | SP SS qw e e SS do M ÉS e V ULA A we I) c c) ex SS re. i ADIT, A ES =40*0'0"N =30%0'0"N Fic. 2. Distribution of Orobanche riparia. e O. riparia Sems e a E T NE S 4S Q 125250 500 750 1,000 Kilometers 110"0'0W 400*00"W 9o*oo"W 80*0'0"W g allergy season). The eastern and western populations of O. riparia stands of ragweed and cockleburs d exhibit slightly different morphologies. Plants of the eastern populations (including the Missouri plants) are generally pallid with very pale lavender or purple pigmentation in the corollas. Pubescence is reduced and sometimes almost absent, especially on the corolla lobes. The primary host is Ambrosia trifida. The habitat of these populations is floodplains with mixed sand and heavy silty loam soils of the midwestern rivers cited above. The western populations (west of eastern Missouri) have more deeply purple-tinted corolla lobes, calyx lobes, and floral bracts. The upper stems, where exposed to sunlight, may be infused with a rosy purple. Pubescence is rather dense on the inflorescence, corolla lobes, and palatal folds. The primary host is Xanthium strumarium. Populations of this variant are concentrated along sand bars of the Platte River in Nebraska, with a few records of populations widely scattered in Kansas, Oklahoma, Texas, Colorado, and New Mexico. Specimens of this species collected along the Rio Grande River from central New Mexico to the Big Bend of Texas have proven problematic for identification throughout the study of Section Nothaphyl- lon. Only recently, when they were compared with the Nebraska plants, did it become apparent that they represent populations of O. riparia. It is tempting to name these two variants formally as subspecies within O. riparia. However, the life history and morphological distinctions are not sufficient to warrant taxonomic status at this time. This problem requires more intensive study of both morphological and genetic variation within and between populations before a conclusion can be reached as to the merits of segregating infrataxa within O. riparia. Host information for Orobanche is generally difficult to assess from herbarium labels. Collectors sometimes list as a host any adjacent plant without confirming a physical connection or list no host at all. However, host data for O. riparia reported here have been observed by all three authors. Only two non-Asteraceous MO Joumalof tani Insti Texas 3( host reports have been confirmed. Some populations in southern Ohio occasionally parasitize cultivated tobacco (Nicotiana tabacum). Likewise, some populations along the Rio Grande in western Texas have been reported on the weedy, introduced N. glauca (the only perennial host with which the species has been associ- ated). It is not possible at this time to judge He a of ak exceptions, if any. Anecdotal evidence suggests that other species of Orobanche lternate hosts at the margins of their ranges. The populations east of the Appalachian Mountains are of particular interest because they represent the extreme eastern portion of the range and may possibly be extirpated. Virginia records are represented by four herbarium specimens from four counties (on the James, Shenandoah and Potomac rivers). The most recent collection we have seen was made in 1933. A single specimen was seen from the New River, Summers County, West Virginia. The present study could not confirm extant populations in the District of Columbia, Virginia, or West Virginia. The species seems reasonably secure elsewhere in its range, although current herbarium vouchers document a highly fragmented range. For example, the single Tennessee record rep- resents a southward disjunction of ca. 325 km from the St. Louis County, Missouri station. However, field experience of the authors suggests that this species is probably more common than the herbarium records indicate. The senior author discovered several previously unreported populations in Nebraska, Illinois, and Indiana. Additionally, a few populations were observed in Illinois, Ohio, and Texas that were located on minor tributaries (creeks) of the major river systems. Some of the historical populations were determined to be extant. Thus, it is likely that concentrated field work will reveal additional populations of O. riparia. The status of the historical populations in the District of Columbia, Virginia, and West Virginia especially needs to be updated by further field work. The habitat of O. riparia is disturbed annually by levents and h tivity. Although it is widely distributed, populations seem to be locally isolated and therefore fragile. The senior author has observed the obliteration of a large population in Posey County, Indiana, as a result of riverbank development. Because of its unique distribution of widely separated populations in regularly disturbed habitats, the species is likely to be threatened on a local or state basis. ACKNOWLEDGMENTS The authors wish to express their appreciation to the staff and curators of the following herbaria for the loan of specimens and/or for providing images and label data from specimens in the Orobanche riparia alli- ance: A, ARIZ, COLO, CS, F, GH, ILL, IND, JEPS, KS, MICH, MIN, MO, NEB, NY, OSC, OKLA, PH, POM, RSA, TENN, TTC, UC, US, UWM, WIS, WTU, and WVA. Collins is grateful to Ken Heil (San Juan College) for accompanying him during field work in the southwestern U.S. Financial support for some of the field work of Collins was provided by a grant from the Evangel University Faculty Summer Research Program. Colwell's initial contributions to this project during her doctoral studies in St. Louis were supported by the Department of Biology at Washington University. The Flora of North America Project additionally provided funding for one of Colwell's trips to the Missouri Botanical Garden. REFERENCES Beck von MANNAGETTA, G. 1890. Monographie der Gattung Orobanche. Biblio. Bot. 4(19):78-85. Beck von MANNAGETTA, G. 1930, Orobanchaceae. In: A. Engler, ed. Das Pflanzenreich: Regni vegetabilis conspectus im Auftrage der Kóngl. preuss. Akademie der Wissenschaften vol. IV (261), issue 96. Wilhelm Engelmann, Leipzig. Pp. 1-348. BENNETT, J.R., AND S. MarraEws. 2006. Phylogeny of the parasitic family Orobanchaceae inferred from phytochrome A. Amer. J. Bot. 93:1039-1051. Cotuns, LT. 1973. Systematics of Orobanche section Myzorrhiza. Ph.D. dissertation, University of Wisconsin, Milwaukee. HeckarD, L.R. 1973. A taxonomic reinterpretation of the Orobanche californica complex. Madroño 22:41-104. Collins et al., Orobanche riparia, a new species 11 KARTESZ, J.T. 1994. A synonymized checklist of the vascular flora of the United States, Canada, and Greenland, 2 vols. Timber Press, Portland, Oregon. Munz, PA. 1930. The North American species of Orobanche, section Myzorrhiza. Bull. Torrey Bot. Club 57: 611-624. Munz, PA. 1974. A flora of southern California. Univ. of California Press, Berkeley. MussLeman, L.J. 1982. The Orobanchaceae of Virginia. Castanea 47: 266-275. NurraLL, T. 1818. The genera of North America plants, and a catalogue of the species, to the year 1817, 2 vols. Published by the author (printed by D. Heartt), Philadelphia, Pennsylvania. OLMSTEAD, R.G., C.W. DePampnitis, A. Wotre, N.D. Young, W.J. ELison, AND P. Reeves. 2001. Disintegration of the Scrophu- lariaceae. Amer. J. Bot. 88:348-361. ScHNEEWEISS, G.M., A. CouweLL, J.-M. Park, C.-G. JANG, AND T.F. Steussy, 2004. Phylogeny of holoparasitic Orobanche (Orobanchaceae) inferred from nuclear ITS sequences. Molec. Phylogen. Evol. 30:465-478. Shreve, F. AND ILL. Wicains. 1964. Vegetation and flora of the Sonoran Desert, 2 vols. Stanford University Press, Stanford, California. SrRAUsBAUGH, P.D. AND E.L. Core. 1978. Flora of West Virginia, ed. 2. Seneca Books, Grantsville, West Virginia. I Al S£ Das anli D L De if — BOOK REVIEW STEPHEN INGRAM. 2008. Cacti, Agaves, and Yuccas of California and Nevada. (ISBN 978-0-9789971-0-6, pbk.; 978-0-9789971-1-3; hbk.). Cachuma Press, PO. Box 560, Los Olivos, California 93441, U.S.A. (Orders: www.cachumapress.com, 805-688-0413). $25.95 pbk., $35.95 hbk., 256 pp., 262 color photographs, 16 botanical watercolors, 52 range maps, 6 3/4" x 9 1/4". What a beautifully illustrated book! The cover feet for the E cover is 14 pou with the binding Smythsewn. The text paper is *Lumi gloss,” acid free, and 150 gsm weight. T liquid stains when in the car, wear and tear when carried in a book bag, or from oue lebris when in use ue ui groundwork. Ar 1 1 * gh de 1 11 3 f, Ihe paperback bl 1 large enough o that color images of landscapes show scenic vistas, habits that o n a oe stem, and d morphological details of HH plants, pee pus nd pps p g p igh-resolution pho t nage lavouts The | J } Ji J 1 f feld ith tiinth page layouts. I extraordinary dept foreg d and i ges in the backg d and both in focus One color image is an outstanding example of Ingram's spectacular photography and is represented by a two-page (176-177) uin out dn end me of dus Adams's most ne: isa rd o New Mene dad nd Be fast Boe [2] o Write images 111 1 Cl : 1 sRBAebesll H ` rp ? IE fl lie f, AA ` 1 J +} € the history of p graphy & ) Bernadino Mountains for the Sangre de Cristo Mountains, and you have a color image that is a magic moment. There are a number of magic moments in this book. ied Bent ud den San more uA He 000 d on e to Pe at the Le DRS at the TRE time to take ng ml Nevada r O E 1 1 ET * [gens | 1 p A ful E Ee Lae 1 + 1 r r Eu ao, e ? r Basin Desert, Mojave Desert, and Colorado Desert (part of the Sonoran Desert). The intent of tl horis to span tl dership f ] public, naturalist, a t li ] iati f nati l h tion, and he deserves five o r T: rr Pul stars for surpassing the highest standards. This book is divided into five chapters: (1) Evolution, Classification, and Botanical Characteristics; (2) Ecology and Habitats of Cacti, Agaves, and Yuccas; (3) HK Profiles; (4) Conservation and d and (5) Exploring the Cactus Country of California and Nevada. The eight appendices a lly informative and increase the value of the book: Cactus, Agave, and Yucca Species of California and Nevada; Men Vegetation Types and ppt for Cacti, Dee dy Yuccas; senec Names ie Non- ates Species; Cholla Species C C y Species List for ME Public tended in Califonia and Nev ada The species profiles for each entry includ name, the scientific name, a phonetic pronunciation guide, recent syn- elie oth common dé a commentary al he plant, a landscape, habit, and flower photograph pue a range map. Ad ing the species. The chapt ipti f the habit, stems, spines or leaves, flowers, and f pprot on oo ation 2 Cultivation”: pecially y ] ion i Siena aves cultivation A E 3111 1 1 1 1 1 g ? li ] ific si 1 lected i be found on bs section on on "Exploring n Cactus Country of calina ae Ney ada ate parks, national parks, nati I sildeenéss areas, wildlife ee and recreation areas. c : 1 wa. : 1 ] id i y common, in igenous rare I new species are th llent | th dagl hat defi ] inol lin the book. The 15 genera of Cactaceae due cli that occur in California. Ru Nevada are a by d ios ao paintings by Eugene O. Murman and th M dd Susan text. For cad “Packrat Middens me Vegetation Histor *Yuccas and Yucca Moths" in both dd to tl iosi iated plants, especially the yucca moth-flower o biology often featured in biology textbooks as obligate un. tho push the Pus are not M in ue text, the pees has more n 200 citati ly published in the last ten Bazell. T *t*r.l ears. J ing ted afr a : 1 decr ] A px by g genus and species li s book must ls on Hee E a did WB. botanist, a: on o landscaper, conservationist, dde] E E E 1 L JE 1 1 L th and national parks E in Western states. Cachuma Press, the editors will be cherished by all bibliophilesi—Harold W. Keller, Ph.D., Research Associate Botanical Research Institute of dois Fort Worth, Texas 76102-4025, U.S.A J. Bot. Res. Inst. Texas 3(1): 12. 2009 BRISTLECONE PINE DWARF MISTLETOE: ARCEUTHOBIUM MICROCARPUM SUBSP. ARISTATAE (VISCACEAE), A NEW SUBSPECIES OF WESTERN SPRUCE DWARF MISTLETOE FROM NORTHERN ARIZONA Jared M. Scott Robert L. Mathiasen School of Forestry School of Forestry Northern Arizona University Northern Arizona University Flagstaff Arizona 86011, U.S.A. Flagstaff, Arizona 8601 1, U.S.A. Jared Scott@nau.edu Robert. Mathiasen@nau.edu ABSTRACT We describe Arceuthobium microcarpum subsp. aristatae (Viscaceae), a parasite of bristlecone pine (Pinus aristata Engelm.) and Engelmann spruce (Picea engelmannii Parry ex Engelm.) in northern Arizona. oe and aer Me data were used to com- pare Á. microcarpum populations parasitizing bristlecone pine on the San Francisco Peaks, A a, with populations on Engelmann ME and a e (Picea pungens a from nto Pid in a and New Mexico. diit ieu e phenological, and Francisco Peaks and Kendrick rr r Peak in northern veneer as a new als RESUMEN Se describe A thobi i p subsp. aristatae (Vi ,un e de Pinus aristata NEN y de Picea engeimannii Parry e Engelm en el norte de Arizona. S 1 fol 5gi y g I p de A. microcarpum que D. 1 r 1 F Shee [PES f cee wees Aéxic Dis fala f, J hili Jad Anal parasitan a Pi istata de San F i , Arizona, blaci i Pi l ti APEA EN Pod 1 mentan la clasificación de las SNAM de A A. microcarpum de San Fand Peaks y de eig Peak en el norte de Arizona como una subespecie nueva Key Wonps: Arceuthobium, Picea engelmannii, Picea pungens, Pinus aristata, Arizona Western spruce dwarf mistletoe (Arceuthobium microcarpum (Engelm.) Hawksw. & Wiens) is an important parasite of blue spruce (Picea pungens Engelm.) and Engelmann spruce (Picea engelmannii Parry ex Engelm.) in the southwestern United States (Mathiasen et al. 1986; Hawksworth & Wiens 1996). It occurs as far south as the Pinaleño Mountains, Arizona and Sacramento Mountains, New Mexico and as far north as the Kaibab Plateau, Arizona. However, western spruce dwarf mistletoe is most common in the White Moun- tains, Arizona and Mogollon Mountains, New Mexico (Mathiasen et al. 1986; Hawksworth & Wiens 1996). On the San Francisco Peaks, Arizona, it also parasitizes Rocky Mountain bristlecone pine (Pinus aristata Engelm.) (Mathiasen & Hawksworth 1980). Initially, the dwarf mistletoe infecting bristlecone pine on the San Francisco Peaks was classified as limber pine dwarf mistletoe (Arceuthobium cyanocarpum (A. Nelson ex Rydb.) Coulter & Nelson) by Hawksworth and Wiens (1972). It was later classified as A. microcarpum based on its host affinities and chemical characters (flavonols) (Crawford & Hawksworth 1979; Mathiasen & Hawksworth 1980; Hawksworth & Wiens 1996). Although Mathiasen and Hawksworth (1980) reported morphological, phenological, and host susceptibility differences between the A. microcarpum populations on the San Francisco Peaks and other A. microcarpum populations in the Southwest, they did not recom- mend giving the San Francisco Peaks populations taxonomic status at that time. In 2006, we began a more detailed analysis of the morphological characteristics of the A. microcarpum populations in Arizona. Our data further substantiated the differences between the A. microcarpum populations on the San Francisco Peaks and those in other mountain ranges of Arizona first reported by Mathiasen and Hawksworth (1980). Because of the differences in plant size, plant color, phenology, host affinities, and geographic isolation of J. Bot. Res. Inst. Texas 3(1): 13 — 21. 2009 14 J t ical Insti Texas 3( the dwarf mistletoe populations on the San Francisco Peaks, and nearby Kendrick Peak, there is sufficient evidence to describe these populations as a new subspecies of A. microcarpum. Arceuthobium microcarpum (Engelm.) Hawksw. & Wiens subsp. aristatae J.M. Scott & Mathiasen, subsp. nov. (Figs. 1-2). Tre: U.S.A. Arizona. Coconino Co.: near summit of Schultz Peak, a southeast subsidiary ridge of the San Francisco Peaks, Coconino National Forest, elev. 3,060 m, Lat. 35? 18' 43" N, Long. 111? 37' 52" W, parasitic on Pinus aristata, 8 Aug 2006, J.M. Scott 2006-6 ( : ASC; IsoTYPES: JEPS, UNM, US). Plantae (1237) lt li principales basi (121.8(—3) mm diam; fructus maturi 3.3 x 2.1 mm; anthesis mense Julio-Augusto; antde i=; ; F F fructus maturitas Augusto—Septembri; in Pinus aristata et Picea engelmannii parasiticae. Male plants (0.8-)2.7(-7.0) cm in height, basal diameter of dominant plants (1.0-)1.8(3.0) mm; female plants (1.4—)3.6(-7.0) cm in height, basal diameter of dominant plants (0.6-)1.8(3.0) mm; male and female plant plants primarily light green, but some brown-green or purple; male flower diameter 2.5 mm, perianth lobe dimensions 1.2 x 1.0 mm, anther diameter 0.5 mm, distance from anther to perianth tip 0.5 mm; fruit dimensions 3.3 x 2.1 mm; seed dimensions 2.4 x 1.1 mm. Common name.—bristlecone pine dwarf mistletoe. Etymology.—We selected aristatae, a name that indicates the principal host of this dwarf mistletoe is Pinus aristata. Phenology.—Anthesis from early July to late-August with peaks in late-July to early-August; seed dis- persal from mid-August to late-September with peaks in mid- to late-August to early-September. Habit.—Parasitic principally on Pinus aristata and occasionally on Picea engelmannii on Schultz Peak, parasitic principally on Picea engelmannii in the Inner Basin of the San Francisco Peaks and on nearby Ken- drick Peak. Not known on Picea pungens. Rarely infects Pinus flexilis James and Abies lasiocarpa (Hook.) Nutt. on Schultz Peak. Distribution.—Bristlecone pine dwarf mistletoe only occurs in three small populations on the San Francisco Peaks (Schultz Peak, Weatherford Trail, and Inner Basin, Fig. 3). It also occurs near the summit of Kendrick Peak (Fig. 4). It ranges in elevation from 2890 to 3130 m on Schultz Peak. METHODS Collection Locations.—1n 2006 and 2007, plants of A. microcarpum were collected from 18 populations in Arizona (Appendix A; Appendices A—F available online at http://www.for.nau.edu/SOFArchive/MSF- Students/JMScott/20090325.pdf). Four populations were sampled near Flagstaff, Arizona (host plants): Schultz Peak (bristlecone pine and Engelmann spruce), Weatherford Trail (bristlecone pine), Inner Basin (bristlecone pine and Engelmann spruce) (Fig. 3), and Kendrick Peak (Engelmann spruce) (Fig. 4). A total of 12 populations of A. microcarpum were sampled in the White Mountains (host plants): Cache Cienega (blue and Engelmann spruce), Coleman Creek (blue spruce), Hannagan Creek (blue spruce), Johns Canyon (blue spruce), Lee Valley (blue spruce), Williams Valley (blue spruce), Forest Road 249 (blue spruce), For- est Road 402 (blue and Engelmann spruce), and Forest Road 72 (blue and Engelmann spruce) (Fig. 4). In addition, two populations were sampled from the North Rim of the Grand Canyon: Point Royal Road (blue spruce) and Route 67 (blue spruce). In 1975 and 1976, plants of A. microcarpum were collected by the junior author from the same loca- tions and hosts on the San Francisco Peaks that were sampled in 2006-07. Plants were also collected from the North Rim of the Grand Canyon (blue spruce), the White Mountains (blue and Engelmann spruce), the Mogollon Mountains (blue spruce), and the Pinaleño Mountains (Engelmann spruce) (Fig. 4). Morphological Measurements.—The following morphological characters of male plants were measured: dominant plant height, dominant plant basal diameter, flower diameter, anther diameter, perianth lobe length, perianth lobe width, and distance from the outer edge of the anther to the tip of the perianth lobe. The following morphological characters were measured for female plants: dominant plant height, dominant plant basal diameter, length and width of both fruits and seeds. The color of plants, fruits, and seeds were recorded. Plant heights and basal diameters were measured with digital calipers to the nearest 0.1 cm. All Cents his ; A thah: : H aricfataa 15 Plants of Arizona VISAS tatae Arceuthobium microcarpam sap. aris da M; "ems Mathiassen rona Tesoro C6 Cocenmo NF Servite Eevu tOnoze Habitat: On birstiecone pine Jared Scot 2006.8 23 August 2005 Northern Arizona University (NAU) Fic. 1. Holotype of Arceuthobium microcarpum subsp. aristatae J.M. Scott & Mathiasen, subsp. nov., J.M. Scott 2006-6, 8 Aug 2006 (ASC). Photo by R.L. Mathiasen. ; Arizona. Photo F f 7,714 2008 hv 1 M Scott (Rott 1 Mala A taken 30 May 2008 by J.M. Scott. Schultz Peak, Arizona. Photo Scott and Mathiasen, Arceuthobium microcarpum subsp. aristatae San Francisco Peaks m Arizona Fic. 3. The four infestations of western spruce dwarf mistletoe on the San Francisco Peaks, AZ. Dark circles indicate an infestation on bristlecone pine, white circles indicate an infestation on Engelmann spruce, half-dark and white circles indicate an infestation on both bristlecone pine and Engelmann Spruce. A LL ÉEc North im » e San D . Francisco Kendrick Peaks Peak @ a New . White Mexico Arizona „Mountains ss” e Mogollon Mountains inaleño Mountains PI te locations for plant collections of western spruce dwarf mistletoe in 1975 and 1976 (gray squares) and in 2006 and 2007 (closed circles). Plant lj 1f, : ER nara dd Sec a ap liqus 2007 (dosi of the Grand Canyon. other morphological characters were measured using a 10x hand lenses with a micrometer to the nearest 0.1 mm. Male plants were collected during peak anthesis and female plants were uda NE fruits were 1 mature. Over E male or female ps were collected for each population and 1 were comp characters that were measured in 2006—07 were measured lou the 1970s ae Characters EXER for both datasets included male and female plant height and basal diameter, flower diameter, perianth lobe length and width, fruit length and width, seed length and width, and plant color. Phenology Observations.—Anthesis and seed dispersal data were recorded when plants were collected for morphological measurements in 1975, 2006, and 2007. Data for the last two years were primarily from the San Francisco Peaks and Kendrick Peak due to frequent visits to these areas. During these years, the White Mountains were only visited twice each year: once during anthesis and again during seed dispersal. Therefore, the precise times of peak anthesis and seed dispersal were estimated for the White Mountains in 2006-07. In 1975, phenology data was collected from the San Francisco Peaks, Kendrick Peak, White Mountains, and Pinaleño Mountains, Arizona. Host Susceptibility.—In 2008, temporary 0.012 ha (radius 6 m) plots tablished around 26 severely infected bristlecone pines on Schultz Peak and 13 severely infected Engelmann spruces in the Inner Basin. In each plot, all trees > 1.4 m in height were examined for dwarf incas dud For each tree, species, diameter at breast height (dbh; to the nearest cm), and a dwarf mistl g R, Hawksworth 1977) were ] qos 1 ] using ten randomly selecrea 1 plants f IOT each led Not el of c 4 I Mash f A hah’ ` | | I aristatae 19 recorded. These data provided information on the relative susceptibility of bristlecone pine and Engelmann spruce to dwarf mistletoe infection on Schultz Peak and for Engelmann spruce in the Inner Basin. Data Analysis. Morphological data were combined from hosts on the San Francisco Peaks (bristlecone pine and Engelmann spruce) and Kendrick Peak (Engelmann spruce) and also for the White Mountains and North Rim (blue and Engelmann spruce) for analyses. Morphological data from 1975 and 1976 were included within these two data sets, but the 1970s data also included morphological measurements from the Mogollon and Pinaleño Mountains. A one-way analysis of variance (ANOVA, P-value < 0.05) was used to test for significant differences between the means of each morphological character measured. RESULTS Morphological Data.—On average, male and female plants of western spruce dwarf mistletoe (WSDM, subsp. microcarpum) were larger than those of bristlecone pine dwarf mistletoe (BPDM, subsp. aristatae) and the differences in size were significantly different (Table 1, Appendices B-E, online). Although the largest plants we measured in 2006 and 2007 were collected from blue spruce in the White Mountains (Appendix B), the largest plants measured in the 1970s were from Engelmann spruce in the Pinaleño Mountains (Appendix E). Plant heights for the 1970s data were slightly larger than those for the 2006-07 data for BPDM, but both sets of data demonstrated that plants of WSDM were larger than BPDM. The mean basal diameter of male plants was similar, but the mean basal diameter of female WSDM plants was larger on average than female plants of BPDM and significantly different. The means for flower diameters, fruit lengths, and seed widths were significantly different, even though the differences were only about 0.1 mm or less (Table 1). The color of male and female plants of BPDM was commonly light green or green-brown. However, many plants on the San Francisco Peaks appeared purple (Fig. 2). The color of male and female plants of WSDM was similar to BPDM, except that some plants in the White Mountains were green-blue. Phenology.—Peak anthesis for BPDM occurred one to two weeks earlier on the San Francisco Peaks than for WSDM in the White Mountains in 2006 and 2007. Seed dispersal of BPDM also starts and ends one to two weeks earlier on the San Francisco Peaks than seed dispersal of WSDM in the White Mountains. The 1970s observations also found that BPDM flowers and disperses seed earlier than WSDM (Mathiasen & Hawksworth 1980) Host Susceptibility.—Infection of bristlecone pine on Schultz Peak was 9696 (n = 111) indicating it is a principal host of bristlecone pine dwarf mistletoe there. However, infection of Engelmann spruce was only 3496 (n = 224) on Schultz Peak placing it in the occasional host class of Hawksworth and Wiens (1996). Engelmann spruce is clearly much less susceptible than bristlecone pine on Schultz Peak. However, infec- tion of Engelmann spruce in the Inner Basin, just 5 km north of Schultz Peak, was 94% (n = 101). All of the bristlecone pines growing near severely infected Engelmann spruces in the Inner Basin were infected, but there are only about 10 bristlecone pines in the area. We also observed rare infection of limber pine (2 trees) and subalpine fir (3 trees) by BPDM on Schultz Peak. Table 2 summarizes the principal morphological and physiological differences between BPDM and WSDM used to distinguish the two subspecies. The shoot heights, colors, and phenology, of BPDM on the San Francisco Peaks and Kendrick Peak were similar enough to group them as one population. Morpho- logical data from Mathiasen and Hawksworth (1980) and our data demonstrated that BPDM plant heights were smaller than for WSDM populations in southern Arizona and central New Mexico. The differences in mean plant height between BPDM and WSDM were statistically significant. Additionally, color differences were found between BPDM (purple plants) and WSDM (blue-green plants). Furthermore, the periods of flowering and seed dispersal for BPDM on the San Francisco Peaks occurred one to two weeks earlier than for populations of WSDM in the White Mountains in 1975-76 and 2006-07. Another important set of A ecc characteristics that separates BPDM from WSDM are their host affinities. Bristlecone pine pal host of BPDM on Schultz Peak (9696 infection), but we found that only 3496 of the Engelmann Soria were infected there. Mathiasen and Hawksworth (1980) also 20 TABLE 1. por : morphological characters of Arceuthobium microcarpum subsp. aristatae and A. microcarpum subsp. microcarpum. Data combine measurements from and 2006—07 for all hosts and are presented as means (ranges) [n]. Characters ul an asterisk had significantly different ica Using ANOVA (P < 0.05). The means for seed width were rounded to the nearest 0.1 mm, but actual values g different. Character A. microcarpum subsp. aristatae A. microcarpum subsp. microcarpum Plant Height (cm) Male* 2.7(0.8-7.0) [152] 5.6(1.8-14.9) [283] Female* Basal Diameter (mm) Male Female* Flower Diameter SUBE 3.6(1.4-7.0) [177] 1.8(1.0-3.0) [121] 1.8(0.6-3.0) [167] 2.5(1.8-4.0) [287] 3:3(3.5-5.1) [281] 2.1(1.7-2.9) [281] 6.4(2.0-15.7) [353] 1.9(0.8-3.4) [157] 2.0(0.8-3.8) [313] 2.4(1.6-3.1) [266] 34(3.4-5.2) [440] 2.2(1.9-3.1) [440] Seed Length (mm) 24(1.5-34) [107] 24(1.3-3.4) [224] Seed Width (mm)* 1.1(0.8-1.4) [107] 1.1(0.7-1.5) [244] Tagle 2. Summary of the principal E and physiological differences between Arceuthobi j 1 ubsp. aristatae and A. microcarpum subsp. microcarp Character A. microcarpum subsp. aristatae A. microcarpum subsp. microcarpum Mean Plant Height (cm) Male Plants 27 5.6 Female Piants 36 64 Plant Color Light green; green-brown; purple Light green; green-brown; blue-green Peak Flowering Period One to two weeks earlier than One to two weeks later than subsp. microcarpum subsp. aristatae Host ad Bristlec ne Principal Unknown ey spruce Principal/Occasional? Principal Blue spruce nknown Principal Limber pine Rare Unknown Subalpine fir Rare Rare Southwestern white pine Unknown Immune 'Host susceptibility classificati j described in Hawl th and Wiens (1996). The host susceptibility designa- tions used for A, nin un micorcarpum are based on information in Hawksworth and Wiens (1996) also. 25 | | fiar | host for Subsp: aristatae in the Inner Basin of the San Francisco Peaks and on Kendrick Peak, Arizona, but it was an sal host of | n Schultz Peak. reported that bristlecone pine was the principal host of BPDM on Schultz Peak, and they reported a similar level of infection for Engelmann spruce (32%), but they used a different method of collecting infection data; a systematic sampling design using 27 rectangular plots. Although they classified Engelmann spruce as a secondary host of BPDM on Schultz Peak, the low incidence of infection we found indicates Engelmann spruce should be classified as an occasional host there. Based on our observations and those of Lynch (2004), blue spruce is more susceptible to infection by WSDM than Engelmann spruce, but both are principal hosts of WSDM. Peas blue spruce has not been reported on Kendrick Peak or the San Francisco Peaks, we were unable to infection data for BPDM on this host. There is also a report of dwarf mistletoe on Chihuahua spruce (Picea chihuahuana Martinez) in northern Mexico by Ledig et al. (2000) which we assumed would be WSDM. However, our examination of several Chihuahua spruce populations in Chihuahua and Durango Scott and Mathi , A thobi i | bsp. aristatae 21 reported to be infested with dwarf mistletoe, indicated that the spruces were infected with spruce broom rust (Chrysomyxa arctostaphyli Dietel) and not dwarf mistletoe. Spruce broom rust induces the formation of witches' brooms on Chihuahua spruce that are similar to those caused by dwarf mistletoes (Cibrián et al. 2007), which was undoubtedly the reason for the report of dwarf mistletoe on Chihuahua spruce in Mexico. Therefore, none of the Mexican species of Picea have been reported as the host of a dwarf mistletoe thus far (Hawksworth et al. 2002). Hawksworth and Wiens (1972, 1996) defined a subspecies as geographically restricted populations having small, but consistent variations. Variations in dwarf mistletoe plant size, color, host range, and phe- nology have been used as the primary characteristics to identify subspecies of dwarf mistletoe (Hawksworth & Wiens 1965; Hawksworth & Wiens 1972, 1977, 1996; Hawksworth et » T Wass & Mathiasen 2003; Mathiasen 2007; Mathiasen & Daugherty 2007). Furthermore, the g lation of the San Francisco Peak/Kendrick Peak BPDM populations is also consistent with Hw and Wiens (1972, 1996) defini- tion that a subspecies should be a "geographically restricted population." ACKNOWLEDGMENTS We thank M. Socorro González Elizondo for the Spanish translation for the Resumen. We also appreciate the field assistance of James White and Laine Smith in northern Arizona and Brian Howell in northern Mexico. Gregory M. Filip and Del Wiens provided helpful suggestions to an earlier version of the manuscript. REFERENCES CIBRIAN, T.D., D. ALVARADO, AND S.E. GARCIA. (EDS.). 2007. Forest diseases in Mexico. Universidad Autonoma Chapingo, Mexico. CRAWFORD, D.J. AND F.G. HawksworTH. 1979. Flavonoid chemistry of Arceuthobium (Viscaceae). Brittonia 31:212-216. HawkswoRTH, F.G. 1977. The 6-class dwarf mistletoe rating system. Gen. Techn. Rep. R.M. U.S. Forest Serv. 48:1-7. Hawksworth, F.G. AND D. Wiens. 1965. Arceuthobium in Mexico. Brittonia 17:213-238. HawksworTH, F.G. AND D. Wiens. 1972. Biology and classification of dwarf mistletoes (Arceuthobium). Agriculture Handbook 401, USDA Forest Service, Washington, D.C. Hawksworth, F.G. AND D. Wiens. 1977. Arceuthobium in Mexico: Additions and range extensions. Brittonia 29: 411-418 HAWKSWORTH, F.G. AND D. Wiens. 1996. Dwarf mist! biology, pathology, and systematics. Agriculture Handbook 709, USDA Forest Service, Washington, D.C. HAwKSWORTH, F.G., D. Wiens, AND B.W. Gas. 2002. Arceuthobium in North America. In: Mistletoes of North American conifers. BW. Giels and T.D. Cibrian, tech. coords. Gen. Techn. Rep. R.M.RS. U.S. Forest Serv. 98:29-56, HAWKSWORTH, F.G., D. Wiens, AND D.L. NickRENT. 1992, New western North American taxa of Arceuthobium (Viscaceae). Novon 2:204-211. LEDIG, Eis M. ebay el B. diia V. HERNANDEZ Reyes, C. FLORES LOPEZ, AND me bin ARTEAGA. 2000. Mexico and the d | f Pi Madroño Locations of g | por graphy of! 47:71-88. LyncH, A.M. 2004. Fate and characteristics of Picea damaged by Elatobium abietinum (Walker) (Homoptera: Aphididae) in the White Mountains of Arizona. W. N. Amer. Naturalist 64:7-17. Mathiasen, R.L. 2007. A new combination for Hawksworth's dwarf mistletoe (Viscaceae). Novon 17:217-221. MATHIASEN, R.L. AND C.M. DaucHerry. 2007. Arceuthobium tsugense subsp. amabilae, a new subspecies of hemlock dwarf mistletoe (Viscaceae) from Oregon. Novon 17:222-227. MATHIASEN, R.L. AND F.G. HAwkswonrH. 1980. Taxonomy and effects of dwarf mistletoe on bristlecone pine on the San Francisco Peaks, Arizona. Res. Pap. R.M. 224:1-10 MATHIASEN, R.L., FG. HAw«swonrH, AND C.B. EDMINSTER. 1986. Effects of dwarf mistletoe on spruce in the White Moun- tains, Arizona. Great Basin Naturalist 46:685-689. Wass, E.F. AND R.L. MATHIASEN. 2003. A new subspecies of A g (Viscaceae) from British Columbia and Washington. Novon 13:268-276. 7 BOOK REVIEW Arnos Farjon. 2008. A Natural History of Conifers. (ISBN 978-0-88192-869-3, hbk.). Timber Press, Inc., The Haseltine Building, 133 S.W. Second Avenue, Suite 450, Portland, Oregon 97204, U.S.A. (Orders: www.timberpress.com, 1-800-827-5622). $34.95, 304 pp., 157 color photos, 39 line drawings, 7 3/8" x 10 3/8". 1 A 3M» 3115 E ‘A J E J 11 x x 1 A x zt plant group y g y This is what we have in this volume. Mr. Farjon, retired head of tł ti f the Kew Herbarium, is an obviously dedicated botanist and always had a working knowledge of the conifers. | i d decidi ite a book on th j d porn) them all over again. io we are dus an extensive T of the conifers - enlightening naper on corrected classificati y methods t y o DNA studies, g ore accurate ob of the group. T} 1 fé if, JL 1 es D 1 AR th lah their viia is Hong and involved. DH course w are e also a pano of the pope Since some are fast-growing and quite dense. their t connecti is very important b 2 2 E ff, ] ¡lod inf, ; hath th 1 ae ne | TI iting should be of i I ienti ientist alike. The photographs by tl I ; some of which depict trees you would have to travel long, difficult ways to see for yourself, make this a wonderful book.—Elaine Bell, Volunteer, Botanical Research Institute of Texas, Fort Worth, Texas 76102-4025, U.S.A. J. Bot. Res. Inst. Texas 3(1): 22. 2009 A NEW CLIFF-DWELLING SPECIES OF ZAMIA (ZAMIACEAE) FROM BELIZE Michael Calonje Montgomery Botanical Center 11901 Old Cutler Road iami, Florida 33156, U.S.A michaelcamontgomerybotanical.org ABSTRACT Zamia meermanii (Zamiaceae), a new cliff-dwelli f l Belize is described. It is distinguished by having 1 to 3 pendent iid EE coriaceous ue that are entire or r erenulately nee on e uis un MM on A gda surface, and ich it most closely resembles, as well as to E d M : Jiff_q 1 š 7 ; hha} Gach Cchut & Deh or r£ O RESUMEN z> do Roli Cantal Pee € PD po : Se describe Zamia meermanii oe j tral q precip p guir porque tiene de l a 3 f T lad ] ee ee eee 2m t T Y en la superficie adii. y persi la superficie abaxial. Esta especi I Zamia furfi L.f., la especie más his 1 T : des "e E Z3 semejante, yt Y A 1 4 A t Vovides, Schut & Dehgan y Z lovalii Nelson INTRODUCTION Zamia is a New World genus ranging from the southeastern USA to Bolivia. It currently comprises 57 species (Hill et al. 2007) and is considered to be the most ecologically and morphologically diverse cycad genus (Norstog & Nicholls 1997). It includes species growing in areas with extremely high rainfall, such as Z. roezlii Linden from the Colombian Chocó region, and adjacent coastal Ecuador to species growing in full sun in extremely dry conditions, such as Z. encephalartoides Stevenson from the Santander region of Colombia. The genus also includes the only known obligately epiphytic cycad, Z. pseudoparasitica Yates, as well as two cliff-dwelling species: Z. cremnophila Vovides, Schutzman & Dehgan, and Z. sandovalii Nelson. Jan Meerman, a Dutch ecologist who resides in Belize, li a third cun dwelling species of Zamia in February of 1999 together with Martin Meadows of the Belize Botanic G during a Rapid Ecological Assessment of a private protected area. Since discovering the plant, Meerman has peen collecting ecologi- cal and distribution data for this species, some of which is used in this description. He was instrumental in bringing a team from Montgomery Botanical Center to Belize to study this taxon and other Belizean Zamiaceae in August and September of 2008. DESCRIPTION Zamia meermanii Calonje, sp. nov. (Figs. 1-2). Tr: BELIZE. Beuze D : 60 m, 28 Aug 2008, M. Calonje, J. Meerman & P Griffith BZ08-152 (moLoryre: BRH; isotypes: FTG, MO, NY, XAL). Species insignis habitui pulicol dex brevis, ape pendulis y litus; foliola coriacea ginil integris vel crenulatis triente apicali, is distinctis adaxialil pagini 3 : pro] Stems globose to cylindrical, to 27 cm long and 18 cm wide, MM or occasionally branching on older plants. Cataphylls chartaceous, stipulate, triangular with lanceolate apex. Leaves 1—3 per stem, pendent, 2-146.5 cm long. Petiole 26-60 cm long with abruptly swollen base 2-3 cm wide, variously unarmed or carrying numerous prickles (50+) up to 3 mm tall. Rachis 50 to 109 cm long, unarmed or bearing a few prickles (« 20). Leaflets 7—23 pairs per leaf on adult plants, jam 3- > cm c from each other with the point of attachment to the rachis 8.5-13.3 mm wide, obovate t narrowly oblong, coriaceous, 12.5-32 cm long, 3.4-7.5 cm wide, with 31-56 veins prominently raised on adaxial surface, visible but not J. Bot. Res. Inst. Texas 3(1): 23 — 29, 2009 Fic.1 (A) Habit of plant 1 M Inge ME fal . ight (Di E ey ee | : gl £4 le, E, " L hs y (0 Detail of trunk and petioles. (D) Variation in leaflet sh (E) à f daxial surface of leaflet showi ised vei d late t r flat L £: " lnc el : ETT RAT fev ct £ al. :4l £. laaflat L duit leaf. A and C rep t Calonje et al. BZ08-152; B and F represent Calonje et al. BZ08-156; D, E and G represent Calonje et al. BZ08-125. Calonje, Zamia meermanii, a new cliff-dwelling species from Belize [M 4 : : vn hl IDVA Emi Ave ile 2 IL 1 f^ id lu £ mat vll Fig. 2 f J t F 3 pi F t ve) b LÁ (PI Ad 1 £ hall /EY RA : tun? a PA (E\ Ra H trnhil $ 1 dehicri (G) Mature V F PM SAT ort J f 3 > 3 1 J uir nne "m el hull Dhatanranhe A E£ m 1 : r RUD + al fall E ire 2 rJ q“ t d 2" kil hab: lew RA f RA J u tc Calanie et al. R708- y j 152. 26 J t ical i Texas 3( protuberant on abaxial surface; margins entire or crenulately notched in the distal third; surface on emergent leaves reddish-brown, but densely white-tomentose, turning cream-colored with saffron-orange tomentum, and gradually maturing to green and glabrous adaxially, with some persistent brown tomentum abaxially along leaflet margins and near the po of Su to the rachis. Eophylls typically carrying a single pair of ovate leaflets. } te strobilus conica l-cylindrical, erect, at pollen release 13-16 cm long, 2.2—2.5 cm diam. pu individually or in groups of 2—5 or more, cream to brown colored; peduncle 9-10 cm long, 0.9-1 cm diam., covered by ferrugineous pubescence. Microsporophylls arranged in 12-18 columns, 17-30 rows; obtrullate, distal face hexagonal to oblong-hexagonal, 5.5-6 mm wide, 2.5-3 mm tall, and 1-2 mm thick ; microsporangia spheroidal, 1-1.6 mm diam., present on abaxial surface, absent on adaxial surface, 22-28 per microsporophyll arranged in two separate groups along margins. Megaspo- rangiate strobilus cylindrical, solitary, erect at maturity, 12-20 cm long, 6-7 cm diam., tan to light-brown tomentose from emergence to maturity; peduncle 8-12 cm long, 1.5-3 cm wide, tan to brown tomentose. Megasporophylls arranged in 6-10 columns, 7-15 or more rows; distal face oblong-hexagonal, 1.5-2.2 cm wide, 1.0—1.5 cm tall, 1-1.2 cm thick, extruded to a narrow, depressed terminal facet. Seeds with sarcotesta ovoid-pyramidal, red at maturity, 2.0-2.5 cm long, 1-1.2 cm wide; sclerotesta ovoid to ovate, 1.5-1.8 cm long, 0.9-1.1 cm wide. ined: BELIZE. Belize District: 40-70m, 25-26 Aug 2008, M. Calonje & J. Meerman BZ08-120 (FTG), BZ08-122 (FTG), BZ08-124 (FTG), BZ08-125 (BRH, FTG, MO, NY, XAL), BZ08-126 (FTG), BZ08-140 (FTG), BZ08-146 (FTG); 60 m, 28 Aug 2008, M. Calonje, J. Meerman & P. Griffith B208-154 (FTG); 70m, 27 Aug 2008, M. Calonje, J. Meerman & P. Griffith BZ08-156 (BRH). Etymology.—The specific epithet honors Jan Meerman, who discovered the species. Distribution and habitat Endemic to Cayo and Belize Districts in Belize, it occurs in seasonally dry tropical evergreen broad-leaved lowland forests on steep karstic hills as defined by Meerman and Sabido (2001), reaching elevations of up to 200 m. The arboreal vegetation is characterized by species such as Co- mocladia guatemalensis Donn. Sm., Metopium brownei (Jacq.) Urb., Plumeria rubra f. acutifolia (Poir.) Woodson, Bursera simaruba (L.) Sarg., Forchhammeria trifoliata Radlk. var. trifoliata , Coussapoa oligocephala Donn.Sm., Caesalpinia gaumeri Greenm., Erythrina standleyana Krukoff, Pseudobombax ellipticoideum A. Robyns, and Thouinia paucidentata Radlk. Disturbed open spots are characterized by the endemic Louteridium donnell- smithii S. Watson. These forests are semi-deciduous, with more than half of the trees being leafless for at least one month during the dry season, and with several species being leafless for four months per year. This semi-deciduous cl implies that the amount of light reaching the forest floor (and under-canopy cliffs) is dramatically different between the dry and rainy seasons. Scattered throughout these hills are vertical limestone cliffs where this species is found. It appears to be an obligate cliff-dwelling species, as it is typically found growing in cracks and crevices on sheer vertical walls and absent from th ling forest floor. The cliff faces are largely bare but depending on the level of hering of the rock and the amount of shading they receive, the accompanying cliff vegetation consists largely of herbaceous plants, hemi-epiphytes and vines such as Adiantum tenerum Sw., Agave angustifolia Haw., Anthurium schlechtendalii Kunth subsp. schlechtendalii, Anthurium verapazense Engl., Philodendron radiatum Schott var. radiatum, Clusia sp., Begonia sericoneura Liebm., Pitcairnia recurvata (Scheidw.) K. Koch, Trades- cantia spathacea Sw., Passiflora cobanensis Killip, and Passiflora xiikzodz J.M. MacDougal subsp. xiikzodz. Climate.—The average annual precipitation within this species’ range is estimated at 2000-2500 mm, with a distinct dry season occurring from February through May. The wettest month is July with an average monthly range of 300-440 mm, and the driest month is April with a range of 50-70 mm. The temperature ranges from 18°C to 31°C, with an annual mean temperature of 25°C. The coldest month is January and the warmest month is May. (Data derived from GIS analysis using Worldclim 1.4 climate layers as described by Hijmans et al. (2005)). Conservation status. —The extent of for thi is estimated to be 750 sq. kmas delimited by the eee range occupied by suitable karst hills in Belize and Cayo Unas The area of occupancy, 1 which incl only these karst formations, is 300 sq. km. Based on an g population density Calonje, Zamia meermanii, a new cliff-dwelling species from Belize 27 of 30 adult plants per km?, the estimated total population size for this species is 9,000 plants (Meerman, unpub. data). The karst hills where this species occurs are often isolated and separated by unsuitable habi- tats such as lowland forest, lowland savannas, and agricultural areas. In addition, this species occurs in discrete populations of only a few individuals, and does not appear to occur on all suitable habitats within its geographic range. As a result of the isolation between karst mountains and the clumped distribution of this species, populations are considerably fragmented. Visits to 12 different localities where this species Occurs appear to indicate that reproduction is occurring and seedling regeneration is healthy. The primary threat to this species appears to be fire, utilized in adjoining areas for slash and burn agriculture. These fires have been observed spreading up surrounding karst hills and decimating native vegetation. Another important threat to this species is mining for construction aggregates and dolomite, with several active and planned quarries in the region. Recent fieldwork in Belize suggests that illegal wild collection of cycad plants and seeds has occurred in the past and may present an additional threat to this species. Hopefully, the difficult terrain and inaccessibility of this cliff dwelling plant, combined with the fact that approximately 5096 of its area of occupancy lies within protected areas will help minimize the risk of illegal harvesting. Specific locality information has been purposefully withheld in order to further minimize this risk. Due to its limited extent of occurrence and area of occupancy, the fragmented nature of its populations, and the threats caused by fire, mining, and illegal harvesting, this species should be listed as Endangered (EN) based on IUCN Red List criteria ab(i-iv) and 2ab(i-iv) (IUCN, 2001). Reproductive phenology.—Elongating microsporangiate strobili have been observed in August and September in habitat, and pollen releasing strobili have been observed in od and OVID: under cultivation at Green Hills Botanical Collections near San Ignacio. A si with about a third of the seeds already dehisced, was collected in September of 2006, ud o en tócally dehisced with most seeds cleaned of sarcotesta, was collected in January 19 of 2006 by Martin Meadows of Belize Botanic Gardens. During the course of fieldwork with this species in August of 2008 our team found several distinct groupings of seedlings with newly-emerged eophylls, indicating that strobili probably disintegrated two to three months earlier. Further work is required to peier A the reproductive phenology of this species. Ecology.—A large number of young seedlings observed during the course of fieldwork indicate that reproduction is healthy and the pollinating agent is active, although it has not yet been observed. Seedlings germinating on the forest floor suffer 100% mortality within a year or two, suggesting the forest floor habitat is unsuitable for this species (J. Meerman, pers. comm.). In August of 2008 our team observed a female plant with its peduncle attached and several seedlings growing directly underneath on the forest floor. Several of these seedlings from this same strobilus had also germinated in a fissure on the cliff directly above the mother plant, indicating that some unknown dispersal agent had moved individual seeds or a piece of the strobilus to this location. Larvae of Eumaeus toxea Godart butterflies were observed feeding on emergent leaves. DISCUSSION Zamia meermanii is most similar in appearance to Zamia furfuracea L.f. from southeastern Veracruz, Mexico. Both species have obovate to oblanceloate or narrowly oblong coriaceous leaflets with persistent tomentum on the abaxial side of leaflets, distinct veins on the adaxial side of leaflets, and a similar number of maxi- mum leaflet pairs per leaf (ca. 25). Zamia meermanii differs from Zamia furfuracea in having usually solitary stems as opposed to freely branching stems; holding 1-3 pendent leaves per crown, rather than 5-7 erect or slightly arching leaves per crown; eophylls with a single pair of leaflets rather than eophylls with two pairs of leaflets, leaves cream-colored and covered with saffron-orange tomentum just prior to maturing as opposed to light green with predominantly white tomentum prior to maturing; coriaceous leaflets to 0.8 mm thick compared to coriaceous or extremely coriaceous leaflets to 1 mm thick; leaflet length to 28 28 1 intel ID t + f Texas 3(1) Taste 1. D lucti | for Za ia furfi (d cliff-dwelli g Zamia species Zamia Zamia Zamía Zamia meermanii cremnophila sandovalii furfuracea Microsporangia per microsporophyll 22-28 14-18 14-18 40-42 Microsporangia diameter (mm) 1-1.6 0.65-0.78 0.6 0.8-1.2 Microsporangiate strobilus 9-10 2.5-3 6.7 8-10 peduncle length (cm) Megasporangiate strobilus length (cm) 12-20 8.5-14 9-17 10-24 Megasporophyll rows 7-15 4-8 4-7 4-15 Megasporophyll width (cm) 1.5-22 1.6-2.6 2.5-3.5 1.5-1.8 Megasporantiate strobilus 8-12 4—5 3.6-6 10-20 eduncle length (cm) Sclerotesta length (mm) 1.5-1.8 1.5-1.7 1.8-2.2 1.1-1.5 Data derived from measurements by author, Whitelock 2002, Nelson 2006, and Schutzman et. al. 1988. cm compared to 18 cm; protuberantly raised veins on adaxial surface to 1.5 mm wide rather than slightly raised, narrower veins to 0.8 mm wide; and 22-28 microsporangia on abaxial side of microsporophylls as opposed to 40 or more on Z. furfuracea Zamia meermanii shares this unusual cliff-dwelling habit with two other Mesoamerican Zamia species: Zamia cremnophila from Mexico and Zamia sandovalii from Honduras. All three species typically carry 1-3 pendent leaves. Zamia meermanii is easily differentiated from the two other species by leaflet features alone (see key below). 1. Leaflets with distincly raised veins on adaxial surface and persistent tomentum on abaxial surface Z. meermanii 1. A | | I 1 : jore QE | f. lue . 1 "sl £. Z. cremnophila Z. sandovalii 2. Adaxial surface of petiole with d 2. Adaxial surface of petiole tod ded without d qd groove It has distinctly raised veins on the adaxial surface, persistent tomentum on the abaxial surface, and mar- gins that are entire or crenulately notched in the upper third, whereas Z. sandovalii and Z. cremnophila both have plane leaflets with no distinct raised veins on the adaxial surface, are glabrous on the adaxial surface, and have margins that are distinctly toothed on the distal half. Zamia sandovalii median leaflets are strongly falcate as opposed to those of Z. meermanii and Z. cremnophila, which are typically straight. Zamia meermanii and Z. cremnophila eophylls typically possess two leaflets, whereas Z. sandovalii eophylls typically possess four leaflets. In addition to the vegetative characters that easily distinguish these three species, reproductive char- acters are also useful (Table 1). Zamia meermanii microstrobili have longer peduncles and possess more and hyll than Z. cremnophila and Z. sandovalii. Megastrobili of Z. meermanii larger i I gi pet can be longer and with more sporophyll rows and have longer peduncles than either Z. cremnophila or Z. sandovalii. Seeds of Z. sandovalii are the largest of the group, and those of Z. meermanii and Z. cremnophila are of a similar smaller size. Until a well-resolved phylogeny of Zamia is available, we have no way to determine if the cliff-dwelling habit evolved more than once in the genus. ACKNOWLEDGMENTS Funding in support of the 2008 Belize expedition which resulted in the description of this species was generously provided by the Association of Zoological Horticulture (AZH), Tim Gregory, and Montgomery Botanical Center (MBC). Jan Meerman of Biodiversity & Environmental Resource Data Systems of Belize introduced me to this species and participated in all fieldwork. Patrick Griffith (MBC) assisted in data and Calonje, Zamia meermanii, a new cliff-dwelling species from Belize 29 specimen collection, and editing the manuscript. Green Hills Botanical Collections donated significant in- kind support by providing lodging, equipment and work infrastructure in Belize. Belize Botanic Gardens provided megasporangiate strobilus images and phenological data. Alan Meerow and Bart Schutzman pro- vided assistance with the Latin diagnosis and comments on the manuscript. Alberto Taylor (PMA), Anders Lindstrom of Nong Nooch Tropical Botanic Garden, Miguel Angel Pérez-Farrera (HEM), and Chad Husby (MBC) provided comments on the manuscript. The Belize Forest Department granted the scientific research and collecting permit (CD/60/3/08(45)) supporting research on Belizean Zamiaceae, as well as herbarium infrastructure for preservation of voucher specimens. REFERENCES HUMANS, R.J., S.E. CAMERON, J.L. PARRA, P.G. JONES, AND A. Jarvis, 2005. Very high resolution interpolated climate surfaces for global land areas. International J. Climatology 25:1965-1978. Hitt, K.D., D.W. STEVENSON, and R. OsgorNE, 2007. The world list of cycads. Proc. 7th Int. Conf. on Cycad Biol. (CYCAD 2005), Xalapa, Mexico, January 2005. Mem. New York Bot. Gard. 97:454—483. IUCN. 2001. IUCN Categories & Criteria, ver. 3.1. IUCN Species Survival Commission, Gland, Switzerland. Website: . Accessed 16 October 2008. Meerman, J. AND W. Sagipo. 2001. Central American ecosystems: Belize. Programme for Belize, Belize City. 2 volumes. Netson, C.H. 2006. Dos plantas del genero Zamia (Gimnosperma) nuevas de Honduras. Ceiba 46:41-44. Norstoc K.J AND T.J. NicHoLLs. 1997. The biology of cycads. Cornell University Press, Ithaca, New York. ScHuTZMAN, B., A.P VoviDEs, AND B. DEHGAN. 1988. Two new species of Zamia (Zamiaceae, Cycadales) from southern Mexico. Bot. Gaz. 149:347-360. Wuitetock, L.M. 2002. The cycads. Timber Press, Portland, Oregon. BOOK REVIEW Lynne CHAPMAN, NOELENE DRAGE, Di DursTON, JENNY JONES, HILLARY MERRIFIELD, AND BILLY West. 2008. Tea Roses: Old Roses for a Warm Climate. (ISBN 9781877058677, hbk.). Rosenberg Publishing Pty Ltd, PO. Box 6125, Dural Delivery Centre, New South Wales 2158, Australia. (Orders: www.rosenbergpub.com. au, 61-2-9654-1502, 61-2-9654-1338 fax). $59.95, 240 pp., color photographs, glossary, appendices, bibliography, index, 9 1/4" x 11 1/4”. This beautiful book is presented by six ladies who love roses with a passi ] realized the lack of hi £d tralia. The tea rose or P Scented China rose hidd China f. ies. They were BUB to Europe in about 1820 and later to Australia. They | d, studied, and hybridized tt hth losing an g over and over again. bia over two hundred pages covering every De of the tea rose, this book is a sioe work and just fascinating for a rose lover. T d in detail, from the horticultural and botanical facts to bs po history. ue teas are presented in all, each with a list of distinguishing features, cultivation background, location, The pepe is quite fantastic as is rud polo by the AE as as ems um close- NUS T aen MS views. and other p Of additio note is the authors’ f di listinguisl ] ] f family ions. Tea rose “imposters” are discussed and a “What not to grow” list from 1922 is given. ook for the beauty of the rose and a book for the factual study of the rose, Tea Roses should strike a chord with a variety of readers.— Elaine Bell, Volunteer, Botanical Research Institute of Texas, Fort Worth, Texas 76102-4025, U.S.A. J. Bot. Res. Inst, Texas 3(1): 30. 2009 A NEW SPECIES OF ZAMIA (ZAMIACEAE) FROM THE MAYA MOUNTAINS OF BELIZE Michael Calonje Jan Meerman Montgomery Botanical Center Green Hills Butterfly Ranch 11901 Old Cutler Road and Botanical Collections Miami Florida 33156, U.S.A Cayo District, BELIZE michaelcamontgomerybotanical.org Patrick Griffith and Geoffrey Hoese Montgomery Botanical Center 11901 Old Cutler Road Miami, Florida 33156, U.S.A ABSTRACT Zamia decumbens (Zamiaceae), a new species from the Maya Mountains of Belize, is described and illustrated. It has affinity with Zamia tuerckheimii, Zamia bussellii/onan-reyesii, and Z. monticola, to which it is compared. RESUMEN mu acce = - q 1 (7 : A le] nac M de Belize M t finidad Zamia ES tuerckheimii, Zamia bussellii/onan-reyesii, y Z. monticola, con las que se compara. INTRODUCTION In December of 1997, while lucting research on wild nutmeg trees, John Janovec and Amanda Neill were led to the bottom of a large sinkhole in the Maya Mountains of Belize by naturalist guide Valentino Tzub to look at an interesting group of plants he called “corn palms.” At the bottom of the sinkhole, they found an old Mayan kiln and many pottery shards, and a dense population of Zamia (Janovec & Neill 2003). The plants did not match the description of any Zamia known from Belize at the time, but shortly after their return, Dr. Dennis Stevenson of the New York Botanical Garden informed them that the holotype of Zamia prasina Bull, a species described in 1881 (Bull 1881), had recently been re-discovered at the Kew herbarium. Janovec and Neill (2003) wrote that they connected these specimens to the sinkhole plants, and the name Zamia prasina has since been applied to this plant in the horticultural trade and several publications (see Balick 2000; Whitelock 2002). In August, 2008, Montgomery Botanical Center (MBC) sponsored an expedition to Belize to shed some light on this poorly-known species and other Belizean Zamiaceae. During the course of the expedition, it became clear that the leaflets of these sinkhole-dwelling plants, which were chartaceous, gradually acumi- nate, with a distinctly raised longitudinal fold, and margins that are entire or with a few minute teeth at the apex, were very different from the leaflets of the Z. prasina holotype, which were coriaceous, abruptly acuminate, strongly serrulate and lacking a longitudinal crease. We determined that Z. prasina was in fact a prior valid name for Z. polymorpha DW. Stev., A. Moretti & Vazq. Torres (see Calonje et al., pages 43 — 49, this volume), and that the cycads found in the sinkhole belonged to an undescribed species. During the cycad research expedition, the authors studied four separate populations of this species in Belize, gathering enough data to formally describe it. SPECIES DESCRIPTION Zamia d I Calonje, Meerman, M.P Griff. & Hoese, sp. nov. (Figs. 1-3). Tree: BELIZE. Torno District: bottom of sinkhole, 350—400 m, 2 Sep 2008, M. Calonje, J. Meerman & P Griffith BZ08-201 (HoLorYrE: BRH; isotypes: FTG, MO, NY, XAL). Species Z. tuerckheimii Donn.Sm. affinis sed caule d bente et pedunculo pollinis longissimo. J. Bot. Res. Inst, Texas 3(1): 31 — 41. 2009 Journal of the Botanical Research Institute of Texas 3(1) Fic. 1. Habit and habitat of Z. decumbens. A. Adult female plant with naturalist guide Valentino Tzub. B. View from sinkhole bottom at type locality of Z. decumbens. tains, Belize Calonie etal.. Z. J} , 16 mm K Fic. 2. Rey j f £7 d b All pl g pl j jfi yl | lity | I lation. A. C f I g bilus, adaxial side. B. C cti f | g bilus, abaxial side. C Rhopalotria sp., put i I ll fZ decumbens. D. Pharaxonotha SP., putative pollinator of Z. decumbens. E Microsporophyll abaxial side. F. Mi I iate strobilus. G. M g gi ti bil | Il | ivi y stag H M gas gi trobilus. |. Seed witt test 1.J. M I l yll f KM poro Í yll [ 1, abaxial side 34 i i Texas 3( Stems globose to cylindrical, to 80 cm long and 6.7-11 cm wide, decumbent and occasionally branch- ing on older plants, in habitat often rotting at base and rooting adventitiously from along the horizontal stems. Vegetative cataphylls tan-tomentose, chartaceous, stipulate, shallowly triangular to triangular with short, brown apex; strobilar cataphylls tan-tomentose, chartaceous, narrowly triangular with elongated light-tan lanceolate apex. Leaves on adult plants 5-17 per crown, erect or arching, 90—175 cm long, light green when emerging, turning dark green when mature. Petiole 30—55 cm long with abruptly swollen base 2.2-3.2 cm wide, moderately to heavily armed with prickles up to 5.3 mm long. Rachis 60-120 cm long, unarmed or sparsely armed in the proximal third. Leaflets to 28 pairs per leaf, 2.8-4.1 cm wide, 17.5-29 cm long, spaced 2—5.5 cm apart from each other with the point of attachment to the rachis 4.5-6 mm wide, chartaceous, lanceolate with gradually tapering apex and distinct longitudinal furrow, straight or slightly falcate, margins entire or with a few teeth restricted to apex. Eophylls typically with a single pair of ovate leaflets. Microsporangiate strobilus conical-cylindrical, emerging erect with peduncles progressively lean- ing outward horizontally, at pollen release 10—16.5 cm long, 2.1—2.5 cm diam., occurring individually or in groups of 2-6 or more, cream to brown; peduncle 12-23 cm long, 0.9-1 cm diam., green with brown pubescence. Microsporophylls arranged in 16-18 columns, 26-36 rows; obtrullate, distal face hexagonal to oblong-hexagonal, 3.0-3.5 mm wide, 4.0—4.5 mm tall; microsporangia spheroidal, 1-1.2 mm diam., present on abaxial nae absent on ue surface, 12-18 per microsporophyll arranged in two separate groups along margins. Meg bilus cylindrical, solitary, erect at maturity, 12-20 cm long, 5.7-6.4 cm diam. , emerging cream- color tomentose and maturing to glabrous light or dark green with patches of tan tomentum ing on inner facet and margins of megasporophylls; peduncle 7-13 cm long, 1.5-2 cm wide, green to brown tomentose. Megasporophylls arranged in 8-12 columns, 4—16 rows; distal face hexagonal to oblong-hexagonal, 2.25-2.9 cm wide, 1.6-1.8 cm tall, 2.5-4 mm thick, with a depressed terminal facet 9.5-12mm wide, 3.5—4.55 tall. Seeds with sarcotesta ovoid to ovoid-pyramidal, red at maturity, 1.7-2.0 cm long, 1.1-1.45 cm wide; sclerotesta ovoid to ovate, 1.45-1.65 cm long, 1.05-1.43 cm wide. Other vouchers examined: BELIZE. Cayo District: 170—200 m, 19 Aug 2008, M. edd J. Meerman, M.A. Perez-Farrera, B. Arevalo BZ08-040 (FTG), BZ08-041 eg BZ08-053 (BRH). Toledo District: 100—200 m, 9,11,12 Mar 1987, Davidse & Brant 32232 (MO); bottom of sinkhole, 700 m, 3 Oct 1999, Janovec & Neill 1185, 1186 (FTG). bottom of sinkhole, 300—320m, 1 Sep 2008, M. Calonje & P. Griffith BZ08-180 (FTG), BZ08-189 (FTG), BZ08-194 (FTG); bottom of sinkhole, Toledo District, 700 m, 2 Sep 2008, M. Calonje, P. Griffith, J. Meerman, & V. Tzub BZ08-222 (FTG), BZ08-224 (FTG), BZ08-225 (FTG), BZ08-227 (FTG), BZ08-231(BRH), BZ08-232 (FTG); 350-400m, 2 Sep 2008, M. Calonje, J. Meerman & P. Griffith BZ08-180 (FTG), BZ08-189 (FTG), BZ08-194 (FTG), BZ08-202 (FTG). Etymology.—The specific epithet refers to the decumbent habit of the stems. Distribution and habitat. —Known from several locations in the Maya Mountains of Belize in Toledo, Cayo, and Stann Creek districts at elevations of 150-700 m. Currently considered a Belizean endemic, but some populations occur near the border with Guatemala, and since the Maya Mountains extend into South- eastern Petén Province, it is likely to occur here as well. These locations were within Tropical Evergreen Broadleaf Lowland Forest variants as defined by Meerman and Sabido (2001) and over a number of geologi- cal formations, most notably late Cretaceous limestones, but also older Triassic metamorphic bedrock to a lesser extent (Cornec d The authors h rved seven different f thi i f 1 throughout the Maya Moun- tains of Belize, encompassing an area of approximately 2600 km". OR each of these populations was extremely small and restricted to rocky mountaintops and ridges or on the bottom of sinkholes. The largest populations found have been inside two steep-walled sinkholes in Toledo district, each approximately 50-80 m wide and 30-60 m in depth, with approximately 150 large plants of reproductive age. The mountaintop and ridge populations observed to date have been much smaller, with a maximum of 20 plants observed, all much smaller in size than those observed in the sinkholes. The factor limiting the distribution of this species appears to be reduced soil moisture. In the sinkholes, the plants are found growing near the sinkhole walls within the sinkhole overhang drip line, where there is less light and they are protected from direct rainfall. Few other plant species are present in these areas, in low densities, with large areas of unoccupied soil (Table 1). The mountaintop populations may benefit from the quick-draining substrate provided by Calonie et al., Z J , Tage 1. À g | wit! t locality for Za d b Taxon Family Justicia sp. Acanthaceae Louteridium donnell-smithii Acanthaceae Adiantum macrophyllum Adiantaceae Syngonium podophyllum Araceae Chamaedorea geonomiformis Arecaceae visa veracruzana Aristolochiaceae Forchhammeria trifoliata var. trifoliata Capparidaceae lis spathacea Commelinaceae Dracaena americana Dracaenaceae ponen Ho URN Fuphorbiaceae Episcia pun Gesneriaceae Calatola d Icacinaceae Dorstenia lindeniana Moraceae icus Sp. oraceae Passiflora lancetillensis Passifloraceae Peperomia sp. Piperaceae Piper spp. Piperaceae Deherainia smaragdina var. smaragdina oo Myriocarpa heterostachya Urticac Myriocarpa longipes i Urera Ho Urticaceae the rocky environment. In Toledo district, other lithophytic species such as Chamaedorea adscendens and Chamaedorea schippii were also observed growing in the same environment. Climate—The average annual precipitation within this species’ range is estimated at 1800-2800 mm, with a distinct dry season occurring from February through May. The wettest month is July, with an aver- age monthly range of 205-580 mm; the driest month is April, with a range of 50-70 mm. The temperature ranges from 16-31°C, with an annual mean temperature of 22-25?C. The coldest month is January and the warmest month is May (data derived from GIS analysis using Worldclim 1.4 climate layers as described by Hijmans et al. (2005). Reproductive phenology.—The reproductive phenology of this little-studied species is not well known and no distinct periodicity is evident based on the authors' visits in September of 2008, as well as the examina- tion of photographic evidence archived at Montgomery Botanical Center from John Janovec and Amanda Neill's visits to the sinkhole populations in August of 1999, and June of 2001. Immature, pollen-releasing, and old microsporangiate strobili were observed during all three visits. Receptive megasporangiate strobili were observed in August and September, and near-mature strobili were observed in June and September. Ecology.—Several microsporangiate strobili at pollen-release stage observed at both sinkhole localities harbored numerous snout weevils of an unknown Rhopalotria species, as well as clavicorn beetles of an unknown Pharaxonotha species (Fig. 2c, d). Both genera are believed to be pollinators of Zamia and have previously been found in mi giate strobili of Zamia furfuracea L.f. (Vovides 1991) and Z. pumila L. (Tang 1987), as well as entering nt exiting both microsporangiate and megasporangiate strobili in many natural Zamia populations in Panama (Alberto Taylor, pers. comm.). A preliminary examination of the Rhopalotria insects suggests that they are either conspecific or close relatives to Rhopalotria mollis, one of the known pollinators of Zamia furfuracea L.f. (William Tang, pers. comm.). Ceratozamia robusta Miq. occurs throughout the range of Z. decumbens and both species were observed growing sympatrically at one location in Northern Cayo district. Zamia decumbens was also observed grow- ing together with Zamia variegata Warsz. at one location in southern Toledo district, with no evidence of hybridization. These two species are unlikely to overlap much geographically, as most observed populations of Z. decumbens occur above 300 m, which is the maximum of the altitudinal range of Z. variegata. — JOU OF Notice f Z. decumbens. c. 3. Di Fi D. Eophylls a Calonie et al., Zamia d b ies from the Maya Mountains, Belize 37 J 7 7 f P 7 DISCUSSION Zamia decumbens appears to be most closely related to Zamia tuerckheimii Donn.Sm. from Guatemala as well as a new species described from Honduras. The taxonomic priority for this Honduran species, described separately as Zamia bussellii Schutzman, R.S. Adams, J.L. Haynes & Whitelock and Z. onan-reyesii C. Nelson & Sandoval is currently in dispute and will poo be aed here as Z. bussellii/onan-reyesii. Zamia decumbens differs from both species in 1 tems to 80 cm long that often rot at the base and re-root horizontally. Zamia tuerckheimii has erect or leaning stems to 3 m tall, and Z. bussellii/onan- reyesii has erect or leaning stems to 2 meters tall. All three species have leaflets with a marked longitudinal crease most prominent near the point of attachment to the rachis, but the leaflets of Z. bussellii/onan-reyesii are characteristic enough to distinguish it vegetatively from the two other species (Fig. 4). Zamia bussellii/ onan-reyesii leaflets are papyraceous and with prominent teeth on the lower margin compared to those of the other two species, which are chartaceous with margins entire or minutely toothed at the apex. Leaflets of Z. decumbens, measuring 2.8—4.1 cm in width, are Mia s narrower than those of Z. tuerckheimii, which typically measure 4—9 cm in width. Another useful veg tinction between the two species is that Z. decumbens has moderately to heavily armed petioles wl | f Z. tuerckheimii are unarmed or sparsely armed. Mature microsporangiate strobili of Zamia decumbens are easily distinguished from those of the two other species as they have long peduncles 11-23 cm in length causing them to lean outward, whereas Z. tuerckheimii has erect to slightly leaning strobili with peduncles to 6.5 cm long and Z. bussellii/onan-reyesii has erect to slightly leaning strobili with peduncles to 8.5 cm long. Zamia bussellii/onan-reyesii has the larg- est microsporangiate strobili, measuring up to 27.5 cm long and 3.8 cm in diameter, whereas those of Z. tuerckheimii measure 14-17 cm in length and 2.5 to 4 cm in diameter, and those of Z. decumbens measure 10-16.5 cm. Zamia bussellii/onan-reyesii is further differentiated from the other two species because it has more than 40 microsporangia arranged in a single group on the abaxial surface of the microsporophyll whereas 2. decumbens has 12-18 ón arranged in two separate groups and Z. tuerckheimii has 18-26 I gia arranged int The microsporophylls of Z. tuerckheimii are elongate triangular, approximately 1. 8-2. 5 times as one as they are wide, whereas those of Z. decumbens and Z. bus- sellii/onan-reyesii are roughly triangular and n a 1.3 to 1.5 times as long as they are wide. TI bili of Z. bussellii/o the largest of the group, measuring 22.4—25.2 cm tall and 10- 11. 2 cm mi wide with sporophylls faces 4. 5-5 cm tall "m 4.3-4.6 cm wide, compared to those of Z. tuerckheimii which are 16.3—22 cm tall and 8.3-10 cm wide with sporophyll faces 1.8-2.2 cm tall and 2.3-3.8 cm wide, and to those of Z. decumbens which are 12-20 cm tall and 5.7-6.4 cm wide with sporophyll faces 1.6-1.8 cm tall and 2.25-2.9 cm wide. See the following dichotomous key and Table 2 for summary of diagnostic characters. The known geographical ranges of the three species do not overlap (Fig. 5). KEY TO Z. BUSSELLII/ONAN-REYESII, Z. DECUMBENS, Z. TUERCKHEIMII . Leaflets papyraceous with prominent teeth on lower margin Z. bussellii/onan-reyesii . Leaflets chartaceous with entire margins Or us toothed at the apex. 2. Stem decumbent | g 11-23 cm long Z. decumbens 2. Stem erect or leaning, ¡ gi bili peduncle to 6.5 cm long Z. tuerckheimii One other species, Zamia monticola Chamb. appears to share some similarities with Z. decumbens in leaf length, leaflet shape and size, and microsporangiate strobilus size. It was described by Charles J. Cham- berlain based on a single male plant cultivated from seed reportedly collected near Naolinco Crater in the vicinity of Xalapa, Mexico. Chamberlain (1926) contends that the single plant sprouted from a batch of seeds thought to be Ceratozamia mexicana. The species has never been found again in the area where it was reportedly collected, and it is thought that it is either extinct, or that Chamberlain's seeds became mixed in his glasshouse and this species was acquired elsewhere (Hill 2004). Material collected in southern Alta Verapaz in Guatemala is currently being identified as this species, but this Guatemalan material is poorly understood at this time. Considering the fact that the species was described from a single cultivated indi- vidual obtained under unusual circumstances, that its present geographic distribution remains unclear, and that its megasporangiate strobili are unknown, a detailed comparison cannot currently be made between this species and Z. decumbens. However, based on Chamberlain's description, this species appears to have an erect stem 14 cm in diameter compared to the stems of Z. decumbens, which are decumbent and up to 11 cm across. In addition, Z. monticola is described as having 20-32 microsporangia per microsporophyll, whereas Z. decumbens has 12-18. Conservation status.—Although we observed seven different populations in the Maya Mountains add- ing up to an extent of occurrence of 2600 km”, the populations were all small, occupying a small area, and limited to the very specialized mountain top and sinkhole habitats. Despite visiting seven populations, only the two sinkhole populations had numerous large plants of reproductive size (about 150 each). The rest of the populations, located on rocky mountaintops, all had less than 15 individual plants of a much smaller size than those observed in the sinkholes. In total, less than 350 plants were observed within a combined area of occupancy of less than a square kilometer. Discussion with local residents throughout the range of Z. decumbens confirms that plants of this species have been extracted commercially in the past and seeds continue to be extracted for commercial purposes. During our fieldwork in the sinkholes, we found evi- dence of past mechanical removal of megasporangiate strobili, and a seemingly low number of seedling and young plants, perhaps suggesting that continued harvesting of seeds may be affecting the health of these populations. Specific locality information has been purposefully withheld from this paper in order to further minimize the risk of illegal harvesting. Considering how fragmented the distribution of this species is, most likely resulting from its specific habitat requirements, the small number of healthy, reproductively active populations observed, the small combined area of occupancy for this species, and the evidence of past and present commercial exploitation, Calonje etal. Z Belize TABLE 2. $ y of di icd Enr 7 tuerckheimii, Z. bussellii/onan-reyesii, and Z. decumbens. Z. tuerckheimii Z. bussellii/onan-reyesii Z. decumbens Microstrobilus length Microstrobilus width Microstrobilar peduncle eng Microsporophyll shape Microsporangia Megastrobilus length Megastrobilus width Petiole armature Leaflet length Leaflet texture Leaflet margins Eophyll leaflet pairs Stem height Stem habit 14-17 cm 2.5-4 cm to 6.5 cm Elongate triangular 18-26 Arranged in two rows Sparse or unarmed 19-30 cm to 14 cm Chartaceous Entire or with a few 27.5 cm 38cm 8.5 cm Broadly triangular ed in single row 22.4-25.2 cm Light to moderate 15.5-36 cm to 23 Papyraceous rominent teeth on minute teeth at distal end the lower margin 1 to3m Erect or leaning 4 to2m Erect or leaning rotting at base and 11-16.5 cm 2.1-2.5 cm 11-23 cm Broadly triangular 12-18 arranged in two rows Moderate to heavy 17.5-29 cm to 28 Chartaceous Entire or with a few minute teeth at distal end ] to 80 cm long. Decumbent, often rooting horizontally Zamia bussellii/onan-reyesii measurements derived from Schutzman et al., 2008. Zamia tuerckheimii measurements derived from Donnell Smith (1903), Standley & Steyermark (1958), Vannini 2008, and measurements taken by primary author. Zamia decumbens measurements taken in-situ by authors. All measurements based on mature plants. we consider this species critically endangered and therefore recommend a Red List Category of CR for this species based on criteria B2ab (i, ii, iii, iv, v) (IUCN 2001). It ue be noted that although the criteria have changed somewhat based on new information, ntinuation of the Red List category currently listed for this species under the misapplied name “Zamia prisa ACKNOWLEDGMENTS The Association of Zoological Horticulture (AZH) funded our research and conservation work with this species through the grant "Zamia prasina in Belize." Tim Gregory and Montgomery Botanical Center sup- ported additional work in Belize. John Janovec and Amanda Neill were the first to scientifically document and collect the sinkhole popu- lations of Z. decumbens, providing the inspiration for our return to Belize on a research expedition. Green Hills Butterfly Ranch and Botanical Collections donated significant in-kind support by providing lodging, equipment and work infrastructure in Belize. The Belize Forest Department granted the scientific research and collecting permit (CD/60/3/08(45)) supporting research on Belizean Zamiaceae, as well as herbarium infrastructure for preservation of voucher specimens. Valentino Tzub and Boris Arevalo provided field as- sistance. Willie Tang offered to identify the possible pollinators of Z. decumbens and provided the images of them. Jay Vannini pru information on Z. tuerckheimii and comments on the manuscript. Bart Schutzman helped in the Latin diag 1 provided comments on the manuscript. Anders Lindstromm, Chad Husby, and Alberto Sidney Taylor provided comments on the manuscript. Nancy Korber provided assistance in locating hard-to-find references. Dennis Stevenson provided herbarium specimen images and references. E E di a § m. 4 BE vd e tn gett Tos Mexico 9o d - j 9 j i a S y d is f d f A iai S E A P ; ë | i y ah aw | Co | Belize a ns í + ES ] ] a ;o. #8 x AS ee z i Fa , = x a X \ | E ^m, ( pos i å "d Sy i y ff E i de i EOM Í » E $ i a? d E = č } a uel NEU cu Som Guatemala i p > | a H i Tr. — ooo «SN mos . h. Mi X we S zi e i SW ° "d 2 e. ^ a en e i p AA (Species | E Honduras | p IE Z busselli/onan-reyesii, | 4 Z decumbens | 4 ; | ates A 100 e Z. tuerckheimii : D km Á Fic. 5. Geographical distribution of Z. bussellii/onan-reyesii, Z. decumbens, and Z. tuerckheimii. REFERENCES BaLick, MJ, M.H. Nee, AND D.E. Atha. 2000. Checklist of the vascular plants of Belize, with common names and uses Mem. New York Bot. Gard. 85:246. Butt, W. 1881. Retail List 20. CHAMBERLAIN, C.J. 1926. Two new species of Zamia. Bot. Gaz. 81:218-227 Cornec, J. 2003. Geology Map of Belize. Private Publication. Donnett SMITH, J. 1903. Undescribed plants from Guatemala and other Central American republics XXIV. (Zamia tuerckheimii). Bot. Gaz. 35:8-10. IUCN 2001. IUCN Red List Categories and Criteria: Version 3.1. IUCN Species Survival Commission. IUCN, Gland Switzerland and Cambridge, U.K. HUMANS, R.J., S.E. CAMERON, J.L. PARRA, PG. JONES, AND A. Jarvis, 2005.Very high resolution interpolated climate surfaces for global land areas V. Int. J. Climat. 25:1965- : Hitt, K. 2004. The Cycad Pages. Zamia monticola. Sydney: Royal Botanic Gardens. http//plantnet.rbgsyd.nsw 8]. gov.au/cgi-bin/cycadpg?taxname=Zamiat+monticola [Accessed 27 December 200 Janovec, J.P. ANO A.K. Neti. 2003. Exploring the palms and cycads of the Maya Mountains of Belize: reflections on MBC-sponsored expeditions to Belize, 1999-2001. The Montgomery News 11(1):5-6. MEERMAN, J. AND W. Sasipo. 2001. Central American ecosystems: Belize. Programme d Belize) Belize City. 7 ZU A A L aceae) aer NELSON-SUTHERLAND, C.H. AND G.G. SANDOVAL-GONZALEZ. 2008. Una especie nueva de Ceiba 49:135-136. aadlil ] lU a Calonie et al., Z J 7 la decumbens, pecies from the Maya Mountains, Belize " SCHUTZMAN, B., R. ADAMS, J.L. HAYNES, AND L.M. WurreLock. 2008. A new endemic Zamia from Honduras (Cycadales: Zamiaceae). Cycad Newsl. 31(2/3):22-26. STANDLEY, P.C. AND J.A. STEYERMARK. 1958. Zamiaceae. In: Flora of Guatemala, Part |. Field Mus. Nat. Hist., Bot, Ser. 24:19-20. TANG, W. 1987a. Insect pollination in the cycad Zamia pumila (Zamiaceae). Amer. J. Bot. 74:90-99. VANINI, J.P. 2008. Notes on the Guatemalan cycad Zamia tuerckheimii Donn. Sm. Cycad Newsl. 30(3):4-7. Vovipes, A.P. 1991. Insect symbionts of some Mexican cycads in their natural habitat. Biotropica 23:102-104. WhrreLock, L.W. 2002. The cycads. Timber Press, Portland, Oregon. BOOK REVIEW Bonnie J. GiseL, with images by STEPHEN J. Joseph. 2008. Nature's Beloved Son: Rediscovering John Muir's Botanical Legacy. (ISBN 978-1-59714-106-2, hbk.). Heyday Books, PO.Box 9145, Berkeley, California, 94709 U.S.A. (Orders: www.heydaybooks.com, 1-510-549-3564, 1-510-549-1889 fax). $45.00, 247 pp., 9 1/2" x 12 1/4". A 1 1 13.21 14 1 ža mi } des fTIAl € ees - C TON, | Es J Lt. P NEA O J 2 I GE and yes, his Eu as s explored and found p f t ifers, all of which he considered friends as well as expressions of divinity. The auth 1 photographs follows Muir's ventures into wilderness with assages from the explorer's own works, drawings from hisi journals and EUM d i d i ross the United p P Į p States. In a way the book is a botanical Bormo of Muir—not definitive, as that would take vol , but llent sampling of his botanical endeavors. Tl 1 1 1 +1 f. iH 1 4l ] ce | fil $4] f, 1 in th « p y I I I Plant Gallery Citations," with luced pi , a herbarium legend, and a quotation from Muir on his discovery of each plant. Di 1 1 T Jis 1 1 lod farh: : 1 i ; [1 1 : ; 1 e r d I I cC adding contrast, and eli 1 f he label d d 1 ] 1 d t Lo e a r Y Y "T 1 iE 1 ben bear es 1| N34 11 3111 e 1 11 1 J CNA. 4 i=) Joann Karges, Texas Christian University Library (retired), Box 298400, Fort Worth, Texas 76129, U.S.A. J. Bot. Res. Inst. Texas 3(1): 42, 2009 WHAT IS ZAMIA PRASINA (ZAMIACEAE: CYCADALES)? Michael Calonje Jan Meerman Montgomery Botanical Center Green Hills Butterfly Ranch 11901 Old Cutler Road and Botanical Collections Miami, Florida 33156, U.S.A. Cayo District, BELIZE ABSTRACT Maya Mountains: in Belize. An examination of the 1 ical ial of Zami ina, herbarium specimens, and wild Zamia popula- tions in Pipe ] ] d t l y to o this taxon but instead is a valid prior name for Zamia ] l 1 A lectotype for Z 4L ra I» Is E 7 : 1 1 Key Worps: Nomenclature, Priority, Belize, Zamia prasina 1 I RESUMEN El nombre Zamia prasina se está aplicando en horticultura y literatura sistemática a un taxó ido de las Montañas Maya de Belize. U del material histórico de Zamia prasina, pecí de herbario, y poblaciones e de Zamia en imd nos 1] 1 1 A 1 1 1 z X: r L yporl i prioridad 1 1. Se designa un lectotipo [ Z 1 INTRODUCTION Zamia prasina Bull was described in 1881 in English nurseryman William Bull's horticultural catalog. Historically, this species has received little mention in the literature, but in recent years the name has been applied in the horticultural industry and systematic literature to a relatively unknown taxon from the Maya Mountains of Belize. We re-examined the historical material related to Zamia prasina as well as herbarium specimens and wild Zamia populations in Belize in order to uncover the true identity of this species, which is presented in the following analysis. HISTORICAL MATERIAL AND EARLY USAGE OF THE NAME ZAMIA PRASINA William Bull's 1881 horticultural catalog pa the ne ias for Zamia prasina: “A hand d distinct-looking Cycad, i luced from H f g h Honduras, now Belize]. The stems in the young plants roundish or roundish-ovate, clothed with the imbri les left by the falling of the 1 The leaves are equally pinnate, the dark Eigen ae terete, with an indistinct PME in genu and Enea with afew small Men prickles. The leaf-blade is oblong-ovate, pinnate, I I entire, and tar base, where they are set on by a oy swollen articulation, denticul 1s tł he upp f f a bright grass-green colour, whence the name. 1 Y? guinea.” Bull also provided a leaf of Zamia prasina, which was subsequently mounted as an herbarium specimen at Kew. In a letter attached to this specimen, dated March 22, 1881, he invites William Turner Thiselton-Dyer, at the time Assistant Director at Kew, to stop by his nursery and collect a leaf of Z. prasina. Although there was no notation on the specimen indicating when it was collected, it was probably shortly after Thiselton- Dyer received the invitation. The specimen at Kew tated ible hol fZ. prasina by Dennis Stevenson of the New York Botanical Garden on September of 1999. However, it has not been designated as a type for the species in any of the printed matter we consulted. This specimen, presumably provided by Bull the same year as he described the species in the horticultural catalog, undoubtedly represents the original material on which the species was based. However, because no reference was made to this specimen in the protologue, it cannot be considered a holotype under article 9.1 of the 2005 International Code of Botanical Nomenclature ("Vienna Code”). We remedy this situation by here designating this specimen as the lectotype for Z. prasina. J. Bot. Res. Inst. Texas 3(1): 43 — 49, 2009 4 tani i Texas 3( Zamia prasina Bull, Hort. Cat. 176:20. 1881. Ter: [BRITISH] HONDURAS [BELIZE]: cultivated from Belize, William Bull s.n., ar 1881 (Lecrorvre, here designated: K!). Zamia polymorpha D.W. Stev., A Moretti & Vázq.Torres, Delpinoa 37—38:3-8. 1998. Tyre: BELIZE. Caro: 22 Jan 1989, D. Stevenson et al. 1119 (uoLorvpe: NY; isotypes: BRH, FTG, MO, NY, U. In addition to this specimen, there is an illustration of a leaf and caudex of Zamia prasina at Kew (dated 1881). A letter attached to the illustration (dated June 15, 1881) provides some background on Bull’s original importation of Zamia prasina. The letter, addressed to “J. Smith Esq.” (presumably Kew curator John Smith), mentions that a “Mr. Watson” (presumably Kew assistant curator William Watson) “talked him out of two plants that he did not want to give up.” He also mentions receipt of two separate shipments of Zamia from Belize and that each of the plants taken by Mr. Watson came from a separate shipment. The first plant, described as having “widely separate” leaflets was the only surviving plant from one of the shipments. The second plant, which Bull called Z. prasina, was one of the larger plants from the second shipment of which only a few plants survived. Mr. Bull clearly did not want to let go of the two plants and was asking Mr. Smith to return them. In return, he promised to provide “a couple” of the smaller surviving plants of Z. prasina for the collection at Kew. The illustration appears to be traced from the original Z. prasina plant provided by Bull before it was returned, as it is on a herbarium-specimen-sized sheet and bears the hand-written words “returned Zamia prasina.” Another inscription on the illustration reads “cf. Z. latifolia Lodd.,” and an inscription on the herbarium specimen reads “Z. latifolia Lodd.” These notes, in the same handwriting, were most likely added by Thiselton-Dyer, as he included Z. prasina in his treatment of Mexican and Central American cycads (Thiselton-Dyer 1884) as a synonym of Z. latifolia Lodd. ex A.DC (see de Candolle 1868). Thiselton-Dyer added a question mark next to his listing of Zamia prasina indicating he was uncertain about its synonymy with Z. latifolia. In fact, today it is not entirely clear what Z. latifolia is, as it was a horticultural name from Loddiges’ catalog for which no type specimen or illustration was provided. Miquel (1843) considered Zamia latifolia a synonym of Z. muricata var. obtusifolia, later bringing it to species rank (1849). In 1847, when he still considered Z. latifolia a variety of Z. muricata, he published an illustration of a leaflet labeled Zamia latifoliae (Linnaea 19(4): Tab. VII, fig. a. 1847), which Stevenson and Sabato (1986) selected as the neotype for Z. muricata var. obtusifolia (Fig. 1a). Since so little is known about the true identity of what was originally called Z. latifolia in Loddiges’s catalog, it must be considered a nomen dubium as well as its synonym, Z. muricata var. obtusifolia. Furthermore, the leaflet illustrated in Miquel's publication is unlikely to have been derived from a Belizean species, as none of the known species have leaflets with strongly serrated rounded leatlet tips. The leaflet shape and serrations in Miquel's illustration match closely those of Z. furfuracea plants at Montgomery Botanical Center grown from seed eed near Alvarado, Veracruz, Mexico (Fig. 1b). Thiselton-Dyer clearly considered Bull's specimen to be di than Z. muricata Willd., as both species are compared on the Zamia prasina illustration in the same handwriting as the "Z. latifolia" annotations. The comparison indicates that Z. prasina has leaflets that are coriaceous, abruptly acuminate, and serrulate at the apex, whereas Z. muricata has leaflets that are chartaceous, gradually acuminate, and spinulose-serrate towards apex. Zamia muricata was originally described from Venezuela, and as currently circumscribed, is a species occurring only in Venezuela and Colombia. In his 1932 cycad treatment, Schuster listed Z. muricata var. obtusifolia as well as Zamia prasina as synonyms of Z. loddigesii var. latifolia (Lodd.) J. Schust., incorrectly citing the publication date for Bull's horticultural catalog as 1822 rather than 1881 (Schuster 1932). MODERN USAGE OF THE NAME ZAMIA PRASINA Since Schuster's synonymization of Zamia prasina, the species name received little use until recent years, as apparently the original description and Bull's specimen were misplaced and not re-discovered until 1998 (Whitelock 2002). Since then, the name has been misapplied to a rare and little known taxon from isolated localities in the Maya Mountains of Belize (see Balick et al. 2000; Whitelock 2002; Janovec & Neill 2005; Hill 2004). This species, formally described in this issue (Calonje et al., page 31), differs from Zamia prasina in having an epigeous stem, leaflets that are chartaceous, gradually acuminate, with a raised longitudinal Lenta I Jin RAS 1(184N RI flat nf 7, DEF AP £ I 4 Marta] Fic. 1. A. Leaflet of Zamia muricata var. obtusi ifolia (= Ze Ji ] ) I near Alvarado, Veracruz, Mexico (MBC Accession # 20011305*A). crease, and margins that are entire or with few minute teeth at the apex, rather than a hypogeous stem and leaflets that are coriaceous, abruptly acuminate, without a raised longitudinal crease, and strongly serrulate on the upper half (Fig. 2). Since it is clear that the plant currently being called Z. prasina in horticult d recent literat not match the original description, lectotype or illustration of this species at Kew, then the logical ensuing question to ask is: What is Zamia prasina? The only other described Zamia species currently known from Belize are Z. variegata Warsz. and Z. polymorpha DW. Stev., A. Moretti & Vázq.Torres. While Zamia variegata has strongly serrulate leaflets as described in the protologue for Z. aiio. ee are pola rather than coriaceous, and are variegated with yellow flecks, a unique and horti that Bull undoubtedly would have men- tioned in his horticultural catalog. However, e coriaceous, strongly serrulate leaflets of Zamia polymorpha exactly match the description of Zamia prasina. A SYNONYMIZATION OF ZAMIA POLYMORPHA Zamia pol } tl ibed ranges from the Yucatan Peninsula in Mexico to Belize (Steven- son et al 1998). It was arene considered to be a form of Z. loddigesii Miq. with a highly variable karyo- Vee (see Mog & Olivares 1996). Stevenson et al. (1998) noted the karyotype variability as an important f this species in addition to its high level of pol; phism in leaf and leaflet morphology. This variability was ascribed to phenotypic plasticity due to dii censes in sun exposure, with plants exposed to full sun having shorter and narrower leaves with lanceolate leaflets, those growing in Mdb shade nas longer, broader leaves with ellit ] late leaflets, and those g gini displaying leaf morphologies that are intermediate between the two extremes. During a cycad research expedition to Belize in August of 2008 sponsored by Montgomery Botanical Center, an extensive survey of populations matching Stevenson et als description of Z. polymorpha with strongly serrulate leaflets uncovered some P ao with remarkably consistent vegetative morphology, and others that were highly variable. Tl ] 11 y consistent populations included 1 narrow-leafleted [| | £ pl D Hi Hi fF 46 Journal of Texas 3( laaflat af 7, a n £ kal fal 1 Mantas £ > A I £ I+ 7 nracina (helow b UL F 7 F Fic. 2. Same Michael Calonje et al. BZ08-201). plants growing in savannas closely resembling Zamia loddigesii, and also other populations growing in tropical evergreen seasonal broad-leaved forests with larger leaves and wider leaflets. In addition to these morphologically consistent populations, others were found to be highly variable, containing the two forms mentioned above as well as intermediate plants. The variability in these mixed populations appeared to be at least partially genetically determined, as plants with wide leaflets were sometimes found growing in full sun, and some narrow-leaflet plants were found growing in more shaded areas. Plants with narrow leaflets collected in an open savanna by the second author have retained their morphological characteristics despite years of growing in a shaded environment at Green Hills Botanical Collections. A survey of one highly variable population halfway between Belmopan and Belize City uncovered in- dividual plants closely matching Bull's holotype (Fig. 3) of Zamia prasina and the drawing of this species at Kew (Fig. 4), as well as the holotype for Z. polymorpha (Fig. 5), indicating that Z. prasina and Z. polymorpha are the same species, and therefore Z. prasina takes nomenclatural precedence over Z. polymorpha. In addition, there is also some circumstantial evidence suggesting that Z. prasina is a prior valid name for the species now considered Z. polymorpha. William Bull did not provide a locality or specific habitat information for where in Belize Z. prasina may have been collected, but two years after its description, Morris (1883) identified plants common on ridges and banks near Point Ycacos as belonging to this species. Morris did not provide a description or illustration of the plants he saw, but the cycad that is common in this area, visited by the second author, matches the description of Zamia polymorpha. Furthermore, this species is the most common cycad throughout Belize and would very likely be the first species encountered by collectors for William Bull. Zamia prasina (now including Z. polymorpha) appears to belong to a species complex that includes Z. loddigesii and other related species such as Z. paucijuga Wieland, and Z. spartea A. DC (Schutzman 1987). Members of this a a display high levels of morphological variation (Gonzalez-Astorga et al. 2006) and may easily hybridize (Schut 1987). Although Schuster’s (1932) taxonomic work with cycads is not highly regarded (see Stevenson & Sabato 1986; De Luca 1990, Taylor et al. 2008), his inclusion of Z. prasina as a synonym of Z. loddigesii suggests that the name at the time may have been applied to plants within this P ne oe l variability observed within populations of Zamia prasina, coupled with the high Ec RI vatisbility between p lations and the la ce a by Stevenson E “1.1 1 populations i d coriaceous, and Sabato (1986), indicate that while it i serrulate leaflets represent one highly poyini species, further research including genetic sampling £ 7, . A [RP f ote r] . " IL Dal 1 DAI Cs late Michnal falania nt X Y Fic. 3. Same-scal al. BZ08-086). across a wide range of populations will be needed to better clarify the relationship between the different forms and their placement within the iia piod d iil complex. Until species nn are PEN we cognition of Z I ith Z. polymorpha included within it ACKNOWLEDGMENTS Funding in support of the 2008 Belize expedition was g ly provided by the Association of Zoological Horticulture (AZH), Tim Gregory, and Montgomery Botanical Center (MBC). Green Hills Botanical Col- lections donated significant in-kind support by providing lodging, equipment and work infrastructure in Belize. Patrick Griffith, Chad Husby, Anders Lindstrom, Walter S. Judd, and Bart Schutzman provided comments on draft versions of the manuscript. The Belize Forest Department granted the scientific research and collecting permit (CD/60/3/08(45)) supporting research on Belizean Zamiaceae, as well as herbarium infrastructure for preservation of voucher specimens. Dennis Stevenson and Andrew Vovides provided im- ages of historical material of Zamia prasina. Nancy Korber of Fairchild Tropical Botanical Garden graciously assisted me in the search of references. 48 I I af sl Das : ID KI FEES {T F AS 1 . £7 5 n J . V. l^ f lant kad hu WR Dull f/eiaht IG. p g I Pr y (right) Dal ] DAI Fitu (late Mic I falanio ot nl D7n0. 72 y Li y i REFERENCES BaLick, M.J., M.H. Nee, AND D.E. ATH. 2000. Checklist of the vascular plants of Belize, with common names and uses. Mem. New York Bot. Gard. 85:246. The New York Botanical Garden Press, Bronx. Butt, W. 1881. Zamia prasina. Bull horticultural catalog. ÉS E De Luca, P. 1990. A historical perspective on cycads fi tl t. Mem. New York Bot. Gard. 57:1—7. DE CANDOLLE, A.P. 1868. Zamia latifolia. In: Prodromus systernati naturalis regni vegetabilis sive enumeratio con- tracta ordinum, generum specierumque plantarum huc usque cognitarum, juxta methodi naturalis normas digesta. Paris. P. 533. GONZÁLES-ASTORGA, J., A.P. VOVIDES, P. OCTAVIO-AGUILAR, D. AGUIRRE-FEY, F. NICOLALDE-MOREJÓN, AND C. IGLESIAS. 2006. Genetic diversity and structure of the cycad Zamia loddigesii Miq. (Zamiaceae): implications for evolution and con- servation. Bot. J. Linnaean Soc. 152:533-544. Hitt, K. 2004. The ean pages Zamia pana Sydney: Royal Botanic Gardens. http://plantnet.rbgsyd.nsw.gov. au/cgi-bin/cycadpg Zamia+prasina [Accessed 27 December 2008]. Janovec, J.P. AND A.K. Nei. 2003. Exploring the palms and cycads of the Maya Mountains of Belize: Reflections on MBC-sponsored Expeditions to Belize, 1999-2001. The Montgomery News 11(1):5-6. McNett, J, ER. Barrie, H.M. Bunper, V. Demoutin, D.L. Hawksworth, K. MARHOLD, D.H. NicoLson, J., PRADO, P.C. Siva, J.E. Skoc, J.H. WIERSEMA, AND N.J. TuRLAND, eds. 2006. International code of botanical nomenclature (Vienna Code) AE YORE PRA TAR d Fi. 5. $ | pari f Zamia polymorpha holotype (right) | bet Imo; d Beli y (left, Michael Calonje et. al BZ08-085). Image of Zamia polymorpha holotyt | with permission from The C.V. Starr Virtual Herbarium of the New York Botanical adopted by the Seventeenth International Botanical Congress Vienna, Austria, July 2005. Gantner Verlag, Ruggell, Liechtenstein. Miquel, FAW. 1843. De cycadeis loddigesianis epistola. Tijdschrift Nat. Ges. Phys. 10(1):68-74. Moue, FAW. 1847. Collectanea nova ad Cycadearum cognitionem. Linnaea 19:411—430. Miquel, FA.W. 1849. Epicrisis systematis cycadearum. Tijdschr. Wis-Natuurk. Wetensch. Eerste Kl. Kon. Ned. Inst. Wetensch. 2:280-302. Morass, D. 1883. The colony of British Honduras, its resources and prospects; with particular reference to its indigenous plants and economic productions. Edward Stanford, 55, Charing Cross, London. SCHUSTER, J. 1932. Cycadaceae. In A. Engler, ed. Das Pflanzenreich 99(4):1-168. SCHUTZMAN, B. 1987. Mesoamerican Zamias. Fairchild Trop. Gard. Bull. 42:16-19. STEVENSON, D.W, A. Moretti, AND L. Gaupio. 1998. A new species of Zamia (Zamiaceae) from Belize and the Yucatan Peninsula of Mexico. Delpinoa 37-38:3-8. STEVENSON, D.W. AND S. SABATO. 1986. Typification of names in Zamia L. and Aulacophyllum Regel (Zamiaceae). Taxon 35:134-144. TAYLOR, A.S. J.L. HAYNES, AND G. HoLzMAN. 2008. Taxonomical, nomenclatural and biogeographical revelations in the Zamia skinneri complex of Central America (Cycadales: Zamiaceae). Bot. J. Linnaean Soc. 158:399-429. THIsELTON-Dyer, WT. 1884. Cycadaceae. In: Helmsley, W.B. Biologia Centrali-Americana, Botany 3(16):190-195. Vovipes, A.P. AND M. Otivares. 1996. Karyotype polymorphism in the cycad Zamia loddigesii (Zamiaceae) of the Yucatan Peninsula, Mexico. Bot. J. Linnaean Soc. 120:77-83. WhrreLock, L.W. 2002. The cycads. Timber Press, Portland, Oregon. BOOK REVIEWS Roy C. Dicks (ed.). 2008. Rhapsody in Green: the Garden Wit and Wisdom of Beverley Nichols. (ISBN 978-03812-9485, hbk.). Timber Press, 133 S.W. Second Ave., Suite 450, Portland, Oregon 97204-3527, U.S.A. (Orders: iip NM com, 1-800-827-5622, 1-503-227-3070 d $17.95, 135 pp., 6" x 7 1/2". tinha] dea] AL Nichols, most 7? n 1 ES 5 zT This little ! the editor has ics cue Readers na a pleasure in his metaphors > “with od white petticoats inde crimson nisu and his lil d gther dan reminds him the weekly washing"). F ] , sage advice, and ing ot i —Joann Rares faa Christian University Library (retired), Box 298400, Fort Worth, esas 76129, U.S.A. BARBARA L. WILSON, RICHARD BRAINERD, DANNA LYTJEN, BRUCE NEWHOUSE, AND NICK OTTING OF THE CAREX WORKING Group. 2008. Field Guide to the Sedges of the Pacific Northwest. (ISBN 9780870711978, pbk.). Oregon State University Press, 121 The Valley Library, Corvallis, Oregon 97331, U.S.A. (Orders: http:// i bed ee 1-800-426- iid $35.00, 431 pp., 6" x 9". E ; i al d guide for al! ee Species i E 2d varieties in the genus HS with d differences 1 e (y ] XA 4 1 D. O hnical charact These differences are uddressed. for each species to btecilitate identification. nd restoration, har x = pa QAM mall, t The Field Guide is divided into ne main Bpurs M information about sedge ecology, habitat management a dG) To accounts. The n portion of the book is the species ethnobotanical uses, and Tue 2 accounts, iption, a distributi mostly color photographs with a few botanical line divides In Addition th llecting tips, a BIS. and bibli hical The Field Guide provi d essential resource for botanists, land nad restoration ne Su le enthusiasts. Ánd, as ie genus Carex becomes increasingly important amongst landscapers, nurseries, and gardeners, the guide will erve as a handy tool for choosing Northwest natives d the PA A very useful book with a binding that Mn i easy opening. MPs for public libraries, schools, and g ut the Authors: The Carex Worki (Bart L. Wilson, Richard Brainerd, Danna Lytjen, B New! and Nick TE is oun up of Oregon botanists di bi sedges and other E identify plant groups. The 1993 to map the distribution of Or d n's OMS and pu La in 2002. That effort SUE in the iuge of Atlas of Mic Carex di In addition to g I graphing lg: CWG I I lants of the western UG) L Jennings, bañan 1 1 CWG came together in Botante Research institute of Texas, Fort Worth, Texas 76102-4025, U.S.A. J. Bot. Res. Inst. Texas 3(1): 50. 2009 FESTUCA ALOHA AND F. MOLOKAIENSIS (POACEAE: LOLIINAE), TWO NEW SPECIES FROM HAWATT Pilar Catalán Robert J. Soreng and Paul M. Peterson Departamento de Agricultura (Botanica) Department of Bota T Satie Superior de el esca National Museum of Naural “a ersidad de Zaragoz mithsonian Instit Ctra. ree um 1, 22071 freu SPAIN Washington, DC oa USA, pcatalangunizar.es sorengr@si.edu, peterson@si.edu ABSTRACT Festuca aloha, sp. nov. of Kaua'i and F. molokaiensis, sp. nov. from Molokz'i are described and illustrated. The two species are en- demic to the Hawaiian archipelago and grow on steep mountain slopes and volcanic cliffs with other native mesic flora. A key to the species of Festuca found in the Hawaiian Islands is given. Key Words: Festuca, new species, Hawaii, Loliinae, key RESUMEN Se describen e ilustran Festuca aloha, sp. nov. de Kaua'i y F. molokaiensis, sp. nov. de Moloka'i. Las d peci démicas del sataj Torr ti 31,1 o lad = J; a f1 A R + E © 4 E mésicos € H 1 da 3 if L da] i Jo Eu Lakit 1 icl Ap] heel r L K e Is E While revising specimens of Festuca deposited at the United States National Herbarium (US) the first author found two specimens of fine-leaved individuals from Kaua'i that were identified as F. rubra L. However, these individuals exhibit several characters that separate them from holarctic red fescues of the F. rubra complex, including; leaf blades flat with involute margins, sheaths open, relatively long ligules, and ovaries with densely hairy apices. After reviewing specimens of Festuca from Molokai'i deposited at the Bishop Museum (BISH) we noticed an additional taxon. Previously, the only species of Festuca and relatives re- corded from Hawai'i and the Pacific include the endemic, F. hawaiiensis Hitchc. (Hitchcock 1922), a robust broad-leaved species placed in F. subg. Drymanthele V.1. Krecz. & Bobrov sect. Banksia E.B. Alexeev (Alexeev 1980), and three Eurasian species: the fine-leaved F. rubra L. (F. subg. Festuca sect. Aulaxyper Dumort.); and the broad-leaved F. arundinacea Schreb. and F. pratensis Huds. [F. subg. Schedonorus (P. Beauv.) Peterm. sect. Schedonorus (P. Beauv.) Endl]. The taxonomy of Festuca s.l., the largest genus of monophyletic subtribe Loliinae Dumort., is i tate of flux. Much additional research is needed to properly align the taxa, althoug preliminary data from DNA sequencing (Catalán et al. 2007; Inda et al. 2008) suggest that two major clades may exist. As elements of the "broad-leaved" clade, the latter two species have been classified within the separate genus Schedonorus P. Beauv., as Schedonorus arundinaceus (Schreb.) Dumort and S. pratensis (Huds.) P. Beauv. (Soreng et al. 2003; Snow 2008). In addition, Festuca subg. Drymanthele is sometimes recognized as the genus Drymochloa Holub. The “fine-leaved” clade of fescues, including F. rubra, are placed in Festuca s.s. (Soreng et al. 2003). However, results are still equivocal concerning the resolution of the division at the base of the two major clades. Specimens ined from the Bishop Museum (BISH), National Tropical Botanic Garden (PTBG), Missouri Botanical Garden (MO), and the United States National Herbarium (US). Festuca aloha Catalán, Soreng & PM. Peterson, sp. nov. (Figs. 1A-J, 2). Tye: U.S.A. Hawar: Kava‘, Hanalei district, Kalalau Rim, Kalala, side bendi and W iun "i d neo. lookout, isolated hanging side-valley of iind diverse mesic forest Peucedanum, Poa mannii, Hedyotis, 1 hia el Melicope surrounded by F pallida, NA dtm. Dubautia. Dodd: POSER Li ta, Lobelia niil is, Lipidium, Neti Hibiscus kokio, Eragrostis, a Myrsine, Acacia, and Psychotris, 790 n m, 13 Mar 1992, K.R. Wood 1701 & S. Verba (HoLOTYPE: PTBG-17679; Iso TYPE: US-32522 J. Bot. Res. Inst. Texas 3(1): 51 — 58. 2009 | | El Dat ¿0UMal OF a N ; ; | "i at bg th he Le = ELE Ae a = > SS == ze LEA AS EE A N X M tA ac = f m An Mh LO SN X N 1 M N ventral view. I. Lodicules, mature o lea, dorsal view. G. Palea with ovary, ventral view. H vary, |l | t of fil ts. J. Lodicule. Fi BISH-728771). K. Spikelet. Fic. 1. Illustration of Festuca aloha Catalán, Soreng & P.M. Peterson (isotype, US-3252239). A. Habit. B. with rachilla. E. Lemma, ventral view. F. Pa Sheath, ligule, and blade. C. Spikelet. D. Floret nC P.M. Pet 0 Fatal J at Catalán et al T : fEncs £ "T. m E A Festuca hawaiiensis Hitchc. paniculis 8-11 cm longis, angustibus, ramis erectis lentibus; pedicellis (11.5—3.5(-5) mm longis; lemmatibus 5.8-8 mm longis; laminis 1.5-2.4 mm longis, recedit. Plants perennial, cespitose, with extravaginal innovations; cataphylls conspicuous, brownish. Culms 45-70 cm tall, erect, slender (ca. 1 mm diam.), nodes 23, smooth and glabrous. Leaf sheath margins fused for 1-2 mm at base, overlapping more than !^ the length below, sparsely villose abaxially when young, becoming glabrous at maturity, purple-brownish at the base, greenish above, becoming fibrous in age at base; auricles absent; collars glabrous; ligules 1-1.5 mm long, scarious, glabrous, brownish, apex obtuse, dentate-erose; leaf blades of vegetative shoots 22-33 cm x (12)1.5-1.7 mm, erect to pendant, flat with involute margins, glabrous and smooth abaxially, hirsute along protruding ribs adaxially, the hairs up to 0.1 mm long, cross section (Fig. 2) about 1 mm bearing 13-16(-20) larger vascular traces or veins, each with prominent ribs adaxially alternating with smaller ribs on the secondary and tertiary veins, scleren- chyma strands forming trabeculae or girders on most veins; leaf blades of culms about 22 cm x 2.4 mm, erect to pendant. Panicles 8-11 x 2.5—4 cm, erect, axis and peduncle smooth; loosely contracted with 34-66 spikelets, spikelets mostly at branch tips; panicle branches 2(3) per node, basal branches 4-6 cm and ca. 1/2 the length of panicle, angled, erect or ascending, smooth proximally to scabrous, angled dis- tally. Spikelets 8.2-13 mm x 3.5-6.0 mm, 4-6-flowered, broadly lanceolate, pale green, imes tinged with purple; pedicels (1-21.5-3.5(-5) mm long; rachilla internodes 0.8-1.2 mm long, scabrous; glumes lanceolate, green, margins very narrowly scarious («0.5 mm), apex acute; lower glumes (3.6-)3.9-4.2(-4.5) mm long, 1-veined; upper glumes 5.2—5.5(-6.0) mm long, 3-veined; lemmas 5.8-8.0 mm long, 5-veined, glabrous below, scabrous near the apex, green or tinged with purple at the apex, mucronate to awned, the mucro or awn 0.5-1.1 mm long, scabrous; callus rounded, obliquely angled (to 45°), smooth to dorsally minutely scabrous at the apex; paleas longer or shorter than lemma, scabrous on and between keels; stamens 3, anthers 3.1-4.2 mm long; ovary densely hairy on upper 1/3. Caryopsis not seen. Additional material examined. U.S.A. Hawaiʻi. Kaua‘i: Hanalei district, Kalalau Rim, N of Kahuamaa flat, 990-1020 m, 3 Mar 1991, K.R. Wood 631, M. Query, S. Montgomery (PTBG-9498); north aspect, 1060—1190 m, 5 Apr 1991, K.R. Wood 901 & M. Query (PTBG-13876); Kalalau side below and W of the first una lookout, 13 Mar a K.R. Wood 1704 & S. Perlman (PTBG-17678, topotype?; Honopu Rim, undisturbed cliffs, r ith old i seen with Panicum lineale, 2800 ft, 6 Nov 1993, K.R. Wood 2832 (PTBG-29690; MO- pened Kalalau valley, Gumping on N facing al cliffs, seen with Poa mannii, Panicum lineale, Eragrostis vari- abilis, Carex Mn ca. 3000 ft, 17 Jun 1994, K.R. Wood 3245 (PTBG-20438); Pohakuao, hanging valley between Kalalau and Hanakoa, and Kaaalahine Ridge, 2000 ft, 4 Jan 1992, K.R. Wood 1761, S. Perlman & J. Lau (BISH-621133, PTBG-12875, US-3250261); Waimea D. with Panicum lineale pn P. FPE 2000 ft, 30 Mar 1993, K.R. Wood 2470, & S. Perlman (PTBG-18788), K.R. Wood 2471 & S. Perlman (PTBG-18779); A alley, N facing slopes above stream, 0.5 mi along trail, growing with Wilkesia gymoxiphium and Lipidium serra, 3300—3500 ft, 18 May 1994, K.R. Wood 3193, S. Perlman & M. Hartley (PTBG-15270); Upper Kawaiiki: Kaluahaulu Ridge PP dh W of ] i iiis into Hager Acacia n Metrosideros iss morpha mixed mesic for- est with D 11 1 Poa ie eee S nd Cl 1149 m, 14 Nov 1996, K.R. Wood 7605 (PTBG-42661). Comments.—Festuca aloha has panicles 8-11 cm long, narrow, branches erect and ascending; lemmas 5.8-8 mm long; and leaf blades 1.5-2.4 mm wide; in contrast F. hawaiiensis has panicles 30-40 cm long, widely open, branches patent or patent-erect; lemmas 9 mm long; and leaf blades 2-3 mm wide. The illustrator Alice Tangerini noticed that all florets with mature anthers of F. aloha contained small, undeveloped ovaries (Fig. 1H). Mature ovaries with well-developed styles were found only in florets that had already shed their anthers (filaments were still present) [Fig. 11]. Therefore, F. aloha is apparently protandrous. Etymology.—The specific epithet F. aloha name derives from the Kalalau mountains of Kauai. Conservation status —Festuca aloha has been found in at least six different localities of Kauaii, covering a distribution area of approximately 700 km?. The number of individuals varies among populations but usually there are less than 1000 individuals per population. Based on these preliminary data, the new spe- cies clearly falls within Vulnerable (VU) category as defined by the IUCN (2001). The major threats to in- dividuals of F. aloha are the allochthonous plants, such as: Cyperus meyenianus Kunth, Erigeron sp., Kalan- choe pinnata (Lam.) Pers., Lantana camara L., Melia sp., Passiflora mollissima (Kunth) L.H. Bailey, Rubus Ec? l £hlad na £ tati L £ Fact Inha Catalán Sorena & PM DP 1110 2353320 Y ). Scale bar = 0.1 mm Li 3 A rosifolius Sm. ex Baker, Setaria gracilis Kunth, Triumfetta sp.; and goats and pigs (K.R. Wood, on herbarium label). Festuca molokaiensis Soreng, PM. Peterson & Catalán, sp. nov. (Fig. 1K). tw U.S.A. Hawari. Moroka‘: Kupaia Gulch, on steep slopes in mesic forest, occasional, associated vegetation: Metrosideros, Coprosma, Leptecophylla, Dodonaea, Alyxia, Viola, Melicope, Myrsine, Diospyros, Dryopteris, Phyllanthus, Luzula, Selaginella, A Carex, 21.119564 N 156.936999 W, 880 m, 3 Apr 2007, H.L. Oppenheimer H40704, S. Perlman & N. Tangalin (noLorvre: BISH-7287 li b tib i bris, arista lon- A Festuca aloha Catalán, Soreng & P.M. Peterson, p giori 1.5-2.8 mm, antheris brevioribus 2-3 mm, shins inferis (s. 0-)5.2- 5. sc 5.7) mm, EET superis (6.0-)6.2-6.8(-7.0) mm, oe Plants perennial, cespitose, with extravaginal innovations; cataphylls conspicuous, brownish. Culms up to 60-70 cm tall, scabrous below panicle. Leaf sheaths with margins fused for 1-2 mm at base, overlapping more than Y the length below, purple-brownish at the base, greenish above, becoming fibrous at base in age; auricles absent; collars glabrous; ligules 1.5—2.5 mm long, scarious, glabrous, brownish, apex obtuse, dentate-erose; leaf blades of vegetative shoots approximately 34 cm x 2 mm, more or less pendant, flat with involute margins, glabrous and smooth abaxially, hirsute along protruding ribs adaxially and on mar- gins and apex, dull green, cross section luu to that of F. aloha, about 2 mm wide bearing 25 larger vas- cular traces or veins, each with prominent ril ially alternating with smaller ribs on the secondary and tertiary veins, bulliform cells between adaxial ribs, sclerenchyma strands present forming trabeculae or girders on most veins, sclerenchyma strands with thin cell walls (adaxially) and thick cell walls (abaxially); leaf blades of culms about 21 cm x 2 mm, flat with involute margins, densely scabrous adaxially. Panicles 8.5-13 x 4 cm, loosely contracted with approximately 75 spikelets, spikelets located at the ends of the branches, 3(4) branches on basal node, 2(3) branches in other nodes, axis and branches strongly scabrous; basal branches as long as 7 cm, more than 1/2 the length of the panicle, angled, suberect to ascending. Spikelets 7-9 x 3-4 mm, 5-6-flowered, broadly lanceolate, pale green; pedicels Q— 4(-9) mm long; ra- chilla internodes scabrous; glumes lanceolate, green, margins very narrow, scarious, apex acute; lower glumes (5.025.2—5.5(5.7) mm long, 1-veined; upper glumes (6.0—)6.2-6.8(-7.0) mm long, 3-veined; lemmas (5.0-)5.5—6.0 mm long, 5-veined, scabrous dorsally and near apex, green, awn 1.5-2.5 mm long, scabrous; callus rounded, obliquely angled (to 459), scabrous; paleas slightly shorter than lemma, bidentate at apex, keels scabrous, between keels scabrous; stamens 3, anthers 2.8-3.0 mm long; ovary densely hairy on upper 1/3. Caryopsis not seen. Comments.—Festuca molokaiensis differs from F. aloha by having more delicate and thinner, more nu- merous and lax branches; scabrous panicles and spikelets; lemmas scabrous throughout, awns longer 1.5-2.8 mm; anthers shorter 2-3 mm; lower glumes (5.0—)5.2—5.5(-5.7) mm; and upper glumes (6.0—6.2— 6.87.0) mm. In comparison to F. aloha, culms are scabrous on the nodes; panicles 8.5-13 cm long, more delicate, axis and peduncles scabrous, branches densely scabrous, angled; spikelets more delicate; rachilla more slender; lower glumes (5.0-)5.2—5.5(-5.7) mm long; upper glumes (6.0—)6.2—6.8(-7.0) mm long; lem- mas 5-6 mm long, evenly scabrous across the back from the base to the apex, 3-veined or infrequently faintly 5-veined (lateral veins obsolete or obscure), awned, the awn 1.5-2.8 mm long; and anthers 2-3 mm long. Catalán et al ; T Lf Etymology.—The specific epithet of F. molokaiensis derives from the island of Moloka'i. Conservation status.—Festuca molokaiensis is only known from the type locality where it has been re- corded as occasional (Oppenheimer, on herbarium label). Due to the scarcity of available data on the po- tential distribution, demography, and population status of F. molokaiensis, we consider to be a Data Deficient (DD) taxon (IUCN 2001). TAXONOMY AND PHYLOGENY Festuca aloha and F. molokaiensi logically similar to the Hawaiian endemic, F. hawaiiensis in ligule size and shape, ovary hairiness, and leal hee y (Hitchcock 1922). However, F. hawaiiensis is a more robust plant up to 150 cm tall with longer (30-40 cm) and wider panicles that are open with 3—5 spreading and drooping branches at the nodes, and larger lemmas up to 9 mm long. Festuca hawaiiensis grows at higher elevations, at ca. 2000 m, in rich soil on moist wooded hills on the island of Hawaii. Festuca aloha and F. molokaiensis resemble the western North American, F. californica ssp. hitchcockiana (E.B. Alexeev) Darbysh. [Darbyshire & Pavlick 2007] in leaf blade cross section, panicle features, and hairy ovary apices. However, F. californica ssp. hitchcockiana differs from the new species by having a more robust and cespitose habit, possession of a continuous sclerenchyma layer on the abaxial side of the leaf blade, a partially hairy collar, and longer ligules with ciliate apices. The new Festuca taxa also show similarities with Macaronesian volcanic cliff dwellers: F. agustinii Lind- ing., F. jubata Lowe, and F. francoi Fern.Prieto, e Aguiar, E. Días & M.I. Gut. (Saint-Yves 1922; Fernández Prieto et al. 2008) because all sh d flat to inrolled (or conduplicate) leaf-blades with some complete sclerenchyma cele However, the latter species all differ from the new species by having glabrous ovary apices and short (« 1.5 mm) ligules. We analyzed the ITS and trnL-F sequences of two samples of Festuca aloha from different populations from Kauai [F. aloha 1: Kalalau, US-3252239 (isotype), GenBank GQ162205 (ITS) and GQ162208 (trnLF) accessions; F. aloha 2: Pohakuao, US-3250261, GenBank GQ162206 (ITS) and GQ162209 (trnLF) accessions] and one sample of F. molokaiensis from the only known population from Molokai [F. molokaiensis 1: Kupai Gulch, BISH-728771 (holotype), GenBank GQ162207 (ITS) and GQ162210 (trnLF) accessions]. We then conducted heuristic parsimony analyses for the combined ITS and trnL-F dataset (10,000 random entry trees, TBR, mulpars off, saving no more than 10 trees of length «10 per replicate; 1000 bootstrap replicates with the same parameters as in the original search) using these three samples and combined them with our previous data for Festuca on a worldwide level (Catalán et al. 2007; Inda et al. 2008). All three samples grouped together within the broad-leaved Festuca clade (Fig.3), nested within the well supported clade Subulatae + Leucopoa p.p. [92% bootstrap(BS)], which were separated from F. califor- nica, F. rubra, and the M ian Festuca sect. Aulaxyper s.l. species. Festuca aloha was monophyletic (9996 BS) and sister to F. molokaiensis (98% BS). The two F. aloha specimens from dn ode on Kauai (F. aloha 1 & F. aloha 2) had similar sequences (9996 BS; 1 trnL-F nucleotid titution difference); however, the sample of F. molokaiensis from Molokai showed several nucleotide diff in both the ITS (five substitutions) and trnL-F (two substitu- tions) regions with respect to F. aloha. These molecular differences also support the taxonomic separation of the two species. Despite the limited infraspecific sampling, the ITS variation found between F. aloha and F. molokaiensis, from the same Hawaiian archipelago, is remarkable as these species show more nucleotide differentiation than that found between other closely related species [e.g. the Pyrenean F. eskia Ramond ex DC., and F. gautieri (Hack.) K. Richt. with only two ITS nucleotide substitutions], and between species from different Macaronesian archipelagos (e.g., Madeiran F. jubata and Azorean F. petraea Guthn. ex Seub. with three ITS nucleotide substitutions). Unfortunately, F. hawaiiensis could not be included in the molecular analysis because it is extremely rare and we have no material other than the type. The closest relative of F. aloha and F. molokaiensis in our analyses (Fig. 3) was the eastern Asian F. parvigluma Steud. (8896 BS), followed by the ] North American F. subulata Trin. (7796 BS), both belonging to F. subg. Subulatae, and then by the Siberian-North American F. altaica Trin. (9296 BS), placed Festuca luciarum Festuca multinodis à Vulpia membranacea Vulpia unilateralis Cienopsis mE Vulpia ci u Fesucrm Festuca violac Vulpía octo Hora gautieri Festuca quadriflora Festuca rieulmis estuca thurbert Festuca arundinacea Festuca atlantigena 1 Festuca leo urnexiana Festuca gincescens Festuca mai Festuca fen Festuca bačtica foe einig pad Tes Festuca aloha2 Festuca moloksiensis s parvigrum [| l af pl JUuiriidi OF Drymanthele s.s. gr. Exaratae p.p./ American Vulpia Leucopoa p.p. + pee S.S. + Aulaxyper s.l./ Vulpia 2x Festuca/ Wangenheimia Micropyrum Narduroides Loretia gr. + Ctenopsis Psilurus/ Vulpia (4x-6x) Exaratae p.p. American II American ] Eskia- Dimorpha gr. Leucopoa p.p. Neozeylandic I./ American Schedonorus gr. Subulbosae Subulatae + Asian-American Leucopoa s.s. Lojaconoa Castellia pa nig Cynosurinae/ Dd Fine-leaved Loliinae Broad-leaved Loliinae conc oly Fic. 3. Strict t ( 7 r. ini 2 fatali , Soreng & PM. Peterson and retention index 0.762) I Catalán et al T iac nf Factura fi T sg T within F. subg. Leucopoa sect. Breviaristatae. In a separate ITS strict consensus tree the F. aloha/F. molohai- ensis clade fell within the broad-leaved clade, and was closely related to the Asian F. parvigluma and F. modesta Steud. [sometimes treated as Drymochloa modesta (Nees ex Steud.) Holub]. In contrast to the ITS data, a trnL-F strict consensus tree placed F. aloha/F. molokaiensis in a polytomy at the base of the fine-leaved clade, among fine-leaved taxa and taxa intermediate between fine- and broad-leaved forms of Festuca. Given the different resolutions between separate nuclear and plastid DNA sequence topologies for our species, it is possible that F. aloha and F. molokaiensis are of allopolyploid origin. On morphological grounds F. aloha, F. molohaiensis, and F. hawaiiensis are all part of the same complex, representing a common pattern of morphological speciation in upland habitats of different islands along the Hawaiian archipelago. KEY TO THE SPECIES OF FESTUCA IN HAWAI 1. Lower leaf blades with auricles; leaf blades usually 5-12 mm wide, flat. 2. Auricles ciliate on margin; lemmas awned (awn 2—3 mm), more or less scabrous distally Festuca arundinacea 2. Auricles not ciliate on margin, lemmas muticous, smooth or only sparsely scabrous distally — baba pratensis 1. All le af bl f blades 0.3- 5m m wide, i involute to flat but when flat 3. Ligules 0.1-0.5 mm long; c ovary apices glabrous; leaf blades conduplicate; leaf sheaths usually closed below (open « Y their length); leaf blades of vegetative shoots in cross section without sclerenchyma girders. estuca rubra . Ligules 1-2.5 mm long; ovary apices densely hairy; leaf blades uu d Tel margina; m Aa generally open Y their length [they are fused only for 1-2 mm att fi in cross d with M dps Sues w 4. Panicles 30-40 cm long, open, t | ling or drooping; leaf blades 3-5 mm wide ; known only dd M. wall Festuca hawaiiensis 4. Panicles 3.5-25 cm long, usually narrow, the branches patent, erect or ascending; leaf blades 0.3-2.4 mm wide. 5, qui scabrous only in the upper third, i liate veins distinct, awns 0.5-1 mm long; anthers 3.1-42 mm long; ligules 1-1.5 mm long; lower glumes (3.6-)3.9—4.2(-4.5) mm long; upper glumes eh 5.5(- is mm long; aca only from pale! i Festuca aloha veins absent or obscure, awns 1.5-2.8 mm; ee 2-3 mm long: ligules 1.5— 25m mm long; lower glumes (5.0-)5.2—5.5(-5.7) mm long; RES gium (6.0-)6.2-6.8(-7.0) mm long; known only from Moloka'i Festuca end: ACKNOWLEDGMENTS This work has been subsidized by the Spanish Ministry of Science and Technology Grant Project CGL2006- 00319/BOS and Sabbatical Research Stay Grant PR2008-0020 to Pilar Catalán. We thank Stephen Darbyshire and Neil Snow for their critical review of an earlier version of the manuscript; Napua Harbottle, Tim Flynn, and Gerrit Davidse, for facilitating for us the study of Hawaiian Festuca materials deposited at BISH, PTBG and MO, respectively; Hank Oppenheimer, Neil Snow, and Cliff Morden for valuable information on the ecology and distribution of F. aloha and F. molokaiensis in Kauai'i and Molokai'i; Sasha Savytskyy, Dai Tsuchiya, and Juan Viruel for helping us with figure preparation; Alain Touwaide for correcting the Latin diagnoses; and Alice R. Tangerini for providing the illustration. REFERENCES ALEXEEV, E.B. 1980. Novye podrody i sekzii ovsjaniz (Festuca L.) severnoj ameriki | meksiki. (Festuca L. subgenera et sectiones novae ex America Boreali et Mexica). Novosti Sist. Vyssh. Rast. 17:42—53.[in Russian]. CATALÁN, P, P. TORRECILLA, J.A. Lopez-Ropricuez, J. MULLER, AND C.A. Stace. 2007. A systematic approach to subtribe Loliinae (Poaceae: Pooideae) based on phylogenetic evidence. Aliso 23:380—405. DaRBYSHIRE, S.J. AND L.E. PAvuck. 2007. 14.01 Festuca L. In: Barkworth, M.E., K.M. Capels, S. Long, L.K. Anderton, and M.B. Piep, eds. Magnoliophyta: Commelinidae (in part): Poaceae, part 1. Flora of North America north of Mexico, volume 24:389—444. Oxford University Press, New York. FERNÁNDEZ PRIETO, J.A., C. AGUIAR, E. Días, AND M.I. Gutiérrez ViLLARÍAS. 2008. On the identity of Festuca jubata Lowe L D H In l.l PA fT, 58 Journal of t| Texas 3( (Poaceae) and the description of a new Festuca species in the Azores Islands. Bot. J. Linnean Soc. 157:493—499. HitcHcock, A.S. 1922. The grasses of Hawaii. Mem. Bernice Pauahi Bishop Mus. 8:100-230. INDA, L.A., J.G. SEGARRA-MORAGUES, J. MULLER, PM. Mine AND P. CATALÁN. e Pains iS biogeography of the temperate Loliinae (Poaceae, Pooideae) grasses in t tl h Molec. Phylogen. Evol. 46:932—957. IUCN. 2001. IUCN Red List Categories and Criteria: Version 3.1. IUCN Species Survival Commission. IUCN, Gland, Switzerland and Cambridge, UK. Saint-Yves, A. 1922. Les Festuca (subgen. Eu-Festuca) de l'Afrique du Nord et de les Isles Atlantiques. Candollea Snow, N. 2008. Notes on grasses (Poaceae) in Hawaii. Bishop Mus. Occas. Pap. 100:38-43. Soreng, RJ, PM. Peterson, G. Davipse, E.J. Jubziewicz, FO. ZULOAGA, AND O. Morrone 2003. Catalogue of the New World grasses (Poaceae): IV Subfamily Pooideae. Contr. U.S. Natl. Herb. 48:1—730. A NEW GENUS, MEXOTIS, FOR FIVE MEXICAN SPECIES OF HEDYOTIDEAE (RUBIACEAE) Edward E. Terrell Harold Robinson Research diy Department of Botany, MRC 166 Department of Botany, MRC 166, National Museum of Natural History National Museum of Natural History 20. PO. Box 37012 Smithsonian Institution Smithsonian Institution Washington, DC 20013-7012, U.S.A. Washington, DC 20013-7012, U.S.A. terr60@ msn.com robinsoh@si.edu ABSTRACT In furtl icti f th t of the g Hedyotis in Mexico, four | pecies, H. galeottii, H. kingii, H. latifolia, and H. terrel- lii, are m inanew eee Mexotis. A new species, M. lorencei, i gni d described. Morphological and distributional dat RESUMEN Enl icción adicional 1 pto del gé Hedyotis en México, lel peci idas, H. galeottii, H. kingii, H. latifolia 9 al + ^ ie. M. lorencei r i i y H. terrellii, y de distribución y se citan las PA representativas. In early 20* century literature on the tribe Hedyotideae of the Rubiaceae the name Hedyotis L. was applied to many members of the tribe in both hemispheres. The main recent proponent of the name Hedyotis was Fosberg (1943) in his treatment of 24 Polynesian species of Hedyotis. He stated that he was adopting Hedyo- tis in the broadest sense because “the criteria to segregate the species involve not very fundamental differ- ences in structure" (Terrell 1996:2—5). Under Hedyotis he recognized five subgenera and commented that many botanists would consider these as genera. Despite his broad view of genera, he favored extreme split- ting for subdivisions of species, e.g., he described 39 M forms of Dou acuminata a & Schltdl.) Steud.(now Kadua acuminata Cham. & Schltdl.). He als idered seed cl because “they vary so much that no two are alike in a single capsule". In subsequent papers rockers continued to use the name Hedyotis. A recent study of Asian Hedyotis shows that many species related to the type, Hedyotis fruticosa L., of Sri Lanka and India, have unique capsule and seed morphology (Terrell & Robinson 2003). These species, in the genus Hedyotis, so far as known, are restricted to Asia and the western Pacific. Excluded from this concept are the mid-Pacific elements more recently treated as Kadua Cham. & Schltdl. (Terrell et al. 2005) and the New World species that have currently been placed in Hedyotis. Many of the American Hedyotideae that were once placed in Hedyotis are now recognized as separate genera on the basis of structural features, particularly seeds and habit. Some of the segregate genera were described as new, including Carterella, with one Mexican species (Terrell 1987), Stenaria with five U.S. and Mexican species (Terrell 2001a), and Stenotis with seven Baja California and one Arizona species (Terrell 2001b). In addition, Houstonia and North American species of Oldenlandia each have five Mexican or Mexican-U.S. species (Terrell 1996; Terrell & Robinson 2006). In recent years further study has included about 12 additional hedyotoid genera. None of these genera were found to have seeds and other characters similar to those of the five species studied here. We provide a further needed reclassification for Mexican Hedyotis species. A new genus, Mexotis, is described, including Hedyotis galeottii, H. kingii, H. latifolia, H. terrellii, and a newly described species, M. lorencei. The removal of Western Hemisphere Hedyotideae from Hedyotis is supported by two recent studies J. Bot. Res. Inst. Texas 3(1): 59 — 70. 2009 60 Journal of tani i Texas 3( using DNA sequencing. In the DNA study of Andersson and Rova (1999), such species once placed in Hedyotis are not closely related to the Asian and eastern Pacific Hedyotis, the latter represented by H. mac- rostegia Stapf or the Central Pacific H. hillebrandii (Fosberg) Wagner & Herbst. (now Kadua axillaris (Wahra) Wagner & Lorence). In a more complete survey, Kárehed et al. (2008) show the division more clearly, and they call for the narrower interpretation of Hedyotis. Thus, the morphological and DNA evidence agree in the need to exclude American species of Hedyotideae from the genus Hedyotis. The Kárehed et al. study subdivides the Hedyotidae (as Spermacoceae) into many discrete clades that are not only distinct from Hedyotis, but distinct from each other. Two such groups are the Arcytophyllum/Houstonia/Stenaria Clade, to which many of the previously studied American Hedyotideae listed above belong and the Manettia/Bou- vardia branch, the latter being close to the Spermacoce Clade. No DNA sequences are available for any of the species placed here in Mexotis, and conclusions regarding which group is closest to Mexotis are based on structural features. A paper by Church (2003) contributed an important molecular study on Houstonia and related genera, but did not include any of the five Mexican species that we investigate here. Mexotís includes four species of large herbs or subshrubs with woody stems, long internodes, large elliptic or ovate leaves, long petioles, filiform pedicels, and generally flattened seeds with central punctiform hila. These species are native to Oaxaca and Veracruz. A fifth smaller species, M. latifolia, has herbaceous or woody-based stems and smaller leaves. It appears closely related to the four more robust species and ranges from Veracruz to Guatemala. This species is the only one of the five with a known chromosome number, n - 17, a number not known to occur in any other species of Hedyotideae. The five species (Table 1; Figs. 1—5) have significant reproductive and vegetative characters. The seed types fall into two groups, the kingii group and the galeottii group. The kingii group is represented by M. kingii, the type species, which has somewhat thicker seeds, sometimes polygonal in outline and being often obtusely angulate or obtusely ridged. The compression is moderate or slight. These exact characters have not appeared in other Hedyotideae species. Mexotis terrellii is also in this group because of its apparent similarity to M. kingii; however, unfortunately the collections have immature seeds. The second seed type group includes M. galeottii, M. latifolia, and M. lorencei. These species have thin, flat or strongly compressed, concavo-convex seeds. In M. galeottii and M. latifolia the seeds are slightly to moderately concavo-convex. In M. lorencei the seeds become strongly concavo-convex and may appear cupulate. In M. galeottii and M. latifolia the seed areoles (cells) are radially elongated, and M. galeottii is unique in having narrow, usually partial, sometimes complete, wings. These same two characters also occur in the genus Bouvardia, which, however, carries the winged character to extremes, as many Bouvardia species have broad, conspicuous wings (and large capsules) (Blackwell (1968). Mexotis Terrell & H. Rob., gen. nov. Tee species: Mexotis kingii (Terrell) Terrell & H. Rob. Plantae UNS saepe Luis b i a herbaceae vel suffrutescentes. Caules ad 1.5 m alti; laminae foliorum ad 11 cm longae vel ellipticae; planatae aut leniter compressae interdum alatae, hilis centralis punctiformibus, ovatae He d PA Dal 1 1: 1 Perennial herbs or subshrubs typically with thick, woody, erect to decumbent, stems ca. 0.4-1.5 m long, in M. latifolia stems 15—55 cm tall, woody only at base. Leaves opposite, blades ovate to elliptic, acute to slightly acuminate, glabrous to sparsely or minutely puberulent or pubescent. Stipules interpetiolate. Inflo- rescences with slender or filiform pedicels, branching alternate or partially opposite. Calyx lobes lancolate. Corollas white or purplish to rarely rose, 3-13 mm long, funnelform, valvate, 4-lobed, not appendaged, glabrous to densely pubescent inside; heterostylous, with pin and thrum forms. Capsules subglobose to broader than long, where stages are first loculicidally then septicidally dehiscent. Seeds numerous, concavo- convex or obtusely ridged, one species with narrow wing, hilum central, punctiform, areoles isodiametric or radially elongated. The genus name, Mexotis, is derived from Mex- as in Mexico and —otis as in Hedyotis, referring to an , Mexotis, g f Rubi 61 KEY TO SPECIES OF MEXOTIS 1. Plants decumbent or erect, slender and only slightly woody at base, height 15-55 cm; leaf blades 4-32(-45) x 3-24 mm; calyx lobes 0.4-1.5 mm long 2. M. latifolia 1. Plants erect, with woody stems and stout woody bases, height 45-150 cm; leaf blades 25-110 x 8-40 mm; calyx lobes 1-6 mm lon 2. Stems and infl i ly hirtell g ls oft bvi ly wil iged 1. M. galeottii 2. Stems and infl lab t winged. 3. Stipules to 10 mm long; calyx lobes 5s 6 mm long; capsules 5.0-5.5 x 6-7 mm 5. M. lorencei 3. Stipules 2-4 mm long; calyx lobes 1-4.7 mm long; capsules 1-4.5 mm in diam. 4. Inflorescences 3-15 x 6-18 cm; pedicels 7-32 long; calyx lobes 1-2 mm long; corollas 5-6.5 mm long; stipule appendages 1-2 mm long 4. M. terrellii 4. Inflorescences to 8 x 11 cm; pedicels 3-13 mm long; calyx lobes 1.5-4.7 mm long; corollas 7-13 mm long; stipule appendages only 0.3-0.5 mm long 3. M. kingii 1. a galcote s ban. Terrell & H. Rob., comb. nov. (Fig. 1A—D). Declieuxia galeottii M. Martens, in d. Roy: Sci. Brux. 11:231. 1844. Hedyotis Minas (M. Martens) Terrell & Lorence, Phytologia 66:1—4. 1989. Tyre: MEXICO. Oaxaca: Llano Verde, 3-7000 ft, Apr 1840, H. Galeotti 2603 BR!; isorvres: BR!, US!) Manettia liebmannii Standley, J. Washington Acad. Sci. 17:337—338. 1927. Tere: MEXICO. Oaxaca: Pelado, Aug 1842, Liebmann 11485 (HOLOTYPE: C; ISOTYPE fragment: US 13157721). Paratype: MEXICO. Oaxaca: Cuesta de San Juan del Estado, Liebmann 11487 (C; frag- ment US 1315771!) Perennial herbs or small shrubs. Stems 60—150 cm tall, to ca. 2.5 mm thick and woody toward base, stout, subterete, erect, glabrous or minutely hirtellous on upper stem and in inflorescence, few-several branched. Leaves with slender petioles 2-7 mm long, blades 40-85 x12—38 mm, ovate or ovate-lanceolate, acuminate glabrous above, glabrous or hirtellous on midrib beneath. Stipules 1-3 mm, broadly deltate, cuspidate with tip 1-2 mm long, glabrous or with few thick marginal teeth. Inflorescence with terminal, few-flowered cymes, branches densely and minutely hirtellous, flowers heterostylous, pedicels 2-7 mm long, very slender or filiform. Hypanthium glabrous; calyx lobes 1-2 x 0,3-0.7 mm, lanceolate, acute or submucronate, gla- brous. Corollas 4.5-9.7 mm long, funnelform or broadly funnelform, white, glabrous externally; tubes 2.8-4 mm long, 1-2.5 mm wide at base, 2.2-4.5 mm wide at throat, densely puberulent within; lobes 2.2—5.2 x1-2.5 mm, ovate, densely downy on adaxial face with white gland-tipped hairs. Pin flowers with stigma lobes 0.6-1.5 mm long, oblong or linear, exserted 1-2 mm beyond throat, anthers located at 3/4-point from base of corolla tube. Thrum flowers with anthers 1.0—1.5 mm long, linear or narrowly oblong, sessile or on filaments 0.5 mm long, attached at sinuses, stigmas located at midpoint of corolla tube. Capsules 2.2—4.0 x 2.2-4.5 mm, subglobose or somewhat compressed, 3/4—4/5 inferior, rather thick-walled, glabrous, loculicidally dehiscent. Seeds 0.8—1.2 x 0.8—1.2 mm, black, strongly compressed, very thin, flat, slightly to moderately concavo-convex, in outline orbicular or suborbicular, margins thin, wings present or absent, complete or partial, very narrow, 0.1-0.2 mm wide, fragile, hilum punctiform, testa finely reticulate, areoles radially elongated (Fig. 1). The pollen was found by Joan Nowicke (Terrell et al. 1986) to be similar to that of Hedyotis kingii. Chromosome number unknown. Distribution and Habitats.—Forests and rocky slopes. México: Oaxaca. Llano Verde area, Sierra de Juarez, north of the city of Oaxaca; Veracruz, in Sierra Madre Oriental northeast and northwest of Xalapa. Notes.—Additional nomenclatural and morphological notes were included in the Terrell and Lorence (1989) description of Hedyotis galeottii. The minute pubescence on the stems and inflorescence branches a the species. Addi i Labelled Hedyotis liebmannii ( l bove) and cited ! don data from D. H. Lorence. MEXICO Veracruz: Mpio. Alto Lucero, El Cerro la Cima, entre Plan de las Hayas y Tierra Blanca, 1600m, Castillo C. & Narave 2158 (XAL); Mpio. Atzalan, Cerro del Aguila, 850 m, 10 Sep 1982, Ventura A. 19740 (ENCB); Mpio. Atzalan, Alseseca, 950 m, 24 Mar 1975, Ventura A. 11118 (ENCB); Mpio. ii La Calavera, 1000m, 27 Apr 1978, Ventura A. 15232 (ENCB); Mpio. Naolinco, Naranjillo, 1250 m, 13 Nov 1976, Ventura A. 13605 (ENCB); Mpio. Yecuatla, Santa Rita, 1450 m, 12 pu 1973, Ventura A. 7824 (ENCB); Mpio. Yecuatla, Loma Santa Rita, 1480 m, 12 Jan hn Ventura A. 4764 (ENCB). Oaxaca: Yolox R de, al oesta de Cuasimulca, Miranda 1094 (MEXU). 2. Mexotis latifolia (M. Martens & Galeotti) Terrell & H. Rob., comb. nov. (Fig. 1E, F). Oldenlandia latifolia M. £ 41 D A ID Li PS £T 62 Journal of Texas 3( £-4- Ot hue latifolia. Purpus 2951 ( NY), Veracruz. A, B, E. Seeds, ventral views, flat, gs. C D, F. Areoles enlarged. Fic. 1. Seeds examined by SEM. A-D. Mexotis galeottii. A. Castillo C. & M 2158 (XAL) V B-D. Ventura A. 11118 (ENCB), Veracruz. E, F. Mexotis E A Il he I Init I tad ch . owina hilum y T Hu JI DAR: , Mexotis, g £ DAR: 63 Martens & Galeotti, Bull. Acad. Roy. Sci. Brux. 11:235. 1844. Hedyotis latifolia (M. Martens & Galeotti) Walp., Rep. 6:55. 1846. Type: MEXICO. Veracruz: Mirador, at 3000 ft, near Jalapa, May 1840, Galeotti 2556 (holotype: BR; isotypes: F! GH! K!). tune: Hedyotis dichotoma Sessé & Moc., Fl. Mexic. ed.2. 20.1893, non H. dichotoma Cav., Icon. 6:53. 1801. Type: Sessé, Mociño, Castillo, <= Mala 2 aang y Houstonia gracilis Brandegee, Zoe 5:238. 1907, non Hedyotis gracilis DC., Prodr. 4:419. 1830. Type: MÉXICO. Veracruz: on rocks near Orizaba, May 1905, C.A. Purpus 1251 (lectotype: UC! isolectotypes: F! MON. Hedyotis exigula WH. Lewis, Rhodora 63:221. 1961. New name. Perennial herbs with roots sometimes woody. Stems 15—55 cm tall, occasionally rather woody at base, tet- ragonal or rounded-tetragonal, slender, erect, decumbent or spreading, glabrous or rarely pubescent, sparsely to much branched. Leaves with petioles to ca.10 mm long, blades 4—32(-45) x 3-14(-24) mm, median blades ovate to broadly elliptic or ovate-lanceolate, upper blades becoming lanceolate or linear- lanceolate, glabrous or puberulent above, glabrous beneath, base rounded to cuneate, apex obtuse or rather acute. Stipules to 2 mm long and wide, ovate or deltate, glabrous or short-pubescent, margin with several glandular teeth to ca. 1.5 mm long, apices truncate to lacerate. Inflorescence with terminal, few- flowered cymes, flowers heterostylous, closed at night, pedicels to 5(-10) mm long, very slender or filiform. Hypanthium glabrous or puberulent; calyx lobes 0.4—1.5 x 0.2-0.7 mm, scarcely to somewhat longer than capsules, erect, deltate to lanceolate, apices acute or obtuse, margins and sinuses glabrous or with white hairs to ca. 0.1 mm long. Corollas 3-7.3 mm long, funnelform or broadly funnelform, white, purplish, or rarely rose, glabrous externally; tube (222.3-3.3(-4) mm long, usually 1-2 times longer than lobes, some- what widened distally, 0.6-1.4 mm wide at base, 1.8-3 mm wide at throat, glabrous or puberulent within; lobes 13.3 x 0.6-2 mm, ovate, glabrous or puberulent within. Pin flowers with stigma lobes 0.4-1.3 mm long, linear, exserted 0.5—2 mm beyond corolla throat, style 2.4-4.3 mm long, filiform, anthers 0.5-0.8 mm long, 0.2 mm wide, narrowly oblong or elliptic, yellow or whitish, included near midpoint of corolla tube, sessile or subsessile. Thrum flowers with anthers 0.7—1.2 x 0.2-0.3 mm, narrowly elliptic or narrowly oblong, yellow or whitish, exserted at throat on filaments 0.2-0.8 mm long, stigma lobes 0.4-0.6 mm long, in- cluded near midpoint of corolla tube on style 1-1.6 mm long, filiform. Capsules 1-2.7 x 1-3.5 mm, usually slightly wider than long, or subequal, usually 3/4 to 9/10 inferior, moderately compressed, thin-walled, glabrous or with few hairs on calyx margins, cuneate or rounded toward base, apex retuse, dehiscing widely loculicidally and then septicidally. Seeds 8-24 per capsule, 0.5-1 x 0.4-0.9 mm, black, strongly compressed, very thin, slightly to moderately concavo-convex, in outline broadly elliptic to suborbicular, margins thin or slightly rounded, hilum punctiform, testa finely reticulate, areoles radially elongated (Fig. 1). Chromosome number: n = 17 (Lewis in Terrell et al. 1986, as Houstonia gracilis). Phenology.—Flowering throughout year. Distribution and Habitats.—Moist soil, shaded places, banks, slopes, among rocks, sides and bases of cliffs, montane rain forest; altitudes recorded for a few collections vary from 330-1430 m (1000-4300 ft). Mexico: Veracruz (mainly from Cofre de Perote and Jalapa south to Orizaba), northern Oaxaca, Chiapas, and west and central Guatemala. Notes.—Stephen Koch reported (pers. comm. 1980) that this species was for sale in Coyoacan, Mexico City, as a house plant. It grows rapidly and continually produces numerous small flowers. Standley (1918) listed Oldenlandia latifolia as a synonym of O. microtheca DC., but the type of O. latifolia (see above) has flat seeds and reticulate testas, as opposed to O. microtheca (Schltdl. & Cham.) DC. which has angular seeds with the areole walls coalescent. The basionym O. latifolia thus supplants the previously accepted names, Hedyotis exigula and Houstonia gracilis. A paper on Oldenlandia (Terrell & Robinson 2006) provided a key to Oldenlandia microtheca and Hedyotis latifolia (Mexotis latifolia), two superficially similar species differing by seed characters and chro- mosome numbers (the former n = 11. latter n = 17). Mexiotis latifolia is a variable species. Its relationships to the other four species in Mexiotis remain to be determined. It resembles Houstonia purpurea L. of eastern U.S. in aspect and leaf shape and size, but differs in seeds and reproductive characters. Das A ID h l titt. ET 1 I £ al 5f 64 Journal of Texas 3( Representative specimens: MEXICO. Chiapas: Lagos de Montebello, 42 km NE of La Triniteria, 23 Oct 1971, Breedlove & Thorne 21112 (DS, MICH, MO). Veracruz: mpio. Ixtaczoquitlan, Parque de Cerveceria Moctezuma, 1 km SE of Fortin, 970 m, 9 Nov 1978, Koch & Fryxell 78185 (CHAPA, ENCB, XAL); Cerro de Chicahuaxtla-Cuauhtlapan, 22 Jan 1968, Marino Rosas R. 1034 (CAS, MO); Mt. Orizaba, Seaton 101 (F, GH, K, NY, US); near Rio Blanco and Orizaba, 4300 ft, 15 Sep 1944, Sharp 44889 (GH, MEXU, TENN, US); side road between Orizaba and Fortin, 3200 ft, 23 May 1973 Terrell & King 4450, (US); Tepejilotla, mpio. Chocaman, 29 Jun 1979, Ventura A. 16282 (ARIZ, CHAPA, ENCB, MO). GUATEMALA. Huehuetenango: 5 mi SE of Barillas, Sierra de los Cuchumatanes, 1150 m, Steyermark 49537 (GH) 3. Mexotis kingii (Terrell) Terrell & H. Rob., comb. nov. (Figs. 2, 3). Houstonia kingii Terrell, Brittonia 32:491. 1980. Hedyotis kingii (Terrell) G.L. Nesom, Syst. Bot. 13:434. 1988. Tre: MEXICO. Oaxaca: E-facing slopes, mountains along rt. 175, 28 km by road S of Valle Nacional on way to Ixtlan de Juarez; rain forest abounding in ferns, bromeliads, orchids, and aroids; elev. ca. 2000 m; not common; full sun; gravelly soil, 30 Jul 1959, R.M. King 2107 (HoLorvre: MICHI, 2 sheets, one flowering, one fruiting). Perennial herbs with thick, woody roots. Stems to 45 cm tall, to ca. 5 mm thick and somewhat woody toward base, tan or straw-colored, subterete, rather stout, erect, glabrous, branched. Leaves with petioles to 10 mm long, blades 40-85 x 8-30 mm, elliptic, ovate, or lanceolate-ovate, acuminate, glabrous, slightly paler be- neath. Stipules to 2 mm long, broadly rounded, lobed, the central lobe with recurved glandular hairs. In- florescence in terminal, few-flowered cymes, widely branching, to 8 x 11 cm, flowers heterostylous, pedicels to 13 mm long, very slender or filiform. Hypanthium glabrous; calyx lobes 1.5-4.7 x 0.3-0.8 mm, 1/2-3/4 as long as corolla tube, linear or narrowly lanceolate, acute or acicular, sometimes divaricately spreading. Corollas 7-13 mm long, broadly funnelform, white or light purple, glabrous externally; tube 3-9 mm, 3-7 mm wide at throat, glabrous or puberulent within; lobes 2.5-7.0 x 1.5-3.3 mm wide, shorter than to slightly longer than the tube, ovate, glabrous or puberulent within. Pin flowers with stigma lobes 0.5—1.3 mm long, narrowly oblong or linear, near corolla throat or slightly exserted, anthers located at 1/3-1/4-point from base of corolla tube. Thrum flowers with anthers 1.3-2.0 mm long, narrowly oblong, purple, attached at corolla throat, subsessile or filaments to 1.5 mm long, stigmas located at 1/4—point from base to midpoint of corolla tube. Capsules (only previous year's dehisced capsules seen) 3-4 x 3-4 mm, subglobose, thin- walled, glabrous, 3/4—7/8 inferior. Seeds 0.6—1.1 x 0.5-0.9 mm, dark brown, moderately to slightly com- pressed, often polygonal in outline, obtusely angulate or obtusely ridged, margins rounded, hilum puncti- form, testa finely reticulate, areoles small, isodiametric. Pollen illustrated in Terrell et al. 1986, figs 21, 22. Plant and seeds illustrated here in Figures 2, 3. Chromosome number unknown. Phenology.—Flowering December to July. Distribution.—México: Oaxaca. Original collection in a cloud forest of the Sierra Juarez range along or near highway 175 from Ixtlán de Juarez to Valle Nacional, 28 km south of Valle Nacional. Additional occur- rences in Oaxaca cited below are mostly along hwy. 175 between Tuxtepec and Oaxaca city; the location of Tarabunde is unknown. On 22 July 1987 Stephen Koch and Terrell, somewhat hampered by a torrential rain, searched unsuccessfully for the species along Highway 175. Additional specimens examined. MEXICO. Oaxaca: Distr. de Ixtlan, roadside, trop. v itl ferns, 24 mi S of Valle Nacio- nal, Hwy. 175, 22 Mar 1978, Poole et al. 1283 (LL); bosque mesofilo perturbado, 1750 m, 38 km S of Valle Nacional, 11 Apr 1976, Rze- dowshi 34092 (ENCB); km 128 entre Llano de Las lores y Tuxtepec, 26 Dec 1965, Delgadillo 121 (MEXU), reported by Lorence; Mpio. de Comaltepec, Hwy. 175, “Vista hermosa,” selva secundaria, km 137, 1500 m, 11 May 1966, Martinez Calderon 801 (ENCB, MEXU, XAL); Tarabunde, 7 Feb 1966, MacDougal 17 (US); 22-26 mi S from Valle Nacional on Hwy. 175, cloud forest, 7 Jan 1989, Todzia et al. 2839 (TEX) 4. Mexotis terrellii (Lorence) Terrell & H. Rob., comb. nov. Hedyotis terrellii Lorence, Novon 7:51.1997. Tyre: MÉXICO. Oaxaca: Distrito de Mixe, Municipio de Totontepec, 3 km al SO de Totontepec, carretera a Mitla, 1910 m, bosque mesófilo, 6 Sep 1986, R. Torres C. & C. Martinez 9067 (HoLorvre: PTBG 12314; isotype: MEXU) The following description is modified from that of David Lorence (1997). Perennial herbs. Stems 45-80 cm tall, branching from the base with several erect stems 2.0-3.5 mm diam. and somewhat woody toward base, terete, pee adi with petioles 3. on mm long, blades (25—)45-100 x 10-40 mm, elliptic to ovate- elliptic, the b decurrent, te, glabrous, chartaceous, margins sometimes with few small teeth. Stipule en to 4 mm ine: MR or deltate, with apical and lateral linear gland-tipped teeth 1-3 , Mexotis, g f Rubi 65 Fe? € de afhalat £ M. n ae A Joy CEA) A R Seeds polygonal, flat, IE ele PIPER] I CD A E el I l, not radially elongated. mm long. Inflorescences terminal, compound cymes 3-15 x 6-18 cm, subtended by reduced leaves, branch- ing dichasially to 2° or 3°, the primary branches 2-5 cm long, the axes slender, glabrous, subtended by stipule-like bracts, flowers on filiform pedicels 7-32 mm long. Hypanthium broadly obconic, glabrous. Calyx lobes 1-2 x 1-2 mm, deltate, glabrous. Corolla in bud quadrangular, rounded at apex, at anthesis 5-6.5 mm long; tube 2-2.5 mm long glabrous, lobes 3-4 mm long, erect or slightly spreading, elliptic, acute at apex, externally glabrous, internally villosulous with crinkled, white trichomes. Stamens attached near middle of tube below sinuses, the filaments 0.5 mm long, anthers 1 mm long, linear-elliptic, style 2-2.5 mm. Stigma lobes 1 mm long, linear. Capsules 2-3 x 3-4 mm, slightly wider than long, 1/2 inferior, thin- walled. Seeds immature. Plant illustrated in Lorence 1997. Phenology.—Collected in flower in March, and in flower and fruit in June and September. Distribution. —Known only from the area around the type locality in cloud forest at ca. 1900 m. near Totontepec in the mountainous Mixe District of Oaxaca. Notes.—Lorence (1997) compared certain characters of M. terrellii with those of M. galeottii and M. kingii. Mexotis terrellii mede resembles M. hingii, as summarized in Table 1 d. P MEXICO. Oaxaca: Distrito Ixtlán, Llano Verde, collector unknown, Herb. Reichenbach fils III-396(W); Dio Mixe, mpio. ded COE Totontepec, bosque mesófilo de montaña, 1900 m, 17?15'N, 9602"W, 11 Mar 1990, Rivera R. & 66 Journal of tl By fo il d y Pz alt ce 9 72 XN Wo b y TÍA AA AS y 1 ae Y 7 KET A "dA iN » 4 A /À E Y ey AQ PLA a A " h á 4 TÁ j 44 2 y x » A h JA ANE. " X AND X y»: ye PV Any u Ny A J f ay E equine Ce ee ae 2. Fic. 3. Mexotis kingii, holotype. A. Habit. B. Flower. C. Corolla with st style, and stigma D Ovary sectioned, ith calyx lol i sti E. In- £1 agt L L L £. + > An^ 19 F. Mature capsules. D i late Regina O. H 3 i DE, 2 2 Terrell and Robinson, Mexotis, a new genus of Rubiaceae 67 gag on : : Jhy CEM A D I £. R Ventral f ith ina-like margin. € 3 3 Fic. typ y Areoles, dorsal surface, isodiametric. D. Areoles, ventral surface, somewhat radially elongated. Martin 1403 (PTBG, UC); Distrito Mixe, mpio-Totontopec, 2 km al SO de Totontepec, carretera a Oaxaca, 1900 m, 17 Jun. 1986, Torres C & Téllez 8643 (MEXU, PTBG). 5. Mexotis jorence! Tere & H. Rob., sp. nov. (Figs. ^ 3» Terr: XEM ea id de pino-encino perturbado, Distr. S Mpio. San Martin Peras, 1 | del t 24 carretera San Sebastián oia Lat. 17.17 N. Long. 98.09 W., Alt. 2655 m, 29 Nov 1994, IL Calzada 19568 ee MEXU; isorvre: PTBG). Ad M. tida bi ot M t sio: 11 di i e ABS >] IH s il dicti t e This description is based only on one E branch in ae condition. Perennial herb to 80 cm tall. Stems 8 mm or more thick, widely branched. Leaves with slender petioles 10-25 mm long, blades 70-110 x 13-30 mm, elliptic or ovate-lanceolate, acuminate. Stipule body to ca. 5 mm long, with several linear apical and lateral gland-tipped teeth to 10 mm long. Inflorescence in fruiting condition only, pedicels to ca. 15 mm long, slender. Calyx lobes ca. 5-6 mm long, linear or narrowly lanceolate. Corollas lacking. Capsules mature, 5-5.5 x 6-7 mm, distinctly wider than long. Seeds (only a few available) 1.4-1.8 x 1.4—1.8 mm, dark brown, orbicular or broadly elliptic in outline, thin, strongly (cupulate) to moderately concave-convex, hilum punc- tiform, testa reticulate, areoles not clearly visible. Fig. 5. Distribution and habitats —The collection label stated “Negro con calizas,” which we suppose referred to black soil with limestone or limey soil. The locality was in a disturbed pine-oak forest on the west side of Oaxaca not far from the Guerrero border. This area is roughly 100 miles or 160 km from the other species and we have not seen any related collections from this part of Oaxaca. 5 Mp! is florence; lesdl[ » MBA, Hs ¡Upa Edvard Terex de 2009 Mesas Ss de Kn rd correo MNA, NY NEN eee Vany UR Fic. 5. Mexotis lorencei, J.I. Calzada 19568, isotype (PTBG) ZR E Sp. w David is Lene An elisdia By daricm t TER) ‘rita T, T i Dah: , Mexotis, £Duhk: 69 Taste 1. Morphological ct M. galeottii M. latifolia M. kingii M. terrellii M. lorencei Stems height cm 60-150 15-55 to 45 45-80 to 80 Leaf blades L xW mm | 40-85 x 12-38 4-32(-45) x 3-24 40-85 x 8-30 25-100x10-40 . 70-110x 13-30 Inflor. L x W cm 3.5 x 2.5-5 no data to8x 11 3-15 x 6-18 no data pedicel Lmm 2-7 to 5(-10) 3-13 7-32 to 15 vesture hirtellous glabrous glabrous glabrous glabrous Calyx lobes L mm 1-2 0.4-1.5 1.5-4.7 1-2 5-6 Corollas L mm 4.5-9.7 3-7.3 7-13 5-6.5 no data tubes L 2.8-4 2-4 3-9 2-2. inner vesture villous glab/puberulent ^ glab/puberulent glabrous lobes L 22-52 1-3.3 2.5-7.0 -4 Capsules L x W mm 2.02-40 X 22-45 | 1-27 x 1-3.5 3-4 x 3-4 2-3x3-4 5-55 x 6-7 Seeds L x W mm 0.8-1.2 x 0.8-1.2 0.5-1.0 x 0.4-0.9 0.6-1.1x0.5-0.9 immature; no data 1.4-1.8 x 1.4-1.8 Notes.—We name this species for David Lorence, who called our attention to the Calzada collection and who has contributed much to knowledge of Rubiaceae in Mexico. This species resembles the other three species with stout woody stems and large leaves. Although only one collection is known, there are capsule, seed and stipule characters for consideration. The capsules and seeds are distinctly larger than in other species. The seeds have a thin, strongly (cupulate) to moderately concavo-convex body. Other species in this genus have flat or slightly concavo-convex seeds. ACKNOWLEDGMENTS David Lorence provided loans and important data on Mexotis galeottii, M. terrellii, and M. lorencei. In earlier times Stephen Koch provided very helpful field assistance. We thank Scott Whittaker, Electron Microscope Laboratory, Smithsonian Institution, for the SEM illustrations, and Marjorie Knowles and Alice Tangerini for preparing the figures. David Lorence and Warren Wagner are thanked for providing in-depth reviews of the manuscript. The curators of the cited herbaria helpfully provided Terrell with loans during a number of years long past. REFERENCES ANDERSSON, L. AND J.H.E. Rova. 1999. The rps16 intron and the phylogeny of the Rubioidae (Rubiaceae). Pl. Syst. Evol. 214:161-186. BLackweLL, W.M. 1968. Revision of Bouvardia (Rubiaceae). Ann. Missouri Bot. Gard. 55:1—30. Church, S.A. 2003. Molecular phylogenetics of Houstonia (Rubiaceae): descending aneuploidy and breeding system evolution in the radiation of the lineage across North America. Molec. Phylogenet. Evol. 27:223-38. FosBERG, F.R. 1943. The Polynesian species of Hedyotis (Rubiaceae). Bernice P Bishop Mus. Bull. 174:1-102. KAREHED, J., |. GROENINCLX, S. Dessein, TJ. MoTLEY, AND B. Bremer. 2008. The phylogenetic utility of chloroplast and nu- clear DNA markers and the phylogeny of the Rubiaceae tribe Spermacoceae. Molec. Phylogenet. Evol. 49:843-866. Lorence, D.H. 1997. New taxa, a new name, and a new combination in Rubiaceae from southern Mexico and Mesamerica. Novon 7:46-58. STANDLEY, P.C.1918. Rubiaceae, Oidenlandieae. In: N. Amer. Fl. 32(1):17-39. TERRELL, E.E. 1987. Carterella (Rubiaceae), a new genus from Baja California, Mexico. Brittonia 39:248-252, TerreLL, E.E. 1996. Revision of Houstonia (Rubiaceae-Hedyotideae). Syst. Bot. Monogr. 48:1-118. TeRRELL, E.E. 2001a. Taxonomy of Stenaria (Rubiaceae: Hedyotideae), a new genus including Hedyotis nigricans. Sida 19:591-614, I afl D : ID hi M. £ T, 70 J t Texas 3( TerreLL, E.E. 20015. Stenotis (Rubiaceae), a new segregate genus from Baja California, Mexico. Sida 19:899-911. TerRELL, EE. AND D.H. Lorence. 1989. Hedyotis galeottii (Rubiaceae), new combination for a Mexican species. Phyto- logia 66:1—4. TerreLt, E.E. AND H. Rosson. 2003. Survey of Asian and Pacific species of Hedyotis and Exallage (Rubiaceae) with nomenclatural notes on Hedyotis types. Taxon 52:775-782. Tengen, E.E. AND H. Rosinson. 2004. Seed and capsule characters in Arcytophyllum, Bouvardia, and Manettia (Rubi- aceae), with notes on A. serpyllaceum. Sida 21:911-927. TerreLt, E.E. AND H. Rosinson. 2006. Taxonomy of North American species of Oldenlandia (Rubiaceae). Sida 22: TERRELL, E.E, W.H. Lewis, H. RoBinson, AND J.W.Nowicke. 1986. Phylogenetic implications of diverse seed types, chromo- some numbers, and pollen morphology in Houstonia (Rubiaceae). Amer. J. Bot.73:103-115. Tengen, E.E., H. Rosson, W.L. Wacner, AND D.H. Lorence. 2005. Resurrection of genus Kadua for Hawaiian Hedyo- tidinae (Rubiaceae), with emphasis on seed and fruit characters and notes on South Pacific species. Syst. Bot. 30:818-833. CAPSICOPHYSALIS: ANEW GENUS OF SOLANACEAE (PHYSALEAE) FROM MEXICO AND CENTRAL AMERICA John E. Averett ee Martinez sic ati Ol P uela de Biologia PO, Box a Autonoma de Querétaro Georgia So es UTER Querétaro, 76010 MEXICO pee Georgia 30460, U.S.A. Mahindaguaq.mx averett@georgiasouthern.edu ABSTRACT B new o from the Moe 2 eastern and southern Mexico and Central America is na da is a m Luis Potosi, Mexico. Áthenaea cernua £1 1 f. E y * A +41 1 1 onymy. The genus hasan condo 5-lobed corolla, red fruit, 5-lobed g cal; y testa, features unknown in related genera of Solanaceae. A p giving the distinguishing psicoy Cha eee is provided. RESUMEN Se descril distrib ] fias del le México, así como en Centro América. Capsicophysalis está basado en iaa potosina, descri de San Luis on México. Athenaea cernua, descrita originalmente de Guatemala y que ha sido transferida a tres géneros, qu nero tiene una corola aaa 5-lobulada, fruto aN un i p D M al pa y eae con una testa tubercalada, T d g 3: f, ] A O a A ] ES r t r T pal Key Wonps: Solanaceae, Athenaea, Chamaesaracha, Physalis, Physaleae, Physalineae, Guatemala, Mexico Hunziker (2001) recognized section Capsicophysalis within Chamaesaracha, a genus of about 10 species distributed in the arid regions of the southwestern United States and adjacent northern Mexico. Within the section he included three species: C. cernua, C. potosina, and C. rzedowskiana. Hunziker listed C. cernua as the type species of the section. All of the species are relatively rare. Hunziker (1980) noted that only seven collections of C. cernua were known. We add an additional 7 for a total of 14. Until this study C. potosina has been known only from the type collection, all other specimens having been referred to C. cernua (= Athenaea cernua). Only two collections of C. rzedowshiana are known to us, both of which are from San Luis Potosi, Mexico. Chamaesaracha cernua also has been included in Physalis and Athenaea and is treated as Athenaea cernua in most recent literature. Waterfall (1967) treated the taxon as a variety of Physalis melanocystis (Robins.) Bit- ter. More recently, Hunziker (1980) agreed with Gentry (1973) that the treatment of this species in Physalis was unacceptable and, in addition, made a very strong argument for its exclusion from Athenaea, which, as now conceived, is a small Brazilian genus of about ten woody species. Careful morphological comparisons indicate that C. cernua and C. potosina are conspecific with po- tosina the oldest specific epithet. The species differs in a number of critical features from Chamaesaracha, Physalis, and Athenaea, and is best treated within a new and separate genus, Capsicophysalis. Chamaesaracha rzedowshiana differs from Capsicophysalis in distribution, flowers, fruit and fruiting calyx and, for now, is retained in Chamaesaracha. Chamaesaracha and Capsicophysalis (as Chamaesaracha cernua) were included by Estrada and Martínez (1999) in their morphology based cladistic analysis of Physalis and related genera. They concluded that C, po- tosina was not closely related to either Physalis or Chamaesaracha. In the strict consensus tree, C. potosina forms a clade with Leucophysalis viscosa which Averett (20092) now recognizes as a distinct genus, Schraderanthus. C. potosina further differs from Chamaesaracha in that it is an annual herb of mesic riparian forests (Table 1). J. Bot. Res. Inst. Texas 3(1): 71 — 75. 2009 Capsicophysalis potosina is relatively rare and is represented by only a few collections in herbaria. Un- fortunately, mature fruiting calyces are not always seen and the feature has not been noted in the literature. However, it clearly is present on a number of specimens, including type material of both C. potosina and C. cernua. The very distinctive irregularly lobed corolla, once seen, is easily observed on herbarium sheets but has not been noted in the literature. The mature calyx also appears to be a dark red on herbarium sheets. More information on these interesting characters would be welcome. Capsicophysalis (Bitter) Averett & M. Martínez, stat. nov. Physalis sect. Capsicophysalis Bitter. Repert. Spec. Nov. Regni 70. 1924. Chamaesaracha sect. Capsicophysalis (Bitter) Hunz., Genera Solanacearum 230. 2001. Tyre spectes, Capsicophysalis potosina (B.L. Rob. & Greenm.) Averett & M. Martínez Annual or weak perennial herbs to 1 m high; herbage glandular pubescent, mixed with longer hairs; leaves petiolate, thin to membranous; flowers 1-2 in axils, campanulate—subrotate, 1.5-3 cm wide, corolla white, yellowish, or yellow-green with villous pads in the throat, unequally 5-lobed and slightly irregular, aestiva- tion plicate; calyx campanulate, 5-lobed, accrescent in fruit, at first loosely investing the berry, then splitting and becoming reflexed below the berry, red, the lobes thickened along the margins; fruit a berry, shiny red or orange-red; seeds discoid 1-1.5 mm long, testa tuberculate. The striking features of this distinctive genus include the irregular corolla, red or orange-red fruit, the red reflexed, deeply lobed structure of the mature calyx, and rod-like projections on the seed testa (Fig. 1). All of these features are uncommon among related genera, and the irregular corolla is completely novel. The latter character is evident in the types of both C. potosina and C. cernua and present in all of the cited specimens with flowers. The red fruit is largely unknown among potential relatives except in Brachistus and Schraderanthus viscosus, both of which have 6—8(-10) flowers arising in fascicles from the axils and other distinguishing features. Capsicophysalis potosina (B.L. Rob. & Greenm.) Averett E M. Da comb. nov. (Figs. 1-2). Basionym, B.L. Rob. & Greenm. Amer. J. Sci. 50:161. 189 (B.L. Rob. € Greenm.) Averett, Ann. Missouri Bot. Gard. 57: 380. 1971. Ere: MEXICO. San Luis Porosr: Tamasopo Puis M 1880, Pringle 3654 (HoLorvrE: VT!; isorvee: GHI). Athenaea cernua Donnell Smith, Bot. Gaz. 48:297. 1909. Physalis melanocystis (Robins.) Bitter var. cernua (Donnell Smith) Waterfall, Rhodora 69:99. 1967. Chamaesaracha cernua (Donn.-Sm.) Hunz., Contr. Gray Herb. 210:25-26. 1980. Tree: GUATEMALA. Derr. in Verapaz: Sasia, 900 m, May 1908, Tuerckheim II 2245 (hoLoTvee: US!; isotype: CORD pl id Repert. Spec. Nov. Regni Veg. 20:371. 1924. Tyr: aaa VERACRUZ: : Papantla, pue 1829, ans Note.— Hunger (1980) terial of thi g thet i tion. We also have not seen ister of this collection. Plants herbaceous annuals or weak perennials to 1 m high; herbage largely glabrous except for a few hairs along the stems and leaf margins; leaves petiolate, blades 2—4 cm long and 1-2 cm wide (about Y as wide as long), narrowly ovate-lanceolate, margins entire, acute-acuminate at the tip, lamina thin, on short petioles; flowers 1-2 from axils on pedicels 3 cm long, flowering calyx ca 15 mm long and 12 mm wide, campanulate and rounded at the base, divided 1/2-3/4 its length, lobes acute to slightly acuminate; corolla ca. 1 cm long, yellow-white, rotate-campanulate; anthers white-yellow, 2 mm long, filaments ca 3 mm long, connected to the base of the anthers; fruiting calyx 8-12 mm long and wide, campanulate, exceeding the berry and becoming reflexed at maturity, berry orange-red to bright red at maturity; seeds dark brown, testa tuber- culate with rod-like projections. Distribution —Chamaesaracha potosina is distributed from Guerrero and Tamaulipas in Mexico south to Guatemala and Honduras (Fig. 2). Additional material examined: BELIZE. a Arenal-Valentine road, Jun-Aug 1936, Lundell 6181 (US); Vaca Plateau, 8 Mar 1980, Whitefoord 2023 (MO). GUATEMALA. Peten: Dolores, 22 Aug 1961, Contreras 2746 (LL); Dolores, 5 km E of village, 30 Aug 1961, Contreras 2830A (TEX); Tikal, 18 Jan 1962, ee 17186 (LL); Tikal Nat'l. Park, 14 Feb 1959, Lundell 15548 (LL); Tikal Nat'l. Park, 9 Mar 1959, Lundell 15805 (LL); Ciebal, Sayaxche, 17 Mar 1970, Contreras 9764 (LL). HONDURAS. Dist. Toledo: Edwards road beyond Columbia, 15 May 1948, Gentle 6535 (LL). MEXICO. chisves: 9km S E oe 6 Apr 1985, E & Cabrera 8168 (TEX); Mpio. Ococingo, 14 Sep 1985 Martínez 13823 (TEX). G le O ril 1938, Hinton 14034 (GH, LL); Dist. Galena, Carrizo-El Río, 20 Oct 1939, Hinton 14689 (GH). Oaxaca: Mpio. Sta. Maria a. 26 July 1985, H. Hernandez G. 1375 (TEX). Tamaulipas: Averett and Martínez, Capsicophysalis, a new genus of Solanacaeae GSU0079 E Fis. 1. Capsicophysalis potosina: A. flower; B. cleared flower showing unequal lobes; C. SEM of seed; D. immature fruit with calyx enclosing the berry; E. mature fruit with red berry and reflexed calyx. na, Mi pr Fic. 2. Distribution of Capsicophysalis potosina. H CG 41 Mpio. Victoria, Cañon de La Libertad, May 1994, Martínez s.n. (UAT); Mpio. de San Nicolás, a 5 km del inicio de la b 28 Nov 1998, Martínez 5329 (QMEX). GENERIC RELATIONSHIPS Capsicophysalis probably is most closely related to Schraderanthus viscosus (Schrad.) Averett which Averett (2009a) recognized as a distinct genus. Hunziker (1991) had included S. viscosus in Leucophysalis and later (1995) in Chamaesaracha. He returned the species to Leucophysalis in his Genera Solanacerum (2001). Nei- ther Capsicophysalis nor Schraderanthus seems to be especially close to Chamaesaracha and certainly not congeneric. Capsicophysalis has a distribution similar to that of Schraderanthus, Brachistus and Tzeltalia, but C. poto- sina extends farther to the north in the Mexican states of San Luis Potosi and Tamaulipas. Morphologically, Capsicophysalis is similar to Schraderanthus and Brachistus which also have orange or red berries, but the flowers are not in fascicles and, at maturity, the fruiting calyx is reflexed under the berry. Table 1 compares Capsicophysalis to Schraderanthus and Chamaesaraccha. Averett (2009a, 2009b) provides further discussion of the history and taxonomy of Schraderanthus and its relationship to Leucophysalis and Brachistus, including supporting molecular data from Olmstead et al. (2008) and Whitson and Manos (2005). Hunziker (2001) placed Brachistus in Tribe Solaneae, subtribe Witheringinae while Olmstead et al., (1999, 2008) place Brachistus in Tribe Physaleae, subtribe Physalineae. Capsicophysalis has a strongly accrescent fruit- ing calyx characteristic of the Physaleae as described by D'Arcy and Averett (1996). We therefore include Capsicophysalis in Physaleae, subtribe Physalineae, near Schraderanthus and Brachistus. All three have bright red or orange-red fruit which is relatively uncommon in the Physalineae. Taste 1. Comparative characters of Capsicophysalis, Schrad hus, and Chamaesaracha. Capsicophysalis Schraderanthus Chamaesaracha Habit Erect, herbaceous, annual Erect, herbaceous to woody, Ascending or spreading perennial annual or perennial herbs Habitat and Mesic forest, southern Mexico, Mesic forest, southern Mexico, Arid regions of southwestern U.S. distribution Guatemala and Honduras Guatemala and northern Mexico Inflorescence 1-2 flowers from axils 6-8 flowers from axils in 7 1-2 flowers from axils fascicles Corolla less than 3 cm wide, w/o 4-5 cm wide, with green broken 2-4 cm wide, with white maculations in the throat maculations in the throa tomentose pads in the throat Fruit Red, fleshy berry Red, fleshy berry Green, dry berr Fruiting calyx Accrescent, reflexing under Accrescent, rotate to slightly Accrescent, appressed to and the berry at maturity, red reflexed under the berry at partially enclosing the maturity, red berry, green Seeds Testa tuburculate with rod- Testa rugose-reticulate, Testa rugose-reticulate, honey- shaped projections honeycombe combed ACKNOWLEDGMENTS Marshall Crosby provided helpful comments about the nomenclature and we appreciate the assistance of those herbaria which were visited and those from which specimens were borrowed, GH, LL, TEX, QMEX, US, UAT, and VT. We thank Alan Harvey for assistance with the photographs and distribution map and 3 anonymous reviewers for their corrections and helpful comments on the manuscript. REFERENCES Averett, J.E. 2009a. Schraderanthus, a new genus of Solanaceae. Phytologia 91:54—61. AvtRETT, J.E. 2009b. Taxonomy of Leucophysalis (Solanaceae, Tribe Physaleae). Rhodora (in press). D'Arcy, W.G. AND J.E. Averett. 1996. Recognition of tribes Capsiceae and Physaleae, subfamily Solanoideae, Solan- aceae. Phytologia 80:273-275. ESTRADA, E. AND M. Martinez. 1999. Physalis L. (Solanoideae: Solaneae) and allied genera: a morphology-based cladistic analysis. In: M. Nee, D.E. Symon, R.N. Lester, L. Jessop, and J.P. Jessop, eds. Solanaceae 4:139-159, Royal Botanic Gardens, Kew. Gentry, J.L. 1973. Studies in Mexican and Central American Solanaceae. Phytologia 26:255-278. Hunziker, A.T. 1980. Studies on Solanaceae XII. Additions to the genus Chamaesaracha. Contr. Gray Herb. Hunziker, A.T. 2001. Genera Solanacearum. Koeltz Scientific Books, Kónigstein, Germany. OLMSTEAD, R.G, J.A. Sweere, R.F. SPANGLER, L. Bons, AND J.D. Pamer. 1999. Phylogeny and provisional classification of the Solanaceae based on chloroplast DNA. In: M. Nee, D.E. Symon, R.N. Lester, L. Jessop, and J.P. Jessop, eds. Solanaceae 4:111-137. Royal Botanic Gardens, Kew. OuusreAD, R.G., L. Bons, H.A. Micip, E. SANTIAGO-VALENTIN, V.F. GARCIA, AND S.M. Cottier. 2008. A molecular phylogeny of the Solanaceae. Taxon 57:1159-1181. WATERFALL, U.T. 1967. Physalis in Mexico, Central America, and the West Indies. Rhodora 69:84- 120. Wuitson, M. AND P. Manos. 2005. Untangling Physalis from the physaloids: a two-gene phylogeny of the Physa- lineae. Syst. Bot. 30:216-230, BOOK REVIEWS Marr Warnock Turner. 2009. Remarkable Plants of Texas. (ISBN 978-0-292-71851-7, hbk.). University of Texas Press, PO. Box 7819, Austin, Texas 78713-7819, U.S.A. (Orders: www.utexas.edwutpress. com, 1-800-252-3206, 1-800-687-6046 fax, email utpress@uts.cc.utexas.edu). $29.95, 320 pp., color images and photographs, 7 1/4" x 10 1/8" Remarkable aed an ae is a fantastic collection of Moss qe on Texas native plants It includes over 100 magnificent color photo- graphs f y plant. The 65 t tain one or more ica pa a are D into plant types: trees, seus, herbaceous pane cacti, M dpt vines, and aquatics. Each pl y common names 1 Mi 1 L 1 fe! 1 1 lies 1 F 2 E > E eor t 1 È a A e Em A1 a Wiel 1 T 1 y gy E : 1 F basics, author Matt Ti c fl £.1£ o * rr di d d quality, medicinal value, culinary uses, fiber, building ials, dyes, fragrances, uses as animal fodder, and otl i d ial kr i? f d S 1 Es 1 1 J 1 1 Tep 1 m Hd 1 i E 1 emarhable Plants of y ] g I y collected all within one text. As such, T has provided a f ic refi 11 i linfi j 1. Itisa J f,.1 1 1. £, e x 1 1 e 1 t q 1 1 g vay peor I y today, but ME historically. Overall, this 1 TE y plant lover would love to ad. —Lee Luckeydoo Dreese, Volunteer, Botanical Research Institute of Texas, Fort Worth, Texas 76102-4025, U.S. ? GRAHAM PAYNE. 2006. Garden Plants for Mediterranean Climates. (ISBN-10: 1861268955, pbk.). The Cro- wood Press Inc., The Stable Block, Crowood Lane, Ramsbury, Marlborough, Wiltshire SN8 2HR, England. (Orders: www.crowoodpress.co.uk, by phone 01-672-520320). $35.00, 240 pp., 8 1/2" x 11". A lavish guide to peana put ] maintaining a Mediterranean garden, featuring an A—Z of more than 1,000 plants and 500 color photos N i pl g colorful shrubs, or lush gardens With the help of this book you can enjoy Medi l in y own n ok includes: an — diterranean d points to consider when planning a garden; key features of a Median ui ME Med pus palms, wn and nd pelas sections on both general care and specific plants; advice to Z of more n 1,000 p pis 500 Spigot color photos. The core of the booki is the A to Z of hos us includes commended nee Each entry includes a general description and growing range, genera 1 p Dp ud ultivation tips The book includes several special PE gofp English , and a gl y d terms. p list for further oe a E ook T : A 1. 1; : f. 1 1 1 + VEN F UE o basin and elsewhere around the globe. He widens the climat lope to include anywt hat is frost-fee with a good water supply. "rl 1 1 J 4 Hd E H uu 1 11 T E i 1 1 1 1 1 i t 1 i 2 4 oro A e R 15, f, E 1:4 z dri 1 1 1 ce Fa Graham ie trained as a honicultarist and followed a career as a pedes Hë became Head oe of ee eee Cambridge, and then Head Gardener of Jesus College, Cambridge, wher gardens and sports grounds, as well as fifty other properties. He has now aum to ues Europe: — Gary L. Jennings, Librarian, t Botas: cal Research Institute of Texas, Fort Worth, Texas 76102-4025, U.S.A. J. Bot. Res. Inst, Texas 3(1): 76. 2009 SYNOPSIS OF GONOLOBUS S.S. (APOCYNACEAE: ASCLEPIADOIDEAE) IN TRINIDAD AND TOBAGO Alexander Krings Herbarium, Department of Plant Biology North Carolina State University Raleigh, North Carolina 27695-7612, U.S.A. ABSTRACT A synopsis of Mauda S.S. rios OR: "e in d and D Nine; a ee descriptions, and illustrations, i I g I [ tobagensis The E 1 $2 1 1; s A a) +h aA 1 2 EI A RESUMEN S i isd lol (Apocynaceae, Asclepiadoid ción en ue y ai Aue incluye una clave, des- ples e codes para f l imi y G. tobagensis. Se discute 1 ] ] la bibliografía y en los espiclinenes. Sirainens About fifty Species of subtribe G lobi (A l loid ) occur on the islands comprised by the Greater and Lesser Antilles, the T Trinidad and Tobago, and Aruba and the Netherland Antilles. Evidence from the chloroplast (Rapini et al. 2003; Liede-Schumann et al. 2005; Rapini et al. 2006; Krings et al. 2008) and nuclear genomes (Krings et al. 2008) supports the monophyly of the subtribe, as well as of Gonolobus s.s. Krings (2008a) recently provided a revision of Gonolobus s.s. in the West Indies. However, Trinidad and Tobago were excluded from that treatment due their South American affinity. The objective of this note is to provide a key to and descriptions of the species of Gonolobus s.s. from Trinidad and Tobago to facilitate their recognition and conservation. This treatment is derived from a larger critical study of ca. 250 specimens of the fifty some known species in West Indian Gonolobinae, obtained in part through: (1) loan requests from ninety herbaria—of which sixty-five responded with either loans, digital images, or negative search results (see Acknowledge- ments), (2) visits to BM, BSC, DUKE, HAC, HAJB, IJ, K, UCWI, UPRRP, US, and P, and (3) forty-eight days of field work by the author in the Bahamas (Long Island), Cuba, Dominica, Jamaica, Puerto Rico, St. Lucia, and St. Vincent. The treatment is also informed by analyses of sequences of portions of the chloroplast (trnL-F, rps16) and nuclear genomes (LEAFY) of selected accessions (see Krings et al. 2008). TAXONOMIC TREATMENT Two species of Gonolobus s.s. are recognized from Trinidad and Tobago. Descriptions of the species are provided following a key to the species. Corona morphological terminology follows Liede and Kunze (1993) and Kunze (1995): Ca = faucal annulus (corolline corolla); Cd = dorsal anther appendage; Ci = interstaminal gynostegial corona; Cs) = fused staminal and interstaminal gynostegial corona; Cs = staminal gynostegial corona. Species are arranged alphabetically. Following Franz et al. (2008), taxon concept mapping is pro- vided to facilitate databasing. The operator =is used to indicate whether a given taxon concept is equal to a previously published concept. The symbol + is used to indicate the misapplication of a name to a concept. KEY TO GONOLOBUS S.S. IN TRINIDAD AND TOBAGO 1. Calyx lobes ovate, 4.1-8.4 mm x 1.9-5.5 mm, btuse t ded; lla lobes glal laxially .G. rostratus 1. Calyx lobes linear-lanceolate, 8.5-13 mm x 1.7-2.6 mm, apices acute to narrowly obtuse; corolla lobes 2 pubescent adaxially . G. tobagensis J. Bot. Res. Inst. Texas 3(1): 77 — 83. 2009 78 tanical h Insti f Texas 3( 1. Gonolobus rostratus (Vahl) Schult. in Roemer & Schultes, Syst. Veg. 6:61. 1820. (Fig. 1). Cynanchum ros- tratum Vahl, Symb. Bot. 3:45. 1794. Tyee: Taina: (von Rohr?) Hb. Liebmann (nororvee: C [IDC microfiche photo: Vahl herbarium nr. 17: III, 5!]; ISOTYPE: B Gonolobus broadwayae Schltr. in Urban, Symb. Antill. 7(3):340. 1912. Tyre: TeiniDap: Lopez 2419 ( , designated by Krings 2008b: Herbaceous perennial vines. Latex unknown, presumably white. Stems pubescent, both short, capitate- glandular, and longer, sharp, eglandular trichomes present, the latter throughout, spreading, or retrorse internodally, and antrorse-appressed along an indistinct, horizontal ridge between two opposing petioles and/or just above the node, to 1.07 mm long; nodes pubescent, gland field apparently absent. Leaf blades ovate to oblong-ovate, 1.9-10.2 cm x 0.9-6.6 cm, apices gradually acuminate with the acumen narrowly obtuse, to 1.2 cm long, bases deeply cordate, margins entire, evenly strigillose on both surfaces, trichomes sharp, eglandular, to 1 mm long, glandular emergences from the surface apparently absent, colleters 23, 0.9-1 mm long; petioles 0.9—6.8 cm long, pubescent on all sides, capitate trichomes to 0.16 mm long, sharp, eglandular trichomes antrorsely-ascending or —appressed, ca. 0.7 mm long; stipular colleters 2, ca. 0.3 mm long, one borne on each side of the petiole base (rarely on the stem), ca. 0.3 mm long. Inflorescences race- miform, peduncles 0.4-2.2 cm long, capitate, as well as sharp, eglandular trichomes present, the latter mostly antrorsely-appressed or -ascending, distributed throughout, ca. 0.4 mm long; pedicels 1.2-4 cm long, evenly pue da apex to base, capitate-glandular trichomes ca 0.1 mm long, longer, sharp, eg- landular tricl ly-ascending, ca. 0.8 mm long, bracts linear-lanceolate, ca. 5.1-5.5 mm x 0.8-1.1 mm long, caducous, KAN surface glabrous, abaxial surface coarsely pubescent, trichomes sharp, eglan- dular, 0.38-0.4 mm long, antrorse. Calyx lobes 5, ovate, 4.1-8.4 mm x 1.9-5.5 mm, apices obtuse to rounded, margins sparsely glabrous or ciliate, abaxial surface sparsely pubescent at the base and glabrous toward the apex, trichomes antrorsely appressed or -ascending, to 0.3 mm long; colleters 1 per sinus. Co- rolla lobes 5, linear-lanceolate or lanceolate, 12-22 mm x 4.3-6.7 mm, slightly lobed at the base, a glandu- lar swelling frequently present in the sinus, adaxial surface glabrous, abaxial surface sparsely pubescent with coarse, sharp, eglandular, antrorsely-appressed or -ascending trichomes, trichomes ca. 0.25 mm long; faucal annulus (corolline corona or Ca) interrupted, a raised bump or indistinct ridge opposite eac h corolla lobe sinus, pubescent or glabrous; gynostegial corona of iuen staminal (Cs) and ea (Ci) parts, prostrate-undulating, single, secondary nectaries in i l position absent or at least not formed into distinct bumps or mounds; anther guiderails without ME laminar dorsal anther appendages (Cd) 1.3-1.5 mm wide, truncate to rounded; style-head ca. 5.4 mm diam, stipe « 0.8 mm long, edentate. Polli- naria: corpuscula ca. 0.23 mm long, pollinia borne horizontally, narrowly ovate, ca. 1.1 mm x 0.4 mm. Follicles appareritly known only from a photo taken by P. and Y.S. Comeau, but this was not obtained on loan. Phenology.—Flowering in March, April, May. Fruiting in March. Distribution and habitat. —1In the Caribbean region, G. rostratus is known only from Trinidad, where it has been found on road banks. Its distribution in South America needs to be re-examined (see discussion below). Etymology.—The specific epithet means “beaked.” Taxon concept mapping.— + G. rostratus sensu Schlechter (1899); = G. broadwayae sensu Schlechter (1912); = G. broadwayi sensu Cheesman (1947). Notes.—Schlechter (1899) mistakenly listed “Gonolobus rostratus Griseb. Flor. (1861'[1864]) p. 420 (nec R.Br)” in synonymy with G. ciliatus Schltr. The lectotype of the latter, i.e., Eggers 5561 (P!; designated by Krings 2008b), is referable to acd ae ata (Vahl) Fontella & E.A. Schwarz (= Gonolobus denticulatus (Vahl) W.D. Stevens, if lato; holotype: von Rohr120, C [IDC microfiche photo: Vahl herbarium nr. 17: III, 11); js. 2008b). For two reasons it seems clear that Grisebach (1864) did in fact mean G. rostratus as based on Cynanchum rostratum Vahl, rather than *Matelea" denticulata, when he penned his description. First, he noted the corolla lobes of “his” G. rostratus to be lanceolate-linear. Corolla lobes of Krinas, $ wm 7 AK Fic. 1. Ink Al Jig Bi. 0} A i f th ll i , and style- head. E. Detail of f; | view). Bii. 0 fl ( baxial view). C. Calvcine colleter. D. Faucal i T i , £ J nt A Life 1: . I EG im segments (style- head removed). Side view vof gynostegial c corona, b Styl I . K. Pollinarium. A, C-K based on Britton & Hazen 1601 (GH). Bi-ii bed on ede 67 faucal annulus of corolla; Ci = interst corpusculum; C V» y) y y PVE. El 80 Journa “Matelea” denticulata are broadly elliptic to ovate, whereas those of G. rostratus are linear-lanceolate to lan- ceolate. Second, Grisebach (1864) cited a plate (i.e., t. 7) in Vahl's Icones illustrationi plantarum America- narum (1798), which is so strikingly similar to the holotype of C. rostratum at C, as to leave little doubt that it must have been based on it. The ovate calyx lobes of the mature flower are clearly visible in the illustra- tion. A contrasting illustration of C. denticulatum (basionym of “M” denticulata) is found on plate 8 of the Icones (Vahl 1798). Perhaps following Schlechter (1899), Cheesman (1947) also mistakenly considered “Gonolobus rostratus R.Br. sens. Griseb Fl. 420" synonymous with the entity today known as “Matelea” denticulata, by submerging the former name under a synonym of the latter: Matelea viridiflora (G.FW. Meyer) Woods. Based on Chees- man's description (e.g., *cal[yx] lobes lanceolate,” "corolla [...] lobes ovate,” “anthers without dorsal append- ages”), his concept of Matelea viridiflora (G.FW. Meyer) Woods undoubtedly corresponds to “M.” denticulata and not Gonolobus rostratus sensu Grisebach (1864) and (Vahl) Schult. as discussed above. The confusion regarding the application of names appears to extend beyond the Caribbean region and affects our understanding of the ranges of G. rostratus, G. tobagensis, and potentially other species. Fontella & Schwarz (1981) proposed that the following names (and combinations based on them) be added to the synonymy of G. rostratus: Cynanchum rotatum Vell. (lectotype: Fl. Flum. Icones t. 79!, designated by Fon- tella & Schwarz 1981), Gonolobus patens Decne. (holotype: Martius s.n., M, n.v. [apparently seen by Fontella and Schwarz, but not relocated by M in searches in Fall 2008]), and G. oxyanthus Turcz. (holotype: Funck 2, G!, PD). As a consequence of this decision, the name G. rostratus, has now been applied widely to plants from Tobago to Brazil, Peru, and Argentina (e.g., Fontella & Schwarz 1981; Morellato & Leitáo-Filho 1999; Hechem & Ezcurra 2006). Plants from West Africa— considered recent introductions (D. Goyder, pers. comm.)— have been referred to G. rostratus, as well. However, based on a preliminary survey of specimen images available to me, including three African collections, it appears that G. rostratus may exhibit a much nar- rower range than presently considered, if not be endemic to Trinidad. The majority! of northern South American and African material available to me bearing the name G. rostratus, G. patens, or G. oxyanthus exhibits calyx lobes with acute to acuminate apices, whether linear-lanceolate or ovate, rather than with obtuse to rounded apices as in G. rostratus (e.g., Argentina: Cabrera et al. 32105, SI; Argentina: Vanni et al. 83, K; Argentina: Venturi 9677, K, SI; Bolivia: Wood €- Goyder 16744, K; Ghana: Merello et al. 1662, K; Ivory Coast: Fosberg 40488, US; Paraguay: Hassler 227, SI; Venezuela: Aristeguieta 6505, US; Venezuela: Funck 2, G, P; Venezuela: Pittier 11787, US). In general appearance this material seems much closer to G. tobagensis, al- though I am not convinced that the latter is the correct name nor that a single taxon is represented. Based on the protologue, Gonolobus patens, for instance, also exhibits similar linear-lanceolate to lanceolate calyx lobes (Decaisne 1844). The holotype of G. patens (Brazil, Martius s.n.) resided at M according to Decaisne's (1844) account, but was not recently relocated (F. Schuhwerk, pers. comm.). Fournier (1885) apparently saw this specimen—or at least cited it. He described the calyx lobes of this species as ovate and provided an illustration (t. 94) depicting calyx lobes ovate to lanceolate, but with acute apices, not obtuse as in G. rostratus (Fournier 1885). Neither Decaisne (1844) nor Fournier (1885) recognized G. rostratus from Brazil. Fontella and Schwarz (1981) were apparently the first to associate G. patens and G. rotatus with G. rostratus. Unfortunately, a complete survey of South American and African entities to which the name G. rostratus has been applied is beyond the scope of the present study and limited by the lack of availability of the type of G. patens. It appears that either G. rostratus displays considerable morphological heterogeneity or that mul- tiple taxa are represented. It is hoped that the key and descriptions provided here will contribute to this much needed survey. Additional specimens examined: Hab. in Ins. Carib., Anonymous s.n. (L). Trinidad. Apr [18?]74, O. Kuntze 881 (E); Morne Cocoa Road, bank, 9 Apr 1920, N.L. Britton and T.E. Hazen 1601 (GH, K); Gaura Old Rd., 19 May 1937, Cheesman 13168 (TRIN); 1877-80, A. Fendler 637 (BM, E); Heights of Aripo Rd., close to small ravine or close to Rhapsey's Estate, Mar 1987, F. Moreau s.n. (TRIN); Mar [18?]85, H. Prestoe s.n. (K). m . £ D. Quipuscoa 378 (Kl) hac! £f, " I n J r f " ypi 1 mn : " " a Krings, Sy a er Lal *a IC | I TAL g 81 Fic. 2. Gonolob b g is. A.l l infl Bi 0j fl ( laxial vi ) Bii 0} fl ( baxial vi ) Biii. Flower bud. C Calycine lleter. D. lla, gynost i , and style-head E Style head and lami | l antl pi | g ( pvi ). F. Detail lami G. Pollinariu m. Based on Sandwith 1840 (K) f ial P tvla-head SQUIRT, NU — Iaimndt Ural GEILE! apperid ys, vuU — CUI puotuluti, S5 — Staal yyliU»tegidi LUTUNI. A »ecymie, rU — pummiit, 311 = style nied’. 82 tani i Texas 3(1) 2. Gonolobus tobagensis Urb., Repert. Spec. Nov. Regni Veg. 16:37. 1919. (Fig. 2). Tye: Tosaco: The Widow, a climbing plant, 28 Apr 1913, WE. Broadway 4467 (vecromvre, designated by Krings 2008b: BM!). Herbaceous perennial vines. Latex unknown, presumably white. Stems pubescent, both short, capitate- glandular, and longer, sharp, eglandular trichomes present, the latter throughout, spreading, or retrorse internodally, and antrorse-appressed along an indistinct, horizontal ridge between two opposing petioles and/or just above the node, to 1.3 mm long; nodes pubescent, gland field apparently absent. Leaf blades ovate to oblong-ovate, 3-10.8 cm x 1—5.7 cm, apices gradually or abruptly acuminate with the acumen nar- rowly obtuse, 0.9—1.9 cm long, bases deeply cordate, margins entire, evenly pubescent on both surfaces, trichomes sharp, eglandular, glandular emergences from the surface apparently absent, colleters 2, ca. 0.9 mm long; petioles 1—4.5 cm long, spreading pubescent on all sides, some trichomes capitate, to 0.2 mm long, but mostly sharp, eglandular, to 1.4 mm long; stipular colleters 2, ca. 0.3 mm long, one borne on each side of the petiole base (rarely on the stem). Inflorescences racemiform, peduncles 0.2-4.7 cm long, capitate, as well as sharp, eglandular trichomes present, the latter mostly spreading to somewhat retrorse, distrib- uted throughout, ca. 1.5 mm long; pedicels ca. 3.6 cm long, relatively evenly pubescent from apex to base, capitate-glandular trichomes ca 0.16 mm long, longer, sharp, eglandular trichomes antrorsely-ascending, ca. 1.5 mm long, bracts linear to linear-oblong, 7.6-16 mm x 1.2-2.4 mm, persistent, adaxial surface gla- brous, abaxial surface coarsely pubescent, trichomes sharp, eglandular, ca. 0.6 mm long, antrorse. Calyx lobes 5, linear-lanceolate, 8.5-13 mm x 1.7-2.6 mm, apices acute to narrowly obtuse, margins sparsely ciliate or glabrous, abaxial surface densely and coarsely pubescent at the base and variously toward the apex, trichomes antrorsely appressed or -ascending, to 0.8 mm long; colleters 1 per sinus. Corolla lobes 5, nar- rowly lanceolate to elongate triangular, 13.8-18.4 mm x 4. n 3 mm, slightly lobed at the base, a [deed swelling frequently present in the sinus, adaxial surf te-pubescent on the right half, ab pubescent with coarse, sharp. eglandular, UN, or -ascending trichomes, particularly at the base and variously toward the apex, ca. 0.38 mm long; faucal annulus (corolline corona or Ca) interrupted, a raised bump or indistinct ridge opposite each corolla lobe sinus, pubescent; gynostegial corona of fused staminal (Cs) and interstaminal (Ci) parts, prostrate-undulating, single, secondary nectaries in interstami- nal position unknown (could not be determined from specimens); anther guiderails without appendages, laminar dorsal anther appendages (Cd) ca. 2.9 mm wide, truncate; style-head ca. 4.3 mm diam, stipe ca. 1 mm long, edentate. Pollinaria: corpuscula ca. 0.3 mm long, pollinia borne horizontally, narrowly ovate, ca. 1.3 mm x 0.29 mm. Follicles unknown. Phenology. Flowering in April, October, and November. Distribution.—Gonolobus tobagensis has been previously considered endemic to Tobago and Trinidad, but was recently cited from Guyana, though not the rest of the Guianan Shield (i.e., not in French Guiana, Surinam, Venezuela: Amazonas, Bolivar, Delta Amacuro; Morillo 2007). In light of the issues presented above, a re-evaluation of the application of this name to South American material is warranted. In Trinidad and Tobago, G. tobagensis is known from only four collections (including the type). Its habitat is unclear, but likely mid-elevation rainforests as for other West Indian congenerics. Etymology.—The specific epithet means “from Tobago." Taxon concept mapping.— = G. tobagensis sensu Cheesman (1947). lditi ined. Tobago: Arima (sp?)-Blanchi Rd. 8^ m. ee oo pls ME Mount E GRO: Castara Rd (sp? i f f Main Ridge, 18 Oct 1937, N.Y. Sandwith 1840 (K). Ti 1 Apr 1928, W.E. Broadway 6876 (MO). ACKNOWLEDGMENTS Ithank the curators and staff of the following herbaria for searching, or providing access to or loans of their collections: B, BG, BH, BKL, BM, BOLO, BR, BREM, BSC, BUF, C, CGE, COLO, CR, DUKE, E, F, FI, FLAS, FR, FTG, G, GH, GOET, H, HAC, HAJB, HBG, IA, IJ, ISC, JBSD, JE, K, L, LD, LE, LINN (Linnean and Smithean herbaria), M, MICH, MIN, MO, MSC, NCU, NEU, NSW, NY, O, OXF, P, PH, RSA, 5, U, UBT, UC, Krings, S . er ILL . o me sJ J J'TAL 83 UCWI, UPRRP, UPS, US, USF, TUR, WILLI, WU, Z. I also thank the following persons for graciously pro- viding images of specimens in their collection cited herein: N. Deginani (SD, L. Gautier (G), D. Goyder (K), A. Lehmann (G), Carolina Loup (P), Marc Pignal (P). The Interlibrary Loan Office of North Carolina State University was instrumental in obtaining titles not in the collection. LITERATURE CITED Cheesman, EE. 1947. Asclepiadaceae. In: EE. Cheesman and RO. Williams, eds. Flora of Trinidad and Tobago. Department of Agriculture, Port-of-Spain. Pp. 162-175. Decaisne, J. 1844. Gonolobae. In: A.P. de Candolle, ed. Prodromus Sy tis Naturalis Regni Vegetabilis 8. Treuttel & Würts, Paris. Pp. 591—605. FowrELLA P, J. Schwarz, AND E.A. Schwarz. 1981. Estudos em Asclepiadaceae, XIII. Novos sinónimos e novas com- binacóes. Bol. Mus. Mun. Curitiba 46:1-10. Fournier, E.PN. 1885. Exolobus. In: CEP. von Martius, ed. Flora Brasiliensis 6(4). F. Fleischer, Monaco. Pp. 318-319. Franz, N.M, R.K. Peet, AND A.S. WeakLev. 2008. On the use of taxonomic concepts in support of biodiversity research and taxonomy. New Taxonomy Proceedings of the 5^ Biennial Meeting of the Systematics Association, Cardiff. GnisEBACH, A.H.R. 1864. Flora of the British West Indian Islands. Reeve & Co., London. Hechem, V. AND C. Ezcurra. 2006. Asclepiadaceae. In: LJ. Novara, ed. Flora del Valle de Lerma. Serie Flora 7 (13). Facultad de Ciencias Naturales, Universidad Nacional de Salta, Salta. Pp. 1—64. Houmcren, PK. AND N.H. HoLmren. 1998-present (continuously updated). Index Herbariorum. New York Botanical Garden. [http//sciweb.nybG.org/science2/IndexHerbariorum.asp] KRINGS, A. 2008a. Revision of Gonolobus s.s. (Apocynaceae, Asclepiadoideae) in the West Indies. J. Bot. Res. Inst. Texas 2:95-138. KRINGS, A. 2008b. Index of names and types in West Indian Gonolobinae (Apocynaceae, Asclepiadoideae), includ- ing fourteen new lectotypifications, one weotyoification: and a new combination. J. Bot. Res. Inst. Texas 2:139-163. KRINGS, A., D.T, Thomas, AND Q.- Y. XIANG. 2008. On the generic circumscription of Gonolobus (Apocynaceae: Ascle- piadoideae): Evidence from molecules and morphology. Syst. Bot. 33:403-415. Kunze, H. 1995. Floral morphology of some Gonolobeae (Asclepiadaceae). Bot. Jahrb. Syst. 117:211—238. Lieve, S. AND H. Kunze. 1993. A descriptive system for c analysis in Asclepiadaceae and Periplocaceae. Pl. Syst. Evol. 185:275-284. LIEDE-SCHUMANN, S., A. Rapini, D.J. Goyder, AND M.W. Chase. 2005. Phylogenetics of the New World subtribes of Ascle- piadeae (Apocynaceae-Asclepiadoideae): Metastelmatinae, Oxypetalinae, and Gonolobinae. Syst. Bot. 30:184-195 MontLLATO, PC. AND H.F. Lerráo-FiHo. 1999. Reproductive phenology of climbers in a southeastern Brazilian forest. Biotropica 28:180-191. Morio, G. 2007. Asclepiadaceae. In: V. Funk, T. Hollowell, P. Berry, C. Kelloff, and S.N. Alexander, eds. Checklist of the plants of the Guiana Shield. Contr. US Natl. Herb. 55:208-212. Rapini, A., M.W. CHASE, AND T.U.P. Konno. 2006. Phylogenetics of South American Asclepiadoideae (Apocynaceae). Taxon 55:119-124. Rapin, A, M.W. Chase, DJ. GOYDER, AND J. GrirritHs. 2003. Asclepiadeae classification: evaluating the phylogenetic relationships of New World Asclepiadoideae (Apocynaceae). Taxon 52:33-50. Schlechter, R. 1899. Asclepiadaceae. In: |. Urban, ed. Symbolae Antillanae. Gebrüder Borntraeger, Berlin. Pp. ScHLEcHrER, R. 1912. Asclepiadaceae. In: |. Urban, ed. Symbolae Antillanae. Gebrüder Borntraeger, Berlin. Pp. Var, M. 1798. Icones illustrationi plantarum Americanarum. Hauniae, Copenhagen. BOOK REVIEWS PuiriP V. Werts. 2000. The Manzanitas of California: also of Mexico and the World. (ISBN: 0-933994- 22-2, hbk.). Originally published by the author, Department of Botany, University of Kansas, Lawrence, Kansas 66045, U.S.A. (Orders: check or money orders by mail to Dr. Anke M. Wells, 4638 Bluebird Lane, McLouth, Kansas 66054-3109). $60.00 incl. ph (plus sales tax for Kansas residents), 151 pp., 8 1⁄2" x 11". Published by the author in 2000, this li d edition book is based han 40 f field and hert ji by the late Philip V. Wells (1928-2004, Professor Emeritus of Ecol 1 Evolutionary Biol he University of Kansas in Lawrence Kansas. His lifelong interest in manzanitas started in the late 1950s while Mods at the neu of California in Santa Barbara and culminated in this 151-page work, the most up-to-date and complete treatise of the genus. The book i dee d synonyms, oe and peas for 61 Audios bd is 150 o 146 are pere for the first tim ludi ital scans of und-breaking work. n attractive PUE book di fu xq rir i DER measures 8.5 x 11 inches and is d on acid- AR paper and bound E red e 7 and gold lettering. Approximately 200 copies are available.— Craig C. Freeman, Curator (R.L. McGregor Herbarium), Associate Scientist (Kansas Biological Survey), University of Kansas, Lawrence, Kansas 66047, U.S.A. Davip Yerman. 2007. The Great Cacti: Ethnobotany & Biogeography. (Arizona-Sonora Desert Museum Studies in Natural History) (ISBN 978-0816524310, hbk.). University of Arizona Press, 355 S. Euclid Ave., Suite 103, Tucson, Arizona 85719, U.S.A. (Orders: www.uapress.arizona.edu, ordersQuapress. arizona.edu, 1-800-426-3797). $59.95, 297 pp., 8 3/4" x 11 1/4". The Great Cacti reviews the more than one hundred species of columnar cacti, with detailed discussions of those that have been the most beneficial to humans or are the most spectacular. The book is divided into three Ae parts. The first covers the great cacti in general and looks at their uses, Scotos esito their origins ev p and tax y. The second, and largest of the three Eos ies of the genus. Some genera are e parts, presents the plants by genera, b grouped by geographical region, Stier; are not. x. The third part anli where the ho spots for he has not personally visited but discusses what he considers to be the most impressive, accessible locations where one can see them d has not listed sites where in their native lands tman focuses particularly on northwestern Mexico and the southwestern United States and South America to some degree. He examines the role of each a in uud society, oo ae cacti have provided ae ae e even Tehsious y significant hallucinogens. great i "elo t d L £ qu 1 AA 1 y ielding their pr duct He draws upon! importance of these TE to the native MCA in these regions. The oral uses are listed with each species, generally in a small portion of the first section. TH 11 2n 1 y I 1 col ps, The Great Cacti i lous book, one which I enjoyed just paging through. The use of color d helm th Ith of sci h he displays throughout his narration. Recommended for public libraries, int d student d ] y levels. David Y a 1 001 he Gaul pea pe eke. reas J] frhe televici ies The Desert Speaks, produced by KUAT Television in T d distril ] nationally by American Public Television. His books include The Organ Pipe Cactus and several bool l d pl He is co-editor of Gentry's Río Mayo Plants: The Tropical Deciduous d f North N , also published e the ais of Arizona Press.—Gary L. Jennings, Librarian, Botanical Research Institute of Texas, Fort Worth, Texas 76102-4025, U.S.A J. Bot. Res. Inst. Texas 3(1): 84. 2009 AN OVERVIEW OF THE DIOSPYROS CAMPECHIANA COMPLEX (EBENACEAE) AND DESCRIPTION OF THREE NEW SPECIES Mitchell C. Provance and Andrew C. Sanders UCR Herbarium m of Botany and Plant Sciences Uni y of California, Riverside uic ‘aloo 9252 i -0124, U. SA. ABSTRACT The Di hi lex (El ) is described in detail. Six speci gnized for the complex in Mesoamerica, and one species in South America. Expanded d ipti ided for the three M i ies, and three new species are de- P I p p E P scribed. O ii,i finedt 11 he Pacific Sl f the Si Madre del Sur, Oaxaca, Mexico, RS two ee are om n only hon rum Rica, including Diospyros crotalaria, en is restricted to the Osa Peninsula, and Diospyros haberi, which is widespread in mountainous regions. Members of the complex are remarkable for having fruiting pedicels that tend to hold fast to the calyx, despite being articulated to the receptacle. The fruit itself is weakly attached to the calyx. Photographs, comparison tables, and a key to the species of the complex are provided. RESUMEN Sed ik detalle el lejo Di } El JS i i M ica, y una en Sur América. Se aportan descripciones en extenso para las tres Ao O y se describen tr peci Una de las especies la Si Madre del Sur, Oaxaca, México, nuevas, Dio pyr p : t d sólo de Costa Rica, iicluyendo Diospyros iota lista: que está restringida a la Península de Osa, y Diospyros aver de = oo montañosas. Los eee de d son Mrd hor tener OB a a pesar de estar Las semillas se dispersan pobbi por pájaros. Se aportan fotografías, tablas ans y una clave de especies del complejo. While ining material during continuing studies of Mesoamerican Diospyros (e.g., Provance & Sanders 2005; Provance & Sanders 2006; Provance et al. 2008), building toward production of a comprehensive treatment of the Mexican species, three new taxa belonging to the Diospyros campechiana complex were discovered. Because this complex has only been discussed briefly (Provance et al. 2008), here we provide a more detailed account. The addition of three new species brings the number of included species to seven. One of the new species, D. camposii, is described from Oaxaca, Mexico, based on collections from a restricted area of the Pacific Slope of the Sierra Madre del Sur, and appears to be a local endemic. Two other species, including D. crotalaria and D. haberi are described for Costa Rica. Diospyros crotalaria appears to be a local endemic restricted to the Osa Peninsula, while D. haberi is fairly widespread in the central mountains. Previ- ously described species include D. campechiana Lundell, D. panamense S. Knapp, D. hartmanniana S. Knapp, and D. juruensis A.C. Smith. The distribution of the complex extends from Oaxaca and Veracruz, Mexico southward to Panama, and likely enters Columbia, since D. panamense has been collected very close to its border. The complex definitely occurs in South America because D. juruensis occurs in the Jurua River Basin of the Amazon watershed in western Brazil, where the only known collection was made. We are unsure about the complete status of this complex in South America because most South Ameri- can specimens are unavailable for examination. In addition, the protologs of twenty-five recently described species of South American Diospyros (Wallnófer 1999, 2000, 2003, 2005) completely lack photos or other illustrations, and are devoid of comparisons to allied, or even similar-looking taxa. This makes many of them difficult or impossible to place with confidence. Identification keys were also not provided, and no attempt was made to fit them into existing keys (e.g., MacBride 1959; Cavalcante 1963a; White 1981; So- thers & Berry 1998). While not violations of the International Code of Botanical Nomenclature (Greuter J. Bot. Res. inst. Texas 3(1): 85 —112. 2009 86 Jour tani i Texas 3( et al. 1994 2000), the figure omissions are inconsistent with Recommendation 32B, and the omission of comparisons to allied taxa is contrary to Recommendation 32C of the Code. Comparisons among taxa are made in virtually all of the other relevant literature (e.g., Miquel 1856; Hiern 1872; Milbraed 1927; Gleason & Smith 1933; Sandwith 1931, 1949, 1963; MacBride 1959; Cavalcante 1963a, 1963b, 1966, 1977; White 1981; Sothers & Berry 1998; Sothers 2003), and the illustration of types has been standard procedure for most modern taxonomists. Consequently, a more thorough examination of South American specimens may reveal that additional members of the D. campechiana complex are present there. However, our examination of the literature and the limited South American Diospyros specimens to which we have had access did not reveal any other such species. The species in this complex are so poorly represented in herbaria that from over fifty herbaria that provided specimens of Diospyros, we received only seventy-three specimens (representing fifty collection events) that were referable to species in it. Little or nothing has been recorded concerning bark texture and color, color of the slash, wood characteristics, fruit color, texture and taste, flower odor, and color of the seeds and guard cells in life for most species. Flowers are completely unknown in D. camposii and D. crotalaria, and male flowers are known from only one or a few collections in the other species. Specimens with female flowers are ially rare: we have seen open female flowers only in D. campechiana, and only a single female flower bud in D. haberi. Regardless of sex, flowering specimens typically have few open flowers. Fruits seem to be unknown in D. juruensis. METHODS All of the relevant literature and specimens, including those types we could obtain, were evaluated. Collec- tions were sorted into preliminary groups based on ue and LAB RUE morphological characters, and these eodd were considered taxonomic hypotheses nvestigation. Comparisons were then made | I igned to each group and ee specimens, specimens determined to be consistent wit! i tol ( ially if from near the type locality) and paratypes determined to be consistent with the al description: We E examined specimens in light of floristic and monographic treatments, and the annotations of previous workers. Expanded descriptions have been constructed for pre- viously described species based on the studied herbarium specimens. In this treatment we use ‘lanceolate’ in the sense of Jackson (1916), being broadest near the lower third of the leaf, not at mid-leaf in the sense of Stearn (2000). Descriptions apply to herbarium material except when indicated, or when obviously referring to fresh material. Map coordinates geo-referenced by the authors are given in brackets. Distribution maps were constructed using The Generic Mapping Tools 4.1.4 (Wessel and Smith 2006) accessed through OMC (Weinelt 2006) and amended using Adobe Photoshop (Ver. CS 8.0, Adobe Systems Incorporated, San Jose, CA). Illustrations are by the first author. TAXONOMIC TREATMENT Diospyros campechiana complex: Members of the Diospyros campechiana complex are evergreen trees with medium to large, more or less evenly spaced leaves. Petioles and midribs are occasionally transversely fissured, and frequently have a bloom of minute white crystals (scintillae) that are probably derived from cuticular wax, though their origin has not been studied. Stems and leaves are glabrous to sparsely hairy, and members of the complex have been variously referred to as “black strigillose,” in reference to D. campe- chiana (Standley & Williams 1967), or “minutely strigillose” in reference to D. hartmanniana (White 1978). At low magnification the vestiture appears to be comprised of minute, mostly black, appressed, fusiform hairs. At higher magnification it becomes evident that most hairs are dark red and 2-armed, with a short basiscopic arm, and a longer acroscopic arm. Less often hairs are simple, and in some speci there may be a high frequency of similar, but paler, yellowish, or reddish brown hairs. In addition, some taxa are minutely hirtellous or minutely pubescent, with hairs much smaller than those of the aforementioned strigillose vestiture. Stems sometimes have localized patches of atypical hairs (e.g., upright, aspergilliform, etc.) which D Ic J E n [ hi [| 87 we think represent endophytic fungi. In this paper the term “strigillose” refers to the common vestiture of dark red to black, simple and 2-armed hairs, with the addition of modifiers for special cases (e.g., “reddish brown strigillose”). The abaxial lamina surfaces typically have some conspicuously darkened guard cells (Fig. 1), and in all species except D. campechiana, the stomatal apparatus is recessed, giving the epidermis a black-puncticulate appearance at 20x magnification. The black dots on the abaxial lamina surface of D. hartmanniana that were interpreted by S. Knapp (1997) to be the bases of broken hairs may have been these pigmented guard cells. We suspect that darkening of guard cells is a result of secondary compound degradation in non-functioning cells of older leaves. Male inflorescences are (1—)2—23(-29)-flowered glomerules, compact cymes, paniculiform cymes, or occasionally fasciculate cymes, in leaf axils of young stems. Their axes are 4-angled, and more densely hairy than leaves and stems, though the hairs are otherwise similar. Female inflorescences are solitary or 2—5(-10)-flowered cymes in leaf axils of young stems, and often more or less 4-angled. Flowers have 4—5 calyx and corolla lobes, though rarely a calyx or corolla will have only 3 lobes. Co- rollas are urceolate and white in life, though they turn dark brown upon drying. Male flowers have 10—28 stamens with their filaments fused into 5-14 pairs (comprised of an inner and outer member). Filaments are adnate to the base of the corolla or inserted on the receptacle. Filaments and/or anther connectives are sericeous. The apex of the anther is tapered to rostrate, and consists mostly of minutely muricate connective tissue. The pistillode in male flowers is markedly lobed, with the number of lobes being equal to, or slightly less than, the number of stamen pairs surrounding it. In D. campechiana the filaments of adjacent stamen pairs are separated by these lobes, and in D. panamense and D. hartmanniana the indentations between the lobes resemble impressions of the filaments, suggesting that pressure from filaments against the pistillode during floral development may be responsible for the lobe pattern on the pistillode. Female flowers in D. campechiana poses eat unpalted. staminodes, with several short hairs on the baxial surf; fth tive, and l 1 les with slits. Open female flowers and their internal structures have only been ea in this complex in this one species, though a single mature flower bud has been observed in D. haberi. Despite female flowers being largely unknown in the complex, styles often persist on fruits, and their number (1-2) and morphology are useful in species identification. The female calyx accresces during enlargement of the developing fruit, though not as markedly as in members of the D. rosei complex. The fruiting calyx tube is patent (“saucer-shaped” sensu White 1978; Knapp 1997) with a short basal protuberance. The apex of the fruiting pedicel is 4-angled, + pyramidal in shape, and is jointed to the receptacle within the protuberance at the base of the calyx. Although Diospyros is typi- cally described as having fruits with a persistent calyx (e.g., Wallnófer 2001), members of the campechiana complex have a fruit that is weakly attached to the calyx. Despite having a pedicel with an articulated apex (as in other Diospyros species), in members of this group the fruit often detaches from the calyx, while the connection of the pedicel apex to the base of the calyx holds firm. The fruits of members of the complex are small to medium-sized, ovoid-ellipsoid or depressed-globose to subglobose, and reportedly green, yellow, orange, or red when mature. Comparing fruit shape between species is not straightforward, since it is highly influenced by the number of seeds that develop. Fruits are usually asymmetric (e.g., Figs. 4b, 6b) when seeds only develop on one side. The fruits have only a small amount of flesh, which is reddish and vitreous when dry, though possibly viscous and clear in life. There are 1-4 locules separated by complete septa in the taxa in which fruit anatomy was studied. Based on fruit lobes, this also seems to be the case in the remaining taxa. Each locule contains a lone seed which lacks adherent pericarp. The large locules include considerable space around the seeds, and when dry fruits are shaken they produce a rattling sound. The combination of brightly colored small to medium-sized fruits, that readily separate from the calyx, and contrasting black or very dark inflorescence branches, suggests avian seed dispersal in the complex. We suspect that dispersal in the campechiana complex is effected by removal of the fruit from the calyx, Journal of the Botanical Research Institute of Texas 3(1) > Fe, f) EA! y fM d a Po AN d s deg ae ae we, are v^ MES € Ey a4 - oh Dy - » et Es LM Bv. Pul tT. Fic. 1. Epidermis details in species of the Diospyros campechiana complex. A. Abaxial leaf surface of D. camposii (scale = 0.5 mm) from A. Campos V. & R. Torres C. 825 (MO). B. Abaxial leaf surface of D. crotalaria (scale = 0.5 mm) from R. Aguilar 2877 (K). C. Pigmented and non-pigmented guard cells on the abaxial leaf surface of D. hartmanniana (scale = 0.125 mm) from G. McPherson & N. Hensold 15310 (B-100152291). D. Pigmented and non-pigmented guard cells on the abaxial leaf surface of D. crotalaria (scale — 0.25 mm) from R. Aguilar 2877 (K). D Ic J , Di hs I 89 Dei r T and swallowing of the fruit whole, unlike in other groups (e.g., the salicifolia complex), where fruits are torn open and seed llowed incidentally along with adherent pericarp (White 1978; Provance pers. obs.). Alternatively, the combination of detachment of the fruit from the calyx and air space surrounding seeds in mature fruit may suggest a hydrochorous dispersal syndrome (Kubitzki & Ziburski 1994). Diospyros campechiana Lundell, Contr. Univ. Michigan Herb. 4:23. 1940. (Fig. 2, 3, 15b). Ber: MEXICO. CAMPECHE: Palizada, [18° 15'N, 92? 05'W 3 m], 25-28 Jul 1939, E. Matuda 3843 (HoLoTYPE: MICH; isotypes: E K!) Trees 6—20 m tall, evergreen; trunk up to 45 cm dbh, bark unknown, latex reportedly yellow to orange; young stems smooth to shallowly furrowed, brownish, usually dark, viscid, sparsely to moderately strigil- lose, also densely hirtellous, the hairs minute, light gray to + colorless, shorter than the appressed hairs; mature stems lenticellate, glabrate. Leaves alternate, simple, entire; PEOS E B mm long, minutely hirtellous, sometimes with a few transverse fissures, rounded below, the ey linally wrinkled, often scintillant, + flat above, somewhat canaliculate distally; lamina subcoriaceous, EOM M to oblong-elliptic, sometimes oblong-oblanceolate, 100—285 mm long, 35-90 mm wide, length to width ratio ca. 2.1-3.5 : 1, base acute to attenuate, sometimes Epes: eee decurrent on the petiole, margin revolute, apex acuminate with a rounded tip; 1 ll brown pure to strigillose, guard cells sometimes conspicuously darkened, put not ee pper |! 11 brown to somewhat glossy, usually papillose. Venation ] ; midrib + rounded below, longitudinally wrinkled, depressed above, d and minutely firtellous ad strigillose near the base of the lamina; lateral veins 7—12 per side, prominent below, slightly raised above; 3? veins slightly raised below, more so above. Laminar extrafloral nectaries on the abaxial surface only, common, round to elliptic, 0.3-0.5 mm long, minutely rimmed, drying dark green to black. Male inflorescences a single (3)6-23(29)-flowered compound cyme in the leaf axil of young stems (sometimes bordering on being a fascicle of 23) cymes), 5-17(31) mm long, initially paniculiform, but the terminal and often the penultimate units dichasia, axes 4-angled, densely strigillose, also minutely hirtellous, sulcate, subtended by ovate to triangular, concave bracts + 1 mm long; peduncles 1.5-5.5 mm long; pedicels 0-1.5 mm long. Male flowering calyx sub- crateriform (urceolate prior to anthesis), with a short basal protuberance, exterior drying dark brown to black, strigillose, papillate, interior glabrate to locally strigillose, 272.4 mm wide, tube + 0.7 mm long, lobes 4—5, depressed-orbicular to widely ovate, 0.8-1 mm long, 0.8-1.4 mm wide, sometimes emarginate, often slightly asymmetric, suberect; male corolla white in life, drying brown, glabrous except for some minute papillae, tube urceolate, 1.2-1.7 mm long, + 2.3 mm wide, lobes (324(—5), 1-1.6 mm long, 1.4-1.9 mm wide, depressed-orbicular to widely ovate, sometimes emarginate, often slightly asymmetric; stamens (12-)16(-20?), with Whitefoord and Knapp (1998—onward) reporting 12-16 stamens and Pacheco (1981) reporting 16 stamens, these inserted on the receptacle, fused near the base of their filaments into (6—)8(-10?) pairs, the outer stamen larger, bowing inward at the middle, anthers yellow, sometimes with minute red spots, lanceolate, outer anthers 1-1.2 mm long (including the connective), sericeous on the abaxial surface along the connective and sometimes on the adaxial surface, hairs yellow to reddish black, inner anthers similar except smaller, less hairy, and the hairs mostly on the adaxial surface, the apex (connective) tapered, minutely muricate, filaments 0.2-0.8 mm long, sericeous near the base of the anther; pistillode rotate, markedly 8(-10)-lobed, 0.7-1 mm wide, glabrous, with 2(-1) broadly rostrate apical structures. Female inflorescences 1-3 compound cymes in leaf axils on young stems, paniculiform, but the terminal units dichasia, (123-5(-10)-flowered, though no more than three fruit seen on a single inflorescence, 5-8 mm long, axes 4-angled, strigillose, also minutely hirtellous, sulcate, subtended by persistent bracts similar to those of male inflorescences; flowering peduncles 0.5-2 mm long; flowering pedicels 0.5-2 mm long; fruiting peduncles 4-6 mm long; fruiting pedicels 2-5 mm long, with a 0.4—0.5 mm long, 4-angled, pyramidal apex, only the distal 0.2-0.3 mm forming a joint within the basal protuberance of the calyx. Female flowering calyx campanulate-infundibuliform to crateriform (urceolate prior to anthesis), with a short basal protuberance, exterior drying dark brown to black, moderately strigillose, minutely papillate, ca. 2.5 mm wide, tube 1.2-1.5 mm long, lobes 4-5, depressed orbicular, (1.121.7-1.9 mm long, 1.5-2.5 mm ES Fic. 3. Diospyros campechiana. A. Male flower (C.L. Lundell & E. Contreras 20739, MO). B. Pistillode (C.L. Lundell & E. Contreras 20739, MO). C. Female flower (L. Pacheco & J.J. Calzada 20, XAL). D. Female flower (L. Pacheco € J.I. Calzada 20, XAL). L Das H ID El A p, ET, ” I £a "^f 92 Journal of Texas 3( wide, erect; female flowering corolla white in life, drying brown, tube 0.9 mm long, 1.3-1.5 mm wide, slightly constricted distally, throat 1.2 mm long, semiglobose to urceolate, 2.3-2.5 mm wide, lobes 4—5, widely ovate, 1.7—1.9 mm long, ca. 1.6 mm wide, erect (L. Pacheco & J.I. Calzada 20) or spreading slightly (W.E. Harmon & J.A. Fuentes 5790); pistil + globose, glabrous; style 12.1 mm long, glabrous, the basal portion 4-angled, divided from the basal 1/3 to 1/2 into two ascending branches (style and stigma [see below] details based on those seen in flowering material and persisting on fruit); stigmas minute, undulat- ing irregularly along the abaxial perimeter of the style branches; staminodes 8, with several short hairs on the abaxial surface of the connective, antherodes lanceolate, with slits. Fruiting calyx not markedly accrescent, relatively thin (not fleshy), drying very dark, exterior glabrate to strigillose, minutely glandular punctate (punctae developing from the minute papillae of flowers), tube patent, with a basal protuberance 0.8-2 mm long and 1.5-2 mm wide, the tube 3.5-6 mm wide between opposing sinuses, interior glabrate to moderately strigillose, sometimes minutely glandular punctate, scattered clavate glandular hairs some- times present, a circular band of short reddish black hairs sometimes pnus at the base, lobes spreading, (222.5-4 mm long, (324—5(-6) mm wide, depressed orbicular, ly revolute, ciliate, adaxial surface glabrate to moderately strigillose. Fruit glabrous, (10213- 17 mm long. 10-17 mm wide, ovoid-ellipsoid, often asymmetric, locules 1—4, separated by complete septa; flesh sparse to negligible, red and vitreous when dry, unknown in life, though likely viscous and clear; epidermis probably smooth in life, usually wrinkled when dry, color progressing from green to yellow to orange with maturity (rarely reported to be red), orange to dark brown when dry; seeds 1—4, reddish brown to brown, loosely contained in the fruit, ovoid to ellipsoid, the shape depending on the number of seeds that develop, circular in x.s (when one seed develops), or with 1—2 flat surfaces (when 2-4 seeds develop), 9.5-12 mm long, 5.7-8.4 mm wide, the dorsal surface with a prominent vascular strand, texture rugulose-foveolate, hilum apical on a minute protuberance. Diospyros hiana is probably most closely related to D. camposii. Separation of these taxa is discussed under the description of the latter. Generally, D. campechiana is an easy species to identify. In addition to the geographical separation of this species from other members of the complex, the narrow oblong leaves are distinct, being reminiscent only of D. crotalaria. However, the minutely hirtellous vestiture of the stems, petioles, midrib, and inflorescence seems to be diagnostic. Fruits in fragment folders that have retained their calyx often have the broken tip of the pedicel apex lodged inside the protuberance of the calyx. Thus, in many cases, fruits and calyces that broke away as a unit during the collection process, did not separate from the mother plant at the point of articulation of the pedicel and receptacle/calyx, but rather the pedicel broke near the opening of the cavity in the protuberance at the base of the calyx. It seems likely that pressure from filaments during floral development creates the lobed pattern on the pistillodes of species in this complex. Because we observed mostly 8-lobed pistillodes in this species, we surmise that the stamen number is id is 16. Distribution and ecology.—Di ana occurs in lowland tropical forests associated with wet- lands and riparian areas of major river m and their tributaries between sea-level and 270 m elevation in the states of Tabasco, Campeche, Chiapas, and Veracruz, Mexico, and Guatemala (Fig. 5). Specimens po MEXICO. Camrecne. Mpio. Palizada: 20 km de dade i e Au a A Pil 07' 19"N, 92? 07' 15"W, [0-4 mi, Dec 2000, P. Sima et al 2555 (MO). CHIAPAS Mpio Ocosingo: a3 Corozal, I Rio Usumacinta, selva alta Rd [near 16? 49'N, 90? 53'W], 120 m, 17 Aug 1984, E. Martinez S. 7326 did Tarasco: “Habitat D margins fluvii Gonzalez,” [18° 15'N, 92° 55'W, 10 m], 8 May 1889, J.N. Rovirosa 482 (K). Mpio Jalapa: 41 pa-Tacotalpa, potrero, [near 17? 50'N, 92? 48'W], 10 m, 22 Nov 1983, F. Ventura A. 20785 (MO, XAL); Mpio. Comalcalco, San Cayetano, potrero, [18° 22'N, 93? 13'W], © m, F. Ventura A. 20796 (MO, XAL). Mpio. Jalapa-Tacotalpa: en el cerco de un potrero, 10 m, 23 Nov 1983, R. Curiel A. & M.A. Guadarrama O. 108 (CHAPA, XAL). Mpio. Macuspana: 8 km de la entrada de Macuspana hacia Escarcega y 1.5 km al N, s FREE asociado con Pachira, Bactris, [near 17° 41'N, 92° 34'W, 45 a 25 Aug 1981, M.A. ue Á. & S. Zamudio 358 (XAL). M cajuca: Tucta, a 2 km de Incunac median caducifolia, 2°, gado, [18° 12'N, 92? 59'W], 26 m, 6 Oct 1978, J.I. Calzada 4906 (XAL). Veracruz. Mpio. Ignacio de la Llave: [probably the town of Ignacio de la Llave], acahual, [18° 43'N, 95? 58’W], 50 m, 13 Nov 1967, G. Martínez C. 1555 (USF); 500 m de d pieza la desviacion Villa Nueva-Zacate Colorado, n AC [| : n: hi I 93 en potrero, 18° 43'N, 95° 59'W, 50 m, 29 Apr 1981, L. Pacheco & J.I. Calzada 20, 21 (XAL). Mpio. Las Choapas: Las Choapas, orillas del Rio Playas, cerca de Abasal (one sheet indicates “Rio Playas, arriba de Abasal”), [near 17° 45'N, 93° 57'W], 30 m, 4 Aug 1984, Miguel Chazaro & Luis Robles 3081 (XAL-2 sheets). GUATEMALA. ALTA VERAPAZ: Sebol, along Rio Sebol, in high forest, [15° 48'N, 89? 57', 153 m], Aug 1964, Elias Contreras 3354 (F, MO); Sebol and vicinity, [15° 48'N, 89? 57', 153 m], Aug 1964, Elias Contreras 5355 (1J, F digital d Chahal, bordering Rio Chiyú, 1km W, in low forest, [near 15? 50'N, 89° 34'N], 29 Oct 1968, Elias Contreras 8032 (F, MO). PETEN Cumbre, is 142 of Cadenas Road, in low forest on wet land, [16? 05'N, 89? 21'W, 270 m], 24 Sep 1966, Elias C 220 (F); Rio Pasion, Al ifi in high forest on bank of river Pd 30'N, 90? 32'W, + 380 ft], 8 Feb 1964, - L. Lundell e (F, If, MO); El Rosario, the FAO-FYDEP Camp at S h li laguna, [16? 31'N, 90? 10'W], 1 19 Jun 1971, W.E. Harmon eLA.F 5790 (UMO); La Cumbre, caserio "Sapurul", bordering a “Sapurul”, in zapotal x a forest, Ade 16? 05'N, 89? 21'W, 270 m], 7 April 1977, C.L. Lundell & Elias Contreras 20739 (WIS, MO); La Cumbre, Pusila Road, 5 km N, in acahual, [16° 07'N, 89? 21"W, 250 ml, 19 Aug 1976, C.L. Lundell & Elias C 20224 e UCR); Brecha Chinajá, 12 km from Laguna Petexbatun, high forest, [16° 18'N, 90° 08'W, 540 ft], 20 May 1965, Elias Contreras 5401 (MO). Diospyros camposii M.C. Provance & A.C. Sanders, sp. nov. (Figs. la, 4, 15c). Tre MEXICO. Oaxaca. Mpio. San Jerónimo Coatlán: 41.5 km al SW de San Jerónimo C.[Coatlán], brecha a Progreso, bosque de pino-encino alterado, suelo amarillo arcilloso, 16? 10'N, 96? 59'W, 1550 m, 29 Nov 1990, A. Campos V. 3452 (HoLotyre: CHAPA!; isotype: MEXU) ae D. n anm oL similis, sed yen IBS, jum ce eee crassiore, intervallo grandiore inter sinos oppositos, lobis fructibus, et st ibus, caulibus, petiolis, costa inflorescentibusque non hirtellis, TUM breviore et marginis minute porcatis, sedi non revolute. ten: Trees 8-25 m tall, pr obably evergreen, t | ] ; stems brownish, shallowly furrowed when young, sometimes smooth, sparsely strigillose, becoming lenticellate and scaly in older stems. ET alternate, simple, entire; petioles 8—14 mm long, with a short marginal wing distally, y fissured, sparsely strigillose, epidermis scintillant; rounded below, epidermis longitudinally wrinkled, + flat above, somewhat canaliculate distally; lamina chartaceous to subcoriaceous, widely ovate to elliptic, 55-152 mm long, 33-69 mm wide, length to SUE ratio ca. 1.5-2.5 : 1, base acute to shortly-attenuate, sometimes oblique, decurrent on the petiole inutely ridged, apex acuminate with a rounded tip; lower lamina surface dull brown, sparsely ls the fiae ] near the leaf base, many guard cells darkened and often recessed; upper lamina surface dull brown, slightly paler than the lower surface, smooth, glabrous. Venation camptodromous to eucamptodromous; midrib subprominent below, + rounded, sparsely strigil- lose, the epidermis longitudinally wrinkled, somewhat raised above with a narrow central canal, glabrous; lateral veins (526—9(-11) per side, fine, but prominently raised below, flush or slightly raised above; 3° veins fine, conspicuously raised below, barely raised above. Laminar extrafloral nectaries on abaxial surface only, uncommon, mostly in the proximal half of the lamina, + round, ca. 0.25 mm wide, green to black. Male inflorescences unknown. Female inflorescences not seen with flowers, apparently cymes, 1-3 in leaf axils of young stems, each 13-flowered, though no more than a single fruit seen on any inflores- cence, axes subterete, moderately to densely strigillose, fruiting peduncle 0.5-2 mm long; fruiting pedicel 1.5-5 mm long, with a 4-angled pyramidal apex jointed to the receptacle within the basal protuberance of the calyx, subtended by 23 lanceolate bracts, ca. 2mm long. Female flowers unknown, except the style at least 2.4 mm long and 0.5 mm wide, hairy, the hairs off-white to reddish, up to 0.7 mm long, + straight, + terete, divided into two appressed, ascending branches at least 0.7 mm long (style details based on those persisting on fruit). Fruiting calyx thick, but not fleshy, exterior moderately strigillose, the epidermis viscid, tube patent, with a basal protuberance 1.3-2 mm long, 6-7.5 mm wide between opposing sinuses, interior with numerous minute, reddish hairs, lobes 4—5, reflexed, roundish, the margins weakly revolute, 4—6 mm long, 4-8 mm wide, interior glabrescent, viscid. Fruit ovoid-ellipsoid, often asymmetric, (15-)18-24 mm long, (11213-15(-17) mm in diameter; hairy near the apex, sometimes with hairs similar to those of the style scattered over the distal third, locules unknown, probably 1-4; flesh unknown; epidermis + smooth, yellow to orange in life when mature, brownish yellow in herbarium material; seeds not observed, loosely contained (audibly so) in the locules of dried specimens. Two of the paratypes (A. Campos V. 1017, MO and A. Campos V. & R. Torres C. 825, MO) were originally identified as D. campechiana, but annotated by R. Durán “No es Diospyros campechiana” in 1995. These specimens were later identified as D. campechiana by B. Wallnófer in 1999. While this species is clearly a n IC J , Di he I 95 close relative of D. campechiana, that taxon is found primarily in lowland tropical forests associated with wetlands and riparian areas of major river systems between sea-level and 270 m elevation in the states of Tabasco, Campeche, Chiapas, and Veracruz, Mexico, and Guatemala, while D. camposii has been found only in Oaxaca in cloud forest between 1200 and 1550 m elevation. There is approximately a 300 km disjunc- tion from the nearest populations of D. campechiana in the coastal lowlands of Veracruz to the mountains in Oaxaca where D. camposii occurs. Diospyros camposii can be separated from D. campechiana by its shorter, widely ovate to elliptic lamina with a minutely ridged, but not at all revolute, margin. Although sometimes elliptic, the lamina of Diospyros campechiana is typically longer and more oblong in outline, and the margin is distinctly revolute. While both species are strigillose, D. camposii lacks the additional hirtellous i always present in D. campechiana. Diospyros camposii also has a larger, thicker, fruiting calyx, with reflexed sepal lobes, in contrast to the thinner calyx with spreading lobes of D. campechiana. Finally, the fruits of D. camposii are longer and have a hairy apex, with the pubescence extending onto the style. The fruits and style of D. campechiana are glabrous. Distribution and ecology.—This species is currently known only from a small area on the Pacific Slope of the Sierra Madre del Sur (Fig. 5), where it is reported to occur in cloud forest (bosque mesófilo) in asso- ciation with Alnus acuminata, Arbutus xalapensis, Chiranothodendron pentadactylon, Clethra mexicana, Cornus disciflora, Dendropanax arboreus, Ficus pertusa, Inga eriocarpa, Ocotea helicterifolia, Oreopanax xalapensis, Or- mosia oaxacana, Parathesis brevipes, Picramnia lindeniana, Quercus candicans, Siparuna andina, Ulmus mexicana, Viburnum elatum and Zinowiewia integerrima (Campos-Villanueva & Villaseñor 1995). The area has a large number of canyons with permanent streams, and soils consisting of a deep layer of organic matter over red clay (Campos-Villanueva and Villaseñor 1995). The holotype was collected in pine-oak forest. Etymology.—This species is named in honor of the Mexican botanist Alvaro Campos-Villanueva, who collected the holotype, and participated in all known collections of the species. He is also an author of a flora of the Municipio de San Jerónimo Coatlán in the Sierra Madre del Sur. Specimens examined. MEXICO. Oaxaca. Mpio. San Jerónimo Coatlán: 41 km al SW de San Jerónimo Coatlán, cañada de bosque dp suelo amarillo dea 16? 10'N, 96? 58'W, pid m, 17 Jan 1988, A. Campos V. 1017 (MO); 12.3 km al N de Piedra Larga, Pr E de Piedral arga, carr. a Miahuatlán, | 5fil bosque de pino, en cañada, 16? 09! 00'N, 97° 01' 00"W, 1200 m, 16 Dec 1987, R. Torres C. & A. Campos 10895 (MO). Mpio. Piedra Larga: 12.5 km al NE de Piedra Larga, sobre el camino a El Progreso, bosque mesófilo, suelo negro, 16? 10'N, 97? 01'W, 1300 m, 15 Dec 1987, A. Campos V. & R. Torres C. 825 (MO). Diospyros crotalaria M.C. Provance & A. E nae qur nov. (Figs: 1b, 1d, 6, 15a). Tree: COSTA RICA. Puntar- ENAS. Cantón de Osa: R.E Golfo Dulce, Pení uemado, Sector Oeste, 08°, 41' 10"N, 83° 35' 10"W, 400 m, 9 Sep 1992, J. Marín 512 (HoLorYrE: MO-5316690!; ISOTYPES: “CR K!). ous D. d S. Knapp a sed e ind bip MD d e oen LERNEN stylibus glabratis, gracilioribus, Trees 10-20 m tall, eats evergreen; trunk recorded up to 15 cm dbh, bark reportedly thick; young stems reddish brown, somewhat compressed, shallowly channeled, or with some minor furrows, glabres- cent to sparsely strigillose, the hairs pale yellow to dark red or nearly black; mature stems terete, rugose, densely lenticellate. Leaves alternate, simple, entire; petioles 4—9(-10) mm long, subterete, epidermis often scintillant, atropurpureous, rugulose, the wrinkles often transverse, sometimes developing into fissures; glabrous to sparsely strigillose, the hairs slightly flexuose, ascending, and pale yellow, or straight, appressed, and dark red to nearly black; shallowly V-grooved above, the groove sometimes with an additional narrow channel down its center, glabrous below, sometimes with a few small longitudinally oriented lenticels; lamina subcoriaceous, oblong to oblong-obovate, sometimes narrowly ovate, 74-150 mm long, 26-49 mm wide, length to width ratio ca. 2.53 : 1, base acute to subacute, margin flat, apex acuminate with a rounded tip; lower lamina surface reddish brown, slightly darker than above, dull, glabrescent or with scattered minute, 0.2-0.5 mm long, straight, appressed, dark red to nearly black, sometimes pale yellow hairs, epidermis sometimes with aggregates of opaque light-colored excrescences, guard cells often very MEE. -94° -92 -90' D. campechiana = O > è $ Mexico km S j | , 0 50 100 A | D. camposii asd 2008 Sop 1222-5543 OMC- artin Wainot Fic. 5. Distrib ti f Di ij d D. d Guat | dark and recessed, upper i f; ish green, not glossy, though slightly more shiny than below, glabrous, epidermis densely papillose, the EM angular. Venation eucamptodromous; midrib narrow but prominently raised below, rounded, longitudinally wrinkled, glabrate to sparsely strigillose; shallowly V-grooved above, glabrous; lateral veins 6—9(-10) per side, fine, raised below, forming inconspicuous loops t laterals, flush or slightly depressed above, papillae absent; 3? veins fine, inconspicuously raised Deu Avperceotible above, or barely so, and then depressed. Laminar extrafloral nectaries on the abaxial surface only, relatively common, round to elliptic, 0.2-0.5 mm wide, black. Male inflorescences unknown. Female inflorescences cymes, 1 per leaf axil on young stems, 1-2-flowered, producing 1-2 fruit, axes « terete, densely strigillose, the hairs flexuose, pale yellow to dark reddish brown, subtended by 2-3 opposite pairs of strigillose, persistent, deltoid bracts; fruiting peduncle 0.5-1.5 mm long; fruiting pedicel 1-2 mm long, with a 1 mm long 4-angled apex that forms a joint within the basal protuberance of the calyx. Female flowers unknown, except styles (2) erect, divided + to the base, 4-angled, at least 0.9 mm long, 0.2 mm wide, essentially glabrous, though a long solitary hair was seen in the style branch fork on two of the known fruit (style details based on remnants from fruit). Fruiting calyx relatively thin (not fleshy), reportedly green in life, drying very dark, exterior strigillose, the hairs pale yellow to dark red, epidermis Vee tube Pa with a ca. 1.5 mm long basal protuberance, 5-6 mm wide between opposing sinuses, with inute, reddish hairs, lobes 4-5, usually reflexed, ovate, 2.4-3 mm long, 3-4 mm wide, the margins flat to weakly tE with a few cilia near the sinuses, moderately pubescent inside, the hairs minute, appressed, pale yellow to dark red. Fruit glabrous, 13-19 mm long, 15-20 mm wide, depressed globose to subglobose, often appearing 2-4 lobed (when dry) due to conformation of the fruit wall to the shape of the (1-4) locules with developing ovules; flesh exiguous; epidermis smooth, yellow in life when nearly mature, drying orangish-brown, red when mature, drying dark reddish brown; seeds 1-4, brown, loosely contained in the fruit, + ellipsoid, 9-11.5 mm long, with 1 or 2 flat surfaces, ca. 6 mm wide, dorsal surface broadly convex, with a prominent vascular strand, texture rugulose-foveolate. Specimens examined. COSTA RICA. PUNTARENAS. Cantónde Osa: R.F. Golfo Dulce, Península de Osa, Cerro Chocuaco, camino el 98 tani i Texas 3( faro, 08? 43' 20"N, dod e oe m, 28 Dec 1993, R. i eal i P Ki, A Cantón de Golfito Jiménez: Dos Brazos de Río Tigre, C N erro Rincón bradi followi g 1 I hl I 11 creeks), 08? 30' 35"N, 83? 28' 15"W, 782 m, 25 Nov 1990, G. Herrera 4647 (CR, K!, MOD. The specimens of D. crotalaria we examined had not previously been determined to species. Diospyros crota- laria is similar to D. hartmanniana, but can be separated (Table 1) by its oblong lamina, glabrous fruits, and more slender, glabrous, erect styles, which are divided nearly to the base. Furthermore, D. crotalaria has a thinner calyx, and smaller, reflexed calyx lobes. Distribution and ecology.—Apparently endemic to hills and canyons with wet tropical evergreen forest between 400 and 800 m elevation on the Osa Peninsula on the coast of southwestern Costa Rica (Fig. 14). Reports of D. hartmanniana from the Osa Peninsula (e.g., Zamora et al. 2004) may have been in reference to this species. The closest population of D. hartmanniana to the Osa Peninsula, that we know of, is about 75 km southeast in Panama. Etymology.—This species i 1 for the rattling noise of the seeds in the locules of fruit on herbarium specimens. Diospyros haberi M.C. Provance & A.C. MR sp. nov. (Figs. X deis Tere: COSTA RICA. AtAjuErA. Cantón de Upala: P. N. Rincón de la Vieja, Cordillera d 7 km de la casa de Administración, 10? 47' 50"N, 85? 18' 19"W, 1500 m, 6 Jul 1991, G. Rivera 1422 (HoLoTYPE: MO}; ISOTYPES: im Arbor D. hartmanniana S IDE siu 1 lobis caly daxialles prope glabratis, 1.5— mm aces ae oe aia oe ES TA LU E Js t T BLISEO Trees 8-28 m tall, BN evergreen; trunk up to 60 cm dbh, the bark reportedly black with white patches; young stems 4—5 angled, somewhat compressed and channeled, sulcate, shiny, minutely black gland-dotted, sparsely strigillose; mature stems terete, glabrate, the epidermis light to dark green, smooth to half-netted, eventually lenticellate. Leaves alternate, simple, entire; petioles 3.5-9 mm long, + terete, epidermis light green to castaneous or atropurpureous, shiny, sometimes transversely fissured, rugulose below, glabrous to sparsely strigillose, rounded ee idees and sparsely strigillose, channeled, often deeply, glabrous within or with scattered black hairs; i , elliptic, 30-96 mm long, 17-40 mm wide, length to width ratio ca. 1.8—2.7 : 1, base acute to attenuate, margin flat, though thickened, apex acute to obtuse, rarely slightly acuminate; lower lamina surface greenish to grayish brown, glabrate to sparsely strigillose, guard cells often dark, the stomatal apparatus often recessed; upper lamina surface greenish brown, shinier and slightly lighter than the lower surface, glabrate, epidermis markedly papillose, the papillae angular. Venation Doe ATEAN to aid midrib narrow below, subpromi- nent, glabrate to sparsely strigillose, epid ften darkened, 1 lly wrinkled, shallowly V-grooved above, glabrous or with some scattered dark appressed hairs; lateral veins 6-9 per side, fine, raised below, barely raised above; 3? veins slightly raised below, inconspicuous above. Laminar extrafloral nectaries on the abaxial surface only, usually near the midrib and towards the base, round to elliptic, 0.1-0.3 mm wide, dark green to dark brown. Male inflorescences solitary compact cymes in leaf axils of young stems, (1323 -flowered, 6 mm long, axes 4-angled, strigillose, subtended by 1-2 pairs of deltate, 0.7 mm long, 0.6 mm wide bracts; peduncles 0.3 mm long; pedicels 0.3-0.8 mm long. Male flowering calyx crateriform to subrotate, exterior drying dark greenish brown, 3 mm wide, tube 1.5 mm long, exterior sparsely black strigillose, interior unknown, lobes 4, widely ovate, 0.5 mm long, 1-1.3 mm wide, erect, exterior glabrate to sparsely black strigillose, margins densely ciliate, interior glabrous; male corolla white in life, drying dark brown, glabrous, tube urceolate, 2 mm long, 2.6 mm wide, lobes 4, quadrate, 1.4 mm long, ca. 1.4 mm wide; only three flowers were seen, none were dissected, thus stamen morphology is unknown, except that anthers have a slender, tapering connective. Female inflorescences solitary, subsessile flowers in leaf axils of young stems, or rarely 2-flowered compact cymes, but no more than a single fruit seen in an inflorescence; axes 4-angled, sparsely strigillose, with 1—2 pairs of persistent, ovate bracts, 0.8 mm long, 1 mm wide; fruiting peduncles 0—0.5 mm long; fruiting pedicels ca. 0.5 mm long, with a 0.3 mm long URI BOTANICAL GARDEN PESBARIL : MECO MENE NO 5600 Fic. 7 Diospyros haberi. A. 1 £. H. L flower bud (G. Rivera 1422, K). D. (W. Haber & S. Daniel 9895, MO). B. B (W. Haber & S. Daniel 9895, MO) ). C. Female 100 Journal of the Botanical R h Institute of Texas 3(1) TABLE 1 MA pt | E | E y d £ my L L ] f, n J n. D. panamense D. hartmanniana D. crotalaria D. haberi Fruiting calyx thick thick thin thick Fruiting calyx tube width 7-10 mm 5-6 mm 5-6 mm 5.1-6.6 mm (distance between opposing sinuses) Fruiting calyx lobe length 2./-5 mm 2./-3.8 mm 24-3 mm 1.5-2 mm Fruiting calyx lobe width 4.5-9 mm (4-)4.4-5.5 mm 3-4mm 3.8-5.3 mm Fruiting calyx lobes spreading to t spreading reflexed + spreading slightly reflexed Fruit length up to 33 mm (14-)17-20 mm 13-19 mm 14-18 mm Fruit vestiture glabrous when usually hairy, glabrous glabrous when mature, hairy at apex and base when immature especially at the base and apex mature, hairy at apex immature Style 1, hairy, minutely 1, hairy, divided 2, + glabrous, divided 1(-2), glabrous, deltoid-bifid nearly to the base, to the base, parallel shortly deltoid-bifid, spreading in the and erect n cleft on form of a 'V' e side Stem, petiole and midrib minutely curved- not curved-acicular not curved-acicular not curved-acicular acicular pubescent pubescent pubescent pubescent Lamina shape oblong-elliptic to elliptic, apex oblong to oblong- elliptic, apex elliptic, apex minate obovate, sometimes rounded acuminate narrowly ovate, apex acuminate Lamina size (96-)127-240 mm (50-)64-163 mm 74-150 mm long, 30-96 mm long long, 45-107 long, 25-70 26-49 mm wide 17-40 mm wide wide mm wide 4-angled pyramidal forming a joint within the basal protuberance of the calyx. Female flowers known from a single flower na and style remnants on immature fruit. Female flowering calyx subcampanulate prior to anthesis, sparsely strigillose, 6.3 mm wide, tube 2.5 mm long, lobes (3-34, thick (ca. 0.8 mm at 1/4 length), + evenly tapered to the apex, depressed-orbicular, 1.2-1.8 mm long, 3.9-4.1 mm wide, some- times emarginate, asymmetric (hooking left viewed abaxially), ciliate on the left margin; female flowering corolla drying dark brown, tube ca. 2.2 mm long, with some scattered dark two-armed hairs, lobes 4, ca. 1.3 mm long, shape and width indiscernible, with some indistinct papillae; pistil unknown; styles 122), 0.8-1.2 mm long, 0.7-1 mm wide, angular, bifid, the branches short, 0.2-0.3 mm long, deltoid, style often grooved on one side, rarely notched apically or grooved on both sides, glabrous, though strigillose near the base on the apex of the fruit (style and stigma [see below] details based on those persisting on fruit); stigmas depressed orbicular, 0.3 mm long, 0.7 mm wide; staminodes unknown. Fruiting calyx barely accrescent, thick, probably fleshy and green in life, drying brown to black, exterior sparsely strigillose, tube patent, with a basal protuberance 1.2-1.5 mm long, 5.1-6.6 mm wide between opposing sinuses, interior glabrous, or with some scattered reddish hairs, lobes + spreading, depressed orbicular, 1.5-2 mm long, 3.8—5.3 mm wide, margins flat to slightly reflexed, ciliate, glabrous inside. Fruit 14-18 mm long, 10-20 mm wide, depressed globose to subglobose, often appearing 2-4 lobed when dry due to conformation of the fruit wall to the (1-4) locules with developing ovules, glabrous except for some dark hairs at the apex in immature fruit; flesh sparse, reddish, translucent and vitreous when dry; epidermis green in life when immature, becoming orange to red when mature, delicate, golden brown to dark reddish brown and loosely wrinkled when dry; seeds 1—4 (based on fruit lobes), reddish brown, loosely contained in the fruit, probably n AC A , Di h; l 101 wedge-shaped, 14-18 mm long (width indeterminable, and shape uncertain, as the only seed observed had been sectioned longitudinally), surface texture rugulose in a maze-like pattern. Distribution and ecology.—This species is associated with moist and windswept forests between 1100 and 1700 m elevation on the Pacific slopes of the Cordilleras de Guanacaste, Tilarán, and Talamanca, in Costa Rica (Fig. 14). Populations occur near, but as far as known do not overlap, populations of D. hartmanniana in the former two mountain ranges. All of the specimens examined had been recently identified as D. hart- manniana, to which it is doubtless closely related. The most obvious differences between the species (Table 1.) involve leaf apex shape, and style number and morphology. In addition to these characters, the interior of the sepal lobes in male and female flowers of D. haberi are glabrous, while in D. hartmanniana they are densely dark strigillose. The color and brilliance of the lamina is also different in these species. In D. haberi the greenish brown upper surface of the lamina is shinier and slightly lighter than the greenish to grayish brown lower surface. Brilliance is reversed in D. hartmanniana, where the grayish brown upper surface is duller than the reddish brown lower surface. Etymology.—This species is named in honor of William A. Haber, a biologist who has to our knowledge of the biota of Costa Rica and who has documented the presence of this new species at Monteverde. tributed greatly O ya Specimens examined. COSTA RICA. GuANAcAsTE. Cantón de Tilarán: Sa Río Cañas, indes Pacifico, 10? ut 84? d 1200 m, 25 Aug 1989, E. Bello 1124 (MO). PUNTARENAS. C g 217 40"N, 83? 11' 50"W, 1700 m, 24 Sep 1989, G. Herrera 3542 (MO). Cantón de Osa: San Luis, Monteverde Buen Amigo, 10? 16' 00'N, 84? 49' 00"W, 1100 m, 10 Nov 1993, Z. Fuentes & E. Fuentes 565 (MO). Cantón de Puntarenas: Monteverde, Bajo Tigre Reserve, 10? 18'N, 84° 48'W, 1200-1300 m, 3 Apr 1991, W. Haber & W. Zuchowski 10600 (MO); Monteverde, above Quebrada Máquina, along Fonseca, Hotel de Montaña and Savage Farms, 10? 18'N, 84? 48'W, 1100-1300 m, 3 Jun 1990, W. Haber & S. Daniel 9895 (MO). Diospyros hartmanniana S. Knapp, Novon 7:256. 1997. (Figs. 8, 9, 15e). Ter: PANAMA. Chiriquí: near Coast Rican border, ca. 13 road-km from Río Sereno, Finca Hartmann, 1550-1750 m, 08? 50'N, 82? 45'W, 23 Oct 1992, G. McPherson & PM. Richardson 15959 (HoLotyee: BM; isotypes: MO, P Trees 7—20 m tall, rarely a 1.5 m tall shrub, apparently evergreen; trunk up to 72 cm dbh, bark smooth and black, wood oxidizing bright yellow; young stems angular, often compressed, sulcate to channeled, shiny, often minutely black gland-dotted, glabrous to sparsely strigillose; mature stems dark, terete, smooth, glabrate to sparsely hairy, later half-netted, scaly and lenticellate. Leaves alternate, simple, entire; peti- oles 4—10 mm long, subterete, epidermis castaneous to atropurpureous, sometimes transversely fissured, sometimes scintillant, longitudinally wrinkled to colliculate below, shiny, glabrous to strigillose; glabrous above, markedly channeled; lamina subcoriaceous, elliptic, (50-)64-163 mm long, 25-70 mm wide, length to width ratio ca. 2-3 : 1, base acute to attenuate, margin flat, though slightly thickened, apex acute to subacute, rarely obtuse, acuminate, the tip rounded; lower lamina surface reddish brown, glabrate to sparsely strigillose, aggregates of opaque off-white excrescences sometimes present, guard cells often dark and the stomatal apparatus often recessed; upper lamina surface grayish brown, duller than the lower surface, glabrous, epidermis conspicuously papillose, the papillae angular. Venation brochidodromous to eucamptodromous; midrib narrow below, though prominently raised, rounded, longitudinally wrinkled, glabrous to strigillose, the hairs pale or dark; V-grooved above, glabrate; lateral veins 6—11 per side, fine, raised below, slightly raised to barely depressed above, lacking papillae; 3° veins slightly raised below, indiscernible above. Laminar extrafloral nectaries on the abaxial surface only, common near the midrib and the base of the lamina, 0.1-0.5 mm wide, castaneous. Male inflorescences solitary compact cymes in the leaf axils of young stems, or several near the base of young stems, (12)3—7-flowered, 7-11 mm long, axes 4-angled, densely reddish brown to black strigillose, subtended by several lanceolate to widely ovate, 1-2 mm long, ca. 1 mm wide bracts; peduncles 0.5-1.5 mm long; pedicels 0.5-1.5 mm long. Male flowers known only from buds. Male flowering calyx subrotate, with a basal protuberance ca. 1 mm long, exterior drying dark greenish brown, moderately strigillose, interior densely strigillose, 3.5-4.1 mm in widest dimension, tube 1-1.2 mm long, lobes 4-5, ovate, 0.9—1.2 mm long, 1.3-1.6 mm wide, suberect to spreading, margins ciliate; male corolla white to cream in life, drying dark brown, thick, glabrous, papillate, tube urceolate, 102 J I| £ 4. JOUTM al OF TERR. HORT, KEW, A Fic. 8. D (G. Herrera 5669, K) B. Fruiti (E Bello 1767, MO). C. Im- mature fruits (£. Bello 2442, MO). 103 Fic. 9. Di D) IR Ecninnza etal 224 MM A Male fl T , bud. C, Stamens £ el Dat tai D khi Fr £ T. 104 Journal of Texas 3( 1.6-2.8 mm long, 3.2-4.5 mm wide, lobes 4—5, widely ovate, 2.1-3.1 mm long, ca. 3 mm wide; stamens ca. 27 (based on R. Espinoza et al. 884 (MO) from Costa Rica) adnate to the base of the corolla, the filaments fused most of their length into ca. 13 pairs, comprised of inner and outer members (Knapp [1997] reports 10-12(20) stamens, and illustrates 14 stamens in pairs based on McPherson & Hensold 15303 (MO) from Panama), anthers cream, the inner anthers often larger than the outer, lanceolate, (2.5)3—3.2 mm long (in- cluding connective), golden to reddish sericeous along the connective on both sides, the apex (connective) rostrate, slightly constricted basally, minutely muricate, filaments 1.2-1.5 mm long (fused portion of the filament pair), glabrate, free portion of the filaments 0.5—0.7 mm long, densely hairy, the hairs minute and curly; pistillode obturbinate, 1.4 mm high, 1.2 mm wide, ca. 8-lobed/grooved, the ‘lobes’ narrower than the ‘grooves’, presence of style-like structures indeterminable due to a dense vestiture of 0.5 mm long, red, flat, twisting hairs in the upper half. Female inflorescences unknown with flowers, apparently a 1-3-flowered compact cyme in leaf axils of young stems, with 1-2 fruit developing per inflorescence, axes + 4-angled, strigillose, with ca. 2 pairs of persistent, widely ovate bracts, 1.2-1.8 mm long, ca. 1.5 mm wide; fruiting peduncles 0—1.8 mm long; fruiting pedicels 0.5-2 mm long, with a 1-1.5 mm long 4-angled pyramidal apex, the distal 1/2 forming a joint within the basal protuberance of the calyx. Female flowers unknown, except style divided nearly to the base, branches angled away from one another in the form of a ‘V’, each 0.7-1.3 mm long, ca. 0.5 mm wide, densely reddish brown strigillose basally and where the style branches diverge (style and stigma [see below] details based on those persisting on fruit), stigmas minute, deltoid. Fruiting calyx thick, probably fleshy, reportedly brown in life, drying dark brown to black, exterior strigillose, the hairs pale yellow to dark, sometimes glabrescent, epidermis viscid, interior Oe strigillose, tube patent, with a basal protuberance ca. 1.5 mm long, 5-6 mm wide between opt , lobes 4—5, spreading, ovate, 2.73.8 mm long, (424.4—5.5 mm wide, the margins flat, n ciliate. Fruit (1417-20 mm long, (12-)17-20 mm wide, depressed globose to subglobose, often appearing 2-4 lobed when dry due to conformation of the fruit wall to the (1—4) locules with developing ovules, usually strigillose basally and at the apex; flesh sparse, reddish; epidermis smooth to wrinkled, in life orange to orangish red when mature, light brown to dark reddish brown when dry; seeds 1-4, loosely contained in the fruit, + wedge-shaped, reddish brown and rugulose-foveolate textured, though only immature seeds were seen. Distribution and ecology—This species is associated with cloud and moist forests between 500 and 2200 m elevation, on slopes of the Cordilleras de Guanacaste and Tilarán in Costa Rica, and Talamanca (La Amistad) in Panama (Fig. 14). Populations occur near, but as far as known do not overlap, populations of D. haberi in the former two mountain ranges. Collections examined. COSTA RICA. ALAJUELA. Cantón de San Carlos: La Fortuna, Finca El Jilguero, 10? 26' 35"N, 84? 41' 25"W, 1140 m, 27 Nov 1992, G. Herrera 5669 (K). Cantón de San R R iol. Monteverde, Valle del Río Peñas Blancas, Fila de Toro, Sendero Pipilacha y Capo Tres, [acca. W3 Tropicos 10? 18' 00"N, 84° 43' mw 900—1100 m, 21 Jan 1991, W. Haber (ex E. Cruz) 10632 (MO); Res. Biol. iuri a Eladio’s, 10? 19'N, 84? 43'W, 820 m, 2 Oct 1990, E. Bello 2442 (MO). Cantón de Upala: P.N. Rincón de la Vieja dill te, sendero a la Quebrada Mora, camino a la Colonia Blanca, 10° 46' 32"N, 85° 15' 10"W, 840 m, 30 Nov 1990, G Rivera 920 (K, MO); P.N. Guanacaste, Cordill Estación, San Ramón, Dos Ríos, Sector La Campana, 10? 52! 50"N, 85? 24! 05"W, 550 m, 1 May 1993, R. Espinoza et al. 884 (K, MO); GUANACASTE. Cantón de La Cruz: De Bahia Salinas a Santa Cecilia Hda. El Oro, 10? 59' 26"N, 85° 25' 40"W, 500 m, 5 Sep 1922, R. Espinoza 531 (K, MO); Puntarenas. Cantón de Coto Brus: Zona Protectora Las Tablas, Cuenca Térraba-Sierpe, 08? 58' abe 82° 50' 14"W, 1500-2000 m, 1 Nov 1996, A. Rojas & E. Navarro & E. Alfaro 1757 (M m de Osa: Reserva Biológica M brada Veracruz, Finca Pablo Morales, 10° 15'N, 84° 48'W, 1600 m, 11 Jan 1990, E. ay 1767 (MO); San Luis, Monteverde, Cerro nude camino a Surtubal, 10? 15' 25"N, 84° 47' 20"W, 1100-1200 m, 22 Mar 1994, Z. Fuentes 704 (MO). PANAMA. Chiriquí. Near border with Costa Rica, ca. 13 road km from Río Sereno, Finca Hartmann, 08? 50'N, 82? 45'W, 1400—1800 m, 12 May 1991, G. McPherson & N. Hensold 15310 (B, 2 sheets, CAS); ridges leading to Cerro Pelota, Parque Nacional Amistad, 08° 52-53'N, 82° 44'W, 1700-2200 m, 21 Aug 2000, 5. Knapp & A. Monro 9254 (MO) € Diospyros juruensis A.C. Smith, Brittonia. 2:163. 1936. (Figs. 10, 11). ‘Tere: BRAZIL. Amazonas: on varzea land, r mouth of Rio Embira (tributary of Rio Tarauaca), 7? 30'S, 70? 15'W, 26 Jun 1933, B.A. Krukoff 5003 (HoLotYrE: NY; ISOTYPES: E MICH, MOD. A lengthy description here of the species based on the single sheet with male flowers we have seen seems premature to us. The characters that lead us to believe that D. juruensis belongs to the D. campechiana complex 105 1 Al ype Specimen QJ iJ / HERA, MLC P e =— P. A. EEUKOTT'S Bh EXPEEFTION Te AAN Ad BAKIN GF MM: JURMUA x Rs ud P due. “E "i amen p Á É t Loe 7: E ato av fd. Embira (tribulary of : near mouth af Rio T^ 2908 ;Jongs te th’ W. as York aries! Gardin varo Ris caca, Fic. 10. The isot is (B.A, Krukoff 5003, MO). A. Fruiti 106 J i i Texas 3( Fic. 11. The isotype of Diospyros juruensis (B.A. Krukoff 5003, MO). A. Pistillodes. B. Stamens. include the large oblong-oblanceolate leaves with frequent darkened guard cells and somewhat recessed stomatal apparatus; male inflorescence axes that are markedly 4-angled; depressed orbicular male calyx lobes; a lobed pistillode (Fig. 11a) very similar to that seen in D. campechiana (Fig. 3b); stamens adnate to the base of the corolla, fused near the base of their filaments into ca. 16 pairs, outer and inner stamens + same length, densely golden sericeous on the filament and connective, anthers reddish orange, lanceolate, the apex (connective) tapered and minutely muricate. Diospyros panamense S. Knapp, Novon 9:524. 1999. (Figs. 12, 13, 15f). Ter: PANAMA. San Bras: Rio Diablo y vecinidad de Duque Sui, a unos 10 km de la costa frente a la Isla de Nargana, ruta hacia Cerro Ibedón, 80-110 m, 09° 22'N, 78° 35'W, [1 Jul 1992 acc. to W3Tropicos], H. Herrera et al. 1175 (noLorvrE: BM; isorvees: MO, PMA) Replaced name: Dios Knapp, Novon 7:258. 1997. Non Diospyros whitei Dows.-Lem. & Pannell, Bull. Jard. Bot. Nat. Belg. 65:399. 1966. t holl OW, Trees 5-20 m tall, evergreen; trunk up to 40 cm dbh, bark black, wood reportedly soft; y distally angular, + compressed and shallowly channeled, terete and smooth proximally, dark. green, brown or gray, occasionally sparsely strigillose, the hairs dark, or sometimes yellowish, minutely pubescent, the hairs colorless, curved-acicular, ca. 0.03 mm long; mature stems with thick, scaly, gray to dark brown bark, lenticels prominent, stemwood reddish brown in stems ca. 10 mm wide. Leaves alternate, simple, entire; petioles 7-14 mm long, subterete, minutely curved-acicular pubescent, epidermis green or atropurpureous, sometimes glaucous and scintillant, sometimes with a few transverse fissures, rounded below, glabrate to sparsely strigillose, + flattened above, sparsely strigillose, with a densely glandular narrow central channel; lamina subcoriaceous to coriaceous, oblong-elliptic to elliptic, though slightly wider in the upper half, (96-)127-240 mm long, 45-107 mm wide, length to width ratio ca. 22.6 : 1, base acute to obtuse, sometimes abruptly decurrent on the petiole, margin flat to revolute, apex acuminate with a rounded tip; lower lamina surface mu pou PR to sparsely strigillose, faintly papillose, guard cells sometimes dark and recessed; upp ish to greenish brown, markedly duller than the | , glabrous to use TE near pb epidermis M a bind papillose, the papillae angular. Venation (sensu Knapp 1997), minutely curved-acicular pubescent below; midrib pionnen below, rande, glabrate to sparsely strigillose, longitudinally wrinkled, concave above, sparsely to moderately strigillose and glandular; lateral veins + evenly spaced, 6-9 per side, fine below, though prominently raised, barely depressed above, the surface smooth, and so, conspicuous adjacent to the papil- late surface of the lamina; 3° veins fine below, though prominently raised, scarcely, if at all apparent above. Laminar extrafloral nectaries on the abaxial surface only, occasional, more frequent near the midrib, dark brown to black, ca. 0.6 mm wide, minutely rimmed. Male inflorescences glomerules, or compact cymes, (3-)10-20-flowered, a few at the base of young stems, or solitary in leaf axils, axes 4-angled, dark reddish brown strigillose, subtended by + 2 strigillose triangular bracts, ca. 0.9 mm long; peduncles 0.1-1 #ISSOURI BOTANICAL GARDEN HERBARIU? il NIE de UE BAIRD B a ay ia ) d LA E Bii y Fic. A. Immature fruit (E. Alfaro 291, MO). B. Fruiting t hlets (R. Aguilar et al. 2727, MO). C. Male infl (R. Aguilar 2749, K). D. ‘Fruits (R. Aguilar et al, 2727, MO). 108 I Ll pnfthn DC 12a n MEER PR JOUTMal OF Fic. 13. Diospyros panamense (R. Aguilar 2749, MO). A. Detail of male infl ith fl buds from fi t folder. B. Interior of male cal with pistillode. C. Stamens. mm long; pedicels 0—1.4 mm long. Male flowers known only from buds; male fl i I brotat with a short basal 1 , exterior drying dark brown to black, strigillose, interior acy strigillose, 3-4.2 mm in widest dimension, tube 1-1.5 mm long, lobes 4-5, depressed orbicular to ovate, 1-1.6 mm long, 1.8-2.4 mm wide, suberect nearing anthesis; male corolla white in life, drying dark brown, thick, glabrous, papillate, tube urceolate, 0.8-1.9 mm long, 2.6-3.2 mm wide, lobes 4-5, ovate, 1.4-2.2 mm long 1.6-1.7 mm wide; stamens 24-28, adnate to the base of the corolla, fused near the base of their filaments into 12-14 pairs, outer stamens bowing inwards, outer and inner stamens + same length, the outer stamen golden seri abaxially on the filament and connective, the hairs long and curved, also golden sericeous D ne J , Di hs [| 109 on the connective adaxially and on both surfaces of the inner stamen, but the hairs shorter, anthers reddish orange, lanceolate, ca. 2 mm long, the apex (connective) tapered and minutely muricate, filaments 0.2-0.4 mm long; pistillode obturbinate, markedly 12-14-lobed, + 1 mm wide, with 2-3 style-like apical struc- tures, each with a few long straight golden hairs. Female inflorescences not seen with flowers, apparently a solitary 1-3-flowered compact fasciculate cyme in the leaf axil of young stems, though no more than one fruit per inflorescence observed, axes 4-angled, sparsely pale strigillose, subtended by several thin, deltoid, strigillose, persistent bracts, ca. 1.5 mm long; fruiting peduncles 0-1 mm long; fruiting pedicels 1-4 mm long, with a 1-1.2 mm long 4-angled apex, the distal 0.5 mm depressed-pyramidal, forming a joint with the receptacle within the basal protuberance of the calyx. Female flowers unknown, except style 1.3-1.6 mm long, 0.9-1.3 mm wide, hairy, the hairs pale, yellowish, slightly wavy, terete to slightly 4-angled, bifid, the branches short, 0.2-0.3 mm long, deltoid (style and stigma [see below] details based on those persisting on fruit), stigmas depressed orbicular, 0.3 mm long, 0.7 mm wide. Based on the holotype (which we have not seen), Knapp (1997) reported, “styles 5, 1-1.5 mm, fused at the base” and “stigmas irregularly lobed.” Fruit- ing calyx thick, possibly fleshy in life, drying dark brown to black, exterior sparsely to densely strigillose, the hairs mostly pale, tube patent, y a short protuberance in immature fruit, this apparently absent at maturity, 7-10 mm wide! j , interior densely golden strigillose, lobes 4-5, spreading to slightly reflexed, depressed- M 2. 7—5 mm long, 4.5-9 mm wide, the margins sometimes ciliate, interior sparsely golden strigillose. Fruit up to 33 mm long, 33 mm wide, globose to subglobose, immature fruit densely hairy near the base and the apex, the remainder sparsely hairy, the hairs golden to dark red, mature fruit glabrous, locules probably 1-4, small damaged fruits with + 2 locules observed, and Whitefoord and Knapp (1998—onward) reporting “lóculos 4?”; flesh unknown in mature fruit; epidermis mature fruit loosely wrinkled, cracked and scaly near the base and apically, the largest fruits seen (R. Aguilar 2727 et al., MO) reportedly brilliant green; seeds unknown. Fruits are suspected to contain up to 4 seeds, though fruits in E. Alfaro 291 (MO), are probably 1 and 2-seeded based on the number of enlarged locules. Distribution and ecology.—Lowland rain forests of Costa Rica, Panama, and possibly Columbia, between 80 and 1200 m elevation (Fig. 14). There are reports (Zamora et al. 2004) of this species from both near the highest peaks (617—745 m) and in the low forests (e.g., 0-300 m) on the Osa peninsula. However, we have not seen collections from either of these regions. Specimens examined. COSTA RICA. PUNTARENAS. Canton de Osa: Fila Costeña, Río Piedras Blancas, junto a la casa, faldas Cerro An- guciana, Fila Cruces. 08? 49' 02"N, 83° 11' 23"W, 900 m, 9 Dec 1993, R. Aguilar et al. 2727 (MO, K); Fila Costeña, Rio Piedras Blancas, cerca de la casa, Cerro Anguciana, Fila Cruces, 08? 49! 02"N, 83° 11' 23"W, 900 m, 10 Dec 1993, R. Aguilar 2749 (MO, K). SAN Jose. Cantón de Perez Zeledon: Cordillera de Talamanca, La Nubes, Santa Elena, 09? 23' 30"N, 83? 35' 50"W, 1210 m, 3 Aug 1995, E. Alfaro 291 (MO). PANAMA. Darien: Cerro Pirre, 4 Aug 1967, N. Briston 1229 (MO-2 sheets). KEY TO THE MESOAMERICAN SPECIES OF THE DIOSPYROS CAMPECHIANA COMPLEX 1. Fruits mostly globose t | ts of Panama and Costa Rica. 2. Styles 2 or 1 style divided nearly to the base (Fig. 15). 3. Lamina oblong to oblong obovate; fruiting calyx thin (not fleshy), the lobes 3-4 mm wide, usually reflexed; TUE apa aud. eam mm s erect, glabrous D. crotalaria 3. | ti | e lobes 3.8-5.5 mm wide, spreading; fruit apex dons styles 0.5-1 mm thick, diverging from near the base i in the form of a'V'to + erect, hairy or glabrou 4. Lamina apex acuminate, upper lamina surface duller and slightly darker than reddish EUM lower P fruiting calyx lobes 2.7-3.8 mm lang ~ dark strigillose inside; style hairy, branches diverging from near the base i i the iudi ofa D. hartmanniana Lamina apex rounded, upper hinier and slightly lighter than the g hto ae brown lower surface; unig calyx lobes 1.5-2 mm long, glabrous inside; style branches glabr A DAT ideo) + erect ( p ypically has one style with some apical notches, or a groove down > D. haberi 2. Style 1 (Fig. 15) 5. Fruiting calyx lobe 2.7-5 mm long, 4.5-9 mm wide; fruiting calyx tube 7-10 mm from sinus to opposite sinus; lamina 96-250 mm long, 45-107 mm wide, the apex acuminate; young stems, petioles, and D. hartmanniana = O D. panamense = © | | D. crotalaria — e | | D. haberi = x po km pum ppp O 50 100 fet ia 2008 Sep 1223:52:38 OHC - Martin Weinatt Fic. 14. Distribution of Di talaria, D. hartmanniana, D. haberi and D. midribs with a sparse, minute, curved-acicular pubescence; fruits up to 32 mm long; style undivided, i 5- | .6 mm long, 1-1.5 mm thic D. panamense alyx lobe 1.5-2 mm long, 3.8-5.3 mm wide; fruiting calyx tube 5.1-6.6 mm from sinus to op- im sinus; lamina 30-96 mm long, 17-40 mm wide, the apex rounded; young stems and midribs without a minute curved-acicular pubescence; fruits 14-18 mm long D. haberi . Fruits mostly ovoid-ellipsoid; plants of Mexico and Guatemala. Stems, petioles, midribs (in addition to being dark api fruit 13-17 mm long, the apex glabrous; style 1-2.1 mm long, glabrous; i calyx thin, the tube 3 from sinus to opposite sinus, the lobes 2-4 mm long, not reflexed; lamina 100-285 mm long, 2n ong to oblong-lanceolate to elliptic, ti oblong- -oblanceolate, the margins ome o D. campechiana 6. Stems, petioles, midribs i minutely ¡ it (15-)18-24 mm long, the apex hairy; style ca. 2.4 mm long, hairy; fruiting x thick, the tube 6- 7. 5 mm from sinus to opposite sinus, the lobes 4-6 mm long, reflexed; lamina 55- 152 mm long, widely ovate to elliptic, the margins minutely ridged . camposii dd ACKNOWLEDGMENTS We would like to express our appreciation to Martha L. Orozco-Cardenas at the UCR Plant Transformation Research Center who kindly allowed access to microscopes a did equipment. We thank Lia Mansfield for providing Latin translations. Ed Plummer g h appreciated technical support. We would especially like to extend our gratitude to the following ebat for specimen loans: B, CAS, CHAPA, F, IJ, K, MO, UCR, UMO, USF, WIS, XAL. Paulo T. Sano and an anonymous reviewer provided helpful comments for improving the manuscript. Finally, we appreciate Giles Waines' continuing support and encouragement of our taxonomic work on the Ebenaceae at the UCR Herbarium. REFERENCES CAMPOS-VILLANUEVA, A. AND J.L. ViLLAsEROR. 1995. Estudio florístico de la porción central del Municipio de San Jerónimo Coatlán, Distrito de Miahuatlán (Oaxaca). Bol. Soc. Bot. México 56:95- 120. Cavalcante, PB., 1963a. Contribucào ao conhecimento do género Diospyros Dalech. (Ebenaceae) na Amazonia. Bol. Mus. Paraense Emilio Goeldi, N. S., Bot. 20:1-53. Fic, 15. Stylar variation in the Diospy pechi plex. A. St (with d 1 apices) isti he frui D. crotalaria, based on the holotype B.B hed tyl (with d g | tig J of D. campechi ana, based on n WE Harmon & LE F Fi 5790 (UMO). C. B damaged apices y i t thi angl ) of D. camposii, k yl Stvle with sti it f D D. haberi, based on G. Rivera 1422 (K). E. Styl itl ) at the fruit | f D. hart ji , based oF Bello 2442 (M0) E Style with sti g at fruit apex of D. panamense, based on E. Alfaro 291 (MO). CAVALCANTE, PB. 1963b. Nova contribuigáo ao conhecimento do género Diospyros Dalech. (Ebenaceae) no Brasil. Bol. Mus. Paraense Emilio Goeldi, N. S., Bot. 21:1-15 (+ estampa I-II). CavALCANTE, PB. 1966. Duas novas espécies do género Diospyros Dalech. (Ebenaceae) da Amazônia. Bol. Mus. Paraense Emilio Goeldi, N. S. Bot. 22:1-5 (+ 2 estampas). CAVALCANTE, PB. 1977. Espécies novas da flora amazónica (Ebenaceae). Acta Amazon. 7:189-197. GLEASON, H.A. AND A.C. SMITH. 1933. Plantae Krukovianae. Bull. Torrey Bot. Club. 60:349-396. GREUTER, W. (and 10 others). 1994. International code of botanical nomenclature (Tokyo Code). Regnum Veg. 131. Koeltz Scientific Books, Kónigstein. GREUTER, W. (and 11 others). 2000. International code of botanical nomenclature (St. Louis Code). Regnum Veg. 138. Koeltz Scientific Books, Königstein. Hiern, W.P. 1873. A monograph of Ebenaceae. Trans. Cambridge Philos. Soc. 12:27—300. Jackson, B.D. 1916. A glossary of botanic terms: with their derivation and accent. 2nd edition. Duckworth (Lon- don), J.B. Lippincott (Philadelphia). P. 209. Knapp, S. 1997. Two new species of Diospyros (Ebenaceae) from Mesoamerica. Novon 7:256-260. KUBITZKI, K. AND A. ZiBURSK. 1994. Seed dispersal in flood plain forests of Amazonia. Biotropica 26:30-43. Macealp:, J.F. 1959. Flora of Peru. Publ. Field Mus. Nat. Hist., Bot. Ser. 13(5/1):205-214. MILDBRAED, J. (Ed.), 1927: Plantae Tessmannianae peruvianae VI. Notizbl. Bot. Gart. Berlin-Dahlem 10:169-197. Moue, FA.G. 1856. Ebenaceae. Flora Brasiliensis. In: C.EP. von Martius, ed. F. Fleischer, Monachii et Lipsiae. 7A:1-10, t. 1-3 112 l Journal of tanical h Instit Texas 3( Pacheco, L. 1981. Ebenaceae. In: Gómez-Pompa, A. and V. Sosa, eds. Fl. Veracruz. Fasc. 16:4-7. Provance, M.C. AND A.C. SANDERS. 2005. Diospyros torresii (Ebenaceae): a new black zapote from tropical Mexico. Sida 21:2045-2050. Provance, M.C. AND A.C. SANDERS. 2006. More American black sapotes: new Diospyros (Ebenaceae) for Mexico and Central America. Sida 22:277-304, Provanc, M.C., |. Garcia R., AND A.C. SANDERS. 2008. The Diospyros salicifolia complex (Ebenaceae) in Mesoamerica. J. Bot. Res. Inst. Texas. 2:1009-1100. SANDWITH, N.Y. 1931. Contributions to the flora of tropical America: VIII. Bull. Misc. Inform. 1931:467-492. SanbwrTH, N.Y. 1949. Contributions to the flora of tropical America: L. Kew Bull. 1949:481-493. SANDWITH, N.Y. 1963. Contributions to the flora of tropical America: LXX. Notes on some Aublet types in the Paris herbarium. Kew Bull. 17:257-262. SoTHERS, C. AND PE, Berry. 1998. Ebenaceae. In: Berry, PE, B.K. Holst, and K. Yatskievych, eds. Flora of the Venezuelan Guayana. Vol. 4:704—712. Missouri Botanical Garden Press, St. Louis. SorHers, C.A. 2003. New species of Diospyros (Ebenaceae) from Brazil. Kew Bull. 58:473-477. STANDLEY, PC. 1924, Trees and shrubs of Mexico. Contr. U.S. Natl. Mus. 23:849-1312. STANDLEY, P.C. AND L.O. WiLLIAMS. 1967. Ebenaceae. In: Standley, P.C. and L.O. Williams, eds. Flora of Guatemala, Part VIII, Number 3. Fieldiana, Bot. 24(8/3):246. Stearn, W.T. 2000. Botanical Latin: history, grammar, syntax, terminology and vocabulary. 4th edition. Timber Press, Portland, Oregon. P. 438. WALLNOFER, B. 1999. Neue Diospyros-Arten (Ebenaceae) aus Súdamerika. Ann. Naturhist. Mus. Wien, B. 101:565— WALLNOFER, B. 2000. Neue Diospyros-Arten (Ebenaceae) aus Südamerika - Il. Ann. Naturhist. Mus. Wien, B. 102:417-433. WALLNOFER, B. 2001. The biology and systematics of Ebenaceae: a review. Ann. Naturhist. Mus. Wien, B. 103:485- WaLLnorEr, B. 2003. A new species of Diospyros from southwestern Amazonia. Ann. Naturhist. Mus. Wien, B. 104:563-566 WALLNOFER, B. 2005. New species of Diospyros (Ebenaceae) from the Neotropics and additional information on D. apeibacarpos. Ann. Naturhist. Mus. Wien, B. O R WeineLT, M. 2006. OMC: Online map creation. http:/ WesseL, P. AnD WH.F. Smit. 2006. GMT: The generic mapping tools (Vers. 4.1.4). Wure, F. 1978. Flora of Panama, Part VIII. Family 155. Ebenaceae. Ann. Missouri Bot. Gard. 65:152-153. Whrre, F. 1981. Ebenaceae. In: Maguire, B. and collaborators, eds. The botany of the Guayana Highland, Part XI. Mem. New York Bot. Gard. 32:323-329. WHITEFOORD, C. AND S. Knapp. 1998-onward. Taxonomic and editorial changes made by the general editors to the internet version of Flora Mesoamericana. Fl. Mesoamer. Internet Version. Davidse, G. M. Sousa S, and S. Knapp, eds. 1998. http://www.mobot.org/MOBOT/FM/. Accessed Jun 2008-Jan 2009. ZAMORA, N., B.E. HAMMEL, AND M.H. Grayum. 2004a. Vegetación/Vegetation. In: B.E. Hammel, M.H. Grayum, C. Her- rera, and N. Zamora, eds. Manual de plantas de Costa Rica, Vol. |. Introducción. Monogr. Syst. Bot. Missouri Bot. Gard. 97:187. - nc/omc. intro.html NOVEDADES Y NOTAS MISCELÁNEAS EN LAS BROMELIACEAE DE MESOAMÉRICA J. Francisco Morales Instituto Nacional de Biodiversidad (INBio) pto 22-3100 Santo Domingo de Heredia, COSTA RICA RESUMEN 1,0 1 q [eae S | Se describe Werauhia luctuosa (Bromeliaceae: Tillandsoideae) con W. pittieri. Adicionalmente, se reporta la presencia de W. brunei y W. tanrtiddda en Panamá. B de herbario, material EDS y taba de O se ne que el material de Mesoamérica previamente identificado como qr. penduliflora y g América) tualmente A. angustifolia y G. blasii respectivamente. n el ectiidi PALABRAS CLAVE: Bromeliaceae, Honduras, Costa Rica, Panama, Aechmea, Werauhia ABSTRACT Werauhia luctuosa Sime peus NA a new species from Honduras is described and its ic affinities with W. pittieri are discussed. Werauhia brunei and W. pua eie are pps for the ieu time in iran is on ME n of cui Speed DB NA m field work, i i 1 ij America), are actually A angustifolia and G. } E Key Wonps: Bromeliaceae, Honduras, Costa Rica, Panama, Aechmea, Werauhia Como resultado de la binación del trabajo de campo y el estudio de colecciones de herbario tendientes al desarrollo de una Guía de Campo para las Bromeliaceae del Sur de Centroamérica, se proponen la siguiente novedad y notas taxonómicas misceláneas. Aechmea penduliflora André, Enum. Bromél. 3. 1888. Tiro: COLOMBIA. SantanveR: Isla Brava, Río Magdalena, entre ondo y Chucuri, 8 dic 1875, André 378 (noLoripo: K, GH-fotografía). Distribución. —Colombia y Venezuela a Ecuador y Perú, donde crece en bosques muy húmedos entre los 50 y 700 m. Aechmea penduliflora ha sido reportada para Nicaragua, Costa Rica y Panamá en diversas ocasiones (Correa et al. 2004; Morales 2003; Utley & Burt-Utley 1994; Utley et al. 2001). Sin embargo, el estudio de su tipo, así como de todos los especimenes citados como testigos en los tratamiento antes mencionados, han revelado que las colecciones (e.g., Espinoza 94, INB, MO; McPherson 8533, MO; Rueda & Mendoza 17152, MO) representan típicos especimenes de A. angustifolia Poepp. & Endl. (concordantes en todo sentido con el tipo de este taxón) y que por lo tanto, A. penduliflora debe considerarse como una especie no presente aún en Mesoamérica. Esta última especie algunas veces puede tener inflorescencias reducidas, en cuyo caso los ejemplares pueden parecer diferentes de la típica variedad con inflorescencias relativamente largas y elon- gadas, pero diferentes grados de variación pueden ser observados en el campo, así como en los especimenes de herbario. Esta misma plasticidad en la morfología y tamaño de la inflorescencia está presente en otras especies del genero en Mesoamerica (e.g., A. tillandsioides (Schult. f. ex Mart.) Baker). Guzmania mitis L.B. Sm., Contr. Gray Herb. 98:31, t. 6, f. 4-5. 1932. Tiro: COLOMBIA. Norte DE SANTANDER: valle de Pica-Pica, sobre Tapata, al N de Toledo, 1-5 mar 1927 (fl), Killip & Smith 20195 ( : GH; isoriro: US, INB-fotografía). Distribución. —Colombia y Venezuela, donde crece en bosques muy húmedos y bosques nubosos entre los 1800 y 2800 m En su revisión monográfica de las Bromeliaceae, Smith y Downs (1977) reportó esta especie para Costa Rica, basado en un especimen recolectado por Foster en la región de Cartago, lo cual ha sido seguido en forma subsecuente por diversos trabajos florísticos (e.g., Luther 1995; Morales 2003; Utley & Burtley-Utley J. Bot. Res. Inst. Texas 3(1): 113 —116. 2009 114 tani i Texas 3( 1994). Ese especimen corresponde a una ejemplar relativamente pobre, con frutos maduros y carece de flores o incluso brácteas florales completas. El pobre ue » ese especimen fue mencionado anteriormente por Utley y Burtley-Utley (1994). Sin embargo, como resultad Be pd Ee oe especimen (Foster 2686, US), asi como del ejemplar tipo de Guzmania mitis y otr de esa especie en herbarios de Estados Unidos, Colombia y Venezuela, es claro que la eos colectada por Foster en Costa Rica, fue mal identificada por Smith y Downs (1977) y que en realidad, corresponde a un ejemplar de Guzmania blassii Rauh, un taxón similar, pero endémico a Costa Rica. Esta última especie se puede reconocer y separar de G. mitis por sus hojas con las láminas con líneas moradas o rojizas longitudinales (vs. concoloras y sin líneas), flores con pedicelos de 4—7 mm de largo (vs. sésiles a subsésiles), pétalos amarillos (vs. blancos) y rangos altitudinales distintos, ya que usualmente G. blasii crece en elevaciones entre 700 y 1700 m, mientras que G. mitis se encuentra entre los 1800 y 2800 m. Por lo tanto, G. mitis debe excluirse como un representante de la flora Mesoamericana y considerarse con un taxón restringido al N de Suramérica. Werauhia brunei (Mez & Wercklé) J. - Grant, D o pue 91:31. 1995. vriesea brunei Mez & Wercklé, Bull. Herb. Boissier, ser. 2, 4(9):865. 1904. Tiro: C Alajuela): Candelaria, oct 1902, Brune in Werckle Brom. Costar 45 ee B, M Distribución.— Costa Rica y el O de Panamá, donde crece en bosques nubosos y robledales en elevaciones de 1700-2750 m Werauhia brunei pertenece a un complejo de especies que se caracterizan por sus inflorescencias simples, las flores siempre con una bráctea adyacente, que Imente cubre gran parte de los sépalos y con las brácteas variadamente imbricadas (Morales, 2003). Este complejo incluye entre otras a W. ampla (L.B. Sm.) J.R. Grant, W. bicolor (L.B. Sm.) J.R. Grant, W. burgeri (L.B. Sm.) J.R. Grant, W. gladioliflora (H. Wendl.) J.R. Grant, W. macrantha (Mez & Werckle) J.F. Morales, W. macrochlamys (Mez & Wercklé) J.F. Morales, W. osaensis (J.F. Morales) J.F. Morales, W. tiquirensis (J.F. Morales) J.F. Morales y W. tonduziana (L.B. Sm.) J.R. Grant, las cuales en general, son superficialmente similares y de difícil distinción (Morales, 1999, 2003). Werauhia brunei se puede separar por sus brácteas florales usualmente dísticas antes y después de la antesis (incluso con frutos maduros), lisas, verdes a verde-canela y de (4.9—)5.2-6 cm de largo. Para mayor infor- mación, véase el trabajo de Morales (2003). Especímenes examinados. PANAMÁ. Chiriquí: cerro Punta, falda NO, camino desde Finca Drácula hacia el último lodge dentro del Parque, 10 ago 2006 (fD, Morales & Santamaría 14625 (INB Werauhia camptoclada (Mez & Wercklé) J.E Morales, Monogr. Syst. Bot. Missouri Bot. Gard. 92:360. 2003. Vriesea camptoclada Mez & Wercklé, Repert. Spec. Nov. Regni Veg. 14(400-404):247. 1916. Tipo: COSTA RICA. San José: en las cercanías de La Palma, oct 1908, Wercklé in Inst. Costaric. 17292 (HoLotiro: B, INB-fotografía; isoro: GH). Distribución. —Costa Rica y el O de Panamá, creciendo en bosques nubosos en elevaciones de 1500-2100 m. Werauhia camptoclada es una especie distintiva que se puede confundir con W. moralesii Luther, W. kupperiana (Suess.) J.R. Grant y W. werckleana (Mez) J.R. Grant, pero que se puede reconocer por su roseta densa y pequeña (en relación al tamaño de la inflorescencia), hojas que no exceden los 35 cm de largo e inflorescencias con las ramas secundarias que raramente exceden los 12 cm de largo. Esta especie ha sido considerada endémica a Costa Rica (e.g., Morales 2003) pero ahora se conoce en el O. de Panamá en las estribaciones de la Cordillera de Talamanca. PANAMÁ. eh cerro Punta, falda NO, camino desde Finca Drácula hacia el ültimo lodge dentro del Pardue 10 ago 2006 (fl), Morales & S 14626 (INB). Werauhia luctuosa J.F Morales, sp. nov. (Fig. 1). Tiro: HONDURAS. Santa BArBara: lago Yojoa, El Sauce, 9 abr 1951 (fl), L. Williams & A. Molina 17704 (moLotipo: EAP, INB-fotografias). A Werauhia pittieri (Mez) J.R. Grant, cui similis, scapo 36-37 cm longo (vs. 45-55 cm), sepalis 3.2-3.5 cm longis (vs. 1.7-2.8 cm) et coma alba (vs. ferruginea) differt. Plantas acaulescentes, con una altura en floración de 50 a 60 cm. Hojas rosuladas, más o menos erectas, Morales, Nueva especie en Bromeliaceae 2cm Pd Ps - 4 "Pe HONDURAS Hentcme al 17704 Facurla Agri Canaria : į > Vriesia A jd ` T ra. V. AN Gau Rain forest area above olot, Bl Sauce Departamento Lows O. Wüll&ms Sta. -arbara Antonio Mola P att 1000 April 9, 1951 na: 115 116 tani Insti Texas 3( 41-46 cm de longitud, densamente punteado-lepidotas en la superficie abaxial, el indumento escaso y más esparcido en la superficie adaxial, Mich verdes p ambas superficies; bases de la lámina anchamente elípticas, 9-11 x 4.5—6 cm, pálidas, usualment las láminas, densamente punteado-lepidotas en la superficie abaxial; lámina de las hojas e abruptamente corto-acuminadas, 31-34.5 x 3.5-4.1 cm. Escapo erecto, 36-37 x 0.5-0.6 cm, diminutamente papilado; brácteas del escapo erectas, 4.3-4.6 cm de longitud, anchamente ovadas, mucho más largas que los internudos, densamente imbricadas, café oscuro basalmente, algo más claras distalmente, sobretodo en la región del ápice, lisas. Inflorescencias erectas, simples, 12-13 cm de longitud; brácteas florales anchamente ovadas, agudas, 3.5-3.8 x 2.9-3.3 cm, relativamente delgadas, ligeramente imbricadas igualando los sépalos en longitud o ligeramente más cortas, café oscuro, los bordes más claros, ecarinadas, lisas. Flores 5 a 6, con los pedicelos de 4-5 mm de longitud, gruesos, erectos, esparcidamente papilados; sépalos 3.2-3.5 cm de longitud, erectos, elípticos, coriáceos, café oscuro. Corola y cápsula desconocidas. Distribución & habitat. —Endémica a Honduras, donde se conoce solo de la localidad tipo, en bosques hümedos en elevaciones de 900 a 1000 m. Werauhia luctuosa se encuentra cercanamente relacionada a W. pittieri (Mez) J.R. Grant, ya que ambas especies comparten inflorescencias simples, con brácteas florales y flores unilaterales, café oscuro, densa- mente imbricadas y de 2.6—3.9 cm de largo, pero la primera especie se puede separar por su escapo más corto (36-37 cm vs. 45-55 cm), sépalos más largos (3.2-3.5 cm vs. 1.7-2.8 cm) y frutos con coma blanca (vs. ferrugínea). Adicionalmente, W. pittieri es una especie predominante en bosques nubosos y robledales sobre los 2000 m (raramente a 1850 m), mientras que W. luctuosa crece en bosques húmedos entre 900 y 1000 m de elevación. Werauhia luctuosa solo es conocida por la colección tipo y a pesar de la búsqueda en otros herbarios no fue posible localizar colecciones adicionales. El epíteto de esta especie está dedicado a la memoria de Rosenda Morales y María Ester Padilla, familiares (tías) fallecidos en forma reciente. AGRADECIMIENTOS Quiero agradecer a los siguientes herbarios por permitir el estudio de sus colecciones: BR, CR, EAP, F, K, LAGU, MHES, MO, Y, P, U, US. Barry Hammel (MO) amablemente realizó el resumen en inglés. Asimismo, la visita de algunas instituciones fue financiada por el proyecto “Compartiendo tecnología y desarrollando capacidades para la gestión de la biodiversidad en Centroamérica" financiada por el Gobierno de Noruega y ejecutado en forma parcial por el Instituto Nacional de Biodiversidad (INBio) REFERENCIAS CORREA A, M.D., C. GALDAMES Y M.S. DE STAPF, 2004, Catálogo de las plantas vasculares de Panamá. Smithsonian Tropical Research Institute, Ciudad de Panamá, Panamá. LurHER H.E. 1995. An annotated checklist of the Bromeliaceae of Costa Rica. Selbyana 16:230-234. Morales, J.F. 1999, Seis nuevas especies de Vriesea section Xiphion (Bromeliaceae: Tillandsoideae) para Costa Rica. Novon 9:401—406. Moraes, J.F. 2003. Bromeliaceae. En: B. Hammel, M. Grayum, C. Herrera y N. Zamora, eds. Manual de plantas de Costa Rica. Monogr. Syst. Bot. Missouri Bot. Gard. 92:297-375. SmrrH, L.B. v RJ. Downs. 1977. Tillandsioideae (Bromeliaceae). Fl. Neotrop. 14(2):663-1492. Urey, J. v K. Bunr.ev-Uriev. 1994. Bromeliaceae. En, G. Davidse, M. Sousa y A. Chater, eds. Fl. Mesoamer. 6:89-156. 1994 UrtEy, J., K. Bunr-Uriev v M.J. Hurr. 2001. Bromeliaceae. En: W.D. Stevens, C. Ulloa Ulloa, A. Pool y O.M. Montiel, eds. Flora de Nicaragua Vol. 1. Mongr. Syst. Bot. Missouri Bot. Gard. 85:460-495. UNA NUEVA ESPECIE Y NOTAS MISCELÁNEAS EN EL GÉNERO OREOPANAX (ARALIACEAE) EN CENTRO AMÉRICA J. Francisco Morales Alvaro Idárraga Instituto Nacional de Biodiversidad (INBio) Herbario, Centro de Investigaciones pto 22-3100 Facultad de Ciencias Exactas y Naturales nro Santo Domingo de Heredia, COSTA RICA Universidad de Antioquia, Apartado Postal 122 Medellín, Antioquia, COLOMBIA RESUMEN aa ee de la revisión de las especies del género Oreopanax (Araliaceae) en Costa Rica, Nicaragua y Panamá, se proponen una yen la descripción de una nueva especie, O. paramicolus, la redefinición del concepto de otras On nica- raguensis), hectificación del rango de distribución geográfica para dos taxones (O. geminatus, O. peltatus) y el reporte d p Panamá (O. nubigenus). PALABRAS CLAVE: Centro América, Costa Rica, Panama, Araliaceae, Oreopanax ABSTRACT A lt of th isi f species of the g O Arali ) in Costa Rica, Nicaragua and Panama, Idi ies | come ght. TI i d ,0 Pe here described. tl O. nicaraguensis and the need oli to adjust the geographical range b three taxa, O. geminatus, O. peltatus, and O. nubigenus, the latter ay reported for Panama. Key Wonps: Central America, Costa Rica, Panama, Araliaceae, Oreopanax Araliaceae es una familia pantropical de plantas, con alrededor de 41 géneros y ca. 1350 especies, distribuidas principalmente en los trópicos (Plunkett et al. 2004). No existen tratamientos monográficos recientes de los géneros del Nuevo Mundo y a nivel de Centroamérica, lo único disponible son sinopsis (Cannon & Can- non 1989) o tratamientos florísticos (e.g., Cannon & Cannon 2001; Standley 1938). La ausencia de trabajos monográficos y la alta variabilidad foliar en algunos géneros (e.g., Dendropanax, Oreopanax) ha provocado una taxonomía confusa, sobretodo en lo que la correcta aplicación de nombres se refiere. Oreopanax, el cual abarca cerca de 80 especies, 24 de ellas presentes en México y Centroamérica (Cannon & Cannon 1989, 2001), tampoco ha escapado de la situación descrita anteriormente, lo que ha traido como consecuencia la incorrecta aplicación del concepto de algunos taxones y datos de distribución incorrectos para otros. Como resultado de la elaboración del tratamiento de Araliaceae para el Manual de Plantas de Costa Rica, se realizó un detallado estudio taxonómico de las especies de Oreopanax presentes en Costa Rica y países aledaños, que incluyó la revisión de especimenes tipo y material en los principales herbarios de Centroamérica, Estados Unidos y Europa. Como resultado, una serie de novedades fueron encontradas, las cuales son descritas o comentadas a continuación. Estas incluyen la descripción de una nueva especie, así como notas misceláneas sobre la incorrecta aplicación del concepto de algunos taxones y la rectificación del rango de distribución geográfica de otros. Algunos datos citados en el manuscrito han sido tomados del tratamiento de Araliaceae para el Manual de Plantas de Costa Rica (Morales et al., datos sin publ.), por lo que pueden diferir de los de otros tratamientos previamente publicados. Milos geminatus Marchal, Bull. Acad. Roy. Sci. Belgique, ser. 2, 47:91. 1879. Tiro: NICARAGUA. Nueva GovIa: "America Centralis, ad Sajonia,” sin fecha (fl), A. Oersted 7 (HoLotIPo: C). Distribución.—México, Belice, Honduras y Nicaragua, en elevaciones de 900—1750(-2000) m. Oreopanax geminatus fue descrita con base en una colección hecha por Oersted en un sitio inespecífico de Centroamérica, con una localidad vaga “Sejovia” (Marchal 1879). Desde entonces, se ha considerado que esta especie está presente en Costa Rica, pues se asume que la localidad tipo corresponde a una localidad en ese país. En general, muchas de las localidades donde recolectó Oersted en Costa Rica son zonas cercanas J. Bot. Res. Inst. Texas 3(1): 117 —121. 2009 118 t tani i Texas 3( al Valle Central, donde se ha conoce relativamente bien la flora. Sin embargo, aunque dicho colector realizó colecciones principalmente en Costa Rica, también recolectó varios centenares de colecciones en Nicaragua entre 1846 y 1848 (Stevens & Montiel 2001). Luego del estudio del especimen tipo de O. geminatus, así como del material de esa especie presente en diferentes herbarios de Europa, Estados Unidos y Centroamérica, hemos llegado a la conclusión de que la localidad "Sejonia" pone es análoga con el departamento de Nueva eu en dd N de Nicaragua, donde E geminatus es te común (R. Abarca, com, pers.). A , este taxón ] els EN Nicaragua ni en ninguna otra región de Costa Rica. Por lo tanto, debemos delimitar que O. geminatus tiene un rango geográfico que se extiende del S de México al N de Nicaragua, cuya colección tipo, fue recolectada en algún sitio del departamento de Nueva Segovia en Nicaragua y que por el momento, basado en la evidencia suminstrada por especimenes de herbario, no se conoce en Costa Rica. Oreopanax nicaraguensis M.F Cannon & Cannon, Ann. Missouri Bot. Gard. 73:482, f. 1-2. 1986. Two: NICARAGUA. Jinoreca: camino a Aranjuez, Santa Elena, 1-30 nov 1983 (fr), S. Vega & J. Quesada 197 (HOLOTIPO: BM; isoripos: HNMN, MO) Distribución —Endémica a Nicaragua, en elevaciones de 100—1300(-1500) m. En la descripción de Oreopanax nicaraguensis Cannon & Cannon (1986) citaron seis especimenes de Nicaragua, cuatro de Costa Rica y uno de Panamá. Desde entonces, se ha manejado un concepto de esta especie que incluye especímenes que crecen casi a nivel del mar (100 m), hasta las zonas más altas de las Cordilleras en Costa Rica y Panamá (3400 m), lo cual ha sido ampliamente seguido en colecciones de her- barios y tratamientos florísticos subsecuentes (e.g., Cannon & Cannon 1989; Cannon & Cannon 2001). Sin embargo, durante el estudio de material para el tratamiento de Araliaceae para el Manual de Plantas de Costa Rica (Morales et al., datos sin publ, llegamos a la conclusión de que en la descripción original tres especies distintas fueron citadas en los paratipos: O. nicaraguensis, a la cual pertenecen todos los especimenes citados de Nicaragua, O. donnell-smithii Standl., a la que corresponde la colección de la cordillera de Tilarán en Costa Rica (Dryer 1360, F, MO) y la última de ellas, una especie sin describir y restringida a las zonas más altas de la cordillera de Talamanca en Costa Rica y Panamá (descrita en este trabajo). De esta forma, O. nicaraguensis debe considerarse un taxón endémico al N de Nicaragua, el cual se encuentra algo relacionado con O. donnell-smithii, O. paramicolus y O. striatus MJ. Cannon & Cannon, pero que se diferencia de este grupo por sus hojas simples, enteras, con la base obtusa redondeada, i o glabrescentes, las hermafroditas con cabezuelas eat pedunculadas, con grupos de 5a 10 flores por cabezuela y frutos globosos o subglobosos al madurar, de 6—7 mm de largo y con los estilos con más de la mitad inserta dentro de la depresión apical del fruto. Oreopanax nubigenus Standl., J. Wash. Acad. Sci. 17:315. 1927. Two: COSTA RICA. San José: Las Nubes, 21 mar 1924 (£D, P Standley 38806 (noLoriro: US). Distribución. —Costa Rica y el O de Panamá, en elevaciones de 155026002850) m. Esta especie era considerada endémica a la Cordillera Central en Costa Rica (Cannon & Cannon 1989) pero es reportada por primera vez para Panamá, donde se conoce en las estribaciones de la Cordillera de Talamanca, en la provincia de Chiriquí. Oreopanax nubigenus puede ser confundida con O. standleyi A. C. Sm., aún no reportada para Panamá (Correa et al. 2004), pero se diferencia por sus láminas foliares usual- mente más pequeñas (6—10(-11.5) x 4-9.5(-11) cm vs. 12-25.5 x 5.5-20 cm), ramitas jóvenes glabras o glabrescentes con el indumento muy esparcido (vs. tomentosas o tomentulosas) e inflorescencias con las flores hermafroditas con 6 a 8 estilos (vs. 10—12) Especímenes examinados. PANAMA. Chiriquí: Cerro Pate Macho, 31 dic 1985 (fr), de Nevers & Chanley 6689 (MO). i oed o. E Morales & A. Idárraga, sp. nov. (Fig. 1). Two: COSTA RICA. San José: cantón de Pérez Cerro de la Muerte, páramo Buena Vista, 19 oct 1993 (fr), J.E Morales, N. Zamora, E. Lépiz & V. Ramírez 1890 (uoLormo: m. isoripo: CR). RA J 1144 ga, N ia daf 119 Fic. 1. 0 icolus. A. Rama con frutos. B. Detalle de los frut 120 A Oreopanax nicaraguensis MJ. Cannon & Cannon, cui similis, inflorescentia hermaphrodita 15—30-floris, fructus 8-10 mm longus et stylis exsertis, non inmersis differt. Árbol o arbusto epífito de 2-10 m de altura, las ramitas glabras o glabrescentes y con el indumento incon- spicuo. Hojas simples, enteras; lámina 7-22 x 9-21 cm, na a a ae ae de forma menos común elíptica, el ápice acuminado o agudo, la base l comúnmente redondeada a obtusa, glabra, pecíolo 4-20 cm de largo. Inflorescencias usualmente paniculadas, con las ramificaciones evidentes, 12-17 cm de e ee o PO y con el qua inconspicuamente pu- berulento justo antes de las cabezuela (González et al. 1592, INB); bea de las a masculinas 8— 12 mm de diámetro, con grupos de 15 a 30 flores, las flores con un estilo solitario; cabezuelas de las inflorescencias hermafroditas 5-7 mm de diámetro, con grupos de 4 a 5(6) flores, las flores con 6-8 estilos. Frutos en grupos de 4-6 por cabezuela, ovoides, con un diámetro de 8-10 mm, los estilos casi totalmente exsertos fuera de la depresión apical del fruto. Distribución, habitat, y ecología.—Restringida a la Cordillera de Talamanca en Costa Rica y el O de Panamá, donde crece en formaciones de robledales (Quercus spp., Fagaceae) y páramos, en elevaciones de 2500-3400 m. Floración se produce entre enero y febrero. Fructificación ocurre de enero a marzo, mayo, y de julio a noviembre. Oreopanax Duomo ha uu confundida con Oreopanax nicaraguensis M.J. Cannon & Cannon, pero se diferencia por sus infl tas con mucho más flores (15 a 30 vs. 5-10), frutos de forma diferente y más grandes (ovoides y de 8-10 mm de largo vs. globosos o subglobosos y de 6-7 mm de largo) y estilos casi totalmente exsertos fuera de la depresión apical en el fruto (vs. estilos con más de la mitad in- cluida). Adicionalmente, O. paramicolus crece en robledales y páramos (i de transición entre ambos ecosistemas), entre 2500 y 3400 m, mientras que O. nicaraguensis crece en bosques muy húmedos, entre los 100 y 1300(-1500) m. Especímenes examinados. COSTA RICA. Cartago: Cerro de la Muerte, La Georgina, 2 ago 1965 (fr), Croat 261 (MO); reserva forestal Rio Macho, estación Ojo de Agua, 12 ene 1996 (fr), Gamboa & Picado 950 (INB, MO); reserva forestal Río Macho, Cerro de la Muerte, 6 sep 1996 (fr), Rodríguez et al. 1470 (INB). Limón: parque nacional Chirripó, 15 feb 1983 (fl), Garwood et al. 1273 (BM, MO); parque internacional La Amistad, Tararia, 12 abr 2002 (fr), González et al. 1592 (INB). Puntarenas: Cordillera de Talamanca, Cerro Echandi, 23 ago 1983 (fr), Davidse et al. 23876 (MO); parque internacional La Amistad, Cerro Echandi, 13 ago 1997 (fr), Quesada et al. 1994 (INB). San José: entre Canaán ny cheapo via pu Angeles, ue El Río a 19-22 enel970 s Burger & Liesner 7381 (F, MO), 24 ago 1971 (fr), d 8328 (F, ^ I hirripó, Indios, 1 may 1997 (fr), Gamboa & Aguilar 1271 (INB) reserva forestal s Santos, camino a Pu 19 jul 1997 (fr), Gamboa 1620 (INB); Pérez eil albergue Cuerecí, 8 jul 2000 (fr), Miller 38 (INB, m, cis: 10 mar 1994 (fr), Ramírez & Hammel 287 (INB, MO). San José-Cartago: Cordillera de Talamanca, Cerros Cuerecí, 15 set 1983 (fr), Davidse 24705 (MO); Copey, cerro Las Vueltas, 28 Mar 1994 (fr), Vargas et al. 1748 (CR). P. MA. Bocas del Toro: cerro Fábrega entre cerros Itamut y Bine, Mar 1984 (fr), Gómez et al. 22461 (MO). Chiriquí: Boquete, camino a Volcán Barú, 18 may 1976 (fr), Croat 34885 (MO); Potero Muleto, Volcán Barú, 27 nov 1975 (st), Volcán de Chiriquí, 20 nov 1975 (fr), Davidse & D'Arcy 10273 (MO); El Volcán, SE de la Nivera, sin fecha (fr), D'Arcy 12505 (MO); Volcán Barú, Potrero Muleto, 6 nov 1978 (fr), Hammel 5642 (MO); Volcán Barú, 24 jul 1975 (fr), Mori & Bolten 7440 (MO); parque nacional Volcán Barú, 15 ene1992 (fl), Vega & Rincón (MO, PMA Oreopanax peltatus Linden, Gartenflora 11:170, t. 363. 1862. Tiro: MÉXICO: Datos perdidos, J. Linden s.n. (notorio: Distribución. —México, Guatemala y El Salvador, en elevaciones de 700—1900(—2400) m. Algunos especimenes de Costa Rica y Panamá, han sido identificados como esta especie (e.g., Liesner et al. 15518, MO), pero estas colecciones en realidad corresponden a especimenes de Dendropanax estériles, género en el cual la presencia de hojas peltadas e irregularmente lobuladas es algo comün en los brotes inmaduros de muchas especies, las cuales se tornan no peltadas con la edad. En Centroamérica Oreopanax peltatus se puede reconocer fácilmente por sus hojas simples, 5—7-lobuladas, peltadas y con la base redon- deada o subcordada a no peltatas y cordadas e inflorescencias con las flores estaminadas de 3-6 mm de diámetro. AGRADECIMIENTOS Deseamos reconocer a los siguientes herbarios por permitir el ingreso a sus colecciones: BM, BR, CGE, COL, , A k An ft 121 CR, CUVC, F, G, HUA, HUQ, INB, JAUM, K, MEDEL, MA, MO, NY, P, Q, QCA, QCNE, TULV, U, W, WAG. Esta investigación fue posible además gracias al convenio de cooperación entre el Ministerio de Ambiente y Energía (MINAE) y el Instituto Nacional de Biodiversidad (INBio) con el fin de completar el Inventario Nacional de Biodiversidad en Costa Rica. REFERENCIAS Cannon, M.J. 8: J.F.M. Cannon. 1986. Studies in the Araliaceae of Nicaragua, and a new widespread species of Oreopanax. Ann. Missouri Bot. Gard. 73:481—485. Cannon, M.J. & J.F.M. CANNON.1989. Central American Araliaceae - A precursory study for the Flora M icana. Bull. Brit. Mus. (Nat. Hist), Bot.19:5-61. Cannon, M.J. & J.F.M. Cannon. 2001. Araliaceae. En: W.D. Stevens, C. Ulloa Ulloa, A. Pool & O.M. Montiel, eds. Flora de Nicaragua Vol. |. Monogr. Syst. Bot. Missouri Bot. Gard. 85:188-192, CORREA, M., C. GALDAMES & M. DE STAPF. 2004. Catálogo de las plantas vasculares de Panamá. Universidad de Panmá, Instituto Smithsonian de Investigaciones Tropicales, Ciudad de Panamá, Panamá. Marcha, E. 1879. Révision des Hédéracées américaines. - Description de dix-huit espéces nouvelles et d'un genre inédit. Bull. Acad. Roy. Sci. Belgique, ser. 2, 47:70-96. PLUNKETT, G.M., J. WEN, AND PP. Lowey II. 2004. Infrafamilial relationships in Araliaceae: insights from plastid (trni-trnF) and nuclear (ITS) sequence data. Pl. Syst. Evol. 245:1-39, STANDLEY, P.C. 1938. Araliaceae. En: P. C. Standley, ed. Flora of Costa Rica. Publ. Field Mus. Nat. Hist., Bot. Ser. 18:851-858. STEVENS, W. & O. Montiet. 2001. Reseña de la exploración botánica. En: W.D. Stevens, C. Ulloa Ulloa, A. Pool & O.M. Montiel, eds. Flora de Nicaragua Vol. |. Monogr. Syst. Bot. Missouri Bot. Gard. 85. 122 [| rnal of tl Das H ID khi HH £ T, BOOK REVIEW JuLes G. Evans. 2008. California Natural History Guides: Natural History of the Point Reyes Peninsula, Second Edition. (ISBN 978-0-520-25467-1, pbk.). University of California Press, 2120 Berkeley Way, Berkeley, California 94704-1012, U.S.A. (Orders: The University of California Press c/o California/ Princeton Fulfillment Services, 1445 Lower Ferry Road, Ewing, New Jersey 08618, U.S.A.; www.ucpress. edu, email ordersOcpfsinc.com, 1-800-777-4726, 1-800-999-1958 fax). $24. i 366 pp, numerous figures, Es images, maps, tables, checklists, index, and references, 5 1/4" x Natural Hi t fl T». T D. L9 | I 11 J 31 1 Tz , geology. | various fauna (e.g. birds, insects marsupials, carnivores pinnipeds, bats whales, M reptiles, hibians) 1fl ities that inhabi this Calif p fthis book (1989, revised 1993), at f hers! jucted studi on Point Reyes, and the inf ion they accumulated ini nint conan this new ie Nw. J ze J D T DP. 1 4, J [94 E + à MES 1 "E Hu Xt hofthereseión'ehabitare- (tr trial, riparian, freshwater r £ É O > Eu and usd wi I I 1 prinklesi hi ical asid fa parti lari I 1 d , and definitions greatly enhancing the overall reader experience. I found this book to be very informative and a good resource for any ici or pum ns is dl to learn more about the diversity of biota found on Point Reyes doa M QUEM The anh size and weig I companion ad any hiker or outdoorsman. The author used easy-to ] der witl d sci The inclusion of a comprehensive checklist of the fauna and ofa with information on whether Pus species is listed as endangered. vulnerable, rare, or threatened further enhanced this book.—Keri McNew, MS Biology, Prog ger, Botanical Research Institute of Texas, 500 E 4" Street, Fort Worth, TX 76102, U.S.A. J. Bot. Res. Inst. Texas 3(1): 122. 2009 FIVE NEW SPECIES OF MYRTACEAE FROM ECUADOR Maria Lúcia Kawasaki Bruce K. Holst Department of Botany Marie Selby Botanical Gardens Field Museum of Natural History 811 South Palm Avenue 00 South Lake Shore Drive Sarasota, Florida 34236-7726, U.S.A. b lb Chicago, Illinois 60605-2496, U.S.A. holst@selby.org lkawasaki@fieldmuseum.org ABSTRACT Five new species of M from Ecuad described and ill d: Calyptranthes compactiflora, C. fusca, C. sparsiflora, Myrcia aequatoriensis, and M. verticillata. RESUMEN Se J } E Y do NM t EE 1 : Calyptranthes compactiflora, C. fusca, C. sparsiflora, Myrcia cea aioe y M. verdeillata: INTRODUCTION Among the Myrtaceae from Ecuador, there are approximately 30 species of Calyptranthes and ca. 40 spe- cies of Myrcia, including many new taxa (Holst 1999; Kawasaki & Holst 2005; Holst & Kawasaki 2006, 2008). In preparation of the Myrtaceae treatment for the Flora of Ecuador, five additional new species of these genera are herein described: Calyptranthes compactiflora, C. fusca, C. sparsiflora, Myrcia aequatoriensis, and M. verticillata. Calyptranthes compactiflora M.L. Kawasaki & B.K. Holst, sp. nov. (Fig. 1). Tre: ECUADOR. Sucumsios (Napo on label): downstream Río Cuyabeno from Laguna de Cuyabeno to Puerto Bolívar, 00°03'S, 76°10'W, 250 m, 14 Apr 1980 (bud), J. Brandbyge, E. Asanza & L. Reib 30543 (noLotYPE: AAU; isotypes: E, QCA, SEL). Arbor, ind to rufo-brunneo , ferrugineo vel flavo-brunneo; ramulis teretis; foliis anguste ellipticis vel ellipticis, chartaceis, 5-8 x s x Te lA TL CEST: if] H lah = 1 E p 3 m longis 1.8-3 cm, abrupto-acuminatis, si | I Tree ca. 4 m tall, the trichomes dibrachiate, reddish-brown, ferrugineous to yellowish-brown; branchlets terete, appressed-pubescent. Leaf blades narrowly elliptic to elliptic, chartaceous, 5-8 x 1.8-3 cm, the up- per surface drying dark-brown, glabrous and lustrous, indistinctly or impressed-punctate, the lower surface reddish-brown in drying, sparsely appressed-pubescent, distinctly dark-punctate; midvein impressed above, convex below; lateral veins 35—40 pairs, parallel, scarcely raised on both surfaces; marginal vein 1, ca. 1 mm from the margin, equaling the lateral veins in prominence; apex abruptly acuminate, the acumen to ca. 1.3 cm long; base cuneate; petioles 5-9 mm long, channeled, puberulous, blackish. Inflorescences of paired, very abbreviate panicles appearing glomerulate by the reduction of the main axes, axillary, with up to ca. 6 flowers, the peduncle 1-2(-10) mm long, pubescent; bracts lanceolate, 3-5 mm long, appressed-pubescent, early deciduous; bracteoles ovate, ca. 2 mm long, appressed-pubescent, deciduous. Flowers: buds closed, obovoid, 3-4 mm long, sessile, appressed-pubescent, glabrescent, the e furfuraceous, yellowish- t; disk ca. 2 mm diam., glabrous; white; calyptra obtuse to shortly apiculate, s parsely pubescent; p stamens numerous, ca. 2-3 mm long, the anthers ca. 0.5 mm long; style ca. 2 mm long; ovary 2-locular; ovules 2 per locule. Fruits not seen. Distribution.—Known only from the type collection from northeastern Ecuador in the Río Cuyabeno region, in riverside forests at 250 m elevation. Calyptranthes compactiflora is readily recognized by the greatly reduced inflorescences, also seen in C. smithii McVaugh, a species known only by the type collection from Guyana. These two species may be dis- tinguished by leaf characters: in C. compactiflora, the leaves are chartaceous, 5-8 x 1.83 cm, the midvein J. Bot. Res. Inst. Texas 3(1): 123 -132. 2009 124 I| | £ JUUurTidi UI Scale = 1 cm FLORA OF ECUADOR Collectad by J. Brandbyye, PF. Mania C, $ L. Peib We. £053 /Á FADEN cam Bie 9 Cuyah no de Cuyabenc to E x C reclivar. Wrasse i Topes E ALE. c. 250 5. 8.13% 903-639) ápral 14, 19 0. all tree 4 u es: upper surface lustrous green, lowe suprace om onareet. puits whitish green, Weed collection in ÁAU. Fic. 1. Calyp h | ctiflora M.L. Kawasaki 8. B.K. Holst (Brandby tal. 30543: Holotype, AAU; i f de + CEI? v L: lua] t, M = £ Mut f, E | 125 is clearly impressed above, and the petioles are 5-9 mm long; in C. smithii, the leaves are coriaceous, 9-12 x 3.5-4.5 cm, the midvein is plane to impressed above, and the petioles are 10-13 mm lon Calyptranthes fusca M.L. Kawasaki € B.K. Holst, sp. nov. (Fig. 2). Type: ECUADOR. CARCHI: Tulcán Canton, Parroquia Tobar Donoso, Sector Sabalera, Reserva Indígena Awa, 01°00'N, 78°24'W, 650-1000 m, 19-28 Jun 1992 (fI), G. Tipaz, J. Zuleta & N. Guanga 1438 (HoLoTiPE: QCNE; isotypes: E MO, SEL) Arbor, ind fl i li is; folii llipticis, ellipticis vel oblanceolatis, chartaceis, 5.5-14 x 3-5 cm, abrupto- E acuminatis, siccati | i iculi ifloris; alal is 1-3 mm longis, petalis nullis; baccis globosis, atropurpureis. Tree 10-20 m tall, 15-40 cm dbh, the trichomes dibrachiate, yellowish; branchlets terete, puberulous. Leaf blades narrowly elliptic to elliptic, or oblanceolate, chartaceous, 5.5-14 x 3-5 cm, drying dark-brown to brownish above, paler below, glabrous on the upper surface, sparsely appressed-pubescent to glabrous on the lower surface, the trichomes located especially along venation; glands distinctly dark-brown, plane above, salient below, also on venation, branchlets, and inflorescences; midvein impressed above, convex below; lateral veins 10—15 pairs, impressed above, salient below; marginal veins 2, the innermost arched, 3-5 mm from the margin, similar to the lateral veins in prominence, the outermost less prominent, paral- lel to the margin, ca. 1 mm from it; apex abruptly acuminate, the acumen 1-1.5 cm long; base cuneate to obtuse; petioles 5-10 mm long, channeled, sparsely pubescent, blackish. Inflorescences of paired panicles, subterminal or axillary, with up to ca. 20 flowers, 2-8 cm long, the branches densely appressed-pubescent, gland-dotted; bracts ovate, ca. 2 mm long, puberulous, early deciduous; bracteoles lanceolate, ca. 1 mm long, puberulous, deciduous. Flowers: buds closed, obovoid to subglobose, of very different sizes, 1-3 mm long, sessile, the lateral flower buds of the terminal triads apparently abortive; hypanthium prolonged ca. 1 mm beyond the ovary, glabrous; calyptra obtuse to minutely apiculate, glabrous; petals absent; disk ca. 2 mm diam., glabrous; stamens numerous, ca. 4—5 mm long, the anthers ca. 0.3 mm long; style ca. 5 mm long; ovary 2-locular; ovules 2 per locule. Fruits berries 1-2 cm diam., globose, crowned by the hypan- thium scar, dark-purple, glabrous, prominently gland-dotted, the pericarp thick and hard; seed 1, ca. 9 x 7 mm, the seed coat membranous; embryo myrcioid, the cotyledons leafy and folded, the radicle elongate, equaling the cotyledons in length. Distribution.—Known from several collections from northwestern Ecuador of the same area (Reserva Etnica Awá) in the provinces of Carchi and Esmeraldas, in humid, lowland to premontane forests at 200-1600 m elevation. The wood is utilized for ax handles and other tools (Tipaz et al. 1290) and the edible fruits to cure diarrhea (Tipaz et al. 1233). Common names are recorded as “ungal nusma” (Tipaz et al. 1290), “unga lusma” (Tipaz et al. 1438), and “pishuilde” (Tipaz et al. 1126). The leaves of Calyptranthes fusca and C. cuspidata DC., a species from Amazonian Brazil and Peru, are morphologically similiar and dark-brown in drying. Calyptranthes fusca is distinguished from C. cuspidata by having the leaves with impressed (vs. convex) midveins above, conspicuous (vs. inconspicuous) and dark glands on both surfaces, along the venation and on the inflorescence axes, and relatively few-flowered (vs. multiflorous) panicles. The varying size of the flower buds is unusual in the family. Additional coliecti ined: ECUADOR. Carchi: Maldonado, Parroquia Tobar Donoso, Reserva Etnica Awá, Sabalera, 00?55'N, 78°32'W, 900 m, 22 Nov 1992 (fr), C. Aulestia et al. 658 (F, MO, QCNE, SEL). Tulcán Cantón, Parroquia Tobar Donoso, Sector El Baboso, Reserva Indígena Awa, 00°53'N, 78°00'W, 1600 m, 3 Oct 1991 (fr), G. Tipaz et al. 310 (F, MO, QCNE, SEL). Tulcán Cantón, Parroquia Chical, Sector Gualpi medio, Reserva Indígena Awá, 01°02'N, 78°16'W, aD 23-27 May 1992 (fr), G. Tipaz et al. 1126 (F, MO, QCNE, SEL). Tulcán Cantón, Parroquia Tobar Donoso, Sector Sabalera, R igena Awá, 01°00'N, 78°24'W, 650-1000 m, 19-28 Jun 1992 (fr), G. Tipaz et al. 1233 (F, MO, QCNE, SEL). Tulcán Cantón, Parroquia Tobar Donoso, Sector Sabalera, Reserva Indígena Awá, 01?00'N, 78°24'W, 650-1000 m, 19-28 Jun 1992 (fl), G. Tipaz et al. 1290 (F, MO, QCNE, SEL). Esmeraldas: San Lorenzo Cantón, Parroquia Mataje, Reserva Etnica Awa, Centro Mataje, on the banks of Rio Mataje, 01°08'N, 78?33'W, 200 m, 21 Sep 1992 (fr), C. Aulestia et al. 551 F, MO, QCNE, SEL). San Lorenzo Cantón, Parroquia Mataje, Reserva Etnica Awd, Centro Mataje, on the banks of Río Mataje, 01°08'N, 78*33'W, 200 m, 21 Sep 1992 (fr), C. Aulestia et al. 568 (F, MO, QCNE, SEL). Calyptranthes sparsiflora M.L. Kawasaki & B.K. Holst, sp. nov. (Fig. 3). Type. ECUADOR. SUCUMBÍOS: Lago Agrio Cantón, Reserva Cuyabeno, Laguna Canangueno, 00%02'S, 76°13'W, 230 m, 18 Nov 1991 (ID, W. Palacios, G. Tipaz, D. Rubio, E. Gudiño & C. Aulestia 9126 (HoLotYrE: QCNE; IsorYPE: MO, SEL) 126 N? 2284711 FIELD MUSEUM NATURAL HISTORY Aa TO nel ield copyright reserved Muscum eon m. de Herr sumi antas. Nubes. ma a d 19-22 dunio 1992 tN. batuta 1438 Gaio ` & HENBARIO Ya IOMA "3 UN "ECUADOR (QCNE HISSOURT BOTANICAL GARDEN HERBAXIUM (MO) & CEIN Kawasaki & B.K. Holst ( Tipaz et dl. 1438: Isotype, F; Fic. 2. G y MISSOUFR BOTANICAL GARDEN HERBA 4 ETUR E Ne 94821285 lsotyg. o; GUAD i Hera MP Rawawkr & BA. Hois, ey nov, iac. Fic. 3. Caly M.L. Kawasaki & B.K. Hol In. t al. 9126: Isotype, MO: i + CED 127 r1 D A D ls lactic £ T. 128 Journal of t t Texas 3( Frutex vel arbor, indumen to rufo-l li inatis; folii llipticis, chartaceis, 4—6.5 x 1.5-2.5 cm, caudato-acuminatis; > o inflorescentiis 1 vel 3 floris, dense appresso-pubescentibus; baccis globosis. Shrub or tree 1.5—4 m tall, the trichomes dibrachiate, reddish-brown, to 0.5 mm long; branchlets bicari- nate, thinly appressed-pubescent to glabrous. Leaf blades narrowly elliptic, chartaceous, 4-6.5 x 1.5-2.5 cm, the upper surface drying olive-green to dark-brown, glabrous, indistinctly or impressed-punctate, the lower surface light-brown in drying, sparsely appressed-pubescent, minutely punctate; midvein impressed above, convex below; lateral veins 30—40 pairs, parallel, scarcely raised on both surfaces; marginal vein 1, 0.5-1 mm from the margin, equaling the lateral veins in prominence; apex caudate-acuminate, the acumen to ca. 1.5 cm long; base obtuse to cuneate; petioles 1-3 mm long, channeled, puberulous. Inflorescences paired, subterminal, with 1 or 3 subsessile flowers, borne at the end of a 1.3-2.8-mm long peduncle, densely appressed-pubescent; bracts and bracteoles early deciduous, not seen. Flowers: buds not seen, the hypan- thium 1.5-2 mm long, densely appressed-pubescent; calyptra ca. 1 mm long, shortly apiculate, densely to loosely appressed-pubescent; petals not seen; disk ca. 2 mm diam., glabrous; stamens not seen expanded, the anthers ca. 0.3 mm long; style ca. 7 mm long. Fruits berries, immature, globose, 8-9 mm diam., yel- low, appressed-puberulous, the trichomes located especially at the base and by the hypanthium scar; seed 1, ca. 8 x 7 mm, the seed coat membranous; embryo myrcioid, the cotyledons leafy and folded, the radicle elongate, equaling the cotyledons in length. Distribution.—Known from eastern Ecuador in Sucumbíos and Pastaza provinces, in primary rain forests at 200—230 m elevation. Calyptranthes sparsiflora is characterized by the very reduced, 1- or 3-flowered inflorescences, that are densely appressed-pubescent with reddish-brown mne ns the species from Ecuador, it resembles C. bipennis O. Berg on branchlet, leaf, and infl y. These two species may be distinguished by the presence of indumentum on young vegetative MN [e inflorescences, flowers, and fruits in C. sparsiflora (ys. mostly glabrous in C. bipennis). Additional collection examined: ECUADOR. Pastaza: Lorocachi, SW of the military camp, 01?38'S, 75°58’W, 200 m, 26 May 1980 (fr), J. Jaramillo et al. 31106 (AAU, F, QCA). Myrcia A id L. Kamarak & B.K. Holst, sp. nov. (Fig. 4). Tre: ECUADOR. Sucuwsios: Reserva Faunistica Cuyabeno, gs, including Río Cuyabeno from Puerto Bolívar to above Laguna Canangueno, 00°00'S, 76°10'W, 2651 m, 11 Mar 1990 (fl), H. Balslev, C.C. Berg, M. Gavilanes, A. Thygesen, D.E. Christensen, L. Ellemann & R. Brucculeri 97496 (HOLOTYPE: AAU; sorres: ASU, E QCA, SE Frutex vel arbor, indumento flavescenti vel cano-flavescenti; ramulis teretis; foliis lanceolatis, chartaceis, 5.6-14 x 1.6—3 cm, longi acuminatis, basi obtusis; paniculis axibus gracilibus; baccis ellipsoideis. Shrub or tree 2—10 m tall, the trichomes simple, yellowish to yellowish-white; branchlets terete, densely pubescent when young, peeling in thin strips when slightly older. Leaf blades lanceolate, 5.6—14 x 1.6-3 cm, membranous to chartaceous, the upper surface glabrous except for the midvein, drying olive-green to brownish, the lower surface paler, sparsely appressed-pubescent, the trichomes present especially along the venation; apex long-acuminate; base obtuse; midvein impressed above, pubescent, convex below; lat- eral veins 20—30 pairs, impressed above, sharply convex below; marginal vein 1, to ca. 1 mm from blade margin, indistinct above, slightly salient below; glands numerous, indistinct or minutely punctiform on both surfaces; petiole 1-3 mm long, channeled, densely pubescent. Inflorescences paniculate, axillary or mostly subterminal, with to ca. 20 flowers, 3—6.5 cm long, the axes appressed-pubescent, filiform; bracts and bracteoles early deciduous, not seen. Flowers 5-merous; buds obovoid to io d mm 1 long; calyx-lobes triangular, to 1 mm long, appressed-pubescent without, glabrous within; p , Cà 2 mm diam., sericeous without; stamens numerous, the filaments ca. 3-4 mm long, the anthers ca. 0.3 mm long; style ca. 4 mm long, the stigma punctiform; hypanthium not prolonged beyond the ovary, sericeous without; disk ca. 2 mm diam., densely pubescent; ovary 2-locular; ovules 2 per locule. Fruits (immature) berries, ellipsoid, ca. 10 x 8 mm, crowned by the persistent calyx lobes, yellow, gland-dotted, puberulous Hire TOR W^ FLORA or. E ECUAD: OR Collected by H. Balsiov, M. Gavilanes, A, Thygesen, D. E. Christ solacia mian BR: Brucculeri 97496 129 Helotype of: Province, SUCUMBIC 5 m PETTEE” OE ; Cae, Pto, Bolívar to above ¡o rola AA Mt (76° 10 W 00° OY S). ALL 265 m. 11 Mar 1990 n. A hu A Shrub 3 m tall, Terra firme north of Laguna nnd NAAT Z AM P — A Se ee e ha Botanical bes Aarhus University, Denmark (AAT: in eo! abotation iversidad Católica, Quita, deus (QCA) is M.L. Kawasaki & B.K. Holst (Balsi |, 97496: Holotype, AAU; inset from isotype at SEL). Fic. 4. Myrcia aeq £ 4L Das s.. 1D Llar ET 130 Journal of Texas 3( to glabrous; seed 1, ca. 8 x 7 mm, the seed coat membranous; embryo myrcioid, the cotyledons leafy and folded, the radicle elongate, equaling the cotyledons in length. Distribution.—This species is known from a few collections from Sucumbíos (Reserva Faunística Cuy- abeno) in northeastern Ecuador, and one collection from Imbabura in northwestern Ecuador; it occurs in lowland tropical rainforests at 220-500 m elevation. Myrcia aequatoriensis belongs to sect. Myrcia, among the group of species related to the M. fallax (Rich.) DC. — M. splendens (Sw.) DC. complex. These species are separated by characters of the indumentum, leaf morphology, including venation and glands, inflorescences, and size of flower buds (McVaugh 1969). From this group, M. aequatoriensis is distinguished by the combination of these characters: lanceolate, thin, sub- sessile leaves, long-acuminate at the apex, obtuse at the base, with impressed lateral veins on the upper surface, and by delicate panicles with filiform axes. Addi l collecti ined: ECUADOR. Imbabura: Lita, 501 m, 28 Apr 1949 (fr), M. Acosta Solís 12296 (F). Sucumbios: Reserva Faunística Cuyabeno, Laguna Grande and surroundings, including Río Cuyabeno from Puerto Bolívar to above Tapuna E m 00°00'S, 76°10'W, 265 m, 11 Mar 1990 (fl), H. Balslev et u 97071 (AAU, QCA). Reserva Faunística Cuyabeno, I roundings, including Río Cuyabeno from Puerto Bolívar to above Laguna Canangueno, 00°00'S, 76°10'W, 265i m, 11 Mar 1990 (f), H. Balslev et al. 97217 (AAU, ASU, F, QCA, SEL). Estación Científica Cuyabeno, 220 m, 21 Jul 1992 (fr), J. Jaramillo 14913 (NY, QCA). Reserva Faunística Cuyabeno, 1 ha plot ca. 1 km N of Laguna Grande and surroundings, 00°00'S, 76°12'W, 265 m, 11 Apr-10 Jun 1988 (st), A.D. Poulsen 78348 (AAU-2 sheets, QCA). Myrcia verticillata M.L. Kawasaki € B.K. Holst, sp. nov. (Fig. 5). Tere: ECUADOR. ORELLANA (Napo on label): Estación Científica Yasuní, Río Tiputini, NW of confluence with Río Tivacuno, 6 km E of Maxus road, Km 44, detour to Tivacuno well, parcela 50 ha, 00938"S, 76930'W, 200—300 m, 21 Mar 1996 (fr), K. Romoleroux & R. Foster 2147 (HoLoTveE: QCA; isotypes: F-2 sheets, SEL). Frutex vel arbor, indumento branie vel O Sa ramulis quadrangulatis; foliis verticillatis, RN MEN coriaceis, 19-29 x lis 5-13 cm longis; baccis globosis, E Is E Shrub or tree qud m tall, to ca. 20 cm dbh, the trichomes simple, brown to yellowish-brown; bark red- dish, the stems lar in cross section with rounded angles, densely pubescent. Leaves subsessile to short-petiolate, verte biet in 4-merous whorls; blades oblanceolate, 19-29 x 4—7.5 cm, coriaceous, the upper surface drying dark brown to brownish, puberulous, the lower surface paler, sparsely hirsute-pubescent, the trichomes present especially on the venation; apex abruptly acuminate; base obtuse; midvein hirsute, impressed above, convex below; lateral veins 15—20 pairs, these and major interconnecting tertiary veins impressed above, convex below; marginal veins 2, the innermost 2-4 mm from blade margin, similar in prominence to the lateral veins, the outermost less than 1 mm from margin; glands numerous, punctiform, indistinct above, evident below; petiole 3-9 mm long, stout, channeled, densely pubescent. Inflores- cences paniculate, axillary, 5-13 cm long, the axes hirsutulous; bracts and bracteoles early deciduous, not seen. Flowers 5-merous; buds not seen; calyx-lobes ca. 1 x 2 mm, truncate to broadly rounded at apex, appressed-pubescent to puberulous without, glabrous within; petals, stamens, and style not seen; hypan- thium not prolonged beyond the ovary; disk ca. 2 mm diam., hirsutulous. Fruits berries, globose, 1-1.7 cm diam., turning from green to yellow-orange to red, and finally purple-black (Acevedo-Rdgz. & Cedeño 7628), puberulous, crowned by the slightly overlapping calyx lobes; seeds 1 or 2, ca. 8-10 x 7-8 mm, the seed coat membranous; embryo myrcioid, the cotyledons leafy and folded, the radicle elongate, equaling the cotyledons in length. Distribution.—Ihis species is known only from northeastern Ecuador in Orellana province, in the region of Yasuní National Park; it occurs in primary, lowland wet forests, at 200—300 m elevation. From all the other species of Myrcia in Ecuador (Holst 1999; Holst & Kawasaki 2008), M. verticillata is promptly recognized, even in sterile condition, by the large, oblanceolate, dark brown leaves (when dry), that are subsessile and verticillate, in tetramerous whorls, and by the quadrangular stems. Species of Myr- cia with both opposite and verticillate leaves were described by Berg (1857) in Martius' Flora Brasiliensis, but at least in northern South America, M. verticillata is the only known species with leaves consistently in tetramerous whorls. Myrcia verticillata may be allied with the Gomidesia group of Myrcia; flowers are needed to confirm the relationship. Lt lll Al s ERA N° 2174920 FELD MUSEUM NATURAL HISTORY id C. BRK. tisk so. iaw om MVECACUM- ESTACION GOES EU A YAZLUNE Yan cs de la carretera Marus, km 44, dos vio hacia - Pi Tano, Purcola 30 Ha. 7630 W, CSRS fere firme; Lomas de 30-50 m, de arcilla rojizo. Sure e) A No. es i Katye pan 2147 21 maa 1996 D: i Resin "este ity an CATOLICA - HERBARIO (QU AY Museum » copyright raserved 131 E] Fic. 5 M) Kawasaki & B.K. Hol ID. Neil et al. 8233, SEL) 132 Journal of the Botanical R h Institute of Texas 3(1) Additional collections examined: ECUADOR. Orellana: Yasuní Forest Reserve, along road between Km 70 and 100, E of PUCE Scien- tific Station, 0°50.014’S, 76220.518"W — 0%54.730'S, 76?13.304"W, 200 m, 2 Jul 1995 (fr), P. Acevedo-Rodriguez & J.A. Cedeño 7628 (SEL, US); Yasuní, Parque Nacional Yasuní, oil well Conoco-Amo 2, 00?57S, 76?13"W, 230 m, 9-19 Jan 1988 (fr), D. Neill et al. 8233 (F, MO, SEL); Aguarico, Yasuní, Reserva Etnica Huaorani, Maxus road and pipeline construction project, Kms 98-99, 00°55’S, 76*13'W, 250 m, 18 Jun 1994 (fr), N. Pitman & G. Romero 299 (F, MO, QCNE, SEL); Estación Científica Yasuni, Río Tiputini, NW of confluence with Río Tivacuno, E of Repsol-YPF road, Km 7, detour to Tivacuno well, parcela 50 ha, column 29-02(2,3), 00°38'S, 76°30"W, 200-300 m, 17 Feb 2001 (fr), G. Villa & L. Velez 912 (F, QCA). ACKNOWLEDGMENTS We are grateful to the curators and staff of the herbaria cited in the manuscript for providing the collections used in this study, especially to Henrik Balslev and Benjamin Dilgaard from AAU, and Ron Liesner and Jim Solomon from MO. Fred Barrie, Les Landrum, and Jim Luteyn offered helpful suggestions to improve this manuscript. REFERENCES Bera, O. 1857-1859. Myrtaceae. In: C.P. von Martius, ed. Fl. Bras. 14(1+suppl.):1-655. Ho st, B.K. 1999. Myrtaceae. In: P.M. Jørg d S. León-Yánez, eds. Catalogue of the vascular plants of Ecuador. Monogr. Syst. Bot. Missouri Bot. Gard. 75:618-622. Hoist, B.K. AND M.L. Kawasaki. 2006. New species of Myrtaceae from Ecuador and Peru. Sida 22:931-934, Ho st, B.K. and M.L. Kawasaki. 2008. New species of Myrtaceae from Ecuador. J. Bot. Res. Inst. Texas 2:297-303. Kawasaki, M.L. AND B.K. Hoist. 2005. Two new species of Calyptranthes (Myrtaceae) from Ecuador. Sida 21:1955- 1960. McVaucn, R. 1969. The botany of the Guayana Highland—Part VIII. Myrtaceae. Mem. New York Bot. Gard. 18(2):55-286. FOLIA TAXONOMICA 12. PARADRYMONIA (GESNERIACEAE: EPISCIEAE) FROM THE GUIANA SHIELD: P. MAGUIREI, A NEW SPECIES FROM AMAZONAS, AND DISTRIBUTION AND FLORAL MORPHOLOGY OF P. MACULATA Christian Feuillet Department of Botany, MRC-166 Smithsonian Institution, PO. Box 37012 Washington, DC 20013-7012, U.S.A. feuillec@si.edu ABSTRACT Parad i irei is described f I fA Venezuela. The distribution of Paradrymonia maculata in Venezuela and the Guiya is enel: as well as its unusual corolla morphology. RÉSUMÉ Paradry i irei est décrit de l'état d'A Venezuela. La répartition de Parad i lata au Venezuela et dans les i M 11 eid ta th i 1 11 Ap f, RESUMEN Se describe P be puli del estada de a Msn bord se documenta la distribución de Paradrymonia maculata en Venezuela y las Hanstein (1854: 206) described the genus Paradrymonia Hanst. with only one species, P. glabra (Benth.) Hanst. (1854, p. 207, fig. 43), which is currently a synonym of P. ciliosa (Mart.) Wiehler. The name Paradrymonia was in use only for a short time as Hanstein (1864) reduced it to a synonym of Episcia Mart. When Wiehler (1973) re-established Paradrymonia, he transferred species from other genera. From Drymonia Mart. he took species that did not have the typical Drymonia anther dehiscence through a basal pore, and from Episcia Mart. he moved species that lacked stolons and did not otherwise belong to Nautilocalyx Linden ex Hanst. Currently Paradrymonia includes 38 species. Recent molecular studies (Clark et al. 2006) have shown that it is likely that the species of Paradrymonia will separate into two natural groups of species, mostly with large leaves (20—50 cm long) and small axillary inflorescences, and a few isolated species. The larger group, true Paradrymonia, would mostly include species with petioles longer than the inflorescences and either with stems 5-10 mm thick and rooting at nodes on the substrate, and the smaller group has thinner and shorter stems and a “rosette-like” habit (Wiehler 1978). Those two groups are likely to stay in Paradrymonia. They are present in continental America from Mexico (Oaxaca) to Bolivia (Cochabamba) and eastward to Brazil (Amapá). Afew Species in eee have tain affinities. P. anisophylla Feuillet & L.E. Skog is an epiphyte with hang gly unequal in a pair; the molecular data (Clark et al. 2006) suggest that it may not be a Paradrymonia. The same data set places outside Paradrymonia an epiphyte with erect thick stems, P. longifolia (Poepp.) Wiehler. Paradrymonia campostyla (Leeuwenb.) Wiehler and P. barbata Feuillet & L.E. Skog from the Guianas are climbers with smaller leaves (3-15 cm long) and 1(-3) axillary flowers. Here two other hard-to-place species are dealt with: P. maguirei, a new species from Amazonas (Venezuela), which is vegetatively unlike other Paradrymonia species with a rosette-like habit, short petioles, and sharply biserrate paper-thin leaves, when dry; and P. maculata (Hook. f.) Wiehler with large condensed inflorescences and large bracts (see below) that is endemic to the Guiana Shield. J. Bot. Res. Inst. Texas 3(1): 133 -138. 2009 134 i itute of Texas 3( A.—Paradrymonia maguirei Feuillet, sp. nov. Although it seems obvious that the genus Paradrymonia will prove to be polyphyletic (Clark et al. 2006), in the absence of a comprehensive molecular study of Paradrymonia, along with Nautilocalyx and Chrysothemis Decne., the only reasonable option at the moment is to place this new species in Paradrymonia as P. magu- irei. It does not belong in any of the other genera of the Episcieae (Weber 2004; Skog & Boggan 2006) and anticipating the split of Paradrymonia by describing a new genus without the proper data would be taking a high risk of creating a generic synonym. Paradrymonia maguirei Feuillet, sp. nov. (Fig. 1). Tree: VENEZUELA. AMAZONAS. Depto. aan od al adc d forest, 1000 m, 3 May 1949, B. Maguire & B. Maguire, Jr. 29185 (Hototyre: NY, pro p I = [P marahuacana Wiehler] invalid: in sched. iin d maguirei ab aliis speciebus a chatactenbus angen o "dus pro parte manie praeter infra foliis intervenia, 10-11 x 7-8 cm, basi asym- vestita, petiolo 2-2.5 cm longo, 1 velutina PI metrica, apice obtuso rotundo, margine bi- serrata, in sicco FRANE Epiphyte or saxicolous. Stem creeping, 0.5 cm thick or more, 5 cm long (in the type collection), with a dense brown-red indumentum, apical few internodes with leaves, about 3 mm long, forming some kind of a loose pauci-leaved rosette. Leaves opposite, strongly unequal in a pair, the smaller about 2 cm long including petiole, ligulate, 2-3 mm wide; the larger with petiole 2-2.5 cm long, thick, covered with dense, long, brown-red trichomes; blade membranous when dry, elliptic, 10—11 x 7-8 cm, asymmetrically acute to obtuse at base, widely rounded at apex, margin sharply biserrate, above velutinous or appressed-pubescent, beneath appressed-pubescent or hirsute on veins. Inflorescence axillary, fasciculate; pedicels up to 2 cm long, with a dense, long, brown-red indumentum. Flowers with sepals lanceolate, long acuminate, 0.8-1.3 x 0.2 cm, with a dense, long, brown-red indumentum; corolla oblique in the calyx, with red trichomes outside, basal gibbosity 1-1.5 x 2 mm, tube cylindric, 1.8-2 cm long, lobes suborbicular, 0.8 x 0.6 mm, undulate at margin. Fruit not seen. Distribution.—Paradrymonia maguirei is known only from the type collection from the area North of La Esmeralda in the Duida-Marahuaca National Park, on a forested slope of the Cerro Marahuaca (Amazonas, Venezuela), 65?24"W 3°40'N according to maps, at 1000 m elevation. It was blooming in May. The color of the corolla is not known. There are two specimens that I marked A and B on the herbarium sheet in the New York herbarium. Specimen A, including the fragments in the pocket, is the type of the new species; it is on the middle left of the sheet and the pocket on the lower right above the label. Specimen B (sterile), on the upper right, is probably Nautilocalyx cordatus (Gleason) L.E. Skog. The only open corolla, preserved in the pocket, is very unlike the corolla of N. cordatus and is the one described here. The affinities of Paradrymonia maguirei in the genus are not clear. This species shows a unique combination of characters: rosette-like habit, petioles short, with dense, long, appressed pubescence, leaf blade drying paper-thin, asymmetric at the base, broadly rounded at the apex, and sharply biserrate at the margin. Other species with short petioles have leaf blades long-decurrent or are long stemmed epiphytic climbers. This species was Paradrymonia “sp. E" in the text and the key (Feuillet & Steyermark 1999). Etymology.—The epithet maguirei refers to the senior collector, a great botanist, collector, and student of the flora of the Guiana Shield. B.—Subgenus Pagothyra Like other infrageneric taxa, Episcia sect. Pagothyra Leeuwenb. coined for E. maculata Hook. f. (Leeuwen- berg 1958: 312) was never transferred to Paradrymonia, although P. maculata stands alone there as well as in Episcia. Paradrymonia subg. Pagothyra (Leeuwenb.) Feuillet, comb. et stat. nov. Basioww: Episcia sect. Pagothyra Leeuwenb., Blumea 7:312. 1958 Feuillet, New species of Paradrymonia The Sow Virk ülantcal Gordon Planie al THE RUNUARDT VESEZEELAN FAPEDIMOY gsis gs Verri Xacahboaza, Teitilnoiia (ee) cman ^a 29155 3 f Kd wA 0o, Sa d i ee LA d 6 X Siape forest, alt. 1000 meters. WBasaedi Ya salia lassi Maguire, Ji Way 5, ar TE —— dai NT E IST Fic. 1. Parad, B. Maguire & B. Maguire, Jr. 29185 (NY) j photograph by € F 136 Journal of i i Texas 3( Paradrymonia maculata (Hook. f.) Wiehler, Selbyana 5:57. 1978. Episcia maculata Hook. f., Bot. Mag. 116: pl. 7131. 1890. [Nautilocalyx maculatus] Wiehler 1970, invalid: in Varie Tere: ee Origin GUYANA: Cult. Hort. Kew.. 2 Sep 1889 (fl), collector unknown s.n. (HOLOTYPE: K; soTyPE: K; pl type BH, NY, U, US, WAG). E Paradrymonia maculata climbs by way of short roots at the nodes and along internodes, similar to Hedera helix L. The stems are tightly applied to the bark and have been reported to grow 1-3 meters high. The leaves are opposite and equal or subequal in a pair with a long petiole; the blade is large and broadly elliptic, with serrate margins. The pedunculate inflorescences are axillary, unilateral cymes, with bracts that are large, greenish yellow with purplish or reddish veins. The corolla is creamy yellow with reddish dots and the ventral lobe acts as a cover closing the tube. Longer descriptions can be found in Leeuwenberg (1958) and Skog & Feuillet (2008). Distribution.—Paradrymonia maculata is known from the forests of French Guiana, Guyana, and Ven- ezuela (Delta Amacuro) at 0—500 m. It has been collected in bloom every month of the year and in fruit in March, June, August, and December. Corolla.—As noted and illustrated in Hooker (1890), the ventral lobe of the corolla of Paradrymonia maculata closes the tube (Fig. 2). The nectary gland is in dorsal position at the base of the ovary, next to the basal gibbosity forming a nectar chamber at the base of the 2.7-3.5 cm long corolla tube. The ventral corolla lobe is effectively a barrier between the nectar produced at the base of the tube and most pollinators. Exert- ing pressure on both sides on the apical third of the tube flips the ventral lobe from its position of convex lid closing the throat to a more classic concave corolla lobe by moving it more than 90°. It allows access to the nectar to strong pollinators, possibly carpenter bees. The bracts and sepals are pale yellow with red or purple veins. In the Guianas similar corolla morphology is found in a Solanaceae, Marhea formicarum Dam- mer, where the lower lobe closes the throat. That species is lacking bracts, but the large calyx is cream- or straw-colored with purple veins, showing a similar color pattern as the bracts and calyces of P. maculata. It might be of interest to note that an Asian Gesneriaceae, Agalmyla chorisepala (C.B. Clarke) Hilliard & B.L. Burtt, has orange corollas with the ventral lobe closing the tube, but in this example the lobe flips at anthesis and opens the access to the tube without further obstacle to pollination. The type of Paradrymonia maculata (Hook. f.) Wiehler from an unknown collector, comes from a plant cultivated at the Royal Botanic Gardens, Kew. It was blooming in September 1889. The plant was grown from material collected in British Guiana, now Guyana. Subsequent collections all came from Guyana and this limited distribution was acknowledged by Leeuwenberg (1958) and Wiehler (1978). During the comple- tion of the treatment of the Gesneriaceae for Flora of the Guianas (Skog & Feuillet 2008), collections of P. maculata from Venezuela and French Guiana came to my attention. That species was not mentioned in the treatment for Flora of the Venezuelan Guayana (Feuillet & Steyermark 1999) but is present in the Antonio Díaz Department, the part of the state of Delta Amacuro neighboring Guyana. No collections from Surinam have been made, but it is likeiy that it is, or has been, present there. Material studied: VENEZUELA. Delta Amacuro. Depto. Antonio Díaz: low forest, 12 km S of San José de Amacuro, 8?28'N 60°27'W, sea level, Feb 1987 (fD, A. Fernandez 3903 (MO, MYF, NY, PORT, US); primary rain forest, Río Grande, 60 km NE of El Palmar, about 8°25'N 61°45'W, 120 m, 15 Mar 1987 W, e ana 5411 (PORT). GUYANA. Arawai Creek, ight! Riv., Feb 1952 (fD, C.A. Persaud 140 = F.D. (Forestry Sd 6897 (K, NY, U); Essequibo River, cain IE Groete Creek, 14 Ap: 1944 (f), B. Maguire & D.B. Fans) 22826 (F, GH, K, NY, U, US); Morawhanna, Oct 1905 (fD, A.W. B 00 (K); Rockstone, 15 July — 1 Aug 1921 (fl), H.A. Gleason 664 (NY); Unabaruka p Aug 1930 (fD, E.B. a 225 (BRG). Barima. Waini: Barima Riv., 15 mi E of Arakaka, 7°37'N 59°54'W, 38 m, 26 July 1986 (fD), J.J. Pipoly 8059 (BRG, CAY, Y, US); Barima Riv., May 1907 (fl), R.D. Ward s.n. (K); Baramita airstrip — Millionaire trail, 7°22'N 60°28'W, 91 m, 3 Apr 1991 (fl), T.D. o et al. 4182 (NY, US); Matthews Ridge, Barima Riv., 23 Jan 1955 (fl), R.S. Cowan 39337 (NY, US); T between Aruau Riv. & Yarikita Riv., 8°00'N 59°55'W, 17 Jan 1920 (fT), A.S. Hitchcock 17601 (GH, K, NY, S, US); Sebai Riv., + 5 km of Sebai Village, 15-20 m, 7°49'N 59°57'W, 16 Dec 1991 (fr), B. Hoffman et al. 615 (US); Upper Aruau Riv., Áruka Riv., Apr-May 1929 > E.B. Martyn 53 (K); Upper Kaituma R.; 3 km w of Port Kaituma, 7°42'N 59°54°W, 0—5 m, 8 Dec Mid (fD, B. Hoffman & H. Benjamin 525 (US); Waini Riv., July 1906 (fn, J. E. Beckett s.n (K, U). Cuyuni Mazaruni: Aurora. heli 1; 6°47'N 59?44"W, 4 Oct 1989 AD, LJ. Gillespie 2084 (US; Bartica, 12-15 mi f t 28 Aug 1935 (fr), D. Potter 5358 (GHD: Essequibo county, near Mazaruni Forest Station, 9 Aug 19 (fD, W.A. Archer 2432 (BRG, K, US); id., July-Sep 1942 (fD, D.B. Fanshawe 785 = F.D. 3521 (K). Essequibo Islands. West Demerara: Macouria Riv., right bank of Lower Essequibo Riv., Nov 1886, G.S. Jenman 2419 (K); Upper White Creek, near Blue Mountain, 6°35'N Feuillet, New species of Paradrymonia 137 Fc? Pp Ir ; lata, phot ph by Chris D 58°43'W, 5-20 m, 14 Apr 1993 (fD, T.W. Henkel et al. 1879 (NY, US). Pomeroon. Supenaam: Pomeroon District, Mt. Russell, Mar 1886 (fl), G.S. Jenman 2097 (K); Pomeroon Riv., 20 Aug 1959 (fI), V. Graham 352 (K); Pomeroon Riv., Yawiami Creek, Aug 1882 (fI), G.S. Jen- man 1939 (K); Pomeroon Riv., Pomeroon apis 17-24 Dec 1922 (fl), J.S. de la Cruz 3124 (F, GH, MO, NY, PH, UC, US); id., 14-20 Jan 1923 (fl&fr), J.S. de la Cruz 3022 (GH H, US); id., Mar 1884 (fD, G.S. Jenman 2002 (K, NY); Abrahms Creek, Mar 1904 (st), "G.S. Jenman” 7808 ER 3km Pu of AM Mission Mig sa 7?15'N 58°45'W, 0-10, 25 Sep 1992 (fl bud), B. Hoffman & L. Roberts 2839 (NY, US). Potaro. S : Garraway stream, 102,5 mi on Bartica — Potaro rd., 5°22'25"N 59°7'20"W, 38 m, 12 Mar 2004 (fl), K.M. Redden 2231 (US); ieee inf R pukari — Annai Road, 4°28'14"N 58?47'16"W, 400—500 m, 21 Mar 1997 (fD, H.D. Clarke, S.A. Mori €: S. Heald 4181 (US); Kaieteur a '5°10'N 59°29'W, 23 Oct - 3 Nov 1923 (fD, J.S. de la Cruz 4395 (F, GH, NY, PH, US, VEN); Potaro, 10 mi S of Potaro landing, 5°10'N 59°00'W, 7-8 Jan 1920 (fl), A.S. Hitchcock 17397 (GH, K, NY, S, US); North Fork Riv., 0.5-1.5 km N of Konawark Riv., 5°9'N 59°8'W, 137 m, 18 May 1991 (fl), T.D. McDowell, C.L. Kelloff & A. Stobey 4819 (US). Upper Demerara. Berbice: Haiowa Falls, Essequibo Riv. basin, 5°7'N 58°49'W, 27 Sep 1937 (fD, A.C. Smith 2123 (F, G, GH, K, NY, S, U, US); Mabura as W inn: compartment, 5°01'95"N 58?37'73"W, 12 Oct 1993 (fI), R.C. Ek, P.].M. Maas, H. Mass & C. Górts 942 (U, US); Mabura region, Holder Falls, 5°20'N 58°10'W, 21 Aug 1993 (fI), R.C. Eh, R. Zagt, L. Brouwer & N. Eernisse 896 (US). Upper Takutu. Upper Esequibo: Maparri R., S bank, 3°20'N 59°15'W, 3 June 1996 (st), H.D. Clarke & T. McPherson 1928 (US); Upper Rupununi Riv., near Dadanawa, 2°45'N 59°31'W, 13 June 1922 (fl&fr), J.S. dela Cruz 1518 (CM, F, MO, NY, PH, US); id., 13 June 1922 (fl), J.S. de la Cruz 1535 (CM, GH, F, MO, NY, PH, UC, US) 138 Journal of the Botani Insti Texas 3(1) FRENCH GUIANA. Approuague Riv. Basin: Crique Cascade, 390 m, 15 Mar 2002 (fl&fr), J.F. Smith, E. Teppe & C. Davidson 4134 (CAY). Matoury: Plateau de Nancibo, 4°40'N 520°30'W, 24 Oct 1983 (fI), F. Billiet & B. Jadin 1857 (BR, CAY). Oyapock Riv. Basin: Crique Gabaret, 3°55'42"N 51?48'7"W, 15 Apr 1988 (fD, G. Cremers 9951 (CAY, NY, P, U, US); Roche Touatou, 130 m, 20 May 1995 (fD, J.-J. de Granville & G. Cremers 13009 (CAY). ACKNOWLEDGMENTS I thank Alain Chautems and an anonymous reviewer for their careful reading of the manuscript. I am grate- ful to Eduardo Garcia-Milagros who translated the abstract into Spanish. This work could not have been completed without the help of the curators of the herbaria CAY, K, MO, NY, P, and VEN who made available to me the material in their care. This is number 144 in the Smithsonian's Biological Diversity of the Guiana Shield Program publication series. REFERENCES CLARK, J.L., PS. HERENDEEN, L.E. SKOG, AND E.A. Zimmer. 2006. Phylogenetic relationships and generic boundaries in the Episcieae (Gesneriaceae) inferred from nuclear, chloroplast, and morphological data. Taxon 55:313-336. FEUILLET, C. AND J.A. STEYERMARK. 1999. Gesneriaceae. In: Steyermark, J.A., PE. Berry, K. Yatskievych, and B.K. Holst, Flora of the Venezuelan Guayana, vol. 5. Missouri Botanical Garden Press, St. Louis. Pp. 542-573. HANSTEN, J. 1854. Die Gesneraceen des Kóniglichen Herbariums und der Gárten zu Berlin, nebst Beobachtungen über die Familie im Ganzen I. Abschnitt. Linnaea 26:145-216; fig. 1-68. HANSTEIN, J. 1864. Gesneraceae. In: Martius, Flora Brasiliensis 8(1):341—428; pl. 58-68. Fleischer, Leipzig. Hooker, J.D. 1890. Paradrymonia maculata. Bot. Mag. 116: pl. 7131. LEEUWENBERG, A.J.M. 1958. The Gesneriaceae of Guiana. Acta Bot. Neerland. 7:291-444. SkoG, L.E. AND J.K. BOGGAN. 2006. A new classification of the Western Hemisphere Gesneriaceae. Gesneriads 56(3):12-17. Skoc, L.E. and C. FEuuLEr. 2008. Gesneriaceae. In M.J. Jansen-Jacobs, ed. Flora of the Guianas ser. A, 26. 136 pages. Royal Botanic Gardens, Kew. WEBER, A. 2004. Gesneriaceae. Pp. 63-158 in K. Kubitzki and J.W. Kadereit, eds. The families and genera of vascular plants, Dicotyledons. Lamiales (except Acanthaceae including Avicenniaceae) vol. 7. Berlin: Springer. Wiener, H. 1973. Seven transfers from Episcia species in cultivation (Gesneriaceae). Phytologia 27:307-308. WienLer, H. 1978. The genera Episcia, Alsobia, Nautilocalyx, and Paradrymonia (G iaceae). Selbyana 5:1 1-60. NEW SPECIES OF BACCHARIS (ASTERACEAE: ASTEREAE) FROM RIO DE JANEIRO STATE, SOUTHEASTERN BRAZIL G. Heiden J.F.A. Baumgratz . Sc, CNPq-PROTAX Fellow Diretoria de Pesquisa Científica Escola pisa de Botánica Tropical Instituto de Pesquisas uto de Pesquisas Jardim Botánico do Rio de Janeiro Jardim as do Rio de Janeiro Rua Pacheco Leão 915 Pacheco Leáo, 915, Jardim Botánico 22460-030, Rio de Janeiro, RJ, BRAZIL Rio de Janeiro, RJ 22460-030, BRAZIL gustavo.heidenggmail.com R.L. Esteves Universidade do Estado do Rio de Janeiro Cen dico Departamento de Biologia o e ~ Rua São Francisco X Maracanã, 20550-900, Rio de A T BRAZIL ABSTRACT m i f NE. 1 pi. det . n +7 4 du A +11 IR 1 sc alti I Heiden, Baumgratz & Esteves, assigned to sect - Caulopterae DC., is characterized by th ] hes, female capitulum with 5 "x | je] Tod 2 n ] i fth urceolate or E glcally P ki ii sgeniseades CORDE * pu friburgensis, assigned to sect. Oblongifoliae DC., is cl ized by tl i l J 1 Af 1 n 1 À f, 1 f1 E > o odas dor 1] d tl } istics diffi iate it f B hylla D Additionally. keys to identify the spe- cies of sect. Caulopterae Anda sect. Oblongifoliae occurring in the state of Rio de Jairo are provided. RESUMO Duas novas espécies de Baccharis do estado do Rio de Janeiro, sudeste do Brasil, sáo descritas e ilustrada RI A ee A & Esteves, pos á sect. E E DC., é ee Bod ramos jid M [Al cer proximas do "(c I 1 R x 11.53 Dp 1 = f...21. g RLW PP | D g e A Oblongifoliae DC., é esa a pelas folhas cc coriáceas, com 2 agudo ou obtuso, gem intei lut panicula een capitulos I ápice d la denticulado, isti ] B. macrophylla Dus é fol imilar. Adicionalmente, são f id } d inação d écies d có e Pr. ocorrentes no estado do Rio de Janeiro. INTRODUCTION Baccharis was never revised taxonomically as a whole. The last worldwide compilation of accepted scientific names and synonyms of the genus was published by Malagarriga (1977). Following Müller (2006), infrage- neric classification of Baccharis is still not well resolved, and, additionally due to the high species number, revisionary work in the genus is only possible with a geographical approach. For Brazil, the last revisionary work of the genus was provided by Barroso (1976), comprising 125 species. Later, Oliveira et al. (2006) published a compilation of 146 names, without citing vouchers to assure some occurrences of species not listed by Barroso (1976) in the country. The genus occurs in all Brazilian biomes and most of its species are concentrated in the central and eastern regions of the country, growing mainly in savannas (cerrado) and grasslands (campos de altitude, campos rupestres and campos sulinos). For the state of Rio de Janeiro information concerning the genus is available only in geographically restricted florulas (Barroso 1957, 1959; Esteves & Barroso 1996) or in the taxonomic revision for Brazil (Barroso 1976). J. Bot. Res. inst. Texas 3(1): 139 —145. 2009 140 i i Texas 3( During the taxonomic revision of Baccharis for the flora of the state of Rio de Janeiro two new species belonging to the sect. Caulopterae DC. and sect. Oblongifoliae DC. were recognized. These new species are described and illustrated, and keys to identify the species of these sections occurring in Rio de Janeiro state are presented. MATERIAL AND METHODS The study has been based on literature revision and the examination of material or photos from herbaria GUA, HAS, HB, HBR, HPNI, HRJ, M, P, R, RB, RBR, RFFP, RUSU and SP. Specimens of the related taxa contrasted to the new taxa are cited just after the description and before the keys. The terminology used in the morphological descriptions is based on Radford et al. (1974) and Muller (2006). The measures were taken by the mean of a digital calliper in the wider portion of the structures. All colours mentioned are of dried mature structures. The illustrations were prepared by stereomicroscopy using a Carl Zeiss Stemi SV6. The conservation status is presented following the IUCN (2001) guidelines. TAXONOMY Baccharis alti t Heiden, B tz & Esteves, sp. nov. (Fig. 1). Tyre: BRAZIL: Rio DE Janeiro: Teresópolis, Serra dos Órgãos, 27 Nov 1933, 9 Brade 12510 E RD. +ollnid qr J s ti-natentibi s (non Ad Baccharis sect. acia DC. pa B. iade (Lam.) Pers. subsp. geni p us (non erectis) munitis j 9-4 mm ee s 4— of mm Hiep dde Pad foemineorum 2.5-3.4 mm longis f] f. (non 3.5-6.8 mm longis), Dioecious subshrubs 0.2—0.4 m tall, erect-patent or prostrate. Stems and branches 3-winged, wings 0.5—6.5 cm long, 0.3-1.6 cm wide, plane to undulate, strongly resinous, seemingly glabrous, indument tufted, tufts appearing under magnification as small resinous dots. Leaves 0.3-2 mm long, 0.2-0.5 mm wide, bract- like. Panicles 18-24 cm long, of pseudospikes 3-18 cm long, capitula solitary or 23, sparsely along the axis, the main pseudospike longer than the axillary ones. Male capitula 4—5.5 mm long, involucre 4.4-6.5 mm long, 4-6.8 mm wide, urceolate to campanulate; pyllaries in 5-8 series, abaxial surface pale yellow to dark, outer and median phyllaries ovate, inner elliptical, apex entire or denticulate, margins light yellow, denticulate, scarious; clinanthium obconical, plane, alveolate, densely covered by biseriate trichomes. Male florets 35-63; corolla 2.9-4 mm long, 5-laciniate, tube 1.8-2.5 mm long, throat 0.2-0.4 mm long, cup- shaped, lobes 1-1.4 mm long, externally with sparse biseriate trichomes on the throat and tube; styles 3.2-4 mm long., apex 1/2-bifidous; pappus 3.2-4.2 mm long, uniseriate, apex slightly broadened, terminal cell ends slightly protruding. Female capitula 6.5-8 mm long; involucre 4.4-6.5 mm long, 4-6.8 mm wide, urceolate to campanulate; phyllaries in 6—9 series, abaxial surface pale yellow to dark brown or olive green, outer ovate, median ovate to elliptical, internal elliptical, apex entire or denticulate, margins light yellow, denticulate, scarious; clinanthium obconical, alveolate, densely covered by biseriate glandular trichomes, paleae absent. Female florets 34-42; corolla 2.5-3.4 mm long, filiform, apex shortly ligulate, 3-5 teeth of unequal size; styles 3.1—5.1 mm long, branches 0.5-0.8 mm long. Cypselae 1.2-1.5 mm long, cylindrical, cal papillose, A pappus 4—4.7 mm long, uniseriate, not-accrescent. S ined. BRAZIL. Rio de J iaia, 4 Nov 1965, Eiten 6612 (RB); 12 Sep 2007, Heiden & Baumgratz 823 (RB); 17 Dec 2007, Heiden 932, 933 (RB); 13 Apr 2008, Heiden 994 (RB); 18 Oct 1977, Landrum 2103 (RB); 5 Dec 1964, Vianna 207 (GUA, RB). Nova Friburgo: 22 Mar 2008, Heiden & Bcumgratz 711 (K, RB). Teresópolis: 27 Jun 2007, Heiden 791, 792 (RB); 26 Sep 2007, Heiden 879, 880 (RB); 21 Mar 2007, Nadruz 1772 (RB); 6 Sep 1981, Ribeiro 138 (GUA); Jan 1952, Vidal 11-173, 11-175 (R); 12 Feb 1952, Vidal 11-551 (R) Feb 1952, Vidal 11-662 (R), 15-20 Dec 1952, Vidal 11-5768 (R); Feb 1953, Vidal 1478 (R). Distribution & ecology.—Baccharis altimontana is restricted to southeastern Brazil, known up to now only from the state of Rio de Janeiro. It occurs in the Atlantic Rain Forest biome, above 1900 m, in the massifs of Itatiaia (Serra da Mantiqueira), Serra dos Órgãos, Pico da Caledônia and Pico do Desengano (all of them belonging to the Serra do Man). It grows in rather small populations in rock grooves on bare rock outcrops along the high altitude grasslands, preferably in places with constant wetness. Flowering and fruiting time is between September and December. Haidan ef al A : £D L ?- 4 a + D A | 7 T 11 NS UNON C" y WA N DUE AU We acres age pein, pa! Bas- ype a , WM te SA * e penna, a é owe aa AE AA r^ ANM Qu ro P) DU ente fon d rot em» rm ead a aa a n e PM ls las: AS qu : LU A DUUM > > 2 > E An "A dae s Coe en A Sia Dire Mte fe mmm z == > E ie ira all ene Trans oqo DA AN py ets rs Fic. 1 D hon DI. m Uni Duce te 0 Estoune: A Hahi f € I lant) D. Mal pit | C. Male fl (I pl 1); D. Female capitulum; E. Phyllaries from female capitulum; F. Femal lla and style; G. Cypsela. Scale I A: 3 cm; B, D: 2 mm; C, E, F, G: 1 mm. (A: Brade 12510; B, C Heiden 879; D, E, F, G Heiden 823). Drawn by Joao Iganci. Das H ID Ll titt £T 142 Journal of ti Texas 3( Vernacular—Carqueja (Ribeiro 138; Heiden 711, 791, 792, 823, 879, 880, 932, 933, 994). Conservation.— Common in high altitude grasslands, the species is considered near threatened (NT) due to the loss of quality of the mountainous environments caused by anthropogenic pressures. Etymology.—The name ban to the habitat along the summits of mountain ranges. Notes.—Baccharis alti igned to the sect. Caulopterae DC., mainly due to presence of winged stems, epaleaceous clinanthia densely red with biseriate glandular hairs, papillose glabrous achenes, and pappus bristles of female flowers enlarged basally and Med into a ring. The species belongs to the “Baccharis genistelloides Complex” ing at least eight } lly close taxa [B. crispa Spreng., B. genistelloides (Lam.) Pers. subsp. ES B. genistelloides subsp. lorentzii Joch. Müll., B. jocheniana Heiden & Macias, B. myriocephala DC., B. opuntioides Mart. ex Baker and B. riograndensis Malag. & J. E. Vidal]. The new taxon is closer related to Baccharis genistelloides subsp. genistelloides, which occurs along the Andes from Colombia to northern Chile and Bolivia (Müller 2006). The studied specimens of B. altimontana were found in iota aa as B. ena P aio: B. myriocephala, B. puntada or B. trimera (= B. crispa). 1 of B. genistelloides subsp. genistelloides by male corollas 2.9—4 mm don female corollas 2. 5 4 mm long and female florets with uniseriate pappus (vs. male corollas 4-6 mm long, female corollas 3.5-6.8 cm long and female florets with multiseriate pappus). Specimens of B. altimontana are commonly identified as B. opuntioides, since Barroso (1976) merged both species in a broad circumscription of B. opuntioides. However, we consider B. opuntioides as endemic to the Serra do Caparaó at the boundary between the states of Espírito Santo and Minas Gerais. This species may be recognized by erect branches with short wings, 0.31.8 cm long and 0.2-0.8 cm wide, pseudospikes with apically crowded capitula (in the apical 1-2.5 cm), with the secondary branches longer than the main axis of the capitulescence, male corolla 4—5.2 mm long, female corolla 4.2—5 mm long and cypselae 0.8-1 =i mm long. Baccharis may be distinguished from B. crispa by erectopatent or prostrate stems and shortly ligulate corollas of female florets with teeth of irregular size (vs. erect stems, corolla of female florets trun- cate or denticulate), and from B. myriocephala by the height (0.2—0.4 m tall), panicles 18-24 cm long with pseudospikes 3-18 cm long, involucre of female capitulum urceolate to campanulate (vs. height 100—250 cm, panicles 25-80 cm long, pseudospikes 10-35 cm long and involucre of female capitulum cylindrical). Ted £ Selected action specimens: Baccharis cri Spreng., Syst. veg. 3:466. 1826. BRAZIL. Rio de Janeiro: Itatiaia, 13 Apr 2008, Heiden 978, 993 (RB); Mar 1894, Ule eo 177 (R); ene 12 Nov 2007, Heiden & Baumgratz 815 (RB); 13 Sep 2007, Heiden €- Baumgratz 833, 834, 850, 851 (RB). URUGUAY. Montevideo: 1821/1822, Sellow d397 GsoTYPE: R). Baccharis genistelloides (Lam.) Pers., Syn. pl. 2:425. 1807. BOLIVIA. Santa Cruz: 13 Nov 1959, Maguire & Maguire 44496 (RB) QUADOR. Cotopaxi: “province de Tacunga,” Jussieau s.n. (P-Lam, photo of holotype). Pichincha: San Juan, 28 Nov 1952, Fagerlind & Wibom 1556 (R) Baccharis myriocephala DC., Prodr. 5:426. 1836. BRAZIL. Minas Gerais: Vauthier 265 (G-DC, photo of lectotype). Rio de Janeiro: Miguel Pereira, 18 Apr 2007, Heiden & Baumgratz 727 (RB). Resende, 26 Jul 1966, Eiten 7447 (RB); 12 Nov 2007, Heiden & Baumgratz 802, 803 (RB); Rio de Janeiro, Jun 2000, Esteves & Esteves 462 (RB); Santa Maria Madalena, 14 May 2007, Heiden 734, 735 (RB); Teresópolis, 27 Jun 2007, Heiden 779, 780 (RB); 26 Oct 2007, Heiden 878 (RB). B h M ex Baker in Martius, Eichler & Urban, Fl. bras. 6(3):39. 1836. BRAZIL. Espírito S /Mi G Caparaó, 15 Naw 1960, Flaster 101 (GUA); Nov 1922, Lobo s.n. (R 37944); 8-12 Mar 1917, Lutz 1197 (R); Manse n. (M, photo of holotype); 21 Oct 1947, Moreira 44 (R); 11 Jul 1998, Oliveira 1468 (GUA); 29 Jun 1950, Santos s.n. (R 52182); 2 Jul 1888, Schwacke s.n. (R 37949 KEY TO BACCHARIS SECT. CAULOPTERAE IN RIO DE JANEIRO STATE, BRAZIL . Stems aptero 2: cm blades c coriaceous, margins enre, m. 3 d Muy basal veins; panicles 1.5-6 cm long B. organensis 2. Leaf b nately veined or with 3 acrodromous 1-5 mm distant from ES base suprabasal veins; panicles 4-16c cm ant B. regnellii inged 1. Stems w 3. Leaves well-developed. 4. Subshrubs (0.5-1 m tall); | lospike branches 5-20 cm long B. junciformis Heiden etal., N peci f Baccharis f ti tern Brazil 143 4. Shrubs (0.5-3 m tall); pseudospike branches 0.5-5 cm lon 5. Wings 1-3 mm wide; leaf blades coriaceous, with attenuate base B. burchellii 5. Wings 5-13 mm wide; leaf blades papiraceous, with cuneate or rounded base B. glaziovii 3. Leaves bract-like. DEGREES EIS ende corolla aN or metan Ada B. crispa E , Scande | | ligulate. 7. Subshrub 02- 04 m tall; S 3-18 cm m een solitary or 2-3; involucre of female capitulum urceolate to campanulate B. altimontana 7. Subshrub 1-2.5 m tall; pseudospikes 10-35 cm long; capitula solitary or gl les with 2-8 zs involucre of female capitulum cylindrical B. myriocephala Baccharis friburgensis Heiden, Baumgratz & Esteves, sp. nov. (Fig. 2). Tre: BRAZIL: ih DE JANEIRO: Nova Friburgo, Macaé de Cima, 16 Aug 1989, 3, Vieira 48 (HoLorYPE: RBI; isotypes: FCAB!, GUA!, NY, RI, RB!, RBR!, SP Ad Baccharis sect. Oblongifoliae DC. pertinens, B. dll ndi Papi Md coriaceo, dbi ai es in d cuin contracta ), e similis, sed capitulis foemineis 2 ad 3 floribus (non 11 ad 21 ; et capitulo masculo 8 ad 12 floribus (non 30 ad 45 floribus) munito differt. Dioecious shrubs 1.5-2 m tall. Stems erect, branching dichotomous, shoots glabrescent, with uniseriate and pedestal trichomes. Leaves spirally alternate, crowded at the apex of the branches, with petioles 0.2-0.8 cm long; blades 30-68 mm long, 7-21 mm wide, coriaceous, narrowly elliptic or oblanceolate to obovate, base attenuate, apex acute to obtuse, margins entire, revolute; pinnately veined; both surfaces seemingly glabrous, abaxial surface with tufted indument appearing as resinous dots, scarce biseriate, uniseriate and pedestal trichomes among the tufts. Panicles 1.9-4.4 cm long, 1.8-5.5 cm wide, corymbose, terminal; peduncles 1-1.8 cm long. Male capitula 2.9-4.8 mm long, involucre 2.9-4.8 mm long, cup-shaped, 2.75 mm wide; phyllaries in 3-5 series, brown, outer and median phyllaries ovate, inner linear-ovate, apex fimbri- ate, margins light-brown, shortly dentate; clinanthium convex, glabrous; paleae absent. Male florets 8-12, apex 5-laciniate; corolla 2.3-2.5 mm long., tube 0.9-1.1 mm long, externally densely covered by biseriate trichomes near the apex, throat 0.5-0.6 mm long, cup-shaped, lobes 0.7-0.8 mm long, apex not revolute; anthers included; styles 2.4—2.6 mm long, apex capitate by sweeping hairs of unequal size, slightly exserted; sterile ovary 0.5-0.7 mm long; pappus 2-2.3 mm long, uniseriate, apex broadened, cell ends erectopat- ent, shortly protruding. Female capitula 5.3-5.8 mm long, involucre 4.3-5 mm long, 2.2-3.1 mm wide, cylindrical; phyllaries in 4—6 series, brown, outer phyllaries ovate, median linear-ovate, inner linear, apex long-fimbriate, margins light-brown, shortly dentate; clinanthium convex, paleae linear, acute, deciduous. Female florets 2-3, filiform; corolla 2.9-3.2 mm long, apex 5-denticulate, distal half with subapical scat- tered biseriate trichomes; styles 3-3.5 mm long., branches 0.3-0.6 mm long. Cypselae 2-2.6 mm long, 0.4-0.6 mm wide, light brown, cylindric, narrowed at both extremities, with 5-6 longitudinal ribs, with scattered biseriate trichomes near the apex; pappus 2.5-3 mm long, slightly shorter to slightly longer than the style, biseriate, persistent, not accrescent at maturity. Additional specimens examined. BRAZIL. Rio de Janeiro: Nova Friburgo, 9, 24 Aug 1986, Leitman 196 (K n.v.; RB-2x); d, 19 Aug 1990, Moraes 414 (RB-2x). Distribution & ecology.—Restricted to southeast Brazil (state of Rio de Janeiro). It occurs in the Atlantic Rain Forest biome, from 1000 to 1500 m above sea level, and is endemic to the summits of the Macaé de Cima Mountains, a part of the Organ Mountains/Serra do Mar Ranges. It grows directly on rock outcrops, along river headwaters or in edges of nebular forest. Flowering and fruiting in August Conservation.—Taking into account the few collections available, the restricted area of occurrence and the presence of anthropogenic pressure in surrounding areas, represented by cattle grazing, agriculture, tourism and urban expansion, the species is considered endangered (EN B2abliiil). Etymology.—The name refers to the type locality in the municipality of Nova Friburgo. Notes. —Baccharis friburgensis is assigned to sect. Oblongifoliae DC. for the pinnately veined leaves, capitate style apex in male flowers due to sweeping hairs of unequal size, paleate female capitula and female flowers with the pappus not elongated at cypsela maturity. The species is remarkable for the few-flowered Fc? R n 25 £pl A] D £5 0 Cop AD le aff, I lene. D AL allan £. f" Mal sel m Mala £l dct AE E I nul t papi Phyllari itul G. Palea: H. Fi ji d style: 1. Cypsela. Scale bar near Ld * £1 A: 2 cm; B, D, F, G, H, I: 1 mm; C, E: 2 mm (A, B, E, F, G, H, I: Leitman 196; C, D: Vieira 48). Drawn by Joao Iganci. Heiden et al., N pecies of Baccharis fi theastern Brazil 145 male (8-12) and female 2-3) capitula, which results in seemingly epaleaceous female clinanthia, because all paleae are, due to the low flower number, more distal then the distalmost flowers and hidden by the innermost phyllary series. Some further distinguishable characters are the coriaceous leaves with acute to obtuse apex, entire and revolute margins, and the corymbose panicles. Baccharis friburgensis is vegeta- tive morphologically similar to the allopatric B. macrophylla Dusén, although this similarity seems to be a convergence to the habitat in sunny habitats on summits of mountains. The later one has leaf margin more commonly with 1—5 subapical teeth, seldom entire, male capitula with 30—45 flowers, female capitulum with 11-21 flowers, and female corollas ligulate. d ane o la Dusén, Arq. Mus. Nac. Rio de Janeiro 13:14. 1905. BRAZIL. Rio de Janeiro: Itatiaia, Dusén 4844 (HOLOTYPE: R); 23 Jul 1960, Handro 941 (SP); 17 Dec 2007, Heiden 930 (RB); 16 Aug 1969, Sucre 5775 (CEPEC); s.d., Tamandaré & Brade 6562 (SP); Resende, 12 Sep 2007, Heiden & Baumgrat. 7, 828 (RB); Jul 1902, Moreira & Teixeira s.n. (R 38027 KEY TO BACCHARIS SECT. OBLONGIFOLIAE IN RIO DE JANEIRO STATE, BRAZIL 1. Panicles pyramidate. 2. Foliar indument ferruginous B. rufidula 2. Foliar indument not ferruginous B. oblongifolia 1. Panicles corymbose. 3. Leaf blade papyraceous, margins not revolute B. grandimucronata 3. Leaf blade mE margins revolute 4. Male capitula 30-45 florets; female capitula 11-21 florets, female corollas with ligulate apex —— B. macrophylla 4. Male capitula 8- E 20 florets, female capitula 3- d florets, female corollas with denticulate apex. B. friburgensis ACKNOWLEDGMENTS The authors acknowledge the consulted herbaria, André Scarambone Zaú, Cecilia Cronemberger de Faria, Denise Pinheiro da Costa, Izar Aximoff, Marcus Nadruz and Marina Wolowski Torres for providing field support, Joào Iganci for the illustrations, Jochen Müller for the valuable comments and an anonymous reviewer. REFERENCES Barroso, G.M. 1957. Flora do Itatiaia—Compositae. Rodriguésia 32:175-241. Barroso, G.M. 1959. Flora da cidade do Rio de Janeiro—Compositae. Rodriguésia 33-34:69-155. Barroso, G.M. 1976. Compositae—Subtribo Baccharidinae Hoffman. Estudo das espécies ocorrentes no Brasil. Rodriguésia 28(40):3-273. Esteves, R.L. AND G.M. Barroso, 1996. Compositae. In: Lima, M.PM. & Gutpes-BRuNI, R.R. Reserva Ecológica de Macaé de Cima: Nova Friburgo—RJ: aspectos florísticos das espécies vasculares. Rio de Janeiro: Jardim Botánico do Rio de Janeiro, 1996, 2:189-245 INTERNATIONAL UNION FOR CONSERVATION OF NATURE AND NATURAL Resources (IUCN). 2001. IUCN Red List Categories and Criteria, Version 3.1. IUCN, Gland, Switzerland and Cambridge, United Kingdom. http://www.iucn.org. MALAGARRIGA Heras, R.P. 1977. Nomenclator baccharidinarum omnium. Mem. Soc. Ci. Nat. La Salle 37:129-224. MÜLLER J. 2006. Systematics of Baccharis (Compositae-Astereae) in Bolivia, including an overview of the genus. Syst. Bot. Monogr. 76:1—341. Ouv&RRA, A.S., L.P. Desle, A.A. SCHNEIDER, AND J.N.C. Mancuion. 2006. Checklist do género Baccharis L. para o Brasil (Asteraceae-Astereae). Balduinia 9:17-27. Raproro, A.E., W.C. Dickison, J.R. Massey, AND CR. BeLL. 1974. Vascular plant systematics. Harper & Row, New York. 146 Journal of the Botanical R h Institute of T BOOK REVIEW Bruce M. Pavuk. 2008 The California Deserts: An Ecological Rediscovery. (ISBN 978-0-520-25145-8, pbk.). University of California Press, 2120 Berkeley Way, Berkeley, California 94704-1012, U.S.A. (Orders: The University of California Press c/o California/Princeton Fulfillment Services, 1445 Lower Ferry Road, Ewing, New Jersey 08618, U.S.A.; www.ucpress.edu, email orders@cpfsinc.com, 1-800- 777-4726, 1-800-999-1958 fax). $27.50, 365 pp, numerous figures, graphs, images, maps, tables, references, index, and art credits, 7" x 10". The California Deserts: An Ecological Rediscovery, is divided i i 1 ifically d ibe tl p y gins, discovery, climate, geology, resources, fauna 1 fl he 25 milli hat f he tł d ts of California: the S Mojave, and Great Basin ac nd Bruce Pavlik thread ies of various early explorati hrougl he fi ion of this book, provid- : 1 1 ST Ta fi £, TT s J 1 Lt ra D. y Ry pe | ca ing the rea Į y : 1 NT v 1 na 1 A ECT 1; E el 1 len f, r 1 mo day oe and how most o are ca between 1 10° and a latitude ae and south of the Go afar M. three dm are in fact very ,andth : ge, precipitation, temperature that are present within Pavlik ao the “remarkable id that can js a Ww unn the California desert region. A pri ple is the Tat ^ f. m Ll TT "m y its abilit y to procure wat book by listing tl 1 (e.g. fragmentation, i i ies) of thi i 1 id ideas f ing tl DOOK DY B 5 5 ? L e F o d deserts 12] 1 3 o, ; 1 acta] f. list, educator, I ] Pi Pi r da. t th J f +1 + California Ib J; th 1 1 e 1 1 1; including A Vaot number f pl f the flora and fauna that are found w a ee regions. Ted Tees that were bus throughout ce PUDE ld the la reader WIA a California deserts. en McNew, MS Biol) Prog Manager, Botanical R h Institute of Texas, 500 E 4th Street, Fort Worth, TX 76102, U.S.A J. Bot. Res. Inst. Texas 3(1): 146. 2009 DIPLYCOSIA INDICA (ERICACEAE): A NEW SPECIES AND A NEW GENERIC RECORD FOR INDIA MR. Debta' and HJ. Chowdhery Botanical Survey of India (Northern Circle) 192, Kaulagarh Road, Dehradun- 248 195, INDIA ABSTRACT Diplycosia Bl (Eri )—a new generi land Dipl ia indica M.R. Debta & HJ. Chowdhery—a new species from India is described and illustrated. Key Worbs: Diplycosia, India, new species, new generic record RESUMEN + Dy] Gyi A 4 e 4 +7 E Tin? iaindica MR Debta @HJ Chowdhery—una nueva o E especie de la India. INIRODUCIION The genus Diplycosia consisting of about 99 species (Mabberley 1997), is widely distributed in the Ma- laysian region. Sleumer (1967) revised the genus for Flora Malesiana treating 97 species. He reported that the genus has its highest diversity in Borneo, while Mt. Kinabalu has the highest concentration of species. Powell and Kron (2001), based on molecular studies, have suggested treating Diplycosia under Gaultheria L. Argent 2002) —Uwhile studying Diplycosia from Borneo and peninsular Malaysia— described seven new species, one new form, and made two new combinations. Diplycosia can be distinguished from Gaultheria with the following key. KEY TO THE GENERA DIPLYCOSIA AND GAULTHERIA 1. Flowers in fascicles; anther cells unawned, not aristate, with short or long apical tubules at apex; tubules connate or free Diplycosia 1. Flowers in racemes; anther cells awned or aristate, not having apical tubules at apex Gaultheria punc a20 Ms plant couecuing = to Singalila National Park e Bengal), in the eastern Himalayan , some i 1 unusual plant speci ing to the family Ericaceae were collected. "Initially ty were iden nea as species of Gaultheria, but dene. ] that some of the specimens belonged to the closely allied genus Diplycosia, which is hitherto unknown from India. Further critical examinations of the specimens have shown that though one specimen had some resemblance to a Malaysian species, Diplycosia aperta JJ. Sm., but differs widely from it in habit, leaf venations, number of flowers, bracteoles, and calyx character. The new collection is therefore described as a new species, Diplyco- sia indica, which is also a new generic record for India. Table 1 shows the differences with its allied species Diplycosia aperta. An illustration of the new species is provided for easy identification. 1; 1 Diplycosia indica M.R. Debta & H.J. Chowdhery, sp. nov. (Fig. 1). Treus: INDIA. West Bencar: Darjeeling district, Singalila National Park, from Kainyakata to Kalapokhri, ca. 2950 m, 02 Jun 2006, M.R. Debta 40813 (HoLorrrus: CAL; isorvrus: BSD) Lan nova b e eae Sm. arte affinis, sed habitu i; infl ii oe floribus multum brevioribus, albis; : 1 [e] E Terrestrial, prostrate shrubs, up to 1.5 m high. Stems terete, us to densely brown-hirsute or setose, reddish. Leaves subcoriaceous, 13-36 mm long x 7-18 mm wide, elliptic to ovate-elliptic, mucronate or apiculate, ‘Corresponding author's e-mail: manas_debta@rediffmail.com J. Bot. Res. Inst. Texas 3(1): 147 —150. 2009 148 £ sl Das Journal of 0.5 mm = Soc EI YA os 327: A W, LJ X 2 Y ORE ASE SU NV Fis. 1. Diplycosia indica. A. Habit. B. Leaf, upper surface. C. leaf, lower surface. D. Flower. E. Bract. F. Bracteole. G. Calyx, dorsal surface. H. Calyx, ventral surface. |. Corolla split open. J. Stamen. K. Pistil (Scale bars: A = 3 cm; B-C = 1 cm; D - land K = 1 mm; J = 0.5 mm). Taste 1. Distinguishing d f Diplycosia indi | D. aperta Diplycosia indica Diplycosia aperta Habit Terrestrial, prostrate shrubs Epiphytic, sometimes scandent shrubs Leaves Margin with short cilia; lateral veins in Margin with long cilia; lateral veins in 2 pairs, 3-4 pairs, distinct beneath obscure beneat Flowers (3-)4-5; pedicel 0.5 mm, glabrous 1 or 2, rarely 3; pedicel 0.8-1.1cm, with bristles Bracteoles Acute to shortly acuminate, 3-3.5 mm long, Obtuse, ca. 1.5 mm long, glabrous throughout very sparsely puberulous in the apical the dorsal surface half along midrib on dorsal surface Calyx Minutely puberulous on both surfaces Glabrous to some fine glandular warts on in apical portion dorsal surface Corolla ca. 4 mm long, white, lobes 0.75 mm 8-9 mm long, red or rose, lobes 1.5 mm rounded to broadly cuneate at base, margin serrulate-ciliate with short cilia, cilia 0.5-1 mm long, laxly softly pilose at the base along mid-vein, dark green above, setulose, light green beneath; veins impressed above, distinct beneath, lateral veins in 3-4 pairs; petiole 1.5-3 mm long, setulose, red. Inflorescence a fascicle, up to 7 mm long, (32)4—6-flowered, glabrous; bracts 1.5-2 mm long x 1.8-2.2 mm wide, basal, ovate, acute, glabrous, margin membranous, ciliate, reddish-green; bracteoles 2, 33.5 mm long x 1.3-1.8 mm wide, ovate, acute to shortly acuminate, opposite, concave, margin membranous, ciliate, very sparsely puberulous in the apical half along the middle on dorsal surface. Flowers 4—5 mm long, urceolate; pedicel ca. 0.5 mm long, inconspicuous, glabrous. Sepals 5-6, 1.5-3 mm long x 1-1.5 mm wide, ovate-triangular, acuminate, minutely puberulous in the apical portion on both the surfaces, margin membranous, thickly puberulous, adnate to the ovary, rosy-pink. Corolla ca. 4 mm long, urceolate, white, 6-lobed; lobes ca. 0.7 mm long, ovate, apex obtuse to rounded, margin obscurely crenulate, glabrous. Stamens 10, ca. 1.5 mm long, loosely epipetalous, caducous, filament ca. 0.8 mm long, oblong, dilated at middle, papillose; anther with tubules 0.6 mm long, very small, 2-lobed, oblong. Pistil 2.5-3.5 mm long; ovary superior, 1 mm long x 1.5 mm wide, subglobose, subglabrous or scattered puberulous, light green; style ca. 2 mm long, slender, sparsely minutely hairy in upper half, disk cupular, 8-10 lobed, pressed against the ovary, papillose. Fruit not seen. Diplycosia indica is closely allied to Diplycosia aperta J.J.Sm. but can be distinguished from it by its ter- restrial habit; inflorescence with 4—6, glabrous pedicelled, much smaller, white flowers; acute to acuminate, EE pd mob and poto poc ] , in association with Viburnum erubescens, Thamnocalamus M etc. and i at an altitude of about 3000 m. Flowering.—June. Distribution —India: Eastern Himalaya (West Bengal). Endemic. Etymology.—Diplycosia indica is named after India, the country of its origin. Conservation status.—The new species could not be traced from anywhere inside the Singalila Na- tional Park except for a small population. Since Singalila National Park—especially from Manebhanjeng to Sandakphu—is a famous trekking route, it attracts a large number of tourists and trekkers throughout the year. The increasing pressure from the tourism industry and large scale cattle grazing in and around park areas, are posing severe threats to the natural habitat of numerous species aaa paa indica. Additional specimen examined: INDIA. West Bengal: Darjeeling district, 32 km N of Manebhanjeng, from K Kalapokhri ca. 2950 m, 02 Jun 2006, M.R. Debta 40813 (CAL, BSD). ACKNOWLEDGMENTS The authors are indebted to the director, Botanical Survey of India for providing facilities. Grateful thanks are due to G.C.G. Argent and DJ. Middleton, Royal Botanic Garden, Edinburgh for their expert suggestions 3 af ob Dat H ID hi bitit. £f T, 150 and helps, to VJ. Nair, ex-joint director, Botanical Survey of India for providing Latin diagnosis of the new species. Thanks are also acknowledged to the officials of the Forest Department, Government of West Bengal for permission to carry out research work inside the Singalila National Park and to Brijesh Kumar, Botanical survey of India, Dehradun for illustrating the new species. We greatly appreciate the helpful review of one anonymous reviewer. REFERENCES Arcent, G.C.G. 2002. New taxa and new combinations in the genus Diplycosia (Ericaceae) of Borneo and Peninsular Malaysia. Gard. Bull. Singapore 54:217-238. Mapsertey, DJ. 1997. The plant book: a portable dictionary of the vascular plants. Cambridge University Press. Pp. 234 PowtLL, E.A. and K.A. Kron. 2001. An analysis of the phy ic relationships in the winterg group (Diplycosia, Gaultheria, Pernettya, Tepuia; Ericaceae). Syst. Bot. 26: :808- 817. SiEuMER H. 1967. Diplycosia Blume. In: van Steenis, C.G.G.S, ed. Fl. Males. 6, ser. 1:696—740. AMORPHA CONFUSA, A NEW NAME FOR AN OLD AMORPHA (FABACEAE: AMORPHEAE) Shannon C.K. Straub Bruce A. Sorrie Cornell University North Carolina ped Heritage Program LH. Bailey Hortorium Office of Natural Resources and Conservation Planning Department of Plant Biology 3076 o Pos thage Roa 412 Mann Library Whispering Pines, North Carolina 28327, U.S.A. Ithaca, New York 14853, U.S.A. ss463@cornell.edu Alan S. Weakley University of North Carolina at Chapel Hill ios of North Carolina Herbarium i U) Nor rolina Botanica den rudi m North Carolina 275 s Ud ABSTRACT bination, Amorph fusa (Fab : Amorpheae), based on A. georgiana var. confusa, is established. The variety is raised in sank due to differences in morphology, distribution, habitat preference, phenology, ploidy, and genetic variation when compared to A. georgiana. RESUMEN Se ao una A nueva, Amorpha cantas a Amorpheae), parana en A. PCM var. confusa. La variedad se , distribución f t, fenología, ploidía, y variación E genética en toiipaba ción. con Á. georgiana. For many years, plants currently known as Amorpha d aid UT var. confusa Wilbur (Fabaceae Juss.: Amorpheae Boriss.) have been a source of tural confusion. Wilbur (1964) elegantly explored the intricacies of the situation in his revision of the dwarf species of Amorpha L. Various names often had been applied incorrectly to this species (e.g., A. caroliniana H. B. Croom, A. cyanostachya M.A. Curtis) or were unavailable (e.g., A. glabra Desf. ex Beadle, nom. illeg.). This left Wilbur no option but recognize it with a new name, which he did at the rank of variety due to its similarity to A. georgiana. At the time, this was the most conservative course of action due to the paucity of collections of both var. georgiana and var. confusa, even though Wilbur recognized that the two differed morphologically. Later, Wilbur (1975 p. 367) commented in his monograph of the genus that he remained "skeptical" that his treatment would prove "satisfactory when more is learned about them" and acknowledged that "future investigation may well dem- onstrate that the two taxa are specifically distinct." Most recent floristic treatments and species checklists (e.g., Isely 1990, 1998; Kartesz 1999) have followed Wilbur's treatments in recognizing two varieties. These varieties differ morphologically in many characters with var. confusa having larger leaflets [((10-)15-25(35) mm long and (7-)9-15(-18) mm wide versus (36-10(-15) mm long and (2—)3-5(-8) mm wide], longer petioles [(6-)8-15(-20) mm versus 1-3(-5) mm] and racemes [10-20(30) cm versus (2-)3-5(6) cm], less numerous leaflets, clustered, panicle-like inflorescences rather than generally solitary racemes, and bright blue rather than reddish-violet vexilla (Fig. 1; Wilbur 1964; Sorrie 1995; Weakley 1995). Recent investigations into the current distribution, conservation status, and genetics of the two varieties have revealed additional differences and supported Wilbur's (1975) notion that they should each be recog- nized at the specific level. The distributions of the varieties do not overlap currently, nor did they histori- cally based on herbarium records (Fig. 2). Amorpha georgiana var. confusa is an endemic of a few counties in J. Bot. Res. Inst. Texas 3(1): 151 —155. 2009 Journal of the Botanical Research Institute of Texas 3(1) Fis. 1. a. Amorpha confusa at Green Swamp Preserve, North Carolina. (Photograph by Andrew Walker). b. Amorpha georgiana at Fort Bragg, North Carolina (Photograph by Bruce Sorrie). Straub et al., A new name for an old Amorpha 153 Amorpha confusa BÉ Current Distribution [2] Historical Distribution Reported Distribution a Amorpha georgiana eS lI Current Distribution [5] Historical Distribution [s] Reported Distribution Atlantic Ocean Fic. 2. Current, historical, extreme southeastern North Carolina and immediately adjacent South Carolina, though it is now believed to be extant in only Brunswick and Columbus counties of North Carolina (Weakley 1995). Amorpha geor- giana var. georgiana is found in the middle and inner Coastal Plain of North Carolina, South Carolina and Georgia (Sorrie 1995). Both varieties are associated with the longleaf pine savanna ecosystem but differ in the details of their habitat preferences with var. confusa occurring in flat, moist to rather dry outer Coastal Plain savannas with loamy soils, especially of the Foreston series (Weakley 1995); whereas var. georgiana occupies more moist to occasionally inundated areas, chiefly sandy river terraces and river banks above blackwater rivers traversing the sandhills of the middle and inner Coastal Plain, and more infrequently the edges of swampy floodplains (Sorrie 1995). Both varieties have suffered habitat loss, fragmentation, and degradation in recent years due to human activities, including fire suppression, agriculture, and land devel- opment, causing them to be of conservation concern (Sorrie 1995; Weakley 1995). In North Carolina, var. confusa is currently considered to be threatened (North Carolina Department of Agriculture & Consumer Services 2008) because its population numbers are estimated to be less than 14,000 individuals, and only those populations occurring in The Nature Conservancy's Green Swamp Preserve are likely to receive long- term conservation-oriented management (Weakley 1995). The two varieties also differ in phenology, with Amorpha georgiana var. confusa flowering from late May to mid-July and var. georgiana flowering from late April to late May (Sorrie 1995). Additionally, recent genetic work has indicated that the genome of var. confusa is likely tetraploid, while that of var. georgiana is diploid (Straub et al. 2009). A comparison of the microsatellite variation observed for the varieties at the population level indicates that they are quite well differentiated genetically and preliminary phylogenetic analyses of chloroplast spacer region and low-copy nuclear gene DNA sequence data indicate that they are likely not each other's closest relative among Amorpha species (S. Straub & J. Doyle, unpublished data). Since the time of Wilbur's (1964) original publications, additional studies of the morphology, distribu- tions, habitat preferences, phenology, and genetics of the two varieties have shown the extent to which they are distinct from one another. These differences warrant the neon an var. confusa at the rank of species (Sorrie 1995; Weakley 1995). Recognition at this rank further empl tion importance and need of both of these imperiled species. 154 t tani i Texas 3( Amorpha confusa (Wilbur) S.C.K. Straub, Sorrie & Weakley, comb. et stat. nov. Amorpha georgiana var. confusa Wilbur, J. Elisha Mitchell Sci. Soc. 80: 58. 1964. Tyre: UNITED STATES. NORTH CAROLINA. Brunswick Co.: Savannah, 7 mi SW of Wilmington, 31 May 1938, R.K. Godfrey & IV. Shunk 4122 (Lectory, designated by Wilbur 1975: GH, digital image!; DUPLICATE OF THE LECTOTYPE: US, digital image!). Amorpha glabra Beadle, Bot. Gaz. 25:279. 1898; EE. Boynton in Smalls Fl. S.E. U.S. 626. 1903, non Poir., Encycl. (Lamarck) Suppl. 1:330. 1810 Amorpha caroliniana sensu Torr. & A. Gray, Fl. N. ea 1:305. 1838 in part; C.K. Schneider, Ill. Handb. Laubholzk. 2:74. Mar 1907 & Bot. Gaz. 43:302. Jun 1907; Rydberg, Fl. N. Amer. 24:29. 1919, non H.B. Croom, Amer. J. Sci. Arts 25:74. 1834 Amorpha cyanostachya sensu E.J. Palmer, J. Arnold pum 12:169. 1931; Small, Man. S.E. Fl. 639. 1933, non M.A. Curtis, Boston J. Nat. Hist. 1:140. 1835 Wilbur (1964) did not choose E Pen two acl listed in the protologue as the holotype, although he later indicated that he considered P GH to be the holotype (Wilbur 1975: 366), here corrected to lectotype pursuant to Article 9.8. of the ICBN ene et al. 2006). Additional collections. NORTH CAROLINA. Bladen C Bil Herbarium 5767B (NY). B k Co.: pineland near Wilmington, 28 Aug 1938, R.K. Godfrey 6233 (GH, US, NCSC- 4939); gs woods along route 17 about 3.5 mi NE of Bolivia, 22 Jun 1947, C.E. Wood, Jr. an D. elle ih (GH, NCU-175175); dry sandy pineland, 20 mi S of Wilmington on route 17, 27 Jun 1950, F.H. Sargent 10008 (GH); 1 Shall d Ash, 6 Jul 1951, R.K. Godfrey & H.L. Blomquist 51177 (FLAS, NCSC); dry, sandy ie: savanna, 2.4 mi W of Shallotte, Highway 130, 10 Jul 1951, H.L. P ds R.K. a ds il R. E Wilbur15225 (PURE, ee in h about 1 mi W of Shallotte on NC 130; 6 Jun 1957, R.L. Wilbur 5991 ( M : f] 4 1 4 ad e bod 6 mi NW of Southport on Route 87, 12 Jun 1957; R.L. Wilbur 6169 (DUKE-14 41699) p route 87-133, 10 Jul 1963, R.L. mee 6954 (DUKE, FSU); open piney woods about 11 mi NW of Supply on Route 211, 18 Aug 1967, R.L. Wilbur 9466 (DUKE-186242); powerline right-of-way, s E state roads 1518 and 1521, N of Funston,14 Jun 1979, DJ. Sieren 1835 i of borrow pit on east side of NC 211, 9.4 km N of US 17, 18 Jul 1986, J.B. Taggart 224 (NCU-557630); Green Swart Preserve, 5.5 mi N of Supply, 22 May 2003, A.S. Weakley & G.T. Chandler 7242 (NCU- 569103, NCU-569104, NCU-569106); Green pie Preserve, W of Big Island Savanna, 22 in 2003, A.S. Weakley & G.T. ene 7244 6-11 Lad Z a Ç Ut Ut = = CO FI z E < a as N eo LY bord be (NCU-569107, NCU-569109, NCU-569111) g Lake Preserve, C d, 11 Aug 2006, J.C. Morris 0 (NCU-584945, WILM). Columbus Co.: ing DE aki 25 TAE dd PO. “Shallert s.n. aye ileal dan near Route 76, 1 mi SE of Delco, 31 Jul 1949, W.B. Fox & R.K. pa o (NCSC-3 : 3.7 mi NW of Old Dock on NC 130, 2 Jul 1968, S.W. Leonard & K. Moore 1720 (FLAS, FSU, GH, NCU sand il along CR 1928 to Shulkins, 7 Jul 1989, J.A. Churchill 89-685 (VDB). New Hanover Co.: Wilmington, 1 Jul 1904, bim DRM 1391-L (NCU): Wilmington, 6 Oct 1908, E.A. Bartram s.n. (PH); Wilmington, 11 Jun 1917, T.G. Harbison 3415 (NCU-3469); low swampy ground, Wilmington, 11 Jun 1917, T.G. Harbison 16 (A). SOUTH CAROLINA. Horry Co Sea, 16 Jun 1941, H.R. Totten s.n. (NCU-22550, NCU-577773). ACKNOWLEDGMENTS The authors thank the curators of A, DUKE, FLAS, FSU, GH, HUH, NCSC, NCU, US, VDB and WILM for access to specimens and digital images, the North Carolina Natural Heritage Program for access to locality data, James L. Reveal (BH) for helpful discussions related to nomenclature issues, oe Tu = Reveal and Guy Nesom for valuable comments which improved the manuscript. Th li the the U.S. Fish and Wildlife Service for status surveys, a Garden Club of America ein H. Beattie Fellowship to S.C.K.S., and National Science Foundation grant DEB-0709960. g from REFERENCES IseLy, D. 1990. Amorpha. In: Vascular flora of the southeastern United States, vol. 3, part 2: Legumi (Fabaceae). University of North Carolina Press, Chapel Hill. Pp. 71-76. IseLy, D. 1998. Amorpha. In: Native and naturalized Leguminosae (Fabaceae) of the United States (exclusive of Alaska and Hawaii). Brigham Young University, Provo, UT. Pp. 132-144. KARTESZ, J.T. 1999. A synonymized checklist and atlas with biological attributes for the vascular flora of the United States, Canada, and Greenland. First Edition. In: Kartesz, J.T. & C.A. Meacham. Synthesis of the North American Flora, Version 1.0. North Carolina Botanical Garden, Chapel Hill, NC. McNELL, J., ER. Barrie, H.M. Bunper, V. Demouun, D.L. HAwKsworTH, K. MARHOLD, D.H. NicoLson, J. PRADO, PC. Sit va, J.E. SKOG, N.J. TURLAND, AND J. Wiersema, eds. 2006. The international code of botanical nomenclature (Vienna Code), July 2005. Regnum Veg. 146:1—568. Straub et al., A new name for an old Amorpha 155 NORTH CAROLINA DEPARTMENT OF AGRICULTURE & CONSUMER Services. 2008. Plant Industry Division — Plant Protection Section - Plant Conservation Program: Protected plant list [online data]. NCDA&CS, Raleigh, NC. Available at http:// www.ncagr.com/plantindustry/ plant/plantconserve/plist.htm, accessed 3 January, 2008. SorRIE, B.A. 1995. Status survey of Amorpha georgiana var. georgiana. Submitted to the US Fish and Wildlife Service, Office of Endangered Species, Asheville, NC, and to the North Carolina Natural Heritage Program, Division of Parks and Recreation, Raleigh, NC. STRAUB, S.C. K,, S.M. BOGDANOWICZ, AND J.J. Dove. 2009. Characterization of twel | hic mic lli | for Georgia false indigo (Amorpha georgiana Wilbur var. georgiana), an endangered species, and their utility in other dwarf Amorpha L. species. Molec. Ecol. Resour. 9:225-228. Weak.ey, A. 1995. Status survey for the savanna Indigo-bush, Amorpha georgiana Wilbur var. confusa Wilbur. Submitted to the US Fish and Wildlife Service, Office of Endangered Species, Asheville, NC, and to the North Carolina Natural Heritage Program, Division of Parks and Recreation, Raleigh, NC. WiLgur, R.L. 1964. A revision of the dwarf species of Amorpha (Leguminosae). J. Elisha Mitchell Sci. Soc. 80: 51-65. WiLBUR, R.L. 1975. A revision of the North American genus Amorpha (Leguminosae-Psoraleae). Rhodora 77:337-409, 156 [] Ingel D saal D L BOOK REVIEW MICHAEL WINK AND BEN-Enik VAN Wyk. 2008. Mind-Altering and Poisonous Plants of the World. (ISBN-13 978-0-88192-952-2, hbk.). Timber m Inc., 133 S. bid eae Avenue, Suite 450, Portland, Oregon 97204-3527, U.S.A. (Orders: j com, com, 800-327-5680, 503-227- 3070 fax). $49.95, 464 pp., ose throughout black/white a ans 6 5/8" x 9 1/2". Mind-Altering and P Pl of the World i k. Its subtitle—A scientifically te guide to 1200 Loss cinia ing plants i fect d inti frbied lightf ] A ful book. I ; 2; NE id this book to be. a valuable aud liabl finf ti on poisonous Slate Not only is this Book for the botanist, but a variety ranchers, teachers, students, homeowners, gardeners, horticulturists, florists, botanists, chemists, medical and ing enforcement personnel, health care professionals, parents, and anyone with a curiosity in the power of green plants of audiences would find ful E It has been said, “Poisons tl he] J 251 I p 1] FA y : g lif y ly ly A <4] iA have ever had a peculiar fascination for mankind." Poi 1 poi pl inly fasci 1 tl thors, Wink and van Wyk; they have done a spectacular job in assimilating a vast t of inf tion int tical and user friendly format. Beautiful but J o F pl f d as gard tals, potted plants, florist fl yet most people are unaware of the de iuis ] ins derived f I Mind-Alteri 1 Poi Plants of the World tal P modern day uses of these toxic plants. The H is followed ar the a of the bake an alphabetical list of about 200 plants. In the authors’ words, “About 200 t d by concise monographs. Essential facts about another 1000 e are provided in the ‘Quick guide to poisonous and mind-altering plants’ at the end of the book.” Eacl l | E 12 sections: 1) Name, 2) plants with similar properties, 3) plant characters, 4) occurrence, 5) cüsefioqion: 6) active elena Li utilization, d onu. 9) upon, 10) M effects, 11) first aid, and 12) systematics. Beautiful color photograpl ] gray ake note, accidental eal happen and therefore I was O to see a list of suggesti f d to avoid pete. or to help p D lin case of ingestion of a toxic plant. The T S ill S ata for 2006—from the rican A tion of Poison Control Centers’ National Poison Data acude MALE accounted for 2.7% of e as most frequently (64,23 ) invol Sadly, deaths do occur. I look 1 wide Poison Help line number (1-800-222- 1222). This toll-free number is available 24 eure a day and will connect you to the nearest poison center poda dn are icum s from in the United States. In an ein call 9-1-1. Increase your knowledge of toxic plants, I I g, and discover I Į gI , plant identification, * 7 Dl RI 1 a | T WM 1l i 1 f, 1] MA 1 1 1 with this book y yingI I vbej Dp g or even save a life. I f j l lical professional lified Mind-Alteri ] Poi Pl f 1 i e ES the World has a Warning dl Disclaimer on the copyright pane but I think it is still li $49.05.— Barney Lipscomb, 8 rZ. : Dar Hey Lip Botanical Research Institute of Texas, Fort Worth, Texas 76102-4025, U.S.A. J. Bot. Res. Inst. Texas 3(1): 156. 2009 A NEW COMBINATION IN NORTH AMERICAN TEPHROSIA (FABACEAE) Guy L. Nesom James L. Zarucchi 2925 Hartwood Drive Missouri Botanica! Garden Fort Worth, Texas 76109, U.S.A guynesom@sbcglobal.net O. Box 299 St. Louis, Missouri 63166-0299, U.S.A. james.zarucchigmobot.org ABSTRACT Tephrosia xintermedia (Small) G.L. Nesom & Zarucchi, comb. n is the correct name for plants regarded as hybrids between Tephrosia espia and T. florida, replacing T. xfloridana (Vail) s sb of typification are given for Cracca intermedia Small and Cracca floridana Vai RESUMEN t iderad mo hí- Tephrosia xintermedia (Small) G.L. Nesom & Zarucchi, comb. nov., es el nombre correcto a a bridos entre Tephrosia chrysophylla y 1. florida, que reemplaza a T. pcne (Vail) Isely. Se aportan det tir intermedia Small y Cracca floridana Vail. Cracca intermedia Small was described (Small 1894) as a new species intermediate between C. chrysophylla (Pursh) Kuntze (= Tephrosia chrysophylla Pursh) and C. ambigua (M.A. Curtis) Kuntze (= Tephrosia florida (F. Dietr.) C.E. Wood). Vail (1895) curiously provided C. intermedia with a new name, Cracca smallii Vail, because she apparently regarded C. intermedia as invalidated by the earlier Tephrosia intermedia Graham in Wall. In the same publication, Vail described C. floridana Vail, noting that it was *very close" to C. smallii. Isely (1982, 1990, 1998) provided the combination Tephrosia xfloridana (Vail) Isely and used it in reference to the putative T. chrysophylla—T. florida hybrids, even though he correctly recognized Graham’s T. intermedia to be a nomen nudum. Tephrosia intermedia Graham was indeed published without a description and is nudum, but it does not block the transfer of Cracca intermedia Small to Tephrosia. Many of the nomina nuda in Wallich's list were validated by description in Don (1831-1834), but Tephrosia intermedia is not among them. *T. intermedia Grah. in Wall. Cat. 5632" was listed at the end of the paragraph for Tephrosia tinctoria Pers. var. intermedia [W & A.] in J.D. Hooker, Fl. Brit. India 2:112. 1879, so the epithet "intermedia" at the rank of species under “Tephrosia” was not there validated. illegiti nomen Tephrosia xintermedia (Smail) al L. Nesom & o comb. nov. Cracca intermedia Small, a d Bot. Club (non Tepl ud.], Numer. List [Wallich] n 2. Between 1831 and 1832). Cracca smallii Vail, Bull. Torrey Bot. Club 22:33. 1895 [nom. nov. e C. intermedia Small]. phim smallii ci. B.L. Rob., Bot. Gaz. 28:198. 1899. Tyre: U.S.A. FLORIDA. [Duval Co.]: near Jacksonville, Florida, 31 May, Fr. 11 Jul 1893, A.H. ahs 4231 M Rn NY AA digital image!; isorEcrorvees: NY 00006578, xd image!; US, US-digital image!) I de "by Mr. A.H. Curtiss in 1893, flowering on do 3lst ne in mature Pie on July 1 Ith” and that it was lee a Sd Y » Alc poor ‘Blackjack thickets’ about Jacksonville, Isely (1982) indicated “Curtiss 4231” at NY to be ihe “holotype” but he did not ere which of o Ao was e Due. Sheet 00006579 has a cie pun “type” bn 00006578 has a printed label “isotype,” bel of 00006578 has “C. smallii A.M. Vail” handwritten below the typed “Cracca intermedia Small. sp. nov.’ designations. T Cracca dicun Vail, Bull. Torrey Bot. Club 22:35. 1895. Tephrosia xfloridana (Vail) Isely, Brittonia 34:340. 1982. Type: U.S.A. FLORIDA. o.: high pineland, vicinity of Eustis, Jul 1894, G.V Nash 1198 (1ecrorvrE, designated by Isely 1982: NY, digital ni Vail me ke Florida, G.V. Nash Nos. 49415, 1198, 1263, 1334, 1552, 1615. Louisiana, New Orleans Ingalls. Tj rb Columbia College.” Photos of all syntypes except Nash 1334 are displayed on the NY web site. As oda " = (1982), e xe collection does not appear to represent a different taxon. The original hypothesis of hybrid origin of Tephrosia intermedia has been accepted by those who have further studied the genus (Vail 1895; Robinson 1899; Wood 1949). Isely (1998, p. 856—857) noted that “As alleged J. Bot. Res. Inst. Texas 3(1); 157 —158. 2009 El D H ID LI Ht f T, 158 Journal of t Texas 3( by Wood (1949), Tephrosia xfloridana probably is a derivative of hybridization between T. chrysophylla and T. florida. Whatever its origin, it is sufficiently widely distributed and consistent in characters to deserve binomial listing.” It has been collected from Alabama, Florida, Georgia, and Mississippi, but the distribution is sporadic and its origin as a recurrent hybrid seems a reasonable hypothesis. It occurs in association with the two putative parents and is intermediate in morphology. ACKNOWLEDGMENTS We are grateful to Kanchi Gandhi (GH) and Richard P. Wunderlin (USF) for comments and analysis. REFERENCES Don, G. 1831-1834. A general history of the dichlamydeous plants, vols. 1—4. sev, D. 1982. New combinations and one new variety among the genera Indigofera, Robinia, and Tephrosia (Leguminosae). Brittonia 34:339-341. Isety, D. 1990. Vascular flora of the southeastern United States, Vol. 3, Part 2: Leguminosae (Fabaceae). Univ. of North Carolina Press, Chapel Hill. sey D. 1998. Native and naturalized Leguminosae (Fabaceae) of the United States (exclusive of Alaska and Hawaii). M.L. Bean Life Science Museum. Univ. of Utah, Provo. Rosinson, B.L. 1899. Revision of the North American species of Tephrosia. Bot. Gaz. 28:193-202. Smal, J.K. 1894. Studies in the botany of the southeastern United States.—ll. Bull. Torrey Bot. Club 21:300-307. Van, A.M. 1895. A revision of the North American species of the genus Cracca. Bull. Torrey Bot. Club 22:25-36. Woon, C.E. 1949. The American barbistyled species of Tephrosia (Leguminosae). Rhodora 51:193-231, 233-302, 305-364, 369—384. A RECOMBINATION FOR VARIETIES OF ANTICLEA ELEGANS (MELANTHIACEAE) Wendy B. Zomlefer Walter S. Judd Department of Plant Biology Department of Biology University of Georgia University of Florida 2502 Plant Sciences PO. Box 118526 Athens, Georgia 30602-7271, U.S.A. Gainesville Florida 32611-8526, U.S.A. ABSTRACT The new bination is 1 le for Anticlea eleg . glaucus, formerly within the polyphyletic Zigad pl Zigad elegans var. glaucus. RESUMEN Se | binació Anticlea elegans var. glaucus, previamente incluida en el complejo polifilético Zigadem Zigadenus elegans var. glaucus. Based on molecular and morphological data (Zomlefer et al. 2001), the authors made recombinations at the generic and species-level for taxa of the former Zigadenus complex (Zomlefer & Judd 2002). While as- sisting in the update of these taxa for the Plants Database (USDA-NRCS 2009) for the National Resources Conservation Service (Cooperative Agreement 68-3H75-3-122 Mod 14; PI Craig C. Freeman, KANU), the first author realized the necessity of making the following infraspecific transfer to Anticlea elegans (formerly Zigadenus elegans), presented below. Anticlea elegans (Pursh) Rydb. var. glaucus (Nutt.) Zomlefer & Judd, comb. nov. Basionvm: Melanthium glaucum Nutt., Gen. 1:232. 1818 Zigadenus elegans Pursh var. glaucus (Nutt.) P Gl & Cronq. Man. Vasc. Pl. NorthE. U.S. Canad., ed. 2, 864. 1991. The wide-ranging Anticlea elegans (Alaska- Canada south to northern Mexico), occurs in various habitats: generally bogs, beaches, and calcareous wetlands in eastern North America, and prairies, coniferous forests, and alpine meadows in the west (Zomlefer 1997). The differences between the eastern element, A. elegans var. glaucus (plants glaucous; leaves blunt or subacute, coriaceous; inflorescence usually paniculate; bracts herbaceous, subulate; tepals intensely colored; capsule ovoid-conic), and the western A. elegans var. elegans (plants green; leaves pointed, more herbaceous; inflorescence usually racemose; bracts scarious margined; tepals pale; capsule lance-conic), as outlined by Fernald (1935), are most evident in the geographical extremes (Preece 1956), and these two taxa require further study in their area of morphological integradation in the eastern Dakotas—western Minnesota region (Gleason & Cronquist 1991; Schwartz 2002). Varietal recogni- tion is appropriate given our current understanding of the pattern of variation. ACKNOWLEDGMENTS We thank Fred Barrie and Guy Nesom for their careful check of this nomenclatural note. REFERENCES FERNALD, M.L. 1935. Critical plants of the upper Great Lakes region of Ontario and Michigan. Rhodora 37:238-262. GLEASON, A. AND A. CRONQUIST. 1991. Manual of vascular plants of northeastern United States and adjacent Canada. The New York Botanical Garden, Bronx. Preece, S.J. 1956. A cytotaxonomic study of the genus Zigadenus (Liliaceae). Ph.D. dissertation. State College of Washington, Pullman. J. Bot. Res. Inst. Texas 3(1): 159 —160. 2009 160 Schwartz, F.C. 2002. 13. Zigadenus Michaux. In: Flora of North America Editorial Committee, eds. Flora of North America north of Mexico, Vol. 26, Magnoliophyta: Liliidae: Liliales and Orchidales. Oxford University Press, New York and Oxford. Pp. 81-8 USDA-NRCS. 2009. The PLANTS Database, National Plant Data Center, Baton Rouge, Louisiana. http://plants. usda.gov. Accessed 23 April 2009, ZomLerer, W.B. 1997. The genera of Melanthiaceae in the southeastern United States. Harvard Pap. Bot. 2:133-177. Zomuerer, W.B. AND W.S. Jupp. 2002. Resurrection of segregates of the polyphyletic genus Zigadenus s.l. (Liliales: Melanthiaceae) and resulting new combinations. Novon 12:299-308. ZomLerer, W.B., N.H. WiLuAMs, W.M. WHITTEN, AND W.S. Jupp. 2001. Generic circumscription and Melanthieae (Liliales, Melanthiaceae), with emphasis on Zigadenus: evidence from ITS and trnl- -F sequence data. Amer. J. Bot. 88:1657-1669. in the tribe TAXONOMIC OVERVIEW OF EURYBIA SECT. HERRICKIA (ASTERACEAE: ASTEREAE) Guy L. Nesom 2925 Hartwood Drive Fort Worth, Texas 76109, U.S.A. www.guynesom.com ABSTRACT hia horrid five closely related j included I ithin Eurybi Horrida. These species are E. glauca, E. horrida, E. pulchra, and E. wasatchensis, as a ipao treated by DE Vs two e cede m nn T o oe as close relatives: a kingii din G.L. Nesom, comb. n g y g y Welsl L. Nesom, b (A. Nels.) G.L. Nesom, comb. nov. The first fi fu peci 1 by Brouill 1 1 1 IT DEN dE 1 1 Tel lin D +11 eee Cu De lan NAAA 1 1 EE 1 isle Herrickia and Trini } 3 g he} frhe Marl 1 lado? hological d of Herrickia and Triniteurybia f Eurybi ivial or non itl p f Eurybia at g AS , 1 N 1 1 : ] hi distribu 1 for the si i f Eurybi Herrickia RESUMEN Herrickia horrida y cinco especies muy relacionadas con ella se incluyen aqui en Eurybia sect. Horrida. Estas especies son E. glauca, E. horrida, E. pulchra, y E. wasatchensis, tal como habían sido tratadas previamente por Nesom, y otras dos especies reconocidas reciente- nte mediante pue E adu como LOMA próximos: Perla King (D.C. Eat.) Nesom, comb. nov. (que inclu e Furybla kingii var. ] N comb. nov.) y E (A. Nels.) Nesom, comb. nov. I especies fueron TN por Brouillet l gé A pli Heridas la última fue bn como hig o tinotibico Trini- teurybia Brouillet, Urbatsch & Roberts. Los d lecul indican que Eurybia, Herrichia, y Tri un grado de tres taxa en la base del clado i las dif i g Herrickia y able in di con n Eurybia son triviales 1 Syi tak d MV LIMLAlSLELLILES y F o o i ibuci Afi ] i ies de Eurybia sect. Herrickia J 5-5 E E Prior to 1994, Herrickia horrida Woot. & Standl. either had been maintained as a mo gens or treated as Aster horridus (Woot. & Standl.) Blake. Nesom (1994) united it with tl lly similar species in the western U.S.A. and treated the group as sect. Herrickia (Woot. & Standl.) son of the genus Eurybia (Cassini) S.F. Gray: Eurybia glauca, E EM E. pulchra, and E. wasatchensis. Eurybia in the sense of the present overview includes a total of 28 species distributed over eastern and western North America. Based on molecular study, Brouillet et al. (2004) reinstated Herrickia Woot. & Standl. at generic rank, including the species noted above but reducing Eurybia pulchra to varietal rank within Herrickia glauca and expanding the group to include Aster/Tonestus hingii. They also discovered Haplopappus/Tonestus aberrans to be a close relative of Herrickia and Eurybia and segregated it as the ions Bi genus Triniteurybia Brouillet, Urbatsch & Roberts (Brouillet et al. 2004). Molecular evidence indicates that this group of “herrickioid” taxa forms a grade at the base of subtribe Machaerantherinae: Oreostemma (Herrickia (Eurybia (Triniteurybia (Machaerantherinae)))) (Brouillet et al. 2004, from ITS and 3'ETS nrDNA; Selliah and Brouillet 2007, from the nuclear CNGC4 gene). The boundaries of Eurybia were further adjusted with the transfer of Eurybia (Aster) pygmaea (Lindl.) Nesom and Eurybia (Aster) chapmanii (Torrey & A. Gray) Nesom to Symphyotrichum (Brouillet & Selliah 2005; Brouillet et al. 2006). The taxonomic repositioning of Symphyotrichum chapmanii and S. pygmaea is justified, based on mo- lecular as well as morphological criteria, and those two species are outside of the immediate relationship of the species of Herrickia/Eurybia/Triniteurybia. The inclusion of Aster kingii and Haplopappus aberrans in this group of species also is a valuable and morphologically justified insight. The implicit rationale of strict monophyly, however, for recognizing three separate genera among these few species of the eurybioid grade J. Bot. Res. Inst. Texas 3(1): 161 —167. 2009 W200 oumalof t tani i Texas 3( is based on a philosophical position not universally shared by systematists (e.g., Nordal & Stedje 2005), whether or not further molecular data may confirm the pattern of cladistic relationship among them. There is no consistent, diagnostic morphological difference to separate Herrickia (sensu Brouillet) from Eurybia, and Triniteurybia is distinguished from Herrickia and Eurybia by a single feature (lack of ray florets), which is sometimes variable within species and at most no greater than sometimes differentiating species among genera of Machaerantherinae sensu stricto. Of course, as among species, there are no consistent or objective criteria specifying the number or kinds of differences that justify the recognition of closely related genera (or see McVaugh's set of recommended criteria [1945]). Similarly, however, there is no constraint that the pattern of descent be mirrored in lat that does not account for phenotypic modification (or lack of it). Were Herrickia, Eurybia, and Triniteurybia distinguished by significant morphological features, combined with a preliminary cladistic hypothesis, their treatment as separate genera would certainly be justified. The present overview examines the generic concepts of Herrichia, Eurybia, and Triniteurybia and shows geographical distributions of species included here within Eurybia sect. Horrida. Map data are taken from specimens at ASC, BRIT-SMU, MO, SJC, and TEX-LL. Morphological distinction of Herrickia Hom abad Eurybia (sensu Nesom 1994, 2000) is di rounded phyllaries with a green, often basally truncate apical patch, leararen aie disc style appendages, cylindric and multinerved cypselae, 2-seriate pappus of equal-length, apically thickened bristles, and base chromosome number of x = 9. Brouillet et al. (2004, 2006) did not provide morphological criteria for the distinction of Herrickia from Eurybia, but for the FNANM treatment, the key to genera (FNANM Editorial Committee 2006, p. 19) used the following contrast. 1 Coe A a | , ciliate-fringed, 1, Stems and leaves usually stipitate-glandular, sometimes eglandular and glaucous; leaves mostly cauline, entire or spinulose-serrate, glabrous or scabrellous; phyllaries sometimes rounded, usually keeled, apices acute to long-acuminate; rays 8-27; disc corolla tubes shorter than limbs (w Cordilleras) Herrickia Stems and leaves usually eglandular, sometimes stipitate-glandular (e North Arnerica only), not glaucous; leaves basal and/or cauline, serrate (teeth sometimes spinulose or spinose, blades then linear, grass- like, se North America) or entire, hairy or glabrous; phyllaries usually rounded, sometimes keeled, apices obtuse to acute; rays 5-60; disc corolla tubes shorter or longer than limbs Eurybia — Each of the comparisons in the key shows broadly overlapping characters and there is no consistent, diag- nostic morphological difference to separate Herrickia (sensu Brouillet) from Eurybia. My rationale for the recognition of sect. Herrickia (Nesom 1994) was similarly lacking in diagnostic cohesiveness, reflecting more a perception of the geographical coherence of the species. Eurybia horrida and E. wasatchensis are similar in their subshrubby habit, sessile and subclasping leaves, subequal phyllaries, and tendency to produce foliaceous bracts immediately subtending the involucre. Eurybia glauca and E. pulchra are similar to the former two in habit and subclasping leaves but differ in graduate phyllaries and lack of foliaceous bracts. Only the subshrubby habit is distinct among other species of the genus and E. glauca and E. pulchra sometimes are more herbaceous than subshrubby. Tonestus aberrans and Aster kingii are similar between themselves (see comments below) and both show features of Eurybia, but they are distinct from other “herrickioids” in their herbaceous habit and reduced infl ; both produce non-clasping leaves and d o pid and od lack foliaceous bracts. Thus, there apparently is no ace morphological 1 group, but tl j glyi interrelated species lust ] their geographic coherence (Figs. 1 and 2) suggest that they are closely related within the larger Eurybia, and molecular data tentatively confirm this. Morphological distinction of Eurybia and Triniteurybia In the original description of Triniteurybia as a new genus (Brouillet et al. 2004), its diagnostic features were noted (in the Latin diagnosis only, as differences from Eurybia) as eradiate heads and stipitate-glandular leaves. The western U.S.A. species Eurybia integrifolia and E. conspicua, however, have stipitate-glandular foliage, as does Herrickia (Eurybia) horrida. Herrickia glauca var. pulchra (= Eurybia pulchra), Herrickia (Eu- TE rT 164 sita af T, a7) T€AdS 311) — O Eurybia glauca A Eurybia pulchra ` O Eurybia horrida Fic. 2. Distributi f Eurybia gl. | E. pulchra, and E. horrida. The record for E, pulchra in Iron Co., Utah, is based on the citation by Welsh (2003). T i i f Eurvbi t. Herrickia 165 , Nesom, rybia) wasatchensis, and the recently joined Herrickia kingii have stipitate-glandular stems, peduncles, and phyllaries. Distal leaves of H. kingii sometimes are minutely stipitate-glandular (e.g., Smith 3508, TEX, from Box Elder Co., Utah). Some species of Eurybia in the eastern U.S.A. also produce stipitate-glandular vestiture: e.g., E. macrophylla, E. schreberi, E. spectabilis. In summary, the vestiture of Triniteurybia aberrans is not distinct among its close relatives. Brouillet later observed (2006, p. 364) that stipitate-glandularity “is most likely a shared primitive feature within [Herrickial and is therefore not indicative of a particular relationship within the group." In the FNANM treatment of Triniteurybia, Brouillet (2006, p. 382) observed that "the cylindro-cam- panulate heads with imbricate phyllaries and a wide green area, and the coarse, dentate foliage, are similar to those of Eurybia. The lack of ray florets clearly distinguishes Triniteurybia.” In fact, lack of rays has been the only morphological feature noted in any discussion that is diagnostic of the new genus. Triniteurybia aberrans is the only rayless species in the eurybioid grade, but within subtribe Machaerantherinae sensu stricto, Dieteria is is variable in ray production, Arida is rayless, and Xanthisma grindelioides is rayless. Ray producti ble among species of Grindelia. Lack of ray florets may iue. recognition of a taxon at specific rank, but there is no other example in the Astereae where it is the sole morphological character for distinction of a genus. Eurybia kingii and Furybia aberrans In the positioning of Aster kingii within Tonestus (Nesom 1991), the species was noted (p. 124) to be most closely similar to Tonestus aberrans: “Indeed, as a pair the two are somewhat set apart from the rest of the genus in the toothed leaves often pais ponies teeth, narrowly lanceolate-attenuate and apically spreading or reflexed phyllaries, styl with more widely arranged sweeping hairs, and purplish disc corolla lobes.” And (p. 125) “in mie a its white rays, it fits more securely in [Tonestus] than in Aster, particularly when placed next to T. aberrans.” Molecular evidence confirms that the two are closely related, and their morphological and geographical similarity (Fig. 1) support this hypothesis. Both are treated here within Eurybia. Taxonomic status of Aster glaucodes subsp. pulcher Blake Nesom (1994) treated Aster glaucodes subsp. pulcher at specific rank within Eurybia, as E. pulchra, distinct from E. glauca. Eurybia pulchra “differs from E. glauca in its smaller leaves, apically acute phyllaries, and well- developed glandularity. ... Eurybia pulchra has a restricted geographic range, and in the specimens I have examined, there appears to be no intermediacy between it and E. glauca” (p. 194). The present study shows that E. pulchra (northern Arizona and southern Utah) is sympatric with the more widespread E. glauca (Fig. 2) and confirms a paucity of intermediates between them. Plants of E. pulchra have distal stems, peduncles, and phyllaries densely stipitate-glandular, while those of E. glauca are completely glabrous. Numerous collections of both taxa have been made in the Grand Canyon, in the area of Havasupai Canyon and roughly between Havasupai Canyon and Bright Angel Canyon, e.g.: Eurybia glauca: Clover 4513, 5136, 7018 (LL), Deaver 1524, 2135, 2677, 2944, 3053 (ASC), Hodgson 16026 (ASC); Eurybia pulchra: Clover 7189 (LL), Deaver 3042 (ASC), Rink 4455, 4477a (ASC), Stevens s.n. [14 Aug 1992], s.n. [27 Sep 1992] (ASC). Among all plants I have studied, the only three that might suggest the possibility of gene flow are from this area: Watters 26 (ASC) and Deaver 4412 (ASC) are typical E. glauca, except that the peduncles of each are minutely and sparsely glandular for about one centimeter immediately below the heads; Stevens s.n. [10 Apr 1991] (MO) is glabrous except for sparsely glandular distal peduncles and glandular phyllary margins. A putative distinction noted by Brouillet et al. (2006) in number of disc florets (12-32 in Eurybia glauca, 29-40 in E. pulchra) is not confirmed here, as E. pulchra has florets evenly distributed in number down to at least 19. Leaf and phyllary morphology also overlap, though E. pulchra does tend to have slightly smaller leaves and more acute phyllaries. Within the area of sympatry, differences in habitat and phenology are not apparent. laftha Dat H ID hl bip, £T, 166 Journal of Texas 3( In his decision to treat Eurybia pulchra at varietal rank within E. glauca, Brouillet (2003, p. 1561) noted that “the ranges of the two taxa appear to be parapatric in southern Utah-northern Arizona with E. pulchra restricted to the vicinity of the Grand Canyon while E. glauca ranges widely to the north and east into the Southern Rocky Mountains; it must be noted however, that populations of the latter are found in Arizona south and east of the Grand Canyon. Nonetheless, no mixed population of the two taxa has been reported. The restricted range and distinct glandularity justify that E. pulchra be recognized as a taxon. Nevertheless, glandularity alone does not seem to justify recognizing E. pulchra as a distinct species.” Brouillet's observation that the lack of glands in E. glauca represents a derived condition is reasonable and probably correct but in itself not pertinent to the decision of rank. The contrasting observation Bre that the ranges are sympatric suggests that the distinct polarity in vestiture i ] through g lation, in which case specific rank for each entity is justified. It is possible that in the southern part of the range (the range of E. pulchra), some kind of genetic switching turns on or off the expression of glands, but similar variation does not occur within other taxa of sect. Herrickia or any other Eurybia species. If E. pulchra is to be recognized at any rank above “forma” the behavior of these two taxa as distinct species is the simplest hypothesis as the basis for assignment of rank. Field study clearly is needed. Welsh (2003) recognized the distinctiveness of Eurybia puchra and treated it as a variety of E. wasatch- ensis rather than a closer relative of E. glauca. The implication by Brouillet et al. (2006) that E. pulchra and E. glauca are sister taxa is better supported by morphology. TAXONOMIC SUMMARY OF EURYBIA SECT HERRICKIA Eurybia sect. Herrickia (Woot. & Standl.) Nesom, Phytologia 77:258. 1994. Herrickia (Woot. & Standl.). Triniteurybia Brouillet, Urbatsch & Roberts, Sida 21:898. Lack of a clear diagnosis is the herrickioid group, apart f I ingly i lated species cl d the geographic col of the species, rend of sect. Herrickia weakly justified. Nevertheless, it i gory in reference to e the group. l. reo aberrans (A. Nels.) G.L. Nesom, comb. nov. Basionym: Macronema aberrans A. Nels.; Haplopappus aberrans (A. 1s.) H.M. Hall; ra pti (A. Nels.) Rydb.; Tonestus aberrans (A. Nels.) Nesom & Morgan; Triniteurybia aberrans (A. v ) Brouillet, Urbat 2. Eurybia glauca (Nutt.) Nesom, Phytologia 11:260. dd Hiephahs glaucus Nutt.; ickia gl (Nutt.) Brouillet; Aster glaucus (Nutt.) Torrey & A. Gray 1841 (non Nees 1818); A od 3. Eurybia horrida (Woot. & Standl.) Nesom, Phytologia 77:260. 1994. Herrickia horrida Woot. & Standi.; Aster hor- ridus (Woot. & Standl.) Blake. 4. Eurybia kingii (D.C. Eaton) G.L. Nesom, comb. nov. Basionvm: Aster kingii D.C. Eaton in S. Watson, Botany (Fortieth Parallel) 141, plate 16, figs. 1-6. 1871. Mni ee (D.C. Eaton) Cronq. & Keck; Tonestus kingii (D.C. Eaton) Nesom; Herrickia kingii (D.C. Eaton) Brouillet, Urbats 4a. ee kingii (D.C. Eaton) G.L. Nesom var. barnebyana (Welsh & Goodrich) G.L. Nesom, comb. NOV. Basionym: Machaeranthera kingii var. barnebyana Welsh & Goodrich, Brittonia 33:299, fig. 5. 1981. Aster kingii var. barnebyana (Welsh & Goodrich) — e iin var. barnebyana (Welsh & Goodrich) Nesom; Herrickia hingii var. barnebyana (Welsh & Goodrich) Brouillet, Urbat 4b. Eurybia kingii (D.C. Eaton) G.L. Nesom var. kingii 5. Eurybia pulchra (Blake) Nesom, Phytologia 77:261. 1994. Aster glaucodes subsp. pulcher Blake; Aster glaucodes var. pulcher (Blake) Blake; Aster wasatchensis var. pulcher (Blake) Welsh; Eurybia glauca var. pulchra (Blake) Brouillet; Herrickia glauca var. pulchra (Blake) Brouillet. 6. Eurybia wasatchensis (M.E. Jones) Nesom, Phytologia 77:262. 1994. Aster glaucus var. wasatchensis M.E. Jones; Aster wasatchensis (M.E. Jones) Blake; Eucephalus wasatchensis (M.E. Jones) Rydb.; Herrichia wasatchensis (M.E. Jones) Brouillet. CONCLUSION While the overall understanding of the herrickioid group is essentially the same as that of Brouillet et al. Nesom, T. i i fE yl i t. Herrickia 167 (2004) and Brouillet (2006), the taxonomic alternative proposed here conveys an equally or more useful picture of the evolutionary standing of these species. Treating them within Eurybia, while explicitly point- ing out evolutionary relationships suggested by the cladistic hypothesis, emphasizes their morphological similarity and geographic coherence. The Brouillet et al. taxonomy emphasizes the evolutionary discrete- ness of species that are morphologically inseparable by characters used elsewhere in the tribe and family to distinguish genera. The formal recognition of a paraphyletic group at generic rank, compared to the cladistically-derived alternative, has merit in this extreme example where morphology plays essentially no role at all in the classification. Further research may resolve the issue with less ambiguity. ACKNOWLEDGEMENTS I am grateful for loans from ASC and SJC, assistance of staff during study at BRIT, MO, and TEX-LL, speci- men data from NMC for Eurybia horrida, and comments on the manuscript from John Pruski and three anonymous reviewers. REFERENCES BrouiLLEr, L. 2003. New combination in Eurybia (Asteraceae: Astereae) from North America. Sida 20:1561-1563. BrouiLter, L. 2006. Eurybia, Herrickia, Triniteurybia. |n: Flora of North America Editorial Committee, eds. Flora of North America north of Mexico. Oxford University Press, New York and Oxford. 20: 365-382; 361-365; 382-383 BrOuILLeT, L, L.E. URBATSCH, AND R.P. Roserts. 2004. Tonestus kingii and T. aberrans are related to Eurybia and the Machaerantherinae (Asteraceae: Astereae) based on nrDNA (ITS and ETS) data: Reinstatement of Herrickia and a new genus, Triniteurybia. Sida 21:889—900. BROUILLET, L. AND S. SELLIAH. 2005. Symphyotrichum pygmaeum: transfer of Eurybia pygmaea from the eurybioid grade to the subtribe Symphyotrichinae (Asteraceae: Astereae). Sida 21:1633-1635. BRouiLLET, L., J.C. SEMPLE, G.A. ALLEN, K.L. CHAMBERS, AND S.D. SUNDBERG. 2006. Symphyotrict In: Flora of North America Editorial Committee, eds. Flora of North America north of Mexico. Oxford University Press, New York and Oxford. 20:465-539. FNANM EbrroriaL Comittee. 2006. Magnoliophyta: Asteridae, part 7: Asteraceae, part 2. Vol. 20. Flora of North America North of Mexico. Oxford University Press, New York and Oxford. Hatt, H.M. 1928. The genus Haplopappus, a phylog ic study in the Compositae. Carnegie Inst. of Washington Publ. No. 389, Washington, D.C. McVauch, R. 1945. The genus Triodanis Rafinesque, and its relationships to Specularia and Campanula. Wrightia 1:13-52. Nesom, G.L. 1991. Transfer of Aster kingii to Tonestus (Asteraceae: Astereae). Phytologia 71:122-127. Nesom, G.L. 1994. Review of the taxonomy of Aster sensu lato (Asteraceae: Astereae), emphasizing the New World species. Phytologia 77:141-297. Nesom, G.L. 2000. Generic conspectus of the tribe Astereae (Asteraceae) in North America, Central America, the Antilles, and Hawaii. Sida, Bot. Misc. 20:i-viii, 1—100. Nesom, G.L. 2009 (in press). Eurybia (Asteraceae). In Heil, K.D., S. O'Kane, and L. Reeves, eds. Flora of the Four Cor- ners Region: vascular plants of the San Juan River Drainage - Arizona, Colorado, New Mexico, Utah. Missouri Bot. Garden Press, St. Louis. Nesom, G.L. ano D.R. Morcan. 1990. Reinstatement of Tonestus (Asteraceae: Astereae). Phytologia 68:174-180. NORDAL, |. AND B. STeDJE [and 148 signatories]. 2005. Paraphyletic taxa should be accepted. Taxon 54:5-8. SELLIAH, S. AND L. BrouiLtet. 2007. Molecular phylogeny of the North American eurybioid asters, Oreostemma, Her- rickia, Eurybia, and Triniteurybia (Asteraceae, Astereae) using a low copy nuclear gene, CNGCA, a. Abstract 1229, Botany & Plant Biology 2007, Joint Congress (Chicago). «www.2007.botanyconference.org» Accessed Oct 2007. SHELLY, J. S. AND M. MANTAS. 1993. Noteworthy collections, Montana. Madrono 40:271-273. WELSH, S.L. 2003. A Utah flora (ed. 3, rev). Brigham Young Univ. Press, Provo. 168 Journal of the Botanical R h Institute of Texas 3( BOOK REVIEW Tuomas G. Barnes, DEBORAH WHITE, AND Marc Evans. 2008. Rare Wildflowers of Kentucky. (ISBN 978-0- 8131-2496-4; hbk.). The University Press of Kentucky, 663 South Limestone, Lexington, Kentucky 40508-4008, U.S.A. (Orders: www.kentuckypress.com, Hopkins Fulfillment Service, RO. Box 50370, Baltimore, Maryland 21211-4370, 1-800-537-5487, 410-516-6998 fax). $39.95, 220 pp., 220 full-color ee 10" x 8 1/2". 1 1 1 11 : 1 : JST d LJ 1 "e ds 1 ls] y 5 y losing aU P P A l. These activi- 1 J > : J; MES 1 Y 1 fl o iy co Hot H 1 El 1 A A th t d wildflowers. T an additional 52 of special concern, and 60 species (historic) have not Mu seen in 20 or more a a color i images were ño with 1 film Men s transition to po cameras. C TS » | J E 1 he text Photographs were by Thomas G. Barnes, mmn, The first sections of the book are divided into a discussion of native species: Reasons for Decline of Native Pro Reasons for Protecting Native Plants, Natural ne of Kentucky ene a map that shows the nae ceps of the pate, Rarity, and J; Extinction and the Preservation Bt | EN ti O any one wl of our native (hh floras. An all too familia I ted: human lifestyl d ] d ) lt in Habitat Loss, Invasive Exotic oe Pollution, bn ci e There is a thoughtful and compelling discourse on reasons for protecting native plants. d by pean a and habitat color images. e section on TI is divided i 96), Prairies and pare as In and Hone Rivers n 13D, Wetlands (133-160), and Cliff Lines and Rock! (161-166). The fl against a a dark or black background. E o of plant names (both common and SCIRE d as well as rare un isu dd a ma a (E= EL T=threatened, S Icon H-histo A list of References and E follows Tl ] 21115 n 3 1 FinlA TH : 1 > PAR | if, 1 1 1 [1 12€ anal 6 but in many cases 1 : Y; eo ide. The threatened biodiversity, esthetic beauty of flowers, and the appeal for more f£] ] f, f, I qu 1 d: 1 1 1 d `} E 1 tat conservation efforts to save Hablas and larger flora of Kentucky.—Harold W. Keller, Ph.D., Research Associate, Botanical Research Institute of Texas, Fort Worth, Texas 76102-4025, U.S.A. J. Bot. Res. Inst. Texas 3(1): 168. 2009 THE PHYLOGENY OF SELENIA (BRASSICACEAE) INFERRED FROM CHLOROPLAST AND NUCLEAR SEQUENCE DATA James B. Beck E of buo Duke University T Not CU 27708, U.S.A. es.be ck@duke.e ABSTRACT Selenia Nutt. (Brassi ) is a North American g f five species distril lf ] land ] U.S.A. to o Moa 17131 hal : y els g 1 1 11 calliehad , Very elsi] 1 ha hin] gy find. or ic relationsl I In this study, DNA sequence iation fi I I i 1 ibed spacer and rr haul Tl Š dom f £1; 1 1 " J four non- coding a regions he v trnL intron; pan the n NOR Miu de n MR ON Lm quus was used A Yd 2l : a 1 1 1 ]; I ] resolved intra pnylogeny hl 1 1 indi 1 sonal hybridization Denda Selenia grandi d Seleni i The phylogenetic distinctiveness of S. mexicana, “knows om none few collections in Nuevo León, combined with the relative lack of collections from Mexi ggested hat add in this group. RESUMEN Selenia Nutt. (Brassicaceae) es un gé i cinco especies que se distribuyen desde el centro y sudoeste de Estados o Unides hasta] te de México. Aunque ! ía bási grup tá bien establecida, se sabe muy poco e la pene de 1 ; 1 Toss f Ati 1 g e E en las secuencias del espaciador de e interna “específico del ADN ribosomal (ADND; dificantes del ADN pue (el intrón trnL 2 I g petå-psbJ, trnQ- eee y srr) Los analisis de máxima p 1 £5 As] Á monia yl o r E r T e completamente resuelta El conflict las topologías obtenid ADN nuclear y cloroplástico indican hibrid histérica entre elenia grandis y listinción filogenética de S. mexicana, ida d pocas localidades en Nuevo León, junto con las p g | pera descubrir más diversidad en este M INTRODUCTION Selenia Nutt. (Brassicaceae) is a distinctive genus of five species distributed from the central and southwest- ern U.S.A. to northern Mexico (Fig. 1). Selenia species are small («50 cm tall), spring flowering, herbaceous annuals found on a wide range of often seasonally wet habitats from sandstone glades (S. aurea Nutt.) to limestone hills (S. dissecta Torr. & A. Gray) to alluvial soils (S. grandis R.F. Martin) (Rollins 1993). Selenia can be easily distinguished from all other genera of the tribe Cardamineae by a combination of an annual habit, fully bracteate inflorescences, yellow flowers, silicles with distinct style, and biseriate, broadly winged seeds. Although S. aurea and S. dissecta can be found in multiple states, known from 57 and 14 counties, EE L S. a as S. jonesii Cory are endemic respectively to southern and western Texas in the L is known only from the Mexican states of Coahuila and Nuevo León, although the individual ranges of this species and of $ dieu, S. jonesii, and S. grandis will surely expand following additional fieldwork in northern Mexico. This | graphic uncertainty is rey ive of a basic lack of knowledge regarding Selenia, and little to no alora oa exists concerning the reproductive biology, ecol- ogy, and phylogenetic relationships within this distinctive group (Al-Shehbaz 1988). Although both morphological and biogeographic patterns within Selenia suggest certain null phyloge- netic hypotheses, no study has addressed these evolutionary relationships. Selenia aurea is morphologically divergent from its congeners, with unappendaged sepals, pinnate (vs. bipinnate) leaves, and relatively long (>5 mm) styles. These features and its disjunct range (Fig. 1) suggest an isolated phylogenetic position for J. Bot. Res. Inst. Texas 3(1): 169 —176. 2009 170 Journal of the Botanical R h Institute of Texas 3(1) - S. aurea ¿ . N E y 5n ma" " B a m" » a "a" lg . tj : E m; E a E rie . o E:N yn Fo S. dissecta i "um H | O i : [ | w a qe : a: = + O ki ctm [| a " a l m a d O O . mf uei ¿ P0006 P jonesii ' i O O O + j iD" un. O + * : Kl E O * : «e í ree + ` t a LA LJ EP + * ^ EI * ros» p^ 4 * PS s . E * * i 1 y > hee * ‘ cr. we * * + A + sans! A 1 t DON ¿e ES $ ul ^49 Ni * . + , t Ei a iN - >» « po o Q o e mexicana Fic 1 TI A O et £ Calan! < All Ii J J . + acp abo eT ot 5 TL tantial EC 2 , rj F bad ret + Tha Ļ r] 1 fin Iahi t. hittan. fh : laf. n L _map.html T V 7 F 2 T 1 2 this species. Within the remaining species, several characters, including possession of a horn-like (vs. pouch-like) sepal appendage and relatively long (>8 mm) sepals and anthers (22.5 mm) suggest that S. dissecta, S. grandis, and S. mexicana form a natural group. These patterns of shared character variation, the morphological cohesiveness of individual species (Al-Shehbaz 1988), and the small size of the genus sug- gest that reconstructing the evolutionary relationships within this distinctive North American taxon will be tractable. This study aims to resolve the phylogenetic relationships between the five recognized species of Selenia using both chloroplast and nuclear DNA sequence variation. MATERIALS AND METHODS Taxon Sampling Cols 1 ] Sample information appears in Appendix 1. Ten d, including two samples from each of the five species recognized in perhaps the most focused examination of the genus (Martin 1940). Three taxa that have been recognized by certain authors were not included, S. jonesii var. obovata Rollins, S. Beck, Phylogeny of Selenia 171 aperta (S. Watson) Small, and 5. ipd Steyerm. pos first taxon is known only from the type collection (Rollins 1993), and the di te fruits) has been considered to be an artifact of pressing the inflated fruits (Al-Shehbaz in ms.). Martin (1940) provided a detailed discussion of the lack of distinctiveness of both S. aperta and S. oinosepala, although both taxa warrant additional study (see discus- sion). Leavenworthia Torr. has been shown to be sister to Selenia, and the monotypic genus Planodes Greene has been shown, along with Barbarea R. Br., to be sister to the Selenia/Leavenworthia clade (Beilstein et al. 2006). Two Leavenworthia samples [L. uniflora (Michx.) Britton and L. alabamica Rollins] and a P. virginicum (L.) Greene sample were therefore used as outgroups. Nine of ten Selenia samples were obtained from her- barium material, and collections made as early as 1958 (see Appendix 1) yielded successful amplifications and sequences. Molecular Methods Extractions were performed with either a Qiagen (Qiagen, Valencia, CA) DNeasy Plant Mini Kit or a Viogene (Viogene U.S.A., Sunnyvale, CA) extraction kit. The nuclear internal transcribed spacer (ITS) region was amplified using the primers "ITS 1" (White et al. 1990) and either "ITS 4" (White et al. 1990) or "ITS2- 265.4” (Rauscher 2002). The trnL"^^ intron was amplified using the primers “C” and *D" (Taberlet et al. 1991). A portion of the chloroplast trnSSU-trnGUY intergenic spacer was amplified with the primers “1F” and “1R” (Säll et al. 2003). A portion of the chloroplast petA-psbJ intergenic spacer was amplified with the primers “5F” and “5R” (Säll et al. 2003). A portion of the chloroplast trnQ""9—rps16 intergenic spacer was amplified with the primers *trnQU"*" and *rpS16x1" (Shaw et al. 2007). All reactions were performed under standard conditions. Products were visualized and purified via agarose gel electrophoresis with a Viogene gel extraction kit. Products were dye-labeled using a Big Dye Terminator Kit (Applied Biosystems, Foster City, CA), and analyzed on either a MJ Research BaseStation (MJ Research, Waltham, MA) or an Applied Biosystems 3130xl Genetic Analyzer. All sequences have been deposited in the EMBL nucleotide sequence database (Appendix 1). Phylogenetic Analyses The ITS and combined chloroplast (trnL, trnS-trnG, petA-psbJ, and trnQ—rps16) datasets were analyzed sepa- rately. Sequences were manually aligned in Se-Al (Rambaut 2002) and the aligned matrix was exported as a NEXUS file. All insertion/deletion (indel) events, both autapomorphic and synapomorphic, were scored except in the case of nucleotide rep resulting in more than two indel character states (which were viewed as likely homoplasious), or in regions of uncertain alignment. In the case of overlapping indel events, the "simple gap coding" method of Simmons and Ochoterena (2000) was used. All positions involved in indels, or situated in regions of uncertain alignment were deleted prior to analysis, and indel events were coded as pagus i dd and added to the end of the NEXUS file. Positions exhibiting poor sequence or additivity (multi ks presumably due to the presence of divergent ITS sequences in a single individual) were coded as bo For each dataset a heuristic maximum parsimony search with 100 random addi- tion replicates was performed using PAUP* 4.0b10 (Swofford 2002) with the following parameters: starting trees obtained by stepwise addition, TBR branch swapping, “MulTrees” turned on, and steepest descent not in effect. Ten thousand bootstrap replicates were cond ] with PAUP* 4.0b10 in order to obtain bootstrap Suppor e s uas. model of sequence evolution for each DNA region (indels and poorly aligned 1 using the Akaike Information Criterion in Modeltest 3.06 (Posada & Crandall 1998), and a Bayesian Markov Chain Monte Carlo analysis was performed on each dataset in MrBayes 3.1.2 (Huelsenbeck & Ronquist 2001). The combined chloroplast data were analyzed as a partitioned dataset, with the best-fitting model of sequence evolution for each separate region assigned to the corresponding partition (see Table 1). For both the ITS and combined chloroplast analyses, the indel characters were as- signed the binary model of character evolution (Nst=1, Coding=Variable) as recommended in the MrBayes documentation. All Bayesian analyses comprised four independent runs, with four chains (one cold and three heated). Flat priors were used, with the exception of the rate prior that was set to allow rates to vary among partitions. Chains were run for 5 million generations, and trees were sampled every 1000 genera- tions. Stationarity was evaluated by examining the standard deviation of split frequencies among runs and 172 Journal of tl tanical h Instit Texas 3( Taste 1. Relative phylogenetic information and model of seq lution chosen in each ofthe five DNA regi lyzed. ‘includes one inversion event. "Only a subset of the Sequence characteristic ITS trL petA-psbJ trnS-trnG trnQ-rps16 Aligned length (bp) 520 578 397 301 523 Analyzed characters, including indels 516 327 286 470 Variable characters, including indels (%) 107 (21%) 23 (5%) 31 (9%) 13 (5%) 35 (7%) Parsimony informative characters, 92 (18%) 9 (2%) 8 (2%) 3 (1%) 12 (3%) including indels (% Number of indels (parsimony informative) 3 (2) 7 (4) 8 (3) 3 (0) 9? (43) substitution model selected by Modeltest TVMef-H K8luf+l — K8luf K81uf+l K81uf substitution model implemented in MrBayes? — GTR+1 GTR-H GTR GTR-H GTR by plotting the log likelihood values from each run using Tracer 1.4 (Rambaut & Drummond 2007). These diagnostics indicated that runs reached stationarity quickly (within 100,000 generations) and I conserva- tively excluded the first 500,000 generations before obtaining a consensus phylogeny and clade posterior probabilities (PP). RESULTS Details regarding the length, variability, and model of sequence evolution chosen for each gene region are presented in Table 1. The analyzed ITS matrix of 516 characters yielded 107 (2196) variable and 92 (1896) parsimony-informative characters. The matrix contained 10 (0.196) cells coded as either missing or ambigu- ous. Additivity, indicated by multiple peaks at a single nucleotide position, was limited to three samples. Two samples (S. aurea sample 1 and S. mexicana sample 2) exhibited multiple peaks at one position each, while S. jonesii sample 2 exhibited multiple peaks at four positions. At each of these four 5. jonesii positions one of the two inferred nucleotides matched that from the other S. jonesii sample, with the other nucleotide a E CÓ typically observed in several congeners and the outgroup. Two of the three insertion/ were parsimony-informative. Each of the 100 random addition replicate parsimony searches using the ITS dataset usi the same island of five most parsimonious trees (MPTS (length = 141, con- sistency index = 0.86, retention index = 0.91). One of the five MPTs, along with bootstrap percentages and Bayesian posterior probabilities, is shown in Figure 2a. The ITS data provided low support for the monophyly of Selenia (0.60 PP, 38% BS), but strong support (1.0 PP, 100% BS) for a “core Selenia" clade comprising S. jonesii, S. grandis, S. dissecta, and S. mexicana. The ITS data also provided strong support (0.97-1.0 PP) for the sister relationship of each pair of conspecific samples. Certain chloroplast regions failed to amplify in four samples (the trnQU"6— rps16 intergenic spacer in S. aurea sample 2, S. mexicana sample 2, and the L. alabamica sample and the trnSGC"-trnG""€ intergenic spacer in S. dissecta sample 2). These samples were excluded from the combined chloroplast analysis. The analyzed chloroplast matrix of 1576 characters yielded 102 (6%) variable and 32 (296) parsimony-informative characters. The chloroplast data matrix contained 29 (0.2%) cells coded as either missing or ambiguous. Eleven of the 27 insertion/deletion events were parsimony- informative. Each of the 100 random addition replicate parsimony searches using the combined chloroplast dataset recovered the same MPT (length = 112, CI = 0.95, RI = 0.88). The MPT, along with bootstrap per- centages and Bayesian posterior probabilities, is shown in Figure 2b. Unlike the ITS data, which provided minimal support for the monophyly of Selenia, the combined chloroplast dataset strongly indicated such a relationship (1.0 PP, 99% BS). Strong support was also provided for the monophyly of “core Selenia” (1.0 PP, 100% BS), and two additional clades nested within this group. Although the four-locus chloroplast dataset was only able to assess the sister relationships of each pair of S. jonesii (1.0 PP, 100% BS) and S. grandis (1.0 PP, 95% BS) samples, a dataset including only trnL intron and petA-psbJ intergenic spacer sequences for all 10 Selenia samples provided strong support for the sister relationship of each pair of conspecific samples: 173 Beck, Phylogeny of Selenia sasueqo £ 68/0"1 68/86'0 E sipunsb *g 58/04 89/660 T — i 0738S ^g 00L/0"L z nsouo(- 001/01 rnsauo(*g F panD ‘S po] fun "1 rn VTL TU Dar Sapounid E 1 DuDonouig z DUDIXAUL cg S6/0 L I DUDOW21U 'g Z DJO2SSIp "S ZB . I DjooSSIpg 99/99°0 soBueyo £ E sipup4B s ESIPUDAB g OOL/O'L 3 gsauoCg 88/090 0010 * rgsauot'g z vano “y | 001/01 tDO4nD 'g n4ojfiun "T ] OOL/O"L DONUDQD]D „l winong Enn SOPOUD]A la MOT a E] 1.1 (Rambaut 2008). r F ET C and potå intron; trnQ—rps16, trn$"-trnG" percentage) Tk 2 3 3 174 t tanical h Insti Texas 3( S. aurea (1.0 PP, 100% BS), S. dissecta (1.0 PP, 90% BS), and S. mexicana (1.0 PP, 86% BS). The only conflict between the chloroplast and nuclear topologies involved the placement of S. mexicana, which was sister to S. dissecta in the nuclear topology and sister to S. grandis in the chloroplast topology. DISCUSSION Selenia phylogeny In general the nuclear and chloroplast datasets provide well-resolved and strongly supported phylogenetic reconstructions that are not only congruent with each other (Fig. 2), but with patterns of shared morpho- logical character states. The core Selenia clade comprising S. jonesii, S. grandis, S. dissecta, and S. mexicana is distinguished by sepals bearing a dorsal appendage, bipinnate (vs. pinnate) leaves, and relatively short («6 mm) styles. A more exclusive clade comprising S. grandis, S. dissecta, and S. mexicana is distinguished by possession of a horn-like (vs. pouch-like in S. jonesii) sepal appendage and relatively long (>8 mm) sepals and anthers (>2.5 mm). Interestingly, even though the Selenia key of Martin (1940) is artificial and there- fore doesn’t necessarily imply relatedness, if it is viewed as a bifurcating tree it perfectly matches the ITS topology presented in Figure 2a. The data clearly support the recognition of S. mexicana as distinct from S. dissecta (see below), suggesting that additional diversity is yet to be documented in this group. The three taxa not analyzed in this study (S. jonesii var. obovata, S. aperta, S. oinosepala) should therefore be subject to future molecular and morphological evaluation. Selenia aperta is a particularly intriguing case. This taxon was originally described as a variety of S. aurea by Watson (1895) based on material from San Augustine County, Texas, which exhibited broadly inflated silicles, a reduced septum, and a relatively long style. The variety was later given species status by Small (1903). Martin (1940) failed to locate the type material but examined both potential types and other specimens exhibiting these character states. He observed the variation described by Watson but found both variation among individuals from a single collection and no specimens that exhibited the full complement of characters. This variation, the potential disjunct range of S. aperta (noted in Fig. 1), and the chromosome number variation noted by Rollins and Rüdenberg (1977) all suggest that additional lineages remain to be identified within S. aurea. Potential Hybridization and Chloroplast Capture The only incongruence between the two topologies is the placement of S. mexicana, and morphological and biogeographical evidence suggest that the nuclear placement (as sister to S. dissecta) is correct and that the anomalous placement of S. mexicana by the chloroplast data is due to historical hybridization between S. mexicana and S. grandis followed by chloroplast capture. Chloroplast capture has been well documented em- pirically (Rieseberg & Soltis 1991), and appears to be possible under a range of biologically realistic situations (Tsitrone et al. 2003; Chan & Levin 2005). From a morphological standpoint, S. dissecta and S. mexicana are difficult to distinguish, with the latter exhibiting shorter («2.5 vs. »3.5 mm) styles and spongy (vs. winged) seed margins, and recent workers (Al-Shehbaz 1988; Rollins 1993) have considered it a synonym of S. dis- secta. Although the ranges of most Selenia taxa are poorly known, the existing biogeographic data also lend support to the proposed hybridization scenario, as known populations of S. mexicana are approximately 250 km closer to known populations of S. grandis than they are to populations of S. dissecta (Fig. 1). This evidence notwithstanding, S. mexicana is by far the most poorly known Selenia species, and additional cyto- logical, genetic, and field studies are needed to thoroughly test this hypothesized gene flow. Selenia dissecta has been reported to be 2n - 14, while S. grandis is known to be 2n = 24 (Warwick & Al-Shehbaz 2006). The currently unknown chromosome number of S. mexicana will therefore reveal if the proposed hybridiza- tion event was via a polyploid or homoploid pathway. Evidence of additivity in the S. grandis or 5. mexicana ITS sequences was limited to a single position in S. mexicana sample 2, indicating that any heterospecific ITS repeats have been largely eliminated due to backcrossing to conspecifics (the homoploid scenario) or concerted evolution (Franzke & Mummenhoff 1999). Particularly in a homoploid hybridization scenario, sequencing both individual cloned ITS sequences and other nuclear loci in an expanded sample set from across both species’ ranges will reveal the extent of proposed introgression. Three individuals exhibiting Beck, Phylogeny of Selenia 175 S. mexicana morphology (short styles in particular) were discovered in a recent survey of specimens from six major herbaria (BRIT, GH, MO, NY, TEX-LL, and US), all from within 100 km of the type locality of Galeana, Nuevo León (Standley 1937). Unfortunately, these collections contain few specimens from Mexico, and additional fieldwork is clearly needed. As noted in the introduction, the range of S. mexicana is but one of many aspects of Selenia species that are poorly documented. Future work, including expanded sampling of all proposed Selenia taxa, is therefore needed to understand the biology of this genus, information that can now be placed in an evolutionary context. APPENDIX 1 Sample information. Taxon (sample number), voucher, year of collection (if herbarium material) try, state, county (if applicable), ITS EMBL, trn£ EMBL, trnS-trnG EMBL, petA-psbJ EMBL, trnQ-rps16 EMBL. oe alabamica Rollins, Beck 486 (MO)—U.S.A. ALaBAMA. Franklin Co.: FM957596, FM957609, FM986404, FM986416, non menso po pal Britton, Beck 516 (MO) —U.S.A. ALasama. Morgan Co.: FM957595, FM957608, FM986403, nive e n pm Al-Shehbaz s.n. (MO)—U.S.A. Missouri: FM957594, FM957607, FM986402, FM986414, 428 Selenia aurea Nutt. (1), Beck 774 (MO)—U.S.A. Missouri. St. Clair Co.: FM957598, FM957611, FM986406, FM986418, F Selenia aurea Nutt. (2), Stephens 29996 (GH) 1969—U.S.A. OxtAHoMA. Nowata Co.: FM957597, FM957610, FM986405, FM9 ,none Selenia dissecta Torr. & A. Gray (1), Worthington 11630 (NY) 1984—MEXICO. Chihuahua, FM957600, FM957613, FM986407, F 20, FM986431 Selenia dissecta Torr. & A. Gray (2), Correll 38395 (TEX-LL) 1970—U.S.A. Texas. Culberson Co.: FM957599, FM957612, none, FM986419, FM986430 Selenia Pas E F. Martin (1), Turner 4323 (TEX-LL) 1958—U.S.A. Texas. Nueces Co.: FM957603, FM957616, FM986410, FM 986435 Selenia ards E F. Mei (2), Correll 36762 (TEX-LL) 1969—U.S.A. Texas. Hidalgo Co.: FM957604, FM957617, FM986411, mm RB Y a jonesii oie (1 pum & Lundell 16958 (GH) 1961—U.S.A. Texas. Dawson Co.: FM957601, FM957614, FM986408, FM986421, FM986432 Selenia jonesii Cory (2), Mahler 8846 (GH) 1981—U.S.A. Texas. Reagan Co.: FM957602, FM957615, FM986409, FM986422, F 643 Selenia mexicana Standl. (1), Hinton 27036 (TEX-LL) 1997—MEXICO. Nuevo León: FM957605, FM957618, FM986412, Eden Eee piso (2), Crutchfieid & Jot (GH) 1960—MEXICO. Nuevo León: FM957606, FM957619, FM986413, FM986426, none ACKNOWLEDGMENTS The author would like to thank the curators of BRIT, GH, MO, NY, TEX-LL, and US for herbarium loans. Special thanks go to Ihsan Al-Shehbaz for critical discussions, to George Yatskievych for help with locating Selenia aurea, to Eric Schuettpelz for help with Bayesian analyses, to Lisa Pokorny for providing the Span- ish abstract, and to Ray Collier for laboratory assistance. Manuel B. Crespo (ABH) and two anonymous reviewers provided helpful comments on the manuscript. This study was funded by the Graduate Studies Program at the Missouri Botanical Garden and the Division of Biology and Biomedical Sciences at Wash- ington University in St. Louis. REFERENCES AL-SHEHBAZ, 1.A. 1988. The genera of Arabideae (Cruciferae, Brassicaceae) in the southeastern United States. J. Arnold Arbor. 69:85-166. AL-SHEHBAZ, I.A. [forthcoming] Selenia (Brassicaceae). In: Flora of North America Editorial Committee, eds., Flora of North America North of Mexico Vol. 7. Oxford Univ. Press, Oxford. Laf al Das e ID ll ip. £T, 176 Texas 3( BeisTein, M.A., LA. At-SHEHBAZ, AND E.A. KELLOGG. 2006. Brassicaceae phylogeny and trichome evolution. Amer. J. Bot. 93:607-619. CHAN, K.M.A. AND S.A. Levin. 2005. Leaky prezygotic isolation and porous genomes: Rapid introgression of mater- nally inherited DNA. Evolution 59:720-729. FRANZKE, A. AND K. MUMMENHOFF, 1999, Recent hybrid speciation in Cardamine (Brassicaceae)-conversion of nuclear ribosomal ITS sequences in statu nascendi. Theor. Appl. Genet. 98:831-834. HueLsensEcK, J.P. AND F. RonquisT. 2001. MrBayes: Bayesian inference of phylogeny. Bioinformatics 17:754-755. Martin, R.F. 1940. A review of the Cruciferous genus Selenia. Amer. Midl. Naturalist 23:455-462 Posapa, D. AND K.A. CRANDALL. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817-818. Rameaut, A. 2002. Se-Al, Version 2.0a11. Available at: ja n Em edac. Ir RAMBAUT, A. 2008. FigTree, Wersion 1.1. Available at: h d.ac.uk/software/figtree/ RAMBAUT, A. AND A.J. DRUMMOND. 2007. Tracer, Version 14. dee from http://beast.bio.ed.ac.uk/Tracer RAUSCHER, J.T. 2002. Molecular phylogenetics of the Espeletia complex (Asteraceae): evidence from nrDNA ITS sequences on the closest relatives of an Andean adaptive radiation. Amer. J. Bot. 89:1074—1084. RIESEBERG, L.H. AND D.E. Souris. 1991. Phylogenetic consequences of cytoplasmic gene flow in plants. Evol. Trends Pl. 5:65-84. RoLuws, R.C. 1993. The Cruciferae of continental North America. Stanford University Press, Stanford, CA. RoLuins, R.C. AND L. RUDENBERG, 1977. Chromosome numbers of Cruciferae Ill. Contr. Gray Herb. 207:101-116. SALL, T., M. JakoBssoN, C. LIND-HALLDEN, AND C. HaLLDEN. 2003. Chloroplast DNA indicates a single origin of the allotet- raploid Arabidopsis suecica. J. Evol. Biol. 16:1019-1029. Shaw, J., E.B. Lickey, E.E. SCHILLING, AND R.L. SMALL. 2007. Comparison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in angiosperms: The tortoise and the hare Ill. Amer. J. Bot. 94:275-288. Simmons, M.P. ano H. OcHoTerENA. 2000. Gaps as characters in sequence-based phylogenetic analyses. Syst. Biol. 49:369-381. SMALL, J.K. 1903. Flora of the southeastern United States. published by the author, New York. STANDLEY, PC. 1937. Studies of American plants- aul Field a Nat. Hist., Bot. Ser. 17:155—284. SwoFFORD, D.L. 2002. PAUP*. Phylogenetic y (*and other methods), Version 4.0b10. Sinauer Associates, Sunderland, MA. TABERLET, P, L. GieLty, G. PAUTOU, AND J. Bouver. 1991. Universal primers for amplification of three non-coding regions of chloroplast DNA. Pl. Molec. Biol. 17:1105-1109. TSITRONE, A., M. KIRKPATRICK, AND D.A. Levin. 2003. A model for chloroplast capture. Evolution 57:1776-1782. WARWICK, S.l. AND |. A. AL-SHEHBAZ. 2006. Brassicaceae: chromosome number index and database on CD-Rom. PI. Syst. Evol. 259:237-248. WATSON, S. 1895. Selenia. In: B.L. Robinson, ed. Synoptical flora of North America. American Book Company, New York P. 151. Whrre, T.J., T. Bruns, S. LEE, AND J. TAYLOR. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: M. Innis, D. Gelfand, J. Sninsky, and T. White, eds., PCR protocols: a guide to methods and applications. Academic Press, San Diego, CA. Pp. 315-322. Á 2r PRELIMINARY INSIGHTS INTO THE PHYLOGENY AND SPECIATION OF SCALESIA (ASTERACEAE), GALÁPAGOS ISLANDS Jeremy D. Blaschke Roger W. Sanders uu Pane #7071 Bryan College # 7802 721 Bryan Dr. Dayton, Rae m U.S.A. Dayton, lennessee 37321, U.S.A. jblaschke0998@bryan.edu rsanders@bryan.edu ABSTRACT Scalesia Arn. (A Heliantl )i lyg f fifi peci demi I lapagos Island ee distribution J : ; C pb RE 3 £. 1 1; 1 1 1 All species of Scalesia Pappo lus SE Blake and Sim sia Pers., along with selected species of Viguiera Kunth, were ejected to phy logenetic analysis (63 characters in 78 taxa). Homoplasy and i g g lting pud resolutio d comparison of speciation events dd: to its eds du dd phologically, species throug! g ked primarily by homoplastic apomorp ies. Base fhabit ph 1 gi lai g f, 1 f, 1 y Lab; pr 1 poorly correlated. cup for f dies should b ded to include other groups in the derived Helianthinae RESUMEN a (A Heli I ) gé lef d I i p ies, dcus t tasta láp g 1 1 5 lógi Lito 1 x T J aie [ad 1 1 Li J Tye Oe P p fs ] * ah seleccionados. Todas | ies de Scalesi TP S.F. Blake, y Simsia Pers., junto con algunas E de Wiewers Kunth, fueron eae 1 un análisis Ei inodo iid caracteres en ids nni La ee e Ud M. entre zs árboles epe idw la E 4 E r E - E ; : fine] leet Bacadecen | pli ización de los hábitat q divergencia 1 f, di Lb r INTRODUCTION 1: L: y . Ef, species, all J a trat? land láp Scalesia Arn. (Asteraceae: Heliantheae: E Islands. Howell (1941) accepted 18 species in four Secun and related it to the Ecliptinae Lessing. Based on extensive field study, Eliasson (1974) recognized only 14 species, avoided the use of sections, discussed aspects of character evolution, and placed the genus in the Helianthinae Dumort. Hamann and Wium- Andersen (1986) described an additional species. Recent studies on Scalesia have focused on intergeneric relationships (Schilling et al. 1994; Spring et al. 1999; Panero 2007), chemical diversity (Adsersen € Baerheim Svendsen 1986; Spring et al. 1997, 1999; Petersen et al. unpubl.), anatomy (Carlquist 1982), autecology (Itow 1995; Kitayama & Itow 1999; Hamann 2001), adaptive reproductive strategies (McMullen & Naranjo 1994; Nielsen et al. 2002, 2007), and popula- tion structure (Nielsen et al. 2003; Nielsen 2004). Chloroplast DNA restriction site analysis suggests that Scalesia belongs to a group of specialized genera, the “derived Helianthinae,” that are embedded within a derived clade of Viguiera Kunth (Schilling et al. 1994). Viguiera, whose taxonomy has been unresolvable on morphological grounds, appears as a paraphyletic assemblage basal to all other genera in the Helianthinae on the basis of cpDNA restriction sites and internal transcribed spacer (ITS) sequences of nuclear ribosomal DNA (Schilling & Jansen 1989; Schilling & Panero 2002). In an analysis in which S. pedunculata Hook.f. and several species of Pappobolus S.F. Blake were sampled (Schilling et al. 1994), Scalesia and Pappobolus were sister groups, and the next closest clade consisted of Simsia Pers. plus Viguiera ser. Pinnatilobatae S.F. Blake. However, the authors noted problems with interpretation of the three restriction sites synapomorphic for Scalesia and Pappobolus and concluded, "Thus, the relative relationships among Scalesia, Pappobolus, Simsia J. Bot. Res. Inst. Texas 3(1): 177 —191. 2009 178 Journal of t tani Insti Texas 3( and Viguiera ser. Pinnatilobatae are not well resolved by these data." Indeed, subsequent work has shown that chloroplast restriction sites, chloroplast genes, and ITS regions lack sufficient variation to resolve relation- ships in the derived Helianthinae (Schilling & Panero 1996, 2002; Petersen et al. unpubl.). However, recent analyses using external transcribed spacer (ETS) regions did find sufficient diversity to resolve species of Helianthus (Timme et al. 2007), a member of the derived Helianthinae, as well as species in other genera of the Heliantheae, such as Montanoa (Plovanich & Panero 2004). Our interest is primarily in patterns of diversification, homoplasy, speciation rates, and degree of ad- aptation using Scalesia because it is the most speciose endemic angiosperm genus in the Galápagos. It is of interest that homoplasy among morphological characters of Simsia species prevented Spooner (1990) from publishing a cladistic analysis in his monograph. Likewise, Panero (1992) chose not to include phylogeny in his monograph of Pappobolus, instead recognizing only phenetic groupings. Thus, notable amounts of unanalyzed data are available in the literature for addressing the issue of homoplasy across Scalesia and relations. While it is our hope in the future to sample ETS regions in Scalesia species to determine their ap- plicability in phylogenetic analysis, as well as use pr Mp: in investigating issues of interest to us, our goal here is to mine the existing pertinent nd ecological data that are available in the literature to provide a comparative context for later molecular ce That is, we seek to provide insights into: 1) sister-group and intrageneric relationships of Scalesia, 2) homoplastic traits, 3) relative amounts of speciation per clade, and 4) directions for future molecular sampling. We anticipate that phylogenetic analysis of morphology may not yield consistent assessments of relationships or be congruent with mo- lecular phylogenies (for example, see Plovanich and Panero [2004] concerning homoplasy in morphological taxonomic criteria in the Heliantheae). However, that result is not certain, for the hand-calculated Wagner parsimony networks of morphological data of Dendroseris and Robinsonia (Sanders et al. 1987), two other island endemics of the Asteraceae (Lactuceae and Senecioneae, respectively), did prove to be cong with later molecular phylogenies (Crawford et al. 1992; Sang et al. 1995). METHODS Data.—Taxon sampling is based on the sister-group conclusions and Figure 1 of Schilling et al. (1994) and availability of supplemental specimens at the Botanical Research Institute of Texas and Bryan College. Morphological traits, coded as binary and multistate unordered characters, were extracted from published monographs of Pappobolus (Panero 1992), Scalesia (Eliasson 1974; Hamann & Wium-Andersen1986), Simsia (Spooner 1990), and species representing Helianthus L. (Schilling 2006), Viguiera sect. Maculatae (S.F. Blake) Panero & Schilling (Panero & Schilling 1988), and the outgroup Bahiopsis Kellogg (Schilling 1990). Selected dried specimens were consulted to verify codings obtained from the literature, supply missing data, and score representative species from Viguiera ser. Grammatoglossae S.F. Blake and ser. Pinnatilobatae (Table 1). Characters were chosen to maximize distinctions within Scalesia, Pappobolus, and Simsia and scored accord- ingly in the remaining taxa (Table 2), resulting in a number of characters being coded as polymorphisms. The compiled data constitute 63 characters in 78 species (Appendix). Phylogenetic Analysis —Parsimony analysis was conducted using PAUP* 4.0b10 (Swofford 1998). Heu- ristic searches were made with character optimization set to both accelerated and delayed transformation and with the following options: character weighting equal, 10 rounds of random addition sequence with 100 trees held at each addition, branch swapping by tree-bisection, MulTrees in effect, MaxTrees=100,000. Bootstrap analysis (10,000 replicates) was conducted using accelerated transformation by heuristic search with 10 trees held at each addition step. A final heuristic search, in which the majority-rule consensus tree from the bootstrap analysis was input for branch swapping only, was conducted using accelerated transfor- mation with options as above. Based on the strict consensus tree from the first heuristic search, a reduced matrix of only the ancestral nodes of the outgroup, Helianthus, Pappobolus, Simsia, and Viguiera grammatoglossa + V. stenophylla; the remaining Viguiera species; and the species of Scalesia was generated. This matrix was subjected to a branch-and-bound search (options: accelerated transformation, equal weighting, MulTrees in effect, furthest addition sequence) and bootstrap analysis as above. Tagle 1. Hert ited to suppl ify y y recognized by p y to Taxon Specimen Locality Herbarium hiopsi B. deltoidea A. Gray S. White Mexico: Baja, California. La Paz SMU B. parishii Greene Mahler & ar USA: Arizona: Maricopa Co: Sagauro Lake SMU Helianthus W. L. Henning Acc. No.B802 USA: Missouri: Boone Co.: W of Columbia BRYAN G. Varga Acc. No. B 1794 USA; Tennessee: Rhea Co, BRYAN H. tuberosus L. W. L. Henning Acc. No. B 804 USA: Missouri: Boone Co: S of Columbia BRYAN Pappobolus P. acutifolius (S.F. Blake) Panero ac 1399 Perd: Ancash: Caráz BRIT P. matthewsii (Hochr) Panero J. Panero 135. P : Pedro Ruiz BRIT P robinsonii Panero Pa cM 1225 Perú: Cajamarca: Celendin BRIT P steubelii (Hieron.) Panero Panero et al. 932 Perú: Cajamarca: Chalhuayaco BRIT Scalesia S. affinis Hook.f. Mears 5296 Ecuador: Galápagos: Floreana BRIT S. helleri B.L. Rob. Mears 5494 Ecuador: Galápagos: Santa Fe BRIT S. stewartii L. Riley Mears 5556 Ecuador: Galápagos: Bartolomé BRIT S. villosa A. Stewart Mears 5226 Ecuador: Galápagos: Gardner BRIT Simsia S. amplexicaulis (Cav.) Pers. A. Cronquist 9611 Mexico: Michoacán: La Piedad SMU S. calva A. Gray J. Rodriguez Mexico: Nuevo León: Vallecillo SMU U. Waterfall 16660 Mexico: Coahuila: Sabinas SMU S. eurylepsis S.F. Blake U. Waterfall 14300 Mexico: San Luis Potosí: Ciudad de Valles SMU S. foetida (Cav.) S.F. Blake Yen & Estrada 6479 Mexico: Chihuahua: Presa Chihuahua BRIT S. fruticulosa (Spreng) S.F.Blake — King € Guevara 5817 Colombia: Cundinamarca. Chipaque MU S. holwayi S.F. Blake R. M. King 7337a Guatemala: Alta Verapaz: San BRIT Cristóbal Verapaz S. sanguinea A. Gray C.G. Pringle 11513 Mexico: Jalisco: Guadalajara SMU Viguiera ser. Grammatoglossae V. cordifolia A. Gray J. Cornelius 227 USA: Texas: Brewster Co.: Black Gap WMA SMU D. S. Correll 15006 USA: Texas: Jeff Davis Co.: Davis Mts. SMU V. grammatoglossa DC. J. Rzedowski 34497 Mexico: Oaxaca. Chilapa de Díaz VDB Viguiera sect. Maculatae V. adenophylla S.F. Blake E. Estrada 1889 Mexico: Nuevo León. Iturbide BRIT Viguiera ser. Pinnatilobatae V. stenoloba S.F. Blake A. Krings 288 USA: Texas: Pind Co.: Big Bend BRIT Ranch State Park Nee & Diggs 25354 Mexico: Edo. pom Mun. San Pedro BRIT A. Treverse 2215 USA: Texas: Brewster Co; Big Bend BRIT National Park Bayesian analysis was conducted using MrBayes v3.1.2 (Huelsenbeck & Ronquist 2001) on both the full and the reduced matrices using the default settings of the standard discrete evolutionary model. Analysis of the full matrix was run for 200,000,000 generations and sampled once every 100,000 generations; the reduced matrix was run for 400,000 generations and sampled every 100 generations. Habitat Characterization.—Geographic distributions and habitat features were estimated from Cron- quist (1971), Eliasson (1976), Hamann and Wium-Andersen (1986), and personal observation of one of us (RWS) 180 Taste 2. Characters and cl states used in data matrix (Appendix). Character states are unordered. 1. Habit 0: shrub, 1:tree, 2: suffrutescent perennial, 3: peren- nial herb, 4: annual 2. Hai 0 to PAN 2: scabrous or strigose 3. Glandular trichomes 0: absent, 1: present 4. hod pubescence color 0: white to gray, 1: yellow or , 1: villous > Leaf Phyllotaxy o: aoe 1: peo is 1:lvs. partially or gradually ERO. 2 pu much reducing into inflor. 7. lla 0: ovate, 1: lanceolate, 2: cordate, 3: triangular, 4: elliptic, 5: linear-oblon 8. Leaf margin lobing 0: unlobed, 1: lobed 1/4 to midrib, 2: lobed 1/2 to midrib, 3: lobed 3/4 to midrib, 4: regularly deeply lobed nearly to midrib 9, Leaf margin serration 0: ai era, 1: crenate or serrulate, indistinctly toothe 10. Leaf margin orientation 0: flat, 1: revolute 11. Leaf adaxially strigose 0: not strigose, scabrous, or sericeous, 1: moderately strigose, scabrous, or sericeous, 2: densely strigose, scabrous, or sericeous 12. Leaf abaxially strigose 0: not strigose, scabrous, or co 1: moderately strigose, scabrous, or sericeous, sely strigose, scabrous, or sericeous 13. Leaf abaxially lanate 0: not villous or lanate, 1: moder- ately villous or lanate, 2: densely villous or lanat 14. Leaf surface reflectance, adaxially 0: dull, 1: A 15. Leaf texture 0: P ous/chartaceous, 1: leathery, 2: thinly m a Leaf venation 0 tripinerved, M PERS 1 a 2: distinctly ICIAL 1: sunken below surface 18. Petiole shape 0: unwinged, 1: wing tapering in apex, 2 wing tapering above 3: wing broad to basal inser- tion, 4: a at base only 19, Petiole length 0: 0-9 mm, 1: > 20. inflorescence reiteration t monochasal, 1: dichasial 2 litary, 1: open panicle, 2: no aggregate ibus 2.1 large c large 15-30 mm, 2: moderate 7- 157 s z lee 23.1 p en lat dea 24. Phyllary series 0: 3-4, 1: 2, 2: (4-)5-6 25. Phyllary shape 0: oblong to obtrullate, oo elliptic, 2: lanceolate, 3: o Suena i le idric, 1: urceo- 26. Phyllary, o outer serie t 1:spatulate 27. Phyll florets 0: sul Ji gf , 1: overtop- pin ng flore 28. Phyllary size ratio, outer/inner 0: outer + inner, 1: outer inner 29. Phyllary pes A oe green, 1: stramineous + with green strip , 3: purple 30. Phyllary o scale- like, 1: foliaceous 31. Phyllary pubescence density 0: revealing surface, 1: obscuring surface 32. Phyllary margin, cilia 0: without cilia, 1: ciliate 33. Phyllary tip shape 0: blunt or abruptly acute, 1: acumi- nate, long acute 4. Phyllary tip ori 1, 1: reflexed UO nyral y up MP Crapredelid 5. Phyllary thicl 0: untl base slightly thickened indurate, 2: base conspicuously thickened indur 36. la jm 1: present in full complement, 2: present in pa 37. is ae ish 0: spreading, 1: strongly reflexed or | |, herbaceous, 1: ved 38. a ligule length 0: «1.5 cm, 1: 1.5-3.0 cm, 2: > 3 cm 39. Ray apex fusion 0: shallowly 2-3 toothed, 1: deeply 2-3 dede 2: Jua. barely fused aa 1: lin . Ray MP WEG 41. Palea length 0: : about equalling dor 1: a above phyllaries, 2: shorter than phyllari 42. Palea apex pubescence 0: Pede 1: E 43. Palea segmentation 0: lacking, 1: shallow, 2: deep 44. Palea segments, shape 0: elliptic, 1: triangular, 2: ovate- rounded, 3: subulate, 4: oblong-ligulate 45. de piel orientation 0: erect, 1: diverging or xed, 2: strongly Xena no! g, 3: a or iud t uch longer than laterals 47. Disk corolla color 0: yellow to orange, 1: brown, 2: pale yellow, 3: white, 4: pink, 5: deep purple 48. Disk corolla tube to throat length ratio 0: -3-4, 1: ~] ~5—10, 2: 49. m corolla tube pubescence 0: glabrous, 1: puberu- len 50 raul n I L f gl | 51 Dial n i L Th ri “al with black pigment, 2: with purple pigm 52. eae yellow, 1: black, 2: galo nine distally, , 1: puberulent dark pigment, 1: 53. Anther appendix coloro. stramineous, 1: all or part black, white 54. ud branch Red i e 0: without black pigment, 1: with black pigm 55. Style branch apex n deltate, 1: attenuate 56. Style branch appendage 0: absent, 1: present 57. Achene length 0: « 3 mm, 1: 3-5 mm, 2: » 5 mm 58. Achene pubescence 0: glabrous, 1: sericeous 59, Achene compression 0: biconvex-lenticular, 1: laterally flat but e ly bi 2: strongly lat. flattened, 3: terete or trigon 60. de developmen 0: absent, 1: callous ring only, 2: ns and/or scales 61. Ba appus, no. awns 0: 0, 1: 1 (often small), 2: 2, 3: multiple 62. Pappus, intervening scales 0: absent, 1: present 63. Pappus persistence 0: persistent, 1: caducous = RESULTS Sister-group relationships.—The first two heuristic searches (random-addition with accelerated vs. de- layed character transformations) resulted in 100,000 shortest trees each (442 steps). These and their strict consensus trees were partially incongruent with the majority-rule tree of the bootstrap analysis. In the delayed transformation search, Viguiera adenophylla was sister to all other ingroup taxa, and Pappobolus was paraphyletic with P. ecuadoriensis sister to all remaining taxa. Of these, one clade contained P. sagasteguii, a subclade of V. stenoloba + Scalesia, and a subclade e the remaining Mrs species, Helianthus, and Simsia as monophyletic genera. The other clad I ppobolus. The accelerated transformation search resulted in V. adenophylla as above but the aida o taxa constituted five clades in an unresolved polytomy: V. cordifolia, Helianthus, Scalesia, Simsia, and one having a monophyletic Pappobolus sister to V. grammatoglossa + V. stenoloba The third heuristic search (bootstrap majority-rule tree input and branches swapped) resulted in all 100,000 trees being congruent with the bootstrap analysis, though one step longer (443) than the trees from the first two searches. In the strict consensus tree of this analysis (Fig. 1), the ingroup formed three major clades. A monophyletic Pappobolus was sister to the remaining ingroup taxa. Of these, one clade consisted of V. adenophylla and Scalesia as sister groups. The other clade contained a tetrachotomy: Simsia, Helianthus, V. cordifolia, and V. grammatoglossa + V. stenoloba. The Bayesian majority-rule consensus tree (analysis final average standard deviation 0.0078) added yet another possible arrangement. Of the ingroup taxa, Scalesia + V. adenophylla were sister to the remainder, which formed a polytomy: V. grammatoglossa, V. stenoloba, nine species of Pappobolus, a clade with all the remaining Pappobolus, and a clade consisting of V. cordifolia, Helianthus, and Simsia. In all of the consensus trees, Simsia was completely unresolved or nearly so, and Pappobolus contained two to three large sets of unresolved species. Scalesia was reasonably well resolved but its topology differed among trees. All heuristic searches found the arboreous species as a resolved clade (S. cordata A. Stewart, S. microcephala B.L. Rob., S. peduculata basal), the lobe-leaved species (S. baurii Robinson & Greenman, S. helleri Robinson, S. incisa Hook.f., S. retroflexa Hemsl.) as a partially or fully resolved clade, the three spe- cies with elongate phyllaries (S. atractyloides Arn., S. stewartii L. Riley, S. villosa A. Stewart basal) as a grade or clade, and a clade of S. divisa Andersson + S. gordilloi OJ. Hamann & Wium-And. In two searches the arboreous clade was sister to the remainder with the elongate-bracted clade deeply imbedded; in the third the elongate-bracted group was a basal grade with the arboreous clade deeply imbedded. In the Bayesian majority rule tree, Scalesia was an eight-way polytomy of the arboreous, elongate-bracted, and lobe-leaved clades, S. affinis Hook.f., S. aspera Andersson, S. crocheri J.T. Howell, S. divisa, and S. gordilloi. Branch-and-bound analysis of the reduced matrix produced 13 trees of equal length (163 steps). In the strict consensus tree (Fig. 2), V. adenophylla was sister to the other ingroup taxa, which formed a tetra- chotomy: Scalesia, Helianthus, Simsia + V. cordifolia, and Pappobolus + the V. grammatoglossa-stenophylla ances- tor. In Scalesia, the arboreous clade (unresolved) was sister to the remainder which formed a polytomy of S. affinis, S. aspera, S. crocheri, S. divisa, S. gordilloi, a partially resolved clade of the lobe-leaved species, and a resolved elongate-bracted clade. However, the Bayesian majority rule tree (analysis final average standard deviation 0.0070) of the reduced matrix differed by being nearly identical to one of the most parsimonious branch-and-bound trees (Fig. 2) except that 1) the arl ies formed a basal grade with S. cordata + S. microcephala sister to the remaining species, 2) there was no PUDE among S. affinis, S. aspera, S. crocheri and s Masc. pa dno 3) S. retroflexa was basal to the other BOO. of the lobe-leaved clade. In one ofthe 100,000 dl lly 1 from the third heuristic Seni the Composite CURRO index (CD was 0. 24 ( (excluding two unif tive ch Dis rescaled CI (RC) was 0.17, and the retention index (RI) was 0.71. In this tree, the onl ies of S with a con nd M over 0. y were el florets absent and anthers black. b p. comparable DE for Pap- [ 1 for Simsia, phyllaries not thickened at laterally flattened pobolus The RE Cl of the branch-and-bound trees (Fig. 2) was 0.49 (including only 51 informative char- 182 84 87 | | 71 98 69 86 86 99 rir B deltoidea B parishii / adenophylla Sc pedunculata c corda Sc microcephala c crockeri c aspera c villosa Sc atractyloides tewartii c helleri c retroflexa urii c incisa cordifolia | annuus 1 tuberosus eurylelpis lagasciformis calva sanguinea setosa villasenorii Fic. 1. Strict ` rr » F F [d 4 Generic abbreviatio ns: B=Bahiopsis, H=Helianthus, P=Pappobolus, Sc=Scalesia, Si=Simsia, V=Viguiera. Bahiopsis 63 56 V adenophylla V gramm steno Pappobolus Helianthus V cordifolia Simsia Sc microcephala "M ie 100 » Ep] LH a (p Sc cordata 1 Sc pedunculata 23 Sc affinis Sc villosa Sc atractyloides Sc stewartii Doa [po Sc crockeri 121 Eo Ee Sc aspera 31 Sc divisa 23 Sc gordilloi 13 Sc retroflexa 11 Sc helleri 7 Sc baurii 1 Sc incisa | 3543] J I é Anek IE probabilities 183 184 J t tani i Texas 3( acters), the RI 0.66, and RC 0.37. In this tree, the synapomorphies for Scalesia supported by a consistency index of 0.4 or greater included: capitula 15-30 cm diameter; involucre hemispheric to urceolate; phyllaries oblong to obtrullate and erect/appressed; ray absent but, when present deeper within the clade, reflexed with irregularly fused lobes; paleae deeply segmented; corollas white; anther appendices white; and achenes glabrous. Black anthers, instead, appeared to be synapomorphic for the ingroup minus V. adenophylla. The third heuristic search of the full matrix resulted in only six characters that were completely con- sistent: four por autapomorphies (or e for species pairs), whereas only three involved gnificant clades. Elev ters had consistency indices of 0.5 or higher. Bout a iliése a base thickening, ray Bm fusion of ray lobes, and shape of palea segments) were pcne or reversals within Scalesia; only two (growth habit, orientation of palea segments) were paral- leli t species of Scalesia and other genera. Forty-six characters had consistency indices lower than 0.5, of which 30 aipsdredd in both Scalesia and other genera, 15 in only other genera, and only one (ratio of corolla tube to limb lengths) just in Scalesia. The branch-and-bound matrix had only 55 variant characters. Sixteen were consistent, and, of these, eleven involved a of significant clades. Twenty-five cl ters were homoplastic with a con- sistency index of 0.5 or higher including ni ing within Scalesia and six in Scalesia and related genera. Only 14 characters were below the 0.5 eorsisieney index level with only one restricted to species of Scalesia (as above), only one Guilde of Scalesia, and the remaining 12 appearing in both Scalesia and other genera. aphic di and ecology.—All species except Scalesia affinis, which is sympatric with S. aspera, S. ae - helleri, S. retroflexa, ide eue are narrowly ad or AO (Fig. 3). Some have disj ions occurring All th (S. pedunculata, S. cordata, and S. cU a) are found in the uos forest zone in mid to upper E and are geographically isolated from each other. Scalesia affinis, the only species with consistently radiate capitula, has the widest distribution and occupies the widest range of habitats; occurring most commonly in the arid zone, it ranges from coastal to lower parts of the moist forest zone. All remaining species are found in the low elevation (littoral, arid, and dry forest zones) (Table 2). Due to overlap of preferences, there appears to be little habitat differentiation among these species. Only the rare species S. crocheri and S. retroflexa are known only from littoral sites. Other species (e.g., S. atractyloides, Mauchamp et al. 1998) are restricted to cliffs due to grazing by feral goats but historically ranged over more littoral and arid sites. DISCUSSION Phylogenetic relationships.—The present ttl hyly of Scalesia, Simsia, and Helianthus. Although molecular data (Schillling et al. 1994) ap ait Panpovoliis as monophyletic, the present data are equivocal in that regard, in some cases placing Scalesia, Simsia, Helianthus, and associated Viguiera species within a paraphyletic Pappobolus. Morphological data do not resolve the sister-group relationships of Scalesia, leaving open the possibility that Scalesia is sister to a group of Viguiera species or that Scalesia arose more or less simultaneously with Simsia, Helianthus, and Pappobolus (with any associated Viguiera species). As Viguiera sect. Maculatae is basal among the derived Helianthinae based on DNA restriction site and ITS data (Schilling & Jansen 1989; Schilling et al. 1994; Schilling & Panero, 1994, 1996, 2002), the sister-group placement of Scalesia and V. adenophylla in some results is due possibly to experimental error in coding or identification. Furthermore, Schilling and Panero’s (1996) molecular analysis suggests that Tithonia Desf. ex Gmelin, Viguiera subg. Amphilepis S.F.Blake, and V. sect. Paradosa S.F Blake are closer to Pappobolus than is Simsia and should be considered as potential sister groups of Scalesia. Within Scalesia there is general support for the lobe-leaved, arboreous, elongate-bracted, and divisa- gordilloi clades. Because the arboreous clade did not receive support from a minority of analyses, it is inter- esting that Eliasson (1974) considered S. pedunculata to have developed arborescence convergently with S. cordata and S. microcephala. If all variant trees based on the various analyses preformed are considered, the only clades receiving total support are the lobe-leaved clade and a terminal clade of S. atractyloides and S. 50 km Galápagos Islands 50 mi baurii m Marchena Genovesa atractyloides LE He 90° Bartolomé |. stewartii Faá Española Floreana E) (Santa María) Fic. 3. Distributi f Scalesi pecies, esti if literat including histori lly | Fine stippli Scalesia affinis, course stippling lnha.l d rlad hatehi h lad tical Ii l " a sta Gd dart Ai HHA clad Ed Inlet lobe g , Vertical lines — elon gate , g ; g gray = remaining species. stewartii, more in line with Eliasson's conclusions. Within the lobe-leaved clade, there is total support for S. baurii and S. incisa as a clade, but only partial support for S. helleri + S. retroflexa. In this case, these clades and all remaining species would radiate from a basal polytomy. If, indeed, Scalesia is an example of radia- tion by the rapid dispersal of founder populations from an initial colonizer, an unresolved basal polytomy may portray more accurately the history of Scalesia than any of the less-supported but more-resolved trees. Homoplasy.—Obviously, the degree of resolution of the particular cladogram examined will affect the level of homoplasy among taxa. Because the branch-and-bound analysis resulted in 13 equally par- simonious well-resolved trees and the comparison of all analyses suggest a minimally resolved polytomy within Scalesia, the level of homoplasy discussed is based on the branch-and-bound consensus tree, which is intermediate in resolution (Fig. 2). Scalesia itself is delimited by five synapomorphies and 11 homoplastic apomorphies (two among Scalesia species, five with external taxa, and four occurring both inside and outside Scalesia). Of the 15 species and 6 clades in Scalesia, only 9 are delimited by unique apomorphies. Of the 90 total character-states apomorphic for clades and species, 14 (16%) are unique, 42 (46%) are homoplastic only within Scalesia, 9 (10%) are homoplastic only between Scalesia and external taxa, and 25 (28%) are homoplastic between Scalesia taxa as well as with external taxa. It will be of inter- est to see the level of morphological homoplasy on DNA-sequence trees when these become available. 186 tani i Texas 3( Tase 3. CI ization of habi f Scalesia species, |f hel Moist Forest Littoral Arid Dry Forest Volcanic Lava Fissured e Zone Soil Gravel lava cordata x x microcephala — x x x pedunculata x x x affinis x X X x x x x villosa X X x x atractyloides X X ? X tewartii X x x x incisa ? x X bautii ? x x ? x retroflexa x x helleri x x x x gordilloi x X X X divisa X X x X aspera X X x x x crockeri X x x The characters that (at least some states of which) are not homoplastic in Scalesia include tree habit, leaf outline, leaf marginal lobing, dense villous hairs on abaxial leaf and phyllary surfaces, petiole shape, phyllary shape, ray orientation, palea segment shape and orientation, disk corolla color, disk corolla shape, anther appendage color, achene pubescence, and pappus development. However, all other characters and some states of the above are homoplastic. Some interesting examples include: 1) the presence of villous hairs in S. villosa and Pappobolus; 2) more or less solitary capitula of most species of Scalesia and Viguiera gram- matoglossa and V. stenoloba; 3) multiple changes in size of capitula in Scalesia; 4) phyllary shape in S. crockeri and Helianthus; 5) palea length in S. microcephala, the elongate bracted clade, and Pappobolus; 6) glabrous paleae in S. aspera, S. baurii, S. incisa, S. microcephala, and Pappobolus; 7) length of the central lobe of the paleae in S. affinis, S. crocheri, S. microcephala, S. villosa; 8) disk corolla tube length in S. affinis, S. baurii, S. stewartii, the lobe-leaved clade, and Viguiera adenophylla; and 9) glabrous disk corolla tubes in S. affinis, S. aspera, S. villosa, the arboreous clade, and Simsia. Many of these characters are associated with the palea and corolla structure. cu to Plovanich and a idi us CODI associated with reproductive in the Helianth tion pressures. Whether this will be true in Scalesia remains to be TENT using molecular data BON In regard to the presence of rays in certain species of Scalesia, Eliasson (1974) concluded that rays were lost in the ancestor of Scalesia, regained as scattered bilabiate disk corollas in the lobed-leaved species, and regained as nearly typical rays in S. affinis. His hypothesis is supported by the results presented here. If the affinis-type rays are the end of a character transformation involving the bilabiate disk corollas or are a reversal to true rays, then this constitutes an additional homoplastic trait. Presumably rays increase insect pollination and should be selected for on islands as the insect fauna diversifies, as suggested by the wider distribution of S. affinis. However, the addition of artificial rays to S. pedunculata did not increase its fitness (Nielsen et al. 2002). Therefore it is not clear that this character has high adaptive value in Scalesia. Despite the species and clades of Scalesia being delimited primarily by unique combinations of ho- moplastic character states as opposed to unique sud shee E species all jaan to be distinct. Moreover, the full data set suggests that there is a real lack of morpl phy within the continental genera because many species groups and es are ase defined 2i by unique combina- tions of homoplastic characters states, not unique states. Tucteibution i relation to nbl 4 th 11 1 R th 14 islands in the ar- chipelago are in the southeast and the youngest in the west and northwest, correlation of geology with the cla- DI Ll le J Dhul " J 224i £Cralaci 187 dograms is not straightforward. Ifthe po diverged from east to west, the basal split should produce a group of eastern species with the west td E However, the main split is between lowland and upland species. This may suggest that the lowland species diverged after the older islands from San Cristóbal west to Santiago were already in puce and the lineage ancestors were able to disperse among islands easily. Among the upland species, Scalesi hala and S ona (basal in some ie occur on pue youngest E islands. Presumeably, Scalesi 1 lata was already distinct and dispersed on t d founded populations on the new ean that later formed Isabela to originate the two other arboreous species. Species of Scalesia are characterized by nearly allopatric distributions in similar habitats (12 spp. in arid communities, 3 spp. in upland moist communities) within the archipelago. The only synapomorphy correlated with the origin of the upland-habitat lineage is the tree growth habit, though four homoplastic characters also accompany the habitat (loss of leaf adaxial strigosity, moderately sized capitula, glabrous paleae, and corolla tube glabrous). The development of arborescence in a moist habitat under reduced competition is easy to understand (see Itow 1995; Hamann 2001), but further study is needed to determine if the other apomorphies are correlated with reproductive ecology. Eight homoplastic o but no as are aa ith origin of the li in the lowland habitat (pul gins, solitary P Blum phyllaries, pe deeply divided into elliptic es en lacking awns). Un] dditi l factors, such idity, ion content, pollinators, or dispersers, etc. significantly differentiate among both lowland and upland habitats, the species within these two elevational zones appear to occupy nearly the same range of habitats. For example, Scalesia villosa is the only Scalesia species having a dense covering of villous hairs, which presumably functions as a protection from high solar radiation. But several species occupying such habitats are not villous even though villous hairs occur in the related genera. Although S. helleri bears pinnatifid leaves as an autapomorphy, the nearly parapatric and perhaps sister species is distinguished only by two homoplastic apomorphies. Only a single autapomorphy (fully winged petioles) and three homoplastic traits distinguish S. crockeri; nearly parapatric with it is S. aspera, which is differentiated by only four | plastic traits. Furthermore, diversification among the lowland species has resulted in some sharing apomorphies with some or all of the upland species and vice versa. Scalesia microcephala of mesic forests shares two palea dud states (see above) with S. villosa but not with its close congeners in the mesic zone. Thus, g adaptation of distinguishing features of these species may prove to be challenging. Speciation Patterns.—Because the sister-group to Scalesia remains obscure, comparison of speciation amounts among clades is not possible. It is apparent that this situation will not change until multiple con- gruent lines of molecular evidence resolve the relationships of the infrageneric groups of Viguiera and other genera in the derived Helianthi However, if a DNA sequence in which there is variation among species of Scalesia can be found and analyzed, then, at least speciation rates within Scalesia should be ipai d Sampling recommendations.— Given the above situation, it is clear that sampling for f ] hould include, in addition to the present taxa, at least e. of Tithonia; Viguiera sube. AD sect Macias, and sect. Parad d other seg g of the derived Helianthinae. When congruent lines of molecular evidence point t to one of these lineages as sister to a a up ene of species should be attempted t to determine whether the wl g E ter to Scalesia. CONCLUSION The present study provides a large morphological data set for comparison with molecular phylogenies of Scalesia and close relatives when the molecular data become available. The results confirm that additional taxa and DNA sequences must be sampled to resolve the interg d internal relationships of Scalesia. Furthermore, divergence of Scalesia from its origin to terminal speciations is characterized by combinations of homoplastic apomorphies. Likewise divergence and EDS geo graphy e to Me poorly correlated. The seeming uniformity within habitat zones, though, appear: plasy associated 1 1 . 1 Lc with divergence in Scalesia. Determining the degree to "iud } logical a Í are adaptive should clarify the process of speciation in this and other mana ae {23} K={24} {14} 1={15} J {12} G={13} H 5 ={03} D={04} E-(05) F= 4 123456789012345678901234567890123456789012345678901234567890123 00100104001100010010011000000000001210211A24N030110120011000007? f(111 RNA O Se een A 3 J APPENDIX (2345) Y2(01234]. 2 024) T=(1 24} U={234} V={01 23} X {023}S {35} N={01 2} P={013} Q={015} R Data matrix. Missing, unknown, or inapplicable=? Sc_helleri Sc atractyloides Taxa M Egl 20001100F01100100001F2002001B001100100A1014A10101102A0011FA22200 Si_fruticulosa Si_santoros. 1000004A10F00200AB0011A0100A101A101EA102A0011FA2NB0O0 U01011CA1000000004A1F2002001C0011001004A101A1010A102A0011FA2000? Silagasciformis 48 1011CYF0OAA00000411FK0A2001C001100100A101AG010A10220011FA2BB00 i i 00 ?? ? | 4 A villasenorii sanguinea . eurylelpis Si chaseae 1:2 1:0 2:0:0:010 001 8.970 1:0:09A.1-0-1-& 1:0 41 2:91 92 500 11 2 A9 $2 X8 100000411 1 CNFO 1 mbeyana Si. do 4B101 2 — — . “Accessed 18 Mar 2009. ni ll Ic J BL J a: J I PEN a £C ] 1 191 PLOVANICH, A.E. AND J.L. Panero. 2004. A phylogeny of the ITS and ETS for Montanoa (Asteraceae: Heliantheae). Molec. Phylogen. Evol. 31:815-821. SANDERS, R.W., T.F. Stuessy, C. MARTICORENA, AND M. SitvA. 1987. Phytogeography and evolution of Dendroseris and Robinsonia, tree-Compositae of the Juan Fernandez Islands. Opera Bot. 92:195-215. SANG, T., D.J. CRAwFORD, T.F. STUESSY, AND M. SiL vA. 1995. ITS sequences and the phylogeny of the genus Robinsonia (Asteraceae). Syst. Bot. 20:55-64. ScHILLING, E.E. 1990. Taxonomic revision of Viguiera subg. Bahiopsis. Madroño. 37:149-170. SCHILUNG, E.E. 2006. Helianthus. In: Flora of North America Editorial Committee, eds. Flora of North America north of Mexico. Oxford University Press, New York and Oxford. 21:141. SCHILLING, E.E. AND R.K, JANSEN. 1989. Restriction fragment analysis of chloroplast DNA and the systematics of Viguiera and related genera (Asteraceae: Helianthinae). Amer. J. Bot. 76:1769-1778. SCHILLING, E.E. AND J.L. Panero. 1996, Phylogenetic reticulation in subtribe Helianthinae. Amer. J. Bot. 83:939-948. SCHILLING, E.E. AND J.L. Panero. 2002. A revised classification of subtribe Helianthinae (Asteraceae: Heliantheae). |. Basal lineages. Bot. J. Linn. Soc. 140:65-76. SCHILLING, E.E., F.B. Da Costa, N.P. Lores, AND PJ. Heise. 2000. Brazilian species of Viguiera (Asteraceae) exhibit low levels of ITS sequence variation. Edinburgh J. Bot. 57:323-332. SCHILLING, E.E., J.E. Panero, AND U.H. ELtasson. 1994. Evidence from Chloroplast DNA restriction site analysis on the relationships of Scalesia (Asteraceae: Heliantheae). Amer. J. Bot. 81:248-254. SPOONER, D.M. 1990. Systematics of Simsia (Compositae-Heliantheae). Syst. Bot. Monogr. 30. SPRING, O, N. He AND U. Eliasson. 1999. Chemosystematic studies on the genus Scalesia (Asteraceae). Biochem. Syst. Ecol. 27:277-288. SPRING, O, N. He, AND B. Voter. 1997. Sesquiterpene lactones and flavanones in Scalesia species. Phytochemistry 46:1369-1373. SWOFFORD, D.L. 1998. PAUP*. Phylogenetic lysis using parsi y (“and Other Methods). Version 4.0610. Sinauer Associates, Sunderland, MA. Timme, R.E., B.B. Simpson, AND C.R. LinDER 2007. High-resolution phylogeny for Helianthus (Asteraceae) using the 185-265 ribosomal DNA external transcribed spacer. Amer. J. Bot. 94:1837-1852. L H Ip hl botit, ET | || £ 4 ¿OU Mal OF 192 BOOK REVIEW Peccy KocuaNorr. 2009. You Can Be a Nature Detective. (ISBN 978-087842556-3, pbk.). Mountain Press Publishing Company, PO. Box 2399, Missoula, Montana 59806, U.S.A. (Orders: www.mountain-press. com, info@mtnpress.com, 1-800-234-5308). $14.00, 63 pp., 9" x 8 3/8". You Can Be a N Detective is a great little book for parents, teachers, grandparents, and any others interested in spending time with ue out Hin doors. It A N with an attention-grabbing opener (“Hmmm... there’s spit all over the grass. Who could have done it?”) p A g ft) ysteri ly found in limited geographi regio ] y winter occurs), al fully engaging and taining. Through the process of solving nature ? Will tl hi i butterfly or moth? Whats that fuzzy stuff on the id (Why are the leaves changing colors? Will that Ep crawly g ground the E has UA dud 12 different ] ] be further i igated by a curious child. A detailed explanation I d with beautiful watercolor die e at iud end a ilie book are some gd *Fun Things to Do in Nature" (each related to one of ip ee n such as na Sue M. owl pee ll O: E I see You Can Bea one DUE more as a vehicle kids are saying, "I'm Ty ; keat dult to spend quality time in nature e with a PA It's a ready-made, go-to source for those times when the d be without.-—Palmer Byerley, Elementary Science Teacher, Crowley, Texas, U.S.A. » 1 bored!, p g Ms. Kochanoff's book is part field guid L It offers good inf ti | variety of pl d ls with well-executed lor ill i HE contrary to the title, 1 tl I ll wi ] t 1 1 A Te 1 1 1 ' P1 1 i dad E 11 A J o T D O " : B 1 FERE d 1 £214 id Ag A } lac, j t Pam © ouo. 6-2 b IUS as a “nature detec Chamberlain, ane Research Institute of Texas, Fort Worth, Texas 76102- 4025, US. A. J. Bot. Res. Inst. Texas 3(1): 192. 2009 INTRODUCED LESSER CELANDINE (RANUNCULUS FICARIA, RANUNCULACEAE) AND ITS PUTATIVE SUBSPECIES IN THE UNITED STATES: A MORPHOMETRIC ANALYSIS Angela R. Post' Alexander Krings go of Horticultural Science Herbarium, Department of de ui North Carolina State University North Carolina State University Raleigh, men Carolina 27695-7609, U.S.A. Raleigh, North Carolina rosca 12, U.S.A. Alexander Kringsncsu.edu Wade A. Wall Joseph C. Neal ie of Plant Biology Eo of Horticultural Science North Carolina State University North Carolina State University Raleigh, M. Carolina 27695-7612, U.S.A. Raleigh, o Carolina 27695-7609, U.S.A. ABSTRACT Ranunculus ficaria is native to Europe, but was introduced to the United States by at least the 19* century as a garden ornamental. Followi ing) x een the m epea un aber Because, in contrast to European floristic treatments, previous North Ámerican I B , Our O for this ÓN to men if and nii 1 1 E) 11 i P 2i P Ls R a z n T IT n ds e ES e J I to th b if foll 1 in Europe, and lyze the distributi habitats d rat f d of each entity. To meet our objectives, we conducted a Borphameue pn based on id of 319 specimiens from forty-seven herbaria, The ecompined dina = I Y e diverged T ld limited f overlap of operational ic units (OTUs) | Principal aoi nates Analysis and Principal Components AROS results, llas i pletel 1OTU i lust d cl tree analyses. Based on our current all fi b i in the United States. They are best adapted to moist sites, y y P exhibit overlapping distributions, and appear to be spreading at DUM rates. RESUMEN Ranunculus ficaria es nativa de re pero fue introducida en los Estados Unidos como ornamental en el siglo XIX. Después de la ultivo. Forau en ee con los tratamientos floristicos europeos, los norteamericanos no b if stros objetivos en este estudio fueron determinar si existen y cuantas enti- la esneci introducción, p p imient habían enfatizado en el recon dades O E de R. caña se dan en los E Unidos, evaluar en que medida Europa E analizar las distribuciones, hábitats, y tasas de ode de cada a Para 1 n epto P ME O lograr t bjetivos, realizamos un análi basado en E estudio de 19 esp ta y siete herbarios. binados indi ] ia de ci idad l I i ] Europa. Si E E bespeci lí diverg i pl dria est una aa ae de esolapamiemto en m Aa t 5 OTUSs) n Análisis de C J r I Eu OTU l denad álisis de cl y árboles de clasificació "Pene en nuestro conocimiento [ ] i 1 peci án en los Estados Unidos. Están j 1 I d l g hú muestran distri d pa | 1 a AN 4 buciones que se Solar yI 1 j INTRODUCTION Ranunculus ficaria L. (Ranunculaceae) is native to Europe (Tutin 1964; Taylor & Markham 1978; Sell 1994; Whittemore 1997), but was introduced to the United States (U.S.) through the garden ornamental trade for its showy flowers (Bailey 1935). It was collected with certainty in the U.S. in 1867 (Philadelphia County, Pennsylvania, Burke s.n., PH) and required 141 years to “spread” from Philadelphia, Pennsylvania to Fort Worth, Texas (Nesom FW08-1, BRIT, MO, NCSC, NCU, TEX)—the southernmost extent of the current known distribution (Nesom 2008). 3) "Present address: Der f Plant Biology, Cornell University, Ithaca, NY 14853, arp232@cornell.edu J. Bot. Res. Inst. Texas 3(1): 193 — 209. 2009 194 Journal of tl tanical Insti Texas 3( In Europe, five subspecies of Ranunculus ficaria are recognized (Tutin 1964; Sell 1994), whose “ecol- ogy and [...] distribution appear to overlap, but tend to be different" (Sell 1994). Flora Europaea currently recognizes: (1) Ranunculus ficaria subsp. ficaria from western Europe, eastward to southern Italy, (2) subsp. bulbilifer Lambinon from northern and central Europe, extending to Spain, Albania and east-central Russia, (3) subsp. calthifolius (Reichenb.) Arcangeli from south-central and eastern Europe, (4) subsp. chrysocephalus PD. Sell from Greece and Crete, and (5) subsp. ficariiformis (FW. Schwartz) Rouy & Fouc from southern Europe. Plants are known to be diploid (2n=16), triploid (2n=24), or tetraploid (2n=32). Diploids have been referred to subspecies calthifolius and ficaria, and tetraploids to bulbilifer, chrysocephalus, and ficariiformis (Greilhuber 1974; Sell 1994). Triploids may represent putative hybrids and have apparently been collected from widely separated localities in Europe (Marchant & Brighton 1974; Sell 1994). Sell (1994) suggested that a large proportion of the pollen of triploids, as well as the tetraploid subsp. bulbilifer, is non-viable and few t. Pollen from diploids and the large-flowered tetraploids (i.e., chrysocephalus and ficariiformis) is apparently viable and many achenes are produced (Sell 1994). Two subspecies are capable of producing bulbils in their leaf axils: bulbilifer and ficariiformis. Subspecies bulbilifer tends to exhibit globose bulbils and subsp. ficariiformis produces ellipsoid bulbils (Sell 1994. In contrast to European accounts, North American floristic treatments either recognized no subspe- cific taxa in R. ficaria (Fernald 1950; Gleason 1952; Gleason & Cronquist 1963; Whittemore 1997) or only variety bulbifera Marsden-Jones (=subsp. bulbilifer Lambinon; Magee & Ahles 1999). However, the recent P of entities poop referable to subsp. ficariiformis in North Carolina (Krings et al. 2005) and Texas (I 2008; bsp. bulbilifer, but with ellipsoid bulbils and flower dimensions within the range of ficariiformis) caused us to question whether additional subspecies may be present in North America that have not been previously recorded and if so, whether these differed in their distributions, habitats, and rates of spread. Because prior North American treatments did not emphasize subspecific recognition and as subspecies are not uniformly accepted, our objectives were to: (1) determine if and how many morpho- logically recognizable entities within R. ficaria occur in the United States, (2) evaluate to what extent such entities correspond to the subspecific concepts followed in Europe (based on the work of Sell 1994), and (3) analyze the distributions, habitats, and rates of spread of each entity. METHODS Distribution and habitat information led from 319 herbari j , requested from the follow- ing forty-seven herbaria based on previous literature reports (Benson 1942; Bell 1945; Gleason & Cronquist 1991; Whittemore 1997): A, AUA, BALT, BH, BKL, BRIT, CONN, CU, DOV, F, FLAS, GA, GH, HNH, ILLS, KE, LGO, LSU, MARY, MASS, MICH, MISS, MO, MOR, MSC, MT, MU, NA, NCSC, NCU, NHA, NY, OS, OSC, PH, POM, TENN, TEX, UNA, US, USF, USCH, VDB, VPI, WTU, WVA, Y (Appendix A). Herbarium label data recorded for each specimen included collector name, collector number, date, habitat, and county and state of collection. Specimens Hone cando were excluded from the study. Collection localities were classified into g nine habitat classes: (1) adjacent to a water source, (2) disturbed areas, (3) dry woods, (4) fields, (5) Bostieuib ss (6) lawns, (7) lowlands, (8) moist areas, and (9) roadsides. A specimen was classified as adjacent to a water source if it was collected along the banks ofa river, stream, or pond. Moist areas were defined as moist or alluvial woods, swamp or bog areas, and other moist shade. The horticultural class was defined as being collected in a nursery or garden under cultivation. Lowlands were defined as low or depressed areas where moisture level was not mentioned on the label. Fields were defined as any open grassy area not maintained as a lawn and in full sun such as pastures and meadows. To determine if and how many morphologically recognizable entities within R. ficaria occur in the United States, each herbarium sheet was treated as an operational taxonomic unit (OTU) for data capture and pertinent subsequent analyses. To evaluate to what extent such entities correspond to the subspecific concepts followed in Europe, each specimen was determined to subspecies following the key constructed by Sell (1994): Post et al., A pl tri lysis of R lus ficari 195 1. Leaf blades to 8 x 9 cm; petioles io 28 cm; MEMES to ne mm diam; achenes 5.0 x 3.5 mm. 2. Stems rather robust, but st bulbil eaf axils after flowering subsp. ficariiformis 2. Stems robust and erect; without bulbils i in leaf axils P flowering subsp. chrysocephalus 1. Leaf blades to4 x : cm; petioles to 15 cm; flowers to 40 mm diam; achenes to 3.5 x 2.2 mm. . Leaves | the base with few on short stems subsp. calthifolius 3. Leaves less crowded at base and more numerous an the ae stems. 4. Buibils not present in leaf axils atte) fl g; V veloped subsp. ficaria 4. Bulbils present in leaf axils after f ing; acl p rly d developed subsp. bulbilifer Sell (1994) recognized the difficulty in identifying Ranunculus ficaria to the subspecific level, noting that specimens should be examined throughout the growing season for positive identifications. We agree with him and others (Whittemore 1997; Nesom 2008) that identification can be challenging and recognize the impact identifications have on analysis results. However, we feel reasonably confident in our subspecies assignments due in part to the quality of specimens, which facilitated taxon assignment, as well as our own field observations and phenological analyses. Of the 319 herbarium specimens examined 232 had at least a month and year date on the label and 9096 of these were collected mid-April through June. Another 596 were collected in the last week of March and the remaining 596 were collected in January through mid-March. Based on date of collection, the majority of specimens examined in this study were collected late enough in the spring that they would exhibit bulbils if uA were i nad capable of producing them. Using a digital caliper, the following mort were taken from each OTU: (1) leaf length from up to ten leaves, Q) leaf width from s to ten leaves, (3) petiole length from up to ten leaves (using same leaves measured for length and width), (4) petal length from up to ten flowers, (5) petal width from up to ten flowers, (6) achene length of all achenes present, (7) achene width for all achenes present. The presence or absence of bulbils was also recorded. Quantitative and qualitative data were studied jointly and separately. Statistic analyses, including ANOVAs and post-hoc tests (Tukey's HSD), were carried out in the statistics package R (Ihaka & Gentleman 1996; R Foundation for Statistical Computing 2008). Prior to multivariate analysis, we tested all quantita- tive univariate variables using the Shapiro-Willks normality test and subsequently log, transformed them to minimize the influence of allometry on the results (Dufréne et al. 1991; Almeida-Pinheiro de Carvalho et al. 2004; Pimentel et al. 2007). Gower's dissimilarity coefficient for mixed data was used to quantify resemblances between OTUs (Gower 1971). The relationships between OTUs were subsequently explored with both hierarchical agglomerative cluster analyses and principal coordinates analyses (PCoA) using the complete set of characters. Three different sorting algorithms were used to help distinguish between data- dependent and potential OR Rd MEN ct in results, following Dickinson & Phipps (1985) and Pimentel et al. (2007): single linkage, g „and average linkage (UPGMA; Sneath & Sokal 1973). Quantitative characters were also MA separately using Principal Components Analysis (PCA). PCA is an objective, correlation-based technique that allows the variance in quantitative characters to be considered simultaneously and the subsequent visualization of dispersion patterns in a number of dimen- sions that explain the greatest amount of variance (Sargent et al. 2004; Joly & Bruneau 2007; Pimentel et al. 2007). A Kaiser-Meyer-Olkin (KMO) test was performed prior to the PCA to assess the suitability of the data for multivariate analysis (see also Almeida-Pinheiro B ee et a dil Pimentel et al. 2007). Classification trees were acci to help identify sy I g hat could reliably separate the OTU group o the fi tati ] j Sell (1994). Classification trees divide datasets vit pre- assigned aie membership into Weee homogeneous subsets in tree-like fashion based on the included hological traits. Recovery proceeds until ps obtained are pure or until a dividing threshold is ed (Joly & Bruneau 2007). For the n tree, all morphological characters were included and quantitative data was not log transformed prior to analysis. The distribution of R. ficaria was mapped based on herbarium specimens and using ArcGIS 9.2 (ESRI 2004). The rate of spread for each subspecies was determined based on the number of counties each was present in during each decade from 1860 to the present. These data were analyzed using proc mixed in SAS 196 Journal of ical Institute of Texas 3( 9.1.3 with a critical value of 0.05 (SAS Institute 2002-2005). Note that we use “spread” in a broad sense, as we suspect that not all new county records are the result of physical movement of propagules from estab- lished parental plants, but that many may reflect novel introduction events. RESULTS Descriptive statistics Table 1 provides non-transformed means and standard deviations for sured for each group of OTUs assignable to one of five putative subspecies sensu Sell (1994) and summarizes the results of the one-way ANOVAs and subsequent post-hoc tests (Tukey's HSD) on log,, transformed data (see superscripts). Figure 1 exhibits box-plots showing the distribution of non-transformed quantitative measurements taken by OTU group. Significant differences in the means for each of the five groups of OTUs assigned to the subspecies sensu Sell (1994) for all seven characters were found (Table 1). Post-hoc tests (Tukey's HSD) to determine which sample means differed from which others showed that means of leaf length and petal width of the OTU group assignable to subsp. chrysocephalus differed significantly (p«0.05) from the respective means of the groups of OTUs assignable to the four other subspecies (Table 1). OTUs assignable to the diploid subsp. calthifolius and ficaria differed significantly (p«0.05) in mean leaf length, mean leaf width, and mean petiole length (Table 1). They did not differ significantly in mean petal length, mean petal width, mean achene length, or mean achene width. OTUS assignable to the tetraploid subsp. bulbilifer, chrysocephalus, and ficariiformis differed significantly (p«0.05) from one another in mean leaf length and mean petal width. Subspecies bulbilifer differed significantly (p«0.05) from both subsp. chrysocephalus and ficariiformis in mean leaf width, mean petiole length, and mean petal length. Subspecies chrysocephalus differed significantly (p<0.05) from subsp. bulbilifer in mean achene length and width, but subsp. ficariiformis differed neither from subsp. chrysocephalus nor bulbilifer in these characters (Table 1). 1 titati 1 f Cluster analyses In all three cluster analyses—average, complete, and single linkage—OTUs were resolved into two large divisions, these corresponding to (1) the bulbil bearing taxa: subsp. bulbilifer and subsp. ficariiformis sensu Sell (1994) and, (2) the non-bulbil bearing taxa: subsp. calthifolius, subsp. chrysocephalus, and subsp. ficaria sensu Sell (1994; Fig. 2). Within these two divisions, the topologies resulting from the three different algorithms differed notably only for those resulting from single linkage. Within the bulbilifer/ficariiformis division, average linkage recovered a cluster predominantly composed of OTUs referable to subsp. ficarii- formis sister to a larger cluster of OTUs predominantly e to denn bulbilifer Both recovered clusters contained OTUs referable to either subspecies. Th 1 three clusters within the bulbilifer/ficariiformis division—one of OTUs predominan ble to ficariiformis, nested within two composed predominantly of OTUs referable to subsp. bulbilifer. Single linkage similarly recovered a cluster of OTUs referable to subsp. ficariiformis nested within OTUs referable to subsp. bulbilifer. In the calthifolius/ chrysocephalus/ficaria division, both average and complete linkage analyses recovered a cluster of OTUs predominantly referable to subsp. ficaría nested within clusters of OTUs predominantly referable to subsp. calthifolius. OTUs referable to subsp. ficaria did not emerge in a distinct cluster in the single linkage analy- sis, but rather were interspersed throughout those referable to subsp. calthifolius. OTUs referable to subsp. chrysocephalus emerged interspersed in grades of OTUs referable to subsp. calthifolius and subsp. ficaria in a sister position to the rest of the division in all three analyses. PCoA and PCA Consistent with cluster analysis results, two non-overlapping clusters of OTUs were recovered in the PCoA corresponding to (1) the bulbil bearing taxa: subsp. bulbilifer and subsp. ficariiformis sensu Sell (1994) and Q) the non-bulbil bearing taxa: subsp. calthifolius, subsp. chrysocephalus, and subsp. ficaria sensu Sell (1994; Fig. 3A). Within both of these clusters, cohesiveness was exhibited by each group of OTUs referable to one of the five subspecies sensu Sell (1994), although each group overlapped with another to some degree. Consistent with expectations for infraspecific entities, distinct but overlapping clusters of OTUs were Post et al., A pl tri lysis of R lus ficari 197 Toate 1.1 No £ J ral Tei | n . i : £ 1 e E. ) y H | COMO | Mt, sacan -values for one-way ANOVAs fto, formed data provided in ulti lumns, Within a row, ith d perscri igni V t 310 7 2f bulbilifer calthifolius chrysocephalus ficaria ficariiformis F p-value Leaf length 2.18 24 (0.41; 82) 1.93?(0.46; 96) 3.545(1.03;6) | 242? (0.39; 37) 2.97 * (0.49; 12) 9.79 «0.001 3.80*2(0.53;12) 10.56 «0.001 — (cm) ani 2.83? (0.53;82) 2.38^(0.54;97) 4.33%(1.15;6) 2.95 %4 (0.48; 37 petole ln 10.41 ? (2.79; 82) 6.55*(1.62; 97) 16.21 (1.44;6) 12.198% (2.16; 37) 14.75 %4 (3.85; 12) 2844 <0.001 mo no 10,23 9(1.67; 75) 11.11 #(1.91;97) 17.43 ©4 (3.79; 6) 12.11 © (1.95;37) 13.91% (1.95,11) 16.89 «0.001 (mm) ous. p 11) 23.62 «0.001 ~ Petal width 3.57 ? (0.91; 75) 4.53? (1.05; 97) 7.38€ (0.45;6) 4,83*%(1.09; 37 (mm Achene length 2.95 ? (0.65; 53) 3.04? (0.65;28) 4.33 £ (0.69;5) 3.27 = (0.41; 6) 3.50 ** (0.90; 7) 530 «0.001 mm Achene width 4.35 è (0.90; 53) 4.262(1.00;28) 622°°(0.77;5) 458**(1.12;6) 4.70 ** (1.31; 7) 383 «001 DN £11 exhibited in the PCA comprising all OTUs when symbol coded for a priori the infraspecific concepts of Sell (1994) (Fig. 3B). Among these, OTUs eee to subsp. bulbil ifer exhibited the most cohesive and least diffuse cluster. In this analysis, 81% of the variation is explained by the first two axes. PCI is positively correlated most strongly with petiole length, leaf width, and leaf length, whereas PC2 is positively correlated most strongly with petal width and petal length (Table 2). A separate analysis of only OTUs referred to the two diploid taxa subsp. calthifolius and ficaria, resulted in two very well-defined clusters with minor overlap (Fig. 3C; Table 3). In this analysis, 73 % of variation is explained by the first two axes. PC] is positively correlated most strongly with petiole length, leaf length, and leaf width, whereas PC2 is positively correlated most strongly with petal width and petal length (Table 3). An analysis of only OTUs referred to the tetraploid taxa—subsp. bulbilifer, chrysocephalus, and ficariiformis— showed evident clustering, but with greater overlap among the three a priori defined subspecies (Fig. 3D; Table 4). In this analysis, 76% of the variation is explained by the first two axes. OTUs defined a priori as subsp. ficariiformis occupied a central coordinate space in the tetraploid analysis, flanked along the primary axis by subsp. chrysocephalus to the left and bulbilifer to the right. PC1 is negatively correlated most strongly with petal width, leaf length, and leaf width, whereas PC2 is positively correlated most strongly with petiole length (Table 4). Among the five subspecies of is ee eee i Sell (1994), only the tetraploid subsp. bulbilifer and ficariiformis are known to prod A y f only OTUs with bulbils, showed two rather cohesive clusters with imed overlap on to a priori assignment to these two subspecies sensu Sell (1994; Fig. 3E; Table 5). In this analysis, 69% of the variation is explained by the first two axes. PC1 is negatively correlated most strongly with petal width, whereas PC2 is positively correlated most strongly with petal width and petal length (Table 5). A separate analysis of OTUs without bulbils (Fig. 3f; Table 6), showed three rather cohesive clusters with limited overlap, corresponding to a priori assigned subspecies. In this analysis, 7996 of the variation is explained by the first two axes. PCL is positively correlated most strongly with petiole length, leaf width, and leaf length, whereas PC2 is positively correlated most strongly with petal width and petal length (Table 6). Classification tree Classification tree analysis showed that 9596 or greater of the OTUs we aa to subsp. calthifolius, subsp. ficaria, and subsp. bulbilifer using Sell (1994) could be placed i Į g g groups (Fig. 4) 198 Leaf Length Leaf Width LJ ir ne = e] i i H H $7 i i , H © H i d H etc zu A — a] T 5 H T É H i = i EXE i i i H ! i T H || E ELE T | [1 : H H H —— —— a m 1 ci T H | H H H 2 i 3 i = | i i — i i T T T T T T T T T T bulbilifer ^ calthifolius chrysocephalus ficaria ficariiformis bulblilfer ^ calthifolius chrysocephalus ficaria ficarlitormis Petiole Length Petal Length a N € t : H o i H : i [| ° 24 y = i H : -— — H H e 1 1 T noct i — : = i | i : i i i | i | ee H H e —— : T E AA ee i o T H i H i 7 i i i i as TEE T T T T^ : ] ' : bulbilifer calthifolius chrysocephalus ficaria ficariiformis bulbilifer calthifolius chrysocephalus ficaria ficarliformis Petal Width Achene Length o H o H H am H H == | i o a H 1 nm > mE i ER o 1 o —— i o- == i H H H i t H q = : H H : Sel s i i i —— i h i 1 ! l ] i +- H : H 1 - E i i i ! 7 7 i " E i i i ole l a H i H i —— i m SRM "ENS T T T T T T T T T bulbilifer calthifolius chrysocephalus — flcarla ficarilformis bulbnifer calthifoliüs: chrysocephalus: , ficaria ficant Ormia ne Width o my o * H e- i j | i 1 H wn - H f <+- i : i F : ] 1 H —— H H a 1 H 1 i i T T T7 T T bulbilifer calthifolius chrysocephalus ficarla ficariitormis las. Ll . Pe eer £ " £ 1 aca L m i» D hie Éixnvia in tha IInitad Chat. t 2 t Law lant eh + which are in cm Post et al A Height 000 005 010 015 020 025 03 | Height Height T MUTO y | — — x— AS 199 ficaria (4); ff = subsp. ficar, llowed by “predom.” bu = subsp. bulbilifer (1); ca = subsp. calthifolius (2); ch = subsp. ch liformis (5). JULI. 200 Journal of the Bot IR h Institute of Texas 3( + 2 _| z = o + 2 e = a x X o ? A, 4,0 Ao, a OX S 7 Š $3 ew a A s ES UA Ky P N Ox 4 o c A^ NE o O+ 3 Aj $ era 7 M * $ A ius $93 oA A xS o $ T A a KARE ARTS a 2 PEE cR Ao Ti o 31 wa pil T s. SET a A ea ay o 8 M oo°o o Bu? o?o So o °? a A À GEA o 5 deg. 6 Bs, e o9 d A , A E A oo, o 9o N o å E o 9 0 2 o0 se A $c ^ A i T 1 1 T B T T T T T T T 02 04 ‘0 i4 02 415 10 05 00 05 10 15 PCoA axis 1 PCA 1 o | 2 o x e A A S64, A EE o 9 20° Aa A 24 A AA x o e A pá x X x e o a a $59 q A A p EN x < o 5 EI 8o ó 4 A A A x oo t ot 23 9 6 p La | AL AS A x > | o ow 9 € o 0 5 A AB a Ana wx et E o 9g, o ^ a "EN a AB X eer + 99 o o ?9 o j A A AMA AX Sak RA x x + + o o % A aA xXx T um o *. S " A A x 6 o o 9 a al C j ! l ' y T D T T T T T T -15 -1.0 -0.5 00 05 10 15 10 05 00 05 10 PCA 1 PCA1 o | E a _| o | in e o o = o o ? o o Y x > li e A A AAA o o o o A^ A A 5 " o o 99 ° " o A Br Sa Âa x a o 9 o o og 5 A + + ` od 9e 9 o e A x do d o o o00 o « & ah A + a? o o o $84 añ A x x o o 90 80 Bo ES 4 pr xXx o £ e$ o o A EY x "Xe Xx o 9 og oo T X An AAA XR i o es x E: -j e o oo o e | 2 [e] : y l y T T T T T T T E 9 05 0.0 05 10 F as ao os of 05 10 15 PCA1 PCA 1 Fic 2 n Je e 1 £ n. Ir J 4 A I fA ATIL- 1 D il Ir | A ly (B F); D NATI! ; f I z dij Laid n TII $ D: putati loid OTUs; E: bulbiliferous OTUs; F bulbilif OTUs. o = subsp. bulbilifer; A = subsp. calthifolius; + = subsp. chrysocephalus; x= subsp. ficaria; 9 = subsp. ficariiformis Post et al A Taste 2, Cl ID + VM | he IDA tall OTH 201 PCI PC PG Leaf length 71 0.06007699 -0.5807531 Leaf width 0.4659988 0.01441147 -0.4945180 Petiole length 0.6942403 -0.45147042 0.5562742 Petal length 0.2116039 0.44013398 0.1484652 Petal width 0.2415118 0.77371804 0,2944399 Table 3. Ct loadi | | PO: DCA nf tt lip "M PCI PQ PG Leaf length 0.4393161 0.1933816 -0.5527152 Leaf width 0.4473662 0.1682932 -0.4790156 Petiole length 0.7371736 -0.4948622 0.4585902 Petal length 0.1785614 0.4539478 0.2102479 Petal width 0.1776393 0.6952149 0.4588473 TABLE 4 f | 1 (PO 5 L NA £ +1 INT PCI PC PG Leaf length -0.4145495 0.2908487 -0.5513932 Leaf width -0.4139542 0.2858958 -0.4349598 Petiole length -0.4282971 0.5777908 0.6924294 Petal length -0.3705260 -0.3867720 0.1113257 Petal width -0.5797091 -0.5918072 0.1221624 TARIE 5 f 1 ]: | DIN f! DC A £ is AT) PC1 PQ PG Leaf length -0.4166727 -0.2045966 0.5826665 Leaf width -0.4395120 -0.2045104 0.4630089 Petiole length -0.4585246 0.6345035 -0.6193181 Petal length -0,3303427 0.3747980 -0.1413127 Petal width -0.5602160 0.6109399 -0.2063936 Taste 6. C] | (PC) in the PCA of th bulbiliferous OTUs. PC1 PQ PG Leaf length 0.4474063 0.1388268 -0.5661383 Leaf width 0.452295] 1139208 -0.4887299 Petiole length 0.6954902 -0.5497825 0.4608629 Petal length 0.2280402 0.4510623 0.2216459 Petal width 0.2440242 0.6797288 0.4232127 Seventy-five percent (N26) and 87.5% (N=7) of OTUs we referred respectively to subsp. chrysocephalus and subsp. ficariiformis using Sell (1994) could be placed into corresponding homogeneous groups. Á quarter of the OTUs (N22) assigned to subsp. chrysocephalus in the analysis were specimens we referred to subsp. ficaria using Sell (1994). The presence of bulbils separated the calthifolius/chrysocephalus/ficaria group from 202 Jour nal of t tani f Texas 3( Bulbils Present No Yes Petiole Petiole Petal Length | Petal Length 29.9 cm «13.5 mm Petal Width | Petal Width >6.9 m «6.9 mm calthifolius bulbilifer ficariiformis 73 (94.8%) bulbilifer 7 (87.5%) ficariiformis 0, PHP A QE 1 (1.3%) calthifolius 1 (12.5%) bulbilifer > 3 (3.9%) ficariiformis ficaria chrysocephalus 31 (96.9%) ficaria 6 (75%) chrysocephalus 1 (3.1%) calthifolius 2 (25%) ficaria is 4. Classification t ly is. Numi gi t node t inals indicat iori classificati Sell (1994) ). For example, the model as- cial to ñ Icaria by us. the dl ied Within the ud iole l liscrimi ] best between OTUs referable to subsp d those referable t onda and EM (oa Petal width discriminated best between euism ficaria a subsp. ola Petal length d best bet bsp. bulbilifer and subsp. ficariiformis. The shape of bulbils was not scored for the classification tree analysis as we were interested in seeing what additional vegetative character distinguished these putative taxa. DISCUSSION Subspecies recognition.—The combined results indicate the presence of five entities that can be reason- ably referred to the subspecies accepted by Sell (1994). If one accepts subspecies as incompletely diverged lineages, one would expect a limited amount of overlap of OTUs as seen in our PCoA and PCA results, as well as incompletely sorted OTUs as seen in our cluster and classification tree analyses (Rosen et al. 2007). The ANOVA results are also informative on this issue, particularly because the assignment of each OTU to a putative subspecies was based exclusively on the key by Sell (1994; see above). In this key, quantitative measurements were used only to distinguish two groups of subspecies (i.e., chrysocephalus/ficariiformis and bulbilifer/calthifolius/ficaria). Qualitative characters are used in Sell's (1994) key to distinguish individual subspecies within these two groups. Thus, contributing evidence of the morphological cohesiveness of the subspecies concept of Sell (1994) is the extent to which differences in quantitative characters are found be- tween all subspecies pairs. Of course, had we found that our OTU groups assigned to the subspecies sensu Sell (1994) did not differ significantly in quantitative characters, it would not necessarily have challenged Sell's concepts, as the taxa may truly differ only in qualitative characters. However, the finding that the OTU groups corresponding to the subspecies sensu Sell (1994) do in fact differ in various combinations of the quantitative characters we examined provides some additional evidence of distinctness. Although the groups differed primarily in the means of quantitative characters and showed overlap in maximum dimen- sions, if one accepts a subspecies as an incompletely diverged lineage, overlap in character states cannot be Post et al., A pt tri lysi fR Jus ficari 203 unexpected. In addition, distinct means in quantitative characters could be viewed as a reflection of partial isolation and potentially emerging distinct evolutionary trajectories, possibly leading to speciation. In rec- ognizing that our analysis is limited to plants introduced to the United States, our results could be biased if our data sets largely contained “non-controversial” individuals (e.g., individuals from subspecific centers of distribution in Europe, rather than regions of overlap). There is no way to know this, except through a broader study. However, we did not explicitly seek to test the subspecies concept sensu Sell (1994), but rather whether plants introduced to the United States could be reasonably referred to that concept—which we believe they can. The recognition of subspecies of R. ficaria has obvious practical consequences in weed management, as not all subspecies may behave in the same manner. Summary of taxon distribution, habitar, and rate eof spread.—Based on our current understanding, the subspecies of R. ficaria exhibit in the United States (Fig. 5). Subspecies calthifo- lius occurs in a states and the District of Columbia. It was apparently first collected in the United States in 1867 (Pennsylvania: Burke s.n., PH). Collections of this subspecies account for 35.5% of specimens examined. Subspecies bulbilifer currently occurs in sixteen states and the District of Columbia. It was appar- ently first collected in the United States in 1891 (New York: Hollick s.n, LGO) and accounts for 31.5% of the specimens examined. Subspecies ficaria occurs in ten states and the District of Columbia. It was apparently first collected in the United States in 1876 (New York: Schrenck s.n., LGO) and accounts for 15% of the col- lections examined. Subspecies ficariiformis currently occurs in Missouri, North Carolina, New York, Ohio, Pennsylvania, and Texas. It was aj ly first collected in the United States in 1910 (Pennsylvania: St. John 111, GH) and accounts for only 8% of specimens examined. Subspecies chrysocephalus currently is known only from Maryland, New York, Oregon, and Washington. It was apparently first collected in the United States in 1975 (Oregon: Hatch s.n., NY, OSC). The narrower range documented for subsp. chrysocephalus vis-a-vis the other subspecies is attributed to the recency of introduction—collections of this subspecies in the United States are unknown prior to 1975. The disjunct populations in the east and west likely resulted from secondary introductions through the horticultural trade. All subspecies of Ranunculus ficaria are best adapted to moist sites (Taylor & Markham 1978). All perform well in irrigated landscapes, such as lawns and horticultural plantings, but occur in a variety of habitats from moist woods to roadsides and lawns (Fig. 6). The majority of collections of all subspecies (45.9%) were made adjacent to a water source such as a river, stream, or pond. An additional 15.5% came from other moist areas. Collections from lawns and horticultural plantings were equal at 8.13% each. A few specimens have been collected in other habitats such as disturbed sites, lowlands, and fields. Twenty-three percent of specimens of subsp. ficariiformis were collected from dry woods suggesting that this subspecies may tolerate more xeric environments than the other four. Habitat distributions of subsp. chrysocephalus and subsp. ficariiformis likely represent only a limited percent of the habitat range of these taxa due to the limited number of collections they are based upon (N = 6 and N = 13, respectively). Vegetative spread occurs through tuberous roots, although subsp. bulbilifer and subsp. ficariiformis also produce axillary bulbils for reproduction (Taylor & Markham 1978; Sell 1994). All subspecies except subsp. bulbilifer produce viable seed which frequently fall adjacent to parent plants (Marsden-Jones 1937). Disper- sal over long distances likely occurs anthropogenically. The subspecies multiply easily along riverbanks, forming dense mats where there is seasonal flooding (Taylor & Markham 1978). Short distance dispersal is effected by seasonal flood waters which may transport tubers or bulbils downstream. This dispersal pattern was confirmed by primary observation on subsp. ficariiformis in Wake Co., North Carolina, in the spring of 2006. The subspecies was distributed along a drainage ditch, through a culvert under the road, and into a local waterway where it colonized banks downstream from the source. Persistence in the landscape is exacerbated by continued use in the nursery trade as a garden plant. Plants may slowly escape from cultiva- tion and spread when tuberous roots, bulbils, or small plants are discarded in yard waste. Figure 7 shows the relative rate of spread of each subspecies by the number of counties in which it was collected by decade. It does not appear that any single subspecies of R. ficaria is more invasive than another 204 Journal Distribution of R lus ficaria and its five subspecies in the United Stat f2008. O = subsp. bulbilifer; 4 = subsp. calthifolius; + = subsp. chrysocephalus; * = subsp. ficaria; Y = subsp. ficariiformis. in the United States. There is no significant difference among the expansion slopes of the five subspecies during the first forty years after each introduction (p=0.0769), suggesting that each pecies behaves simi larly, at least in the early phases of expansion. It may be expected, therefore, that more recently introduced taxa, such as subsp. chrysocephalus, subsp. ficaria, and subsp. ficariiformis, will follow a similar pattern to that of the earlier introductions, subsp. bulbilifer and subsp. calthifolius, and extend their ranges at similar exponential growth rates in the next hundred years. All subspecies of R. ficaria should be expected to persist where introduced throughout most of Canada, New England to Iowa, possibly northern California, and as far south as Texas. We have not seen collections from California, Iowa, Indiana, Maine, Rhode Island, or Vermont, although expect that subspecies will persist there, as well. The Midwestern states of Arizona, Colorado, Idaho, Kansas, North Dakota, Nebraska, Nevada, Oklahoma, South Dakota, Utah, and Wyoming are likely too arid for widespread naturalization. However, subspecies may establish locally in irrigated areas or wetlands following introduction. APPENDIX A List of exsiccatae of Ranunculus ficaria in the United States. Arranged alphabetically by subspecies. * 2 handwriting difficult to decipher; d» = double flowers. Ranunculus ficaria L. subsp. bulbilifer Lambinon U.S.A. CONNECTICUT. New Haven Co.: 12 May 1992, Mehrhoff 15469 (CONN, Y); 4 Jun 1997, Morehead Il! 3561 (CONN DELAWARE. New Castle Co.: 24 Apr 2004, Clancy 5905 (DOV); Apr 1978, Lindtner 109 (DOV); Schuyler 7210 (PH). DISTRICT OF COLUMBIA: 27 Apr 1993, Redman 6651 (BALT). ILLINOIS. Cook Co.: 6 May 1987, Evert 11832 (MOR); 6 May 1987, Evert 11833 (MOR); 17 May 1989, Evert 16279 (MOR, NA); 8 May 1997, Hick 514 (MOR); 26 May 1978, Kamin 945-3000 (MOR); 2 May 1982, Lace s.n. (MOR); 5 May 1998, Masi, Epting, & ! ich 817 (ILLS); 1 May 1960, Venrick 122 (MO). Dupage Co.: 27 Apr 2003, Kobal FPD03-02 (MOR); 5 May 1995, Lampa 95-03 (MOR). Lake Co.: 27 May 1998, Fiest 24 (LLS). MARYLAND. Baltimore Co.: 26 Apr 1975, Beach 0107 (MARY); 17 Apr 1984, Hill 13552 (BRIT, GH, MARY, MO, MSC, NY-2 sheets, POM); 22 Mar 1989, Redman 6320 (BALT); 30 Apr 1993, Redman 6492 (BALT); 14 Apr 1974, Romeo 41 (MARY). Howard Co.: 30 Apr 1965, Engh s.n. (BKL, MARY). — 205 Post et al i 9=N SIUJOJIL82H 'deqns Ze= WET Bally JSON 0) Jueselpy snjeudo»2os/uuo ‘dsqns "S08 Pauly ISIOIN WEE JenipnomnoH %LL spoom Aig 8H82IJ 'dsqns L1 £9JV ISON JSJEM 0} jusoeípy 403111G/Nq ‘dsqns snyojiyzjeo *deqns o} jusoeípy 966p Jej8M a} usoelpy A? gay pequnisiq \ 9M yg 9PspeoM puejmo7 piti Fic. 6. P. 90 7 subsp. calthifolius Ê B —n— subsp. bulbilifer / P o —&— subsp. ficaria 2 35 1 subsp. ficariiformis [a S 30 | subsp. chrysocephalus o 25 = 20 7 > 15 ¿A e * 10 DE E D erxRERELTVT 0 4 1860 1880 1900 1920 1940 1960 1980 2000 decade F F Etha £i L A TO hA £ £ O A i n n7coV M VU US) Montgomery Co.: 25 Apr 1937, Benedict Jr. 3679 (NA); 18 Apr 1976, Morris 1 1 (MARY). Prince Georges Co.: 19 Apr 1974, Bunn 37 (MARY). MASSACHUSETTS. Suffolk Co.: 24 May 2004, Mehrhoff 21169 (CONN). MICHIGAN. Clinton Co.: 9 May 1982, Gereau 966 (MICH, MSC). Eaton Co.: 22 May 1984, Blouch s.n. (MSC); 31 May 1984, Blouch s.n. (MSC). lonia Co.: 21 May 2003, Reznicek & Kogge 11470 (MICH). MISSOURI. Saint Louis Co.: 21 Apr 1985, Brant 55 1 (MO); 26 Apr 1972, Brown s.n. (MO); 12 Apr 1992, Ochs 19 (MO). NEW HAMPSHIRE. Hillsborough Co.: 13 May 1962, Stiff s.n. (NHA). NEW JERSEY. Essex Co.: 25 Apr 2003, Glenn 8181 (BKL). Hunterdon Co.: 26 Apr 2001, Glenn 5425 (BKL). Mercer Co.: 22 Apr 2002, Glenn 8988 (CONN, BKL). Passaic Co.: 17 Ma 1939, Clausen s.n. (BH); 6 May 1941, Langmuir & Lawrence s.n. (BH). Somerset Co.: 23 Apr 2003, Glenn 8149 (BKL). NEW YORK. Bronx Co.: 4 May 1964, Bennett s.n. (NY); 28 Apr 1939, Swift R432/37 (NY); 4 May 1996, Walker 1868 (NY); 19 Apr 1989, Yost 355 (DOV). Cayuga Co.: 23 May 1932, Hazard 17883 (CU). Nassau Co.: 27 Apr 2001, Steward 416 (BKL). Richmond Co.: 3 May 1937, Guiler s.n. (CU); 19 Apr 1891, Hollick s.n. (LGO). Tompkins Co.: 10 May 1935, Burnham 18847 (CU); 22 May 1935, Clausen s.n. (Bm), 22 May ud 7, Gershoy 8104 (CU). OHIO. Franklin Co.: 17 Apr 1977, Carr 68 (OS); 27 Apr 1980, Carr 2645 (OS); 28 Apr 1987, 4 (OS); 16 Apr 1992, Lowden 4908 (OS). Hamilton Co.: 21 Apr 1982, Cusick 21469 (OS); 12 Apr 1989, Cusick 27953 (OS).M Co.: 6 May 1978, Cusick 18074 (KE, OS); 23 Apr 1998, McCormac 6484 (MICH). OREGON. Multnomah Co.: 26 Apr 1962, Sib (OSC). PENNSYLVANIA. Berks Co.: 5 May 1972, Brumbach 7910 (BH, NA-2 sheets, NY). Blair Co.: 1 May 1987, k 8564 (PH). Bucks Co.: 3 May 1959, Forman s.n. (PH); 18 Apr 1998, Mehrhoff 20039 (CONN). Chester Co.: 18 Apr 1959, Webb & Wherry s.n. (PH). Delaware Co.: 6 Apr 1937, Blaser s.n. (CU); 7 May 1944, Carter 5084 (DOV); Apr 1908, Painter s.n. (MO, NA); 16 Apr 1942, Wheeler 5600 (POM). LeHigh Co.: 26 Apr 1959, Schaeffer Jr. 58388 (PH). Montgomery Co.: 22 Apr 1922, Dreisbach 868 (F, MICH); 24 Apr 1954, Wherry s.n. (PH). Philadelphia Co.: 26 Apr 1911, Eckfelds s.n. (PH-2 sheets); 1 May 1912, Fackenthall s.n. (NA); 20 Apr 1911, St. John s.n. (GH); 12 Apr 1908, Van Pelt s.n. (MICH, PH). VIRGINIA. Fairfax Co.: 13 Apr 1976, Bradley & Frederickson 9954 (WVA). WASHINGTON. King Co.: 17 Mar 2002, Zika & Jacobson 16885 (WTU). WEST VIRGINIA. Ritchie Co.: 15 May 1970, Elliott s.n. (WVA). Wood Co.: 8 Apr 2003, Grafton s.n. (WVA). m a on. | subsp el: iD L hy Arcan nge U.S.A. CONNECTICUT. Fairfield Co.: s Apr 1985, Mehrhoff 11192 (CONN). New Haven Co.: 20 Apr 2001, Murray 01-001 (CONN). pilin co 29 Apr 2 í 0744 (CONN). DELAWARE. New Castle Co.: 4 Apr 1985, Meyer & Mazzeo 20693 (NA). DISTRICT :20 Mar 1983, Flemming s.n. (MARY); 14 Apr 1983, Fleming s.n. (NA); 3 Apr 1986, Fleming 4 (NA); 13 Apr 1899, qs 72 (NA); de 1874, McCarthy s.n. (NA); 26 Apr 1884, McCarthy s.n. (NA); 18 Apr 1897, Topping s.n. (NA). ILLI- NOIS. Cook Co.: 2 May 1996, Antonio & Masi 7577 (ILLS); 24 Apr 1966, Argent M. D. s.n. (ILLS). Lake Co.: 24 Apr 1988, Snydacker lysis of R lus ficari 207 Post et al., A | [ 561 (F). KENTUCKY. Campbell Co.: 12 Apr 1981, Buddell I| 108 (NY). MARYLAND. Anne Arundel Co.: 4 Apr 1991, Longbottom 1460 (MARY); 15 Apr 1993, Longbottom 3473 (MARY). Baltimore Co.: 23 Apr 1971, Chanoski 043 (BALT); 3 Apr 1980, Critikos 9 (BALT); 4 Apr 1977, Lears s.n. (MARY); 13 Apr 1974, Ness 33 Mp 1 Mod idi Redman 4029 (BALT). Caroll Co.: 13 Apr 1963, Burroughs 31 (MARY). Howard Co.: 17 Apr 1965, Stolze 386 (F-2 sheets). A y Co.: 25 Jan 1950, Cross s.n. (NA); 27 Mar 1976, Mora 22 (MARY); 1 ii e Sappington s.n. (MARY); 6 Apr en Schlossberg 0002 (MARY); 19 Mar 1983, Zastrow 1 (OSC). Prince Georges Co.: 9 Apr 1987, Bowman 377 (MARY); 13 Apr 1970, Thompson Jr. s.n. (MARY). MASSACHUSETTS. Middlesex Co.: 8 May 1982, Wood 4690 m MICHIGAN. lonia Co.: 23 Apr 1989, Penskar 1085 (MICH); 24 Apr 1989, Penskar 1086 (MICH). MISSOURI. Saint Louis Co.: 4 Apr 1989, Yatskievych, Yatskievych, iod 89-05 (MO); 1 Apr 1999, a ied Harris, Harris, & S 99-04 (MO). NEW JERSEY. Burlington Co.: 16 Apr 1932, Stokes M.D. s.n. a : 1898, Saunders s.n. (PH). Middlesex Co.: 12 Apr 2003, P 354 (CONN). Somerset Co.: 7 Apr ian. dd pu (BKL). Union Co.: 3 Apr 2002, Glenn & Steward 6197 (BKL). NEW YORK. Bronx Co.: 11 Apr 1976, n Ne (BKL). Dutchess Co.: May 1941, Van Melle s.n. (BH). Queens Co.: Apr 1876, Schrenk s.n. (LGOJ; 1876, Schrenk s.n. (LGO); 25 Apr 1877, Schrenk s.n. (BKL, MOJ; Apr 1877, Schrenk s.n. (PH); 11 Apr 1878, Schrenk s.n. (MICH); Apr 1878, Schrenk s.n. (BKL, F-2 sheets, HNH, NA, NY-2 sheets, POM); Apr 1878, Schrenk s.n. (LGO-2 sheets, NA); May 1882, Bisky s.n. (BKL). Richmond Co.: 13 Apr 1898, Coheu* s.n. ini 2 sheets). Tompkins Co.: 7 May 1937, Anderson s.n. (MASS); e 19 Apr 1959, Dress 5984 (BH); 1 May 1996, Dress 19996 (BH). OHIO. Clark Co.: 31 Mar 1992, Cusick 30056 (OS). Clermont Co.: 2 Apr 1996, Cusick 32886 (MO, OS). Clinton Co.: 3 Apr 1989, inr 27946 (OS). Franklin Co.: 5 Apr 1986, Cooperband 5 (OS); 29 Mar bu Cusick 26207 (NY). Greene Co.: 31 Mar 1992, gu es (OS). Lake Co.: 2 Jun 1901, Hacker s.n. (OS). Van Wert Co.: 22 Apr 1946, Brooks 1393 (OS); 28 Apr 1947, Brooks REGON. Multnomah Co.: 27 Mar 1991, Zika 11064 (OSC). IAN Bucks Co.: 18 Apr 1962, Wherry s.n. (Ph Delaware Co.: 13 Apr 1934, Fogg Jr. 6316 (PH); 12 Apr 1938, Fogg Jr. 14021 (GH); 7 Apr 1894, MacElwee Jr. s.n. (PH); 12 Apr 1920, Meredith M.D. s.n. (NY); 15 Apr e Schaeffer Jr. 16953 (PH-2 sheets); 19 Apr 1936, Thompson Jr. 17 (PH); Greene Co.: 4 Apr 1953, Buker s.n. (PH). M y Co.: e 19 Apr 1963, Fogg Jr. 22220 (A); 3 Apr 1921, Long 23784 (PH); 20 Apr 1937, Long 49720 (PH); 18 Feb 1954, Long 77721 o 1 May 1985, Weaver s.n. (PH). Philadelphia Co.: 1867, Burke s.n. (PH); e10 Apr 1954, Fogg Jr. 21460 (PH-2 sheets); e 18 Apr 1954, Fogg Jr. 21474 (PH); 3 Apr 1933, Hermann 3953 (NA); 18 Apr 1974, Jers* (PH); 27 Apr 1924, Lang 112 (GH); %1 May 1920, Meredith M.D. (NY); 30 Mar 1909, St. John 110 (GH). TENNESSEE. Knox Co.: 17 Mar 1977, BA (TENN); 25 Mar 1966, Thomas s.n. (BRIT); 31 Mar 1966, Thomas & Rogers s.n. (TENN). VIRGINIA. Albemarle Co.: 25 Mar 19 25629 (VPI). Chester Co.: 3 Apr 1999, Huber 1 (ILLS, OS). Fairfax Co.: 13 Apr 1947, Sargent s.n. (NCSC); 23 Mar 1974, rd (MARY). WASHINGTON. Whatcom Co.: 5 Apr 1968, Sundquist 1550 (POM). WEST VIRGINIA. Ritchie Co.: 15 Apr 1963, Stonestreet s.n. (WVA). WISCONSIN. ohh Co.: May 1970, Larkin s.n. (MOR). R ficaria L. subsp. cl halus PD. Sell U.S.A. MARYLAND. diss Georges Co.: 27 Mar 1977, Wirick 05 (MARY). NEW YORK. Tompkins e MAR Dress d (CU). OREGON. Benton Co.: 31 Mar 1991, Zika 11065 (OSC). Lane Co.: 22 Feb 1975, Hatch s.n. (NY, Co.: 17 May 2002, pois & Jacobson 13733 (WTU). Ranunculus beris L. subsp. ficaria » S.A. DISTRICT OF COLUMBIA: e 29 Apr 1896, Pollard 776 (NY); #29 Apr 1896, Pollard 776 (MSC); %29 Apr e Pollard n. (POM); 417 Apr 1898, Pollard s.n. (NY); 417 Apr 1898, Steele s.n. (MSC); #18 Apr 1897, Steele s.n. (GH); de 18 Apr 1897, med s.n. (NA); e 18 Apr 1897, Steele s.n. (NY). ILLINOIS. Lake Co.: 24 Apr 1988, Snydacker 560 (MOR). aÑo Baltimore Co.: 13 Apr 1980, Caruso 22 (BALT); 5 Apr 1980, King 10 (MARY). Howard Co.: 25 Mar 1989, Redman 6321 (BALT). Montgomery Co.: 19 Apr 1975, Hollenberg 8 pea 19 Apr 1979, Trumball 4 (BRIT); 13 Apr 1975, Yinger 17 (MOR). Prince Georges Co.: 10 Apr 1977, Dochtermann 29 (MARY); 16 Apr 1994, Hedge 13 (MARY); 2 Apr 1982, Kunowsky 10 (MARY); 4 Apr 1964, Weigel Jr. s.n. ane MASSACHUSETTS. Hampshire Co.: 24 Apr 2002, Mehrhoff 21611 (CONN, MASS). MISSOURI. Saint rs Co.: 1 990, Christ s.n. (MO); NEW JERSEY. Mercer Co.: 18 Apr 2001, Glenn 5386 (BKL). Somerset Co.: 19 Apr a Ps 5401 (BKL). NEW YORK. Cayuga Co.: 7 Jun 1935, Petry 18846 (CU). Dutchess Co.: 1932, Van ma s.n. (BH). Queens Co.: 22 Apr 1990, sel s.n. (BKL); 1921, Martin s.n. (PH); Apr 1877, Redfield 10896 (MO); Apr 1876, Schrenk s.n. (LGO). OHIO. Butler C Apr 1997, Turner 31 (MU). PENNSYLVANIA. Chester Co.: 4922 Apr 1941, Terrell Jr. 375 (PH); %6 May 1954, idus n. (PH). Delaware Co.: 30 Apr 1892, Brinton M. D. (PH-3 sheets); Apr 1903, Conard s.n. (PH); 5 Apr 1935, Fogg Jr. 7996 (PH); 1 May 1904, Jahn s.n. (PH); 4 Apr 1946, Proctor 1716 (NHA). Northampton Co.: 2 May 1969, Tucker s.n. (DOV). Philadelphia Co.: %20 Apr 1921, Hur s.n. (PH); 23 Apr 1932, Hermann 2752 (NA); e May 1878, Martindale s.n. (NA, LGO); %12 May 1904, Van Pelt s.n. (PH). WASHINGTON. King Co.: 8 Mar 2000, Zika 6 Jacobson 14827 (WTU). WEST VIRGINIA. Wood Co.: 12 Apr 2003, Grafton s.n. (WVA) R ficariiformis (FW. Schwartz) Rouy & Fouc U.S.A. MISSOURI. Sui Du Co.: 19 May 1994, Ladd 18515 (MO). NEW YORK. Bronx Co.: 17 Apr 1988, Mori & Gracie "o (GH, MO, NY). Nassau Co.: 8 May 1950, Abbott s.n. (CU). Suffolk Co.: 1 May 2003, Glenn 8215 (BKL). Westchester Co.: 2 May 1994, Walker 684 (NY). NORTH CAROLINA. Wake Co.: 11 Apr 2005, Krings 1271 (AUA, F, FLAS, GA, LSU, MISS, NCSC-2 rub NCU, TEX, UNA, US, USF, USCH, VDB). OHIO. Butler Co.: 5 Apr 1988, Cusick 27224 (OS). PENNSYLVANIA. Delaware Co.: 16 May £ 4L D H ID hl EP. FT, 208 Journal of Texas 3( 1915, Cramfondi* s.n. (PH); 26 May 1920, Long 23006 (PH). Philadelphia Co.: 7 May 1910, St. John 111 (GH). TEXAS. Tarrant Co.: 8 Mar 2008, Nesom FW08-1 (NCSC) Ranunculus ficaria L. (subspecific determination not feasible) U.S.A. CONNETICUT. Middlesex Co.: 27 Apr 1991, Swan s.n. (Y). New Haven Co.: 18 Apr 1995, Brown 1 (Y); 10 May 1993, Souther s.n. (Y). MARYLAND. Baltimore Co.: 25 Apr 1948, Moudry s.n. (MARY); 21 Apr 1968, Redman s.n. (BALT). Howard Co.: 17 Apr 1966, Engh s.n. (MARY). Prince Georges Co.: 19 Apr 1979, Mills Jr. 14 (BRIT). MASSACHUSETTS. Barnstable Co.: 27 Apr 2001, Mehrhoff 21129 (CONN). MICHIGAN. Ingham Co.: 10 Apr 1991, Stephenson s.n. (MSC). NEW JERSEY. Union Co.: 3 Apr 1977, Moldenke & Moldenke 31285 (NY); 10 May 1978, Moldenke & Moldenke 31642 (NY). NEW YORK. Nassau Co.: 4 May 2004, Bennett 030031 (BKL). Queens Co.: 27 Apr 1918, Ferguson 7 (NY); 17 Apr 1921, Martin s.n. (NY); 30 Apr 1876, Schrenck s.n. (CU). OHIO. MONTGOMERY Co.: 18 Mar 1990, McCormac 1977 (MICH, OS). OREGON. Multnomah Co.: 4 Apr 1991, Zika 11066 (OSC). PENNSYLVANIA. Chester Co.: 15 Apr 1957, Wilkens 9129 (PH). Delaware Co.: 7 May 1977, Brown 199 (Y); 24 Apr 1894, MacElwee Jr. s.n. (PH). Philadelphia Co.: 11 Apr 1954, Fogg Jr. 21461 (PH). WEST VIRGINIA. Monongalia Co.: 12 Apr 1995, Baer s.n. (WVA). Ritchie Co.: 25 Apr 1971, Elliott s.n. (WVA); 27 Apr 1991, Grafton s.n. (WVA). — ACKNOWLEDGMENTS We thank the following herbaria and their staff for ilable to us for study: A, AUA, BALT, BH, BKL, BRIT, CONN, CU, DOV, F, FLAS, GA, GH, HNH, ILLS, KE, LGO, LSU, MARY, MASS, MICH, MISS, MO, MOR, MSC, MT, MU, NA, NCSC, NCU, NHA, NY, OS, OSC, PH, POM, TENN, TEX, UNA, US, USF, USCH, VDB, VPI, WTU, WVA, Y. We also thank David Schuman and Steve Stanislav for statistical support. Three anonymous reviewers offered constructive comments on an earlier draft of the ms. REFERENCES ALMEIDA-PINHEIRO DE CARVALHO, M.A., C.C. WiLcock, T.M. Marques Dos SANTOS, IC. VatE-Lucas, J.F. TEIXEIRA-GANANGA, E. FRANCO, D. THANGADURAI, D. MURALIDHARA-RAO, AND D. FrEITAS-SOUSA. 2004. A review of the genus Semele (Ruscaceae) sys- tematics in Madeira. Bot. J. Linn. Soc. 146:483-497. BAiLEY, L.H. 1935. Cyclopedia of horticulture. MacMillan Co., New York. Bett, FH.1945. The genus Ranunculus in West Virginia. Amer. Midland Naturalist 34:735—743. Benson, L. 1942. North American Ranunculi-V. Bull. Torrey Bot. Club 69:373-386. Breiman, L., J.H. FRIEDMAN, R.A. OLSHEN, AND C.J. STONE, 1984. Classification and regression trees. Wadsworth & Brooks/ Cole, Monterrey. Dickinson, T.A. AND J.B. Phipps. 1985. Studies in Crataegus L. (Rosaceae: Maloideae).XIll. Degree and pattern of phe- notypic variation in Crataegus sect. dud ul in DUO Syst. Bot. 10: a 337. DurRENE, M, J.L. GATHOYE, AND D. TYTECA. 1991 | Dactylorhiza (Orchidaceae)— the D. maculata group. PI. Syst. Evol. 175:55-72. ESRI. 2004. ArcGIS Version 9.1. Environmental Systems Research Institute, Inc., Redlands. GLEASON, H.A. AND A. CronquisT.1991. Manual of vascular plants of northeastern United States and adjacent Canada. 2™ ed. New York Botanical Garden, New York. Gower, J.C. 1971. A general coefficient of similarity and some of its properties. Biometrics 27:857-871. GREILHUBER, J. 1974. Ein chromosomensatz von Ranunculus ficaria subsp. calthifolius. Mitt. Bot. Arbeitsgem. Oberósterr. Landesmus. Linz 6:3-6 IHAKA, R. AND R. GENTLEMAN. 1996. R: A language for data analysis and graphics. J. Computat. Graphical Statistics 5(3):299-314. Jovy, S. AND A. Bruneau. 2007. Delimiting species boundaries in Rosa Sect. Cinnamomeae (Rosaceae) in Eastern North America. Syst. Bot. 32:819-836. KRINGS, A, A.S. WEAKLEY, J.C. NEAL, AND E.C. Swag. 2005. Ranunculus ficaria (Ranunculaceae) new to North Carolina and an updated key to Carolina congeners. Sida 21:2429-2437. Marchant, CJ. AND C.A. BnicHTON. 1974. Cytological diversity and triploid frequency in a complex population of Ranunculus ficaria L. Ann. Bot.38:7-15. Marosen—Jones, E.M.1937. Ranunculus ficaria Linn.: life-history and pollination. J. Linn. Soc. Bot. 50:39-55. Nesom, G.L. 2008. Ranunculus ficaria (Ranunculaceae), naturalized in Texas. J. Bot. Res. Inst. Texas 2:/41—742. Post et al., A pt tri lysi fR lus ficari 209 Pimentel Pereira, M., G. Estévez Pérez, AND E. SAHUQUILLO BALBUENA. 2007. European sweet vernal grass (Anthoxanthum: Poaceae, Pooideae, Aveneae): A morphometric taxonomical approach. Syst. Bot. 32:43-59. R FOUNDATION FOR STATISTICAL COMPUTING, 2008. R version 2.7.1. http://www.r-project.org/foundation/ Rosen, D.J., S.R. Hatch, AND R. Carter. 2007. Infraspecific taxonomy and nomenclature of Eleocharis acutangula (Cyperaceae). J. Bot. Res. Inst. Texas 1:875-889. SARGENT, D.J., M. Gege, J.A. HAWKINS, MJ. WILKINSON, N.H. Batey, AND D.W. Simpson. 2004. Quantitative and qualitative differences in morphological characters revealed between diploid Fragaria species. Ann. Bot. 94:787—796. SAS INsrirUTE Inc. 2002-2005. SAS Version 9.1.3. SAS Institute Inc., Cary. SeLt, PD. 1994. Ranunculus ficaria L. sensu lato. Watsonia 20:41-50. SNEATH, PH.A. AND R.R. Sokar. 1973. Numerical taxonomy: The principles and practice of numerical classification. W.H. Freeman and Company, San Francisco. TAYLOR, K. AND B. Markham. 1978. Biological flora of the British Isles Ranunculus ficaria L. (Ficaria verna Huds.; F. ra- nunculoides Moench). J. Ecol. 66(3):1011-1031. Tun, T.G. 1964. Ranunculus L. In T.G. Tutin et al, eds. Flora Europaea, Vol. 1. Cambridge University Press, Cam- bridge. Pp. 223-237. USDA, NRCS. 2007. The PLANTS Database, Version 3.5 (http://plants.usda.gov). Data compiled from various sources by Mark W. Skinner. National Plant Data Center, Baton Rouge. WhrtTemMORE, A.T. 1997. Ranunculus. In Flora North America Editorial Committee, eds. Flora North America, Vol. 3. Oxford University Press, New York. Pp. 88-135. 210 BOOK REVIEW WILLIAM A. WEBER AND RONALD C. WITTMANN. 2007. Bryophytes of Colorado: Mosses, Liverworts, and Hornworts. (ISBN 978-0979090912, pbk.). Pilgrims Process, Inc., PO. Box 32597, Santa Fe, New Mexico 87594-2597, U.S.A. (Orders: www.pilgrimsprocess.com, 1-720-937-6641). $29.95, 231 pp., 8 1/2" x 11". Almost too small to bud NE PAPER en masse, De are not pe When viewed with a hand lens or microscope some of the unusual, lare revealed e r D The genesis for the cur 1 1 with a preliminary led by Weber in 1973. He presented a joi del i keys B us mosses, at an time e identifying shou 290 pe itl i g d hornworts. The current I is th prehensive study of ‘Colorado mosses and liverworts to date. It en b ful in adj f hl d intl i gi ftl western e States. Bernd: and students will find this bool luable ref though it d t Keys to the genera and pe i mbined s notes on E in the dae th trate on n distinguishing characteristics of idi 1 for the identificati but the authors J g I have been undis to produce a useful ke to he eae They hat tl family | ilabl the internet, but that for the seasoned MISCERE family id d that for the beginner tl | 11 d D pa | $, A 1 id f, de 1 11 f A Only a few e&t e E 1] D 13 1 ET A t y and anatomy. Recommended for 1 1 ] ig 11 1 El J 1 Should h 215411 7 j 11 RI +L, f. 1 x E XA7AL e LEN. f, E. E 1 Td E fr^ J n 14 J +l f, fel a PONE x Ir William A. y olorado Museum Herbarium. He is tl I f bool fl ing plants, ferns, and lichens as well as bryophytes, concentrating on Colorado m E d M E Pon in scope as well Wi plished amateur botanist, and is the co-autl f the l versions of Ronald the Colorado Flora.—Gary L. prem Librarian Botanical Resear Institute of Texas, Fort Worth, Texas 76102-4025, U.S.A. J. Bot. Res. Inst. Texas 3(1): 210. 2009 THE TYPES OF ASTRAGALUS SECTION DIPHYSI (FABACEAE), A COMPLEX ENDEMIC TO WESTERN NORTH AMERICA, PART I: LECTOTYPIFICATIONS, EPITYPIFICATIONS, AND NEW COMBINATIONS OF SEVERAL TAXA J. Andrew Alexander Curator, Utah Valley University Herbarium (UVSC) and Research Associate Pu A as Utah Valley University Wesley E. Niles Herbarium (UNLV) st University Parkway as of Nevada, Las Vegas prn Utah, 84058, U.S.A. 4505 Maryland Parkwa alexanja@uvu.edu Las Vegas, o 4004 U.S.A. RESUMEN This is the first of several Pipe discussing typification issues Pipes JURE snis leading to a monograph of Astragalus L. Section Diphysi A. Gray. À lectot designat Hook. Original material designated here as the nol ype of Ass diphysus A. Gray var. s albiflorus- A. nid [sAsirzgalus lentiginiosus var. albiflorus (A. Gray) Schoener] was rediscovered at GH over y used name, A. lentiginosus var. diphysus (A. oy M.E. Jones, is discussed. ds are deere for. lennon var. micans Barneby and A. lentiginosus var. oropedii Barneby. Three new combinations, previ- are proposed: A g ar. bryantii (Barneby) J.A. Alexander, A. lentiginosus var. iodanthus (S. Watson) J A. Alexander and A. lentiginosus var. pseudiodanthus (Barneby) J.A. Alexander. RESUMEN T 1 - > : 1 dz +1 de tin Es A ey J 1 As 43 1 F I £ T ph ZA Gra 1 Al x um T* I1 T. a | A sidus L.S e como Rd de Astragalus dps A idi var. t albiflorus 5 Gray [Astragalus lennan Si var. albiflorus (A. Gray) Sdhóeed fue i lentiginosus var. diphysus (A. Gray) M.E. ones Son deste d epitipos de A. lentiginosus y var. micans Barneby y A. lentiginosus var lentiginosus var. bryantii (Barneby) J. A. aa A. lentiginosus var. iodanthus (S. watsen) A. Alexander, y A. nia var. pseudiodanthus (Barneby) J.A. Alexander. INTRODUCTION Astragalus L. Section Diphysi A. Gray, a section native to Western North America, is a problematic species complex composed of taxa that were originally described as or have been split as species in the past 150 years. Astragalus lentiginosus Douglas ex Hook. (in Hooker 1831) and A. diphysus A. Gray (in Gray 1849) were the core species of Section Diphysi in the first comprehensive monograph of Astragalus by Asa Gray (1863). Sereno Watson (1871) expanded Gray's concept of Section Diphysi to include A. coulteri Benth. and A. platytropis A. Gray, both of which have inflated pods similar to A. lentiginosus. Gray's monograph (1863) and Watson's (1871) revision were the primary references for Astragalus taxonomy until Marcus E. Jones began publication of treatments in the genus in the late 1890s. In his 1898 publication, Jones proposed that species from Gray's (1863) Section Diphysi should be com- bined, as varieties, into a greatly expanded concept of A. lentiginosus. His full treatment of these species was not widely known until he published his Revision of North-American Species of Astragalus in 1923 (Barneby 1964). Jones was a field botanist of unmatched experience. As a taxonomist, he had a disdain for rules of priority and nomenclature, and was known for his often frustrating brevity. Barneby (1945) found “that many of his names covered unreasonable extremes of variation, or that his descriptions and indications of range were either inaccurate or actually misleading" (p.65). Barneby was able to decipher Jones' species only after extensive study of his annotations and vouchers at POM and duplicates at other herbaria (see discussion of A lancearius A. Gray and A. episcopus S. Watson, Barneby 1964, p. 267; compare descriptions of varieties of A. lentiginosus of Jones 1923, p.124-125 and Barneby 1945, p. 65-152; and discussion of the taxonomic J. Bot. Res. Inst. Texas 3(1): 211 — 218. 2009 BR oummalof ical I i Texas 3( confusion between Jones’ species and hybrids which are now synonyms of A. argophyllus Nutt. var. martinii M.E. Jones, A. marianus Barneby, and A. desereticus M.E. Jones, Barneby 1964 p.629—635). Concurrently, Rydberg (1929) delimited all members of Section Diphysi as separate species of the genus Cystium Steven. Rydberg appeared to have an idealized concept of each of his species. He regularly composed his morphological descriptions and keys from features exclusively on a single type. As a result, his keys and descriptions sometimes did not match the morphology on all specimens he annotated for his monograph. Despite this, Rydberg for his time was more precise than all previous workers in the genus (Barneby 1964). Barneby (1945) was the first to comprehensively evaluate Jones’ and Rydberg’s hs and found that only one overall classification scheme fit the taxa related to A. lentiginosus well: he don of all former species into varieties. Barneby's (1945) treatment and his later Monograph (1964) still serve as a precise baseline from which all taxonomic treatments in this group are based. Different taxonomic interpretations of the degree of morphological differentiation among the many widespread, sympatric, and geographically isolated taxa within this complex have been the major source of disagreement between the revisions of Rydberg (1929), Jones (1923), Barneby (1964, 1989), Isely (1998), and Welsh (2007). Ultimately Ene major B in varietal delimitations poe these workers stem from their individual i of ty] d selection of Į | i g the nomenclaturally ‘ ‘typical” variant. Despite Barneby’s decades of work in sorring out the enon issues leading up to his Monograph, it is still difficult to determine which variant is nomenclaturally “typical” due to the often poor quality of the type material. Barneby M Qs for is adr nature of many ee by 1 examining specimens from the vicinity of the type locality. Th the informal basis for his concept of the nomenclaturally typical variant and ds sia morphological descriptions. The most recent monographs, Isely (1998) and Welsh (2007), have made significant additions to the knowledge of taxonomic boundaries in this section, but each has added their own entangled varietal de- limitations. However, both still use Barneby (1964) as a nomenclatural foundation. For a monograph of the section that is more comprehensive and thorough than Barneby's, nomenclatural issues that have resulted in conflicting varietal delimitations between the major monographs must be resolved. Fortunately, there are more formal processes in use today that seek refine the delimitation of taxa with problematic types. The Linnaean Plant Name Typification Project () were examined for examples of specific applications of the Code. Other typification papers, especially those resulting from the various tropical flora projects, were also examined. Turland and Jarvis (1997) and Krings (2008) were most frequently consulted for example typifications. Vander Kloet (1989) was consulted to determine how other Douglas taxa described by Hooker were lectotypified (i.e., Vaccinium membranaceum Douglas ex Hook., p.133). RESULTS ps m lentiginosus us ex Hook Fl Bor. bend 1:151. 1831. Tragacantha cie (Douglas ex Hook.) e, Revisio Gen. Pl. 2:946. 1891. P} (D Hook.) Piper, Contr. U.S. Natl. Herb. 11:368. 1906. d n (Douglas ex mee ae Bull. Torte Bot. Club 40: 50. 1913. PROTOLOGUE: = Ee ranges a des E Mountains North West America. Douglas." Tyre: [U.S.A. OrEGON]: in the valley of the Bl awallah us Walla] and Utala [Umatilla] Rivers, [no date], David Douglas s.n. (LecromwE [first step] designated by xis 1964: K; LECTOTYPE ee ae undas mid K diga d smaller, fruiting, Di wd ao on the oe left hand side of the sheet ight); [U.S.A. OREGON] thern branches of the Goiaba oey collected at ihe same pie Ad as a lectoypel, ED cate David Douglas s.n. (zPrrYre, designated here: K 264945}, the mostly flowering, E 1 1 3; f£, > 1 £] 1 SA 1 1 d us. P: 11 f .L E sel MM p yI ig p TNR Ine Tri : hiato] ich f fi i ftheD As with MR Douglas opener described as new qe ~ Hooker (1831 El none of the pe spectes have a fone Ga matches the From an examination of the annotations on the three type sheets, Asa Gray in 1868 was the first to annotate the A. lentiginosus specimens at K. Gray did not indicate which of the sheets was the type nor did he annotate every sheet, but he was the first to recognize that the central element (K264945) on one of the types was A. lentiginosus. However, he did not recognize that the other two elements on this sheet were type material for A. diaphanus Douglas ex Hook. After Gray, the delimitation of typical Astragalus lentiginosus has largely been based on the morphology of specimens found in the vicinity of the Blue Mountains in Oregon. Since both A. lentiginosus var. salinus (Howell) Barneby and A. lentiginosus var. platyphyllidius (Rydb.) Peck can be found in southern Blue Mountains of Oregon, some populations of these taxa have been misinterpreted as typical A. lentiginosus. Barneby (1945) was the first since Gray to critically analyze the types of the A. lentiginosus complex and refine typical A. lentiginosus morphologically. The only Douglas specimen of A. lentiginosus he saw for this revision was the fragmentary type at GH. Barneby (1964, p.917) indicated a specimen a K found “near the source of the Wallahwallah and Utala rivers” was the holotype. This is the first step lectotypifica- tion of Astragalus lentiginosus (McNeill et al. 2006, Article 9.8). Unquestionably, Barneby selected a fruiting specimen at K as the lectotype of A. lentiginosus since he knew that Hooker (1831, p.151) indicated that "the flowers of this do not appear to have been seen by Mr. Douglas” and that "floribus - ?” was the only reference to flowers in the diagnosis. However, there are two sheets in fruit from the same individual at K: one from the Herbarium Benthamianum (K 264012) and one from Herbarium Hookerianum (K 264017). It is likely that Barneby chose the specimen from the Hooker Herbarium, however no specimens at K are annotated by him. The individual, K264017, is mounted with a specimen of A. lentiginosus var. lentiginosus collected by Dr. Lyall (K 264018). After an examination of the diagnosis and all the types of A. lentiginosus, the second step lectotypification was made herein to unambiguously specify which sheet, Kew accession number, and elements mounted on the sheet correspond to the lectotype in accordance with McNeill et al. (2006) Article 9.15. The second specimen, K 264012, is an isolectoype of A. lentiginosus. Since the fruit of typical Astragalus lentiginosus is not by itself agree precise taxon delimitation based on the lectotype is not possible. Both fruit type and flower si ic features that distinguish typical A. lentiginosus from other sympatric varieties. Astragalus lentiginosus var. salinus has small whitish flowers (keel «9 mm) and thin walled, bladdery inflated pods. Astragalus lentiginosus var. platyphyllidius has £L Dag H ID ll II P, £T, 214 Journal of Texas 3( larger whitish flowers (keel 11-15 mm long) and thick walled, curved pods inflated only towards the base. Throughout its range, the fruit of A. lentiginosus var. lentiginosus can take both of these forms. The lectotype is a late season specimen with only mature, thick walled, curved pods inflated only at the base. The size of the flowers is unknown. For a precise interpretation of nomenclaturally typical Astragalus lentiginosus, a flowering specimen was chosen from the original Douglas specimens. The epitype of Astragalus lentiginosus is designated herein (K264945; McNeill et al. 2006, Article 9.7) as the larger individual (keels 8-9 mm long) in the center of the sheet. It is mounted with two other fragmentary elements. Element 2 (K264015) is a fruiting specimen of A. diaphanus. Element three is mounted in two different places on the sheet, has received different accession numbers (K264016 and K264014), and is a flowering specimen of A. diaphanus. This entire sheet was from the Herbarium Hookerianum. Elements two and three are potential lectotypes or isolectotypes of A. diaphanus and will be discussed in a future publication (Alexander, in prep). The Douglas label is associated with a portion of element three and indicates it was collected “on the banks of streams on the southern branches of the Columbia." Presumably, the flowering A. lentiginosus element was collected at the same locality as the A. do elements. Alternatively, the epitype and the lectotype could have been collected from the same l vicinity since the Walla Walla and Umatilla Rivers are two of several southern branches of the Columbia River in northeastern Oregon. desa braces var. albiflorus A. Gray, Pl. Fendler. Novo-Mexicanae, Mem. Amer. Acad. Arts II. 4:34. 1849 bib em lentiginosus var. albiflorus (A. Gray) Schoener, Great Basin Naturalist 34:180. 1974. ProroLoGue: “with the preced- ing [Fend r 146 from “around Santa Fe,” New Mexico]. No. 147.” Tyre: U.S.A. Plantae Novo-Mexicanae [New Mexico]: [locality not a on label], 1847, A. Fendler 147 (novorvee: GH 112365! H 587151, K 2640231, MO). Notes.—When Barneby (1945) ined the Fendl fA dipl t GH, he only found one type specimen of Astragalus diphysus var. albiflorus A. Gray (Fendler 147 GH58715) This sheet was nn of a collection transferred from the herbarium of the Boston Society of Natural History to GH in 1941. pod pr aa P. i noted mu iioi pene “was received long after Gray’s death and not annotated by him,” and sel YE d, Barneby may have chosen the MO sheet because GH58715 has an uncertain history and was mounted wi i b p galus dipl vat. diphysus (Fendler 146 GH58716). In this study: e of Du Bes aus ee rues dl pu to uate institutions revealed mal d SUD. E ¿E Novo-Mexicanae, ] he other instituti ] inati heir dupli based his manuscript. The Fendler types i] LI! 1 1 : : $ : 2 | J 1 n 1 Mi na 1 ab 1 In 2002, a thorough search was conducted at GH of all the Monk: American fold 1 types of Astragalus. The origina ELA of A. diphysus var. albiflorus (Fendler 147 GH112365) with th j ] d the d rediscovered. This type specimen was recently divided from another ipic collection and remounted on a new shee which is the likely reason that Barneby was not able to find the original type. Currently, GH112365 is mounted with an unaccessioned, non-type specimen of A. iiie Sn by pm in 1883 in mn Abuqurqe New Mexico. The emia of the holotype of A. diphysus var. albifl bys (1 leill et al. 2006, Article 9.17. aa pane var. diphysus (A. Gray) M.E. Jones, Proc. Calif. Acad. Sci. H 5:673. 1895. Astragalus physus A. Gray, Pl. Fendler. Novo-Mexicanae, Mem. Amer. Acad. Arts IL 4:34. 1849. Gona diphysa (A. Gray) Kuntze, Revisio Gen. Pl. 2:944. 1891. Cystium diphysum (A. Gray) Rydb. Bull. Torrey Bot. Club 32:659. 1905. ProroLoGUE: “around Santa Fe [New Mexico]...No. 146." rr: U.S.A. Plantae Novo-Mexicanae [NEw js ei! not ud on label], 1847, A. Fendler 146 (noLorvee: GH 587141; isotypes: BM, E GH 587161, K! [3 Sheets], MO, NY Notes.—The ar name iid what dini been nina i = as id. cate A var. diphysus (A. Gray) M.E. Jones has been con- Barneby (1964, p. 941) stated “strict adherence E Rules of N ] ire that tl lier i varietal rank [in his opinion A. diphysus var. albiflorus] takes precedence q. A. diphysus var. diphysus]." He rejected making a new Po A. ene var. albiflorus, since it would result in “an absurdity and runs counter to common sense” of using th Based on her interpretation of the Article 11 of the 1972 edition of the International Code, Scheer (1974) concisa: “the name, var. diphysus, is anteceded by the name A. diphysus var. albiflorus...[and] the older name in the same rank has poene. (p. 180). Schoener (1974) then made the new combination, A. lentiginosus var. albiflorus (A. Gray) Schoener. S (Welsh 1978; Isely odd. have iini pd Schoeners combination, while the most recent (Welsh 2007) has used Jones’. The most recent I j Il et al. 2006, Article 11.6 and 26.3) resolves this controversy. Two taxa were described by Gray (1849): first, A. diphysus (based on ae 146); and Alexander, Types of Astragalus sect. Diphysi 215 second, A. diphysus var. albiflorus ind on "iid 147). The HE taxon, Á. iain var. albiflorus ically creates the priorable tion, A. lentiginosus var. diphysus, that M.E. autonym, A. diphysus var. diphysus. Jones made in 1895 has priority over Schoeners 1974 Eom BA: A. Top var. albiflorus. Astragalus lentiginosus var. bryantii (Barneby) J.A. Alexander, comb. nov. Astragalus bryantii Barneby, Proc. Calif. Acad. Sci. 4, 25:156. 1944. PROTOLOGUE: “ARIZONA: at the head of Phantom Canyon in the Grand Canyon of the Colorado River, Coconino Co., 15 Dec. 1939. Collected by Dr. H.C. Bryant...” Tyre: U.S.A. Arizona: [Coconino Co.]: head of Phantom Canyon, in Grand Canyon, 15 Dec 1939, H.C. Bryant s.n. (HOLOTYPE: CAS 2939401; sorvre: US 1769041, internet image. 5 fn by) LA AI EI 1 1 A Pee | A itl A lentiginosus var. palans Notes.—Astragalus lentiginosus var. } (M.E. Jones) M.E. Jones or A. lentiginosus var. mkd (A. Gray) M.E. joes in hist (Barneby 1944, Barneby ae any ie Isely 1998, Welsh 2007). Welsh et al. (2003) and mes e cra that Pies in Utah prev iously ryantii Barneby are A. piles laa var. palans All full 1 I been published, } 1 by Barneby (1964) and Wi Ish (2007) ines 828, Gaines 1005, O E 4 JM 4 and Gaines 1009, all fom NAU. Although t! pecific speci ined, Al 1 (2008) ae not BER ay en from Utah previously d d as A. bryantii Barneby or A. lentiginosus var. pal hat has tl found in 1 1 $ e 1 Dl D 1 E $ f. A Á Jaji * 11 AT 1 K^ rA QN por y G Canyon y " non var. boat is s » Fa 1 = 1 -1 c fl 333 J D e | f£ Dh t p Ranch, Coconino Co., Arizona. As a as N these voptladions are long-dist disjuncts from that of tl population of A. lentiginosus var. palans. In addition, the Lin variation d in i specimens of A. up var. TUM is n e that of A A. lentiginosus lentiginosus var. palans. However. var. wilsonii (Greene) Barneby and A. bici var. ursinus de Gray) inia Bameby ea theorized u that oui taxon was closely related to A. lentiginosus var. mokiacensis, a rel y , Isely 1998, Welsh et al. 2003, Welsh 2007). Further favestination of habitat in the canyons of the Colorado River eastward ni the nearest populations of A. lentiginosus var. mokiacensis at Emory Falls (Clover 6079, CAS!) and Quartermaster Canyon (Goodding 15-41, RM!) may reveal populations of A. lentiginosus var. bryantii intermediate to A. lentiginosus var. mokiacensis. Astragalus lentiginosus var. iodanthus (S. Watson) J.A. Alexander, comb. nov. Astragalus iodanthus S. Watson, Bot. g. 70. 1871. ProTOLOGUE: “in the foothills of Western Nevada from the Virginia to the West H ; 4500-6000 feet altitude... [S. Watson] 269.” Tyre: U.S.A. Nevapa: [Pershing Co.]: West Humboldt Mountains [Humboldt Range, NE of Lovelock], Jun 1868, S. Watson 269 [in part] (LecrorvrE, designated by Barneby 1964:962: US 469091; isoLecroryres: GH!, NYi, YU). Astragalus iodanthus var. diaphanoides Barneby, Leafl. W. Bot. 4:50. 1944. Tire: U.S.A. Nevapa: Washoe Co.: between Reno and Dewey, 10 : RSA) J 3, H.D. Ripley & R.C. Barneby 5659 Astragalus iodanthus var. vipereus Barneby, T4 New York Bot. Gard. 13:963. 1964. Tree: U.S.A. IDAHO: e Co.: on cobblestone bluffs near Bruneau, 31 May 1945, H.D. ecd a R.C. ao Tum eee is ISOTYPES: n RSA). AL Notes.—In his taxonomic revision, Al gh deg F E ferentiation between A. ingre var salis. A. iodanthus and A. pseudiodanth 1 continued separation of th I species is not supported. The new lin Alexander (2008) are made formal herein. Traditionally, differences in pod morphology have been the basis for the separation of Astragalus iodanthus S. Watson and A. pseudiodanthus Barneby from A. lentiginosus. Both have a deciduous, mostly uniloculate to partially biloculate pod with a septum less than half the width of the locule. This feature does differentiate these two taxa from the varieties of A. lentiginosus with bladdery inflated, completely biloculate pods. The contrast appears so great that in other sections of Astragalus, these differences have been considered species- level indicators. When these two taxa are merged into the complex folds of A. lentiginosus, along with the widespread and morphologically diverse taxa allied with A. lentiginosus var. palans (some of which were also originally recognized as a species), they form a continuum of variation—morphologically and geographi- cally. In the Mojave and Sonoran Deserts, A. lentiginosus var. mokiacensis and A. lentiginosus var. maricopae Barneby form the southern end of the continuum with persistent, mostly straight, tubular, scarcely inflated pods and a septum extending from one-half to slightly over three quarters the width of the locule. In the Colorado Plateau vicinity, A. lentiginosus var. palans, forms the central and eastern axis of the continuum with deciduous, mostly straight to nearly 180° curved, tubular to triquetrous, scarcely inflated pods and a septum from one-half to slightly over three quarters the width of the locule. In the Great Basin, A. lentiginosus var. iodanthus (S. Watson) J.A. Alexander and A. lentiginosus var. pseudiodanthus (Barneby) J.A. Alexander form the western axis of the continuum with deciduous, mostly 180° curved, triquetrous, scarcely inflated pods and a septum from one-quarter to less than one-half the width of the locule. Barneby (1964) was the 216 J i i fTexas 3(1) first to recognize this similarity. In his uniquely succinct style, he stated "this pair of species [A. iodanthus and A. pseudiodanthus] posses no character which cannot be matched somewhere in A. lentiginosus” (p. 911) and *it is often difficult or nearly impossible to separate flowering material of [A. iodanthus] from the polymorphic A. lentiginosus, from which it differs principally in a tendency to dorsiventral and triquetrous compression of the fruit, which is never inflated and commonly very strongly incurved... however in A. lentiginosus var. palans, the lace-ellipsoid, little inflated pod varies from erect to decurved and its section varies from round to triangular, so that sometimes the only technical differential character that remains is the broader septum" (p. 959-960). Other varieties of A. iodanthus have been recognized in Barnbey (1964), Isely (1998) and Welsh (2007), but they are treated herein and in Alexander (2008) as synonyms. Results of population level morphological analyses in progress may warrant recognition of A. iodanthus var. diaphanoides Barneby and A. iodanthus var. vipereus Barneby at the varietal-level within A. lentiginosus. Astragalus lentiginosus var. micans Barneby, Leafl. W. Bot. 8:22. 1956. ProroLocur: “CALIFORNIA: lower slopes of sand dunes at southeast end of Eureka Valley, east of Inyo Mts., Inyo County, elevation 3050 ft. May 13, 1955 (fr.) John C. Roos 6354, and at the same place, ae ae i ape 9, Te = ), Munz & Roos 20851. o Rancho Santa Ana Bot. Gard.” Type: U.S.A. CALIFORNIA: 13 May 1955, J.C. R 354 ( , designated here: RSA Invo C » of Inyo 1001791; isoLecToTYPES: GHI, Kl, NY! [2 sheets], OSC!, RSA 1146941, SD, ya ip sheets], US 2483071, internet image!); U.S.A. CALIFORNIA: Inyo Co.: on sand dunes at S end of Eureka Valley, E of Inyo Mts., 9 Apr 1955, P Munz & J.C. Roos 20581 (errrvre, designated here: RSA 100180!; isoEprrvees: NY!, OSC!, RSA 1099741, RSA 115975!, UCD). Notes.—Barneby (1956) eed two Ones (as p in the protelogue: one dios dim as C. Roos oi RSA) e one with e e & Roos 2058 I rneby t as es dAsd 7" lectotype Maleh (9007) al 1 R hire f, 1 u 4 hoth * r KDCAIT/An 1 L 1 (Me Nlei3ll et al 1 as npe. ap mene ee Roos a 1 79) is designated yp 2006, Article 9.2 and 0] 1 s 1 1 1 f Leal c i f, J ES fq 4E. on the T syntype. To keep all of tl ial used in the original type description (and the original i rol theanthon intact, the flowering specimen Mite & Roos 20581 (RSA100180) is selected h the epitype (McNeill et al. 2006, Article 9.7) It should be noted that in his original type publication, Barneby (1956) mistakenly cited “Munz & Roos 20851” as the syntype at RSA. Also Welsh (2007) mistakenly stated that the RSA syntype was “Munz & Roos 20815”. All syntype specimens at RSA bear the correct collection number, Munz & Roos 20581. Also all isosyntypes at various institutions listed above are labeled with the collection number, Munz & Roos 20581. There do not appear to be any types with labels bearing the typographical errors published by Barneby or Welsh. Astragalus lentiginosus var. oropedii Barneby, Leafl. W. Bot. 4:135. 1945. Prorotocue: “Kaibab Trail to Roaring Springs, Grand Canyon National Park, Coconino County, Arizona, 22 September 1938, fruct., Eastwood & Howell No. 7064 (Herb. Calif. Acad. Sci. No. 262056). Also ibid., 23 June 1933, flor., Eastwood & Howell No. 1054 (Herb. Calif. Acad. Sci. No. 211208, co- TyPE).” Type: U.S.A. Arizona: Coconino Co.: Kaibab Trail to Roaring Springs, 22 Sep 1938, A. Eastwood & J.T. Howell 7064 (LECTOTYPE, designated by Barneby 1989:158: CAS 2620561); U.S.A. Arizona: Coconino Co.: Kaibab Trail to Roaring Springs, 23 Jun 1933, A. Eastwood & J.T. Howell 1054 (errrveE, designated here: CAS 211208!) Notes. mr (1945) designated two syntypes (as cotypes) in the protologue: one us i ee & ee oe baie. and one with flowers (Eastwood & Howell 1054, CAS). In his monograph, pao (19 p pet For the Intermountain Flora, Barneby (1989) listed E & Ho oe 7064 as the iE for A. lentiginosus var. oedi bini winch rami isa latory pintaron (McNeill et al 2006, Article 9.8). The flowers of A. lentiginosus var. oropedii are the À pityr led for precise taxonomic application of this rua cn D of A. lentiginosus var. d herein as the floweri designated by Barneby as a sy twood & Howell 1054 (CAS; McNeill 1 1 ; 1 2a] Jin R hw ioinal d rintion Y fa A E C» et al. 2006, Article 9.7), a d c Astragalus lentiginosus var. pseudiodanthus (Barneby) J.A. Alexander, comb. nov. Astragalus pseudiodanthus TT iur W. Bot. H a m s iodanthus var. jum die enr Isely, Syst. Bot. 8:422. 1983. PRorOLOGUE: "NEVA f Cactus Peak outheast of Tonopah, Nye Co.: alt. 5500 ft, 3 Jun 1941, Ripley & Barneby N 3725." Tre: U.S.A. Nevapa: Nye Co.: plateau N vet Cactus Peak, about 20 mi SE of Tonopah, 3 Jun 1941, H.D. Ripley, R.C. Barneby 3725 (notorvee: CAS 2904051; isotype: RSA, PO fd l } t I d A. lentiginosus the al t of A. lentigi- F F Agent ast m Notes. discussion t nosus var. iodanthus. Alexander, Types of Astragalus sect. Diphysi 217 Barneby (1964) was the first to describe in detail of the gas 2 nen of A. io den var. pseudio- danthus and some populations of A. lentiginosus var. iodanthus. T por but spread throughout its range. It is likely that A. lentiginosus var. pseudiodanthus recently diff iated from several of these relictual, intermediate populations and became adapted to a stabilized sand dune habitat. Similar adaptations have occurred in populations of A. lentiginosus var. variabilis Barneby, A. lentiginosus var. fremontii (A. Gray ex Torr.) S. Watson and A. lentiginosus var. stramineus (Rydb.) Barneby (which may itself be a sand dune derivative of A. lentiginosus var. fremontii or A. lentiginosus var. vitreus Barneby). Population level morphological analyses in progress, leading to an overall monograph of Section Diphysi, may provide more details on the taxonomic status of these sand dune variants. ACKNOWLEDGMENTS Numerous individuals provided advice, funding and encouragement during various stages of this project including: Wesley E. Niles; Aaron Liston; Richard Halse; Kathryn Birgy of the UNLV Herbarium; Arnold Tiehm and the Northern Nevada Native Plant Society (and the NNNPS Small Grants Program); Lisa Decesare of the Library of the Gray Herbarium, Harvard University; and the Moldenke Fund for Plant Systematics and the Hardman Fund at Oregon State University. Additional thanks go to the late R.C. Barneby and the ond dun for SUM in Legume Systematics, which funded a month of research at NY. The following lted for this project: CAS, DS, GH, JEPS, K, MO, NY, OSC, ORE, RSA, SD, UC, UCR, UCSB, UNIV, US, WILLU. I thank the curators and staff of the following herbaria for access to their collections or loans of their specimens: BRY, CAS, DS, GH, JEPS, K, NY, OSC, ORE, RENO, RM, RSA, UC, UNLV, US, WILLU. Arnold Tiehm and one anonymous reviewer provided useful reviews. REFERENCES ALEXANDER, J.A. 2008. A taxonomic revision of Astragalus mokiacensis and allied taxa within the Astragalus lentigi- nosus complex of Section Diphysi. Ph.D. dissertation. Oregon State University, Corvallis, Oregon. Barnesy, R.C. 1944. Pugillus Astragalorum alter. Proc. Calif. Acad. Sci. 4:147-167 Barnesy, R.C. 1945. Pugillus Astragalorum - IV: the section Diplocystium. Leafl. W. Bot. 4:65-147. Barney, R.C. 1956. Pugillus Astragalorum - XVII: four new species and one variety. Leafl. W. Bot. 8:14-23. Barnesy, R.C. 1964. Atlas of North American Astragalus. (2 volumes). Mem. New York Bot, Gard. 13:1—1188. Barney, R.C. 1989. Intermountain flora. Volume III Part B. A. Cronquist, A.H. Holmgren, N.H. Holmgren, J.L. Reveal, PK. Holmgren, eds. The New York Botanical Garden. Bronx, New York. Gray, A. 1849. Plantae Fendlerianae Novi-Mexicanae. Mem. Amer. Acad. Arts, Ser. 2, 1:1-116. Gray, A. 1863. A Revision and arrangement (mainly by the fruit) of the North American species of Astragalus and Oxytropis. Proc. Amer. Acad. Arts 6:188-237 HARVARD University HErBARIA. 2008. Index of botanical specimens. . Accessed Apr 2002, Jul 2007. New York BOTANICAL GARDEN. 2008. The CV. Starr Virtual Herbarium. . Accessed April 2002, July 2007, August 2008. UNIVERSITY AND JEPSON HERBARIA. 2008a. Consortium of California herbaria. . Accessed August 2008. UNIVERSITY AND JEPSON HERBARIA. 2008b. Type specimens at the herbaria.